Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease

undergoing Surgery

Mr Pritesh S Morar MBBS MRCS PgCert (Medical Education)

SURGICAL EPIDEMIOLOGICAL TRIALS AND OUTCOMES CENTRE (SETOC)

DIVISION OF SURGERY, DEPARTMENT OF SURGERY AND CANCER,

IMPERIAL COLLEGE LONDON

Thesis towards Doctor of Philosophy (PhD) 2016

Supervisors:

Mr Janindra Warusavitarne

Professor Omar Faiz

Professor Ailsa Hart

1

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

CONTENTS

ABSTRACT ...... 17

ACKNOWLEDGEMENTS ...... 18

ABBREVIATIONS ...... 19

PUBLICATIONS AND PRESENTATIONS RESULTING FROM THIS THESIS ...... 21

STATEMENT OF CONTRIBUTION AND DECLARATION OF ORIGINALITY ..... 27

THESIS HYPOTHESIS, AIMS AND OBJECTIVES ...... 28

THESIS STRUCTURE ...... 29

CHAPTER 1: INTRODUCTION AND BACKGROUND ...... 31

1.1. The Nature of Crohn’s Disease ...... 31

1.1.1. History ...... 31

1.1.2. Epidemiology...... 33

1.1.3. Aetiology ...... 34

1.1.4. The behaviour of Crohn’s Disease ...... 36

1.1.5. Classification systems: a historical perspective...... 38

1.1.6. The Montreal Classification ...... 39

1.2. Management strategies ...... 39

1.2.1. An overview ...... 39

1.2.2. Medical strategies ...... 40

1.2.3. Endoscopic balloon dilatation ...... 42

1.2.4. Surgical strategies ...... 44

2

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

1.3. Surgical outcomes ...... 51

1.3.1. Introduction ...... 51

1.3.2. Short-term outcome measures for small bowel and ileocolonic resection ...... 51

1.3.3. Predictors of intra-abdominal septic complication ...... 52

1.3.4. Long term outcome following ileocaecal resection – the recurrent concern ...... 54

1.4. Variability and standards of care ...... 58

1.4.1. Variability in decision-making ...... 58

1.4.2. The patient perspective of surgery...... 59

1.4.3. The quality of care within an IBD service provision...... 60

1.4.4. The centralisation of care ...... 62

1.4.5 Standardising multidisciplinary team driven care ...... 63

CHAPTER 2: METHODS AND MATERIALS ...... 65

2.1. Materials ...... 65

2.1.1. The Crohn’s surgical database ...... 65

2.1.2. Database screening ...... 66

2.1.3. Data extraction ...... 67

2.1.4. Outcome measures ...... 68

2.1.5. Data validation ...... 71

2.1.6. Statistical analyses ...... 71

2.1.7. Missing data ...... 71

2.2. Qualitative methodology ...... 72

3

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

2.2.1. Rationale ...... 72

2.2.2. Semi-structured interviews ...... 73

2.3. Achieving consensus through Delphi ...... 74

CHAPTER 3: THE INSTITUTIONAL INCIDENCE AND PREDICTORS OF INTRA-

ABDOMINAL SEPTIC COMPLICATIONS (IASC) FOLLOWING ILEOCOLONIC

RESECTION FOR ILEAL CROHN'S DISEASE ...... 76

3.1. Abstract ...... 76

3.2. Aims ...... 77

3.3. Methodology ...... 77

3.3.1. Patient selection ...... 77

3.3.2. Outcome measures………………………………………………………………..78

3.3.3. Statistical analysis……………………………………………………………… 79

3.4. Results ...... 78

3.4.1. Patient and operative demographics ...... 78

3.4.2. The incidence of IASCs and subsequent management ...... 79

3.4.3. One-stage procedures ...... 82

3.4.4. Two-stage procedures ...... 89

3.5. Discussion ...... 90

3.5.1. Summary of findings ...... 90

3.5.2. Limitations ...... 91

4

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

CHAPTER 4: THE INSTITUTIONAL INCIDENCE AND PREDICTORS OF POST-

OPERATIVE CLINICAL RECURRENCE FOLLOWING ILEOCOLONIC

RESECTION FOR ILEAL CROHN'S DISEASE ...... 92

4.1 Abstract ...... 92

4.2. Aims ...... 93

4.3. Methodology ...... 93

4.3.1. Patient selection ...... 93

4.3.2. Outcome measures ...... 94

4.3.3. Statistical analysis...... 94

4.4. Results ...... 95

4.4.1. Patient and operative demographics ...... 95

4.4.2. The incidence of clinical recurrence ...... 97

4.4.3. Pre-operative factors ...... 99

4.4.4. Peri-operative factors ...... 103

4.4.5. Multivariate analysis...... 105

4.5. Discussion ...... 106

4.5.1. Summary of findings ...... 106

4.5.2 Limitations ...... 108

CHAPTER 5: THE EFFICACY OF ENDOSCOPIC BALLOON DILATATION IN

AVOIDING SURGERY FOR CROHN'S STRICTURES ...... 109

5.1. Abstract ...... 109

5.2. Aims ...... 110

5

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

5.3. Methodology ...... 110

5.3.1. Protocol and registration ...... 110

5.3.2. Eligibility criteria ...... 110

5.3.3. Information sources ...... 111

5.3.4. Search ...... 111

5.3.5. Data collection process ...... 111

5.3.6. Data items ...... 112

5.3.7. Risk of bias in individual studies ...... 112

5.3.8. Summary measures ...... 113

5.3.9. Synthesis of results ...... 113

5.3.10. Risk of bias across studies ...... 114

5.3.11. Additional analyses ...... 114

5.4. Results ...... 114

5.4.1. Study selection ...... 114

5.4.2. Study characteristics ...... 120

5.4.3. Synthesis of results ...... 122

5.5. Discussion ...... 135

5.5.1. Summary of evidence ...... 135

5.5.2. Limitations ...... 137

CHAPTER 6: THE DEVELOPMENT AND VALIDATION OF AN EVIDENCE-

BASED, END-USER INFORMED RADIOLOGICAL STAGING TOOL FOR

ILEOCOLONIC CROHN’S DISEASE ...... 139

6

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

6.1. Abstract ...... 139

6.2. Aims ...... 140

6.3. Methodology ...... 140

6.3.1. Tool construction ...... 140

6.3.2. Validity and reliability testing ...... 143

6.4. Results ...... 145

6.4.1. Literature review ...... 145

6.4.2. Semi-structured interviews ...... 149

6.4.3. Tool construction ...... 152

6.4.4. Validity testing ...... 154

6.4.5. Reliability testing ...... 166

6.5. Discussion ...... 168

6.5.1. Summary of findings ...... 168

6.5.2. Limitations ...... 171

CHAPTER 7: DEFINING KEY PERFORMANCE INDICATORS (KPIs) TO

MONITOR AND ENHANCE THE QUALITY OF CARE IN THE SURGICAL

MANAGEMENT OF ILEOCOLONIC CROHN’S DISEASE ...... 172

7.1. Abstract ...... 172

7.2. Aims ...... 173

7.3. Methodology ...... 173

7.3.1. Stage 1: semi-structured interviews ...... 173

7.3.2. Stage 2: Delphi consensus ...... 174

7

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

7.4. Results ...... 176

7.4.1. Participants ...... 176

7.4.2. Outcome measures of a surgical service provision ...... 178

7.4.3. Structure and organisation of IBD surgical services ...... 190

7.3.4. Clinical care processes ...... 198

7.4. Discussion ...... 202

7.4.1. Summary of findings ...... 202

7.4.2. Limitations ...... 205

CHAPTER 8: ESTABLISHING THE AIMS, FORMAT AND FUNCTION FOR

MULTIDISCIPLINARY TEAM DRIVEN CARE WITHIN AN INFLAMMATORY

BOWEL DISEASE SERVICE: A MULTI-CENTRE QUALITATIVE EXPERT-BASED

CONSENSUS STUDY ...... 207

8.1. Abstract ...... 207

8.2. Aims ...... 208

8.3. Methodology ...... 208

8.3.1. Stage 1: semi-structured interviews...... 208

8.3.2. Stage 2: Delphi consensus ...... 209

8.4. Results ...... 211

8.4.1. Participants ...... 211

8.4.2. Aims of the IBD MDT ...... 214

8.4.3. The role of key specialists in the IBD MDT ...... 221

8.4.4. Structural and organisational requirements for an effective IBD MDM ...... 225

8

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

8.4.5. Eligibility criteria for case discussion ...... 232

8.4.6. Outcome measures for the IBD MDT ...... 238

8.5. Discussion ...... 241

8.5.1. Summary of findings ...... 241

8.5.2. Limitations ...... 246

CHAPTER 9: THESIS DISCUSSION AND FUTURE WORK ...... 247

9.1. Summary of thesis ...... 247

9.2. Limitations ...... 250

9.3. Future work ...... 252

REFERENCES ...... 255

APPENDIX ...... 293

i. Semi-structured interview protocol: construct of an ileocolonic staging system for Crohn’s

disease ...... 293

ii. Semi-structured interview protocol (for expert participants): development of key

performance indicators for an IBD surgical service provision ...... 295

iii. Delphi survey: Devising Key Performance Indicators to Measure A Surgical Service

Provisions in Inflammatory Bowel Disease……………………………………………...295

iv. Semi-structured interview protocol (for expert participants): development of key

performance indicators for an IBD multidisciplinary team …………………………... 318

v. Delphi survey: Setting Standards for The IBD MDT - Obtaining Consensus Through a

Delphi Survey Approach…………………………………………………………………316

9

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

LIST OF FIGURES

Figure 1.1 Diagram representing basic concept of an ileocolonic resection followed by an anastomosis between A and B ...... 46

Figure 1. 2 Diagram representing basic concept of Heineke–Mikulicz strictureplasty ...... 48

Figure 1. 3 Diagram representing basic concept of Finney strictureplasty ...... 49

Figure 1. 4 Diagram representing basic concept of side-to-side isoperistaltic strictureplasty. 50

Figure 2. 1 Amalgamation of three prospectively maintained databases ...... 66

Figure 2. 2 Flow diagram demonstrating screening process and final number of patients (and procedures) included and excluded from the database...... 67

Figure 3. 1 Flow diagram demonstrating the final patient and procedure breakdown across one- stage and two-stage produces...... 79

Figure 4. 1 Flow diagram demonstrating the screening process and final number of procedures included and excluded from the study...... 95

Figure 4. 2 Pie chart demonstrating relative proportions of ileocolonic surgical procedures across the 142 included patients ...... 96

Figure 4. 3 Pie chart demonstrating proportions of patients who developed post-operative clinical recurrence and those who maintained remission out of the 142 patients ...... 97

Figure 4. 4 Life table analysis demonstrating the proportion of patients in remission (each drop in the curve represents a patient developing clinical recurrence) ...... 98

Figure 4. 5 Kaplan-Meier curves of months in remission for patients with ileal (L1) Crohn’s and ileocolonic (L3) Crohn’s disease ...... 101

Figure 4. 6 Kaplan-Meier curves of time in remission for patients undergoing a primary ileocaecal resection and redo ileocolonic resection ...... 102

Figure 4. 7 Kaplan-Meier curves of time in remission for patients with and without 30-day morbidity with IASCs...... 104

10

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

Figure 5. 1 Flow chart demonstrating the search strategy in accordance with PRISMA ...... 115

Figure 5. 2 Forrest plot reporting on the rate of surgical intervention in the event of a failed clinical outcome ...... 123

Figure 5. 3 Forest plot for studies reporting on symptomatic response...... 129

Figure 5. 4 Funnel plot for studies reporting on symptomatic response rate ...... 130

Figure 5. 5 Forest plot for studies reporting on technical response...... 131

Figure 5. 6 Forrest plot reporting on perforation rates ...... 133

Figure 6. 1 Flow chart demonstrating the search strategy in accordance with PRISMA guidance

...... 142

Figure 6. 2 – End-user opinion demonstrating items (n = number of interviewees) encompassing features consistent with the dichotomy of early and advanced disease (solid outline boxes) and features requiring surgical versus medication therapy (dashed outlined boxes) ...... 150

Figure 6. 3 The ileocolonic staging tool consists of four stages of escalating disease advancement mirroring radiological features identified from the evidence (i.e. literature review) and end-user opinion (i.e. semi-structured interviews) ...... 153

Figure 6. 4 Pie chart representing relative proportions per stage based on intra-operative and histopathological findings ...... 155

Figure 6. 5 Bar chart representing trends in males and females with each increasing stage . 159

Figure 6. 6 Bar chart representing trends in the number of patients with and without a pre- operative anaemia per stage ...... 160

Figure 6. 7 Bar chart representing trends in the number of patients with and without a pre- operative leukocytosis per stage ...... 161

Figure 6. 8 Bar chart representing the number of patients with and without a thrombocytosis per stage ...... 162

11

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

Figure 6. 9 Bar chart representing the number of patients with and without a raised CRP per stage ...... 163

Figure 6. 10 Bar chart representing trends in the number of patients who had a concomitant sigmoidectomy...... 164

Figure 6. 11 Bar chart representing trends in the number of patients who had a pre-emptive stoma formation ...... 165

Figure 6. 12 Bar chart representing trends in the number of patients who had a post-operative

IASC ...... 166

Figure 6. 13 Bar chart representing the number of patients based on the intra-operative and histopathological stage and the radiological stage...... 167

Figure 7. 1 Eligibility criteria for inclusion as an expert Delphi panellist ...... 175

Figure 7. 2 Pie chart representing relative proportions of participants within the interview study.

...... 176

Figure 7. 3 Pie chart demonstrating consensus on the number of major IBD resections performed per year by individual consultant colorectal surgeons to ensure a high quality service provision ...... 195

Figure 7. 4 Pie chart demonstrating consensus on the number of major IBD resections performed per year by institutions to ensure a high-quality service provision ...... 196

Figure 8. 1 Eligibility criteria for inclusion as a Delphi panellist...... 210

Figure 8. 2 Pie chart demonstrating relative proportion of participants within the multidisciplinary sample...... 212

Figure 8. 3 Perceived responses (and number of respondents) as to who should be core members, extended members and non-members of the IBD MDT ...... 223

12

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

LIST OF TABLES

Table 2. 1 OPCS codes and procedures relating to the ileum and ileocolonic bowel ...... 66

Table 2. 2 The Clavien-Dindo classification ...... 69

Table 2. 3 Pre-operative, peri-operative and post-operative variables obtained during data extraction...... 70

Table 3. 1 Demographics and operative profiles of 17 patients who demonstrated post-operative

IASCs ...... 81

Table 3. 2 Frequency data and univariate analyses for patient and disease related characteristics across 142 patients undergoing a one-stage procedure ...... 83

Table 3. 3 Data and univariate analysis for pre-operative medications across 142 patients undergoing a one-stage procedure ...... 85

Table 3. 4 Frequency data and univariate analysis for peri-operative blood parameters across

142 patients undergoing a one-stage procedure ...... 86

Table 3. 5 Frequency data and univariate analysis for peri-operative patient, disease and operative variables across 142 patients undergoing a one-stage procedure ...... 87

Table 3. 6 Multivariate analysis adjusting for smoking status, concomitant UGI Crohn’s, peri- operative biological therapy, peri-operative anaemia, peri-operative hypoalbuminaemia, intra- operative sepsis ...... 88

Table 4. 1 Data, univariate (log rank analyses) for patient and disease related pre-operative variables ...... 99

Table 4. 2 Data, univariate (log rank analyses) for peri-operative variables ...... 103

Table 4. 3 Multivariate analysis for all pre-operative, peri-operative and post-operative variables found to be significant (p < 0.1) on univariate analysis ...... 105

Table 5. 1 Quality assessment of studies using the Newcastle-Ottawa scale ………………113

13

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

Table 5. 2 Study demographics, frequency of population size, strictures and dilatations. Study demographics, frequency of population size, strictures and dilatations ...... 118

Table 5. 3 The relationship between response rates (symptomatic and technical) and adverse event rates (perforation) with categorical variables ...... 125

Table 5. 4 Meta-analysis in studies reporting on outcomes for active and quiescent strictures

...... 127

Table 5. 5 Meta-analysis in studies reporting on outcomes for anastomotic and de novo strictures ...... 127

Table 6. 1 Description of studies included in the literature review…………………………148

Table 6. 2 Development of the staging tool ...... 152

Table 6. 3 Frequency data, proportions and trend analyses for patient and disease related, pre- operative medications, pre-operative blood parameters, peri-operative and post-operative factors for each stage ...... 157

Table 7. 1 Expert consensus panellists (ECPs) detailing years in post, institution and country

………………………………………………………………………………………178

Table 7. 2 Outcome measures specific to small bowel and ileocolonic Crohn’s disease surgery

...... 186

Table 7. 3 Outcome measures for the overall quality of an IBD surgical service provision . 187

Table 7. 4 Quality assurance mechanisms to ensure a high quality IBD surgical service ..... 189

Table 7. 5 IBD team members necessary for an IBD surgical service provision ...... 194

Table 7. 6 Facilities necessary for an IBD surgical service provision ...... 198

Table 7. 7 Themes, items and respective Delphi Likert ratings for (a) Pre-operative and (b) post-operative considerations in the processes of delivering high quality IBD care ...... 201

Table 8. 1 Members of the multidisciplinary Expert Consensus

Panel………………………………………………………………………………………...213

14

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

Table 8. 2 Themes, parameters, number of respondents and representative quotations to the role and purpose of the MDT in IBD care from semi-structured interviews across 28 participants ...... 218

Table 8. 3 Delphi Likert ratings for items that describe the aim of the IBD MDT ...... 220

Table 8. 4 Items and respective responses N [%] for specialist and patient role as core, extended or non IBD MDT members ...... 225

Table 8. 5 Themes, parameters, number of respondents and representative quotations to factors required for an effective IBD MDT to occur, from semi-structured interviews across 28 participants ...... 229

Table 8. 6 Delphi Likert ratings for items ensure the IBD MDM runs smoothly and effectively

...... 231

Table 8. 7 Themes, parameters, number of respondents and representative quotations to the need of an eligibility criterion for case selection for the IBD MDT, from semi-structured interviews across 28 participants ...... 236

Table 8. 8 Delphi Likert ratings for items ensuring eligible cases for discussion in the IBD

MDM...... 237

Table 8. 9 Themes, parameters, number of respondents and representative quotations for outcome measures of the IBD MDT, from semi-structured interviews across 28 participants.

...... 240

Table 8. 10 Delphi Likert ratings for the outcome measure of an IBD MDT ...... 241

15

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

“The world as we have created it is a process of our thinking. It cannot be changed

without changing our thinking.”

Albert Einstein

16

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

ABSTRACT

This thesis provides a comprehensive overview of the variability seen at time of surgery for patients with ileocolonic Crohn’s disease. It questions the validity of the Montreal classification in the setting of surgery. It provides a consensus for how a surgical service provision and the multidisciplinary team can be standardised to enhance the quality of care for patients with ileocolonic Crohn’s disease. Through meta-analysis, this thesis demonstrates that surgery is an inevitable end-point in 75% of patients who undergo endoscopic balloon dilatation for Crohn’s strictures within a 5-year follow-up period. It raises the profile of surgery as an intervention for ileocolonic Crohn’s disease, as opposed to a ‘failure of medical therapy’ and goes on to devise key performance indicators to monitor and enhance the quality of surgery for patients with ileocolonic Crohn’s disease.

This thesis demonstrates that Montreal B3 (penetrating disease) and peri-operative biological therapy were both independent predictors of intra-abdominal septic complication. Further clarification is required to determine if fistulating or perforating disease specifically contributes to this effect. The newly devised, end-user informed staging tool provides distinct stages to aid this effect. Validity and reliability testing of this staging tool against a cohort of surgical patients has demonstrated it can reliably be used to pre-operatively stage patients and inform surgical decision-making and estimate post-operative morbidity.

The thesis has also found that re-resectional surgery and Montreal L3 disease (ileocolonic disease) are independent predictors of recurrent disease following surgery. Montreal L3 disease is a novel finding previously un-reported and should be considered in risk stratification models for post-operative prophylactic therapy.

17

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

ACKNOWLEDGEMENTS

Firstly, my thanks to my supervisor Mr Janindra Warusavitarne for taking me on as a PhD student in studies relating to Crohn’s disease. His experience in this field was pivotal in steering the direction of my thesis. I would like to also thank his support throughout this journey, clinical and pastoral, as well as the academic. My thanks also go to Professor Ailsa

Hart who has provided guidance on the gastroenterological and multidisciplinary perspective.

I would like to thank my supervisor Professor Omar Faiz who provided direction to the thesis specifically relating to quality improvement for patients undergoing ileocolonic surgery. His

‘global’ perspective on healthcare research and outcomes provided insight into how quality can be improved for patients with IBD undergoing surgery. His concepts on phenotypic variability in ileocolonic Crohn’s disease led to the staging tool described within, alongside recognition to homogenise the reporting of outcome measure following surgery.

My sincere thanks also to Professor Nick Sevdalis and Professor James Green for their input relating to the qualitative methodology, and their vast experience in the setting of multidisciplinary team driven care. Dr Cathryn Edwards for her input in multidisciplinary team driven care in the setting of inflammatory bowel disease. I would like to also thank Dr

Naila Arebi for her guidance in systematic review and meta-analysis which was required in assessing the efficacy of endoscopic balloon dilatation.

Finally, I would like to thank my wife Hansa, my daughter, Aryana and son, Rihaan, for their endless love, support, and for keeping me grounded. My sister, Dr Sonali Shah, and our parents

(Mr Sumant P Morar and Mrs Kumud S Morar), for their advice, and guidance. Finally, to

Dilip Damniwala and Kumud Morar – your strength and resolve in overcoming challenges are inspirational. Rest in peace Dilipmama.

18

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

ABBREVIATIONS

ASA AMERICAN SOCIETY OF ANASTHESIOLOGY

5-ASA 5-AMINOSALICYLIC ACID

6-MP 6 MERCAPTOPURINE

BMI BODY MASS INDEX

BWT BOWEL WALL THICKNESS

CARD15 CASPASE RECRUITMENT DOMAIN-CONTAINING PROTEIN 15

CD CLAVIEN-DINDO

CDAI CROHN’S DISEASE ACTIVITY INDEX

CI CONFIDENCE INTERVAL

CNS CLINICAL NURSE SPECIALIST

CRP C-REACTIVE PROTEIN

CTE COMPUTED TOMOGRAPHIC ENTEROGRAPHY

ECF ENTERO-CUTANEOUS FISTULA

GI GASTROINTESTINAL

HR HAZARD RATIO

I2 HETEROGENEITY ACROSS STUDIES

IASC INTRA-ABDOMINAL SEPTIC COMPLICATIONS

IBD INFLAMMATORY BOWEL DISEASE

IBDQ INFLAMMATORY BOWEL DISEASE QUESTIONNAIRE

ICD -10 INTERNATIONAL CLASSIFICATIONS OF DISEASES

IQR INTERQUARTILE RANGE

KPI KEY PERFORMANCE INDICATIOR

MCAR MISSING COMPLETELY AT RANDOM

19

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

MDM MULTI-DISCLINARY MEETING

MDT MULTI-DISCLINARY TEAM

MRE MAGNETIC RESONANCE ENTEROGRAPHY

NOD2 NUCLEOTIDE-BINDING OLIGOMERIZATION DOMAIN-

CONTAINING PROTEIN 2

OR ODDS RATIO

OPCS OPERATING PROCEDURE CODE SUPPLEMENT

PMER POOLED MEAN EVENT RATE

POCR POST-OPERATIVE CLINICAL RECURRENCE

PRISMA PREFERRED REPORTING ITEMS FOR SYSTEMATIC REVIEWS

AND META-ANALYSES

P VALUE PROBABILITY VALUE

QUIP QUALITY IMPROVEMENT PROJECT

SD STANDARD DEVIATION

SPSS STATISTICAL PRODUCT AND SERVICE SOLUTIONS

UGI UPPER GASTROINTESTINAL

USS ULTRASOUND SCAN

χ2 CHI SQUARED TEST

20

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

PUBLICATIONS AND PRESENTATIONS RESULTING FROM THIS THESIS

Publications directly resulting from thesis

Morar PS, Wasmann K, Fareleira A, Sahnan K, Adegbola SO, Mainta E, Ilangovan R, Arora

S, Sevdalis N, Koysombat K, Hart A, Burling D, Edwards C, Warusavitarne J, Gupta A,

Bemelman W & Faiz O. A Novel Ileocolonic Crohn’s Staging Tool - the development and validation of an evidence-based, end-user informed radiological decision-aid, (submitted, under peer review).

Lightner A, McKenna N, Warusavitarne J, Adegbola S, Morar P, Fleshner F, Spinelli A. Intra-

Abdominal Sepsis following Ileocecal Resection for Crohn’s Disease: Are the Risk Factors

Site Specific? (submitted, under peer review)

Morar PS, Sevdalis N, Warusavitarne J, Hart A, Green J, Edwards C, Faiz O. Establishing the aims, format and function for multidisciplinary team-driven care within an inflammatory bowel disease service: a multicentre qualitative specialist-based consensus study. Frontline

Gastroenterology, 2017 Aug 10.

Morar PS, Hollingshead J, Bemelman W, Sevdalis N, Pinkney T, Wilson G, Dunlop M, Justin

Davies R, Guy R, Fearnhead N, Brown S. Establishing Key Performance Indicators [KPIs] and

Their Importance for the Surgical Management of Inflammatory Bowel Disease–Results from a Pan-European, Delphi Consensus Study. Journal of Crohn's and Colitis. 2017 Jul 20.

21

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

Morar P, Faiz O, Warusavitarne J, Brown S, Cohen R, Hind D, Abercrombie J, Ragunath K,

Sanders DS, Arnott I, Wilson G, Bloom S, Arebi N. A systematic review and meta-analysis - endoscopic balloon dilatation for Crohn’s disease strictures. Alimentary Pharmacology and

Therapeutics, 2015 Sep 11.

Morar P, Read J, Arora S, Hart A, Warusavitarne J, Green J, Sevdalis N, Edwards C, Faiz O.

Defining the optimal design of the inflammatory bowel disease multi-disciplinary team – results from a multi-centre qualitative expert-based study. Frontline Gastroenterology. 2015

March 26.

Morar P, Faiz O, Hodgkinson J, Zafar N, Koysombat K, Purcell M, Hart AL, Warusavitarne

J. Concomitant Colonic Disease (Montreal L3) and Re-resectional Surgery are Predictors of

Clinical Recurrence Following Ileocolonic Resection for Crohn’s Disease. Colorectal Disease,

2015 Aug 20.

Morar P, Hodgkinson J, Thalayasingam S, Koysombat K, Purcell M, Hart AL, Warusavitarne

J, Faiz O. Determining predictors for intra-abdominal septic complications following ileocolonic resection for Crohn’s disease – considerations in pre-operative and peri-operative optimisation techniques to improve outcome. Journal of Crohn’s and Colitis. 2015 June 9.

Associated with work from thesis

Morar P, Hart A, Warusavitarne J. Issues Surrounding Post-operative Therapy in Crohn's

Disease to Prevent Recurrence. Clinical Gastroenterology & Hepatology. October 2014.

Volume 12, Issue 10, Pages 1763–1764.

22

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

Morar P, Bagnall M, Faiz O. Pre-operative Magnetic Resonance Enterography in Predicting

Findings and Optimizing Surgical Approach in Crohn's Disease. Journal of Gastrointestinal

Surgery (2014) 18:1721

Presentations

Morar P S, Wasmann KA, Fareleira ACT, Sahnan K, Adegbola SO, Mainta E, Ilangovan R,

Arora S, Sevdalis N, Koysombat K, Hart A, Burling D, Edwards C, Warusavitarne J, Gupta A,

Bemelman WA, Faiz OD. A Novel Ileocolonic Crohn’s Staging Tool - the development and validation of an evidence-based, end-user informed radiological decision-aid (A-1546). 14th

Congress of ECCO - Inflammatory Bowel Diseases 2019 (accepted).

Morar P, Warusavitarne J, Sevdalis N, Faiz O. The role of multidisciplinary team driven care in cancer management. Digestive Disorders Federation. 2015, London.

Morar P, Sevdalis N, Read J, Arora S, Warusavitarne J, Hart A, Green J, Edwards C, Faiz O.

The Role of the Multidisciplinary Team in IBD. Medicine and me: living with Crohn's and

Colitis. Royal Society of Medicine, 2015.

Morar P, Mainta E, Arora S, Ilangovan R, Hart A, Gupta A, Warusavitarne J, Faiz O.

Predicting surgical strategy in ileal Crohn’s disease – the construction and validation of an

23

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery evidence-based, end-user informed radiological staging tool (A-1170) 11th Congress of ECCO.

2016, Amsterdam.

Morar P, Mainta E; Hart A, Arora S; Gupta A; Warusavitarne J; Faiz O. Subcategorization of complex ileal Crohn’s disease in the aid of management algorithms – the development of a multidisciplinary, evidence-based, end-user informed radiological staging system. The 102nd

General Meeting of the Japanese Society of Gastroenterology, 2016, Tokyo.

Morar P, Sevdalis N, Read J, Arora S, Warusavitarne J, Hart A, Green J, Edwards C, Faiz O.

Establishing eligibility for case discussion in multidisciplinary team care within an inflammatory bowel disease service provision - results from a qualitative two-stage expert based study. Digestive Disorders Federation, 2015 London

Morar P, Faiz O, Warusavitarne J, Brown S, Cohen R, Hind D, Abercrombie J, Ragunath K,

Sanders DS, Arnott I, Wilson G, Bloom S, Arebi N. A systematic review and meta-analysis - endoscopic balloon dilatation for Crohn’s disease strictures. Digestive Disorders Federation,

2015 London

Morar P, Sevdalis N, Warusavitarne J, Green J, Hart A, Edwards C, Faiz O. Setting standards for multi-disciplinary team driven care in inflammatory bowel disease service provision - an expert consensus on key specialists to be involved. 10th European Crohn’s & Colitis Congress

(ECCO), 2015 Barcelona.

24

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

Morar P, Sevdalis N, Warusavitarne J, Green J, Hart A, Edwards C, Faiz O. Defining the aims of multi-disciplinary team driven care within an inflammatory bowel disease service provision – results from a Delphi consensus-building methodology. 10th European Crohn’s &

Colitis Congress (ECCO), 2015 Barcelona.

Morar P, Read J, Arora S, Hart A, Sevdalis N, Faiz O, Edwards C. Setting Standards by

Defining the Aims & Optimal Design of The Inflammatory Bowel Disease (IBD)

Multidisciplinary Team (MDT) Meeting. British Society of Gastroenterology, 2014

Manchester.

Morar P, Faiz O, Hart A, Warusavitarne J. The 5-year incidence of post-operative clinical recurrence following ileocolonic resection for terminal ileal Crohn’s disease. 10th European

Crohn’s & Colitis Congress (ECCO), 2015 Barcelona.

Morar P, Hodgkinson J, Thalayasingam S, Koysombat K, Faiz O, Hart A, Warusavitarne J.

Determining patient & disease related Risk factors for The Development of Clinical

Recurrence Following Ileocolic Resection for Crohn’s Disease within A tertiary IBD Unit. 9th

European Crohn’s & Colitis Congress (ECCO), 2014 Copenhagen.

25

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

Morar P, Purcell M, Faiz O, Hart A, Warusavitarne J. Determining predictors for the development of clinical recurrence following ileocolonic resection for Crohn’s disease within a specialist institution. Tripartite, 2014 Birmingham.

Morar P, Hodgkinson J, Hart A, Warusavitarne J, Faiz O. Determining predictors for intra- abdominal septic complications following ileocolonic resection for Crohn’s disease within a specialist institution. Tripartite, 2014 Birmingham.

Morar P, Purcell M, Hart A, Warusavitarne J, Faiz O. Determining Pre-operative Risk Factors for Intra-Abdominal Septic Complications Following Ileocolonic Resection for Crohn’s

Disease within a Tertiary Institution. International Surgical Congress of the Association of

Surgeons of Great Britain and Ireland. 2014 Harrogate.

26

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

STATEMENT OF CONTRIBUTION AND DECLARATION OF ORIGINALITY

The work contained in this thesis is my own and was performed by myself. In the process of carrying out this work, other individuals were involved. My supervisors, Mr Janindra

Warusavitarne, Mr Omar Faiz and Dr Ailsa Hart provided direction and advice with conception, design and editing of the studies included in this thesis. In addition, Professor Nick

Sevdalis, Professor James Green and Dr Cathryn Edwards provided direction on the design and editing of Chapters 7 and 8, with Dr James Read aiding data collection. Dr Arun Gupta, Dr

Raj Ilangovan, Dr Evgena Mainta and Miss Sonal Arora provided advice on the design and editing of Chapter 6. Dr Naila Arebi provide direction and advice with conception, design and editing of Chapter 5. All other work is appropriately referenced.

COPYRIGHT DECLARATION

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.

27

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

THESIS HYPOTHESIS, AIMS AND OBJECTIVES

Hypothesis

The quality of surgery for patients with ileocolonic Crohn’s disease can be enhanced by appropriate case-mix adjustment of intra-abdominal disease phenotypes, the development of key performance indicators and multidisciplinary team driven care.

Aims

The aim of this thesis is to determine factors that contribute to unfavourable outcome for patients with Crohn’s disease who undergo ileocolonic resection or endoscopic balloon dilatation, and to develop a staging tool and key performance indicators to improve the quality of surgical and multidisciplinary team driven care for these patients.

Objectives

1. To validate the Montreal classification against short-term and long-term post-operative

outcome following ileocolonic resection for Crohn’s disease.

2. To investigate the outcomes of endoscopic alternatives to surgery (endoscopic balloon

dilatation) in patients with ileal Crohn’s disease strictures.

3. To construct and validate a staging tool for ileocolonic Crohn’s disease to adequately

categorise pre-operative disease phenotypes and facilitate decision-making.

4. To construct metrics to monitor and enhance the performance of IBD surgical service

provision and multidisciplinary team care.

28

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

THESIS STRUCTURE

This thesis has been structured with a view to meeting the stated objectives. Chapter 1 explores the historic perspective of Crohn’s disease and provides a background detailing the disease burden, including epidemiology, risk factors, interventions and treatments to date. It also explores the role of quality and centralisation, and its importance in developing a surgical service provision. Chapter 2 provides a broad outline of the materials that were used including the database construct, and quantitative and qualitative methodologies used in the subsequent studies. Additional study specific methodologies are described in each subsequent chapter.

Chapters 3 and 4 seek to address objective 1. Chapter 3 examines the ileocolonic Crohn’s surgical database to explore the incidence of short term post-operative complications and variables that contribute to poor outcome following surgery. Chapter 4 examines the same database to explore the incidence of disease recurrence following surgery and variables that contribute to early recurrence.

Chapter 5 seeks to address objective 2. With varying reports on the efficacy of endoscopic balloon dilatation for Crohn’s strictures of the terminal ileum, Chapter 5 uses meta-analytical techniques to determine the short-term and long-term outcomes of this intervention, and its role in avoiding surgery.

Chapter 6 seeks to address objective 3, by developing a staging tool through qualitative techniques to categorise disease complexity in the pre-operative setting and provide validity and reliability evidence of the tool from the database.

29

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

Chapter 7 and 8 seek to address objective 4. Qualitative methodologies including semi- structured interview and Delphi consensus are used in both chapters, with Chapter 7 developing metrics to monitor and enhance the quality of the IBD surgical service provision, and Chapter

8 developing metrics for multidisciplinary team driven care, for patients with ileocolonic

Crohn’s disease.

Chapter 9 provides an overall summary of the finding relating to Chapters 3 to 8, in respect to the initial objectives, and any further research that could arise from this work.

30

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

CHAPTER 1: INTRODUCTION AND BACKGROUND

1.1. The Nature of Crohn’s Disease

1.1.1. History

The earliest descriptions of Crohn’s disease date back to the 18th century, where Italian physician Giovanni Battista Morgagni wrote De Sedibus et Causis Morborum (translated as

“The Seats and Causes of Diseases”). Specifically, Morgagni described the post-mortem anatomy and pathology of a man in his twenties with a longstanding history of diarrhoea with associated griping “tortures of the bowel” and “frequent bloody stools”. Further descriptions a perforation in the extremity of the ileum, within two hand breaths of the colon, which also demonstrated “ulcerations” and the “glands of the mesentery having grown out into a tumour”.

It was on May 13, 1932, before the Section of Gastroenterology and Proctology at the 83rd

Session of the American Medical Association where the landmark paper by Burrill B. Crohn

(1884 – 1983), Leon Ginzburg (1898 – 1988) and Gordon D. Oppenheimer (1900 – 1974) was presented. The paper was entitled “Regional Ileitis, a Pathological and Clinical Entity”, and it proposed:

“to describe, in its pathologic and clinical details, a disease of the terminal ileum,

affecting mainly young adults, characterized by a subacute or chronic necrotizing and

cicatrizing inflammation. The ulceration of the mucosa is accompanied by a

disproportionate connective tissue reaction of the remaining walls of the involved

intestine, a process which frequently leads to stenosis of the lumen of the intestine,

associated with the formation of multiple fistulas” (1).

31

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

The series of 14 patients described in the paper that shared the clinico-pathological descriptions were those of A. A. Berg, a senior surgeon at the Mount Sinai Hospital. Berg was invited to be an author however he declined on account of not having been involved in the study from the beginning. The eponym therefore fell to Crohn, as it was the policy of JAMA (Journal of the American Medical Association) at the time to list authors alphabetically. Had Berg accepted the invitation to be an author, the condition would have been known as Berg’s disease instead (2). Support for the eponym Crohn’s disease grew over time to avoid confusion over different terms being used, including “regional ileitis” and “cicatrizing enteritis”, to describe the same clinical entity (2).

The identification of the tubercle bacillus by Heinrich Robert Koch on March 24, 1882 was a turning point in differentiating Crohn’s disease as inflammatory as opposed to infective in aetiology (3). Sir Thomas Kennedy Dalziel (1861 – 1924) described nine patients with clinico- pathological features involving the small bowel and colon (19 years before Crohn and colleagues) at the British Medical Association conference in Glasgow in 1913. The pathologist observed eosinophils, giant cells and granulomas with no infectious agents. Specimens from these patients had also tested negative for tuberculosis. He also noted that despite successful colonic lavage, which may have contributed to the absence of micro-organisms, the disease continued to steadily progress.

A major distinction between the descriptions by Crohn and Dalziel was the presence of colonic disease. Crohn and colleagues described predominantly small bowel disease, whereas Dalziel described colonic involvement as well. Numerous reports in later years began to recognise

Crohn’s disease with oesophageal and gastric manifestations (4, 5). The pathological entity of

32

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

Crohn’s disease was vastly being recognised as being beyond the regional ileitis originally described, and that of a pan enteric condition.

There was, however, a large degree of uncertainty as to its association with ulcerative colitis and if Crohn's disease may give rise to pathological changes in the colon like those found in the terminal ileum. Hugh Evelyn Lockhart-Mummery (1960) provided a clear distinction of colonic Crohn’s disease as a separate entity to ulcerative colitis. The examination of colonic and enteric specimens taken from 25 patients from St Mark’s Hospital, London, demonstrated key clinical, radiographic and pathological differences between the two conditions (6).

Crohn’s Disease is today described as a chronic disease process that can affect the gut at any point from the mouth to the anus. It is categorised by the presence of transmural inflammation and skip lesions.

1.1.2. Epidemiology

Variation in disease location is one factor in determining the true incidence of Crohn’s disease.

There is also evidence of a geographical variation and gradient in incidence. From a global perspective, non-equatorial regions appear to have lower incidences compared to regions of a higher latitude. There is a growing incidence in countries that have rarely documented Crohn’s disease, and a plateau in the incidence in countries with the high incidences (> 10/10,000) (7).

The incidence across eight northern European centres demonstrated an incidence rate of 6.3 per 100,000 for all patients above the age of 15. The same study demonstrated a lower incidence of 3.6 per 100,000 across twelve centres in southern Europe. The same study went on to demonstrate higher rates in northern European centres when adjusting for age and gender

(8). Other studies have quoted annual incidence rates within Europe ranging from 0.7 – 9.8 /

33

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

100,000 (9). Numerous reports suggest the annual UK incidence of Crohn’s disease range from 5.6 / 10,000 (1991 – 1995) to 8.3 / 100,000 per year (10, 11). The 20th century has seen a rise in the incidence of Crohn’s disease over time, with an annual incidence of 0.7 / 100,000

(1951 – 1955) to 6.67 / 100,000 (1981 – 1985) (12).

There are also suggestions that the prevalence of Crohn’s disease is rising. European studies have quoted approximately 40 / 100,000 of those living in Europe that have Crohn’s disease

(13). The disease can affect persons at any age, but is greatest in those aged between 15-30 years, with 20-30% of affected patients being younger than 30-40 years (14). There are also reports of a second peak at 60–70 years (7). In areas that appear to have a high incidence of

Crohn’s disease, there is a 20-30% higher incidence in women, compared to men. In developing countries however, the incidence of Crohn’s disease in men appears to be equivalent or higher than that of women (7).

1.1.3. Aetiology

There is no clear causative agent for the development of Crohn’s disease, but it is postulated that there is an interplay between various factors which result in chronic inflammation within a genetically susceptible host: genetic susceptibility being influenced by the luminal microbiota, which provide antigens and adjuvants that stimulate either pathogenic or protective immune responses. Environmental triggers are necessary to initiate or reactivate disease expression in those who have genetic susceptibility (15).

1.1.3.1. Heritability and genetic susceptibility

Heritability is the proportion of phenotypic variance that can be attributed to genetic variance.

A family history of IBD is reported in 5% to 16% of patients with Crohn’s disease. Familial

34

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery aggregation is common in Crohn’s disease, and having one or more affected first degree relatives is a strong risk factor in the development of Crohn’s disease (16, 17). Twin studies have demonstrated a higher rate of concordance in the development of Crohn’s disease among monozygotic twins (range 20-55%) compared to dizygotic twins (0 – 3.6%) (16).

Mutations within the NOD2/CARD15 gene loci, are associated with Crohn’s disease (18), however these only account for twenty percent of the pathogenesis. NOD2 encodes for an intracellular protein receptor within CD4+ monocytes. It recognises antigens from Gram negative bacteria, driving the inflammatory process (18, 19) by releasing the NFyB transcription factor, which regulates transcription of chemokines and cytokines. Mutations in

NOD2 are thought to be involved in the altered immunological tolerance to gut flora seen in ileal Crohn’s, though the mechanism is not clear (20). Approximately 50% the Caucasian population with Crohn’s disease have mutations of the NOD2 gene, and mutations are more common in southern Europeans in comparison to northern Europeans. This is a paradoxical correlation with European epidemiological studies on Crohn’s disease (19, 21), which sees a higher incidence of Crohn’s disease in northern European countries. Furthermore, mutations associated with disease risk in Europe are not found in Asian or African populations (19).

Further complexities have been introduced in recent years with genome wide association studies. Over 163 single nucleotide proteins are found to have associations with IBD, of which the NOD2/CARD15 variant is one. Crohn’s disease is very much a polygenic condition.

Between 19 – 26% of these explain variation in liability of inheritance (16, 22), indicating environmental influences.

35

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

1.1.3.2. Micro and macro environmental triggers

The increasing incidence of Crohn’s disease in developing countries indicates that there are environmental triggers in genetically susceptible people. Understanding the aetiology of ileal

Crohn’s has been aided by numerous studies (23, 24) in particular those that have studied post- operative recurrence and predisposing risk factors (25). Patient related factors show smoking as the strongest independent risk factor to developing recurrence (26). The mechanism of this is unclear, and perplexing considering it has a protective effect in patients with ulcerative colitis. Dietary factors, stress, non-steroidal anti-inflammatory drugs (NSAID) usage and infections have also been implicated as environmental triggers but the specific mechanisms are not fully understood (9, 15).

There is increasing evidence that the microenvironment, or microbiota, is implicated in the development of inflammation. Commensal bacteria act in symbiosis and have several important functions within the gut, including nutrient absorption and intestinal maturation. The mechanisms responsible for ensuring tolerance to these commensal bacteria are complex and poorly understood. Therefore it is widely accepted that Crohn’s disease results from an inappropriate response of a defective mucosal immune system to commensal bacteria and other luminal antigens (27).

1.1.4. The behaviour of Crohn’s Disease

The chronic inflammatory process seen in Crohn’s disease is relapsing and remitting in nature.

Flare-ups are usually followed by a period of clinical remission in the early phase of the disease process, which are usually associated with evidence of bowel inflammation demonstrated endoscopically, radiologically or intra-operatively. A pre-clinical inflammatory phase often pre-dates the onset of symptoms and often continues after a relapse despite being in clinical

36

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery remission. Disease therefore often progresses to form either a fibro-stenotic stricture or penetrating lesions of the bowel (28), or in combination (1). The initial presentation, with advanced disease, can, and often does, still occur (29).

Strictures in Crohn’s disease usually develop during the course of the disease (30, 31) but in 5-

27% of cases symptomatic strictures may be the presenting feature of the condition (31-34).

Strictures develop as part of the disease process i.e. de novo or due to surgical healing i.e. anastomotic. Either may be predominantly inflammatory or fibrotic; the majority (98.8%) are found in the colon, ileo-colonic region, and ileum after 10 years of disease (35). The pathogenesis of fibrosis in Crohn’s disease is not well understood. Chronic inflammation leads to thickening of the mucosa and narrowing of the gut lumen (35-37). Thereafter, disruption in the normal extracellular matrix and irregular activity of fibroblasts contribute to an imbalance of collagen deposition. Anastomotic strictures, on the other hand, develop through a combination of local and technical factors, such as bacterial stasis from post-operative narrowing of the lumen, high intraluminal pressures, or vascular compromise resulting in tissue ischemia, leaking, or infection which drive healing by tissue fibrosis (38, 39). Smoking was associated with an increased rate of progression from inflammatory to structuring disease (40).

In separate studies, mutations in the NOD2 gene were associated with small bowel fibro- stenosing Crohn’s disease (41) whereas the NOD2/CARD15 genotype was an independent risk factor for early surgical intervention due to strictures (42). There remain several unanswered questions about fibrosis namely why some patients are protected, how early diagnosis can be detected and which factors modulate the pathways to arrest or reverse the process. Animal models of intestinal fibrosis may provide some answers and further our understanding of pathogenesis of strictures (37, 43).

37

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

Penetrating disease is represented by the presence of fistulae, sinus tract formation and perforation, the latter often presenting with abscess formation. Penetrating disease develops rapidly over time from a non-stricturing non-penetrating disease state (31). Recurrent flares and smoking are causative factors (44). The pathophysiology originates from the formation of deep penetrating ulcers which traverse all layers of the bowel wall. The eventual perforation forms adjacent abscesses or collection, which progress to form fistulae. In one epidemiological study, 88 fistulising episodes were present across a cohort of 59 patients. Of them, 54% were peri-anal, 24% were entero-enteric, 9% were rectovaginal, 6% were enterocutaneous, 3% were entero-vesical (45).

1.1.5. Classification systems: a historical perspective

Various models exist that attempt to account for the heterogeneity observed in disease phenotype and presentation with Crohn’s disease. One of the earliest classification systems was based on the anatomical location of the disease: Farmer and colleagues demonstrated 4 distinct anatomical locations based on 615 consecutive patients with Crohn’s disease: ileocolonic, small intestine, colon and anorectal (46). Greenstein and colleagues introduced the concept of distinct disease phenotypes based on indications for surgery: perforating and non-perforating. They also observed that the indications for repeat surgery were similar to the disease entity indicating the primary surgical procedure (47). This was refined with the Rome classification which added further dimensions including operative history and the extent of disease, however it could result in 756 subgroups of Crohn’s disease and was therefore never widely adopted. The Vienna classification, in 1998, aimed to develop a simplified classification based on objective variables. As well as disease phenotype and location, the Working Party for the World Congress of Gastroenterology established the age of disease onset as a significant

38

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery variable as well (48). The discovery of a series of genetic and serological markers associated with disease susceptibility warranted revisions in the Vienna classification.

1.1.6. The Montreal Classification

A Working Party of the 2005 Montreal World Congress of Gastroenterology devised the

Montreal classification, which placed a further subdivision to the age of disease onset, and as well a consideration for isolated upper gastrointestinal Crohn’s disease and perianal disease

(49). The Montreal Classification recognises 3 phenotypes in Crohn’s disease: non- stricturing/non-penetrating, stricturing and penetrating disease (49). The Montreal classification has been adopted and is widely used in clinical practice, however it is not without its controversies (50). Firstly, its view of surgery as a marker of severity and an adverse outcome measure, as opposed to an intervention, may be secondary to a lack of surgical expertise within the working party. Secondly, it does not distinguish between fistulating and perforating disease in its penetrating (B3) subset. There are continuing efforts to develop a model that can predict disease progression as well as classify disease phenotype into homogenous groups, with multidisciplinary input including that of IBD colorectal surgeons.

1.2. Management strategies

1.2.1. An overview

Variability in the aetiology, epidemiology, anatomical location and phenotypic behaviour of

Crohn’s disease has, unsurprisingly, led to difficulties in deciding on the optimal management strategy for patients. A number of different therapeutic strategies therefore exist that aim to induce mucosal healing. Medical therapies can be divided into agents that induce remission and agents that maintain remission. Agents of induction include antibiotics, corticosteroids,

39

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery anti-tumour necrosis alpha (anti-TNF), and exclusive enteral feeding. Agents that maintain remission include thiopurines, 6-metacaptopurines, and anti-TNF.

The second European evidence-based consensus on the diagnosis and management of Crohn's disease (51), suggests the medical management of localised ileocaecal Crohn’s disease should be based on disease severity as defined by the Crohn’s Disease Activity Index (CDAI): mild disease managed with mild steroid based regimens, moderate disease with stronger steroid regimens with or without maintenance immunomodulatory therapy and severe disease with biological therapy. Surgical resection in the setting is considered a last resort in the failure of medical therapy. Evidence of its efficacy compared to anti-TNF therapy is limited. The LIR!C trial (Laparoscopic Ileocaecal Resection versus Infliximab Therapy) is a multi-centre randomised trial which aims to address the supremacy of these therapies with quality of life as the primary outcome measure in these patients (52). European consensus considers surgical management as a primary indication in the presence of obstruction or abscess formation

(following antibiotic therapy and drainage) (51).

1.2.2. Medical strategies

1.2.2.1. Steroid- based therapies

Synthetic corticosteroids are primarily indicated for the induction of remission in patients with active Crohn’s disease. They work by inhibiting inflammation through mechanisms that reduce intestinal permeability (53). Side effects of long-term corticosteroid therapy make them undesirable as a maintenance regiment. There has been a significant reduction in the use of long term corticosteroid therapy (greater than 1-year usage) within the first 5 years of diagnosis

(54).

40

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

Budesonide is a milder corticosteroid-based therapy which has demonstrated efficacy as an inducer of remission during an acute flare. It’s efficacy as a maintainer of remission, however, is limited. Following surgery, it did not demonstrate any significant benefit against placebo in the development of endoscopic or clinical recurrence (55).

1.2.2.2. 5-ASA-based therapies

The efficacy of 5-aminosalcyate acid has been unclear, with studies demonstrating heterogeneous results in the role of remission maintenance in Crohn’s. A meta-analysis by

Camma and colleagues demonstrated therapy with mesalamine significantly reduced the risk of symptomatic relapse in patients with Crohn’s ileitis (56). More recent studies however have shown no clear benefits compared with placebo (25, 57).

1.2.2.3. Thiopurine and 6-mercaptopurine therapies

Azathioprine is a prodrug which is metabolised in vivo to form 6-mercaptopurine, which goes on to form a number of thionucleotide metabolites with eventually are incorporated into DNA as a false base (58). Evidence for the early use of thiopurines to maintain remission has increased significantly over time (54). Thiopurine agents have been demonstrated to be superior in preventing post-operative recurrence, when compared with 5-ASA based regiments and placebo (59). They are currently indicated in moderate to severe Crohn’s disease.

1.2.2.4. Antibiotic therapies

There is evidence that antibiotic therapy has a favourable response in patients with Crohn’s colitis and following ileocaecal resection for Crohn’s ileitis (15). Metronidazole therapy has demonstrated efficacy in reducing recurrence rates following ileocaecal resection (60, 61).

