Development of a Deep Vein Valve Replacement

Development of a Deep Vein Valve

Replacement

Thesis submitted to Imperial College London for the degree of Doctor of

Philosophy

Miss Hayley Moore MA (Cantab), MBBS (AICSM), MRCS (Eng)

Registration 1st October 2010

Supervisors

Professor Alun H Davies (Vascular Surgery)

Professor Molly Stevens (Biomedical Materials)

Professor Yun Xu (Biofluid Mechanics)

Mr Manj Gohel (Vascular Surgery)

CONTENTS

Contents ...... 2

Abstract ...... 8

Acknowledgements ...... 10

Statement of Originality ...... 11

List of Tables ...... 12

List of Figures ...... 14

Abbreviations...... 22

1. Introduction ...... 26

1.1 Background ...... 26

1.2 The Aetiology of deep venous disease ...... 27

1.2.1 Anatomical locations of deep venous obstruction ...... 29

1.2.2 Anatomical locations of deep venous reflux ...... 34

1.3 Post-thrombotic syndrome ...... 42

2. Aims ...... 61

3. Deep venous disease ...... 63

3.1 Current Treatment options for Patients with Deep venous Disease ...... 63

3.2 Implantable Venous Valves for Deep Venous Insufficiency – the Story so Far ...... 72

4. The Anatomy of Deep Vein Valves ...... 91

4.1 Review of the Literature ...... 91

4.1.1 Introduction ...... 91

4.1.2 Methods ...... 91

4.1.3 Results ...... 92

4.1.4 Discussion...... 96

5. Damaged venous valves ...... 99

5.1 A Survey on the management of patients with complex venous reflux and obstruction ...... 99

5.1.1 Introduction ...... 99

5.1.2 Methods ...... 100

5.1.3 Results ...... 100

5.1.4 Discussion...... 118

6. Imaging the Venous System to Develop A Computational Model of A Deep Vein Valve ...... 119

6.1 Introduction...... 119

6.1.1 Venous Haemodynamics ...... 119

6.1.2 Flow Direction ...... 120

6.1.3 Venous Pressure (Normal, supine, stationary, walking) ...... 124

6.2 Imaging in Venous Disease ...... 136

6.2.1 Introduction ...... 136

6.2.2 Ultrasound ...... 136

6.2.3 Laser Doppler ...... 145

6.2.4 Plethysmography ...... 145

6.2.5 Venography ...... 147

6.2.6 Intravascular Ultrasound (IVUS) ...... 149

6.2.7 Computed Tomography Venography (CTV) ...... 150

6.2.8 Magnetic Resonance Venography (MRV) ...... 151

6.2.9 Image Guidance for venous interventions ...... 154

6.2.10 Venous Malformations ...... 155

6.2.11 Other techniques ...... 155

6.3 Ultrasound Imaging of Deep Vein Valves ...... 155

6.3.1 Background ...... 155

6.3.2 Methods ...... 156

6.3.3 Results ...... 160

6.3.4 Discussion...... 169

6.4 Magnetic Resonance Imaging of Deep Vein Valves ...... 169

6.4.1 Introduction ...... 169

6.4.2 Methods ...... 170

6.4.3 Results ...... 175

6.4.4 Discussion...... 178

6.5 Computational Modelling of Deep Vein Valves ...... 179

6.5.1 Introduction ...... 179

6.5.2 Methods ...... 181

6.5.3 Results ...... 185

6.5.4 Discussion...... 195

7. Development of a Novel Material for Use in the Venous System ...... 197

7.1 Cardiac Valve & Vascular Graft Engineering: previous work ...... 197

7.2 Selection of Polymer ...... 204

7.3 Aims ...... 207

7.4 Patent Search ...... 207

7.5 Polymer synthesis ...... 208

7.5.1 Introduction ...... 209

7.5.2 Methods ...... 209

7.5.3 Results ...... 214

7.6 Electrospinning MPC-HPMA-TMSPMA ...... 220

7.6.1 Scanning Electron Microscopy Imaging of Deep Vein Valves ...... 220

7.6.2 Optimisation of Electrospinning the Polymer ...... 223

7.6.3 Electrospinning a Bicuspid Valve Shape ...... 229

7.7 Haemocompatibility Testing ...... 230

7.7.1 Thrombogenicity Testing of MPC-HPMA-TMSPMA ...... 230

7.7.2 Complement Convertase Assay ...... 241

7.7.3 Haemolytic Assay ...... 247

7.8 Fourier Transform Infrared Spectroscopy ...... 253

7.8.1 Background ...... 253

7.8.2 Methods ...... 253

7.8.3 Results ...... 253

7.9 Mechanical Property Testing of materials ...... 256

7.9.1 Background ...... 256

7.9.2 Methods ...... 256

7.9.3 Results ...... 258

7.10 Discussion ...... 260

8. Development of A laboratory Venous Flow Model ...... 264

8.1 Introduction...... 264

8.2 Methods ...... 264

8.2.1 Set up of Laboratory Flow Model ...... 264

8.2.2 Validation of Laboratory Flow Model ...... 269

8.3 Results ...... 270

8.3.1 Flow Meter Calibration...... 270

8.3.2 Pressure Gauge Calibration ...... 272

8.3.3 Validation of the Laboratory Flow Model...... 273

8.3.4 Discussion...... 278

9. Future Work ...... 280

9.1 Imaging of the Venous System ...... 280

9.2 Development of a Computational Model of Venous Flow ...... 280

9.3 Development of a Laboratory Flow Model ...... 281

9.4 Development of MPC-HPMA-TMSPMA ...... 282

9.5 Potential Secondary Research Avenues ...... 283

Summary of Thesis Achievements ...... 284

Publications ...... 284

Presentations ...... 286

INTERNATIONAL ...... 286

NATIONAL ...... 290

10. References ...... 293

Appendices ...... 308

ABSTRACT

Background

Chronic venous disease is a common, distressing and significant cause of health care expense. There have

been few developments in the treatment of deep venous disease as the understanding of the clinical and

pathophysiological significance of deep vein reflux and valve failure remains poor. Previous attempts to

develop a prosthetic vein valve implant have been disappointing. Difficulties with early thrombosis led

researchers to abandon their efforts many years ago. Attempts to create a valve implant should be

revisited.

Aims

The aims of this project are to: evaluate variables around normal deep vein valves, to develop validated

computational and laboratory flow models for deep venous function, and to develop and investigate a

novel material to engineer a prototype bioprosthetic deep vein valve replacement.

Methods

Functional Anatomy: This is a prospective observational study evaluating subjects with normal deep

veins. B and M Mode ultrasound, contrast (microbubble) enhanced ultrasound and dynamic magnetic

resonance imaging of normal subjects was carried out. This has given the flow, velocity data and

anatomical images required for the project.

Modelling: A preliminary computational flow model has been developed using the data obtained from the

imaging stage of the project. This is a 2-dimensional model incorprating flexible valve leaflets. A laboratory model of venous function, in the form of a flow rig has been created.

Materials: Presently, polymers and polymer coated metal stents, used in the vascular system have several

problems: they are very thrombogenic and they lack haemocompatibilty and biocompatibility, in addition they lack the required mechanical properties. A novel material that is biocompatible, a copolymer of methacrylolyoxyethyl phosphorylcholine (MPC), trimethylsilyl-2-propyl methacrylate (TMSPMA) and

Hydroxypropyl methacrylate (HPMA), has been synthesised. Its properties have been modified by electrospinning and crosslinking to change its solubility and mechanical properties, without altering its biocompatibility.

Impact This project aims to guide the development of a treatment for patients, for whom few options are available. Chronic venous disease and venous ulceration are painful and debilitating, potentially requiring years of treatment. Effective, minimally invasive treatment options could result in accelerated ulcer healing and improvements in symptoms and quality of life as well as reduced costs.

ACKNOWLEDGEMENTS

I would like to thank the many people who have helped me both with the project and through the PhD.

I could not have done it without you. Firstly, my Professor, my supervisor, my mentor, and most importantly my friend, Professor Alun Davies; no one else would have got me through this and, if you’ll have me back, it would be fantastic to work for you again. I am grateful to my other supervisors

Professors Molly Stevens and Yun Xu and Mr Manj Gohel for their time, support and expertise. I thank all the members of the Academic Section of Vascular Surgery; Mr Ian Franklin, Mr Joseph Shalhoub, Mr

Ankur Thapar, Mr Muzzafar Anwar, Mr Tristan Lane, Mr Brahman Dharmarajah, Miss Amanda

Shepherd, Mr Chung Lim and Miss Sarah Onida for their encouragement. For their help with all of the ultrasound imaging, I thank Mary Ellis and Professor Edward Leen. The MR Imaging would not have been possible without the expertise and time of Dr Iain Pierce, Dr Peter Gatehouse, Dr Mark Haake and

Dr David Utrainen. For the Computational modelling I thank Dr Ying Wang and Miss Patrizia Lee and for the laboratory modelling I thank Mr John Wilson and Professor James Moore Jr. Finally for the materials project I am eternally grateful to Dr Seth McCullen, Mr Joseph Steele, Mr Gabriel Mecklenburg,

Dr Jon Weaver, Mr Anthony Macon and Miss Rachel Harbron, and of course the whole of the Stevens

Group, for their technical help and polymer expertise.

An extra special thank you goes to my parents who have lived through this with me, worried when I have and not slept sometimes, and who will be more relieved than anyone that this is finished! My husband

John Harrison, from whom I am now sadly separated, I love you dearly for all of the help and support that you have given me and for still just being there. My best friend Kate Williams, without you I would not have made it through the last few years and certainly not through the last couple of months – thank you. And finally to my brother, sister-in-law, nephews and all my family and friends who have been at the end of the phone with a friendly word or a pint when things got tough.

STATEMENT OF ORIGINALITY

The material presented in this report is the original work of the author. Where results or diagrams have been reproduced from the work of others, the source is clearly stated.

The copyright of this thesis rests with the author and is made available under a Creative Commons

Attribution Non-Commercial No Derivatives licence. Researchers are free to copy, distribute or transmit the thesis on the condition that they attribute it, that they do not use it for commercial purposes and that they do not alter, transform or build upon it. For any reuse or redistribution, researchers must make clear to others the licence terms of this work.

LIST OF TABLES

Table 1: Shows the 18 segments of the deep venous system9 ...... 27

Table 2: Shows the clinical severity of venous disease and the occluded deep venous segments from the literature ...... 32

Table 3: Shows the clinical severity of venous disease and the distribution of venous reflux ...... 37

Table 4: Shows the risk of developing PTS in female patients. Pooled analysis of available data...... 47

Table 5: Shows the risk of developing PTS in DVT patients with proven thrombophilia. Analysis of available data80...... 49

Table 6: Shows the incidence of PTS in patient with and without asymptomatic deep venous thrombosis80...... 50

Table 7: Shows the comparison of the properties of the different scoring systems for PTS98 ...... 53

Table 8: Shows a summary of elastic compression stocking for the prevention of PTS. Description of study designs and results ...... 56

Table 9: Shows the in vitro studies utilising a natural valve material ...... 77

Table 10: Shows the in vitro studies utilising a synthetic valve material ...... 80

Table 11: Shows the animal studies utilising a natural valve material...... 84

Table 12: Shows the in human clinical studies ...... 88

Table 13: Shows a summary of the number and location of valves within the femoral veins ...... 93

Table 14: Shows the summary of number and location of valves within the popliteal veins ...... 95

Table 15: Shows the position of the valve measured and summary of the maximum velocities obtained in both recumbent and standing positions for each subject ...... 163

Table 16: Shows the RV spiral scan parameters ...... 173

Table 17: Shows the diameter of vein, sinus and length of the valve leaflets ...... 185

Table 18: Shows the comparison between subject-specific experimental and computed velocity values ascertained at different positions relative to the deep vein valve ...... 194

Table 19: Shows the ratios of monomers combined to make both linear and branched versions of MPC-

HPMA-TMSPMA ...... 211

Table 20: Shows the known molecular weights and densities of the monomers used, provided in the product specification information ...... 214

Table 21: Shows the weights and volumes of monomers, anhydrous methanol and initiator used for each synthesis ...... 215

Table 22: Shows the diameter of fibres seen on the luminal surface of a bovine vein valve ...... 223

Table 23: Shows the variables for electrospinning polymer in methanol ...... 226

Table 24: Shows the samples tested with TGA assay, material, time in TEA and time in water ...... 233

Table 25: Shows the thrombin concentrations at 1, 2, 4 and 6 minutes corrected for negative controls . 235

Table 26: Shows the thrombin generation of the reference samples and sample materials ...... 237

Table 27: Shows the CCA results for reference materials and sample materials ...... 243

Table 28: Shows the haemoglobin concentration for reference materials and sample materials...... 249

LIST OF FIGURES

Figure 1: Shows the summary of the search strategy for the anatomical locations of chronic deep venous occlusions ...... 31

Figure 2: Shows the percentage of obstructions at the Inferior Vena Cava (IVC), Iliac Vein (IV), Femoral

Vein (FV), Popliteal Vein (PV) and other occlusions relative to the total number of obstructions for C0-3 and C4-6 disease ...... 33

Figure 3: Shows the summary of the search strategy for the anatomical locations of venous reflux ...... 36

Figure 4: Shows the anatomical location of venous reflux for C0-3 and C4-6 disease. Isolated superficial

(IS), isolated deep (ID), isolated perforator (IP), deep and superficial (DS), perforator (P), perforator and superficial (PS), perforator and deep (PD) and perforator, deep and superficial (PDS) ...... 38

Figure 5: Shows the total frequency of superficial, deep and perforator reflux for patients with C0-3 and

C4-6 disease. Superficial (S), Deep (D), Perforator (P)...... 39

Figure 6: Shows the relative involvement of the superficial, deep and perforating systems in patients with superficial venous reflux. Isolated superficial (IS), deep and superficial (DS), perforator and superficial

(PS) and perforator, deep and superficial (PDS)...... 39

Figure 7: Shows the relative involvement of the superficial, deep and perforating systems in patients with deep venous reflux. Isolated deep (ID), deep and superficial (DS), perforator and deep (PD) and perforator, deep and superficial (PDS)...... 40

Figure 8: Shows the relative involvement of the superficial, deep and perforating systems in patients with perforator reflux. Isolated perforator (IP), perforator and superficial (PS), perforator and deep (PD) and perforator, deep and superficial (PDS)...... 40

Figure 9: Shows the summary of the search strategy for literature regarding the post-thrombotic syndrome...... 44

Figure 10: Shows a summary of the management choices for patients suitable for deep venous surgery197

...... 71

Figure 11: Shows a summary of the search strategy for all studies of implantable deep venous valves ...... 74

Figure 12: Shows the most constant locations of valves within the femoral and popliteal veins ...... 97

Figure 13: Shows the duplex ultrasound imaging for Reflux Case 1 ...... 101

Figure 14: Shows the results for Reflux Case 1 - 1st Choice of management ...... 102

Figure 15: Shows the results for Reflux Case 1 – 2nd Choice of management ...... 102

Figure 16: Shows the duplex ultrasound imaging for Reflux Case 2 ...... 103

Figure 17: Shows the results for Reflux Case 2 - Investigations ...... 104

Figure 18: Shows the results for Reflux Case 2 - Management of the patient’s left leg ...... 104

Figure 19: Shows the results for Reflux Case 2 - Management of patient’s right leg ...... 104

Figure 20: Shows the duplex ultrasound imaging for Reflux Case 3 ...... 105

Figure 21: Shows the results for Reflux Case 3 - 1st Choice of management ...... 106

Figure 22: Shows the results for Reflux Case 3 – 2nd Choice of management ...... 106

Figure 23: Shows the duplex ultrasound imaging for Reflux Case 4 ...... 107

Figure 24: Shows the results for Reflux Case 4 - 1st Choice of management ...... 108

Figure 25: Shows the results for Reflux Case 4 – 2nd Choice of management ...... 108

Figure 26: Shows the duplex ultrasound imaging for Reflux Case 5 ...... 110

Figure 27: Shows the results for Reflux Case 5 - 1st Choice of management ...... 110

Figure 28: Shows the results for Reflux Case 5 – 2nd Choice of management ...... 111

Figure 29: Shows the duplex ultrasound imaging for Obstruction Case 1 ...... 112

Figure 30: Shows the results for Obstruction Case 1 – 1st Choice of management ...... 112

Figure 31: Shows the results for Obstruction Case 1 – 2nd Choice of management ...... 113

Figure 32: Shows the CT Venogram for Obstruction Case 2 ...... 114

Figure 33: Shows the duplex ultrasound imaging for Obstruction Case 2 ...... 114

Figure 34: Shows the results for Obstruction Case 2 – 1st Choice of management ...... 115

Figure 35: Shows the results for Obstruction Case 2 – 2nd Choice of management ...... 115

Figure 36: Shows the duplex Ultrasound Imaging for Obstruction Case 5 ...... 116

Figure 37: Shows the results for Obstruction Case 3 – 1st Choice of management ...... 117

Figure 38: Shows the results for Obstruction Case 3 – 1st Choice of management ...... 117

Figure 39: Show the velocity of blood flow in the femoral vein during quiet respiration in the supine position of a normal subject ...... 122

Figure 40: Shows the venous pressure-volume curve, reproduced from Katz et al, 1969256 ...... 126

Figure 41: Shows the venous hydrostatic pressures in the supine position (adapted from Meissner et al238)

...... 127

Figure 42: Shows the venous hydrostatic pressures in the standing position (Adapted from Meissner et al238) ...... 128

Figure 43: Shows the femoral vein flow variation according the respiratory pump and posture. Solid line:

Supine position, dotted line: upright standing position. (This image is adapted from an unpublished image provided by Claude Franceschi) ...... 132

Figure 44: Shows the velocity in the femoral vein in the supine position during respiration ...... 133

Figure 45: Valsalva manoeuvre (normal individual): The blocked expiration (BE) reverses strongly the pressure gradient that induces only a negligible short and small reflux thanks to the correct competence of

16 the underlying valve. (This image is adapted from an unpublished image provided by Claude Franceschi)

...... 134

Figure 46: Valsalva manoeuvre: pathological reflux (valsalva positive). Valsalva is Positive when valves are incompetent. Reverse Flow appears when blowing (BE) and at release (BR). (This image is adapted from an unpublished image provided by Claude Franceschi) ...... 134

Figure 47: Shows the measurement of venous velocity with ultrasound distal (a), within (b) and proximal to (c), valve leaflets and the velocity curve traced by one observer on ImageJ (d)...... 160

Figure 48: Shows the ultrasound images of venous valves. A - Normal respiration, standing position, B -

Standing position, calf squeeze, C - Semi recumbent position, normal respiration, D - Still of B mode video image showing valve leaflets...... 162

Figure 49: Contrast enhanced ultrasound of venous valve ...... 162

Figure 50: Shows the Bland-Altman analysis of the inter-observer reliability between observer 1 and observer 2 for mean velocity measurements ...... 165

Figure 51: Shows the Bland-Altman analysis of the inter-observer reliability between observer 1 and observer 2 for measurements of the duration of calf squeeze ...... 165

Figure 52: Shows the Bland-Altman analysis of the inter-observer reliability between observer 1 and observer 2 for maximum velocity measurements ...... 166

Figure 53: Shows the Bland-Altman analysis of the intra-observer reliability for observer 1 for maximum velocity measurements ...... 166

Figure 54: Shows the average of the maximum velocity for the 7 subjects relative to the valve leaflets using ultrasound ...... 167

Figure 55: Shows the average of the mean velocities for each of the 7 subjects relative to the valve leaflets using ultrasound ...... 168

Figure 56: Shows a volunteer ready for MRI, showing clockwise from top left; expiratory breath hold,

Kendal SCD express system calf pump, heel raised slightly from the bed to prevent calf compression and surface coil...... 171

Figure 57: Shows a transverse slice from the bSSFP 3D volume within the imaging region used, with vessels labelled, also noted are the femur and testicle for orientation. An artefact from the carotid coil can be seen above the anterior surface of the leg and the coil sensitivity can also be observed...... 172

Figure 58: Shows the average velocity across the femoral vein during a calf squeeze ...... 176

Figure 59: Shows the average of the maximum velocity for the 7 subjects relative to the valve leaflets using MRI ...... 176

Figure 60: Shows a comparison of the maximum velocity in the femoral vein measured by both MRI and ultrasound in the form of a Bland-Altman analysis ...... 178

Figure 61: Shows the contour plots of velocity distribution (cm/s) ...... 188

Figure 62: Shows the vector plots (vector size scaled to magnitude) for the velocity profile along the vein

(cm/s) ...... 190

Figure 63: Shows the vector plot of flow pattern for a) minimal opening of valve, b) maximum opening of valve; arrows indicate the location of vortices in one side of the pocket...... 191

Figure 64: Shows the contour plot of Wall Shear Stress (dyne/cm2) ...... 193

Figure 65: Shows a summary of the search strategy for previous cardiac valve and vascular graft engineering ...... 199

Figure 66: Shows that MPC reduces platelet adhesion, reproduced from Ye et al, 2003 395 ...... 206

Figure 67: Shows the structure of MPC ...... 209

Figure 68: Shows the structure of HPMA ...... 209

Figure 69: Shows the structure of TMSPMA ...... 210

Figure 70: Shows the laboratory set up for free radical polymerisation...... 213

Figure 71: Shows the clear polymer solution ...... 216

Figure 72: Shows the predicted structure of the polymer MPC-HPMA-TMSPMA ...... 217

Figure 73: Shows the NMR spectrum for the monomer MPC ...... 218

Figure 74: Shows the NMR spectrum for the monomer HPMA ...... 218

Figure 75: Shows the NMR spectrum for the monomer TMSPMA ...... 219

Figure 76: Shows the NMR spectrum for the linear version of the polymer MPC-HPMA-TMSPMA .... 219

Figure 77: Shows the scanning electron microscopy image of a bovine vein valve (luminal surface) ...... 222

Figure 78: Shows a diagrammatic representation of the electrospinning set up ...... 224

Figure 79: Shows the MPC-HPMA-TMSPMA fibres electrospun with 23G needle showing beading .... 229

Figure 80: Shows the electrospun bicuspid valve shape ...... 230

Figure 81: Shows the calibration curve of thrombin concentrations of the calibrators against the OD405

...... 234

Figure 82: Shows a line graph of thrombin concentrations at 1, 2, 4 and 6 minutes corrected for negative controls ...... 236

Figure 83: Shows the thrombin generation of reference samples and materials ...... 238

Figure 84: Shows the comparison of the thrombin generation of the reference materials and MPC50-

HPMA15-TMSPMA35 and MPC50-HPMA5-TMSPMA45 ...... 239

Figure 85: Shows the comparison of the thrombin generation of MPC-HPMA-TMSPMA cross-linked in

TEA for 1 hour, 24 hours and 48 hours ...... 240

Figure 86: Shows the comparison of the thrombin generation of MPC-HPMA-TMSPMA left in water for

0 days, 1 day or 7 days after cross-linking in TEA ...... 241

Figure 87: Shows the comparison of the CCA results of the reference materials and MPC50-HPMA15-

TMSPMA35 and MPC50-HPMA5-TMSPMA45 ...... 244

Figure 88: Shows the comparison of the CCA results of MPC-HPMA-TMSPMA cross-linked in TEA for

1 hour, 24 hours and 48 hours ...... 245

Figure 89: Shows the comparison of the CCA results of MPC-HPMA-TMSPMA left in water for 0 days,

1 day or 7 days after cross-linking in TEA ...... 246

Figure 90: Shows the comparison of the haemoglobin concentration of the reference materials and

MPC50-HPMA15-TMSPMA35 and MPC50-HPMA5-TMSPMA45 ...... 250

Figure 91: Shows the comparison of the haemoglobin concentrations of MPC-HPMA-TMSPMA cross- linked in TEA for 1 hour, 24 hours and 48 hours ...... 251

Figure 92: Shows the comparison of the haemoglobin concentrations of MPC-HPMA-TMSPMA left in water for 0 days, 1 day or 7 days after cross-linking in TEA ...... 252

Figure 93: Shows the effect of TEA on MPC50-HPMA15-TMSPMA35 (A) and MPC50-HPMA5-

TMSPMA45 (B) ...... 254

Figure 94: Shows the effect of water on MPC50-HPMA15-TMSPMA35 (A) and MPC50-HPMA5-

TMSPMA45 (B) ...... 255

Figure 95: Shows the tensile modulus of vein and MPC-HPMA-TMSPMA ...... 259

Figure 96: Breakpoint stress of vein and MPC-HPMA-TMSPMA ...... 260

Figure 97: Shows a schematic diagram of the laboratory flow model: Data Acquisition Assistant (DAQ);

Pressure Gauge (PG); Ultrasound (US) ...... 266

Figure 98: Shows an example of the DAQ assistant system which controls and collects data from the

model ...... 267

Figure 99: Shows the flow model set up in the laboratory ...... 268

Figure 100: Shows an example of the voltage input in graphical form. This represents subject 4, in the standing position with a calf squeeze and breath hold...... 269

Figure 101: Shows the flow meter calibration results comparing the pump RPM and the measured flow rate in ml/min ...... 271

Figure 102: Shows the flow meter calibration results comparing the measured flow rate in ml/min and the

RPM of output voltage of the flow meter ...... 271

Figure 103: Shows the calibration results for Pressure Gauge 1 showing the output voltage compared to the known pressure generated by the column of water ...... 272

Figure 104: Shows the calibration results for Pressure Gauge 2 showing the output voltage compared to the known pressure generated by the column of water ...... 273

Figure 105: Shows the calibration of the flow meter for voltage and RPM ...... 274

Figure 106: Shows a Bland-Altman analysis of maximum velocity measurements comparing the ultrasound data obtained from the subjects and the flow model ...... 275

Figure 107: Shows a Bland-Altman analysis of the mean velocity during the calf squeeze measurements comparing the ultrasound data obtained from the subjects and the flow model ...... 275

Figure 108: Shows a Bland-Altman analysis of the mean velocity measurements comparing the ultrasound data obtained from the subjects and the flow model ...... 276

Figure 109: Shows a graph of the maximum velocity from the subject data and the flow model showing the mean and 95% confidence intervals demonstrating a Spearman correlation r=0.7 ...... 277

Figure 110: Shows a graph of the mean velocity from the subject data and the flow model showing the mean and 95% confidence intervals demonstrating a Spearman correlation r=0.38 ...... 278

Figure 112: Shows the ultrasound image of the valve in the flow model ...... 282

ABBREVIATIONS

ADVT Asymptomatic Deep Vein Thrombosis

AIBN Azoisobutyronitrile

BMI Body Mass Index

BMP Bone Morphogenic Protein

BVV Bioprosthetic vein valve

BVV2 2nd generation bioprosthetic vein valve

CCA Complement Convertase Assay

CDI Carbodiimide

CDT Catheter Directed Thrombolysis

CDU Colour Duplex Ultrasound

CEAP Clinical severity, aetiology, anatomy and pathophysiology

CFD Computational Fluid Dynamics

CFV Common femoral vein

CJV Common jugular vein

CT Computed tomography

DFV Deep femoral vein

DICOM Digital imaging and communications in medicine

DS Deep Superficial

DVT Deep Vein Thrombosis

ECM Extra Cellular Matrix

ECS Elastic Compression Stockings

EJV External jugular vein

EMS Electrical muscle stimulation

FEM Finite Element Method

FOV Field of view

FSE Fast Spin Echo

FSI Fluid structure interaction

FV Femoral vein

HMDS Hexamethyldisilazane

HPMA Hydroxypropyl methacrylate

ID Isolated Deep

INR International Normalised Ratio

IS Isolated Superficial

IPC Intermittent pneumatic compression

IVC Inferior vena cava

LDPE Low-density polyethylene

LMWH Low Molecular Weight Heparin

MMP Matrix Metalloproteinase

MPC Methacryloyloxyethyl phosphorylcholine

MPFF Micronised purified flavonoid fraction

MR Magnetic resonance

NMR Nuclear magnetic resonance

NYHA New York Health Association

P Perforator

PCL Polycaprolactone

PDE Partial Differential Equations

PDMS Polydimethylsiloxane

PCU Polycarbonate urethane

PD Perforator Deep

PDS Perforator Deep Superficial

PEU Polyester urethane

PLLA poly(L-lactic acid)

POSS Polyhedral oligomeric silsesquioxane

PPMT Percutaneous Pharmacomechanical Thrombolysis

PRISMA Preferred reporting items for systematic reviews and meta-analyses

PS Perforator Superficial

PTFE Polytetrofluoroethylene

PV Popliteal vein

PVA Polyvinyl Alcohol

PVC Polyvinyl Chloride

PVVB Percutaneous vein valve bioprosthesis

RCT Randomised Controlled Trial

REC Research ethics committee

ROI Region of interest

SEM Scanning electron microscopy

SIBS styrene-block-iso butylenes-block-styrene

SIS Small intestinal submucosa

SFJ Saphenofemoral junction

SPJ Saphenopopliteal junction

TGF-β Transforming Growth Factor - beta

TMSPMA 2-trimethylsilyl-2-propyl methacrylate

TOF Time of Flight

TUS Triplex Ultrasound

UV Ultraviolet

VEGF Vascular Endothelial Growth Factor

VENC Velocity Encoding

VKA Vitamin K Antagonist

VVP Venous valve pocket

WSS Wall Shear Stress

1. INTRODUCTION

1.1 BACKGROUND

Chronic venous disease is common, distressing and a significant cause of health care expense1. It affects a third of the UK adult population and is costs the NHS approximately 2% of the annual budget. The cause is failure of the valves in the superficial or deep veins of the legs due to primary or secondary failure. Primary valve failure is caused by weakening of the vessel wall leading to dilatation and separation of the valve leaflets. Secondary valve failure is due to damage to the valves commonly due to venous thrombosis. Both of these result in refluxing disease. Symptoms include leg swelling, ‘bursting’ pain on walking (venous claudication), and venous ulceration. Although uncommon, amputation may be necessary in severe cases. The prevalence of venous ulceration in the UK and USA in the adult population is approximately 0.5-1%, and in patients over 80 years it increases to 3%2, 3.

Recent advances in the management of venous disease have focused on patients with superficial reflux,

particularly the use of endothermal technologies to ablate superficial veins4. Treatments can be performed

in an office-based setting and excellent early outcomes have been reported 5, 6. Duplex studies have shown that reflux (or less commonly, occlusion) of deep veins is seen in around 40% of patients with chronic

venous insufficiency and may be significant in 15-20%7, 8. Most vascular specialists will have managed

these challenging patients. There have been few developments in the treatment of deep venous reflux as

our understanding of the clinical and pathophysiological significance of deep vein reflux and valve failure

remains poor.

Attempts to create a valve implant must and are being revisited. Crucially, a non-thrombogenic,

biocompatible material with the necessary required properties and mechanical integrity for implantation

into the venous system is essential as well as a design that will minimise turbulence and zone stasis around

the valve leaflets and valve pocket.

1.2 THE AETIOLOGY OF DEEP VENOUS DISEASE

Veins are thin walled capacitance vessels that return blood from the peripheries towards the heart. The venous system consists of three systems; the deep venous system, the superficial venous system and perforating veins. This is divided into 18 segments as described in Table 1 below:

Table 1: Shows the 18 segments of the deep venous system9

Superficial veins Deep veins Perforating veins

1 Telangiectasias/reticular 6 Inferior vena cava 17 Thigh veins 2 Above knee Great 7 Common Iliac 18 Calf saphenous vein 3 Below knee Great 8 Internal Iliac saphenous vein 4 Small saphenous vein 9 External Iliac

5 Non-saphenous 10 Pelvic, gonadal, broad ligament 11 Common Femoral 12 Deep Femoral 13 Superficial Femoral 14 Popliteal 15 Tibial (anterior, posterior, or peroneal) 16 Muscular (gastrocnemial, soleal, other)

Superficial veins drain blood from the skin and subcutaneous fat, then via the perforating veins, this blood passes to the deep venous system and is returned to the heart10. Flow is maintained by the movement of the diaphragm during respiration, pressure gradients created by the thoracic and abdominal cavities and by peripheral pumps including the foot, calf and thigh pumps. Unidirectional flow is ensured by the presence of valves, which are usually bicuspid, throughout the system preventing retrograde flow, and of course patency of the veins ensuring no resistance to flow is essential. Undoubtedly, the venous system is more complex than the arterial system9.

The action of the foot, calf and thigh muscle pumps is essential to maintain normal venous function, as

during ambulation approximately 90% of venous return is attributed to these muscular structures11. The

calf muscle pump is the most important, as it has the largest capacitance and generates the highest

pressure, with an ejection fraction (EF) of approximately 65%. The thigh muscle pump has an EF of

approximately 15%11. During ambulation there is a pressure increase of 92% in the deep posterior compartment of the calf muscle pump, 104% in the superficial posterior compartment, and 18% in the anterior tibial compartment12. When the calf muscle relaxes, the venous pressure falls to between 15 and

30mmHg, when the presence of well functioning bicuspid valves is necessary to prevent retrograde flow13. Dysfunction or impairment of either the valves within the venous system or of the muscle pumps,

particularly the calf muscle pump, can lead to chronic venous disease13. In addition, any obstruction to the flow of blood towards the heart will result in increased resistance to the flow of blood and increased pressure within the veins, both during muscle contraction and at rest. This will also result in the clinical manifestation of venous disease.

It is unclear which of these factors considered to potentially be the cause of venous disease is the initiating factor, as many patients with venous disease have all of these conditions. Some patients have a clear trigger, such as a deep vein thrombosis, trauma, an anatomical narrowing or chronic inflammation, which is particularly the case in those patients with obstructive venous disease, however for many patients none of these is present. In patients with valve reflux, maybe a valve defect is the primary initiator or potentially a condition that affects the vein wall primarily, such as altered collagen composition and reduced elastin14. However, whatever the initiating factor, the underlying pathophysiology of CVD is chronic venous hypertension, causing oedema, pigmentation, fibrosis and ulceration of the skin and subcutaneous tissues13.

The CEAP classification system, clinical class (C), etiology (E), anatomical (A) distribution of disease and underlying pathophysiology (P) is commonly used to sub-classify patients with venous disease. The clinical (C) classification is the most commonly used, which includes patients with no disease (C0) telangiectasiae (C1), varicose veins (C2), oedema (C3), liperdermatosclerosis (C4), healed ulcers (C5) and active ulcers (C6)15. This chapter will review the literature to determine if there is a correlation between

the anatomical locations of deep venous reflux and obstruction with clinical disease severity and review

the available literature on the Post Thrombotic Syndrome. The locations of reflux are divided into its

presence in the deep, superficial and perforating veins, where as the locations of obstruction are

specifically assessed as to their location in the deep veins only.

1.2.1 Anatomical locations of deep venous obstruction

Chronic deep venous obstruction impairs outflow from the lower extremities16, 17. It can be caused by the

absence of or incomplete recanalisation of the vein following a DVT, which can be the case in up to 80%

of patients with an iliac vein thrombosis16, or by anatomical variants such as May-Thurner syndrome, venous agenesis associated with conditions such as Klippel-Trenaunay syndrome, trauma, iatrogenic causes, and tumours18. Here, the anatomical location of deep venous obstruction in the deep venous

system is summarised and correlated with disease severity, from the evidence available in the literature.

Methods

The review was conducted according to the preferred reporting items for systematic reviews and meta-

analyses (PRISMA) guidelines19. Relevant papers were identified through systematic searches of Pubmed

(1948 to June 2013). The search terms “chronic venous obstruction”, “chronic venous occlusion”, “deep

vein thrombosis” AND “lower limbs”, “legs” were used. Article titles and abstracts were screened for

inclusion. A manual reference list search was also carried out for further appropriate studies to be

considered for inclusion, ensuring relevant papers published before 1966 were included. Relevant full text

articles were scrutinised and all studies reporting details of cross-sectional and longitudinal studies

regarding venous obstruction or occlusion in the lower extremities including the prevalence within the

deep venous segments were included. Case studies, reviews and letters were excluded, as were articles concerning the arterial system, acute venous occlusion, acute deep vein thrombosis or venous reflux or incompetence, and animal studies. No filters, limits or language exclusions were applied.

Results

The majority of studies used duplex ultrasound to establish the location of venous occlusion. The results

of the search are shown in Figure 1. From 8620 studies identified at the initial search, 2 were included, as only 2 allowed correlation between disease severity and the anatomical location in the context of chronic obstruction. Many studies included detailed assessments of the anatomical locations of the obstruction, but did not correlate this with clinical severity. Several studies followed up patients with acute DVT of known anatomical location and assessed clinical severity of their ongoing venous disease, however, they did not assess the extent of the ongoing occlusion at these follow up visits.

Figure 1: Shows the summary of the search strategy for the anatomical locations of chronic deep venous occlusions

Search terms entered: “chronic venous obstruction” OR “chronic venous occlusion” OR “deep vein thrombosis” AND “lower limbs” OR “legs”

Records identified through database Additional records identified through searching other sources (n = 8620 publications (n =5) between 1948 and 2013)

Records excluded Records screened (title) (n =8589) (n = 8625) Exclusion criteria: Case studies, reviews and letters, arterial system, acute venous occlusion, acute deep vein thrombosis or venous reflux or incompetence

Records excluded (n = 26) Abstracts and full-text Excluded: Articles lacking articles assessed for prevalence data eligibility (n = 36) Records excluded (n = 8) Excluded: Articles lacking clinical severity data

Studies included (n = 2)

The occluded segments in the IVC, IV, FV, and PV relative to the total number of limbs in that group are

shown as well as these numbers as a percentage of the total number of isolated obstructions depending

on the severity of chronic venous disease in Figure 2. All other occluded deep vein segments were

included in the category of other obstructions, these were occlusions of the tibial and muscular veins.

Johnson et al20 had 49 patients in the asymptomatic group and 34 patients in the symptomatic group who

were diagnosed with PTS. Labropoulos et al21 included 120 limbs of 105 patients all of which were

followed up with duplex ultrasound at least one year following the acute DVT. 85 limbs demonstrated

C0-C3 disease and 35 limbs had C4-C6 disease.

Table 2 summarises these studies and describes the individual occluded venous segments. The

summation of these papers is shown in Figure 2.

Table 2: Shows the clinical severity of venous disease and the occluded deep venous segments from the

literature

Reference Number IVC IV obstruction FV obstruction PV obstruction Other

of patients obstruction occlusions

(limbs) C0-3 ≥C4 C0-3 ≥C4 C0-3 ≥C4 C0-3 ≥C4 C0-3 ≥C4

Johnson et 78 (83) - - 5/49 5/34 24/49 18/34 14/49 18/34 5/49 9/34 al, 1995 20

Labropoulos 105 (120) 10/85 5/35 22/85 15/35 58/85 30/35 50/85 27/35 39/85 17/35 et al, 2008 21

Figure 2: Shows the percentage of obstructions at the Inferior Vena Cava (IVC), Iliac Vein (IV), Femoral

Vein (FV), Popliteal Vein (PV) and other occlusions relative to the total number of obstructions for C0-3 and C4-6 disease

40 C0-3 C4-6

segment (%) 30

10 Percentage of Percentage of occludedlimbs with 0 IVC IV FV PV Other Occluded Venous Segment

From the literature, the percentage of occluded segments is higher in patients with more severe CVD, as shown in

Figure 2. Labropoulos et al21 were the only group to distinguish between isolated vein thrombosis and multi- segment occlusions. They found that patients with multi-segment occlusions had more advanced PTS. Femoral and popliteal veins are the most commonly occluded segments.

Discussion

This demonstrates that there is a paucity of literature available on the anatomical locations of chronic

venous obstruction. There are many studies of acute DVT, its location and ongoing follow up and

assessment of patients with post thrombotic syndrome, but few perform duplex imaging in the follow up

assessment to ascertain the ongoing presence of chronic occlusion. There are several studies which

describe the locations of chronic occlusions, however, these do not provide any correlation with clinical

disease severity. It would seem logical that the presence of venous obstruction contributes to increased

venous disease severity and furthermore that the more proximal the occlusion, the more severe the

venous disease. This has been demonstrated in the 2 studies here. Multisegmental occlusions appear to

be more common than isolated obstructions, which result in more severe disease. It could be argued that

as one of the causes of chronic venous obstruction is an acute DVT with failure of recanalisation, that it

the location of the initial DVT correlates well with disease severity, and there are studies which

demonstrate that the more proximal the DVT, the worse the severity of venous disease22. However, many DVTs may be asymptomatic and development and recanalisation rates for patients in the studies of acute DVT that were excluded were not known as duplex imaging was not performed, this data cannot be extrapolated for the purpose of chronic occlusions discussed here. Isolated venous obstruction is rare and more commonly obstruction and reflux are present in the same subject. In patients with

multisegmental obstructions there is an increased incidence of venous reflux23.

1.2.2 Anatomical locations of deep venous reflux

Venous reflux occurs when the return of blood to the heart flows in a retrograde direction. This may be

secondary to valve damage, or an abnormality of the vein wall or if obstruction is present concurrently

that the occluded veins dilate preventing the valve leaflets meeting and serving as a functional valve, as suggested by Van Bemmelen et al24. By whichever mechanism it occurs, when the valves fail, venous insufficiency results. The available literature is reviewed here to evaluate the anatomical locations of

venous reflux and correlate this with the clinical severity of venous disease, specifically the presence of

reflux in the three compartments; the deep, superficial and perforating veins.

Methods

The review was conducted according to the preferred reporting items for systematic reviews and meta-

analyses (PRISMA) guidelines19. Relevant papers were identified through systematic searches of Pubmed

(1948 to June 2013). The search terms “chronic venous insufficiency”, “chronic venous reflux”, AND

“lower limbs”, “legs” OR “lower extremities” AND “vein* OR venous” AND “reflux OR

incompetence” were used. Article titles and abstracts were screened for inclusion. A manual reference list

search was also carried out for further appropriate studies to be considered for inclusion, ensuring

relevant papers published before 1948 were included. No filters or limits were applied to the search.

Relevant full text articles were scrutinised and all studies reporting details of cross-sectional and

longitudinal studies regarding the prevalence of venous disease in the lower extremities, the severity of

venous disease and a study population age group of greater than 18 years old were included. Case studies,

reviews and letters were excluded, as were articles concerning the arterial system, acute venous occlusion,

acute deep vein thrombosis or chronic venous occlusion, and animal studies. The data was grouped

according to the anatomical location of venous reflux and clinical disease severity according to the CEAP

classification, as described previously.

Results

As shown in Figure 3, from 1728 titles screened 14 were included. The data from these studies was extracted, and the presence of reflux in the deep, superficial and perforating veins was compared in those patients with C0-C3 disease and C4-C6 disease. The studies that were excluded at the full text stage generally contained data which could not be interpreted, or related only to the superficial system, thus not allowing adequate comparison, related only to perforating veins or combined both reflux and obstruction which could not be separated. The patterns of reflux that were considered for this study were isolated superficial (IS), isolated deep (ID), deep and superficial (DS), perforator (P), perforator and superficial

(PS), perforator and deep (PD) and perforator, deep and superficial (PDS). This resulted in a total of

6581 patients and 7081 limbs included, with 5244 limbs (74.1%) in the C0-3 and 1837 limbs (25.9%) in the C4-6 group. These results are shown in Table 3.

Figure 3: Shows the summary of the search strategy for the anatomical locations of venous reflux

Search terms entered: “chronic venous insufficiency”, “chronic venous reflux”, AND “lower limbs”, “legs” OR “lower extremities” AND “vein* OR venous” AND “reflux OR incompetence”

Records identified through database Additional records identified through searching other sources (n = 1728 publications (n =16) between 1948 and 2013)

Records excluded Records screened (title) (n =1683) (n = 1744) Exclusion criteria: Case studies, reviews and letters, arterial system, acute venous occlusion, deep vein thrombosis, chronic venous occlusion, animal studies

Records excluded (n = 30) Abstracts and full-text Excluded: Articles not relating articles as sessed for to reflux alone eligibility (n = 61)

Records excluded (n = 17) Excluded: Articles lacking adequate clinical severity data

Studies included (n = 14)

Number Number No Total Authors and publication year CEAP IS ID DS Total S PS PD PDS Total P patients limbs reflux D IP

Cornwall et al, 198625 C4-6 100 117 12 40 22 9 49 32

Darke and Penfold, 199226 C4-6 213 232 90 81 90 81 9 180

C0-3 202 45 132 17 8 8 Lees and Lambert, 199327 153 C4-6 98 2 57 39 2 2

Shami, 199328 C4-6 59 118 42 12 25 67 37 0

Labropoulos et al, 199529 C4-6 94 112 4 26 7 13 93 56 23 5 31 3 62 C0-3 776 192 429 18 137 566 155 356 107 78 441 Myers et al, 199530 1114 C4-6 166 20 63 13 70 228 142 95 59 20 110

C0-3 319 168 129 7 12 144 19 3 0 0 0 3 Labropoulos et al, 199631 465 C4-6 275 10 57 12 48 246 163 42 4 99 3 148 C0-3 218 80 92 14 14 92 28 Welch et al, 199632 320 C4-6 282 44 67 30 59 126 89

Labropoulos et al, 199733 C0-3 110 220 91 108 4 7 125 13 8 0 2 0 10 C0-3 694 537 144 413 16 16 511 48 76 10 6 Giannoukas et al, 200234 C4-6 117 11 3 30 92 72 36 24 15

Danielsson et al, 200435 C4-6 83 98 1 3 4 11 84 72 19 6 51 3 79 C0-3 121 66 44 21 12 Ibegbuna et al, 200636 C4-6 66 16 50 42

C0-3 3014 2906 406 353 Maurins et al, 200837 C4-6 108 74 33

Kanchanabat et al, 201038 C4-6 41 48 4 8 32 36 40

Key: Isolated superficial (IS), isolated deep (ID), isolated perforator (IP), deep and superficial (DS), perforator (P), perforator and superficial (PS), perforator and deep (PD) and perforator, deep and superficial (PDS)

Table 3: Shows the clinical severity of venous disease and the distribution of venous reflux

Once this data is combined as shown in Figure 4, it is demonstrated that in the presence of isolated

superficial reflux, C0-3 disease, that is the presence of no clinical venous disease, telangiectasia, varicose

veins or leg oedema, is more common. However, if deep venous reflux, perforator reflux, or two or

more of the three venous systems are involved, then C4-6 disease is more likely. For all stages of venous

disease, for all CEAP classifications, reflux occurred most commonly in the superficial veins

compartments (46.3% of the total limbs). The frequency of superficial, deep and perforator reflux was

higher in all cases for patients with C4-6 disease when compared to patients with C0-3 disease, as shown in Figure 5. For patients with C0-3 disease, only 12% had an element of deep venous reflux and 15% had perforator reflux, whereas for patients with C4-6 disease, 49% and 36% had deep and perforator reflux respectively. Only 0.38% of patients with reflux in all 3 lower limb venous system (PDS) had C0-3 disease, but this was seen in 17% of patients with C4-6 disease.

Figure 4: Shows the anatomical location of venous reflux for C0-3 and C4-6 disease. Isolated superficial

(IS), isolated deep (ID), isolated perforator (IP), deep and superficial (DS), perforator (P), perforator and superficial

(PS), perforator and deep (PD) and perforator, deep and superficial (PDS)

C0-3 10 C4-6

5 Percentage of total limbs (%) total of Percentage

0 IS ID IP DS PS PD PDS Anatomical location of reflux

Figure 5: Shows the total frequency of superficial, deep and perforator reflux for patients with C0-3 and C4-6 disease. Superficial (S), Deep (D), Perforator (P).

40 C0-3 30 C4-6 20

Percentage of total limbs (%) total of Percentage 10

0 Total S Total D Total P Anatomical location of reflux

Figure 6: Shows the relative involvement of the superficial, deep and perforating systems in patients with superficial

venous reflux. Isolated superficial (IS), deep and superficial (DS), perforator and superficial (PS) and perforator,

deep and superficial (PDS)

(%) 30 C0-3 C4-6 20

Percentage of Percentage of Superficial Total Reflux 0 IS DS PS PDS Anatomical location of reflux

20 C0-3 15 C4-6 10

Percentage of Percentage of DeepTotal Reflux (%) 0 ID DS PD PDS Anatomical location of reflux

(%) 30 C0-3 C4-6 20

Percentage of Percentage of PerforatorTotal Reflux 0 IP PS PD PDS Anatomical location of relux

If the involvement of each venous compartment is compared individually, for patients with superficial venous reflux, this most commonly (58%) occurs in isolation in patients with C0-3 disease, in 22% of patients with superficial reflux and C0-3 disease occurs with perforator reflux and in 9% with deep venous reflux. In contrast in patients with C4-6 disease, of the total patients with superficial reflux 24% of patients have an association with either deep or perforator disease and in 25% of patients it occurs in isolation or with both perforator and deep venous reflux. This is shown in Figure 5.

For patients with deep venous reflux, this most commonly (35%) occurs with both perforator and superficial reflux in patients with C4-6 disease, in 33% of patients with deep reflux and C4-6 disease occurs with superficial reflux and in 11% with perforator reflux. It occurs least frequently (12%) in isolation in these patients. In contrast in patients with C0-3 disease, of the total patients with an element of deep venous reflux 28% of patients had concurrent superficial reflux. This is shown in Figure 6.

For patients with perforator reflux, this most commonly (67%) occurs with superficial venous disease in

patients with C0-3 disease. It occurs in isolation in these patients in only 13% of cases. In contrast, in patients with C4-6 disease and perforator reflux, 40% and 42% has an association with either superficial or superficial and deep reflux. This is shown in Figure 8.

Discussion

This has shown that in CVI of the lower extremities, particularly for patients with less severe C0-3 disease, the majority of the reflux involves superficial veins only. Once disease progresses, reflux in the superficial veins remains the most common, however, involvement of the deep and perforator veins increases in frequency. There is currently no longitudinal study that demonstrates whether in primary venous disease superficial reflux simply progresses to involve more venous segments and clinical disease severity increases, or whether the initial underlying abnormality involves all of the systems resulting in worse disease severity. In the case of venous reflux, however, it is clear that although superficial reflux remains the most common for all grades of severity, involvement of the deep venous system is apparent

more frequently in patients with C4-6 disease. It is these patients for whom deep venous interventions would be potentially beneficial to improve symptoms. Deep venous reflux does for the most part coexist with superficial reflux, and when considering treatment for patients with venous disease, treatment of the superficial venous disease will remain the first line of treatment. However, given that the deep venous and perforator systems are involved with more severe cases of CVI demonstrates these as potential targets for intervention. Treatment options for patients with deep venous disease are limited and this will be

discussed in greater detail in Chapter 2.

There are limitations to this limited study reviewing the literature regarding patients with venous reflux.

There is heterogeneity in the data collected. The eligibility criteria for patients in each study were variable;

some papers included patients with C0 disease, which were included here, and others did not, although

the majority excluded patients with a history of previous varicose vein or other venous intervention,

others did not. This may have resulted in the underestimation of the frequency of superficial venous

reflux. Papers were excluded if they only focussed on patients with reflux in one system, such as the

saphenous or perforator systems, as this limited the information available on the excluded patients. There

were studies that included both patients with reflux and obstruction. If these could be separated they

were included but otherwise were not.

1.3 POST-THROMBOTIC SYNDROME

Introduction

Deep vein thrombosis (DVT) has a prevalence of approximately 0.2% in the community and up to 25%

in hospitalised patients39 and is a significant healthcare and socioeconomic burden. Post-thrombotic syndrome (PTS) is a frequent complication of acute DVT and occurs in approximately half of patients within the first 1-2 yrs40, particularly after more proximal DVT. Severe PTS occurs in 5-10% of patients

following iliofemoral DVT and at its worse may result in venous ulceration, which causes significant

disability and impaired quality of life41. Not only does it have a significant impact on patients, but it also

provides a significant financial burden and societal cost42. When compared to other chronic diseases, patients with PTS have a poorer quality of life than those with chronic lung disease, diabetes or arthritis43

and the quality of life of patients with severe PTS is in line with patients with congestive heart failure or

cancer41. This part of the chapter reviews the pathophysiology, risk factors for developing PTS, as well as

diagnosis and treatment of PTS, as it is a common cause for secondary venous disease.

Methods

Relevant papers were identified through systematic searches of Pubmed (1948 to June 2013). The search

terms “postthrombotic syndrome”, “post-thrombotic syndrome” and “post phlebitic syndrome” in all

combinations were used. Article titles and abstracts were screened for inclusion. A manual reference list

search was also carried out for further appropriate studies to be considered for inclusion, ensuring

relevant papers published before 1966 were included. An English language filter was applied to the

search. Relevant full text articles were scrutinised and all studies published in English, with human study

participants, study data relating to adult (>18 yrs) patients, and relevant data on lower limb post-

thrombotic syndrome were included. Case studies, non-systematic reviews and letters were excluded.

Full text articles were obtained if from the title or abstract the inclusion criteria were satisfied. These

were then further evaluated for relevance and methodological rigor.

Results

Although the initial search strategy yielded 1190 results, following exclusion of non-relevant publications

and the addition of records identified from other sources, a total of 145 articles were included. This is

shown in Figure 9.

Figure 9: Shows the summary of the search strategy for literature regarding the post-thrombotic syndrome.

Search terms entered: “posthrombotic syndrome”, “post-thrombotic syndrome”, “post phlebitic syndrome”

Records identified through database Additional records identified through searching other sources (n = 1190) (n = 33)

Records excluded (n = 935) Excluded: Articles not Records screened (title) published in English, (n = 1223) animal and in-vitro trials, paediatric trials, upper limb PTS, non-systematic reviews, single patient case

reports

Records excluded (n =143) Abstracts and full-text Excluded: Data not relevant articles assessed for or insufficient, upper-limb eligibility (n = 288) post thrombotic syndrome included in data, review of previous material, included patients with no history of DVT

Studies included (n = 145)

Incidence and natural history of PTS

The incidence of PTS reported in the literature is widely variable ranging from 20-80%40, 44-48. However,

throughout the literature there is marked heterogeneity in the diagnostic criteria, patient populations,

study designs and time of measurement from the initial insult of acute DVT. Those studies which use

validated diagnostic criteria consistently estimate the incidence of PTS within the first year following a

DVT, to be around 50%, even with treatment of the initial DVT with anticoagulation48, 49. Early studies

described PTS developed some 5-10 years after the initial DVT, however, current prospective studies report that the cumulative incidence of PTS plateaus within 1-2 years of the first presentation40, 49. The incidence and severity of PTS continues to accumulate, however, with up to one-third of patients affected at six years49 and in around half of these patients the disease has progressed from mild or moderate PTS to severe PTS. Half of patients have severe PTS at their initial presentation50.

Pathophysiology

An acute DVT results in a partial or complete occlusion of a vein. Following this a process of change to

the size, shape and structure of the thrombus occurs. This may result in recanalisation of the vein, that is

the venous outflow is restored. This can be apparent at 6 weeks following the initial diagnosis of DVT51,

however, approximately half of all legs at 3 years on imaging still demonstrate some residual thrombus

causing at least a partial obstruction52. Within the first three months after an acute thrombus, there is an approximately 50% reduction in thrombus load53, but the rate of recanalisation varies. It appears to be

related to the initial thrombus load52 and thrombus site, with distal thrombi undergoing more rapid and

complete resolution54. The process of recanalisation is less well understood, but seems to involve both

complex intrinsic mechanisms within the thrombus and systemic activation of fibrinolysis55. A correlation between the presence of fibrinolytic inhibitors and the degree of recanalisation has been described55 and it was suggested by Killewich et al suggested that regression of an acute DVT is related to an increase in endogenous fibrinolysis56.

Importantly, the response to an acute thrombosis and recanalisation are inflammatory processes which cause damage to the vein wall and venous valves. As well as the partial or complete obstruction to the

vein, this also results in reflux. This process seems to occur early following a DVT, and at 1 week

following a DVT 17% of patients demonstrate venous reflux, however, this increases to 69% of patients

at 1 year following an acute DVT57, and the degree of occlusion is related to the likelihood of developing reflux57. Indeed, rapid resolution of thrombus seems to result in greater preservation of valvular function58, 59. Eventually, the combination of reflux and obstruction results in venous hypertension. With

no reduction in the standing pressure during ambulation, chronically elevated deep venous pressure

adversely affects venous haemodynamics and results in calf muscle pump dysfunction. These elevated

pressures get transmitted to the capillary beds. Large molecules including plasma proteins and fluid are

forced out of the capillaries to the tissues resulting in tissue oedema, subcutaneous fibrosis and finally

tissue hypoxia and ulceration60-62.

Risk Factors for developing Post-thrombotic Syndrome

Recurrent DVTs and DVT location

Recurrent ipsilateral DVT is the most significant risk factor for the development of PTS, resulting in a 3-6

fold increased risk46, 50, 63. This logically follows as it is likely to worsen pre-existing resistance to venous outflow and valve damage. The anatomical location of the DVT seems to have an impact on the future development of PTS. More proximal DVTs have been associated with an increased risk of developing

PTS in several studies22, 50, and in the acute scenario are generally associated with more severe signs and

symptoms21, 64. However, distal thrombi can propagate to more proximal thrombi and in fact, it has been

observed that approximately 9%-20% of below knee DVTs will involve more proximal veins at further

imaging65 and over 20% of patients with more distal DVTs will go on to develop PTS66. More distal

DVTs do demonstrate improved rates of thrombus regression and recanalisation and reduced rates of

recurrence, and it maybe this that explains the lower incidence of PTS in these patients23.

Age and gender

There is an unclear association between gender and age on the development of PTS. Studies performed

have demonstrated discrepancies. Several studies failed to find an association with gender63, 67, a large trial

of 1668 patients identified a 1.5 fold increased risk for women22, and one reported an increase in men54.

The data from these studies is shown in Table 4. Similarly with regard to age, there is no definite association and further comparison of these studies is difficult due to heterogeneity between the studies.

As with gender, several studies have failed to find an association48, 67, two studies show a correlation with increasing age and PTS development50, 63, but conversely other studies have found a reduced risk of developing PTS with increasing age22, 68.

Table 4: Shows the risk of developing PTS in female patients. Pooled analysis of available data.

Author, Year No. of patients with PTS (n) Relative risk (RR) p-value Female Male [n/N (%)] [n/N (%)] Ageno, 200368 * 10/35 (29) 10/48 (21) RR 1.29; 95% CI 0.59, 2.83 0.53 Kahn, 200369 * 10/17 (59) 9/23 (39) RR 1.32; 95% CI 0.63, 2.76 0.47 Gabriel, 200448 * 35/57 (61) 49/78(63) RR 0.98; 95% CI 0.70, 1.39 0.94 Kahn, 200567 * 27/64 (42) 27/81 (33) RR 1.19; 95% CI 0.75, 1.87 0.46 Stain, 200570 ** 96/246 (39) 80/160 (50) RR 0.84; 95% CI 0.67, 1.08 0.17 Prandoni, 200571 * 42/97 (43) 25/83 (30) RR 1.31; 95% CI 0.85, 2.00 0.22 Tick, 200822 * 235/880 (27) 129/788 (16) RR 1.50; 95% CI 1.23, 1.82 <0.001 Tick, 201054 ** 19/47 (40) 30/52 (58) RR 0.79; 95% CI 0.49, 1.26 0.32 Latella, 201072 * 85/172 (49) 70/172 (41) RR 1.14; 95% CI 0.88, 1.49 0.32 Total 559/1615 (35) 429/1485 (29) RR 1.15; 95% CI 1.03, 1.28 0.01 *Villalta scale used to define presence or absence of PTS **CEAP classification used to define presence or absence of PTS

Obesity

Studies that have observed the risk of developing PTS in obese patients are small. Obesity may be a risk

factor although the underlying mechanism for this is not clear. It may be associated with a reduced

exercise capacity73 and increased venous pressures. An association was observed in this study of 83 patients, in that for those patients that developed PTS following an acute DVT, the mean BMI was significantly higher73 and Tick et al found a 1.5-fold increased risk of PTS in patients with body mass index greater than 30 kg/m222.

Genetics

Patients with inherited thrombophilias have a predisposition to developing DVT, although this is often not discovered until after the initial event. As such, it would be assumed that inherited thrombophilias are associated with the development of PTS, which has been demonstrated with Factor V Leiden in some studies74, 75. However, other studies have failed to find an association63, 70, 76 and some have found that there is a reduced risk67. The data from these studies is shown in Table 5. The causes for this may be related to the duration of anticoagulation therapy, although some studies corrected for this, or may in fact represent underlying differences in thrombus biology, as patients with Factor V Leiden mutation may be more likely to develop acute DVTs involving the popliteal or calf veins77, rather than more proximal

DVT and are less likely to embolise78 and are less likely to be recurrent79.

Table 5: Shows the risk of developing PTS in DVT patients with proven thrombophilia. Analysis of available data80.

Table 2: Risk of developing PTS in DVT patient with proven Thrombophilia. Analysis of available data Author, Year No. of patients with PTS (n) Relative risk (RR) p-value Thrombophilia No Thrombophilia [n/N (%)] [n/N (%)] Hafner, 200174 * 16/42 (38) I 2/45 (4.4) RR 6.48; 95% CI 1.57, 26.8 0.01 Gaber, 200175 * 19/24 (79) I 34/125 (27) RR 2.06; 95% CI 1.32, 3.24 0.002 Kahn, 200567 13/49 (27) II 38/ 76 (50) RR 0.62; 95% CI 0.36, 1.09 0.10 Stain, 200570 71/175 (41) II 105/231 (45) RR 0.92; 95% CI 0.72, 1.19 0.53 Spieza , 201081 63/191 (33) III 110/339 (32) RR 1.01; 95% CI 0.77,1.32 0.93 Total 182/481 (38) 289/816 (35) RR 1.05; 95% CI 0.90, 1.23 0.55 *Outcome measure is the development of PTS ulcers. Remainder of studies included all severities of PTS classified using Villalta or CEAP systems. Type of Thrombophilia detected: I-Factor V Leiden mutation II- Factor V Leiden mutation and prothrombin gene mutation III-Factor V Leiden, prothrombin gene mutation, lupus anticoagulant, protein C/S deficiency

Asymptomatic DVTs

Asymptomatic DVTs (ADVT) are complex as they are difficult to study in the general population. In

post-operative patients, a reduction in ADVT is associated with a reduction in DVT, suggesting the 2 are

linked82. A recent meta-analysis pooling 7 clinical trials suggested that ADVT is associated with a 58% increased risk of developing late PTS [RR 1.58, 1.24-2.02, p <0.0005]83. As shown in Table 6, it is not conclusive from these studies whether ADVT affects either the incidence or severity of PTS.

Table 6: Shows the incidence of PTS in patient with and without asymptomatic deep venous

thrombosis80.

First Author, Year Villalta ≥5 or equivalent (n) DVT Treatment Treatment Duration ADVT No DVT RR [95% CI] n/N (%) n/N (%) Ginsberg, 200084 ʅ 5/91 (5.5) 7/164 (4.3) 1.27 [0.4,3.9] Warfarin 6-12 weeks

Persson, 200985 ʃ 3/38 (7.8) 2/45 (4.4) 1.7 [0.3,9.8] Warfarin and ECS 12 weeks Andersen, 199186 * 8/25 (32) 0/16 (0) 8.5 [0.5, 138] Warfarin 12 weeks Siragusa, 199787 ** 11/46 (24) 2/52 (3.8) 5.2 [1.2, 22] Warfarin 12 weeks

Schindler, 200588 ǂ 7/42 (17) 0/75 (0) 23[1.3, 390] Warfarin or 12 weeks Aspirin Mudge, 198889 * 2/26 (7.7) 1/108 (0.9) 7.8 [0.7, 83] Not stated N/A Lindhagen, 198490 31/137 (23) 135/625 1.0 [0.73,1.48] None N/A ** (22) Total 67/405 147/1085 1.2[0.91,1.56] Definition of PTS used in studies: *non-validated scoring system based on clinical examination ** non-validated scoring system based on subjective symptoms and clinical examination ǂ CEAP classification ʃ Villalta scale ʅ non-validated scoring system based on subjective symptoms and evidence of valvular incompetence

Diagnosis of PTS

PTS is complex, and patients’ symptoms vary widely ranging from pain, swelling, restless or heavy legs

and itching, and these symptoms are often subjective. In more severe cases they can experience venous

claudication during mobilisation91. This makes the diagnosis of PTS difficult, which may explain the

heterogeneity between studies on PTS. Signs are more objective and include oedema, telangiectasias,

venous eczema, hemosiderin deposition and lipodermatosclerosis and in more severe cases leg ulceration,

which can be slow to heal and can recur92. The initial symptoms of the acute DVT may be similar to that of early or mild PTS, and moreover may take some time to resolve, therefore the timing of diagnosis of

PTS is crucial. If patients are observed 1 month following the initial DVT and classified as having

moderate or severe PTS, over half are reclassified at 6 months as having mild or resolved PTS, but this

changes less if patients are diagnosed at 4 or 12 months50. The standard at present is that a diagnosis of

PTS is delayed until at least 3 months after the initial acute DVT93.

There are several imaging modalities that can be used in the diagnosis and ongoing management of patients with PTS. Associated venous reflux, venous obstruction and calf muscle pump function can be measured using Doppler ultrasound and plethysmography. If residual occlusion of the vein is present resulting in obstruction to venous outflow, or if venous reflux has developed, then there is an increased risk of developing PTS, and if both are present concurrently then this tends to result in more severe

PTS20, 54, 71. This has been discussed previously in this chapter. Roumen-Klappe et al49 used imaging to

prospectively assess 93 patients and found that if residual thrombus and reflux were present three months

after an acute DVT, these were the most prognostically important variables with an area under the ROC

curve of 0.77 [95% CI; 0.65-0.90 p<0.01]. Their presence also seems to be an independent risk factor for

developing severe PTS21, 72. As previously discussed, however, not all patients with haemodynamic venous disease go on to develop symptoms or signs71. However, in patients who do not develop PTS,

venous abnormalities can also be demonstrated.

Following an acute DVT, inflammatory processes are occurring, but it is not clear whether markers of

inflammation can predict the development of PTS and indeed, whether this is clinically useful. There

may be an association between elevated D-dimer levels 3 months after DVT and the subsequent

development of PTS70, however, the converse is true for other markers of inflammation such as IL-6 and

CRP94, 95. The Bio-SOX study, a randomised prospective trial of 800 patients, aims to further delineate

the role of inflammatory makers in prediction of PTS96.

Numerous classification systems have been produced to aid diagnosis, grade severity and enable

standardisation. An ideal scoring system for post-thrombotic syndrome should be easy to use, valid,

specific, have good inter-observer reliability, be acceptable to patients, have a clear association with the pathophysiological mechanism, and be able to categorize according to disease severity and identify improvement or deterioration in the condition. The Villalta score is able to lay claim to all these features,

as shown in Table 7. It does contain several subjective criteria which could reduce its reliability, and perhaps further investigation into inter- and intra-observer reliability would be useful, however, the

Villalta score appears to be the most commonly used, due to its ease and versatility of use in diagnosis, categorisation of severity, and in follow up. In 2008 it was recommended at the International Society on

Thrombosis and Haemostasis in Vienna that the Villalta scoring system should be utilised in clinical trials as the scoring system of choice for the diagnosis and grading of PTS. An assessment of the impact of venous disease on a patient’s quality of life should also be part of a PTS score. As demonstrated in the study by Kahn et al97, patients with PTS have a significantly decreased quality of life, and this in itself is a

symptom that contributes to the severity of the condition. Consequently, a quality of life questionnaire, in

combination with the Villalta score, may aid in further clarifying the differences in the categories of

severity of the condition, and standardise some of the subjective criteria of the Villalta score98.

