Design and Evaluation of Novel Devices to Facilitate Long Bone Fracture Reconstruction

Hamid Ebrahimi

A thesis submitted in conformity with the requirements for the degree of Doctor of Philosophy Institute of Biomaterials and Biomedical Engineering University of Toronto

Hamid Ebrahimi

Doctor of Philosophy

Institute of Biomaterials and Biomedical Engineering University of Toronto

2016 Abstract

Intramedullary (IM) nailing is the standard of care for adult lower extremity long bone fracture stabilization. In IM nailing, obtaining the correct entry point for nail insertion has been identified as the most important technical aspect of the operation [1]. Upon accurate entry point selection and opening the intramedullary canal, intra-operative reduction is necessary to insert a long guide wire through the proximal and distal fragments to enable subsequent IM reaming and nail insertion. Despite widespread usage of IM nailing, significant surgical challenges arise in accurate entry point selection and obtaining adequate provisional reduction to allow conventional intramedullary guide wire insertion. Such challenges can significantly impede the surgical workflow, introduce surgical complications, requiring additional operative time and radiation exposure to both patients and medical staff, as well as elevating surgical frustration levels.

This thesis focuses on analyzing the IM nailing surgical process and the development and evaluation of two novel surgical tools, FAST, to facilitate entry point selection and FLEX FiRST

ii wire, to aid long bone fracture reduction. FAST (Femoral Antegrade Starting Tool) is a surgical tool that enables maintenance of Kirschner (K) wire anteroposterior (AP) alignment when lateral images are acquired to obtain accurate K-wire positioning in the sagittal plane. FLEX FiRST

(FLEXible Fracture Reduction Steerable Telescoping) Wire is a flexible endoscopic device whose insertion is guided by a proximal joy-stick like controller which enables navigation of the device tip through a malreduced fracture site under standard intra-operative fluoroscopy.

Ultimately, the design of these novel tools can address the lack of connectivity in utilizing sequential 2D fluoroscopic images to achieve 3D alignment and may facilitate the overall surgical workflow in IM nailing of femoral shaft fractures.

iii

Acknowledgments

First and foremost, I would like to thank the people of Canada for their contributions toward my OGS and CIHR scholarships in my entire graduate studies. I have always tried to find a way to pay back to the society by teaching and helping the undergraduate students across three universities of Toronto, Ryerson, and McMaster. I hope the devices I have designed and tested, would be utilized in the operating rooms and help the patients undergo surgery in a more efficient way.

I would like to specially acknowledge the help, support, and guidance of my supervisors, Dr. Cari Whyne and Dr. Albert Yee. Dr. Whyne, has always been an inspiration for all her students. Her trust on my project allowed me to fully enjoy taking responsibility for my Ph.D. project. She has been extremely supportive and helped me flourish an idea to a working prototype in my Ph.D. project. Dr. Albert Yee, my co-supervisor, truly facilitated my understanding of the clinical needs in my project and also provided financial support for me to attend different scientific conferences.

I am grateful for the guidance and invaluable comments from my doctoral committee members Dr. Emil Schemitsch and Dr. Emily Seto. In addition, I would like to extend my gratitude to my external examiners Dr. Paul Kuzyk and Dr. James Johnson for their constructive feedback.

I also want to thank Mr. Mohammad Kazem and Mr. Michael Pozzobon in the device development laboratory at the Sunnybrook Research Institute. Thank you for teaching me machining and helping me optimize my designs.

My words of gratitude also go to my undergraduate supervisor at Ryerson University, Dr. Marcello Papini, who gave me an opportunity to explore the world of biomedical engineering for the first time.

I wanted to thank my fantastic team of five engineers during my eighteen-month internship at Celestica, who taught me the fundamentals of industrial design, Mr. Dale Warner, Mr. Warner Wong, Mr. Andrew Smith, and Mr. David Lekx. The internship definitely helped me become a better designer and apply the experience to my Ph.D. project.

I wanted to thank my lab members at Orthopaedic Biomechanics Lab for their help, support and great memories. My special thanks go to Dr. David Burns, Dr. Ayelet Atkin, Dr. Patrick Henry and Dr. Normand Robert for their help and consultations in my cadaveric testing.

Last but not least, I wanted to thank my parents for providing me with an opportunity to study at a world - class University, and supporting me throughout my entire life and being a motivation to conquer new opportunities in my life.

I would like to dedicate this thesis to my parents. May this serve as a small thank - you for their sacrifices they made throughout their lives for me.

Contents

Acknowledgments...... iv

List of Tables ...... x

List of Figures ...... xi

Chapter 1: Introduction ...... 1

1.1 Anatomy of Femur ...... 1

1.2 Proximal Femur ...... 1

1.3 Femoral Shaft ...... 2

1.4 Distal Femur...... 3

1.5 Femoral Fractures ...... 4

1.5.1 Fracture Types ...... 4

1.5.2 Standard of Care for the Treatment of Diaphyseal Femoral Fractures ...... 7

1.5.3 IM nailing, External Fixation and Plating...... 16

1.6 Methodology ...... 17

1.6.1 Introduction ...... 17

1.6.2 User-centered Design ...... 17

1.6.3 CAD and FEA ...... 18

1.6.4 Experimental Design ...... 18

1.6.5 Performance Evaluation ...... 19

1.7 Motivation and Objectives ...... 19

1.8 Thesis Organization ...... 20

Chapter 2: Surgical Process Analysis ...... 21

2.1 Abstract ...... 21

2.2 Introduction ...... 22

2.3 Research Design and Methods ...... 23

2.3.1 Semi-structured interviews ...... 23

2.3.2 Surgical observations ...... 24

2.4 Results ...... 24

2.4.1 Semi-structured interviews ...... 24

2.4.2 Surgical observations ...... 26

2.5 Discussion ...... 32

2.5.1 Entry point selection analysis ...... 32

2.5.2 Reduction analysis ...... 35

2.5.3 Common challenges and limitations ...... 37

2.6 Conclusion ...... 37

Chapter 3: Femoral Access Starting Tool – FAST ...... 39

3.1 Abstract ...... 39

3.2 Introduction ...... 41

3.3 Methods...... 42

3.3.1 User-centred design ...... 42

3.3.2 Device requirements ...... 42

3.4 Existing technology ...... 44

3.5 Device design features ...... 45

3.5.1 Fixed curved frame...... 45

3.5.2 Rotatable Arm ...... 47

3.5.3 Locking mechanism ...... 48

3.5.4 Cover ...... 49

3.6 Device use protocol...... 49

3.7 Testing...... 51

3.7.1 Formative usability testing ...... 51 vii

3.7.2 Cadaveric Testing ...... 54

3.8 Conclusion ...... 72

Chapter 4: Flexible Fracture Reduction Steerable Telescoping Wire – FLEX FiRST Wire ...... 73