Rutgeert demonstrated in his case control study that 52% of patients following curative ileal

41

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery resection in the 3-month metronidazole group demonstrated endoscopic recurrence after surgery compared with 75% in the placebo group (62). Therapeutic responses suggest that dominant bacterial stimuli are different in ileal and colonic Crohn’s disease. Side effects associated with metronidazole therapy, however, have raised concerns of compliance.

Combination therapy with thiopurines demonstrates a significantly lower endoscopic recurrence rate at 3 months and 1 year compared with metronidazole therapy alone (63).

1.2.2.5. Biological therapies

Anti TNF-α (Infliximab or Adalimumab) is a human-murine monoclonal antibody that bind with high affinity to tumour necrosis factor alpha (a cytokine that drives the process of systemic inflammation) inhibiting its functional activity. They were licenced for use 1999 in management of patient with severe and active Crohn’s disease, who have not responded to a full and adequate course of corticosteroids and/or immunomodulatory therapies (64, 65).

Evidence, though limited, does suggest biologics are highly effective in prevention of post- operative recurrence. Regueiro demonstrated significantly lower endoscopic, clinical and histological recurrence rates across 24 patients who were randomized and blinded to receiving either infliximab or placebo following surgery (66). Savarino demonstrated significantly lower

2 year endoscopic and clinical recurrence rates with adalimumab compared with azathioprine and mesalamine across 51 patients randomised into each therapy (67).

1.2.3. Endoscopic balloon dilatation

The first description of hydrostatic balloon dilatation in Crohn’s disease was provided in 1986.

R. A. Brower described a report of a successful hydrostatic balloon dilatation of a terminal ileal

Crohn’s stricture in a 24-year-old woman with refractory symptoms of bowel obstruction who refused surgery (68). Since this description the procedure has been widely adopted. The

42

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

European consensus statement for localised ileocolic disease in a post-surgical recurrence scenario, states that endoscopic balloon dilatation could be tried before moving to intestinal resection (51). The choice is determined by stricture characteristics: accessible, shorter and anastomotic strictures are intuitively considered for dilatation whereas endoscopically inaccessible, multiple and >5cm in length are suited to strictureplasty as a bowel preserving measure. Although balloon dilatation seems an attractive first therapy option because of ease of administration and low costs, a key limitation is access to some strictures either due to location identified by imaging before endoscopic attempts or during endoscopy because of stricture characteristics thereby exposing patient to risks of colonoscopy without any therapeutic benefit. Other risks relate to the dilatation procedure; short term risks include perforation and bleeding whereas long term risks are related to disease recurrence which may warrant further dilatation or surgery (69).

Several studies have reported on the outcomes of endoscopic balloon dilatation in Crohn’s disease strictures. Most of these studies are based on adult populations. There is much variability in the characteristics of the strictures, the reported techniques and outcome measures. It is therefore difficult to draw any meaningful conclusion on effectiveness of endoscopic balloon dilatation leaving uncertainty about whether it prevents or merely delays surgery in the long term, and whether certain stricture characteristics may be more suitable for the intervention. There are no randomised controlled trials on the efficacy of endoscopic balloon dilatation compared with surgery. These outcomes were collated in two systematic literature reviews. The first was published in 2007 which reported technical success, clinical success and a major complication rate of 86%, 58% and 2% respectively (69). The second was published in 2010 (70) which reported the technical success, surgical recurrence and major complication rates as 90%, 27.6% and 3% respectively. The lack of any pooled analyses

43

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery regarding events rates, and non-conformity with the Preferred Reporting Items for Systematic

Reviews and Meta-analyses (PRISMA) guidelines are discernible weaknesses of both reviews.

Moreover, one of the reviews included outcomes of double balloon enteroscopy as well as strictureplasty (70). The optimal management pathway of strictures in Crohn’s disease therefore relies on multidisciplinary team discussions on a case by case basis. Further analysis of current literature is needed to guide treatment decisions and inform the design of future studies.

1.2.4. Surgical strategies

Surgical strategy in the setting of small bowel Crohn’s disease can be summed up by five key points (71), summarised as follows:

➢ Crohn’s disease is a pan-intestinal disease, with intermittent activity and the potential of

focal exacerbations throughout the patient’s life,

➢ it is impossible to cure Crohn’s disease by excision. The surgeon is required only to treat

the consequences and improve quality of life,

➢ surgical treatment should be carried out safely, minimising morbidity, but preservation

should be the primary focus of surgery to reduce the impact of further surgery if required.

➢ all diseased bowel need not be excised, only that part leading to symptoms,

➢ if only stenotic complications are being treated, perhaps the stenosis can be simply widened

by strictureplasty or dilatation.

The probability of undergoing an ileocaecal resection is 20 - 40% in the first year following diagnosis. At 10 years the probability of surgical resection increases to 70%, and can be as high as 90% at 15 years (72, 73). Considering the care of patients with Crohn’s disease is

44

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery predominantly led by gastroenterologists, an understanding of surgery in terms of symptom relief and how this balances against the risk of the procedure is necessary to ensure the best therapy is offered at the optimal time (51). The multidisciplinary team meeting provides a useful forum to discuss patient management. It is, however, still difficult to decide on the

‘optimal time’ for surgery. Delaying the operation with on-going medical therapies could lead to complex disease and technically difficult surgery. This can increase the risk of anastomotic leakage, stoma formation, and overall morbidity.

Improved medical therapy is thought to have reduced the time to first surgery significantly. In a population-based cohort study, the proportion of patients having intestinal surgery at 5 years following diagnosis was significantly lower at 25% (1990 - 2003), compared with 59% at an earlier period (1986 – 1991) (54). The added benefit of improved quality of life and clinical remission observed after surgery, suggests a growing argument for minimally invasive surgery at an earlier stage in the natural history of disease progression, as opposed to delaying until disease is at an advanced stage, which may be associated with a poor surgical outcome (73,

74). Preservation of bowel length and laparoscopic techniques are important strategic considerations when considering surgery, and they become challenging targets to meet in the setting of advanced disease (75). Indications for elective surgery include fistulating disease

(with or without abscess formation), bowel obstruction, a failure of medical therapy, malignancy and growth retardation. Emergency indications include perforation or haemorrhage. Three approaches, therefore, have become commonplace to ensure the above key factors are covered: strictureplasty, resection and bypass (76, 77).

45

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

1.2.4.1. Resection

Advanced disease, defined by the presence of fistulae or obstructing fibro-stenotic disease, often require surgery in the form of a small bowel, ileocaecal or ileocolonic resection (the latter in the setting of post-operative recurrence). A standard approach to laparoscopic surgery includes a 3-trocar approach, where the surgeon stands between the legs of the patient. A surgeon-assistant will stand on the left of the patient and will manipulate the caecum with a retractor. The right colon will be mobilized and the distal ileum and caecum will be exteriorized by a 4 to 5 cm vertical incision through the umbilicus. Vascular ligation, bowel division, and a sutured end-to-side anastomosis can be performed extra corporeally or intra corporeally (figure 1.1).

Figure 1.1 Diagram representing basic concept of an ileocolonic resection followed by an anastomosis between A and B (tracing is reproduced from the Johns Hopkins Gastroenterology and Hepatology Resource Centre, Johns Hopkins University)

Laparoscopy is considered the standard approach to performing an ileocolic resection where appropriate expertise is available. Its role in the setting of more complex disease is limited

(51). A randomised trial comparing post-operative outcomes between laparoscopic assisted

46

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery versus open ileocolic resection in patients with Crohn’s disease demonstrated significantly lower morbidity, hospital stay and costs with laparoscopic surgery (78). A meta-analysis also demonstrated lower morbidity and a lower incidence of recurrence with laparoscopic surgery compared with open surgery. Benefits of laparoscopy also included a more rapid recovery of bowel function and a shorter duration of hospitalisation. These benefits were set against the longer operative time to perform a laparoscopic procedure (79).

1.2.4.2. Strictureplasty

Strictureplasty has gained popularity over resection in view of its ability to spare the bowel, preventing short bowel syndrome. It is primarily indicated in patients with multiple fibrostenotic small bowel strictures over a large length of bowel, previous significant small bowel resection, and in patients with small bowel syndrome. It is contraindicated in patients who are malnourished (albumin < 2.0 g/dl), who have perforated bowel, the presence of multiple strictures over small lengths, or if a stricture is present a short distance from the area of resection, or if there is bleeding from a planned strictureplasty site (80). The first descriptions of strictureplasty were seen in patients with tuberculosis (81). Since then numerous studies have provided descriptions of the use of strictureplasty in patients with

Crohn’s disease which have demonstrated their safety and efficacy (82, 83).

There are three common techniques recognised in the setting of strictureplasty: Heineke–

Mikulicz, Finney, and the side-to-side isoperistaltic strictureplasty. The Heineke–Mikulicz strictureplasty is performed with longitudinal enterotomy over the stricture on the antimesenteric border of the bowel and is extended 1 to 2 cm onto either side of normal bowel.

The enterotomy is then closed transversely with interrupted, seromuscular absorbable sutures.

The closure may be performed in one or two layers (figure 1.2) (80).

47

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

Figure 1. 2 Diagram representing basic concept of Heineke–Mikulicz strictureplasty (tracing reproduced from Gut 2013 BMJ Publishing Group Ltd & British Society of Gastroenterology

- http://gut.bmj.com/content/62/7/1072)

The Finney strictureplasty is performed for intermediate length strictures. First, a stay suture is placed at the midpoint of the stricture. The enterotomy is made through the stricture as in the

Heineke–Mikulicz, again extending 1 to 2 cm onto normal bowel, then the strictured segment is folded onto itself to form a U-shape. Another stay suture is placed on the normal side of the bowel to hold the U-shape in place. The posterior edges are then sutured in a continuous fashion using an absorbable suture. The anterior edges are then closed with an absorbable suture in an interrupted fashion (figure 1.3) (80).

48

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

Figure 1. 3 Diagram representing basic concept of Finney strictureplasty (tracing reproduced from Gut 2013 BMJ Publishing Group Ltd & British Society of Gastroenterology - http://gut.bmj.com/content/62/7/1072)

The side-to-side isoperistaltic strictureplasty, as described by Michelassi (84), is performed for longer length strictures, usually greater than 20 to 25 cm. In this procedure, the length of strictured bowel is lifted and the mesentery for this region is divided at the midpoint. Next, the diseased bowel is divided between atraumatic bowel clamps at the midpoint of the stricture.

The proximal end of the cut bowel is brought over the distal end in a side-to-side fashion. The two loops of bowel are then approximated with a single layer of interrupted non-absorbable sutures. Next, the enterotomy is made longitudinally for the length of the stricture. The ends of the bowel are spatulated to avoid blind ends. Next, an inner layer of running, full-thickness absorbable sutures are placed and continued anteriorly as a running Connell stitch. This anterior layer is then followed by a layer of interrupted, non-absorbable sero-muscular sutures (figure

1.4) (80).

49

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

Figure 1. 4 Diagram representing basic concept of side-to-side isoperistaltic strictureplasty

(tracing reproduced from Gut 2013 BMJ Publishing Group Ltd & British Society of

Gastroenterology - http://gut.bmj.com/content/62/7/1072).

1.2.4.3. Bypass

There are two different techniques recognised in the setting of advanced Crohn’s disease with the presence of an inflammatory mass, plegmon or abscess. An exclusion bypass can be performed when the bowel proximal to the macroscopically disease segment is transected and the cut end of bowel is anastomosed to the transverse colon in an end-to-side fashion. A mucus fistula may or may not be constructed to the distally transected small bowel. A simple bypass involves the formation of a side-to-side ileocolonic anastomosis within a similar setting (77).

50

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

1.3. Surgical outcomes

1.3.1. Introduction

NHS England have established initiatives to create greater transparency and more choice for patients and commissioners. Colorectal surgery, among many other specialties, has been highlighted as an area of analysis, with the publication of institutional and individual consultant operative outcomes (85). The Association of Coloproctology of Great Britain and Ireland published crude 90-day mortality data following colorectal cancer resections for individual consultant colorectal surgeons and institutions.

There is a need to introduce this into the IBD setting. IBD has been identified as a risk factor for increased 30-day re-operation rates nationally (86). Advanced disease at the time of operation can increase the incidence of intra-abdominal septic complications secondary to anastomotic dehiscence or enterotomies, post-operative ileus or entero-cutaneous fistula formation. A lack of clarity however exists as to what outcomes measure would be suitable in the setting of ileocolonic and/or small bowel Crohn’s resections.

1.3.2. Short-term outcome measures for small bowel and ileocolonic resection

The primary concern following ileocolonic or intestinal resection with a primary anastomosis for Crohn’s disease is the development of intra-abdominal septic complication. The definition of this is variable throughout the literature, but usually consists of an anastomotic leak, an intra- abdominal collection or the presence of enterocutaneous fistulae. Numerous observational studies (publication year: 2000 to 2012) have reported on a variety of outcome measures based on the presence or absence of intra-abdominal septic complication following a ileocolonic or small bowel resection with primary anastomosis for Crohn’s disease. Variability in the reporting of outcome measures makes it difficult to ascertain the true incidence of this as a

51

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery complication. The presence of an anastomotic leak or dehiscence is the most commonly reported outcome measures in recent literature, with rates ranging from 0% - 15% (87-101).

The presence of an intra-abdominal abscess or collection ranges from 0 – 20% across recent literature (91, 93, 96, 98, 99, 102). The description of fistulae formation following surgery is present in three studies and range from 0% – 2%, (87, 98, 100) all of which were enterocutaneous fistulae.

A more clinically meaningful outcome measure reported across most recent literature relates to the intervention following the development of intra-abdominal septic complications. Re- operation has been reported across 16 studies and ranges from 0% - 10.3% (87-89, 91-103).

The use of radiologically guided drainage of intra-abdominal septic complication following surgery has been reported across five studies and ranges from 0.6% - 10% (87, 88, 96, 102,

104). Conservative management of intra-abdominal septic complication has been reported across six studies and ranges from in 0.5% - 8.4% (87, 92-94, 97, 98).

1.3.3. Predictors of intra-abdominal septic complication

Generic risk factors contributing to post-operative intra-abdominal septic complications for colorectal surgery as a whole can be classified into two entities: patient-specific and intraoperative risk factors (105). Patient-specific risk factors include malnutrition, obesity, smoking, previous abdominal or pelvic irradiation, ASA scores greater than 2 and steroid use.

Intraoperative factors include low anastomosis, prolonged operation duration, bowel obstruction, anastomotic vascularity, peri-operative blood transfusion and intraoperative septic conditions (105-107).

52

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

In the setting of inflammatory bowel disease, the peri-operative well-being of a patient may be crucial in determining 30-day post-operative morbidity. Similar studies have demonstrated the presence of active disease and intra-abdominal sepsis does correlate with post-operative morbidity. Hulten was one of the first to describe a higher proportion of early septic intra- abdominal complications following surgery for patients with small bowel or ileocolonic

Crohn’s disease in the presence of pre-operative intra-abdominal sepsis compared to those without usually attributable to the presence of abscesses, perforation and or fistulation (49% and 12% respectively). Hulten also argued the likelihood of these pre-operative features to be secondary to advanced disease (108). Recent studies have also reported similar findings of pre-operative intra-abdominal sepsis and its association with post-operative septic complications (91, 98, 109-115). Penetrating disease, particularly in the presence of an intraoperative abscesses (110, 111, 116, 117) or fistulae (111, 118) do increase the incidence of post-operative IASCs. Peri-operative percutaneous drainage has been demonstrated to improve post-operative outcome (119).

Some institutions advocate corticosteroid therapy to induce remission in the setting of active

Crohn’s. Numerous reports, however, have demonstrated a significant association with this and the development of post-operative IASCs (87, 110, 111, 117, 120). These findings are consistent with another prospective study which demonstrated a significant increase in intra- abdominal septic complications in patients receiving pre-operative corticosteroids, compared to those without corticosteroids at 11.8% and 2.4% respectively (OR 8.7; 95% CI 1.2-45.1; p

= 0.015) (9). The use of newer agents, such as anti-TNF alpha biological therapies, and their association with post-operative IASC were somewhat controversial, until a recent meta- analysis that demonstrated associations with non-local septic complications [OR 2.07 95% CI

1.30-3.30], as opposed to a specific association with post-operative IASCs (121).

53

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

The role of pre-operative nutritional status has been implicated in the development of IASCs after surgery. A six-fold increased risk in the development of IASCs was demonstrated by

Alves following a multivariate analysis for patients who had a greater than 10% weight loss at the time of surgery (87). The role of serum albumin has often been considered a marker of nutrition. Low serum albumen levels in the setting of Crohn’s disease may also be a marker of disease activity. Serum albumen levels of less than 30 g/L have been demonstrated as an independent risk factor for the development of IASCs (111, 122).

Identifying the risks associated with post-operative intra-abdominal septic complications is necessary if earlier ileocolonic resection is being considered. As well as employing laparoscopic techniques, consideration of a two-stage procedure, with initial resection and stoma formation, followed by a later stoma reversal, may be appropriate in patients deemed high risk for developing post-operative intra-abdominal septic complications.

1.3.4. Long term outcome following ileocaecal resection – the recurrent concern

Surgery is not curative, and a large majority of cases will develop recurrent disease. Early recrudescent lesions can be detected endoscopically or radiologically within the pre-clinical phase. The onset of symptoms from active disease defines the onset of clinical recurrence.

Surgical recurrence defines the need for a re-resectional surgery.

1.3.4.1. Endoscopic recurrence

Much of the earlier work relating to the natural history of recurrent disease was performed by

Rutgeert and colleagues. A crude endoscopic recurrence rate of 72% was demonstrated across

54

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

114 patients at one year after ‘curative’ resectional surgery for ileocaecal Crohn’s disease.

Furthermore, lesions appeared to develop within the neo-terminal ileum at the site of the anastomosis in 88% of those who developed recrudescent lesions (123). A prospective longitudinal study on a cohort of 89 patients who had ileal resection, done by the same author, determined that the severity of recrudescent disease at one year was able to predict the onset of symptoms (124).

1.3.4.2. Clinical recurrence

Clinical recurrence is defined by the presence of recurrent Crohn’s symptoms, with radiological or endoscopic evidence of active disease. Post-operative clinical recurrence rates ranged from

30-55% at 5 years (74, 125-127), 49-76% at 10 years (74, 126, 128) and 59-79% at 15 years

(74, 128) after surgery. There is a difficulty in distinguishing symptomatic patients with recurrent disease from those without recurrence. Several objective markers exist for identifying and monitoring recurrent disease and the efficacy of post-operative prophylactic therapies. The CDAI may be useful but is time consuming and complicated and requires the patient to fill a diary over seven days detailing symptoms. Furthermore, the CDAI has not been validated in the post-operative setting and has limited predictive power for patients with fistulating and stenotic disease (129). C-reactive protein levels are usually increased in patients with active disease. Its values are needed to calculate the CDAI, but it can also prognosticate recurrence in patients (130). Another non-invasive biomarker is faecal Calprotectin which has demonstrated better correlation with endoscopic assessments in the prognostication of recurrent disease (131).

55

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

1.3.4.3. Surgical recurrence

Numerous studies have reported on re-resectional surgical rates following an index ileocolonic or ileal resection. Most studies, however, have been performed in the pre-biological era. Crude re-resectional surgical rates at 5, 10, 15 and 20 years range from 11-27%, 22-44%, 38-53% and

52–55% respectively (74).

A recent paper has reported the incidence of re-sectional surgery within the biological era. The study provided a retrospective review of a cohort of 105 patients with long-term follow-up.

Re-resectional surgical rates at 5 and 10 years after index surgery was reported as 9.5% (95%

CI, 6.2%–14.5%) and 18.6% (95 % CI, 13.2%–25.9%) respectively. Eighteen patients (56.3%) had recurrence at the anastomotic site, 6 patients (18.8%) in the colon, 7 (21.9%) in the small bowel, and 1 (3.1%) in the rectum. Twenty-eight patients (13.8%) from this group were exposed biologics post-operatively. Evidence of the efficacy of biological therapy in reducing surgical recurrence rates is limited however the lower crude incidences at the 5 and 10-year time-points may be suggestive of improved post-operative strategies to maintain remission

(132).

1.3.4.4. Predictors of recurrence

Numerous studies have reported on identifying risk factors for the development of recurrence disease following surgery. These have been summarised extensively in two systematic reviews. De Cruz and colleagues analysed the natural history of post-operative luminal recurrence in Crohn’s Disease and identified patient, disease and operative factors contributing to recurrent disease. Smoking is the strongest independent risk factor for the development of recurrent disease (26). A history of prior resection, a penetrating disease phenotype and short

56

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery disease duration have all been implicated in the development of earlier recurrent disease.

Buisson and colleagues determined the same risk factors in addition to the presence of perianal

Crohn’s disease (23, 25).

There have been no demonstrable relationships identified for any operative technique that may predispose to the development of recurrent disease. Two randomized controlled trials have demonstrated no significant relationship between recurrent disease between laparoscopic and open surgery (78, 133). Laparoscopy has therefore become the gold standard for suitable patients. A meta-analysis by Simillis and colleagues demonstrated that the configuration of the anastomosis (end-to-end versus side-to-side) did not influence recurrence rates (134). One randomised trial, although under-powered, demonstrated that there was no significant relationship between the involvement of disease at the resection margin, and the development of recurrence (135)

There has been much work on the role of post-operative medical prophylaxis to prevent recurrent disease following surgery. Considering compliance is maintained post-operative prophylactic medical therapy does reduce recurrence rates (136). The use of mesalamine has been demonstrated to reduce the incidence of clinical recurrence compared to placebo, although inferior in effect when compared to azathioprine or 6-MP therapies (137). Its current use is therefore primarily aimed at low risk groups. Thiopurine therapies have been demonstrated to have a more potent effect at maintaining remission and are commonly aimed at moderate risk groups (138).

There is strong evidence to support the role of biological therapies in the maintenance of post- operative remission (66, 139) however with a standard average dosage of anti-TNF therapy of

57

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

£12,584 over one year for the average patient, treating all Crohn’s patients with anti-TNF in the post-operative phase will have major cost implications (140). Despite this that National

Institute of Clinical Excellence has recognised the importance of biologics in maintaining remission for patients with Crohn’s disease (64). Stratifying and trying to predict those patients who are likely to have a more difficult onward course, and assessing and monitoring with faecal

Calprotectin, other objective markers of inflammation and ileocolonoscopy at given time points would create a more cost-effective algorithm of care in the post-operative phase.

1.4. Variability and standards of care

1.4.1. Variability in decision-making

The vast degree of variation observed across outcomes has led to differences in opinion between gastroenterologists, colorectal surgeons and patients in treatment algorithms in the context of ileal Crohn’s disease surgery (141, 142). The reasons for such variations in opinion are unknown, but may arise from anxieties of post-operative complications, including anastomotic leak, recurrent disease, future short bowel and a need for stoma formation.

Difficulties in predicting surgical outcome in patients with Crohn’s disease can lead to differences in opinion between physicians in planning definitive care. Differences in opinion between surgeons and gastroenterologists when planning definitive management for patients with inflammatory bowel disease do exist. A study from Byrne and colleagues in Sydney,

Australia demonstrated gastroenterologists tend to avoid recommending surgery on a preference measure for specific Crohn’s related scenarios when compared with both patients and surgeons (141). Paradoxically, studies in the UK have shown an opposite trend. Shariff et al. demonstrated that in a specific scenario of terminal ileitis discovered at surgery during a

58

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery case of suspected appendicitis, surgeons were more conservative compared with gastroenterologists, avoiding bowel resection (142). The lack of clear guided standards in the setting of complex IBD may result in clinical decisions being made from an intuitive mix of probability judgements, experience and sometimes bias, combined with implicit judgments on the importance of possible outcomes. While attempts are made to elicit patient preferences in relation to the latter, the success of such patient centred algorithms have yet to be tested. The reason for these differences in opinion between clinicians is yet to be determined.

1.4.2. The patient perspective of surgery

A questionnaire survey of 1067 patients with IBD has demonstrated 81% of respondents reported it as ‘very important’ to be actively involved in the decision-making process regarding therapy (143). The perception of therapy therefore has an important role in management for patients with Crohn’s disease. Although numerous studies have explored the perceptions and preferences of medical therapies (144-149), few have explored the perceptions of surgical intervention.

Divergent preferences exist between gastroenterologists and patient regarding open or laparoscopic ileocolonic resection. In a scenario based questionnaire survey across 123 patients with Crohn’s disease, 127 colorectal surgeons and 272 gastroenterologists, 76% of gastroenterologist were ‘willing to gamble’ to avoid surgery. This was significantly higher compared with the preferences of patients (39%) and surgeons (37%) when faced with the same scenario. Further subgroup analysis demonstrated the willingness to gamble was lower in patients who had undergone surgery compared to those who had not (29% and 47% respectively), although despite this large difference being present, no significant difference was

59

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery demonstrated. The need for a much larger sample size was discussed to explore whether significant differences in perceptions truly exist (141).

One questionnaire study spanning 1220 patients explored the perceptions of surgery. Of those who had intestinal surgery relating to their Crohn’s disease, the impact surgery had to their health was explored. Over half of respondents report better short term (56.1%) and long term

(55.8%) outcome. Nearly half also recognised a fear of surgery (52.6%). Surgery also appeared to have a psychological impact on patients with 33.2% reporting low self-esteem,

22.8% reporting high self-esteem, and 25.1% feeling mutilated. The role of a stoma was also explored across the 36% of respondents who had a stoma formation. Primary concerns included the interference with clothing (66.2%), impact on intimacy (53%), physical appearance (50.5%), and physical appearance (51.8%). Stoma related complications were the lowest ranking concern (39.2%) (150).

1.4.3. The quality of care within an IBD service provision

The quality of care, as established by Lord Darzi’s ‘High Quality Care For All – NHS Next

Stage Review Final Report’, is described as multi-dimensional, and consists of a need for patient care to be safe, that care is clinically effective and care provides a positive experience for patients (151, 152). The NHS outcomes framework provides a means to measure quality across health and care through key performance indicators. The Donabedian model provides an evaluation of quality and has been widely adopted throughout healthcare. The model provides three related concepts. Firstly, the structure of healthcare, as defined by the physical and organisational aspects of the care setting. Secondly, the processes required to improve patient health and outcomes. Lastly, the outcomes of healthcare (153).

60

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

In recent years there have been significant steps made towards improving the quality of care offered to patients with IBD in the United Kingdom. IBD surgery is rapidly being accepted as a separate surgical specialty, but standards and acceptable outcomes are yet to be determined.

The quality of surgical care received by this patient group requires further attention to ensure that there are appropriate and equitable standards being accessed by patients throughout the

United Kingdom (154).

In 2009 the IBD Standards Group published a set of service standards for the healthcare of people with IBD (154). The aim of these standards were to ensure those who have inflammatory bowel disease receive healthcare that is safe, effective and of consistently high quality. Standard A7 of this document describe elements for patients’ who undergo IBD surgery:

➢ IBD surgery should be undertaken by recognised colorectal surgeons who are core

members of the IBD Team, or their supervised trainees, in a unit where the operations are

done regularly.

➢ Expert pathological assessment before surgery is important. This may involve referral of

cases to a nationally recognised expert in the diagnosis, and differential diagnosis, of

chronic inflammatory bowel disease.

Despite these standards there is still variability and heterogeneity nationally in surgical provisions delivered in IBD care (154). IBD service provisions and care are being analysed as part of a general drive to reduce variation in standards of healthcare and to improve the quality of services that patients receive (154-156).

61

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

1.4.4. The centralisation of care

The relationship between surgical volume and clinical outcomes in the setting of cancer has been analysed extensively. The first study to demonstrate this effect was by Luft et al (1997) who analysed mortality rates across 1498 hospitals and 12 surgical procedures and found lower mortality with higher volume centres (157). Further evidence to support lower mortality outcomes with increasing operative volume has been demonstrated in subsequent studies (157-

163), particularly in the setting of high risk, complex or rare disease (164). Morbidity analysis has also demonstrated lower complication rates in high volume centres across prostatic, pancreatic and oesophageal surgery (165-168). The cost-effectiveness of centralisation is further highlighted by lower inpatient length of hospital stay for patients operated in high volume centres (167) (167, 169, 170)

Determining the cut-off threshold critical volume for high quality include both the surgeon and the institution, and variations in set cut-off thresholds exist and depend of the operation being performed. There is little evidence to demonstrate appropriate methods to calculate appropriate thresholds (171). Current guidelines for colorectal cancer recommend a minimal volume of 20 procedures per surgeon and 50 per centre performed annually (172, 173).

The predominant criticisms surrounding studies relating to the volume-outcome effect relate to the use of large dataset collection. Limitations exist in determining the severity of disease and how disease severity impacts on outcome. Furthermore, clinical engagement with coding is largely administrator lead, questioning the accuracy of data entered (86, 174, 175).

62

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

1.4.5 Standardising multidisciplinary team driven care

The concept of MDT-driven care has been widely implemented for the clinical decision- making and management of complex diseases. The basic premise of MDT-driven care is to involve all key professional groups in the consideration of complex patients and/or diagnostic dilemmas to create a clear care plan. It is a forum where clinical cases can be discussed among a variety of healthcare professionals and care recommendations are made.

In the United Kingdom, the Calman-Hine report was introduced in 1995 in the cancer setting to ensure a change from a generalist model of care to a specialist model (176). Although there has been some criticism in the implementation of this change (177, 178), the presence of specialist care has demonstrated an improvement in survival across various cancer specialties

(176, 179-183). Efforts have been made to standardise the organisational structure and design of MDT-driven care to further improve this effect. Recognition of core and extended members of this meeting has been suggested in the cancer setting (184).

MDT-driven care is now being introduced into IBD centres (185). The UK National IBD Audit demonstrated that 75% of participating institutions undertake a weekly multidisciplinary meeting (MDM) for IBD patients (155, 156). European centres have also demonstrated variability in MDT driven care (186). There is, however, little evidence of its efficacy in this context and currently there is no guidance on how this intervention may be standardised and used effectively (154-156). Variability in the workings of the cancer MDT have been demonstrated and protocols for the structure of this meeting are being designed and implemented. Providing a standardised framework for the IBD MDM may enhance its capacity to establish effective quality improvement.

63

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

In 2013 the IBD Standards were established in the United Kingdom, with the primary aim to ensure that IBD patients receive consistent, high quality care. Furthermore, European consensus guidelines have recognised IBD multidisciplinary team driven care in the management of complex IBD cases (51, 154). There is little evidence to support the purpose and construct of MDT driven care, and therefore difficulty in meeting standards have been noted. The UK national report for the audit of inflammatory bowel disease service provision demonstrated up to 73% of institution did not meet standards for the IBD team. Providing consensus-derived standards for the IBD MDT and MDMs can guide managers, policy makers and departmental leads to adhere to standards. There is recognition that IBD MDMs have a pivotal role in the management of complex IBD cases. A formal footing is now required to ensure that they occur regularly and with appropriate structure and resource (187).

64

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

CHAPTER 2: METHODS AND MATERIALS

2.1. Materials

2.1.1. The Crohn’s surgical database

Institutional approval from the local board research and development committee was obtained.

Data were obtained by cross-referencing three prospectively maintained databases within our institution from over a six-year period (1st January 2005 to 31st December 2010). A total of

1589 ICD-10 (International Classification of Diseases, 10th revision (188)) established diagnoses relating to admissions for Crohn’s disease from the hospital coding department were obtained. A total of 988 positively diagnosed Crohn’s disease samples from our histopathology department was also obtained. The patient identification codes for each group were cross- referenced and duplicate cases were removed. A further cross-referencing with 7298 small and large bowel surgical procedures was performed within the same time-period, obtained from

OPCS-4.6 (Classification of Interventions and Procedures) codes (189) (table 2.1). A final figure of 236 ileocolonic Crohn’s surgical procedures was obtained (figure 2.1).

OPCS Code OPCS Procedure G693 ILEECTOMY AND ANASTOMOSIS OF ILEUM TO ILEUM G694 ILEECTOMY AND ANASTOMOSIS OF ILEUM TO COLON G698 OTHER SPECIFIED EXCISION OF ILEUM G699 UNSPECIFIED EXCISION OF ILEUM G731 REVISION OF ANASTOMOSIS OF ILEUM G734 RESECTION OF ILEOCOLIC ANASTOMOSIS G741 CREATION OF CONTINENT ILEOSTOMY G742 CREATION OF TEMPORARY ILEOSTOMY G743 CREATION OF DEFUNCTIONING ILEOSTOMY G748 OTHER SPECIFIED CREATION OF ARTIFICIAL OPENING INTO ILEUM H06 EXTENDED EXCISION OF RIGHT HEMI-COLON EXTENDED RIGHT HEMICOLECTOMY AND END TO END H061 ANASTOMOSIS EXTENDED RIGHT HEMICOLECTOMY AND ANASTOMOSIS OF H062 ILEUM TO COLON H063 EXTENDED RIGHT HEMICOLECTOMY AND ANASTOMOSIS NEC H064 EXTENDED RIGHT HEMICOLECTOMY AND ILEOSTOMY HFQ RIGHT HEMICOLECTOMY AND END TO END ANASTOMOSIS OF H071 ILEUM TO COLON INCLUDES: ILEOCAECAL RESECTION

65

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

RIGHT HEMICOLECTOMY AND SIDE TO SIDE ANASTOMOSIS OF H072 ILEUM TO TRANSVERSE COLON H073 RIGHT HEMICOLECTOMY AND ANASTOMOSIS NEC H074 RIGHT HEMICOLECTOMY AND ILEOSTOMY HFQ Table 2. 1 OPCS codes and procedures relating to the ileum and ileocolonic bowel

Figure 2. 1 Amalgamation of three

prospectively maintained databases (i) ICD-10

(International Classification of Diseases), (ii)

OPCS (Classification of Interventions and

Procedures) & (iii) Institutional

Histopathology Database.

2.1.2. Database screening

The Crohn’s surgical database was screened to identify patients who had undergone ileocolonic surgery (figure 2.2). Two hundred and thirty-six patients were selected from this database. The clinical case records were screened to ensure the following eligibility criteria were met:

Inclusion criteria:

i) All patients who underwent a resection for ileal Crohn’s disease.

Exclusion criteria

i) Previous segmental or subtotal colectomy.

ii) Histopathology not suggestive of Crohn’s disease.

66

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

iii) Pure non-resectional surgery (e.g. strictureplasty alone)

Number of patient records included in dataset: n = 163 [n = 175 Small bowel Number of patient (strictureplasty) resection: records required procedures] screening: n = 1 n = 358 Number of patient Non Crohn's histology: records excluded n = 21 from dataset: n = 73 Previous colectomy: n = 51

Figure 2. 2 Flow diagram demonstrating screening process and final number of patients (and procedures) included and excluded from the database.

2.1.3. Data extraction

Following the screening stage, all pre-operative and peri-operative variables were extracted from included cases. Pre-operative covariates were categorised as patient and disease related.

Peri-operative covariates were categorised into medications, blood parameters, patient and disease related, and operative factors. Patient-related variables extracted were gender, smoking status and the presence of a family history of IBD. Disease-related variables extracted included the age of disease onset, disease location, disease phenotype and the Crohn’s surgical history surgery as a result of Crohn’s disease. Peri-operative medications (received within four weeks of surgery) included steroid-based regimes (prednisolone reducing regimes start at 40mg once a day and reduced by 5mg a week), two week broad spectrum antibiotics course (co-amoxiclav

67

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

625mg three time a day and/or metronidazole 400mg three time a day), thiopurine (azathioprine

2-2.5mg/kg daily) / 6-mercaptopurine 1-1.5mg/kg daily (6-MP), 5-aminosalicylic acid (5-

ASA) (mesalamine 2.4 grams daily), anti-TNFα (tumour necrosis factor alpha) biologics

(Infliximab 5mg/kg infusion every 8 weeks or Adalimumab 40mg weekly-fortnightly injection), methotrexate 15mg weekly and nutritional supplementation with either enteral or parenteral (intravenous nutrition). Peri-operative blood parameters (obtained within four weeks of surgery) included indicators of anaemia (defined as a peri-operative haemoglobin less than 13 g/dL in males or less than 12 g/dL in females), haematocrit levels (defined as low when less than 41% in males or less than 35% in females), inflammatory parameters including the presence of leucocytosis (defined with a leucocytosis greater than 11 x109 /L), C-reactive protein (CRP) levels (considered raised if greater than 40mg/L) and serum thrombocyte count

(considered raised if greater than 450 x109 /L). Serum albumin (defined as low when less than

25 g/L) and serum corrected calcium (defined as low when less than 2.20 mmol/L) were also obtained. Peri-operative patient and disease related factors included the ASA (American

Society of Anaesthesiology) grade (<3 or ≥ 3), scheduling (elective or emergency), intra- operative findings of sepsis, access (laparoscopic or open), anastomosis (stapled or hand- sewn), histological margin (clear or involved).

2.1.4. Outcome measures

Post-operative complications were also extracted and included the presence of 30-day intra- abdominal septic complications (IASCs), defined by the presence of an anastomotic leak (with radiological or intra-operative evidence of a peri-anastomotic collection), intra-abdominal collection (with radiological or intra-operative evidence of an intra-abdominal collection) or enterocutaneous fistulae formation. Post-operative complications were also assessed with the

Clavien-Dindo classification system (table 2.2) (190).

68

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

Grades Definition Grade 1 Any deviation from the normal post-operative course without the need for pharmacological treatment or surgical, endoscopic and radiological interventions Allowed therapeutic regimens are: drugs as antiemetics, antipyretics, analgesics, diuretics and electrolytes and physiotherapy. This grade also includes wound infections opened at the bedside. Grade 2 Requiring pharmacological treatment with drugs other than such allowed for grade I complications. Blood transfusions and total parenteral nutrition are also included. Grade 3 Requiring surgical, endoscopic or radiological intervention - 3a Intervention not under general anaesthesia - 3b Intervention under general anaesthesia Grade 4 Life-threatening complication (including central nervous system complications) * requiring IC/ICU-management - 4a single organ dysfunction (including dialysis) - 4b multi-organ dysfunction Grade 5 Death of a patient Suffix "d" If the patient suffers from a complication at the time of discharge, the suffix "d" (for 'disability') is added to the respective grade of complication. This label indicates the need for a follow-up to fully evaluate the complication. *brain haemorrhage, ischemic stroke, subarachnoid bleeding, but excluding transient ischemic attacks (TIA); IC: Intermediate care; ICU: Intensive care unit Table 2. 2 The Clavien-Dindo classification

The long-term post-operative outcomes were also measured as time to the development of recurrence or time in remission until last known follow-up. The development of clinical recurrence was defined as the presence of recurrent Crohn’s symptoms, with radiological or endoscopic evidence of active disease, warranting an initiation or change in medical or nutritional therapy (table 2.3).

69

Improving the Quality of Care for patients with Ileal Crohn’s disease undergoing Surgery

Pre-operative patient-related Pre-operative disease - Pre-operative Peri-operative Operative Post-operative data related data medication patient/disease related outcome Gender: Disease phenotype: Corticosteroid. ASA grade: ≤ 2 or ≥ 3 Access: Short-term: ➢ male ➢ stricturing (B2) laparoscopy or ➢ 30-day IASC (anastomotic ➢ female ➢ penetrating (B3) open leak or intra-abdominal Smoking status: Concomitant perianal Antibiotic therapy. Scheduling: Anastomosis: collection or ECF formation) ➢ never smoker (no documented Crohn’s disease (‘p’ ➢ elective ➢ stapled ➢ 30-day morbidity (as per the evidence of smoking history or modifier): ➢ urgent ➢ hand-sewn Clavien-Dindo stated documented non-smoker ➢ present classification) throughout patient record) ➢ absent ➢ smoker (a documented smoking history at any time point in the pre-operative, peri-operative or post-operative period)

Family History of IBD (first or Concomitant upper GI 5-aminosalicylic acid Intraoperative findings: Anastomosis: Long-term: second degree relative): Crohn’s disease (L4 therapy (5-ASA) / ➢ primary ➢ Time in remission until last ➢ present modifier): sulphasalazine: ➢ abscess/fistulae resection documented follow-up or the ➢ absent ➢ present ➢ strictures ➢ stoma reversal development of clinical ➢ absent recurrence. Age of disease onset: Thiopurine therapy / 6- Stoma: ➢ <16 (A1) mercaptopurine (6-MP) ➢ present ➢ 16-40 (A2) ➢ absent ➢ >40 (A3) Re-resection for Biological therapy: Resection margin: recurrent disease: ➢ Infliximab ➢ clear ➢ yes ➢ Adalimumab ➢ involved ➢ no Disease location: Methotrexate ➢ ileal (L1) Nutrition: ➢ ileocolonic (L3) ➢ enteral ➢ parenteral Table 2. 3 Pre-operative, peri-operative and post-operative variables obtained during data extraction

70

Improving the Quality of Care for patients with Ileal Crohn’s disease undergoing Surgery

2.1.5. Data validation

Data extraction involved examining clinical records and coding data to allow statistical analysis. To ensure accuracy, each clinical record was validated by a second medically trained person. Any discrepancy was discussed among the two data entry personnel until agreement and consensus was achieved.

2.1.6. Statistical analyses

All statistical analyses were carried out with SPSS version 22.0 (SPSS, Chicago, Illinois,

USA). For continuous variables: the mean alongside standard deviation values, and the median alongside range values, were given. Data were assessed for being parametric or non-parametric visually and using the Shapiro-Wilks test. For parametric data, the mean and standard deviation (SD) values are reported. For non-parametric data, the median and interquartile range

(IQR) values are reported.

2.1.7. Missing data

Frequency and percentage values are provided for missing data. Data were assumed missing completely at random (MCAR) if present in < 5% across all variables within multivariate analyses. Missing data were assessed visually using missing value patterns in datasets with >

5% missing values to determine if data were missing at random (MAR) or missing not at random (MNAR) (191). Complete-case analyses were used in subsequent univariate and multivariate analyses incorporating a listwise deletion of cases.

71

Improving the Quality of Care for patients with Ileal Crohn’s disease undergoing Surgery

2.2. Qualitative methodology

2.2.1. Rationale

Qualitative research aims to study things in its natural setting, attempting to make sense of, or interpret, phenomena in terms of the meaning people bring to them. A holistic perspective is maintained that preserves the complexities of human behaviour (192, 193). It differs from quantitative research in that it does not seek to measure. It deals with textual data as opposed to numerical data. It seeks to analyse and evaluate complex systems and accounts for, as opposed to controlling for, bias. It is explorative and allows the researcher an in-depth analysis of respective problems. Inductive reasoning is utilised in the qualitative researcher, such that observations and theories are proposed at the end of the research process from observations of qualitative data, recognitions of patterns, and the subsequent development of theories or explanations. The process is in contrast with deductive reasoning, where observations are collected with a view to testing a hypothesis which has been generated from a theory. Inductive reasoning is more open and exploratory whereas deductive reasoning is narrow and specific.

Biomedical research has historically been quantitative and experimentally driven. Qualitative research has therefore received much criticism within the healthcare profession. Firstly, that qualitative research is mainly anecdote and subject to bias. Secondly, that it lacks reproducibility, and thirdly that is lacks generalisability. Strategies therefore do exist to ensure rigour in qualitative approaches (194).

Non-probability sampling is utilised to identify specific groups of people who possess characteristics of the relevant to study. It differs from probability sampling, in which a random sample is collected to ensure statistical representation. Non-probability sampling allows the

72

Improving the Quality of Care for patients with Ileal Crohn’s disease undergoing Surgery researcher to select a purposive sample to include a wide range of informants, as well as key informants. Meticulous documentation of observations, and the process of analysis is required to ensure the reliability. This can be achieved through audio recordings and a coding framework. Triangulation refers to the process of data collection that ensures validity to the findings. This occurs through the collection of data from a range of independent sources, by differing means (194).

2.2.2. Semi-structured interviews

Semi-structured interviews are conducted in Chapters 6, 7 and 8 of this thesis with emergent theme analysis based upon grounded theory methodology. This approach was chosen as it offers detailed information from individual participants and is well suited to explore complex and potentially controversial issues from which interventions can be generated and tested further (195, 196). To ensure credibility, transferability, dependability and confirmability appropriate to qualitative research, the following established measures were employed: the use of a standardised interview protocol and clearly defined coding framework, and training of the interviewer before interviews (194, 196-198).

The sample size is determined on the basis of the ‘saturation’ of the themes that emerge from participant interviews: that is, when similar themes are being extracted, the sample was deemed adequate (199). This is a standard sample size estimator used in this type of qualitative research

(195). Standard recommended analytical techniques were used for analysis of the qualitative data that emerged from the study (195). The purposive sampling method employed is opportunistic. It primarily involves inviting members from the target audience to participate in the study.

73

Improving the Quality of Care for patients with Ileal Crohn’s disease undergoing Surgery

Standard recommended analytical techniques were used for analysis of the qualitative data that emerged from the study (195). All interviews were audiotaped and transcribed verbatim.

Transcripts were cross-checked with the original recordings to ensure accuracy. Each transcript was subsequently analysed for content to identify emergent themes by a coder. Emergent themes were reviewed by a second person who was blind to the theme extraction process, and key themes were tabulated. To establish confirmability in accordance with qualitative research, all themes were finally reviewed by a senior member of the research team with background in psychology and patient safety (197). Participation was voluntary, and informed consent was obtained from all participants prior to data collection. Anonymity was ensured throughout the study.

2.3. Achieving consensus through Delphi

There are many decisions within the context of healthcare that need to be made without adequate objective information. Decision-making in this setting often require a combination of both qualitative and quantitative techniques. Issues, or items, deemed relevant through the qualitative process can be ranked and structured in an explicit and transparent quantitative form. Key features for formal consensus methods require individual consensus group members to be provided with a synthesis of all evidence relating to the subject matter; privacy, such that individual members can express thoughts and opinions that other members of the consensus group remain unaware off; the opportunity for individual members to change their views in light of initial views of all group members; and explicit and transparent derivation of the group’s decision, based on pre-arranged statistical methods of aggregation and analysis (200).

74

Improving the Quality of Care for patients with Ileal Crohn’s disease undergoing Surgery

The Delphi methodology uses a systematic process of consulting, collecting, evaluating and tabulation of expert opinion on a specific topic without bringing experts together. Questions and statements are posed to experts and answered anonymously within a round. The benefit being it can sample the opinion of a group of experts without being altered by the opinions of influential persons. Exposures to the responses are provided and members revise their opinions on consecutive rounds until convergence and consensus is reached (201, 202). This method has been applied by authors of this study with success to complex surgical issues (203).

2.3.1. Materials and survey design

The survey in Chapters 7 and 8 was designed using an online survey tool freely available to our research team (Qualtrics) (see appendix iii and v for survey). The survey was designed to obtain the specific information from participants in both a free text, dichotomous response entry format (i.e. yes/no) and Likert scales. Likert scale were categorised from 1 (= not important) to 5 (= very important). Successive rounds were carried out, with participants being informed of aggregated responses, until formal consensus was reached. Survey responses were analysed using the Statistical Package for the Social Sciences (SPSS) v23.

2.3.2. Data analyses

Descriptive analyses were performed. For Likert scales, median scores and inter-quartile range values are reported. Consensus was defined with an IQR ≤ 1. Items with a median score > 3, were considered eligible for inclusion. For dichotomous responses, consensus was defined a priori by a pre-defined agreement of greater than 60% across panellists. Standards agreement of the cut off for Delphi studies is set at 70% across panellists. For multinomial outcomes, rather than continuous scales, this cut off has been recommended to be set at 60% (201-204).

75

Improving the Quality of Care for patients with Ileal Crohn’s disease undergoing Surgery

CHAPTER 3: THE INSTITUTIONAL INCIDENCE AND PREDICTORS OF INTRA-

ABDOMINAL SEPTIC COMPLICATIONS (IASC) FOLLOWING ILEOCOLONIC

RESECTION FOR ILEAL CROHN'S DISEASE

3.1. Abstract

Intra-abdominal septic complications following ileocolonic resection for Crohn’s disease are common. Determining predictors can aid pre-operative and peri-operative strategies to reduced morbidity. This chapter aims to determine the incidence and predictors of intra-abdominal septic complications following ileocolonic resection for Crohn’s disease. A single centre, retrospective study was conducted. The clinical case notes of patients with histopathologically proven Crohn’s disease who underwent an ileocolonic resection as a one-stage or two-stage procedure were reviewed. The primary endpoint was the formation of intra-abdominal septic complications within a 30-day post-operative timeframe. Overall 163 patients underwent 175 ileocolonic procedures. Post-operative intra-abdominal septic complications were demonstrated in 9% [13/132] of one-stage procedures and 12% [4/33] of two-stage procedures

[p = 0.2]. Post-operative IASCs following a one-stage procedure demonstrated associations with smokers [p = 0.004], intraoperative abdominal sepsis [p = 0.005], concomitant upper gastrointestinal Crohn’s [p = 0.015], the presence of peri-operative anaemia [p = 0.037], hypoalbuminaemia (< 25 g/L) [p = 0.04] and histologically involved margins [p = 0.001].

Multivariate analysis demonstrated the presence of intra-abdominal sepsis [HR 8.6, 95% CI:

1.2 – 60.1] and the use of peri-operative biologics [HR 24.6, 95% CI: 2.0 – 298] as independent predictors of post-operative intra-abdominal septic complications. These findings may be important when optimising patients for surgery, as well as planning an appropriate operative strategy. It highlights a need to further categorise fistulating and perforating disease and validate these sub-stages.

76

Improving the Quality of Care for patients with Ileal Crohn’s disease undergoing Surgery

3.2. Aims

The aim of this chapter was to identify the incidence of post-operative IASCs in a specialist

IBD unit and identify pre-operative risk factors for the development of IASCs following ileocolonic resection for Crohn’s Disease within this unit.

3.3. Methodology

3.3.1. Patient selection

The clinical case records identified following database screening from the Crohn’s surgical database (see section 2.1) were reviewed further to ensure the following inclusion criteria were met to meet the stated objectives:

➢ All patients who underwent a ‘one-stage’ resection and anastomosis (without stoma

formation) for terminal ileal Crohn’s disease.

➢ All patients who underwent a ‘two-stage’ procedure: (i) resection with ileostomy formation

and (ii) stoma reversal and restoration of intestinal continuity following an earlier

ileocolonic resection with stoma formation.

3.3.2. Outcome measures

The primary outcome measure was the development of post-operative intra-abdominal septic complication. This was defined by the presence of an anastomotic leak (with radiological or intra-operative evidence of a peri-anastomotic collection), intra-abdominal collection (with radiological or intra-operative evidence of an intra-abdominal collection) or enterocutaneous fistulae formation.

77

Improving the Quality of Care for patients with Ileal Crohn’s disease undergoing Surgery

3.3.3. Statistical analysis

Section 2.1.3 provides a further detail on the assessment of data normality and subsequent reporting. Section 2.1.7 describes how missed data was handled. Categorical variables were investigated with the Chi squared (χ2) test. Logistic regression analysis was used to investigate categorical independent variables against the dependant outcome (the presence or absence of post-operative IASC). Variables with a p value of < 0.1 following univariate analysis were including into the multivariate logistic regression analysis. Significance was determined using a p value of < 0.05.

3.4. Results

3.4.1. Patient and operative demographics

Over a six-year period, 163 patients underwent 175 ileocolonic surgical procedures for Crohn’s disease within our institution. Surgical procedures were carried out across a team consisting of 7 consultant colorectal surgeons, and included 142 one-stage procedures, and 33 two-stage procedures (figure 3.1). The mean age at surgery was 36 years [SD 13.1]. Seventy-nine patients were female [79/163; 48%]. The mean length of stay following surgery was 9.5 days

(SD 8).

78

Improving the Quality of Care for patients with Ileal Crohn’s disease undergoing Surgery

Figure 3. 1 Flow diagram demonstrating the final patient and procedure breakdown across one- stage and two-stage produces.

3.4.2. The incidence of IASCs and subsequent management

The presence of post-operative IASCs was observed within 30-days following 17 [9.7%] surgical procedures [n = 13 one-stage procedures and n = 4 two-stage procedures; p = 0.201].

Four procedures developed post-operative IASCs that required eventual surgical re- intervention [4/175; 2%] as definitive management in the long term, six procedures developed post-operative IASCs that required radiological intervention [6/175; 3%], and 7 procedures developed post-operative IASCs that required only conservative management [7/175; 4%] as definitive management in the short term (table 3.1).