Table 7: Shows the comparison of the properties of the different scoring systems for PTS98

Villalta Ginsberg Brandjes Widmer

Inter-observer Rodger et al 200899 No evidence No evidence No evidence reliability

Association with Kolbach et al Kahn et al 2006101 Kolbach et al Kolbach et al pathophysiology 2005100 2001005 2005100

Kahn et al 2006101

Validity Kahn et al 200297 Kahn et al 2006101 Kolbach et al Kolbach et al 2005100 2005100 Kahn et al 200843

Kolbach et al 2005100

Kahn et al2006101

Diagnosis Prandoni et al Ginsberg et al Brandjes et al Widmer et al 199646 200084 199740 1981103

van Dongen et al Kahn et al 200796 Kolbach et al 2005102 2003104

Tick et al 200822

Kahn et al 200541

Roumen-Klappe et al 200994

Kahn et al 200567

Categorisation of Kahn et al 200297 No Yes Yes severity Kahn et al 2011105

Identify change in Yes No No No condition

Evaluation of O’Donnell et al Ginsberg et al No evidence No evidence treatment for PTS 2008106 2001107

Prevention of PTS

Anticoagulation

In the UK, patients diagnosed with an acute DVT start anticoagulation in the form of Vitamin K

antagonists such as warfarin, with LMWH until the INR is therapeutic. In the absence of a triggering

factor, this is continued lifelong, but if a trigger is identified and can be removed or controlled then 3

months is usually sufficient, although each case is assessed individually108. The duration and type of

anticoagulation should perhaps be studied further, as a longer duration of anticoagulation would reduce

recurrence of DVT and this has been shown in several studies109-111, but for warfarin therapy no

additional reduction in PTS was found with patients receiving 6 months versus 6 weeks therapy112. Trials have studied the use of LMWH for the duration of anticoagulation, rather than vitamin K antagonists.

Patients treated with 12 weeks of tinzaparin reported fewer symptoms of PTS [OR 0.77 95% CI, 0.67-

0.69], fewer venous ulcers [OR 0.12 95% CI, 0.01-0.97] and improved patient satisfaction compared with warfarin113. This could be explained either by the fact that LMWH increases the rate of thrombus regression114, 115 or its anti-inflammatory properties116. Although there is no trial evaluating the effect of anticoagulation on the incidence of PTS, logically as its use prevents recurrent DVT, and a subtherapeutic

INR is predictive of PTS then its use should reduce the incidence of PTS67, 102. In the peri-operative

period, bridging anticoagulation reduces the risk of peri-operative recurrent DVT117, 118. As recurrent

ipsilateral DVT is a major risk factor for the development of PTS, this will reduce the incidence of PTS in

the future.

Elastic Stocking Compression Therapy

There is little consistency between practices on the current use of elastic compression stockings (ECS) following an acute DVT, and only around one third of physicians who regularly treat DVT prescribe ECS to prevent PTS in asymptomatic patients114, 119. Their application reduces signs and symptoms during the

acute episode and by reducing transcapillary filtration, increasing fibrinolytic activity and stimulation of

collateral formation may reduce the incidence of PTS120. Following an acute DVT Elastic compression

stockings (ECS) are prescribed both to alleviate venous symptoms and to prevent PTS121. Studies

performed have demonstrated conflicting results, with several randomised control trials (RCTs) showing

that daily use of ECS can prevent PTS40, 63, and others failed to demonstrate any benefit107, 122. A meta- analysis pooling 5 RCTs123 and a Cochrane review124 concluded that ECS reduced the incidence of PTS

from 46 to 26% [RR 0.54, 95% CI, 0.44-0.67] with the greatest benefit in severe PTS [RR 0.38, 95% CI,

0.22-0.68]. These results are summarised in Table 8. A recent study by Kahn el al122, a large multicentre randomised trial of patients with a first proximal DVT failed to demonstrate a benefit of wearing ECS for the prevention of PTS.

Table 8: Shows a summary of elastic compression stocking for the prevention of PTS. Description of study designs and results

Author, Participants Stocking type Scoring System Stocking Median Follow- Incidence of PTS (%) Year (n) Duration (Yrs) up [RR with 95% CI] Mild/Moderate Severe Total Brandjes Stocking: 98 Stocking: Below- Villalta Scale ≥ 2 76 months Stocking: 20 Stocking: 11 Stocking: 31 [RR 199740 Control : 96 knee, customised, [RR 0.50; 95% [RR 0.54; 95% 0.56; 95% CI 30-40mmHg CI 0.31,0.81; p= CI 0.28,1.06; 0.39, 0.80; Control: No 0.004] p=0.07] p=0.002] intervention Control: 47 Control: 23 Control: 70 Ginsberg Stocking: 24 Stocking: Below- Custom 1.4-4.6 57 months Below knee: 2001107 Control: 23 knee, 20-30 mmHg definition* Stocking: 0 Control: Placebo [RR 1.04; 95% stocking CI 0.46,2.34; p= 0.92] Placebo: 4.3 Partsch Stocking: 26 Stocking: Thigh Villalta Scale Bed rest or 2 yr Stocking: 54 Stocking: 0 Stocking: 54 200473 Control: 11 length, inelastic compression for Control: 82 Control: 0 [RR 0.78; 95% compression (n=13) 9 days and then CI 0.41,1.48; or 30-40mmHg (n= stockings for 2 p= 0.44] 13) years Control:82 Control: Bed rest Prandoni Stocking: 90 Stocking: Villalta Scale 2 4yr Stocking: 22 Stocking: 3 [RR Stocking: 26 [RR 200463 Control: 90 Below-knee, [RR 0.66; 95% 0.32; 95% CI 0.62; 95% CI 30-40mmHg CI 0.41,1.08; p= 0.09,1.14; p= 0.40,0.96; p= Control: No 0.09] 0.08] 0.03] intervention Control: 38 Control: 11 Control: 49 Aschwanden Stocking: 84 Stocking: Below- CEAP 3.2 3.2 Stocking: 13 Stocking: 0 Stocking: 13 2008125 Control: 85 knee, 26-36 mmHg classification 2.9 Control: 20 Control: 0 [RR 0.69; 95% Control: No CI 0.34,1.41; intervention p= 0.3] Control: 20 *Typical pain plus swelling of > 1 month duration occurring ≥6 months after acute deep venous thrombosis 56

If compression stockings are used, both the type used and duration of therapy are a subject of debate.

With regard to the duration guidelines recommend the use of ECS for 2 years following an acute DVT108, but a study of 169 patients demonstrated that only women had significant benefit from extending compression therapy from 6 months to 2 years [hazard ratio 0.11, 95% CI, 0.02-0.91]125. The type of ECS used has been studied widely. Grade II graduated knee length compression stockings are the most commonly used in clinical practice and in studies. With regard to DVT prevention in hospitalised patient, above knee ECS do not confer a benefit over below knee ECS126. Grade III graduated ECS offer greater compression and therefore should offer greater benefit in reducing PTS, but no difference between the 2 has been found for development of PTS107 or recurrence of ulceration, although this may be related to

patient compliance as patients appear to be more compliant with grade II compression ECS over grade

III, perhaps due to comfort and ease of application127. Early application of compression and mobilisation has been found to be of benefit in the reduction of PTS in one study73, but this could not be corroborated further in this bandaging study128, however, early compression significantly increases vein patency the amount of recanalisation at 3 months compared with the application of ECS 2 weeks the initial DVT129. No trials have yet investigated whether other methods of improving venous return, such

as pneumatic compression devices or electrical muscle and nerve stimulation can prevent PTS, although

they have been studied for its treatment.

Exercise

There is no clear correlation between exercise and the development of PTS, and exercise may exacerbate

symptoms in patients with obstruction to venous outflow of the lower limb. In one study early

mobilisation reduced the risk of developing PTS73, however, another found no correlation between the

reported level of physical activity 1 month after the initial DVT and the subsequent development of

PTS130, 131. Although exercise should help the development of a collateral circulation132 structured

exercise programmes for patients with DVT improve recanalisation rates of occluded veins133.

Thrombolysis

As early recanalisation of the vein reduces the incidence of PTS, if thrombus is removed early the venous

outflow would be restored, reducing venous hypertension and reducing PTS59 and patients who underwent surgical thrombectomy had improved venous patency and a reduction in PTS when compared with the use of anticoagulation alone134. In the presence of residual thrombus, the rate of

recurrent DVT is higher135, and recurrent DVT is a risk factor for developing PTS. It should therefore

follow that reducing the thrombus load early is beneficial to the patient.

Systemic thrombolysis for acute DVT is efficacious136-139. In a meta-analysis on the use of thrombolysis for acute DVT there was a significant improvement in complete clot lysis [RR 4.14; 95% CI 1.22 to 14.01] which is maintained [RR 2.71; 95% CI 1.84 to 3.99]140, and this appears to follow through to result in a reduction in the symptoms and signs of PTS141, 142, but as expected the risk of major bleeding is increased

with systemic thrombolysis [RR 2.9; 95% CI 1.1 to 8.1 p=0.04] compared to heparin alone134. Catheter-

directed thrombolysis (CDT) infuses the thrombolytic agent locally to the affected limb, which reduces

the overall dose of the drug infused and should reduce the rate of major bleeding complications. Three

randomised trials of CDT reported benefit; an improved rate of complete lysis in patients undergoing

CDT (72% vs. 12% p<0.001)143; a reduction in the rate of PTS development in patients treated with CDT

compared to anticoagulation alone (3.4% vs. 27.2%, p=0.001) (41.1% vs. 55.6%, p=0.047)144, 145; a

significant reduction in recurrent VTE (2.3% vs. 14.8%, p=0.003)144; a reduction in haemodynamic

venous obstruction [20% vs. 49%; RR 29.1% (95%CI:20-38% p=0.004)]145. However, major bleeding complications were higher in the treatment than control group (3 vs. 0)145.

Percutaneous pharmacological mechanical thrombectomy (PPMT) may be used in conjunction with

CDT, to further reduce the thrombus load following an acute DVT. This technique has recently been developed and involves mechanical removal of thrombus in conjunction with local delivery of thrombolysis medication, further reducing exposure to thrombolytic agents and perhaps improving efficiency and effectiveness of thrombus removal. Small studies have shown encouraging results and a

recent systematic review confirmed safety and efficacy but highlighted the need for high level evidence

from large randomised controlled trials146.

Treatment of PTS

Once the diagnosis of PTS is established, treatment relies on alleviation of patient symptoms. There is

little in the way of trial data to guide the treatment of established PTS. Most physicians recommend the

use of ECS, however only 1 randomised trial using placebo stockings that were too large for the patients

has evaluated this and found no significant benefit to active compression, however the numbers were

small107, however, even large stockings would exert some compression104. Compression therapy does, however, accelerate healing of venous ulceration147. Intermittent pneumatic compression (IPC) may be

an alternative to ECS, and appear to improve symptoms in patients with PTS148, however, the

practicalities of long term use are limiting. Other pumps such as the VenowaveTM device, a small peristaltic calf pump, in small trials have demonstrated an improvement in PTS scores and patients quality-of-life106. Further evaluation of novel electrical calf muscle and nerve stimulation devices, which

have been seen to improve venous haemodynamics may be of benefit to patients with PTS, but this has

yet to be further evaluated149-151.

There are several pharmacological agents that have been evaluated for the treatment of patients with PTS and venous ulceration, however, non are currently in widespread use in the UK as trials have been small and heterogenous. Pharmacological treatments which treat the processes of inflammation and venous hypertension may be of benefit152. The use of Aspirin 300mg has been shown to improve ulcer healing

rates153. Flavonoid derivatives have both anti-inflammatory and venoactive properties and micronised

purified flavonoid fraction (MPFF) significantly improved symptoms in a large trial154, as well as in a metanalysis155. Some other flavonoid derivatives including the rutosides156 have improved patients

symptoms but others have shown no additional benefit over compression therapy alone157. Horse

chestnut seed extract158, calcium dobesilate159, 160 and pentoxifylline161 may provide additional benefit over compression therapy alone.

Surgical treatment options may benefit patients with PTS. Treatment of superficial veins improves deep

venous haemodynamics and improves calf muscle pump function162, 163 but is less effective in preventing

ulcer recurrence in PTS than primary valvular incompetence164, 165. Deep venous surgery including valve repair, transposition or transplantation have been performed and are discussed in more detail in Chapter

2. For patients undergoing valve reconstruction in a 25 year follow up study of 51 limbs there was a higher rate of symptom and ulcer recurrence in patient with PTS compared with primary venous incompetence [43% vs. 73% p=0.029]166. Surgical167, 168 and percutaneous169, 170 recanalisation of

chronically obstructed veins improves venous haemodynamics and symptoms of chronic venous disease

their value in the treatment of PTS has not yet been fully evaluated.

Discussion

PTS occurs in around half of patients following a DVT. Patients are often left with residual obstruction to venous outflow and venous reflux as a result of the underlying pathophysiological processes. The resulting valvular damage and venous hypertension causes severe symptoms and signs, including venous ulceration and impairs patients’ quality of life. Treatment options are limited and the evidence for them is lacking. Compression therapy remains the mainstay of treatment. Aggressive treatment of the initial

DVT to restore venous flow and reduce thrombus load may be beneficial in reducing the severity of PTS, followed by adequate anticoagulation. Prevention of PTS is more beneficial to patients than treatment once it is established, however, once it has been diagnosed symptoms can be further improved by surgical and radiological treatment to restore venous outflow and valve function. These management options are discussed further in Chapter 2.

2. AIMS

Venous disease is a healthcare burden with a detrimental effect on the patient’s quality of life. While

there have been advances in the treatment of superficial venous disease, there has been very little change

in the management of patients with deep venous disease over the last few decades. Currently the only

management option for these patients is compression or major surgery. A minimally invasive treatment

option for this group of patients should be available.

The main aim of this project is to:

• Design and develop a valve for implantation into the lower limb deep venous system to treat

patients with deep venous insufficiency

In order to achieve the main aim of the overall project, each stage of this proposal has a specific objective

and there is immense potential to generate novel secondary research avenues as the subject of deep

venous disease remains under researched. These specific objectives are:

• To further evaluate anatomical and haemodynamic variables in normal subjects

• To provide data required for the development of computational and experimental flow models

for deep venous function

• To develop computer models of venous flow in normal deep venous function and validate this

using data from duplex ultrasound and MRI scanning

• To set up and validate a laboratory model of deep vein flow in the form of a flow rig

• To investigate previously used and novel materials to engineer a prototype bioprosthetic deep

vein valve replacement

• To design a valve utilising the models

• To test this valve in the validated laboratory model

3. DEEP VENOUS DISEASE

3.1 CURRENT TREATMENT OPTIONS FOR PATIENTS

WITH DEEP VENOUS DISEASE

Introduction

The treatment options and recommendations for patients with PTS have been discussed, but the majority of patients have primary venous disease. Current treatments for chronic deep venous insufficiency are limited, with variable success rates. The management options for patients with deep venous disease are discussed further here.

Conservative Management Options

Leg Elevation

Conservative treatments for chronic deep venous insufficiency are limited. Measures including leg elevation, weight reduction, walking, exercise and physiotherapy act to improve venous haemodynamic return and reduce venous hypertension. This alone, however, is invariably not sufficient and it impacts significantly on patients’ daily lives.

Compression

Compression in some form is recommended for patients with venous disease by most surgeons and physicians. Contraindications to compression include concurrent peripheral arterial disease, severe

peripheral neuropathy and contact dermatitis, however, for the majority of patients it remains the first line of treatment. Compression may exacerbate symptoms in patients with occlusion, reliant on superficial

venous channels. The type of compression used varies between countries, centres and even individuals

within the same institution and there is little in the way of consensus. However, a recent Cochrane review

in 2009 concluded that venous leg ulcers heal more quickly when compression is applied147.

Hosiery

Graduated compression stockings can be applied and removed by the patient and are therefore the most commonly used form of compression. As with all compression devices, however, the amount of compression applied depends upon measurements taken and the individual fitting the stocking, and can be reduced over time or indeed as the patients swelling improves, for example. The pressures indicated by the manufacturer can also vary between brands of stocking. Meissner et al, 2007 recommended that the 1st choice for compression pressure for patients with C0-C2 disease is 10-20mmHg (Grade 1), for C3-

C4 disease is 20-30mmHg (Grade 2) and for patients with C5-C6 disease is 30-40mmHg (Grade 3). For patients with deep chronic venous disease, the recommendation is that the compression is lifelong18.

Although it improves symptoms and prevents ulceration, patient compliance is poor6.

Four layer compression

Multi-component compression systems improve healing of ulcers compared with single component systems147. However, the recent randomised controlled Canadian Bandaging trial171 involving 424 individuals failed to show a significant difference in ulcer healing times, recurrence or health related quality of life between four layer bandaging and short stretch bandage.

Intermittent Pneumatic Compression (IPC)

A Cochrane review in 2011 of 7 randomised controlled trials concluded that IPC use reduces ulcer healing time and increases the number of healed ulcers when compared to dressings alone, but not conclusively when compared to other forms of compression172. The use of IPC is well documented for use in venous thromboembolism prophylaxis and for treating lymphoedema and it is known that it improves peripheral haemodynamics.

Electrical stimulation of calf muscle pump

Repeated contraction of the calf muscle pump improves venous velocity and reduces the incidence of peri and postoperative deep vein thrombosis. Dysfunction of the calf muscle pump either due to joint immobility or paralysis reduces venous return and worsens chronic venous insufficiency. Stimulation of the calf muscle to contract either by direct stimulation of the muscle or by indirect nerve stimulation may improve symptoms and ulcer healing in patients with chronic deep venous disease. This may be especially beneficial for patients in whom external compression is contraindicated due to peripheral arterial disease, and others who are not able to tolerate it due to discomfort. Venous haemodynamics are improved as calf muscle stimulation, which generates contraction of the muscle will eject blood from the lower limb veins.

Electrical Muscle Stimulation

An electrical current applied directly to the calf from a conducting pad applied to the surface of the skin directly overlying the muscle causes the calf muscle to contract. A recent study has demonstrated a possible added benefit over and above the improvement in venous haemodynamics seen in patients who wear this for a short period of time. Patients who use this system daily over a week may have an additional habituation effect which increases ejected venous volume and peak venous velocity173. As yet, there is no data to suggest this improves clinical outcomes.

Electrical Nerve Stimulation

Transcutaneous nerve stimulation to induce calf muscle contraction by stimulation of the peroneal nerve

has been shown to enhance venous flow including flow volume and peak velocity comparable to EMS

and IPC174. In addition when compared to EMS there was less pain and discomfort, however, these differences were not significant. As yet, there is no data to suggest this improves clinical outcomes.

Drug therapy

There are several drugs available which improve symptoms of venous disease, however, none are

available in the UK. Daflon®, for example, is popular in Europe, but trials have been small and heterogeneous and it is not available across the UK. It is a micronized purified flavonoid fraction

(MPFF) consisting of 90% diosmin and 10% flavonoids expressed as hesperdin. It improves the

symptoms reported by patients with venous disease, including reducing calf or ankle circumference and

improving quality of life, by reducing capillary permeability, improving lymphatic drainage and venous

tone. In addition, when combined with standard treatment for lower limb venous ulcers, it has been

shown to reduce ulcer healing time 175. The use of other drugs including Oxerutins, rutosides and other flavonoid derivatives, which have similar mechanisms of action to MPFF, as well as diuretics, naftidrofuryl, Pentoxifylline, calcium dobesilate, and prostaglandin E1 have been shown in trials to either improve symptoms or reduce the duration of ulcer healing, but their use remains limited.

Surgery

Deep venous surgery is performed, but its widespread use is restricted. Valvuloplasty, venous

transposition, valve autotransplantation and neovalve procedures have all been performed176. Although

isolated small studies and case series have shown promising short term results from these invasive

procedures, consistently good long-term results are lacking. Moreover, a number of other limitations of

deep vein surgery have emerged.

• Deep vein surgery is highly invasive, involving often long and difficult procedures.

• Elderly patients with extensive co-morbidity are often unfit or unwilling for such interventions.

• Long term (often lifelong) anticoagulation therapy is usually required.

• Deep venous surgery is only offered in a few specialist centres worldwide.

• Patients are likely to require extensive hospital admission after deep venous procedures, resulting

in significant health care expenses and time off work.

Long-term success has not been reliably demonstrated. Consequently, deep venous surgery is performed

in a small number of centres worldwide and is available for a minority of patients. A Cochrane review in

2004 concluded that there is not enough evidence on the effects of surgery for deep venous

incompetence, but recommended that more well-designed, randomised trials with long term follow up are

needed177. The options for the surgical management of deep venous disease are discussed in more detail.

Valve reconstruction

Valve reconstruction, or valvuloplasty, aims to restore valve competence and is the surgical treatment of

choice in patients with primary venous insufficiency. This can either be carried out by the internal

method178, under direct vision, which was first carried out by Kistner and subsequently modified, or by

the external method by restoring the diameter of the vein lumen using sutures at the point of insertion of

the valve cusps179-181. An alternative to external valvuloplasty is external valve banding, in which an

external cuff of material is applied, thus avoiding a venotomy. Many materials have been used including

PTFE, silastic, Dacron, and live fascia182. The disadvantage of both of these external techniques is that the valve leaflets are not repaired, so remain incompetent if they are damaged, however, the benefits are the avoidance of a venotomy which may causing an inflammatory process and possible subsequent thrombosis.

Venous transplant

A segment of vein with a competent valve matched for size, for example the axillary vein, is transplanted

or relocated to the incompetent deep vein. This procedure is often carried out in combination with an

external valve banding in an attempt to preserve competence. These transplanted veins are however,

prone to failure either due to thrombosis, or more commonly for no identifiable reason. The reported

results show a wide discrepancy, ranging from a 75% success rate and a 92% technical success rate at 5

years183, 184 to 50% at 10 years.

Neovalve formation

A new valve is constructed from the patient’s own tissue. The techniques that have been described for

these procedures are plentiful, but the most commonly employed are Maleti’s technique using the vein

intima to create the valve leaflets, and Plagnol’s technique using the invaginated termination of the long

saphenous vein to create the valve. The patient’s own tissue is used and transposition of a vein from

another location is not required, thus reducing morbidity compared to venous transplant. The results

from Italy of autologous valve reconstruction have shown similar success, with a median follow-up of 22

months demonstrated a clinical improvement in 89% of cases, with an overall vein patency rate of

83.3%185. Short-term results of cryopreserved vein valves in humans have not been so favourable with

patency rates of 41% at 2 years and a competency rate of 27%186.

Venous Transposition

A segment of long saphenous vein or deep femoral vein containing a competent valve is transposed onto the incompetent femoral vein, a technique first described by, Kistner and Sparkuhl in 1979187. This

technique has the appeal of a single anastomosis between the two adjacent segments. Early outcomes for

the patients were good, but at 3 years many had recurrence of both symptoms and reflux. Queral et al

performed venous transposition alone in 12 limbs of 9 patients, early results were again promising, but at

1 year 9 of the 12 limbs had reverted to the preoperative abnormal venous refilling time, with many

having recurrent leg ulceration188, 189. Between 1983 and 1997, Perrin performed 18 transpositions, with

the superficial femoral vein transposed onto the great saphenous vein in 11 cases and onto the deep

femoral vein in 7 cases190, however, concomitant superficial or perforator vein surgery was performed in

14 of these cases. Of 9 patients in the study group who had active or previous ulceration, 2 had recurrent

ulceration. Of the 17 patient followed up with duplex scanning post operatively, 29% had major reflux

and 6% had minor reflux. This procedure is now rarely performed.

Surgical and Interventional treatments for deep vein obstruction

Occlusive disease causing lower limb outflow obstruction occurs more commonly in the iliac vein. It can

also result from more distal segmental occlusions after a failure of recanalisation following a deep vein

thrombosis. Correction of the outflow obstruction improves patients’ symptoms, reduces ulcer healing

times and improves quality of life191.

Venous Stenting

Venous stenting for occlusion has been proposed, but is only performed in a few centres in significant

numbers16. However, complete occlusion is present in only a minority of patients with deep venous disease, with valvular incompetence responsible for the majority. In one single centre study, 304 limbs

with symptomatic chronic venous insufficiency were treated, resulting in primary and secondary stent

patency rates at 24 months of 71% and 90%, respectively with symptomatic relief in over 70% of patients

192. Hartung et al reported they were able to reproduce this with follow up at 36 months demonstrating primary and secondary patency rates of 73% and 90%193.

Bypass procedures

Patients with occlusive disease may benefit from venous bypass procedures, such as femoro-femoral crossover bypass (Palma procedure)194 and iliocaval bypass with autologous ipsilateral or contralateral

long saphenous vein, or reinforced PTFE195. Saphenopopliteal bypass is infrequently performed, but published results show a 77% functional improvement but variable and often poor patency of 5-

100%196.This does, however, necessitate major surgery and often lifelong anticoagulation, and reported outcomes have been variable, and consequently, relatively few procedures are carried out.

Summary of surgical treatment options for patients with deep venous disease

Deep vein reconstruction procedures are highly challenging, operator dependent surgical procedures, and

require meticulous surgical technique and comes with it a recognised learning curve. There is no single

operation applicable to every patient, and every patient requires a thorough clinical and diagnostic

workup. Deep vein reconstruction (either by surgery or by radiological intervention) should be

considered early in the management of selected patients. The interventional and surgical choices for

patient management are shown in Figure 10.

Figure 10: Shows a summary of the management choices for patients suitable for deep venous surgery197

'((39(1286',6($6(

385( 0,;(' 385( 2%6758&7,21 5()/8;

2EVWUXFWLRQ 5HIOX[ GRPLQDQW GRPLQDQW

,QWHUQDOYDOYXORSODVW\ 9HQRSODVW\ 6WHQW ([WHUQDOYDOYXORSODVW\ %\SDVVSURFHGXUH 7UDQVSRVLWLRQ 7UDQSODQWDWLRQ 1HRYDOYH

Bioprosthetic valves

Bioprosthetic valves could be implanted to treat patients with both deep venous reflux and obstruction.

None are currently in clinical use. Early prototypes which reached the stages of animal studies yielded poor outcomes as the majority of stents occluded within a few weeks. A complete review of previous work carried out in there development is considered in the next part of this chapter.

3.2 IMPLANTABLE VENOUS VALVES FOR DEEP

VENOUS INSUFFICIENCY – THE STORY SO FAR

Introduction

Previous attempts to develop a percutaneous prosthetic vein valve implant have been disappointing198.

For over 40 years, researchers have attempted to create prostheses. Results from early animal studies were

poor as the majority of stents occluded within a few weeks199, 200. More recent prostheses involved single,

double or triple cusp leaflets usually mounted on a self-expanding metal frame. Animal studies showed

promising early patency rates, but poor long-term results201. Nevertheless, 2 clinical studies have been

conducted in recent years. In one small series, early occlusion was seen in 4/5 implants, thought to be due

to inappropriate valve design202. More recent work by Pavcnik et al demonstrated early success and deep

vein competence in 14/15 studied patients with early clinical improvements. However, stent valve

occlusion was seen in 4/15 patients by 7 months and by 12 months, all the remaining valves had failed

and become incompetent203.

Presently, polymers and polymer coated metal stents, used in the vascular system have several problems: they are very thrombogenic, resulting in a deep vein thrombosis and an occlusion of the vein, they lack haemocompatibilty and biocompatibility, which causes an inflammatory reaction to occur, their presence results in intimal hyperplasia, which subsequently causing vessel narrowing and material failure. All of these would serve to worsen the symptoms and signs of venous disease. In addition they lack the required mechanical properties, as ideally, they should have the ability to expand and contract along with the vein, which can double in size with changes in the patient position from lying to standing.

This review aims to describe previous work that has been carried out and to summarise the reasons for failure to guide future work to develop a successful deep vein valve replacement for the treatment of patients with chronic venous insufficiency.

Methods

The review was conducted according to the preferred reporting items for systematic reviews and meta-

analyses (PRISMA) guidelines19. Relevant papers were identified through systematic searches of Pubmed

(1966 to October 2010). The search terms “bioprosthetic vein valve”, “prosthetic vein valve”,

“implantable vein valve”, “vein valve replacement”, “percutaneous vein valve” and these search terms repeated with “vein” replaced by “venous” were used. Article titles and abstracts were screened for inclusion. A manual reference list search was also carried out for further appropriate studies to be considered for inclusion, ensuring relevant papers published before 1966 were included. Relevant full text articles were scrutinised and all studies reporting details of implantable deep vein valves at any stage were included. Articles concerning deep vein thrombosis, deep venous valve open surgery or replacement, superficial vein valves and cardiac valves were excluded.

Results

Although the initial search strategy yielded 1486 results, with 13 articles included from elsewhere,

following exclusion of non-relevant publications, a total of 19 articles were included (Figure 11). All

selected studies aimed towards developing a minimally invasive technique to implant a natural or synthetic

valve into the deep venous system to treat patients with deep venous insufficiency. To highlight the

development process, the results are categorised into in vitro studies, animal studies and in human clinical

studies. The problems encountered with the valves and suggestions for improvement are also

highlighted.

Figure 11: Shows a summary of the search strategy for all studies of implantable deep venous valves

Search terms entered: “bioprosthetic vein valve” OR “prosthetic vein valve” OR “implantable vein valve” OR “vein valve replacement” OR “percutaneous vein valve” OR “bioprosthetic venous valve” OR “prosthetic venous valve” OR “implantable venous valve” OR “venous valve replacement” OR “percutaneous venous valve”

Records identified through database Additional records identified through searching other sources (n = 1486 publications (n =13) between 1963 and 2010)

Records excluded (n = 1445) Records screened (title) Exclusion criteria: Articles on (n = 1499) deep vein thrombosis, deep venous valve surgery or replacement, superficial vein valves and cardiac valves, title of paper not within scope of review

Abstracts and full-text articles assessed for Records excluded eligibility (n = 54) (n = 35) Excluded: duplicate records, open surgical techniques cardiac valve studies

Studies included (n = 19)

In vitro studies

Natural Valve material

Geselschap et al204 in 2006 carried out an in vitro evaluation of an autologous valve-stent device. They used a Palmaz stent, which is a balloon expandable stent made from woven surgical grade steel, initially developed for use in arteries. By sawing half way along the length, so when expanded, half is cylindrical and half collapsed, and attaching a segment of vein without valves placed inside the stent and folded back to cover both the inside and outside of stent, the valves were made. During prograde flow the valve opened, then reversal of flow caused the unsupported part of vein to collapse against the stent. The major advantages of this design were that it was able to be deployed percutaneously and it is made of autologous tissue. The valve was mounted vertically for testing in a laboratory flow model. To mimic the calf muscle, rapid prograde flow generated by a pump for 3 seconds is followed by allowing 3 seconds to allow column to fall back. Duplex scanning was performed using a solution containing 10 micron polyamide microparticles. Twenty valves were tested. They demonstrated a closure pressure of 1-3 cmH2O. Nine of twelve valves tolerated flow up to 1000ml/min and the closure time was 0.04-0.33s.

There were however, some issues with the valves tested. The valve started to deform with repetitive cycles in 3 valves due to saline filling between layers which was resolved by eliminating side branches.

No further results on additional testing of this valve have been published.

Gluteraldehyde fixing of veins and valves has been considered as an option for the open surgical replacement of deep venous valves for many decades. DeLaria et al205 carried out a haemodynamic

evaluation of a venous valve implant in vitro, which they aimed to subsequently mount on a stent after

further testing. They used gluteraldehyde fixation of 9 bovine jugular vein segments containing a valve in

0.2-0.5% solution and tested in a flow circuit. When compared to fresh vein segments, the fixed veins

with valves demonstrated significantly higher opening times, significantly reduced expansion with

increasing pressures and a significant reduction in orifice area at higher flow rates. However, they did

show less reflux when subjected to higher back pressures. Thrombogenicity and biocompatibility are not

tested, and the author raises concerns regarding the flexibility of the valve leaflets and their suitability for

the venous system.

Pavcnik and his group to date have the most published literature on the subject of percutaneously

implantable deep venous valve replacements. Since 1999, they have developed several bioprosthetic vein

valves (BVV) and are now in the process of developing a 4th generation valve. In 2000, the first generation BVV1 was developed, consisting of a square stainless steel stent with leaflets of small intestinal submucosa (SIS). SIS is described as an acellular, non-immunogenic, biodegradable, xenogenic, collagen, predominantly type 1, based biomaterial from submucosal layer of porcine small intestine, which is pliable when wet. It attracts host cells, is a good bioscaffold and rapidly is remodelled by the host vessel and degraded. Two leaflets of SIS were stretched over the stent and once deployed via catheter, two valve sinuses were created206. The valves were evaluated in vitro using a flow circuit and a back pressure of

60mmHg created by a column of water, which resulted in valve leaflet closure. Forward flow was

generated in a constant manner by a pump, and in a pulsatile manner by a reservoir bag with manual

compression to simulate the calf pump. The valve leaflets opened with forward flow and with valve

closure the pressure below the valve reduced. Subsequently the valve was evaluated in both a canine and

ovine model and in a phase I clinical trial as discussed below. Due to the problems associated with the

valve a second generation valve (BVV2) was developed. These in vitro studies are summarised in Table

Table 9: Shows the in vitro studies utilising a natural valve material

Reference Materials used Design Method of testing Results of testing Further improvements needed

Geselschap et al204, 2006 Great saphenous vein on Stent sawed along half Mounted vertically, saline Closure pressure 1- Saline filling between the surgical steel Palmaz way, saphenous vein filled system, 3 second 3cmH2O walls – reduced by side stent sutures around it to cover prograde flow, 3 second Opening pressure 2- branches both the inside and fall back, assessed by 4cmH2O outside of the stent duplex scanning Closure times 0.04-0.33s

DeLaria et al205, 1993 Gluteraldehyde preserved Bileaflet bovine valve Respiratory flow pump in Gluteraldehyde fixing Requires mounting on a bovine jugular vein circuit with video camera results in: stent and flow meter. Tested reduction in orifice area Smaller sizes may before and after fixing at higher flow increase thrombosis with gluteraldehyde reduced expansion with Safety and increasing pressure biocompatibility need to prolonged opening times be demonstrated

Pavcnik et al207, 2000 Stainless steel stent with Square based stainless In flow circuit , Pressure below valve Further evaluation in leaflets of SIS (Pavcnik steel stent with 2 leaflets retrograde pressure reduced animal studies carried out BVV1) of SIS sutures with 60mmHg, valve closed at Valve closed at rest prolene, deployed via rest, both pulsatile and Opening pressures high catheter non pulsatile forward flow

Synthetic Valve Material

Oberdier and Rittgers208 in 2008 designed a prosthetic valve consisting of a solid frame and flexible leaflet, with a circular base and 2 opposing struts, with shoulders for reinforcement, projecting with an anchored flexible leaflet. In order maintain native anatomy, a bevelled leaflet attachment was created to allow the formation of the valve sinuses. These features evolved over several stages following in vitro testing. Zones of stasis and excessive shear stress were diminished to minimise blood damage and thrombus formation. The resting valve position was open. The prototype was made utilising a photo- activated polymer resin of bis-(epoxycyclohexly)-methylcarboxylate and 2,2-dimethoxy-2- phenylacetophenone (RenShape®) for the frame, cured in an Ultraviolet (UV) oven, for the frame . The leaflets were created using a hand fabrication technique with 0.1270mm thick sheets of ‘Biospan’

(segmented polyether urethane), cut with a surgical scalpel and superglued to the flange. The initial problems encountered including high closure pressure, leaflet sagging and frame breakage, were corrected prior to quantitative testing by bevelling the proximal base edge and by incorporating a supporting shoulder on the flange. Testing was carried out in a flow system209 with pulsatile flow via roller pump and two hydrostatic pressure towers, controlled by hydraulic communications to simulate body positions, including standing and supine positions with both breathing and ankle flexion. The results from the flow system indicated that testing the valves in the standing, breathing simulation gave the best discrimination between the valves. All valves gave significantly improved haemodynamics compared to “no device” controls, including regurgitation, percent reflux, reflux energy loss and energy retention, however, valves sized at 2.50 and 3.75mm were significantly superior to 1.25 and 5.00mm suggesting that there were issues with scalability.

Sathe and Ku, in 2006210 developed a flexible prosthetic vein valve designed with the criteria that it must

be able to withstand 300mmHg back pressure with <1ml/min leakage, the initial and second opening

pressures must be less than 5mmHg, and it must be able to withstand 500 cycles opening and closing.

The valve was fabricated using Salubria, a freeze thawed organic polymer containing poly (vinyl alcohol,

fixed with polymeric glue, injected into a 2 part silicone cavity mould, fixed and polished. The tests

carried out included initial opening pressure, second opening pressure, reflux leakage, burst pressure and

cyclic life functionality. 11 valves were tested and 5 showed good competency, but it was discovered that

valve opening pressures increased after exposures to high back pressures, however, only 5 of the 11

valves underwent testing of the opening pressures after proximal pressure testing as they demonstrated

poor competency testing. Only one valve was tested for cyclic life functionality and was able to withstand

500,000 cycles and still maintain competency. Burst pressure testing also yielded results that were well

above physiological pressures. Further studies have not yet been published, however, this valve is

undergoing testing in an animal model.

Most recently in 2011, Moriyama et al211 produced fibre scaffolds by electrospinning polyurethane over cobalt-chromium stents. This material has previously been shown to support endothelial cell growth212.

Electrospinning is a technique whereby a voltage is applied across a solution in a syringe in a syringe

pump. The needle is positively charged, and the collecting plate is grounded, drawing out a fibre of

polymer from the solution. They developed an in vitro model to simulate standing, standing with ankle

flexion and supine positions. In the standing positions, both open and biomimetic bicuspid polyurethane

valve prevented the increase in distal pressure with small amounts of leakage, however, the open valve

had greater, 0.514ml/s against 0.034ml/s, leakage than the biomimetic valve. It was noted however, that

large stagnation zones were present between the leaflets of the valves and the tubing walls. This could

represent a potential site for thrombus formation. These are summarised in Table 10.

Table 10: Shows the in vitro studies utilising a synthetic valve material

Reference Materials used Design How tested Results of testing Further improvements needed

Oberdier and Rittgers208, Frame – photo-activated Circular base with 2 Pulsatile flow pump, Improved haemodynamic Authors suggested drug 2008 resin RenShape® opposing shouldered pressure towers, 3.45% compared to “No device” reservoir incorporation Leaflet – polyether struts, projecting with an dextran fluid. Standing / Intra-design variability Thickness scaling for polyurethane anchored flexible leaflet supine / breathing /ankle Valve size had effect on different sizes flexion included. performance Compared to “No device”

Sathe and Ku210, 2006 Poly(vinyl alcohol) Bicuspid valve with Withstand back pressure 5/11 good competency Additional flow testing hydrogel, “Salubria” leaflets open in resting 300mmHg with leakage opening pressures needed to determine Polymer injected into position <1ml/min, initial and 2nd increased after high back optimum valve shape mould valve opening pressure pressures <5mmHg, withstand Good cyclic functionality 500,000 cycles

Moriyama et al211, 1993 Electrospun polyurethane Bicuspid valves with In vitro model to simulate Valves withstood Additional studies Cobalt-chromium stent leaflets in closed 3 positions, flow fields 150mmHg pressure with needed to characterise (biomimetic) and open visualised and recorded by minimal leakage material resting positions video Open valve regurgitation Large stagnation zones > closed valve in noted between leaflets standing position and tubing

Animal studies

Natural Valve material

By far the most extensive work has been carried out using natural valve material in animal studies. As yet,

no results have been published using synthetic material in animal studies. De Borst et al213 from the

Netherlands used a percutaneous venous valve bioprosthesis (PVVB), from VenPro Inc., USA, and

implanted it into the porcine venous system, 18 valves into the iliac systems of 10 pigs. PVVB is a

gluteraldehyde preserved valve bearing venous xenograft from segment of bovine jugular vein containing

competent bicuspid native valve sutured inside laser cut memory-coded nitinol frame, sutured in place

with prolene. In order to maintain the position within the frame, it was relatively oversized, and when

inserted to common iliac vein, it was deployed endovenously to 2mm larger than the vein itself. The

valves were tested in vivo at 4 weeks to assess patency and competence with antegrade and retrograde

phlebography, the animal was sacrificed and the valves underwent histological examination to assess

thrombus formation and venous wall. They were graded 0 (no changes) to 4 (severe changes). In pigs

receiving vitamin K antagonists, sacrificed at 2 weeks due to bleeding complications, 7 of 8 valves were

patent and 3 of 8 valves were competent. In pigs receiving aspirin and clopidogrel, sacrificed at 4 weeks,

5 of 10 valves were patent and 3 of 10 were competent. Clearly long term patency and competency of

these valves needs to be improved. Gomez-Jorge et al201, also studied a gluteraldehyde fixed segment of

bovine EJV with valve sutured in a self expanding nitinol stent from Boston Scientific. They deemed the

procedure to be technically feasible, however of the 11 valve-stent devices implanted either into the inferior vena cave or external iliac veins, 2 were unsuccessful immediately due to deployment in the renal vein and rupture of the inferior vena cava. Animals were sacrificed immediately or at 1 week or 2 weeks..

Of the 5 successful deployments which were sacrificed immediately, all were patent and competent, and of the 4 sacrificed at 1-2 weeks, of which 1 died in the cage, all were patent and competent, however, they all showed granulomatous change and foreign body reaction, and one showed bacterial contamination.

Boudjemline et al214 did a similar study in 2004, again assessing a segment of bovine jugular vein containing a valve, preserved in gluteraldehyde and fixed in a platinum-iridium stent from Numed Inc.

The gluteraldehyde was removed prior to use by soaking in saline solution. They focussed on the longer

term outcomes, in contrast to Gomez-Jorge et al. The valve stent devices were deployed endovenously into the IVC of 6 lambs and assessed 2 months after implantation. In all 6 cases, the valves and IVC were completely occluded. They questioned appropriateness of the use of the IVC to site the valve as the pressure gradient is so low, but in addition the material itself needs improvement as it is thrombogenic.

Kucher et al215 made a direct comparison between segments of jugular vein containing valves from dogs implanted into canine femoral veins, either by open surgical technique or by a percutaneous technique after mounting on a nitinol stent. In total, 7 dogs each had open surgery on one side, and the valve stent device inserted to the contralateral side, intraoperative venograms confirmed patency in all cases. 6 valves in each category remained patient at 120 days, reflux pressure was slightly higher in the stent grafts than the transplanted side. This did not demonstrate that stenting was superior as expected, but it showed it as an alternative to valve transplant. Ofenloch et al216 in 1997 performed a similar study in goats, taking a

segment of external jugular vein and valve and mounting it, by suturing with polypropylene, within a

Wallstent device, a self expanding stent made of stainless steel. The terminal 1-2mm of the stent was not

covered by EJV in the 1 week study, leaving exposed metal, however, for the 6 week study the stent was

completely covered with tissue. The devices were deployed endovenously back into the same animal, into

the contralateral EJV. The 5 animals tested at 1 week, which had exposed metal from the device, all had

thrombus on the stent struts, however, the valves themselves remained both patent and competent.

Those tested at 6 weeks, which had no exposed metal remained patent in all cases, however, 1 of the 6

valves was incompetent. Autologous valves clearly have good compatibility, however, may not be

practical in clinical terms. Dalsing et al217, 1996 carried out a feasibility study assessing 5 valve-stent devices of canine external jugular vein over a stainless steel stent, deployed into the same animals iliac venous system. Of the 5 devices placed, 1 migrated to the lung and only one that was placed in the common iliac vein was patent at 4 weeks, with 2 showing clot and all valves showing scarring.

Pavcnik et al’s first generation valve, BVV1, was tested in both a long and short term animal model, inserted into canine and ovine Inferior vena cava (IVC) for the short term studies and into the external jugular veins (EJV) bilaterally of 12 sheep for longer studies with 3 animals receiving an extra BVV1 in the common jugular vein (CJV). All were implanted via the percutaneous route. In the short term study207, all venous valves implanted showed good patency for 6 weeks and good valve function and

histologically, SIS was incorporated smoothly into the vessel wall and a good covering of vascular

endothelial cells covered the valve leaflets. The longer term study demonstrated tilting on deployment in

3 of the 25 valves in total implanted, but no reflux in 24 of the 25 valves, and 22 of the 25 valves were

functioning at the time of animal sacrifice on venographic assessment, only one valve was thrombosed218.

All valves demonstrated complete endothelialisation and neovascularsation histologically, one animal

showed dystrophic calcification at 6 months. Subsequently the BVV1 was inserted into 3 patients, as

discussed below. Following problems with tilting, Pavcnik et al developed the second generation BVV2.

This consisted of 2 different stent designs and 2 different materials, implanted as a short term study in

sheep EJV. In total, 16 Z-stents with SIS were placed, 8 of the stents were stainless steel and 8 were

nitinol. All were placed without tilt and were competent and patent at 6 weeks. Histological analysis was

not carried out on the Z-stents. 32 double stents were sited, 4 stainless steel and 28 nitinol. All were

patent at 6 weeks. However, 2 of 4 stainless steel double-stents became incompetent and had severe

intimal hyperplasia on histological analysis and 2 of 28 nitinol double stents became incompetent.

Histology of 4 of the valves showed endothelial cells and remodelling of the SIS. Pavcnik et al219 also carried out studies in autologous venous valves harvested from sheep jugular veins, mounted in a stainless steel double stent and deployed into the contralateral jugular vein of the same animal. The outcomes were very good, and at 3 months 8 of 9 were competent, and the only incompetent valve showed immediate reflux due to a technical problem, 2 of 9 had perivalvular leak immediately. However, the sutures and wires all showed extensive foreign body reaction and chronic inflammation. These results are summarised Table 11.

Table 11: Shows the animal studies utilising a natural valve material

Reference Materials used / valve Animal Model used How did it work How tested Results of testing Further improvements needed

Gomez-Jorge et al201 Gluteraldehyde fixed Porcine External iliac Bicuspid valve in Animals scarified All had granulomatous Technically feasible bovine EJV with valve vein or Inferior vena nitinol stent deployed immediately, at 1 week change and foreign but issues with sutured in self cava endovenously or at 2 weeks, 1 animal body reaction deployment, vein expanding nitinol stent died in the cage Immediate sacrifice – rupture and (Boston Scientific) Venographic for reflux 5/5 patent and inflammatory response Histopathology post competent mortem 1-2 week sacrifice 4/4 competent and patent 2/11 unsuccessful immediately after deployment, 1 had bacterial contamination De Borst G et al213, Gluteraldehyde Porcine iliac veins Bicuspid valve in Venography to assess 12/18 patent, 6/18 Long term patency and 2003 preserved bovine nitinol frame, deployed patency and histology competent., no competency jugular vein valve endovenously after animal sacrifice inflammation and inside nitinol frame, slight fibrosis in all VenPro USA valves

Boudjemline et al214, Gluteraldehyde Porcine inferior vena Tricuspid valve in Venography to assess 6/6 completely Improvements in 2004 preserved bovine cava platinum-iridium stent patency and histology occluded at 2 months material jugular vein valve in a deployed after animal sacrifice thrombogenicity platinum-iridium stent endovenously needed from Numed Inc

Kucher et al215, 2005 Jugular vein over Canine femoral vein Jugular vein harvested Venography to assess 6/7 patent at 120 days Further testing needed nitinol stent from dog, stretched patency, histology and Valve structures intact as model does not over nitinol frame, electron microscopy to Endothelial lining subject valves to blood native bileaflet valve assess valve structure intact column contained in segment

Ofenloch et al216, 1997 Goat external jugular Autologous goat Jugular vein harvested Venography to assess 1 week – 5/5 patent Autologous transplant vein over stainless steel external jugular vein from goat, sutured in patency and histology and competent with not practical for stent stainless steel stent, after animal sacrifice thrombus on struts clinical use native bileaflet valve 6 weeks – 6/6 patent, contained in segment 5/6 competent

Dalsing et al217, 1996 Dog EJV over Autologous dog Jugular vein harvested Venography to assess Animal sacrificed Needs to improve Stainless steel stent, common or external from dog, stretched patency and histology between 1-4 weeks scarring and prevent Cook Company iliac vein over stainless steel after animal sacrifice 1/5 migrated to lung migration stent, native valve 5/5 Scarring contained in segment, 1/5 patent at 4 weeks placed endovenously

Pavcnik et al206, 2001 BVV1 (See above) Canine IVC BVV1 as above Venography and Good valve function Longer term studies Ovine IVC pressure measurement, and patency in all needed and histological animals for 6 weeks, assessment after SIS incorporated into sacrifice` vein wall, histology – leaflets covered with endothelial cells Pavcnik et al218, 2002 BVV1 Ovine EJV BVV1 as above Venography in live 1/25 thrombosed Issues with valve tilting animals, and histology 22/25 good function need addressing when animals 25/25 endothelialised sacrificed at 1, 3 and 6 1/25 dystrophic months calcification

Pavcnik et al220, 2004 SIS membrane on Z- Ovine EJV SIS membrane sutured Venography in live All stents placed no tilt Longer term study stent or double stent over square nitinol animals immediate and 16/16 Z stents needed of nitinol or stainless stent , with additional at 6 weeks, and competent and patent steel (BVV2) stent to prevent tilting histology when animals at 6 weeks sacrificed 28/32 Double stents competent and patent at 6 weeks

Pavcnik et al219, 2007 Jugular vein from Ovine Jugular Vein Autologous bicuspid Venography at 3 2/9 immediate The use of autologous sheep into double valve from jugular vein months, competence perivalvular leak tissue difficult in square stainless steel sutured in double tested in vitro after 8/9 competent at 3 clinical practice stent square stent explants, histology of 4 months samples

In Human Clinical studies

Three clinical studies have been conducted in recent years. The phase I clinical trial in the first 2 patients

was published by Gale et al202. They observed that with their percutaneous venous valve bioprosthesis

(PVVB), consisting of gluteraldehyde preserved bovine venous bicuspid valve on a nitinol frame,

implanted into the femoral vein by a percutaneous method, 1 patient had a successful competent and

patent valve with ulcer healing. However, the second patient developed post insertion thrombosis and

required a second procedure to restore patency, but most importantly also had a healed ulcer at 2 months.

Suggestions were made for a second generation device for the future, including a shorter device, with less

tissue, 3 cusps, made with porcine pericardium and increased radial strength to reduce compliance. There

are no further publications, however, data has been presented that the complete series consisted of 5

patients, however, early occlusion was seen in 4 of the 5 implants, thought to be due to inappropriate

valve design202, 221. According to investigators reports, 1 valve embolised to the pulmonary artery198.

Pavcnik et al, tested their early first generation BVV1 in human studies222, by implanting the valve by the percutaneous method via a right internal jugular approach into 3 patients with severe deep venous insufficiency. Valves were assessed by descending venography in the first instance and at 2 months, then subsequently by duplex ultrasound up to 12 months. Clinical assessment was also carried out. All patients were fully anticoagulated. 1 patient had a significant perivalvular leak, no reflux through the valve, but importantly, complete ulcer healing, 1 patient had an oversized BVV with valve leak and persistence of clinical symptoms and 1 patient had good improvement in clinical symptoms, but due to tilting had valve leak. The issues with tilting were addressed resulting in the development of second and third generation valves.

More recent work by Pavcnik et al demonstrated early success and deep vein competence with early clinical improvements in 12 of 15 studied patients into whom the BVV3 was implanted. The BVV3 was deployed percutaneously by the transjugular approach into the femoral vein. However, stent valve occlusion was seen in 4 of the 15 patients by 7 months and by 12 months, all the remaining valves had failed and become incompetent198, 203. These are summarised in Table 12.

Table 12: Shows the in human clinical studies

Reference Materials used / valve How did it work How tested Results of testing Further improvements needed

Gale et al202, 2004 Gluteraldehyde fixed Bileaflet valve in nitinol Patency and competency Thrombosis in 4/5 Improved valve design Serino221, 2002 bovine valve in nitinol stent, functioning as assessed by venous valves Improved self expanding stent native valve, implanted duplex and ulcer healing 1 embolised to thrombogenicity (PVVB, VenPro) percutaneously rates noted pulmonary artery

Pavcnik et al223, 2003 BVV1 as above BVV1 as above Percutaneous 2/3 patient improved Issues with tilting need implantation to clinical symptoms addressing 222 Pavcnik et al , 2005 proximal femoral vein, 3/3 valve leak assessed by venography and duplex ultrasound

Pavcnik et al198, 2008 Square stent with frame As previous BVV but Clinical examination, 4/15 occluded – at 12 Geometry needs of nitinol tubing, round with geometry altered to Intravascular US, duplex months non functional optimising geometry, valve leaflets prevent contact of US, venography, air due to incompetency of SIS (Pavcnik, BVV3) leaflet with venous wall plethysmography due to thickening and rigidity of leaflets

Discussion

Deep vein valve replacement by the open surgical technique with autologous or allogenic vein grafts has

been considered a viable option for the treatment of patients with deep venous insufficiency for many

decades. Results from early animal studies were poor as the majority of stents and grafts occluded within

a few weeks199, 200. There have been attempts to develop a prosthetic vein valve implant over 40 years,

with many prelimary models being developed. These include methods to create a new in situ valve, to

create an artificial valve with stainless steel and platinum, the use of a cryo-preserved pulmonary heart

valve, and attempts to transplant cryo-preserved vein valves. More recent developments have seen vein

prostheses involving single, double or triple cusp leaflets mounted on a self-expanding metal frame to develop a percutaneous method of insertion. Animal studies showed promising early patency rates, but poor long-term results201. Nevertheless, 2 clinical studies have been conducted in recent years, but again problems with thrombosis and valve competency resulted in the studies being abandoned202, 203. In

addition to the published works, there are reports of conference proceedings and a plethora of patents for

artificial vein valves, however, the evidence for how these were developed and evaluated is lacking.

Presently, the use of polymers and polymer coated metal stents, used in the vascular system is receiving

much attention, however, there are several problems, particularly for use in the venous system: they are

very thrombogenic, resulting in a deep vein thrombosis and an occlusion of the vein, they lack

haemocompatibilty and biocompatibility, which causes an inflammatory reaction to occur, their presence

results in intimal hyperplasia, which subsequently causing vessel narrowing and material failure. All of

these would serve to worsen the symptoms and signs of venous disease. In addition they lack the

required mechanical properties, as ideally, they should have the ability to expand and contract along with

the vein, which can double in size with changes in the patient position from lying to standing.

The scope of this review excluded valves and experimental models using open surgical techniques using

autologous valves without stents for both animal and in human studies as this would not further

contribute to the development a deep vein valve replacement, however, it is important to consider some

of the tissue engineering techniques employed to study valve transplantation for use in the future

development of a percutaneous deep vein valve replacement.

The long term aims must be to develop a material that is non-thrombogenic, biocompatible,

haemocompatible, is flexible, does not fatigue and avoids intimal hyperplasia. The valve design itself should minimise the zones of stasis of blood, reduce turbulent flow and have low shear stress, as well as remain patent and competent over time. The method of implantation should ideally be via a minimally invasive percutaneous technique and additional thought must be given to the number of valves and their location. Attempts to create a valve implant should and are being revisited here.

4. THE ANATOMY OF DEEP VEIN VALVES

4.1 REVIEW OF THE LITERATURE

4.1.1 Introduction

The deep veins of the lower limb accompany all major arteries and their branches. Blood flows from the

peripheries towards the heart and from the superficial to the deep veins, maintained by unidirectional224, usually bicuspid225, 226, valves which close passively as the valve leaflets move towards the centre of the vein 227. Early anatomists including Charles Estienne (1539), Giovanni Battista Canano (1551)

demonstrated venous valves, but were ridiculed as their work could not be repeated 228 and now, their discovery is generally attributed to Fabricus ab Aquapendente in 1579.

The number, location and consistency of deep vein valves in normal subjects are poorly understood. As an understanding of these details is essential to inform and guide the development of novel interventions and treatments for patients with deep venous disease, a literature review was performed229.

4.1.2 Methods

Relevant papers were identified through systematic searches of Pubmed (1966 to October 2010). The

search terms “femoral vein valve”, “popliteal vein valve”, “femoral vein”, “venous valves” and “venous

valve AND lower limb” were used. English language and human limits were applied and article titles and

abstracts were screening for inclusion. A manual reference list search was also carried out for further

appropriate studies to be considered for inclusion. Relevant full text articles were scrutinised and all

anatomical studies reporting details of deep vein valves were included. Articles concerning deep vein

thrombosis, the treatment of deep venous disease, deep venous valve surgery or replacement, superficial

vein anatomy and cardiac valves were excluded.

4.1.3 Results

Although the initial search strategy yielded 7472 results, following exclusion of non-relevant publications, a total of 7 articles were included. The junction of the femoral vein (FV) and Deep Femoral vein (DFV) was chosen as the reference point for valve positions for several reasons:

• It is readily identified on colour duplex ultrasound imaging, in contrast to the inguinal ligament

• It is unlikely to have been surgically altered or removed, in contrast to superficial venous

reference points

In the studies, if other venous reference points were used, for example, the saphenofemoral junction

(SFJ) and saphenopopliteal junction (SPJ), the distances and locations obtained from the publications were adjusted, using the average values: from the mouth of DFV to the inguinal ligament is 9cm and from the mouth of the long saphenous tributary to the inguinal ligament is 5cm226. A summary of the studies is shown in Table 13 and Table 14 which show the number and location of valves within the femoral and popliteal veins respectively.

Table 13: Shows a summary of the number and location of valves within the femoral veins

Reference Number Age of Sex of Number Location of valves in femoral vein of subjects subjects subjects of valves (distance from Deep Femoral vein) (limbs) in femoral vein

Genovese, 1(2) Not Not 3-4 - circa known known 2002230

Muhl- 32(63) 56.5- 17 male 1-6 0-2 valves proximal 0-9cm berger et 103.25 al, 2008231 15 female 1-3 valves distal 1.9-11.2cm

Powell 27(54) Still born 14 male 2-6 1-2 valves proximal 44 limbs, last at and Lynn, foetus - 84 inguinal ligament 9cm 1951225 13 female 50 limbs valve "just below" 1-3cm distal, then every 2.5-16cm

Banjo, 92(184) All adults 73 male Not 169 limbs ≥1 valve proximal at 5- 1987232 stated: 9cm (study in 19 female black All had a 13 limbs ≥1 valve proximal at 0-5cm Africans) least 1 valve 184 limbs ≥1 valve distal at 0-3cm

Basmajian, 38(76) 31-94 38 male 0-6 51 limbs 1 valve 5-9cm proximal 1952226 2 limbs 1 valve 0-5cm proximal

68 limbs 1 valve 1-3cm distal

Kwakye, 25(50) 60-85 20 male 2-6 Most constant sites: 5-9cm proximal 1971233 and 1-3cm distal. 5 female 6cm distal often 2nd valve (exact figures not given)

Eger and 100(200) 91 adults, Not stated Not stated 138 limbs ≥1 valve at 4-9cm Wagner, 4 children, proximal 1949234 5 infants 62 limbs no valves at 4-9cm proximal

Femoral vein valves

Muhlberger et al 231 used the saphenofemoral junction as a reference point and investigated 63 limbs (32

patients), of which 83% of legs had at least one valve proximal to the junction of the DFV and FV and

distal to the inguinal ligament within the common femoral vein (CFV). Distally, 76-100% of legs had a

valve distal to the junction between DFV and FV. A further cadaveric study of 54 legs from Powell, 1951

reported that a valve in the common femoral vein (CFV) was present in nearly three quarters of patients.

The CFV valve was located consistently at the level of the inguinal ligament. The FV contained 1-4 valves in all cases, with at least two valves seen in over 90% of cases. The most common site for a valve was just distal to the junction of the DFV and FV (93% of cases).

Fifty legs were studied by Kwakye233, who found that a mean of 4 valves were present in the FV, ranging from 2-6. Consistently, a valve was seen just distal to the DFV opening, and just proximal to the SFJ.

This may correspond to the valve just below or at the inguinal ligament, consistently reported in other papers. The femoral vein was more likely to contain valves proximally than distally, unlike the popliteal vein, where valves were more frequently cited in the distal part. Using a reference point of the inguinal ligament, a further study reported that 76% of the 76 legs studied had a valve in the five centimetres inferior to the inguinal ligament and a valve immediately distal to the DFV / FV junction was present in nearly 90% of legs226. One limb had no valves throughout the femoral vein, the median number

identified was 3 with a maximum of six.

Banjo et al232 divided veins into 7 segments in a study of 92 black African subjects. In keeping with other

reports, all patients in this study had a valve in the 3cm distal to the DFV junction of the FV and a valve

was present in the CFV in 93% of legs.

Eger and Wagner, 1949234, carried out the largest study of 200 limbs (100 cadavers) by dissection and found that 69% of the 200 limbs studied had valves present within the common femoral vein proximal to the saphenofemoral junction, that is within 9cm proximal to the junction of the FV and DFV. There was complete absence of valves within this region in 31%. This study did, however, only focus on the region above the saphenofemoral junction, excluding the 4cm immediately proximal to the FV and DFV junction, which is a limitation, as well as not discussing the exact locations of these valves.

Popliteal vein valves

The popliteal vein (PV) is defined as the segment of vein between the adductor hiatus and confluence of

anterior and posterior tibial veins. Studies previously discussed including, Powell, 1951225 and Kwayke,

1971233, have consistently reported the presence of a valve just distal to the adductor hiatus. A second

valve was commonly seen in the very distal segment of the PV, just proximal to anterior and posterior

tibial veins, Santilli, 2000235 and Kwayke, 1971233.

Table 14: Shows the summary of number and location of valves within the popliteal veins

Reference Number Age of Sex of Number Location of valves in popliteal vein of subjects subjects subjects of valves (distance adductor hiatus) (limbs) in popliteal vein

Genovese, 1(2) Not Not 1-2 - circa known known 2002230

Santilli et 39(47) Not 13 male 1.8+/-0.5 1st valve: 7.0 +/-5.0cm al., 2000235 known 26 female Last valve: 14.1+/-3.0cm

Powell 27(54) Still born 14 male 1 in upper Just distal to adductor hiatus (1-3cm) and Lynn, foetus - 84 1/3 (rest 1951225 13 female not visualised)

Kwakye, 25(50) 60-85 20 male 0-4, Most constant sites: above junction 1971233 average 1- anterior and posterior tibial veins 5 female 2 and near adductor ring

4.1.4 Discussion

The validity of these finding is, however, limited by the small number and heterogeneity in reporting of studies. Fixed anatomical reference points, including the inguinal ligament and DFV tributary, have been used to amalgamate with some confidence, the studies, although as the distances used to adjust the distances used were average values this will have caused some inaccuracy. Moreover, many of the reports were of cadaveric studies in patients without venous disease or the presence or absence of venous disease was not discussed, so the application of these data to a patient population with deep venous disease is suboptimal.

In summary, in the studies included, venous valves are consistently located in specific sites in the common femoral, femoral and popliteal veins of the leg. This is summarised in Figure 12.

Figure 12: Shows the most constant locations of valves within the femoral and popliteal veins

The most consistently reported sites of valves are:

• in the CFV, near the inguinal ligament

• in the FV just distal to the DFV tributary

• in the PV near the adductor hiatus

As a result of these findings the clinical studies in this report, including the ultrasound and magnetic

resonance (MR) imaging of deep vein valves and their function has specifically target these anatomical

97 locations. In addition, this information may also be of value once a minimally invasive treatment for patients with deep vein valve failure has been developed, these are sites that should be considered for location of the device.

5. DAMAGED VENOUS VALVES

5.1 A SURVEY ON THE MANAGEMENT OF PATIENTS

WITH COMPLEX VENOUS REFLUX AND

OBSTRUCTION

5.1.1 Introduction

As discussed in the previous chapters, the management of venous disease is variable both between centres and even between different specialists within the same institution. This often depends on their personal expertise and experience. Large trials are difficult and due to the heterogenous nature of patients the exclusion rates are often high236 and therefore often not applicable to the patients that we see in daily practice. For these patients, the underlying haemodynamic abnormalities are variable and similar underlying disease patterns do not results in the same signs and symptoms33 in different patients. For this reason patients are treated on an individual basis, and assessed by each lead physician. Many patients have complex underlying venous disease with mixed venous reflux and obstruction, and many patients have had intervention, particularly to their superficial veins previously237. Following this they will have different degrees of recurrence of their superficial venous disease and different effects on deep venous haemodynamics. Vascular specialists manage patients such as this with complex deep venous reflux and obstruction in their everyday practice. Therefore, by performing a survey of venous experts, this study aims to guide the management of these challenging patients.

5.1.2 Methods

Eight patient scenarios with descriptions of their symptoms and signs as well as their vascular and

imaging demonstrating complex venous reflux (5 patients) and obstruction (3 patients) or mixed disease,

were developed. Questions were on the preferred choice for the management of these patients, and the

experts were asked to rank the choices given in the order in which they would choose them. The

questionnaire was sent to 10 venous experts for their review. Each of them gave feedback on the

questions and requested changes. These were made and the same 10 specialists were given the

questionnaire again for their review. This was to ensure all possible management options that would be

considered in clinical practice were listed and that the questionnaires were easy to understand and

complete. Of the ten complex cases selected, all five patients in the reflux cases had C5-6 disease and

mixed superficial and deep reflux with or without chronic deep venous obstruction, and of the

obstruction cases, 3 were relating to the lower limbs, and 2 to the upper limbs. The upper limb cases are not discussed further in this paper as they are not relevant to the management of lower limb deep venous disease. The cases are described in complete detail in the results section for ease of understanding. An online survey program, www.freeonlinesurveys.com, was employed and the anonymous surveys were

distributed by e-mail to vascular specialists. The surveys sent can be seen in Appendix 1. This was done through the administration staff of each society to member of the American Venous Forum, the

European Venous Forum, the UK Venous Forum and the Vascular Society of Great Britain and Ireland.

The first and second preferred management options were taken and the results are shown below.

5.1.3 Results

For the deep venous reflux survey, to date, there have been 77 responses, including 38 Consultants. For

patients with deep venous obstruction, there have been 35 responses, including 25 Consultants. Each

individual case is discussed below along with the preferred first and second choice management options

for patients with these conditions.

100

Reflux

Case 1

A 64 year old female, who is currently taking HRT presents with a one year history of an aching, painful right leg, with heaviness and swelling in right leg, despite adequate compression dressings. On examination she has C6 disease with lipodermatosclerosis and an active 7x3 centimetre ulcer present in the medial gaiter area, which has been non-healing despite wearing compression for 6 months. Her ABPI was greater than 1 bilaterally. Her duplex ultrasound imaging is shown in

Figure 13, and the thrombus in the iliac vein appeared chronic. The options for the initial management of this patient were; compression only, anticoagulation with warfarin, anticoagulation with treatment dose heparin alone, open surgery, iliac vein stent, treatment of the superficial veins or pharmacomechanical thrombolysis. The results are shown in Figure 14 and Figure 15. The preferred first choice management option by over half of respondents was to proceed with an iliac vein stent and the second choice was to treat the superficial venous reflux.

Figure 13: Shows the duplex ultrasound imaging for Reflux Case 1

101

Figure 14: Shows the results for Reflux Case 1 - 1st Choice of management

Reflux Case 1 - 1st choice Compression only

4% Anticoagulation 3% 3% Anticoagulation Heparin 19% Open surgery

10% Iliac vein stent

53% 7% 1% Treatment superficial veins Pharmacomechanical thrombolysis

Other

Figure 15: Shows the results for Reflux Case 1 – 2nd Choice of management

Reflux Case 1 - 2nd choice Compression only

4% 2% Anticoagulation

Anticoagulation Heparin 15% Open surgery 29% Iliac vein stent 23% Treatment superficial veins 13% 10% Pharmacomechanical thrombolysis 4% Other

Case 2

A 53 year old female, who had previous had bilateral great saphenous vein ablation GSV ablation 3 years ago, presented with a 6 month history of cramping, aching, heavy, throbbing legs with healed a ulcer on the right and an active 5x4 centimetre venous ulcer on the left. Her symptoms were persisting despite compression stockings. Her

ABPI was greater than 1 bilaterally. Her duplex ultrasound imaging is shown in Figure 16. Respondents were ask whether they would carry out any further investigations to further guide the management of this patient; no

102 additional investigations; CT Venography; MR Venography; conventional venography. The results of this are shown in Figure 17. Although the majority (31%) preferred to perform MRV to guide further management of this patient, 22% opted for CTV, and 26% did not require any further imaging of this patient. Further they were asked how they would best manage the patient’s left leg, that is the leg with active venous ulceration. The options given were compression only, drug therapy such as daflon, open surgical options including transposition, valve transplant, external valvuloplasty, internal valvuloplasty, or another preferred option. The results of this are shown in Figure

18. 68% of respondents recommended continuing with compression therapy despite the fact that the patient had not responded to it in 6 months, and 32% of respondents recommended surgery, of which valve transposition was the preferred option. With regard to the right leg, that is the leg without ulceration present, 76% of respondents recommended continuing with compression therapy, and 13% recommended the use of medication, only 11% in this case opted for any surgical treatment options.