4.1 Abstract ...... 73

4.2 Introduction ...... 75

4.3 Device Requirements ...... 76

4.3.1 Geometry...... 76

4.3.2 Device Functionality ...... 76

4.3.3 Materials and Manufacturing ...... 77

4.4 Existing technology ...... 78

4.4.1 Manual Techniques ...... 78

4.4.2 Navigation Systems ...... 80

4.4.3 Intramedullary Devices ...... 81

4.4.4 Patents ...... 81

4.5 Device Design features ...... 82

4.6 Finite Element Analysis ...... 89

4.6.1 Methods...... 90

4.6.2 Results ...... 94

4.7 Design Parameter Selection ...... 98

4.8 Experimental Evaluation ...... 98

4.9 Results and Discussion ...... 101

Chapter 5: Summary and Future Work ...... 104

5.1 Summary of work ...... 104

5.2 Future Work (FAST)...... 105

5.2.1 Design Optimization ...... 105

5.2.2 Instruction Materials ...... 108 viii

5.2.3 Cadaveric Testing ...... 109

5.2.4 Translation ...... 110

5.3 Future Work (FLEX FiRST Wire) ...... 110

5.3.1 Design optimization – Insertion ...... 110

5.3.2 Design - Reduction ...... 111

5.3.3 Translation ...... 113

5.4 Applications ...... 113

5.5 Contributions...... 114

5.6 Significance...... 114

References ...... 115

Appendices: Patent Submission ...... 129

List of Tables

Table 1 Summary of the indications and contraindications for the three main surgical approaches for femoral diaphyseal fractures ...... 16

Table 2 Guiding questions used in the semi-structured interviews ...... 23

Table 3 Reported number of IM nailing surgeries conducted per year (career average) by interviewed surgeons working at community and teaching hospitals ...... 24

Table 4 Time in minutes for individual phases associated with the IM nailing procedure ...... 31

Table 5 Time spent during the entry point selection and reduction steps and overall fluoroscopy time for the entire surgical procedure in minutes ...... 31

Table 6 List of identified design issues and corresponding mitigation plans ...... 53

Table 7 List of surgical steps read to each surgeon prior to the testing ...... 58

Table 8 Number of acquired fluoroscopic images and required time in minutes to complete the surgical task for three surgeons with varied levels of experience. The highlighted rows show the cases where the device/bone attachment was not sufficiently secured...... 61

Table 9 Observed AP and Lateral K-wire misalignment with respect to IM canal and the results for comparing the device with the free-hand approach. The difference is calculated as the Free- hand angle minus the device angle, a positive value representing an improved orientation outcome using the device...... 62

Table 10 Summary of the qualitative results obtained from short semi-structured interview after the surgeries ...... 63

Table 11 This table shows the high and low factor levels for width, height, pitch, E and force utilized in this study to investigate the geometric, physical effects and their interactions on the radius of curvature ...... 90

Table 12 Design-of-Experiments modeling results ...... 96

List of Figures

Figure 1 Proximal femur anatomy (posterior view) [125] ...... 1

Figure 2 Femoral shaft anatomy [125]...... 3

Figure 3 Distal femur anatomy [125]...... 4

Figure 4 Winquist et al. classification of the degree of comminution shown on a middle third fractured femur. Reproduced with permission from [19] ...... 5

Figure 5 AO classification for proximal femur fracture A) Extraarticular, trochanteric B) Extraarticular, neck C) Articular, head [22] ...... 6

Figure 6 AO classification for diaphyseal fracture A) Simple B) Wedge C) Complex [22] ...... 6

Figure 7 AO classification for distal fracture A) Extraarticular B) Partial articular C) Complete articular [22]...... 6

Figure 8 Femur neurovascular structure [57] ...... 14

Figure 9 Frame External fixation frame with interconnecting tubes for joy-stick like control of displaced fragments [57] ...... 14

Figure 10 Preliminary reduction using traction table [58] ...... 15

Figure 11 Bridge Plating [58] ...... 15

Figure 12 Challenging steps identified by the interviewed surgeons along with the mean difficulty rating on a scale of 1 to 10 ...... 25

Figure 13 Hierarchical decomposition of IM nailing. The seven phases of IM nailing are represented in light grey boxes, with the associated steps below in white boxes ...... 27

Figure 14 Activities associated with the a) Entry point selection and b) Reduction steps in the guide wire insertion phase ...... 28

Figure 15 Direct manipulation of fracture fragments can be performed by making an small incision close to the fracture site and inserting (a) Schanz traction pins or (b) bone hooks to realign fracture fragments ...... 30

Figure 16 K-wire positioning with (a) correct entry point location and orientation on the AP image. Perpendicular lateral images show: (b) correct location, incorrect orientation, requiring AP rotation; (c) incorrect location, correct orientation, requiring AP translation; and (d) incorrect location and orientation of the entry point, requiring AP rotation and translation to obtain correct lateral entry point location and orientation (e) ...... 34

Figure 17 (a) a trocar with multiple parallel holes can be place on the femur head through (b) a cannulated tissue protector to facilitate entry point selection step. (c) an awl can also be used alternatively to open the intramedullary canal and allow the passage of a guide wire...... 35

Figure 18 a) The final prototype and b) the SolidWorks model of FAST showing: the fixed curved frame (1), the rotatable arm (2), (3) the locking mechanism, and the cover (4) ...... 45

Figure 19 The exploded assembly view of the device proximal end. The screw (1) is tightened on the connecting plate (4). The connecting plate (4) is secured to the rotating arm (5) with two screws (3). The rotating arm (5) rotates around the pin (2) which is press-fitted onto the device tip and has a clearance fit on the connecting plate side ...... 46

Figure 20 The pin joint exploded view assembly showing the connection between the rotating arm (1) distal end and the rotating shaft (4). The screw (2) is used to connect the rail (3) to the rotating shaft (4). The rail itself is secured to the rotating arm (1) with four screws. A 0.5 mm gap between the screw (2) and the inner surface of the rotating arm allows a smooth sturdy movement of the connecting shaft (4) ...... 47

xii

Figure 22 The exploded assembly view of the locking mechanism. The shoulder screw (1) is tightened inside the connecting shaft (5). The spring (2) is placed between the shoulder screw (1) and the outer gear (3) to engage both gears in the locked position. The set screw (6) couples the rotational movement of the connecting shaft (5) and the outer gear (3). The axial movement of the outer gear is still possible through a small gap (4) when the trigger (7) is engaged ...... 48

Figure 23 a) Inserting FAST on the greater trochanter area and b) aligning the orientation with the IM canal under image guidance ...... 50

Figure 24 The device is securely hammered into the bone and a cannula is selected for optimal AP entry point location ...... 50

Figure 25 Adjusting the rotatable arm to obtain optimal lateral entry point orientation, cannula adjustment for position and K-wire advancement ...... 51

Figure 26 Synthetic bone set-up with surrounding foam to simulate soft tissue ...... 52

Figure 27 FAST before the implementation of the trigger mechanism for the engagement/disengagement of the outer gear. The outer gear (1) had to be moved laterally with two fingers to disengage from the inner gear (2) and then rotated to manipulate the rotatable arm ...... 53

Figure 28 The device tip showing the broken connecting plate which connects the tip and the rotating arm ...... 54

Figure 30 A dummy torso made out of foam rapped by a rubber insulation (part number 9349K1 McMaster-Carr) (a) and cadaveric testing setup with the leg in the adducted position secured on a custom made extension radiolucent fame with the dummy torso (b)...... 56

Figure 31 This image illustrates the angle between the IM canal and the K-wire in the AP view58

xiii

Figure 32 Placement of the device pins on the GT for trochanteric fossa (piriformis fossa) entry point ...... 67

Figure 33 a) Correct rotatable arm orientation prior to the device insertion versus (b) incorrect orientation of the rotatable arm and (c) misalignment of pins (with 2 pins visible on the AP image) ...... 68