79

Improving the Quality of Care for patients with Ileal Crohn’s disease undergoing Surgery

Intra- Patient Operative Procedure Complication Intervention Findings Inflammat Elective open ileocolonic re- Drained under Right para-colic 1. 51 Male ory TI resection for recurrent radiological collection stricture disease guidance Emergency open ileocolonic Drained under Fibrotic TI re-resection for recurrent Peri-anastomotic 2. 50 Male radiological stricture disease (end to end hand- collection guidance sewn anastomosis) Open right hemicolectomy Drained under Complex Right para-colic 3. 16 Male (hand-sewn side to end radiological TI mass collection anastomosis) guidance Mesenteric abscess from peri- Emergency open ileocolonic anastomoti Drained under re-resection for recurrent Peri-anastomotic 4. 39 Male c sinus radiological disease and defunctioning collection tract guidance stoma formation coming from the ileum Complex Elective open ileocaecal re- Pelvic collection Drained under 5. 30 fistulating resection for recurrent from anastomotic radiological Female TI mass disease leak guidance Emergency right Drained under 6. 29 Complex hemicolectomy (end to end Pelvic collection radiological Female TI mass hand-sewn anastomosis) guidance Fibrotic TI Elective open ileocolonic re- stricture, resection for recurrent ileo- disease (with side to side Healed with Enterocutaneous 7. 49 Male jejunal and stapled anastomosis), repair antibiotic therapy fistulae ileo- of jejunal fistula and Roux- alone duodenal en-Y repair of duodenal fistulae fistula. Treated with Elective ileocolonic Fibrotic TI Peri-anastomotic intravenous 8. 39 Male resection (with stapled side stricture collection antibiotic therapy to side anastomosis) alone TI fibrotic Elective open right Treated with stricture hemicolectomy (with Enterocutaneous intravenous 9. 29 Male and psoas stapled side to side fistulae antibiotic therapy abscess anastomosis) alone Enterocuta neous fistula Surgical excision of following Elective open ileocolonic re- persistent an open Enterocutaneous 10. 29 Male resection for recurrent enterocutaneous ileocolonic fistulae disease fistula two years resection later. for TI fistulating mass

80

Improving the Quality of Care for patients with Ileal Crohn’s disease undergoing Surgery

Mesenteric Emergency laparoscopic Treated with and inter- converted to open ileocaecal Peri-anastomotic intravenous 11. 28 Male loop resection (side to side collection antibiotic therapy abscess stapled anastomosis) alone Enterocuta Elective open ileocolonic 12. 57 Enterocutaneous Surgical excision of neous resection and repair of ileo- Female fistulae ECF 11 months later fistula cutaneous fistulae Restoratio Healed with 13. 31 n of Elective open restoration of Enterocutaneous antibiotic therapy Female intestinal intestinal continuity fistulae alone continuity Inflammat ory TI stricture, Elective open ileocolonic re- Healed with ileo- resection for recurrent Enterocutaneous 14. 24 Male antibiotic therapy cutaneous disease and concomitant fistulae alone fistula and sigmoidectomy sigmoid colitis Restoratio Healed with n of Elective open restoration of Enterocutaneous 15. 11 Male antibiotic therapy intestinal intestinal continuity fistulae alone continuity Laparotomy six months later with Inflammat Anastomotic redo right ory Elective laparoscopic 16. 31 Leak and hemicolectomy, stricture ileocaecal resection (with Female enterocutaneous repair of and sinus stapled anastomosis) fistulae enterocutaneous tract fistulae and sigmoid colectomy. Ileocaecal resection for TI Laparotomy six perforation months later with and split 17. 39 Elective open restoration of Enterocutaneous repair of stoma Female intestinal continuity fistulae enterocutaneous formation fistulae and sigmoid done colectomy. elsewhere 6 months prior Table 3. 1 Demographics and operative profiles of 17 patients who demonstrated post-operative

IASCs - Five ileocolonic resections underwent concomitant strictureplasty for proximal small bowel strictures [5/175; 2.9%] and eight patients underwent concomitant sigmoidectomy.

81

Improving the Quality of Care for patients with Ileal Crohn’s disease undergoing Surgery

3.4.3. One-stage procedures – ileocolonic resection and anastomoses [n = 142]

There were 104 patients who underwent an elective one-stage procedure, 73.1% [76/104] in view of intractible medical therapy and 26.9% [28/104] of whom were on no medical therapy.

Twenty patients underwent an emergency one-stage procedure, 70% [14/20] were on medical therapy and 30% [6/20] were on no medical therapy.

Post-operative IASCs occurred in 13 out of a total of 142 [9.2%] one-stage ileocolonic resections with anastomoses, 7 of which were ECF formations and 6 were anastomotic leak requiring radiological or operative re-intervention (table 3.1: patients 1-3, 5-12, 14 and 16).

82

Improving the Quality of Care for patients with Ileal Crohn’s disease undergoing Surgery

3.4.3.1. Patient and disease related variables

Total No IASC 3 IASC 3 Chi Patient and disease related variables n [%] n [%] n [%] squared 142 [100] 129 [91] 13 [9] [p value] Female 71 [50] 67 [94] 4 [6] Gender 0.15 Male 71 [50] 62 [87] 9 [13] Never smoker 60 [42] 59 [98] 1 [1.7] 0.004 Smoking Smoker 73 [51] 61 [84] 12 [16] Missing 9 [6] No 90 [63] 83 [92] 7 [8] 0.39 Family History IBD Yes 22 [16] 19 [86] 3 [14] Missing 30 [21] <16 24 [17] 22 [92] 2 [8] 16-40 99 [70] 89 [90] 10 [10] 0.46 Age of disease onset >40 14 [10] 14 [100] 0 [0] Missing 5 [3] Terminal ileal 82 [58] 76 [93] 6 [7] Disease location 0.38 Ileocolonic 60 [42] 53 [88] 7 [12] Absent 94 [66] 84 [89] 10 [11] 0.41 Concomitant perianal Present 47 [33] 44 [94] 3 [6] Missing 1 [1] Absent 136 [96] 125 [92] 11 [8] 0.01 Concomitant UGI 1 CD 2 Present 5 [3] 3 [60] 2 [40] Missing 1 [1] Stricturing 76 [54] 73 [96] 3 [4] Disease phenotype 0.02 Penetrating 66 [46] 56 [85] 10 [15] Absent 71 [50] 66 [93] 5 [7] 0.50 Re-resection for recurrent Present 68 [48] 61 [90] 7 [10] disease Missing 3 [2] UGI = Upper Gastrointestinal; CD = Crohn’s Disease; IASC = intra-abdominal septic complication

Table 3. 2 Frequency data and univariate analyses for patient and disease related characteristics

across 142 patients undergoing a one-stage procedure – A significantly high proportion of

IASC rates were seen across smokers [p = 0.004], patient with concomitant upper GI Crohn’s

disease [p = 0.01] and penetrating disease [p = 0.02].

83

Improving the Quality of Care for patients with Ileal Crohn’s disease undergoing Surgery

There were a greater proportion of smokers who developed post-operative IASCs (12/73 [16%] and 1/60 [2%] respectively; p = 0.004). Two patients with concomitant UGI CD also developed post-operative IASCs (2/5 [40%] and 11/136 [8%] respectively; p = 0.015). Three patients with stricturing disease (Montreal ‘B2’ subtype) developed post-operative IASCs and 10 patients with a penetrating disease (Montreal ‘B3’ subtype) developed post-operative IASCs

(3/76 [4%] and 10/66 [15%] respectively; p = 0.021).

3.4.3.2. Peri-operative medications

Total No IASC 4 IASC 4 Chi squared Peri-operative medication n [%] n [%] n [%] [p value] 142 [100] 129 [91] 13 [9] Absent 104 [73] 97 [93] 7 [7] Corticosteroids Present 34 [24] 29 [85] 5 [15] 0.15 Missing 4 [3] Absent 120 [85] 110 [92] 10 [8] Antibiotics Present 10 [7] 8 [80] 2 [20] 0.22 Missing 12 [8] Absent 68 [48] 60 [88] 8 [12] Thiopurine / 6- Present 58 [41] 54 [93] 4 [7] 0.35 MP 2 Missing 16 [11] Absent 82 [58] 72 [88] 10 [12] 5ASA 1/ Present 43 [30] 41 [95] 2 [5] 0.17 Sulphasalazine Missing 17 [12] Anti-TNF 3 / Absent 126 [89] 116 [92] 10 [8] Biological Present 4 [3] 2 [50] 2 [50] 0.004 Therapy Missing 12 [8] Absent 125 [88] 114 [91] 11 [9] Methotrexate Present 6 [4] 5 [83] 1 [17] 0.51 Missing 11 [8] Absent 113 [80] 102 [90] 11 [10] Nutritional Present 16 [11] 15 [94] 1 [6] 0.65 Supplementation Missing 13 [9] 5-ASA = 5-aminosalicylic acid; 6-mercaptopurine = 6-MP; TNF = tumour necrosis factor; IASC = intra- abdominal septic complication

84

Improving the Quality of Care for patients with Ileal Crohn’s disease undergoing Surgery

Table 3. 3 Data and univariate analysis for pre-operative medications across 142 patients undergoing a one-stage procedure – A significantly high proportion of IASC rates were seen across patients receiving peri-operative biological therapy [p = 0.004].

There were five instances of post-operative IASCs in patients who received peri-operative corticosteroid therapy compared with seven incidences in patients without peri-operative corticosteroid therapy (5/34 [15%] and 7/104 [7%] respectively; p = 0.152). A higher proportion of patients receiving peri-operative anti-TNF biologics demonstrated post-operative

IASCs compared to those not receiving anti-TNF biologics (2/4 [50%] and 10/126 [8%] respectively, p = 0.004).

85

Improving the Quality of Care for patients with Ileal Crohn’s disease undergoing Surgery

3.4.3.3. Peri-operative blood parameters

Total No IASC 1 IASC 1 Peri-operative Blood n [%] n [%] n [%] Chi squared Parameters [p value] 142 [100] 129 [91] 13 [9] Present 54 [38] 46 [85] 8 [15] Anaemia Normal 86 [61] 82 [95] 4 [5] 0.04 Missing 2 [1] Low 58 [41] 52 [89] 6 [10] Haematocrit Normal 81 [57] 75 [93] 6 [7] 0.54 Missing 3 [2] Leucocytosis Absent 113 [80] 106 [94] 7 [6] [count > 11 Present 27 [19] 22 [81] 5 [19] 0.04 x109 / L] Missing 2 [1] CRP < 40 58 [41] 53 [91] 5 [9] C-Reactive CRP ≥ 40 24 [17] 21 [88] 3 [12] 0.59 Protein Missing 60 [42] Thrombocytosi Absent 110 [78] 102 [93] 8 [7] s [Platelet Present 30 [21] 26 [87] 4 [13] 0.29 count > 450 x 109 / L] Missing 2 [1] Hypoalbumina Present 7 [5] 5 [71] 2 [29] emia [Serum Absent 130 [92] 121 [93] 9 [7] 0.04 albumin < 25 x 109 / L] Missing 5 [3] Hypocalcaemia Present 5 [4] 5 [100] 0 [0] [Serum Absent 97 [68] 87 [90] 10 [10] corrected 0.45 Calcium < 2.20 Missing 40 [28] mmol / L] IASC = intra-abdominal septic complication

Table 3. 4 Frequency data and univariate analysis for peri-operative blood parameters across

142 patients undergoing a one-stage procedure – A significantly high proportion of IASC rates were seen across patients with anaemia [p = 0.04], leucocytosis [p = 0.04] and hypoalbuminaemia [p = 0.04].

Eight patients with peri-operative anaemia developed post-operative IASCs (8/54 [15%] and

4/86 [5%] respectively; p = 0.037). Five patients with a peri-operative leucocytosis developed

86

Improving the Quality of Care for patients with Ileal Crohn’s disease undergoing Surgery post-operative IASCs (5/27 [18%] and 7/113 [6%] respectively; p = 0.04). Two patients with serum hypoalbuminaemia (serum albumin less than 25g/L) developed post-operative IASCs

(2/7 [29%] and 9/130 [7%] respectively; p = 0.04).

3.4.3.4. Peri-operative variables

Chi Total n [%] No IASC 2 n [%] IASC 2 n [%] squared Peri-operative variables [p value] 142 [100] 129 [91] 13 [9] ASA < 3 117 [83] 106 [91] 11 [9] ASA ASA ≥ 3 12 [8] 10 [83] 2 [17] 0.43 Missing 13 [9] Elective 118 [83] 109 [92] 9 [8] Scheduling Urgent 23 [16] 19 [83] 4 [17] 0.14 Missing 1 [1] Intra- Present 57 [40] 47 [82] 10 [18] operative Absent 84 [59] 81 [96] 3 [4] 0.005 abdominal sepsis Missing 1 [1] Laparoscopic 28 [20] 27 [96] 1 [4] Access Open 113 [80] 101 [89] 12 [11] 0.25 Missing 1 [1] Stapled 87 [62] 79 [91] 8 [9.2] Anastomosis Handsewn 22 [15] 18 [82] 4 [18] 0.23 Missing 33 [23] Clear 52 [37] 52 [100] 0 [0] Resection Involved 62 [43] 51 [82] 11 [18] 0.001 Margins Missing 28 [20] ASA - American Society of Anaesthesiologists; IASC = intra-abdominal septic complication Table 3. 5 Frequency data and univariate analysis for peri-operative patient, disease and operative variables across 142 patients undergoing a one-stage procedure – A significantly high proportion of IASC rates were seen across patients with intra-operative sepsis [p = 0.005] and involved histological resection margins [p = 0.001].

87

Improving the Quality of Care for patients with Ileal Crohn’s disease undergoing Surgery

Ten patients with intraoperative sepsis, and three patients with stricturing disease, developed post-operative IASCs, (10/57 [17%] and 3/84 [4%] respectively; p = 0.005). Eleven patients with involved histological inflammatory developed post-operative IASCs (11/62 [18%] and

0/52 [0%] respectively; p = 0.001).

3.4.3.5. Multivariate analysis

Variable Multivariate Analysis P Value Odds 95% Ratio Confidence Interval Smoking Never Smoker * - Smoker 0.06 9.3 0.9 - 96.3 Concomitant UGI Absent * - Present 0.36 7.2 0.1 - 499 Biological Therapy Absent * - Present 0.01 24.6 2.0 - 298.1 Anaemia Present 0.79 1.3 0.2 - 6.5 Normal * - Hypoalbuminaemia Present 0.43 5.9 0.1 - 476.3 (Serum albumin < Absent * - 25 x 109 / L) Intra-abdominal Abscess +/- 0.03 8.6 1.2 - 60.1 sepsis Fistula Stricturing * - * = reference categories Footnote: - A list-wise deletion of cases was performed on the assumption of ‘MCAR data’ based on 28 (3%) of all variables included within Logistic Regression model. - Resection margin was excluded from multivariate analysis because of its large missing data subset (20%). - Leucocytosis and disease phenotype were excluded from multivariate analysis because of their collinearity with intra-abdominal sepsis.

Table 3. 6 Multivariate analysis adjusting for smoking status, concomitant UGI Crohn’s, peri- operative biological therapy, peri-operative anaemia, peri-operative hypoalbuminaemia, intra- operative sepsis – The presence of intra-abdominal sepsis (OR 8.6, 95% CI 1.2 – 60.1; P =

0.03), and peri-operative biological therapy (OR 24.6, 95% CI: 2 – 298; P = 0.01) were

88

Improving the Quality of Care for patients with Ileal Crohn’s disease undergoing Surgery demonstrated as independent risk factors for the development of post-operative intra- abdominal septic complications.

The presence of intra-abdominal sepsis (OR 8.6, 95% CI 1.2 – 60.1; p = 0.03), and peri- operative biological therapy (OR 24.6, 95% CI: 2 – 298; p = 0.01) were associated with post- operative IASCs when adjusting for smoking status, concomitant UGI Crohn’s, peri-operative anaemia, and peri-operative hypoalbuminaemia.

3.4.4. Two-stage procedures

3.4.4.1. Primary procedure - ileocolonic resection with stoma formation

There were 20 ileocolonic resections with either split stoma (without an anastomosis) (n = 15) or proximal defunctioning stoma formations (with an anastomosis) (n = 5). In the latter group, one case demonstrated IASC (1/20, 5%) in the form of an intra-abdominal collection that required operative re-intervention (table 2: patient 4). A large proportion demonstrated intra- operative findings of fistulating and/or abscess formation (17/20, 85%). Twelve cases underwent an elective stoma reversal procedure at a later date.

3.4.4.2. Secondary procedure - elective stoma reversal and restoration of intestinal continuity

There were 13 cases that underwent a stoma reversal procedure following an early ileocolonic resection with stoma formation, 12 of which the primary procedure (ileocolonic resections with stoma formation) was performed within our cohort of patients and one where the primary procedure was performed elsewhere. IASC was demonstrated in 3 cases (3/13, 23.1%) (table

2: patient 13, 15 and 17).

89

Improving the Quality of Care for patients with Ileal Crohn’s disease undergoing Surgery

3.5. Discussion

3.5.1. Summary of findings

This chapter demonstrates the overall incidence of 30-day IASCs following ileocolonic resection within a specialist IBD unit over a six-year period as 9.7%. There is currently no literature to provide the standard incidence of post-operative IASCs, however a number of single-centre reports have demonstrated rates as high as 17% and as low as 3% (205, 206). A two-stage procedure may be an important operative strategy in reducing the incidence of post- operative IASCs.

The peri-operative well-being of a patient may be crucial in determining 30-day post-operative outcome. Similar studies have demonstrated that the presence of active disease and intra- abdominal sepsis does correlate with post-operative morbidity. The results have demonstrated the presence of intra-abdominal, intra-operative sepsis is an independent predictor of post- operative IASCs. This finding is consistent with previous reports (110, 111, 116-118). This does suggest a need to consider pre-operative antibiotic therapy to improve 30-day outcome, as well as percutaneous drainage of underlying sepsis. Some centres advocate corticosteroid therapy to induce remission in the setting of active Crohn’s. Intra-abdominal, intra-operative sepsis, in this study broadly defined as the presence of an abscess/collection or fistulae or both at the time of surgery. Further studies should aim to identify which specific subsets of penetrating disease (fistulation or perforation) impact on morbidity. Chapter 6 analyses this through the development and validation of a radiological staging tool for ileocolonic Crohn’s disease. The findings within this chapter did not demonstrate a significant association between peri-operative corticosteroid administration and post-operative IASC formation. The limitations of the study may be a reason for this contradictory finding (87, 110, 111, 117, 120).

90

Improving the Quality of Care for patients with Ileal Crohn’s disease undergoing Surgery

Most studies have not demonstrated a significant association between peri-operative ant-TNFα therapy and the development of post-operative IASCs (207-211). The use of anti-TNF as a

‘last resort’ in the escalation of medical treatment of Crohn’s disease may be contributory to the findings of this study, as patients come to surgery with advanced disease and are more prone to complications.

Determining the patient’s individual risk of post-operative IASCs is necessary when obtaining informed consent prior to surgery. Studies have demonstrated the incidence of post-operative

IASC increases with the number of risk factors present. The incidence of post-operative IASCs was 40% if the patient had both pre-operative corticosteroids and intra-abdominal abscesses present at the time of surgery.

Setting guidelines in pre-operative optimisation and operative practice is required in order to improve patient outcome and reduce the incidence of post-operative morbidity following ileocolonic resection for Crohn’s disease. Highlighting specific predictors of post-operative

IASCs is necessary for this to occur.

3.5.2. Limitations

The retrospective nature of the study and relatively small sample size may explain the lack of other significant findings, including pre-operative corticosteroid use, smoking status and biologic as risk factors for post-operative IASC. Further limitations relating to the retrospective validity design, and lack of sensitivity testing around pre-defined categorical blood parameter cut-offs are discussed in section 9.2.

91

Improving the Quality of Care for patients with Ileal Crohn’s disease undergoing Surgery

CHAPTER 4: THE INSTITUTIONAL INCIDENCE AND PREDICTORS OF POST-

OPERATIVE CLINICAL RECURRENCE FOLLOWING ILEOCOLONIC

RESECTION FOR ILEAL CROHN'S DISEASE

4.1 Abstract

Post-operative clinical recurrence can occur in up to 55% of patients within 5 years. Predicting the risk of recurrence is key in determining appropriate treatment strategies. This study aims to determine the incidence of post-operative clinical recurrence and determine predictors of recurrence within a specialist institution. The clinical case records of 142 patients who underwent either a one-stage or two-stage procedure from 1st January 2005 to 31st December

2010 for ileocolonic Crohn’s disease were reviewed. Pre-operative, peri-operative and post- operative variables were extracted. Post-operative clinical recurrence was defined as an initiation or change in medical treatment for recurrent symptoms with endoscopic or radiological evidence of active disease. Time to clinical recurrence was measured in months after surgery. Over a six-year period (1st January 2005 to 31st December 2010), follow-up data were obtained on 142 patients. The median follow-up was 28.5 months. Clinical recurrence was demonstrated in 59 patients [41.5%]. The proportion of patients with clinical recurrence at 5 years was 48.2%. Predictors of recurrence included a re-resection for recurrent disease [HR 1.9; 95% CI: 1.1 – 3.3; P = 0.02], and ileocolonic disease [HR1.7; 95% CI: 1.0 –

2.9; p = 0.05]. The chapter discusses the relevance these finding may have in current risk stratification models and the catering post-operative prophylactic therapy. Identification of predictors for post-operative clinical recurrence is important in determining the post-operative strategy. This study provides a unique perspective of the incidence of recurrence and associated predictors from the perspective of a specialist unit.

92

Improving the Quality of Care for patients with Ileal Crohn’s disease undergoing Surgery

4.2. Aims

This chapter aims to determine the incidence of post-operative clinical recurrence and to determine pre-operative and peri-operative predictors of post-operative clinical recurrence from a specialist IBD institution.

4.3. Methodology

4.3.1. Patient selection

The clinical case records identified following screening from the Crohn’s surgical database

(see section 2.1) were reviewed further to ensure the following eligibility criteria were met to meet the stated objectives:

Inclusion criteria:

1. All patients who underwent a ‘one-stage’ resection and anastomosis (without stoma

formation) for terminal ileal Crohn’s disease.

2. All patients who underwent a ‘two-stage’ procedure with an anastomosis including (i)

resection with defunctioning ileostomy formation and (ii) stoma reversal and restoration of

intestinal continuity following an earlier ileocolonic resection with stoma formation.

Exclusion criteria:

1. No follow-up data

93

Improving the Quality of Care for patients with Ileal Crohn’s disease undergoing Surgery

4.3.2. Outcome measures

The primary outcome measure was the development of clinical recurrence. This was defined as the presence of recurrent Crohn’s symptoms, with radiological or endoscopic evidence of active disease, warranting an initiation or change in medical or nutritional therapy. The follow- up was measured in months after surgery until two events: the time to clinical recurrence and the time in remission until the last documented correspondence.

4.3.3. Statistical analysis

Section 2.1.3 provides a further detail on the assessment of data normality and subsequent reporting. Section 2.1.7 describes how missed data was handled. Life table analysis was used to determine the five-year incidence of post-operative clinical recurrence. Univariate analysis was performed using the Log-Rank test. Visual analyses were performed using the Kaplan-

Meier method. Multivariate analysis was performed using Cox Proportional-

Hazards Regression models. Significance was determined using a p value of < 0.05.

94

Improving the Quality of Care for patients with Ileal Crohn’s disease undergoing Surgery

4.4. Results

4.4.1. Patient and operative demographics

The flow diagram of the study is shown in figure 4.1.

Figure 4. 1 Flow diagram demonstrating the screening process and final number of procedures included and excluded from the study.

236 patients were identified for eligibility screening over a six-year period (1st January 2005 to

31st December 2010). Follow-up data were obtained on 142 patients: 130 patients underwent a one-stage procedure (ileocolonic resection with anastomosis, no stoma) [130/142; 91.5%] and

12 patients underwent a two-stage procedure [12/142; 8.5%]. Of those who underwent a two- stage procedure, patients with a stoma reversal included those who had a resection for fistulating disease [n=4], fibrotic disease with hypoalbuminaemia [n=1] and intra-abdominal

95

Improving the Quality of Care for patients with Ileal Crohn’s disease undergoing Surgery abscess formation [n=5] at the resectional stage six months prior. Two patients underwent an ileocolonic resection with anastomosis and proximal defunctioning stoma formation for intra- abdominal abscess [n=2] (figure 4.2).

One stage resection and anastomosis [N = 130] Stoma reversal and delayed anastomosis [N = 10] Ileocolonic resection and stoma formation [N = 2]

Figure 4. 2 Pie chart demonstrating relative proportions of ileocolonic surgical procedures across the 142 included patients – There were 130 patients (92%) who underwent a one-stage procedure, 2 patients (1%) who underwent stoma formation and 10 patients (7%) who underwent restoration of continuity.

A total of 94 patients were excluded from the analysis for (i) having undergone an ileocolonic strictureplasty with no bowel resection [n = 1], (ii) no evidence of Crohn’s disease on histopathology [n = 21], (iii) previous colectomy [n=51], (iv) no long-term follow-up data

[n=8] and (v) resection with split stoma formation [n=13]. The mean age at surgery was 37.5 years [SD 13.6]. The median follow-up was 28.5 [1 – 101] months.

96

Improving the Quality of Care for patients with Ileal Crohn’s disease undergoing Surgery

4.4.2. The incidence of clinical recurrence

Clinical recurrence was demonstrated in 59 [42%] patients and 83 [59%] remained in remission up to the last documented follow-up (figure 4.3).

59 Remission [N = 83]

Post-operative clinical 83 recurrence [N = 59]

Figure 4. 3 Pie chart demonstrating proportions of patients who developed post-operative clinical recurrence and those who maintained remission out of the 142 patients – There were

59 patients who demonstrated clinical recurrence (42%) and 83 patients who maintained remission (58%).

97

Improving the Quality of Care for patients with Ileal Crohn’s disease undergoing Surgery

The median number of months to clinical recurrence following surgery was 24 [2 – 92]. The proportion of patients with clinical recurrence at 5 years was 48.2% (figure 4.4).

Figure 4. 4 Life table analysis demonstrating the proportion of patients in remission (each drop in the curve represents a patient developing clinical recurrence) - The proportion of patients in remission at 1, 2, 3, 4 and 5 years following ileocolonic resection for Crohn’s disease are

91.7%, 77.3%, 64.6%, 60.5% and 51.8%.

98

Improving the Quality of Care for patients with Ileal Crohn’s disease undergoing Surgery

The median number of months until the last documented time in remission following surgery was 34 [1 – 101].

4.4.3. Pre-operative factors

Patient related Total Remission Recurrence Log Rank test N (%) N (%) N (%) (P value) 142 (100) 83 (58) 59 (42) Gender Female 70 (49) 42 (60) 28 (40) 0.18 Male * 72 (51) 41 (57) 31 (43) Smoking status Never Smoker * 60 (42) 38 (63) 22 (37) 0.52 Smoker 73 (51) 41 (56) 32 (44)

Missing data 9 (6) Family History Absent * 93 (65) 57 (61) 36 (39) 0.91 Present 22 (15) 13 (59) 9 (41)

Missing data 27 (19)

Disease related Age of Disease <16 (A1) 24 (17) 13 (54) 11 (46) 0.45 Onset 16-40 (A2) 101 (71) 59 (58) 42 (42) >40 (A3) * 14 (10) 9 (64) 5 (36)

Missing data 3 (2) Disease Location Terminal Ileum 73 (51) 48 (66) 25 (34) 0.05 (L1) * Ileocolonic (L3) 69 (49) 35 (51) 34 (49) Concomitant No * 94 (66) 55 (59) 39 (41) 0.50 Perianal Yes 47 (33) 28 (60) 19 (40)

Missing data 1 (1) Concomitant No * 136 (96) 80 (59) 56 (41) 0.55 UGI Yes 6 (4) 3 (50) 3 (50) Disease Stricturing (B2) * 72 (51) 41 (57) 31 (43) 0.70 Phenotype Penetrating (B3) 70 (49) 42 (60) 28 (40) Re-resection for No (Primary 66 (46) 42 (64) 24 (36) 0.02 recurrent resection) * disease Yes 74 (52) 40 (54) 34 (46)

Missing data 2 (1) * = Reference Category

Table 4. 1 Data, univariate (log rank analyses) for patient and disease related pre-operative variables - – Log rank analyses demonstrate a higher proportion of patients with ileocolonic

99

Improving the Quality of Care for patients with Ileal Crohn’s disease undergoing Surgery

(Montreal L3) disease [p = 0.049] and a re-resection for recurrent disease [p = 0.016] with post- operative clinical recurrence.

The proportion of smokers and non-smokers was 51% (73/142) and 42% (60/142). The development of clinical recurrence was 44% (32/73) in smokers and 37% (22/60) in non- smokers (p = 0.517). The proportion of patients with ileal (Montreal L1) and ileocolonic

(Montreal L3) disease was 51% (73/142) and 49% (69/142). The presence of clinical recurrence was significantly greater in patients with ileocolonic disease 49% (34/69) compared with ileal disease 34% (25/73) (p = 0.049) (figure 4.5).

Ileal (Montreal L1)

Ileocolonic (Montreal L3)

Log Rank: p = 0.049

100

Improving the Quality of Care for patients with Ileal Crohn’s disease undergoing Surgery

Figure 4. 5 Kaplan-Meier curves of months in remission for patients with ileal (L1) Crohn’s and ileocolonic (L3) Crohn’s disease – Kaplan-Meier curves of months in remission for patients with ileal (L1) Crohn’s and ileocolonic (L3) Crohn’s disease – patients with ileocolonic (Montreal L3) disease developed earlier episodes of post-operative clinical recurrence compared to those with ileal (Montreal L1) disease.

The proportion of patients with stricturing disease (Montreal B2) and a penetrating disease

(Montreal B3) was 51% (72/142) and 49% (70/142). The disease phenotype was not associated with clinical recurrence (stricturing: 31/72 [43%]; penetrating: 28/70 [40%]; p = 0.703). The proportion of patients undergoing re-resection of recurrent disease and primary ileocaecal resection was 52% (74/142) and 46% (66/142). Re-resection of recurrent disease was associated with the development of clinical recurrence (re-resection of recurrent disease: 34/74

[46%]; primary: 24/66 [36%]; p = 0.016) (figure 4.6).

101

Improving the Quality of Care for patients with Ileal Crohn’s disease undergoing Surgery

Log Rank: p = 0.016

Figure 4. 6 Kaplan-Meier curves of time in remission for patients undergoing a primary ileocaecal resection and redo ileocolonic resection - Patients who underwent a redo procedure demonstrated earlier post-operative clinical recurrence compared to those who underwent a primary procedure.

Of those who had re-resectional surgery the first re-resection was performed in 38 patients, the second in 15, the third in 13, the forth in 4 and the fifth re-resection in 3. One patient (with intestinal failure) underwent the thirteenth re-resectional operation. When adjusting for all patient and disease-related pre-operative covariates, re-resection for recurrent disease was an

102

Improving the Quality of Care for patients with Ileal Crohn’s disease undergoing Surgery independent risk factor for the development of clinical recurrence (p = 0.037, HR 2.1, 95% CI:

1 – 4.3).

4.4.4. Peri-operative factors

Total Remission Recurrence Log Rank n (%) n (%) n (%) (p value)

142 (100) 83 (58) 59 (42) ASA grade ASA < 3 * 114 (80) 71 (62) 43 (38) 0.9 ASA ≥ 3 14 (10) 8 (57) 6 (43) Missing data 14 (10) Scheduling Elective * 119 (84) 70 (59) 49 (41) 0.6 Emergency 22 (15) 13 (59) 9 (41) Missing data 1 (1) Access Laparoscopic * 25 (18) 16 (64) 9 (36) 0.5 Open 117 (82) 67 (57) 50 (43) Anastomosis Stapled * 85 (60) 52 (61) 33 (39) 0.2 Handsewn 20 (14) 8 (40) 12 (60)

Missing data 37 (26) Intra-operative Penetrating 53 (37) 32 (60) 21 (40) 0.5 findings Non-penetrating * 78 (55) 45 (58) 33 (42) Missing data 11 (8) Resection Clear * 53 (37) 32 (60) 21 (40) 0.3 Margins Involved 63 (44) 32 (51) 31 (49) Missing data 26 (18) Post-operative Absent * 129 (91) 79 (61) 50 (39) 0.04 30-day IASCs Present 13 (9) 4 (31) 9 (69) * = reference category

Table 4. 2 Data, univariate (log rank analyses) for peri-operative variables – Log rank analyses demonstrated a higher proportion of patients with 30-day post-operative intra-abdominal septic complications (IASCs) with clinical recurrence [p = 0.04].

Most patients underwent a one-stage procedure as opposed to a delayed stoma reversal

(132/142 [93%] and 10/142 [7%]). The proportion of patients demonstrating clinical

103

Improving the Quality of Care for patients with Ileal Crohn’s disease undergoing Surgery recurrence following stoma reversal and a one-stage procedure was 50% (5/10) and 41%

(54/132) (p = 0.181). One patient who had an ileocolonic resection with anastomosis and a proximal defunctioning ileostomy developed recurrence 19 months after surgery.

The proportion of patients without post-operative intra-abdominal septic complications (IASC) within 30-day was 91% (129/142) and the proportion of patients with a post-operative IASC was 9% (13/142). The proportion of patients with post-operative IASCs who demonstrated clinical recurrence was 69% (9/13). The proportion without post-operative IASCs who did not demonstrate clinical recurrence was 39% (50/129) (p = 0.043) (figure 4.7).

IASC Absent

IASC Present

Log Rank: p = 0.043

Figure 4. 7 Kaplan-Meier curves of time in remission for patients with and without 30-day morbidity with IASCs (Kaplan-Meier curves of time in remission for patients with and without

30-day morbidity with IASCs) – Patients who developed intra-abdominal septic complications

104

Improving the Quality of Care for patients with Ileal Crohn’s disease undergoing Surgery within a 30-day timeframe following surgery where significantly more likely to demonstrate earlier clinical recurrence compared to those who did not [p = 0.043].

4.4.5. Multivariate analysis

When adjusting for those patients who demonstrated 30-day a post-operative IASC, both ileocolonic (Montreal L3) disease and re-resection of recurrent disease were independent predictors of clinical recurrence. There was a 1.7-fold increased risk in patients who had ileocolonic disease (HR1.7, 95% CI: 1.0 – 2.9), and a 1.9-fold increased risk in patients who had undergone re-resection of recurrent disease (HR1.7, 95% CI: 1.1 – 3.3) (table 4.3).

Significance Hazard 95% Confidence (Cox regression Ratio (HR) Interval for HR - p value) Disease Terminal Ileum (L1) * 0.05 1.7 1.0 - 2.9 Location Ileocolonic (L3) Re-resection No * for recurrent 0.02 1.9 1.1 - 3.3 Yes disease Post- Absent * operative 30- 0.2 1.6 0.8 - 3.3 day IASCs Present * = reference category Footnote: List-wise deletion of cases performed on assumption of MCAR data based on 2 (0.4%) of all variables included within Cox Proportional-Hazards Regression model. Table 4. 3 Multivariate analysis for all pre-operative, peri-operative and post-operative variables found to be significant (p < 0.1) on univariate analysis – Patients who underwent a re-resection for recurrent disease had a 1.9 fold increased risk of developing clinical recurrence

[HR 1.9, 95% CI 1.1-3.3; p = 0.02], and patients who had ileocolonic (Montreal L3) disease demonstrated a 1.7 fold increased risk of developing clinical recurrence [HR 1.7, 95% CI: 1.0

– 2.9; p = 0.05], when adjusting for those who demonstrated post-operative 30-day IASCs.

105

Improving the Quality of Care for patients with Ileal Crohn’s disease undergoing Surgery

4.5. Discussion

4.5.1. Summary of findings

This chapter reports the annual institutional incidence of clinical recurrence over five post- operative years, and pre-operative and peri-operative predictors of clinical recurrence following ileocolonic resection for patients with Crohn’s disease from a specialist IBD institution. Smoking, a history of previous surgical resection and a penetrating disease phenotype are all considered independent risk factors for the development of clinical recurrence (25, 212). Smoking has been demonstrated to have a two-fold increasing effect in the development of clinical recurrence compared with non-smokers (OR = 2.15; 95% CI =

1.42, 3.27; p < 0.001) (26). This chapter demonstrated a similar trend, but its retrospective nature and the inability to extract specific pre-operative, peri-operative and post-operative smoking status may have contributed to the lack of a significant finding. A history of a previous ileocolonic resection has also been demonstrated in numerous studies to be a risk factor for clinical recurrence (136, 212), including similar analyses of previous work from our institution from earlier years (2002 - 2005) (213). The findings from this chapter are consistent with these reports. Recent studies are in favour of perforating disease phenotypes being a risk factor for clinical recurrence when compared with non-perforating disease (212, 214, 215). Interestingly the findings of the present chapter were not consistent with this trend which, again, may reflect its retrospective design.

The occurrence of a post-operative IASC was associated with clinical recurrence, a finding consistent with other published studies (216, 217). This group of patients may represent a distinct phenotype that may benefit from early targeted therapy to maintain remission. In this chapter (and Chapter 3) post-operative IASCs were defined by the presence of anastomotic leakage, intra-abdominal collection or enterocutaneous fistula formation. Further prospective

106

Improving the Quality of Care for patients with Ileal Crohn’s disease undergoing Surgery studies should investigate whether a specific penetrating or perforating subset may be associated with recurrence disease. A consideration should also be given to operative strategy during a Crohn’s ileocolonic resection, as certain techniques may influence clinical recurrence.

A two-stage procedure, with an initial resection and split stoma formation, followed by stoma reversal at a later date, may be considered a safer means of avoiding a post-operative septic complication when a patient is already unwell with complex Crohn’s disease (218).

There has been much work on the role of post-operative medical prophylaxis to prevent recurrence after surgery. Compliance with post-operative maintenance therapy is predictive of a lower risk of symptomatic recurrence (136). Mesalamine has been demonstrated to reduce the incidence of clinical recurrence compared with placebo, although inferior in effect when compared with Azathioprine or 6-MP (137). Its current use is therefore primarily aimed at patients at low risk of recurrence. Indeed thiopurine therapy has been demonstrated to have a more potent effect at maintaining remission and is commonly aimed at moderate risk groups

(138).

There is strong evidence to support the role of biological therapies in the maintenance of post- operative remission (66, 139) but with a standard average annual cost of anti-TNF therapy of

£12,584 (140). Despite this cost, the National Institute of Clinical Excellence has recognised the importance of biologics in maintaining remission in patients with Crohn’s disease (64).

The POCER (Post-operative Crohn’s Endoscopic Recurrence) study was a multi-centre randomised trial that aimed to compare strategies to prevent disease recurrence and identify optimal strategies at preventing recurrence. Patients were randomised into an active care group

(receiving colonoscopy at the 6-month post-operative time-point) and standard care (no colonoscopy). Endoscopic assessment of recurrence was determined using the Rutgeert’s

107

Improving the Quality of Care for patients with Ileal Crohn’s disease undergoing Surgery score. All patients received metronidazole within 3 months following surgery. Post-operative drug therapy was then given to those who were at a high risk of recurrence (smokers, perforating disease, or 2 or more previous resectional surgery). A preliminary report of the study demonstrated adalimumab was superior to thiopurines in high-risk patients in maintaining remission (Rutgeert’s score i0 or i1) at six months following surgery (219). The final report demonstrated a significantly higher proportion of patients with recurrence

(Rutgeert’s i2 – i4) at an 18-month colonoscopy in the standard care group when compared to those receiving active care. An earlier opportunity for treatment escalation may have led to this effect however long term surveillance was deemed necessary for low risk patients as well as high risk (220). Further analyses of clinical predictors of recurrence, with inclusion of disease location highlighted in this chapter, may be necessary to cater an appropriate long-term surveillance strategy in patients with Crohn’s disease.

4.5.2 Limitations

A small sample size may explain the failure to identify smoking status and penetrating disease as risk factors for recurrence. Another limitation was the difficulty in determining what maintenance regime was used following surgery. Patients on biological therapy or thiopurines would certainly experience a prolonged time in remission. Further limitations relating to the retrospective validity design, and lack of sensitivity testing around pre-defined categorical blood parameter cut-offs are discussed in section 9.2.

108

Improving the Quality of Care for patients with Ileal Crohn’s disease undergoing Surgery

CHAPTER 5: THE EFFICACY OF ENDOSCOPIC BALLOON DILATATION IN

AVOIDING SURGERY FOR CROHN'S STRICTURES

5.1. Abstract

Endoscopic balloon dilatation (EBD) is recognised treatment for symptomatic Crohn’s strictures. Several case series report its efficacy. A systematic analysis for overall efficacy can inform the design of future studies. This chapter aims to examine the pooled incidence of surgery, technical (TR), clinical response (CR) and complications following EBD for Crohn’s strictures in adults. Stricture characteristics were also explored. A systematic search strategy of COCHRANE, MEDLINE and EMBASE was performed. All original studies reporting outcomes of EBD for Crohn’s strictures were included. Surgical response was defined as patients with recurrent symptoms of a Crohn’s stricture following balloon dilatation requiring surgical intervention. CR was defined as obstructive symptom-free outcome at the end of follow-up, TR as post-dilatation passage of the endoscope through a stricture, and adverse event as the presence of complication (perforation and/or bleeding). Pooled event rates across studies were expressed with summative statistics. 25 studies included 1089 patients and 2664 dilatations. The cumulative surgery rate at 5-year follow-up was 75%. The proportion of patients who had further surgery at 1, 2- and 5-year follow-up was 21.5% (153/712), 18.7%

(118/632), and 8.4% (38/455) respectively. Pooled event rates for surgery, SR, TR, complications and perforations were 20.2%, 70.2% (95% CI: 60-78.8%), 90.6% (95% CI: 87.8-

92.8%), 6.4% (95% CI: 5.0-8.2) and 3% (95% CI: 2.2-4.0%), respectively. Outcomes between

2 stricture types were no different on sub-group meta-analysis. Efficacy and complication rates for EBD were higher than previously reported. Surgery is still a long-term inevitable outcome for the majority of patients undergoing EBD. Future studies are needed to determine whether

EBD has significant long-term benefits.

109

Improving the Quality of Care for patients with Ileal Crohn’s disease undergoing Surgery

5.2. Aims

This systematic review was performed to describe the outcomes of endoscopic balloon dilatation for Crohn’s disease strictures. The primary aim was to examine the pooled incidence of a surgical, technical, clinical response and complications following endoscopic balloon dilatation for Crohn’s strictures in adults. The secondary aim was to explore the impact of stricture characteristics on outcomes.

5.3. Methodology

5.3.1. Protocol and registration

The protocol for this study was registered on PROSPERO (CRD42015015758).

5.3.2. Eligibility criteria

All original studies, from 1991 to October 2014, reporting outcomes of endoscopic balloon dilatation for Crohn’s disease intestinal strictures in the adult population (age ≥ 18) were included in the review. Randomised controlled trials, observational reports and case series with sample size more than 10 were all included. Case reports, studies reporting on multiple diagnoses and conference proceedings were excluded from the review. Patients undergoing double balloon dilatation for deep seated intestinal strictures and children (age < 18) were more likely to require a general anaesthesia for the required intervention. Studies reporting exclusively on these were also excluded.

110

Improving the Quality of Care for patients with Ileal Crohn’s disease undergoing Surgery

5.3.3. Information sources

A three-step search strategy was employed. Initially a limited search was performed using

PUBMED to identify keywords and index terms contained in the title or abstract. The second step involved an extensive search using all identified keywords and extensive terms. Studies were identified by searching the following databases: COCHRANE, MEDLINE and

EMBASE.

5.3.4. Search

The final search terms were ("Crohn’s Disease" OR Crohn’s) AND (strictu* OR “Montreal

B2”) AND (endoscopy OR endoscopic OR ileocolonoscopy OR ileoscopic OR colonoscopy

OR colonoscopic) AND (“balloon therapy” OR “balloon dilatation” OR balloon dilation” OR dilatation OR “balloon strictureplasty”). The final step was a hand search of reference lists and bibliographies from previously retrieved studies to identify further relevant trials.

5.3.5. Data collection process

The first reviewer screened the titles and abstracts that were identified in the search strategy.

The papers were then evaluated by two reviewers according to the eligibility criteria outlined above. Discrepancies were resolved by consensus between the two reviewers. Data from selected studies were extracted by the first reviewer and this was followed by a further, unblended, check by the second reviewer. Extracted data was entered into an Excel

(Microsoft® software) database.

111

Improving the Quality of Care for patients with Ileal Crohn’s disease undergoing Surgery

5.3.6. Data items

The following variables were extracted: study demographics (year and country of publication, study design, and sample size), nature of the stricture (stricture characteristics including location, activity as active or quiescent fibrotic, type as de novo or anastomotic, length and diameter), pre-operative radiographic assessment, intervention technique (dilatation time, balloon dilator size and endoscopic accessibility), follow-up time period and outcome measures

(symptomatic response, technical response, overall complication and perforation rates).

5.3.7. Risk of bias in individual studies

The quality of studies was assessed by using the Newcastle-Ottawa Scale. The quality of studies was evaluated by examining three items: patient selection, comparability and outcome

(table 5.1).

Author, year Selection Comparability Outcome Score Ajlouni Y 2006 (221) *** - ** *** Atreja 2014 (222) *** - ** ** Bhalme 2013 (223) *** - ** **** Blomberg 1991(224) ** - * **** Breysem 1992(225) * - * ***** Brooker 2003(226) **** - ** ****** Couckuyt 1995 (227) *** - * ***** De Angelis 2013 (228) *** - ** **** East J.E. 2007 (229) ** - * **** Endo 2013 (230) *** - ** **** Ferlitsch 2006 (231) *** - * ***** Foster 2008 (232) *** - * **** Gustavsson 2012 (233) *** - ** **** Honzawa 2013 (234) *** - ** ***** Hunter 2001 (235) *** - ** ***** Mueller 2010 (236) *** - ** ***** Nanda 2013 (237) *** - ** *****

112

Improving the Quality of Care for patients with Ileal Crohn’s disease undergoing Surgery

Ramboer 1995 (238) ** - ** ***** Sabate 2003 (239) *** - * ***** Scimea 2011(240) *** - ** ***** Singh 2005(241) *** - * ***** Stienecker 2009 (242) ** - ** ***** Thomas-Gibson 2003 (243) *** - ** *** Van Assche 2010 (244) *** - ** **** Williams A.J.K. 1991 (245) *** - * ***** Table 5. 1 Quality assessment of studies using the Newcastle-Ottawa scale – The maximum number of stars each study can receive is 9 (maximum 4 for selection, 2 for comparison and 3 for outcome. The median number of stars across studes was 5 (range: 2 – 6).

5.3.8. Summary measures

Surgical response was defined as patients with recurrent symptoms of a Crohn’s stricture following balloon dilatation requiring surgical intervention. Symptomatic response was defined as patients with an obstructive symptom-free outcome at the end of follow-up, technical response by the passage of the scope following endoscopic balloon dilatation and adverse events by the proportion of patients who develop complications. Outcomes are expressed as pooled event rates (with 95% confidence interval limits), or as unweighted proportions of the size of the population studied.

5.3.9. Synthesis of results

Continuous numerical data are expressed as means (with standard deviations) or as medians

(with range values). A per patient analysis was used to determine the cumulative proportion of patients within a group and expressed as crude, unweighted proportions and percentages. A per study analysis was used to assess pooled event rates across studies. The random effects model was used, and results were expressed with forest plots and summative statistics.

113

Improving the Quality of Care for patients with Ileal Crohn’s disease undergoing Surgery

5.3.10. Risk of bias across studies

Heterogeneity across studies was assessed visually with forest plots and numerically (I2 < 25% indicates low heterogeneity). Evidence of publication bias was assessed visually using funnel plots. Comprehensive Meta-analysis (CMA, Biostat, Inc.) programme was used. To assess the relationship of continuous variables on outcome, the pooled event rate for each outcome was transformed into a dichotomous form (less than or greater than the stated pooled event rate).

5.3.11. Additional analyses

To determine association between stricture characteristics and outcome subgroup analyses were performed. The pooled event rates and 95% confidence interval were expressed per outcome for each categorical variable (e.g. balloon diameter, duration of inflation, geography and pre-interventional imaging). To compare the effect of the proportion of patients within each group (e.g. stricture activity and stricture type) on outcome, sub-group meta-analyses were performed, and pooled group differences were summarised by the pooled relative risk and corresponding confidence intervals.

5.4. Results

5.4.1. Study selection

Figure 5.1 details the study selection flow chart.

114

Improving the Quality of Care for patients with Ileal Crohn’s disease undergoing Surgery

Figure 5. 1 Flow chart demonstrating the search strategy in accordance with PRISMA – Two hundred and three (n = 203) records were identified following duplicate removal. Fifty-three

(n = 53) records were removed after limits were applied. One hundred and fifty (n = 150) records underwent screening and one hundred and twenty records were excluded (n = 120).

Thirty records (n = 30) were assessed for eligibility and twenty-five (n = 25) articles were included in our quantitative analysis.

115

Improving the Quality of Care for patients with Ileal Crohn’s disease undergoing Surgery

Two hundred and three studies were identified following both the initial and secondary database search. Studies were screened according to the above eligibility criteria and 30 studies were included as part of a full text review. A total of 25 studies were included in the final review (table 5.2).

116

Improving the Quality of Care for patients with Ileal Crohn’s disease undergoing Surgery

Surgical Technical Complication Study Number of Number of Study Author Country Population Size Response Response Occurrence Design Strictures (S) Dilatations (D) Outcome Occurrence (S) Occurrence(E) (C) Ajlouni Y Australia R 37 83 113 2 75 s 1 E, C 2006 (221) Atreja 2014 USA R 128 169 430 42 154 s 4 E, C (222) Bhalme 2013 United Kingdom R 79 93 191 18 75 3 E, S, C (223) Blomberg Sweden P 27 ns ns 1 ns 3 S, C 1991(224) Breysem Belgium P 18 20 24 5 16 0 E, S, C 1992(225) Brooker United Kingdom R 14 14 26 3 ns 0 S, C 2003(226) Couckuyt Belgium P 55 59 78 11 70 d 6 E, S, C 1995 (227) De Angelis France R 26 27 46 2 46 d - E, S 2013 (228) East J.E. United Kingdom RCT 13 ns ns - 12 - E 2007(229) Endo Japan P 30 47 83 14 s 154 s 5 s E, C 2013(230) Ferlitsch Austria P 46 ns 73 11 ns 3 d C 2006(231) Foster USA R 24 29 71 2 ns 2 S, C 2008(232) Gustavsson Sweden R 125 ns 594 22 533 d 41 d E, C 2012(233) Honzawa Japan R 25 29 83 6 22 - E 2013(234) Hunter United Kingdom R 22 ns 71 6 ns 0 S, C 2001(235)

117

Improving the Quality of Care for patients with Ileal Crohn’s disease undergoing Surgery

Mueller Germany P 55 74 93 10 52 1 E, S, C 2010(236) Nanda Ireland P 31 ns 55 17 55 d 0 E, S, C 2013(237) Ramboer Belgium P 13 15 53 2 ns 0 S, C 1995(238) Sabate France R 38 41 53 11 32 E, S 2003(239) Scimea Italy P 37 39 72 2 31 0 E, S, C 2011(240) Singh USA R 17 20 29 1 28 d 4 E, S, C 2005(241) Stienecker Germany P 25 31 50 3 24 - E 2009(242) Thomas- Gibson United Kingdom R 59 ns 124 - 101 d 8 d E, S, C 2003(243) Van Assche Belgium R 138 ns 237 - 134 12 d E, S, C 2010(244) Williams A.J.K. 1991 United Kingdom R 7 ns 15 2 5 1 E, C (245) Total 1089 790 2664 179 /849

Pooled unweighted event rate (%) (21.1%) 403 / 435 (93%) 25/564 (4%) RCT = Randomised Control Trials; R = Retrospective; P = Prospective; ns = not specified Study Outcome Measures: S = Surgical response, E = Endoscopic / Technical response, C = Complication, s = number of strictures, d = number of dilatations. Table 5. 2 Study demographics, frequency of population size, strictures and dilatations. Study demographics, frequency of population size,

strictures and dilatations - Description of studies included in the literature review. A total of 1089 subjects were described in the literature with

790 strictures undergoing 2664 interventions. Twenty-one studies (n = 179) reported surgical outcome per patient and one study per stricture

118

Improving the Quality of Care for patients with Ileal Crohn’s disease undergoing Surgery population (n = 14 strictures, n = 30 patients). A total of 179 (21.1%) patients reported surgery with dilatation. Ten studies reported technical response per patient. The number of patients reporting a technical response is shown. A total of 403 (93%) reported technical response with dilatation. Fifteen studies reported the occurrence of a complication following dilatation. The number of patients reporting complications in each study is shown. A total of 25 (4%) patients reported perforation following dilatation. For all cumulative analyses, studies reporting on the number of strictures dilated (s) and the number of dilatations performed (d) were excluded.