Figure 16: Shows the duplex ultrasound imaging for Reflux Case 2

103

Figure 17: Shows the results for Reflux Case 2 - Investigations

Reflux Case 2 - Investigation No Investigations

7% CTV 14% 26% MRV

31% 22% Venography

Other

Figure 18: Shows the results for Reflux Case 2 - Management of the patient’s left leg

Left leg Compression

3% 3% 3% Daflon

Transposition 15%

8% Transplant 68% External valvuloplasty

Internal Valuloplasty

Figure 19: Shows the results for Reflux Case 2 - Management of patient’s right leg

Right leg Compression

0% 2% 2% Daflon 7% Transposition 13%

Transplant

76% External valvuloplasty

Internal Valuloplasty

104

Case 3

A 69 year old gentleman with a high BMI of 40 and heart failure with a poor performance status NYHYA grade III and limited mobility due to shortness of breath presents following a DVT 20 years ago with a 3 month history of worsening pain and swelling of the left leg and ankle. He has tried compression, but this worsens his symptoms and so has not been compliant with its use. On examination he has lipodermatosclerosis in left leg and a non healing ulcer over the medial gaiter area which is 6x6 centimetres in size, and has been present for 1 year. He has bilateral

ABPIs greater than 1 and his duplex ultrasound assessment is shown in Figure 20. The further options for the management of this patient that were offered included compression therapy, anticoagulation with warfarin, anticoagulation with treatment dose heparin alone, open surgery, treatment of the superficial veins, or pharmacomechanical thrombolysis. The preferred first choice management option for this patient was to persist with compression therapy (45%) as far as the patient could tolerate, and several respondents stressed that this was their preferred and most conservative management option due to the patients multiple comorbidities. 34% of respondents did suggest treating the superficial veins as either the first or second preferred treatment option, but others commented that if compression exacerbated his symptoms then treating the superficial veins would be ill advised. This is shown in Figure 21 and Figure 22.

Figure 20: Shows the duplex ultrasound imaging for Reflux Case 3

105

Figure 21: Shows the results for Reflux Case 3 - 1st Choice of management

Reflux Case 3 - 1st choice Compression

Warfarin 6% Heparin 19% 45% Open surgery 8%

Treatment Superficial Veins 5% 17% Pharmacomechanical

Figure 22: Shows the results for Reflux Case 3 – 2nd Choice of management

Reflux Case 3 - 2nd choice Compression

Warfarin 6% 15% 29% Heparin

9% Open surgery 9% Treatment Superficial Veins 32%

Pharmacomechanical

Case 4

A 49 year old lady with a history of breast cancer and left sided DVT 4 years ago, and previous ablation of the left great saphenous vein 2 years ago presented with a 1 year history of left leg pain, itching, oedema and recurrent leg ulceration. She had multiple recurrence of the same ulcer in the left, totaling 4

106

ulcers in the last year despite wearing compression stockings continually. On examination in clinic, there

was evidence of a healed ulcer around the left medial malleolus which was 2x2 centimetres. Her ABPIs

were greater than 1 bilaterally. Her duplex ultrasound assessment is shown in Figure 23. The options for the management of this patient that were offered were compression only, anticoagulation with warfarin, anticoagulation with treatment dose heparin alone, open surgery to treat the venous reflux, open surgery to treat obstruction, stenting of the iliac vein or pharmacomechanical thrombolysis with one of the novel devices such as Trellis®. The results are shown in Figure 24 and Figure 25. 50% of respondents considered that the preferred first choice management option was to stent the iliac vein to treat the obstruction in the first instance. A further 35% recommended continuing compression therapy only, and

7% opted for open surgery to treat the obstruction as a first line treatment. As a second choice option, all of the given choices were considered to be possible options.

Figure 23: Shows the duplex ultrasound imaging for Reflux Case 4

107

Figure 24: Shows the results for Reflux Case 4 - 1st Choice of management

Reflux Case 4 - 1st choice Compression

2% Warfarin

Heparin 35% Surgery - reflux

50% Surgery - obstruction

7% 4% Iliac vein stent 0% 2% Pharmacomechanical

Figure 25: Shows the results for Reflux Case 4 – 2nd Choice of management

Reflux Case 4 - 2nd choice Compression

Warfarin 10% 21% Heparin 17% Surgery - reflux

Surgery - obstruction 15% 25% Iliac vein stent 8% 4% Pharmacomechanical

108

Case 5

A 30 year old gentleman, who was a lifelong heavy smoker of 40 cigarettes per day presented with a 9 month history of heaviness and throbbing pain in the right leg with oedema, and a 6 x6 centimetre ulcer in gaiter area. This had failed to heal despite 9 months of well applied 4 layer bandage therapy. He had a previous right femoral vein DVT 5 years ago following flight from Australia. He took warfarin at that time for 6months, and had not been followed up with regard to this since. Bilaterally his ABPIs were greater than 0.9. His duplex imaging is shown in Figure 26. A total of 79% of respondents preferred to continue with compression therapy alone as a first or second choice option, despite this having failed to heal the ulcer for 9 months. A total of 58% of respondents treated the superficial veins as a first or second line for this gentleman, and only 23% of respondents thought that open surgery to treat the deep venous obstruction would be the best option for this patient management.

The options offered to those taking the survey for the ongoing management of this patient were: continuing compression therapy, anticoagulation with warfarin, anticoagulation with treatment dose heparin alone, open surgery, treatment of his superficial veins or pharmacomechanical thrombolysis. A total of 79% of respondents preferred to continue with compression therapy alone as a first or second choice option, despite this having failed to heal the ulcer for 9 months. A total of 58% of respondents treated the superficial veins as a first or second line for this gentleman, and only 23% of respondents thought that open surgery to treat the deep venous obstruction would be the best option for this patient management.

109

Figure 26: Shows the duplex ultrasound imaging for Reflux Case 5

Figure 27: Shows the results for Reflux Case 5 - 1st Choice of management

Reflux Case 5 - 1st choice Compression

3% Warfarin

Heparin 27%

51% Open surgery 13% Treatment Superficial Veins 6% 0% Pharmacomechanical

110

Figure 28: Shows the results for Reflux Case 5 – 2nd Choice of management

Reflux Case 5 - 2nd choice Compression

3% Warfarin

28% Heparin 31% Open surgery

10% 23% Treatment Superficial Veins

5% Pharmacomechanical

Obstruction

Case 1

A 34 year old lady, who is 13 weeks pregnant presented with an 8 day history of acute, painful left lower

leg swelling. She had no history of a previous DVT, although her mother had a DVT while she was

pregnant and now has a post phlebitic limb and venous ulceration. The patient’s only history was that of mild asthma. Her duplex imaging is shown in Figure 29. Respondents were asked to rate the following management options in their order of preference: Compression, anticoagulation with warfarin, anticoagulation with treatment dose heparin alone, open thrombectomy, catheter directed thrombolysis or pharmacomechanical thrombolysis with one of the novel devices such as Trellis®. The preferred first choice management option was anticoagulation with heparin, in two thirds of respondents. All of the other options were considered as possible methods of treatment as a second line therapy including interventional treatment, such as pharmacomechanical or catheter directed thrombolysis and open surgery. Several commented that they would only perform this, however, if heparin therapy failed. These results are summarized in Figure 30 and Figure 31.

111

Figure 29: Shows the duplex ultrasound imaging for Obstruction Case 1

Figure 30: Shows the results for Obstruction Case 1 – 1st Choice of management

Obstruction Case 1 - 1st Choice Anticoagulation - warfarin 3% 0% Anticoagulation - heparin

Open thrombectomy 3% 15% 2% Catheter directed thrombolysis 10% Other 67% Pharmacomechanical

Compression

112

Figure 31: Shows the results for Obstruction Case 1 – 2nd Choice of management

Obstruction Case 1 - 2nd Choice Anticoagulation - warfarin

0% Anticoagulation - heparin

21% 17% Open thrombectomy Catheter directed thrombolysis 12% 21% Other 8% Pharmacomechanical 21% Compression

Case 2

A 48 year old male rock singer with a 3 year history of a painful swollen left leg after a DVT 10 years before has

recurrent left leg ulceration. Wearing compression stockings does improve his symptoms slightly, but do not

prevent ulcer recurrence. On examination he has left leg swelling, evidence of a recently healed ulcer in the left

gaiter area and lipodermatosclerosis. He has palpable pedal pulses and his ABPIs bilaterally are greater than 1. The

results of his CT venograms and duplex ultrasound are shown in Figure 32 and Figure 33 respectively. As shown in the imaging, the patient had developed collaterals to bypass the obstruction in the iliac vein, this could be called a

“self-Palma” procedure. The management options that were offered for consideration included compression,

anticoagulation with warfarin, anticoagulation with treatment dose heparin alone, open surgery such as a Palma

procedure or in line bypass or an interventional procedure such as an iliac stent. 66% of respondents preferred iliac

stenting as a first or second choice for the preferred management of this patient. 37% considered that open surgery

should be considered as a first or second choice, and 29% would continue compression alone as the first line

treatment.

113

Figure 32: Shows the CT Venogram for Obstruction Case 2

Figure 33: Shows the duplex ultrasound imaging for Obstruction Case 2

114

Figure 34: Shows the results for Obstruction Case 2 – 1st Choice of management

Obstruction Case 2 - 1st Choice Warfarin

Heparin 15% 3% 29% Open surgery (bypass) 7% Intervention (stent)

2% 44% Other

Compression

Figure 35: Shows the results for Obstruction Case 2 – 2nd Choice of management

Obstruction Case 2 - 2nd Choice Warfarin

Heparin

19% 0% 22% Open surgery (bypass)

7% Intervention (stent) 22%

Other 30%

Compression

Case 3

A 55 year old male suffered a subdural haematoma following a road traffic accident 4 months ago, which was treated with craniotomy and surgical evacuation. Subsequently the patient had a thrombosis of the

115

IVC managed with an IVC filter. Initially, the symptoms resolved for a week then recurred. On

examination the patient had bilateral lower extremity oedema with varicose veins. The duplex imaging of

this patient is shown in Figure 36. The options offered for the management of this patient included

anticoagulation with warfarin, anticoagulation with treatment dose heparin, pharmacomechanical

thrombolysis, open surgical thrombectomy, open surgical bypass, stenting or catheter directed thrombolysis. 56% of respondents considered that anticoagulation with heparin or warfarin was the preferred first line management despite the head injury. 28% considered interventions such as catheter directed or pharmacomechanical thrombolysis or stenting to be more appropriate in the first instance.

These results are shown in Figure 37 and Figure 38.

Figure 36: Shows the duplex Ultrasound Imaging for Obstruction Case 5

116

Figure 37: Shows the results for Obstruction Case 3 – 1st Choice of management

Obstruction Case 3 - 1st Choice Warfarin

Heparin

Open Thromectomy 13% Bypass 10% 36% Stent 10% Pharmacomechanical 8% 20% Catheter directed thrombolysis 0% 3% Compression

Figure 38: Shows the results for Obstruction Case 3 – 1st Choice of management

Obstruction Case 3 - 2nd Choice Warfarin

Heparin

0% Open Thromectomy 16% 4% 28% Bypass 4% Stent

8% 20% Pharmacomechanical

20% Catheter directed thrombolysis

Compression

117

5.1.4 Discussion

The majority of cases had a clear preferred option for initial management. For patients with a combination of reflux and obstruction, if the obstruction is proximal, such as in the iliac, this survey found that the majority of venous specialists would treat the obstruction in the first instance. If the obstruction is more distal, such as in the femoral or popliteal veins, the majority of respondents treated the superficial venous reflux first. For patients with combined superficial and deep venous reflux, the preferred option was to treat the superficial venous system first before considering any treatment of the deep venous disease. Compression was considered a reasonable option for all cases of venous disease, and all combinations of obstruction and reflux. In the presence of deep venous reflux and leg ulceration, surgical management options were considered. In the presence of malignancy or pregnancy heparin was the preferred method of anticoagulation and surgical or interventional management options were less popular. For a patient 4 months following a head injury, anticoagulation was considered a safe option by more than half of those taking the survey.

Preferred management options for patients with complex deep vein reflux, deep venous obstruction or mixed reflux and obstruction are variable across the world. A consensus would be valuable to determine the best management options for these patients. In general, anticoagulation and more conservative management options are the preferred options. Interventional and open surgical procedures were considered second or third line.

118

6. IMAGING THE VENOUS SYSTEM TO

DEVELOP A COMPUTATIONAL MODEL

OF A DEEP VEIN VALVE

6.1 INTRODUCTION

6.1.1 Venous Haemodynamics

The circulatory system moves blood around the body and the venous circulation is responsible for returning blood back to the heart for recirculation. For the most part, the veins carry deoxygenated blood from the peripheries to the right side of the heart, except the pulmonary vein, which bring oxygenated blood from the pulmonary circulation to the left side of the heart. In order to maintain antegrade flow from the peripheries towards the heart, there are several requirements:

• A pressure gradient

• A venous pump, that is a muscle pump or respiratory pump

• Competent unidirectional venous valves

The venous system has deep, superficial and perforating veins. The deep veins run below the muscular fascia, the superficial veins run above the muscular fascia and the perforating veins which link the two systems and perforate the fascia. The direction of blood flow in the normal circumstances is from superficial to deep veins and from distal to proximal, that is from the limbs towards the heart, in a prograde fashion.

119

Blood flow in veins is regulated by several different mechanisms, including gravity, peripheral muscular pumps, the respiratory pump and ultimately, venous return must equal cardiac output in homeostatic conditions, therefore, this will also have an effect on flow. However, veins are compliant and have the capacity to pool blood for a short time. Venous capacity can be controlled passively by gravity, causing an increase in transmural pressure, resulting in expansion of the veins. There is also a hormonal and neural control of veins which is regulated both centrally and at a local level.

The sympathetic nervous system innervates the smooth muscle in the vein wall via α1-adrenoceptors. An increase in activity of these receptors causes venoconstriction, and a decrease below the baseline tonic level of activity causes venodilatation. At the general level, and increase in circulating hormones such as adrenaline or angiotensin II also cause venoconstriction.

6.1.2 Flow Direction

Blood flow in healthy veins the blood flow in veins is directed towards the heart. In general that means

from superficial to deep veins and from foot to groin. This flow direction is called antegrade or

retrograde. Normally functioning valves in the lower limb veins prevent retrograde flow. Reflux is the flow in the opposite direction to physiological flow.

Antegrade flow in the vasculature is generated by the pumping of the heart, which generates a pressure to propel blood through the arteries. This pressure is dissipated gradually as blood flows through the arteries and capillaries, vein and right atrium, therefore, in the supine position, antegrade flow will continue down this pressure gradient, in the veins and return to the right side of the heart. In the lower limb venous system, the return of blood to the right atrium is augmented by the work of respiration and the lower limb muscle pumps, most notably the calf muscle pump. Retrograde flow is prevented by the presence of unidirectional valves. These mechanisms ensure that in the normal healthy state, the flow of blood in the lower limbs is from the superficial to the deep system and from the feet towards the heart, maintaining antegrade flow.

120

The pressure gradient between the capillaries and the right atrium is the dominant driving force for the

venous blood flow. In the supine position, at the venous end of the capillary, the pressure is

approximately 12-18mmHg238. The pressure in the larger veins, such as the femoral vein in approximately

8mmHg and in the right atrium is 4-7mmHg, therefore, in the supine position, antegrade flow will continue down this pressure gradient, in the veins and return to the right side of the heart. In the standing position, the gradient between the artery and vein at any level is maintained as the gravitational pressure affects both the artery and vein to the same hydrostatic pressure value. At the entrance to the right atrium the pressure is also pulsatile. As blood enters the right atrium, it is subjected to a pulse as the right ventricle contracts against the closed tricuspid valve. This augments the entry of blood from the vena cava to the right atrium. It is possible that this pulsatile pressure wave is transmitted throughout the venous system239, however, this would have only a very weak effect on the flow and remains a controversial issue.

The respiratory pump has an effect on venous return. During inspiration and expiration, changes in intrathoracic and intra-abdominal pressures are transmitted to the right atrium, vena cava and intra- abdominal vessels and have an effect on venous return. During inspiration, intra-thoracic pressure falls and this is transmitted to the right atrium. In turn, the venous pressure gradient along which blood flows back to the heart is increased, promoting venous return. When the diaphragm contracts and descends, the intra-abdominal pressure increases, compressing the inferior vena cava, driving blood into the thorax, but as inspiration continues and intra-abdominal pressure increases further, the IVC can collapse and cause a reduction or complete cessation of flow from the lower limbs240-244. During expiration the reverse

is true. An example of the velocity of blood flow in the femoral vein in the supine position during quiet

respiration in a normal subject demonstrated using ultrasound is shown in Figure 39 below.

121

Figure 39: Show the velocity of blood flow in the femoral vein during quiet respiration in the supine position of a

normal subject

The muscle pumps of the lower limb are also a major driving factor for venous return. These include the

calf muscle pump, the foot pump and the thigh pump. Contraction of the surrounding skeletal muscle

compresses the vein, forcing blood out of he vein and towards the heart, as in the healthy state, reflux is

prevented by bicuspid valves. During relaxation, the pressure is reduced in the deep veins, causing blood

to flow from the superficial veins, via the perforating veins, to the deep venous system along a pressure

gradient. Again, reflux back to the superficial system is prevented by valves in the perforating veins238.

The calf muscle pump, consisting of the gastrocnemius and soleus muscles, is the most effective of these

muscle pumps14 and in the normal subject should expel more than 60% of the volume of blood in the

calf veins as determined by air plethysmography245. During a normal ambulation, the pressure is approximately at 20 – 30 mmHg in normal subjects246. The Cockett perforators (posterior tibial perforating veins) contain valves which close during muscle systole to prevent high pressure from the deep system being transferred to the superficial, low pressure chamber. This is the reason why the system is sometime referred as the “peripheral heart”. Conversely, in muscle diastole, the valves prevent backflow and allow drainage from superficial to deep chambers via the perforating veins aided by

122

temporarily lowered pressure in the deep veins14. The foot pump also make a contribution expelling

approximately 25 millilitres of blood during each step in ambulation247.

There have been many studies to establish which is dominant, and as such, muscle contractions during

varying respiratory exercises have been investigated 242, 243, 248, 249. This remains controversial and is

obviously dependent on position, activity and respiration of the subject, however, Miller 243 concluded that respiration maintains a persistent effect during both mild and moderate calf contraction.

Retrograde flow occurs if the blood flows from deep to superficial veins, or from the direction central to peripheral. Pathological reflux is considered to be occurring if the duration of retrograde flow is more than 0.5s. There was a suggestion by Labropoulos et al that the cut off value for reflux in the superficial and deep calf veins should be 0.5s, but the cut off value for the femoropopliteal veins should be greater than 1s250. However, a more recent study by Lurie et al concluded that a universal cut off of 0.5s throughout the venous system to describe pathological reflux would improve reliability251. These measurements were all performed using duplex ultrasound which offers information about venous flow patterns and reflux, it is safe, reliable and readily available in the clinic setting. Reflux time itself, however, does not correlate well with clinical severity 252. The cause is failure of the valves, which should maintain unidirectional flow, present in the superficial, perforating and deep veins of the legs. This may be due to

primary or secondary failure. Primary valve failure is caused by weakening of the vessel wall leading to

dilatation and separation of the valve leaflets. Secondary valve failure is due to damage to the valves

commonly due to venous thrombosis. Both of these result in refluxing disease. Duplex studies have shown that reflux (or less commonly, occlusion) of deep veins is seen in around 40% of patients with

chronic venous insufficiency and may be significant in 15-20%7, 8.

123

6.1.3 Venous Pressure (Normal, supine, stationary, walking)

Introduction

Vascular haemodynamics is governed by blood flow, pressure and vascular resistance. Veins are thin

walled capacitance vessels which contain approximately two-thirds of the total circulating blood

volume253, depending on the tone of the vein wall smooth muscle. Their function is to return blood to the heart. The return of this large reservoir of blood to the heart is carried out by a combination of:

• peripheral muscle pumps which squeeze blood from the veins during muscle contraction

• gravity, which assists venous return in the raised upper limb and cerebral circulation and opposes

return from the lower limbs in the standing position

• respiration, which changes intra-thoracic pressure, contributing to a pressure gradient from the

peripheral to the central veins

Uni-directionality from peripheries towards the heart and from superficial to deep veins is ensured by the

presence of many, mostly bicuspid, venous valves. Valves are more frequent in the more distal veins and

are not present in the vena cava229. Pressure in the venous system, in contrast to the arterial system in

which it is generated by the heart, is determined by forces which oppose venous return and forces which

promote venous return.

Factors opposing venous return from the lower limbs

The pressure generated by gravity, known as the hydrostatic pressure, is the weight of the column of

blood from the right atrium. In the standing position, in the lower limbs, this opposes the return of

blood to the heart from the peripheries, in the raised upper limb it promotes it. It is calculated by

measuring the vertical distance from the right atrium to the point under consideration254, 255. The pressure is calculated by taking into consideration the density of blood (1.056g/cm3) and the acceleration of gravity

(980cm/s2) and expressed in mmHg238, according to the following equation:

124

Hydrostatic Pressure = Density of blood (g/ml) x Acceleration of gravity (cm/s2)

(mmHg) Hg constant (dynes/cm2)

Hydrostatic Pressure = 1.056 x 960

1333

Hydrostatic Pressure = 0.77 mmHg (per cm below the right atrium)

Therefore the effect of gravity increases the pressure by 0.77mmHg/cm below the level of the right

atrium.

As the pressure increases, the volume of the vein increases. Between 0 and 10 mmHg, the pressure-

volume curve for the vein is steep as the vein wall distends easily. Above 10 mmHg the curve flattens as

the vein resists further distension. In addition, the change in shape of the vein can be observed, at very

low pressures (<1 mmHg) the vein adopts a dumbbell shape, and as the pressure increases it becomes

more elliptical, and then spherical to accommodate a larger volume of blood. This is shown in Figure 40

below.

125

Figure 40: Shows the venous pressure-volume curve, reproduced from Katz et al, 1969256

This increase in pressure does not occur instantly, but gradually increases over 30-60 seconds during standing as blood from the arterial system passes into the venous system254.

Importantly, the effect of gravity cannot and must not be explained simply because there is less flow in

the veins in the standing position in the lower limbs or the lowered arm position in the upper limbs. In

fact, the flow volume itself must remain the same in both the arteries and veins as the system is closed

and gravity will have an equal effect on both veins and arteries in the peripheries, so that in the mean the

inflow will be equal to the outflow in any position. The pressure in the veins and velocity of flow will

however demonstrate changes. Any overall changes in flow volume must occur in both the arteries and

the veins simultaneously due to changes in cardiac output, which can be altered by venous return,

according to the Starling law.

In the supine position, without the effect of the muscle pump, the pressure in the lower limb veins is 0-

10mmHg as there is little or no vertical difference between the peripheral veins238 and the right atrium.

The venous pressures in the supine position are summarised in the Figure 41 below. Any pressure gradient is generated by changes in intra-thoracic and intra-abdominal pressures due to the work of respiration, which is discussed further later in the chapter.

126

Figure 41: Shows the venous hydrostatic pressures in the supine position (adapted from Meissner et al238)

Supine Venous Pressures (mmHg)

4.6 7.5 10

In the standing position, if no muscle contraction is taking place, the pressure at the distal calf is

94mmHg in an average 174cm man257, which is generated by the hydrostatic pressure alone. Venous pressures in the standing position are summarised in Figure 42 below. During calf contractions, either during walking or heel raising, the pressure decreases to a mean of 22mmHg within 7-12 steps254. This is

the ambulatory venous pressure which ranges from 0-25mmHg. Once static, the pressure increases again to resting hydrostatic pressure in 14+/-11 seconds after tiptoeing or 19+/-13 seconds after knee

flexion258.

127

Figure 42: Shows the venous hydrostatic pressures in the standing position (Adapted from Meissner et al238)

Standing Venous Pressures (mmHg)

-32

-15

Factors promoting venous return in the lower limbs

Muscle Pumps

At the venous end of the capillaries, there is a pressure generated by the pumping action of the heart,

called vis a tergo. Most of the energy is absorbed by the arteries, but in the supine state, this pressure,

usually 12-18mmHg, is sufficient to generate a pressure gradient, so blood flows to the right atrium. In the standing position, however, the pressure is used in the work of dilating the veins so that blood progression is delayed until a maximal volume, just before producing an appreciable progression. The role of the muscle pumps is to work in the lower part of the curve of compliance, with low volume and low transmural pressure. (Work=Energy=Transmural Pressure x Volume change).

In the limbs, contraction of the skeletal muscle squeezes the veins within it and expels the blood towards

the heart. The veins are emptied, decreasing the volume in the veins, the pressure decreases, then the

veins fill from the more distal veins during muscle relaxation. The presence of venous valves, which are

128

more frequently located distally229 where the muscle pump has more effect on the venous return, prevent

reflux of blood. If the valves are incompetent, the muscle pump becomes ineffective259, the veins distend and the pressure increases in the vein (venous hypertension). The pressure is transmitted to the capillary beds, fluid moves out of the vessels into the surrounding tissues, causing oedema. If this is persistent, skin changes occur including pigmentation and ulceration.

The lower limb muscle pump, or “the peripheral heart” as it is sometimes known, consists of the calf muscle pump, the foot pump and the thigh pump. The calf muscles, consisting of gastrocnemius and soleus on one contraction expel over 60% of the volume they contain260, 261, which can be up to 100-

150ml in the normal limb. During walking, repeated contractions of the calf muscle will reduce the pressure in the dorsal veins of the foot from resting hydrostatic pressure to the ambulatory venous pressure, which is in the range of 5 to 30mmHg254. These values can be measured invasively or indirectly by foot volume, air or strain gauge plethysmography. The calf muscle pump can also be augmented extrinsically using either intermittent pneumatic compression, in which applies a direct squeeze to the calf, or using electrodes applied to the skin which either stimulate the calf muscle to contract, known as electrical muscle contraction, or in which a skin electrode stimulates a peripheral nerve which cause contraction of the calf muscle indirectly262. In addition, the thigh and foot pumps make a contribution to

venous return from the lower limb, but to a lesser extent than the calf muscle pump. The thigh has been

shown to have an ejection fraction of 20% using plethysmography260, this is likely to be because the deep veins are not compressed as much by each muscle contraction. Compression of the plantar venous plexus on weight bearing forces blood from the plantar venous arch of the foot to the calf263.

The muscle pump fills the veins with distal blood when relaxing (diastole) and propels it upwards when contracting (systole) thanks to the upstream and downstream valves that alternately open and close ensuring unidirectional antegrade flow. This creates Gravitational Hydrostatic Pressure Fractioning

(GHPF)264, 265. In the upright position, the GHPF generates a pressure which is lower below the knee than above, or a retrograde PG. This causes a retrograde flow in both the superficial and deep veins.

129

Muscle Pump Tests

There are several muscle pump tests that have been suggested in research and clinical practice. Muscle pump tests are designed to measure both the rate and direction of venous flow during the systolic and diastolic phases of MP activity, particularly the calf pump, which is the most important of the lower limb muscular pumps. The ideal test conditions which mimic daily activities such as walking and running are not readily available in daily practice. Some manoeuvres try to recreate these conditions in a reproducible fashion.

• The Parana manoeuvre consists of a slight push-pull at the waist of the patient triggering a

proprioceptive isometric reflex contraction of the calf to restore balance and has been used in

different studies to detect deep and superficial vein incompetence266, 267.

• The oscillation manoeuvre consists of accompanying a forward slight oscillation of the body with

a hand on the patient’s waist.

• The Wunstorf manoeuvre consists of a voluntary flexion of the toes, causing the calf muscle

veins to empty passively268. It has been the basis for different studies to investigate venous

insufficiency

• A calf squeeze is commonly used and can be performed manually or by inflation of a cuff. It is

considered that this is less likely to mimic physiological conditions as the MP is not stimulated. It

may compress veins not physiologically squeezed by the muscles, for example the superficial

veins and in addition, may not be feasible due to pain, lipodermatosclerosis, skin changes or

ulceration. However, it can be useful in the supine and sitting positions when the manoeuvres

described above are not possible. A calf squeeze by insufflations of a cuff would, however, be a

consistent and reproducible method of applying the same squeeze for repeated reproducible

measurements.

130

Muscle Pump Diastolic Reflux Grading

A reliable grading of the muscle pump diastolic reflux would improve the diagnostic reproducibility of reflux and allow a means of quantifying haemodynamic outcomes of different treatments. However the haemodynamic pathological significance of the reflux should be interpreted according its cause. So far, various methods have been proposed to define pathological:

• Refluxing time: As a little retrograde flow is necessary to allow veins valves to close269, it was

considered that “The cut-off value for reflux in the superficial and deep calf veins is greater than

500 ms. However, the reflux cut-off value for the femoropopliteal veins should be greater than

1000 ms. Outward flow in the perforating veins should be considered abnormal at greater than

350 ms. Reflux testing should be performed with the patient standing”250. However, more

recently it has been considered that in order to improve reproducibility in the diagnosis of reflux,

a uniform cut-off value of 500ms should be used. This was used in the studies later described to

ensure subjects did not have significant reflux.

• Reflux ratio: Psatakis proposed a ratio, known as the Psatakis Index (PI)270, calculated by dividing

the amount of diastolic reflux by the amount of systolic antegrade flow. RT doesn’t take into

account the amount of reflux and PI doesn’t take in account the reflux time. The Reflux

Dynamic Index (RDI) which uses the calculation:

RDI = (mean reflux velocity² x reflux time)

(mean systolic velocity² x systolic time)

takes into account both reflux time and amount. The reflux can be “total” when all of the valves above the measured reflux are absent or incompetent, “partial” when valves are not completely incompetent and

“segmental” when the valves overlying the incompetent one store part of the systolic outflow.

131

Respiration

Normally, spontaneous respiration activates the respiratory pump that affects intrathoracic pressure, which is transmitted to the blood in the vena cava, meaning that venous blood flow varies according to breathing depth and strength. Its effect is maximal proximally and decreases distally. The flow variation induced by breathing changes with the posture. In the upright position, this flow is antegrade and persists

only during the inspiratory phase, whilst in the supine position; it persists for the complete respiratory cycle, with a peak at the end of the expiration, as shown diagrammatically in Figure 43.

Figure 43: Shows the femoral vein flow variation according the respiratory pump and posture. Solid line:

Supine position, dotted line: upright standing position. (This image is adapted from an unpublished image provided by Claude Franceschi)

During inspiration, the intrathoracic pressure decreases, and due to diaphragmatic contraction,

intrabdominal pressure increases. This increases the flow from the central veins in the abdomen to the

thorax. The reverse happens during expiration. This is transmitted to the veins in the limbs in addition.

Figure 44 below shows the fluctuation in the velocity of blood moving through the femoral veins

associated with respiration in the supine position in a healthy subject.

132

Figure 44: Shows the velocity in the femoral vein in the supine position during respiration

In the presence of a venous obstruction, in the distal patent vein, the amplitude of the respiratory variation in the flow decreases and the minimum velocity increases proportionally to the resistance caused by the obstruction. This is particularly evident at the groin in the case of iliac vein or caval obstruction.

The valsalva manoeuvre is a forced expiration against a closed glottis, or blocked expiration. This manoeuvre compresses the right atrium and the vena cava so that it reverses the antegrade venous pressure gradient. This effect is proportional to the strength and the speed of the manoeuvre, which is difficult to standardise and measure unless the patients blow into a gauge. In daily practice, the patient can blow into a blocked straw or for improved reproducibility; the valsalva could be quickly performed after a deep inspiration resulting in a fast and strong retrograde gradient. The blocked expiration (BE) is followed by a release (BR). In normal individuals, the blocked expiration reverses the pressure gradient, dilates the veins, closes the valve leaflets and can produce a short and low flow. The release restores the antegrade pressure gradient and triggers a consecutive antegrade flow (Figure 45). In the presence of valve incompetence, blocked expiration induces a longer and faster pressure gradient than below the competent valves (Figure 46).

133

Figure 45: Valsalva manoeuvre (normal individual): The blocked expiration (BE) reverses strongly the pressure

gradient that induces only a negligible short and small reflux thanks to the correct competence of the underlying

valve. (This image is adapted from an unpublished image provided by Claude Franceschi)

Figure 46: Valsalva manoeuvre: pathological reflux (valsalva positive). Valsalva is Positive when valves are

incompetent. Reverse Flow appears when blowing (BE) and at release (BR). (This image is adapted from an

unpublished image provided by Claude Franceschi)

The rate of flow during BE also depends on the capacitance of the underlying venous bed. Therefore, a

VM will be more demonstrative when the distal venous volume is first reduced. However, if competent valves are interposed between the RP and the incompetent veins, a VM will not demonstrate reflux

134

during BE, as is the case for incompetent superficial veins in the presence of competent perforators, SFJ

and SPJ.

Venous Diameters

In general vein size is related to transmural pressure, which is derived from the inner hydrostatic pressure

and outer interstitial or tissue pressure. Measurements are taken in the standing position and with the

probe transversely to the long axis of the vein271. GSV diameter measurements may correlate with clinical severity272, with terminal valve competence or incompetence273, as well as with ilio-femoral valve

competence, incompetence or absence. Vein wall compliance, defined as the ability of vein wall to change

diameter between the lying and standing positions, is another factor that plays a role in changes in vein

diameter. In addition their changes after treatment are a tool to observe vein disease evolution.

For the deep veins, the diameter of the common femoral vein is usually measured immediately below the

confluence with GSV. The femoral vein may be measured at the mid thigh or just distal to the confluence

with the deep femoral vein. The popliteal vein measurement is conditioned by the anatomical variability

of the popliteal fossa and no consensus is available yet.

For the superficial veins, there is still no consensus about the optimal place to measure the great and

small saphenous vein diameters. The UIP Consensus 2006274 proposes the levels of the groin and mid-

thigh for GSV and at the proximal and mid calf levels for the SSV. Most published data in interventional

studies use measurements at the groin (stripping, endovenous procedures or foam) or at the proximal

thigh275. Furthermore when comparing measurements of the GSV at the groin and proximal thigh level,

there is a higher correlation between diameter and presence of reflux; similarly proximal thigh GSV

diameter measurements correlate with the clinical condition of the limb272.

For the perforator veins, the diameters of perforators are usually taken at the fascial level271.

135

6.2 IMAGING IN VENOUS DISEASE

6.2.1 Introduction

Detailed evaluation of the venous system providing information on the location, presence and extent of

valvular incompetence or obstruction is important to choose appropriate percutaneous procedures and

assess haemodynamic improvement. Conventionally, in the past, ambulatory venous pressure measurement and descending venography have been the methods of choice complementing the localisation and haemodynamic assessment. These treatments are largely superseded by less invasive techniques. Continuous wave Doppler (CWD) examinations have little accuracy on identifying specific incompetent or obstructed vein segments and differentiating duplicated or collateral veins however can provide segmental physiology. Duplex ultrasound on the other hand is the most useful modality for measuring valvular incompetence and venous obstruction276. Other haemodynamic investigations include air Plethysmography (APG) and photo-plethysmography (PPG) these tests are used to compliment duplex ultrasound to obtain more detailed measurements of haemodynamic severity. These along with computed tomography and magnetic resonance imaging of the venous system are discussed in more detail here.

6.2.2 Ultrasound

Ultrasound, from the continuous wave (CW) doppler that first allowed a live, non-invasive and

repeatable blood flow velocity and direction measurement almost 40 years ago, to today’s pulsed wave

Doppler and CDU that simultaneously shows the flow (colour and B-Flow) and the vessels (B-Mode) have advanced our understanding of venous anatomy and physiology. It is a tool which has improved both our understanding of the haemodynamics and diagnosis of the venous disease, which is yet to be surpassed, and is increasingly considered to be the gold standard in the evaluation of the morphology, haemodynamics and surrounding tissues limb veins.

136

Continuous Wave Doppler

Continuous Wave Doppler (CWD) is commonly used to complete a clinical examination. Deep veins are

ideally studied with lower frequencies (5MHz) and superficial veins have low velocity blood flow thus are

better heard with higher frequencies (10MHz). At each specific location the presence of a spontaneous

signal is documented. Small veins with low flow may require manoeuvres to hear the signal by

compressing the leg above or below the Doppler probe.

Flow reversal in a segment demonstrates valvular incompetence. Presence of a signal during the Valsalva

manoeuvre diagnoses the presence of common femoral vein incompetence. More distally compression

techniques are commonly used. The patient in an upright position gives a better indication of the

physiological conditions that are normally present and is ideal for assessing venous reflux. The presence

of perforator vein incompetence can be demonstrated by the presence of bidirectional flow in the medial

calf when the calf is compressed above the Doppler probe277.

Duplex

Duplex ultrasonography is a safe and reliable diagnostic imaging which assesses the venous system

effectively. In the last ten years many alternatives to duplex ultrasound have been investigated with

potential to replace this established modality. Duplex ultrasound with compression is commonly regarded

as highly sensitive and specific in the determination of acute DVT. In chronic DVTs, with patients who

are obese and oedematous it is less effective. It is also considered more challenging to visualise the iliac

veins and inferior vena cava278.

Duplex ultrasonography has many advantages over continuous wave doppler. B-mode imaging visualises the vein directly and accurate image guided placement of the probe. This results in better identification of non axial veins, paired veins and distinguishes collateral veins from major veins more easily274, 279.

Coleridge-Smith et al, 2006271 outlined in a consensus report that patient with chronic venous disease should routinely have the following evaluated with duplex ultrasound:

137

• The saphenous junctions to assess which are incompetent, their locations and diameters.

• The extent of reflux in the saphenous veins of the lower limbs and their diameters.

• The number, location, diameter and function of incompetent perforating veins.

• Other relevant veins that show reflux.

• The source of filling of all superficial varices if not from the veins already described.

• Veins that are hypoplastic, atretic, absent or have been removed.

• The state of the deep venous system including competence of valves and evidence of previous

venous thrombosis.

This information is then used to determine the type of treatment offered. Those with sapheno-popliteal

junction or sapheno-femoral junction incompetence can be offered ultrasound guided sclerotherapy,

endothermal ablation including laser or radio frequency ablation, tumescent free ablation techniques, or

surgery. However, others with isolated incompetence of saphenous tributaries maybe offered

sclerotherapy or phlebectomy. Inadequate identification of all areas of venous filling can lead to early

recurrence271, 280.

Colour Duplex Ultrasound

Colour spectrum analsysis enables visualisation of presence, direction and patterns of blood flow. Colour

coded flow maps can be added to simplify this281. Colour duplex also distinguishes between veins and arteries and flow in multiple simultaneous vessels. It shows the flow over long vessels so a larger area can be seen more easily compared with continuous wave doppler. Colour immediately shows disturbance of flow by a thrombus and the presence of reflux by inverting the colour pattern. Conventionally, the direction is set so that blue indicates antegrade venous blood flow back to the heart and red indicates

arterial blood away from the heart281.

138

In cases of venous obstruction the whole venous system from the iliac to the ankle can be assessed by

duplex scanning. On identification of a vein, compressibility assesses its patency. A complete thrombosis

of the lumen will be non compressible. Chronic obstruction will be evident by the presence of marrowed

lumina with thick walled veins that can be echogenic with fibrosis. Calcification can be present and will

also contribute to decreased compressibility238.

The diagnosis of venous valvular incompetence is undertaken with the patient standing, similar to the

continuous wave doppler examination. Manoeuvres described in the previous section to induce

antegrade venous flow are performed. The veins are imaged using B-mode imaging and character of the

Doppler signal is audible. Subsequently colour flow velocity can be used to visualise the long axis view

and a colour flow map or inverted spectrum can demonstrate retrograde flow thus venous incompetence.

Many clinicians use compression above or below the probe to identify suspected distal reflux disease.

However others prefer the distal cuff deflation method, when the velocity spectrum is monitored during

inflation and for 4 seconds after deflation as they consider this to be more of a reliable method to identify

reflux disease282. The variability of venous reflux patterns complicate the management of venous disease,

they are best assessed by complete whole leg venous duplex mapping283.

Much literature is available on the quantification of reflux and its relationship with clinical symptoms in chronic venous insufficiency. This has included investigating factors including: the number of incompetent valves, mean flow rates, duration of reflux. Although mean reflux times correlate well with clinical disease severity they provide unreliable evidence or prediction of ulcer healing or clinical improvement post treatment284.

Generally incompetence of the deep veins is associated with worse clinical symptoms and signs of venous insufficiency285. In contrast incompetent distal superficial veins are associated with lower risk of ulcer

development. Two thirds of patients with C4-6 disease have incompetent perforator veins. Concomitant obstruction with reflux disease can lead to patient’s complaining of both venous claudication and superficial ulcers286. Adopting a standardised protocol of clinical and duplex assessment can ensure that

the underlying aetiology of the leg ulcer can be established in 97% of cases287.

139

In each case the clinical question that needs to be answered should be clear. In situations when non-

operative management is considered duplex may be employed to exclude co-existing arterial disease prior

to the use of compression stockings. In contrast, if surgical management is being considered a more

thorough knowledge of the anatomical location and extent of the underlying venous disease may be

required to plan the optimum surgical approach. Valvular incompetence, wall abnormalities and

perforator incompetence should be examined. Multiple sites of reflux can exist and if the patient has

symptoms of venous claudication, further investigations should be carried out to exclude venous

obstruction18. In general DUS is a more accessible modality than other forms of imaging278. Venous

reflux is measured by distal manual compression and rapid release. A reflux time of more than 0.5 s is

generally accepted as the definition of venous reflux38.

CDU and Haemodynamic Changes

The velocity and direction of the flow are valuable hemodynamic data because they are related to the

variations in the pressure gradient. The changes in calibre of the vein correlate with the variations of the transmural pressure and the wall compliance. According to the location of the obstruction and incompetent valves, specific patterns of each pathological haemodynamic configuration can be classified.

CDU gives an accurate and detailed evaluation of vein dimensions and haemodynamics in deep and superficial veins, but is limited by both current ultrasound device technology, and the venous haemodynamics model that defines the imaging protocol. It is obvious that improving the quality of the

CDU protocol will improve the treatment strategy259, 264, 288. Unlike arteries, veins do not have a consistent

regulated pump that drives blood flow. In the static standing position very little blood flow may be

registered at DUS, and a totally immobile standing subject would have almost zero flow in the lower limb

veins. Manoeuvres which affect blood flow include respiration and muscle pump activation, particularly

the calf muscle pump. Different manoeuvres, to deliver reproducibility, have been developed to allow

investigation of blood flow, as has been discussed previously.

CDU and haemodynamic changes

140

Normal subjects

Altered venous haemodynamics can be found in asymptomatic individuals and can be normal in

symptomatic patients. Obese individuals, prolonged standing, high levels of physical activity and patients

with cardiac insufficiency can have a normal CDU examination, but have a functional impairment of their

venous drainage.

Venous obstruction

Obstruction increases the resistance to flow. It increases the residual pressure and consequently the

transmural pressure. The resistance depends on its extent and location and decreases proportionally to the

number and calibre of the collateral pathways. B-mode imaging can demonstrate the obstruction and its

cause, the dilated underlying veins and the overloaded collateral veins while colour doppler imaging can

visualize the stenotic lumen.

Pulsed wave and continuous wave doppler measure the direction and velocity of the flow upstream, in the collaterals and in the stenosis. Distal to the obstruction, the effect of the respiratory pump is reduced in proportion to the resistance. The pressure measurement at the ankle is performed in the horizontal supine position in order to select the respiratory pump value, independently of the gravitational hydrostatic pressure, assumed to be 0mmHg in this position. The normal values are around 20 mmHg.

An upright posture is necessary to assess the flow activated by the muscle pump systole.

Valve incompetence

Valvular incompetence can be seen in B-mode imaging, when the leaflets are seen to reverse. The grade can be assessed by the reflux time. Colour Doppler shows the changes of the flow direction but not its grade.

Lower limb deep venous incompetence

141

Deep veins are assessed directly in the upright position from the ankle up the groin while the iliac

incompetence is assessed indirectly by observing reflux at the proximal common femoral vein during

valsalva manoeuvre. The incompetence grade is assessed by various tests and measured according to the

elected measurement method described above. A thorough assessment can allow delivery of an efficient

therapeutic strategy: the best location for valve repair or transplant, prosthetic valve or deep venous shunt

disconnection.

Lower limb superficial venous incompetence

An incompetent valve will not cause any reflux if there is no refluxing tributary or any entering perforator

interposed with an underlying competent valve. On the other hand, reflux can occur between two

competent valves if an interposed perforator or a tributary is incompetent. Two different types of

superficial venous incompetence are discussed here: with and without competent connections, including

the SFJ, SPJ and perforators.

Superficial venous incompetence with competent connections with deep veins

The connections between superficial and deep veins are competent when the incompetent superficial

veins reflux during muscle pump diastole down to the deep veins through perforators, but do not reflux

during the valsalva manoeuvre. This may occur when the SFJ valve is competent, meaning the superficial veins are not overloaded by the deep veins. When saphenous trunk reflux is not induced by valsalva manoeuvre, the re-entry perforators are often located on the same saphenous trunk or distant from the trunk were it connects through a tributary289. Sometimes, reflux at the SFJ terminal valve can be

triggered by the valsalva and not present during muscle pump diastole. This is because the pressure gradient produced by the valsalva is higher than that of the muscle pump test. This phenomenon has been named “terminal valve dissociation”.

142

Superficial venous incompetence with incompetent connections with deep veins

Superficial venous incompetence with incompetence of one or more of the SFJ, SPJ and perforators

demonstrate reflux during diastole with the muscle pump dynamic tests and the valsalva manoeuvre.

Perforator vein incompetence

Perforators have often been considered the cause of varices, ulcers and recurrence. However, the

assumption that all perforators need ablating in these circumstances is haemodynamically contradicted in

both studies using invasive methods of assessment290, 291 and more recent CDU data264, 265, 292. Two

combined conditions must exist to produce perforator reflux, both valve incompetence and pressure

gradient reversal. CDU can show clearly inflow and outflow through the perforators during valsalva, dynamic muscle pump and squeezing tests, and allows the understanding of their significance according to the normal and abnormal functioning of the deep and superficial veins.

Combination of deep and superficial venous incompetence in the same limb

Both superficial and deep venous incompetence can be responsible for increased transmural pressure and its clinical consequences and their relative contributions can be determined by CDU. The specific haemodynamic effects of each condition have been described above, however, their combination can change the haemodynamics. Deep venous competence is a necessary condition for a proper functioning muscle pump, and can activate superficial venous reflux, however if the pump is not efficient, superficial reflux may not occur. The most demonstrative case of “deep venous competitive reflux” is the absence of

Doppler observable diastolic reflux in a large incompetent GSV, in which the valves have been destroyed, when the deep venous incompetence is severe. More than a century ago, Perthes asked his patients to walk with a tourniquet at the proximal end of the varicose GSV, stating that the deep veins were competent when the GSV collapsed and incompetent when it did not. Therefore, in such cases, GSV reflux that becomes apparent or increased after a deep valve incompetence repair, attests for the success of the operation. It only remains to treat the GSV reflux.

143

Combination of valve incompetence and obstruction in the same limb

The combination of obstruction and valve incompetence in the same limb is most commonly found in the postphlebitic disease. The findings on CDU are a combination of those described above for both conditions. The clinical challenge is to determine the proportion of increased transmural pressure they produce respectively, in order to predict the positive effect their respective treatment. Therefore, an additional CDU challenge is to recognize their common and independent contribution in order to treat the patient effectively.

Common Challenges in Duplex US Imaging

Many studies have shown the sensitivity and specificity of venous ultrasound to be between 90-100% however all three set of calf veins are seen in only 60-90% of studies293. Other challenges in duplex imaging are correct identification of the vein, for example, in the lower limb the great saphenous can be mistaken for the femoral vein. However this is reduced with identification of anatomical landmarks and experience. Image quality can also be a problematic with obese subjects and those with soft tissue oedema, however transducer selection, identifying imaging planes and optimal instrumentation setting are needed to optimise this. Areas of poor visualisation are often eluded to in ultrasound reports (N

Labropoulos, Kang, et al., 1999).

Venous compression is difficult in certain anatomical positions: iliac veins, adductor area of the femoral vein and the calf veins, and this can lead to false positive results when assessing for compressibility. This is generally overcome by the use of colour flow Doppler. In addition unusual anatomy, such as venous duplications may cause a DVT to be overlooked within the duplicated segment. This should be considered particularly when the vein is of smaller calibre or is in an atypical position.

Colour flow Doppler is a useful technique but it too has diagnostic problems. If the colour Doppler settings are too high colour artefacts may arise leading to masking of small to moderately sized thrombi.

The opposite may occur, that is false positives will arise, if the Doppler angle is incorrect, the wrong

144

velocity range is set or if examination is in a cool room. A higher incidence of false negatives has also

been reported in patients with a lack of venous distension secondary to hypovolaemia or systemic illness.

Characterising acute from chronic thrombus can be difficult with the use of many modalities but generally

heterogeneity and echogenicity of the thrombus increase with time since onset. Chronic occlusions are

also more likely to have reduced vein diameters, venous collateral development and flow channels within

the lumen compared with acute thromboses which have a more dilated appearance. The presence of a

free floating edge also suggests a new thrombus.

6.2.3 Laser Doppler

Laser Doppler flowmetry is a technique which can be used to assess the haemodynamics of blood flow.

A laser is directed at an area, and the beam is scattered by the presence of red blood cells, and returned to

a detector. It is most useful in superficial areas, such as the skin and superficial muscle, and is

predominantly used to assess a change in blood flow in these areas in response to an intervention. A

study comparing laser Doppler fluximetry (LDF) and laser Doppler imaging (LDI) demonstrated that

both can be used to assess the microcirculation by different methods, where LDF electrodes measure a small area with electrodes applied directly to the skin and LDI measures a large skin surface, but does not directly touch the skin294. LDI provides an image of a perfusion end point whereas LDF provides a constant measure of blood flow. Previous studies have struggled with reproducibility of the two techniques295.

6.2.4 Plethysmography

Plethysmography provides a non-invasive global evaluation of lower limb vein physiology, including

venous capacity and outflow, providing an assessment of both reflux and obstruction and their impact on

venous function and a measure of calf muscle pump function296, 297. Veins are compliant vessels and in

response to changes in circulating volume and central haemodynamics can pool volumes of blood. In

145

normal conditions, there is a large reserve capacity, but in the presence of venous obstruction this reserve

is reduced and resistance to flow increases. Plethysmography assesses the degree of venous obstruction by

measuring capacitance and resistance within the venous system. In the case of chronic obstruction,

collaterals may have formed that can increase the resistance which can affect the results298. There have

been issues with reproducibility, due to inter-observer differences, but results are also affected by changes

in environmental temperature and pressure, the presence of cool peripheries, and more proximal

obstructions.

Strain gauge plethysmography

Strain gauge plethysmography (SGP) measures changes in the calf in response to different conditions.

The plethysmograph consists of a compliant tube containing a liquid, which is placed around the widest

part of the calf, which will detect a change in volume and circumference of the calf, which translates to a

change in electrical resistance of the gauge, picked up by a detector. It makes the assumption that the leg

is a cylindrical shape with a uniform distribution of volume change. A tourniquet placed proximal to the

gauge can be inflated above venous pressure, but lower than arterial pressure, and deflated to halt and

restart blood flow. The rate the limb swells is a measure of arterial flow. Release of the venous

tourniquet will cause the limb swelling to decrease, and based on the principle that arterial blood supply

and transcapillary fluid exchange are constant, the change in extremity volume is due to emptying and

filling of veins.

Air plethysmography

Air plethysmography (APG) measures changes in cuff pressure standardised to reflect volume changes.

In one study, for patients with a venous filling index of more than 7, it showed a 73% sensitivity and

100% positive predictive value of identifying pathological venous reflux299 and it has been shown to have some correlation with CEAP classification300. Both SGP and APG give quantitatively different

information, however, it has been shown to be qualitatively similar301. The diagnosis of chronic venous

146

insufficiency by venous duplex correlates well with the diagnosis by air plethysmography302. APG has been shown to be useful in the diagnosis of complex congenital anomalies cases, such as Klippel-

Trenaunay Syndrome with advanced venous disease303. Inconsistencies of results can be due to many factors; systemic changes in central venous pressure, changes in arterial supply to the leg, variability in

performing exercises, changes in posture and weight distribution, as well as temperature changes which

alter vein lumen size and venous capacitance. For these reasons care must be taken when analysing the

results18.

Photoplethysmography

An alternative form of this method is photoplethysmography (PPG). This measures the changes in tissue density by quantifying the amount of light reflected from the extremity. Although skin is opaque, in the infra-red spectrum it transmits a small amount of light, which can be detected by a light sensitive diode.

Red blood cells absorb the most light when pressure in the veins is higher and they are full. The probe if fixed to the leg either close to the medial malleolus or dorsal foot. In some studies it has been shown to be relatively unreliable technique as there is low specificity and difficulties in calibration with poor light penetration through the skin304 but he VRT measured by PPG correlates well with direct pressure

measurements305. When PPG is used with a tourniquet, it can distinguish limbs with and without venous pathology, and limbs with superficial and deep venous incompetence, however, it is not considered to be a good indicator severity of deep venous disease306 .

6.2.5 Venography

Ascending venography

Contrast venography has been considered historically the gold standard for defining venous anatomy, and

making the diagnosis of DVT venous incompetence, since it was first described by Berberich and Hirsch

in 1923. It is usually carried out and observed in real time in a fluoroscopy suite. Contrast injected into

147

the dorsum of the foot allows good visualisation of the deep veins of the calf and thigh, improved by the

use of a tourniquet around the thigh to slow the movement of the contrast. Preferential filling of the

deep veins can be aided through compression, a tilt table and the Valsalva manoeuvre. Iliac veins and the

IVC are best visualised by injecting contrast into the femoral vein238. It has now largely been replaced by the use of duplex ultrasound as venography is less reproducible, difficult to interpret in 9-14% of patients, associated with inter-observer disagreements in up to 10% of studies and is invasive, and it does not give quantitative information on the collateral circulation307-310. Nevertheless it is still used in situations when

the duplex is equivocal or if the duplex is negative but the clinical suspicion for DVT remains high,

particularly if atypical venous anatomy is suspected, with one study demonstrating that 12% of the

patients had a truncular venous malformation of the femoral vein311. However, for the investigation of new techniques for the diagnosis of DVT, ascending is still widely considered the gold standard for comparison312, 313 and the diagnosis of a DVT is demonstrated by the presence of a lack of opacification of a vessel also known as a filling defect312. More recently, however, with the increased use of

interventional procedures to treat DVTs, including chemical thrombolysis and pharmacomechanical

thrombectomy, the se of venography is again on the rise.

Descending venography

Descending venography is used to demonstrate reflux in the superficial, deep and pelvic vein, as well as

determine the location of venous valves. Contrast is injected through either the brachial vein or the

ipsilateral or contralateral femoral vein or, for assessing for distal disease, the ipsilateral popliteal vein.

The patient is usually tilted towards the upright position, and injection is performed during a forced VM

to determine the extent and site of venous reflux. This is graded accorded to Kistner’s classification, with

4 grades of abnormal reflux according to the amount of leakage and the distal extent of leakage in each of

the SFV, DFV and GSV314, as below:

• Normal – No reflux occurs through any valve during VM

148

• Grade 1 – A wisp of reflux through one or more valves during VM, which is not sufficient to

affect the direction of flow in the iliac veins

• Grade 2 – Considerable reflux in the thigh veins, which does not extend to the calf due to the

presence of a competent valve in the distal thigh, with some effect on flow direction in the iliac

veins

• Grade 3 – Considerable leakage, as grade 2, but with extension into the calf, with an effect on the

iliac vein flow

• Grade 4 – Cascading reflux in the thigh and calf and complete reversal of flow in the iliac veins

Although it may still have a role in situations such as venous reconstruction, it is an invasive test with

recognized limitations. There are associated complications including a risk of contrast allergy or

extravasation, thrombosis, and nephrotoxicty, therefore, its use has largely been replaced by duplex

ultrasound.

6.2.6 Intravascular Ultrasound (IVUS)

IVUS is an invasive imaging modality whereby a catheter with an ultrasound probe is inserted into the

vein to be assessed. The vein can be visualised from the inside allowing an accurate assessment of any

stenosis or obstructions present. There have been several studies into its use in the venous system,

particularly for IVC filter placement and the assessment of iliac vein stenosis or obstruction. In one

study, the insertion of IVC filters using transabdominal DUS, fluoroscopy and IVUS were compared,

IVUS was only used if DUS was inadequate. Of the 28 procedures performed with IVUS, 26 were

successful at the time of placement and 24 confirmed to be correctly located post-procedure315. Other studies have looked at the use of IVUS for assessing iliofemoral veins for stenosis, caused either by a partial obstruction, or extravascular compression. One study assessing the role of IVUS in endovascular

149

surgery for the treatment of iliac vein obstruction, suggested that IVUS should be used generously in both

the diagnosis and treatment as venography was unreliable, as the venograms significantly underestimated

the degree of iliac vein obstruction by 20% in patients with post-thrombotic syndrome and 28% in

patients with May-Thurner syndrome170.

The advantages of IVUS over other imaging modalities are that it can be used at the bedside, which

contrast venography cannot, and it allows improved visualisation and assessment of stenoses, which

cannot always be detected haemodynamically on DUS. It will also show intraluminal features, such as

webs and trabeculations, which other modalities will not. However, its disadvantage over other

modalities is that it can only detect collateral vessels close to the imaged vessel. The complications of

IVUS include thrombosis at the insertion site, pulmonary embolism, infection,

6.2.7 Computed Tomography Venography (CTV)

The advent of spiral-computed tomography (CT) has greatly extended the range of venous imaging.

Intravenous contrast administered though an arm vein is usually necessary for optimisation of evaluation.

Within a matter of second large areas of the body can be imaged rapidly. In addition different scans can show opacity of various veins depending on their filling time based on the organ, blood flow and proximity to central veins. However these factors (blood and contrast flow) themselves can cause artefacts. Furthermore if homogenous mixing of the contrast within the vessels does not occur filling defects may appear. This is more of a pitfall imaging the thoracic region when filling of central veins occurs unilaterally – therefore bilateral infusions in each ante-cubital vein is preferred18.

CT venography is shown to be as accurate as duplex ultrasound (DUS) for detecting DVTs below the

inguinal ligament and more accurate than DUS for detecting thrombus that extends above into the pelvic

veins and inferior vena cava316, 317.

Colour Doppler ultrasonography is an easy to use, widely available and robust screening tool for diagnosing DVT. It is particularly more accurate in patients who are symptomatic. The benefits of CT

150

venography over Duplex are that there is less operator variability, shorter investigation time, and better

imaging of the pelvic veins. CT venography does however require a large volume of contrast and subjects

the patient to high dose radiation making it inappropriate for use in patients with impaired renal function

and those who are pregnant.

Marston et al 2011318 retrospectively reviewed 78 patients with CEAP 5-6 to look at the incidence of iliocaval obstruction. They found that DUS is a useful screening tool however 23% of patients with high grade ICVO confirmed on US and MRI had a normal DUS. The study did acknowledge that CT and

MRI are limited in visualising intraluminal webs and chronic abnormalities that may contribute to venous obstruction.

Some studies show that magnetic resonance venography is more accurate than Duplex or conventional venography for identifying pelvic extension of DVT319-321. Both MRA and CT have been proven effective in helping clinician’s make decisions and accurately plan further interventions. Data from CT and MR can be further manipulated in a multi-planar and 3D fashion to give exacting detail. CT results are generally equivocal to MR, and both compare favorably with contrast invasive imaging. Ultimately the individual use of different imaging modalities depends on experience of operator, local availability and cost322.

6.2.8 Magnetic Resonance Venography (MRV)

Although duplex ultrasonography is a well-established standard of care and an excellent screening test for

the detection of venous disease its accuracy of displaying the venous system is limited, in particular when

imaging the pelvic veins323. Duplex has a high sensitivity and specificity in the popliteal and femoral veins however is operator dependent and its accuracy is reduced in the distal extremities due to low velocity and volume of blood324.

Magnetic resonance imaging has been employed to detect obstruction in both upper and lower extremities over the last decade325. It provides high-resolution images and decreases the need for ionising

151

radiation. It has been shown to be capable of detecting DVTs in the upper leg, pelvic and abdominal

regions278.

Other modalities, such as time-of-flight (TOF) and phase contrast MR venography have been extensively

investigated and considered reliable for the assessment of the deep venous system324. Despite this, these techniques are limited in the identification of paired veins and valves, are time consuming and can be

prone to artefacts326.

Flow-enhanced multi-shot TOF echo-planar imaging results in improved imaging of the calf veins by exploiting transient flow-augmenting with venous occlusion, but this method is prone to edge blurring due to T2 decay and poorly differentiates arterial-venous circulations327.

Small veins with slow flow can be difficult to image, however fast-spin-echo (FSE) techniques have a high sensitivity for these but have a low sensitivity for large veins with high flow rates328. In the presence of

oedema, a high signal occurring in flow dependent MR venography can lead to suboptimal visualisation

of venous anatomy329.

Contrast-enhanced MRV is a more recent investigation used for the assessment of venous anatomy330-332, which involves the injection of contrast material. On injection of the bolus the arterial system is enhanced then subsequently venous structures are displayed. In the lower limbs diluted contrast can be injected by the pedal veins and evaluated by single phase, which enhances the lower limbs and pelvic veins330.

Multi-phase MR can show a difference in arterial and venous systems, by using a subtraction algorithms; the early arterial phase can be subtracted to show a selective venogram331. Most dynamic imaging

produces time resolved imaging that can give poor temporal resolution. Higher temporal resolution is

better to illustrate contrast dynamics. High contrast may also assist with image processing, filtering and

correlation analysis. This can help generate higher quality venograms.

Venous contrast dynamics are effective at visualising venous structures particularly in areas where the soft

tissue enhancement can obscure the vasculature325. Time resolved data in this modality can help the

identification of areas of subclinical tissue injury with contrast dynamics333.

152

Du et al, 2006325 demonstrated the feasibility of performing three-dimensional DCE-MRV in the distal extremities. Time-resolved hybrid PR scanning resulted in images with high temporal and spatial resolution. This gave good delineation of venous valves and separation from background tissue enhancement. They also effectively were able distinguish veins and arteries by using contrast BAT thresholds. Region-specific arterial and venous CATs can be automatically generated with an iterative CE and CAT threshold algorithm.

Magnetic resonance venography (MRV) has the advantage over CT in avoiding the need for nephrotoxic contrast. It gives high-resolution images of adjacent structures and allows for a variety of views, both coronal and sagittal, of the area. The radio-frequency generated by the magnetic resonance allows for increased contrast between the venous system and its surrounding structures. MRV is particularly advantageous in determining thrombus in smaller branched vessels. It is useful in the diagnosis of pelvic vein thrombosis and Kippel-Trenaunay Syndrome334. It is also the preferred modality for the identifying

the venous drainage of other soft tissues.

Unique to DUS is the dynamic manipulation of the patient’s position to assess flow and valve patency in

the deep venous system. Arnoldussen found that twice as many collateral systems were identified with

DUS compared with MRV which was more difficult to ascertain their patency. However their anatomical

position was more easily detected with MRV278.

MRV is also advantageous for the reviewing of images by multiple radiologists and allows comparison between cases. It gives excellent information of the anatomical location and size of vessels this is particularly important in the abdominal and pelvic regions where no valves are present so there is less emphasis on hemodynamic information. Both DUS and MRV identified the majority of deep veins and analysis was performed using the LOVE score278.

Edelman et al, 2009323 describe a pilot study in a cohort of healthy subjects with lower limb venous

disease, to assess the clinical feasibility of a new method to image the deep and superficial veins of the

lower extremities. The signal targeting using alternative radiofrequency and flow-independent relaxation

enhancement (STARFIRE) is a flow-independent method for performing MR angiography that can selectively depict arteries or veins. The technique preferentially suppresses signal intensities of fat and

153

muscle whilst preserving that of blood. STARFIRE can be used in patients at risk of developing

nephrogenic systemic fibrosis, as it does not require the use of paramagnetic contrast agents. The

technique has potential in the evaluation of acute and chronic deep venous thrombosis and venous

mapping for the planning of varicose vein treatment. The vessel detail was competitive with MR

angiography however as with MRA is prone to high signal defects from fluids and oedema. Subcutaneous

and joint fluid can be edited out however the deep tissue fluid cannot. In the case of DVT, the thrombus

was low intensity and the oedema surrounding was high intensity. 3D MRA has been used to image

recurrent varicose veins, STARFIRE may provide potentially, equivalent diagnostic information without

the need for contrast administration. However this needs to be further evaluated with larger studies.

6.2.9 Image Guidance for venous interventions

Often various clinical scenarios require the need to ‘road map’ the venous system by CT or MRI imaging

prior to an intervention. As in some cases DUS alone may not be sufficient. This is particularly pertinent

in the pelvic area, for obese subjects or the presence of bowel gas or if atypical anatomy is suspected.

These modalities can further guide the approach to treatment as well as giving detailed information on the

location and extent of disease. This includes feasibility of the proposed intervention, puncture site, the

need for IVC filters, other devices or the requirement for further medical therapy335. In the case of placement of IVC filters, methods of image guidance include intravascular ultrasound336, trans-abdominal ultrasound337, venography338 and fluoroscopy. Aberrant venous anatomy that can affect filter placement

include duplicated IVC, accessory renal veins, intra-caval thrombus, caval tortuosity, caval compression

and mega or micro cava. CT and venography together are able to reliably identify these anomalies.

However if venography is not undertaken, careful review of CT scans must be undertaken. A large review

of abdominal CTs by showed that there was a 5.6% incidence of major venous anomaly339.

Ultrasound is a modality that is well established for the guidance of various interventions in venous disease, for example ultrasound guided sclerotherapy is a minimally invasive procedure that is well established for the treatment of perforator vein incompetence340.

154

6.2.10 Venous Malformations

As MRI provides the most comprehensive information of the anatomical distribution of venous malformation and it has become the ‘gold standard’ in these clinical situations. These problems can be complex and planning of interventions from radical surgery to micro embolisation is crucial for satisfactory outcomes. MRI is also superior at visualising high flow malformations, arteriovenous shunting and communications between the deep and superficial systems. CT in comparison can underestimate the extent of soft tissue and bony involvement in venous malformations238.

6.2.11 Other techniques

Capillary microscopy is effective in showing capillary convolution is associated with lipodermatosclerosis and ulceration in severe chronic venous disease341 (Howlader & Smith, 2003).

6.3 ULTRASOUND IMAGING OF DEEP VEIN VALVES

6.3.1 Background

As previously discussed, ultrasound duplex imaging offers information about deep venous flow patterns and reflux, it is safe, reliable and readily available in the clinic setting. Importantly for this study, velocity data can be obtained, and the movement of the valve leaflets visualised in real time, giving essential information for understanding the venous haemodynamics and modelling the deep venous system.

Contrast enhanced ultrasound has been well documented in its use in the vascular system, including the assessment of hepatic haemodynamics342 and atherosclerosis in carotid arterial disease343. This may provide additional information on the flow through in the deep vein and through the valve. Normal subjects were selected as the aim is to develop a model of a normal vein valve with a view to developing a deep vein valve replacement, although this will be implanted into patients with venous disease.

155

The aims of this part of the project are to use ultrasound and contrast enhanced ultrasound imaging343 to evaluate anatomical and haemodynamic variables in patients with normal deep veins and to compare this data with that obtained from magnetic resonance imaging20 of the same segment of vein, which is

discussed in Section 6.4. This ultrasound data will then be used for the development of the

computational and experimental flow models for deep venous function, which are described in Section

6.5 and Chapter 8 respectively.

6.3.2 Methods

Ethical approval, study registration and regulation

Ethical approval for this study to carry out ultrasound and magnetic resonance imaging of normal

subjects has been obtained from the West London Research Ethics Committee 2, REC Reference

Number 10/H0711/79 (Appendix 2).

The research has been registered with Imperial College Joint Research Office. Informed written and

verbal consent has been obtained from all patients and volunteers involved in this study. A prospective

observational study to evaluate patients with normal deep veins has been carried out.