Figure 34 The region shows an approximate insertion site for the device pins ...... 68

Figure 35 The device must be inserted fully into the bone as shown in (a), insufficient insertion may result in loose device/bone attachment, as shown in (b) ...... 68

Figure 36 The device cover is shown with hatched lines and the dotted arrow shows its direction of removal ...... 69

Figure 37 a) The petal design with entry points stacked on top of one another in 3 columns and b) a K-wire (1) inserted through the column 2 which is aligned with the IM canal axis ...... 69

Figure 38 A lateral image showing a portion of the K-wire (solid black line) visible under fluoroscopy with a satisfactory K-wire trajectory (dotted line) ...... 70

Figure 39 Pushing the trigger as shown in the direction of the red arrow will unlock the rotatable arm ...... 70

Figure 40 K-wire trajectory adjustment using the hand holding the drill to indirectly manipulate the rotatable arm ...... 70

Figure 41 Release of the trigger will move it to its original default position due to the spring loaded outer knob and lock the rotatable arm position ...... 71

Figure 42 a) The petal design with entry points in 5 rows allows entry point location adjustment to obtain satisfactory AP K-wire alignment with respect to the IM canal. b) 5 different channels can be selected to achieve an optimal K-wire AP location. The rotatable arm is not moved ensuring maintenance of the satisfactory K-wire trajectory...... 71

Figure 43 Technical challenges associated with the reduction step are shown in a) where AP alignment is achieved, however, in the lateral view the segments are not aligned. b) Further xiv

attempts to reduce the fracture in lateral view (2) may still result in AP malalignment (1). The isometric view (3) in both views (a) and (b) are presented as a reference, visualizing the relative location of the proximal and distal fragments...... 76

Figure 44 The device configuration after navigation through the distal fragment with two independent bend radii ...... 77

Figure 45 Fracture reduction using a) a wrap b) a crutch to elevate the distal fragment c) a mallet to lower the proximal fragment by external pressure [34] ...... 79

Figure 46 A fracture reduction device with an F shape (1) for manipulating the proximal (2) and the distal fragments (3) to obtain alignment ...... 79

Figure 47 Direct manipulation of the displaced fragments via a) a Schanz pin and b) a bone hook [34] ...... 80

Figure 48 the outer tube and a rotatable tip shown in a) the desired configuration presented as a SolidWorks model composed of the b) the machined outer tube and c) outer tube tip making d) the final assembly of the pin joint. Two 1mm diameter pins of 1.5mm length are used to secure either side of the tip and assembled through laser jet welding ...... 83

Figure 49 The final prototype of FLEX FiRST Wire showing: the controller (1), the rigid outer tube (2), (3) the outer tube tip, and the inner (4)...... 84

Figure 50 a) The front view of the assembled prototype and b) the side view of the device with the top cover removed showing 1) knob 1 for rotating the outer tube, 2) knob 2 for bending the outer tube 3) knob 3 for advancing the inner tube and finally 4) knob 4 for bending the inner tube tip and 5) the rotating disc facilitating the bending mechanism ...... 85

Figure 51 a) the SolidWorks model and b) the final device prototype showing 1) the slotted shaft, 2) knob 2 allowing bending mechanism 3) knob 3 responsible for inner tube advancement, 4) the pin allowing independent advancement and bending of the inner tube, 5) threaded shaft advancing the carriage (6) and 7) the pins limiting the carriage (6) degree of freedom, press fitted into the knob 1 with a slide-fit in the carriage (6) ...... 86

Figure 52 The control cable’s tip (1) is attached to the outer surface of the coil (2) ...... 86 xv

Figure 53 Cables (1) attachment on both sides of the outer tube tip (2) and supported with two 1.8 mm tubes (3) ...... 87

Figure 54 Inner tube advancement through the outer tube ...... 87

Figure 55 Transverse view of the device showing a) the outer and inner tubes aligned with the distal end of the fracture viewed on an AP image and b) misalignment of the device with respect to the distal end due to an error introduced by the device rotation ...... 88

Figure 56 The diagram illustrates the error associated with the rotation of the device to be proportional to the distance X and the rotation angle α ...... 89

Figure 57 Illustration showing the generalized endoscope design configuration where a) the cable tip is connected to the helical spring and b) the cable distal end is placed inside the helical spring. The rectangular cross-section is composed of a 3x3 matrix of elements (height x width) ...... 91

Figure 58 The mesh convergence analysis shows the change in the radius of curvature and maximum deformation as a function of the number elements ...... 92

Figure 59 Flowchart illustrating the automation procedure ...... 93

Figure 60 Automated analysis of 32 different endoscopic tip design configurations with pitch, height, E, applied force and width at defined high and low levels. The resultant model enables optimization of these parameters to minimize the radius of curvature outcome measure (enabling tighter bends to be achieved)...... 95

Figure 61 A Pareto chart showing the main effects as hollow parts and excluded effects considered as model errors as solid bars ...... 96

Figure 62 Illustration showing the response surface based on pitch and width factors. An increase in pitch reduces the cable width effect on increasing the radius of curvature...... 97

Figure 63. Experimental testing setup for consistent fracture displacement and rotation simulation shown in a) the lateral and b) the AP views ...... 99

Figure 64. Simulated proximal and distal displaced fragments in a proximal (a, b), mid-shaft (c, d), and distal (e, f) fractures ...... 100 xvi

Figure 65. AP and later images based on common displaced proximal and distal fragment positions in a proximal (a, b), mid-shaft (c, d) and distal fracture (e, f). FLEX FiRST Wire was utilized in all three configurations to traverse the fracture gap ...... 101

Figure 66 A pin design with partial rectangular cross section to minimize unintended rotation 105

Figure 67 An alternative tip design to allow sagittal movement of the rotatable arm ...... 107

Figure 68 The shadow of the handle is shown in this lateral image (the borders of the shadow is shown with a dotted line) ...... 107

Figure 69 Applying a force through a pulley system to the proximal fragment to simulate muscle forces ...... 113

xvii Chapter 1: Introduction

1.1 Anatomy of Femur

The femur or thighbone is the longest, strongest, and fastest growing bone in the skeleton [2,3]. It is located in the upper section of the leg and connects the hip to the knee joint (Figure 1). The dimensions of the femur such as length, global diameter, intramedullary canal diameter and cortex thickness vary according to individual height, sex, age and ethnic group [4].

1.2 Proximal Femur

The proximal femur consists of a head, a neck, a greater, and a lesser trochanter (Figure 1, Figure 2) [2]. The head’s surface is smooth and coated with cartilage in the fresh state forming roughly two-thirds of a sphere and articulates with the pelvis at the acetabulum. The neck connects the head to the body at an angle of about 125 deg., which varies with age, stature, and width of the pelvis. The greater trochanter is located at the proximal and lateral part of the femoral shaft and gives attachment to a number of muscles in gluteal region. Lateral and medial surfaces of the greater trochanter serve the insertion of the tendon of the gluteus minimus and medius, tendon of the piriformis, obturator externus, internus, quadratus femoris and gemelli. The lesser trochanter is a conical eminence situated at the lower and back part of the base of the neck and gives insertion to the tendon of the iliopsoas at its summit.