119

Improving the Quality of Care for patients with Ileal Crohn’s disease undergoing Surgery

They included 10 prospective studies (n = 10), 14 retrospective studies (n = 14) and 1 randomised control trial (n = 1) (221-245). Publication dates ranged from 1991 to October

2014 and originated from European (223-229, 231, 233, 235-240, 242-245) (6 studies from the United Kingdom) (223, 226, 229, 235, 243, 245), (222, 232, 241), Japanese (230, 234), and

Australian (221) institutions.

5.4.2. Study characteristics

The cumulative data for the 25 studies (196-220) included 1089 patients, 790 strictures and

2664 dilatations. Fifty-one percent (557/1089) were females, 43% (466/1089) were males and for the remaining 8% (66/1089) gender was unspecified. The median age at first dilatation reported across 17 studies (determined from the mean age at first dilatation per study) was 41.1

(range = 32.5 - 50) (222, 224-228, 230, 232, 234, 235, 237-242, 244). Surgical response was reported as the outcome measure in 5 studies (224, 226, 231, 232, 235), technical response in

3 studies (229, 243, 244), and both surgical and technical response in 11 studies (table 5.1)

(221-223, 225, 227, 228, 230, 233, 234, 236, 237, 239-242, 245).

Follow-up duration was reported in 24 studies with wide variation (221-225, 227-236). The median maximum follow-up time period was 83.5 months (range 12-172). The median minimum follow-up time period was 4 months (range 0 – 84).

Nine studies (n = 468) did not report number of strictures per patient. All but 14 of the remaining 621 cases from 16 studies, had >1 stricture documented. Fifteen studies examined lower gastrointestinal strictures only (224, 229, 231, 233, 235, 237, 243, 244) while ten

120

Improving the Quality of Care for patients with Ileal Crohn’s disease undergoing Surgery included both upper and lower GI strictures. The proportion of upper GI strictures was 3.8%

(though due to missing data in 9 studies this was an estimation).

5.4.2.1. Stricture activity

The proportion of patients with active (223-227, 238, 239) and quiescent (223-227, 230, 231,

238, 239) strictures was 44.9% (155/345) and 47.2% (151/320) respectively. In the majority of cases 82.2% (447/544) data relating to stricture activity was not reported (223, 225, 226,

233-237, 240, 244).

5.4.2.2. Stricture type

Across thirteen studies (n = 565) (223-227, 231, 235, 237-239, 242-244), most patients (79.1%;

447/565) had anastomotic strictures and 19.6% (111/565) had de novo strictures (223-227, 231,

235, 237-239, 242-244).

5.4.2.3. Intervention technique

A maximum balloon diameter was reported across all 25 studies: 18mm (225, 228, 235-238,

242-244), 20mm (221-223, 226, 229-232, 234, 240, 241, 245), and 25mm (224, 227, 233, 239).

There was variation in the maximum inflation time across 23 studies with maximal inflation periods of one (223, 234, 236), two (221, 225, 227, 228, 232, 238-245), three (229, 230, 235), four (224, 233, 237), and five minutes (231).

121

Improving the Quality of Care for patients with Ileal Crohn’s disease undergoing Surgery

5.4.2.4. Imaging

The majority of studies describe pre-interventional imaging (18/20; 90%) (222, 223, 226-231,

233, 235-237, 239-244) and two studies did not (2/20; 10%) (221, 232). The median maximum length of strictures reported across 21 studies was 7cm (range 3 - 25cm).

5.4.3. Synthesis of results

5.4.3.1. Surgery

Surgery was required for one or more of the following two events: (i) inaccessible strictures during endoscopy and (ii) persistent or recurrent symptoms i.e. failed repeated dilatation.

(i) Endoscopic inaccessibility

Seven studies reported 12.9% of cases (33/256) where endoscopic balloon dilatation could not be completed during endoscopy either because the stricture was too narrow or there was acute angulation (225, 227, 231, 236, 239, 240, 245). Two studies (n = 130) reported this event 6.9%

(9/130) by number of strictures (221, 230).

(ii) Symptomatic disease

Twenty one studies reported surgical outcomes for ongoing recurrent disease despite repeated balloon dilatation according to the number of patients (n = 849) (221-228, 231-242, 245). The mean length of follow-up from studies where surgery rates are reported was 19.7 months. The proportion of patients who underwent surgery within a 5 year follow-up was 75% (341/455).

The proportion of patients who had further surgery at 1, 2 and 5 year follow-up was 21.5%

(153/712), 18.7% (118/632), and 8.4% (38/455) respectively. Pooled data demonstrated a

122

Improving the Quality of Care for patients with Ileal Crohn’s disease undergoing Surgery mean surgical event rate of 20.2% (95% CI: 15.7-25.6), with low heterogeneity demonstrated across studies (I2: 13.2%) (figure 5.2).

Figure 5. 2 Forrest plot reporting on the rate of surgical intervention in the event of a failed clinical outcome – A random effects model demonstrating a pooled surgical intervention rate of 20.2% (95% CI: 15.7-25.6; I2: 13.2%) across reported outcomes expressed according to the number of patients (221-228, 231-242, 245).

123

Improving the Quality of Care for patients with Ileal Crohn’s disease undergoing Surgery

One study reported the requirement of surgery according to the number of strictures (n = 47)

(230). The proportion of strictures requiring surgery in this study was 29.8% (14/47).

124

Improving the Quality of Care for patients with Ileal Crohn’s disease undergoing Surgery

The relationship between a surgical outcome and variables is shown in table 5.3.

Variable Surgical response Symptomatic response Technical response Perforation Categorical variables represented as pooled event rates % (95% Confidence Intervals) 18 22.4 (12.6 – 26.8) 61.9 (47.4 – 74.4) 94.7 (87-97.9) 3.1 (1.9-5.1) Balloon diameter (millimetres) 20 20.3 (14.5 – 27.6) 79.5 (72.6-85) 90.6 (87.4-93.0) 3.8 (2.2-6.5) 25 19.4 (12.7 – 28.6) 71 (48 – 86.6) 89.4 (86.9-91.5) 2.2 (1.3-3.6) Duration of 2 16.8 (12.2 – 22.7) 70.6 (58.1-80.6) 92.2 (86.6-95.5) 3.1 (1.8-5.3) inflation (minutes) 5 25.8 (15.6 – 39.6) 67.1 (41.9 – 85.2) 92.6 (84.2-96.7) 3.6 (1.6-8.2) European 21.1 (16.1 – 27.2) 67.9 (57-77.2) 90.6 (86.7-93.5) 2.3 (1.6-3.3) North Geography 14.8 (3.9 – 43.0) 84.5 (62.6 – 94.7) 91.6 (86.8-94.8) 5.0 (1.3-17.7) American Japanese 91.3 (81.9-96.0) Pre-interventional Reported 22.7 (17.6 – 28.8) 70.4 (58.8 – 79.8) 92.3 (85.8-95.9) 2.7 (1.8-4.0) imaging Table 5. 3 The relationship between response rates (symptomatic and technical) and adverse event rates (perforation) with categorical variables -

The categorical variables (balloon diameter, duration of inflation, geography and pre-interventional imaging) are presented as pooled event rates across studies alongside their 95% confidence intervals.

125

Improving the Quality of Care for patients with Ileal Crohn’s disease undergoing Surgery

Balloon inflation diameters of 18mm (225, 228, 235-238, 242-244), 20mm (221-223, 226, 229-

232, 234, 240, 241, 245), and 25mm (224, 227, 233, 239) demonstrated a pooled surgical outcome rate of 22.4% (95% CI: 12.6 – 26.8%), 20.3% (95% CI: 14.5 – 27.6%) and 19.4%

(95% CI: 12.7 – 28.6%), respectively. An inflation time of up to 2 minutes (223, 225, 227,

228, 232, 236, 238, 240, 241) and 5 minutes (224, 230, 231, 233, 235, 237) demonstrated a pooled surgical outcome rate 16.8% (95% CI: 12.2 – 22.7%), and 25.8% (95% CI: 15.6 –

39.6%) respectively. The pooled surgical outcome rate across 16 European studies was 21.1%

(95% CI: 16.1 – 27.2%). The pooled surgical outcome rate across three North American studies was 14.8% (95% CI: 3.9 – 43.0%) (222, 232, 241). The use of pre-interventional imaging was described across 14 studies where the pooled surgical outcome rate was 22.7% (95% CI: 17.6

– 28.8%) (222, 223, 226-228, 230, 231, 233, 235-237, 239-242).

One study (225) reported subgroup data on stricture activity for surgery. Sub-group meta- analysis demonstrated no evidence of a difference of active versus quiescent (reference) strictures [RR2.0, 95% CI: 0.5 – 8.0, p = 0.33] (table 5.4).

126

Improving the Quality of Care for patients with Ileal Crohn’s disease undergoing Surgery

Outcome Number Heterogeneity Group difference (*) studies P-value I2 RR (95% CI) P-value Technical response 3 0.80 0% 0.98 (0.69, 1.38) 0.89 Symptomatic response 3 0.71 0% 0.91 (0.63, 1.32) 0.71 Complications 0 - - - - Perforation 0 - - - - Surgery 1 - - 2.00 (0.50, 8.00) 0.33 (*) Pooled group differences reported as outcome for active group relative to quiescent group

Table 5. 4 Meta-analysis in studies reporting on outcomes for active and quiescent strictures

Eight studies (221, 222, 225, 232, 234, 240, 241, 243) reported subgroup data on stricture type

for surgery with minimal evidence of heterogeneity across studies [I2 15%]. Sub-group meta-

analysis demonstrated no evidence of a difference of anastomotic versus de novo (reference)

strictures [RR1.1, 95% CI 0.77 – 1.5, p 0.61] (table 5.5).

Outcome Number Heterogeneity Group difference (*) studies P-value I2 RR (95% CI) P-value Technical response 7 0.11 43% 1.01 (0.92, 1.11) 0.81 Symptomatic 10 0.72 0% 1.09 (0.96, 1.25) 0.18 response Complications 1 - - 1.83 (0.77, 4.32) 0.17 Perforation 4 0.41 0% 0.87 (0.29, 2.60) 0.80 Surgery 8 0.31 15% 1.09 (0.77, 1.50) 0.61 (*) Pooled group differences reported as outcome for the anastomotic group relative to the de novo group. Table 5. 5 Meta-analysis in studies reporting on outcomes for anastomotic and de novo

strictures

127

Improving the Quality of Care for patients with Ileal Crohn’s disease undergoing Surgery

5.4.3.2. Symptomatic response

The population sample size across all 16 studies reporting on symptomatic response was 653.

Fifteen studies reported on symptomatic response for patient numbers (n=615) (223-228, 232,

235-238, 240, 241, 243, 244) which was 63.9% (393/615) (table 5.2). Fourteen studies were from Europe (612/653 [93.7%]) spanning 7 different countries (four from the UK) (223-228,

235-240, 243, 244) and 2 were North American studies (41/653 [6.3%]) (232, 241). The proportion of females and males was 51.8% (338/653) and 39.2% (256/653) respectively. One study did not report on gender proportions (n=59) (243). The median age at first dilatation reported across 13 studies reporting on symptomatic response (determined from the mean age at first dilatation per study) was 42.6 (range 33.7-50) (199-203, 207, 210, 212-216, 219). The proportion of patients who require further dilatation at 1, 2- and 5-year follow-up was 31.6%

(160/506), 25.9% (117/451) and 1.7% (5/299). The cumulative proportion of patients requiring further dilatation over 5 years was 80.6% (241/299).

Analysis of pooled study outcomes demonstrated a symptomatic response rate of 70.2% [95%

CI: 60-78.8%] with evidence of moderate to high heterogeneity between studies [I2: 63.8%]

(figure 5.5).

128

Improving the Quality of Care for patients with Ileal Crohn’s disease undergoing Surgery

Figure 5. 3 Forest plot for studies reporting on symptomatic response. (excluding Sabate et al

(239) – Random effects model demonstrating a pooled event rate for symptomatic response as

70.2% (95% CI: 60.0 – 78.8%; I2: 63.8%).

The relationship between symptomatic response and variables are shown in tables 5.3, 5.4 and

5.5. Three studies (225, 226, 241) reported subgroup data on stricture activity for symptomatic response with no evidence of heterogeneity across studies [I2 0%]. Sub-group meta-analysis demonstrated no evidence of a difference of active versus quiescent (reference) strictures

[RR0.91, 95% CI: 0.6 – 1.3, p = 0.7] (table 5.4). Ten studies (222, 225, 230, 232, 238, 240-

243, 245) reported subgroup data on stricture type for symptomatic response with no evidence of heterogeneity across studies [I2 0%]. Sub-group meta-analysis demonstrated no evidence of a difference of anastomotic versus de novo (reference) strictures [RR 1.1, 95% CI 0.96 – 1.2, p 0.2] (table 5.5).

129

Improving the Quality of Care for patients with Ileal Crohn’s disease undergoing Surgery

Sensitivity analysis

In view of heterogeneous data, a sensitivity analysis was performed. Visual inspection of the funnel plot (figure 5.6) demonstrated six outlier publications, (225, 228, 232, 237, 243, 244) and evidence of asymmetry indicating overestimation of the effect size as well as the possibility of publication bias. The symptomatic response rate in one study was measured according to the number of dilatations done and was reported as 89% (47/53) (239).

Figure 5. 4 Funnel plot for studies reporting on symptomatic response rate - Heterogeneity across studies was demonstrated (I2: 63.8%). Six studies (225, 228, 232, 237, 243, 244) were outliers and determined as sources for publication bias on sensitivity analysis. Evidence of asymmetry in the funnel plot was also present.

130

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

5.4.3.3. Technical response

A total of 19 studies reported on technical response as an outcome measure. Analysis for pooled study results demonstrated an event rate of 90.6% (95% CI: 87.8-92.8%) with low heterogeneity between studies (I2: 11.7%) (figure 5.5).

Figure 5. 5 Forest plot for studies reporting on technical response – A random effects model demonstrating a pooled technical response event rate of 90.6% (95% CI: 87.8 – 92.8%; I2:

11.7%) across 19 studies.

Fourteen studies originated from Europe across seven different countries. The median age at first dilatation reported across 12 studies was 40.1 years (range 32-49). The proportion of females and males were 49.8% (470/943) and 43.2% (407/943) respectively.

131

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

Reported outcomes were expressed either for patients (n = 435) (223, 225, 229, 234, 236, 239,

240, 242, 244, 245), stricture (n = 299) (221, 222, 230), and intervention/dilatation (n = 926)

(table 5.2) (227, 228, 233, 237, 243, 246). The proportion of patients demonstrating technical response was 92.6% (403/435). The pooled event analysis results were similar due to low heterogeneity.

The relationship between technical response rates and variables are shown in tables 5.3, 5.4 and 5.5. Three studies (225, 226, 241) reported subgroup data for active and quiescent strictures for technical response with no evidence of heterogeneity across studies [I2 0%]. Sub- group meta-analysis demonstrated no evidence of a difference of active versus quiescent

(reference) strictures [RR 0.98, 95% CI: 0.7 – 1.4, p = 0.89] (table 5.4). Seven studies (222,

225, 228, 232, 241, 242, 245) reported subgroup data on stricture type for technical response with moderate evidence of heterogeneity across studies [I2 43%]. Sub-group meta-analysis demonstrated no evidence of a difference of anastomotic versus de novo (reference) strictures

[RR 1.0, 95% CI 0.92 – 1.1, p 0.81] (table 5.5).

5.4.3.4. Complications and perforations

Fifteen studies reported complications according to the number of patients (n = 564) (221-227,

232, 235-238, 240, 241, 245) and four studies reported complications based on the number of dilatations (n = 1228) (231, 233, 243, 244). The proportion of patients with complications was

132

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

4.4% (25/564) (table 5.1). Analysis of pooled study data demonstrated an overall complication rate of 6.4% (95% CI: 5.0 – 8.2; I2: 4.0%).

Eighteen studies reported on perforation for patients (n = 654) (221-228, 234-242, 245) and four studies for number of dilatations (n = 1281) (231-233, 243, 244). The proportion of patients who had perforation was 2.4% (16/654). Study data analysis showed no heterogeneity across studies with a pooled perforation rate of 3% (95% CI: 2.2- 4.0%; I2: 0%) (figure 5.6).

Figure 5. 6 Forrest plot reporting on perforation rates– A random effects model demonstrating a pooled perforation rate of 3% (95% CI: 2.2-4.0%; I2: 0%) across 22 studies with reported

133

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery outcomes expressed according to number of patients (221-228, 234-242, 245), balloon dilatation (231-233, 243, 244) and strictures (230).

The relationship between perforation rates and variables are shown in tables 5.3, 5.4 and 5.5.

Balloon inflation diameters of 18mm (225, 228, 234-238, 242-244), 20mm (221-223, 226, 231,

234, 240, 241, 245), and 25mm (224, 227, 233, 238) demonstrated pooled perforation rates of

3.1% (95% CI: 1.9 – 5.1%; I2 0%), 3.8% (95% CI: 2.2 – 6.5%; I2 0%), and 2.2% (95% CI: 1.3

– 3.6%; I2 0%) respectively. An inflation time of up to 2 minutes (223, 225, 227, 228, 232,

235-238, 240, 241, 243, 244) and 5 minutes (224, 230, 233, 236) demonstrated a pooled perforation rate of 3.1% (95% CI: 1.8 – 5.3%; I2 0%) and 3.6% (95% CI: 1.6 – 8.2%; I2 0%) respectively. The mean perforation rate across 18 European studies that reported perforation according to the number of patients was 2.3% (95% CI: 1.6 – 3.3%; I2 0%) (223-228, 231, 233,

235-240, 242-245). The mean perforation rate across three North American studies was 5.0%

(95% CI: 1.3 – 17.7%; I2 0%) (table 5.3) (222, 232, 241). The use of pre-interventional imaging was described across 14 studies (222, 223, 227, 228, 230, 233, 236, 237, 239-244) where the pooled perforation rate was 2.7% (95% CI: 1.8 – 4.0%; I2 0%). The perforation rate in one study that did not use pre-interventional imaging was 1.3% (95% CI: 0.1 – 17.8%) (221) (table

5.2). The median maximum stricture length reported across 20 studies was 7cm (range 2 –

25cm).

Four studies (232, 234, 241, 243) reported subgroup data on stricture type for perforation with no evidence of heterogeneity across studies [I2 0%]. Sub-group meta-analysis demonstrated

134

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery no evidence of a difference of anastomotic versus de novo (reference) strictures [RR 0.87, 95%

CI 0.29 – 2.6, p 0.8] (table 5.5).

5.5. Discussion

This is the most comprehensive systematic review of endoscopic balloon dilatation for the management of Crohn’s strictures to date.

5.5.1. Summary of evidence

The pooled event rate for a surgical outcome following endoscopic balloon dilatation for ongoing symptoms was 20.4%. This figure is lower than the 28.5% reported by a previous systematic review (70). Since the pooled summative effect addresses weight and heterogeneity between studies, it reflects a more accurate measure of efficacy albeit subject to bias.

Furthermore, the proportion of patients who underwent surgery within a 5-year follow-up was

75%. This suggests endoscopic balloon dilatation may be more efficacious as a short-term measure. The pooled event rate for symptomatic response of 70.2% includes 6 studies identified as outliers from the funnel plot. However, the distribution of the studies within the plot also suggests publication bias may be over-estimating the effect size. Furthermore, the source of heterogeneity across studies may also be secondary to differences in the intervention performed. In this review it was only reported in 24% of the studied population with a failure rate of 13%. Failure of endoscopic access is particularly relevant as most strictures undergoing dilatation are likely to be anastomotic and associated with adhesions and fibrosis. In contrast, a previous systematic review on endoscopic balloon dilation indicated 58 % response

135

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery comparable to our 63.9% expressed as proportion of patients but a lower value than the pooled effect (69).

The review reveals other relevant observations pertaining to the intervention. Where the studies examined both symptomatic and technical response, the former was consistently less than the technical rate of 90.6%. The inference is that passage of the endoscope through the stricture is an inadequate outcome for patients. There was a wide variation in dilation techniques suggesting an overriding need for standardisation of endoscopic procedures. Balloon diameter of 20mm seems to be commonest and most effective size limit, consistent with the internal small bowel diameter of 25mm. Two minutes of dilatation is the commonest duration used and may be associated with better outcomes. The commonest reported dilatation technique was the three-step technique increase in diameter with regular repeat dilatations until resolution of symptoms on a normal diet.

The study focused on an adult population with a mean age at first dilatation of 41 years, which reflects their aetiology as a complication of the disease or surgery. Moreover, sub-group analysis of disease activity and type of stricture did not reveal differences in outcomes for stricture characteristics. With respect to dilatation technique maximum balloon diameter and duration of inflation did not seem to show different outcomes, except that 20mm size was accompanied by higher symptomatic response rate than 18mm (80% vs 52%). There was no evidence of higher perforation rate with dilatation diameters of 25mm. North America symptomatic response rates were higher than Europe but were associated with higher complication rates.

136

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

Overall pooled complication rate was 6.4% [95% CI: 5.0 – 8.2], much higher than complication rates of 2% reported by Hassan (69), and 3% reported by Wilber (70). In contrast, the perforation rate, which represents the most significant complication, was 3% [95% CI: 2.2 –

4.0%] for pooled analysis and similar to that expressed as proportion of patient in this study

(2.6%). The previous reviews did not report perforation and a separate event (69, 70).

5.5.2. Limitations

Firstly, because of the absence of control groups for comparison, we created sub-groups to examine the effect of stricture characteristics on outcome. However, as the number of studies where results were consistently described for stricture types was small, the lack of effect may be related to a type 2 statistical error due to the exclusion of a large proportion of the data. The exclusion of 39 conference abstracts may have contributed to this effect, however the lack of methodological information would have made it difficult to assess the quality of data in this format. The second limitation was the diversity of the populations studied in terms of stricture characteristics, techniques and expression of results according to sample size, stricture numbers or number of interventions. This made comparisons across the studies difficult particularly for outcomes which expressed results according to sample size, stricture numbers and/or number of interventions. We used population size for primary outcomes of this review. Thirdly incomplete and variable reporting of some population and interventions characteristics means analyses was conducted on data that was available and may not be generalizable to other studies or populations. This limitation explains why the number of strictures was less than the sample number of the review (790 and 1089). Fourthly, most studies were reported by

137

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery gastroenterologists, with a bias towards showing endoscopic benefit through both performance and reporting bias. Also, none of the studies mention dietary restrictions on follow-up: low fibre diet will be associated with better and sustained response than a resumption to a full diet at the expense of quality of life.

138

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

CHAPTER 6: THE DEVELOPMENT AND VALIDATION OF AN EVIDENCE-

BASED, END-USER INFORMED RADIOLOGICAL STAGING TOOL FOR

ILEOCOLONIC CROHN’S DISEASE

6.1. Abstract

Consensus guidelines emphasise the importance of multidisciplinary team driven care for patients with complex Crohn’s disease. There are, however, no clear definitions of complex ileocolonic Crohn’s beyond the presence or absence of pre-operative intra-abdominal sepsis.

This chapter aims to develop a staging tool for complex ileocolonic Crohn’s disease based upon best evidence and end–user expert opinion. The secondary aim is to investigate validation and reliability evidence for this staging tool. A 4-stage tool was derived including items through literature review and semi-structured interviews of IBD experts: Stage 1, predominantly inflammatory ileal stricture; stage 2, predominantly fibrotic ileal stricture; stage 3, fistulating disease; and stage 4, intraabdominal abscess or collection. Intra-operative and histopathologically-assigned stages 1, 2, 3 and 4 were present in 41 [25.6%], 43 [26.9%], 52

[29.7%] and 24 [13.7%] patients, respectively. Validity testing confirmed a greater proportion of males (p=0.02), patients undergoing re-resectional surgery (p=0.01), patients with pre- operative anaemia (p<0.001), leukocytosis (p=0.03), thrombocytosis (p<0.001), a raised CRP

(p=0.002), emergent surgery (p=0.003), concomitant sigmoidectomy (p=0.03) and stoma formation (p<0.001), and IASC (p=0.02) with increasing stage, respectively. Eight four patients had pre-operative cross-sectional imaging. Good reliability was demonstrated between pre-operative radiological stages and intra-operative and histopathology assigned stages (R=0.8; p<0.001). This novel staging tool has been validated to show greater morbidity with increasing stage severity. It can reliably be used to pre-operatively stage patients and

139

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery inform surgical decision-making and estimate peri-operative morbidity. Further multi-center prospective validation of this tool is required for both clinical and research purposes.

6.2. Aims

The primary objective of this chapter is to devise a staging tool for ileocolonic Crohn’s disease based upon best evidence and end–user opinion. The secondary objective is to provide validation and reliability evidence for the staging tool against a cohort of patients who have undergone ileocolonic resection.

6.3. Methodology

A mixed-methodology approach was adopted using qualitative approaches for tool construction and quantitative approaches for validity and reliability testing, as per current best psychometric practice.

6.3.1. Tool construction

The primary objective of this study sought to establish key radiological features in ileocolonic

Crohn’s disease, and to determine if these features could be categorised by disease severity.

To achieve this objective, tool construction progressed through the identification of items and sub-divisions through literature review and expert end-user opinion to determine purpose and provisional categorisation of identified items.

140

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

6.3.1.1. Literature review

A systematic review of the literature was conducted (March to July 2013) to identify the current best evidence on radiological features of ileocolonic Crohn’s disease. The databases of

Medline and PubMed were searched using the following keywords and their combinations:

Radiolog*, Ultrasound, MRI, CT, Crohn’s, “Crohn’s Disease”, “Inflammatory Bowel

Disease”.

When combining the three search categories, 277 records were identified. Limits were applied

(English language, human subjects, all citations from 1980 to 2012) and 95 records were excluded. A title screen was performed, and a further 110 records were excluded. The remaining 72 records underwent a full text review and a further 64 records were excluded based on the following eligibility criteria:

Inclusion:

➢ Studies detailing radiological features relating to ileocolonic Crohn’s disease against a

surgical and/or histopathological gold standard in the adult population.

Exclusion:

➢ Review articles, case reports, editorials and letters

➢ Comparative studies (radiological gold standard)

➢ Sample size < 5

➢ Population age < 18

141

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

In total, 8 studies were included in the final evidence synthesis (figure 6.1).

Records identified through

database searching:

(n = 277) Identification

No of records excluded after limits applied: Studies in English, Human subjects, 1980 – 2012.

(n = 95)

Records screened: Records excluded:

(n = 182) (n = 110) Screening

Records excluded, with reasons: (n = 64) Records assessed for ➢ Incorrect study design: n=9

eligibility: (n = 72) ➢ No clear histological/intra- operative gold standard Eligibility comparator: n=47

➢ No data on radiological classification / features: n=3

➢ Incorrect population: n=5

Records included in qualitative synthesis:

(n = 8) Included

Figure 6. 1 Flow chart demonstrating the search strategy in accordance with PRISMA guidance

- Two hundred and seventy-seven (n=277) records were identified following duplicate removal. Ninety-five (n=95) records were removed after limits were applied. One hundred and eighty-two (n=182) records underwent screening and one hundred and ten records were

142

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery excluded (n=110). Seventy-two (n=72) were assessed for eligibility and eight (n=8) studies were included in our qualitative analysis.

6.3.1.2 Semi-structured interviews with end-user IBD experts

A comprehensive qualitative methodology was used for this section of the study and is detailed in section 2.2.2. Fourteen participants were purposively recruited from St Mark’s Hospital specialising in intestinal and colorectal disease. The participants consisted of 5 consultant colorectal surgeons (CS), 4 consultant radiologists (CR), 3 consultant gastroenterologists (CG) and 2 consultant histopathologists (CP) (median 11 [IQR 11 – 19] years in service). This unit was chosen for the basis of this study because of its internationally recognised academic track record in the field of IBD. Interviews were carried out using a standardised and previously piloted semi-structured interview protocol (Appendix i). The interview protocol explored key themes including the importance and purpose of a radiological staging system, radiological indicators of disease severity, and the constituents of potential stages. For qualitative data sampling and analytical methods used, see section 2.2.2.

6.3.2. Validity and reliability testing

6.3.2.1. Patient selection

The clinical case records identified following database screening were reviewed further to ensure the following inclusion criteria were met to meet the stated objectives for validity testing:

➢ All patients who underwent a one-stage resection and anastomosis for ileocolonic Crohn’s

disease.

143

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

➢ All patients who underwent a resection with ileostomy formation (with or without an

anastomosis) for ileocolonic Crohn’s disease.

In addition to this all patients who have had pre-operative magnetic resonance enterography or computer tomographic enterography were analysed for reliability testing.

6.2.2.2. Data extraction

The database of prospectively collected data on patients who underwent ileocolonic resection for Crohn’s disease was retrospectively reviewed to extract relevant pre-operative, peri- operative and post-operative data (see section 2.1.3 for variables extracted). For validity testing, stages were applied to each case following a review of the intra-operative and histopathological findings. For reliability testing, the pre-operative radiology of each patient was reviewed by two consultant gastrointestinal radiologists. Magnetic resonance enterography and computer tomographic enterography images were prospectively staged accordingly to the devised staging tool. Staging radiologists were blinded to the histopathological and intra-operative stages, and patient outcome to avoid bias.

6.3.2.2. Outcome measures

To evaluate validity of the staging tool, trend analyses against intra-operative and histopathologically assigned stages were assessed against the following outcome measures: the peri-operative and post-operative morbidity as defined by the formation of a stoma, concomitant surgery, the presence of an intra-abdominal abscess or an anastomotic leak following surgery, and the pre-operative pre-morbid condition of the patient as defined by the

144

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery pre-operative patient and disease related characteristics, immunosuppressant usage and blood parameters (as extracted in section 2.1.3). To evaluate the reliability of the staging tool, the pre-operative CTE and MRE images were assigned specific stages and a correlation analysis was performed against intra-operative and histopathological stages.

6.3.2.3. Statistical analyses

Section 2.1.3 provides a further detail on the assessment of data normality and subsequent reporting. Section 2.1.7 describes how missed data was handled. For validity testing, categorical variables were investigated with the chi square [χ2] test for trend. Kendall’s Tau- b correlation coefficient was used to analyse trends across median values with increasing intra- operative and histopathological stage. For reliability testing, Spearman's Rank-Order correlation coefficient was used to determine reliability of the pre-operative radiological stages against the intra-operative and histopathological stages. Correlation was demonstrated with a co-efficient ‘R’ of great than 0.6. Statistical significance was set at a p value of less than 0.05.

Statistical analyses were carried out with SPSS version 23.0 [SPSS, Chicago, IL, USA] and

GraphPad Prism version 7.00 for Windows [GraphPad Software, La Jolla California USA, www.graphpad.com].

6.4. Results

6.4.1. Literature review

A total of eight studies were included in this review, of which four reported exclusively on radiological features visible on MRE (247-250), one reported exclusively on radiological features visible on CTE (251), one reported exclusively on the radiological features visible on

145

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery ultrasound (252) and two studies reported on the radiological features visible on both CTE and

MRE (253, 254). Five studies were prospective (248, 251-254) and three were retrospective

(247, 249, 250). Publication dates ranged from 2004 to 2012. The gold standard comparator to determine diagnostic accuracy of each modality was intra-operative evaluation across six studies (247-249, 252-254) and histopathological surgical specimen across two studies (250,

251). Sensitivities for MRE, CTE and ultrasound modalities were 95%-81% (247, 253), 95%-

76% (251, 253, 254) and 99% (252) respectively. Specificities for MRE, CTE and ultrasound were 86%-82% (247, 253), 89%-70% (251, 253, 254) and 96% respectively.

Descriptions of mucosal, mural and extra-mural features were provided (n = number of studies). Mucosal changes visible on radiology included ulceration [n=6] (248, 249, 253, 254) and mucosal hyper-enhancement [n=6] (248-251, 253, 254). Mural changes visible on radiology included bowel wall thickening [n=9] (247-254), stenosis [n=8] (248, 249, 251-254), hyper-enhancement [n=7](248-251, 253, 254), stratification [n=1] (250) and skip lesions [n=1]

(251). Extra-mural changes visible on radiology included penetrating disease (abscess or fistula) [n=8] (248-254), creeping fat [n=5] (250, 251, 253, 254), Comb’s sign [n=5] (250, 251,

253, 254), fat stranding [n=1] (250), intra-peritoneal fluid [n=1] (248) and lymphadenopathy

[n=1] (250).

Bowel wall thickening, stenosis and penetrating disease (abscess or fistulae) were features identifiable on all modalities: ultrasonography (252), CTE (251, 253, 254), and MRE (247-

250, 253, 254).

146

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

Mucosal Mural Extra-mural

TI TI

-

fistula

ated vasa vasa ated

-

BWT

recta)

Settings

Stenosis

standard standard

peritoneal fluid peritoneal

segments)

-

Skip lesions Skip

Creeping fat Creeping

Stratification

Fatstranding

enhancement

Study Design Study

Author /Year Author

Mucosal hyper hyper Mucosal

Abscess +/ Abscess

Lymphadenopathy

Mucosal ulceration Mucosal

Intra

Subjects (TI /neo (TI Subjects

Comparison with gold gold with Comparison

Comb sign (dil sign Comb Mural hyper enhancement hyper Mural Parente F et al. 2004 Prospective, Intraoperative Italy 102 (82) ˗ ˗ ˗ α ˗ α ˗ α ˗ ˗ ˗ ˗ ˗ (252) comparative evaluation +/- ICS Histopathology Paparo et al. 2011 Prospective, Italy 22 (168) evaluation from ˗ ∞ ∞ ∞ ˗ ∞ ∞ ∞ ∞ ˗ ∞ ˗ ˗ (251) descriptive surgical specimens Prospective, Jensen M et al. 2011 Denmar Intraoperative blinded, 35 (34) ∞ § ∞ ∞ ∞ ˗ ∞ § ˗ ∞ ∞ ˗ ∞ ˗ ˗ (254) k evaluation +/- ICS multi-centre Multicenter, Intraoperative Jensen M et al. 2012 Denmar prospective, 21 evaluation +/- CE ∞ § ∞ ∞ § ∞ § ˗ ∞ § ˗ ∞ § ∞ § ˗ ∞ § ˗ ˗ (253) k blinded +/- ileocolonoscopy Histopathologic Retrospectiv evaluation from Oto A et al 2011 (250) USA e, single 18 (18) surgical or ˗ § § § § ˗ ˗ § § § § ˗ § centre endoscopic specimens Retrospectiv Surgical +/- Oto A. et al. 2009 e, endoscopic USA 11 (53) ˗ ˗ ˗ § ˗ ˗ ˗ ˗ ˗ ˗ ˗ ˗ ˗ (247) comparative, evaluation + evaluation histopathology Retrospectiv Intraoperative or Negaard et al 2008 Norway e, 35 endoscopic § § § § ˗ § ˗ § ˗ ˗ ˗ ˗ ˗ (249) comparative evaluation 147

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

Prospective, Intraoperative or Negaard et al. 2007 Norway randomized 40 endoscopic § § § § ˗ § ˗ § ˗ ˗ ˗ § ˗ (248) comparative evaluation Total 6 6 7 9 1 8 1 8 5 1 5 1 1 ICS = ileocolonoscopy; CE = capsule endoscopy; BWT = bowel wall thickening; TI = terminal ileal; α = ultrasonographic features; ∞ = CTE features; § = MRE features

Table 6. 1 Description of studies included in the literature review - A total number of eight studies reported on radiological features visible on

ultrasound (n=1), CTE (n=3) and MRE (n=6). Studies reported on 13 radiological features detailing mucosal (n=2), mural (n=5) and extra-mural

(n=6) changes. Two mural (bowel wall thickening [n=9] and stenosis [n=8]) and one extra-mural feature (penetrating disease [n=8]) were common

to all three radiological modalities (highlighted grey).

148

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

6.4.2. Semi-structured interviews

Thematic saturation was achieved after 14 semi-structured interviews (four consultant radiologists, five consultant colorectal surgeons, three consultant gastroenterologists, and two consultant histopathologists). All interviews took place face-to-face from March 2013 to July

2013. The themes extracted for the interview transcripts are reported in sections 6.4.2.1 and

6.4.2.2 below (n = number of interviewees); verbatim quotes illustrate each theme.

6.4.2.1. Role and purpose of radiology and a radiological staging system for ileocolonic

Crohn’s disease

Participants (n = number of interviewees) reported that radiology plays an important role in the setting of ileocolonic Crohn’s disease (n=12), particularly in the setting of initial diagnoses and disease mapping (n=7), disease monitoring (n=4) and guiding management (n=3). One participant highlighted the limitations of endoscopic assessment from the viewpoint of disease mapping: “endoscopically traditionally you can get to a lesion from below and may be able to confirm that there’s ileocecal disease without colonic disease but the small bowel, so from the proximal end it’s more difficult and Crohn’s, being a pan-enteric condition, you need to isolate that” (CG1).

Many participants agreed there is a requirement for a radiological staging system (n=12). In particular it would allow uniformity of treatment for specific Crohn’s pathology and research

(n=9): “What one’s really trying to work out is what things influence practice, what things influence outcome, what things influence the need for adjuvant therapy … and if you have those

149

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery things then knowing about them, which is staging, then becomes really useful” (CS3), “it could also help enormously in terms of research, in terms of actually homogenising groups that you’re actually studying. Currently I think you have everybody under one umbrella of ileocaecal Crohn’s disease and that houses very, very mild disease to very, very severe disease” (CS1).

6.4.2.2. Determining severity (figure 6.2)

Early disease Disease indicating primarily surgical therapy • localised isolated sites (n=5) • abscess formation (n=5) ¥ • mild inflammatory change • fistulation (n=4) ¥ (n=5) * • perforation (n=3) ¥ • mucosal ulceration (n=4) • limited length of disease (n=1)

Disease indicating primarily Advanced disease medical therapy • penetrating disease (n=11) • inflammatory changes ¥ alone (n=4) * • abscess/perforation • the absence of penetrating formation (n=11) ¥ disease or obstruction • fistulae (n=8) ¥ (n=3) • stricturing disease with • long segments of disease obstruction (n=10) (n=1)

Figure 6. 2 – End-user opinion demonstrating items (n = number of interviewees) encompassing features consistent with the dichotomy of early and advanced disease (solid

150

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery outline boxes) and features requiring surgical versus medication therapy (dashed outlined boxes) - High ranking items are in bold italics. Items with * are common to both early disease and disease requiring medical therapy as a primary indication. Items with ¥ are common to both advanced disease and disease requiring surgical therapy as a primary indication.

Items (n = number of interviewees) demonstrating advanced disease included penetrating disease (n=11) such as localised perforation (n=4), abscess formation (n=7) and fistulae (n=8), with the presence of: “an abscess, that’s a poor prognostic sign in that you might well need a stoma temporarily” (CS2). Stricturing disease with obstruction was also suggestive of advanced disease (n=10). Items requiring surgery as a primary intervention included abscess formation (n=5), fistulation (n=4), localised perforation (n=3), limited length of disease (n=1) and active disease causing obstruction (n=2): “… as a general rule if the bowel is perforated you’re more likely to operate…when there is fistulation you’re more likely to operate… when there is obstruction, i.e. radiological dilatation… you will go towards (an) operation” (CS3).

Items (n = number of interviewees) detailing early disease included localised isolated sites

(n=5), mild inflammatory change (n=5), and mucosal ulceration (n=4): “a small amount of inflammation involving a small segment of ileum, and at its minimalistic point simply an ileitis”

(CP1) or “Simple short stricture without abscess or fistula” (CS2). Items that were identified with medical therapy as the primary form of treatment include inflammatory changes alone

(n=4), the absence of penetrating disease or obstruction (n=3), and long segments of disease

(n=1): “… the absence of sepsis …you could have all of those factors and it would really come down to luminal narrowing…it would be for medical treatment” (CS1), and “relatively short

151

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery area of inflammation that isn’t causing any complications, and complications I mean associated sepsis, stricturing causing obstructive features on radiology in this case, or fistulation” (CG1).

6.4.3. Tool construction

The emergent staging tool consists of four stages, with each stage mirroring the core radiological features as identified from the evidence (literature review) and populated through expert end-user opinion (semi-structured interviews) (table 6.2)

Stages (populated from end-user opinion)

1 2 3 4 Bowel Ileitis or localised wall inflammation with ˗ ˗ ˗ thickening mild wall Core (247-254) thickening (n=8) components Fibrosis with Stenosis Fibrosis with derived or without (248, 249, ˗ obstruction ˗ from obstruction 251-254) (n=2) literature (n=6) review Fistulae Fistulae Abscess ˗ ˗ (n=6) (n=4) +/- fistula Perforation Perforation (248-254) ˗ ˗ (n=1) (n=5) Table 6. 2 Development of the staging tool – key components of the staging tool were extracted from the evidence (i.e. literature review) and populated into stages through end-user opinion

(i.e. semi-structured interviews). High frequency items (n = number of interviewees) for each stage were incorporated into the final staging tool (shaded grey).

152

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

•Predominantly inflammatory ileal stricture evident by bowel STAGE 1 wall thickening.

•Predominantly fibrotic ileal stricture evident by bowel wall STAGE 2 thickening with stenosis.

•Penetrating disease evident by fistulae. STAGE 3

•Perforating disease evident by intra-abdominal abscess STAGE 4 formation, or collection with or without fistulae formation.

Figure 6. 3 The ileocolonic staging tool consists of four stages of escalating disease advancement mirroring radiological features identified from the evidence (i.e. literature review) and end-user opinion (i.e. semi-structured interviews) - Stage 1 reflects early disease which becomes increasingly advanced with stage 4 reflecting ileocolonic Crohn’s disease at its most advanced stage.

High frequency items for each stage were incorporated into the final staging tool (figure 6.3):

➢ Stage 1: ileitis or localised inflammation with mild bowel wall thickening (n=8):

“relatively short segment … no associated sepsis or fistulation” (CG1), and “… new

inflammation and ulceration” (CP2).

153

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

➢ Stage 2: fibrosis with obstruction (n=6): “I think you’d be looking at general fibrotic

sort of change of the small bowel, so obstructive elements to it” (CS1), and “stricturing

in evidence of upstream hold-up dilatation but still no fistulation” (CG1)

➢ Stage 3 – fistulating disease (n=6): “… would be progression of those (previous), so

you’ve got obvious fistulae” (CR1), or “… more about your advancement in terms …

entero enteric fistulae” (CS1).

➢ Stage 4 –perforating disease (n=5) with or without fistulae (n=4): “Abscesses, probably

talking about a stage four… complications with a fistula…bladder, skin…an inter-

abdominal collection associated” (CG2), “you get perforations into the psoas...You

could have a big Crohn’s mass with lots of loops pulled into it” (CS2) and “perforation

we’d associate it (with an…intestinal mass with fistulation and abscess” (CP2).

6.4.4. Validity testing

A total of 160 patients contributed data to the validity testing, with 41 (23.4%), 43 (24.6%), 52

(29.7%), and 24 (13.7%) having stage 1, 2, 3 and 4 disease, based on intra-operative and histopathological findings, respectively (figure 6.4).

154

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

Stage 4 - perforating 15% Stage 1 - Stage 1 - predominantly predominantly inflammatory inflammatory Stage 2 - predominantly 26% fibrotic Stage 3 - fistulating

Stage 3 - fistulating Stage 4 - perforating 32% Stage 2 - predominantly fibrotic 27%

Figure 6. 4 Pie chart representing relative proportions per stage based on intra-operative and histopathological findings - There were 41 (23.4%) with stage 1 disease, 43 (24.6%) with stage

2 disease, 52 (29.7%) with stage 3 disease, and 24 (13.7%) with stage 4 disease.

155

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

Intra-operative and histopathological stages Stage 1 - Stage 2 - Stage 3 - Stage 4 - predominantly predominantly fistulating perforating inflammatory fibrotic Chi squared N (%) N (%) N (%) N (%) test for trend (p value; x2) Total number of patients (%) 41 (100) 43 (100) 52 (100) 24 (100) Female 23 (56) 23 (54) 27 (52) 5 (21) Gender 0.02; 5.2 Male 18 (44) 20 (47) 27 (48) 19 (79) Never 16 (40) 19 (51.4) 23 (45.1) 11 (47.8) Tobacco use Smoker 0.6; 0.2 Smoker 24 (60) 18 (49) 28 (55) 12 (52) Ileocaecal 16 (39) 18 (44) 30 (58) 15 (68) resection Re-resection for Re- 0.01; 6.4 recurrent disease resectional 25 (61) 23 (56.1) 22 (42.3) 7 (31.8) surgery Pre-operative corticosteroids * 11 (28) 7 (17) 14 (29) 8 (36) 0.3; 0.8 Pre-operative antibiotic therapy * 0 (0) 1 (3) 6 (13) 5 (26) <0.001; 12.8 Pre-operative thiopurines and 6- 17 (47) 19 (50) 20 (43) 9 (50) 0.8; 0.03 Metocaptopurine * Pre-operative 13 (35) 17 (46) 14 (31) 4 (21) 0.2; 1.5 5ASA/sulphazalizine Pre-operative biological therapy 2 (5) 2 (5) 2 (4) 0 (0) 0.4; 0.7 * Pre-operative methotrexate * 2 (5) 3 (8) 0 (0) 1 (5) 0.4; 0.6 Pre-operative nutritional 6 (16) 5 (13) 9 (19) 1 (5) 0.6; 0.2 supplementation * Pre-operative anaemia present (Hb < 13 in males and Hb < 12 12 (29) 15 (36) 25 (49) 17 (71) <0.001; 11.5 in females) * Pre-operative leucocytosis 5 (12) 9 (21) 9 (18) 10 (42) 0.03; 4.3 present (WCC ≥ 12) * Pre-operative thrombocytosis 3 (7) 6 (14) 18 (35) 13 (54) <0.001; 21.3 present (Platelets ≥ 450) * Pre-operative hypoalbuminaemia 1 (3) 3 (7) 5 (10) 3 (14) 0.08; 2.9 present (Albumin < 25) * Pre-operative CRP ≥ 40 * 4 (19) 6 (22) 15 (46) 10 (59) 0.002; 9.2 Scheduling Elective 39 (95) 34 (81) 43 (83) 15 (63) 0.03; 8.8 Emergency 2 (5) 8 (19) 9 (17) 9 (38) ASA grade ASA ≤ 2 38 (95) 33 (89) 38 (83) 19 (91) 0.2; 1.4 ASA > 2 2 (5) 4 (11) 8 (17) 2 (10) Laparoscopic 9 (22) 9 (21) 7 (14) 3 (13) Access 0.2; 1.6

Open 32 (78) 33 (79) 45 (87) 21 (88)

156

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

Stapled 24 (77) 25 (83) 27 (82) 13 (77) Anastomosis anastomosis 0.9; < 0.01 Handsewn 7 (23) 5 (17) 6 (18) 4 (24) anastomosis Concomitant strictureplasty * 1 (2) 4 (9) 0 (0) 0 (0) 0.2; 1.6 Concomitant sigmoidectomy * 0 (0) 0 (0) 7 (14) 1 (4) 0.03; 4.6 Stoma formation * 1 (2) 2 (5) 10 (19) 6 (25) < 0.001; 11.2

Histopathological Clear 21 (62) 17 (44) 13 (34) 9 (50) 0.1; 2.2 margin Involved 13 (38) 22 (56) 25 (66) 9 (50) Post-operative intra-abdominal 1 (2) 2 (5) 7 (14) 4 (17) 0.02; 5.8 abscess * Post-operative anastomotic leak 1 (2) 2 (5) 5 (10) 3 (13) 0.07; 3.3 * Post-operative IASC * 1 (2) 2 (5) 7 (14) 4 (17) 0.01; 5.8 Clavien-Dindo ≥ 3 * 1 (2) 2 (5) 3 (6) 2 (8) 0.3; 1.1 Kendall’s Median Median Tau-b Median (IQR) Median (IQR) (IQR) (IQR) Correlation (p value; R) 21 (16 - 27 (19 - Age at diagnosis (years) 25 (20 - 33) 23 (18 - 27) 0.3; -0.06 27) 30) Number of previous resectional 1 (0 - 2) 1 (0 - 2) 0 (0 - 1) 0 (0 - 1) 0.02; -0.16 surgeries Post op length of stay (days) 7 (5 - 10) 7 (5 - 9) 8 (6 - 11) 7 (5 - 10) 0.2; 0.09

* = Reference categories are the absence of the stated variable Hb = haemoglobin; WCC = white cell count; CRP = C-Reactive Protein; ASA = American Society of Anaesthesiology

Table 6. 3 Frequency data, proportions and trend analyses for patient and disease related, pre- operative medications, pre-operative blood parameters, peri-operative and post-operative factors for each stage - Patient and disease related categorical variables were gender (male

[n=84] and female [n=78]), tobacco use (smokers [n=82] and never smoker [n=69]), and previous surgery (ileocaecal resection [n=74] and re-resectional surgery [n=77]). Pre-operative medication variables were corticosteroid (present [n=40] and absent [n=112]), antibiotic therapy (present [n=12] and absent [n=131]), thiopurine/6-metocaptopurines (present [n=65] and absent [n=74]), 5ASA/sulphasalazine (present [n=48]) and absent [n=90]), biologics

157

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

(present [n=6] and absent [n=137]), methotrexate (present [n=6] and absent [n=138]) therapy, and nutritional supplementation (present [n=21] and absent [n=121]). Pre-operative blood parameters were anaemia (present [n=69] and absent [n=89]), leukocytosis (present [n=33] and absent [n=125]), hypoalbuminaemia (present [n=12] and absent [n=142]), thrombocytosis

(present [n=40] and absent [n=118]), and CRP (≥ 40 [n=35] and < 40 [n=63]). Peri-operative variables were scheduling (elective [n=131] and emergency [n=28]), ASA grade (ASA ≤ 2

[n=128] and ASA > 2 [n=16]), access (laparoscopic [n=28] and open [n=131]), anastomosis

(stapled [n=89] and hand-sewn [n=22]) and concomitant strictureplasty (absent [n=155] and present [n=5]), concomitant sigmoidectomy (absent [n=152] and present [n=8]), stoma formation (absent [n=141] and present [n=19]), and histopathological resection margin

(involved [n=69] and clear [n=60]). Post-operative variables were intra-abdominal septic complications (absent [n=146] and present [n=14]), anastomotic leak (absent [n=149] and present [n=11]), and Clavien-Dindo (<3 [n=151] and ≥3 [n=8]). Non-parametric numerical variables were age at diagnosis (recorded for 154 patients) and the number of previous resectional surgeries performed (recorded for 155 patients) and post-operative length of stay in days (recorded for 151).

158

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

6.4.4.1. Trends in patient and disease related characteristics with increasing stage (table 6.3)

There were 18 (43.9%), 20 (46.5%), 25 (48.1%) and 19 (79.2%) males with increasing increment stage, respectively (p = 0.02; x2 = 5.2) [figure 6.5].

Figure 6. 5 Bar chart representing trends in males and females with each increasing stage -

There were 23 (56.1%), 23 (53.5%), 27 (51.9%) and 5 (20.8%) females with stage 1, 2, 3 and

4 disease respectively, and there were 18 (43.9%), 20 (46.5%), 25 (48.1%) and 19 (79.2%) with stage 1, 2, 3 and 4 disease respectively (p = 0.02; x2 = 5.2).