Volunteer and patient recruitment

The target study population is 1 group, containing 7 patients. In view of little previous research in this area, it has not been possible to calculate a target sample size using power calculations. The chosen target provides a plausible, realistic sample size within the clinical setting in order to achieve the study aims.

Volunteers have been recruited from the Department of Vascular Surgery, Charing Cross Hospital.

Clinical assessment was performed to ensure that either no venous disease was clinically evident or the presence of telangiectasia only. A medical history and clinical evaluation is carried out. Subjects were eligible and approached for recruitment if they had no clinical evidence of venous disease and no deep

156 venous reflux or obstruction evident on ultrasound imaging. They are then given an information leaflet if they are willing to consider participating.

Inclusion Criteria

• C0 or C1 venous disease

• Ankle brachial pressure index >0.85

• Age >18 years

Exclusion Criteria

• Patients under the age of 18 years

• Those unable to consent

• Subjects with significant peripheral arterial disease, ankle brachial pressure index <0.85

• Previous allergy to sulphur or contrast media used for MRI (gadolinium)

• Subjects who are pregnant, planning pregnancy or breastfeeding

• Subjects who have had a myocardial infarction within the last week, have an artificial heart valve,

cardiac pacemaker or defibrillator or have had a recent coronary angioplasty.

• Subjects with cerebral aneurysm clips, metallic shards in their eyes, cochlear implants, other

electronic implants or nerve simulation units.

Image storage

All images taken are immediately anonymised and stored on a “WD My Book Essential” 2 terabyte external hard drive in a locked drawer in a locked office.

157

Ultrasound Imaging

Ultrasound scans are performed using the standard Venous presets on a Philips iU22 scanner. A 12-

5MHz 3D linear array probe is used. Contrast imaging is obtained with non-linear imaging contrast mode.

The patient is positioned standing with the weight on the left leg, the right hip externally rotated to 30° and the right knee flexed to 30°. The exact location of the first visible valve distal to the confluence of the right DFV and FV is marked on the surface of the patient’s skin while the patient is standing. The distance from the base of the valve leaflets to the junction of the DFV and FV is recorded. A device that can apply a standard calf squeeze, the Kendall SCD express system, which inflates to a pressure of

45mmHg, is applied with the bladder that is first to inflate around the widest section of the calf. The following images and videos are then acquired.

• B-mode / 30 second video / standing at rest

• M-mode / Standing / 2cm Proximal to valve / No calf squeeze / Normal respiration

• M-mode / Standing / within valve / No calf squeeze / Normal respiration

• M-mode / Standing / 2cm Distal to valve / No calf squeeze / Normal respiration

• M-mode / Standing / 2cm Proximal to valve / calf squeeze / Normal respiration

• M-mode / Standing / within valve / calf squeeze / Normal respiration

• M-mode / Standing / 2cm Distal to valve / calf squeeze / Normal respiration

• M-mode / Standing / 2cm Proximal to valve / calf squeeze / No respiration

• M-mode / Standing / within valve / calf squeeze / No respiration

• M-mode / Standing / 2cm Distal to valve / calf squeeze / No respiration

The patient is then positioned supine on the couch, with a leg-torso angle of 15°, right hip externally rotated to 30° and right knee flexed to 30°. With the patient in this position, the same images and videos were acquired as described above.

158

1ml of “SonoVue” microbubbles is then injected into the right antecubital fossa of the patient through a

20 Gauge cannula. The following images are then obtained.

• Video / semi recumbent, at rest

• Video / semi recumbent, calf squeeze

• Video / standing, at rest

• Video / standing, calf squeeze

B and M Mode images of the same segment are acquired for use subsequently to build the geometry of

the venous valves and to assess wall compliance and motion. “SonoVue” microbubbles are typically of

the order of 2 to 6 microns consisting of a lipid shells filled with a high molecular weight gas, sulphur hexafluoride. They are true intravascular agents and remain within the vascular system, and have been used in both non-targeted and targeted forms for clinical diagnostic, prognostic and therapeutic purposes in oncology and cardiovascular diseases. Their resonance frequency, at which compression and expansion occurs most readily, is within the range of frequencies used for clinical ultrasound, therefore detectable signals are returned from the microbubbles.

All images are recorded to allow subsequent validated assessment and extraction of geometric and flow parameters. The Digital Imaging and Communications in Medicine (DICOM) videos are converted to

AVI and JPEG format for review. Image analysis is carried out using ImageJ analysis software. The curve depicting the velocity of blood within the vein is traced manually by one observer on the software program, and measurements including the maximum velocity, mean velocity and duration of increased flow during calf compression can be calculated. This was then repeated by a second observer, and again by the first observer to establish inter and intra observer reliability. For clarity, the locations that were considered distal (a), within (b) and proximal to (c), valve leaflets are considered anatomically. This is demonstrated in Figure 47.

159

Figure 47: Shows the measurement of venous velocity with ultrasound distal (a), within (b) and proximal to (c), valve leaflets and the velocity curve traced by one observer on ImageJ (d).

6.3.3 Results

Seven normal subjects have been studied, of which 6 were male. The mean age was 31.85. Three subjects had C1 disease, with mild asymptomatic telangiectasia, 4 had C0 disease, that is no signs of venous disease. None of he subjects had any evidence of deep venous reflux or incompetence. In the case of subject 4, the 2nd valve distal to the confluence was used as the first was less than 1cm distal, making velocities difficult to measure as distal to the valve they were taken in the CFV. Examples of the ultrasound images of deep venous valves are shown in Figure 48 and Figure 49. In Figure 48, the difference between the flow waveforms and velocities within the vein in both the standing (A) and lying

(C) positions can be observed. Image B demonstrates the flow waveform and transient increase in velocity during a calf squeeze to 45mmHg which simulates walking.

160

161

Figure 48: Shows the ultrasound images of venous valves. A - Normal respiration, standing position, B - Standing position, calf squeeze, C - Semi recumbent position, normal respiration, D - Still of B mode video image showing valve leaflets.

Figure 49: Contrast enhanced ultrasound of venous valve

162

Figure 49 shows still images of the contrast enhanced microbubble ultrasound using SonoVue microbubbles. In image A, behind the valve leaflet is a flow void, and image B demonstrates that the microbubbles can adhere to the valve leaflet and highlight its shape.

The maximum velocities obtained in both the lying and standing positions for each subject are shown in

Table 15 below. The maximum and mean velocities and the duration of increased flow following calf squeeze are shown for each subject in Appendix 3 for observer 1 and Appendix 4 for observer 2. These show the maximum and mean velocities over the total duration of the recording as well as the mean velocity during the calf squeeze only, if one was applied.

Table 15: Shows the position of the valve measured and summary of the maximum velocities obtained in both recumbent and standing positions for each subject

163

Subject Distance of valve Maximum velocity Maximum velocity Mean duration of

from DFV/FV standing position recumbent position calf squeeze (s)

junction (cm) (cm/s) (cm/s)

1 4.8 59.00 91.05 2.08

2 5.0 40.35 71.93 2.42

3 15.0 41.68 76.50 2.6

4 6.0 54.9 105.88 1.63

5 12.0 41.11 93.23 2.88

6 2.9 29.37 51.43 2.52

7 2.7 52.94 127.06 2.14

The intra and inter observer reliability data is demonstrated by Bland-Altman analysis.

Figure 50: Shows the Bland-Altman analysis of the inter-observer reliability between observer 1 and

observer 2 for mean velocity measurements demonstrating a bias of -2.0 (SD 7.6) and 95% limits of agreement from -17 to 13.

Figure 51: Shows the Bland-Altman analysis of the inter-observer reliability between observer 1 and observer 2 for measurements of the duration of calf squeeze demonstrating a bias of -0.099 (SD 0.63) and

95% limits of agreement from -1.3 to 1.1.

164

Figure 52: Shows the Bland-Altman analysis of the inter-observer reliability between observer 1 and observer 2 for maximum velocity measurements demonstrating a bias of -2.9 (SD 16) and 95% limits of agreement from -34 to 28. Figure 53: Shows the Bland-Altman analysis of the intra-observer reliability for observer 1 for maximum velocity measurements demonstrating a bias of -0.61 (SD 1.1) and 95% limits of agreement from -2.8 to 1.5.

Figure 50: Shows the Bland-Altman analysis of the inter-observer reliability between observer 1 and observer 2 for mean velocity measurements

Figure 51: Shows the Bland-Altman analysis of the inter-observer reliability between observer 1 and observer 2 for measurements of the duration of calf squeeze

165

Figure 52: Shows the Bland-Altman analysis of the inter-observer reliability between observer 1 and observer 2 for maximum velocity measurements

Figure 53: Shows the Bland-Altman analysis of the intra-observer reliability for observer 1 for maximum velocity measurements

166

The maximum velocities obtained in the lying position ranged from 51.43-127.06 cm/s and in the standing positions from 29.37-59.00 cm/s. The average of all the maximum velocities 2cm distal to the valve, within the valve and 2cm proximal to the valve are shown in Figure 54. Using a one-way Anova test to compare the three values there was no significant difference (p=0.401), and an unpaired student t- test demonstrated no significant difference between each of the three columns. The average of all the mean velocities 2cm distal to the valve, within the valve and 2cm proximal to the valve are shown in

Figure 55. Using a one-way Anova test to compare the three values there was no significant difference

(p=0.217), and an unpaired student t-test demonstrated no significant difference between each of the three columns, however the difference between the mean velocity measurements 2cm distal to the valve compared with the mean velocity measurements within the valve and 2cm proximal to the valve to the valve approached significance with p=0.131 and p=0.119 respectively.

Figure 54: Shows the average of the maximum velocity for the 7 subjects relative to the valve leaflets using

ultrasound

167

Figure 55: Shows the average of the mean velocities for each of the 7 subjects relative to the valve leaflets using ultrasound

168

6.3.4 Discussion

This study investigated the use of ultrasound imaging to provide the flow profile around the femoral vein

valve distal to the confluence of FV and DFV in normal subjects. The mean duration of calf squeeze was

2.32s. Although the velocities were very variable between subjects, the average maximum velocity and

the average mean velocity increases as blood flows through the valve leaflets. The information obtained

can be compared to other imaging modalities to assess its reliability. Given the good inter and intra

observer reliability obtained, this information can be used to develop computational models and

laboratory flow models

6.4 MAGNETIC RESONANCE IMAGING OF DEEP VEIN

VALVES

6.4.1 Introduction

Magnetic Resonance (MR) Venography is increasingly used to obtain information on the deep veins that

are less accessible to ultrasound, for example, the pelvic veins. It is able to give accurate anatomical data

and information on obstruction of the deep venous system278, 325 and confers an advantage over

Computed Tomography Venography as it does not involve the use of ionising radiation. The use of dynamic MRI has expanded in many clinical areas recently as real-time imaging can be obtained of vessels. Contrast is not required as the blood acts as a contrast as it is moving. MR Velocity Mapping has been previously used predominantly in arterial system, when it is gated with the cardiac cycle to build up

169

the acquired images. It is not however, often used in the venous system. More recently it has been used

to measure flow in the deep veins of the calf in volunteers with and without compression hosiery, or with

intermittent pneumatic compression344, 345. It confers advantage as it is able to give simultaneous anatomical and velocity mapping. One of the major disadvantages, however, is that it is not possible to image valves directly using MR due to spatial resolution restrictions and incredibly thin valve leaflets, however high resolution ultrasound can achieve this as shown in Figure 48.

This part of the study aimed to image the same subjects who had been assessed with ultrasound and obtain dynamic MR images across the same segment of vein to obtain accurate velocity data.

6.4.2 Methods

IPC Device

AS for the ultrasound study, the same IPC device, Kendall SCD EXPRESS Compression System, was

used to drive an impulse of venous blood through the deep veins of the leg. The device delivers

45mmHg through a cuff consisting of an airbladder with non-elastic Velcro strapping transferring the pressure over the whole circumference of the leg. The device inflates separate air bladders sequentially; however the cuff was attached so that only the first bladder inflation would have a compression effect on the leg, and this was applied to the widest point of the calf.

Subject setup

The same 7 normal subjects that were scanned using ultrasound to locate the first venous valve in the

common femoral vein in the right leg and to confirm the competency of the valve were imaged. The

distance from the confluence of the femoral vein with the deep femoral vein to the base of the first valve

was measured and used as a marker for centring the MR imaging volume, this was performed in the same

subject positioning as used for MR scanning to reduce any anatomical movement and allow approximate

location with a pen mark drawn on the subjects skin.

170

The subjects were set up for the MR scan as follows; Using a 1.5T Avanto Siemens scanner the subject

lay supine on the MR table with a foam support (~ 10 cm high) beneath the ankles, this allowed the calf

to remain above the table reducing any unwanted compression of the leg and also allowed the application

of the compression cuff around the muscle mass of the calf without causing excessive motion of the leg

during inflation. A further foam support (generally used for knee and lower leg support) was used with

pillows to raise the subject into a low semi-recumbent position, this increased hydrostatic pressure acting on the veins in the leg increasing vein size and improving venous filling between compressions. A small surface coil, Carotid Coil, Machnet, was then taped over the pen mark of the valve location, near the proximal inner thigh. This set up is shown in Figure 56.

Figure 56: Shows a volunteer ready for MRI, showing clockwise from top left; expiratory breath hold, Kendal SCD

express system calf pump, heel raised slightly from the bed to prevent calf compression and surface coil.

Then, bSSFP multi-slice localisers allowed visualisation of the bifurcation and, using the distance

measured from the ultrasound scan, the estimated location of the valve was moved to the scanner iso-

171 centre. A high resolution 3D bSSFP scan allowed further confirmation of the imaging region. This is shown in Figure 57.

The 3D images also allowed a check of the trajectory of the vein through the imaging slice, fortunately the vein in this location tends to run axially through the leg, major offsets were corrected by repositioning the subject. The subject was asked to perform an end expiratory breath-hold, to reduce any respiratory induced venous flow variations between scans, with the IPC device activated during the breath-hold instructions as it had a delay of approximately 15s from turning on to first inflation. The RV spiral sequence was then run continuously over the compression period. A preparation scan allowed a check of

172 the RV spiral setup and image reconstruction, but also flushed any excess blood in the venous reservoir and therefore reduced variance between the initial scans. This was repeated for 10 slices through the isocentre with Z offsets within the range +25 mm to -25 mm with 5 mm slice separation with approximately 45 s between scans to further regularise the venous blood reservoir filling. After a short break the same 10 slice were repeated again. The parameters for the sequence are shown in Table 16 below.

Table 16: Shows the RV spiral scan parameters

Parameter Value

FOV 150

Resolution (acquired/reconstructed) 1 x 1 -> 0.5 x 0.5 mm

Slice Thickness 4 mm

VENC 85 cm/s

Water Excitation / flip angle (1,1) / 30 °

Interleaves 4

No. Ref scans 3

RV Spiral Scanning Parameters

The RV spiral sequence at 1 x 1 mm spatial resolution was used to achieve maximum temporal resolution.

The fixed design of the FOV and the imaging location presented a problem of wrap artefact in the

173 images; the small carotid coil reduces this artefact with its sensitivity. The slice thickness was reduced to 4 mm to reduce partial volume effects. A large VENC range was chosen such that all velocity maps could be resolved with no phase wrap around due to the number of slices being acquired and the lack of an off- line reconstruction, pilot scans agreed with preliminary Doppler ultrasound measurements showing a peak venous velocity ~50 – 70 cm/s over the flow impulse.

Images were reconstructed using modelling of the background muscle signal for the phase subtraction because of vessel size and shape changes over the flow impulse from compression; also because of potential variance between scans a separate reference scan subtraction could not be used. ROIs for modelling of the muscle signal were drawn, independently for each of the 2 elements in the carotid coils, on a magnitude image averaged from 3 Ref scans at the beginning of each RV spiral acquisition, avoiding any residual fat signal and leaving room around vessels for shape changes. The same ROI was used throughout each acquisition applied to each velocity image for an individual background phase map subtraction. This sequence was developed by Dr Pierce and Dr Gatehouse at the Royal Brompton

Hospital, optimised there by with Dr Pierce for imaging the FV / DFV confluence and all scanning was carried out at Charing Cross Hospital, with Dr Pierce and with the help of Anastasia Papadaki and Lesley

Honeyfield.

Image analysis

Images were loaded into CMR Tools for inspection and analysis. Using SPIN (signal processing in NMR) software, obtained from Professor Mark Haake, The MRI Institute, Detroit, USA, the images were analysed. Briefly, both the phase and magnitude images are opened within the software. Using flow mode, the vessel to be assessed was defined as a region of interest by one observer and the software then follows this through the images. The VENC and R-R intervals are then manually inputted and the software calculates the velocity data for the images across the defined vessel. The maximum velocity during the calf squeeze was recorded by the software.

174

6.4.3 Results

Velocity data from MRI

Images have been reconstructed using modelling of the background muscle signal for the phase subtraction as the vessel size and shape changes over the flow impulse. An early problem with the reconstruction sequence on Matlab and the selection of the region of interest for digital subtraction meant that alternative software was chosen. The reconstruction was performed by the SPIN software. A region of interest for modelling of the muscle signal was drawn by one observer. The software then produces a graph of the average velocity across the femoral vein as against time. The coefficient of variation for 5 readings taken was 3.21%. Data was acquired to excel and the mean velocity waveform was used to identify the frame with the temporal peak mean velocity for each slice, an example of this is shown in Figure 58.

175

Figure 58: Shows the average velocity across the femoral vein during a calf squeeze

Figure 59: Shows the average of the maximum velocity for the 7 subjects relative to the valve leaflets using MRI

176

The maximum velocities for each subject during the calf squeeze were recorded 2cm distal to the valve,

within the valve and 2cm proximal to the valve. Using a one-way Anova test to compare the three values

there was a significant difference (p=0.013), and an unpaired student t-test demonstrated a significant

difference (p=0.004) between the maximum velocity 2cm distal to the valve and the maximum velocity

within the valve and a significant difference (p=0.013) between the maximum velocity 2cm distal to the

valve and the maximum velocity 2cm proximal to the valve. There was no significant difference

(p=0.887) between the maximum velocity within and 2cm proximal to the valve. This demonstrates that

the maximum velocity of blood increases as it passes through the valve leaflets.

Comparison of Ultrasound and MRI Results

The data regarding the maximum velocities from both the ultrasound and MRI data was compared by

Bland-Altman analysis. Figure 60: Shows a comparison of the maximum velocity in the femoral vein

measured by both MRI and ultrasound in the form of a Bland-Altman analysis. This shows a bias of -50

(SD 22) and 95% limits of agreement from -93 to -7.8. The velocities obtained from MRI were lower than those obtained from ultrasound in all cases.

177

Figure 60: Shows a comparison of the maximum velocity in the femoral vein measured by both MRI and ultrasound

in the form of a Bland-Altman analysis

6.4.4 Discussion

This study investigated the potential of both Ultrasound imaging with and without contrast and real-time

spiral MR image sequences to provide the flow profile around the first femoral vein valve distal to the

confluence of femoral and deep femoral vein in normal subjects. The information obtained from these

seven patients has been further analysed and used to contribute to the development of both the

computational and laboratory flow models described later in this Chapter and in chapter 8. Data obtained from both US and MRI demonstrates that the velocity of blood flow increases as it passes from distal to the valve, through the valve and then proximally towards the heart. This did not reach significance for the US modality, however.

There are limitations to this study. The US and MRI data was obtained from the same subjects but on different days and at different times. Venous haemodynamics can vary with time and physiological status346 The accuracy of US is limited not only by the equipment and software but also to operator related problem such as identification of the correct vessel, correct insonation angle relative to vessel cross section and ability to replicate this over multiple scans.

178

The MRI used here also had significant limitations. The valves themselves were not visible using MRI, therefore the location was estimated using the measurements obtained from US, measuring the distance distally from the confluence of the femoral vein and deep femoral vein. The subjects used for the experiment had to breath hold in expiration for the duration of the study, relying on their compliance. In addition, a 1.5T MRI scanner was used for this study. Sacrifices between temporal and spatial resolution that were made could have been minimised by using higher field strength, such a 3T scanner.

6.5 COMPUTATIONAL MODELLING OF DEEP VEIN

VALVES

6.5.1 Introduction

Background

There have been few previous studies to develop computational models of the venous system with and without a venous valve, in comparison to the studies of the arterial system, the work is very limited. Basic venous models require a resistor to represent viscosity of the blood, a capacitor to incorporate the vessel wall compliance and an inductor347. Pressure and flow data, as well as the tendency for smaller venules to collapse is harder to incorporate into these models. The model must also be changed with posture of the subject, as gravity will have an impact on all factors incorporated348. Other venous models have focussed on modelling the cardiopulmonary vasculature. Using the assumption that blood is a Newtonian fluid, a model of cardiopulmonary bypass was developed349, but this is not applicable to the lower limb venous system. Only one study focussed specifically on the modelling of a venous valve, however, the tubes were rigid, and no experimental subject data was used350. Zervides et al351 and Diaz-Zuccarini et al352 have

179

further advance this work and are developed a model into which a valve can be incorporated and have

studies the venous haemodynamics around the valve, but they did not include the valve sinuses.

Experience gained in computational analysis of large vessel haemodynamics combined with patient-

specific images from a previous study of the effect of external compression on the deformation and wall

shear stress in the superficial and deep veins in the lower limb 353, 354 laid a solid foundation for further

development of the computational model to incorporate the deep vein valves. Using computational fluid

dynamics, more information on the haemodynamics around the venous valve can be understood,

including the velocity, wall shear stress and flow patterns. Computational Fluid Dynamics (CFD) is a

method which uses algorithms to solve a system of partial differential equations by discretisation in order

to describe fluid flow355. Using CFD in blood flow studies can reduce the number of experiments on patients which are time consuming, and costly in terms of resources, as well as exposing patients to unnecessary radiation if CT is used. There have been great improvements in computing power and speed which have allowed more advanced calculations and therefore more comprehensive analysis of data. For flow data, this has allowed more comprehensive analysis of flow condition at in the vascular system.

Biofluid Mechanics

Viscosity is an inherent property that determines the fluid’s resistance to flow and is affected by factors

such as density (mass per unit volume) and shear rate (the velocity gradient across the diameter of the

vessel)356. When viscosity remains constant irrespective of the shear rate, the fluid is termed a

Newtonian’ fluid and when it changes with the shear rate, the fluid is termed ‘Non-Newtonian’. Non-

Newtonian fluids may be ‘shear thinning’, where the viscosity decreases at higher shear rates or ‘shear

thickening’ where viscosity increases at higher shear rates. Water is a Newtonian fluid while blood and

aqueous foams are examples of Non-Newtonian shear thinning fluids357.

180

Aims

This part of the project aims to develop a computer model of venous flow in normal deep venous function and validate the computational model using data obtained from ultrasound scanning.

6.5.2 Methods

Governing Equations

Assuming that blood is a Newtonian fluid, the model was governed by 2 equations. The first is the continuity equation, which derives as it is assumed that energy is neither lost nor gained, but conserved locally within a system, so any energy entering, must leave the system at the end. Therefore in this case:

Δρ + . (ρu) = 0

∇ Δt

Where:

= Divergence

∇ ρ = fluid density, assumed to be 1.60g/cm3 for blood t = time u = flow velocity vector field

As we have made the assumption that ρ is constant, as we have assumed incompressible flow, then the equation is simplified to:

. u = 0

∇

181

The second is the Navier-Stokes equation describing the motion of fluid substances simplified for

incompressible fluid355, which is:

ρ Δv + v . v = - p + µ 2v

∇ ∇ ∇ Δt

Where: v = flow velocity p = pressure

µ = viscosity of blood, assumed to be 0.035g/cm/s for blood

Ultrasound Imaging

The ultrasound imaging obtained in section 6.3 was used. One subject was excluded as the valve leaflets

could not be visualised throughout the complete process of opening and closing. Therefore 6 normal

subjects, all male, mean age 32 were included. AVI Videos of B-mode ultrasound images of the first

valve distal to the confluence of the femoral vein and deep femoral vein obtained while the subjects were

standing during normal respiration were used. If both leaflets could not be visualised, the imaging

acquired focussed on one leaflet, and geometry was duplicated. The averaged cross sectional velocity

from the velocity waveforms at the locations, 2cm distal to the valve, within the valve leaflets, and 2cm

proximal to the valve were used.

182

Geometry classification

DICOM images of the ultrasound scans were used. MATLAB (Mathworks), was used to open the

images and extract the geometry. As discussed, one of the limitations of the ultrasound imaging is that

both valve leaflets cannot be visualised at all times, however, where one leaflet was visible, symmetry

between the two leaflets was assumed, in terms of both length and movement, and symmetry of the valve

sinus was assumed. The valve was open for the majority of the time

Meshing

Once the geometry was obtained, models of the valve leaflet positions were created. A mesh was generated for each model, as part of the pre-processing of CFD, which ensures the discretisation of the geometric domain of the venous valve. This results in the formation of finite elements using the finite element method (FEM). In view of the geometric issue characterised, the components inside the grid comprise of partial differential equations (PDEs) that could predict haemodynamic variables. FEM transform, by discretisation, can then be approximated with algebraic equations. For this simulation,

PDEs are governed by the 2D Navier-Stokes comparison for incompressible liquid.

The denser the mesh, the more accurate the fluid behaviour will be, however, this will result in a longer computational time. Therefore a compromise must be reached. The mesh density can be changed at different locations within the vein and valve. In this case, it was increased around the valve leaflets and within the valve sinus to obtain more accurate results. Both quadrilateral and triangle mesh structures were employed. A grid dependency test was performed to ensure mesh quality. A mesh size of 9307 elements, which gave a change of <0.41 cm/s relative to the 17961 element mesh was chosen, as a mesh size of 5135 elements gave a velocity change of 1.33 cm/s.

183

Boundary conditions

Reliable boundary conditions are essential for reproducible and accurate results, as they provide the

solution to the PDEs. The inlet and outlet location of the models were defined to ensure the model was

directional. Other conditions that were applied include: a no-slip condition to the interior of the vessel, that is it was assumed that the velocity at the wall was 0cm/s, flow was laminar, and at the outlet the pressure was 0. Subject-specific velocity was initially applied at the inlet, however, no difference in the spatial distribution of the variables was found between subjects, an average velocity of 7.83 cm/s was applied at the inlet for all models, unless specified.

Steady-flow simulation

The dynamic viscosity of blood was 0.035 g/(cm/s) and the density was 1.60 g/cm3 respectively. The

Reynolds number ranges between 34.43 and 344.33 depending on the flow rate in the deep vein and

therefore laminar flow could be assumed354.

ADINA (Automatic Dynamic Incremental Non-Linear Analysis, v 8.8, ADINA R&D Inc., USA), finite analysis software was used to create the model. This is able to analyse slid, fluids and fluid-structure interactions for both linear and non-linear systems, and each CFD model was created and meshed using this.

Summary of Methods

To summarise the methods:

1. The geometry of the vein and valve leaflets was extracted from the ultrasound imaging.

2. The geometry was reconstructed for different valve leaflet positions, that is closed, minimal

opening, intermediate opening and maximal opening.

184

3. A mesh was generated using the Adina software, boundary conditions applied and a model

constructed.

This work was done in collaboration with Patrizia Lee, as part of her MSc project, under the supervision of Professor Yun Xu.

6.5.3 Results

Valve, Sinus and Vein Wall Geometry

The measurements taken from the ultrasound imaging including the diameter of the vein proximal to the valve, at the base of the valve leaflets, at the widest point of the valve sinus and distal to the valve, as well as the visible length of valve leaflet are described in Table 17. These represent the mean values from all of the subjects, except subject 2 who was not included as the valve leaflets could not be seen adequately to establish the valve geometry.

Table 17: Shows the diameter of vein, sinus and length of the valve leaflets

Diameter Diameter at Length of Leaflet basal Length of

proximal to valve sinus sinus (cm) length (cm) valve leaflet

valve (cm) (cm) (cm)

Median 0.93 1.06 1.41 0.14 0.83

Average 0.94 1.07 1.45 0.13 0.88

Standard error 0.05 0.04 0.10 0.02 0.12 of mean

Percentage 9.61 6.95 13.34 27.14 27.64 difference (%)

185

Using this information and the assumptions previously made, Adina finite analysis software was used. A

mesh was overlain onto structures with the geometry of the subjects’ veins as described above. There is a difference of up to 27% between individuals.

Valve motion

The AVI videos of the ultrasound of real-time flow through the valve demonstrate the stages of the

“valve cycle” previously described by Lurie et al, 2002224. They described the valve passing through

closed, opening, equilibrium and closing phase. In the standing position with normal respiration, the majority of the time the valve is in at the maximal opening stage. The valve closes only occasional and usually for less than 2 seconds. One of the issues with this study is that the valve leaflets are dynamic and during the valve cycle can move into and out of the visual plane.

Velocity profile

Derived from Ultrasound Imaging

The velocity waveforms observed on the ultrasound during relaxed respiration are similar to those

demonstrated previously in other studies of lower limb veins using other imaging modalities354, 358. The

velocity waveforms do appear to have a very low magnitude pulsatility, which was observed in the

ultrasound imaging, and given its frequency seemed to represent changes caused by respiration. During

inspiration, the changes in intrathoracic and intra-abdominal pressures contribute to flow in the deep vein of the lower limbs.

Steady-state flow analysis

Although blood flow through the body is not constant, as in the venous system it varies with respiration,

this study assumes that a state of steady flow exists in the deep veins. Between respirations, this is the

case, and therefore the model is that of venous flow during this steady state flow.

186

CFD simulation for each of the six included subjects was performed. This is shown in Table 18. No

spatial changes in flow, pressure and WSS patterns were found. The model was therefore computed from

during the different stages of the valve cycle as previously discussed:

• minimal opening

• intermediate opening

• maximal opening

with an averaged maximum velocity load of 7.83±0.98 cm/s.

Velocity profile

The contour plots represent the different velocities of blood flow across the vessel. These are shown in

Figure 61. These are comparable with those found by Lurie et al224, who showed that the velocity of blood flow in the femoral vein increases close to the valve, demonstrated in the contour plots by he change in colour moving through the valve. The maximum velocity is achieved just proximal to the apex of the leaflets, and this is further increased when the valve is in the minimal opening phase, when the maximum velocity at this point is 162 cm/s, gradually decreasing to a maximum velocity of 12.89 cm/s at

this point in the maximal opening phase. In the contour plots, the lowest velocities are seen in the

venous valve pocket, between the base of the valve leaflet and the sinus.

187

Figure 61: Shows the contour plots of velocity distribution (cm/s)

188

Flow pattern

Vector plots of the velocity profile were developed for the different stages of the valve cycle, as shown in

Figure 62. As shown, blood flow follows the wall, but at the apices of the leaflets, separates resulting in a flow void in the area behind the valve leaflet, in the venous valve pocket. Some distance beyond the leaflets, the flow resumes the original velocity profile. To further evaluate the venous valve pocket, the vector plots in these areas were replotted. This is shown in Figure 63. Three separate vortices were observable within a single venous pocket. This could be relevant as if the valves are damaged these mechanisms will fail.

189

Figure 62: Shows the vector plots (vector size scaled to magnitude) for the velocity profile along the vein (cm/s)

190

Figure 63: Shows the vector plot of flow pattern for a) minimal opening of valve, b) maximum opening of valve; arrows indicate the location of vortices in one side of the pocket.

191

Wall Shear Stress (WSS)

The distribution of wall shear stress in the model is shown in Figure 64, and it varies at different points in the valve cycle. During minimal opening, the maximum WSS is at the apex of the leaflets and then shifts as the valve leaflets open further along the leaflets. There was decrease in maximum WSS when comparing minimum to maximum opening of the valve by 97.1%. The WSS in the venous valve pocket was close to 0 g/cm/s2).

192

Figure 64: Shows the contour plot of Wall Shear Stress (dyne/cm2)

193

Model validation

Table 18: Shows the comparison between subject-specific experimental and computed velocity values ascertained at different positions relative to the deep vein valve

Subject Inlet Velocity (cm/s) taken from between Velocity (cm/s) taken from 2cm

No. Velocity leaflets proximal to leaflets

Experi- Computed % Diff. Experi- Computed % Diff.

mental mental

1 5 8.5 8.32 2.16 7.5 6.8 9.33

3 10 17.5 16.2 8.02 18 14.3 20.6

4 10 19 16.2 17.3 18 14.3 20.6

5 10 12 16.2 -25.9 20 14.3 28.5

6 6 10 9.89 1.11 17 8.3 51.2

7 6 12 9.89 21.3 12 8.3 30.8

Median 8 12 13.0 8.02 17.5 11.3 24.5

Mean 7.83 14.1 12.8 3.99* 15.4 11.1 26.8*

SEM 0.980 1.60 1.55 7.63 1.93 1.47 5.76

The velocities from the computational model were compared with those from the ultrasound images of the six subjects’ deep vein. The ultrasound image velocities taken between the leaflets and 2 cm proximal to the valve leaflets were compared with the computed values. This is shown in Table 18. There is some discrepancy, but the maximum measurable velocity for each subject was applied to the model, which

194

could allow for this. Therefore, the comparison was made between this value and the maximum value on

the computational model. These differences are small and do not reach significance, for the

measurements between the leaflets p=0.75 and proximal to the valve p=0.25.

6.5.4 Discussion

A “typical” model of the femoral vein including the valve sinus and a pair of crescent shaped flexible

leaflets attached at a point to the vein wall has been constructed. A model to simulate the dynamic

movement of the valve leaflets has been developed and incorporated into the venous flow model. The

geometry was extracted from the ultrasound imaging carried out on normal subjects and the model based

on average dimensions measured from these subjects, all of which have normal deep venous function.

Fluid-structure interaction (FSI) simulations of venous flow in the typical normal vein model has been

performed and analysed to establish flow patterns in the venous valve pocket (VVP) where thrombi are

likely to form. There are several considerations that have been accounted for:

 The valves have flexible leaflets

 Blood consists of cells, platelets and plasma, making it behave in a non-Newtonian way, however,

for this model it is assumed that blood behaves as an incompressible Newtonian fluid

 Flow is turbulent, especially around special areas such as valves and bifurcations

This is the first study that has used CFD analysis based on realistic geometry extraction and velocity data from ultrasound images of the femoral vein in human subjects to demonstrate flow dynamics around the valve and valve sinus, the distribution of velocity changes and WSS exerted by the flow. The findings are consistent with previous flow patterns in other veins, both ex vivo and in vivo224, 359. The vortices in the venous valve pocket as well as the acceleration of the blood as it passes through the valve leaflets, which act as a relative stenosis. This keeps the valve leaflets separate from the wall of the vein.

195

The model developed here differs from a previous study which showed that the separated part of the

flow should reach the sinus wall, followed by vortex formation359, however, this study was based on dog

at post mortem. This model demonstrates the flow forms vortices further downstream.

The valvular motion can thus be described according to Bernoulli’s principle, that is an increase in the

speed of the fluid occurs simultaneously with a decrease in pressure or a decrease in the fluids potential

energy. When the valves are minimally open, the pressure gradient is at its highest, as is the velocity

gradient. This exerts a pressure against the luminal surface of the leaflet, forcing the valve to open. The

vortices are formed once the valve is maximally open, which increases the pressure within the sinus,

resulting in closing of the valve.

The vector plots revealed three vortices in the venous valve pocket. This could result in cell aggregation

and if this gets isolated from the flow in the vein then this could be a potential site for the initiation of

thrombus formation within the vein360.

This study has many limitations. It corresponded well with the data obtained from the ultrasound imaging, however, the assumptions that were made initially will affect the results. The fluid properties and geometry of the valves that were obtained from the ultrasound imaging will lead to inaccuracies.

Ultrasound imaging itself has limitations, there is inherent user variability. Out images were all performed by the same experienced vascular scientist to make an attempt to overcome this. The model was developed under the assumption of steady state flow, thus providing a snapshot of flow through the vein valve, however this could be further developed to incorporate respiration and the action of the calf squeeze, which was not incorporated into this model.

In summary, by utilising CFD and FSI, computational models could potentially provide more data to guide the design of a valve or be used to develop patient specific models to accurately calculate the size and location that the valves developed should be sited.

196

7. DEVELOPMENT OF A NOVEL

MATERIAL FOR USE IN THE VENOUS

SYSTEM

7.1 CARDIAC VALVE & VASCULAR GRAFT

ENGINEERING: PREVIOUS WORK

Introduction

To date, no material is suitable ideal for use in the venous system, as it is a low flow, low velocity, and

therefore high thrombogenicity system. As demonstrated in Section 3.2, there is a paucity of literature on

the use of bioengineered or synthetic materials in the venous system. However, there has been similar

work done in the cardiac valve and vascular graft engineering fields. This section will review this

literature as well as the literature available on a novel material which is considered may be of sufficiently

low thrombogenicity for use in the venous system.

Cardiac valve disease is a major worldwide health issue and cardiac valve replacement using metallic or

tissue valves is common. Each has its advantages and disadvantages, but neither provides the ideal

solution, therefore research has continued to try to develop an alternative valve which would provide the

answer. Increasing work has also been carried out in the field of tissue engineering of vascular grafts, as

atherosclerosis continues to increase, and if autologous vein graft is not available for bypass, the

expanded PTFE, first introduced in 1973361, remains the most commonly used synthetic prosthesis. In the medium term it develops intimal hyperplasia and eventually graft occlusion. The previous work done in these fields, not including these common practices, is discussed here.

197

Methods

Relevant papers were identified through systematic searches of Pubmed (1966 to August 2013). The search terms “cardiac valve”, “heart valve”, “aortic valve”, “mitral valve”, “vascular graft” AND

“polymer”, “tissue engineer”, “scaffold” were used. Article titles and abstracts were screened for

inclusion. A manual reference list search was also carried out for further appropriate studies to be

considered for inclusion, ensuring relevant papers published before 1966 were included. Relevant full text

articles were scrutinised and all studies reporting details of tissue engineered cardiac valves at any stage

were included. Articles concerning native cardiac valves, peri-operative outcomes, aortic stenosis,

superficial vein valves and deep venous valves were excluded.

Results

Although the initial search strategy yielded 1039 results, with 4 articles included from elsewhere, following exclusion of non-relevant publications, a total of 25 articles were included, as shown in Figure 65.

198

Figure 65: Shows a summary of the search strategy for previous cardiac valve and vascular graft engineering

Search terms entered: (((((cardiac valve[Title]) OR heart valve[Title]) OR aortic valve[Title]) OR mitral valve[Title])) AND (((polymer) OR tissue engineer) OR scaffold)

Records identified through database Additional records identified through searching other sources (n = 1039 publications (n = 4) between 1963 and 2013)

Records excluded (n = 1003) Records screened (title) Exclusion criteria: Articles on (n = 1043) native cardiac valves, peri- operative outcomes, aortic stenosis, superficial vein valves and deep venous valves title of paper not within scope of review

Abstracts and full-text articles assessed for Records excluded eligibility (n = 40) (n = 15) Excluded: duplicate records, conventional open surgical techniques, venous valve studies

Studies included (n = 25)

Cardiac Valves

199

Biological Work

Badylak et al362 in 2009 noted that at that time there did not appear to be a commercially available vessel

biological scaffold and carried out some of the most extensive work with several published studies using

ECM as a scaffold for a tissue engineered cardiac valve. These include a study by Reing et al363 to assess

the effects of processing methods on the mechanical and biological properties of porcine dermal

extracellular matrix scaffolds, including growth factor content, glycosaminoglycan content, the ability to

support cell growth in vitro and the mechanical strength. Different decellularisation protocols are

successful, but have different effects on detectable VEGF, TGF-beta1, glycosaminoglycans, ball-burst

strength, and the ability of the ECM to support cell growth in vitro. Further work by the same group by

Tottey et al364 demonstrated differences in the properties of the scaffold depending on the age of the source animal, using porcine SIS. They found that SIS from 3 week old animals failed at lower stress, 52 week old SIS failed at higher strain and both 3 week and 52 week SIS had a lower elastic modulus. It is known that SIS becomes thicker with age, and age increases the resistance to enzymatic degradation

(collagenase), which could provide an explanation for this. Brown et al365 characterised the surface

properties of extracellular matrix scaffolds using SEM and ToF-SIMS, to include the luminal and

abluminal surfaces (luminal sides relatively smooth) of urinary bladder matrix and SIS, which were found

to be different and liver ECM, which showed no difference. It is possible that source of ECM, as well as

the surface used may modulate the phenotype of cells that are seeded onto it. It was also found that the

presence of a basement membrane also affects growth of cells. When undergoing the reseeding process

Tottey et al 366 found that using perivascular stem cells harvested from foetal muscle tissue by flow cytometry and ECM harvested from porcine urinary bladder matrix, that ECM degradation products and low oxygen conditions (6% rather than 21%) enhance the regenerative potential of perivascular stem cells.

Cells cultured in 6% oxygen had 7 times more DNA synthesised and 8 times more cells in number, possibly due to higher intracellular reactive oxygen species and higher superoxide dismutase.

When utilising biological material as a scaffold many other factors are involved and have an effect on the cells growth and differentiaton. Chiu et al367 determined via in situ hybridisation and antisense RNA

probes that TGF-β, BMP-2 and VEGFA was expressed in chick embryo valvulogenesis at different times

and that TGF-β and VEGFA modulate migration of AV valvular progenitor cells but BMP2 does not.

200

TGF-β 3 and BMP2 increase cushion invasion and different growth factors differentially regulate valve

progenitor phenotype. Gwanmesia et al368 found that TGF-beta1 and VEGF had opposite effects on the degeneration of aortic valvular interstitial cell, which could be modified by the ECM protein fibronectin.

VEGF inhibited calcific nodule formation independent of ECM protein coating, where as TGF-β1 increased calcific nodules if the ECM protein was coated with collagen, laminin or was uncoated, but a fibronectin coating caused reduced nodule formation.

Sacks et al369 considered the aortic valve cellular and ECM components: Endothelial cells are important for thromboresistance and mediation of inflammation, interstitial cells modulate durability and synthesis of the ECM and express enzymes including MMPs and elastin, and in the ventricularis layer have a mechanical role, glycosaminoglycans in the spongiosa layer cushion against shock and have a mechanical role, collagen provides strength and stiffness and is the most important structural component. Valentin et al370 studied the effect of macrophage depletion by injection of clodronate encapsulated in liposomes

on porcine SIS crosslinked by CDI, and not crosslinked, placed in rat abdominal wall. They found that

cross-linking increases resistance to degradation by collagenase by depletion of macrophages at the site.

Brown et al371 using porcine urinary bladder ECM, found different subgroups of macrophages present will also modify the phenotype of the cells.

Polymer work

There has been extensive work on the development of a cardiac valve using a polymer. A review by

Ghanbari et al372 in 2009 reviewed the literature to date regarding polymeric cardiac valves. Required

characteristics of the material to be used for the valve that were essential were; the need for the same

durability of a mechanical valve, combined with the haemocompatibility and low thrombogenicity of a

tissue valve. Flexible polyurethane was not sufficiently durable. Silicone demonstrated good flexibility

and biocompatibility, but a short durability with early structural failure. PTFE resulted in

thromboembolism and major complications. PEU and PCU were susceptible to biodegradation and

calcification, as the soft segment of the polymer degrades quickly. PVA, SIBS, polydimethylsoloxane –

polyhexamethylene oxide PU and PDS-PCU nanocomposite all need further assessment for biostability.

201

In addition, the effect of incorporating anti-calcific agents or heparin needs further investigation.

Paediatric patients have additional complex requirements for cardiac valve replacement, as they will grow

over the lifetime of the valve. Sachweh and Daebritz373 noted that manufacturing methods were very

relevant when considering Silicone, PTFE, Dacron, PVC and PU to be used as polymeric heart valves,

and when they tested PU valves in the aortic and mitral valve positions found them to be comparable

with existing metallic and tissue valves used. Christenson et al carried out both in vivo studies on

degradation374 and in vivo and in vitro correlations on the oxidative mechanisms of PCU and PEU375.

They found that PEU and PCU had different soft and hard segments, and different mechanical properties, and that PCU was more susceptible to degradation. However, PCU undergoes less degradation when subjected to hydrogen peroxide and cobalt chloride solution. Kannan et al376 found

that the resistance of PU to degradation could be improved by the incorporation of a polyhedral

oligomeric silsesquioxane nanocore and in this case the addition of carbonate improved the resistance to

degradation when compared to bovine pericardium. Silsesquioxane nanocomposite polymer was also

found to have anti-calcification potential in in-vitro conditions by Ghanbari et al377. Calcification is due to reaction between aldehyde groups of phospholipids and circulating calcium ions causing calcium phosphate, which causes cracking. Modification of polyurethane with covalently attached bisphosphonate diethylamino moieties reduces calcification and reduces mechanical stress. Incorporating

POSS into the backbone gives reduced hydrophilicity, improving surface integrity, and reducing thrombogenicity, that is reducing platelet and protein adsorption.

Poly(SIBS) is a biostable thermoplastic elastomer, which has been extensively investigated for use in many biomedical applications including cardiac valves378. It was considered suitable for use in the vascular system as it does not substantially activate platelets, has few polymorphonuclear implants and myofibroblasts, embrittlement was not observed over time, neither was calcification or degradation.

However, it was susceptible to stress cracking with organic solvents, had a tensile strength less than polyurethane and sterilisation was very difficult due to poor gas permeability. After early promising studies of SIBS by Gallocher et al379, Wang et al380 further studied SIBS as a trileaflet heart valve in an in- vivo sheep model, however, it exhibited cracking and severe tissue reactions, after becoming calcified and the material failed.

202

After earlier work, Schmidt et al381 used a minimally invasive method of implantation of trileaflet tissue

engineered heart valves consisting of a scaffold of non-woven polyglycolic acid mesh coated with poly-4-

hydroxybutyrate, fixed in nitinol stent, seeded with myofibroblasts and endothelial cells into 16 sheep.

The valves functioned, but became rapidly thickened.

Vascular Grafts

Much of the work on vascular grafts has focussed on the development of grafts which are pre-seeded

with cells prior to implantation. L’Heureux et al382 used a tissue engineered blood vessel utilising

autologous fibroblasts and endothelial cells harvested from small biopsies of the skin and superficial

veins. Sheets of living fibroblasts were wrapped around a stainless steel mandrel, after 10 weeks of

culture, and after a further 10 weeks, the luminal surface was devitalised by air drying and the endothelial

cells seeded and conditioned to flow and pressure for 7 days. Of the 6 patients, 1 thrombosed and 5

remained patent, however, this method is costly and time consuming and unlikely to become widely

adopted. McAllister et al383 demonstrated that these tissue engineered sheets could be used effectively for

haemodialysis access in 10 patients, however, although most remained patent, the grafts failed due to

aneurysm. Teebken et al384 used ECM from decellularised vessels such as pig aortas, and reseeded them

in vitro using endothelial cells harvested with collagenase from saphenous vein and smooth muscle cells

harvested from saphenous vein using collagenase A and elastase. Cells were seeded and the construct was

inserted into a perfusion circuit. This was successful in that endothelial cells formed a single layer, where

as myofibroblasts showed a tendency to build more than one cell layer.

Other groups focussed more on the use of polymers to develop a vascular graft. Sun et al385 studied a material which is thermoplastically similar to polyethylene, with a strength and strain suitable for vascular tissues, namely Poly(3-hydroxybutyrate-co-3-hydroxyhexanoate). It has a good affinity for vascular smooth muscle cells and endothelial cells and after modification with silk fibroin showed an improved call affinity. Other materials have been investigated for use as biodegradeable coronary stents, including Poly-

L-Lactide386, tyrosine poly(desaminotyrosyl_tyrosineethyl ester) carbonate, and poly(anhydride ester)

203

salicylic acid, but all have shown issues with restenosis and intimal hyperplasia and therefore will not be

considered for the development of vascular grafts.

Discussion

Polymers provide consistent, reproducible material from which to manufacture the valve, in contrast to

biological scaffolds which are much more variable in their properties. ECM for example varies with the

age of the source donor, different methods of decelularisation, residual MMPs, elastin, collagen, growth

factors and other constituents. Once implanted, it is important to consider the effect cells, such as

macrophages, cytokines and other circulating factors from the recipient patient and the effect that these

will have. When developing a suitable polymer, its properties must be fully optimised and characterised

to include; thrombogenicity, haemocompatibility, biocompatibility, mechanical properties and the effect

of the material on host cells. Also, surface and structural modifications can improve the properties of the

polymer, ensuring that these do not adversely affect the other properties.

7.2 SELECTION OF POLYMER

Any patient who undergoes deep venous surgery or intervention requires a period of anticoagulation with

heparin or warfarin387, and if a foreign material is implanted, this can sometimes be a lifelong requirement.

A material for use in the venous system as a valve should ideally have the following properties:

• Flexible – to allow for leaflet movement

• Non-biodegradable – in order that it remains in situ for full endothelialisation and

maintains the properties of the valve over time

• non-thrombogenic - to prevent early or late occlusion

204

• Biocompatible – to prevent excessive tissue reaction, which may cause thrombosis

• Support cell growth - allow growth of 1-2 layers of vascular endothelial cells only

• non-fatigable and non calcifying – to maintain its mechanical properties and flexibility

over time

• have option for drug releasing component e.g. heparin – reducing the need for the

patient to take systemic anticoagulation

• potential for dip coating in decellularised vascular endothelium extracellular matrix – to

promote endothelialisation in vivo

• Ability to expand and contract with the vein, which can change its diameter two-fold

with postural changes, to prevent a relative stenosis and alteration in haemodynamics

Presently, polymers and polymer coated metal stents, used in the vascular system have several problems:

they are very thrombogenic and they lack haemocompatibilty and biocompatibility. In addition they lack

the required mechanical properties. A non-thrombogenic material would have implications not only for

the valve, but also for a venous stent to treat patients with deep venous obstruction and potential uses in

the arterial system as well.

MPC has been demonstrated be biocompatible and to have antifouling and antithrombogenic properties.

It has been shown to reduce the non specific absorption of plasma proteins and cells and for proteins

that are immobilised on it, appears to improve their longevity and stability388. MPC copolymers are

compatible with cell growth and have been suggested they would be beneficial for cell and tissue

engineering, in this study when it was polymerised with butylmethacrylate389.

In vitro studies demonstrated the ability of MPC to reduce platelet adhesion onto surfaces after contact with ovine blood, as shown in Figure 66. Recently, there have been further studies looking at methacrylolyoxyethyl phosphorylcholine (MPC) and its use in arterial stents 390, 391. Its haemocompatibilty

205 has been demonstrated in several studies392, 393. Modifying surfaces of other polymers with MPC polymers has been demonstrated to protect from thrombus formation when blood is passed over these surfaces394.

Figure 66: Shows that MPC reduces platelet adhesion, reproduced from Ye et al, 2003 395

Therefore MPC was chosen as the main polymer on which to base this novel copolymer from which to develop a novel deep vein valve replacement. The other monomers selected were Hydroxypropyl methacrylate (HPMA), which was selected as it is very hydrophilic, and has some crosslinking sites.

Methacrylate based polymers are biocompatible and Hydroxypropyl methacrylate is hydrophilic, non toxic and non-immunogenic, and has been documented to be useful for developing polymers for drug delivery396-398. Trimethylsilylpropylmethacrylate (TMSPMA) will hydrolyse in the presence of acid or base through the Si-OH group, creating a stable insoluble polymer. During the crosslinking process, this will form a backbone and if fibres are formed, will move towards the centre of the fibre forcing the more hydrophilic biocompatible MPC to the surface of the fibres, thus improving the haemocompatibility of the final polymer.

206

7.3 AIMS

This part of the project aims to synthesise and investigate this novel biocompatible polymer, MPC-

HPMA-TMSPMA, for use in the venous system and develop a scaffold from which a valve could

subsequently be made.

7.4 PATENT SEARCH

A full patent search was carried out to ensure that there were no issues with intellectual property. The

results of this are described below: A freedom-to-operate search on “MPC coated implantable valve” for all in-force patent documents filed since 01/01/1990 was executed. The search yielded 10 patent documents pertaining to MPC coated implantable valve:

• EP2326283 A1 BIOABSORBABLE BLEND FOR TEMPORARY SCAFFOLDING OF THE

BLOOD VESSEL WALL

• US20100003303 A1 POLYMERS OF FLUORINATED MONOMERS AND

HYDROCARBON MONOMERS

• US20090238854 A1 PLASTICIZERS FOR COATING COMPOSITIONS

• US20050208093 A1 PHOSPHORYLCHOLINE COATING COMPOSITIONS

• EP1784434 A1 POLYMERS OF FLUORINATED MONOMERS AND HYDROPHILIC

MONOMERS

• WO2006025858A2MICROFABRICATED CELLULAR TRAPS BASED ON THREE-

DIMENSIONAL MICROSCALE FLOW GEOMETRIES

• WO2007005253A1 BIODEGRADABLE POLYMER FOR COATING

• WO2006118808A1 AMORPHOUS POLY (D,L-LACTIDE) COATING

207

• WO2005118018A1 HEPARIN CONTAINING BLOCK COPOLYMERS COATED ON

IMPLANTS LIKE STENTS

• WO2006026325A2 IMPLANTABLE TISSUE COMPOSITIONS AND METHOD

Patent Document “US20100003303 A1” discloses a method of treating a disorder in a patient comprising

implanting in the patient an implantable device not limited to artificial vein grafts. The implantable device

includes a coating comprising at least one biocompatible polymer not limited to 2- Methacryloyloxyethyl

phosphorylcholine (MPC).

Patent Document “EP1784434 A1”discloses a biocompatible polymer blend comprising a biocompatible

polymer selected from 2-methacryloyloxyethylphosphorylcholine (MPC), coated on an implantable device used for the treatment of disorders such as anastomotic proliferation for vein and artificial grafts.

Patent Document “WO2006025858 A2” discloses a method of cross-linking an implantable tissue comprising steps of treating the tissue under effective cross-linking conditions with a monomer capable of polymerization being selected from the group of sulfonate or phosphorylcholine group. The tissue prepared includes vascular patch, heart valve, venous valve bioprosthesis and vascular grafts etc.

These are being taken into consideration in the development of the polymer, and an Invention Disclosure

Form has been submitted to Imperial Innovations, who are continuing the search.

7.5 POLYMER SYNTHESIS

208

7.5.1 Introduction

The monomers were selected as previously described. Free radical polymerisation is a method of forming a polymer by the progressive addition of free radicals which are formed by the addition of an initiator.

The building blocks of monomers are then added to the polymer as the radicals are transferred from the initiator to the monomers.

7.5.2 Methods

Monomers

The materials used are, MPC, Hydroxypropyl methacrylate (HPMA) and Trimethoxysilyl propyl methacrylate (TMSPMA). The structure of these is shown below in Figure 67, Figure 68 and Figure 69 below. These were drawn using ChemDraw®, PerkinElmer.

Figure 67: Shows the structure of MPC

Figure 68: Shows the structure of HPMA

209

Figure 69: Shows the structure of TMSPMA

Both linear and branched versions of the polymer have been synthesised. In order to do this the monomers above are combined with EGDMA (the branching agent) and DDT (the termination agent) for the branched version of the polymer and with DDT only for the linear version. In order to vary the degree of branching, the amount of EGDMA can be varied and to vary the molecular weight of the polymer, the amount of DDT added can be varied, which reduces the chance of the polymerisation being terminated early. If the amount of DDT is reduced, the molecular weight of the polymer should increase.

The following ratios have been tried in the synthesis of the polymer, as shown in the Table 19 below:

210

Table 19: Shows the ratios of monomers combined to make both linear and branched versions of MPC-HPMA-

TMSPMA

Sample MPC HPMA TMSPMA EGDMA DDT

Linear

1 50 25 25 - 2

2 50 25 25 - 1

3 50 25 25 - -

4 50 15 35 - -

5 50 5 45 - -

Branched

1 50 25 25 10 10

2 50 25 25 10 7.5

3 50 25 25 10 5

Free Radical Polymerisation

Free radical polymerisation is a method of forming a polymer by the progressive addition of free radicals, which in this case are formed by the monomers. The free radicals are formed by the addition of an initiator, which in this case was AIBN. The building blocks of monomers are then added to the polymer as the radicals are transferred from the initiator to the monomers. The radical in this case is created by heating the initiator so that it produces a radical.

211

The materials were purchased as follows:

• 2-Methacryloyloxyethyl phosphorylcholine, 730115 Aldrich

• Hydroxypropyl methacrylate, 268542, Aldrich

• 3-(Trimethoxysilyl) propyl methacrylate, M6514, Sigma

• Ethylene glycol dimethacrylate, 335681, Aldrich

• 1-Dodecanethiol, 471364, Aldrich

• Methanol, anhydrous, 322415, Sigma-Aldrich

• Azobisisobutyronitrile, 441090, Aldrich

Prior to the free radical polymerisation process, the monomers are prepared to remove stabilising agents.

The MPC was placed in a 500ml volume round bottom flask in powder form with a magnetic stirrer and the anhydrous methanol added, under anhydrous conditions. All monomers are passed in their liquid form through an aluminium oxide column prior to use to remove any impurities or inhibitors. These liquid monomers are then added to the flask while maintaining anhydrous conditions, without oxygen, under nitrogen gas only, to make a 10% solution, according to the ratios described in Table 19, with the molar weights and volumes calculated, as shown in the results section. After adding the dry MPC, the anhydrous methanol is added next, followed by the HPMA, TMSPMA and the branching and termination agent, ensuring anhydrous conditions at all times. The initiator, AIBN, dissolved in anhydrous methanol under anhydrous conditions and then added to the monomer solution and the mixture is heated to 70°C in an oil bath, with “burping” to ensure the pressure inside the flask is released at 45°C and 60°C. This temperature is then maintained for 48 hours and the mixture is stirred constantly with the magnetic stirrer inside the flask. This process is free radical polymerisation. The set up is shown in Figure 70 below.

212

Figure 70: Shows the laboratory set up for free radical polymerisation

Once this process is complete at 48 hours, the heat is turned off and the resulting polymer solution is allowed to cool for 24 hours. If anhydrous conditions are maintained, this solution can be stored in the flask at room temperature for up to 1 month before the next stage is carried out. The methanol solvent is evaporated using a rotary evaporator, which combines heat and a vacuum to evaporate the solvent. This was required to form a viscous solution ready for electrospinning.

The final polymer solutions underwent NMR spectroscopy in anhydrous deuterated methanol (methanol- d4), 569534 Aldrich, to determine the rates of conversion of the monomers to the final polymer. This was performed in the Cross-faculty NMR Centre, using the Avance II 800 spectrometer. 5mg of the polymer was dissolved in 700ml of methanol-d4 and placed in an NMR analysis tube, and analysis was performed using Topspin software. Subsequently, MestReNova analytical chemistry software was used to perform the integration and calculations from the NMR spectra.

213

7.5.3 Results

Monomers

The molecular weights and densities of the monomers are shown in Table 20.

Table 20: Shows the known molecular weights and densities of the monomers used, provided in the product specification information

Monomer Molecular Weight Density (g/ml)

MPC 295.27 N/A - Solid

HPMA 144.17 1.066

TMSPMA 248.35 1.045

EGDMA 198.22 1.051

DDT 202.40 0.845

Using these molecular weights, the volume of the liquid monomers required for each of the syntheses described in Table 19, as well as the total solid, volume of anhydrous methanol, weight of AIBN and volume of methanol used to dissolve the AIBN is shown in Table 21. 2g of MPC was always used for each synthesis.

214

Table 21: Shows the weights and volumes of monomers, anhydrous methanol and initiator used for each synthesis

Sample MPC (g) HPMA (ml) TMSPMA EGDMA DDT (ml) Total weight Methanol AIBN (g) Methanol for

(ml) (ml) of solid (g) polymer AIBN (ml)

solution (ml)

Linear

1 2 0.458 0.805 - 0.065 3.384 33.84 0.034 3.38

2 2 0.458 0.805 - 0.032 3.357 33.57 0.034 3.36

3 2 0.458 0.805 - - 3.329 33.29 0.033 3.33

4 2 0.275 1.127 - - 3.471 34.71 0.035 3.47

5 2 0.092 1.449 - - 3.612 36.12 0.036 3.61

Branched

1 2 0.458 0.805 0.256 0.324 3.872 38.72 0.039 3.87

2 2 0.458 0.805 0.256 0.243 3.804 38.04 0.038 3.80

3 2 0.458 0.805 0.256 0.162 3.735 37.35 0.037 3.74

215

Free Radical Polymerisation

The resulting polymer, MPC-HPMA-TMSPMA, should be a clear solution as shown below in Figure 71.

However, the branched polymer, whenever the EGDMA was introduced became cloudy, and this did not resolve as the polymerisation process progressed. A second batch of EGDMA was purchased and this did not resolve the issue. Branched versions of the polymer were therefore not used for electrospinning or for further development of the valve.

Figure 71: Shows the clear polymer solution

The predicted structure of the resulting polymer is shown in the Figure 72 below. The polymerisation

process should stop by either combination or disproportionation. Both involve the combining of 2 free

radicals and the cessation of the polymerisation process.

216

Figure 72: Shows the predicted structure of the polymer MPC-HPMA-TMSPMA

Nuclear magnetic resonance (NMR) spectroscopy was carried out for the monomers MPC, HPMA,

TMSPMA and the linear versions of the polymer. All were performed in methanol-d4. The branched version was not further investigated due to the cloudiness of the polymer solutions. The resulting NMR spectra are shown in Figure 73, Figure 74, Figure 75, and Figure 76 below. The structure of the

monomers are shown on the NMR spectra, and labelled with letters which correspond to the peaks in the

spectra, which are labelled in each NMR spectra with the same letter. Figure 74 shows the spectra for

HPMA. This is, in fact, a mixture of Hydroxypropyl and hydroxyisopropyl methacrylate. “d”, “e” and

“f” apply to the hydroxypropyl methacrylate and “d*”, “e*” and “f*” apply to the hydroxyisopropyl

methacrylate.

217

Figure 73: Shows the NMR spectrum for the monomer MPC

MPC

a d f

b e g c c

f a b e d g

Figure 74: Shows the NMR spectrum for the monomer HPMA

HPMA

a c g b

c f

a b

d e* e f* d*

218

Figure 75: Shows the NMR spectrum for the monomer TMSPMA

TMSPMA g a f d b g e c g c g a b d f

Figure 76: Shows the NMR spectrum for the linear version of the polymer MPC-HPMA-TMSPMA

MPC-HPMA-TMSPMA (Linear)

219

The relevant peaks are circled in blue. Due to the complexity of the polymer, however, it was not possible to fully characterise the exact structure of the polymer, however, based on resulting weight, the conversion of the monomers to polymer was >99% for all the linear versions of the polymer. It was therefore assumed that the resulting polymer had all the monomers combined in the rations that were put into the solution for synthesis.

The linear versions of the polymer in solution were then prepared ready for electrospinning. This involved removal of the solvent to create a viscous solution using a Rotary Evaporator or “Rotovap”.

The solvent could only be partially evaporated, as if it was fully evaporated, the dry polymer could not be dissolved in either methanol or HFIP for electrospinning. Samples of the linear polymer solution 1 and

2, that is those containing the termination agent DDT did not form an adequately viscous solution until more than 95% of the solvent was evaporated. This suggests that the molecular weight of the resulting polymer is too low. This did not leave sufficient polymer solution for electrospinning. Therefore, Linear samples 3, 4 and 5 were used for further evaluation for creating electrospun fibres.

7.6 ELECTROSPINNING MPC-HPMA-TMSPMA

7.6.1 Scanning Electron Microscopy Imaging of Deep Vein

Valves

Background

There are many techniques that can be employed to vary the microstructure of a scaffold, including electrospinning and micropatterning. If the native valve structure itself can be mimicked then this may reduce the thrombogenicity of the material by reducing flow variation as blood passes through the valve, preventing platelet adhesion or stasis of blood. If the microstructure is known, this may guide

220 manufacturing techniques. This study aims to determine the underlying microstructure of a deep vein valve using scanning electron microscopy techniques.

Methods

Fresh bovine veins were obtained from a 2 whole calves aged 2 weeks and 6 weeks harvested and processed within 12 hours of death. The calves had died of natural causes; one due to a chest infection and one having been trampled by its mother. 4 legs were removed from the calf at Lye Cross Farm,

Bristol, packed in ice and returned to the laboratory. 10 cm of deep vein were harvested from each of the

4 legs. Aseptic technique was used and a 2-0 braided suture was used to tie all tributaries. Samples were stored in phosphate-buffered saline (PBS) at 4° C for 2 hours. The vein was opened along its length so the valves inside the lumen were exposed. The vein was cut so that only the segment containing the leaflet remained. Each sample was approximately 9mm x 4mm in size. The samples were fixed in 4%

Formalin for 1 hour. 6 concentrations of ethanol were prepared by diluting by volume 100% ethanol with deionised water, to allow serial dehydration of the samples. Each of the samples was placed sequentially for 1 hour, in 30%, 45%, 60%, 75%, 90% and 100% ethanol. Finally each sample was placed in HMDS for 1 hour, then left to dry in the open air for 48 hours. They were then stored in uncoated tissue culture well plates. Scanning Electron Microscopy was then carried out of the dried samples.

Results

All valves seen were macroscopically bicuspid. An example of the images obtained of the luminal surface of the valves is shown in Figure 77. The surface appears to have a fibrous structure with the fibres running parallel to the direction of flow of blood. A few red blood cells can be seen on the surface of the valve, as they were not all washed off at the time of preparation.

221

Figure 77: Shows the scanning electron microscopy image of a bovine vein valve (luminal surface)

5µm

The sizes of the fibres were measured on the x3000 images using ImageJ software. The mean size of the fibres was 1.13µm, the measurements are shown in Table 22 below.

222

Table 22: Shows the diameter of fibres seen on the luminal surface of a bovine vein valve

Sample number Fibre diameter (µm)

1 1.305

2 1.092

3 1.331

4 0.984

5 0.946

6 0.941

7 1.421

8 1.083

9 1.114

10 1.068

7.6.2 Optimisation of Electrospinning the Polymer

Background

I have already demonstrated the microstructure of the valve suggesting that there is an alignment of a fibrous structure. Perhaps this is generated by the flow of blood. By simulating this with a material it

223 may be possible to optimise cell growth and the haemodynamics. The process of electrospinning generates fibres which can be aligned in a similar fashion to recreate this micropatterning structure.

Electrospinning is a process by which a voltage is applied across a polymer in solution. The polymer, in a syringe, is expelled using a syringe pump, and a positive charge is applied across the needle. A negatively charged collecting plate is placed at a set distance from the needle. Once the voltage is applied, the solution is drawn from the tip of the needle towards the collecting plate to make a fibre. This is shown in

Figure 78 below.

Figure 78: Shows a diagrammatic representation of the electrospinning set up

Methods

The polymer solution made by free radical polymerisation is dehydrated using a Rotary Evaporator or

“Rotovap” to a known concentration of solution ready for electrospinning. A concentration of solution which yields consistent fibres of approximately 1.1µm in diameter without beading is required. The variables in the electrospinning process include; the solvent, the concentration of solution, the needle

224

gauge, the distance of collection plate from needle tip, the rate of syringe pump, and voltage applied

across the needle are listed in the tables below. Only the linear version of the polymer was used as

previously discussed. The polymer was dissolved in solvent, to a known concentration. The solution was

then loaded into a syringe and onto a programmable syringe pump (KD Scientific model KDS 100,

Sandback, Cheshire), which was set to expel a known volume of solution per hour. A hypodermic needle

of known gauge, which was varied to change the fibre size, was attached to the syringe. The voltage

power supply (Glassman high voltage power supply series WR, Bramley, Hampshire) was connected to

the needle. The collecting plate can be in the form of either a plate, or a rotating mandrill, but for the

optimisation process a static metallic plate to collect randomly aligned fibre scaffolds. Electrospinning

was carried out under standard laboratory conditions of 20 ± 3 °C and relative humidity of 50 ± 5%.

Scanning electron microscopy images of each of the samples were then obtained and then the diameter of

the fibres was measured using Image J software as previously described.

Results

A viscous solution of polymer was obtained once 30-40% of the original weight of the polymer solution dissolved in methanol remained. It was not possible to completely evaporate all of the solvent to a dry polymer as it would then not dissolve back into solution. Further assessment between these values demonstrated that it was optimum to evaporate the solvent so that 35% of the initial weight of polymer and solvent was remaining. A lower concentration that this resulted in beading of the fibres and a higher concentration resulted in sputtering of the solution during the electrospinning process. This consistently provided a solution of polymer and solvent which could be used for the subsequent electrospinning process. The addition of other solvents to the polymer solution, including HFIP resulted in the solution becoming cloudy, therefore all further electrospinning was performed from anhydrous methanol only.