Figure 1 Proximal femur anatomy (posterior view) [125]

The terminology associated with some anatomical landmarks in the proximal femur is unclear and multiple studies have focused on clarifying these inconsistencies [5–7]. The piriformis fossa, digital fossa, trochanter fossa and tip of the greater trochanter have all been identified as anatomical landmarks for intramedullary (IM) nail insertion. The first three terms are used to describe a single region in the proximal femur. Digital fossa and trochanteric fossa are synonymous terms, with trochanteric fossa being used in more recent literature. The trochanteric fossa is described as the insertion site of obturator externus. This site is further labeled as a deep depression on the inner surface of the greater trochanter. The piriformis fossa, however, is the insertion site of piriformis muscle located in a small, shallow depression on the tip of the greater trochanter. The term “piriformis fossa” has been widely used to describe the trochanteric fossa region [7]. This may explain part of the controversies in the literature with regards to different entry point selection for IM nailing.

The tip of the greater trochanter was recommended by Kuntscher [8] in 1967 as an ideal entry point for IM nail insertion and numerous studies have studied the correct entry point ever since. The exact location of the greater trochanter is defined as the site of the insertion of piriformis muscle [5,6,9]. However, the vertex point of GT [10] and the most proximal point of the GT in the AP radiograph [11] has also been identified as the tip of the greater trochanter.

1.3 Femoral Shaft

The femoral shaft is almost cylindrical in form (Figure 2). It is, narrowest in the middle, with little expansion as it traces proximally, widening noticeably near the distal end. It contains a slight arch (curvature) that is convex anteriorly and concave posteriorly (Figure 2). The shaft has been defined as consisting of anterior, lateral, and medial surfaces, demarcated by three borders [2]. The anterior surface is bounded by the rounded lateral border and the medial border. The lateral surface lies between the lateral border and the posterior border. The posterior border is formed by the linea aspera, a posterior ridge (crest) located within the middle third of the bone. The linea aspera is an attachment site for various muscles including the vastus medialis, vastus lateralis, adductors longus, magnus and brevis, and the biceps femoris - short head. The medial surface is bounded in front by the medial border and by the linea aspera behind. The average femoral diaphyseal canal diameters for European and Afro-Caribbean males have been reported as 14.3 (11-19) mm and 13.4 (11-15.6) mm, respectively [4]. Average femoral canal diameters

are estimated to be smaller in women [4]. With aging the femur expands and the diameter of medullary canal increases which in turn affects the canal shape.

Figure 2 Femoral shaft anatomy [125] 1.4 Distal Femur

The distal femur articulates with the proximal tibia and patella at the knee joint (Figure 3). The distal femur consists of two condyles, lateral and medial, which provide stability and protection from shear in articulation with the proximal tibia. The patellar surface separates the condyles; the interval between the condyles forms the intercondyloid fossa. The medial condyle gives insertion to the tendon of the adductor magnus muscle inserts at the medial condyle at the adductor tubercle, a small projection on its medial aspect. The lateral ligament of the knee-joint and the popliteus muscle attach to the lateral condyle.

Figure 3 Distal femur anatomy [125] 1.5 Femoral Fractures

1.5.1 Fracture Types

Femoral fractures are generally caused by high-energy forces such as motor vehicle crashes in the young adult population, but can also occur under lower energy forces in the elderly due to poor bone quality. Fractures may be closed (skin intact) or open (the bone has punctured the skin) [12]. In both cases, the surrounding musculature may cause the fragment to become misaligned (displaced). Fractures of the human femur occur at an annual rate of approximately 30/100,000 [13]. In the United States it has been estimated that approximately 250,000 proximal femoral fractures occur each year [14]. Such fractures generally require operative intervention to yield sufficient internal stabilization for healing.

There are number of different systems proposed for fracture classification [15–18]. Femoral fractures are commonly classified in terms of the level of the fracture, configuration and/or degree of comminution. Winquest et al. in 1984 introduced a classification system based on the fracture level (proximal third, middle third and distal third) [1]. The proximal and distal thirds were further divided based on the configuration into transverse, oblique and comminuted fracture groups. The middle third was divided into segmental, segmental oblique and comminuted and spiral fracture groups. The degree of comminution was categorized on a scale of 1 to 4 as follows: Grade 1 - one small broken piece; Grade 2 - butterfly fragment with 50 per cent intact cortical contact; Grade 3 - large butterfly fragment; and Grade 4 - severe comminution with no direct contact between the main fragments [1] (Figure 4).

Figure 4 Winquist et al. classification of the degree of comminution shown on a middle third fractured femur. Reproduced with permission from [19]

The Arbeitsgemeinschaft fur Osteosynthesefragen (AO) system is widely accepted for the long bone fracture classification [18,20,21]. In the AO classification of femur fractures, the bone is divided into proximal, diaphyseal and distal sections. Proximal fractures (“hip fractures”) are generally divided into, trochanteric (31-A), Neck (31-B) and head fractures (31-C) [22] as shown in Figure 5. Hip fracture is an injury characteristic for older patients with osteoporosis [23].

One of the most common major injuries that an orthopaedic surgeon will treat is femoral shaft fractures [2]. In the AO classification system, all diaphyseal (subtrochanteric, midshaft and distal shaft) femoral fractures are designated as 32 (3 representing the femur and 2 representing the diaphyseal segment). General sub classifications for diaphyseal fractures in adults are simple (A), Wedge (B) and Complex (C) as shown in Figure 6.

The majority of distal femoral fractures occur during sports activities in individuals over 10 years of age [3]. Based on the AO classification, there are three distinct fracture types of distal femoral fractures: Extra-articular (33-A), Partial articular (33-B) and Complex articular (33-C) (Figure 7).

A B C Figure 5 AO classification for proximal femur fracture A) Extraarticular, trochanteric B) Extraarticular, neck C) Articular, head [22]

A B C

Figure 6 AO classification for diaphyseal fracture A) Simple B) Wedge C) Complex [22]

A B C Figure 7 AO classification for distal fracture A) Extraarticular B) Partial articular C) Complete articular [22]

1.5.2 Standard of Care for the Treatment of Diaphyseal Femoral Fractures

1.5.2.1 Non-Operative Treatment

Non-operative treatment is generally considered as the only therapy necessary for paediatric diaphyseal femur fractures [24]. Even though operative treatment has been proposed in some recent studies [25], no superior difference in long-term outcomes has been found for these fractures between surgical and non-surgical approaches [26].

Gallows traction is the common treatment of choice exclusively for children under the age of two; this involves the application of skin traction to both legs. The use of cast splintage and walking spica casts have also been reported for treatment of low energy femoral shaft fractures for children one to six years old [27]. Based on the type of injury, casts may extend across the abdomen and down both legs after the initial balanced skeleton traction.

1.5.2.2 Operative Treatment

Operative treatment of diaphyseal femoral fractures with secure internal fixation has great advantages in reducing both hospitalization periods and the risk of non-union [26]. However, operative procedures expose the patient to common surgical risks such as complications from anesthesia, bleeding problems (injury to nerves and blood vessels), fat emboli, infection and the potential for implant failure.