There were 24 (60.0%), 18 (48.6%), 28 (54.9%) and 12 (52.2%) smokers with increasing increment stage, respectively (p = 0.6). There were 25 (61.0%), 23 (56.1%), 22 (42.3%) and 7

(31.8%) of patients who had a prior history of resectional surgery with each increment stage, respectively (p = 0.01; x2 = 6.5). The median number (IQR) of resections per stage increment were 1 (0 - 2), 1 (0 - 2), 0 (0 - 1) and 0 (0 - 1).

159

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

6.4.4.2. Trends in pre-operative medication usage with increasing stage (table 6.3)

There were 11 (28%), 7 (17%), 14 (29%) and 8 (36%) patients who received pre-operative corticosteroids with increasing increment stage respectively (p = 0.4). There were 0 [0%], 1

[3%], 6 [13%] and 5[26%]) patients who received pre-operative antibiotic therapy, with increasing increment stage, respectively (p=0.0004; x2 = 12.8). There were 2 (5%), 2 (5%), 2

(4%), 0(0%) patients who received pre-operative biological therapy with increasing stage, respectively (p = 0.4).

6.4.4.3. Trends in pre-operative blood parameters with increasing stage (table 6.3)

There were 12 (29%), 15 (36%), 25 (49%) and 17 (71%)] patients with a pre-operative anaemia, with increasing increment stage, respectively (p = 0.0007; x2 = 11.5) [figure 6.6].

Figure 6. 6 Bar chart representing trends in the number of patients with and without a pre- operative anaemia per stage - There were 12 (29%), 15 (36%), 25 (49%) and 17 (71%) with a

160

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery pre-operative anaemia and 71 (63%), 41 (63%), 41 (45%) and 11 (26%) without a pre-operative anaemia with each incremental stage, respectively (p < 0.001, x2 = 19.2).

There 5 (12%), 9 (21%), 7 (14%) and 10 (42%) patients with a pre-operative leucocytosis, with increasing increment stage, respectively [p = 0.04; x2 = 4.3] (figure 6.7).

Figure 6. 7 Bar chart representing trends in the number of patients with and without a pre- operative leukocytosis per stage - There were 5 (12%), 9 (21%), 7 (14%) and 10 (42%) patients with a pre-operative leukocytosis and 36 (88%), 33 (79%), 44 (86%) and 14 (58%) without a pre-operative leukocytosis with increasing increment stage, respectively (p = 0.04; x2 = 4.3).

161

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

There were 1 (3%), 3 (7%), 5 (10%), 3 (14%) patients with a hypoalbuminaemia, with increasing increment stage, respectively (p = 0.008, x2 = 2.9). There were 3 (7%), 6 (14%), 18

(35%) and 13 (54%) patients with thrombocytosis, with increasing increment stage, respectively (p< 0.001, x2 = 22.0).

Figure 6. 8 Bar chart representing the number of patients with and without a thrombocytosis per stage - There were 38 (93%), 36 (86%), 33 (65%) and 11 (46%) patients without a pre- operative thrombocytosis and 3 (7%), 6 (14%), 18 (35%) and 13 (54%) with a pre-operative thrombocytosis with each incremental stage, respectively (p < 0.001, x2 = 20.0).

There were 4 (19%), 6 (22%), 15 (46%), and 10 (59%) patients with a CRP ≥ 40, with increasing increment stage, respectively (p < 0.001, x2 = 32.2) [figure 6.9].

162

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

Figure 6. 9 Bar chart representing the number of patients with and without a raised CRP per stage - There were 4 (19%), 6 (22%), 15 (46%), and 10 (59%) with a raised CRP, and 17 (81%),

21 (78%), 18 (55%) and 7 (41%) without a raised CRP, with increasing increment stage, respectively (p < 0.001, x2 = 32.2).

6.4.4.4 Trends in peri-operative factors with increasing stage (table 6.6)

There were 39 (95%), 34 (81%), 43 (83%) and 15 (63%) patients who underwent an elective operation and 2 (5%), 8 (19%), 9 (17%) and 9 (38%) patients who underwent an emergency operation, with increasing increment stage, respectively (p = 0.08, x2 = 3.4). There were 9

(22%), 9 (21%), 7 (14%) and 3 (13%) patients undergoing a laparoscopic procedure and 32

(78%), 33 (79%), 45 (87%) and 21 (88%) undergoing an open procedure, with increasing increment stage, respectively (p = 0.2). There were 0 (0%), 0 (0%), 7 (14%) and 1 (4%) patients undergoing a concomitant sigmoidectomy, with increasing increment stage respectively (p =

0.03; x2 = 4.6) [figure 6.10].

163

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

Figure 6. 10 Bar chart representing trends in the number of patients who had a concomitant sigmoidectomy - There were 0 (0%), 0 (0%), 7 (14%) and 1 (4%) patients who had a concomitant sigmoidectomy and 41 (100%), 43 (100%), 45 (87%) and 23 (96%) patients who did not have a concomitant sigmoidectomy, with increasing increment stage, respectively (p =

0.03; x2 = 4.6).

There were 1 (2%), 2 (5%), 10 (19%) and 6 (25%) patients who had pre-emptive stoma formation, and 40 (98%), 41 (95%), 42 (81%) and 18 (75%) patients who had a one-stage procedure, with increasing increment stage, respectively (p = 0.0008; x2 = 11.2) [figure 6.11]

164

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

Figure 6. 11 Bar chart representing trends in the number of patients who had a pre-emptive stoma formation - There were 1 (2%), 2 (5%), 10 (19%) and 6 (25%) patients who had a stoma formation and 40 (98%), 41 (95%), 42 (81%) and 18 (75%) patients who had a one-stage procedure, with increasing increment stage, respectively (p = 0.0008; x2 = 11.2).

6.4.4.5 Trends in post-operative factors with increasing stage (table 6.7)

There were 1 (2%), 2 (5%), 7 (14%) and 4 (17%) patients who developed post-operative IASCs, with increasing increment stage, respectively (p = 0.01; x2 = 5.8) [figure 6.12].

165

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

Figure 6. 12 Bar chart representing trends in the number of patients who had a post-operative

IASC - There were 1 (2%), 2 (5%), 7 (14%) and 4 (17%) patients who had a post-operative

IASC with increasing increment stage, respectively (p = 0.01; x2 = 5.8).

There were 0 (0%), 0 (0%), 6 (12%) and 1 (4%) patients with entero-cutaneous fistulae formation, with increasing increment stage, respectively (p = 0.04; x2 = 4.1). The post- operative length of stay in days (median, IQR) was 7 (5-10), 7 (5-9), 7 (5-9) and 8 (6-11) and

7 (5-10) days, per increment stage, respectively (p = 0.2).

6.4.5. Reliability testing

Stages were applied to the pre-operative images of 84 patients. There were 44 (52%), 6 (7%),

25 (30%) and 9 (11%) patients with stage 1, 2, 3 and 4 disease, respectively. In the same cohort, there were 26 (39%), 14 (21%), 20 (30%), and 7 (10%) patients with stage 1, 2, 3 and

166

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

4 disease, based on the intra-operative and histopathological findings (R=0.8; p < 0.001) [figure

6.12].

50

45 44

40

35

30 Histopathology 26 25 and intra- 25 operative stages 20 20 CTE and MRE 14

15 stages Number of patients (n) patients of Number 10 9 7 6 5

0 Stage 1 - Stage 2 - Stage 3 - Stage 4 - predominantly predominantly fistulating perforating inflammatory fibrotic

Figure 6. 13 Bar chart representing the number of patients based on the intra-operative and histopathological stage and the radiological stage - There were 44 (52%), 6 (7%), 25 (30%) and 9 (11%) patients with stage 1, 2, 3 and 4 based on the pre-operative CTE and MRE, and there were 26 (39%), 14 (21%), 20 (30%), and 7 (10%) patients with stage 1, 2, 3 and 4 disease based on the intra-operative and histopathological findings, respectively (R=0.8; p < 0.001).

167

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

6.5. Discussion

6.5.1. Summary of findings

This chapter has described for the first time to the best of our knowledge four distinct clinic- pathologic-radiological stages based on literary evidence and expert opinion to further define ileocolonic Crohn’s disease. The combined quantitative approach in validity and reliability testing provides an additional strength for this study. Preliminary validation of stages based on intra-operative and histopathological findings of a surgical cohort has demonstrated the association between more advanced stages of disease with poorer pre-operative state (i.e. hypoalbuminaemia, anaemia, markers of sepsis and inflammation), emergent surgery and additional surgical procedures (including pre-emptive stoma formation and concomitant sigmoidectomy) and post-operative outcome (including IASC). Broad reliability between staging derived from pre-operative imaging and subsequent pathological examination has been established. The described staging tool may therefore be applied to pre-operative imaging to facilitate surgical decision-making and estimate morbidity.

The deliberate use of qualitative approaches to construct the staging tool offered detailed information from individual IBD experts, and was well suited to explore a complex and heterogeneous issue. (195, 196). In its first step to establish a staging tool, features common to all validated radiological modalities (CTE, and MRE) comprising of the core components of the staging system through literature review following PRISMA guidelines. These features were assessed against a surgical or resected histopathological specimen reference. Studies reporting on endoscopy exclusively as a reference standard were excluded as they would not provide information on mural or extra-mural features. Further limitations to endoscopy were

168

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery highlighted in the interviews. End-user opinion allows further definitions of these features, with a view to establishing severity in a staged fashion.

Validation against intra-operative and histopathological staging demonstrated several applications for this tool. Analysis of patient and disease related factors demonstrated that male gender was associated with a more advance disease stage, in particular, a higher proportion of men appear to develop stage 4 disease. Crohn’s severity has often been considered equivocal between gender types, however some studies have demonstrated a higher proportion of males with abscess formation (255, 256). Smoking is considered the strongest predictor of disease recurrence following surgery (26). This chapter, however, suggests that tobacco use has no role in disease advancement with increasing stage. This may suggest tobacco use is independent of disease phenotypic severity and is more important in the setting of relapsing and recurrent disease.

Disease advancement with increasing stage is demonstrated with an increase in the proportion of patients with a pre-operative septic state and disease activity. These are evident with the presence of anaemia, leukocytosis, and thrombocytosis with increasing stage, further validating the ordinal severity index of the staging tool. Anaemia, leucocytosis and raised CRP are all recognised markers of disease activity, though their role in phenotypic variance before surgery is less well understood (257-263). Although CRP is a recognised marker for disease activity, it is not known to correlate with disease phenotype or radiological activity (e.g. stricturing vs penetrating) (261, 262). The findings from this study however do show correlation between

169

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery pre-operative CRP and increasing intra-operative and histopathological stage, which may suggest its importance in highlighting pre-operative intra-abdominal sepsis prior to surgery.

Greater peri-operative morbidity with increasing stage was demonstrated with greater emergent operations and greater concomitant procedures alongside the ileocolonic resection, including sigmoidectomy and stoma formation. Furthermore, greater post-operative morbidity with increasing stage was demonstrated with greater post-operative intra-abdominal septic complications. This chapter, therefore highlights the importance of targeted surgery with early disease to improve outcome. Stoma formation is often a necessary procedure either to avoid the breakdown of the newly formed anastomosis through a proximal faecal diversion or through a split stoma formation (264).

Broad correlation between radiological and histopathological and/or intra-operative assigned stages was demonstrated in the study, highlighting its use as a decision-aid pre-operatively, in the MDT setting. There was, however, an under-estimation of stage 1 and over-estimation of stage 2 disease. Predominantly inflammatory strictures are due to acute transmural inflammation and oedema and are typically treated with anti-inflammatory medications. Pre- dominantly fibrotic strictures are caused by chronic mural deposition of extracellular matrix proteins and are treated with mechanical therapies consisting of surgical resection or endoscopic dilation (265). This study therefore highlights a potential greater proportion of patients with predominantly fibrotic disease receiving immunomodulatory therapy but may benefit with earlier endoscopic balloon dilatation or surgery.

170

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

6.5.2. Limitations

This chapter has limitations. Firstly, in its construction, a distinction was made between the type of stricture being either predominantly inflammatory or predominantly fibrotic. Although broad correlation was demonstrated across all stages between actual and radiological stages, there is under estimation of stage 1 disease and over estimation of stage 2 disease. The retrospective review of histopathological reports and pre-operative imaging would likely have led to this effect and would be limited with prospective staging of actual specimens by trained histopathologists. Secondly, it is a single centre study describing the experience of a single tertiary specialist institution and cannot exclude bias in end-user responses particularly relating to shared misconceptions of disease severity. Thirdly, testing variables against radiologically assigned stages (as opposed to histopathological) may further delineate the predictive power of the staging tool in the pre-operative setting. Further limitations relating to the qualitative purposive sampling, the Hawthorne effect, the retrospective validity design, and lack of sensitivity testing around pre-defined categorical blood parameter cut-offs are discussed in section 9.2.

171

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

CHAPTER 7: DEFINING KEY PERFORMANCE INDICATORS (KPIs) TO

MONITOR AND ENHANCE THE QUALITY OF CARE IN THE SURGICAL

MANAGEMENT OF ILEOCOLONIC CROHN’S DISEASE

7.1. Abstract

KPIs help to monitor outcomes, reduce variation in services, and drive up standards. General indicators of performance have been previously put forward for IBD care, but none specifically cover provision of the surgical IBD service for patients Crohn’s disease undergoing ileocolonic surgery. This is a prospective, multi-centre, expert-based consensus study, using a Delphi formal consensus-building methodology online survey. Initial items were developed from semi-structured interviews. These were then fed through a Delphi process to achieve consensus. Items were rated on a Likert scale from 1 (not important) to 5 (very important).

Consensus was defined with an interquartile range ≤ 1, and items with a median score of > 3 were considered for inclusion. A panel of 21 experts, consisting of 14 surgeons and 7 gastroenterologists was recruited across Europe. Consensus was achieved on 12 surgery- specific outcome measures: with 6 items relating to morbidity, and 1 item relating to quality of life. Consensus was achieved on 5 overall measures of quality: 2 institutional measures and 3 measures relating to patient function and satisfaction. A further 9 items detailing quality assurance mechanisms achieved consensus. There was consensus on the organisational structure with 7 items relating to the IBD team, 11 items relating to facilities and consensus detailing the critical volume for the individual surgeon and the institution. Consensus on 6

KPIs detailing clinical care processes was achieved. These indicators cover the governance and structure of the surgical services. Monitoring of IBD surgical services with these KPIs may help to reduce variation across services and to improve quality.

172

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

7.2. Aims

The primary aim of this study is to obtain expert consensus on outcome measures for patients undergoing ileocolonic Crohn’s resections. The secondary aim is to obtain expert consensus- based recommendations for the organisational structure and clinical care processes to improve these outcomes.

7.3. Methodology

A two-stage, multi-centre, Delphi methodology was used. Stage 1 involved item construction through semi-structured interviews. Items were then incorporated into stage 2, which used a

Delphi to obtain consensus. The protocol for the study was reviewed by a by a Research Ethics

Committee in London, UK and approval was given prior to data collection (Research Ethics

Committee reference: 13YH 0175).

7.3.1. Stage 1: Semi-structured interviews

As detailed in section 2.2.2, a full and comprehensive qualitative methodology was applied to this section of the study.

7.3.1.1. Procedure

Interviews were carried out using a standardised and piloted semi-structured interview protocol

(Appendix ii) delivered by two trained interviewers of clinical background. The interview protocol explored key themes encompassing the quality of an IBD surgical service provision in accordance with Donabedian’s quality framework model (see section 1.4.3 for more detail

173

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery on this) (266). Specifically, the interviews explored the markers of quality in IBD surgery, the markers of quality from the organisational structure, and clinical care processes of an IBD surgical service provision, and the outcomes measure that are important as quality indicators of IBD surgery. Themes were then refined and made into items by the research team and incorporated into the Delphi survey.

7.3.1.2. Participants

An opportunistic multidisciplinary sample was obtained and included consultant colorectal surgeons (CS), consultant gastroenterologists (CG), service managers (SM), IBD nurse specialists (NS), and IBD patients (PP). Expert participants were recruited from twelve different institutions across the United Kingdom. Participation was voluntary, and informed consent was obtained from all participants prior to data collection. Anonymity was ensured throughout the study. Invitations were sent via the charity Crohn’s and Colitis UK to recruit patients with IBD into the interview study.

7.3.2. Stage 2: Delphi consensus

As detailed in sections 2.3.1 and 2.3.2, a full and comprehensive Delphi methodology was applied to this section of the study.

7.3.2.1. Research procedure

The second stage was a prospective, multi-centre, pan-European standard Delphi methodology.

The survey was emailed to a multidisciplinary specialist sample recruited through the European

Crohn’s and Colitis Organisation (ECCO). Participants responded to statements about what

174

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery outcome can be used to measure a surgical service provision for patients with ileocolonic

Crohn’s disease. Furthermore, statements on the organisation structures and clinical care processes necessary for these outcome measures to be reached. In addition, space for free text comments was made available (see Appendix iii for survey).

7.3.2.2. Establishing the multidisciplinary expert consensus panel members

Surgeons and gastroenterologists from nineteen institutions across Europe were recruited to the expert panel for the Delphi process. Specific eligibility criteria were used (figure 7.1.) to ensure an appropriate level of expertise on the panel and credibility of the subsequent consensus (204)

(see section 2.3.2 for consensus inclusion criteria).

Figure 7. 1 Eligibility criteria for inclusion as an expert Delphi panellist

175

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

An invitation was sent to members of the European Crohn’s and Colitis Organisation to recruit

Delphi consensus panellists. Reminders were issued to all non-responders at 2 and 4 weeks after initial contact. There were a total of twenty-five respondents.

7.4. Results

7.4.1. Participants

7.4.1.1. Stage 1: semi-structured interviews.

Thematic saturation was achieved after 27 semi-structured interviews, consisting of 9 consultant colorectal surgeons (CS), 4 IBD nurse specialists (NS), 5 consultant gastroenterologists (CG), 4 service managers (SM) and 5 IBD patients (PP) (figure 7.2).

Patients 19% Consultant Surgeons 33%

Service Managers 15%

IBD Nurse Gastroenterologist Specialists 18% 15%

Figure 7. 2 A pie chart representing relative proportions of participants within the interview study – There were 9 (33%) consultant colorectal surgeons (CS), 4 (15%) IBD nurse specialists

(NS), 5 (18%) consultant gastroenterologists (CG), 4 (15%) service managers (SM) and 5

(19%) IBD patients (PP).

176

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

Individual interviews took place face-to-face (n = 16) and via telephone (n = 11) between June and October 2013. Participants were recruited from 12 hospitals which were a mixture of community hospitals and specialist IBD centres. In the Southwest 11 participants were interviewed: 4 consultant colorectal surgeons, 4 consultant gastroenterologists, 1 IBD nurse specialist, and 2 service managers. In the London region 11 participants were interviewed: 5 consultant colorectal surgeons, 1 consultant gastroenterologist, 3 IBD nurse specialist, and 2 service managers. Five patients who had recent (within 5 years) luminal Crohn’s surgery were recruited through Crohn’s and Colitis UK from 3 different hospitals.

7.3.1.2. Stage 2: Expert consensus panel

Of the twenty-five respondents, twenty-one were clinicians who met the eligibility criteria set out in table 1 and were subsequently recruited to the expert panel. Fourteen of the panel members were consultant colorectal surgeons, and seven were consultant gastroenterologists.

The median number of years’ experience in consultant post across the group was 10 years.

Participants were recruited across 18 institutions spanning 10 European countries. Consensus was reached following two iterations of the Delphi process (table 7.1).

Consultant gastroenterological Consultant colorectal surgical panellists panellists (n = 7) (n = 14) Years in City/Country Years in City/Country post post 12 Ankara / Turkey 16 Amsterdam / Netherlands 4 Bern / Switzerland 6 Athens / Greece 7 Heraklion / Crete 18 Oporto / Portugal

177

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

5 Abergavenny / Wales 6 Linköping / Sweden 21 Cardiff / UK 3 Muenster / Germany 21 Sheffield / UK 10 Harrow / UK 30 Angera / Italy 16 Plymouth / UK 3 Birmingham / UK 6 London / UK 12 London / UK 2 Chelmsford / UK 13 Sheffield / UK 15 London / UK 1 Rozzano Milano / Italy

Table 7. 1 Expert consensus panellists (ECPs) detailing years in post, institution and country -

Seven consultant gastroenterologists (CG) and fourteen consultant colorectal surgeons (CS) were recruited as ECP members. Participants were recruited across 18 institutions spanning 10

European countries.

7.4.2. Outcome measures of a surgical service provision

Tables 7.2 and 7.3 demonstrate consensus-derived surgery specific and overall quality outcome measures respectively (with the median and IQR for each iteration) derived from thematic analysis of the semi-structured interviews.

178

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

7.4.2.1. Stage 1: semi-structured interviews

7.4.2.1.1. Surgery specific outcome measures of quality

Pre-operative morbidity (PrM)

Considering the vast phenotypic heterogeneity observed in ileocolonic Crohn’s surgery, it was suggested by three participants to subdivide patients into those with early and late disease, and to cater outcome measures accordingly: “…(regarding) the ones who had late ileocecal, and when nutritionally depleted, what proportion received nutritional therapy…In the early

(disease), what proportion received a laparoscopic operation?” (CS4). Nutritional depletion with advanced disease is known to have an impact on outcomes following surgery. The proportion of patients who received pre-operative nutritional optimisation was highlighted as an outcome measure across interviewees: "… nutrition has to be high up on the list of making sure that you’ve got no areas that you can improve on before surgery” (CS5).

Peri-operative morbidity (PeM)

A measure of cases performed laparoscopically without conversion to open was suggested by four participants: “…with a younger age group and with the advent of laparoscopic surgery, that’s what people expect, almost in the same way as cancer now, it is a standard of care to offer that” (CS1), furthermore “… as a fundamental principle if you offer laparoscopic surgery and can do a majority of an operation laparoscopically it has (a) significant benefit for the patient because a lot (will have re-resectional) surgery” (CS4). The formation or avoidance of a stoma was considered a difficult measure of quality: “…quality can involve avoidance of stoma but also formation of stoma…” (CS5), “…a good IBD surgeon tailors what he does to that patient, and it’s very hard to put it in black and white…no stomas (doesn’t) equal high

179

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery quality” (CS4). The importance of reversibility was therefore highlighted as a measure “I give a period with a stoma, so they know life’s not the end of the world with a stoma, but also know it’s reversible..." (CS5).

Post-operative morbidity (PoM)

Fourteen participants highlighted a measure of morbidity as a measure of quality: “...patients tend to be even more prone to complications; they have a problem with healing, so quality indicators are when you have low morbidity, low infection post-operatively, quickly recovery, minimal malnutrition and minimal complications” (CS5). In particular, the ability to capture re-operation rates secondary to intra-abdominal septic complications through existing classification systems: “…such as Clavien-Dindo" (CS3). The occurrence of fistulae formation was also considered important as a specific measure in the setting of Crohn’s disease: “…

(wound) dehiscence (and) fistulae… after surgery… are higher in Crohn’s resections” (CS1).

Long-term morbidity (LtM)

Presence of clinical or surgical recurrence was highlighted by six participants as an outcome measure of quality: "A significant proportion will recur; therefore, you need to have some sort of a medical input immediately afterwards to prevent this or reduce the recurrence rate or reduce the morbidity from recurrences, (or) reduce the re-operation data" (CS5). The influence this would have on surgical recurrence however was uncertain: “I don’t think anyone has data of any direct reduction in … surgery for recurrences following medical therapy”

(CS5). In relation to re-resectional surgery and the formation of short-bowel, it was also considered important to “...be reporting residual, remaining bowel as a marker…" (CS6).

180

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

Furthermore, a marker of quality was suggested indicating the proportion of those started on prophylactic medical therapy: “…looking at recurrence rates … (although) not directly attributable to the surgery but the whole process of surgical care and the follow-up and decision to or not to start maintenance therapy afterwards… prevention therapy, it’s all part of that package" (CG4).

Quality of life (QoL)

A measure of quality of life was highlighted as an outcome measure with emphasis that it

“…has to be the number one important outcome. The rest are all, not to say not important, but all come secondary to that because if you have not achieved an improvement in that patient’s quality of life there was no point you are doing that surgery" (CS4). This was further highlighted in that “…hard markers (such as stool frequency) might not be relevant to the patient because they might not be worried about that. But if you’ve got a sick patient who’s been made to feel a lot better but actually isn’t yet back to normal, then again they may be very satisfied because they’re not sick anymore" (CG1).

7.4.2.1.2. Overall markers of quality

Institutional measures (InM)

Ten participants highlighted the length of stay and the readmission rates as an overall measure of quality for an IBD surgical service provision. The usefulness length of stay adds as a measure of quality was also contested across participants: “…you can measure the duration of the inpatient admission plus any subsequent pre-admission of a staged operation…I would put is as a low priority measure in IBD surgery” (CS3). A reason for this included its influence

181

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery on the patient’s pre-operative condition: “I’m not sure if it’s very useful but it is an indicator, a target indicator of things going well, but it is influenced by pre-operative state” (CS5). A similar recognition was highlighted for 28-day readmission rates: “…a readmission should be considered a treatment failure in that grand scheme of things. You are trying to get patients home so… that they function better at home. Your easiest way to prevent a readmission is to keep them in longer” (CS3). A measure of overall 30-day post-operative morbidity (such as

Clavien-Dindo) was also highlighted to provide: “…an evaluation of surgical outcome…”

(CS3).

Return to function (RtF)

Nineteen participants highlighted quality of life as an overall measure of quality. A suggested time-point of measure after surgery included a quality of life check at “… six month(s) one might be very good” (CS3), with "(measurement through) ... quality of life scales that one can use that are very specific to IBD" (CS4). It was highlighted that the aim of surgery should be to return the patient to some functional normality such as a return to work: “but the notion of having surgery is to improve one’s lifestyle… looking at the number of people that have returned to work or back to their former employment or their sickness record has gone from being off work sick to back at work ...” (SM3).

Patient satisfaction (PS)

Fifteen participants suggested that patient satisfaction assessments through surveys would be important: “…you can learn an awful lot by just asking the patients, about your processes as well, you know, they’ll soon uncover parts of your processes that you might think are great,

182

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery that actually you’re wasting your time on. Because either they’re not important to the patient or they’re not very good, so either they need to be improved or you put your resources elsewhere” (CS1).

7.4.2.1.3. Quality assurance

Local meeting / forum (LM)

Thirteen participants highlighted the importance of discussing outcomes within the context of a meeting. Suggested forums were dependant on the outcome measure requiring discussion.

In the setting of surgical death: “…mortality is a very poor outcome in IBD surgery because it’s so low... it should be measured, and it certainly should be captured” (CS3). Potential forums for case discussions include the morbidity and mortality meeting: “…(it’s) a good potential forum for the immediate things… (however) I don’t believe gastroenterologists can peer review morbidity, especially when they are responsible in part for some of it… you have to look at this on a regional level, actually, and a national level also..." (CS3); or the IBD MDT

“… it should be an extension of the multidisciplinary meeting…it’s no different than a cancer

MDT where you discuss outcomes” (CS4). Discussions relating to significant morbidity were considered necessary within the IBD MDT setting: “…it’s actually very important in ileocecal

Crohn's because I think it can be avoided” (CS3). Furthermore, capturing and discussing readmissions and prolonged length of stay were highlighted: “the nature of IBD is that a lot of patients come back. It’s high morbidity, they do come back usually with small things, hopefully with small things…” (CS3).

183

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

Reporting outcomes (RO)

Seven participants highlighted the need for data collection through a local registry of surgical

IBD patients, with a view: “…to get prospective data on what happens to the segmental resections, the stricture in Crohn’s Disease and that sort of thing, so re-operation rates over time, and gut failure rates and all this sort of thing.” (CS6). It was also highlighted to submit data into national audits or registries "…reporting of your data, actually submitting to registries of outcome for IBD work I think is essential … that will highlight where there are problems."

(CS3). Furthermore, reporting ensuring open publication of this data: “…if you are talking about a national database… open publication of results sharpens the mind a little” (CS1); and if met with apprehension “…part of that is extremely justified because the perception when a problem happens following a surgery is always related to the surgeon. I think we know with

IBD that a lot of it actually comes down to some of the treatment and delay in getting to surgery, and we have to be more clear about that and upright, and actually reflect back on our gastroenterological practice and say… in your attempt to avoid surgery for these patients you are actually exposing some of them to significant risk.” (CS3). Registries and open publication would also aid credentialing of IBD services though governing bodies: “The way that the

Trusts are changing is because the mirror is being held up to them as to what better Trusts are doing elsewhere. And so, this is with the IBD QUIP, people are being given grading status as to what their service is. So, if they come back with a grading status of ‘D’ and it’s because the things that are deemed high quality aren’t being implemented then it’s reflected back to the

Trust” (CS3).

184

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

7.4.2.2. Stage 2: Delphi consensus on outcome measures

7.4.2.2.1. Surgery-specific outcome measures

A consensus on key performance indicators [median; IQR] that were considered very important

(Likert rating 5) in Crohn's small bowel or ileocolonic surgery included the proportion of patients who require re-operations within 30 days secondary to intra-abdominal septic complications [5; 0], the proportion of patients who develop enterocutaneous fistulae within

90 days [5; 1], and the proportion of patients who develop subsequent short bowel [5; 1]. A consensus on items that were considered important (Likert rating 4) were the proportion of patients who receive pre-operative nutritional optimization where indicated [4; 1], quality of life measures at 6 months post-operatively using the Crohn's Life Impact Questionnaire (CLIQ)

[4; 1], and the proportion of patients who develop clinical recurrence within 12 months [4; 1].

Delphi Likert ratings (median; interquartile range)

Emergent Outcome measures specific to Crohn's Round 1 Round 2 themes small bowel or ileocolonic surgery include: the proportion of patients who require re- PoM operations within 30 days secondary to intra- 5; 1 5; 0 abdominal septic complications. the proportion of patients who develop PoM 4; 1 5; 1 enterocutaneous fistulae within 90 days. the proportion of cases performed PeM 3; 0 3; 0 Items laparoscopically without conversion to open. derived from the proportion of patients who are stoma free PeM 4; 1 4; 2 semi- after 12 months. structured the proportion of patients who develop interviews LtM 4; 1 4;1 clinical recurrence within 12 months. the proportion of patients who develop LtM 3; 1 3; 1 surgical recurrence within 36 months. the proportion of patients who receive pre- PrM operative nutritional optimisation where 4; 1 4; 1 indicated.

185

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

the proportion of patients who develop LtM 5; 1 5; 1 subsequent short bowel. the proportion of patients who are started on LtM prophylactic immunomodulatory therapy 4; 1 4; 1 within 6 months after surgery. quality of life measures at 6 months post- QoL operatively using the Crohn's Life Impact 4; 2 4; 1 Questionnaire (CLIQ). Footnote: PrM = Pre-operative Morbidity; PeM = Peri-operative Morbidity; PoM = Post-operative Morbidity; LtM = Long-term Morbidity; QoL = Quality of Life Table 7. 2 Outcome measures specific to small bowel and ileocolonic Crohn’s disease surgery

(shaded areas indicate items that reached consensus following two iterations) - There were 10 items submitted into the first Delphi iteration, derived from emergent themes obtained through semi-structured interviews: 1 item relating to pre-operative morbidity (PrM), 2 items relating to peri-operative morbidity (PeM), 2 items relating to post-operative morbidity (PoM), 4 items relating to long-term morbidity (LtM) and 1 item relating to quality of life (QoL). Following two iterations, 7 items achieved consensus.

7.4.2.2.2. Outcome measures for the overall quality of the IBD surgical service provision (table

7.3)

A consensus on indicators that were considered important (Likert rating 4) were the proportion of readmissions within the six month post-operative period [4; 1], the proportion of patients that return to work within a six month post-operative period [4; 1], distribution of patient satisfaction surveys (through IBD specific patient panels, departmental open days, or patient opinion websites) [4; 1], overall 30-day post-operative morbidity (as graded by the Clavien-

Dindo classification) [4; 1], and quality of life measures 12 months post-operatively by use of the Inflammatory Bowel Disease Questionnaire (IBDQ) [4; 1].

186

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

Delphi Likert ratings Emergent Outcome measures required in measuring (median; interquartile the overall quality of an IBD surgical range) themes service include: Round 1 Round 2

the proportion of readmissions within the six- InM 4; 1 4; 1 month post-operative period. InM the length of post-operative inpatient stay. 4; 2 4; 2 the proportion of patients that return to work RtF 4; 1 4; 1 Items within a six-month post-operative period. derived distribution of patient satisfaction surveys from (through IBD specific patient panels, PS 4; 2 4; 1 semi- departmental open days, or patient opinion structured websites). interviews overall 30-day post-operative morbidity (as InM 4; 1 4; 1 graded by the Clavien-Dindo classification). quality of life measures 12 months post- RtF operatively by use of the Inflammatory Bowel 4; 1 4; 1 Disease Questionnaire (IBDQ). Footnote: InM = Institutional Measures; RtF = Return to function; PS = Patient satisfaction

Table 7. 3 Outcome measures for the overall quality of an IBD surgical service provision

(shaded areas indicate items that reached consensus following two iterations) – There were six items submitted into the first Delphi iteration, derived from emergent themes from stage 1: 3 items relating to institutional measures (InM), 2 items relating to return to function (RtF), and

1 item relating to patient satisfaction (PS). Following two iterations, six items achieved consensus.

7.4.2.2.3. Quality assurance (table 7.4)

A consensus on KPIs [median; IQR] that were considered very important (Likert rating 5) in

187

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery quality assurance included an IBD team meeting discussion for all IBD deaths within 12 months of surgery with the outcome of the discussion recorded and submitted to national data collection databases [5; 0], an IBD team meeting discussion in the event of significant post- operative morbidity (Clavien-Dindo grade ≥ 3) [5; 1], discussion in the surgical Morbidity &

Mortality meeting in the event of surgical deaths within 30 days of surgery [5; 0], submission of surgical outcome data into national IBD audits or registries [5; 1], and a local registry of surgical IBD patients [5; 1]. A consensus on indicators that were considered important (Likert rating 4) were open publication or registry reporting for overall morbidity and mortality [4; 0], a process of credentialing from a national or international governing body for the IBD unit to deliver surgical services [4; 1], an IBD team meeting discussion in the event of all IBD-related post-operative readmissions [4; 0], and an IBD team meeting discussion considering the length of post-operative inpatient stay is greater than 14 days [4; 1].

188

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

Delphi Likert ratings Quality assurance mechanisms for the (median; interquartile Emergent delivery of a high quality IBD surgical range) themes service should include: Round 1 Round 2

discussion in the surgical Morbidity & LM Mortality meeting in the event of surgical 5; 0 5; 0 deaths within 30 days of surgery. an IBD team meeting discussion for all IBD deaths within 12 months of surgery with the LM 5;1 5; 0 outcome of the discussion recorded and submitted to national data collection. an IBD team meeting discussion in the event LM of significant post-operative morbidity 4; 1 5; 1 (Clavien-Dindo grade 3). submission of surgical outcome data into Themes national IBD audits or registries (including RO 5; 1 5; 1 and items the IBD audit, ileal pouch registry, IBD derived registry). from semi- RO a local registry of surgical IBD patients. Ω 5; 1 structured interviews an IBD team meeting discussion in the event LM of all IBD-related post-operative 4; 2 4; 0 readmissions. open publication or registry reporting for RO 4; 2 4; 0 overall morbidity and mortality. an IBD team meeting discussion considering LM the length of post-operative inpatient stay is 4; 1 greater than 14 days. 4; 1 a process of credentialing from a national or international governing body for the IBD unit RO 4; 2 4; 1 to deliver surgical services.

Footnote: LM = local meeting/forums; RO = reporting outcomes

Table 7. 4 Quality assurance mechanisms to ensure a high quality IBD surgical service - There were 8 items submitted into the first Delphi iteration, derived from emergent themes from stage

1: 5 items relating to local meetings (LM) and 3 items relating to reporting outcomes (RO). A ninth item (Ω) relating to reporting outcomes (RO), was submitted into the second Delphi iteration (after deliberation within the study group). Consensus was achieved for all nine items.

189

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

7.4.3. Structure and organisation of IBD surgical services

7.4.3.1. Stage 1: semi-structured interviews

7.4.3.1.1. The IBD team

The surgical expertise (SuE)

Twenty-two interviewees recognised the need of a good, comprehensive IBD team to ensure a high-quality surgical service provision. It was recognised that this team should include a consultant colorectal surgeon who is trained and experienced in laparoscopic surgery: “…it has to be an accredited colorectal surgeon… with a younger age group and with the advent of laparoscopic surgery, that’s what people expect… it is a standard of care to offer that” (CS1), and a member of the IBD MDT: “is that person involved as a surgical representative in IBD

MDT?... I think that would be how it would work in most cases, with a genuine IBD surgeon”

(CS3). The ability for that surgeon to provide on-call cover was also considered necessary as:

“…you’d obviously want to have surgeons with a sub-specialist interest in providing the IBD, certainly elective, care and preferably offering a sub-specialist emergency service for the emergency work” (CS6).

The wider team (WT)

Participants agreed that further IBD MDT core members needed to deliver an IBD service provision would include a consultant radiologist, histopathologist and gastroenterologist:

“…you can’t make good decisions without first class radiology… you need someone who can interpret that well, or someone who is motivated to learn and expose themselves to a tight learning curve” (CS2), “…an experienced IBD pathologist is essential to getting (the)

190

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery diagnosis right…and you need good pathology input for assessment and pathology prior to consideration to surgery", “…a comprehensive IBD team is absolutely essential…with that comes adequately experienced radiologists, strong gastroenterologists, and I think some degree of sub-specialisation within that” (CS3). The involvement of an IBD nurse specialist was considered vital not just for the psychosocial considerations but delivery of the service overall “…I think an IBD nurse really binds the service, and it’s hard to provide that without an IBD nurse” (CS3), “…need to have an adequate number of nurse specialists and almost get them to run the service, you know, in terms of they can coordinate the service that... you know, get the clinicians to make the decisions…but in terms of the administration of it…you’ve got to rely on the CNSs to do that.” (CS8).

7.4.3.1.2. Surgical volume

Centralisation

Nine participants described the importance of the need to centralise surgical IBD services to ensure high quality. The need to be performing a critical minimal volume of procedures annually was considered necessary both for the individual consultant colorectal surgeon, as well as at an institutional level, “… a critical volume of IBD work is essential. The good decision-making, I believe, follows sufficient experience at making these decisions" (CS3).

There was however some uncertainty expressed as to the value of this critical volume required, and the impact this would have on smaller institutions: “…that’s my anxiety about all of this, because you can deliver a wonderful service that patients love but some bright spark will say:

‘Well, you’re not doing enough’.” (CS9).

191

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

7.4.3.1.3. Facilities

Capacity and services (CS)

The purpose of required facilities was to aid the provision of multidisciplinary working, and a suggestion that the required numbers should be stratified by the population: “…you could stratify either by population in brackets of 100,000 or 200,000 to 250,000 and 500,000” (CS2).

Fourteen participants agreed for the need of a dedicated gastroenterology ward: “… co-locating everyone is useful. Ward rounds involving the surgeon but also the physician in charge is useful” (CS9). Participants suggested that to ensure a quality IBD surgical service provision there would be a need for “…top of the range, high spec theatre facilities” (NS1), “…work up where there can be close interaction (with) endoscopy” (CG1), and interventional radiology to

"…optimise as best they can before the operation (and) drain any sepsis…” (CS3).

Joint care (JC)

Ten participants agreed for the need of joint clinics: “There needs to be, ideally, a joint clinic

…where you do have a gastroenterologist and a nutritionist next door, with all the nursing team available to ask questions" (CS5). An emergency 24-hour joint surgery & gastroenterology IBD on-call service: “For emergency patients the thing is we’re not equipped... patients that come up to our day unit, ‘Nurse, I’m feeling unwell, can I see a doctor?’ And they can’t see a doctor because the doctor’s assigned to a clinic or an endoscopy list, so we have to send them to A&E” (NS2).

192

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

Access and contactability (AC)

Twelve participants agreed that accessibility of the IBD MDT was essential and could be facilitated through an urgent referral pathway: “…for new patients, you would want… a two- week pathway such that they can get investigated if the GP is worried about something, and then have a very clear pathway of care to take them through to (specialist care), when IBD has been diagnosed” (CS9). Suggested means of access to care was also suggested by the need of an IBD ‘hotline’: “you don’t really want to have to go back in just through the accident and emergency route … It’s better if you could have some sort of direct point of contact either on the ward or through a CNS or through a helpline …” (SM4).

7.4.3.2. Stage 2: Delphi consensus

7.4.3.2.1. IBD Team (table 7.4)

A consensus on the members [median; IQR] that were considered very important (Likert rating

5) to be core members of the IBD team included a consultant colorectal surgeon who is trained and experienced in laparoscopic surgery [5; 0], a consultant radiologist [5; 1], a consultant gastroenterologist [5; 0], and a clinical nurse specialist with a special interest and competency in stoma therapy [5; 1]. It was considered important (Likert rating 4) for the team to include a consultant histopathologist [4; 1] as core member of the team.

193

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

Delphi Likert ratings The IBD team members required to Emergent (median; interquartile deliver a high quality IBD surgical service, themes range) include: Round 1 Round 2 a consultant colorectal surgeon trained and SuE 5; 1 5; 0 experienced in laparoscopic surgery. a consultant colorectal surgeon who is a core SuE 5; 0 5; 0 member of the IBD multidisciplinary team. a consultant colorectal surgeon who is on a SuE dedicated IBD on-call rota providing 4; 2 4; 2 Themes emergency IBD surgery when required. and items derived a consultant radiologist who is a core WT 5; 1 5; 1 from member of the IBD multidisciplinary team. semi- a consultant gastroenterologist who is a core structured WT 5; 0 5; 0 interviews member of the IBD multidisciplinary team. a clinical nurse specialist with a special WT 5; 1 5; 1 interest and competency in stoma therapy. a consultant histopathologist who is a core WT 4; 1 4; 1 member of the IBD multidisciplinary team. SuE a nominated IBD surgical lead. 4; 2 4 ;1 Footnote: SuE = Surgical expertise; WT = Wider team

Table 7. 5 IBD team members necessary for an IBD surgical service provision (shaded areas indicate items that reached consensus following two iterations) – There were 8 items submitted into the first Delphi iteration derived from emergent themes from stage 1: 4 items relating to surgical expertise (SuE), and 4 items relating to the wider team (WT). Consensus following two iterations, were achieved for 7 items.

7.4.3.2.2. Surgical Volume

There was consensus [i.e. over 70% agreement across the panel] that an individual consultant colorectal surgeon should be performing >15 major IBD resections per year (including proctocolectomy, subtotal colectomy, ileo-anal pouch formation & ileocolonic/small bowel

194

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

Crohn's surgery) [85%] (figure 7.3), and that an institution delivering a high quality surgical service should be performing >20 major IBD resections per year [95%] (figure 7.4).

Figure 7. 3 Pie chart demonstrating consensus on the number of major IBD resections performed per year by individual consultant colorectal surgeons to ensure a high quality service provision – Following the second iteration there was a 9.5% consensus on 5 to 10 major resections, 4.8% consensus on 11 to 15 major resections, 33.3% consensus on 16 to 20 major resections, and 52.4% consensus on more than 20 major resections performed per year by individual consultant colorectal surgeons.

195

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

Figure 7. 4 Pie chart demonstrating consensus on the number of major IBD resections performed per year by institutions to ensure a high-quality service provision – 95.2% of Delphi participants agreed that following two iterations more than 20 major resections should be performed per institution per year, and 4.8% of Delphi participants agreed that between 5 to 10 major resections should be performed per institution per year.

7.4.3.2.3. Facilities (table 7.6)

There was a consensus [median; IQR] that was very important (Likert rating 5) for an institution delivering a high quality IBD surgical service to a defined population of 250,000 to have 2 whole time equivalent consultant surgeons (dedicated IBD time) minimum [5; 0] , 1 dedicated laparoscopic operating facility [5; 0], 1 IBD-specific endoscopy facility with dye spray, double-balloon enteroscopic and dilatation capabilities [5; 0], 1 dedicated gastroenterology ward [5; 0], once a month minimum parallel or joint gastroenterology and

196

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

surgical IBD clinics [5 ; 0], 2 whole time equivalent consultant gastroenterologists [5; 0], 1.5

whole time equivalent clinical nurse specialists with competency in stoma therapy [5; 1], an

IBD advice line manned during working hours [5; 1], and a interventional radiology service

(for percutaneous intra-abdominal abscess drainage) [5; 1].

There was a consensus that it was important (Likert rating 4) for the institution to have

emergency 24-hour joint surgery and gastroenterology IBD on-call service [4; 1], and an urgent

referral pathway for new IBD patients (new referrals to be seen within two weeks) [4; 1].

An institution delivering a high quality IBD Delphi Likert ratings Emergent surgical service, for a defined population of (median; themes approximately 250,000 inhabitants, should interquartile range) have: Round 1 Round 2 2 whole time equivalent consultant surgeons CS 5; 1 5; 0 (dedicated IBD time) minimum. CS 1 dedicated laparoscopic operating facility. 5; 2 5; 0 1 IBD-specific endoscopy facility with dye CS spray, double-balloon enteroscopic and dilatation 5; 1 5; 0 capabilities. CS 1 dedicated gastroenterology ward. 5; 1 5; 0 an emergency 24-hour joint surgery & Items JC 4; 2 4; 1 derived gastroenterology IBD on-call service. from semi- an urgent referral pathway for new IBD patients AC 4; 1 4; 1 structured (new referrals to be seen within two weeks). interviews once a month minimum parallel or joint JC 5; 0 5; 1 gastroenterology and surgical IBD clinics. 2 whole time equivalent consultant CS 5; 1 5; 0 gastroenterologists. 1.5 whole time equivalent clinical nurse CS 5; 1 5; 1 specialists with competency in stoma therapy. an IBD advice line manned during working AC 5; 2 5; 1 hours.

197

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

an interventional radiology service (for percutaneous intra-abdominal abscess drainage). ─ 5; 1 Ω Footnote: CS = capacity and services; JC = joint care; AC = access and contactability Table 7. 6 Facilities necessary for an IBD surgical service provision (shaded areas indicate items that reached consensus following two iterations) – There were 10 items submitted into the first Delphi round derived from emergent themes from stage 1: 7 items detailing capacity and services (CS), 2 items detailing joint care (JC), and 2 items detailing access and contactability (AC). A further eleventh item (Ω) was submitted into the second Delphi round

(following deliberation within the study group). All items following the second iteration achieved consensus.

7.3.4. Clinical care processes

7.3.4.1. Stage 1: Semi-structured interviews

7.3.4.1.1. Pre-operative

Optimisation (O)

Ten participants shared the viewpoint that the pre-operative clinical processes should be focused on optimising the patient in preparation for surgery. Specifically, there is a need to optimise the patient’s nutritional and medication status: “These are sick patients, you just need to make them as fit as you possibly can by altering their nutrition, adjusting their medications as best you can, but not compromising their well-being in the run up to surgery. So it’s a fine balance” (CS4). Addressing nutritional optimisation was considered best facilitated by a referral to a dietitian review. A gastroenterology review to optimise the patient’s medication status was suggested in view of “…immuno-suppressants and surgery…aren’t good

198

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery bedfellows. But on the other hand, by the time they’ve reached the need for surgery they probably are going to be inevitably on some form of immunosuppressant…” (CS1).

Managing expectations (ME)

Twenty participants agreed that the pre-operative setting is an opportunity to manage the patient expectations. There was a need from patient participants to be fully informed about the operation and aftercare: “…give as much information and maybe also written information…maybe also put them in touch with a support group” (PP3). Those suggested to be in the best position to facilitate this is the IBD nurse specialist, alongside the surgical consultation: “…if any patient is referred to me for surgery following a discussion, I will involve my nurse specialist …it’s the case usually that the IBD nurse will come with them to the consultation (to address) the psychological aspects… social issues, childcare, fertility issues …” (CS5).

7.3.4.1.2. Post-operative

Shared care (SC)

Twelve participants suggested adherence to enhance recovery protocols as a marker of quality in the post-operative setting: “The well proven enhanced recovery pathways work very well with not just cancer and elderly patients but IBD patients, they should be involved” (CS5). Ten participants highlighted the importance of joint care as a marker of quality, facilitated either by a discussion of all post-operative patients in the IBD team meeting by “…making sure the best plan for surgical follow-up but also medical follow-up, locking the patient back into appropriate medical follow-up once the surgical episode is finished” (CS9), or by a

199

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery gastroenterology inpatient review to “…comment on further early management and timing of follow-up because you don't want them to enter a sort of void where all they're having is post- operative follow-up.” (CG3). Eight participants agreed in the importance of accurate clinical record facilitating joint care through shared care IBD bundles or booklets incorporating “… checklists for the sort of things you worry about, steroids and medications, DVT prophylaxis and bowel prep if people feel that’s appropriate, just make sure things are done as they should be” (CS9).

7.3.4.2. Stage 2: Delphi consensus

7.3.4.2.1. Pre-operative care (table 7.6a)

A consensus on the key performance indicator [median; IQR] that was considered very important (Likert rating 5) in pre-operative care was an outpatient gastroenterology review with a view to optimise the patient's medication status [5; 1]. A consensus on indicators that were considered important (Likert rating 4) were the presence of an IBD nurse specialist during surgical consultations to address social & psychological concerns and manage expectations

[4; 1], and a referral to a dietitian for optimisation of the patient's nutritional status where indicated [4; 1].

7.3.4.2.2. Post-operative care (table 7.6b)

A consensus on key performance indicators [median ; IQR] that were considered important

(Likert rating 4) in post-operative care included the use of a shared care IBD bundle / booklet

[4 ; 1], a discussion of all post-operative inpatients in the IBD team meeting [4 ; 1], and a

200

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery gastroenterology inpatient review to assess the need for medical prophylaxis after surgery [4 ;

1].

(a) Pre-operative considerations in Delphi Likert ratings Emergent the process of delivering high (median; interquartile themes quality IBD surgical care should range) include: Round 1 Round 2 the presence of an IBD nurse specialist during surgical ME consultations to address social & 4; 2 4; 1 Themes psychological concerns and manage and items expectations. derived from an outpatient gastroenterology review semi- O with a view to optimise the patient’s 5; 1 5; 1 structured medication status. interviews a referral to a dietitian for O optimisation of the patient’s 5; 1 4; 1 nutritional status where indicated. (b) Post-operative considerations required in delivery of a high quality IBD surgical service include adherence to enhanced recovery SC 4; 2 4; 2 protocols. Themes the use of a shared care IBD bundle / and items SC 4; 1 4; 1 derived booklet. from a discussion of all post-operative SC 4; 1 4; 1 semi- inpatients in the IBD team meeting. structured interviews a gastroenterology inpatient review to SC assess the need for medical 4; 1 4; 1 prophylaxis after surgery.

Footnote: ME = Managing expectations; O = Optimisation; SC = Shared-care

Table 7. 7 Themes, items and respective Delphi Likert ratings for (a) Pre-operative and (b) post-operative considerations in the processes of delivering high quality IBD care (shaded areas indicate items that reached consensus following two iterations) – There were 7 items submitted into the first Delphi iteration derived from emergent themes from stage 1: pre-operatively two

201

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery items detailing optimisation (O) and managing expectations (ME), and post-operatively 4 items detailing shared-care (SC). Following two iterations, consensus was achieved for six items.

7.4. Discussion

7.4.1. Summary of findings

This chapter has provided for the first time to our knowledge a consensus driven set of metrics that will make effective KPIs for the delivery of IBD surgery for patients with ileocolonic

Crohn’s disease. Measuring performance on an individual measure may fail to appreciate the complexity of a healthcare system, and risks failing to assess performance appropriately for a number of reasons (267). The process of measuring an individual outcome variable and using it as an indicator of quality may alter behaviour in order to improve just that single outcome without improving overall care (268). This is especially the case where there are trade-offs, for example between length of post-operative stay and readmission rates. Measuring variables from both sides of the trade-off may reduce the risk of this effect. (268) The KPIs for overall provision of an IBD surgery service include having a local registry of patients undergoing surgery for IBD, and submitting data to national databases or registries. Participation in national data collection and having robust collection locally have proved important as indicators of quality in their own right, with examples including The Bowel Cancer Audit (269) and The Inflammatory Bowel Disease Audit, (155) but also produce the data required to monitor performance in other areas. (86, 270)

A number of more frequently occurring adverse outcomes have been suggested as KPIs by this expert panel. Re-operation within 30 days was considered an important indicator for all the

202

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery intra-abdominal and pelvic operations. Development of short bowel syndrome and enterocutaneous fistulae are fortunately relatively rare complications of Crohn's disease, but the severity of impact on a patient's quality of life warrants their inclusion here. (271)

Some of the initial statements were rejected as useful KPIs by the Delphi panel. The rate of laparoscopic converted to open surgery did not meet the inclusion criteria. The use of laparoscopy and reasons for conversion are based on clinical and intra-abdominal circumstances (e.g. previous abdominal surgery). These findings may therefore be reflective of the questionable use of laparoscopic surgery as measure of surgical quality. Laparoscopic surgery for IBD does result in improved cosmesis and faster recovery than open surgery, (272,

273) and current guidance suggests it should be the treatment of choice in ileocolonic resection for Crohn's disease (51, 274).