The variables that were changed for optimisation of the electrospinning process are shown in Table 23,

alongside the x3000 SEM images of the resulting electrospun polymer. All were performed with the plate

at a distance of 10cm from the needle tip and at the same concentration of polymer solution from

methanol.

225

Table 23: Shows the variables for electrospinning polymer in methanol

Needle (G) Rate (ml/h) Voltage (Kv) X3000 SEM image Mean fibre size

(µm)

16 1 10 0.566

16 1 12 0.814

16 1 15 0.540

16 1 20 0.676

16 2 10 0.696

226

Needle (G) Rate (ml/h) Voltage (Kv) X3000 SEM image Mean fibre size

(µm)

16 2 12 0.955

16 2 15 0.825

16 2 20 0.752

19 1 10 0.478

19 1 12 0.515

9 1 15 0.696

227

Needle (G) Rate (ml/h) Voltage (Kv) X3000 SEM image Mean fibre size

(µm)

19 1 20 0.623

19 2 10 0.814

19 2 12 0.846

19 2 15 1.013

19 2 20 1.092

228

Fibres electrospun with all variables using a 23G needle showed marked beading and therefore are not

included in the table and the fibre sizes were not measured. An example of this beading is shown in

Figure 79.

Figure 79: Shows the MPC-HPMA-TMSPMA fibres electrospun with 23G needle showing beading

The variables which resulted in the fibres which were the most uniform and of mean fibre size 1.092µm used the variables: 19G needle, 10cm from the plate, with 20kV applied across the needle and the pump at a rate of 2ml/hour. These were used for all subsequent experiments.

7.6.3 Electrospinning a Bicuspid Valve Shape

A bicuspid valve shape that could be attached to the rotating mandrill for electrospinning was designed

according to the dimensions measured in Section 6.5. This was made by the Blackett Laboratory,

229

Imperial College. The design sketches for this are shown in Appendix 5. Using the parameters described above, the valve shape was spun, with the mandrill rotating at 100 rotations per minute. The resulting shape is shown in Figure 80.

Figure 80: Shows the electrospun bicuspid valve shape

7.7 HAEMOCOMPATIBILITY TESTING

7.7.1 Thrombogenicity Testing of MPC-HPMA-TMSPMA

Background

Materials which come into contact with the blood must be biocompatible, non thrombogenic and not cause haemolysis. Any material to be implanted into the vascular system must undergo tests to ensure it fulfils these criteria. Thrombin is an enzyme involved in the coagulation of blood and it can be measured

230

in laboratory conditions. This will provide information on the thrombogenicity of the material tested. In vivo, thrombin is rapidly inactivated as soon as it is captured in the fibrin mesh by antithrombin III, making it difficult to accurately measure. A thrombin generation assay (TGA) kit developed by

HaemoScan® determines the level of thrombin activity in the medium following exposure to the tested material and compares this to other materials, or reference materials, of known thrombogenicity399. This

part of the project aims to test the haemocompatibility of MPC-HPMA-TMSPMA, both MPC50-

HPMA15-TMSPMA35 and MPC50-HPMA5-TMSPMA45 and establish whether the length of time the material was left in the crosslinking solvent TEA and subsequently in water had any effect.

Methods

The electrospun scaffold of MPC50-HPMA15-TMSPMA35 and MPC50-HPMA5-TMSPMA45 were cross linked to form the Si backbone to ensure insolubility. This was done by placing the electrospun samples in TEA for 1hour, 24 hours or 48 hours. These samples were then either not placed in water, placed in water for 1 day or for 7 days. The hydrophilic monomers should over time migrate to the surface of the electrospun fibres. The samples under the various conditions are described in Table 24.

Thrombogenicity of the materials was tested using a thrombin generation assay (TGA), which is a plasma product that will enable the determination of thrombin activity in a sample after it has been exposed to a material. The concentration of thrombin is determined by enzymatic reaction with a thrombin-specific chromogenic substrate, which yields a yellow colored product proportional to the amount of thrombin.

This can be compared to other reference materials of known high and low thrombogenicity, in this case one type of metal and two types of polymers. A TGA kit was obtained from HaemoScan®.

In the first instance, a calibration curve was developed using known concentrations of thrombin dissolved in buffer solution. The materials to be tested were cut by hand to a size of 5mm x 5mm to ensure it was of the same size as the reference materials included in the kit.

The reference and test materials were placed in micro-centrifuge vials and fixed between the walls of the vials to prevent floating. An empty vial without material was used as a negative control. For each vial, 4

231

labelled micro-centrifuge vials were prepared and labelled SPL1-4. Into each 490µl of Buffer B was placed and the vials were placed on ice. 350µl of TGA plasma was added to each of the vials containing the reference material, material to be tested or the negative control vial ensuring that the materials were completely immersed in the plasma. The vials were then incubated for 15 minutes in a water bath at

37°C. At 11 minutes, a vial containing 450µl of TGA Reagent A was placed in the water bath at 37 °C and at 14 minutes, 50µl of TGA Reagent B was added to the vial containing TGA Reagent A and mixed.

At 15 minutes, 175µl of the mixture of TGA Reagent A and B was added to the vial containing plasma and mixed. At 16 minutes, a sample of 10µl was taken out of the reaction mixture after mixing it, and placed into the vial labelled SPL1, vortexed and store on ice. This sampling procedure was repeated at 2 minutes, 4 minutes and 6 minutes and placed into the vials labelled SPL2, SPL3 and SPL4, respectively.

All calibrators and diluted samples collected in Buffer B were then removed from the ice, vortexed and placed at room temperature. 150µl of each of the calibrator and diluted samples was then transferred to a

96-well plate and incubated for 2 minutes in an incubator at 37 °C. 50µl of diluted Substrate solution was

added to each well of the plate, the plate was covered with foil and incubated for 20 minutes at 37 °C.

50µl of Stop Solution was added to each well, and the optical density was measured using a plate reader at

405 nm, using 540 nm as a reference wavelength. The thrombin concentration of all diluted samples and

then the velocity of thrombin generation for each test material, reference material and negative control

were then calculated. The material samples tested, the length of time spent in TEA and the time left in

water subsequently are listed in Table 24.

232

Table 24: Shows the samples tested with TGA assay, material, time in TEA and time in water

Sample Number Material Time in TEA (hours) Time in water (days)

1 MPC50-HPMA15-TMSPMA35 1 0

2 MPC50-HPMA15-TMSPMA35 24 0

3 MPC50-HPMA15-TMSPMA35 48 0

4 MPC50-HPMA15-TMSPMA35 1 1

5 MPC50-HPMA15-TMSPMA35 24 1

6 MPC50-HPMA15-TMSPMA35 48 1

7 MPC50-HPMA15-TMSPMA35 1 7

8 MPC50-HPMA15-TMSPMA35 24 7

9 MPC50-HPMA15-TMSPMA35 48 7

10 MPC50-HPMA5-TMSPMA45 1 0

11 MPC50-HPMA5-TMSPMA45 24 0

12 MPC50-HPMA5-TMSPMA45 48 0

13 MPC50-HPMA5-TMSPMA45 1 1

14 MPC50-HPMA5-TMSPMA45 24 1

15 MPC50-HPMA5-TMSPMA45 48 1

16 MPC50-HPMA5-TMSPMA45 1 7

17 MPC50-HPMA5-TMSPMA45 24 7

18 MPC50-HPMA5-TMSPMA45 48 7

233

Results

The optical density (OD) was plotted for OD405, corrected for OD540 as this was used as the reference

wavelength, against the thrombin concentrations of the calibrators. The calibration curve is shown in

Figure 81. The resulting straight line indicates that the kit is working to detect increasing thrombin

concentrations. The correlation coefficient value was 0.9962, the slope 0.0037 and the intercept 0.0113.

This allowed the calculation of the thrombin concentration for each of the materials.

Figure 81: Shows the calibration curve of thrombin concentrations of the calibrators against the OD405

1.600 1.400 1.200

1.000 0.800 0.600 OD 405 nm 0.400 0.200 0.000 0 100 200 300 400 500

Thrombin (mU/mL)

At each time point, samples of each material were tested and the mean calculated. The mean thrombin concentrations for the reference material, the negative control, and the material samples at each of the time points 1,2,4 and 6 minutes is shown in Table 25, and represented by a line graph in Figure 82. The resulting curves demonstrate that MPC-HPMA-TMSPMA is comparable with the lowest thrombogenicity reference material sent with the kit, LDPE.

234

Table 25: Shows the thrombin concentrations at 1, 2, 4 and 6 minutes corrected for negative controls

Reference Materials Sample Number

MS 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 Incubation time time (min) Incubation Control Negative LDPE PDMS 1 27.9 0.0 7.5 -6.9 -6.6 -5.0 -6.2 -6.3 -5.6 -3.0 -1.7 -0.1 7.4 0.3 -3.6 -2.6 -17.3 -3.0 -3.0 -30.9 -3.1 -0.5

2 26.9 -4.6 3.4 48.0 -2.3 -2.8 -4.3 -4.3 -5.2 0.1 -0.3 0.4 -2.3 -1.1 1.5 -3.2 -16.7 19.2 -1.2 -30.0 3.9 -2.0

4 22.1 3.4 77.5 96.8 3.8 2.2 3.5 1.5 0.8 5.1 3.0 3.6 7.0 4.0 3.1 5.4 -10.2 27.0 3.1 -25.1 3.6 3.5

6 26.8 23.2 174 66.0 28.4 17.2 27.7 21.4 11.0 36.0 4.6 12.1 5.0 23.8 22.7 22.3 22.0 97.4 24.1 -29.8 18.8 12.4

235

Figure 82: Shows a line graph of thrombin concentrations at 1, 2, 4 and 6 minutes corrected for negative controls

200.0 LDPE PDMS MS 1 150.0 2 3 4 5

100.0 6 7 8 9 10 50.0 11

Thrombin (mU/mL) Thrombin 12 13 14 0.0 15 0 2 4 6 8 16 17 18 -50.0

Time (min)

The thrombin generation curve for each of the samples is shown in Figure 82, by plotting the mean thrombin concentration at each sampling time points 1,2,4 and 6 minutes, corrected for the thrombin concentration of the negative control. The highest velocity of thrombin generation in between any 2 measured time points, that is the point of the steepest slope on the curve, was calculated and this was corrected for the dilution factor of the assay, and hence the thrombin generation for each material and reference material was calculated. No correction for surface area was required as all materials tested has a surface area of 1cm2. The thrombin generation for each of the samples and reference materials is shown

in Table 26 and Figure 83. The sample numbers correspond to those shown in Table 24.

236

Table 26: Shows the thrombin generation of the reference samples and sample materials

Sample Surface Area Thrombin Generation Corrected Thrombin (mU/ml/min) Generation (mU/ml/min/cm2)

LDPE 1.00 497 497

PDMS 1.00 2402 2402

MS 1.00 2741 2741

1 1.00 615 615

2 1.00 376 376

3 1.00 605 605

4 1.00 497 497

5 1.00 255 255

6 1.00 773 773

7 1.00 81 81

8 1.00 212 212

9 1.00 232 232

10 1.00 494 494

11 1.00 490 490

12 1.00 423 423

13 1.00 806 806

14 1.00 1760 1760

15 1.00 524 524

16 1.00 121 121

17 1.00 380 380

18 1.00 222 222

237

Figure 83: Shows the thrombin generation of reference samples and materials

3000.00 ) ) 2 2500.00

2000.00

1500.00

1000.00

500.00 Thrombin (mU/mL/min/cm Generation Thrombin

0.00 LDPEPDMS MS 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 Reference Material and Sample Number

Further comparisons were then made between the thrombin generation of the reference materials and all samples of MPC50-HPMA15-TMSPMA35 and MPC50-HPMA5-TMSPMA45. The error bars show the

95% confidence interval. For MPC50-HPMA15-TMSPMA35 the corrected mean thrombin generation was 405mU/ml/min/cm2 (SD 230.1, 95% CI; 228.2-581.9) and for MPC50-HPMA5-TMSPMA45 the corrected mean thrombin generation was 580mU/ml/min/cm2 (SD 483.1, 95% CI; 208.8-951.4). Using an unpaired t-test, there was no significant difference between these groups, p=0.3409. The material

MPC-HPMA-TMSPMA is less thrombogenic than the reference materials MS and PDMS, and comparable with the lowest thrombogenicity material in the assay LDPE. These results are shown in

Figure 84.

238

Figure 84: Shows the comparison of the thrombin generation of the reference materials and MPC50-HPMA15-

TMSPMA35 and MPC50-HPMA5-TMSPMA45

Comparison between the thrombin generation of the all samples of MPC-HPMA-TMSPMA depending on the time left for crosslinking in TEA, for 1 hour, 24 hours or 48 hours was performed. These results are shown in Figure 85. The error bars shown are the 95% confidence interval. For MPC-HPMA-

TMSPMA in TEA for 1 hour the corrected mean thrombin generation was 435.7mU/ml/min/cm2 (SD

283.3, 95% CI; 138.4-733.0), for MPC-HPMA-TMSPMA in TEA for 24 hours the corrected mean

thrombin generation was 578.8mU/ml/min/cm2 (SD 587.1, 95% CI; -37.24-1195) and for MPC-HPMA-

TMSPMA in TEA for 48 hours the corrected mean thrombin generation was 463.2mU/ml/min/cm2 (SD

215.9, 95% CI; 236.6-689.7). Using an ANOVA one-way analysis of variance to compare the three groups, there was no significant difference between these groups, p=0.8048.

239

Figure 85: Shows the comparison of the thrombin generation of MPC-HPMA-TMSPMA cross-linked in TEA for 1

hour, 24 hours and 48 hours

Comparison between the thrombin generation of the all samples of MPC-HPMA-TMSPMA depending on the time left in water following crosslinking in TEA, for 0 days, 1 day or 7 days was performed. These results are shown in Figure 86. The error bars shown are the 95% confidence interval. For MPC-

HPMA-TMSPMA not placed in water the corrected mean thrombin generation was

500.5mU/ml/min/cm2 (SD 95.59, 95% CI; 400.2-600.8), for MPC-HPMA-TMSPMA in water for 1 day the corrected mean thrombin generation was 769.2mU/ml/min/cm2 (SD 525.7, 95% CI; 217.5-1321) and for MPC-HPMA-TMSPMA in water for 7 days the corrected mean thrombin generation was

208.0mU/ml/min/cm2 (SD 104.0, 95% CI; 98.87-317.1). Using an ANOVA one-way analysis of variance to compare the three groups, there was a significant difference between these groups, p=0.0247, and using the unpaired t-test to compare each of the groups, there was a significant difference between

MPC-HPMA-TMSPMA in no water and in water for 7 days (p=0.0005) and between MPC-HPMA-

TMSPMA in water for 1 day and in water for 7 days (p=0.0281).

240

Figure 86: Shows the comparison of the thrombin generation of MPC-HPMA-TMSPMA left in water for 0 days, 1 day or 7 days after cross-linking in TEA

7.7.2 Complement Convertase Assay

Background

Materials that have contact with blood will activate the complement system, which is part of the innate immune system. Activation of complement proteins amplifies the immune response which can have a detrimental effect to the patient. It results in the attraction of white blood cells, which release proteases than cleave proteins to produce cytokines, which in turn lead to an exaggerated inflammatory response within tissue. Any foreign material implanted into the body can cause instability of the enzymes and uncontrolled activation resulting in a foreign body reaction. Therefore, testing of complement convertase activity by materials used for the construction of medical devices is needed to ensure the use of materials with as low complement activation as possible.

241

Measurement of complement convertase determines the complement activation by biomaterials. By incubating the material with plasma, complement factors which bind the surface of the material result in the formation of a complement convertase complex. This can cleave a chromogenic substrate, which in turn can be quantified by measuring the optical density. This can be compared to a negative control and reference materials.

Methods

A Complement Convertase Assay Kit was obtained from HaemoScan®, The Netherlands. Samples were prepared as described in Table 24 and cut to size 10x5mm. The reference and test materials were placed in micro-centrifuge vials using forceps, and fixed vertically between the walls of the vials to prevent floating. 650 µl of CCA plasma to each vial containing the material, and 650 µl of 0.9% sodium chloride solution was added to the negative control. The vials were incubated for 15 minutes at room temperature after which 1ml of wash buffer was added. The materials were then removed, blotted dry and each placed in separate vial containing 1.5ml of wash buffer. The materials were then placed in micro- centrifuge vials containing 0.5ml of substrate solution and incubated for 24 hours at room temperature.

The rate of colour development depends on how much convertase has been generated on the surface of the biomaterial. 225 µl from each vial was transferred to a 96-well plate and the optical density read with a plate reader at 405 nm (OD405).

Results

CCA results are described in OD45/24h/cm2 and are corrected for the negative control as shown in

Table 27. CCA results are classified as Inactive: OD405/24 h/cm2 ≤ 0.06, Low: 0.06 < OD405/24 h/cm2 ≤ 0.20, Medium: 0.20 < OD405/24 h/cm2 ≤ 0.60, High: OD405/24 h/cm2 > 0.60.

242

Table 27: Shows the CCA results for reference materials and sample materials

Sample Mean CCA Result Corrected CCA Result

(OD405/24h/cm2) (OD405/24h/cm2)

LDPE 0.393 0.320

PDMS 0.792 0.719

Negative Control 0.073 -

1 0.744 0.671

2 0.643 0.570

3 0.598 0.525

4 0.652 0.579

5 0.629 0.556

6 0.616 0.543

7 0.713 0.64

8 0.670 0.597

9 0.613 0.540

10 0.664 0.591

11 0.667 0.594

12 0.719 0.646

13 0.578 0.505

14 0.649 0.576

15 0.679 0.606

16 0.690 0.617

17 0.718 0.645

18 0.733 0.660

243

Comparisons were made between the CCA results of the reference materials and all samples of MPC50-

HPMA15-TMSPMA35 and MPC50-HPMA5-TMSPMA45. The error bars show the 95% confidence interval. For MPC50-HPMA15-TMSPMA35 the corrected mean CCA result was 0.580 OD405/24 h/cm2 (SD 0.081, 95% CI; 0.540-0.620) and for MPC50-HPMA5-TMSPMA45 the corrected mean CCA result was 0.610 OD405/24 h/cm2 (SD 0.0760, 95% CI; 0.571-0.650). Using a one-way ANOVA test

there was a significant difference between the columns and with Bonferroni’s Multiple Comparison Test,

this reached significance p<0.05 for LDPE vs. MPC50-HPMA15-TMSPMA35 and LDPE vs. MPC50-

HPMA5-TMSPMA45. There was no significant difference between PDMS and the sample materials.

Using an unpaired t-test, there was no significant difference between MPC50-HPMA15-TMSPMA35 and

MPC50-HPMA5-TMSPMA45, p=0.266. The material MPC-HPMA-TMSPMA has no significant difference in complement activation to the reference material PDMS and has a medium complement convertase activation overall according to this assay. These results are shown in Figure 87.

Figure 87: Shows the comparison of the CCA results of the reference materials and MPC50-HPMA15-TMSPMA35

and MPC50-HPMA5-TMSPMA45

244

Comparison between the CCA results of the all samples of MPC-HPMA-TMSPMA depending on the time left for crosslinking in TEA, for 1 hour, 24 hours or 48 hours was performed. These results are shown in Figure 88. The error bars shown are the 95% confidence interval. For MPC-HPMA-TMSPMA in TEA for 1 hour the corrected CCA results were 0.609 OD405/24 h/cm2 (SD 0.070, 95% CI; 0.562-

0.656), for MPC-HPMA-TMSPMA in TEA for 24 hours the corrected CCA results were 0.589

OD405/24 h/cm2 (SD 0.073, 95% CI; 0.5473-0.636) and for MPC-HPMA-TMSPMA in TEA for 48 hours the mean corrected CCA results were 0.586 OD405/24 h/cm2 (SD 0.096, 95% CI; 0.526-0.647).

Using an ANOVA one-way analysis of variance to compare the three groups, there was no significant difference between these groups, p=0.771.

Figure 88: Shows the comparison of the CCA results of MPC-HPMA-TMSPMA cross-linked in TEA for 1 hour, 24

hours and 48 hours

245

Comparison between the CCA results of the all samples of MPC-HPMA-TMSPMA depending on the time left in water for 0 days, 1 day or 7 days following crosslinking in TEA, was performed. These results are shown in Figure 89. The error bars shown are the 95% confidence interval. For MPC-HPMA-

TMSPMA not placed in water the corrected mean CCA results were 0.599 OD405/24 h/cm2 (SD 0.067,

95% CI; 0.557-0.642), for MPC-HPMA-TMSPMA in water for 1 day the corrected mean CCA results were 0.566 OD405/24 h/cm2 (SD 0.079, 95% CI; 0.512-0.619) and for MPC-HPMA-TMSPMA in water for 7 days the corrected mean CCA results were 0.616 OD405/24 h/cm2 (SD 0.088, 95% CI; 0.561-

0.672). Using an ANOVA one-way analysis of variance to compare the three groups, there was no significant difference between these groups, p=0.303.

Figure 89: Shows the comparison of the CCA results of MPC-HPMA-TMSPMA left in water for 0 days, 1 day or 7

days after cross-linking in TEA

246

7.7.3 Haemolytic Assay

Background

Erythrocytes that come into contact with toxins, metal ions and materials can haemolyse, therefore the

haemolytic activity of materials that will come into contact with blood is essential, particularly when the

contact will be prolonged as is the case with a venous valve. Excess haemolysis would induce anaemia

and the release of iron. Optimal conditions to perform a haemolysis assay have been described400

previously using diluted human blood in an erythrocyte suspension incubated with the test material.

Methods

A biomaterial haemolytic assay was obtained from Haemoscan®, The Netherlands. Samples were

prepared as described in Table 24 and cut to size 5x5mm, giving a total surface area of 0.5cm2. 5ml of

was buffer was added to the erythrocyte suspension and mixed, then centrifuged for 10 minutes at 400g.

The supernatant was removed and the procedure repeated. The same wash procedure was performed

with 5ml of the dilution buffer. The final pellet was then resuspended in 5ml of the dilution buffer. The

test materials were placed in a 2ml syringe and the erythrocyte suspension added and the air removed

from the syringe and sealed with parafilm. The negative control is a syringe without any material. The

haemoglobin calibrators were prepared according to the manufacturer’s instructions by diluting with lysis

fluid. The materials to be tested including the reference materials were placed in 2ml syringes and 0.5ml

of erythrocyte suspension was added to each, air expelled and the outlet closed with parafilm. 10µl of

erythrocyte suspension was added to 1ml of lysis fluid to ascertain the total haemoglobin concentration.

The vials were then incubated with end-over end rotating at 37°C for 24 hours. The erythrocyte suspension was then transferred to an eppendorf tube and centrifuged at 4000g for 1 minute. 20µl of supernatant from the sample was added to the well plate, and 180µl of assay buffer was added and mixed on a plate shaker. The OD was measured at 415 nm. The haemoglobin concentration for all test and

247

reference materials was corrected for the haemoglobin concentration of the negative control and

calculated per 1cm2.

Results

The mean haemoglobin concentrations are described in mg/cm2 and are corrected for the negative control as shown in Table 28.

248

Table 28: Shows the haemoglobin concentration for reference materials and sample materials

Sample Mean Haemoglobin Corrected Haemoglobin

Concentration (mg/cm2) Concentration (mg/cm2)

Silicon Elastomer 0.638 0.076

Buna-N 0.685 0.123

Negative Control 0.562 -

1 0.763 0.201

2 0.725 0.163

3 0.923 0.361

4 1.071 0.508

5 0.778 0.215

6 0.703 0.141

7 0.921 0.359

8 0.833 0.270

9 0.850 0.288

10 0.826 0.264

11 0.764 0.202

12 0.756 0.194

13 0.826 0.264

14 1.205 0.642

15 0.867 0.305

16 0.916 0.354

17 1.075 0.513

18 1.022 0.460

249

Comparisons were made between the haemoglobin concentrations of the reference materials and all

samples of MPC50-HPMA15-TMSPMA35 and MPC50-HPMA5-TMSPMA45, corrected for the negative control. The error bars show the 95% confidence interval. For MPC50-HPMA15-TMSPMA35 the corrected mean haemoglobin concentration was 0.281 mg/cm2 (SD 0.175, 95% CI; 0.194-0.367) and for

MPC50-HPMA5-TMSPMA45 the corrected mean haemoglobin concentration was 0.359 mg/cm2 (SD

0.184, 95% CI; 0.264-0.454). Using a one-way ANOVA test there was no significant difference between

the reference materials and sample materials, although significance was approached, p=0.080. Using an

unpaired t-test, there was no significant difference between MPC50-HPMA15-TMSPMA35 and MPC50-

HPMA5-TMSPMA45, p=0.206. These results are shown in Figure 90.

Figure 90: Shows the comparison of the haemoglobin concentration of the reference materials and MPC50-

HPMA15-TMSPMA35 and MPC50-HPMA5-TMSPMA45

250

Comparison between the haemoglobin concentration of the all samples of MPC-HPMA-TMSPMA depending on the time left for crosslinking in TEA, for 1 hour, 24 hours or 48 hours was performed.

These results are shown in Figure 91. The error bars shown are the 95% confidence interval. For MPC-

HPMA-TMSPMA in TEA for 1 hour the corrected mean haemoglobin concentration was 0.327 mg/cm2

(SD 0.177, 95% CI; 0.209-0.446), for MPC-HPMA-TMSPMA in TEA for 24 hours the corrected mean haemoglobin concentration was 0.337 mg/cm2 (SD 0.211, 95% CI; 0.203-0.470) and for MPC-HPMA-

TMSPMA in TEA for 48 hours the corrected mean haemoglobin concentration was 0.293 mg/cm2 (SD

0.166, 95% CI; 0.187-0.398). Using an ANOVA one-way analysis of variance to compare the three

groups, there was no significant difference between these groups, p=0.830.

Figure 91: Shows the comparison of the haemoglobin concentrations of MPC-HPMA-TMSPMA cross-linked in

TEA for 1 hour, 24 hours and 48 hours

251

Comparison between the haemoglobin concentrations of the all samples of MPC-HPMA-TMSPMA depending on the time left in water for 0 days, 1 day or 7 days following crosslinking in TEA, was performed. These results are shown in Figure 92. The error bars shown are the 95% confidence interval.

For MPC-HPMA-TMSPMA not placed in water the corrected mean haemoglobin concentration was

0.233 mg/cm2 (SD 0.163, 95% CI; 0.129-0.336), for MPC-HPMA-TMSPMA in water for 1 day the corrected mean haemoglobin concentration was 0.348 mg/cm2 (SD 0.214, 95% CI; 0.211-0.484) and for

MPC-HPMA-TMSPMA in water for 7 days the corrected mean haemoglobin concentration was 0.381 mg/cm2 (SD 0.133, 95% CI; 0.292-0.470). Using an ANOVA one-way analysis of variance to compare the three groups, there was no significant difference between these groups, p=0.114.

Figure 92: Shows the comparison of the haemoglobin concentrations of MPC-HPMA-TMSPMA left in water for 0

days, 1 day or 7 days after cross-linking in TEA

252

7.8 FOURIER TRANSFORM INFRARED SPECTROSCOPY

7.8.1 Background

Fourier Transform Infrared Spectroscopy (FTIR) is a method used to collect data on a solid, liquid or gas

regarding its functional groups and surface characteristics. The FTIR spectrometer produces a light in the

infrared spectrum and the sample placed in its path absorbs the light according to its chemical properties.

This data is collected by a detector and compared to a reference energy source. The measured signal is

sent to the computer and undergoes Fourier transformation, producing an infrared spectrum which can

be interpreted. In this part of the study, by using FTIR spectroscopy the aim is to gain further

information about MPC50-HPMA15-TMSPMA35 and MPC50-HPMA5-TMSPMA45 and the effects of cross-linking with TEA and effect of water on the samples.

7.8.2 Methods

Electrospun scaffolds of MPC50-HPMA15-TMSPMA35 and MPC50-HPMA5-TMSPMA45 were prepared as previously described. Samples were then placed in TEA for 1hour, 24 hours or 48 hours and placed in water for 1 day or for 7 days. The samples under the various conditions are described in Table

24. FTIR spectroscopy was performed using an FTIR Spectrometer, Bruker Optics. Spectra were collected and plotted as absorption.

7.8.3 Results

FTIR analysis was carried out on the electrospun scaffolds of the samples in Table 24. None of the

results are significant an therefore no graphs are included here, however, the images contribute towards

establishing the differences between MPC50-HPMA15-TMSPMA35 and MPC50-HPMA5-TMSPMA45

253 and the different amounts of time spent in TEA and water. The spectra obtained are shown in Figure 93 and Figure 94.

The peaks were assigned as follows:

• 1715 cm-1: C=O

• 1480 cm-1: CH2

• 1230 cm-1: C-O

• 1160 cm-1: C-O + C-C (backbone of the polymer)

• 1080 cm-1: O=P-O-

• 1050 cm-1: Si-O-Si

• 1020 cm-1: Si-O-CH3

• 960 cm-1: Si-O- (non- bridging oxygen)

• 925 cm-1: Si-OH (Silanol)

• 870 cm-1: Si-C

• 770 cm-1: CH3 (Rocking backbone of the polymer)

Figure 93: Shows the effect of TEA on MPC50-HPMA15-TMSPMA35 (A) and MPC50-HPMA5-TMSPMA45 (B)

A B

254

Figure 94: Shows the effect of water on MPC50-HPMA15-TMSPMA35 (A) and MPC50-HPMA5-TMSPMA45 (B)

A B

Immersion of MPC50-HPMA15-TMSPMA35 and MPC50-HPMA5-TMSPMA45 in TEA appears to result in a change in the spectra overt time, with more crosslinking of the Si-O-Si groups with increased time in TEA. At 48 hours there was more crosslinking of these groups which should result in a more stable polymer.

With regard to the immersion of the MPC50-HPMA15-TMSPMA35 and MPC50-HPMA5-TMSPMA45 that was cross-linked in TEA for 48 hours, it was expected that MPC50-HPMA5-TMSPMA45 would be more stable as there are more Si-O-SI groups, however, the converse was the case. MPC50-HPMA5-

TMSPMA45 demonstrated a desorption of TEA after 1 day in water, which was further demonstrated at

7 days, resulting in a disruption of the silica network as the peak Si-O-Si decreased and the Si-O- peak increase. For MPC50-HPMA15-TMSPMA35 there is an initial same desorption but the effect, but only of the Si-OH peaks, followed by a condensation of the silica network. This results in a scaffold which is more stable than before immersion in water.

255

7.9 MECHANICAL PROPERTY TESTING OF MATERIALS

7.9.1 Background

Mechanical properties of materials that will be implanted as part of a medical device must be considered.

Particularly for the case of a valve leaflet, the material should be flexible and have a similar strength to that of the native tissue. There are several ways of testing the mechanical properties of materials, and in the laboratory setting to test the stress-strain relationships of materials, a tensiometer is commonly used.

Several studies have been carried out to establish the behaviour of blood vessels under tension with a tensiometer. The majority of studies have assessed arteries, including carotid401, aorta402 and coronary arteries403, and Rossmann et al, 2010404, assessed jugular and bovine veins. However, there is a paucity of literature regarding the mechanical properties of the deep veins. This study aims to compare the mechanical properties of deep veins with those of the material MPC-HPMA-TMSPMA.

7.9.2 Methods

Bovine Vein

Fresh bovine veins were obtained from a whole calf aged 6 weeks harvested and processed within 12 hours of death. The calf had died of natural causes having been trampled by its mother.4 legs were removed from the calf at Lye Cross Farm, Bristol, packed in ice and returned to the laboratory. 10cm of superficial vein and 10 cm of deep vein were harvested from each of the 4 legs. Aseptic technique was used, 2-0 braided suture was used to ties all tributaries. Samples were stored in phosphate-buffered saline

(PBS) at 4° C overnight. The segments of vein were opened along their length to form a flat sheet

50mm in length. The cross-sectional height of each sample, that is the height of the vein wall was measured in 3 different places across the length of vein, using an electronic digital calliper, and a mean obtained. Samples were mounted in the Instron, Model 5540, testing machine with a 50N load cell at

256

crosshead speed of 10mm/min. 10mm at each end of the 50mm vein lengths was held fixed in the

machine, hence a testing length of approximately 30mm was used, then after slack was removed from the

sample, the gauge length was set. From the results obtained, the breakpoint stress was calculated by

observing the maximum load that could be tolerated and the modulus was calculated using the equation:

Modulus (E) = Stress (σ)

Strain (ε) where:

Stress (σ) = Force / N

Area / mm2

And:

Strain (ε) = Change in length (ΔL)

Initial length (Li)

This was then normalised to the mean of 3 cross sectional measurements for each segment of vein and

compared to the same mechanical property testing of the material MPC-HPMA-TMSPMA materials.

Electrospun MPC-HPMA-TMSPMA

Randomly aligned electrospun scaffolds of MPC50-HPMA15-TMSPMA35 and MPC50-HPMA5-

TMSPMA5 were made, as described in Section 7.6.2. These were mechanically tested in tension to failure using an Instron Model 5540 testing machine, equipped with a 50 N load cell at a crosshead speed of 10 mm/min. Samples were cut to size 50mm x 10mm. The cross-sectional height of each sample was measured in 3 different places across the length of the sample, using an electronic digital calliper, and a

257

mean obtained. Samples were mounted in the Instron, Model 5540, testing machine with a 50N load cell

at crosshead speed of 10mm/min. 10mm at each end of the 50mm samples was held fixed in the

machine, hence a testing length of approximately 30mm was used, then after slack was removed from the

sample, the gauge length was set. From the results obtained, the breakpoint stress and modulus was

calculated as described above and normalised to the mean of 3 cross sectional measurements.

7.9.3 Results

The mean tensile modulus of vein was 4.66 N/mm2 (SD 1.87, 95% CI; 1.69-7.63). This was compared with the tensile modulus of MPC50-HPMA15-TMSPMA35 and MPC50-HPMA5-TMSPMA45 normalised to the diameter of the samples. The results are shown in Figure 95. The error bars show the

95% confidence interval. For MPC50-HPMA15-TMSPMA35 the tensile modulus was 0.13 N/mm2 (SD

0.02, 95% CI; 0.10-0.16) and for MPC50-HPMA5-TMSPMA45 the tensile modulus was 0.12 N/mm2

(SD 0.06, 95% CI; 0.03-0.21). Using a one-way ANOVA test there was a significant difference between

the three groups, p=0.0003 and using an unpaired t-test, there was a significant difference between vein

and MPC50-HPMA15-TMSPMA35 (p=0.003) and vein and MPC50-HPMA5-TMSPMA45 (p=0.003).

Using an unpaired t-test, there was no significant difference between MPC50-HPMA15-TMSPMA35 and

MPC50-HPMA5-TMSPMA45, p=0.665.

258

Figure 95: Shows the tensile modulus of vein and MPC-HPMA-TMSPMA

The mean breakpoint stress of vein was 7.05 N/mm2 (SD 4.07, 95% CI; 0.58-8.33). This was compared with the breakpoint stress of MPC50-HPMA15-TMSPMA35 and MPC50-HPMA5-TMSPMA45 normalised to the diameter of the samples. The results are shown in Figure 96. The error bars show the

95% confidence interval. For MPC50-HPMA15-TMSPMA35 the mean breakpoint stress was 5.11

N/mm2 (SD 2.03, 95% CI; 1.88-8.33) and for MPC50-HPMA5-TMSPMA45 the mean breakpoint stress was 7.89 N/mm2 (SD 5.29, 95% CI; -0.53-16.30). Using a one-way ANOVA test there was no significant

difference between the three groups, p=0.621.

259

Figure 96: Breakpoint stress of vein and MPC-HPMA-TMSPMA

7.10 DISCUSSION

The venous system is a low flow, low velocity, system and therefore highly thrombogenic. Previous work on materials in other parts of the vascular system, such as for cardiac valves or vascular graft, can be utilised to guide further research in developing a material for use in the venous system. Autologous tissue remains the gold standard for grafts but if it is not available then an optimised prosthetic material with low thrombogenicity would be ideal. In the biological scaffold field ECM has been extensively studied362, with different properties363 depending upon the processing methods and origin of the ECM364.

260

There has been extensive work on the development of a cardiac valve using a polymer, stressing the need

for durability haemocompatibility and low thrombogenicity372. PEU, PCU, silicone, PTFE, Dacron,

PVC and PU have been extensively investigated373, as well as Silsesquioxane nanocomposite polymer377 and Poly(SIBS)378. Much of the work on vascular grafts has focussed on the development of grafts which

are pre-seeded with cells, such as epithelial cells382 prior to implantation. Polymers provide consistent,

reproducible material from which to manufacture the valve, in contrast to biological scaffolds which are

much more variable in their properties. ECM for example varies with the age of the source donor,

different methods of decelularisation, residual MMPs, elastin, collagen, growth factors and other

constituents. Once implanted, it is important to consider the effect cells, such as macrophages, cytokines

and other circulating factors from the recipient patient and the effect that these will have. When

developing a suitable polymer, its properties must be fully optimised and characterised to include;

thrombogenicity, haemocompatibility, biocompatibility, mechanical properties and the effect of the

material on host cells. Also, surface and structural modifications can improve the properties of the

polymer, ensuring that these do not adversely affect the other properties.

A material for use in the venous system as a valve should be flexible, non-biodegradable, be of low thrombogenicity and be biocompatible. Presently, polymers and polymer coated metal stents are

thrombogenic and they lack haemocompatibilty and biocompatibility. MPC has been demonstrated be

biocompatible and to have antifouling and antithrombogenic properties in previous studies388 and copolymers of MPC have been suggested to be beneficial for cell and tissue engineering389. Its

haemocompatibilty has been demonstrated in several studies392, 393 and its use reduces the

thrombogenicity of other polymers394 . This has been further demonstrated in this study.

2 polymers of MPC-HPMA-TMSPMA have been synthesised in this study by free radical polymerisation, and the ratios of HPMA and TMSPMA were varied for further experiments. The conversion of the monomers to polymer was >99% for all the linear versions of the polymer. Scanning electron microscopy images of bovine vein valves demonstrated a fibrous underlying microstructure with a mean

261

fibre diameter of 1.13µm. The material developed has tried to mimic the structure using the technique of

electrospinning which to form a fibrous scaffold. The linear versions of the polymer in solution were

then electrospun to make this fibrous scaffold, which underwent cross-linking in TEA. The variables

which resulted in the fibres which were the most uniform and of mean fibre size 1.092µm used the

variables: 19G needle, 10cm from the plate, with 20kV applied across the needle and the pump at a rate

of 2ml/hour. These variables were then used to create a bicuspid valve shape by electrospinning directly

onto a shaped mandrill.

The thrombogenicity, haemocompatibility and biocompatibility of the material were further tested.

Materials which come into contact with the blood must be biocompatible, non thrombogenic and not

cause haemolysis. The haemocompatibility of MPC-HPMA-TMSPMA, both MPC50-HPMA15-

TMSPMA35 and MPC50-HPMA5-TMSPMA45 was further tested. For MPC50-HPMA15-TMSPMA35 the corrected mean thrombin generation was 405mU/ml/min/cm2 (SD 230.1, 95% CI; 228.2-581.9) and for MPC50-HPMA5-TMSPMA45 the corrected mean thrombin generation was 580mU/ml/min/cm2

(SD 483.1, 95% CI; 208.8-951.4), but the difference is not significant, however, MPC-HPMA-TMSPMA is less thrombogenic than the reference materials MS and PDMS, and comparable with the lowest thrombogenicity material in the assay LDPE. There was no significant difference between the thrombin generation of the all samples of MPC-HPMA-TMSPMA depending on the time left for crosslinking in

TEA, for 1 hour, 24 hours or 48 hours. However, there was a significant difference between MPC-

HPMA-TMSPMA in no water and in water for 7 days (p=0.0005) and between MPC-HPMA-TMSPMA in water for 1 day and in water for 7 days (p=0.0281), suggesting that the more haemocompatible monomers may migrate to the surface of the fibres. The biocompatibility of the material MPC-HPMA-

TMSPMA, tested with a complement convertase assay demonstrated no significant difference in complement activation to the reference material PDMS and has a medium complement convertase activation overall according to this assay. Comparison between the CCA results of the all samples of

MPC-HPMA-TMSPMA depending on the time left for crosslinking in TEA, for 1 hour, 24 hours or 48 hours and the time left in water for 0 days, 1 day or 7 days following crosslinking in TEA showed no

262

significant difference between these groups. With regard to haemolysis of erythrocytes in contact with

MPC-HPMA-TMSPMA, there was no significant difference between the reference materials and sample

materials, and no significant difference between MPC50-HPMA15-TMSPMA35 and MPC50-HPMA5-

TMSPMA45. These studies demonstrate that the thrombogenicity, biocompatibility and haemocompatibility of MPC-HPMA-TMSPMA is good, and improves once the material is left in water for 7 days.

FTIR analysis of the electrospun scaffolds, although it did not give any statistically significant data, did demonstrate that immersion of both MPC50-HPMA15-TMSPMA35 and MPC50-HPMA5-TMSPMA45 in TEA appears to result in a change in the spectra overt time, with more crosslinking of the Si-O-Si groups at 48 hours which should result in a more stable polymer. Immersing cross-linked MPC50-

HPMA5-TMSPMA45 in water, resulted in a decomposition of the silica network upon immersion, but in the case of the MPC50-HPMA15-TMSPMA35 the immersion in water actually reinforced the silica network, resulting in increased stability.

The mean tensile modulus of vein was 4.66 N/mm2 (SD 1.87, 95% CI; 1.69-7.63), compared to 0.13

N/mm2 (SD 0.02, 95% CI; 0.10-0.16) for MPC50-HPMA15-TMSPMA35 and 0.12 N/mm2 (SD 0.06,

95% CI; 0.03-0.21) for MPC50-HPMA5-TMSPMA45. The mean breakpoint stress of vein was 7.05

N/mm2 (SD 4.07, 95% CI; 0.58-8.33), compared with 5.11 N/mm2 (SD 2.03, 95% CI; 1.88-8.33) for

MPC50-HPMA15-TMSPMA35 and 7.89 N/mm2 (SD 5.29, 95% CI; -0.53-16.30) for MPC50-HPMA5-

TMSPMA45. The tensile modulus of vein is different from the material tested, howeer, the breakpoit stress is consistently in the correct range. Vein has a relatively low stiffness but high upper limit of tensile strength.

263

8. DEVELOPMENT OF A LABORATORY

VENOUS FLOW MODEL

8.1 INTRODUCTION

A laboratory model of deep venous function, in the form of a flow rig, would be a useful dynamic

assessment rig for the simulation of deep venous flow. Once developed, the conduit to be evaluated

could be easily assessed by ultrasound in order to obtain detailed flow dynamic measurements to compare

to in-vivo observations, or to provide further information on the behaviour of a valve and its longevity.

Although there are many in vitro models of arterial flow, arterialised great saphenous vein and models of venous thrombosis, there is a paucity of literature regarding lower limb venous flow models. In vitro models of venous flow have been used in previous studies. Wang et al, 1992405 used a flow loop which functioned by adjusting the height of the head tank and outflow chamber with a sinusoidal pressure pump to simulate respiratory function, but this was not a validated model. There is no validated in vitro flow model in the literature of lower limb venous flow.

This part of the study aims to develop and validate a laboratory flow model of venous flow.

8.2 METHODS

8.2.1 Set up of Laboratory Flow Model

Flow Meter Calibration

Before setting up the complete flow model, the various components were calibrated to ensure that the

output voltages received by the computer could be translated to relevant flow and pressure data. The

flow meter, Transonic TS410, was connected to the MCP-Z Standard gear pump, Ismatec with Z-120

264 pump head, Ismatec. Both the pump and the flow meter were connected to a PC via the Data

Acquisition System (DAQ), National Instruments, through Labview software, to control the pump and receive voltage information from the flow meter. The rate of flow was measured manually in millilitres per minute, by collecting the liquid passing through the pump and flow meter for 1 minute, using a measuring flask. The flow medium used was water.

Pressure Gauge Calibration

Two compound pressure transducers, Ashcroft G2, Cole-Parmer, UK were used. These were calibrated manually. A vertical column containing a known height of water was attached above the pressure transducer. The height of the column of water was gradually increased in a stepwise manner from 5cm to

135cm, in 5cm increments. The voltage reading from each pressure transducer was detected by the computer via the Data Acquisition System (DAQ), National Instruments, through Labview software.

Laboratory Flow Model

The laboratory flow model was set up as shown schematically in Figure 97, below.

265

Figure 97: Shows a schematic diagram of the laboratory flow model: Data Acquisition Assistant (DAQ);

Pressure Gauge (PG); Ultrasound (US)

Direction of Flow

In summary an MCP-Z Standard gear pump, Ismatec with Z-120 pump head, Ismatec, was connected to a reservoir containing water by 1/4" silicon tubing. The pump was controlled using Labview software and Data Acquisition System (DAQ), National Instruments. An example of the DAQ assistant is shown in Figure 98.

266

Figure 98: Shows an example of the DAQ assistant system which controls and collects data from the model

The flow of water then passed through ¼” silicon tubing passing through the first pressure gauge,

Ashcroft G2, Cole-Parmer, the flow meter, Transonic TS410, then the second pressure gauge, Ashcroft

G2, Cole-Parmer. This then passed through a pulse dampener. This was made by drilling three holes into a rubber bung fitted tightly on a glass beaker. A 1/8" to 1/4" male luer adaptor attached to the silicon tubing was inserted through the bung. A second 1/8" to 1/4" male luer adaptor attached to another length of silicon tubing was inserted through another hole in the bung. The tubing inside the dampener reached the bottom of the liquid. The third hole in the bung is an air vent. The tubing leaving the dampener then passes to the reservoir, thus completing the flow circuit. The actual laboratory set up is shown below in Figure 99.

267

Figure 99: Shows the flow model set up in the laboratory

268

8.2.2 Validation of Laboratory Flow Model

The velocity data obtained by ultrasound for each subject, described in Section 6.3 was used to validate

the flow model. The MCP-Z Standard gear pump, Ismatec with Z-120 pump head, Ismatec, was calibrated so that the input voltage, RPM and flow rate were known. Subject specific simulations were performed. The velocities obtained from the ultrasound images, those taken distal to the valves were used, and the voltage input to the gear pump calculated. As such an input voltage which changed with time was given to the pump via the DAQ assistant. An example of the voltage input in graphical form is shown in Figure 100.

Figure 100: Shows an example of the voltage input in graphical form. This represents subject 4, in the standing

position with a calf squeeze and breath hold.

3.50 3.00 2.50

2.00 1.50

Voltage (V) 1.00 0.50 0.00

-0.50 0.01 0.23 0.49 0.71 0.89 1.09 1.27 1.47 1.75 2.04 2.24 2.42 2.61 2.85 3.12 3.37 3.61 3.84 4.02 4.23 4.45 4.68 4.85 4.98 5.11 5.23 5.37 5.51 5.73 6.02 6.19 6.36 6.58 Time (s)

269

The actual flow velocity in the segment of the flow model was measured using a CX50 Philips ultrasound

machine with a linear transducer with venous protocol. The pressure gauges and flow meter collected

data regarding the pressure and flow velocities in the subject specific models. The voltage readings from

both the pressure gauges and the flow meter were detected by the computer via the Data Acquisition

System (DAQ), National Instruments, through Labview software. These measured readings were then

compared to the actual readings from the original subjects.

8.3 RESULTS

8.3.1 Flow Meter Calibration

The flow meter, Transonic TS410, was calibrated to ensure that the output voltages received by the

computer could be translated to the relevant flow data. The RPM of the gear pump was generated by an

input voltage from the computer through Labview software. The voltage from the flow meter was

received via the Data Acquisition System (DAQ), National Instruments, through Labview software. And

the rate of flow was measured manually in millilitres per minute. These values were compared as shown

in Figure 101 and Figure 102 below. The derived equations that were subsequently inputted to the

labview software to calculate the measured velocity data are shown in each figure. The R2 values for all 3 measurements were >0.97 demonstrating consistent data readings.

270

Figure 101: Shows the flow meter calibration results comparing the pump RPM and the measured flow rate in ml/min

600 y = 0.3723x + 190.96 500 R² = 0.9727

400

300

200

100 Measured flow rate (ml/min) 0 0 200 400 600 800 1000 RPM

Figure 102: Shows the flow meter calibration results comparing the measured flow rate in ml/min and the RPM of output voltage of the flow meter

600 y = 684.25x - 32.705 500 R² = 0.9753

400

300

200

100 Measured flow rate (ml/min) 0 0 0.2 0.4 0.6 0.8 1 Voltage (V)

271

8.3.2 Pressure Gauge Calibration

The pressure gauge, Ashcroft G2, Cole-Parmer, was calibrated to ensure that the output voltages received

by the computer could be translated to the relevant pressure data. The vertical column containing a

known height of water, attached above the pressure transducer, was measured. This was compared to the

the voltage reading from each pressure transducer, detected by the computer via the Data Acquisition

System (DAQ), National Instruments, through Labview software. These values were compared as shown

in Figure 103 and Figure 104 below. The derived equations that were subsequently inputted to the

labview software to calculate the measured velocity data are shown in each figure. The R2 values for the measurements were >0.99 demonstrating consistent data readings.

Figure 103: Shows the calibration results for Pressure Gauge 1 showing the output voltage compared to the known

pressure generated by the column of water

160 140 y = 783.84x - 720.28

R² = 0.9998 120 100 80 60

Pressure (cmH2O) 40 20 0 0.9 0.95 1 1.05 1.1 Voltage

272

Figure 104: Shows the calibration results for Pressure Gauge 2 showing the output voltage compared to the known

pressure generated by the column of water

160 140 y = 780.49x - 709.65

R² = 0.9998 120 100 80 60

Pressure (cmH2O) 40 20 0 0.9 0.95 1 1.05 1.1 Voltage (V)

8.3.3 Validation of the Laboratory Flow Model

The velocity data obtained by ultrasound for each subject, described in Section 6.3 was used to validate

the flow model. The MCP-Z Standard gear pump, Ismatec with Z-120 pump head, Ismatec, was calibrated so that the input voltage, RPM and flow rate were known. These results are shown in Figure

105.

273

Figure 105: Shows the calibration of the flow meter for voltage and RPM

0.5 y = 0.0008x + 0.0059 0.45 R² = 0.9999 0.4

0.35 0.3 0.25 0.2 0.15 Flow Meter RPM 0.1 0.05 0 0 100 200 300 400 500 600 Voltage

The subject specific simulations were performed. The actual flow velocity in the segment of the flow

model was measured using a CX50 Philips ultrasound machine with a linear transducer with venous

protocol. These were repeated five times by one observer. The coefficient of variation for the maximum

velocity measurements was 5.51% and for the mean velocity measurements was 6.67%. These measured

readings were then compared to the actual readings from the original subjects. These have been

compared with the data obtained from patients by Bland-Altman analysis. Figure 106: Shows a Bland-

Altman analysis of maximum velocity measurements comparing the ultrasound data obtained from the subjects and the flow model demonstrating a bias of -25 (SD 34) and 95% limits of agreement from -92 to 42. Figure 107: Shows a Bland-Altman analysis of the mean velocity during the calf squeeze measurements comparing the ultrasound data obtained from the subjects and the flow model demonstrating a bias of -12.34 (SD 23) and 95% limits of agreement from -57.41 to 23. Figure 108:

Shows a Bland-Altman analysis of the mean velocity measurements comparing the ultrasound data obtained from the subjects and the flow model demonstrating a bias of -8.2 (SD 12) and 95% limits of agreement from -32 to 16.

274

Figure 106: Shows a Bland-Altman analysis of maximum velocity measurements comparing the ultrasound data obtained from the subjects and the flow model

Figure 107: Shows a Bland-Altman analysis of the mean velocity during the calf squeeze measurements comparing the ultrasound data obtained from the subjects and the flow model

275

Figure 108: Shows a Bland-Altman analysis of the mean velocity measurements comparing the ultrasound data

obtained from the subjects and the flow model

Figure 109: Shows a graph of the maximum velocity from the subject data and the flow model showing the mean and 95% confidence intervals demonstrating a Spearman correlation r=0.7, with p<0.0001 reaching statistical significance. This demonstrates that the agreement between the data obtained experimentally from subjects and frm the flow model regarding maximum velocity is good.

276

Figure 109: Shows a graph of the maximum velocity from the subject data and the flow model showing the mean and 95% confidence intervals demonstrating a Spearman correlation r=0.7

Figure 110: Shows a graph of the mean velocity from the subject data and the flow model showing the mean and 95% confidence intervals demonstrating a Spearman correlation r=0.38, with p=0.04 reaching statistical significance. This demonstrates that there is agreement between the data obtained experimentally from subjects and from the flow model, however this is less than the maximum velocity.

277

Figure 110: Shows a graph of the mean velocity from the subject data and the flow model showing the mean and

95% confidence intervals demonstrating a Spearman correlation r=0.38

8.3.4 Discussion

Using readily available flow generators and variables obtained from in Chapter 6, a rig for the simulation of deep venous flow has been been developed. It includes a segment into which the conduit or implant to be tested may be incorporated into the circuit. An image of the valve in the flow model is shown in the discussion regarding future work. The subject specific simulations performed showed good consistency for the observer obtaiing the ultrasound images from the flow model. When the measurements obtained from ultrasound of the subjects were then compared to the readings taken from the flow model, it was demonstrated that there was good agreement for the maximum and mean velocities, but there was a consistent bias whereby the data from the flow model seemed to overestimate the velocity when compared to the subject data. This may be an inhernet error in the system as in the subjects, the velocity of blood was occasionally 0 cm/s, however, in our flow model, the minimum input voltage to the pump was 0.06 which generated slow flow in the model and therefore the velocity was never 0cm/s. The

278 accuracy of the US is limited as two different US machines were used. The patient data was obtained using a Phillips IU22, however, it was not possible to transport this to the laboratory and the smaller more mobile CX50 Philips ultrasound machine was used. The standardised venous protocols on these machine were used, which may differ between the machines. The operators were differwnt in both cases, which can lead to error. In addition the readings taken by the flow model were taken directly from the tubing in the flow model. In patients there are layers of skin, fat and muscle, the varying depth of which can have an impact on the results.

This study has developed a laborator model of deep venous flow which has been validated using ultrasound and compared to subject specific data from subjects with normal deep venous function.

279

9. FUTURE WORK

At each stage of the project there is further work that could be considered, and the results have

demonstrated new avenues for further research.

9.1 IMAGING OF THE VENOUS SYSTEM

This study has used normal subjects to model deep venous function. This project could be repeated with

patients with deep venous disease. The optimised US and MRI protocols could be repeated and used in

the laboratory flow model. This information could be used in the computational model and flow model

to give further information about the management and even the pathophysiology of venous disease. In

addition, it was observed that the background venous flow when neither a calf squeeze nor respiration

were playing a part, was pulsatile. This will be further studied in the literature and from the imaging

performed in this study.

9.2 DEVELOPMENT OF A COMPUTATIONAL MODEL

OF VENOUS FLOW

A computational flow model of deep vein valve function has been completed but further work is needed

to develop a 3-dimensional model which will account for movements of the vein wall. Further experiments can then begin by varying the valve structure and function of valves to assess the effect on

flow dynamics. Further collaboration will begin with Professor Jean-Francois Uhl, Paris Descartes

University, France, who has extensive experience in CT venography, and utilising this for 3-Dimensional modelling of the venous system. There is potential to combine the two models for further work. The computational model developed can be extended to incorporate patients with abnormal deep veins to develop a “typical” model of the femoral vein with an incompetent valve to simulate abnormal venous

280

function. Further patient-specific simulations could then be performed and deep venous reflux will be

calculated and compared with ultrasound measurements for validation purpose. Differences in flow

characteristics between normal and abnormal veins will be assessed in order to identify haemodynamic

conditions that predispose to thrombus formation. This model will be developed into a 3-dimensional

model. Finally, prosthetic venous valves with a single cusp and double cusp leaflets will be built into a

“typical” model of the femoral vein. Numerical experiments will be performed by varying the valve

geometry (such as valve depth, leaflet to leaflet contact) and material properties and subsequently

different valve-stent designs will be analysed. The first step involves simulation of stent deployment in

deep veins using computational solid mechanics in order to predict the deformed configuration of the

veins following stent insertion. A number of parameters will be examined; these include stent type and

dimension, stent inflation pressure, and the way valve leaflets are attached to stent. The deformed

configurations following stent deployment will then be used in the second step where haemodynamics

analysis will be performed. The haemodynamic conditions identified as thrombus prone will be used to

evaluate different stent designs and identify an optimal stent design that is likely to minimise the risk of

thrombus formation can be determined. This will require further funding and expertise in computational

modelling to progress.

9.3 DEVELOPMENT OF A LABORATORY FLOW MODEL

The laboratory flow model can be further developed. For example if it could be validated in the vertical

position to incorporate the effects of gravity, this would further improve the study. The image below

shows the bicuspid valve of MPC-HPMA-TMSPMA in the flow model. However the fixation into the

model was problematic and further work is needed to improve its fixation into the model to allow it to be further investigated.

281

Figure 111: Shows the ultrasound image of the valve in the flow model

9.4 DEVELOPMENT OF MPC-HPMA-TMSPMA

The material development will continue as it may be possible to further optimise it. Its material properties could be improved, and its thrombogenicity may be further optimised, perhaps by incorporating a heparin into the material which would be slowly released over time. Extracellular matrix could also be incorporated to encourage cell growth on the material. Further cell studies are needed to ensure that the material can facilitate cell growth.

282

Once this has been completed, it is likely that industrial involvement will be necessary. This will be to

develop a valve stent device that can be deployed endovascularly. This can be tested initially in the

laboratory flow model and then animal studies can take place. Following this, cell studies will be carried out to ensure that the material will support cell growth.

9.5 POTENTIAL SECONDARY RESEARCH AVENUES

In view of the huge incidence of chronic venous disease and deep vein valve failure, the potential future

clinical impact of the work is significant. The development of a valve stent prosthesis as a minimally

invasive clinical treatment modality will provide a treatment where few therapeutic options are currently

available. There is immense potential to generate novel secondary research avenues as the subject of deep

venous disease remains under researched. Some potential secondary research projects are listed below:

• Identification of anatomical and other risk factors in patients with deep venous failure to develop

clinical management algorithms

• Adaptation of vein flow computer models to application in deep venous thrombosis

• Exploration of new venous and arterial applications of materials developed for use in the valve

stent prosthesis

283

SUMMARY OF THESIS ACHIEVEMENTS

PUBLICATIONS

Lane TRA, Moore HM, Franklin IJ, Davies AH. Retrograde inversion stripping as a complication of the

Clarivein mechanochemical venous ablation procedure. Ann R Coll Surg Engl. 2015;97(2):18-20

Rowland, SP, Dharmarajah B, Moore HM, Davies AH. IVC Filters for Prevention of Venous

Thromboembolism in Obese Patients undergoing Bariatric Surgery: A Systematic review. Annals of

Surgery. 2015;261(1):35-45

Sounderajah V, Moore HM, Thapar A, Lane TRA, Fox K, Franklin IJ, Davies AH. Acoustic reflectors are visible in the right heart during radiofrequency ablation of varicose veins. Phlebology. 2014 (Epub ahead of print)

Rowland SP, Dharmarajah B, Moore HM, Dharmarajah K, Davies AH. Venous Injuries in pediatric trauma: Systematic review of injuries and management. J Trauma Acute Care Surg. 2014;77(2):356-63

Varatharajan L, Willliams K, Moore HM, Davies AH. The effect of footplate neuromuscular electrical stimulation on venous and arterial haemodynamics. Phlebology: 2014 (Epub ahead of print)

Dharmarajah B, Lane TRA, Moore HM, Neumann HM, Wittens CH, Davies AH. The Future of

Phlebology in Europe. Phlebology: 2014;29(1 suppl):181-5

Moore HM, Lane TRA, Kelleher D, Franklin IJ, Davies AH. Cyanoacrylate Glue for the treatment of

Great Saphenous Vein Incompetence in the Anticoagulated Patient. Journal of Vascular Surgery: Venous and

Lymphatic Disorders. 2014, (Epub ahead of print)

Williams KJ, Ayekoloye O, Moore HM, Davies AH. The Calf Muscle Pump Revisited. Journal of

Vascular Surgery: Venous and Lymphatic Disorders. 2014, (Epub ahead of print)

William K, Moore HM, Davies AH. Haemodynamic changes with the use of neuromuscular electrical stimulation compared to intermittent pneumatic compression. Phlebology;2014 (Epub ahead of print)

284

Nghiem AZ, Rudarakanchana N, Moore HM, Davies AH. Percutaneous Pharmacomechanical

Thrombectomy for Acute Iliofemoral Deep Vein Thrombosis: a Suitability Study. Phlebology. 2014,

(Epub ahead of print)

Moore HM, Lane TRA, Franklin IJ, Davies AH. Retrograde mechanochemical ablation of the small

saphenous vein for the treatment of a venous ulcer. Vascular. 2014; 22(5):375-7

Moore HM, Williams KJ, Onida S, Davies AH. Chronic Deep Venous Valve Management in the

Future. Italian Journal of Vascular and Endovascular Surgery. 2014, in press.

Baldwin M, Moore HM, Rudarakanchana N, Gohel M, Davies AH. Post-thrombotic Syndrome: a

Clinical Review. Journal of Thrombosis and Haemostasis. 2013, 11(5):795-805.

Qureshi MI, Lane TRA, Moore HM, Franklin IJ, Davies AH Patterns of short saphenous vein incompetence. Phlebology: 2013, Suppl 1:47-50.

Moore HM, Lane TRA, Thapar A, Franklin IJ, Davies AH. The European Burden of Primary Varicose

Veins, Phlebology. 2013, Suppl 1:141-7.

Soosaninathan A, Moore HM, Gohel MS, Davies AH. Scoring systems for the post-thrombotic

syndrome. Journal of Vascular Surgery. 2013;57(1):254-61

Thapar A, Moore HM, Golden D, Davies AH. Peri-operative anticoagulation – bridging the gap.

Annals of the Royal College of Surgeons: 2012;94:142-5

Moore HM, Gohel M, Davies AH, The Number and Location of Deep Venous Valves of the lower limb. Journal of Anatomy: 2011;219(4):439-43

Moore HM, Thapar A, Davies KJ, Herbert P, Davis AH. Delayed diagnosis of mesenteric artery rupture. European Journal of Vascular and Endovascular Surgery Extra:2011;22:e72-5

Moore HM, Gohel M, Davies AH. The forgotten burden of deep venous disease. Phlebology:

2010;25(2):53

285

PRESENTATIONS

INTERNATIONAL

April 2014 Deep Venous Valve Management in the Future, Moore HM, Davies AH, Charing Cross

International Symposium, invited oral presentation

Feb 2014 IVC Filters for Prevention of Venous Thromboembolism in Obese Patients undergoing

Bariatric Surgery: A Systematic review. Rowland, SP, Dharmarajah B, Moore HM,

Davies AH. American Venous Forum, Poster Presentation

Feb 2014 The Effects of detergent sclerosants on white blood cells, Moore HM, Willenberg T,

Coleridge-Smith P, Chiappini C, Bertazzo S, Stevens MM, Davies AH, American Venous

Forum, Poster Presentation

Feb 2014 Development of a Novel Material for use in the Venous System. Moore HM, Steele J,

Macon A, Harbron R, Stevens M, Davies AH. American Venous Forum, Poster

Presentation

Feb 2014 Patient Follow-Up after Varicose Vein Interventions in the UK – the View of Surgeons,

General Practitioners and Patients, Moore HM, Adesina-Georgiadis N, Lane TRA,

Davies AH, American Venous Forum, Poster Presentation

Sept 2013 The anatomy of venous reflux. Qureshi MI, Moore HM, Ellis M, Franklin IJ, Davies

AH. XVII International Union of Phlebology World Meeting, Boston USA, Oral

prrsentation

Sept 2013 Mathematical Tools: HRQoL Derived Burden, QALY, DALY, Burden, Moore

HM, Lane TRA, Davies AH, XVII International Union of Phlebology World Meeting,

Boston USA, invited oral presentation

286

Sept 2013 The UK Varicose Vein Story 1989-2010, Moore HM, Head K, Shalhoub J, Lane TRA,

Davies AH, XVII International Union of Phlebology World Meeting, Boston USA,

Poster and short oral presentation

Sept 2013 In vitro effects of detergent sclerosants on white blood cells, Moore HM, Willenberg T,

Coleridge-Smith P, Chiappini C, Bertazzo S, Stevens MM, Davies AH, XVII

International Union of Phlebology World Meeting, Boston USA, Oral presentation

Sept 2013 Haemodynamic measurements around a femoral vein valve by duplex ultrasound,

Moore HM, Dharmarajah B, Gohel MS, Davies A H, XVII International Union of

Phlebology World Meeting, Boston USA, Oral presentation

Sept 2013 Deep Venous Reconstruction – A Case Series, Kosasih S, Moore HM, Davies AH,

XVII International Union of Phlebology World Meeting, Boston USA, Oral presentation

Sept 2013 Investigating Peripheral Haemodynamics during Neuromuscular Stimulation,

Varatharajan L, Moore HM, Williams KJ, Davies AH, XVII International Union of

Phlebology World Meeting, Boston USA, Poster and short oral presentation

Sept 2013 The Impact of Depression and Anxiety on the Perception of Success and Satisfaction

following Varicose Vein Interventions, Moore HM, Shalhoub J, Lane TRA, Davies AH,

XVII International Union of Phlebology World Meeting, Boston USA, Oral

presentation

Sept 2013 How should we treat complex deep venous reflux and obstruction? A survey, Moore

HM, Meththananda I, Lane TRA, Franklin, IJ, Davies A H, XVII International Union

of Phlebology World Meeting, Boston USA, Poster and short oral presentation

Sept 2013 Patient Follow-Up after Varicose Vein Interventions in the UK – the View of Surgeons,

General Practitioners and Patients, Moore HM, Adesina-Georgiadis N, Lane TRA,

287

Davies AH, XVII International Union of Phlebology World Meeting, Boston USA,

Poster and short oral presentation

Sept 2013 A UK View - Where Does Venous Disease Fit in the Vascular Surgeon's Workload?

Salem J, Thapar A, Moore HM, Davies AH, XVII International Union of Phlebology

World Meeting, Boston USA, Poster and short oral presentation

July 2013 Microemboli are visible in the right heart during endothermal ablation of varicose veins.

Moore HM, Sounderajah V, Thapar A, Lane TRA, Fox KF, Franklin IJ, Davies AH.

Surgical Research Society of Southern Africa Annual Meeting, Oral presentation.

June 2013 Post-thrombotic syndrome: A review. Moore HM, Baldwin MJ, Rudarakanchana N,

Gohel M, Davies AH, European Venous Forum, Serbia, invited oral presentation

June 2013 Comparing the venous haemodynamic effect of a neuromuscular stimulation device to

intermittent pneumatic compression in healthy subjects. Williams KJ, Moore HM, Ellis

M, Davies AH. European Venous Forum, Serbia, oral presentation

April 2013 The Need for More Deep Venous Valve Replacement, Moore HM, Davies AH,

Charing Cross International Symposium, invited oral presentation

Mar 2013 Haemodynamic measurements around a femoral vein valve by duplex ultrasound,

Moore HM, Dharmarajah B, Gohel MS, Davies A H, Australasian College of

Phlebology Annual Scientific meeting, Poster presentation

Mar 2013 The Burden of Superficial Venous Disease and the Disparity in Treatment, Poster

presentation. Moore HM, Lane TRA, Thapar A, Lifsitz A, Franklin IJ, Davies AH,

Australasian College of Phlebology Annual Scientific meeting, Poster presentation

Mar 2013 Microemboli are visible in the right heart during endothermal ablation of varicose veins,

Oral presentation. Moore HM, Sounderajah V, Thapar A, Lane TRA, Fox KF, Franklin

IJ, Davies AH. Australasian College of Phlebology Annual Scientific meeting, Oral

presentation

288

Mar 2013 Venous haemodynamics and Electrical Neuromuscular Stimulation – Preliminary data,

Poster presentation. Williams KJ, Moore HM, Davies AH, Australasian College of

Phlebology Annual Scientific meeting, Poster presentation

Jan 2013 The Disparity in Healthcare Provision in Patients with Varicose Veins. Moore HM.