1.5.2.2.1 IM Nailing

Gerhardt Küntscher developed diaphyseal IM nailing and his first nail was inserted for the treatment of a patient with subtrochanteric fracture in 1939. The design of the IM nails has changed from initial thick, straight, slotted, cloverleaf shaped, stainless steel nails to thinner, curved, unslotted, solid or cannulated nails made of enhanced stainless steel, new titanium alloys or polymers enforced by carbon fiber. Despite these design changes, Küntscher’s focus on the biological advantages of IM nailing for fracture care remains as a fundamental principal in modern IM nailing [19]. Today, IM nailing is the standard treatment for adult diaphyseal femoral fractures with union rates approaching 97% [28]. This approach consists of a long nail placed

surgically from the proximal (antegrade) or distal (retrograde) end of the femur, down the centre of the bone and fixed to stabilize the fractured fragments of the femur.

In antegrade nailing, the skin incision is located by marking the intersection point of trochanter’s tip line and the femoral axis. A 3 - 7 cm incision based on the type of the nail is necessary for manual insertion of a K-wire under image intensifier control [29]. The K-wire can be either inserted on the tip of the greater trochanter or trochanteric fossa to open the intramedullary canal for subsequent insertion of a guide wire. Extensive literature has investigated the advantages/disadvantages of both approaches and a summary is presented in chapter 3.

In the retrograde nailing, the guide wire is inserted through a predrilled entry point located in the middle of the intercondylar notch. Advancement through the proximal fragment is up to the area of the greater trochanter where good purchase can be achieved to prevent guide wire extrusion during reamer exchange. Guide wire insertion during antegrade or retrograde nailing is generally performed under real time fluoroscopic visualization.

Once the guide wire is inserted through the proximal fragment, the fracture must be aligned (reduced) so that the guide wire can be advanced to the distal end. Various techniques have been proposed to facilitate obtaining and maintaining the reduction. A summary of these techniques is presented in chapter 4.

Upon successful advancement of the guide wire, reaming may be performed. Reaming is the process of enlarging the bone canal to allow insertion of a larger nail. IM nails may be inserted with or without reaming. Studies have shown that reamed IM nailing leads to lower rates of delayed union and non-union [30]. Reaming is a safe procedure with low incidence of systemic embolism compared to the unreamed insertion of IM nails [31]. It has also been found that limited reaming may be beneficial to the acute management of shaft fractures [32]. Additionally, with reamed nailing, a larger diameter implant can be used which may provide greater initial stability to the fractured femur. If a reamed nailing procedure is chosen, reaming begins with a small diameter reamer and progresses incrementally to a size 0.5 to 1 mm bigger than the selected nail diameter [19]. The progressive use of larger reamer heads avoids the introduction of sudden stresses into the bone canal [33].

The nail length is determined by marking the proximal and distal end of the nail on the uninjured femur (if available). Subsequently, the nail diameter is determined through templating based on fluoroscopic images obtained over the femoral isthmus [33].

Following reaming (if completed), the nail is advanced manually along the guide wire. It is recommended that the nail is rotated approximately 90 degrees between the entry point and its final position so that it follows the anticurvature of the femur [34]. In order to secure the nail, proximal and distal locking screws are inserted through the nail under fluoroscopic image guidance.

Radiation Exposure in IM nailing

The average fluoroscopic use time for femur fracture reduction is reported to be between 4 to 5 min for experienced surgeons, and approximately 14 min for surgeons-in-training [35]. However, in complicated cases the fluoroscopic use time can increase up to 30 minutes [36].

The International Commission on Radiological Protection (ICRP) has recommended a maximum permissible dose (MPD) of 50mSv per year for the whole body, head, neck, trunk, eyes, bone marrow and gonads [37]. The radiation dose to patients’ gonads was reported to be 1.36mSv and between 0-1.62 mSv in two separate studies [35,38] during femoral intramedullary nailing. The radiation dose to the surgeon’s hand was reported to be 2.02 mSv [39] and 1.272 mSv [38] in femoral IM nailing. This suggests that a patient with polytrauma and a surgeon performing approximately 50 surgeries (considering the higher reported value) may exceed the annual limit for radiation exposure.

Entry point

The trochanteric fossa (also known as “piriformis fossa”) and the tip of the greater trochanter are two anatomical landmarks which are often used as a reference point for insertion of IM nails. An IM nail designed for use with a trochanteric fossa entry point is uniplanar with an anticurvature (available in different radii of curvatures to match the patient’s bone). The surgeon often chooses the appropriate nail by placing it next to the contralateral leg under image guidance to observe the fit with respect to the bone anticurvature. The trochanteric fossa is widely recommended for the entry point location as this site is believed to be aligned with the IM canal. Yet, a recent study of 23 femora found that the trochanteric fossa was aligned with the IM canal only in 68%

of cases [6]. Hence, the trochanteric fossa may not be assumed to be universally aligned with the IM canal. In addition, in utilizing uniplanar nails the proximal anatomy must be analyzed to estimate the correct location to ensure it is aligned with the IM canal. The device position and orientation must be checked on both AP and lateral views to confirm the entry location. Using a trochanteric fossa entry point, additional challenges may exist in obtaining access to the surgical site particularly when the patient is obese or extremely muscular. Techniques such as leg adduction or lateral positing of the patient can be implemented to facilitate access to the entry site.

A more lateral approach, using an entry point at the greater trochanter, when the patient is in a supine position may be preferable with respect to ease of access and reduced soft tissue damage. IM nails have been designed for this scenario with an additional proximal lateral bend to account for a more lateral entry point [40]. In addition, less soft tissue damage is reported to be associated with a greater trochanter entry point in comparison to the trochanteric fossa [41].

Regardless of the site of entry, for a given IM nail design the entry point must be accurately located. An incorrectly chosen entry point may lead to complications such as iatrogenic fractures [42], with a too medial entry point increasing the risk of iatrogenic fracture of femoral neck and a too lateral entry point resulting in a blast out of the medial cortex. In a more proximal fracture a too lateral entry point could also lead to a varus deformity [19] and fracture healing issues.

Delayed union, Nonunion, Malunion

Fracture healing can be divided into primary (direct, cortical) bone healing and secondary (indirect, spontaneous) bone healing. Primary bone healing is a biological process for internal remodeling which occurs only under rigid stabilization. Primary bone gap and contact healing can lead to bone union without external callus formation and any fibrous tissue or cartilage formation within the fracture gap [43]. In contrast, secondary healing occurs via callus formation occurs under conditions of relative stability. The classical four stages of secondary healing are inflammation, soft callus formation, hard callus formation, followed by remodeling [44]. In this, low levels of interfragmentary movement have been shown to stimulate callus formation and in turn accelerate healing [45]. However, excessive interfragmentary movement results in instability which leads to nonunion.

Aseptic nonunions are divided into hypertrophic and atrophic nonunions. Hypertrophic nonunion is associated with instability at the fracture site whereas atrophic nonunion is associated with inadequate or poor blood supply [46]. In order to classify a non-union, callus formation is assessed. The absence of progressive fracture healing for three months is defined as delayed union and lack of healing after 6 months as nonunion [47]. The treatment of fracture non-union highly depends on its causation. Hypertrophic nonunion must be fixed with fixation sufficient to provide the required stability at the fracture site. For the treatment of the atrophic nonunion, fracture stability as well as enhancement of the poor biologic environment must be provided to facilitate bone formation [46].