The involvement of other members of the MDT (e.g. specialist IBD nursing staff) were recommended as important rather than very important indicators. This may in part represent the intentional decision not to include these core and extended members of the MDT on the expert consensus panel in order to focus the key performance indicators on the surgery itself

(275, 276). Involvement of the full range of members of the MDT is vital in the quality of an

IBD service as a whole,(277) is recognised in other guidelines,(51, 274) and will be picked up in audits of a unit's overall IBD performance, but for specifically assessing the quality of the surgical service there are other indicators which are more useful.

Some of the initial statements were rejected as useful performance indicators by the Delphi

203

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery panel. Rates of laparoscopic surgery, conversion rates, and the use of enhanced recovery programs divided opinion across the panel, with no consensus being reached as to their usefulness as performance indicators. All panel members felt it was important for laparoscopic surgery to be available within the core IBD team, and the lack of consensus therefore relates to either whether a laparoscopic approach is correct for all patients, or whether measuring rates of laparoscopic surgery is a useful measure of surgical quality. There are substantial benefits with laparoscopic surgery in Crohn’s disease. There is improved cosmesis and faster recovery than open surgery in the short-term. Furthermore, long-term benefits have been demonstrated with reduced adhesiolysis and incisional hernia surgery. (272, 273, 278) Current guidance suggests it should be the treatment of choice in ileocolonic resection for Crohn's disease. (51)

There are often valid clinical reasons why a case requires open rather than laparoscopic surgery.

Measuring laparoscopic and conversion rates may therefore have been felt to be an inadequate measure of performance.

Institutional and surgeon volumes have been shown to correlate with outcomes for a range of surgical procedures, including ileo-anal pouch surgery, yet pouch surgery volumes are low in many units, with 30% carrying out fewer than 2 pouches per year.(279) There is however no evidence of a relationship between surgical volume and outcomes in other areas of inflammatory bowel disease surgery. High volume does correlate to high quality performance.

The technical expertise required to operate on complex Crohn’s disease, and the peri-operative decision-making of when to operate and whether to fashion an anastomosis or to create a stoma require highly trained surgeons. The consensus of the panel was that an IBD unit should be carrying out more than 20 major intra-abdominal IBD operations each year, and that an individual surgeon should be carrying out more than 15. Although there is no UK national

204

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery consensus on what defines high volume, currently European consensus states high volume centres should perform at least 10 pouch formations per year. This would leave the remaining ten to include the other major IBD surgical procedures, including small bowel or ileocolonic resections for Crohn’s disease. (154, 280) This low number should be considered an initial threshold for the process of centralisation, supported by the necessary organisational structure and clinical care processes to support this. From the standpoint of the individual surgeon, colorectal surgeons carrying out cancer resections in the UK are required to carry out at least

20 cancer resections each year. (172) Depending on other clinical commitments it may not be realistic for a surgeon to achieve both of these targets, driving centralisation, and the development of IBD surgery as a separate subspecialty interest within colorectal surgery.

7.4.2. Limitations

A number of limitations need to be considered with this chapter. Firstly, those outcome measures that achieved consensus as either very important or important represent metrics in the ideal IBD surgical service provision. Many of these are undefined proportions and, unlike the clinical process KPIs, may be difficult to measure. Secondly, the study was limited by selection bias as there were only 21 respondents to a survey invitation sent to all the European

Crohn’s and Colitis Organisation members. Thirdly, coverage bias may exist considering representation from only 10 European countries and skewing of opinion if some countries are represented more across panellists then others. Fourthly, a lack of patient representation as

Delphi panellists questions the relevance or importance of these KPIs at a patient level. Further limitations relating to the qualitative purposive sampling, the process of member checking, the

205

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

Hawthorne effect, and limitation with Delphi including outlier analyses are discussed in section

9.2.

206

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

CHAPTER 8: ESTABLISHING THE AIMS, FORMAT AND FUNCTION FOR

MULTIDISCIPLINARY TEAM DRIVEN CARE WITHIN AN INFLAMMATORY

BOWEL DISEASE SERVICE: A MULTI-CENTRE QUALITATIVE EXPERT-

BASED CONSENSUS STUDY

8.1. Abstract

MDT driven care is evolving intuitively within the IBD setting. Providing a standardised framework, with a clear definition of key members and cases eligible for discussion, may enhance its capacity to establish effective quality improvement. The purpose of this chapter is to obtain an expert-based consensus on the aims, role of key specialists and case eligibility for

MDT driven care within an IBD service. This is a prospective, multi-centre, expert-based consensus study, using a Delphi formal consensus-building methodology online survey. An eligibility criterion was established to ensure panel members had recognised expertise in the field of IBD. Items were obtained from semi-structured interviews performed by the researchers. Panellists were asked to rank each item on a Likert scale which was categorised from 1 (=not important) to 5 (=very important). Consensus was defined with an IQR ≤ 1. Items with a median score > 3 were considered eligible for inclusion. Consensus for dichotomous responses was defined by an agreement of greater than 60% across panellists. A multidisciplinary sample of 24 experts were recruited as panellists. A consensus was achieved for 8 items that described the aims of the IBD MDT. A consensus for being a core IBD MDT member was demonstrated for 7 specialists. Seven items demonstrated consensus for case discussion eligibility in the IBD MDT. Consensus-derived statements can focus the IBD MDT for core members and aid a contractual recognition to ensure attendance and pro-active contribution.

207

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

8.2. Aims

The purpose of this study is to obtain an expert-based consensus on the aims, role of key specialists and case eligibility for MDT driven care within an IBD service.

8.3. Methodology

A two-stage, prospective, multi-centre, expert-based consensus study using a Delphi formal consensus-building methodology online survey was performed. Stage 1 involved item construction through semi-structured interviews. Items were then incorporated into stage 2, which used a Delphi to obtain consensus. The protocol for the study was reviewed by a by a

Research Ethics Committee in London, UK and approval was given prior to data collection

(Research Ethics Committee reference: 13YH 0175).

8.3.1. Stage 1: Semi-structured interviews.

As detailed in section 2.2.2, a full and comprehensive qualitative methodology was applied to this section of the study.

8.3.1.1. Procedure

Interviews were carried out using a standardised and piloted semi-structured interview protocol

(see Appendix iv) delivered by two trained interviewers of clinical background. The interview protocol explored key themes encompassing key elements for an effective IBD MDT, including an understanding of the role and purpose of the IBD MDT, structural inputs required for an effective IBD MDM, logistical considerations for an effective IBD MDM and what the overall

208

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery design of an effective IBD MDM should entail. Themes and were then refined and made into items by the research team and incorporated into the Delphi survey (see section 2.2.2 for data analysis).

8.3.1.2. Participants

The opportunistic multidisciplinary sample included consultant colorectal surgeons, consultant gastroenterologists, consultant gastrointestinal (GI) histopathologists, consultant GI radiologists and IBD nurse specialists, and were recruited from two separate regions in the UK

(the Southwest region and London), across multiple institutions. Participation was voluntary, and informed consent was obtained from all participants prior to data collection. Anonymity was ensured throughout the study.

8.3.2. Stage 2: Delphi consensus

As detailed in sections 2.3.1 and 2.3.2, a full and comprehensive Delphi methodology was applied to this section of the study.

8.3.2.1. Research procedure

The second stage was a prospective, multi-centre, study using a standard Delphi methodology.

The survey was emailed to a multidisciplinary specialist sample. Participants responded to statements about what aims of the IBD MDT should be, structural requirements for the IBD

MDM to function effectively, the role of key specialists’ and patients in the context of the IBD

MDT and eligibility of appropriate cases to be discussed. In addition, space for free text

209

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery comments was made available (see Appendix v for survey). Reminders were issued to all non- responders at 2 and 4 weeks after initial contact. The response rate was 63.8% (30/47).

8.3.2.2. Establishing the multidisciplinary expert consensus panel members

Forty-seven specialists (n = 47) were invited by members of the research team to participate in the first round of the Delphi process including 12 gastroenterologists, 13 colorectal surgeons,

7 radiologists, 7 IBD nurse specialists and 8 histopathologists. Thirty participants responded and included 6 gastroenterologists, 7 colorectal surgeons, 5 IBD nurses, 5 histopathologists and

7 radiologists. Participants were asked if they met any of the following eligibility criteria

(figure 8.1).

Figure 8. 1. Eligibility criteria for inclusion as a Delphi panellist.

210

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

This eligibility criterion was established to ensure the final panel members had recognised expertise in the field of IBD. This would also ensure credibility of the consensus study (204).

8.3.2.3. Survey design (see Appendix v for survey)

Section 2.3 of this thesis refers to software and materials used to design the survey. For this section, specifically, the definition of a ‘core member’ was a regular attendee with a contractual obligation to participate in the IBD MDT, an ‘extended member’ has a contractual obligation to contribute to the IBD MDT if invited to participate by a core member, and a ‘non-member’ being a someone can attend and participate, without obligation, if invited to by a core member.

8.4. Results

8.4.1. Participants

8.4.1.1. Stage 1: semi-structured interviews.

Thematic saturation was achieved after 28 semi-structured interviews, consisting of 6 consultant colorectal surgeons (CS), 6 IBD nurse specialists (NS), 7 consultant gastroenterologists (CG), 5 consultant GI radiologists (CR) and 4 consultant GI histopathologists (CP).

211

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

Histopathologists, 4, Colorectal 14% Surgeons, 6, 22% Radiologists, 5, 18% Colorectal Surgeons IBD Nurse Specialists Gastroenterologists Radiologists Histopathologists

IBD Nurse Specialists, 6, 21%

Gastroenterolo gists, 7, 25%

Figure 8. 2 Pie chart demonstrating relative proportion of participants within the multidisciplinary sample. There were 6 colorectal surgeons, 6 IBD nurse specialists, 7 gastroenterologists, 5 radiologists & 4 pathologists recruited.

Twenty responding experts were regular attenders to an IBD MDM within their institution.

These comprised of 6 IBD nurse specialists, 6 consultant gastroenterologists, 2 consultant GI histopathologists, 2 consultant gastrointestinal radiologists and 3 consultant colorectal surgeons. Seven experts described a variable attendance to the IBD MDM. They consisted of

2 consultant colorectal surgeons, 3 consultant GI radiologists and 2 consultant GI histopathologists. There was 1 consultant colorectal surgeon who did not attend an IBD MDM.

212

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

8.4.1.2. Multidisciplinary expert consensus panel members

The final 24 participants who were included into the Delphi Expert Consensus Panel are represented in table 8.1.

Years in Specialty Grade post Consultant 2 Consultant 5 Colorectal Consultant & Professor 11 Surgeon Consultant & Senior Lecturer 7 Consultant 13 Consultant 20 Consultant 29 Consultant & Senior Lecturer 7 Consultant 20 Gastroenterologist Consultant & Professor 22 Consultant & Professor 20 Consultant 8 Consultant 9 Consultant 9 Radiologist Professor 10 Consultant 12 Consultant 1 Consultant 17 Histopathologist Consultant 4 Consultant 5 Nurse consultant 10 IBD Nurse Nurse team leader 9 Specialist Nurse team leader 3 Nurse consultant 10

Table 8. 1 Members of the multidisciplinary Expert Consensus Panel – There were six consultant colorectal surgeons, six consultant gastroenterologists, five consultant radiologists, three consultant histopathologists and four IBD nurse specialists.

213

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

All expert panel members met the eligibility criteria (figure 8.1). They included 6 consultant colorectal surgeons, 6 gastroenterologists, 5 consultant radiologists, 3 consultant histopathologists and 4 IBD nurse specialists.

8.4.2. Aims of the IBD MDT

8.4.2.1. Stage 1: semi-structured interviews

Fifteen participants, across all healthcare disciplines (1 consultant surgeon, 4 consultant gastroenterologists, 3 consultant pathologists, 3 consultant radiologists and 4 IBD nurse specialists), thought there was a role for MDT-driven care in IBD. Perceived reasons for this included a recognition of IBD being a complex disease (n = 11) and decision-making not being straightforward (n = 10). One participant thought there was no role for the MDM in the care of patients with IBD because “resources, time and money are a huge issue; on that basis it’s very difficult in my mind to justify MDTs” (CS2).

Team-driven factors

Twenty-eight participants highlight predominantly team driven factors whilst describing the purpose of the IBD MDT. There was a recognition that IBD is complex with a necessary requirement for multidisciplinary input: “…you have to have multiple individuals who understand the patient and who understand the disease to be able to advise on the right options.

And there are multiple options available in IBD care and sometimes not all of the options are evidence-based and therefore a multidisciplinary approach and a consensus approach is always a good way of dealing with this…” (CS3).

214

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

The IBD MDT therefore provides a forum to obtain shared experience and expertise such that

“…the presence of an MDT where there’s multiple different professionals, all experienced, regularly seeing large groups of patients with similar complex disease should provide an enhanced assurance to the patient that their treatment options have been properly considered by an expert panel” (CR3). This in turn also provides a basis for support and shared decision- making: “potentially they were more junior consultants who looked for support in more difficult cases from the more senior members of the team, so it was a, sort of, formalisation of that support and mentoring” (CG8).

A further team-driven purpose of the IBD MDT was to provide a forum for research: “…it does because it is the place where you can discuss difficult patients and introduce the concept of trials for people who don't understand, or who aren't aware of some of the trials that are ongoing at the time… And so, it is the cornerstone of research” (CS4). The educational standpoint stems reflect back to its ability to provide shared experience and expertise: “…it produces a good format for education in decision-making in IBD” (CS4).

Patient driven factors

Twenty-three participants highlighted patient-driven factors when describing the purpose of the IBD MDT. Specifically, the need to advance patient care and deliver the best possible care for the patient: “…to bring the right people together to move patients forward, to prevent, sort of, letter-based ping pong: this person's struggling could you see them in clinic? They wait up

215

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery to four weeks then see a registrar who says, ‘I'll get my consultant to consider’. I think its key benefit is senior, timely, decisive decision-making” (CG2).

The need to improve patient outcome was also highlighted as a purpose of the IBD MDT. From a surgical standpoint, this would include minimising urgent or emergency surgery: “…to minimise the urgent operation and the emergency operations. I think that's it's chief role. We know that the outcomes are much worse in that group of patients” (CS5). Further to this, outcomes are improved through early recognition of disease, and preventing patients being missed: “there may be patients that haven't been seen for a while, for instance, who've slipped through the net, and it's actually quite helpful to bring them back to that discussion to say, you know, they've been stable for a year” (CG4).

216

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

NUMBER OF THEMES PARAMETER REPRESENTATIVE QUOTATIONS FROM MDT MEMBERS RESPONDANTS Question: Is there a role for an MDT in the care of patients with IBD? "IBD is a complex medical condition (that) involves the input from a wide range of IBD is complex 11 disciplines, hence 'multidisciplinary', and the decision-making is often not straightforward" CG7 YES (R) Requires "more than any other discipline IBD care has to be multidisciplinary. It revolves multidisciplinary 17 around controlling the disease so that the patients can get on with their lives. That input is the primary aim of management of IBD, be it medical or surgical" CS4 Question: If so, what should the purpose of the IBD MDT be? “…the presence of an MDT where there’s multiple different professionals, all To share collective experienced, regularly seeing large groups of patients with similar complex disease 19 experience/expertise should provide an enhanced assurance to the patient that their treatment options have been properly considered by an expert panel” CR3. To provide support “potentially they were more junior consultants who looked for support in more and shared decision- 22 difficult cases from the more senior members of the team, so it was a, sort of, TEAM making formalisation of that support and mentoring” CG8 DRIVEN To provide FACTORS "It engenders a team-working approach and where decisions are difficult it's consensus on 6 (T) always good to bounce ideas off of colleagues…" CS4 decision-making “…it produces a good format for education in decision-making in IBD” CS4 To provide a forum “…it does because it is the place where you can discuss difficult patients and for research and 5 introduce the concept of trials for people who don't understand, or who aren't education aware of some of the trials that are ongoing at the time… And so it is the cornerstone of research” CS4. PATIENT “…to bring the right people together to move patients forward, to prevent, sort of, DRIVEN To advance patient letter-based ping pong: this person's struggling could you see them in clinic? They 10 FACTORS care wait up to four weeks then see a registrar who says, ‘I'll get my consultant to (P) consider’. I think its key benefit is senior, timely, decisive decision-making” CG2 217

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

"...to give the best quality care to the patients...in IBD surgery it's to minimise Improve patient emergency operations. I think that's its chief role. We know that the outcomes are outcome and deliver much worse in that group of patients." CS4 15 the best possible “there may be patients that haven't been seen for a while, for instance, who've care slipped through the net, and it's actually quite helpful to bring them back to that discussion to say, you know, they've been stable for a year” CG4 MDT = Multidisciplinary Team; IBD = Inflammatory Bowel Disease; CS = Consultant Colorectal Surgeon; CG = Consultant Gastroenterologist Table 8. 2 Themes, parameters, number of respondents and representative quotations to the role and purpose of the MDT in IBD care from semi- structured interviews across 28 participants – In response to “should there is a role for the IBD MDT?”, one theme arose with two parameters. In response to “what should the purpose of the IBD MDT be?” two themes arose surrounding team and patient drive factors: team-driven factors encompassed four parameters and patient driven factors encompassed two parameters.

218

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

8.3.2.2. Stage 2: Delphi consensus

A consensus on items that described the aims of the IBD MDT that were considered very important (Likert rating 5) included to advance patient care, provide multidisciplinary input for the patient’s care plan, provide shared experience and expertise, improve patient outcome, deliver the best possible care for the patient, and to obtain consensus on management for a patient with IBD. A consensus on stems that were considered important (Likert ranking 4) included reducing emergency surgical procedures and providing a forum for research and education. Providing a safety net so patients are not missed obtained consensus for having some importance (table 8.3).

219

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

Likert rating (median; interquartile range) Round 1 Round 2 P advance patient care. 5, 4-5 5; 5-5 R provide multidisciplinary team input for the patient care plan. 5; 5-5 5; 5-5 Themes T provide shared experience and expertise. 5; 5-5 5; 5-5 and items P improve patient outcome. 5; 4.8-5 5; 5-5 identified P deliver the best possible care for the patient 5; 5-5 5; 5-5 from T obtain consensus on management for a patient with IBD. 5; 4-5 5; 4-5 semi- T provide a basis of support and shared decision-making. 5; 4-5 5; 4-5 structured P reduce emergency surgical procedures. 4; 3-4 4; 3-4 interviews T provide a forum for research and education. 4; 3-4.3 4; 3-4 P provide a safety net so patients are not missed. 4; 3-5 3; 3-4 Table 8. 3 Delphi Likert ratings for items that describe the aim of the IBD MDT – Ten items derived from semi-structured interviews and subsequent thematic analysis (see table 8.2) were incorporated into the Delphi and consensus for inclusion (Likert ranking > 3; IQR ≤ 1) was obtained for nine items (shaded blue and green). Seven items achieved consensus and ranked as ‘very important’ (shaded blue), two items achieved consensus and ranked as ‘important’ (shaded green).

220

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

8.4.3. The role of key specialists in the IBD MDT

8.4.3.1. Stage 1: item construction (figure 8.3).

Core members

Twenty-seven participants agreed the need to define the roles of those who attend the IBD

MDM. There was a need to define a core member: “Anybody who contributes a valuable opinion to the decision-making process of an IBD patient has to be a core member.” (CS4).

The core group were largely comprised on the following: “… a clinician with a specialist interest and a surgeon with a specialist interest at a meeting to make it quorate, if you like; a radiologist, obviously, again with a specialist interest; for sure, the IBD nurses. They would be probably the key individuals...” (CG1).

Extended members

There was also a need to recognise a non-core group who may still provide important information on patients on occasions: “you’d have core members and you’d have affiliates, and I think increasingly importantly you don’t have people just hanging around and wasting time and wasting resource. So, I think you decide what is appropriate for that patient mix in that centre...” (CR1). There was, however, some variation across participants as to which specialist would have an extended role: “I think the IBD MDT would be an opportunity to discuss dysplasia, so patients with UC and dysplasia, but then you need a histopathologist there and you might not have those cases every week, so does a histopathologist need to be there every week? Possibly not...” (CG1). In contrast, most histopathologist interviewees considered themselves core members: “The pathologist has to be there because if he has diagnosed it he can explain things” (CP4).

221

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

Patient representative

Fifteen participants discussed the need for patient representation. There was variation in determining who may be best placed to provide that: “…their input and their feelings should be represented, either by a specialist nurse, or by the clinician” (CG1). The patient representing themselves in the MDM was also mentioned: “…ideally the patient…there have certainly been MDTs where the patient’s been there, but I think that required a finesse” (CR1).

Facilitators were mentioned in ensuring the patient’s presence in the MDM: “probably the best way to do that, I would have thought, is through teleconferencing with a nurse perhaps present in the room with the patient guiding and supporting them” (CR5). Barriers were also highlighted in the logistical considerations in having the patient within the MDM: “the idea of patients waiting nicely outside on chairs… I don’t know. It may stifle both patients’ ability to communicate and also the team’s ability to have frank and honest discussions.” (CR5).

222

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

Figure 8. 3 Perceived responses (and number of respondents) as to who should be core members, extended members and non-members of the IBD MDT – The diagram demonstrates fifteen suggested core members of the IBD MDT (represented by the inner purple circle), six suggested extended members of the IBD MDT (represented by the outer larger green circle), and one suggested non-member (represented by the smaller outside blue circle).

223

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

8.4.3.2. Stage 2: Delphi consensus

After two iterations, a consensus for being a core IBD MDT member was demonstrated for colorectal surgeons [24/24], radiologists [24/24], gastroenterologists [24/24], IBD nurse specialists [24/24], dieticians [14/23], histopathologists [21/23] and the MDT co-ordinator

[21/24]. A consensus for being an extended IBD MDT member was demonstrated for the paediatrician [22/23], the research fellow [20/23], the junior doctor [20/23], the pharmacist

[18/23], the dermatologist [19/23], and the rheumatologist [19/23]. The patient was considered as a non-member of the IBD MDT [15/23] (table 8.4).

Delphi Panellist Responses N [%]

Extended Non- Core Member Member Member Colorectal Surgeon 24 [100] 0 [0] 0 [0]

IBD nurse specialist 24 [100] 0 [0] 0 [0]

Gastroenterologist 24 [100] 0 [0] 0 [0]

Radiologist 24 [100] 0 [0] 0 [0]

MDT Coordinator 21 [87.5] 3 [12.5] 0 [0]

Paediatrician 1 [4.3] 22 [95.6] 0 [0] structured interviews structured - IBD Research Fellow 1 [4.3] 20 [86.9] 2 [8.7] Junior Doctor/Resident 4 [17.4] 17 [73.9] 2 [8.7] Histopathologist 21 [91.3] 1 [4.3] 1 [4.3] Dietician 14 [60.8] 8 [34.8] 1 [4.3] Pharmacist 4 [17.4] 18 [78.3] 1 [4.3]

Dermatologist 0 [0] 19 [82.6] 4 [17.4] Items identified from semi from identified Items Rheumatologist 1 [4.3] 19 [82.6] 3 [13.1] Patient 1 [4.3] 7 [30.4] 15 [65.2]

224

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

Table 8. 4 Items and respective responses N [%] for specialist and patient role as core, extended or non IBD MDT members – Fourteen items were identified for inclusion within the Delphi and consensus (≥ 60%) was achieved following two iterations. A consensus on specialists considered core members are shaded in blue. A consensus for specialists considered extended members are shaded in green. The patient (shaded grey) was considered a non-member.

8.4.4. Structural and organisational requirements for an effective IBD MDM

8.4.4.1. Stage 1: item construction (table 8.5).

Attendance

Twenty-five participants highlighted the importance of attendance of core members to the IBD

MDM. The attendance of a consultant colorectal surgeon often highlighted as sporadic:

“Surgeons are always difficult to get if they’re not part of the core group, they’re just special guest stars if you will” (CS3). This was considered necessary to ensure a multidisciplinary contribution to the patients care plan and MDT outcome. Ensuring core members have a contractual obligation to the IBD MDT was suggested as a means of ensuring the attendance for core members: “It should be job-planned for unprotected time, much in the same way that cancer (MDT) is…” (CS1).

Multidisciplinary contribution and focused discussion

Eighteen participants highlighted the importance of a multidisciplinary contribution: “…a forum where you can have the input of multiple different specialities and different views from the same speciality, so maybe different consultant gastroenterologists, different consultant

225

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery surgeons, and then hopefully get as much expert opinion and then come to a reasonable consensus…” (CR2). Barriers to ensuring multidisciplinary contribution exist: “I think you’re always going to have personalities who are quieter in a bigger group.” (NS4).

A means of facilitating multidisciplinary contribution by assigning a chair was highlighted:

“that would help with the issue of whose patients get discussed etc. and to ensure that if there's an opinion that they feel it would be beneficial to hear that individual is heard… you have to have a chair....” (CG1). The option of alternating the chairperson was also suggested as a means of sharing MDT duties across the core member group: “Would you consider… deliberately alternating the chair so that it isn't the same person every week?" (CG1).

Further duties of the chairperson would include prioritising the patients being discussed: “…if there is an urgent decision, say a sick inpatient or what have you, then somebody will just say, look, I've got somebody I really need to discuss, can we discuss them first? And it works fine.

I guess as we get busier, which inevitably we will, then there may have to be a more organised approach to that aspect of it.” (CG1). Furthermore, highlighting the need to discuss those who may have been missed in the prior meeting because of time constraints: “you don’t want to overrun because everyone wants to get away somewhere else, and then it makes you just a bit more focussed if you just know you’ve got from 8 to 9 to discuss this today” (NS6).

The MDT Co-ordinator

Eighteen participants highlighted the importance of an MDT co-ordinator with designated administrative responsibilities to ensure the effective running of the IBD MDT and MDMs:

226

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

“There should be a coordinator who ensures that the cases are meeting all of the relevant bits of material brought together on the day for discussion, and that notes are taken and things and all the information is gathered into the system” (CP4). Their duties would include bringing cases to the MDM and ensure each case has a specific question to be addressed: “a more formalised process where you have to submit the names in advance, a bit like the cancer MDT, and the notes are got by a coordinator and the question is highlighted” (CG1). Facilitating the documentation was considered an important role, particularly with a view to ensuring an electronic recording of the MDT outcome in the patient clinical record: “documentation and a way of doing that easily, which should be audited, is something that I think we definitely should improve upon” (CG1). Furthermore, ensuring a correspondence of the MDT outcome is relayed to the primary care physician: “…communicate that in writing in the letter to the GP which is copied to the patient” (CG1).

It was also highlighted that additional duties of the MDT co-ordinator should include working and regularly maintained technological resources and a confidential meeting space: “You need some kind of IT that allows interpretation of X-rays, so with a radiologist there. Ideally, someone who can do a histology set-up … and access endoscopy…" (CS1).

227

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

NUMBER OF THEMES PARAMETER REPRESENTATIVE QUOTATIONS FROM MDT MEMBERS RESPONDANTS Question: What factors are required for an effective IBD MDT to occur? Contractual “It should be job-planned for unprotected time, much in the same way that cancer recognition of 17 (MDT) is…” CS1. MDT duties Ensuring "…surgical attendance can be quite variable.... we have radiology, pathology, attendance (EA) Preventing medical gastro and occasional attendance of colorectal surgery... we definitely sporadic 8 need to work towards getting more colorectal presence there because (of) joined attendance up decision-making." CG6 Proactive “I think you’re always going to have personalities who are quieter in a bigger multidisciplinary 19 group.” NS4. contribution “…if there is an urgent decision, say a sick inpatient or what have you, then somebody will just say, look, I've got somebody I really need to discuss, can we Multidisciplinary Prioritising cases 7 discuss them first? And it works fine. I guess as we get busier, which inevitably contribution and for discussion we will, then there may have to be a more organised approach to that aspect of it.” focused discussion CG1 (MDTcFD) "you’re dealing with a lot of personalities.... various methods (include) ... alternating the Chairperson so that it moves from one consultant to the next," NS4 Chair person 9 “that would help with the issue of whose patients get discussed etc. and to ensure that if there's an opinion that they feel it would be beneficial to hear that that individual is heard… you have to have a chair....” (CG1) The MDT co- “There should be a coordinator who ensures that the cases are meeting all of the ordinator as the 18 relevant bits of material brought together on the day for discussion, and that notes facilitator are taken, and things and all the information is gathered into the system” (CP4) Administration (A) Ensuring a "...in trying to make it an efficient process …there should be criteria of those specific question 13 patients who are clearly going to benefit …there should be a specific reason or a is addressed key question to be answered that can really only be answered by the MDT" CR2

228

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

Ensuring "MDT pro-forma, which is really useful and has everything about the patients all appropriate 9 on a couple of sheets with all of their charts, history and everything else and up-to- documentation date X-rays and everything else that we need, so that's really good." NS2. "...infrastructure that worked. It would be incredibly frustrating if that were Ensuring working inadequate in any way... from a radiologist’s perspective really good facilities for and regularly reviewing cases. ...so just a workstation with a means of beaming that...the maintained 26 pathologist would have their stipulations around microscopes and all that sort of technological thing...you need a coordinator, somebody who can record things. You want resources. previous results flashed up." CR5 MDT = Multidisciplinary Team; IBD = Inflammatory Bowel Disease; CS = Consultant Colorectal Surgeon; CR = Consultant GI Radiologist; CG = Consultant Gastroenterologist, NS = IBD Nurse Specialist; CP = Consultant Pathologist

Table 8. 5 Themes, parameters, number of respondents and representative quotations to factors required for an effective IBD MDT to occur, from

semi-structured interviews across 28 participants – There were three themes that arose from the interviews: 1. Ensuring attendance (A), with two

parameters detailing the prevention of sporadic attendance and a contractual recognition of MDT duties; 2. Multidisciplinary contribution and

focused discussion (MDTcFD), with three parameters including the need for a chairperson to facilitate multidisciplinary contribution and

prioritising cases for discussion; 3. Administration (A), with four parameters including the need of an MDT co-ordinator to facilitate case

submission with a highlighted question, documentation and working and regularly maintained technological resources.

229

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

8.4.4.2. Stage 2: Delphi consensus

A consensus on items that described structural and organisational requirements for an effective

IBD MDM that were considered very important (Likert rating 5) included a clear electronic documentation of the MDT discussion outcome in the patients clinical records, organisational recognition of the IBD MDT and implementation into the job plan for core members, a specific question to be addressed, a designated MDT coordinator with designated administrative responsibilities, working and regularly maintained technological resources. A consensus on items that were considered important (Likert rating 4) were a confidential meeting space (bleep free environment, away from public or clinical areas), a chair person who is also a core member of the IBD MDT, a priority to discuss urgent cases and/or IBD inpatients first, a priority to discuss patients who were missed at the previous meeting and a letter to the patient and primary care physician detailing the outcome of the MDT discussion (table 8.6).

230

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

Likert rating (median; To ensure the IBD MDM runs smoothly and interquartile range) effectively, there needs to be: Round 1 Round 2

clear electronic documentation of the MDT 5; 4-5 5; 4-5 A discussion outcome in the patients’ clinical records. organisational recognition of the IBD MDT and 5; 5-5 5; 5-5 EA implementation into the job plan for core members. MDTcF 5; 4-5 5; 4-5 a specific question to be addressed. D a designated MDT coordinator with designated 5; 4-5 5; 4-5 A administrative responsibilities. working and regularly maintained technological 5; 4-5 5; 4-5 A resources. a confidential meeting space (bleep free 4; 4-5 4.5; 4-5 A Themes environment, away from public or clinical areas). and Items a chair person who is also a core member of the IBD - 4; 4-5 identified - MDT. ¥ from semi- structured MDTcF a priority to discuss urgent cases and/or IBD 4; 4-5 4; 4-5 interviews D inpatients first. MDTcF a priority to discuss patients who were missed at the 4; 3-5 4; 4-5 D previous meeting. a letter to the patient and primary care physician - 4; 4-5 - detailing the outcome of the MDT discussion. ∞ a submission of clinical cases no later than three 3; 3-4 3; 3-3 A working days in advance. a need to alternate chairing responsibilities across - 3; 3-4 - IBD MDT core members. α an alternating chair person who is 4; 4-5 - MDTcF trained/experienced in chairing MDT meetings and D who is also a core member of the IBD MDT. § Table 8. 6 Delphi Likert ratings for items ensure the IBD MDM runs smoothly and effectively

– Ten items derived from semi-structured interviews and thematic analysis (see table 8.5) were

231

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery incorporated into the first iteration of the Delphi process. Item § was separated into two different items (α and ¥) and incorporated into the second Delphi iteration. A further item (∞) was added into the second Delphi iteration. Following the second iteration, consensus for inclusion (Likert ranking > 3; IQR ≤ 1) was obtained for ten items (shaded blue and green).

Six items achieved consensus and ranked as ‘very important’ (shaded blue), four items achieved consensus and ranked as ‘important’ (shaded green). Two items achieved consensus for ‘some importance’ without inclusion (shaded grey).

8.4.5. Eligibility criteria for case discussion

8.4.5.1. Stage 1: item construction (table 8.7).

Safety-netting

Twenty-two participants highlighted the benefits and disadvantages of defining eligibility criteria to identify cases for discussion by the IBD MDT. Disadvantages to case selection were raised: “…you run the risk of missing cases, especially if some people function very much to protocol; if it doesn't fit the criteria of the protocol, they'll avoid bringing those cases and that might do more harm than good for your patient." (CS4). Instead of an eligibility criterion, cases should be brought at the discretion of the team: “… the team should be given the option to submit a case which he thinks deserves to be discussed. If we did that then we’d catch them all.” (CP4). This could be further facilitated by being a consultant led selection processes, not avoid unnecessary referrals into the IBD MDT: “…the cases to be brought should always be selected by the most experienced or the senior consultants, so the … let’s say specialist registrar, we will meet to discuss it, to see whether this is worth discussing, rather than becoming just another case, and with no specific question in mind” (CP2).

232

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

To further facilitate a safety netting approach, and to ensure patients are not missed, participants suggested protocols that ensure all patients are discussed. Suggested means would be to ensure that all IBD cases are discussed at least once a year: “…physicians have codes for surveillances reviewed regularly so the lists to scrutinise people, people don’t drop off … so that would be a task of the MDT to make sure that the follow-up for people in the patch is appropriate for them.” (CS1). Furthermore, a discussion of all new diagnoses which would have its role certain cases of IBD where management algorithms may differ: “Crohn's: you could discuss all new diagnoses. All ileocolics…or new presentations ileocolics…” (CS6).

Timely discussions

Twenty-one participants agreed that there should be a selective process with a view to facilitating time-keeping and ensuring appropriate utilisation of the IBD MDT. Specifically it was decided that cases should be referred to the IBD MDT if there is a change or step up in management: “You could say that if they are going to incur a step change in cost and personal risk, so either surgery or anti TNF’s, at that point the MDT need to sanction that and review it to make sure that the right decision is made” (CR3). Other justifications include complexity of the case at hand, which may include those who are a consideration for surgery or following surgery to review histopathology and decide on appropriate post-operative medical therapy:

“so patients who are post-op, you go through all of the pathology. You know, all inoperative specimens get reported by the pathologist and then an MDT decision is made as to the most appropriate follow-up therapy and who's going to see them in the clinic." (CS6). The need for an eligibility was highlighted in view of time constraints, and suggested eligibility criteria included: “candidates who are new diagnoses, inpatient severe, patients coming to surgery,

233

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery patients’ post-surgery looking for surveillance, and I think down to your time constraints you'd, kind of, want to look at it almost in those categories.” (CS6). As further means to facilitate timely discussions with the IBD MDT, it was suggested to split the IBD MDT, so that patients on biological therapy are discussed separately. The problem however is the lack of multidisciplinary input this could raise: “…we’ve got a biologics meeting which doesn’t have surgeons there …maybe that could be helpful to put every anti-TNF patient through the MDT in case the surgeons felt that they could offer an alternative maybe” (NS5). A further means to facilitate time-keeping would be to set a threshold of cases to discussion, with participants suggesting thresholds as low as 5 cases per meeting to 12 cases per meeting: “We’d probably only be able to discuss 12 patients per meeting and that’s about it. And there’s no point in discussing people who are doing fine or who have no particular issues to discuss.” (CS3).

234

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

NUMBER OF THEMES PARAMETER REPRESENTATIVE QUOTATIONS FROM MDT MEMBERS RESPONDANTS Question: For the IBD MDT to run effectively should there be a selection criterion for cases? “…you run the risk of missing cases, especially if some people function very much to protocol; if it doesn't fit the criteria of the protocol, they'll avoid bringing those cases and that might do more harm than good for your patient." CS4 Team or "… the team should be given the option to submit a case which he thinks deserves consultant 5 to be discussed. If we did that then we’d catch them all.” CP4 discretion “…the cases to be brought should always be selected by the most experienced or the Safety senior consultants, so the … let’s say specialist registrar, we will meet to discuss it, netting (SN) to see whether this is worth discussing, rather than becoming just another case, and with no specific question in mind” CP2. “…physicians have codes for surveillances reviewed regularly so the lists to scrutinise people, people don’t drop off … so that would be a task of the MDT to All cases 6 make sure that the follow-up for people in the patch is appropriate for them.” CS1 discussed “Crohn's: you could discuss all new diagnoses. All ileocolics…or new presentations ileocolics” CS6 “You could say that if they are going to incur a step change in cost and personal risk, so either surgery or anti TNF’s, at that point the MDT need to sanction that and review it to make sure that the right decision is made” CR3. Case complexity “…so, patients who are post-op, you go through all of the pathology. You know, requiring all inoperative specimens get reported by the pathologist and then an MDT disciplinary 21 Timely decision is made as to the most appropriate follow-up therapy and who's going to management discussions see them in the clinic." CS6. decisions (TD) “…we’ve got a biologics meeting which doesn’t have surgeons there …maybe that could be helpful to put every anti-TNF patient through the MDT in case the surgeons felt that they could offer an alternative maybe” NS5 Limit threshold “We’d probably only be able to discuss 12 patients per meeting and that’s about it. of cases per 3 And there’s no point in discussing people who are doing fine or who have no MDM particular issues to discuss.” CS3.

235

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

Table 8. 7 Themes, parameters, number of respondents and representative quotations to the need of an eligibility criterion for case selection for the IBD MDT, from semi-structured interviews across 28 participants – There were two themes that arose from the interviews: 1.

Safety-netting, with two parameters detailing need for consultant or team discretion in bringing cases to the IBD MDT and the need to ensure all cases are discussed at least once by the IBD

MDT; and 2. Timely discussions within the IBD MDM, facilitated by two parameters detailing the need of an eligibility criteria for specific cases and a minimum number of cases for discussion per MDM.

8.4.5.2. Stage 2: Delphi consensus

Following two iterations, a consensus on items deemed important (Likert rating 4) as eligible for discussion in the IBD MDM included complex cases requiring surgery, all patients on biologics, all new diagnoses, and all patients who have undergone recent IBD surgery (table

8.8).

236

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

Likert rating (median; interquartile range) The eligibility criteria for selecting patients for discussion in the IBD MDT meeting should include: Round 1 Round 2

TD complex cases requiring surgery 5; 4-5 4; 4-5 Themes and TD all patients on biologics 4; 3-5 4; 3-4 items SN all new diagnoses 4; 3-5 4; 3-4 identified ― all patients who have undergone recent IBD surgery α ― 3.5; 3-4 from semi- structured ― nothing – no need for an eligibility criterion ∞ ― 5; 3-5 interviews SN any case at the discretion of the named IBD physician 5; 4-5 4; 3-5 SN all cases discussed once a year 2; 2-3 2; 2-3.7

Table 8. 8 Delphi Likert ratings for items ensuring eligible cases for discussion in the IBD MDM - Five items were incorporated into the first iteration of the Delphi from semi-structured interviews and subsequent thematic analysis (see table 8.7). A further two items (α and ∞) were incorporated into the second iteration of the Delphi. Consensus for inclusion (Likert ranking > 3; IQR ≤ 1) was obtained for four items (shaded) and ranked as ‘important’ (shaded). Three items did not achieve consensus.

237

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

8.4.6. Outcome measures for the IBD MDT

8.4.6.1 Stage 1: item construction (table 8.9)

Fourteen participants highlighted the need for a feedback process to reflect on outcomes for the IBD MDT: “…there should be feedback. I think decisions need to be recorded and outcomes looked at periodically, and that will be a function of the MDT, as it were, business meeting, but feeding back to the MDT about the quality of decisions, what was found, correcting radiology aspects, so there needs to be a feedback element to it” (CS1). Parameters to measure focused around the organisation structure of the IBD MDT, and patient outcome:

“there should be a means of us auditing or feeding back in some governance type way of decisions that were made previously and learning from those because otherwise we’re never going to be able to improve, and we haven’t actually addressed that” (NS4).

238

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

NUMBER OF THEMES PARAMETER REPRESENTATIVE QUOTATIONS FROM MDT MEMBERS RESPONDANTS Question: Is there a measure the effectiveness of the IBD MDT? Measuring the “…it looks at how many patients who have had decisions made, whether implementation those decisions were taken up…So it all comes down to audit...” CR5 of MDT “...there ought to be some kind of audit to see whether these decisions are decisions actually implemented…” CP4 Measuring "…have the waiting time directives been breached or met?... the hardest part patient waiting is showing people who don’t really understand IBD, the people who times from essentially govern the resources, about how much better the decision is that submission to will be made if a group looks at a particular problem rather than if they Organisational structure (O) 5 MDT discussion don’t." CS3 "If you decide that it’s essential to have a surgeon there and they are missing every second meeting, that is a major issue… insisting that those Measuring core members are there at every meeting otherwise … those patients who are attendance of discussed that week are not having the full attention they deserve". CR5 core members "…measure against infrastructure things that you think are important...How many clinicians do they actually have from the allied disciplines who are actually in there with an IBD focus? Attendance...." CS5 “…it must set aside a meeting where it actually looks at the effectiveness of the last three months/six months/year… (and audit) whether the patient has Measuring got better…” CR5 patient outcomes “…IBD MDT has to look at both new patients and follow-up patients Patient outcomes (P) 11 and impact on because realistically its role will be to find if we achieve good outcomes for quality of life our patients and particularly the quality of life...It's essentially about actually looking at what's discussed and what are the decisions that come out of those and what are the outcomes.” CS4

239

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

Table 8. 9 Themes, parameters, number of respondents and representative quotations for outcome measures of the IBD MDT, from semi-structured interviews across 28 participants –

There were two themes that arose from the interviews: Outcome measures based on 1. the organisational structural requirements of the IBD MDT, with three parameters including a measure of the implementation of decisions, the measure of waiting times associated with the

MDT discussion of patients and a measure of attendance of core members; and 2. patient outcome, including a measure of individual patient outcomes following MDT discussions.

8.4.6.2 Stage 2: Delphi consensus

A consensus on items for the outcome measures for an IBD MDT important (Likert rating 4) included a record of attendance of all designated core members of the IBD MDM, a measure of elective versus emergency IBD surgery - to ensure early IBD MDM leads to planned surgery, and a record of the number of weeks awaiting case discussion following case submission for

MDM. A consensus on items considered having some importance (Likert rating 3) includes a record of the number of weeks from MDM to surgical review - if requested (table 8.10).

240

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

Likert rating (median; interquartile range) To ensure the IBD MDM runs smoothly Round 1 Round 2 and effectively, there needs to be: O a record of attendance of all designated 5; 4-5 4; 4-5 core members of the IBD MDM. Themes O a measure of elective versus emergency and items IBD surgery - to ensure early IBD MDM 4; 3-4 4; 3-4 identified discussion leads to planned surgery. from O a record of the number of weeks awaiting semi- case discussion following case 4; 3-5 4; 3-4 structured submission for the MDM interviews O a record of the number of weeks from MDM discussion to surgical review - if 3; 3-4 3; 3-4 requested.

Table 8. 10 Delphi Likert ratings for the outcome measure of an IBD MDT – Four items were incorporated into the Delphi from semi-structured interviews and subsequent thematic analysis

(see table 8.9) and following two iterations, consensus for inclusion (Likert ranking > 3; IQR

≤ 1) was obtained for three items (shaded green) and ranked as ‘important’ (shaded green).

One item achieved consensus and ranked as ‘some importance’ (shaded grey).

8.5. Discussion

8.5.1. Summary of findings

This chapter has provided consensus derived statements for the aims, structural and organisational requirements, and outcome measures for effective multidisciplinary team driven care for patients with ileocolonic Crohn’s disease. The strengths of this study lie in its multi- centre design, two-stage qualitative approach, the inclusion of IBD experts across multiple clinical specialities and allied care disciplines as panellists, and the utilisation of multiple

241

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery rounds (iterations) to achieve consensus. The identification and incorporation of items from a semi-structured interview add to the robustness of the qualitative approach used (276).

Consensus derived primary and secondary aims of the IBD MDT have been defined in this chapter. The primary aim is to deliver the best possible care for the patient, improve patient outcome, and advance patient care by providing multidisciplinary team input for the patient care plan, through a basis that provides support, shared experience and expertise, and shared decision-making. Secondary aims are to reduce emergency surgery and provide a forum for research and education. This consensus derived definition provides a focus for key specialists, enhancing the MDT as a tool that delivers a high quality IBD service provision.

In recognition of the need to define key member roles, this chapter has demonstrated that colorectal surgeons, radiologists, gastroenterologists, IBD nurse specialists, dieticians, histopathologists and the MDT co-ordinator should all be considered ‘core’ members of the

IBD MDT, such that they have a regular contractual obligation to participate in the IBD MDT.

Extended members, or those who are invited to participate and contribute to the IBD MDT, without a contractual obligation, include the paediatrician, the research fellow, the junior doctor, the pharmacist, the dermatologist, and the rheumatologist. Definitions for core and extended members are in keeping with current standards (154). Providing clarity on the role of the core and extended member in this context will aid managerial and contractual recognition and implementation into the job roles, particularly for core members.

242

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

The IBD patient was a non-member of the IBD MDT, in that they should have no participation or contribution to the IBD MDT, unless invited to by a core member. Considering the goal of multidisciplinary team care should be patient-centred, the recognition of the patient as a non- member of the IBD MDT provides a paradoxical dynamic to the meeting. A previous study in the setting of cancer has demonstrated patients have limited opportunities to input or influence the decision-making process in MDMs. Reasons for this include patients having inconsistent information and MDT members having variable definitions for patient centeredness in the context of MDTs (281). Patient involvement within the MDM may be possible as they enter at the time of their case being discussed. Potential drawbacks to this include a restriction of the free-flow of information, and limited understanding of medical terminology (282). This is perceived to impact on the level and pace of discussion (276). It still remains controversial, considering patient choice, and additional information, available after the MDT has been demonstrated as a major factor in the discordance of treatment decisions following MDT discussion within the MDM (283, 284). Considering that MDTs represent a costly intervention, patient involvement, through representation by a key patient advocate, in the

MDT may be necessary to improve concordance to MDT treatment decisions and ensure economical returns (285-287).

The importance of good attendance was also highlighted across interviewees as a necessity for the establishment of an effective IBD MDT, to facilitate effective decision-making. From our participants, consultant colorectal surgeons admitted to being variable or non-attenders to the

IBD MDT. Frustrations were also raised by consultant gastroenterologists with the variable attendance of certain specialists, particularly consultant colorectal surgeons, and the impact this has on shared decision-making. The respective experts were also viewed as core members

243

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery of the IBD MDT (Figure 8.3). Non-attendance of key members is a barrier to effective decision-making in the IBD MDT setting (288, 289). The consensus statement to ensure organisational recognition of the IBD MDT and implementation into the job plan for core members, will facilitate attendance records to the MDM.

The interviews highlighted a requirement for multidisciplinary contribution for an effective

IBD MDT to occur. Studies in UK based cancer MDTs, which have been implemented for almost 20 years, have shown contributions from physicians outweigh those from nurses (290) and an overall bias of the MDT towards the biological side of cancer at the expense of the psychosocial circumstances of the patient – which are often left to the surgeon who sees the patient in clinic to manage alone. The presence of different personalities within the IBD MDT was viewed as source of varying contribution. A means to improve this was by introducing an alternating formal chair person who can lead discussions and ensure multidisciplinary contribution. Consensus for the need of a chair person who is also a core member of the IBD

MDT was determined, but not for the need to alternate the chairperson. The later findings may be in recognition of the skill requirements and training of a good chairperson (291).

An emergent theme that arose was the need for a selection process to limit the demands on the

IBD MDT. Numerous suggestions were made including discussing no more than five patients in one session, splitting the meeting so those on biological therapy are discussed separately, or discussing only ‘complex’ IBD cases, with complex IBD defined by one participant as those requiring surgery. Providing eligibility criteria can ensure a focus for core members to discuss appropriate cases. This chapter has provided a consensus derived statement that patients

244

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery eligible for discussion in the IBD MDT should have complex IBD requiring surgery, be on biologics, be newly diagnosed with IBD, or have undergone recent IBD surgery. This provides a focus for core members, enhancing the MDT as a tool that delivers a high quality IBD service provision. This eligibility criterion would ensure all candidates for surgery are considered, and ensures patients remain in remission following surgery with post-operative medications. It does however run the risk of not allowing for clinician discretion, and missing cases that may not meet the criterion for MDT discussion.

Measuring the effectiveness of the MDT through audit of organisational structure and patient reported or clinical outcomes was highlighted by interviewees. A consensus on outcome measures included a record of attendance of all designated core members of the IBD MDM, a measure of elective versus emergency IBD surgery (to ensure early IBD MDM discussion leads to planned surgery), and a record of the number of weeks awaiting case discussion following case submission for the MDM. There is a paucity of evidence on how best to measure the effectiveness of an MDT, but further reported measures include the implementation of an MDT decision, and the ability of an MDT to reach a decision (292, 293). The measure of clinical outcome measure may have a role, but are subject to a number of disease and treatment related factors, and should be looked at in conjunction with other short term measures including the time from first referral to diagnosis, time from diagnosis to the first treatment, and costs reductions (294).

245

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

8.5.2. Limitations

Limitations need to be considered with regards to the current chapter. Firstly, although we have characterised the role of core members, further detail is required as the role of the chairperson to guide the meeting and ensure proactive contribution from all members.

Secondly, the inclusion of IBD patients into the survey may have provided a useful viewpoint in term of how best to represent their opinions to ensure the MDT remains patient-centred.

Thirdly, the duration of the interviews were short and lacks ‘prolonged engagement’ which is a recognised criterion to ensure credibility in qualitative research (197). MDT-driven care is arising throughout several European IBD centres (185). Representation from IBD experts from these centres may have provided useful international perspective. Further limitations relating to the qualitative purposive sampling, the process of member checking, the Hawthorne effect, and limitation with Delphi including outlier analyses are discussed in section 9.2.

246

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

CHAPTER 9: THESIS DISCUSSION AND FUTURE WORK

9.1. Summary of thesis

This thesis provides a description of the incidence and possible causative factors of post- operative intra-abdominal septic complications and clinical recurrence following ileocolonic resection within a specialist IBD institution over a six-year period. It goes on to explore the role of endoscopic balloon dilatation in avoiding surgery. The complexity and variability in disease phenotype for ileocolonic Crohn’s disease has demonstrated a need to categorise and validate these into simplified stages. Furthermore, qualitative techniques have been used to obtain a consensus on outcome measures and components of multidisciplinary team driven care to improve the overall quality of care for patients undergoing ileocolonic resection for Crohn’s disease.