Royal Society of Medicine Surgery Section, Winter Meeting, Italy.

Dec 2012 Development of a Deep Vein Valve Implant. Moore HM, Gohel MS, Xu Y, Stevens

MM, Davies AH. The MRI Instritute for Biomedical Research, Detroit USA, Oral

presentation.

June 2012 A European Disparity in Healthcare Provision in Patients with Varicose Veins. Moore

HM, Lifsitz A, Lane TRA, Thapar A, Franklin IJ, Davies AH. European Venous Forum,

Oral presentation

Jan 2012 Disparity in Healthcare Provision in Patients with Varicose Veins. Moore HM, Lifsitz

A, Gaweesh AS, Thapar A, Shalhoub J, Davies AH. American Venous Forum, Oral

Presentation

Jan 2012 Systematic Review of Sonographic CCSVI findings in Muliple Sclerosis. Thapar A, Lane

T, Moore HM, Nicholas R, Friede T, Ellis M, Assenheim J, Franklin IJ, Davies AH.

American Venous Forum, Poster Presentation, with Oral Presentation

Jan 2012 The High Incidence of Depression in Chronic Venous Disease. Sritharan K, Moore

HM, Davies AH. American Venous Forum, Poster Presentation, with Oral Presentation

Jan 2012 Despite Informed Consent Patients understanding of the Risks and Benefits of

Endovenous therapy is Poor. Sritharan K, Moore HM, Davies AH. American Venous

Forum, Poster Presentation, with Oral Presentation

July 2011 A study to evaluate the role of contrast enhanced ultrasound and dynamic magnetic

resonance imaging for deep venous disease. Moore HM, Gohel MS, Davies AH.

European Venous Forum, Oral Presentation

289

Apr 2011 Delayed presentation of mesenteric artery rupture. Moore HM, Thapar A, Davies K J,

Herbert P, Davies A H. CX Vascular Symposium, EVST Session. Oral presentation.

Jan 2011 Novel Techniques for Deep Venous Imaging. Moore HM, Pierce I. Royal Society of

Medicine Surgery Section, Winter Meeting. Oral presentation.

NATIONAL

Oct 2014 Venous Artificial Neovalve: True or Fiction and Who Will Benefit? Moore HM, Davies

AH. London Cardiovascular Symposium, invited oral presentation.

April 2013 Patient Related Outcome Measures in Varicose Vein Interventions and the Impact of

Depression and Anxiety on the Perception. Shalhoub JS, Moore HM, Lane TRA,

Davies AH. UK Venous Forum, Poster presentation

April 2013 The Effect of Neuromuscular stimulation on Peripheral Haemodynamics. Varatharajan

L, Moore HM, Williams KJ, Davies AH. UK Venous Forum, Poster presentation

April 2013 Haemodynamic Measurements around the Femoral Vein Valve by Duplex Ultrasound.

Moore HM, Dharmarajah B, Gohel MS, Davies AH. UK Venous Forum, Poster

presentation

April 2013 How should we treat complex deep venous reflux and obstruction? A survey. Moore

HM, Meththananda I, Lane TRA, Franklin IJ, Davies AH. UK Venous Forum, Poster

presentation

April 2013 Patient follow-up after Varicose Vein Interventions – a Survey of Surgeons and General

Practitioners. Moore HM, Adesina-Georgiadis N, Lane TRA, Davies AH. UK Venous

Forum, Poster presentation

290

April 2013 Intermittent pneumatic compression and neuromuscular electrical stimulation of the leg:

venous haemodynamic effects. Williams KJ, Moore HM, Ellis M, Davies AH. UK

Venous Forum, Poster presentation

March 2013 Insight into Surgical Training: Tips for Early Career Development in Surgery: A

Programme for Surgical Trainees, Trainers and Aspirant Surgeons, Moore HM, Royal

Society of Medicine Education Meeting, Oral Presentation (Invited)

Jan 2013 Microemboli are visible in the right heart during endothermal ablation of varicose veins.

Moore HM, Sounderajah V, Thapar A, Lane TRA, Fox KF, Franklin IJ, Davies AH.

Society of Academic and Research Surgery Annual Meeting, Oral presentation.

Jan 2013 A UK Disparity in the Treatment of Patients with Varicose Veins has been Identified.

Moore H M, Lane T R A, Franklin I J, Davies A H. Society of Academic and Research

Surgery Annual Meeting, Poster of Distinction and Oral presentation

Nov 2012 Microemboli are visible in the right heart during endothermal ablation of varicose veins.

Moore HM, Sounderajah V, Thapar A, Lane TRA, Fox KF, Franklin IJ, Davies AH.

Vascular Society Annual meeting, Oral presentation.

Apr 2012 The Disparity in Healthcare Provision in Patients with Varicose Veins. Moore HM,

Lifsitz A, Gaweesh AS, Thapar A, Shalhoub J, Davies AH. UK Venous Forum, Oral

Presentation

Apr 2012 A study to evaluate the role of contrast enhanced ultrasound and dynamic magnetic

resonance imaging for deep venous disease. Moore HM, Gohel MS, Davies AH. UK

Venous Forum, Oral Presentation

Apr 2012 The location and number of deep venous valves of the lower limb. Moore HM, Gohel

M, Davies AH. UK Venous Forum, Poster Presentation.

Apr 2012 Scoring systems for the Post Thrombotic Syndrome. Soosainathan A, Moore HM,

Gohel M, Davies AH. UK Venous Forum, Poster Presentation

291

May 2011 The Number and Location of Deep Venous Valves of the Lower Limb. Moore HM,

Gohel MS, Davies AH. ASBGI meeting, Bournemouth, Poster presentation.

292

10. REFERENCES

1. Van den Oever R, Hepp B, Debbaut B, Simon I. Socio-economic impact of chronic venous insufficiency. An underestimated public health problem. Int Angiol. 1998; 17(3): 161-7. 2. Lawrence P, Gazak C. Epidemiology of Chronic Venous Insufficiency. In: Gloviczki P, Bergan J, editors. Atlas of Endoscopic Perforator Vein Surgery. London: Springer-Verlag; 1998. p. 31-44. 3. Nelzen O, Bergqvist D, Lindhagen A. The prevalence of chronic lower-limb ulceration has been underestimated: results of a validated population questionnaire. Br J Surg. 1996; 83(2): 255-8. 4. van den Bos R, Arends L, Kockaert M, Neumann M, Nijsten T. Endovenous therapies of lower extremity varicosities: a meta-analysis. J Vasc Surg. 2009; 49(1): 230-9. 5. Shepherd AC, Gohel MS, Brown LC, Metcalfe MJ, Hamish M, Davies AH. Randomized clinical trial of VNUS ClosureFAST radiofrequency ablation versus laser for varicose veins. Br J Surg. 2010; 97(6): 810-8. 6. Raju S, Neglen P. Clinical practice. Chronic venous insufficiency and varicose veins. N Engl J Med. 2009; 360(22): 2319-27. 7. Grabs AJ, Wakely MC, Nyamekye I, Ghauri AS, Poskitt KR. Colour duplex ultrasonography in the rational management of chronic venous leg ulcers. Br J Surg. 1996; 83(10): 1380-2. 8. Hanrahan LM, Araki CT, Rodriguez AA, Kechejian GJ, LaMorte WW, Menzoian JO. Distribution of valvular incompetence in patients with venous stasis ulceration. J Vasc Surg. 1991; 13(6): 805-11; discussion 11-2. 9. Meissner MH. Lower extremity venous anatomy. Semin Intervent Radiol. 2005; 22(3): 147-56. 10. Alguire PC, Mathes BM. Chronic venous insufficiency and venous ulceration. J Gen Intern Med. 1997; 12(6): 374-83. 11. Meissner MH. Lower extremity venous anatomy. Seminars in Interventional Radiology. 2005; 22(3): 147- 56. 12. Alimi YS, Barthelemy P, Juhan C. Venous pump of the calf: a study of venous and muscular pressures. Journal of Vascular Surgery. 1994; 20(5): 728-35. 13. Eberhardt RT, Raffetto JD. Chronic venous insufficiency. Circulation. 2005; 111(18): 2398-409. 14. Beebe-Dimmer JL, Pfeifer JR, Engle JS, Schottenfeld D. The epidemiology of chronic venous insufficiency and varicose veins. Ann Epidemiol. 2005; 15(3): 175-84. 15. Porter JM, Moneta GL. Reporting standards in venous disease: an update. International Consensus Committee on Chronic Venous Disease. J Vasc Surg. 1995; 21(4): 635-45. 16. Neglen P. Chronic deep venous obstruction: definition, prevalence, diagnosis, management. Phlebology. 2008; 23(4): 149-57. 17. Gaweesh AS, Kayed MH, Gaweesh TY, Shalhoub J, Davies AH, Khamis HM. Underlying deep venous abnormalities in patients with unilateral chronic venous disease. Phlebology. 2012. 18. Meissner MH, Eklof B, Smith PC, Dalsing MC, DePalma RG, Gloviczki P, et al. Secondary chronic venous disorders. J Vasc Surg. 2007; 46 Suppl S: 68S-83S. 19. Moher D, Liberati A, Tetzlaff J, Altman DG. Preferred reporting items for systematic reviews and meta- analyses: the PRISMA statement. BMJ. 2009; 339: b2535. 20. Johnson BF, Manzo RA, Bergelin RO, Strandness DE, Jr. Relationship between changes in the deep venous system and the development of the postthrombotic syndrome after an acute episode of lower limb deep vein thrombosis: a one- to six-year follow-up. J Vasc Surg. 1995; 21(2): 307-12; discussion 13. 21. Labropoulos N, Waggoner T, Sammis W, Samali S, Pappas PJ. The effect of venous thrombus location and extent on the development of post-thrombotic signs and symptoms. J Vasc Surg. 2008; 48(2): 407-12. 22. Tick LW, Kramer MH, Rosendaal FR, Faber WR, Doggen CJ. Risk factors for post-thrombotic syndrome in patients with a first deep venous thrombosis. J Thromb Haemost. 2008; 6(12): 2075-81. 23. Yamaki T, Nozaki M. Patterns of venous insufficiency after an acute deep vein thrombosis. J Am Coll Surg. 2005; 201(2): 231-8. 24. van Bemmelen PS, Bedford G, Beach K, Strandness DE, Jr. Status of the valves in the superficial and deep venous system in chronic venous disease. Surgery. 1991; 109(6): 730-4. 25. Cornwall JV, Dore CJ, Lewis JD. Leg ulcers: epidemiology and aetiology. Br J Surg. 1986; 73(9): 693-6. 26. Darke SG, Penfold C. Venous ulceration and saphenous ligation. Eur J Vasc Surg. 1992; 6(1): 4-9. 27. Lees TA, Lambert D. Patterns of venous reflux in limbs with skin changes associated with chronic venous insufficiency. Br J Surg. 1993; 80(6): 725-8. 28. Shami SK, Sarin S, Cheatle TR, Scurr JH, Smith PD. Venous ulcers and the superficial venous system. J Vasc Surg. 1993; 17(3): 487-90. 29. Labropoulos N, Leon M, Geroulakos G, Volteas N, Chan P, Nicolaides AN. Venous hemodynamic abnormalities in patients with leg ulceration. Am J Surg. 1995; 169(6): 572-4.

293

30. Myers KA, Ziegenbein RW, Zeng GH, Matthews PG. Duplex ultrasonography scanning for chronic venous disease: patterns of venous reflux. J Vasc Surg. 1995; 21(4): 605-12. 31. Labropoulos N, Delis K, Nicolaides AN, Leon M, Ramaswami G. The role of the distribution and anatomic extent of reflux in the development of signs and symptoms in chronic venous insufficiency. J Vasc Surg. 1996; 23(3): 504-10. 32. Welch HJ, Young CM, Semegran AB, Iafrati MD, Mackey WC, O'Donnell TF, Jr. Duplex assessment of venous reflux and chronic venous insufficiency: the significance of deep venous reflux. J Vasc Surg. 1996; 24(5): 755-62. 33. Labropoulos N, Giannoukas AD, Delis K, Mansour MA, Kang SS, Nicolaides AN, et al. Where does venous reflux start? J Vasc Surg. 1997; 26(5): 736-42. 34. Giannoukas AD, Tsetis D, Ioannou C, Kostas T, Kafetzakis A, Petinarakis I, et al. Clinical presentation and anatomic distribution of chronic venous insufficiency of the lower limb in a typical Mediterranean population. Int Angiol. 2002; 21(2): 187-92. 35. Danielsson G, Arfvidsson B, Eklof B, Kistner RL, Masuda EM, Satoc DT. Reflux from thigh to calf, the major pathology in chronic venous ulcer disease: surgery indicated in the majority of patients. Vasc Endovascular Surg. 2004; 38(3): 209-19. 36. Ibegbuna V, Delis KT, Nicolaides AN. Haemodynamic and clinical impact of superficial, deep and perforator vein incompetence. Eur J Vasc Endovasc Surg. 2006; 31(5): 535-41. 37. Maurins U, Hoffmann BH, Losch C, Jockel KH, Rabe E, Pannier F. Distribution and prevalence of reflux in the superficial and deep venous system in the general population--results from the Bonn Vein Study, Germany. J Vasc Surg. 2008; 48(3): 680-7. 38. Kanchanabat B, Wongmahisorn Y, Stapanavatr W, Kanchanasuttirak P, Manomaiphiboon A. Clinical presentation and patterns of venous reflux in Thai patients with chronic venous insufficiency (CVI). Eur J Vasc Endovasc Surg. 2010; 40(3): 399-402. 39. Clagett GP, Reisch JS. Prevention of venous thromboembolism in general surgical patients. Results of meta-analysis. Ann Surg. 1988; 208(2): 227-40. 40. Brandjes DP, Buller HR, Heijboer H, Huisman MV, de Rijk M, Jagt H, et al. Randomised trial of effect of compression stockings in patients with symptomatic proximal-vein thrombosis. Lancet. 1997; 349(9054): 759-62. 41. Kahn SR, Ducruet T, Lamping DL, Arsenault L, Miron MJ, Roussin A, et al. Prospective evaluation of health-related quality of life in patients with deep venous thrombosis. Arch Intern Med. 2005; 165(10): 1173-8. 42. MacDougall DA, Feliu AL, Boccuzzi SJ, Lin J. Economic burden of deep-vein thrombosis, pulmonary embolism, and post-thrombotic syndrome. Am J Health Syst Pharm. 2006; 63(20 Suppl 6): S5-15. 43. Kahn SR, Shbaklo H, Lamping DL, Holcroft CA, Shrier I, Miron MJ, et al. Determinants of health-related quality of life during the 2 years following deep vein thrombosis. J Thromb Haemost. 2008; 6(7): 1105-12. 44. Kakkar VV, Lawrence D. Hemodynamic and clinical assessment after therapy for acute deep vein thrombosis. A prospective study. Am J Surg. 1985; 150(4A): 54-63. 45. Monreal M, Martorell A, Callejas JM, Valls R, Llamazares JF, Lafoz E, et al. Venographic assessment of deep vein thrombosis and risk of developing post-thrombotic syndrome: a prospective study. J Intern Med. 1993; 233(3): 233-8. 46. Prandoni P, Lensing AW, Cogo A, Cuppini S, Villalta S, Carta M, et al. The long-term clinical course of acute deep venous thrombosis. Ann Intern Med. 1996; 125(1): 1-7. 47. Saarinen J, Kallio T, Lehto M, Hiltunen S, Sisto T. The occurrence of the post-thrombotic changes after an acute deep venous thrombosis. A prospective two-year follow-up study. J Cardiovasc Surg (Torino). 2000; 41(3): 441-6. 48. Gabriel F, Labios M, Portoles O, Guillen M, Corella D, Frances F, et al. Incidence of post-thrombotic syndrome and its association with various risk factors in a cohort of Spanish patients after one year of follow-up following acute deep venous thrombosis. Thromb Haemost. 2004; 92(2): 328-36. 49. Roumen-Klappe EM, den Heijer M, Janssen MC, van der Vleuten C, Thien T, Wollersheim H. The post- thrombotic syndrome: incidence and prognostic value of non-invasive venous examinations in a six-year follow-up study. Thromb Haemost. 2005; 94(4): 825-30. 50. Kahn SR, Shrier I, Julian JA, Ducruet T, Arsenault L, Miron MJ, et al. Determinants and time course of the postthrombotic syndrome after acute deep venous thrombosis. Ann Intern Med. 2008; 149(10): 698-707. 51. van Ramshorst B, van Bemmelen PS, Hoeneveld H, Faber JA, Eikelboom BC. Thrombus regression in deep venous thrombosis. Quantification of spontaneous thrombolysis with duplex scanning. Circulation. 1992; 86(2): 414-9. 52. Markel A, Meissner M, Manzo RA, Bergelin RO, Strandness DE, Jr. Deep venous thrombosis: rate of spontaneous lysis and thrombus extension. Int Angiol. 2003; 22(4): 376-82. 53. Meissner MH, Caps MT, Zierler BK, Polissar N, Bergelin RO, Manzo RA, et al. Determinants of chronic venous disease after acute deep venous thrombosis. J Vasc Surg. 1998; 28(5): 826-33. 54. Tick LW, Doggen CJ, Rosendaal FR, Faber WR, Bousema MT, Mackaay AJ, et al. Predictors of the post- thrombotic syndrome with non-invasive venous examinations in patients 6 weeks after a first episode of deep vein thrombosis. J Thromb Haemost. 2010; 8(12): 2685-92.

294

55. Meissner MH, Zierler BK, Bergelin RO, Chandler WL, Strandness DE, Jr. Coagulation, fibrinolysis, and recanalization after acute deep venous thrombosis. J Vasc Surg. 2002; 35(2): 278-85. 56. Killewich LA, Macko RF, Cox K, Franklin DR, Benjamin ME, Lilly MP, et al. Regression of proximal deep venous thrombosis is associated with fibrinolytic enhancement. J Vasc Surg. 1997; 26(5): 861-8. 57. Markel A, Manzo RA, Bergelin RO, Strandness DE, Jr. Valvular reflux after deep vein thrombosis: incidence and time of occurrence. J Vasc Surg. 1992; 15(2): 377-82; discussion 83-4. 58. O'Shaughnessy A M, Fitzgerald DE. The patterns and distribution of residual abnormalities between the individual proximal venous segments after an acute deep vein thrombosis. J Vasc Surg. 2001; 33(2): 379-84. 59. Meissner MH, Manzo RA, Bergelin RO, Markel A, Strandness DE, Jr. Deep venous insufficiency: the relationship between lysis and subsequent reflux. J Vasc Surg. 1993; 18(4): 596-605; discussion 6-8. 60. Nicolaides AN, Hussein MK, Szendro G, Christopoulos D, Vasdekis S, Clarke H. The relation of venous ulceration with ambulatory venous pressure measurements. J Vasc Surg. 1993; 17(2): 414-9. 61. Araki CT, Back TL, Padberg FT, Thompson PN, Jamil Z, Lee BC, et al. The significance of calf muscle pump function in venous ulceration. J Vasc Surg. 1994; 20(6): 872-7; discussion 8-9. 62. Welkie JF, Comerota AJ, Katz ML, Aldridge SC, Kerr RP, White JV. Hemodynamic deterioration in chronic venous disease. J Vasc Surg. 1992; 16(5): 733-40. 63. Prandoni P, Lensing AW, Prins MH, Frulla M, Marchiori A, Bernardi E, et al. Below-knee elastic compression stockings to prevent the post-thrombotic syndrome: a randomized, controlled trial. Ann Intern Med. 2004; 141(4): 249-56. 64. Philbrick JT, Becker DM. Calf deep venous thrombosis. A wolf in sheep's clothing? Arch Intern Med. 1988; 148(10): 2131-8. 65. McLafferty RB, Moneta GL, Passman MA, Brant BM, Taylor LM, Jr., Porter JM. Late clinical and hemodynamic sequelae of isolated calf vein thrombosis. J Vasc Surg. 1998; 27(1): 50-6; discussion 6-7. 66. Meissner MH, Caps MT, Bergelin RO, Manzo RA, Strandness DE, Jr. Early outcome after isolated calf vein thrombosis. J Vasc Surg. 1997; 26(5): 749-56. 67. Kahn SR, Kearon C, Julian JA, Mackinnon B, Kovacs MJ, Wells P, et al. Predictors of the post-thrombotic syndrome during long-term treatment of proximal deep vein thrombosis. J Thromb Haemost. 2005; 3(4): 718-23. 68. Ageno W, Piantanida E, Dentali F, Steidl L, Mera V, Squizzato A, et al. Body mass index is associated with the development of the post-thrombotic syndrome. Thromb Haemost. 2003; 89(2): 305-9. 69. Kahn SR, Azoulay L, Hirsch A, Haber M, Strulovitch C, Shrier I. Effect of graduated elastic compression stockings on leg symptoms and signs during exercise in patients with deep venous thrombosis: a randomized crossover trial. J Thromb Haemost. 2003; 1(3): 494-9. 70. Stain M, Schonauer V, Minar E, Bialonczyk C, Hirschl M, Weltermann A, et al. The post-thrombotic syndrome: risk factors and impact on the course of thrombotic disease. J Thromb Haemost. 2005; 3(12): 2671-6. 71. Prandoni P, Frulla M, Sartor D, Concolato A, Girolami A. Vein abnormalities and the post-thrombotic syndrome. J Thromb Haemost. 2005; 3(2): 401-2. 72. Latella J, Desmarais S, Miron MJ, Roussin A, Joyal F, Kassis J, et al. Relation between D-dimer level, venous valvular reflux and the development of post-thrombotic syndrome after deep vein thrombosis. J Thromb Haemost. 2010; 8(10): 2169-75. 73. Partsch H, Kaulich M, Mayer W. Immediate mobilisation in acute vein thrombosis reduces post- thrombotic syndrome. Int Angiol. 2004; 23(3): 206-12. 74. Hafner J, Kuhne A, Schar B, Bombeli T, Hauser M, Luthi R, et al. Factor V Leiden mutation in postthrombotic and non-postthrombotic venous ulcers. Arch Dermatol. 2001; 137(5): 599-603. 75. Gaber Y, Siemens HJ, Schmeller W. Resistance to activated protein C due to factor V Leiden mutation: high prevalence in patients with post-thrombotic leg ulcers. Br J Dermatol. 2001; 144(3): 546-8. 76. Biguzzi E, Mozzi E, Alatri A, Taioli E, Moia M, Mannucci PM. The post-thrombotic syndrome in young women: retrospective evaluation of prognostic factors. Thromb Haemost. 1998; 80(4): 575-7. 77. Bjorgell O, Nilsson PE, Nilsson JA, Svensson PJ. Location and extent of deep vein thrombosis in patients with and without FV:R 506Q mutation. Thromb Haemost. 2000; 83(5): 648-51. 78. Bounameaux H. Factor V Leiden paradox: risk of deep-vein thrombosis but not of pulmonary embolism. Lancet. 2000; 356(9225): 182-3. 79. Christiansen SC, Cannegieter SC, Koster T, Vandenbroucke JP, Rosendaal FR. Thrombophilia, clinical factors, and recurrent venous thrombotic events. JAMA. 2005; 293(19): 2352-61. 80. Spiezia L, Campello E, Giolo E, Villalta S, Prandoni P. Thrombophilia and the risk of post-thrombotic syndrome: retrospective cohort observation. J Thromb Haemost. 2010; 8(1): 211-3. 81. Hull RD, Pineo GF, Stein PD, Mah AF, MacIsaac SM, Dahl OE, et al. Extended out-of-hospital low- molecular-weight heparin prophylaxis against deep venous thrombosis in patients after elective hip arthroplasty: a systematic review. Ann Intern Med. 2001; 135(10): 858-69. 82. Wille-Jorgensen P, Jorgensen LN, Crawford M. Asymptomatic postoperative deep vein thrombosis and the development of postthrombotic syndrome. A systematic review and meta-analysis. Thromb Haemost. 2005; 93(2): 236-41.

295

83. Ginsberg JS, Turkstra F, Buller HR, MacKinnon B, Magier D, Hirsh J. Postthrombotic syndrome after hip or knee arthroplasty: a cross-sectional study. Arch Intern Med. 2000; 160(5): 669-72. 84. Persson LM, Lapidus LJ, Larfars G, Rosfors S. Asymptomatic deep venous thrombosis is associated with a low risk of post-thrombotic syndrome. Eur J Vasc Endovasc Surg. 2009; 38(2): 229-33. 85. Andersen M, Wille-Jorgensen P. Late complications of asymptomatic deep venous thrombosis. Eur J Surg. 1991; 157(9): 527-30. 86. Siragusa S, Beltrametti C, Barone M, Piovella F. [Clinical course and incidence of post-thrombophlebitic syndrome after profound asymptomatic deep vein thrombosis. Results of a transverse epidemiologic study]. Minerva Cardioangiol. 1997; 45(3): 57-66. 87. Schindler OS, Dalziel R. Post-thrombotic syndrome after total hip or knee arthroplasty: incidence in patients with asymptomatic deep venous thrombosis. J Orthop Surg (Hong Kong). 2005; 13(2): 113-9. 88. Mudge M, Leinster SJ, Hughes LE. A prospective 10-year study of the post-thrombotic syndrome in a surgical population. Ann R Coll Surg Engl. 1988; 70(4): 249-52. 89. Lindhagen A, Bergqvist D, Hallbook T. Deep venous insufficiency after postoperative thrombosis diagnosed with 125I-labelled fibrinogen uptake test. Br J Surg. 1984; 71(7): 511-5. 90. Pesavento R, Villalta S, Prandoni P. The postthrombotic syndrome. Intern Emerg Med. 2010; 5(3): 185-92. 91. Prandoni P, Kahn SR. Post-thrombotic syndrome: prevalence, prognostication and need for progress. Br J Haematol. 2009; 145(3): 286-95. 92. Kahn SR, Partsch H, Vedantham S, Prandoni P, Kearon C. Definition of post-thrombotic syndrome of the leg for use in clinical investigations: a recommendation for standardization. J Thromb Haemost. 2009; 7(5): 879-83. 93. Roumen-Klappe EM, Janssen MC, Van Rossum J, Holewijn S, Van Bokhoven MM, Kaasjager K, et al. Inflammation in deep vein thrombosis and the development of post-thrombotic syndrome: a prospective study. J Thromb Haemost. 2009; 7(4): 582-7. 94. Shbaklo H, Holcroft CA, Kahn SR. Levels of inflammatory markers and the development of the post- thrombotic syndrome. Thromb Haemost. 2009; 101(3): 505-12. 95. Kahn SR, Shbaklo H, Shapiro S, Wells PS, Kovacs MJ, Rodger MA, et al. Effectiveness of compression stockings to prevent the post-thrombotic syndrome (the SOX Trial and Bio-SOX biomarker substudy): a randomized controlled trial. BMC Cardiovasc Disord. 2007; 7: 21. 96. Kahn SR, Hirsch A, Shrier I. Effect of postthrombotic syndrome on health-related quality of life after deep venous thrombosis. Arch Intern Med. 2002; 162(10): 1144-8. 97. Soosainathan A, Moore HM, Gohel MS, Davies AH. Scoring systems for the post-thrombotic syndrome. J Vasc Surg. 2013; 57(1): 254-61. 98. Rodger MA, Kahn SR, Le Gal G, Solymoss S, Chagnon I, Anderson DR, et al. Inter-observer reliability of measures to assess the post-thrombotic syndrome. Thromb Haemost. 2008; 100(1): 164-6. 99. Kolbach DN, Neumann HA, Prins MH. Definition of the post-thrombotic syndrome, differences between existing classifications. Eur J Vasc Endovasc Surg. 2005; 30(4): 404-14. 100. Kahn SR, Desmarais S, Ducruet T, Arsenault L, Ginsberg JS. Comparison of the Villalta and Ginsberg clinical scales to diagnose the post-thrombotic syndrome: correlation with patient-reported disease burden and venous valvular reflux. J Thromb Haemost. 2006; 4(4): 907-8. 101. van Dongen CJ, Prandoni P, Frulla M, Marchiori A, Prins MH, Hutten BA. Relation between quality of anticoagulant treatment and the development of the postthrombotic syndrome. J Thromb Haemost. 2005; 3(5): 939- 42. 102. Widmer L, Stathelin H, Nissen C. Sd. Venen-, Arterien-, Krankheiten, koronaire Herzkrankhei bei berufstatigen. Bern: Verlag Hans Huber; 1981. 103. Kolbach DN, Sandbrink MW, Neumann HA, Prins MH. Compression therapy for treating stage I and II (Widmer) post-thrombotic syndrome. Cochrane Database Syst Rev. 2003; (4): CD004177. 104. Kahn SR, Shrier I, Shapiro S, Houweling AH, Hirsch AM, Reid RD, et al. Six-month exercise training program to treat post-thrombotic syndrome: a randomized controlled two-centre trial. CMAJ. 2011; 183(1): 37-44. 105. O'Donnell MJ, McRae S, Kahn SR, Julian JA, Kearon C, Mackinnon B, et al. Evaluation of a venous-return assist device to treat severe post-thrombotic syndrome (VENOPTS). A randomized controlled trial. Thromb Haemost. 2008; 99(3): 623-9. 106. Ginsberg JS, Hirsh J, Julian J, Vander LaandeVries M, Magier D, MacKinnon B, et al. Prevention and treatment of postphlebitic syndrome: results of a 3-part study. Arch Intern Med. 2001; 161(17): 2105-9. 107. Kearon C, Kahn SR, Agnelli G, Goldhaber S, Raskob GE, Comerota AJ. Antithrombotic therapy for venous thromboembolic disease: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines (8th Edition). Chest. 2008; 133(6 Suppl): 454S-545S. 108. Kearon C, Gent M, Hirsh J, Weitz J, Kovacs MJ, Anderson DR, et al. A comparison of three months of anticoagulation with extended anticoagulation for a first episode of idiopathic venous thromboembolism. N Engl J Med. 1999; 340(12): 901-7. 109. Agnelli G, Prandoni P, Santamaria MG, Bagatella P, Iorio A, Bazzan M, et al. Three months versus one year of oral anticoagulant therapy for idiopathic deep venous thrombosis. Warfarin Optimal Duration Italian Trial Investigators. N Engl J Med. 2001; 345(3): 165-9.

296

110. Schulman S, Rhedin AS, Lindmarker P, Carlsson A, Larfars G, Nicol P, et al. A comparison of six weeks with six months of oral anticoagulant therapy after a first episode of venous thromboembolism. Duration of Anticoagulation Trial Study Group. N Engl J Med. 1995; 332(25): 1661-5. 111. Schulman S, Lindmarker P, Holmstrom M, Larfars G, Carlsson A, Nicol P, et al. Post-thrombotic syndrome, recurrence, and death 10 years after the first episode of venous thromboembolism treated with warfarin for 6 weeks or 6 months. J Thromb Haemost. 2006; 4(4): 734-42. 112. Hull RD, Pineo GF, Brant R, Liang J, Cook R, Solymoss S, et al. Home therapy of venous thrombosis with long-term LMWH versus usual care: patient satisfaction and post-thrombotic syndrome. Am J Med. 2009; 122(8): 762-9 e3. 113. Gonzalez-Fajardo JA, Martin-Pedrosa M, Castrodeza J, Tamames S, Vaquero-Puerta C. Effect of the anticoagulant therapy in the incidence of post-thrombotic syndrome and recurrent thromboembolism: Comparative study of enoxaparin versus coumarin. J Vasc Surg. 2008; 48(4): 953-9. 114. Hull RD, Pineo GF, Brant RF, Mah AF, Burke N, Dear R, et al. Self-managed long-term low-molecular- weight heparin therapy: the balance of benefits and harms. Am J Med. 2007; 120(1): 72-82. 115. Mousa SA. The low molecular weight heparin, tinzaparin, in thrombosis and beyond. Cardiovasc Drug Rev. 2002; 20(3): 199-216. 116. Kearon C, Hirsh J. Management of anticoagulation before and after elective surgery. N Engl J Med. 1997; 336(21): 1506-11. 117. Kearon C. Management of anticoagulation in patients who require invasive procedures. Semin Vasc Med. 2003; 3(3): 285-94. 118. Roche-Nagle G, Ward F, Barry M. Current prescribing patterns of elastic compression stockings post-deep venous thrombosis. Phlebology. 2010; 25(2): 72-8. 119. Brakkee A, Kuiper J. THe Influence of Compressive Stockings on the Haemodynamics in the Lower Extremities. Phlebology. 1988; 3: 147-53. 120. Kahn SR, Elman E, Rodger MA, Wells PS. Use of elastic compression stockings after deep venous thrombosis: a comparison of practices and perceptions of thrombosis physicians and patients. J Thromb Haemost. 2003; 1(3): 500-6. 121. Kahn SR, Shapiro S, Wells PS, Rodger MA, Kovacs MJ, Anderson DR, et al. Compression stockings to prevent post-thrombotic syndrome: a randomised placebo-controlled trial. Lancet. 2014; 383(9920): 880-8. 122. Musani MH, Matta F, Yaekoub AY, Liang J, Hull RD, Stein PD. Venous compression for prevention of postthrombotic syndrome: a meta-analysis. Am J Med. 2010; 123(8): 735-40. 123. Kolbach DN, Sandbrink MW, Hamulyak K, Neumann HA, Prins MH. Non-pharmaceutical measures for prevention of post-thrombotic syndrome. Cochrane Database Syst Rev. 2004; (1): CD004174. 124. Aschwanden M, Jeanneret C, Koller MT, Thalhammer C, Bucher HC, Jaeger KA. Effect of prolonged treatment with compression stockings to prevent post-thrombotic sequelae: a randomized controlled trial. J Vasc Surg. 2008; 47(5): 1015-21. 125. Sajid MS, Tai NR, Goli G, Morris RW, Baker DM, Hamilton G. Knee versus thigh length graduated compression stockings for prevention of deep venous thrombosis: a systematic review. Eur J Vasc Endovasc Surg. 2006; 32(6): 730-6. 126. Nelson EA, Harper DR, Prescott RJ, Gibson B, Brown D, Ruckley CV. Prevention of recurrence of venous ulceration: randomized controlled trial of class 2 and class 3 elastic compression. J Vasc Surg. 2006; 44(4): 803-8. 127. Roumen-Klappe EM, den Heijer M, van Rossum J, Wollersheim H, van der Vleuten C, Thien T, et al. Multilayer compression bandaging in the acute phase of deep-vein thrombosis has no effect on the development of the post-thrombotic syndrome. J Thromb Thrombolysis. 2009; 27(4): 400-5. 128. Arpaia G, Cimminiello C, Mastrogiacomo O, de Gaudenzi E. Efficacy of elastic compression stockings used early or after resolution of the edema on recanalization after deep venous thrombosis: the COM.PRE Trial. Blood Coagul Fibrinolysis. 2007; 18(2): 131-7. 129. Shrier I, Kahn SR. Effect of physical activity after recent deep venous thrombosis: a cohort study. Med Sci Sports Exerc. 2005; 37(4): 630-4. 130. Shrier I, Kahn SR, Steele RJ. Effect of early physical activity on long-term outcome after venous thrombosis. Clin J Sport Med. 2009; 19(6): 487-93. 131. Stick C, Grau H, Witzleb E. On the edema-preventing effect of the calf muscle pump. Eur J Appl Physiol Occup Physiol. 1989; 59(1-2): 39-47. 132. Isma N, Johanssson E, Bjork A, Bjorgell O, Robertson F, Mattiasson I, et al. Does supervised exercise after deep venous thrombosis improve recanalization of occluded vein segments? A randomized study. J Thromb Thrombolysis. 2007; 23(1): 25-30. 133. Plate G, Eklof B, Norgren L, Ohlin P, Dahlstrom JA. Venous thrombectomy for iliofemoral vein thrombosis--10-year results of a prospective randomised study. Eur J Vasc Endovasc Surg. 1997; 14(5): 367-74. 134. Hull RD, Marder VJ, Mah AF, Biel RK, Brant RF. Quantitative assessment of thrombus burden predicts the outcome of treatment for venous thrombosis: a systematic review. Am J Med. 2005; 118(5): 456-64.

297

135. Browse NL, Thomas ML, Pim HP. Streptokinase and deep vein thrombosis. Br Med J. 1968; 3(5620): 717- 20. 136. Kakkar VV, Flanc C, O'Shea MJ, Flute PT, Howe CT, Clarke MB. Treatment of deep-vein thrombosis with streptokinase. Br J Surg. 1969; 56(3): 178-83. 137. Porter JM, Seaman AJ, Common HH, Rosch J, Eidemiller LR, Calhoun AD. Comparison of heparin and streptokinase in the treatment of venous thrombosis. Am Surg. 1975; 41(9): 511-19. 138. Seaman AJ, Common HH, Rosch J, Dotter CT, Porter JM, Lindell TD, et al. Deep vein thrombosis treated with streptokinase or heparin. A randomized study. Angiology. 1976; 27(10): 549-56. 139. Watson LI, Armon MP. Thrombolysis for acute deep vein thrombosis. Cochrane Database Syst Rev. 2004; (4): CD002783. 140. Elliot MS, Immelman EJ, Jeffery P, Benatar SR, Funston MR, Smith JA, et al. A comparative randomized trial of heparin versus streptokinase in the treatment of acute proximal venous thrombosis: an interim report of a prospective trial. Br J Surg. 1979; 66(12): 838-43. 141. Arnesen H, Heilo A, Jakobsen E, Ly B, Skaga E. A prospective study of streptokinase and heparin in the treatment of deep vein thrombosis. Acta Med Scand. 1978; 203(6): 457-63. 142. Elsharawy M, Elzayat E. Early results of thrombolysis vs anticoagulation in iliofemoral venous thrombosis. A randomised clinical trial. Eur J Vasc Endovasc Surg. 2002; 24(3): 209-14. 143. Sharifi M, Mehdipour M, Bay C, Smith G, Sharifi J. Endovenous therapy for deep venous thrombosis: the TORPEDO trial. Catheter Cardiovasc Interv. 2010; 76(3): 316-25. 144. Enden T, Klow NE, Sandvik L, Slagsvold CE, Ghanima W, Hafsahl G, et al. Catheter-directed thrombolysis vs. anticoagulant therapy alone in deep vein thrombosis: results of an open randomized, controlled trial reporting on short-term patency. J Thromb Haemost. 2009; 7(8): 1268-75. 145. Karthikesalingam A, Young EL, Hinchliffe RJ, Loftus IM, Thompson MM, Holt PJ. A systematic review of percutaneous mechanical thrombectomy in the treatment of deep venous thrombosis. Eur J Vasc Endovasc Surg. 2011; 41(4): 554-65. 146. O'Meara S, Cullum NA, Nelson EA. Compression for venous leg ulcers. Cochrane Database Syst Rev. 2009; (1): CD000265. 147. Ginsberg JS, Magier D, Mackinnon B, Gent M, Hirsh J. Intermittent compression units for severe postphlebitic syndrome: a randomized crossover study. CMAJ. 1999; 160(9): 1303-6. 148. Rosenberg IL, Evans M, Pollock AV. Prophylaxis of postoperative leg vine thrombosis by low dose subcutaneous heparin or peroperative calf muscle stimulation: a controlled clinical trial. Br Med J. 1975; 1(5959): 649-51. 149. Bostrom S, Holmgren E, Jonsson O, Lindberg S, Lindstrom B, Winso I, et al. Post-operative thromboembolism in neurosurgery. A study on the prophylactic effect of calf muscle stimulation plus dextran compared to low-dose heparin. Acta Neurochir (Wien). 1986; 80(3-4): 83-9. 150. Doran FS, Drury M, Sivyer A. A Simple Way to Combat the Venous Stasis Which Occurs in the Lower Limbs during Surgical Operations. Br J Surg. 1964; 51: 486-92. 151. Gohel MS, Davies AH. Pharmacological treatment in patients with C4, C5 and C6 venous disease. Phlebology. 2010; 25 Suppl 1: 35-41. 152. Layton AM, Ibbotson SH, Davies JA, Goodfield MJ. Randomised trial of oral aspirin for chronic venous leg ulcers. Lancet. 1994; 344(8916): 164-5. 153. Shoab SS, Porter JB, Scurr JH, Coleridge-Smith PD. Effect of oral micronized purified flavonoid fraction treatment on leukocyte adhesion molecule expression in patients with chronic venous disease: a pilot study. J Vasc Surg. 2000; 31(3): 456-61. 154. Smith PC. Daflon 500 mg and venous leg ulcer: new results from a meta-analysis. Angiology. 2005; 56 Suppl 1: S33-9. 155. Jantet G. Chronic venous insufficiency: worldwide results of the RELIEF study. Reflux assEssment and quaLity of lIfe improvEment with micronized Flavonoids. Angiology. 2002; 53(3): 245-56. 156. Prandoni P. Elastic stockings, hydroxyethylrutosides or both for the treatment of post-thrombotic syndrome. Thromb Haemost. 2005; 93(1): 183-5. 157. Pittler MH, Ernst E. Horse chestnut seed extract for chronic venous insufficiency. Cochrane Database Syst Rev. 2006; (1): CD003230. 158. Ciapponi A, Laffaire E, Roque M. Calcium dobesilate for chronic venous insufficiency: a systematic review. Angiology. 2004; 55(2): 147-54. 159. Martinez-Zapata MJ, Moreno RM, Gich I, Urrutia G, Bonfill X. A randomized, double-blind multicentre clinical trial comparing the efficacy of calcium dobesilate with placebo in the treatment of chronic venous disease. Eur J Vasc Endovasc Surg. 2008; 35(3): 358-65. 160. Jull A, Arroll B, Parag V, Waters J. Pentoxifylline for treating venous leg ulcers. Cochrane Database Syst Rev. 2007; (3): CD001733. 161. Rhodes JM, Gloviczki P, Canton L, Heaser TV, Rooke TW. Endoscopic perforator vein division with ablation of superficial reflux improves venous hemodynamics. J Vasc Surg. 1998; 28(5): 839-47.

298

162. Proebstle TM, Weisel G, Paepcke U, Gass S, Weber L. Light reflection rheography and clinical course of patients with advanced venous disease before and after endoscopic subfascial division of perforating veins. Dermatol Surg. 1998; 24(7): 771-6. 163. Gloviczki P, Bergan JJ, Rhodes JM, Canton LG, Harmsen S, Ilstrup DM. Mid-term results of endoscopic perforator vein interruption for chronic venous insufficiency: lessons learned from the North American subfascial endoscopic perforator surgery registry. The North American Study Group. J Vasc Surg. 1999; 29(3): 489-502. 164. Tenbrook JA, Jr., Iafrati MD, O'Donnell T F, Jr., Wolf MP, Hoffman SN, Pauker SG, et al. Systematic review of outcomes after surgical management of venous disease incorporating subfascial endoscopic perforator surgery. J Vasc Surg. 2004; 39(3): 583-9. 165. Masuda EM, Kistner RL. Long-term results of venous valve reconstruction: a four- to twenty-one-year follow-up. J Vasc Surg. 1994; 19(3): 391-403. 166. Comerota AJ, Grewal NK, Thakur S, Assi Z. Endovenectomy of the common femoral vein and intraoperative iliac vein recanalization for chronic iliofemoral venous occlusion. J Vasc Surg. 2010; 52(1): 243-7. 167. Puggioni A, Kistner RL, Eklof B, Lurie F. Surgical disobliteration of postthrombotic deep veins-- endophlebectomy--is feasible. J Vasc Surg. 2004; 39(5): 1048-52; discussion 52. 168. Kolbel T, Lindh M, Akesson M, Wasselius J, Gottsater A, Ivancev K. Chronic iliac vein occlusion: midterm results of endovascular recanalization. J Endovasc Ther. 2009; 16(4): 483-91. 169. Neglen P, Berry MA, Raju S. Endovascular surgery in the treatment of chronic primary and post- thrombotic iliac vein obstruction. Eur J Vasc Endovasc Surg. 2000; 20(6): 560-71. 170. Harrison MB, Vandenkerkhof EG, Hopman WM, Graham ID, Carley ME, Nelson EA, et al. The Canadian Bandaging Trial: Evidence-informed Leg Ulcer Care and the Effectiveness of Two Compression Technologies. BMC Nurs. 2011; 10(1): 20. 171. Nelson EA, Mani R, Thomas K, Vowden K. Intermittent pneumatic compression for treating venous leg ulcers. Cochrane Database Syst Rev. 2011; (2): CD001899. 172. Corley GJ, Breen PP, Birlea SI, Serrador JM, Grace PA, Olaighin G. Hemodynamic effects of habituation to a week-long program of neuromuscular electrical stimulation. Med Eng Phys. 2011. 173. Izumi M, Ikeuchi M, Mitani T, Taniguchi S, Tani T. Prevention of venous stasis in the lower limb by transcutaneous electrical nerve stimulation. Eur J Vasc Endovasc Surg. 2010; 39(5): 642-5. 174. Lyseng-Williamson KA, Perry CM. Micronised purified flavonoid fraction: a review of its use in chronic venous insufficiency, venous ulcers and haemorrhoids. Drugs. 2003; 63(1): 71-100. 175. Perrin M. [Surgery for deep venous reflux in the lower limb]. J Mal Vasc. 2004; 29(2): 73-87. 176. Hardy SC, Riding G, Abidia A. Surgery for deep venous incompetence. Cochrane Database Syst Rev. 2004; (3): CD001097. 177. Maleti O, Perrin M. Reconstructive surgery for deep vein reflux in the lower limbs: techniques, results and indications. Eur J Vasc Endovasc Surg. 2011; 41(6): 837-48. 178. Kistner RL. Valve reconstruction for primary valve insufficiency. In: Bergan J, Kistner RL, editors. Atlas of Venous Surgery. Philadelphia: Saunders; 1992. p. 125-30. 179. Sottiurai V. Supravalvular incision for valve repair in primary valvular insufficiency: method of Sottiurai. In: Bergan J, Kistner RL, editors. Atlas of Venous Surgery. Philadelphia: Saunders; 1992. p. 137-44. 180. Hoshino S, Satokawa H, Takase S, Midorikawa H, Igar T, Iwaya F. External valvuloplasty for primary valvular incompetence of the lower limbs using angioscopy. International Journal of Angiology. 1997; 6(2): 137-41. 181. Jessup G, Lane RJ. Repair of incompetent venous valves: a new technique. J Vasc Surg. 1988; 8(5): 569-75. 182. Taheri SA, Lazar L, Elias S, Marchand P, Heffner R. Surgical treatment of postphlebitic syndrome with vein valve transplant. Am J Surg. 1982; 144(2): 221-4. 183. Iafrati M, O'Donnell TF. Surgical reconstruction for deep venous insufficiency. J Mal Vasc. 1997; 22(3): 193-7. 184. Puggioni A, Lurie F. Advances in the surgical treatment of postthrombotic syndrome. Phlebolymphology. 2008; 15(1): 99. 185. Neglen P, Raju S. Venous reflux repair with cryopreserved vein valves. J Vasc Surg. 2003; 37(3): 552-7. 186. Kistner RL, Sparkuhl MD. Surgery in acute and chronic venous disease. Surgery. 1979; 85(1): 31-43. 187. Johnson ND, Queral LA, Flinn WR, Yao JS, Bergan JJ. Late objective assessment of venous value surgery. Arch Surg. 1981; 116(11): 1461-6. 188. Queral LA, Whitehouse WM, Jr., Flinn WR, Neiman HL, Yao JS, Bergan JJ. Surgical correction of chronic deep venous insufficiency by valvular transposition. Surgery. 1980; 87(6): 688-95. 189. Perrin M. Reconstructive surgery for deep venous reflux: a report on 144 cases. Cardiovasc Surg. 2000; 8(4): 246-55. 190. Neglen P, Hollis KC, Olivier J, Raju S. Stenting of the venous outflow in chronic venous disease: long- term stent-related outcome, clinical, and hemodynamic result. J Vasc Surg. 2007; 46(5): 979-90. 191. Raju GS, Ahmed I. Team approach to the management of complex esophageal obstruction. Gastrointest Endosc. 2002; 55(2): 304.

299

192. Hartung O, Otero A, Boufi M, Decaridi G, Barthelemy P, Juhan C, et al. Mid-term results of endovascular treatment for symptomatic chronic nonmalignant iliocaval venous occlusive disease. J Vasc Surg. 2005; 42(6): 1138- 44; discussion 44. 193. Smith DE, Trimble CE. Surgical management of obstructive venous disease of the lower extremity. Vascular Surgery. Philadelphia: Saunders; 1977. p. 1247-68. 194. Eklof B, Albrechtson U, Einarsson E, Plate G. The temporary arteriovenous fistula in venous reconstructive surgery. Int Angiol. 1985; 4(4): 455-62. 195. AbuRahma AF, Robinson PA, Boland JP. Clinical, hemodynamic, and anatomic predictors of long-term outcome of lower extremity venovenous bypasses. J Vasc Surg. 1991; 14(5): 635-44. 196. Kelleher D, Moore HM, Davies AH. Debate: Deep venous reconstruction is worthwhile. In: Greenhalgh RM, editor. Vascular and Endovascular Controversies Update. London: Biba Publishing; 2012. p. 610-8. 197. Pavcnik D, Uchida B, Kaufman J, Hinds M, Keller FS, Rosch J. Percutaneous management of chronic deep venous reflux: review of experimental work and early clinical experience with bioprosthetic valve. Vasc Med. 2008; 13(1): 75-84. 198. McLachlin AD, Carroll SE, Meads GE, Amacher AL. Valve replacement in the recanalized incompetent superficial femoral vein in dogs. Ann Surg. 1965; 162(3): 446-52. 199. Hill R, Schmidt S, Evancho M, Hunter T, Hillegass D, Sharp W. Development of a prosthetic venous valve. J Biomed Mater Res. 1985; 19(7): 827-32. 200. Gomez-Jorge J, Venbrux AC, Magee C. Percutaneous deployment of a valved bovine jugular vein in the swine venous system: a potential treatment for venous insufficiency. J Vasc Interv Radiol. 2000; 11(7): 931-6. 201. Gale SS, Shuman S, Beebe HG, Pigott JP, Comerota AJ. Percutaneous venous valve bioprosthesis: initial observations. Vasc Endovascular Surg. 2004; 38(3): 221-4. 202. Pavcnik D. Update on venous valve replacment: long-term clinical results. Vascular. 2006; 14: 106. 203. Geselschap JH, van Zuiden JM, Toonder IM, Wittens CH. In vitro evaluation of a new autologous valve- stent for deep venous incompetence. J Endovasc Ther. 2006; 13(6): 762-9. 204. DeLaria GA, Phifer T, Roy J, Tu R, Thyagarajan K, Quijano RC. Hemodynamic evaluation of a bioprosthetic venous prosthesis. J Vasc Surg. 1993; 18(4): 577-84; discussion 84-6. 205. Pavcnik D, Uchida B, Timmermans H, Keller FS, Rosch J. Square stent: a new self-expandable endoluminal device and its applications. Cardiovasc Intervent Radiol. 2001; 24(4): 207-17. 206. Pavcnik D, Uchida BT, Timmermans H, Corless CL, Keller FS, Rosch J. Aortic and venous valve for percutaneous insertion. Minim Invasive Ther Allied Technol. 2000; 9(3-4): 287-92. 207. Oberdier MT, Rittgers SE. The design, development, and evaluation of a prototypic, prosthetic venous valve. Biomed Eng Online. 2008; 7: 25. 208. Rittgers SE, Oberdier MT, Pottala S. Physiologically-based testing system for the mechanical characterization of prosthetic vein valves. Biomed Eng Online. 2007; 6: 29. 209. Sathe RD, Ku DN. Flexible Prosthetic Vein Valve. Journal of Medical Devices. 2007; 1: 105-12. 210. Moriyama M, Kubota S, Tashiro H, Tonami H. Evaluation of prosthetic venous valves, fabricated by electrospinning, for percutaneous treatment of chronic venous insufficiency. J Artif Organs. 2011; 14(4): 294-300. 211. Kuraishi K, Iwata H, Nakano S, Kubota S, Tonami H, Toda M, et al. Development of nanofiber-covered stents using electrospinning: in vitro and acute phase in vivo experiments. J Biomed Mater Res B Appl Biomater. 2009; 88(1): 230-9. 212. de Borst GJ, Teijink JA, Patterson M, Quijano TC, Moll FL. A percutaneous approach to deep venous valve insufficiency with a new self-expanding venous frame valve. J Endovasc Ther. 2003; 10(2): 341-9. 213. Boudjemline Y, Bonnet D, Sidi D, Bonhoeffer P. Is percutaneous implantation of a bovine venous valve in the inferior vena cava a reliable technique to treat chronic venous insufficiency syndrome? Med Sci Monit. 2004; 10(3): BR61-6. 214. Kucher T, Brener B, Marak M, Parsonnet V. Endovascular delivery of vein segments with valves versus direct anastomosis. J Endovasc Ther. 2005; 12(3): 366-70. 215. Ofenloch JC, Chen C, Hughes JD, Lumsden AB. Endoscopic venous valve transplantation with a valve- stent device. Ann Vasc Surg. 1997; 11(1): 62-7. 216. Dalsing MC, Sawchuk AP, Lalka SG, Cikrit DF. An early experience with endovascular venous valve transplantation. J Vasc Surg. 1996; 24(5): 903-5. 217. Pavcnik D, Uchida BT, Timmermans HA, Corless CL, O'Hara A, Toyota N, et al. Percutaneous bioprosthetic venous valve: A long-term study in sheep. Journal of Vascular Surgery. 2002; 35(3): 598-602. 218. Pavcnik D, Yin Q, Uchida B, Park WK, Hoppe H, Kim MD, et al. Percutaneous autologous venous valve transplantation: short-term feasibility study in an ovine model. J Vasc Surg. 2007; 46(2): 338-45. 219. Pavcnik D, Kaufman J, Uchida B, Correa L, Hiraki T, Kyu SC, et al. Second-generation percutaneous bioprosthetic valve: a short-term study in sheep. J Vasc Surg. 2004; 40(6): 1223-7. 220. Serino F. Preliminary clinical experiences with the VenPro PVVB. Updating Course in vascular pathology of surgical interest; 2002; Pisa; 2002. 221. Pavcnik D, Kaufman J, Machan L, Uchida B, Keller FS, Rosch J. Percutaneous therapy for deep vein reflux. Semin Intervent Radiol. 2005; 22(3): 225-32.

300

222. Pavcnik D, Machan L, Uchida B, Kaufman J, Keller FS, Rosch J. Percutaneous prosthetic venous valves: current state and possible applications. Tech Vasc Interv Radiol. 2003; 6(3): 137-42. 223. Lurie F, Kistner RL, Eklof B. The mechanism of venous valve closure in normal physiologic conditions. J Vasc Surg. 2002; 35(4): 713-7. 224. Powell T, Lynn RB. The valves of the external iliac, femoral, and upper third of the popliteal veins. Surg Gynecol Obstet. 1951; 92(4): 453-5. 225. Basmajian JV. The distribution of valves in the femoral, external iliac, and common iliac veins and their relationship to varicose veins. Surg Gynecol Obstet. 1952; 95(5): 537-42. 226. Lurie F, Kistner RL, Eklof B, Kessler D. Mechanism of venous valve closure and role of the valve in circulation: a new concept. J Vasc Surg. 2003; 38(5): 955-61. 227. Scultetus AH, Villavicencio JL, Rich NM. Facts and fiction surrounding the discovery of the venous valves. J Vasc Surg. 2001; 33(2): 435-41. 228. Moore HM, Gohel M, Davies AH. Number and location of venous valves within the popliteal and femoral veins: a review of the literature. J Anat. 2011; 219(4): 439-43. 229. G G. Venous anatomy of the lower limb. SocietàItaliana di Flebologia; c2002; c2002. 230. Muhlberger D, Morandini L, Brenner E. An anatomical study of femoral vein valves near the saphenofemoral junction. J Vasc Surg. 2008; 48(4): 994-9. 231. Banjo AO. Comparative study of the distribution of venous valves in the lower extremities of black Africans and Caucasians: pathogenetic correlates of prevalence of primary varicose veins in the two races. Anat Rec. 1987; 217(4): 407-12. 232. LB K. Frequency and Distribution of the Valves in the large deep and superficial veins of the lower limb in the aged. ActaMorpholNeerl-Scand. 1971; 9: 41-6. 233. Eger SA, Wagner FB, Jr. Etiology of varicose veins. Postgrad Med. 1949; 6(3): 234-8. 234. Santilli SM, Lee ES, Wernsing SE, Diedrich DA, Kuskowski MA, Shew RL. Superficial femoral popliteal vein: An anatomic study. J Vasc Surg. 2000; 31(3): 450-5. 235. Nordon IM, Hinchliffe RJ, Brar R, Moxey P, Black SA, Thompson MM, et al. A prospective double-blind randomized controlled trial of radiofrequency versus laser treatment of the great saphenous vein in patients with varicose veins. Ann Surg. 2011; 254(6): 876-81. 236. Ciostek P, Michalak J, Noszczyk W. Improvement in deep vein haemodynamics following surgery for varicose veins. Eur J Vasc Endovasc Surg. 2004; 28(5): 473-8. 237. Meissner MH, Moneta G, Burnand K, Gloviczki P, Lohr JM, Lurie F, et al. The hemodynamics and diagnosis of venous disease. J Vasc Surg. 2007; 46 Suppl S: 4S-24S. 238. Abu-Yousef MM, Mufid M, Woods KT, Brown BP, Barloon TJ. Normal lower limb venous Doppler flow phasicity: is it cardiac or respiratory? AJR Am J Roentgenol. 1997; 169(6): 1721-5. 239. Wexler L, Bergel DH, Gabe IT, Makin GS, Mills CJ. Velocity of blood flow in normal human venae cavae. Circ Res. 1968; 23(3): 349-59. 240. Willeput R, Rondeux C, Troyer AD. Breathing affects venous return from legs in humans. J Appl Physiol. 1984; 57(4): 971-6. 241. Kwon O-Y, Jung D-Y, Kim Y, Cho S-H, Yi C-H. Effects of ankle exercise combined with deep breathing on blood flow velocity in the femoral vein. Aust J Physiother. 2003; 49(4): 253-8. 242. Miller JD, Pegelow DF, Jacques AJ, Dempsey JA. Skeletal muscle pump versus respiratory muscle pump: modulation of venous return from the locomotor limb in humans. J Physiol. 2005; 563(Pt 3): 925-43. 243. Downie SP, Raynor SM, Firmin DN, Wood NB, Thom SA, Hughes AD, et al. Effects of elastic compression stockings on wall shear stress in deep and superficial veins of the calf. Am J Physiol Heart Circ Physiol. 2008; 294(5): H2112-H20. 244. Padberg FT, Jr., Johnston MV, Sisto SA. Structured exercise improves calf muscle pump function in chronic venous insufficiency: a randomized trial. J Vasc Surg. 2004; 39(1): 79-87. 245. Chauveau M, Gelade P, Cros F. The venous return simulator: comparison of simulated with measured ambulatory venous pressure in normal subjects and in venous valve incompetence. Vasa. 2011; 40(3): 205-17. 246. Uhl JF, Gillot C. Anatomy of the foot venous pump: physiology and influence on chronic venous disease. Phlebology. 2012; 27(5): 219-30. 247. Stewart JM, Medow MS, Montgomery LD, McLeod K. Decreased skeletal muscle pump activity in patients with postural tachycardia syndrome and low peripheral blood flow. Am J Physiol Heart Circ Physiol. 2004; 286(3): H1216-H22. 248. O'Donovan KJ, O'Keeffe DT, Grace PA, Lyons GM. Accelerometer based calf muscle pump activity monitoring. Med Eng Phys. 2005; 27(8): 717-22. 249. Labropoulos N, Tiongson J, Pryor L, Tassiopoulos AK, Kang SS, Ashraf Mansour M, et al. Definition of venous reflux in lower-extremity veins. J Vasc Surg. 2003; 38(4): 793-8. 250. Lurie F, Comerota A, Eklof B, Kistner RL, Labropoulos N, Lohr J, et al. Multicenter assessment of venous reflux by duplex ultrasound. J Vasc Surg. 2012; 55(2): 437-45. 251. Neglen P, Egger JF, 3rd, Olivier J, Raju S. Hemodynamic and clinical impact of ultrasound-derived venous reflux parameters. J Vasc Surg. 2004; 40(2): 303-10.

301

252. Shepherd JT. Role of the veins in the circulation. Circulation. 1966; 33(3): 484-91. 253. Pollack AA, Wood EH. Venous pressure in the saphenous vein at the ankle in man during exercise and changes in posture. J Appl Physiol. 1949; 1(9): 649-62. 254. Arnoldi CC. Venous pressure in the leg of healthy human subjects at rest and during muscular exercise in the nearly erect position. Acta Chir Scand. 1965; 130(6): 570-83. 255. Katz AI, Chen Y, Moreno AH. Flow through a collapsible tube. Experimental analysis and mathematical model. Biophys J. 1969; 9(10): 1261-79. 256. Padberg F. The Physiology and Haemodynamics of the Normal Venous Circulation. In: Gloviczki P, Yao JS, editors. Handbook of Venous Disorders: Arnold; 2002. p. 25-34. 257. Kugler C, Strunk M, Rudofsky G. Venous pressure dynamics of the healthy human leg. Role of muscle activity, joint mobility and anthropometric factors. J Vasc Res. 2001; 38(1): 20-9. 258. Nicolaides AN. Investigation of chronic venous insufficiency: A consensus statement (France, March 5-9, 1997). Circulation. 2000; 102(20): E126-63. 259. Ludbrook J. The musculovenous pumps of the human lower limb. Am Heart J. 1966; 71(5): 635-41. 260. Christopoulos DG, Nicolaides AN, Szendro G, Irvine AT, Bull ML, Eastcott HH. Air-plethysmography and the effect of elastic compression on venous hemodynamics of the leg. J Vasc Surg. 1987; 5(1): 148-59. 261. Maffiuletti NA. Physiological and methodological considerations for the use of neuromuscular electrical stimulation. Eur J Appl Physiol. 2010; 110(2): 223-34. 262. Gardner AM, Fox RH. The venous pump of the human foot--preliminary report. Bristol Med Chir J. 1983; 98(367): 109-12. 263. Franceschi C, Zamboni P, editors. Principles of venous haemodynamics. New York: Novapublishers; 2010. 264. Franceschi C. Dynamic fractionizing of hydrostatic pressure, closed and open shunts, vicarious varicose evolution: how these concepts made the treatment of varices evolve? . Phlebologie. 2003; 56(1): 61-6. 265. Mdez-Herrero A, Gutierrez J, Camblor L, Carreno J, Llaneza J, Rguez-Olay J, et al. The relation among the diameter of the great saphenous vein, clinical state and haemodynamic pattern of the saphenofemoral junction in chronic superficial venous insufficiency. Phlebology. 2007; 22(5): 207-13. 266. Escribano JM, Juan J, Bofill R, Maeso J, Rodriguez-Mori A, Matas M. Durability of reflux-elimination by a minimal invasive CHIVA procedure on patients with varicose veins. A 3-year prospective case study. Eur J Vasc Endovasc Surg. 2003; 25(2): 159-63. 267. Mendoza E, Berger HA. Provokationsmanöver für die duplex-sonographische Diagnostik der Varikosis. Gefässchirurgie. 2001; 6(1): 43-6. 268. van Bemmelen PS, Beach K, Bedford G, Strandness DE, Jr. The mechanism of venous valve closure. Its relationship to the velocity of reverse flow. Arch Surg. 1990; 125(5): 617-9. 269. Franceschi C. [Measures and interpretation of venous flow in stress tests. Manual compression and Parana manoeuver. Dynamic reflux index and Pstakis index]. J Mal Vasc. 1997; 22(2): 91-5. 270. Coleridge-Smith P, Labropoulos N, Partsch H, Myers K, Nicolaides A, Cavezzi A. Duplex ultrasound investigation of the veins in chronic venous disease of the lower limbs--UIP consensus document. Part I. Basic principles. Eur J Vasc Endovasc Surg. 2006; 31(1): 83-92. 271. Mendoza E, Blattler W, Amsler F. Great Saphenous Vein Diameter at the Saphenofemoral Junction and Proximal Thigh as Parameters of Venous Disease Class. Eur J Vasc Endovasc Surg. 2013; 45(1): 76-83. 272. Cappelli M, Molino Lova R, Ermini S, Zamboni P. Hemodynamics of the sapheno-femoral junction. Patterns of reflux and their clinical implications. Int Angiol. 2004; 23(1): 25-8. 273. Cavezzi A, Labropoulos N, Partsch H, Ricci S, Caggiati A, Myers K, et al. Duplex ultrasound investigation of the veins in chronic venous disease of the lower limbs--UIP consensus document. Part II. Anatomy. Eur J Vasc Endovasc Surg. 2006; 31(3): 288-99. 274. Cappelli M, Molino Lova R, Ermini S, Turchi A, Bono G, Franceschi C. [Comparison between the CHIVA cure and stripping in the treatment of varicose veins of the legs: follow-up of 3 years]. J Mal Vasc. 1996; 21(1): 40-6. 275. Fleming BP. Innervation of the microcirculation. J Reconstr Microsurg. 1988; 4(3): 237-40. 276. Parsi K, Partsch H, Rabe E, Ramelet AA. Reticulate eruptions. Part 1: Vascular networks and physiology. Australas J Dermatol. 2011; 52(3): 159-66. 277. Delis KT, Nicolaides AN. Effect of intermittent pneumatic compression of foot and calf on walking distance, hemodynamics, and quality of life in patients with arterial claudication: a prospective randomized controlled study with 1-year follow-up. Ann Surg. 2005; 241(3): 431-41. 278. Folse R, Alexander RH. Directional flow detection for localizing venous valvular incompetency. Surgery. 1970; 67(1): 114-21. 279. Arnoldussen CW, Wittens CH. An imaging approach to deep vein thrombosis and the lower extremity thrombosis classification. Phlebology. 2012; 27 Suppl 1: 143-8. 280. McMullin GM, Coleridge Smith PD. An evaluation of Doppler ultrasound and photoplethysmography in the investigation of venous insufficiency. Aust N Z J Surg. 1992; 62(4): 270-5. 281. Nicolini P. Treatment of primary varicose veins by endovenous obliteration with the VNUS closure system: results of a prospective multicentre study. Eur J Vasc Endovasc Surg. 2005; 29(4): 433-9.