In a study of 520 closed intramedullary nailing of femoral fractures femoral nonunions following reamed IM nailing has been reported to be as low as 0.9 percent [1]. In a separate multicenter randomized clinical trial, femoral shaft fractures without reaming were found to be associated with a significantly higher rate of nonunion (7.5 percent) compared to IM nailing with reaming (1.7%) [48]. Improved implant mechanical purchase, greater fracture stability [48] and altered blood flow to the bone and local muscles [15] in reamed femurs were identified to result in higher union rates. The creation of diastasis between proximal and distal fragments by the nail in unreamed femurs [19] has also been presented as a cause of nonunion/delayed union rates in unreamed femurs.

Reamed locked intramedullary nailing is the preferred treatment for most femoral nonunions [46]. However, indirect reduction and plating of femoral shaft nonunions have also shown promising results [49]. The use of plate fixation is limited in the treatment of femoral fractures due to the potential introduction of increased damage to vascularity [46]. If however, plating is selected as the primary treatment option, use of minimally invasive techniques, avoidance of fragment devitalization with clamps and retractors and bridging techniques are recommended to decrease the incidence of nonunion rates [15].

The definition of malunion is diverse among various studies in the literature. Malunion is often used to express the degree of femoral length shortening, angular and rotational malalignments. The most commonly cited definition is presented by Ricci et al. as greater than 5 degrees of angular deformity, greater than one centimeter of length deformity or greater than or equal to 10 degrees of rotational deformity [47]. In this, the angular deformity is defined as angulation in the coronal or sagittal plane. The angular deformity is often seen in proximal and distal thirds of the femur due to the absence of cortical contact with the nail. In addition, this deformity may be more prominent in proximal and distal fractures due to the higher dependency of angular reduction on the starting point. In proximal fractures, accurate entry point selection has been identified to be associated with minimizing angular deformities [15].

In femoral rotational malalignment the angle between the femoral neck axis and the posterior condyles of the femurs is measured. If the measured angle is above 10 [50] or 15 [15] degrees, the femur is considered to be in rotational malalignment. This method of rotational alignment measurement can be utilized intraoperatively on the intact contralateral femur and adapted to the fractured side. The bone diameter and differences in the cortical thickness at the fracture level can also be used to correct rotational deformity [19].

1.5.2.2.2 Fixion Nails

Carbo-Fix offers the Fixion nail as an alternative to traditional locking nails. The Fixion Nail is an intramedullary, expandable, and inflatable nail that can be inserted into the tibia, femur or humerus via a typical access. Prior to insertion the fractured bone must be reduced but no reaming is required. The nail is inserted into the canal in its folded state through the proximal bone into the distal fragment. Saline is driven into the nail with a pressure of 70 bar which expands the nail within the canal [51]. The press-fit is designed to prevent the nail from rotation,

removing the need for lateral locking. This is proposed to lower operative time, radiation exposure, blood loss, and to require less extensive soft tissue dissection [52]. Multiple studies have been conducted on the Fixion Nail including computational modeling and cadaveric testing.

Clinical testing has also been conducted to compare traditional locked intramedullary nailing to the Fixion Nail in age and sex matched controls (n=43 per group). The reported mean surgery time was significantly shorter for the Fixion nailing and exposure to radiation was minimized [53]. In another study, the expandable Fixion nail was compared with locked intramedullary nailing for treatment of 46 patients experiencing AO type 32A and 32B1 femoral mid-shaft fractures. The operative time, estimated blood loss, radiation exposure and healing time were significantly lower when using the Fixion nails compared with locked nailing [54]. It is important to mention that the inflatable expandable nail is significantly more expensive (as high as five times) compared to traditional nails [53].

1.5.2.2.3 External Fixation

The use of external fixation has been reported in multiple studies for distraction osteogenesis in treatment of limb-length discrepancy, non-unions, bone defects, deformity correction, infections and traumatic injuries [55]. While this method offers several advantages including minimal blood loss and limited surgical exposure, less than 10% of all external fixators are used to treat femoral lesions and they are generally utilized only for pediatric lesions [56]. This technique has also been suggested in cases of extensive soft-tissue damage, severe contamination and prolonged patient length of stay in intensive care units [57].

External fixation stabilizes the femur fracture by inserting pins into the fractured fragments attached through a longitudinal frame support. Prior to pin insertion, provisional manual reduction and realignment of fracture fragments are conducted while maintaining axial traction for accurate pin placement. Subsequently, three pins, widely separated from each other, are inserted into each main fragment 2-3 cm away from the fracture line to obtain optimal frame construction [57]. The placement of the pins in each fragment is crucial in preventing injuries to principal neurovascular structures that surround the femur such as femoral nerve, sciatic nerve and hip and knee joints. Therefore, in order to prevent injuries to neurovascular structures the anterolateral and direct lateral regions are considered as safe zones for pin insertion Figure 8.

Figure 8 Femur neurovascular structure [57]

Figure 9 Frame External fixation frame with interconnecting tubes for joy-stick like control of displaced fragments [57] After pin insertion, based on the fracture location and surgeon’s preference, a frame configuration is constructed. In order to reduce the fracture, two small tubes are utilized to connect the three pins in each main fracture fragment (Figure 9). The tubes are connected loosely via an interconnecting tube which provides a joy-stick like control for appropriate manipulation of the fracture fragments under visual guidance via an image intensifier. Once reduction is achieved the connecting tube’s clamp is fully tightened. The frame configuration can be further supported using a secondary connecting tube [57].

1.5.2.2.4 Plating In the cases where indirect reduction is not possible due to fracture complexity or access to an image intensifier is not available / not recommended, plating can be used for femoral fracture fixation. Bridge plating is a less demanding procedure from a technical perspective as it allows direct reduction. However, plating leads to greater blood loss and less appealing cosmetic results. The preoperative planning for plating consists of the choice of the implant and determining the required length and number of screws for plate fixation. Additionally, conventional plates need to be contoured to match the surface of proximal and distal femur. The preliminary reduction is conducted by using a traction table or manual traction to restore bone length, axis, rotation, and tension in the soft tissue (Figure 10). For this open procedure an incision is made between the greater trochanter and epicondyle. An external fixator or distractor is used to maintain the fracture fragment alignment prior to planting. The plate is secured to the main proximal fragment using verbrugge forceps with titanium or stainless steel screws. The rotation assessment is conducted by comparing the shape of lesser trochanter with the contralateral side. Finally, the plate is fixed to both fragments with at least three screws on either side (Figure 11) [58].

Figure 10 Preliminary reduction using traction table [58]

Figure 11 Bridge Plating [58]

1.5.3 IM nailing, External Fixation and Plating

A summary of indications and contraindications for IM nailing, external fixation and plating for treatment of diaphyseal femoral fractures is presented in Table 1 [15].