Chapter’s 3 and 4 sought to validate the Montreal classification against short-term and long- term post-operative outcomes following ileocolonic resection for Crohn’s disease. Its strengths lie in the database construct and subsequent patient selection through a specific eligibility criterion limiting selection bias. Both Chapters demonstrated that there is much diversity in populations across the cohort in terms of patient and disease related variables, pre-operative and peri-operative variables, and post-operative outcomes. Chapter 3 demonstrated the presence of an intra-operative intra-abdominal septic bed (Montreal B3), and peri-operative biologics as independent predictors of post-operative intra-abdominal septic complications.

The former finding is in keeping with numerous studies including that by Hulten (108). It highlighted the need to further discriminate between fistulating and perforating disease in the pre-operative or peri-operative setting (addressed in Chapter 6). Chapter 4 provided an analysis of variables impacting on post-operative clinical recurrence and demonstrated re-resectional

247

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery surgery for recurrent disease as an independent predictor for early recurrence. It also demonstrated that ileocolonic (Montreal L3) disease rather than ileal disease (Montreal L1) is an independent predictor for the development of clinical recurrence. This was a novel finding and suggested the need to incorporate disease location into risk stratification models for gastroenterologists to start medical therapy post-operatively.

Chapter 5 sought to investigate the outcomes of endoscopic balloon dilatation, and its role in the avoidance of surgery in patients with ileal Crohn’s disease strictures. The strength of this review lies in the systematic manner in which it was conducted in accordance with PRISMA guidelines and methods for narrative reviews (295). It demonstrated variation in disease activity being either predominantly inflammatory or predominantly fibrotic. Furthermore, through meta-analysis, it demonstrated that 75% of patients undergo surgery within a five year follow-up following endoscopic balloon dilatation. It questions the long-term efficacy of endoscopic balloon dilatation in avoiding surgery in patients with small bowel stricturing

Crohn’s disease. It strengthens the argument that many patients require surgery despite medical management. Considering there is a symptomatic response rate of 70.2% in the short term, the long-term role of endoscopic balloon dilatation in this setting may need to be redefined.

Chapter 6 sought to construct and validate a staging tool for ileocolonic Crohn’s disease to adequately categorise pre-operative disease phenotypes. Its strengths lie in its mixed- methodological approach and triangulation, through literature review and semi-structured interviews, and providing validity and reliability data through quantitative techniques from the database (296, 297). The difficulty in developing a risk stratification model lies in the

248

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery variability in disease phenotype observed at surgery as observed in Chapters 3 and 4. Although

Montreal B3 is a predictor of post-operative IASC, it is unclear if fistulating disease and/or perforating disease phenotypes are contributory to this. The Working Group of Montreal did not include colorectal surgical expertise (49). The staging tool therefore has several implications. Clinical applications include its ability to pre-operatively counsel patients on the likelihood of operative strategy, including stoma formation, concomitant sigmoidectomy and post-operative morbidity. Research applications include analyses of patient and disease related factors per stage, including a possible association with male gender and perforating disease.

Furthermore, surgical outcome analyses can adjust for disease complexity. In the event of open reporting of outcomes, as witnessed in colorectal cancer surgery, this staging tool can allow for case mix adjustment if imbedded into national audit datasets.

Chapters 7 and 8 sought to construct metrics to monitor and enhance the quality for an IBD surgical service provision and multidisciplinary team care. The strength in these studies are in their dual qualitative methodologies, national (Chapter 8) and international (Chapter 7) multi- centre design. There is a growing argument for centralising IBD surgery with a view to improving the quality of care for these patients. Chapter 7 provided a consensus on procedure- specific outcome measures for ileocolonic Crohn’s surgery and outcome measures for the overall quality of an IBD surgical service provision. Chapter 8 provided consensus-derived statements for the role of key specialists in the context of the IBD MDT and provided a focus for core members to discuss appropriate cases through the construction of eligibility criteria.

These consensus-derived statements can aid a contractual recognition of responsibilities for core members to ensure attendance and pro-active contribution.

249

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

9.2. Limitations

This thesis has some limitations. In Chapters 3, 4 and 6, the capture of pre-operative medications use within four weeks of surgery may not necessarily highlight its impact on post- operative outcome, as patients may have been on this prior to the four-week pre-operative period and stopped within, or before, this period. The relatively small sample size within

Chapters 3, 4 and 6 would have limited the statistical power, leading to type 2 statistical errors.

Efforts were made to limit this effect and improve the robustness of data with a validation process during the extraction phase (see section 2.1.5). Despite this, p values greater than 0.05 still cannot imply a definitive insignificant association.

Blood parameters extracted from the database required conversion from numerical to categorical form as they were analysed in Chapters 3, 4, and 6. The ‘cut-offs’ (e.g. serum albumin defined as low when < 25 g/L) were based on the reporting of normal and abnormal reference ranges and previously published reports. Regarding hypo-albuminaemia, higher cut- offs have also been previously reported in the literature (< 30 g/L). The lower threshold was chosen as it was more reflective of the state of disease severity seen within a tertiary centre.

Sensitivity testing, however, may have provided more robustly defined cut-offs.

The retrospective and non-random design of Chapters 3, 4 and 6 expose them to sample selection bias. The reporting of missing data and its handling is generally poor in across healthcare literature (298). Alternative measures of managing missing data could have been employed to prevent such bias. Multiple imputation provides an estimate of missing data and can improve the validity of findings (299). Adherence to STROBE (Strengthening the

250

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

Reporting of Observational Studies in Epidemiology) guidelines can provide further quality assurance within cohort studies.

In Chapters 6, 7 and 8, the process of member-checking during semi-structured interviews, whilst considered a good practice by some, it does carry the potential risks of changing the spontaneous response offered during the interview process (300) (although this was not observed, as none of the participants edited or amended their prior responses when member- checking was carried out). Furthermore, the Hawthorne effect, which describes an observer effect where individuals modify their behaviour in response to being aware that they are being observed, may have also had an impact on findings (301). In Chapters 6, 7 and 8, opportunistic sampling was utilised in the semi-structured interviews. This purposive approach to qualitative sampling is commonly utilised in psychology research because of its economical and sampling ease. The predominant disadvantage, however, is that it predisposes the study to selection bias.

The use of Delphi in Chapters 7 and 8 are generally advantageous when compared to other forms of consensus development methods (such as nominal group techniques) because they allow for greater numbers and anonymity across panellists. The primary limitation with this is that panellists are unable to understand reasons for divergent views on the second Delphi iteration. A further limitation is when panellists may deliberately provide extreme ratings out of line with the rest of the group. Outlier analysis of items that did not achieve consensus may have provided further insight into whether this effect may have occurred.

251

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

9.3. Future work

There is a need for further prospective work to explore predictors of post-operative IASCs and clinical recurrence. Further studies should be prospective, adequately powered, and include a control group and multivariate analyses that can adjust for stages, as determined in Chapter 6, and outcome measures as determined in Chapter 7. The goal of future research should be to risk stratify patients into either a one-stage approach or a two-stage approach. The role of other predictors, such as smoking status, hypo-albuminaemia, and corticosteroid therapy also need to be implemented into this model. The protective role of a proximal defunctioning ileostomy as part of a two-stage procedure also needs to be explored, as opposed to split stoma formation.

Currently, risk stratification models do not recognise Montreal L3 as a predictor of recurrence.

Future work should also aim to validate this finding through prospectively collected data and adequately powered studies.

Future work needs to account for how best to predict who responds favourably to endoscopic balloon dilatation. Approximately 30% at 5 years do not require surgery according to the findings of Chapter 5. Meta-regression analysis did not demonstrate any stricture or intervention related factors contributing to the need for eventual surgery. Therefore, an analysis of patient or demographic related factors may be necessary to explore this further response to treatment.

Further prospective reliability testing of the staging tool is required in a multi-centre setting to ensure generalisability and statistical power. Pre-operative radiological imaging should be prospectively staged, and correlation analyses should be tested against histopathologically-led, prospectively staged specimens, and surgically-led, prospectively staged, intra-operative

252

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery findings. Should there be a true radiological over-estimation of patients with stage 1 disease, and under-estimation of patients with stage 2 disease, further studies should seek to explore why this could be. This may include studies addressing the diagnostic accuracy of radiological modalities in differentiating stricture activity in the pre-operative setting. Validation against primary resection and anastomosis without stoma formation (one-stage procedure) may demonstrate an association with consensus-derived outcome measures (as described in Chapter

7). Furthermore, validation against non-surgical interventions (biologics or endoscopic therapy) should also be explored.

The preliminary evidence base for the consensus-derived metrics for the IBD MDT (Chapter

8) and key performance indicators (Chapter 7) should be utilised in 2 ways. Firstly, to encourage the provision of resource at an institutional level and to embed multidisciplinary team driven care into institutional clinical processes, framed through mechanisms such as a

PDSA (Plan-Do-Study-Act) cycle, and through the recognition of the roles of the MDT’s core members in job-planning. Secondly, having determined outcome measures, the question of establishing an evidence-base for improved outcomes should be driven by multi-stakeholder initiatives through our National Societies. This could be achieved by embedding simple outcomes into national data collection initiatives (the National IBD Registry) and mandating participation by institutions on patient safety grounds (monitored by the National IBD

Audit/QUIPP processes and institutional governance structures). In addition to this benchmarking exercise, a formal prospective multi-centre trial to compare the speed of clinical decision-making, time to definitive ileocolonic Crohn’s surgery or treatment escalation could provide a definitive validation of the model. Consideration should be given to defining proportions and whether the creation of a minimum set should be implemented into practice.

253

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

These could be achieved through implementation of these KPIs into national audit datasets to provide a benchmark on proportions and volumes.

Finally, further qualitative studies need to explore the reasons for decision-making when presented with treatment options for ileocolonic Crohn’s disease. Patient fears and anxieties of surgery and stoma formation are recognised barriers to surgery. These perceptions may also be re-enforced by medical gastroenterologists caring for the patient, who often view surgery as an adverse event or ‘failure of medical therapy’. Considering surgery for Crohn’s disease is often an inevitable long-term event, an understanding is required of the perceptions of surgery and a need to develop interventions to improve the patients understanding of the potential benefits of timely surgery may be necessary. Further studies can also determine patient-related or psycho-social factors that impact on patient preferences for intervention choice.

254

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

REFERENCES

1. Crohn BB, Ginzburg L, Oppenheimer GD. Regional ileitis: A pathologic and clinical

entity. Journal of the American Medical Association. 1932;99(16):1323-9.

2. Kirsner JB. Crohn's disease: Yesterday, today, and tomorrow. Gastroenterology.

1997;112(3):1028-30.

3. Koch R. The etiology of tuberculosis. Review of Infectious Diseases. 1982;4(6):1270-

4.

4. Ross JR. Cicatrizing enteritis, colitis and gastritis; a case report. Gastroenterology.

1949;13(4):344-50.

5. Franklin RH, Taylor S. Nonspecific granulomatous (regional) esophagitis. The Journal

of thoracic surgery. 1950;19(2):292-7.

6. Lockhart-Mummery HE, Morson BC. Crohn's Disease (Regional Enteritis) of the Large

Intestine and its Distinction from Ulcerative Colitis. GUT. 1960;1(2):87-105.

7. Cosnes J, Gower-Rousseau C, Seksik P, Cortot A. Epidemiology and Natural History

of Inflammatory Bowel Diseases. Gastroenterology. 2011;140(6):1785-U118.

8. Shivananda S L-JJ, Logan R, Fear N. Incidence of inflammatory bowel disease across

Europe: is there a difference between north and south? Results of the European

collaborative study on inflammatory bowel disease (EC-IBD). GUT. 1996;39:690-7.

9. Loftus Jr EV. Clinical epidemiology of inflammatory bowel disease: incidence,

prevalence, and environmental influences. Gastroenterology. 2004;126(6):1504-17.

10. Yapp TR, Stenson R, Thomas GA, Lawrie BW, Williams GT, Hawthorne AB. Crohn's

disease incidence in Cardiff from 1930: an update for 1991-1995. European Journal Of

Gastroenterology & Hepatology. 2000;12(8):907-11.

255

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

11. Rubin GP HA, Kelly PJ, Ling J. In Inflammatory bowel disease epidemiology and

management in an English general practice population. Aliment Pharmacol Ther.

2000;14:1553-9.

12. Fellows IW, Freeman JG, Holmes GK. Crohn's disease in the city of Derby, 1951-85.

GUT. 1990;31(11):1262-5.

13. Trallori G, Palli D, Saieva C, Bardazzi G, Bonanomi AG, D'Albasio G, et al. A

Population-Based Study of Inflammatory Bowel Disease in Florence over 15 Years

(1978-92). Scandinavian Journal of Gastroenterology. 1996;31(9):892-9.

14. Van Assche G, Dignass A, Panes J, Beaugerie L, Karagiannis J, Allez M, et al. The

second European evidence-based Consensus on the diagnosis and management of

Crohn's disease: Definitions and diagnosis. J Crohns Colitis. 2010;4(1):7-27.

15. Sartor RB. Mechanisms of Disease: pathogenesis of Crohn's disease and ulcerative

colitis. Nat Clin Pract Gastroenterol Hepatol. 2006;3(7):390-407.

16. Gordon H, Trier Moller F, Andersen V, Harbord M. Heritability in Inflammatory Bowel

Disease: From the First Twin Study to Genome-Wide Association Studies.

Inflammatory Bowel Diseases. 2015;21(6):1428-34.

17. Peeters M, Nevens H, Baert F, Hiele M, de Meyer AM, Vlietinck R, et al. Familial

aggregation in Crohn's disease: increased age-adjusted risk and concordance in clinical

characteristics. Gastroenterology. 1996;111(3):597-603.

18. Gaya DR, Russell RK, Nimmo ER, Satsangi J. New genes in inflammatory bowel

disease: lessons for complex diseases? The Lancet. 2006;367(9518):1271-84.

19. Nussbaum M, Willard. Genetics in Medicine. 7th Edition ed: Thompson & Thompson;

2007.

256

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

20. Petnicki-Ocwieja T, Hrncir T, Liu YJ, Biswas A, Hudcovic T, Tlaskalova-Hogenova

H, et al. Nod2 is required for the regulation of commensal microbiota in the intestine.

Proceedings of the National Academy of Sciences of the United States of America.

2009;106(37):15813-8.

21. Shivananda SL-J, J. Logan, R. et. al. Incidence of IBD across Europe. GUT.

1996;39:690-7.

22. Chen GB, Lee SH, Brion MJ, Montgomery GW, Wray NR, Radford-Smith GL, et al.

Estimation and partitioning of (co)heritability of inflammatory bowel disease from

GWAS and immunochip data. Human Molecular Genetics. 2014;23(17):4710-20.

23. Buisson A, Chevaux JB, Allen PB, Bommelaer G, Peyrin-Biroulet L. Review article:

the natural history of postoperative Crohn's disease recurrence. Aliment Pharmacol

Ther. 2012;35(6):625-33.

24. Unkart JT, Anderson L, Li E, Miller C, Yan Y, Gu CC, et al. Risk factors for surgical

recurrence after ileocolic resection of Crohn's disease. Diseases of the Colon and

Rectum. 2008;51(8):1211-6.

25. De Cruz P, Kamm MA, Prideaux L, Allen PB, Desmond PV. Postoperative recurrent

luminal Crohn's disease: a systematic review. Inflamm Bowel Dis. 2012;18(4):758-77.

26. Reese GE, Nanidis T, Borysiewicz C, Yamamoto T, Orchard T, Tekkis PP. The effect

of smoking after surgery for Crohn's disease: a meta-analysis of observational studies.

Int J Colorectal Dis. 2008;23(12):1213-21.

27. Baumgart DC, Carding SR. Inflammatory bowel disease: cause and immunobiology.

The Lancet.369(9573):1627-40.

257

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

28. Pariente B, Cosnes J, Danese S, Sandborn WJ, Lewin M, Fletcher JG, et al.

Development of the Crohn's disease digestive damage score, the Lemann score.

Inflamm Bowel Dis. 2011;17(6):1415-22.

29. Freeman HJ. Natural history and long-term clinical course of Crohn’s disease. World

Journal of Gastroenterology : WJG. 2014;20(1):31-6.

30. Cosnes J, Cattan S, Blain A, Beaugerie L, Carbonnel F, Parc R, et al. Long-term

evolution of disease behavior of Crohn's disease. Inflamm Bowel Dis. 2002;8(4):244-

50.

31. Louis E, Collard A, Oger AF, Degroote E, Aboul Nasr El Yafi FA, Belaiche J.

Behaviour of Crohn's disease according to the Vienna classification: changing pattern

over the course of the disease. GUT. 2001;49(6):777-82.

32. Solberg IC, Vatn MH, Hoie O, Stray N, Sauar J, Jahnsen J, et al. Clinical course in

Crohn's disease: results of a Norwegian population-based ten-year follow-up study.

Clinical gastroenterology and hepatology : the official clinical practice journal of the

American Gastroenterological Association. 2007;5(12):1430-8.

33. Thia KT, Sandborn WJ, Harmsen WS, Zinsmeister AR, Loftus EV, Jr. Risk factors

associated with progression to intestinal complications of Crohn's disease in a

population-based cohort. Gastroenterology. 2010;139(4):1147-55.

34. Wolters FL, Russel MG, Sijbrandij J, Ambergen T, Odes S, Riis L, et al. Phenotype at

diagnosis predicts recurrence rates in Crohn's disease. Gut. 2006;55(8):1124-30.

35. Rieder F, Zimmermann EM, Remzi FH, Sandborn WJ. Crohn's disease complicated by

strictures: a systematic review. GUT. 2013;62(7):1072-84.

258

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

36. Graham MF, Diegelmann RF, Elson CO, Lindblad WJ, Gotschalk N, Gay S, et al.

Collagen content and types in the intestinal strictures of Crohn's disease.

Gastroenterology. 1988;94(2):257-65.

37. Rieder F, Karrasch T, Ben-Horin S, Schirbel A, Ehehalt R, Wehkamp J, et al. Results

of the 2nd scientific workshop of the ECCO (III): basic mechanisms of intestinal

healing. J Crohns Colitis. 2012;6(3):373-85.

38. Scarpa M, Angriman I, Barollo M, Polese L, Ruffolo C, Bertin M, et al. Role of stapled

and hand-sewn anastomoses in recurrence of Crohn's disease. Hepato-gastroenterology.

2004;51(58):1053-7.

39. Yamamoto T, Allan RN, Keighley MR. Strategy for surgical management of

ileocolonic anastomotic recurrence in Crohn's disease. World Journal of Surgery.

1999;23(10):1055-60; discussion 60-1.

40. Louis E, Michel V, Hugot JP, Reenaers C, Fontaine F, Delforge M, et al. Early

development of stricturing or penetrating pattern in Crohn's disease is influenced by

disease location, number of flares, and smoking but not by NOD2/CARD15 genotype.

GUT. 2003;52(4):552-7.

41. Abreu MT, Taylor KD, Lin YC, Hang T, Gaiennie J, Landers CJ, et al. Mutations in

NOD2 are associated with fibrostenosing disease in patients with Crohn's disease.

Gastroenterology. 2002;123(3):679-88.

42. Alvarez-Lobos M, Arostegui JI, Sans M, Tassies D, Plaza S, Delgado S, et al. Crohn's

disease patients carrying NOD2/CARD15 gene variants have an increased and early

need for first surgery due to stricturing disease and higher rate of surgical recurrence.

Annals of Surgery. 2005;242(5):693-700.

259

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

43. Rigby RJ, Hunt MR, Scull BP, Simmons JG, Speck KE, Helmrath MA, et al. A new

animal model of postsurgical bowel inflammation and fibrosis: the effect of commensal

microflora. GUT. 2009;58(8):1104-12.

44. Louis E, Michel V, Hugot JP, Reenaers C, Fontaine F, Delforge M, et al. Early

development of stricturing or penetrating pattern in Crohn’s disease is influenced by

disease location, number of flares, and smoking but not by NOD2/CARD15 genotype.

GUT. 2003;52(4):552-7.

45. Schwartz DA, Loftus Jr EV, Tremaine WJ, Panaccione R, Harmsen WS, Zinsmeister

AR, et al. The natural history of fistulizing Crohn's disease in Olmsted County,

Minnesota. Gastroenterology. 2002;122(4):875-80.

46. Farmer R. Clinical Patterns In Crohns Disease: Statistical Study Of 615 Cases. In:

Gastroenterology 1974(66);4,690-690. Wb Saunders Co..

47. Greenstein AJ, Lachman P, Sachar DB, Springhorn J, Heimann T, Janowitz HD, et al.

Perforating and non-perforating indications for repeated operations in Crohn's disease:

evidence for two clinical forms. GUT. 1988;29(5):588-92.

48. Gasche C, Scholmerich J, Brynskov J, D'Haens G, Hanauer SB, Irvine EJ, et al. A

simple classification of Crohn's disease: report of the Working Party for the World

Congresses of Gastroenterology, Vienna 1998. Inflamm Bowel Dis. 2000;6(1):8-15.

49. Silverberg MS, Satsangi J, Ahmad T, Arnott ID, Bernstein CN, Brant SR, Caprilli R,

Colombel JF, Gasche C, Geboes K, Jewell DP. Toward an integrated clinical, molecular

and serological classification of inflammatory bowel disease: Report of a Working

Party of the 2005 Montreal World Congress of Gastroenterology. Canadian Journal of

Gastroenterology and Hepatology. 2005;19(Suppl A):5A-36A.

260

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

50. Satsangi J, Silverberg MS, Vermeire S, Colombel JF. The Montreal classification of

inflammatory bowel disease: controversies, consensus, and implications. GUT.

2006;55(6):749-53.

51. Dignass A, Van Assche G, Lindsay JO, Lemann M, Soderholm J, Colombel JF, et al.

The second European evidence-based Consensus on the diagnosis and management of

Crohn's disease: Current management. J Crohns Colitis. 2010;4(1):28-62.

52. Eshuis EJ, Bemelman WA, van Bodegraven AA, Sprangers MAG, Bossuyt PMM, de

Wit AWMvM, et al. Laparoscopic ileocolic resection versus infliximab treatment of

distal ileitis in Crohn's disease: a randomized multicenter trial (LIR!C-trial). BMC

Surgery. 2008;8:15-.

53. Wild GE, Waschke KA, Bitton A, Thomson ABR. The mechanisms of prednisone

inhibition of inflammation in Crohn's disease involve changes in intestinal

permeability, mucosal TNFα production and nuclear factor kappa B expression.

Alimentary Pharmacology & Therapeutics. 2003;18(3):309-17.

54. Ramadas AV, Gunesh S, Thomas GAO, Williams GT, Hawthorne AB. Natural history

of Crohn's disease in a population-based cohort from Cardiff (1986–2003): a study of

changes in medical treatment and surgical resection rates. Gut. 2010.

55. Hellers G, Cortot A, Jewell D, Leijonmarck CE, Lofberg R, Malchow H, et al. Oral

budesonide for prevention of postsurgical recurrence in Crohn's disease. The IOIBD

Budesonide Study Group. Gastroenterology. 1999;116(2):294-300.

56. Camma CA, Giunta MA, Rosselli MA, Cottone MA. Mesalamine in the maintenance

treatment of Crohn's disease: a meta-analysis adjusted for confounding variables.

Gastroenterology. 1997 Nov 30;113(5):1465-73

261

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

57. Lochs H, Mayer M, Fleig WE, Mortensen PB, Bauer P, Genser D, et al. Prophylaxis of

postoperative relapse in Crohn's disease with mesalamine: European Cooperative

Crohn's Disease Study VI. Gastroenterology. 2000;118(2):264-73.

58. Lennard L. TPMT in the treatment of Crohn's disease with azathioprine. Gut.

2002;51(2):143-6.

59. Hanauer SB, Korelitz BI, Rutgeerts P, Peppercorn MA, Thisted RA, Cohen RD, et al.

Postoperative maintenance of Crohn's disease remission with 6-mercaptopurine,

mesalamine, or placebo: a 2-year trial. Gastroenterology. 2004;127(3):723-9.

60. Rutgeerts P, Hiele M, Geboes K, Peeters M, Penninckx F, Aerts R, Kerremans R.

Controlled trial of metronidazole treatment for prevention of Crohn's recurrence after

ileal resection. Gastroenterology. 1995;108(6):1617-21.

61. Rutgeerts P, Van Assche G, Vermeire S, D’Haens G, Baert F, Noman M, Aerden I, De

Hertogh G, Geboes K, Hiele M, D’Hoore A. Ornidazole for prophylaxis of post-

operative Crohn’s disease recurrence: a randomized, double-blind, placebo-controlled

trial. Gastroenterology. 2005;128(4):856-61

62. Rutgeerts P, Hiele M, Geboes K, Peeters M, Penninckx F, Aerts R, Kerremans R.

Controlled trial of metronidazole treatment for prevention of Crohn's recurrence after

ileal resection. Gastroenterology. 1995;108(6):1617-21..

63. D'Haens GR, Vermeire S, Van Assche G, Noman M, Aerden I, Van Olmew G, et al.

Therapy of metronidazole with azathioprine to prevent postoperative recurrence of

Crohn's disease: A controlled randomized trial. Gastroenterology. 2008;135(4):1123-9.

64. Infliximab (review) and adalimumab for the treatment of Crohn's disease. National

Institute for Clinical Excelence: technology appraisal guidance 2010

262

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

65. Hanauer SB, Feagan BG, Lichtenstein GR, Mayer LF, Schreiber S, Colombel JF, et al.

Maintenance infliximab for Crohn's disease: the ACCENT I randomised trial. Lancet.

2002;359(9317):1541-9.

66. Regueiro M, Schraut W, Baidoo L, Kip KE, Sepulveda AR, Pesci M, et al. Infliximab

Prevents Crohn's Disease Recurrence After Ileal Resection: Gastroenterology. 136 (2)

(pp 441-450.e1), 2009. Date of Publication: February 2009.; 2009.

67. Savarino E, Bodini G, Dulbecco P, Assandri L, Bruzzone L, Mazza F, et al.

Adalimumab is more effective than azathioprine and mesalamine at preventing

postoperative recurrence of Crohn's disease: A randomized controlled trial: American

Journal of Gastroenterology. 108 (11) (pp 1731-1742), 2013

68. Brower RA. Hydrostatic balloon dilation of a terminal ileal stricture secondary to

Crohn's disease. Gastrointest Endosc. 1986;32(1):38-40.

69. Hassan C, Zullo A, De Francesco V, Ierardi E, Giustini M, Pitidis A, et al. Systematic

review: Endoscopic dilatation in Crohn's disease. Aliment Pharmacol Ther.

2007;26(11-12):1457-64.

70. Wibmer AG, Kroesen AJ, Grone J, Buhr HJ, Ritz JP. Comparison of strictureplasty and

endoscopic balloon dilatation for stricturing Crohn's disease--review of the literature.

Int J Colorectal Dis. 2010;25(10):1149-57.

71. Alexander-Williams J, Haynes IG. Up-to-date management of small-bowel Crohn's

disease. Advances in Surgery. 1987;20:245-64.

72. Bernell O, Lapidus A Fau - Hellers G, Hellers G. Risk factors for surgery and

postoperative recurrence in Crohn's disease.231(1):38-45.

263

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

73. Aratari A, Papi C, Leandro G, Viscido A, Capurso L, Caprilli R. Early versus late

surgery for ileo-caecal Crohn's disease. Aliment Pharmacol Ther. 2007;26(10):1303-

12.

74. Borley NR, Mortensen NJ, Jewell DP. Preventing postoperative recurrence of Crohn's

disease. The British Journal of Surgery. 1997;84(11):1493-502.

75. Shaffer VO, Wexner SD. Surgical management of Crohn's disease: Langenbeck's

Archives of Surgery. 398 (1) (pp 13-27), 2013. Date of Publication: January 2013.;

2013.

76. Holzheimer RG, Mannick JA, Strong SA. Surgical management of Crohn's disease.

2001.

77. Yamamoto T, Watanabe T. Surgery for luminal Crohn’s disease. World Journal of

Gastroenterology: WJG. 2014;20(1):78-90.

78. Maartense S, Dunker MS, Slors JF, Cuesta MA, Pierik EG, Gouma DJ, et al.

Laparoscopic-assisted versus open ileocolic resection for Crohn's disease: a

randomized trial. Annals of Surgery. 2006;243(2):143-9; discussion 50-3.

79. Tan JJ, Tjandra JJ. Laparoscopic surgery for Crohn's disease: a meta-analysis. Diseases

of the Colon and Rectum. 2007;50(5):576-85.

80. Jobanputra S, Weiss EG. Strictureplasty. Clinics in colon and rectal surgery.

2007;20(4):294-302.

81. Katariya RN, Sood S, Rao PG, Rao PL. Stricture-plasty for tubercular strictures of the

gastro-intestinal tract. The British journal of Surgery. 1977;64(7):496-8.

82. Fazio VW, Galandiuk S, Jagelman DG, Lavery IC. Strictureplasty in Crohn's disease.

Annals of Surgery. 1989;210(5):621-5.

264

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

83. Lee EC, Papaioannou N. Minimal surgery for chronic obstruction in patients with

extensive or universal Crohn's disease. Annals of the Royal College of Surgeons of

England. 1982;64(4):229-33.

84. Michelassi F. Side-to-side isoperistaltic strictureplasty for multiple Crohn's strictures.

Diseases of the Colon and Rectum. 1996;39(3):345-9.

85. Everyone Counts - Planning for Patients 2013/14. Leeds: NHS Commissioning Board.

2012

86. Burns EM, Bottle A, Aylin P, Darzi A, Nicholls RJ, Faiz O. Variation in reoperation

after colorectal surgery in England as an indicator of surgical performance:

retrospective analysis of Hospital Episode Statistics. BMJ. 2011;343:d4836.

87. Alves A, Panis Y, Bouhnik Y, Pocard M, Vicaut E, Valleur P. Risk factors for intra-

abdominal septic complications after a first ileocecal resection for Crohn's disease: a

multivariate analysis in 161 consecutive patients. Diseases of the Colon and Rectum.

2007;50(3):331-6.

88. Bemelman WA, Slors JF, Dunker MS, van Hogezand RA, van Deventer SJ, Ringers J,

et al. Laparoscopic-assisted vs. open ileocolic resection for Crohn's disease. A

comparative study. Surgical Endoscopy. 2000;14(8):721-5.

89. Bergamaschi R, Haughn C, Reed JF, 3rd, Arnaud JP. Laparoscopic intracorporeal

ileocolic resection for Crohn's disease: is it safe? Diseases of the colon and rectum.

2009;52(4):651-6.

90. Campbell MJ, Paull NB, Thirlby RC. Transverse incisions for resection of ileocolic

Crohn's disease. The American surgeon. 2013;79(3):279-83.

91. Chaudhary B, Glancy D, Dixon AR. Laparoscopic surgery for recurrent ileocolic

Crohn's disease is as safe and effective as primary resection. Colorectal disease : the

265

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

official journal of the Association of Coloproctology of Great Britain and Ireland.

2011;13(12):1413-6.

92. Evans J, Poritz L, MacRae H. Influence of experience on laparoscopic ileocolic

resection for Crohn's disease. Diseases of the Colon and Rectum. 2002;45(12):1595-

600.

93. Fichera A, Peng SL, Elisseou NM, Rubin MA, Hurst RD. Laparoscopy or conventional

open surgery for patients with ileocolonic Crohn's disease? A prospective study.

Surgery. 2007;142(4):566-71.

94. Maeda K, Noda E, Nagahara H, Inoue T, Takii M, Watanabe K, et al. A comparative

study of single-incision versus conventional multiport laparoscopic ileocecal resection

for Crohn's disease with strictures: Asian Journal of Endoscopic Surgery. 5 (3) (pp 118-

122), 2012.

95. Resegotti A, Astegiano M, Farina EC, Ciccone G, Avagnina G, Giustetto A, et al. Side-

to-side stapled anastomosis strongly reduces anastomotic leak rates in Crohn's disease

surgery. Diseases of the Colon and Rectum. 2005;48(3):464-8.

96. Rijcken E, Mennigen R, Argyris I, Senninger N, Bruewer M. Single-incision

laparoscopic surgery for ileocolic resection in Crohn's disease. Diseases of the Colon

and Rectum. 2012;55(2):140-6.

97. Riss S, Bittermann C, Zandl S, Kristo I, Stift A, Papay P, et al. Short-term complications

of wide-lumen stapled anastomosis after ileocolic resection for Crohn's disease: who is

at risk? Colorectal disease : the official journal of the Association of Coloproctology of

Great Britain and Ireland. 2010;12(10 Online):e298-303.

98. Shental O, Tulchinsky H, Greenberg R, Klausner JM, Avital S. Positive histological

inflammatory margins are associated with increased risk for intra-abdominal septic

266

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

complications in patients undergoing ileocolic resection for Crohn's disease. Diseases

of the Colon and Rectum. 2012;55(11):1125-30.

99. Soop M, Larson DW, Malireddy K, Cima RR, Young-Fadok TM, Dozois EJ. Safety,

feasibility, and short-term outcomes of laparoscopically assisted primary ileocolic

resection for Crohn's disease. Surgical Endoscopy. 2009;23(8):1876-81.

100. Vangeenberghe N, De Vogelaere K, Haentjens P, Delvaux G. Laparoscopically assisted

ileocolectomy in patients with Crohn's disease: a study of 50 consecutive patients.

Surgical Endoscopy. 2009;23(8):1797-801.

101. Zurbuchen U, Kroesen AJ, Knebel P, Betzler MH, Becker H, Bruch HP, et al.

Complications after end-to-end vs. side-to-side anastomosis in ileocecal Crohn's

disease--early postoperative results from a randomized controlled multi-center trial

(ISRCTN-45665492). Langenbeck's archives of surgery / Deutsche Gesellschaft fur

Chirurgie. 2013;398(3):467-74.

102. Eshuis EJ, Voermans RP, Stokkers PC, van Berge Henegouwen MI, Fockens P,

Bemelman WA. Laparoscopic resection with transcolonic specimen extraction for

ileocaecal Crohn's disease. British Journal of Surgery. 2010;97(4):569-74.

103. Bergamaschi R, Pessaux P, Arnaud JP. Comparison of conventional and laparoscopic

ileocolic resection for Crohn's disease. Diseases of the colon and rectum.

2003;46(8):1129-33.

104. Duepree HJ, Senagore AJ, Delaney CP, Brady KM, Fazio VW. Advantages of

laparoscopic resection for ileocecal Crohn's disease. Diseases of the colon and rectum.

2002;45(5):605-10.

105. Kingham TP, Pachter HL. Colonic anastomotic leak: risk factors, diagnosis, and

treatment. Journal of the American College of Surgeons. 2009;208(2):269-78.

267

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

106. Alves A, Panis Y, Trancart D, Regimbeau JM, Pocard M, Valleur P. Factors associated

with clinically significant anastomotic leakage after large bowel resection: multivariate

analysis of 707 patients. World Journal of Surgery. 2002;26(4):499-502.

107. Iancu C, Mocan LC, Todea-Iancu D, Mocan T, Acalovschi I, Ionescu D, et al. Host-

related predictive factors for anastomotic leakage following large bowel resections for

colorectal cancer. Journal of Gastrointestinal and Liver Diseases : JGLD.

2008;17(3):299-303.

108. Hulten L. Surgical Treatment of Crohn's Disease of the Small Bowel or Ileocecum.

World J Surg. 1988;12:180-5,.

109. Kanazawa A, Yamana T, Okamoto K, Sahara R. Risk factors for postoperative intra-

abdominal septic complications after bowel resection in patients with Crohn's disease.

Diseases of the Colon and Rectum. 2012;55(9):957-62.

110. Tzivanakis A, Singh JC, Guy RJ, Travis SP, Mortensen NJ, George BD. Influence of

risk factors on the safety of ileocolic anastomosis in Crohn's disease surgery. Diseases

of the Colon and Rectum. 2012;55(5):558-62.

111. Yamamoto T, Allan RN, Keighley MR. Risk factors for intra-abdominal sepsis after

surgery in Crohn's disease. Diseases of the Colon and Rectum. 2000;43(8):1141-5.

112. Bellolio F, Cohen Z, Macrae HM, O'Connor BI, Huang H, Victor JC, et al. Outcomes

following surgery for perforating Crohn's disease. The British Journal of Surgery.

2013;100(10):1344-8.

113. Post S, Herfarth C, Bohm E, Timmermanns G, Schumacher H, Schumann G, et al. The

impact of disease pattern, surgical management, and individual surgeons on the risk for

relaparotomy for recurrent Crohn's disease. Annals of Ssurgery. 1996;223(3):253-60.

268

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

114. Wang WY, Chen CL, Chen GL, Wu CJ, Li HG, Luan SM, et al. Analysis of risk factors

for anastomotic infectious complications following bowel resection for Crohn disease.

Zhonghua wei chang wai ke za zhi = Chinese journal of gastrointestinal surgery.

2013;16(4):328-31

115. Wu JS, Birnbaum EH, Kodner IJ, Fry RD, Read TE, Fleshman JW. Laparoscopic-

assisted ileocolic resections in patients with Crohn's disease: are abscesses, phlegmons,

or recurrent disease contraindications? Surgery. 1997;122(4):682-8; discussion 8-9.

116. Myrelid P, Olaison G, Sjodahl R, Nystrom PO, Almer S, Andersson P. Thiopurine

therapy is associated with postoperative intra-abdominal septic complications in

abdominal surgery for Crohn's disease. Diseases of the Colon and Rectum.

2009;52(8):1387-94.

117. Post S, Betzler M, von Ditfurth B, Schurmann G, Kuppers P, Herfarth C. Risks of

intestinal anastomoses in Crohn's disease. Annals of Surgery. 1991;213(1):37-42.

118. Yamamoto T, Shiraki M. Risk factors for anastomotic complications after resection for

Crohn's disease. Diseases of the Colon and Rectum. 2012;55(10):e346-7; author reply

e7.

119. Muller-Wille R, Iesalnieks I, Dornia C, Ott C, Jung EM, Friedrich C, et al. Influence of

percutaneous abscess drainage on severe postoperative septic complications in patients

with Crohn's disease. Int J Colorectal Dis. 2011;26(6):769-74.

120. El-Hussuna A, Andersen J, Bisgaard T, Jess P, Henriksen M, Oehlenschlager J, et al.

Biologic treatment or immunomodulation is not associated with postoperative

anastomotic complications in abdominal surgery for Crohn's disease. Scand J

Gastroenterol. 2012;47(6):662-8.

269

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

121. Kopylov U, Ben-Horin S, Zmora O, Eliakim R, Katz LH. Anti-tumor necrosis factor

and postoperative complications in Crohn's disease: systematic review and meta-

analysis. Inflamm Bowel Dis. 2012;18(12):2404-13.

122. Yang SS, Yu CS, Yoon YS, Yoon SN, Lim SB, Kim JC. Risk factors for complications

after bowel surgery in Korean patients with Crohn's disease. Journal of the Korean

Surgical Society. 2012;83(3):141-8.

123. Rutgeerts P, Geboes K, Vantrappen G, Kerremans R, Coenegrachts JL, Coremans G.

Natural-History Of Recurrent Crohns-Disease At The Ileocolonic Anastomosis After

Curative Surgery. Gut. 1984;25(6):665-72.

124. Rutgeerts P, Geboes K, Vantrappen G, Beyls J, Kerremans R, Hiele M. Predictability

of the postoperative course of Crohn's disease. Gastroenterology. 1990;99(4):956-63.

125. Wettergren A, Christiansen J. Risk of recurrence and reoperation after resection for

ileocolic Crohn's disease. Scand J Gastroenterol. 1991;26(12):1319-22.

126. Heen LO, Nygaard K, Bergan A. Crohn's disease. Results of excisional surgery in 133

patients. Scand J Gastroenterol. 1984;19(6):747-54.

127. Pallone F, Boirivant M, Stazi MA, Cosintino R, Prantera C, Torsoli A. Analysis of

clinical course of postoperative recurrence in Crohn's disease of distal ileum. Digestive

Diseases and Sciences. 1992;37(2):215-9.

128. Wettergren A, Christiansen J. Risk of recurrence and reoperation after resection for

ileocolic crohns-disease. Scandinavian Journal of Gastroenterology.

1991;26(12):1319-22..

129. Sostegni R, Daperno M, Scaglione N, Lavagna A, Rocca R, Pera A. Review article:

Crohn's disease: monitoring disease activity. Aliment Pharmacol Ther. 2003;17 Suppl

2:11-7.

270

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

130. Sostegni R, Daperno M, Scaglione N, Lavagna A, Rocca R, Pera A. Crohn's disease:

monitoring disease activity. Alimentary Pharmacology & Therapeutics. 2003;17:11-7.

131. Schoepfer AM, Beglinger C, Straumann A, Trummler M, Vavricka SR, Bruegger LE,

et al. Fecal Calprotectin Correlates More Closely With the Simple Endoscopic Score

for Crohn's Disease (SES-CD) than CRP, Blood Leukocytes, and the CDAI. The

American Journal of Gastroenterology. 2009;105(1):162-9.

132. Riss S, Schuster I, Papay P, Mittlbock M, Stift A. Repeat intestinal resections increase

the risk of recurrence of crohn's disease: Diseases of the Colon and Rectum. 56 (7) (pp

881-887), 2013.

133. Milsom JW, Hammerhofer KA, Bohm B, Marcello P, Elson P, Fazio VW. Prospective,

randomized trial comparing laparoscopic vs. conventional surgery for refractory

ileocolic Crohn's disease. Diseases of the Colon and Rectum. 2001;44(1):1-8;

discussion -9.

134. Simillis C, Purkayastha S, Yamamoto T, Strong SA, Darzi AW, Tekkis PP. A meta-

analysis comparing conventional end-to-end anastomosis vs. other anastomotic

configurations after resection in Crohn's disease. Diseases of the Colon and Rectum.

2007;50(10):1674-87.

135. Fazio VW, Marchetti F, Church M, Goldblum JR, Lavery C, Hull TL, et al. Effect of

resection margins on the recurrence of Crohn's disease in the small bowel. A

randomized controlled trial. Annals of Surgery. 1996;224(4):563-71; discussion 71-3.

136. McLeod RS, Wolff BG, Ross S, Parkes R, McKenzie M. Recurrence of Crohn's disease

after ileocolic resection is not affected by anastomotic type: Results of a multicenter,

randomized, controlled trial: Diseases of the Colon and Rectum. 52 (5) (pp 919-927),

2009.

271

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

137. Doherty G, Bennett G, Patil S, Cheifetz A, Moss AC. Interventions for prevention of

post‐operative recurrence of Crohn's disease. The Cochrane Library. 2009 Jan 1.

138. Peyrin-Biroulet L, Deltenre P, Ardizzone S, D'Haens G, Hanauer SB, Herfarth H, et al.

Azathioprine and 6-mercaptopurine for the prevention of postoperative recurrence in

Crohn's disease: a meta-analysis. The American journal of Gastroenterology.

2009;104(8):2089-96.

139. Regueiro M, Kip KE, Baidoo L, Swoger JM, Schraut W. Post-operative therapy with

infliximab prevents long-term Crohn’s disease recurrence. Clinical Gastroenterology

and Hepatology. 2014;12(9):1494-502.

140. Dretzke J, Edlin R, Round J, Connock M, Hulme C. A systematic review and economic

evaluation of the use of tumour necrosis factor-alpha (TNF-α) inhibitors, adalimumab

and infliximab, for Crohn’s disease. Health Technology Assessment. 2011;15(6):250.

141. Byrne CM, Solomon MJ, Young JM, Selby W, Harrison JD. Patient preferences

between surgical and medical treatment in Crohn's disease. Diseases of the Colon and

Rectum. 2007;50(5):586-97.

142. Shariff U, Narula H, Speake W, Brown S. Terminal ileal Crohn’s disease: conservative

surgeon and aggressive physician?. Colorectal Disease. 2009;11(5):522-3.

143. Baars JE, Markus T, Kuipers EJ, van der Woude CJ. Patients' preferences regarding

shared decision-making in the treatment of inflammatory bowel disease: results from a

patient-empowerment study. Digestion. 2010;81(2):113-9.

144. Borruel N, Castro J, Riestra S, Costi M, Casellas F. Treatment preferences of patients

with Crohn's disease: Development of the IMPLICA questionnaire. Revista espanola

de enfermedades digestivas : organo oficial de la Sociedad Espanola de Patologia

Digestiva. 2014;106(6):372-80.

272

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

145. Johnson FR, Hauber B, Ozdemir S, Siegel CA, Hass S, Sands BE. Are

gastroenterologists less tolerant of treatment risks than patients? Benefit-risk

preferences in Crohn's disease management. Journal of managed care pharmacy :

JMCP. 2010;16(8):616-28.

146. Kennedy ED, To T, Steinhart AH, Detsky A, Llewellyn-Thomas HA, McLeod RS. Do

patients consider postoperative maintenance therapy for Crohn's disease worthwhile?

Inflamm Bowel Dis. 2008;14(2):224-35.

147. Johnson FR, Ozdemir S, Mansfield C, Hass S, Miller DW, Siegel CA, et al. Crohn's

disease patients' risk-benefit preferences: serious adverse event risks versus treatment

efficacy. Gastroenterology. 2007;133(3):769-79.

148. Esrailian E, Spiegel BM, Targownik LE, Dubinsky MC, Targan SR, Gralnek IM.

Differences in the management of Crohn's disease among experts and community

providers, based on a national survey of sample case vignettes. Aliment Pharmacol

Ther. 2007;26(7):1005-18.

149. Montgomery AA, Fahey T. How do patients' treatment preferences compare with those

of clinicians? Quality in health care : QHC. 2001;10 Suppl 1:i39-43.

150. Zutshi M, Hull TL, Hammel J. Crohn's disease: a patient's perspective. Int J Colorectal

Dis. 2007;22(12):1437-44.

151. Quality Governance in the NHS National Quality Board ‐ A guide for provider boards.

2011.

152. Darzi A. High quality care for all: NHS next stage review final report: The Stationery

Office; 2008.

153. Donabedian A. Evaluating the Quality of Medical Care. The Milbank Quarterly.

2005;83(4):691-729.

273

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

154. Quality Care: Service Standards for the healthcare of people who have Inflammatory

Bowel Disease (IBD). IBD Standards Group 2009.

155. UK IBD Audit 2nd Round (2008) Report - Executive Summary of the National Results

for the Organisation & Process of Adult IBD Care in the UK. 2009.

156. UK IBD Audit 2006 - Executive Summary of the National Results for the Organisation

& Process of IBD Care in the UK. 2007.

157. Luft HS, Bunker JP, Enthoven AC. Should operations be regionalized? The empirical

relation between surgical volume and mortality. The New England Journal of Medicine.

1979;301(25):1364-9.

158. Tingulstad S, Skjeldestad FE, Hagen B. The effect of centralization of primary surgery

on survival in ovarian cancer patients. Obstetrics and Gynecology. 2003;102(3):499-

505.

159. Forshaw MJ, Gossage JA, Stephens J, Strauss D, Botha AJ, Atkinson S, et al.

Centralisation of oesophagogastric cancer services: can specialist units deliver? Annals

of the Royal College of Surgeons of England. 2006;88(6):566-70.

160. Lieberman MD, Kilburn H, Lindsey M, Brennan MF. Relation of perioperative deaths

to hospital volume among patients undergoing pancreatic resection for malignancy.

Annals of Surgery. 1995;222(5):638-45.

161. Begg CB, Cramer LD, Hoskins WJ, Brennan MF. Impact of hospital volume on

operative mortality for major cancer surgery. JAMA. 1998;280(20):1747-51.

162. Birkmeyer JD, Finlayson SR, Tosteson AN, Sharp SM, Warshaw AL, Fisher ES. Effect

of hospital volume on in-hospital mortality with pancreaticoduodenectomy. Surgery.

1999;125(3):250-6.

274

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

163. Hannan EL, Radzyner M, Rubin D, Dougherty J, Brennan MF. The influence of

hospital and surgeon volume on in-hospital mortality for colectomy, gastrectomy, and

lung lobectomy in patients with cancer. Surgery. 2002;131(1):6-15.

164. Weitz J, Koch M, Friess H, Büchler MW. Impact of Volume and Specialization for

Cancer Surgery. Digestive Surgery. 2004;21(4):253-61.

165. Begg CB, Riedel ER, Bach PB, Kattan MW, Schrag D, Warren JL, et al. Variations

in Morbidity after Radical Prostatectomy. New England Journal of Medicine.

2002;346(15):1138-44.

166. Dimick JB, Pronovost PJ, Cowan JA, Lipsett PA. Surgical volume and quality of care

for esophageal resection: do high-volume hospitals have fewer complications? The

Annals of Thoracic Surgery. 2003;75(2):337-41.

167. Swisher SG, Deford L, Merriman KW, Walsh GL, Smythe R, Vaporicyan A, et al.

Effect of operative volume on morbidity, mortality, and hospital use after

esophagectomy for cancer. The Journal of Thoracic and Cardiovascular Surgery.

2000;119(6):1126-32.

168. Edge SB, Schmieg RE, Jr., Rosenlof LK, Wilhelm MC. Pancreas cancer resection

outcome in American University centers in 1989-1990. Cancer. 1993;71(11):3502-8.

169. Ke KM, Hollingworth W, Ness AR. The costs of centralisation: a systematic review of

the economic impact of the centralisation of cancer services. European Journal of

Cancer Care. 2012;21(2):158-68.

170. Posnett J. Is bigger better? Concentration in the provision of secondary care. BMJ.

1999;319(7216):1063-5.

275

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

171. Christian CK, Gustafson ML, Betensky RA, Daley J, Zinner MJ. The Leapfrog volume

criteria may fall short in identifying high-quality surgical centers. Annals of Surgery.

2003;238(4):447-55; discussion 55-7.

172. Colorectal Cancer Surgery Standards ACPGBI Advisory. 2012.

173. Beets G, Sebag-Montefiore D, Andritsch E, Arnold D, Beishon M, Crul M, et al. ECCO

Essential Requirements for Quality Cancer Care: Colorectal Cancer. A critical review.

Critical Reviews in Oncology/Hematology. 2017;110(Supplement C):81-93.

174. Ihse I. The Volume-Outcome Relationship in Cancer Surgery: A Hard Sell. Annals of

Surgery. 2003;238(6):777-81.

175. Spencer SA, Davies MP. Hospital episode statistics: improving the quality and value

of hospital data: a national internet e-survey of hospital consultants. BMJ Open.

2012;2(6).

176. Calman K, Hine D. A policy framework for commissioning cancer services. London:

Department of Health. 1995 Apr;18.

177. Haward RA. The Calman-Hine report: a personal retrospective on the UK's first

comprehensive policy on cancer services. The Lancet Oncology. 2006;7(4):336-46.

178. Morris E, Haward RA, Gilthorpe MS, Craigs C, Forman D. The impact of the Calman-

Hine report on the processes and outcomes of care for Yorkshire's colorectal cancer

patients. British Journal of Cancer. 2006;95(8):979-85.

179. Kersten C. Does in-house availability of multidisciplinary teams increase survival in

upper gastrointestinal-cancer? World Journal of Gastrointestinal Oncology.

2013;5(3):60.

276

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

180. Kesson EM, Allardice GM, George WD, Burns HJG, Morrison DS. Effects of

multidisciplinary team working on breast cancer survival: retrospective, comparative,

interventional cohort study of 13 722 women. BMJ. 2012;344(apr26 1):e2718-e.

181. Eaker S. Regional Differences in Breast Cancer Survival Despite Common Guidelines.

Cancer Epidemiology Biomarkers & Prevention. 2005;14(12):2914-8.

182. Gomella LG, Lin J, Hoffman-Censits J, Dugan P, Guiles F, Lallas CD, et al. Enhancing

prostate cancer care through the multidisciplinary clinic approach: a 15-year

experience. Journal of Oncology Practice. 2010;6(6):e5-e10.

183. MacDermid E, Hooton G, MacDonald M, McKay G, Grose D, Mohammed N, et al.

Improving patient survival with the colorectal cancer multi-disciplinary team.

Colorectal disease : the official journal of the Association of Coloproctology of Great

Britain and Ireland. 2009;11(3):291-5.

184. Taylor C, Ramirez A-J. Defining the characteristics of effective MDT working in

cancer care. BMJ Supportive & Palliative Care. 2011;1(Suppl 1):A23.