302

282. Weingarten MS, Branas CC, Czeredarczuk M, Schmidt JD, Wolferth CC, Jr. Distribution and quantification of venous reflux in lower extremity chronic venous stasis disease with duplex scanning. J Vasc Surg. 1993; 18(5): 753-9. 283. van Bemmelen PS, Bedford G, Beach K, Strandness DE. Quantitative segmental evaluation of venous valvular reflux with duplex ultrasound scanning. J Vasc Surg. 1989; 10(4): 425-31. 284. Wong JK, Duncan JL, Nichols DM. Whole-leg duplex mapping for varicose veins: observations on patterns of reflux in recurrent and primary legs, with clinical correlation. Eur J Vasc Endovasc Surg. 2003; 25(3): 267-75. 285. Neglen P, Raju S. A comparison between descending phlebography and duplex Doppler investigation in the evaluation of reflux in chronic venous insufficiency: a challenge to phlebography as the "gold standard". J Vasc Surg. 1992; 16(5): 687-93. 286. Labropoulos N, Mansour MA, Kang SS, Gloviczki P, Baker WH. New insights into perforator vein incompetence. Eur J Vasc Endovasc Surg. 1999; 18(3): 228-34. 287. Gooley NA, Sumner DS. Relationship of venous reflux to the site of venous valvular incompetence: implications for venous reconstructive surgery. J Vasc Surg. 1988; 7(1): 50-9. 288. Adam DJ, Naik J, Hartshorne T, Bello M, London NJ. The diagnosis and management of 689 chronic leg ulcers in a single-visit assessment clinic. Eur J Vasc Endovasc Surg. 2003; 25(5): 462-8. 289. Labropoulos N, Leon LR, Jr. Duplex evaluation of venous insufficiency. Semin Vasc Surg. 2005; 18(1): 5-9. 290. Zamboni P, Gianesini S, Menegatti E, Tacconi G, Palazzo A, Liboni A. Great saphenous varicose vein surgery without saphenofemoral junction disconnection. Br J Surg. 2010; 97(6): 820-5. 291. Bjordal R. Simultaneous pressure and flow recordings in varicose veins of the lower extremity. A haemodynamic study of venous dysfunction. Acta Chir Scand. 1970; 136(4): 309-17. 292. Recek C, Pojer H. Ambulatory pressure gradient in the veins of the lower extremity. Vasa. 2000; 29(3): 187-90. 293. Franceschi C, editor. Théorie et pratique de la cure Conservatrice et hémodynamique de l’insuffisance veineuse en ambulatoire: Precy-sous-Thil France Editions de l’Armançon; 1988. 294. Masuda EM, Kistner RL, Eklof B. Prospective study of duplex scanning for venous reflux: comparison of Valsalva and pneumatic cuff techniques in the reverse Trendelenburg and standing positions. J Vasc Surg. 1994; 20(5): 711-20. 295. Klonizakis M, Lingam K, Manning G, Donnelly R. Characterising the time-course of microvascular vasodilator responses in humans using Laser Doppler Fluximetry and iontophoresis. J Pharmacol Toxicol Methods. 2011; 63(1): 115-8. 296. Klonizakis M, Manning G, Donnelly R. Assessment of lower limb microcirculation: exploring the reproducibility and clinical application of laser Doppler techniques. Skin Pharmacol Physiol. 2011; 24(3): 136-43. 297. Christopoulos D, Nicolaides AN, Szendro G. Venous reflux: quantification and correlation with the clinical severity of chronic venous disease. Br J Surg. 1988; 75(4): 352-6. 298. Miyazaki K, Nishibe T, Kudo F, Miyazaki YJ, Nishibe M, Ando M, et al. Hemodynamic changes in stripping operation or saphenofemoral ligation of the greater saphenous vein for primary varicose veins. Ann Vasc Surg. 2004; 18(4): 465-9. 299. Barnes RW, Collicott PE, Sumner DS, Strandness DE, Jr. Noninvasive quantitation of venous hemodynamics in the postphlebitic syndrome. Arch Surg. 1973; 107(5): 807-14. 300. Harada RN, Katz ML, Comerota A. A noninvasive screening test to detect "critical" deep venous reflux. J Vasc Surg. 1995; 22(5): 532-7. 301. Eidson JL, 3rd, Bush RL. Diagnosis and current management of incompetent perforator veins. Semin Vasc Surg. 2010; 23(2): 113-7. 302. Louisy F, Cauquil D, Andre-Deshays C, Schroiff P, Lazerges M, Lafaye C, et al. Air plethysmography: an alternative method for assessing peripheral circulatory adaptations during spaceflights. Eur J Appl Physiol. 2001; 85(3-4): 383-91. 303. Weingarten MS, Czeredarczuk M, Scovell S, Branas CC, Mignogna GM, Wolferth CC, Jr. A correlation of air plethysmography and color-flow-assisted duplex scanning in the quantification of chronic venous insufficiency. J Vasc Surg. 1996; 24(5): 750-4. 304. Delis KT, Gloviczki P, Wennberg PW, Rooke TW, Driscoll DJ. Hemodynamic impairment, venous segmental disease, and clinical severity scoring in limbs with Klippel-Trenaunay syndrome. J Vasc Surg. 2007; 45(3): 561-7. 305. Bays RA, Healy DA, Atnip RG, Neumyer M, Thiele BL. Validation of air plethysmography, photoplethysmography, and duplex ultrasonography in the evaluation of severe venous stasis. J Vasc Surg. 1994; 20(5): 721-7. 306. Darvall KA, Sam RC, Bate GR, Adam DJ, Silverman SH, Bradbury AW. Photoplethysmographic venous refilling times following ultrasound guided foam sclerotherapy for symptomatic superficial venous reflux: relationship with clinical outcomes. Eur J Vasc Endovasc Surg. 2010; 40(2): 267-72. 307. van Bemmelen PS, van Ramshorst B, Eikelboom BC. Photoplethysmography reexamined: lack of correlation with duplex scanning. Surgery. 1992; 112(3): 544-8.

303

308. Fletcher JP, Kershaw LZ, Barker DS, Koutts J, Varnava A. Ultrasound diagnosis of lower limb deep venous thrombosis. Med J Aust. 1990; 153(8): 453-5. 309. Sauerbrei E, Thomson JG, McLachlan MS, Musial J. Observer variation in lower limb venography. J Can Assoc Radiol. 1981; 32(1): 28-9. 310. de Valois JC, van Schaik CC, Verzijlbergen F, van Ramshorst B, Eikelboom BC, Meuwissen OJ. Contrast venography: from gold standard to 'golden backup' in clinically suspected deep vein thrombosis. Eur J Radiol. 1990; 11(2): 131-7. 311. Raju S, Fredericks R. Venous obstruction: an analysis of one hundred thirty-seven cases with hemodynamic, venographic, and clinical correlations. J Vasc Surg. 1991; 14(3): 305-13. 312. Uhl JF, Gillot C, Chahim M. Anatomical variations of the femoral vein. J Vasc Surg. 2010; 52(3): 714-9. 313. Gaitini D. Multimodality imaging of the peripheral venous system. Int J Biomed Imaging. 2007; 2007: 54616. 314. Orbell JH, Smith A, Burnand KG, Waltham M. Imaging of deep vein thrombosis. Br J Surg. 2008; 95(2): 137-46. 315. Kistner RL, Ferris EB, Randhawa G, Kamida C. A method of performing descending venography. J Vasc Surg. 1986; 4(5): 464-8. 316. Garrett JV, Passman MA, Guzman RJ, Dattilo JB, Naslund TC. Expanding options for bedside placement of inferior vena cava filters with intravascular ultrasound when transabdominal duplex ultrasound imaging is inadequate. Ann Vasc Surg. 2004; 18(3): 329-34. 317. Baldt MM, Zontsich T, Stumpflen A, Fleischmann D, Schneider B, Minar E, et al. Deep venous thrombosis of the lower extremity: efficacy of spiral CT venography compared with conventional venography in diagnosis. Radiology. 1996; 200(2): 423-8. 318. Loud PA, Katz DS, Klippenstein DL, Shah RD, Grossman ZD. Combined CT venography and pulmonary angiography in suspected thromboembolic disease: diagnostic accuracy for deep venous evaluation. AJR Am J Roentgenol. 2000; 174(1): 61-5. 319. Marston W, Fish D, Unger J, Keagy B. Incidence of and risk factors for iliocaval venous obstruction in patients with active or healed venous leg ulcers. J Vasc Surg. 2011; 53(5): 1303-8. 320. Oguzkurt L, Tercan F, Pourbagher MA, Kizilkilic O, Turkoz R, Boyvat F. Computed tomography findings in 10 cases of iliac vein compression (May-Thurner) syndrome. Eur J Radiol. 2005; 55(3): 421-5. 321. Stern JB, Abehsera M, Grenet D, Friard S, Couderc LJ, Scherrer A, et al. Detection of pelvic vein thrombosis by magnetic resonance angiography in patients with acute pulmonary embolism and normal lower limb compression ultrasonography. Chest. 2002; 122(1): 115-21. 322. Wolpert LM, Rahmani O, Stein B, Gallagher JJ, Drezner AD. Magnetic resonance venography in the diagnosis and management of May-Thurner syndrome. Vasc Endovascular Surg. 2002; 36(1): 51-7. 323. Cao P, Eckstein HH, De Rango P, Setacci C, Ricco JB, de Donato G, et al. Chapter II: Diagnostic methods. Eur J Vasc Endovasc Surg. 2011; 42 Suppl 2: S13-32. 324. Edelman RR, Koktzoglou I. Unenhanced flow-independent MR venography by using signal targeting alternative radiofrequency and flow-independent relaxation enhancement. Radiology. 2009; 250(1): 236-45. 325. Evans AJ, Sostman HD, Witty LA, Paulson EK, Spritzer CE, Hertzberg BS, et al. Detection of deep venous thrombosis: prospective comparison of MR imaging and sonography. J Magn Reson Imaging. 1996; 6(1): 44- 51. 326. Du J, Thornton FJ, Mistretta CA, Grist TM. Dynamic MR venography: an intrinsic benefit of time- resolved MR angiography. J Magn Reson Imaging. 2006; 24(4): 922-7. 327. Lanzer P, Gross GM, Keller FS, Pohost GM. Sequential 2D inflow venography: initial clinical observations. Magn Reson Med. 1991; 19(2): 470-6. 328. Holtz DJ, Debatin JF, McKinnon GC, Unterweger M, Wildermuth S, von Schulthess GK, et al. MR venography of the calf: value of flow-enhanced time-of-flight echoplanar imaging. AJR Am J Roentgenol. 1996; 166(3): 663-8. 329. Bluemke DA, Wolf RL, Tani I, Tachiki S, McVeigh ER, Zerhouni EA. Extremity veins: evaluation with fast-spin-echo MR venography. Radiology. 1997; 204(2): 562-5. 330. Gallix BP, Achard-Lichere C, Dauzat M, Bruel JM, Lopez FM. Flow-independent magnetic resonance venography of the calf. J Magn Reson Imaging. 2003; 17(4): 421-6. 331. Ruehm SG, Wiesner W, Debatin JF. Pelvic and lower extremity veins: contrast-enhanced three- dimensional MR venography with a dedicated vascular coil-initial experience. Radiology. 2000; 215(2): 421-7. 332. Lebowitz JA, Rofsky NM, Krinsky GA, Weinreb JC. Gadolinium-enhanced body MR venography with subtraction technique. AJR Am J Roentgenol. 1997; 169(3): 755-8. 333. Fraser DG, Moody AR, Davidson IR, Martel AL, Morgan PS. Deep venous thrombosis: diagnosis by using venous enhanced subtracted peak arterial MR venography versus conventional venography. Radiology. 2003; 226(3): 812-20. 334. Zhang ZQ, Liu QH. Three-dimensional nonlinear image reconstruction for microwave biomedical imaging. IEEE Trans Biomed Eng. 2004; 51(3): 544-8.

304

335. Ziyeh S, Spreer J, Rossler J, Strecker R, Hochmuth A, Schumacher M, et al. Parkes Weber or Klippel- Trenaunay syndrome? Non-invasive diagnosis with MR projection angiography. Eur Radiol. 2004; 14(11): 2025-9. 336. Godshall CJ. Computed tomographic angiography allows accurate planning of the setting and technique of open and percutaneous vascular interventions. Am J Surg. 2005; 190(2): 218-20. 337. Gamblin TC, Ashley DW, Burch S, Solis M. A prospective evaluation of a bedside technique for placement of inferior vena cava filters: accuracy and limitations of intravascular ultrasound. Am Surg. 2003; 69(5): 382-6. 338. Conners MS, 3rd, Becker S, Guzman RJ, Passman MA, Pierce R, Kelly T, et al. Duplex scan-directed placement of inferior vena cava filters: a five-year institutional experience. J Vasc Surg. 2002; 35(2): 286-91. 339. Danetz JS, McLafferty RB, Ayerdi J, Gruneiro LA, Ramsey DE, Hodgson KJ. Selective venography versus nonselective venography before vena cava filter placement: evidence for more, not less. J Vasc Surg. 2003; 38(5): 928-34. 340. Aljabri B, MacDonald PS, Satin R, Stein LS, Obrand DI, Steinmetz OK. Incidence of major venous and renal anomalies relevant to aortoiliac surgery as demonstrated by computed tomography. Ann Vasc Surg. 2001; 15(6): 615-8. 341. Masuda EM, Kessler DM, Lurie F, Puggioni A, Kistner RL, Eklof B. The effect of ultrasound-guided sclerotherapy of incompetent perforator veins on venous clinical severity and disability scores. J Vasc Surg. 2006; 43(3): 551-6; discussion 6-7. 342. Howlader MH, Smith PD. Microangiopathy in chronic venous insufficiency: quantitative assessment by capillary microscopy. Eur J Vasc Endovasc Surg. 2003; 26(3): 325-31. 343. Cobbold JF, Patel D, Fitzpatrick JA, Patel N, Crossey MM, Abdalla MS, et al. Accuracy and reliability of microbubble ultrasound measurements for the non-invasive assessment of hepatic fibrosis in chronic hepatitis C. Hepatol Res. 2012; 42(5): 515-22. 344. Shalhoub J, Owen DR, Gauthier T, Monaco C, Leen EL, Davies AH. The use of contrast enhanced ultrasound in carotid arterial disease. Eur J Vasc Endovasc Surg. 2010; 39(4): 381-7. 345. Pierce IT, Gatehouse PD, Kalodiki E, Lattimer C, Geroulakos G, Xu XY, et al. Magnetic resonance venous velocity mapping during intermittent pneumatic compression of the calf and foot. Phlebology. 2011. 346. Pierce IT, Gatehouse PD, Xu XY, Firmin DN. MR phase-contrast velocity mapping methods for measuring venous blood velocity in the deep veins of the calf. J Magn Reson Imaging. 2011; 34(3): 634-44. 347. Veerman DP, Imholz BP, Wieling W, Wesseling KH, van Montfrans GA. Circadian profile of systemic hemodynamics. Hypertension. 1995; 26(1): 55-9. 348. Mehlsen J. Models of the venous system. Stud Health Technol Inform. 2000; 71: 109-17. 349. Snyder MF, Rideout VC. Computer simulation studies of the venous circulation. IEEE Trans Biomed Eng. 1969; 16(4): 325-34. 350. Brown SM, Fennigkoh L, Gerrits R, Hietpas M, Tritt C. A model of venous return while utilizing vacuum assist during cardiopulmonary bypass. J Extra Corpor Technol. 2003; 35(3): 224-9. 351. Buxton GA, Clarke N. Computational phlebology: the simulation of a vein valve. J Biol Phys. 2006; 32(6): 507-21. 352. Zervides C, Narracott AJ, Lawford PV, Hose DR. The role of venous valves in pressure shielding. Biomed Eng Online. 2008; 7: 8. 353. Diaz-Zuccarini V, Narracott AJ, Burriesci G, Zervides C, Rafiroiu D, Jones D, et al. Adaptation and development of software simulation methodologies for cardiovascular engineering: present and future challenges from an end-user perspective. Philos Trans A Math Phys Eng Sci. 2009; 367(1898): 2655-66. 354. Downie SP, Firmin DN, Wood NB, Thom SA, Hughes AD, Wolfe JN, et al. Role of MRI in investigating the effects of elastic compression stockings on the deformation of the superficial and deep veins in the lower leg. J Magn Reson Imaging. 2007; 26(1): 80-5. 355. Downie SP, Raynor SM, Firmin DN, Wood NB, Thom SA, Hughes AD, et al. Effects of elastic compression stockings on wall shear stress in deep and superficial veins of the calf. Am J Physiol Heart Circ Physiol. 2008; 294(5): H2112-20. 356. Chandran K, Yoganathan A, Rittergs S. Biofluid Mechanics: The Human Circulation: Taylor and Francis; 2006. 357. Papaioannou TG, Stefanadis C. Vascular wall shear stress: basic principles and methods. Hellenic J Cardiol. 2005; 46(1): 9-15. 358. Briceno M, Joseph D. Self-lubricated transport of aqueous foams in horizontal conduits. International Journal of Multiphase Flow. 2003; 29(12): 1817-31. 359. Wang Y, Pierce I, Gatehouse P, Wood N, Firmin D, Xu XY. Analysis of flow and wall shear stress in the peroneal veins under external compression based on real-time MR images. Med Eng Phys. 2012; 34(1): 17-27. 360. Karino T, Motomiya M. Flow through a venous valve and its implication for thrombus formation. Thromb Res. 1984; 36(3): 245-57. 361. Sevitt S. The structure and growth of valve-pocket thrombi in femoral veins. J Clin Pathol. 1974; 27(7): 517-28. 362. Matsumoto H, Hasegawa T, Fuse K, Yamamoto M, Saigusa M. A new vascular prosthesis for a small caliber artery. Surgery. 1973; 74(4): 519-23.

305

363. Badylak SF, Freytes DO, Gilbert TW. Extracellular matrix as a biological scaffold material: Structure and function. Acta Biomater. 2009; 5(1): 1-13. 364. Reing JE, Brown BN, Daly KA, Freund JM, Gilbert TW, Hsiong SX, et al. The effects of processing methods upon mechanical and biologic properties of porcine dermal extracellular matrix scaffolds. Biomaterials. 2010; 31(33): 8626-33. 365. Tottey S, Johnson SA, Crapo PM, Reing JE, Zhang L, Jiang H, et al. The effect of source animal age upon extracellular matrix scaffold properties. Biomaterials. 2011; 32(1): 128-36. 366. Brown BN, Barnes CA, Kasick RT, Michel R, Gilbert TW, Beer-Stolz D, et al. Surface characterization of extracellular matrix scaffolds. Biomaterials. 2010; 31(3): 428-37. 367. Tottey S, Corselli M, Jeffries EM, Londono R, Peault B, Badylak SF. Extracellular matrix degradation products and low-oxygen conditions enhance the regenerative potential of perivascular stem cells. Tissue Eng Part A. 2011; 17(1-2): 37-44. 368. Chiu YN, Norris RA, Mahler G, Recknagel A, Butcher JT. Transforming growth factor beta, bone morphogenetic protein, and vascular endothelial growth factor mediate phenotype maturation and tissue remodeling by embryonic valve progenitor cells: relevance for heart valve tissue engineering. Tissue Eng Part A. 2010; 16(11): 3375-83. 369. Gwanmesia P, Ziegler H, Eurich R, Barth M, Kamiya H, Karck M, et al. Opposite effects of transforming growth factor-beta1 and vascular endothelial growth factor on the degeneration of aortic valvular interstitial cell are modified by the extracellular matrix protein fibronectin: implications for heart valve engineering. Tissue Eng Part A. 2010; 16(12): 3737-46. 370. Sacks MS, Schoen FJ, Mayer JE. Bioengineering challenges for heart valve tissue engineering. Annu Rev Biomed Eng. 2009; 11: 289-313. 371. Valentin JE, Stewart-Akers AM, Gilbert TW, Badylak SF. Macrophage participation in the degradation and remodeling of extracellular matrix scaffolds. Tissue Eng Part A. 2009; 15(7): 1687-94. 372. Brown BN, Valentin JE, Stewart-Akers AM, McCabe GP, Badylak SF. Macrophage phenotype and remodeling outcomes in response to biologic scaffolds with and without a cellular component. Biomaterials. 2009; 30(8): 1482-91. 373. Ghanbari H, Viatge H, Kidane AG, Burriesci G, Tavakoli M, Seifalian AM. Polymeric heart valves: new materials, emerging hopes. Trends Biotechnol. 2009; 27(6): 359-67. 374. Sachweh JS, Daebritz SH. Novel "biomechanical" polymeric valve prostheses with special design for aortic and mitral position: a future option for pediatric patients? ASAIO J. 2006; 52(5): 575-80. 375. Christenson EM, Dadsetan M, Wiggins M, Anderson JM, Hiltner A. Poly(carbonate urethane) and poly(ether urethane) biodegradation: in vivo studies. J Biomed Mater Res A. 2004; 69(3): 407-16. 376. Christenson EM, Anderson JM, Hiltner A. Oxidative mechanisms of poly(carbonate urethane) and poly(ether urethane) biodegradation: in vivo and in vitro correlations. J Biomed Mater Res A. 2004; 70(2): 245-55. 377. Kannan RY, Salacinski HJ, Odlyha M, Butler PE, Seifalian AM. The degradative resistance of polyhedral oligomeric silsesquioxane nanocore integrated polyurethanes: an in vitro study. Biomaterials. 2006; 27(9): 1971-9. 378. Ghanbari H, Kidane AG, Burriesci G, Ramesh B, Darbyshire A, Seifalian AM. The anti-calcification potential of a silsesquioxane nanocomposite polymer under in vitro conditions: potential material for synthetic leaflet heart valve. Acta Biomater. 2010; 6(11): 4249-60. 379. Pinchuk L, Wilson GJ, Barry JJ, Schoephoerster RT, Parel JM, Kennedy JP. Medical applications of poly(styrene-block-isobutylene-block-styrene) ("SIBS"). Biomaterials. 2008; 29(4): 448-60. 380. Gallocher SL, Aguirre AF, Kasyanov V, Pinchuk L, Schoephoerster RT. A novel polymer for potential use in a trileaflet heart valve. J Biomed Mater Res B. 2006; 79B(2): 325-34. 381. Wang Q, McGoron AJ, Bianco R, Kato Y, Pinchuk L, Schoephoerster RT. In-vivo assessment of a novel polymer (SIBS) trileaflet heart valve. J Heart Valve Dis. 2010; 19(4): 499-505. 382. Schmidt D, Dijkman PE, Driessen-Mol A, Stenger R, Mariani C, Puolakka A, et al. Minimally-invasive implantation of living tissue engineered heart valves: a comprehensive approach from autologous vascular cells to stem cells. J Am Coll Cardiol. 2010; 56(6): 510-20. 383. L'Heureux N, McAllister TN, de la Fuente LM. Tissue-engineered blood vessel for adult arterial revascularization. New Engl J Med. 2007; 357(14): 1451-3. 384. McAllister TN, Maruszewski M, Garrido SA, Wystrychowski W, Dusserre N, Marini A, et al. Effectiveness of haemodialysis access with an autologous tissue-engineered vascular graft: a multicentre cohort study. Lancet. 2009; 373(9673): 1440-6. 385. Teebken OE, Bader A, Steinhoff G, Haverich A. Tissue engineering of vascular grafts: Human cell seeding of decellularised porcine matrix. Eur J Vasc Endovasc. 2000; 19(4): 381-6. 386. Sun M, Zhou P, Pan LF, Liu S, Yang HX. Enhanced cell affinity of the silk fibroin- modified PHBHHx material. J Mater Sci-Mater M. 2009; 20(8): 1743-51. 387. Ramcharitar S, Serruys PW. Fully Biodegradable Coronary Stents Progress to Date. Am J Cardiovasc Drug. 2008; 8(5): 305-14. 388. Maleti O, Lugli M. Neovalve construction in postthrombotic syndrome. J Vasc Surg. 2006; 43(4): 794-9.

306

389. Park J, Kurosawa S, Takai M, Ishihara K. Antibody immobilization to phospholipid polymer layer on gold substrate of quartz crystal microbalance immunosensor. Colloids Surf B Biointerfaces. 2007; 55(2): 164-72. 390. Byambaa B, Konno T, Ishihara K. Detachment of cells adhered on the photoreactive phospholipid polymer surface by photoirradiation and their functionality. Colloids Surf B Biointerfaces. 2013; 103: 489-95. 391. Kim HI, Ishihara K, Lee S, Seo JH, Kim HY, Suh D, et al. Tissue response to poly(L-lactic acid)-based blend with phospholipid polymer for biodegradable cardiovascular stents. Biomaterials. 2011; 32(9): 2241-7. 392. Soletti L, Nieponice A, Hong Y, Ye SH, Stankus JJ, Wagner WR, et al. In vivo performance of a phospholipid-coated bioerodable elastomeric graft for small-diameter vascular applications. J Biomed Mater Res A. 2011; 96(2): 436-48. 393. Ishihara K, Nomura H, Mihara T, Kurita K, Iwasaki Y, Nakabayashi N. Why do phospholipid polymers reduce protein adsorption? J Biomed Mater Res. 1998; 39(2): 323-30. 394. Ishihara K, Ziats NP, Tierney BP, Nakabayashi N, Anderson JM. Protein adsorption from human plasma is reduced on phospholipid polymers. J Biomed Mater Res. 1991; 25(11): 1397-407. 395. Liu Y, Inoue Y, Mahara A, Kakinoki S, Yamaoka T, Ishihara K. Durable modification of segmented polyurethane for elastic blood-contacting devices by graft-type 2-methacryloyloxyethyl phosphorylcholine copolymer. J Biomater Sci Polym Ed. 2014: 1-16. 396. Ye SH, Johnson Jr CA, Woolley JR, Oh H-I, Gamble LJ, Ishihara K, et al. Surface modification of a titanium alloy with a phospholipid polymer prepared by a plasma-induced grafting technique to improve surface thromboresistance. Colloids and Surfaces B: Biointerfaces. 2009; 74(1): 96-102. 397. Duncan R. Polymer conjugates as anticancer nanomedicines. Nat Rev Cancer. 2006; 6(9): 688-701. 398. Vicent MJ, Greco F, Nicholson RI, Paul A, Griffiths PC, Duncan R. Polymer therapeutics designed for a combination therapy of hormone-dependent cancer. Angew Chem Int Ed Engl. 2005; 44(26): 4061-6. 399. Barz M, Tarantola M, Fischer K, Schmidt M, Luxenhofer R, Janshoff A, et al. From defined reactive diblock copolymers to functional HPMA-based self-assembled nanoaggregates. Biomacromolecules. 2008; 9(11): 3114-8. 400. van Oeveren W, Haan J, Lagerman P, Schoen P. Comparison of coagulation activity tests in vitro for selected biomaterials. Artif Organs. 2002; 26(6): 506-11. 401. Henkelman S, Rakhorst G, Blanton J, Van Oeveren W. Standardization of incubation conditions for hemolysis testing of biomaterials. Materials Science and Engineering C - Biomimetic and Supramolecular Systems. 2009; 29: 1650-4. 402. Teng Z, Tang D, Zheng J, Woodard PK, Hoffman AH. An experimental study on the ultimate strength of the adventitia and media of human atherosclerotic carotid arteries in circumferential and axial directions. J Biomech. 2009; 42(15): 2535-9. 403. Assoul N, Flaud P, Chaouat M, Letourneur D, Bataille I. Mechanical properties of rat thoracic and abdominal aortas. J Biomech. 2008; 41(10): 2227-36. 404. Holzapfel GA, Sommer G, Gasser CT, Regitnig P. Determination of layer-specific mechanical properties of human coronary arteries with nonatherosclerotic intimal thickening and related constitutive modeling. Am J Physiol Heart Circ Physiol. 2005; 289(5): H2048-58. 405. Rossmann JS. Elastomechanical properties of bovine veins. J Mech Behav Biomed Mater. 2010; 3(2): 210- 5. 406. Wang SK, Qiu YC, Phifer TJ, Delaria GA, Tu R, Quijano RC, et al. In vitro performance of venous valve prostheses. An experimental model study. ASAIO J. 1992; 38(3): M213-5.

307

APPENDICES

Appendix 1

308

309

310

311

312

313

314

315

316

317

318

319

320

321

322

323

324

325

326

327

328

329

330

331

Appendix 2

332

Appendix 3

Table 1: Subject 1 maximum and mean velocities from ultrasound data

Subject position Calf Respiration Position Maximum Mean Mean

squeeze relative to velocity velocity velocity in

valve (cm/s) (cm/s) calf squeeze

(cm/s)

Standing No Normal 2cm proximal 6.60 4.40 -

Standing No Normal Within leaflets 17.46 4.92 -

Standing No Normal 1cm distal 5.02 2.70 -

Standing Yes Normal 2cm proximal 57.50 18.39 27.35

Standing Yes Normal Within leaflets 53.50 18.20 27.88

Standing Yes Normal 1cm distal 43.50 14.17 20.82

Standing Yes Breath hold 2cm proximal 59.00 18.73 31.84

Standing Yes Breath hold Within leaflets 48.50 16.69 27.94

Standing Yes Breath hold 1cm distal 40.00 13.16 22.13

Semi-recumbent No Normal 2cm proximal 21.91 15.14 -

Semi-recumbent No Normal Within leaflets 21.43 12.65 -

Semi-recumbent No Normal 1cm distal 20.48 10.33 -

Semi-recumbent Yes Normal 2cm proximal 87.62 32.21 42.35

Semi-recumbent Yes Normal Within leaflets 65.24 28.42 38.16

333

Semi-recumbent Yes Normal 1cm distal 74.76 25.51 40.75

Semi-recumbent Yes Breath hold 2cm proximal 91.05 27.88 45.93

Semi-recumbent Yes Breath hold Within leaflets 91.05 32.36 47.82

Semi-recumbent Yes Breath hold 1cm distal 63.50 23.96 29.59

Table 2: Subject 2 maximum and mean velocities from ultrasound data

Subject position Calf Respiration Position Maximum Mean Mean

squeeze relative to velocity velocity velocity in

valve (cm/s) (cm/s) calf squeeze

(cm/s)

Standing No Normal 2cm proximal 18.60 6.41 -

Standing No Normal Within leaflets 15.44 7.16 -

Standing No Normal 1cm distal 11.00 4.09 -

Standing Yes Normal 2cm proximal 36.67 10.18 26.22

Standing Yes Normal Within leaflets 40.35 9.96 21.39

Standing Yes Normal 1cm distal 32.63 7.98 16.67

Standing Yes Breath hold 2cm proximal 34.04 8.36 21.76

Standing Yes Breath hold Within leaflets 37.54 10.99 23.37

Standing Yes Breath hold 1cm distal 26.32 7.47 17.19

Semi-recumbent No Normal 2cm proximal 26.32 12.08 -

334

Semi-recumbent No Normal Within leaflets 25.61 13.11 -

Semi-recumbent No Normal 1cm distal 29.83 20.93 -

Semi-recumbent Yes Normal 2cm proximal 58.95 22.28 35.48

Semi-recumbent Yes Normal Within leaflets 54.74 17.32 29.62

Semi-recumbent Yes Normal 1cm distal 42.81 16.10 26.14

Semi-recumbent Yes Breath hold 2cm proximal 47.37 16.50 30.23

Semi-recumbent Yes Breath hold Within leaflets 71.93 25.92 36.15

Semi-recumbent Yes Breath hold 1cm distal 62.81 21.23 33.87

Table 3: Subject 3 maximum and mean velocities from ultrasound data

Subject position Calf Respiration Position Maximum Mean Mean

squeeze relative to velocity velocity velocity in

valve (cm/s) (cm/s) calf squeeze

(cm/s)

Standing No Normal 2cm proximal 16.84 6.51 -

Standing No Normal Within leaflets 18.11 8.72 -

Standing No Normal 1cm distal 12.21 5.67 -

Standing Yes Normal 2cm proximal 27.79 11.19 16.64

Standing Yes Normal Within leaflets 41.68 17.36 25.89

Standing Yes Normal 1cm distal 15.58 8.17 11.57

335

Standing Yes Breath hold 2cm proximal 28.63 9.87 13.08

Standing Yes Breath hold Within leaflets 30.32 13.85 18.56

Standing Yes Breath hold 1cm distal 21.77 9.88 1.65

Semi-recumbent No Normal 2cm proximal 28.21 24.53 -

Semi-recumbent No Normal Within leaflets 24.00 18.11 -

Semi-recumbent No Normal 1cm distal 12.00 8.42 -

Semi-recumbent Yes Normal 2cm proximal 76.50 29.83 47.85

Semi-recumbent Yes Normal Within leaflets 60.00 22.29 32.19

Semi-recumbent Yes Normal 1cm distal 36.00 17.05 21.16

Semi-recumbent Yes Breath hold 2cm proximal 67.00 26.43 44.06

Semi-recumbent Yes Breath hold Within leaflets 64.00 26.41 41.19

Semi-recumbent Yes Breath hold 1cm distal 58.00 21.49 38.55

Table 4: Subject 4 maximum and mean velocities from ultrasound data

Subject position Calf Respiration Position Maximum Mean Mean

squeeze relative to velocity velocity velocity in

valve (cm/s) (cm/s) calf squeeze

(cm/s)

Standing No Normal 2cm proximal 11.37 5.48 -

Standing No Normal Within leaflets 12.94 7.99 -

336

Standing No Normal 1cm distal 18.43 3.12 -

Standing Yes Normal 2cm proximal 54.90 6.83 25.63

Standing Yes Normal Within leaflets 52.94 7.97 28.20

Standing Yes Normal 1cm distal 49.41 7.47 22.32

Standing Yes Breath hold 2cm proximal 51.77 5.92 23.16

Standing Yes Breath hold Within leaflets 47.84 6.77 21.59

Standing Yes Breath hold 1cm distal 42.69 6.03 20.40

Semi-recumbent No Normal 2cm proximal 28.35 15.11 -

Semi-recumbent No Normal Within leaflets 37.26 21.47 -

Semi-recumbent No Normal 1cm distal 29.02 14.36 -

Semi-recumbent Yes Normal 2cm proximal 100.39 22.14 45.57

Semi-recumbent Yes Normal Within leaflets 102.75 24.27 46.46

Semi-recumbent Yes Normal 1cm distal 85.49 27.59 48.21

Semi-recumbent Yes Breath hold 2cm proximal 105.88 23.81 51.37

Semi-recumbent Yes Breath hold Within leaflets 91.37 23.06 49.08

Semi-recumbent Yes Breath hold 1cm distal 91.37 18.71 43.28

Table 5: Subject 5 maximum and mean velocities from ultrasound data

Subject position Calf Respiration Position Maximum Mean Mean

squeeze relative to velocity velocity velocity in

337

valve (cm/s) (cm/s) calf squeeze

(cm/s)

Standing No Normal 2cm proximal 16.86 8.28 -

Standing No Normal Within leaflets 17.65 6.80 -

Standing No Normal 1cm distal 11.77 4.07 -

Standing Yes Normal 2cm proximal 36.03 11.97 17.80

Standing Yes Normal Within leaflets 32.67 11.33 18.37

Standing Yes Normal 1cm distal 17.60 4.08 10.19

Standing Yes Breath hold 2cm proximal 41.11 10.80 23.07

Standing Yes Breath hold Within leaflets 37.78 12.13 24.56

Standing Yes Breath hold 1cm distal 25.30 6.44 14.35

Semi-recumbent No Normal 2cm proximal 31.11 14.34 -

Semi-recumbent No Normal Within leaflets 15.93 9.42 -

Semi-recumbent No Normal 1cm distal 22.59 16.60 -

Semi-recumbent Yes Normal 2cm proximal 93.23 39.24 61.32

Semi-recumbent Yes Normal Within leaflets 85.85 32.79 44.65

Semi-recumbent Yes Normal 1cm distal 80.00 27.46 50.25

Semi-recumbent Yes Breath hold 2cm proximal 85.93 44.58 58.96

Semi-recumbent Yes Breath hold Within leaflets 57.78 25.04 39.49

Semi-recumbent Yes Breath hold 1cm distal 68.89 24.91 46.97

338

Table 6: Subject 6 maximum and mean velocities from ultrasound data

Subject position Calf Respiration Position Maximum Mean Mean

squeeze relative to velocity velocity velocity in

valve (cm/s) (cm/s) calf squeeze

(cm/s)

Standing No Normal 2cm proximal 11.14 5.62 -

Standing No Normal Within leaflets 13.17 6.14 -

Standing No Normal 1cm distal 9.62 3.96 -

Standing Yes Normal 2cm proximal 27.85 7.33 18.31

Standing Yes Normal Within leaflets 29.00 11.91 19.99

Standing Yes Normal 1cm distal 24.81 7.95 13.67

Standing Yes Breath hold 2cm proximal 16.71 4.80 11.56

Standing Yes Breath hold Within leaflets 29.37 8.55 19.58

Standing Yes Breath hold 1cm distal 27.85 6.89 15.97

Semi-recumbent No Normal 2cm proximal 16.20 9.67 -

Semi-recumbent No Normal Within leaflets 18.73 9.03 -

Semi-recumbent No Normal 1cm distal 17.72 8.89 -

Semi-recumbent Yes Normal 2cm proximal 48.57 17.64 24.84

Semi-recumbent Yes Normal Within leaflets 45.31 13.06 23.50

339

Semi-recumbent Yes Normal 1cm distal 34.69 13.34 21.33

Semi-recumbent Yes Breath hold 2cm proximal 49.39 19.56 33.66

Semi-recumbent Yes Breath hold Within leaflets 51.43 19.69 33.24

Semi-recumbent Yes Breath hold 1cm distal 41.63 13.70 20.75

Table 7: Subject 7 maximum and mean velocities from ultrasound data

Subject position Calf Respiration Position Maximum Mean Mean

squeeze relative to velocity velocity velocity in

valve (cm/s) (cm/s) calf squeeze

(cm/s)

Standing No Normal 2cm proximal 22.75 15.10 -

Standing No Normal Within leaflets 15.69 10.42 -

Standing No Normal 1cm distal 16.47 9.78 -

Standing Yes Normal 2cm proximal 40.78 12.67 21.05

Standing Yes Normal Within leaflets 34.90 11.04 18.03

Standing Yes Normal 1cm distal 30.20 7.50 15.06

Standing Yes Breath hold 2cm proximal 52.94 15.38 24.68

Standing Yes Breath hold Within leaflets 33.73 11.04 17.78

Standing Yes Breath hold 1cm distal 37.65 6.87 16.71

Semi-recumbent No Normal 2cm proximal 18.43 12.30 -

340

Semi-recumbent No Normal Within leaflets 17.65 10.93 -

Semi-recumbent No Normal 1cm distal 15.69 8.70 -

Semi-recumbent Yes Normal 2cm proximal 100.00 26.13 54.57

Semi-recumbent Yes Normal Within leaflets 101.57 29.73 53.34

Semi-recumbent Yes Normal 1cm distal 94.51 24.27 48.48

Semi-recumbent Yes Breath hold 2cm proximal 113.53 31.04 59.24

Semi-recumbent Yes Breath hold Within leaflets 127.06 39.16 69.84

Semi-recumbent Yes Breath hold 1cm distal 125.88 33.35 72.10

Appendix 4

Table 1: Subject 1 maximum and mean velocities from ultrasound data

Subject position Calf Respiration Position Maximum Mean Mean

squeeze relative to velocity velocity velocity in

valve (cm/s) (cm/s) calf squeeze

(cm/s)

Standing No Normal 2cm proximal 6.04 4.01 -

Standing No Normal Within leaflets 17.12 4.99 -

Standing No Normal 1cm distal 5.56 2.29 -

341

Standing Yes Normal 2cm proximal 55.12 20.70 26.12

Standing Yes Normal Within leaflets 26.19 9.52 12.58

Standing Yes Normal 1cm distal 41.95 19.07 22.67

Standing Yes Breath hold 2cm proximal 59.51 20.13 32.58

Standing Yes Breath hold Within leaflets 46.83 16.07 31.76

Standing Yes Breath hold 1cm distal 38.05 17.88 24.22

Semi-recumbent No Normal 2cm proximal 18.18 13.73 -

Semi-recumbent No Normal Within leaflets 25.26 13.93 -

Semi-recumbent No Normal 1cm distal 25.00 16.21 -

Semi-recumbent Yes Normal 2cm proximal 73.68 32.95 48.00

Semi-recumbent Yes Normal Within leaflets 71.80 32.75 44.51

Semi-recumbent Yes Normal 1cm distal 85.41 31.60 46.77

Semi-recumbent Yes Breath hold 2cm proximal 110.00 37.98 52.98

Semi-recumbent Yes Breath hold Within leaflets 92.00 29.26 60.13

Semi-recumbent Yes Breath hold 1cm distal 64.00 24.56 36.30

Table 2: Subject 2 maximum and mean velocities from ultrasound data

Subject position Calf Respiration Position Maximum Mean Mean

squeeze relative to velocity velocity velocity in

valve (cm/s) (cm/s) calf squeeze

342

(cm/s)

Standing No Normal 2cm proximal 19.26 5.96 -

Standing No Normal Within leaflets 17.36 10.79 -

Standing No Normal 1cm distal 12.36 6.70 -

Standing Yes Normal 2cm proximal 40.71 13.98 26.53

Standing Yes Normal Within leaflets 31.34 11.51 14.72

Standing Yes Normal 1cm distal 45.28 12.36 22.48

Standing Yes Breath hold 2cm proximal 37.36 12.77 22.34

Standing Yes Breath hold Within leaflets 40.00 14.85 24.93

Standing Yes Breath hold 1cm distal 29.23 9.77 18.58

Semi-recumbent No Normal 2cm proximal 29.06 16.26 -

Semi-recumbent No Normal Within leaflets 23.64 13.61 -

Semi-recumbent No Normal 1cm distal 32.22 22.68 -

Semi-recumbent Yes Normal 2cm proximal 64.62 28.46 41.77

Semi-recumbent Yes Normal Within leaflets 60.00 20.38 34.83

Semi-recumbent Yes Normal 1cm distal 45.39 18.56 33.55

Semi-recumbent Yes Breath hold 2cm proximal 49.81 19.00 33.30

Semi-recumbent Yes Breath hold Within leaflets 78.49 28.07 51.38

Semi-recumbent Yes Breath hold 1cm distal 70.77 23.12 44.61

343

Table 3: Subject 3 maximum and mean velocities from ultrasound data

Subject position Calf Respiration Position Maximum Mean Mean

squeeze relative to velocity velocity velocity in

valve (cm/s) (cm/s) calf squeeze

(cm/s)

Standing No Normal 2cm proximal 17.55 8.68 -

Standing No Normal Within leaflets 19.57 7.77 -

Standing No Normal 1cm distal 11.92 6.58 -

Standing Yes Normal 2cm proximal 41.36 15.63 27.70

Standing Yes Normal Within leaflets 34.43 15.34 24.35

Standing Yes Normal 1cm distal 17.67 5.96 10.30

Standing Yes Breath hold 2cm proximal 44.80 14.72 25.84

Standing Yes Breath hold Within leaflets 40.40 15.11 27.78

Standing Yes Breath hold 1cm distal 28.80 8.85 15.17

Semi-recumbent No Normal 2cm proximal 23.20 16.01 -

Semi-recumbent No Normal Within leaflets 17.20 9.00 -

Semi-recumbent No Normal 1cm distal 23.92 17.57 -

Semi-recumbent Yes Normal 2cm proximal 99.68 49.31 83.16

Semi-recumbent Yes Normal Within leaflets 89.23 32.18 55.07

Semi-recumbent Yes Normal 1cm distal 93.33 35.04 61.52

344

Semi-recumbent Yes Breath hold 2cm proximal 79.49 28.93 55.98

Semi-recumbent Yes Breath hold Within leaflets 64.26 26.35 42.81

Semi-recumbent Yes Breath hold 1cm distal 91.61 49.06 67.87

Table 4: Subject 4 maximum and mean velocities from ultrasound data

Subject position Calf Respiration Position Maximum Mean Mean

squeeze relative to velocity velocity velocity in

valve (cm/s) (cm/s) calf squeeze

(cm/s)

Standing No Normal 2cm proximal 24.83 15.32 -

Standing No Normal Within leaflets 16.21 10.48 -

Standing No Normal 1cm distal 18.28 10.29 -

Standing Yes Normal 2cm proximal 45.97 13.92 23.88

Standing Yes Normal Within leaflets 37.90 11.94 21.66

Standing Yes Normal 1cm distal 33.45 9.43 19.88

Standing Yes Breath hold 2cm proximal 57.19 18.39 36.44

Standing Yes Breath hold Within leaflets 35.00 11.50 19.22

Standing Yes Breath hold 1cm distal 40.00 9.11 21.53

Semi-recumbent No Normal 2cm proximal 19.31 12.89 -

Semi-recumbent No Normal Within leaflets 17.00 11.83 -

345

Semi-recumbent No Normal 1cm distal 17.93 10.52 -

Semi-recumbent Yes Normal 2cm proximal 109.31 29.86 64.94

Semi-recumbent Yes Normal Within leaflets 112.98 36.83 65.16

Semi-recumbent Yes Normal 1cm distal 100.00 24.83 52.54

Semi-recumbent Yes Breath hold 2cm proximal 137.44 33.57 74.56

Semi-recumbent Yes Breath hold Within leaflets 138.50 40.45 75.62

Semi-recumbent Yes Breath hold 1cm distal 122.53 33.11 62.39

Table 5: Subject 5 maximum and mean velocities from ultrasound data

Subject position Calf Respiration Position Maximum Mean Mean

squeeze relative to velocity velocity velocity in

valve (cm/s) (cm/s) calf squeeze

(cm/s)

Standing No Normal 2cm proximal 10.91 6.86 -

Standing No Normal Within leaflets 13.33 6.15 -

Standing No Normal 1cm distal 10.91 4.43 -

Standing Yes Normal 2cm proximal 28.93 9.12 17.68

Standing Yes Normal Within leaflets 32.00 12.81 23.37

Standing Yes Normal 1cm distal 28.15 7.81 15.71

Standing Yes Breath hold 2cm proximal 15.00 5.31 11.68

346

Standing Yes Breath hold Within leaflets 31.64 10.66 19.10

Standing Yes Breath hold 1cm distal 29.29 8.26 15.64

Semi-recumbent No Normal 2cm proximal 17.82 10.89 -

Semi-recumbent No Normal Within leaflets 19.27 9.25 -

Semi-recumbent No Normal 1cm distal 17.90 8.58 -

Semi-recumbent Yes Normal 2cm proximal 52.17 18.58 29.86

Semi-recumbent Yes Normal Within leaflets 50.59 14.86 25.79

Semi-recumbent Yes Normal 1cm distal 36.06 14.40 22.59

Semi-recumbent Yes Breath hold 2cm proximal 45.00 16.03 27.39

Semi-recumbent Yes Breath hold Within leaflets 54.12 22.62 33.84

Semi-recumbent Yes Breath hold 1cm distal 52.46 20.74 35.44

Table 6: Subject 6 maximum and mean velocities from ultrasound data

Subject position Calf Respiration Position Maximum Mean Mean

squeeze relative to velocity velocity velocity in

valve (cm/s) (cm/s) calf squeeze

(cm/s)

Standing No Normal 2cm proximal 18.31 9.95 -

Standing No Normal Within leaflets 17.25 9.58 -

Standing No Normal 1cm distal 11.93 5.61 -

347

Standing Yes Normal 2cm proximal 16.67 8.07 11.38

Standing Yes Normal Within leaflets 44.14 18.65 28.79

Standing Yes Normal 1cm distal 29.12 11.49 15.42

Standing Yes Breath hold 2cm proximal 30.18 10.58 18.36

Standing Yes Breath hold Within leaflets 30.69 15.00 19.59

Standing Yes Breath hold 1cm distal 22.69 10.71 19.37

Semi-recumbent No Normal 2cm proximal 28.00 25.82 -

Semi-recumbent No Normal Within leaflets 26.43 19.11 -

Semi-recumbent No Normal 1cm distal 12.36 9.10 -

Semi-recumbent Yes Normal 2cm proximal 38.57 17.64 24.90

Semi-recumbent Yes Normal Within leaflets 62.50 24.64 40.14

Semi-recumbent Yes Normal 1cm distal 82.18 31.89 50.43

Semi-recumbent Yes Breath hold 2cm proximal 74.18 31.13 47.49

Semi-recumbent Yes Breath hold Within leaflets 71.43 29.71 44.60

Semi-recumbent Yes Breath hold 1cm distal 63.27 24.99 40.74

Table 7: Subject 7 maximum and mean velocities from ultrasound data

Subject position Calf Respiration Position Maximum Mean Mean

squeeze relative to velocity velocity velocity in

valve (cm/s) (cm/s) calf squeeze

348

(cm/s)

Standing No Normal 2cm proximal 12.542 5.83 -

Standing No Normal Within leaflets 12.203 8.16 -

Standing No Normal 1cm distal 19 5.33 -

Standing Yes Normal 2cm proximal 49.84 9.80 23.56

Standing Yes Normal Within leaflets 56.95 9.74 30.47

Standing Yes Normal 1cm distal 61.03 10.30 24.57

Standing Yes Breath hold 2cm proximal 56.21 11.43 25.22

Standing Yes Breath hold Within leaflets 51.00 8.50 21.18

Standing Yes Breath hold 1cm distal 53.22 8.20 22.63

Semi-recumbent No Normal 2cm proximal 30.17 16.49 -

Semi-recumbent No Normal Within leaflets 41.40 22.86 -

Semi-recumbent No Normal 1cm distal 30.51 16.17 -

Semi-recumbent Yes Normal 2cm proximal 92.54 29.49 53.45

Semi-recumbent Yes Normal Within leaflets 112.54 27.40 52.21

Semi-recumbent Yes Normal 1cm distal 110.69 28.69 50.42

Semi-recumbent Yes Breath hold 2cm proximal 120.70 29.86 55.02

Semi-recumbent Yes Breath hold Within leaflets 97.63 22.65 42.85

Semi-recumbent Yes Breath hold 1cm distal 102.07 24.99 50.53

349

Appendix 5

350

351

4. van den Bos R, Arends L, Kockaert M, Neumann M, Nijsten T. Endovenous therapies of lower extremity varicosities: a meta-analysis. J Vasc Surg. 2009; 49(1): 230-9. 5. Shepherd AC, Gohel MS, Brown LC, Metcalfe MJ, Hamish M, Davies AH. Randomized clinical trial of VNUS ClosureFAST radiofrequency ablation versus laser for varicose veins. Br J Surg. 2010; 97(6): 810-8. 6. Raju S, Neglen P. Clinical practice. Chronic venous insufficiency and varicose veins. N Engl J Med. 2009; 360(22): 2319-27. 7. Grabs AJ, Wakely MC, Nyamekye I, Ghauri AS, Poskitt KR. Colour duplex ultrasonography in the rational management of chronic venous leg ulcers. Br J Surg. 1996; 83(10): 1380-2. 8. Hanrahan LM, Araki CT, Rodriguez AA, Kechejian GJ, LaMorte WW, Menzoian JO. Distribution of valvular incompetence in patients with venous stasis ulceration. J Vasc Surg. 1991; 13(6): 805-11; discussion 11-2. 9. Meissner MH. Lower extremity venous anatomy. Semin Intervent Radiol. 2005; 22(3): 147-56. 10. Alguire PC, Mathes BM. Chronic venous insufficiency and venous ulceration. J Gen Intern Med. 1997; 12(6): 374-83. 11. Meissner MH. Lower extremity venous anatomy. Seminars in Interventional Radiology. 2005; 22(3): 147- 56. 12. Alimi YS, Barthelemy P, Juhan C. Venous pump of the calf: a study of venous and muscular pressures. Journal of Vascular Surgery. 1994; 20(5): 728-35. 13. Eberhardt RT, Raffetto JD. Chronic venous insufficiency. Circulation. 2005; 111(18): 2398-409. 14. Beebe-Dimmer JL, Pfeifer JR, Engle JS, Schottenfeld D. The epidemiology of chronic venous insufficiency and varicose veins. Ann Epidemiol. 2005; 15(3): 175-84. 15. Porter JM, Moneta GL. Reporting standards in venous disease: an update. International Consensus Committee on Chronic Venous Disease. J Vasc Surg. 1995; 21(4): 635-45. 16. Neglen P. Chronic deep venous obstruction: definition, prevalence, diagnosis, management. Phlebology. 2008; 23(4): 149-57. 17. Gaweesh AS, Kayed MH, Gaweesh TY, Shalhoub J, Davies AH, Khamis HM. Underlying deep venous abnormalities in patients with unilateral chronic venous disease. Phlebology. 2012. 18. Meissner MH, Eklof B, Smith PC, Dalsing MC, DePalma RG, Gloviczki P, et al. Secondary chronic venous disorders. J Vasc Surg. 2007; 46 Suppl S: 68S-83S. 19. Moher D, Liberati A, Tetzlaff J, Altman DG. Preferred reporting items for systematic reviews and meta- analyses: the PRISMA statement. BMJ. 2009; 339: b2535. 20. Johnson BF, Manzo RA, Bergelin RO, Strandness DE, Jr. Relationship between changes in the deep venous system and the development of the postthrombotic syndrome after an acute episode of lower limb deep vein thrombosis: a one- to six-year follow-up. J Vasc Surg. 1995; 21(2): 307-12; discussion 13. 21. Labropoulos N, Waggoner T, Sammis W, Samali S, Pappas PJ. The effect of venous thrombus location and extent on the development of post-thrombotic signs and symptoms. J Vasc Surg. 2008; 48(2): 407-12. 22. Tick LW, Kramer MH, Rosendaal FR, Faber WR, Doggen CJ. Risk factors for post-thrombotic syndrome in patients with a first deep venous thrombosis. J Thromb Haemost. 2008; 6(12): 2075-81. 23. Yamaki T, Nozaki M. Patterns of venous insufficiency after an acute deep vein thrombosis. J Am Coll Surg. 2005; 201(2): 231-8. 24. van Bemmelen PS, Bedford G, Beach K, Strandness DE, Jr. Status of the valves in the superficial and deep venous system in chronic venous disease. Surgery. 1991; 109(6): 730-4. 25. Cornwall JV, Dore CJ, Lewis JD. Leg ulcers: epidemiology and aetiology. Br J Surg. 1986; 73(9): 693-6. 26. Darke SG, Penfold C. Venous ulceration and saphenous ligation. Eur J Vasc Surg. 1992; 6(1): 4-9. 27. Lees TA, Lambert D. Patterns of venous reflux in limbs with skin changes associated with chronic venous insufficiency. Br J Surg. 1993; 80(6): 725-8. 28. Shami SK, Sarin S, Cheatle TR, Scurr JH, Smith PD. Venous ulcers and the superficial venous system. J Vasc Surg. 1993; 17(3): 487-90. 29. Labropoulos N, Leon M, Geroulakos G, Volteas N, Chan P, Nicolaides AN. Venous hemodynamic abnormalities in patients with leg ulceration. Am J Surg. 1995; 169(6): 572-4. 30. Myers KA, Ziegenbein RW, Zeng GH, Matthews PG. Duplex ultrasonography scanning for chronic venous disease: patterns of venous reflux. J Vasc Surg. 1995; 21(4): 605-12. 31. Labropoulos N, Delis K, Nicolaides AN, Leon M, Ramaswami G. The role of the distribution and anatomic extent of reflux in the development of signs and symptoms in chronic venous insufficiency. J Vasc Surg. 1996; 23(3): 504-10. 32. Welch HJ, Young CM, Semegran AB, Iafrati MD, Mackey WC, O'Donnell TF, Jr. Duplex assessment of venous reflux and chronic venous insufficiency: the significance of deep venous reflux. J Vasc Surg. 1996; 24(5): 755-62. 33. Labropoulos N, Giannoukas AD, Delis K, Mansour MA, Kang SS, Nicolaides AN, et al. Where does venous reflux start? J Vasc Surg. 1997; 26(5): 736-42. 34. Giannoukas AD, Tsetis D, Ioannou C, Kostas T, Kafetzakis A, Petinarakis I, et al. Clinical presentation and anatomic distribution of chronic venous insufficiency of the lower limb in a typical Mediterranean population. Int Angiol. 2002; 21(2): 187-92.

352

35. Danielsson G, Arfvidsson B, Eklof B, Kistner RL, Masuda EM, Satoc DT. Reflux from thigh to calf, the major pathology in chronic venous ulcer disease: surgery indicated in the majority of patients. Vasc Endovascular Surg. 2004; 38(3): 209-19. 36. Ibegbuna V, Delis KT, Nicolaides AN. Haemodynamic and clinical impact of superficial, deep and perforator vein incompetence. Eur J Vasc Endovasc Surg. 2006; 31(5): 535-41. 37. Maurins U, Hoffmann BH, Losch C, Jockel KH, Rabe E, Pannier F. Distribution and prevalence of reflux in the superficial and deep venous system in the general population--results from the Bonn Vein Study, Germany. J Vasc Surg. 2008; 48(3): 680-7. 38. Kanchanabat B, Wongmahisorn Y, Stapanavatr W, Kanchanasuttirak P, Manomaiphiboon A. Clinical presentation and patterns of venous reflux in Thai patients with chronic venous insufficiency (CVI). Eur J Vasc Endovasc Surg. 2010; 40(3): 399-402. 39. Clagett GP, Reisch JS. Prevention of venous thromboembolism in general surgical patients. Results of meta-analysis. Ann Surg. 1988; 208(2): 227-40. 40. Brandjes DP, Buller HR, Heijboer H, Huisman MV, de Rijk M, Jagt H, et al. Randomised trial of effect of compression stockings in patients with symptomatic proximal-vein thrombosis. Lancet. 1997; 349(9054): 759-62. 41. Kahn SR, Ducruet T, Lamping DL, Arsenault L, Miron MJ, Roussin A, et al. Prospective evaluation of health-related quality of life in patients with deep venous thrombosis. Arch Intern Med. 2005; 165(10): 1173-8. 42. MacDougall DA, Feliu AL, Boccuzzi SJ, Lin J. Economic burden of deep-vein thrombosis, pulmonary embolism, and post-thrombotic syndrome. Am J Health Syst Pharm. 2006; 63(20 Suppl 6): S5-15. 43. Kahn SR, Shbaklo H, Lamping DL, Holcroft CA, Shrier I, Miron MJ, et al. Determinants of health-related quality of life during the 2 years following deep vein thrombosis. J Thromb Haemost. 2008; 6(7): 1105-12. 44. Kakkar VV, Lawrence D. Hemodynamic and clinical assessment after therapy for acute deep vein thrombosis. A prospective study. Am J Surg. 1985; 150(4A): 54-63. 45. Monreal M, Martorell A, Callejas JM, Valls R, Llamazares JF, Lafoz E, et al. Venographic assessment of deep vein thrombosis and risk of developing post-thrombotic syndrome: a prospective study. J Intern Med. 1993; 233(3): 233-8. 46. Prandoni P, Lensing AW, Cogo A, Cuppini S, Villalta S, Carta M, et al. The long-term clinical course of acute deep venous thrombosis. Ann Intern Med. 1996; 125(1): 1-7. 47. Saarinen J, Kallio T, Lehto M, Hiltunen S, Sisto T. The occurrence of the post-thrombotic changes after an acute deep venous thrombosis. A prospective two-year follow-up study. J Cardiovasc Surg (Torino). 2000; 41(3): 441-6. 48. Gabriel F, Labios M, Portoles O, Guillen M, Corella D, Frances F, et al. Incidence of post-thrombotic syndrome and its association with various risk factors in a cohort of Spanish patients after one year of follow-up following acute deep venous thrombosis. Thromb Haemost. 2004; 92(2): 328-36. 49. Roumen-Klappe EM, den Heijer M, Janssen MC, van der Vleuten C, Thien T, Wollersheim H. The post- thrombotic syndrome: incidence and prognostic value of non-invasive venous examinations in a six-year follow-up study. Thromb Haemost. 2005; 94(4): 825-30. 50. Kahn SR, Shrier I, Julian JA, Ducruet T, Arsenault L, Miron MJ, et al. Determinants and time course of the postthrombotic syndrome after acute deep venous thrombosis. Ann Intern Med. 2008; 149(10): 698-707. 51. van Ramshorst B, van Bemmelen PS, Hoeneveld H, Faber JA, Eikelboom BC. Thrombus regression in deep venous thrombosis. Quantification of spontaneous thrombolysis with duplex scanning. Circulation. 1992; 86(2): 414-9. 52. Markel A, Meissner M, Manzo RA, Bergelin RO, Strandness DE, Jr. Deep venous thrombosis: rate of spontaneous lysis and thrombus extension. Int Angiol. 2003; 22(4): 376-82. 53. Meissner MH, Caps MT, Zierler BK, Polissar N, Bergelin RO, Manzo RA, et al. Determinants of chronic venous disease after acute deep venous thrombosis. J Vasc Surg. 1998; 28(5): 826-33. 54. Tick LW, Doggen CJ, Rosendaal FR, Faber WR, Bousema MT, Mackaay AJ, et al. Predictors of the post- thrombotic syndrome with non-invasive venous examinations in patients 6 weeks after a first episode of deep vein thrombosis. J Thromb Haemost. 2010; 8(12): 2685-92. 55. Meissner MH, Zierler BK, Bergelin RO, Chandler WL, Strandness DE, Jr. Coagulation, fibrinolysis, and recanalization after acute deep venous thrombosis. J Vasc Surg. 2002; 35(2): 278-85. 56. Killewich LA, Macko RF, Cox K, Franklin DR, Benjamin ME, Lilly MP, et al. Regression of proximal deep venous thrombosis is associated with fibrinolytic enhancement. J Vasc Surg. 1997; 26(5): 861-8. 57. Markel A, Manzo RA, Bergelin RO, Strandness DE, Jr. Valvular reflux after deep vein thrombosis: incidence and time of occurrence. J Vasc Surg. 1992; 15(2): 377-82; discussion 83-4. 58. O'Shaughnessy A M, Fitzgerald DE. The patterns and distribution of residual abnormalities between the individual proximal venous segments after an acute deep vein thrombosis. J Vasc Surg. 2001; 33(2): 379-84. 59. Meissner MH, Manzo RA, Bergelin RO, Markel A, Strandness DE, Jr. Deep venous insufficiency: the relationship between lysis and subsequent reflux. J Vasc Surg. 1993; 18(4): 596-605; discussion 6-8. 60. Nicolaides AN, Hussein MK, Szendro G, Christopoulos D, Vasdekis S, Clarke H. The relation of venous ulceration with ambulatory venous pressure measurements. J Vasc Surg. 1993; 17(2): 414-9.

353

61. Araki CT, Back TL, Padberg FT, Thompson PN, Jamil Z, Lee BC, et al. The significance of calf muscle pump function in venous ulceration. J Vasc Surg. 1994; 20(6): 872-7; discussion 8-9. 62. Welkie JF, Comerota AJ, Katz ML, Aldridge SC, Kerr RP, White JV. Hemodynamic deterioration in chronic venous disease. J Vasc Surg. 1992; 16(5): 733-40. 63. Prandoni P, Lensing AW, Prins MH, Frulla M, Marchiori A, Bernardi E, et al. Below-knee elastic compression stockings to prevent the post-thrombotic syndrome: a randomized, controlled trial. Ann Intern Med. 2004; 141(4): 249-56. 64. Philbrick JT, Becker DM. Calf deep venous thrombosis. A wolf in sheep's clothing? Arch Intern Med. 1988; 148(10): 2131-8. 65. McLafferty RB, Moneta GL, Passman MA, Brant BM, Taylor LM, Jr., Porter JM. Late clinical and hemodynamic sequelae of isolated calf vein thrombosis. J Vasc Surg. 1998; 27(1): 50-6; discussion 6-7. 66. Meissner MH, Caps MT, Bergelin RO, Manzo RA, Strandness DE, Jr. Early outcome after isolated calf vein thrombosis. J Vasc Surg. 1997; 26(5): 749-56. 67. Kahn SR, Kearon C, Julian JA, Mackinnon B, Kovacs MJ, Wells P, et al. Predictors of the post-thrombotic syndrome during long-term treatment of proximal deep vein thrombosis. J Thromb Haemost. 2005; 3(4): 718-23. 68. Ageno W, Piantanida E, Dentali F, Steidl L, Mera V, Squizzato A, et al. Body mass index is associated with the development of the post-thrombotic syndrome. Thromb Haemost. 2003; 89(2): 305-9. 69. Kahn SR, Azoulay L, Hirsch A, Haber M, Strulovitch C, Shrier I. Effect of graduated elastic compression stockings on leg symptoms and signs during exercise in patients with deep venous thrombosis: a randomized crossover trial. J Thromb Haemost. 2003; 1(3): 494-9. 70. Stain M, Schonauer V, Minar E, Bialonczyk C, Hirschl M, Weltermann A, et al. The post-thrombotic syndrome: risk factors and impact on the course of thrombotic disease. J Thromb Haemost. 2005; 3(12): 2671-6. 71. Prandoni P, Frulla M, Sartor D, Concolato A, Girolami A. Vein abnormalities and the post-thrombotic syndrome. J Thromb Haemost. 2005; 3(2): 401-2. 72. Latella J, Desmarais S, Miron MJ, Roussin A, Joyal F, Kassis J, et al. Relation between D-dimer level, venous valvular reflux and the development of post-thrombotic syndrome after deep vein thrombosis. J Thromb Haemost. 2010; 8(10): 2169-75. 73. Partsch H, Kaulich M, Mayer W. Immediate mobilisation in acute vein thrombosis reduces post- thrombotic syndrome. Int Angiol. 2004; 23(3): 206-12. 74. Hafner J, Kuhne A, Schar B, Bombeli T, Hauser M, Luthi R, et al. Factor V Leiden mutation in postthrombotic and non-postthrombotic venous ulcers. Arch Dermatol. 2001; 137(5): 599-603. 75. Gaber Y, Siemens HJ, Schmeller W. Resistance to activated protein C due to factor V Leiden mutation: high prevalence in patients with post-thrombotic leg ulcers. Br J Dermatol. 2001; 144(3): 546-8. 76. Biguzzi E, Mozzi E, Alatri A, Taioli E, Moia M, Mannucci PM. The post-thrombotic syndrome in young women: retrospective evaluation of prognostic factors. Thromb Haemost. 1998; 80(4): 575-7. 77. Bjorgell O, Nilsson PE, Nilsson JA, Svensson PJ. Location and extent of deep vein thrombosis in patients with and without FV:R 506Q mutation. Thromb Haemost. 2000; 83(5): 648-51. 78. Bounameaux H. Factor V Leiden paradox: risk of deep-vein thrombosis but not of pulmonary embolism. Lancet. 2000; 356(9225): 182-3. 79. Christiansen SC, Cannegieter SC, Koster T, Vandenbroucke JP, Rosendaal FR. Thrombophilia, clinical factors, and recurrent venous thrombotic events. JAMA. 2005; 293(19): 2352-61. 80. Baldwin MJ, Moore HM, Rudarakanchana N, Gohel M, Davies AH. Post-thrombotic syndrome: a clinical review. J Thromb Haemost. 2013; 11(5): 795-805. 81. Spiezia L, Campello E, Giolo E, Villalta S, Prandoni P. Thrombophilia and the risk of post-thrombotic syndrome: retrospective cohort observation. J Thromb Haemost. 2010; 8(1): 211-3. 82. Hull RD, Pineo GF, Stein PD, Mah AF, MacIsaac SM, Dahl OE, et al. Extended out-of-hospital low- molecular-weight heparin prophylaxis against deep venous thrombosis in patients after elective hip arthroplasty: a systematic review. Ann Intern Med. 2001; 135(10): 858-69. 83. Wille-Jorgensen P, Jorgensen LN, Crawford M. Asymptomatic postoperative deep vein thrombosis and the development of postthrombotic syndrome. A systematic review and meta-analysis. Thromb Haemost. 2005; 93(2): 236-41. 84. Ginsberg JS, Turkstra F, Buller HR, MacKinnon B, Magier D, Hirsh J. Postthrombotic syndrome after hip or knee arthroplasty: a cross-sectional study. Arch Intern Med. 2000; 160(5): 669-72. 85. Persson LM, Lapidus LJ, Larfars G, Rosfors S. Asymptomatic deep venous thrombosis is associated with a low risk of post-thrombotic syndrome. Eur J Vasc Endovasc Surg. 2009; 38(2): 229-33. 86. Andersen M, Wille-Jorgensen P. Late complications of asymptomatic deep venous thrombosis. Eur J Surg. 1991; 157(9): 527-30. 87. Siragusa S, Beltrametti C, Barone M, Piovella F. [Clinical course and incidence of post-thrombophlebitic syndrome after profound asymptomatic deep vein thrombosis. Results of a transverse epidemiologic study]. Minerva Cardioangiol. 1997; 45(3): 57-66. 88. Schindler OS, Dalziel R. Post-thrombotic syndrome after total hip or knee arthroplasty: incidence in patients with asymptomatic deep venous thrombosis. J Orthop Surg (Hong Kong). 2005; 13(2): 113-9.