Table 1 Summary of the indications and contraindications for the three main surgical approaches for femoral diaphyseal fractures

Approach/ Indications- IM nailing External Fixation Plating Contraindications Indications - Most fractures - Soft tissue damage - Narrow IM canal (isolated, - Soft tissue / IM - Fracture around a comminuted, open) Contamination previous malunion - Stabilization prior to - Vascular injury definitive treatment (temporary fixation) - Vascular injury [19] (temporary fixation) - Periprosthetic and [19] peri-implant fractures

Contraindications - Narrow IM canal No contraindications - Due to the success of - Ipsilateral femoral uncommon treatment IM nailing, it is not neck and acetabular commonly used fractures - Open growth plates - Previous malunion - IM infection - Multi-trauma patients with thoracic injury

1.6 Methodology

1.6.1 Introduction

In this clinical engineering thesis, needs assessment followed by device design and evaluation were conducted to meet the identified requirements. While distinct from hypothesis driven research, this approach follows the process of medical technology innovation as described through biodesign [59]. A user-centered design approach was chosen for the development of the surgical tools which in turn, necessitated the understanding of the surgical process to determine end-users’ needs. Computer Assisted Design (CAD) was used to create the initial tool prototypes. Finite Element Analysis (FEA) was also used in the design process to evaluate the performance of specific components under various loading conditions in combination with a Design-of-Experiments (DOE) approach in order to understand the effects of input parameters and their interactions on the response outcomes of the simulation. Prototyping was conducted using a combination of traditional machining, water jet technology, laser welding and additive manufacturing (3D printing). Finally, formative usability testing was conducted using synthetic and cadaveric specimens to evaluate device usability and performance. In the following sections a brief introduction of these methodologies are presented.

1.6.2 User-centered Design

Throughout the development of the surgical tools in this project, extensive attention was paid to end-users’ needs and limitations. This process involved systematic surgical observations and semi-structured interviews at the early stages of this project to understand the clinical and end users’ needs. The ideation and brainstorming with clinicians and engineers during the design and the early stages of prototyping allowed the evaluation of relevant design concepts suited to the intended device use context. By conducting usability testing, the device ease-of-use and functionalities were tested and surgeons’ comments along with quantitative results were utilized in further design optimization. This optimization process was iterative with changes recorded to provide a platform for further technology development and its ultimate use in the operating room.

1.6.3 CAD and FEA

Computer Aided Design enables the creation of individual parts and integrated components generating accurate technical drawings and file formats required for automated manufacturing. SolidWorks (Dassault Systemes SolidWorks Corp.) was utilized as a CAD software to model all the necessary components of the devices presented in this thesis.

While the use of analytical methods is important in design and in understanding simple devices, this is often performed in isolation from surrounding complex physical interactions. The development of complex models for the analysis of more sophisticated assemblies it is often carried out using numerical methods. Such computational analysis of device components is often important in evaluating and optimizing a design prior to the prototyping stage. Finite Element Analysis enables the structural evaluation of complex parts and systems. FEA is a numerical approach by which complex problems can be solved in an approximate manner but instead of seeking approximate solutions that hold over an entire region, the region is subdivided into smaller parts, so called elements. The approximation is carried out over each element and these results combine to give an approximate solution for the entire body. FEA can be used to understand the behavior of a system under specific load conditions in which the model is meant to operate, and under extreme load conditions for risk assessment. FEA can reduce the need, time and cost of repeated physical prototyping by using computational simulations to direct design optimization.

1.6.4 Experimental Design

There are various methodologies which can be used to understand the effects of input variables on the response measures of a system. Generally, a system can be treated as a black box, receiving input variables and outputting response measures. The one-factor-at-a-time (OFAT) approach makes adjustments in only a single factor for each analysis while all other factors remain fixed. This approach is limited when multiple input factors may be considered and is unable to analyze potential interactions between individual input variables. The Design-of Experiments (DOE) approach enables the adjustment of multiple factors simultaneously (considering high, low and mid-level values in a multidimensional design space) in order to increase the efficiency of a multi factor analysis and to yield information on the relative impact of factors and their interactions.

1.6.5 Performance Evaluation

In the development of medical technology performance evaluation is needed at multiple levels prior to clinical testing. While some scalable tools and devices may be suitable for preclinical testing in ex and in vivo animal models, synthetic and human cadaveric bones are often used in the testing of orthopaedic tools and devices. Synthetic bone is often used to simulate human bone during initial in vitro testing for testing orthopaedic implants and devices, based on its lower relative cost, availability, and uniformity in shape and material properties. Follow-up cadaveric testing is important in evaluating the impact of geometric and material heterogeneity, including the presence of soft tissue and muscles, on the performance, usability and functionality of orthopaedic devices. It is also important to consider formative usability testing within a simulated operating room environment, to understand the potential barriers to use and performance, which may not be evident under laboratory conditions.

1.7 Motivation and Objectives

Traumatic musculoskeletal injury represents a significant cost to health care systems in Canada and worldwide. As operative resources are stretched, improving the surgical workflow in fracture treatment is becoming increasingly important. The proposed research is focused on the most common treatment for adult diaphyseal femoral fracture fixation, IM nailing. The general hypothesis of this proposal is that surgical process analysis can elucidate the underling causation of challenges in the IM nailing procedure and guide the development of novel surgical instruments that can address these clinical barriers. Hence, the objectives of the proposed research are to describe and identify challenges in the IM nailing workflow and to design and evaluate surgical tools to facilitate these challenging steps in IM nailing of diaphyseal femoral fractures.

Specific Aim 1: To analyze the IM nailing surgical process for identification and better understanding of the underlying causation of its associated challenges.

The findings of this study motivated the development of two surgical tools to address the identified clinical needs associated with the entry point selection and the reduction steps.

Specific Aim 2: To design, prototype and test a surgical instrument to facilitate entry point guidance in IM nailing.

Specific Aim 3: To design, simulate and prototype a device that can be navigated through the fracture fragments in diaphyseal femoral fractures.

1.8 Thesis Organization

The femoral IM nailing process is analyzed to better understand common techniques, barriers and opportunities to optimize the surgical procedure in chapter 2. Based on this study, two steps were found to be associated with high levels of frustration and extended operative time. Two devices are proposed to address the underlying causation of these challenges. Chapter 3 describes the design, prototyping and testing of FAST, a Femoral Antegrade Starting Tool. Subsequently, Chapter 4 describes the development of the FLEX FiRST Wire, a Flexible Fracture Reduction Steerable Telescoping Wire. Finally, in Chapter 5 a summary of the thesis contributions and future development of both devices are presented.

Chapter 2: Surgical Process Analysis

This chapter is based on the manuscript entitled: “Surgical process analysis identifies lack of connectivity between sequential fluoroscopic 2D alignment as a critical impediment in femoral intramedullary nailing” published in the International Journal for Computer Assisted Radiology and Surgery.

2.1 Abstract

Purpose: Identifying key steps and barriers within complex and simple surgical procedures can be accomplished in a structured and rigorous manner using surgical process modeling. For lower extremity long bone fracture stabilization, the current standard of care is closed intramedullary (IM) nailing, which despite its widespread use is associated with challenges that greatly impact operative time and lead to the frustration of medical staff. The aim of this study was to identify challenging surgical steps in IM nailing and understand their underlying causation.

Methods: Eight semi-structured interviews with staff orthopaedic surgeons and eight detailed surgical observations were conducted to understand the surgical steps, challenges and adapted techniques used in IM nailing. Hierarchical decomposition was then utilized to structure the IM nailing surgical procedure into phases, steps and activities.

Results: In the developed IM nailing surgical process model, the most challenging steps were identified as fracture reduction (75%) and entry point selection (25%), both of which were associated with high levels of frustration in the observed surgeries. Both of these steps utilize 2D fluoroscopic imaging to guide 3D alignment. Challenges arise when the alignment in one plane is lost while adjusting the alignment in the perpendicular plane. This leads to unpredictable repetition of activities which can be time consuming and frustrating.

Conclusions: Identifying the causation of surgical challenges in IM nailing through surgical process modelling forms a knowledge base that can be used to guide future improvements to techniques and surgical instrumentation.