185. Panes J, O'Connor M, Peyrin-Biroulet L, Irving P, Petersson J, Colombel JF. Improving

quality of care in inflammatory bowel disease: What changes can be made today? J

Crohns Colitis. 2014;8(9):919-26.

186. Cassinotti A, Keshav S, Ardizzone S, Mortensen N, Sampietro G, Fociani P, et al. IBD

care in Europe: A comparative audit of the inpatient management of Crohn's disease

and ulcerative colitis using the national UK IBD audit tool. J Crohns Colitis.

2009;3(4):291-301.

187. Physicians RCo. National Audit of Inflammatory Bowel Disease (IBD) Service

Provision. September 2014.

277

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

188. International Statistical Classification of Diseases and Related Health Problems 10th

Revision. WHO. 2010.

189. NHS Connecting for Health, Department of Health, Social Care Information Centre.

OPCS Classifications of Interventions and Procedures. The Stationery Office; 2006 Feb

27..

190. Dindo D, Demartines N, Clavien P-A. Classification of Surgical Complications: A New

Proposal With Evaluation in a Cohort of 6336 Patients and Results of a Survey. Annals

of Surgery. 2004;240(2):205-13.

191. Little RJ, Rubin DB. Statistical analysis with missing data: John Wiley & Sons; 2014.

192. Denzin NK, Lincoln YS. The SAGE handbook of qualitative research. Thousand Oaks:

Sage Publications; 2005.

193. Green J, Britten N. Qualitative research and evidence based medicine. BMJ.

1998;316(7139):1230-2.

194. Mays N, Pope C. Qualitative Research: Rigour and qualitative research. BMJ.

1995;311(6997):109-12.

195. Pope C, Ziebland S, Mays N. Analysing qualitative data. BMJ. 2000;320(7227):114-6.

196. Britten N. Qualitative Research: Qualitative interviews in medical research. BMJ.

1995;311(6999):251-3.

197. Lincoln YS, Guba EG. Naturalistic Inquiry: SAGE Publications; 1985.

198. Mays N, Pope C. Assessing quality in qualitative research. BMJ. 2000;320(7226):50-

2.

199. Kneebone R, Fry H. Principles and Methods in Qualitative Research. In: Athanasiou T,

Debas H, Darzi A, editors. Key Topics in Surgical Research and Methodology: Springer

Berlin Heidelberg; 2010. p. 243-53.

278

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

200. Mays, N. and Pope, C. (2007). Quality in Qualitative Health Research. In Qualitative

Research in Health Care (eds C. Pope and N. Mays).

201. Stefanidis D, Cochran A, Sevdalis N, Mellinger J, Phitayakorn R, Sullivan M, et al.

Research priorities for multi-institutional collaborative research in surgical education.

The American Journal of Surgery. 2015;209(1):52-8.

202. Dalkey NC. The Delphi Method: An Experimental Study of Group Opinion. Santa

Monica, CA: RAND Corporation, 1969. 1969.

203. Hull L, Arora S, Symons NRA, Jalil R, Darzi A, Vincent C, et al. Training Faculty in

Nontechnical Skill Assessment: National Guidelines on Program Requirements. Annals

of Surgery. 2013;258(2):370-5.

204. Fink A, Kosecoff J, Chassin M, Brook RH. Consensus methods: characteristics and

guidelines for use. American Journal of Public Health. 1984;74(9):979-83.

205. Stewart DB, Messaris E. Early experience with single-site laparoscopic surgery for

complicated ileocolic Crohn's disease at a tertiary-referral center. Surgical Endoscopy.

2012;26(3):777-82.

206. Bergamaschi R, Haughn C, Reed JF, III, Arnaud J-P. Laparoscopic Intracorporeal

Ileocolic Resection for Crohn's Disease: Is It Safe? Diseases of the Colon & Rectum.

2009;52(4):651-6.

207. Canedo J, Lee SH, Pinto R, Murad-Regadas S, Rosen L, Wexner SD. Surgical resection

in Crohn's disease: is immunosuppressive medication associated with higher

postoperative infection rates? Colorectal disease : the official journal of the Association

of Coloproctology of Great Britain and Ireland. 2011;13(11):1294-8.

208. Colombel JF, Loftus EV, Jr., Tremaine WJ, Pemberton JH, Wolff BG, Young-Fadok

T, et al. Early postoperative complications are not increased in patients with Crohn's

279

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

disease treated perioperatively with infliximab or immunosuppressive therapy. The

American Journal of Gastroenterology. 2004;99(5):878-83.

209. Kasparek MS, Bruckmeier A, Beigel F, Muller MH, Brand S, Mansmann U, et al.

Infliximab does not affect postoperative complication rates in Crohn's patients

undergoing abdominal surgery. Inflamm Bowel Dis. 2012;18(7):1207-13.

210. Nasir BS, Dozois EJ, Cima RR, Pemberton JH, Wolff BG, Sandborn WJ, et al.

Perioperative anti-tumor necrosis factor therapy does not increase the rate of early

postoperative complications in Crohn's disease. Journal of gastrointestinal surgery :

official journal of the Society for Surgery of the Alimentary Tract. 2010;14(12):1859-

65; discussion 65-6.

211. Norgard BM, Nielsen J, Qvist N, Gradel KO, de Muckadell OB, Kjeldsen J. Pre-

operative use of anti-TNF-alpha agents and the risk of post-operative complications in

patients with Crohn's disease--a nationwide cohort study. Aliment Pharmacol Ther.

2013;37(2):214-24.

212. Swoger JM, Regueiro M. Preventive therapy in postoperative Crohn's disease. Current

Opinion in Gastroenterology. 2010;26(4):337-43.

213. Ng SC, Lied GA, Arebi N, Phillips RK, Kamm MA. Clinical and surgical recurrence

of Crohn's disease after ileocolonic resection in a specialist unit. European journal of

Gastroenterology & Hepatology. 2009;21(5):551-7.

214. Simillis C, Yamamoto T, Reese GE, Umegae S, Matsumoto K, Darzi AW, et al. A

meta-analysis comparing incidence of recurrence and indication for reoperation after

surgery for perforating versus nonperforating Crohn's disease. The American Journal

of Gastroenterology. 2008;103(1):196-205.

280

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

215. Sachar DB, Lemmer E, Ibrahim C, Edden Y, Ullman T, Ciardulo J, et al. Recurrence

patterns after first resection for stricturing or penetrating Crohn's disease: Inflammatory

Bowel Diseases. 15 (7) (pp 1071-1075), 2009.

216. Scarpa M, Ruffolo C, Bertin E, Polese L, Filosa T, Prando D, et al. Surgical predictors

of recurrence of Crohn's disease after ileocolonic resection: International Journal of

Colorectal Disease. 22 (9) (pp 1061-1069), 2007.

217. Holzheimer RG, Molloy RG, Wittmann DH. Postoperative complications predict

recurrence of Crohn's disease. The European Journal of Surgery. 1995;161(2):129-35.

218. Morar PS, Hodgkinson JD, Thalayasingam S, Koysombat K, Purcell M, Hart AL,

Warusavitarne J, Faiz O. Determining Predictors for Intra-abdominal Septic

Complications Following Ileocolonic Resection for Crohn’s Disease—Considerations

in Pre-operative and Peri-operative Optimisation Techniques to Improve Outcome.

Journal of Crohn's and Colitis. 2015;9(6):483-91.

219. De Cruz P, Kamm MA, Hamilton AL, Ritchie KJ, Gorelik A, Liew D, et al. 1161

Adalimumab Prevents Post-Operative Crohn's Disease Recurrence, and is Superior to

Thiopurines: Early Results From the POCER Study. Gastroenterology. 2012;142(5):S-

212.220. De Cruz P, Kamm MA, Hamilton AL, Ritchie KJ, Krejany EO, Gorelik

A, et al. Crohn's disease management after intestinal resection: a randomised trial. The

Lancet. 2015;385(9976):1406-17.

221. Ajlouni Y, Iser JH, Gibson PR. Endoscopic balloon dilatation of intestinal strictures in

Crohn's disease: safe alternative to surgery. Journal of Gastroenterology and

Hepatology. 2007;22(4):486-90.

281

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

222. Atreja A, Aggarwal A, Dwivedi S, Rieder F, Lopez R, Lashner BA, et al. Safety and

efficacy of endoscopic dilation for primary and anastomotic Crohn's disease strictures.

J Crohns Colitis. 2014;8(5):392-400.

223. Bhalme M, Sarkar S, Lal S, Bodger K, Baker R, Willert RP. Endoscopic balloon

dilatation of Crohn's disease strictures: results from a large United Kingdom series.

Inflamm Bowel Dis. 2014;20(2):265-70.

224. Blomberg B, Rolny P, Jarnerot G. Endoscopic treatment of anastomotic strictures in

Crohn's disease. Endoscopy. 1991;23(4):195-8.

225. Breysem Y, Janssens JF, Coremans G, Vantrappen G, Hendrickx G, Rutgeerts P.

Endoscopic balloon dilation of colonic and ileo-colonic Crohn's strictures: long-term

results. Gastrointest Endosc. 1992;38(2):142-7.

226. Brooker JC, Beckett CG, Saunders BP, Benson MJ. Long-acting steroid injection after

endoscopic dilation of anastomotic Crohn's strictures may improve the outcome: a

retrospective case series. Endoscopy. 2003;35(4):333-7.

227. Couckuyt H, Gevers AM, Coremans G, Hiele M, Rutgeerts P. Efficacy and safety of

hydrostatic balloon dilatation of ileocolonic Crohn's strictures: a prospective longterm

analysis. Gut. 1995;36(4):577-80.

228. de'Angelis N, Carra MC, Borrelli O, Bizzarri B, Vincenzi F, Fornaroli F, et al. Short-

and long-term efficacy of endoscopic balloon dilation in Crohn's disease strictures.

World J Gastroenterol. 2013;19(17):2660-7.

229. East JE, Brooker JC, Rutter MD, Saunders BP. A pilot study of intrastricture steroid

versus placebo injection after balloon dilatation of Crohn's strictures. Clinical

gastroenterology and hepatology : the official clinical practice journal of the American

Gastroenterological Association. 2007;5(9):1065-9.

282

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

230. Endo K, Takahashi S, Shiga H, Kakuta Y, Kinouchi Y, Shimosegawa T. Short and

long-term outcomes of endoscopic balloon dilatation for Crohn's disease strictures.

World J Gastroenterol. 2013;19(1):86-91.

231. Ferlitsch A, Reinisch W, Puspok A, Dejaco C, Schillinger M, Schofl R, et al. Safety

and efficacy of endoscopic balloon dilation for treatment of Crohn's disease strictures.

Endoscopy. 2006;38(5):483-7.

232. Foster EN, Quiros JA, Prindiville TP. Long-term follow-up of the endoscopic treatment

of strictures in pediatric and adult patients with inflammatory bowel disease. J Clin

Gastroenterol. 2008;42(8):880-5.

233. Gustavsson A, Magnuson A, Blomberg B, Andersson M, Halfvarson J, Tysk C.

Endoscopic dilation is an efficacious and safe treatment of intestinal strictures in

Crohn's disease. Aliment Pharmacol Ther. 2012;36(2):151-8.

234. Honzawa Y, Nakase H, Matsuura M, Higuchi H, Toyonaga T, Matsumura K, et al. Prior

use of immunomodulatory drugs improves the clinical outcome of endoscopic balloon

dilation for intestinal stricture in patients with Crohn's disease. Digestive endoscopy:

Official Journal of the Japan Gastroenterological Endoscopy Society. 2013;25(5):535-

43.

235. Dear KL, Hunter JO. Colonoscopic hydrostatic balloon dilatation of Crohn's strictures.

J Clin Gastroenterol. 2001;33(4):315-8.

236. Mueller T, Rieder B, Bechtner G, Pfeiffer A. The response of Crohn's strictures to

endoscopic balloon dilation. Aliment Pharmacol Ther. 2010;31(6):634-9.

237. Nanda K, Courtney W, Keegan D, Byrne K, Nolan B, O'Donoghue D, et al. Prolonged

avoidance of repeat surgery with endoscopic balloon dilatation of anastomotic strictures

in Crohn's disease. J Crohns Colitis. 2013;7(6):474-80.

283

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

238. Ramboer C, Verhamme M, Dhondt E, Huys S, Van Eygen K, Vermeire L. Endoscopic

treatment of stenosis in recurrent Crohn's disease with balloon dilation combined with

local corticosteroid injection. Gastrointest Endosc. 1995;42(3):252-5.

239. Sabate JM, Villarejo J, Bouhnik Y, Allez M, Gornet JM, Vahedi K, et al. Hydrostatic

balloon dilatation of Crohn's strictures. Aliment Pharmacol Ther. 2003;18(4):409-13.

240. Scimeca D, Mocciaro F, Cottone M, Montalbano LM, D'Amico G, Olivo M, et al.

Efficacy and safety of endoscopic balloon dilation of symptomatic intestinal Crohn's

disease strictures. Digestive and liver disease : official journal of the Italian Society of

Gastroenterology and the Italian Association for the Study of the Liver.

2011;43(2):121-5.

241. Singh VV, Draganov P, Valentine J. Efficacy and safety of endoscopic balloon dilation

of symptomatic upper and lower gastrointestinal Crohn's disease strictures. J Clin

Gastroenterol. 2005;39(4):284-90.

242. Stienecker K. Long-term results of endoscopic balloon dilatation of lower

gastrointestinal tract strictures in Crohn’s disease: A prospective study. World Journal

of Gastroenterology. 2009;15(21):2623.

243. Thomas-Gibson S, Brooker JC, Hayward CM, Shah SG, Williams CB, Saunders BP.

Colonoscopic balloon dilation of Crohn's strictures: a review of long-term outcomes.

European Journal of Gastroenterology & Hepatology. 2003;15(5):485-8.

244. Thienpont C, D'Hoore A, Vermeire S, Demedts I, Bisschops R, Coremans G, et al.

Long-term outcome of endoscopic dilatation in patients with Crohn's disease is not

affected by disease activity or medical therapy. Gut. 2010;59(3):320-4.

284

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

245. Williams AJ, Palmer KR. Endoscopic balloon dilatation as a therapeutic option in the

management of intestinal strictures resulting from Crohn's disease. The British journal

of Surgery. 1991;78(4):453-4.

246. Singh VV, Draganov P, Valentine J. Efficacy and safety of endoscopic balloon dilation

of symptomatic upper and lower gastrointestinal Crohn's disease strictures: Journal of

Clinical Gastroenterology. 39 (4) (pp 284-290), 2005. Date of Publication: April 2005.;

2005.

247. Oto A, Zhu F, Kulkarni K, Karczmar GS, Turner JR, Rubin D. Evaluation of diffusion-

weighted MR imaging for detection of bowel inflammation in patients with Crohn's

disease. Academic Radiology. 2009;16(5):597-603.

248. Negaard A, Paulsen V, Sandvik L, Berstad AE, Borthne A, Try K, et al. A prospective

randomized comparison between two MRI studies of the small bowel in Crohn's

disease, the oral contrast method and MR enteroclysis. Eur Radiol. 2007;17(9):2294-

301.

249. Negaard A, Mulahasanovic A, Reisaeter LA, Aasekjaer K, Sandvik L, Klow NE.

Crohn's disease evaluated with magnetic resonance enteroclysis: diagnostic

performance of experienced and inexperienced readers before and after training. Acta

radiologica (Stockholm, Sweden : 1987). 2008;49(9):967-74.

250. Oto A, Kayhan A, Williams JT, Fan X, Yun L, Arkani S, et al. Active Crohn's disease

in the small bowel: evaluation by diffusion weighted imaging and quantitative dynamic

contrast enhanced MR imaging. Journal of Magnetic Resonance Imaging: JMRI.

2011;33(3):615-24.

251. Paparo F, Bacigalupo L, Garello I, Biscaldi E, Cimmino MA, Marinaro E, et al. Crohn's

disease: prevalence of intestinal and extraintestinal manifestations detected by

285

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

computed tomography enterography with water enema. Abdom Imaging.

2012;37(3):326-37.

252. Parente F, Greco S, Molteni M, Anderloni A, Sampietro GM, Danelli PG, et al. Oral

contrast enhanced bowel ultrasonography in the assessment of small intestine Crohn's

disease. A prospective comparison with conventional ultrasound, x ray studies, and

ileocolonoscopy. Gut. 2004;53(11):1652-7.

253. Jensen MD, Nathan T, Rafaelsen SR, Kjeldsen J. Diagnostic accuracy of capsule

endoscopy for small bowel Crohn's disease is superior to that of MR enterography or

CT enterography. Clinical gastroenterology and hepatology : the official clinical

practice journal of the American Gastroenterological Association. 2011;9(2):124-9.

254. Jensen MD, Kjeldsen J, Rafaelsen SR, Nathan T. Diagnostic accuracies of MR

enterography and CT enterography in symptomatic Crohn's disease. Scand J

Gastroenterol. 2011;46(12):1449-57.

255. Yamaguchi A, Matsui T, Sakurai T, Ueki T, Nakabayashi S, Yao T, et al. The clinical

characteristics and outcome of intraabdominal abscess in Crohn’s disease. Journal of

gastroenterology. 2004;39(5):441-8.

256. Lee H, Kim YH, Kim JH, Chang DK, Son HJ, Rhee PL, et al. Nonsurgical treatment of

abdominal or pelvic abscess in consecutive patients with Crohn's disease. Digestive and

Liver Disease.38(9):659-64.

257. Portela F, Lago P, Cotter J, Gonçalves R, Vasconcelos H, Ministro P, et al. Anaemia in

Patients with Inflammatory Bowel Disease - A Nationwide Cross-Sectional Study.

Digestion. 2016;93(3):214-20.

286

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

258. Cabral VL, de Carvalho L, Miszputen SJ. Importance of serum albumin values in

nutritional assessment and inflammatory activity in patients with Crohn's disease.

Arquivos de Gastroenterologia. 2001;38(2):104-8.

259. Gisbert JP, Gomollón F. Common misconceptions in the diagnosis and management of

anemia in inflammatory bowel disease. The American Journal of Gastroenterology.

2008;103(5):1299.

260. Nakarai A, Kato J, Hiraoka S, Inokuchi T, Takei D, Morito Y, et al. Slight increases in

the disease activity index and platelet count imply the presence of active intestinal

lesions in C-reactive protein-negative Crohn's disease patients. Internal Medicine.

2014;53(17):1905-11.

261. Solem CA, Loftus EV, Jr., Tremaine WJ, Harmsen WS, Zinsmeister AR, Sandborn WJ.

Correlation of C-reactive protein with clinical, endoscopic, histologic, and radiographic

activity in inflammatory bowel disease. Inflamm Bowel Dis. 2005;11(8):707-12.

262. Henriksen M, Jahnsen J, Lygren I, Stray N, Sauar J, Vatn MH, et al. C-reactive protein:

a predictive factor and marker of inflammation in inflammatory bowel disease. Results

from a prospective population-based study. Gut. 2008;57(11):1518-23.

263. Vermeire S, Van Assche G, Rutgeerts P. Laboratory markers in IBD: useful, magic, or

unnecessary toys? Gut. 2006;55(3):426-31.

264. Morar PS, Hodgkinson JD, Thalayasingam S, Koysombat K, Purcell M, Hart AL, et al.

Determining Predictors for Intra-abdominal Septic Complications Following

Ileocolonic Resection for Crohn's Disease-Considerations in Pre-operative and Peri-

operative Optimisation Techniques to Improve Outcome. J Crohns Colitis.

2015;9(6):483-91.

287

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

265. Moy MP, Sauk J, Gee MS. The role of MR enterography in assessing Crohn’s disease

activity and treatment response. Gastroenterology research and practice. 2016;2016.

266. Donabedian A. Explorations in quality assessment and monitoring. Vol. II. The criteria

and standards of quality. Ann Arbor, MI: Health Administration Press. 1982.

267. Almoudaris AM, Burns EM, Bottle A, Aylin P, Darzi A, Vincent C, et al. Single

measures of performance do not reflect overall institutional quality in colorectal cancer

surgery. Gut. 2013;62(3):423-9.

268. Bevan G, Hood C. What’s measured is what matters: targets and gaming in the English

public health care system. Public administration. 2006;84(3):517-38..

269. National Bowel Cancer Audit. 2016.

270. Burns EM, Bottle A, Aylin P, Clark SK, Tekkis PP, Darzi A, et al. Volume analysis of

outcome following restorative proctocolectomy. British Journal of Surgery.

2011;98(3):408-17.

271. Carlsson E, Bosaeus I, Nordgren S. Quality of life and concerns in patients with short

bowel syndrome. Clinical Nutrition. 2003;22(5):445-52.

272. Eshuis EJ, Slors JFM, Stokkers PCF, Sprangers MAG, Ubbink DT, Cuesta MA, et al.

Long-term outcomes following laparoscopically assisted versus open ileocolic

resection for Crohn's disease. British Journal of Surgery. 2010;97(4):563-8.

273. Spinelli A, Bazzi P, Sacchi M, Danese S, Fiorino G, Malesci A, et al. Short-term

outcomes of laparoscopy combined with enhanced recovery pathway after ileocecal

resection for Crohn's disease: a case-matched analysis. Journal of gastrointestinal

surgery. 2013;17(1):126-32; discussion p 32.

288

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

274. Oresland T, Bemelman WA, Sampietro GM, Spinelli A, Windsor A, Ferrante M, et al.

European evidence based consensus on surgery for ulcerative colitis. J Crohns Colitis.

2015;9(1):4-25.

275. Morar P SN, Warusavitarne J, Green J, Hart A, Edwards C, Faiz O P400. Setting

standards for multi-disciplinary team driven care in inflammatory bowel disease service

provision - an expert consensus on key specialists to be involved. Journal of Crohn's

and Colitis. 2015;9(suppl 1):S279-S.

276. Morar P, Read J, Arora S, Hart A, Warusavitarne J, Green J, et al. Defining the optimal

design of the inflammatory bowel disease multidisciplinary team: results from a

multicentre qualitative expert-based study. Frontline Gastroenterology. 2015.

277. Bortoli A, Daperno M, Kohn A, Politi P, Marconi S, Monterubbianesi R, Castiglione

F, Corbellini A, Merli M, Casella G, D'Incà R. Patient and physician views on the

quality of care in inflammatory bowel disease: Results from SOLUTION-1, a

prospective IG-IBD study. Journal of Crohn's and Colitis. 2014;8(12):1642-52.

278. Hull T, Joyce M, Geisler D, Coffey J. Adhesions after laparoscopic and open ileal

pouch–anal anastomosis surgery for ulcerative colitis. British Journal of Surgery.

2012;99(2):270-5.

279. Burns EM, Bottle A, Aylin P, Clark SK, Tekkis PP, Darzi A, et al. Volume analysis of

outcome following restorative proctocolectomy. The British journal of Surgery.

2011;98(3):408-17.

280. Øresland T, Bemelman WA, Sampietro GM, Spinelli A, Windsor A, Ferrante M, et al.

European evidence based consensus on surgery for ulcerative colitis. Journal of Crohn's

and Colitis. 2015;9(1):4-25.

289

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

281. Taylor C, Finnegan-John J, Green JS. "No decision about me without me" in the context

of cancer multidisciplinary team meetings: a qualitative interview study. BMC Health

Services Research. 2014;14:488.

282. Sidhom MA, Poulsen M. Group decisions in oncology: doctors' perceptions of the legal

responsibilities arising from multidisciplinary meetings. Journal of Medical Imaging

and Radiation Oncology. 2008;52(3):287-92.

283. Wood JJ, Metcalfe C, Paes A, Sylvester P, Durdey P, Thomas MG, et al. An evaluation

of treatment decisions at a colorectal cancer multi-disciplinary team. Colorectal

Disease. 2008;10(8):769-72.

284. Rajan S, Foreman J, Wallis MG, Caldas C, Britton P. Multidisciplinary decisions in

breast cancer: does the patient receive what the team has recommended? British Journal

of Cancer. 2013;108(12):2442-7.

285. Taylor C, Munro AJ, Glynne-Jones R, Griffith C, Trevatt P, Richards M, et al.

Multidisciplinary team working in cancer: what is the evidence? 2010;340.

286. Fosker C. The Cost of the MDT. BMJ. 2010;340(c951).

287. Lamb BW, Taylor C, Lamb JN, Strickland SL, Vincent C, Green JS, et al. Facilitators

and barriers to teamworking and patient centeredness in multidisciplinary cancer teams:

findings of a national study. Annals of Surgical Oncology. 2013;20(5):1408-16.

288. Jalil R, Ahmed M, Green JS, Sevdalis N. Factors that can make an impact on decision-

making and decision implementation in cancer multidisciplinary teams: an interview

study of the provider perspective. International Journal of Surgery (London, England).

2013;11(5):389-94.

290

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

289. Lamb BW, Brown KF, Nagpal K, Vincent C, Green JS, Sevdalis N. Quality of care

management decisions by multidisciplinary cancer teams: a systematic review. Annals

of Surgical Oncology. 2011;18(8):2116-25.

290. Lamb BW, Sevdalis N, Mostafid H, Vincent C, Green JS. Quality improvement in

multidisciplinary cancer teams: an investigation of teamwork and clinical decision-

making and cross-validation of assessments. Annals of Surgical Oncology.

2011;18(13):3535-43.

291. Lamb B, Payne H, Vincent C, Sevdalis N, Green JS. The role of oncologists in

multidisciplinary cancer teams in the UK: an untapped resource for team leadership?

Journal of Evaluation in Clinical Practice. 2011;17(6):1200-6.

292. Fleissig A, Jenkins V, Catt S, Fallowfield L. Multidisciplinary teams in cancer care: are

they effective in the UK? The Lancet Oncology. 2006;7(11):935-43.

293. Lamb BW, Brown KF, Nagpal K, Vincent C, Green JSA, Sevdalis N. Quality of Care

Management Decisions by Multidisciplinary Cancer Teams: A Systematic Review.

Annals of Surgical Oncology. 2011;18(8):2116-25.

294. Lamb BW, Sevdalis N, Taylor C, Vincent C, Green JSA. Multidisciplinary team

working across different tumour types: analysis of a national survey. Annals of

Oncology. 2012;23(5):1293-300.

295. Akers J, Aguiar-Ibáñez R, Baba-Akbari Sari A. CRD’s guidance for undertaking

reviews in health care. York, UK: Centre for Reviews and Dissemination (CRD). 2009

296. Lingard L, Albert M, Levinson W. Grounded theory, mixed methods, and action

research. BMJ. 2008;337:a567.

297. Sandelowski M. Triangles and crystals: on the geometry of qualitative research.

Research in Nursing & Health. 1995;18(6):569-74.

291

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

298. Karahalios A, Baglietto L, Carlin JB, English DR, Simpson JA. A review of the

reporting and handling of missing data in cohort studies with repeated assessment of

exposure measures. BMC Medical Research Methodology. 2012;12:96.

299. Sterne JA, White IR, Carlin JB, Spratt M, Royston P, Kenward MG, et al. Multiple

imputation for missing data in epidemiological and clinical research: potential and

pitfalls. BMJ. 2009;338:b2393.

300. Hagens V, Dobrow MJ, Chafe R. Interviewee Transcript Review: assessing the impact

on qualitative research. BMC Medical Research Methodology. 2009;9:47.

301. McCarney R, Warner J, Iliffe S, van Haselen R, Griffin M, Fisher P. The Hawthorne

Effect: a randomised, controlled trial. BMC Medical Research Methodology.

2007;7:30-.

292

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

APPENDIX i. Semi-structured interview protocol: construct of an ileal staging system for Crohn’s disease

THEME QUESTIONS Introduction to study Introduce yourself. Duration. i.e. My name is Pritesh Morar. I am a clinical research fellow at St Mark’s studying outcomes in ileocaecal Crohn’s disease. Explain what you are doing and why? i.e. We want to devise a Crohn’s ileocaecal staging system, like Hinchey for diverticular disease or TNM for cancer, in order to look at the impact of radiological stage on medical and surgical management and consequent outcome. i.e. In TNM we know how we would manage things according to its pre-operative staging. We don’t have anything like that in inflammatory bowel disease and it might be quite useful to have that in terms of knowing who you survey, who you give biological therapy too, and who might consider for surgery? Do you agree? Establish whether a radiological Do you think radiology has an important role in guiding staging system is needed or not. management of ileocaecal Crohn’s disease? Why is it important? Free discussion If a radiological staging system were to be developed do you think this would help in management decisions related to ileocaecal Crohn’s disease?

Radiological Stage on Prognosis Do you think it is possible to use a radiological staging system for ileocaecal Crohn’s disease in prognosticating (PROMPT – is it possible to tell someone the likelihood of requiring surgery / biological therapy given the radiological features they will display?) What features do you think would lead to a poor prognosis? What features do you think would lead to a good prognosis? Is it possible to subdivide these features into stages?

Radiological Stage on What specific factors on radiology would lead you to consider Management surgery as a primary option in patients presenting with ileocaecal Crohn’s disease? (PROMPT – What features are important to go straight for surgery) What specific factors on radiology would lead you to consider medical therapy as a primary option in patients presenting with ileocaecal Crohn’s disease? (PROMPT – What features are important to go straight to medical therapy)

293

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

What specific factors on radiology would make it difficult for you to make a decision on surgery versus medicine? (PROMPT – What features make the management decisions difficult). Of all the things that you have mentioned, what features do you think are most relevant in clinical decision-making? If based on the clinical criteria that you mentioned a radiological staging system was developed, how would this impact on your clinical practice? (PROMPT: i.e. would there be an algorithm on the wall, would you still use clinical discretion, what patient factors would influence the decision- making process)

Type of imaging If we were going to do imaging to look at ileocaecal Crohn’s Disease, what modality would you use? (PROMPT – Plain (Questions for radiologists) film, CT, MRI, USS).

(If they choose more than one) Which would you choose in the first instance? Why would you choose that modality in the first instance? Does it matter which imaging modality we use? (PROMPT: example Hinchey can be derived from CT or MRI)

Radiological Features of (Modality of their choice) Ileocaecal Crohn’s (Questions for Can you describe the radiological features of ileocaecal radiologists) Crohn’s disease for your chosen radiological modality?

In your experience which radiological features predict early disease? In your experience which radiological features predict advanced disease? Can we subdivide these radiological features into stages? (PROMPT: like Hinchey). If you were to divide these radiological features into four stages, what would go into each stage? (PROMPT: If Stage 1 was mild inflammation and Stage 4 was free perforation within a complex Crohn’s mass, what would you place in the stages in-between?) All questions above for other two modalities (CT/MRI/USS)

Uses of the staging system If we were to develop this staging system, how could we prove that it works as a staging system? (PROPMT: How

would we validate it?)

294

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

How would you use this radiological ileocaecal staging system? (PROPMT: What would you envisage using this for?)

ii. Semi-structured interview protocol (for expert participants): development of key performance indicators for an IBD surgical service provision

THEMES QUESTIONS Introduction Introduce yourself (name, role)

Obtain consent – (mention that recordings will go to a transcription agency - please maintain anonymity).

Development of KPI to We all aim to provide the highest quality of care to our patients. I measure quality of service am going to talk to you about what defines quality in the care of provisions in IBD Surgery ‘IBD’ patients in particular. Although we have a national ‘IBD’ audit, the standards against which they measure performance are not necessarily the right ones that make a difference in ‘IBD’ care.

The quality of surgical care in So, in the first instance, what are the markers of quality in ‘IBD’ IBD surgery? Probe: What defines quality? What does high quality ‘IBD’ surgery look like? Prompt: From your perspective and from the perspective of the patient?

The structure of surgical care Now I am going to split this into structure, processes and outcomes in IBD to try and determine the ‘KPI’ for each stage. ‘KPI’ are metrics for measuring how well we are doing? Prompt – example is length of stay / morbidity / mortality So, regarding the overall organisational structure in delivering care to IBD patients, what factors are important in ensuring quality: Prompts: - Staffing levels - Facilities’ on wards to deal with these patients. - Dedicated nursing staff - Pathology and radiology services - Equipment - Access to specialists - i.e. IBD specialists, nutritionists, IBD specialist nurses.

295

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

How do we measure them? Prompt: How do we ensure these things are in place to ensure high quality care How can we improve them? Prompt: is there anything you have done in your centre

The processes of surgical care So, regarding the surgical processes integral to delivering care to in IBD IBD patients, which ones are important in ensuring quality pre- operatively, peri-operatively and post-operatively: Prompts (include patient perspective): - Pre-operative factors? (informed consent process / nutrition / timing of surgery / optimising medical therapy) - Perioperative factors? (anaesthesia / consultant led operation / Two consultants?) - Post-operative factors? (Adjuvant medical therapy / Nutrition / Stoma care)

How do we measure them? How can we improve them?

The processes of surgical care So, regarding the surgical processes integral to delivering care to in IBD IBD patients, which ones are important in ensuring quality pre- operatively, peri-operatively and post-operatively: Prompts (include patient perspective): - Pre-operative factors? (informed consent process / nutrition / timing of surgery / optimising medical therapy) - Peri-operative factors? (anaesthesia / consultant led operation / Two consultants?) - Post-operative factors? (Adjuvant medical therapy / Nutrition / Stoma care)

How do we measure them? How can we improve them?

The outcomes of surgical care Which outcome measures are important as quality indicators of in IBD IBD surgery? Prompt: Prospective database collection / 30-day morbidity / mortality

How might we be able to measure outcomes following IBD surgery Prompt: patient experience questionnaires / feedback

Should there be a team meeting reflecting on these outcomes (M&M meeting)

296

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

END of interview: These are all the questions I wanted to ask. Is there anything else that you would like to say or mention?

Adapted Interview Protocol (for patient participants)

Themes Questions Introduction Introduce yourself (name, role)

Obtain consent – remember to mention that recordings will go to a transcription agency - please maintain anonymity.

Development of KPI to measure quality of There are a lot of differences in the quality of care service provisions in IBD Surgery patient receive who undergo inflammatory bowel disease related surgery. Nationally we are trying to improve and standardise the care of patients’ who undergo surgery relating to their inflammatory bowel disease. Although we are collecting information on patients with inflammatory bowel disease, we haven’t got the right standard by which to measure this information. I am going to talk to you about what defines quality of care in patients with inflammatory bowel disease.

The quality of surgical care in IBD What do you think would improve the quality of care in inflammatory bowel disease surgery? The structure of surgical care in IBD I want to talk to you about each stage in the package of care you/patients receive relating to inflammatory bowel disease surgery. How do you think we can arrange the structure of care you receive so that it enhances the quality of care? What do you think are the most important aspects of the care being given? How can we measure these factors that you have mentioned? How can we improve these factors that you have mentioned?

297

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

The processes of surgical care in IBD What do you think is important before the surgery or days/weeks leading up to your surgery?

What do you think is important on the day of your surgery or just before?

What do you expect will make it easier after the surgery done?

How can we measure these factors that you have mentioned? How can we improve these factors that you have mentioned? The processes of surgical care in IBD What do you think is important before the surgery or days/weeks leading up to your surgery?

What do you think is important on the day of your surgery or just before?

What do you expect will make it easier after the surgery done?

How can we measure these factors that you have mentioned? How can we improve these factors that you have mentioned? The outcomes of surgical care in IBD What outcomes following surgery do you feel are important in measuring quality of overall surgical care? (PROMPT: Risk of getting worse? Developing a complication following surgery?)

How do you think we could measure these outcomes that you have mentioned?

If there was a team meeting involved in your care would you like to be involved?

END of interview: These are all the questions I wanted to ask. Is there anything else that you would like to say or mention?

298

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery iii. Delphi survey: Devising Key Performance Indicators to Measure A Surgical Service

Provisions in Inflammatory Bowel Disease

Inflammatory Bowel Disease (IBD) service provision and care is being analysed as part of a general drive to reduce variation in standards of healthcare and to improve the quality of services that patients receive. Standards for surgical treatment of IBD have not been formalised, but the identification of Key Performance Indicators (KPIs) for surgical IBD services will provide consensus-derived standards thereby delivering a tool for measuring quality.

The following stems underlying each statement/question were extracted from semi-structured interviews from a multidisciplinary sample of colorectal surgeons, gastroenterologists, IBD nurse specialists, service provision managers and patients who have undergone recent IBD surgery.

Please provide your rankings for each question or statement.

299

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

Demographics Please fill in details:

Name:

Grade & Post (e.g. consultant surgeon /

gastroenterologist etc.):

Years in current

post:

Centre / Hospital /

Organisation:

City/Town:

Email Address:

300

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

Do you have at least 5 peer reviewed publications in IBD or an accredited academic appointment?

o Yes

o No

Do you have an active or past role (past 5 years) on any one of the following national/international committee on management of IBD: National standards group National audit Registry group IBDQIP (Inflammatory Bowel Disease Quality Improvement Programme) Groups dealing with guidelines or commissioning member of BSG (British Society of Gastroenterology) IBD section NICE (National Institute for Health & Care Excellence) guidance working groups on IBD related drugs or disease specific. i.e. the recent CD and UC NICE appraisal groups ECCO (European Crohn's & Colitis Organisation) Guidelines European working groups National pharmacological advisories

o Yes

o No

301

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

Are you an IBD clinical lead for your trust AND/OR national IBD audit lead AND/OR medical advisor to local NACC (National Association for Colitis and Crohn’s Disease) group?

o Yes

o No

302

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

Outcome measures Not Minimal Some Highly specific to Crohn's Important important importance importance important small bowel or (4) (1) (2) (3) (5) ileocolonic surgery include: the proportion of patients who require re- operations within 30 days secondary o o o o o to intra-abdominal septic complications. the proportion of patients who develop enterocutaneous o o o o o fistulae within 90 days. the proportion of cases performed laparoscopically without conversion o o o o o to open the proportion of patients who are stoma free after 12 o o o o o months the proportion of patients who develop clinical recurrence within o o o o o 12 months. the proportion of patients who develop surgical recurrence within o o o o o 36 months

303

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

the proportion of patients who receive pre- operative nutritional o o o o o optimisation where indicated. the proportion of patients who develop subsequent o o o o o short bowel. the proportion of patients who are started on prophylactic immunomodulatory o o o o o therapy within 6 months after surgery. quality of life measures at 6- month postoperatively using the Crohn's o o o o o Life Impact Questionnaire (CLIQ).

304

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

Outcomes measures required in Not Minimal Some Highly measuring the Important important importance importance important overall quality (4) (1) (2) (3) (5) of an IBD surgical service include: the proportion of readmissions within the six- month o o o o o postoperative period. the length of post-operative inpatient stay. o o o o o the proportion of patients who return to work at six months o o o o o post- operatively. patient satisfaction surveys through IBD specific patient panels, departmental o o o o o open days, or patient opinion websites. overall 30-day morbidity as graded by the Clavien-Dindo o o o o o classification. overall mortality rates 12 months following o o o o o surgery.

305

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery quality of life measures 12 months post- operatively by use of the Inflammatory o o o o o Bowel Disease Questionnaire (IBDQ).

306

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

Quality assurance mechanisms for the delivery of a Not Minimal Some Highly Important high quality important importance importance important (4) IBD surgical (1) (2) (3) (5) service should include: an IBD team meeting discussion for all IBD deaths within 12 months of surgery with the outcome o o o o o of the discussion recorded and submitted to national data collection. an IBD team meeting discussion in the event of significant post- operative o o o o o morbidity (Clavien- Dindo grade ≥ 3). an IBD team meeting discussion in the event of all IBD- o o o o o related post- operative readmissions.

307

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery an IBD team meeting discussion considering the length of post- o o o o o operative inpatient stay is greater than 14 days. discussion in the surgical Morbidity & Mortality meeting in the event of o o o o o surgical deaths within 30 days of surgery. open publication or registry reporting for overall o o o o o morbidity and mortality. a process of credentialing from a national or international governing body for the o o o o o IBD unit to deliver surgical services.

308

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

submission of surgical outcome data into national IBD audits or registries (including o o o o o the IBD audit, ileal pouch registry, IBD registry).

Q15 An individual consultant colorectal surgeon should be performing the following number of IBD major resections per year (including proctocolectomy, subtotal colectomy, ileoanal pouch formation & ileocolonic/small bowel Crohn's surgery):

o Less than 5 (1)

o 5 - 10 (2)

o 11-15 (3)

o 16-20 (4)

o Greater than 20 (5)

Q17 An institution delivering a high quality surgical service should be performing the following

309

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery minimum number of IBD major resections per year (including proctocolectomy, subtotal colectomy, ileoanal pouch formation and ileocolonic/small bowel Crohn's surgery):

o Less than 5 (1)

o 5-10 (2)

o 11-15 (3)

o 16-20 (4)

o Greater than 20 (5)

310

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

Q16 The IBD team members required to Not Minimal Some Highly Important deliver a high important importance importance important (4) quality IBD (1) (2) (3) (5) surgical service, include: a consultant colorectal surgeon trained and experienced o o o o o in laparoscopic surgery. (1) a consultant colorectal surgeon who is a core member of the IBD o o o o o multidisciplinary team. (2) a consultant colorectal surgeon who is on a dedicated IBD on-call rota providing o o o o o emergency IBD surgery when required. (3) a consultant radiologist who is a core member of the IBD o o o o o multidisciplinary team. (4) a consultant gastroenterologist who is a core member of the IBD o o o o o multidisciplinary team. (5)

311

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

a clinical nurse specialist with a special interest and competency in stoma therapy o o o o o and ileoanal pouch surgery. (6) a consultant histopathologist who is a core member of the IBD o o o o o multidisciplinary team. (7) a nominated IBD surgical lead. (8) o o o o o

312

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

Q18 An institution delivering a high quality IBD surgical service, Not Minimal Some Highly for a defined Important important importance importance important population of (4) (1) (2) (3) (5) approximately 250,000 inhabitants, should have: 2 whole time equivalent consultant surgeons o o o o o (dedicated IBD time) minimum. 1 dedicated laparoscopic operating facility. o o o o o 1 IBD-specific endoscopy facility with dye spray, double-balloon enteroscopic and o o o o o dilatation capabilities. 1 dedicated gastroenterology ward. o o o o o an emergency 24- hour joint surgery & gastroenterology IBD on-call o o o o o service. an urgent referral pathway for new IBD patients (new referrals to be seen o o o o o within two weeks).

313

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

once a month minimum parallel or joint gastroenterology o o o o o and surgical IBD clinics. 2 whole time equivalent consultant o o o o o gastroenterologists. 1.5 whole time equivalent clinical nurse specialists with competency in stoma therapy o o o o o and ileoanal pouch surgery. an IBD advice line manned during working hours. o o o o o

Page Break

314

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

Q19 Pre- operative considerations No Minimal Some Highly in the process of Important importance importance importance important delivering high (4) (1) (2) (3) (5) quality IBD surgical care should include: the presence of an IBD nurse specialist during surgical consultations to address social & o o o o o psychological concerns and manage expectations. (1) an outpatient gastroenterology review with a view to optimise the patient's o o o o o medication status. (2) a referral to a dietitian for optimisation of the patient's nutritional status o o o o o where indicated. (3)

315

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

Q20 Post- operative considerations No Minimal Some Highly required in Important importance importance importance important delivery of a (4) (1) (2) (3) (5) high quality IBD surgical service include: adherence to enhanced recovery o o o o o protocols. (1) the use of a shared care IBD bundle / o o o o o booklet. (2) a discussion of all post- operative inpatients in the o o o o o IBD team meeting. (3) a gastroenterology inpatient review to assess the need for medical o o o o o prophylaxis after surgery. (4)

Page Break

316

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

Q21 The first round of the survey is completed.

Your responses will be analysed and incorporated into a second Delphi round with amended stems.

Thank you for your contribution.

Q26 Please feel free to add any further comments or suggestions below:

______

______

______

______

______

End of Block: Default Question Block

317

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery iv. Semi-structured interview protocol (for expert participants): development of key performance indicators for an IBD multidisciplinary team

Themes Questions Construction of the ideal MDT Many healthcare professionals feel the Define Aims, Optimal Design, Format and multidisciplinary team (MDT) meeting a useful Function of an IBD MDT forum to discuss patients and plan on-going medical care. I am going to talk to you about how we might go about constructing the ideal MDT, with a view to defining the aims of the IBD MDT, and optimising the design, format and function of the IBD MDT to meet these aims.

The MDT purpose Is there a role for MDTs in IBD care?

If interviewee has regular MDT: Why did you introduce MDTs in IBD care?

If yes, what is the purpose of the MDT in IBD? The MDT process in IBD How do you feel the MDT process in IBD works at the moment?

Do you think it is good? What factors work well?

What are the characteristics of an Effective MDT: • The team (PROMPT: people present / contribution to discussion) • Infrastructure for Meetings (PROMPT: scans and pathology information present and discussed) • Meeting Organization & Logistics (PROMPT: dedicated time slot and adequate time) • Patient-Centred Clinical Decision-Making (PROMPT: patient view and social situation represented) • Team Governance (PROMPT: Feedback process, peer review)

Do you think there is room for improvement? What areas do not work well?

How can we improve the MDT process?

318

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

How can we measure the improvement? What should we specifically be measuring?

Design of the IBD MDT How can we re-design the IBD MDT to maximise potential? Logistics of the IBD MDT How long should it last?

When should it be scheduled? Protected time?

Who should be present? Who should participate?

What cases should be discussed (PROMPT: Should every case be discussed or just selective cases): - If selected cases only, how do we select these cases? Should there be a protocol?

Should each hospital hold their own IBD MDT or should there be a teleconference links to a central IBD unit?

319

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery v. Delphi survey: Setting Standards for The IBD MDT - Obtaining Consensus Through a

Delphi Survey Approach

Inflammatory Bowel Disease (IBD) service provision and care is being analysed as part of a general drive to reduce variation in standards of healthcare and to improve the quality of services that patients receive. Standards for the IBD MDT (multidisciplinary team) meeting have not been formalised, but the identification of consensus-derived standards will provide a tool for measuring the quality and improving standards of the IBD MDT meeting. The following stems underlying each statement/question were extracted from semi-structured interviews from a multidisciplinary sample of colorectal surgeons, gastroenterologists, histopathologists, radiologist and IBD nurse specialists. Please provide your rankings for each question or statement.

Q2 DEMOGRAPHICS

o NAME: (1) ______

o GRADE & POST: (2) ______

o YEARS IN CURRENT POST: (3) ______

o CENTRE / HOSPITAL / ORGANISATION: (4) ______

o CITY / TOWN: (5) ______

o EMAIL ADDRESS: (6) ______

320

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

Q18 Do you have at least 5 peer reviewed publications in IBD OR an accredited academic appointment?

o Yes (1)

o No (2)

Q19 Do you have an active or past role (past 5 years) on any one of the following national/international committee on management of IBD:

National standards group National audit Registry group IBDQIP (Inflammatory Bowel Disease Quality Improvement Programme) Groups dealing with guidelines or commissioning member of BSG (British Society of Gastroenterology) IBD section NICE (National Institute for Health & Care Excellence) guidance working groups on IBD related drugs or disease specific. i.e. the recent CD and UC NICE appraisal groups ECCO (European Crohn's & Colitis Organisation) Guidelines European working groups National pharmacological advisories

o Yes (1)

o No (2)

Q20 Are you an IBD clinical lead for your trust AND/OR national IBD audit lead AND/OR medical advisor to local NACC (National Association for Colitis and Crohn’s Disease) group?

o Yes (1)

o No (2)

321

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

Q3 The aims of Minimal Some the IBD MDT Not Highly relevance relevance Relevant (4) meeting should relevant (1) relevant (5) (2) (3) be to: advance patient care. (1) o o o o o provide multidisciplinary team input for the patient's care o o o o o plan. (2) provide shared experience and expertise. (3) o o o o o obtain consensus on management for a patient o o o o o with IBD. (4) provide a basis of support and shared decision- o o o o o making. (5) deliver the best possible care for the patient. (6) o o o o o reduce emergency surgical o o o o o procedures. (7) improve patient outcome. (8) o o o o o provide a safety net so patients are not missed. o o o o o (9) provide a forum for research and education. (10) o o o o o

322

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

Q4 The following individuals should be considered as either Extended Member core members, Core Member (1) Non-member (3) (2) extended members or non-members of the IBD MDT meeting: The Consultant Gastroenterologist (1) o o o The Consultant Colorectal Surgeon (2) o o o The Consultant Paediatrician (3) o o o The Consultant Gastrointestinal Radiologist (4) o o o The Consultant Endoscopy Specialist (5) o o o The IBD Nurse Specialist (6) o o o The Research Fellow (7) o o o Junior doctors (Foundation/Core Trainee/ ST3+/StR) o o o (8) The Consultant Histopathologist (9) o o o

The Dietician (10) o o o

The Pharmacist (11) o o o The Psychologist (12) o o o The Outpatient Sister (13) o o o

323

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

The Consultant Dermatologist (14) o o o The Consultant Rheumatologist (15) o o o

The IBD Patient (16) o o o The MDT Coordinator (17) o o o

324

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

Q5 To ensure the IBD MDT meeting Minimal Some Not Relevant Highly runs smoothly and relevance relevance relevant (1) (4) relevant (5) effectively, there (2) (3) needs to be: an alternating chair person who is trained/experienced in chairing MDT meetings and who o o o o o is also a core member of the IBD MDT. (1) a specific question to be addressed. (2) o o o o o a priority to discuss urgent cases and/or IBD inpatients o o o o o first. (3) a submission of clinical cases no later than three working days in o o o o o advance. (4) a designated MDT coordinator with designated administrative o o o o o responsibilities. (5) clear electronic documentation of the MDT discussion outcome o o o o o in the patients’ clinical records. (6) working and regularly maintained technological o o o o o resources. (7)

325

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

a confidential meeting space (bleep free environment, away o o o o o from public or clinical areas). (8) organisational recognition of the IBD MDT and implementation into the job plan o o o o o for core members. (9) a priority to discuss patients who were missed at the previous meeting. o o o o o (10)

Q6 The ideal duration of the IBD MDT meeting should be no longer than:

Not relevant Minimal Some Highly Relevant (4) (1) relevance (2) relevance (3) relevant (5) half an hour (30 minutes). (1) o o o o o one hour (60 minutes). (2) o o o o o one and a half hours (90 minutes). o o o o o (3) two hours (120 minutes). (4) o o o o o

326

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

Q7 There should be an eligibility criterion for selecting patients for discussion in the IBD MDT meeting.

o No (1)

o Yes (2)

Skip To: Q9 If There should be an eligibility criterion for selecting patients for discussion in the IBD MDT meet... = No

327

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

Q8 The eligibility criteria for selecting patients for discussion in the IBD MDT meeting should include:

Not relevant Minimal Some Highly Relevant (4) (1) relevance (2) relevance (3) relevant (5) all complex IBD cases that may warrant o o o o o surgery. (1) all patients on biological therapy. (2) o o o o o all newly diagnosed IBD cases. o o o o o (3) any IBD case, at the discretion of the named consultant clinician o o o o o (who should also be a core member). (4) an annual review of all IBD cases - all IBD cases to be o o o o o discussed in the space of a year. (5)

328

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

Q9 The overall design of the IBD MDT meeting should ensure:

Not relevant Minimal Some Highly Relevant (4) (1) relevance (2) relevance (3) relevant (5) every trust within the UK can hold its own IBD o o o o o MDT meeting. (1) smaller non- specialist institutions should be linked with specialist institutions o o o o o holding IBD MDT meetings (a Hub & Spoke Model). (2) that only trusts that have passed an accreditation process o o o o o should hold IBD MDT meetings. (3)

329

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

Q10 Outcomes measures of the IBD MDT meeting should include:

Not relevant Minimal Some Highly Relevant (4) (1) relevance (2) relevance (3) relevant (5) a record of the number of weeks from MDT meeting discussion to o o o o o surgical review (if requested). (1) a record of attendance of all designated core members of o o o o o the IBD MDT meeting. (2) a measure of elective versus emergency IBD surgery (to ensure early IBD o o o o o MDT discussion leads to planned surgery). (3) a record of the number of weeks awaiting case discussion following case o o o o o submission for MDT discussion. (4)

330

Improving the Quality of Care for patients with Ileocolonic Crohn’s disease undergoing Surgery

Q11 Thank you for your time.

Your answers will be combined with the rest of the participating specialist panel and analysed. Following this a second round Delphi will be sent detailing the consensus of this first round.

Q17 Please feel free to add any further comments in the text box below:

______

______

______

______

______

331