354

89. Mudge M, Leinster SJ, Hughes LE. A prospective 10-year study of the post-thrombotic syndrome in a surgical population. Ann R Coll Surg Engl. 1988; 70(4): 249-52. 90. Lindhagen A, Bergqvist D, Hallbook T. Deep venous insufficiency after postoperative thrombosis diagnosed with 125I-labelled fibrinogen uptake test. Br J Surg. 1984; 71(7): 511-5. 91. Pesavento R, Villalta S, Prandoni P. The postthrombotic syndrome. Intern Emerg Med. 2010; 5(3): 185-92. 92. Prandoni P, Kahn SR. Post-thrombotic syndrome: prevalence, prognostication and need for progress. Br J Haematol. 2009; 145(3): 286-95. 93. Kahn SR, Partsch H, Vedantham S, Prandoni P, Kearon C. Definition of post-thrombotic syndrome of the leg for use in clinical investigations: a recommendation for standardization. J Thromb Haemost. 2009; 7(5): 879-83. 94. Roumen-Klappe EM, Janssen MC, Van Rossum J, Holewijn S, Van Bokhoven MM, Kaasjager K, et al. Inflammation in deep vein thrombosis and the development of post-thrombotic syndrome: a prospective study. J Thromb Haemost. 2009; 7(4): 582-7. 95. Shbaklo H, Holcroft CA, Kahn SR. Levels of inflammatory markers and the development of the post- thrombotic syndrome. Thromb Haemost. 2009; 101(3): 505-12. 96. Kahn SR, Shbaklo H, Shapiro S, Wells PS, Kovacs MJ, Rodger MA, et al. Effectiveness of compression stockings to prevent the post-thrombotic syndrome (the SOX Trial and Bio-SOX biomarker substudy): a randomized controlled trial. BMC Cardiovasc Disord. 2007; 7: 21. 97. Kahn SR, Hirsch A, Shrier I. Effect of postthrombotic syndrome on health-related quality of life after deep venous thrombosis. Arch Intern Med. 2002; 162(10): 1144-8. 98. Soosainathan A, Moore HM, Gohel MS, Davies AH. Scoring systems for the post-thrombotic syndrome. J Vasc Surg. 2013; 57(1): 254-61. 99. Rodger MA, Kahn SR, Le Gal G, Solymoss S, Chagnon I, Anderson DR, et al. Inter-observer reliability of measures to assess the post-thrombotic syndrome. Thromb Haemost. 2008; 100(1): 164-6. 100. Kolbach DN, Neumann HA, Prins MH. Definition of the post-thrombotic syndrome, differences between existing classifications. Eur J Vasc Endovasc Surg. 2005; 30(4): 404-14. 101. Kahn SR, Desmarais S, Ducruet T, Arsenault L, Ginsberg JS. Comparison of the Villalta and Ginsberg clinical scales to diagnose the post-thrombotic syndrome: correlation with patient-reported disease burden and venous valvular reflux. J Thromb Haemost. 2006; 4(4): 907-8. 102. van Dongen CJ, Prandoni P, Frulla M, Marchiori A, Prins MH, Hutten BA. Relation between quality of anticoagulant treatment and the development of the postthrombotic syndrome. J Thromb Haemost. 2005; 3(5): 939- 42. 103. Widmer L, Stathelin H, Nissen C. Sd. Venen-, Arterien-, Krankheiten, koronaire Herzkrankhei bei berufstatigen. Bern: Verlag Hans Huber; 1981. 104. Kolbach DN, Sandbrink MW, Neumann HA, Prins MH. Compression therapy for treating stage I and II (Widmer) post-thrombotic syndrome. Cochrane Database Syst Rev. 2003; (4): CD004177. 105. Kahn SR, Shrier I, Shapiro S, Houweling AH, Hirsch AM, Reid RD, et al. Six-month exercise training program to treat post-thrombotic syndrome: a randomized controlled two-centre trial. CMAJ. 2011; 183(1): 37-44. 106. O'Donnell MJ, McRae S, Kahn SR, Julian JA, Kearon C, Mackinnon B, et al. Evaluation of a venous-return assist device to treat severe post-thrombotic syndrome (VENOPTS). A randomized controlled trial. Thromb Haemost. 2008; 99(3): 623-9. 107. Ginsberg JS, Hirsh J, Julian J, Vander LaandeVries M, Magier D, MacKinnon B, et al. Prevention and treatment of postphlebitic syndrome: results of a 3-part study. Arch Intern Med. 2001; 161(17): 2105-9. 108. Kearon C, Kahn SR, Agnelli G, Goldhaber S, Raskob GE, Comerota AJ. Antithrombotic therapy for venous thromboembolic disease: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines (8th Edition). Chest. 2008; 133(6 Suppl): 454S-545S. 109. Kearon C, Gent M, Hirsh J, Weitz J, Kovacs MJ, Anderson DR, et al. A comparison of three months of anticoagulation with extended anticoagulation for a first episode of idiopathic venous thromboembolism. N Engl J Med. 1999; 340(12): 901-7. 110. Agnelli G, Prandoni P, Santamaria MG, Bagatella P, Iorio A, Bazzan M, et al. Three months versus one year of oral anticoagulant therapy for idiopathic deep venous thrombosis. Warfarin Optimal Duration Italian Trial Investigators. N Engl J Med. 2001; 345(3): 165-9. 111. Schulman S, Rhedin AS, Lindmarker P, Carlsson A, Larfars G, Nicol P, et al. A comparison of six weeks with six months of oral anticoagulant therapy after a first episode of venous thromboembolism. Duration of Anticoagulation Trial Study Group. N Engl J Med. 1995; 332(25): 1661-5. 112. Schulman S, Lindmarker P, Holmstrom M, Larfars G, Carlsson A, Nicol P, et al. Post-thrombotic syndrome, recurrence, and death 10 years after the first episode of venous thromboembolism treated with warfarin for 6 weeks or 6 months. J Thromb Haemost. 2006; 4(4): 734-42. 113. Hull RD, Pineo GF, Brant R, Liang J, Cook R, Solymoss S, et al. Home therapy of venous thrombosis with long-term LMWH versus usual care: patient satisfaction and post-thrombotic syndrome. Am J Med. 2009; 122(8): 762-9 e3.

355

114. Gonzalez-Fajardo JA, Martin-Pedrosa M, Castrodeza J, Tamames S, Vaquero-Puerta C. Effect of the anticoagulant therapy in the incidence of post-thrombotic syndrome and recurrent thromboembolism: Comparative study of enoxaparin versus coumarin. J Vasc Surg. 2008; 48(4): 953-9. 115. Hull RD, Pineo GF, Brant RF, Mah AF, Burke N, Dear R, et al. Self-managed long-term low-molecular- weight heparin therapy: the balance of benefits and harms. Am J Med. 2007; 120(1): 72-82. 116. Mousa SA. The low molecular weight heparin, tinzaparin, in thrombosis and beyond. Cardiovasc Drug Rev. 2002; 20(3): 199-216. 117. Kearon C, Hirsh J. Management of anticoagulation before and after elective surgery. N Engl J Med. 1997; 336(21): 1506-11. 118. Kearon C. Management of anticoagulation in patients who require invasive procedures. Semin Vasc Med. 2003; 3(3): 285-94. 119. Roche-Nagle G, Ward F, Barry M. Current prescribing patterns of elastic compression stockings post-deep venous thrombosis. Phlebology. 2010; 25(2): 72-8. 120. Brakkee A, Kuiper J. THe Influence of Compressive Stockings on the Haemodynamics in the Lower Extremities. Phlebology. 1988; 3: 147-53. 121. Kahn SR, Elman E, Rodger MA, Wells PS. Use of elastic compression stockings after deep venous thrombosis: a comparison of practices and perceptions of thrombosis physicians and patients. J Thromb Haemost. 2003; 1(3): 500-6. 122. Kahn SR, Shapiro S, Wells PS, Rodger MA, Kovacs MJ, Anderson DR, et al. Compression stockings to prevent post-thrombotic syndrome: a randomised placebo-controlled trial. Lancet. 2014; 383(9920): 880-8. 123. Musani MH, Matta F, Yaekoub AY, Liang J, Hull RD, Stein PD. Venous compression for prevention of postthrombotic syndrome: a meta-analysis. Am J Med. 2010; 123(8): 735-40. 124. Kolbach DN, Sandbrink MW, Hamulyak K, Neumann HA, Prins MH. Non-pharmaceutical measures for prevention of post-thrombotic syndrome. Cochrane Database Syst Rev. 2004; (1): CD004174. 125. Aschwanden M, Jeanneret C, Koller MT, Thalhammer C, Bucher HC, Jaeger KA. Effect of prolonged treatment with compression stockings to prevent post-thrombotic sequelae: a randomized controlled trial. J Vasc Surg. 2008; 47(5): 1015-21. 126. Sajid MS, Tai NR, Goli G, Morris RW, Baker DM, Hamilton G. Knee versus thigh length graduated compression stockings for prevention of deep venous thrombosis: a systematic review. Eur J Vasc Endovasc Surg. 2006; 32(6): 730-6. 127. Nelson EA, Harper DR, Prescott RJ, Gibson B, Brown D, Ruckley CV. Prevention of recurrence of venous ulceration: randomized controlled trial of class 2 and class 3 elastic compression. J Vasc Surg. 2006; 44(4): 803-8. 128. Roumen-Klappe EM, den Heijer M, van Rossum J, Wollersheim H, van der Vleuten C, Thien T, et al. Multilayer compression bandaging in the acute phase of deep-vein thrombosis has no effect on the development of the post-thrombotic syndrome. J Thromb Thrombolysis. 2009; 27(4): 400-5. 129. Arpaia G, Cimminiello C, Mastrogiacomo O, de Gaudenzi E. Efficacy of elastic compression stockings used early or after resolution of the edema on recanalization after deep venous thrombosis: the COM.PRE Trial. Blood Coagul Fibrinolysis. 2007; 18(2): 131-7. 130. Shrier I, Kahn SR. Effect of physical activity after recent deep venous thrombosis: a cohort study. Med Sci Sports Exerc. 2005; 37(4): 630-4. 131. Shrier I, Kahn SR, Steele RJ. Effect of early physical activity on long-term outcome after venous thrombosis. Clin J Sport Med. 2009; 19(6): 487-93. 132. Stick C, Grau H, Witzleb E. On the edema-preventing effect of the calf muscle pump. Eur J Appl Physiol Occup Physiol. 1989; 59(1-2): 39-47. 133. Isma N, Johanssson E, Bjork A, Bjorgell O, Robertson F, Mattiasson I, et al. Does supervised exercise after deep venous thrombosis improve recanalization of occluded vein segments? A randomized study. J Thromb Thrombolysis. 2007; 23(1): 25-30. 134. Plate G, Eklof B, Norgren L, Ohlin P, Dahlstrom JA. Venous thrombectomy for iliofemoral vein thrombosis--10-year results of a prospective randomised study. Eur J Vasc Endovasc Surg. 1997; 14(5): 367-74. 135. Hull RD, Marder VJ, Mah AF, Biel RK, Brant RF. Quantitative assessment of thrombus burden predicts the outcome of treatment for venous thrombosis: a systematic review. Am J Med. 2005; 118(5): 456-64. 136. Browse NL, Thomas ML, Pim HP. Streptokinase and deep vein thrombosis. Br Med J. 1968; 3(5620): 717- 20. 137. Kakkar VV, Flanc C, O'Shea MJ, Flute PT, Howe CT, Clarke MB. Treatment of deep-vein thrombosis with streptokinase. Br J Surg. 1969; 56(3): 178-83. 138. Porter JM, Seaman AJ, Common HH, Rosch J, Eidemiller LR, Calhoun AD. Comparison of heparin and streptokinase in the treatment of venous thrombosis. Am Surg. 1975; 41(9): 511-19. 139. Seaman AJ, Common HH, Rosch J, Dotter CT, Porter JM, Lindell TD, et al. Deep vein thrombosis treated with streptokinase or heparin. A randomized study. Angiology. 1976; 27(10): 549-56. 140. Watson LI, Armon MP. Thrombolysis for acute deep vein thrombosis. Cochrane Database Syst Rev. 2004; (4): CD002783.

356

141. Elliot MS, Immelman EJ, Jeffery P, Benatar SR, Funston MR, Smith JA, et al. A comparative randomized trial of heparin versus streptokinase in the treatment of acute proximal venous thrombosis: an interim report of a prospective trial. Br J Surg. 1979; 66(12): 838-43. 142. Arnesen H, Heilo A, Jakobsen E, Ly B, Skaga E. A prospective study of streptokinase and heparin in the treatment of deep vein thrombosis. Acta Med Scand. 1978; 203(6): 457-63. 143. Elsharawy M, Elzayat E. Early results of thrombolysis vs anticoagulation in iliofemoral venous thrombosis. A randomised clinical trial. Eur J Vasc Endovasc Surg. 2002; 24(3): 209-14. 144. Sharifi M, Mehdipour M, Bay C, Smith G, Sharifi J. Endovenous therapy for deep venous thrombosis: the TORPEDO trial. Catheter Cardiovasc Interv. 2010; 76(3): 316-25. 145. Enden T, Klow NE, Sandvik L, Slagsvold CE, Ghanima W, Hafsahl G, et al. Catheter-directed thrombolysis vs. anticoagulant therapy alone in deep vein thrombosis: results of an open randomized, controlled trial reporting on short-term patency. J Thromb Haemost. 2009; 7(8): 1268-75. 146. Karthikesalingam A, Young EL, Hinchliffe RJ, Loftus IM, Thompson MM, Holt PJ. A systematic review of percutaneous mechanical thrombectomy in the treatment of deep venous thrombosis. Eur J Vasc Endovasc Surg. 2011; 41(4): 554-65. 147. O'Meara S, Cullum NA, Nelson EA. Compression for venous leg ulcers. Cochrane Database Syst Rev. 2009; (1): CD000265. 148. Ginsberg JS, Magier D, Mackinnon B, Gent M, Hirsh J. Intermittent compression units for severe postphlebitic syndrome: a randomized crossover study. CMAJ. 1999; 160(9): 1303-6. 149. Rosenberg IL, Evans M, Pollock AV. Prophylaxis of postoperative leg vine thrombosis by low dose subcutaneous heparin or peroperative calf muscle stimulation: a controlled clinical trial. Br Med J. 1975; 1(5959): 649-51. 150. Bostrom S, Holmgren E, Jonsson O, Lindberg S, Lindstrom B, Winso I, et al. Post-operative thromboembolism in neurosurgery. A study on the prophylactic effect of calf muscle stimulation plus dextran compared to low-dose heparin. Acta Neurochir (Wien). 1986; 80(3-4): 83-9. 151. Doran FS, Drury M, Sivyer A. A Simple Way to Combat the Venous Stasis Which Occurs in the Lower Limbs during Surgical Operations. Br J Surg. 1964; 51: 486-92. 152. Gohel MS, Davies AH. Pharmacological treatment in patients with C4, C5 and C6 venous disease. Phlebology. 2010; 25 Suppl 1: 35-41. 153. Layton AM, Ibbotson SH, Davies JA, Goodfield MJ. Randomised trial of oral aspirin for chronic venous leg ulcers. Lancet. 1994; 344(8916): 164-5. 154. Shoab SS, Porter JB, Scurr JH, Coleridge-Smith PD. Effect of oral micronized purified flavonoid fraction treatment on leukocyte adhesion molecule expression in patients with chronic venous disease: a pilot study. J Vasc Surg. 2000; 31(3): 456-61. 155. Smith PC. Daflon 500 mg and venous leg ulcer: new results from a meta-analysis. Angiology. 2005; 56 Suppl 1: S33-9. 156. Jantet G. Chronic venous insufficiency: worldwide results of the RELIEF study. Reflux assEssment and quaLity of lIfe improvEment with micronized Flavonoids. Angiology. 2002; 53(3): 245-56. 157. Prandoni P. Elastic stockings, hydroxyethylrutosides or both for the treatment of post-thrombotic syndrome. Thromb Haemost. 2005; 93(1): 183-5. 158. Pittler MH, Ernst E. Horse chestnut seed extract for chronic venous insufficiency. Cochrane Database Syst Rev. 2006; (1): CD003230. 159. Ciapponi A, Laffaire E, Roque M. Calcium dobesilate for chronic venous insufficiency: a systematic review. Angiology. 2004; 55(2): 147-54. 160. Martinez-Zapata MJ, Moreno RM, Gich I, Urrutia G, Bonfill X. A randomized, double-blind multicentre clinical trial comparing the efficacy of calcium dobesilate with placebo in the treatment of chronic venous disease. Eur J Vasc Endovasc Surg. 2008; 35(3): 358-65. 161. Jull A, Arroll B, Parag V, Waters J. Pentoxifylline for treating venous leg ulcers. Cochrane Database Syst Rev. 2007; (3): CD001733. 162. Rhodes JM, Gloviczki P, Canton L, Heaser TV, Rooke TW. Endoscopic perforator vein division with ablation of superficial reflux improves venous hemodynamics. J Vasc Surg. 1998; 28(5): 839-47. 163. Proebstle TM, Weisel G, Paepcke U, Gass S, Weber L. Light reflection rheography and clinical course of patients with advanced venous disease before and after endoscopic subfascial division of perforating veins. Dermatol Surg. 1998; 24(7): 771-6. 164. Gloviczki P, Bergan JJ, Rhodes JM, Canton LG, Harmsen S, Ilstrup DM. Mid-term results of endoscopic perforator vein interruption for chronic venous insufficiency: lessons learned from the North American subfascial endoscopic perforator surgery registry. The North American Study Group. J Vasc Surg. 1999; 29(3): 489-502. 165. Tenbrook JA, Jr., Iafrati MD, O'Donnell T F, Jr., Wolf MP, Hoffman SN, Pauker SG, et al. Systematic review of outcomes after surgical management of venous disease incorporating subfascial endoscopic perforator surgery. J Vasc Surg. 2004; 39(3): 583-9. 166. Masuda EM, Kistner RL. Long-term results of venous valve reconstruction: a four- to twenty-one-year follow-up. J Vasc Surg. 1994; 19(3): 391-403.

357

167. Comerota AJ, Grewal NK, Thakur S, Assi Z. Endovenectomy of the common femoral vein and intraoperative iliac vein recanalization for chronic iliofemoral venous occlusion. J Vasc Surg. 2010; 52(1): 243-7. 168. Puggioni A, Kistner RL, Eklof B, Lurie F. Surgical disobliteration of postthrombotic deep veins-- endophlebectomy--is feasible. J Vasc Surg. 2004; 39(5): 1048-52; discussion 52. 169. Kolbel T, Lindh M, Akesson M, Wasselius J, Gottsater A, Ivancev K. Chronic iliac vein occlusion: midterm results of endovascular recanalization. J Endovasc Ther. 2009; 16(4): 483-91. 170. Neglen P, Berry MA, Raju S. Endovascular surgery in the treatment of chronic primary and post- thrombotic iliac vein obstruction. Eur J Vasc Endovasc Surg. 2000; 20(6): 560-71. 171. Harrison MB, Vandenkerkhof EG, Hopman WM, Graham ID, Carley ME, Nelson EA, et al. The Canadian Bandaging Trial: Evidence-informed Leg Ulcer Care and the Effectiveness of Two Compression Technologies. BMC Nurs. 2011; 10(1): 20. 172. Nelson EA, Mani R, Thomas K, Vowden K. Intermittent pneumatic compression for treating venous leg ulcers. Cochrane Database Syst Rev. 2011; (2): CD001899. 173. Corley GJ, Breen PP, Birlea SI, Serrador JM, Grace PA, Olaighin G. Hemodynamic effects of habituation to a week-long program of neuromuscular electrical stimulation. Med Eng Phys. 2011. 174. Izumi M, Ikeuchi M, Mitani T, Taniguchi S, Tani T. Prevention of venous stasis in the lower limb by transcutaneous electrical nerve stimulation. Eur J Vasc Endovasc Surg. 2010; 39(5): 642-5. 175. Lyseng-Williamson KA, Perry CM. Micronised purified flavonoid fraction: a review of its use in chronic venous insufficiency, venous ulcers and haemorrhoids. Drugs. 2003; 63(1): 71-100. 176. Perrin M. [Surgery for deep venous reflux in the lower limb]. J Mal Vasc. 2004; 29(2): 73-87. 177. Hardy SC, Riding G, Abidia A. Surgery for deep venous incompetence. Cochrane Database Syst Rev. 2004; (3): CD001097. 178. Maleti O, Perrin M. Reconstructive surgery for deep vein reflux in the lower limbs: techniques, results and indications. Eur J Vasc Endovasc Surg. 2011; 41(6): 837-48. 179. Kistner RL. Valve reconstruction for primary valve insufficiency. In: Bergan J, Kistner RL, editors. Atlas of Venous Surgery. Philadelphia: Saunders; 1992. p. 125-30. 180. Sottiurai V. Supravalvular incision for valve repair in primary valvular insufficiency: method of Sottiurai. In: Bergan J, Kistner RL, editors. Atlas of Venous Surgery. Philadelphia: Saunders; 1992. p. 137-44. 181. Hoshino S, Satokawa H, Takase S, Midorikawa H, Igar T, Iwaya F. External valvuloplasty for primary valvular incompetence of the lower limbs using angioscopy. International Journal of Angiology. 1997; 6(2): 137-41. 182. Jessup G, Lane RJ. Repair of incompetent venous valves: a new technique. J Vasc Surg. 1988; 8(5): 569-75. 183. Taheri SA, Lazar L, Elias S, Marchand P, Heffner R. Surgical treatment of postphlebitic syndrome with vein valve transplant. Am J Surg. 1982; 144(2): 221-4. 184. Iafrati M, O'Donnell TF. Surgical reconstruction for deep venous insufficiency. J Mal Vasc. 1997; 22(3): 193-7. 185. Puggioni A, Lurie F. Advances in the surgical treatment of postthrombotic syndrome. Phlebolymphology. 2008; 15(1): 99. 186. Neglen P, Raju S. Venous reflux repair with cryopreserved vein valves. J Vasc Surg. 2003; 37(3): 552-7. 187. Kistner RL, Sparkuhl MD. Surgery in acute and chronic venous disease. Surgery. 1979; 85(1): 31-43. 188. Johnson ND, Queral LA, Flinn WR, Yao JS, Bergan JJ. Late objective assessment of venous value surgery. Arch Surg. 1981; 116(11): 1461-6. 189. Queral LA, Whitehouse WM, Jr., Flinn WR, Neiman HL, Yao JS, Bergan JJ. Surgical correction of chronic deep venous insufficiency by valvular transposition. Surgery. 1980; 87(6): 688-95. 190. Perrin M. Reconstructive surgery for deep venous reflux: a report on 144 cases. Cardiovasc Surg. 2000; 8(4): 246-55. 191. Neglen P, Hollis KC, Olivier J, Raju S. Stenting of the venous outflow in chronic venous disease: long- term stent-related outcome, clinical, and hemodynamic result. J Vasc Surg. 2007; 46(5): 979-90. 192. Raju GS, Ahmed I. Team approach to the management of complex esophageal obstruction. Gastrointest Endosc. 2002; 55(2): 304. 193. Hartung O, Otero A, Boufi M, Decaridi G, Barthelemy P, Juhan C, et al. Mid-term results of endovascular treatment for symptomatic chronic nonmalignant iliocaval venous occlusive disease. J Vasc Surg. 2005; 42(6): 1138- 44; discussion 44. 194. Smith DE, Trimble CE. Surgical management of obstructive venous disease of the lower extremity. Vascular Surgery. Philadelphia: Saunders; 1977. p. 1247-68. 195. Eklof B, Albrechtson U, Einarsson E, Plate G. The temporary arteriovenous fistula in venous reconstructive surgery. Int Angiol. 1985; 4(4): 455-62. 196. AbuRahma AF, Robinson PA, Boland JP. Clinical, hemodynamic, and anatomic predictors of long-term outcome of lower extremity venovenous bypasses. J Vasc Surg. 1991; 14(5): 635-44. 197. Kelleher D, Moore HM, Davies AH. Debate: Deep venous reconstruction is worthwhile. In: Greenhalgh RM, editor. Vascular and Endovascular Controversies Update. London: Biba Publishing; 2012. p. 610-8.

358

198. Pavcnik D, Uchida B, Kaufman J, Hinds M, Keller FS, Rosch J. Percutaneous management of chronic deep venous reflux: review of experimental work and early clinical experience with bioprosthetic valve. Vasc Med. 2008; 13(1): 75-84. 199. McLachlin AD, Carroll SE, Meads GE, Amacher AL. Valve replacement in the recanalized incompetent superficial femoral vein in dogs. Ann Surg. 1965; 162(3): 446-52. 200. Hill R, Schmidt S, Evancho M, Hunter T, Hillegass D, Sharp W. Development of a prosthetic venous valve. J Biomed Mater Res. 1985; 19(7): 827-32. 201. Gomez-Jorge J, Venbrux AC, Magee C. Percutaneous deployment of a valved bovine jugular vein in the swine venous system: a potential treatment for venous insufficiency. J Vasc Interv Radiol. 2000; 11(7): 931-6. 202. Gale SS, Shuman S, Beebe HG, Pigott JP, Comerota AJ. Percutaneous venous valve bioprosthesis: initial observations. Vasc Endovascular Surg. 2004; 38(3): 221-4. 203. Pavcnik D. Update on venous valve replacment: long-term clinical results. Vascular. 2006; 14: 106. 204. Geselschap JH, van Zuiden JM, Toonder IM, Wittens CH. In vitro evaluation of a new autologous valve- stent for deep venous incompetence. J Endovasc Ther. 2006; 13(6): 762-9. 205. DeLaria GA, Phifer T, Roy J, Tu R, Thyagarajan K, Quijano RC. Hemodynamic evaluation of a bioprosthetic venous prosthesis. J Vasc Surg. 1993; 18(4): 577-84; discussion 84-6. 206. Pavcnik D, Uchida B, Timmermans H, Keller FS, Rosch J. Square stent: a new self-expandable endoluminal device and its applications. Cardiovasc Intervent Radiol. 2001; 24(4): 207-17. 207. Pavcnik D, Uchida BT, Timmermans H, Corless CL, Keller FS, Rosch J. Aortic and venous valve for percutaneous insertion. Minim Invasive Ther Allied Technol. 2000; 9(3-4): 287-92. 208. Oberdier MT, Rittgers SE. The design, development, and evaluation of a prototypic, prosthetic venous valve. Biomed Eng Online. 2008; 7: 25. 209. Rittgers SE, Oberdier MT, Pottala S. Physiologically-based testing system for the mechanical characterization of prosthetic vein valves. Biomed Eng Online. 2007; 6: 29. 210. Sathe RD, Ku DN. Flexible Prosthetic Vein Valve. Journal of Medical Devices. 2007; 1: 105-12. 211. Moriyama M, Kubota S, Tashiro H, Tonami H. Evaluation of prosthetic venous valves, fabricated by electrospinning, for percutaneous treatment of chronic venous insufficiency. J Artif Organs. 2011; 14(4): 294-300. 212. Kuraishi K, Iwata H, Nakano S, Kubota S, Tonami H, Toda M, et al. Development of nanofiber-covered stents using electrospinning: in vitro and acute phase in vivo experiments. J Biomed Mater Res B Appl Biomater. 2009; 88(1): 230-9. 213. de Borst GJ, Teijink JA, Patterson M, Quijano TC, Moll FL. A percutaneous approach to deep venous valve insufficiency with a new self-expanding venous frame valve. J Endovasc Ther. 2003; 10(2): 341-9. 214. Boudjemline Y, Bonnet D, Sidi D, Bonhoeffer P. Is percutaneous implantation of a bovine venous valve in the inferior vena cava a reliable technique to treat chronic venous insufficiency syndrome? Med Sci Monit. 2004; 10(3): BR61-6. 215. Kucher T, Brener B, Marak M, Parsonnet V. Endovascular delivery of vein segments with valves versus direct anastomosis. J Endovasc Ther. 2005; 12(3): 366-70. 216. Ofenloch JC, Chen C, Hughes JD, Lumsden AB. Endoscopic venous valve transplantation with a valve- stent device. Ann Vasc Surg. 1997; 11(1): 62-7. 217. Dalsing MC, Sawchuk AP, Lalka SG, Cikrit DF. An early experience with endovascular venous valve transplantation. J Vasc Surg. 1996; 24(5): 903-5. 218. Pavcnik D, Uchida BT, Timmermans HA, Corless CL, O'Hara A, Toyota N, et al. Percutaneous bioprosthetic venous valve: A long-term study in sheep. Journal of Vascular Surgery. 2002; 35(3): 598-602. 219. Pavcnik D, Yin Q, Uchida B, Park WK, Hoppe H, Kim MD, et al. Percutaneous autologous venous valve transplantation: short-term feasibility study in an ovine model. J Vasc Surg. 2007; 46(2): 338-45. 220. Pavcnik D, Kaufman J, Uchida B, Correa L, Hiraki T, Kyu SC, et al. Second-generation percutaneous bioprosthetic valve: a short-term study in sheep. J Vasc Surg. 2004; 40(6): 1223-7. 221. Serino F. Preliminary clinical experiences with the VenPro PVVB. Updating Course in vascular pathology of surgical interest; 2002; Pisa; 2002. 222. Pavcnik D, Kaufman J, Machan L, Uchida B, Keller FS, Rosch J. Percutaneous therapy for deep vein reflux. Semin Intervent Radiol. 2005; 22(3): 225-32. 223. Pavcnik D, Machan L, Uchida B, Kaufman J, Keller FS, Rosch J. Percutaneous prosthetic venous valves: current state and possible applications. Tech Vasc Interv Radiol. 2003; 6(3): 137-42. 224. Lurie F, Kistner RL, Eklof B. The mechanism of venous valve closure in normal physiologic conditions. J Vasc Surg. 2002; 35(4): 713-7. 225. Powell T, Lynn RB. The valves of the external iliac, femoral, and upper third of the popliteal veins. Surg Gynecol Obstet. 1951; 92(4): 453-5. 226. Basmajian JV. The distribution of valves in the femoral, external iliac, and common iliac veins and their relationship to varicose veins. Surg Gynecol Obstet. 1952; 95(5): 537-42. 227. Lurie F, Kistner RL, Eklof B, Kessler D. Mechanism of venous valve closure and role of the valve in circulation: a new concept. J Vasc Surg. 2003; 38(5): 955-61.

359

228. Scultetus AH, Villavicencio JL, Rich NM. Facts and fiction surrounding the discovery of the venous valves. J Vasc Surg. 2001; 33(2): 435-41. 229. Moore HM, Gohel M, Davies AH. Number and location of venous valves within the popliteal and femoral veins: a review of the literature. J Anat. 2011; 219(4): 439-43. 230. G G. Venous anatomy of the lower limb. SocietàItaliana di Flebologia; c2002; c2002. 231. Muhlberger D, Morandini L, Brenner E. An anatomical study of femoral vein valves near the saphenofemoral junction. J Vasc Surg. 2008; 48(4): 994-9. 232. Banjo AO. Comparative study of the distribution of venous valves in the lower extremities of black Africans and Caucasians: pathogenetic correlates of prevalence of primary varicose veins in the two races. Anat Rec. 1987; 217(4): 407-12. 233. LB K. Frequency and Distribution of the Valves in the large deep and superficial veins of the lower limb in the aged. ActaMorpholNeerl-Scand. 1971; 9: 41-6. 234. Eger SA, Wagner FB, Jr. Etiology of varicose veins. Postgrad Med. 1949; 6(3): 234-8. 235. Santilli SM, Lee ES, Wernsing SE, Diedrich DA, Kuskowski MA, Shew RL. Superficial femoral popliteal vein: An anatomic study. J Vasc Surg. 2000; 31(3): 450-5. 236. Nordon IM, Hinchliffe RJ, Brar R, Moxey P, Black SA, Thompson MM, et al. A prospective double-blind randomized controlled trial of radiofrequency versus laser treatment of the great saphenous vein in patients with varicose veins. Ann Surg. 2011; 254(6): 876-81. 237. Ciostek P, Michalak J, Noszczyk W. Improvement in deep vein haemodynamics following surgery for varicose veins. Eur J Vasc Endovasc Surg. 2004; 28(5): 473-8. 238. Meissner MH, Moneta G, Burnand K, Gloviczki P, Lohr JM, Lurie F, et al. The hemodynamics and diagnosis of venous disease. J Vasc Surg. 2007; 46 Suppl S: 4S-24S. 239. Abu-Yousef MM, Mufid M, Woods KT, Brown BP, Barloon TJ. Normal lower limb venous Doppler flow phasicity: is it cardiac or respiratory? AJR Am J Roentgenol. 1997; 169(6): 1721-5. 240. Wexler L, Bergel DH, Gabe IT, Makin GS, Mills CJ. Velocity of blood flow in normal human venae cavae. Circ Res. 1968; 23(3): 349-59. 241. Willeput R, Rondeux C, Troyer AD. Breathing affects venous return from legs in humans. J Appl Physiol. 1984; 57(4): 971-6. 242. Kwon O-Y, Jung D-Y, Kim Y, Cho S-H, Yi C-H. Effects of ankle exercise combined with deep breathing on blood flow velocity in the femoral vein. Aust J Physiother. 2003; 49(4): 253-8. 243. Miller JD, Pegelow DF, Jacques AJ, Dempsey JA. Skeletal muscle pump versus respiratory muscle pump: modulation of venous return from the locomotor limb in humans. J Physiol. 2005; 563(Pt 3): 925-43. 244. Downie SP, Raynor SM, Firmin DN, Wood NB, Thom SA, Hughes AD, et al. Effects of elastic compression stockings on wall shear stress in deep and superficial veins of the calf. Am J Physiol Heart Circ Physiol. 2008; 294(5): H2112-H20. 245. Padberg FT, Jr., Johnston MV, Sisto SA. Structured exercise improves calf muscle pump function in chronic venous insufficiency: a randomized trial. J Vasc Surg. 2004; 39(1): 79-87. 246. Chauveau M, Gelade P, Cros F. The venous return simulator: comparison of simulated with measured ambulatory venous pressure in normal subjects and in venous valve incompetence. Vasa. 2011; 40(3): 205-17. 247. Uhl JF, Gillot C. Anatomy of the foot venous pump: physiology and influence on chronic venous disease. Phlebology. 2012; 27(5): 219-30. 248. Stewart JM, Medow MS, Montgomery LD, McLeod K. Decreased skeletal muscle pump activity in patients with postural tachycardia syndrome and low peripheral blood flow. Am J Physiol Heart Circ Physiol. 2004; 286(3): H1216-H22. 249. O'Donovan KJ, O'Keeffe DT, Grace PA, Lyons GM. Accelerometer based calf muscle pump activity monitoring. Med Eng Phys. 2005; 27(8): 717-22. 250. Labropoulos N, Tiongson J, Pryor L, Tassiopoulos AK, Kang SS, Ashraf Mansour M, et al. Definition of venous reflux in lower-extremity veins. J Vasc Surg. 2003; 38(4): 793-8. 251. Lurie F, Comerota A, Eklof B, Kistner RL, Labropoulos N, Lohr J, et al. Multicenter assessment of venous reflux by duplex ultrasound. J Vasc Surg. 2012; 55(2): 437-45. 252. Neglen P, Egger JF, 3rd, Olivier J, Raju S. Hemodynamic and clinical impact of ultrasound-derived venous reflux parameters. J Vasc Surg. 2004; 40(2): 303-10. 253. Shepherd JT. Role of the veins in the circulation. Circulation. 1966; 33(3): 484-91. 254. Pollack AA, Wood EH. Venous pressure in the saphenous vein at the ankle in man during exercise and changes in posture. J Appl Physiol. 1949; 1(9): 649-62. 255. Arnoldi CC. Venous pressure in the leg of healthy human subjects at rest and during muscular exercise in the nearly erect position. Acta Chir Scand. 1965; 130(6): 570-83. 256. Katz AI, Chen Y, Moreno AH. Flow through a collapsible tube. Experimental analysis and mathematical model. Biophys J. 1969; 9(10): 1261-79. 257. Padberg F. The Physiology and Haemodynamics of the Normal Venous Circulation. In: Gloviczki P, Yao JS, editors. Handbook of Venous Disorders: Arnold; 2002. p. 25-34.

360

258. Kugler C, Strunk M, Rudofsky G. Venous pressure dynamics of the healthy human leg. Role of muscle activity, joint mobility and anthropometric factors. J Vasc Res. 2001; 38(1): 20-9. 259. Nicolaides AN. Investigation of chronic venous insufficiency: A consensus statement (France, March 5-9, 1997). Circulation. 2000; 102(20): E126-63. 260. Ludbrook J. The musculovenous pumps of the human lower limb. Am Heart J. 1966; 71(5): 635-41. 261. Christopoulos DG, Nicolaides AN, Szendro G, Irvine AT, Bull ML, Eastcott HH. Air-plethysmography and the effect of elastic compression on venous hemodynamics of the leg. J Vasc Surg. 1987; 5(1): 148-59. 262. Maffiuletti NA. Physiological and methodological considerations for the use of neuromuscular electrical stimulation. Eur J Appl Physiol. 2010; 110(2): 223-34. 263. Gardner AM, Fox RH. The venous pump of the human foot--preliminary report. Bristol Med Chir J. 1983; 98(367): 109-12. 264. Franceschi C, Zamboni P, editors. Principles of venous haemodynamics. New York: Novapublishers; 2010. 265. Franceschi C. Dynamic fractionizing of hydrostatic pressure, closed and open shunts, vicarious varicose evolution: how these concepts made the treatment of varices evolve? . Phlebologie. 2003; 56(1): 61-6. 266. Mdez-Herrero A, Gutierrez J, Camblor L, Carreno J, Llaneza J, Rguez-Olay J, et al. The relation among the diameter of the great saphenous vein, clinical state and haemodynamic pattern of the saphenofemoral junction in chronic superficial venous insufficiency. Phlebology. 2007; 22(5): 207-13. 267. Escribano JM, Juan J, Bofill R, Maeso J, Rodriguez-Mori A, Matas M. Durability of reflux-elimination by a minimal invasive CHIVA procedure on patients with varicose veins. A 3-year prospective case study. Eur J Vasc Endovasc Surg. 2003; 25(2): 159-63. 268. Mendoza E, Berger HA. Provokationsmanöver für die duplex-sonographische Diagnostik der Varikosis. Gefässchirurgie. 2001; 6(1): 43-6. 269. van Bemmelen PS, Beach K, Bedford G, Strandness DE, Jr. The mechanism of venous valve closure. Its relationship to the velocity of reverse flow. Arch Surg. 1990; 125(5): 617-9. 270. Franceschi C. [Measures and interpretation of venous flow in stress tests. Manual compression and Parana manoeuver. Dynamic reflux index and Pstakis index]. J Mal Vasc. 1997; 22(2): 91-5. 271. Coleridge-Smith P, Labropoulos N, Partsch H, Myers K, Nicolaides A, Cavezzi A. Duplex ultrasound investigation of the veins in chronic venous disease of the lower limbs--UIP consensus document. Part I. Basic principles. Eur J Vasc Endovasc Surg. 2006; 31(1): 83-92. 272. Mendoza E, Blattler W, Amsler F. Great Saphenous Vein Diameter at the Saphenofemoral Junction and Proximal Thigh as Parameters of Venous Disease Class. Eur J Vasc Endovasc Surg. 2013; 45(1): 76-83. 273. Cappelli M, Molino Lova R, Ermini S, Zamboni P. Hemodynamics of the sapheno-femoral junction. Patterns of reflux and their clinical implications. Int Angiol. 2004; 23(1): 25-8. 274. Cavezzi A, Labropoulos N, Partsch H, Ricci S, Caggiati A, Myers K, et al. Duplex ultrasound investigation of the veins in chronic venous disease of the lower limbs--UIP consensus document. Part II. Anatomy. Eur J Vasc Endovasc Surg. 2006; 31(3): 288-99. 275. Cappelli M, Molino Lova R, Ermini S, Turchi A, Bono G, Franceschi C. [Comparison between the CHIVA cure and stripping in the treatment of varicose veins of the legs: follow-up of 3 years]. J Mal Vasc. 1996; 21(1): 40-6. 276. Delis KT, Nicolaides AN. Effect of intermittent pneumatic compression of foot and calf on walking distance, hemodynamics, and quality of life in patients with arterial claudication: a prospective randomized controlled study with 1-year follow-up. Ann Surg. 2005; 241(3): 431-41. 277. Folse R, Alexander RH. Directional flow detection for localizing venous valvular incompetency. Surgery. 1970; 67(1): 114-21. 278. Arnoldussen CW, Wittens CH. An imaging approach to deep vein thrombosis and the lower extremity thrombosis classification. Phlebology. 2012; 27 Suppl 1: 143-8. 279. McMullin GM, Coleridge Smith PD. An evaluation of Doppler ultrasound and photoplethysmography in the investigation of venous insufficiency. Aust N Z J Surg. 1992; 62(4): 270-5. 280. Nicolini P. Treatment of primary varicose veins by endovenous obliteration with the VNUS closure system: results of a prospective multicentre study. Eur J Vasc Endovasc Surg. 2005; 29(4): 433-9. 281. Weingarten MS, Branas CC, Czeredarczuk M, Schmidt JD, Wolferth CC, Jr. Distribution and quantification of venous reflux in lower extremity chronic venous stasis disease with duplex scanning. J Vasc Surg. 1993; 18(5): 753-9. 282. van Bemmelen PS, Bedford G, Beach K, Strandness DE. Quantitative segmental evaluation of venous valvular reflux with duplex ultrasound scanning. J Vasc Surg. 1989; 10(4): 425-31. 283. Wong JK, Duncan JL, Nichols DM. Whole-leg duplex mapping for varicose veins: observations on patterns of reflux in recurrent and primary legs, with clinical correlation. Eur J Vasc Endovasc Surg. 2003; 25(3): 267-75. 284. Neglen P, Raju S. A comparison between descending phlebography and duplex Doppler investigation in the evaluation of reflux in chronic venous insufficiency: a challenge to phlebography as the "gold standard". J Vasc Surg. 1992; 16(5): 687-93.

361

285. Labropoulos N, Mansour MA, Kang SS, Gloviczki P, Baker WH. New insights into perforator vein incompetence. Eur J Vasc Endovasc Surg. 1999; 18(3): 228-34. 286. Gooley NA, Sumner DS. Relationship of venous reflux to the site of venous valvular incompetence: implications for venous reconstructive surgery. J Vasc Surg. 1988; 7(1): 50-9. 287. Adam DJ, Naik J, Hartshorne T, Bello M, London NJ. The diagnosis and management of 689 chronic leg ulcers in a single-visit assessment clinic. Eur J Vasc Endovasc Surg. 2003; 25(5): 462-8. 288. Labropoulos N, Leon LR, Jr. Duplex evaluation of venous insufficiency. Semin Vasc Surg. 2005; 18(1): 5-9. 289. Zamboni P, Gianesini S, Menegatti E, Tacconi G, Palazzo A, Liboni A. Great saphenous varicose vein surgery without saphenofemoral junction disconnection. Br J Surg. 2010; 97(6): 820-5. 290. Bjordal R. Simultaneous pressure and flow recordings in varicose veins of the lower extremity. A haemodynamic study of venous dysfunction. Acta Chir Scand. 1970; 136(4): 309-17. 291. Recek C, Pojer H. Ambulatory pressure gradient in the veins of the lower extremity. Vasa. 2000; 29(3): 187-90. 292. Franceschi C, editor. Théorie et pratique de la cure Conservatrice et hémodynamique de l’insuffisance veineuse en ambulatoire: Precy-sous-Thil France Editions de l’Armançon; 1988. 293. Masuda EM, Kistner RL, Eklof B. Prospective study of duplex scanning for venous reflux: comparison of Valsalva and pneumatic cuff techniques in the reverse Trendelenburg and standing positions. J Vasc Surg. 1994; 20(5): 711-20. 294. Klonizakis M, Lingam K, Manning G, Donnelly R. Characterising the time-course of microvascular vasodilator responses in humans using Laser Doppler Fluximetry and iontophoresis. J Pharmacol Toxicol Methods. 2011; 63(1): 115-8. 295. Klonizakis M, Manning G, Donnelly R. Assessment of lower limb microcirculation: exploring the reproducibility and clinical application of laser Doppler techniques. Skin Pharmacol Physiol. 2011; 24(3): 136-43. 296. Christopoulos D, Nicolaides AN, Szendro G. Venous reflux: quantification and correlation with the clinical severity of chronic venous disease. Br J Surg. 1988; 75(4): 352-6. 297. Miyazaki K, Nishibe T, Kudo F, Miyazaki YJ, Nishibe M, Ando M, et al. Hemodynamic changes in stripping operation or saphenofemoral ligation of the greater saphenous vein for primary varicose veins. Ann Vasc Surg. 2004; 18(4): 465-9. 298. Barnes RW, Collicott PE, Sumner DS, Strandness DE, Jr. Noninvasive quantitation of venous hemodynamics in the postphlebitic syndrome. Arch Surg. 1973; 107(5): 807-14. 299. Harada RN, Katz ML, Comerota A. A noninvasive screening test to detect "critical" deep venous reflux. J Vasc Surg. 1995; 22(5): 532-7. 300. Eidson JL, 3rd, Bush RL. Diagnosis and current management of incompetent perforator veins. Semin Vasc Surg. 2010; 23(2): 113-7. 301. Louisy F, Cauquil D, Andre-Deshays C, Schroiff P, Lazerges M, Lafaye C, et al. Air plethysmography: an alternative method for assessing peripheral circulatory adaptations during spaceflights. Eur J Appl Physiol. 2001; 85(3-4): 383-91. 302. Weingarten MS, Czeredarczuk M, Scovell S, Branas CC, Mignogna GM, Wolferth CC, Jr. A correlation of air plethysmography and color-flow-assisted duplex scanning in the quantification of chronic venous insufficiency. J Vasc Surg. 1996; 24(5): 750-4. 303. Delis KT, Gloviczki P, Wennberg PW, Rooke TW, Driscoll DJ. Hemodynamic impairment, venous segmental disease, and clinical severity scoring in limbs with Klippel-Trenaunay syndrome. J Vasc Surg. 2007; 45(3): 561-7. 304. Bays RA, Healy DA, Atnip RG, Neumyer M, Thiele BL. Validation of air plethysmography, photoplethysmography, and duplex ultrasonography in the evaluation of severe venous stasis. J Vasc Surg. 1994; 20(5): 721-7. 305. Darvall KA, Sam RC, Bate GR, Adam DJ, Silverman SH, Bradbury AW. Photoplethysmographic venous refilling times following ultrasound guided foam sclerotherapy for symptomatic superficial venous reflux: relationship with clinical outcomes. Eur J Vasc Endovasc Surg. 2010; 40(2): 267-72. 306. van Bemmelen PS, van Ramshorst B, Eikelboom BC. Photoplethysmography reexamined: lack of correlation with duplex scanning. Surgery. 1992; 112(3): 544-8. 307. Fletcher JP, Kershaw LZ, Barker DS, Koutts J, Varnava A. Ultrasound diagnosis of lower limb deep venous thrombosis. Med J Aust. 1990; 153(8): 453-5. 308. Sauerbrei E, Thomson JG, McLachlan MS, Musial J. Observer variation in lower limb venography. J Can Assoc Radiol. 1981; 32(1): 28-9. 309. de Valois JC, van Schaik CC, Verzijlbergen F, van Ramshorst B, Eikelboom BC, Meuwissen OJ. Contrast venography: from gold standard to 'golden backup' in clinically suspected deep vein thrombosis. Eur J Radiol. 1990; 11(2): 131-7. 310. Raju S, Fredericks R. Venous obstruction: an analysis of one hundred thirty-seven cases with hemodynamic, venographic, and clinical correlations. J Vasc Surg. 1991; 14(3): 305-13. 311. Uhl JF, Gillot C, Chahim M. Anatomical variations of the femoral vein. J Vasc Surg. 2010; 52(3): 714-9.

362

312. Gaitini D. Multimodality imaging of the peripheral venous system. Int J Biomed Imaging. 2007; 2007: 54616. 313. Orbell JH, Smith A, Burnand KG, Waltham M. Imaging of deep vein thrombosis. Br J Surg. 2008; 95(2): 137-46. 314. Kistner RL, Ferris EB, Randhawa G, Kamida C. A method of performing descending venography. J Vasc Surg. 1986; 4(5): 464-8. 315. Garrett JV, Passman MA, Guzman RJ, Dattilo JB, Naslund TC. Expanding options for bedside placement of inferior vena cava filters with intravascular ultrasound when transabdominal duplex ultrasound imaging is inadequate. Ann Vasc Surg. 2004; 18(3): 329-34. 316. Baldt MM, Zontsich T, Stumpflen A, Fleischmann D, Schneider B, Minar E, et al. Deep venous thrombosis of the lower extremity: efficacy of spiral CT venography compared with conventional venography in diagnosis. Radiology. 1996; 200(2): 423-8. 317. Loud PA, Katz DS, Klippenstein DL, Shah RD, Grossman ZD. Combined CT venography and pulmonary angiography in suspected thromboembolic disease: diagnostic accuracy for deep venous evaluation. AJR Am J Roentgenol. 2000; 174(1): 61-5. 318. Marston W, Fish D, Unger J, Keagy B. Incidence of and risk factors for iliocaval venous obstruction in patients with active or healed venous leg ulcers. J Vasc Surg. 2011; 53(5): 1303-8. 319. Oguzkurt L, Tercan F, Pourbagher MA, Kizilkilic O, Turkoz R, Boyvat F. Computed tomography findings in 10 cases of iliac vein compression (May-Thurner) syndrome. Eur J Radiol. 2005; 55(3): 421-5. 320. Stern JB, Abehsera M, Grenet D, Friard S, Couderc LJ, Scherrer A, et al. Detection of pelvic vein thrombosis by magnetic resonance angiography in patients with acute pulmonary embolism and normal lower limb compression ultrasonography. Chest. 2002; 122(1): 115-21. 321. Wolpert LM, Rahmani O, Stein B, Gallagher JJ, Drezner AD. Magnetic resonance venography in the diagnosis and management of May-Thurner syndrome. Vasc Endovascular Surg. 2002; 36(1): 51-7. 322. Cao P, Eckstein HH, De Rango P, Setacci C, Ricco JB, de Donato G, et al. Chapter II: Diagnostic methods. Eur J Vasc Endovasc Surg. 2011; 42 Suppl 2: S13-32. 323. Edelman RR, Koktzoglou I. Unenhanced flow-independent MR venography by using signal targeting alternative radiofrequency and flow-independent relaxation enhancement. Radiology. 2009; 250(1): 236-45. 324. Evans AJ, Sostman HD, Witty LA, Paulson EK, Spritzer CE, Hertzberg BS, et al. Detection of deep venous thrombosis: prospective comparison of MR imaging and sonography. J Magn Reson Imaging. 1996; 6(1): 44- 51. 325. Du J, Thornton FJ, Mistretta CA, Grist TM. Dynamic MR venography: an intrinsic benefit of time- resolved MR angiography. J Magn Reson Imaging. 2006; 24(4): 922-7. 326. Lanzer P, Gross GM, Keller FS, Pohost GM. Sequential 2D inflow venography: initial clinical observations. Magn Reson Med. 1991; 19(2): 470-6. 327. Holtz DJ, Debatin JF, McKinnon GC, Unterweger M, Wildermuth S, von Schulthess GK, et al. MR venography of the calf: value of flow-enhanced time-of-flight echoplanar imaging. AJR Am J Roentgenol. 1996; 166(3): 663-8. 328. Bluemke DA, Wolf RL, Tani I, Tachiki S, McVeigh ER, Zerhouni EA. Extremity veins: evaluation with fast-spin-echo MR venography. Radiology. 1997; 204(2): 562-5. 329. Gallix BP, Achard-Lichere C, Dauzat M, Bruel JM, Lopez FM. Flow-independent magnetic resonance venography of the calf. J Magn Reson Imaging. 2003; 17(4): 421-6. 330. Ruehm SG, Wiesner W, Debatin JF. Pelvic and lower extremity veins: contrast-enhanced three- dimensional MR venography with a dedicated vascular coil-initial experience. Radiology. 2000; 215(2): 421-7. 331. Lebowitz JA, Rofsky NM, Krinsky GA, Weinreb JC. Gadolinium-enhanced body MR venography with subtraction technique. AJR Am J Roentgenol. 1997; 169(3): 755-8. 332. Fraser DG, Moody AR, Davidson IR, Martel AL, Morgan PS. Deep venous thrombosis: diagnosis by using venous enhanced subtracted peak arterial MR venography versus conventional venography. Radiology. 2003; 226(3): 812-20. 333. Zhang ZQ, Liu QH. Three-dimensional nonlinear image reconstruction for microwave biomedical imaging. IEEE Trans Biomed Eng. 2004; 51(3): 544-8. 334. Ziyeh S, Spreer J, Rossler J, Strecker R, Hochmuth A, Schumacher M, et al. Parkes Weber or Klippel- Trenaunay syndrome? Non-invasive diagnosis with MR projection angiography. Eur Radiol. 2004; 14(11): 2025-9. 335. Godshall CJ. Computed tomographic angiography allows accurate planning of the setting and technique of open and percutaneous vascular interventions. Am J Surg. 2005; 190(2): 218-20. 336. Gamblin TC, Ashley DW, Burch S, Solis M. A prospective evaluation of a bedside technique for placement of inferior vena cava filters: accuracy and limitations of intravascular ultrasound. Am Surg. 2003; 69(5): 382-6. 337. Conners MS, 3rd, Becker S, Guzman RJ, Passman MA, Pierce R, Kelly T, et al. Duplex scan-directed placement of inferior vena cava filters: a five-year institutional experience. J Vasc Surg. 2002; 35(2): 286-91. 338. Danetz JS, McLafferty RB, Ayerdi J, Gruneiro LA, Ramsey DE, Hodgson KJ. Selective venography versus nonselective venography before vena cava filter placement: evidence for more, not less. J Vasc Surg. 2003; 38(5): 928-34.

363

339. Aljabri B, MacDonald PS, Satin R, Stein LS, Obrand DI, Steinmetz OK. Incidence of major venous and renal anomalies relevant to aortoiliac surgery as demonstrated by computed tomography. Ann Vasc Surg. 2001; 15(6): 615-8. 340. Masuda EM, Kessler DM, Lurie F, Puggioni A, Kistner RL, Eklof B. The effect of ultrasound-guided sclerotherapy of incompetent perforator veins on venous clinical severity and disability scores. J Vasc Surg. 2006; 43(3): 551-6; discussion 6-7. 341. Howlader MH, Smith PD. Microangiopathy in chronic venous insufficiency: quantitative assessment by capillary microscopy. Eur J Vasc Endovasc Surg. 2003; 26(3): 325-31. 342. Cobbold JF, Patel D, Fitzpatrick JA, Patel N, Crossey MM, Abdalla MS, et al. Accuracy and reliability of microbubble ultrasound measurements for the non-invasive assessment of hepatic fibrosis in chronic hepatitis C. Hepatol Res. 2012; 42(5): 515-22. 343. Shalhoub J, Owen DR, Gauthier T, Monaco C, Leen EL, Davies AH. The use of contrast enhanced ultrasound in carotid arterial disease. Eur J Vasc Endovasc Surg. 2010; 39(4): 381-7. 344. Pierce IT, Gatehouse PD, Kalodiki E, Lattimer C, Geroulakos G, Xu XY, et al. Magnetic resonance venous velocity mapping during intermittent pneumatic compression of the calf and foot. Phlebology. 2011. 345. Pierce IT, Gatehouse PD, Xu XY, Firmin DN. MR phase-contrast velocity mapping methods for measuring venous blood velocity in the deep veins of the calf. J Magn Reson Imaging. 2011; 34(3): 634-44. 346. Veerman DP, Imholz BP, Wieling W, Wesseling KH, van Montfrans GA. Circadian profile of systemic hemodynamics. Hypertension. 1995; 26(1): 55-9. 347. Mehlsen J. Models of the venous system. Stud Health Technol Inform. 2000; 71: 109-17. 348. Snyder MF, Rideout VC. Computer simulation studies of the venous circulation. IEEE Trans Biomed Eng. 1969; 16(4): 325-34. 349. Brown SM, Fennigkoh L, Gerrits R, Hietpas M, Tritt C. A model of venous return while utilizing vacuum assist during cardiopulmonary bypass. J Extra Corpor Technol. 2003; 35(3): 224-9. 350. Buxton GA, Clarke N. Computational phlebology: the simulation of a vein valve. J Biol Phys. 2006; 32(6): 507-21. 351. Zervides C, Narracott AJ, Lawford PV, Hose DR. The role of venous valves in pressure shielding. Biomed Eng Online. 2008; 7: 8. 352. Diaz-Zuccarini V, Narracott AJ, Burriesci G, Zervides C, Rafiroiu D, Jones D, et al. Adaptation and development of software simulation methodologies for cardiovascular engineering: present and future challenges from an end-user perspective. Philos Trans A Math Phys Eng Sci. 2009; 367(1898): 2655-66. 353. Downie SP, Firmin DN, Wood NB, Thom SA, Hughes AD, Wolfe JN, et al. Role of MRI in investigating the effects of elastic compression stockings on the deformation of the superficial and deep veins in the lower leg. J Magn Reson Imaging. 2007; 26(1): 80-5. 354. Downie SP, Raynor SM, Firmin DN, Wood NB, Thom SA, Hughes AD, et al. Effects of elastic compression stockings on wall shear stress in deep and superficial veins of the calf. Am J Physiol Heart Circ Physiol. 2008; 294(5): H2112-20. 355. Chandran K, Yoganathan A, Rittergs S. Biofluid Mechanics: The Human Circulation: Taylor and Francis; 2006. 356. Papaioannou TG, Stefanadis C. Vascular wall shear stress: basic principles and methods. Hellenic J Cardiol. 2005; 46(1): 9-15. 357. Briceno M, Joseph D. Self-lubricated transport of aqueous foams in horizontal conduits. International Journal of Multiphase Flow. 2003; 29(12): 1817-31. 358. Wang Y, Pierce I, Gatehouse P, Wood N, Firmin D, Xu XY. Analysis of flow and wall shear stress in the peroneal veins under external compression based on real-time MR images. Med Eng Phys. 2012; 34(1): 17-27. 359. Karino T, Motomiya M. Flow through a venous valve and its implication for thrombus formation. Thromb Res. 1984; 36(3): 245-57. 360. Sevitt S. The structure and growth of valve-pocket thrombi in femoral veins. J Clin Pathol. 1974; 27(7): 517-28. 361. Matsumoto H, Hasegawa T, Fuse K, Yamamoto M, Saigusa M. A new vascular prosthesis for a small caliber artery. Surgery. 1973; 74(4): 519-23. 362. Badylak SF, Freytes DO, Gilbert TW. Extracellular matrix as a biological scaffold material: Structure and function. Acta Biomater. 2009; 5(1): 1-13. 363. Reing JE, Brown BN, Daly KA, Freund JM, Gilbert TW, Hsiong SX, et al. The effects of processing methods upon mechanical and biologic properties of porcine dermal extracellular matrix scaffolds. Biomaterials. 2010; 31(33): 8626-33. 364. Tottey S, Johnson SA, Crapo PM, Reing JE, Zhang L, Jiang H, et al. The effect of source animal age upon extracellular matrix scaffold properties. Biomaterials. 2011; 32(1): 128-36. 365. Brown BN, Barnes CA, Kasick RT, Michel R, Gilbert TW, Beer-Stolz D, et al. Surface characterization of extracellular matrix scaffolds. Biomaterials. 2010; 31(3): 428-37.

364

366. Tottey S, Corselli M, Jeffries EM, Londono R, Peault B, Badylak SF. Extracellular matrix degradation products and low-oxygen conditions enhance the regenerative potential of perivascular stem cells. Tissue Eng Part A. 2011; 17(1-2): 37-44. 367. Chiu YN, Norris RA, Mahler G, Recknagel A, Butcher JT. Transforming growth factor beta, bone morphogenetic protein, and vascular endothelial growth factor mediate phenotype maturation and tissue remodeling by embryonic valve progenitor cells: relevance for heart valve tissue engineering. Tissue Eng Part A. 2010; 16(11): 3375-83. 368. Gwanmesia P, Ziegler H, Eurich R, Barth M, Kamiya H, Karck M, et al. Opposite effects of transforming growth factor-beta1 and vascular endothelial growth factor on the degeneration of aortic valvular interstitial cell are modified by the extracellular matrix protein fibronectin: implications for heart valve engineering. Tissue Eng Part A. 2010; 16(12): 3737-46. 369. Sacks MS, Schoen FJ, Mayer JE. Bioengineering challenges for heart valve tissue engineering. Annu Rev Biomed Eng. 2009; 11: 289-313. 370. Valentin JE, Stewart-Akers AM, Gilbert TW, Badylak SF. Macrophage participation in the degradation and remodeling of extracellular matrix scaffolds. Tissue Eng Part A. 2009; 15(7): 1687-94. 371. Brown BN, Valentin JE, Stewart-Akers AM, McCabe GP, Badylak SF. Macrophage phenotype and remodeling outcomes in response to biologic scaffolds with and without a cellular component. Biomaterials. 2009; 30(8): 1482-91. 372. Ghanbari H, Viatge H, Kidane AG, Burriesci G, Tavakoli M, Seifalian AM. Polymeric heart valves: new materials, emerging hopes. Trends Biotechnol. 2009; 27(6): 359-67. 373. Sachweh JS, Daebritz SH. Novel "biomechanical" polymeric valve prostheses with special design for aortic and mitral position: a future option for pediatric patients? ASAIO J. 2006; 52(5): 575-80. 374. Christenson EM, Dadsetan M, Wiggins M, Anderson JM, Hiltner A. Poly(carbonate urethane) and poly(ether urethane) biodegradation: in vivo studies. J Biomed Mater Res A. 2004; 69(3): 407-16. 375. Christenson EM, Anderson JM, Hiltner A. Oxidative mechanisms of poly(carbonate urethane) and poly(ether urethane) biodegradation: in vivo and in vitro correlations. J Biomed Mater Res A. 2004; 70(2): 245-55. 376. Kannan RY, Salacinski HJ, Odlyha M, Butler PE, Seifalian AM. The degradative resistance of polyhedral oligomeric silsesquioxane nanocore integrated polyurethanes: an in vitro study. Biomaterials. 2006; 27(9): 1971-9. 377. Ghanbari H, Kidane AG, Burriesci G, Ramesh B, Darbyshire A, Seifalian AM. The anti-calcification potential of a silsesquioxane nanocomposite polymer under in vitro conditions: potential material for synthetic leaflet heart valve. Acta Biomater. 2010; 6(11): 4249-60. 378. Pinchuk L, Wilson GJ, Barry JJ, Schoephoerster RT, Parel JM, Kennedy JP. Medical applications of poly(styrene-block-isobutylene-block-styrene) ("SIBS"). Biomaterials. 2008; 29(4): 448-60. 379. Gallocher SL, Aguirre AF, Kasyanov V, Pinchuk L, Schoephoerster RT. A novel polymer for potential use in a trileaflet heart valve. J Biomed Mater Res B. 2006; 79B(2): 325-34. 380. Wang Q, McGoron AJ, Bianco R, Kato Y, Pinchuk L, Schoephoerster RT. In-vivo assessment of a novel polymer (SIBS) trileaflet heart valve. J Heart Valve Dis. 2010; 19(4): 499-505. 381. Schmidt D, Dijkman PE, Driessen-Mol A, Stenger R, Mariani C, Puolakka A, et al. Minimally-invasive implantation of living tissue engineered heart valves: a comprehensive approach from autologous vascular cells to stem cells. J Am Coll Cardiol. 2010; 56(6): 510-20. 382. L'Heureux N, McAllister TN, de la Fuente LM. Tissue-engineered blood vessel for adult arterial revascularization. New Engl J Med. 2007; 357(14): 1451-3. 383. McAllister TN, Maruszewski M, Garrido SA, Wystrychowski W, Dusserre N, Marini A, et al. Effectiveness of haemodialysis access with an autologous tissue-engineered vascular graft: a multicentre cohort study. Lancet. 2009; 373(9673): 1440-6. 384. Teebken OE, Bader A, Steinhoff G, Haverich A. Tissue engineering of vascular grafts: Human cell seeding of decellularised porcine matrix. Eur J Vasc Endovasc. 2000; 19(4): 381-6. 385. Sun M, Zhou P, Pan LF, Liu S, Yang HX. Enhanced cell affinity of the silk fibroin- modified PHBHHx material. J Mater Sci-Mater M. 2009; 20(8): 1743-51. 386. Ramcharitar S, Serruys PW. Fully Biodegradable Coronary Stents Progress to Date. Am J Cardiovasc Drug. 2008; 8(5): 305-14. 387. Maleti O, Lugli M. Neovalve construction in postthrombotic syndrome. J Vasc Surg. 2006; 43(4): 794-9. 388. Park J, Kurosawa S, Takai M, Ishihara K. Antibody immobilization to phospholipid polymer layer on gold substrate of quartz crystal microbalance immunosensor. Colloids Surf B Biointerfaces. 2007; 55(2): 164-72. 389. Byambaa B, Konno T, Ishihara K. Detachment of cells adhered on the photoreactive phospholipid polymer surface by photoirradiation and their functionality. Colloids Surf B Biointerfaces. 2013; 103: 489-95. 390. Kim HI, Ishihara K, Lee S, Seo JH, Kim HY, Suh D, et al. Tissue response to poly(L-lactic acid)-based blend with phospholipid polymer for biodegradable cardiovascular stents. Biomaterials. 2011; 32(9): 2241-7. 391. Soletti L, Nieponice A, Hong Y, Ye SH, Stankus JJ, Wagner WR, et al. In vivo performance of a phospholipid-coated bioerodable elastomeric graft for small-diameter vascular applications. J Biomed Mater Res A. 2011; 96(2): 436-48.

365

392. Ishihara K, Nomura H, Mihara T, Kurita K, Iwasaki Y, Nakabayashi N. Why do phospholipid polymers reduce protein adsorption? J Biomed Mater Res. 1998; 39(2): 323-30. 393. Ishihara K, Ziats NP, Tierney BP, Nakabayashi N, Anderson JM. Protein adsorption from human plasma is reduced on phospholipid polymers. J Biomed Mater Res. 1991; 25(11): 1397-407. 394. Liu Y, Inoue Y, Mahara A, Kakinoki S, Yamaoka T, Ishihara K. Durable modification of segmented polyurethane for elastic blood-contacting devices by graft-type 2-methacryloyloxyethyl phosphorylcholine copolymer. J Biomater Sci Polym Ed. 2014: 1-16. 395. Ye SH, Johnson Jr CA, Woolley JR, Oh H-I, Gamble LJ, Ishihara K, et al. Surface modification of a titanium alloy with a phospholipid polymer prepared by a plasma-induced grafting technique to improve surface thromboresistance. Colloids and Surfaces B: Biointerfaces. 2009; 74(1): 96-102. 396. Duncan R. Polymer conjugates as anticancer nanomedicines. Nat Rev Cancer. 2006; 6(9): 688-701. 397. Vicent MJ, Greco F, Nicholson RI, Paul A, Griffiths PC, Duncan R. Polymer therapeutics designed for a combination therapy of hormone-dependent cancer. Angew Chem Int Ed Engl. 2005; 44(26): 4061-6. 398. Barz M, Tarantola M, Fischer K, Schmidt M, Luxenhofer R, Janshoff A, et al. From defined reactive diblock copolymers to functional HPMA-based self-assembled nanoaggregates. Biomacromolecules. 2008; 9(11): 3114-8. 399. van Oeveren W, Haan J, Lagerman P, Schoen P. Comparison of coagulation activity tests in vitro for selected biomaterials. Artif Organs. 2002; 26(6): 506-11. 400. Henkelman S, Rakhorst G, Blanton J, Van Oeveren W. Standardization of incubation conditions for hemolysis testing of biomaterials. Materials Science and Engineering C - Biomimetic and Supramolecular Systems. 2009; 29: 1650-4. 401. Teng Z, Tang D, Zheng J, Woodard PK, Hoffman AH. An experimental study on the ultimate strength of the adventitia and media of human atherosclerotic carotid arteries in circumferential and axial directions. J Biomech. 2009; 42(15): 2535-9. 402. Assoul N, Flaud P, Chaouat M, Letourneur D, Bataille I. Mechanical properties of rat thoracic and abdominal aortas. J Biomech. 2008; 41(10): 2227-36. 403. Holzapfel GA, Sommer G, Gasser CT, Regitnig P. Determination of layer-specific mechanical properties of human coronary arteries with nonatherosclerotic intimal thickening and related constitutive modeling. Am J Physiol Heart Circ Physiol. 2005; 289(5): H2048-58. 404. Rossmann JS. Elastomechanical properties of bovine veins. J Mech Behav Biomed Mater. 2010; 3(2): 210- 5. 405. Wang SK, Qiu YC, Phifer TJ, Delaria GA, Tu R, Quijano RC, et al. In vitro performance of venous valve prostheses. An experimental model study. ASAIO J. 1992; 38(3): M213-5.

366