2.2 Introduction

IM nailing is a minimally invasive surgical (MIS) procedure typically performed under a general anaesthetic and with fluoroscopic image guidance. The surgical phases associated with this type of surgery are well defined: patient preparation, access to the bone entry site, IM guide wire insertion (including fracture reduction for guide wire placement), IM nail placement, locking of the nail to control rotation and length, final clinical and radiographic assessment of fracture reduction/restoration of length/alignment/rotation, and surgical wound closure. Yet, despite widespread usage of IM nailing, significant surgical challenges may arise. Such challenges can significantly impede the surgical workflow, requiring additional operative time and radiation exposure to both patients and medical staff. Fracture reduction and proper localization for initial IM access are particularly challenging areas in the workflow pathway [60,61]. Lengthy delays in the procedure and unacceptable fracture reduction or stabilization can also significantly endanger patient safety, particularly in those who may suffer from poly-trauma and/or acute respiratory issues.

Clear description and analysis of the surgical process can expose challenging steps and activities that can also be used to develop a surgical process model [62]. The use of surgical process modeling may guide development of new surgical technologies/methodologies, improve surgical planning, assist surgical teams during the intervention and enhance surgical training and education [63].

Various methodologies have been proposed to generate individual surgical process models from collected intraoperative data [62,64]. Generated individual surgical process models can then be utilized to develop generic surgical process models [65–67], evaluate a navigation system [68], compare procedure parameters such as operating time, adapted techniques, senior vs. junior surgeon performance [69] and the sequentiality of activities during surgeries [70]. However, implementation of identified surgical models to understand surgical barriers and their root sources have received little attention [71]. Hence, the aim of this study was to scientifically map the IM nailing process, identify surgical challenges that impede the surgical process and elucidate their underlying causes.

2.3 Research Design and Methods

2.3.1 Semi-structured interviews

This study consisted of semi-structured interviews and surgical observations, conducted under institutional Research Ethics Board approval (#122-2013). Orthopaedic surgeons were identified by the research team and invited to participate in a semi-structured interview designed to allow participants to describe their work and experiences (Table 2).

Table 2 Guiding questions used in the semi-structured interviews

 How many femoral IM nailing surgeries do you perform per year?  Do you follow the proposed IM nailing procedure as listed below? If you are not following the same procedure, can you specify the steps you add/eliminate or modify?  What is/are the most challenging step(s)?  Can you rate the difficulty of the steps on a 1-10 scale? (1 = easiest and 10 = most difficult)  How many surgeons/ assistants/ residents generally assist you in this type of surgery?  What do you think about the reduction step? Have you experienced any difficulty/frustration during this step?  How many femoral reductions do you perform per year? How long does it usually take for an easy/ difficult/ average case?  What techniques do you use during this type of surgery to facilitate the process? What happens if you experience difficulties with these techniques?

Proposed IM Nailing Procedure: 1- Patient positioning 2- Entry point selection 3- Use of straight or ball tip 4- Bending the tip (as necessary) 5- Provisional reduction 6- Passing the guide wire 7- Reaming 8- Nail selection (choice of nail diameter and length) 9- Nail insertion 10- Proximal locking 11- Distal locking 12- Final assessment for length & rotation

Surgeons were selected with varying levels of experience and also including representation from community and academic teaching hospitals. Face-to-face interviews were conducted that lasted approximately 30-45 min in a private room. Participants were interviewed to identify the IM nailing surgical steps, surgical challenges, and their use of adapted surgical techniques when encountering challenges. Eight interviews were conducted until the saturation of the study [72] was reached and each interview was audio recorded and transcribed. Data collection and analysis were done in an iterative fashion. Data was transcribed concurrently with interviewing to allow for the interview guide to be refined. The study authors (HE, AY, CW) debriefed regularly to determine when saturation had been reached [73].

2.3.2 Surgical observations

Next, eight IM nailing surgeries of diaphyseal femoral fractures conducted by three interviewed staff surgeons were observed to gather additional information on the surgical procedure. In this, the term activities, steps and phases were used to represent different levels of granularity in the surgical process. An activity was defined by a physical task. A step was comprised of a sequence of activities which accomplish a specific surgical objective. A phase was used to describe a major event occurring during the surgical procedure [63] .

In order to better understand challenges and barriers associated with this surgery, observations were recorded to identify individual activities, steps and phases focusing on surgeons as the operator. In addition, the total time for completing each phase of the procedure and the amount of radiation/fluoroscopy used were recorded for the entire surgery. In those steps identified as challenging, the individual activities were analyzed to elucidate their underlying causes. Finally, the observed data was formalized using hierarchical decomposition [74].

2.4 Results

2.4.1 Semi-structured interviews

Semi-structured interviews were conducted with eight surgeons with varying levels of experience, three from community hospitals and five from academic teaching hospitals. On average, these surgeons perform 25 (range: 5-75) IM nailing surgeries per year as shown in Table 3.

Table 3 Reported number of IM nailing surgeries conducted per year (career average) by interviewed surgeons working at community and teaching hospitals

Centres Community Teaching Hospital

Surgeons A B C D E F G H

Number of 10 5 75 5 22 30 30 25 Surgeries/Year

The sequence of surgical steps associated with IM nailing was discussed. The difficulty of surgical steps was then rated on a scale of 1 to 10 (with 1 being the easiest and 10 the most

challenging). Seventy-five percent of surgeons identified fracture reduction as the most challenging step of IM nailing (Figure 12). The greatest challenges were manipulation of fracture fragments for realignment. Adapted manipulation techniques used alteration of patient positioning and/or traction, reducer devices, mallets, surgical wraps, bone hooks and Schanz traction pins.

Entry point selection was identified as the most difficult step by 25% of surgeons (Figure 12) due to challenges in surgical access and K-wire trajectory for bone canal starting point entry. Adapted techniques employed altering of patient positioning (torso abduction and/or knee adduction) or the elimination of the need for a straight access line to the IM canal through the use of a curved awl in place of the traditional K-wire. Lateral decubitus patient positioning instead of traditional patient supine positioning was suggested to facilitate entry point access in obese or muscular patients.

6 Most Challenging Step 5 Mean Rating: 9.4 (out of 10 ) 4 Second Most Challenging Step 3

Mean Rating: 7.8 (out of 10) Number of Surgeons ofNumber 2

0 Reduction Entry Point Selection Surgical Steps

Figure 12 Challenging steps identified by the interviewed surgeons along with the mean difficulty rating on a scale of 1 to 10

2.4.2 Surgical observations

In the second part of this study, eight IM nailing surgeries for diaphyseal femoral fractures were observed. A short description of observed surgeries and identified phases/steps are presented below.

Once in the operating room and under anesthesia, each patient is transferred to a fracture or a radiolucent surgical table. This marks the beginning of the IM nailing procedure as described in this study. Patient positioning, skin preparation, draping and provisional reduction steps constitute the first phase of the operation. At the end of this phase the X-ray imaging is conducted to verify proper C-arm positioning with respect to the area of interest. The second phase is focused on gaining access to the surgical site and included localization/marking the incision site, performing the initial incision and dissection of the soft tissue. The third phase is focused on guide wire insertion. In this, following palpation of the greater trochanter, a straight short sharp tipped rod (i.e. K-wire) or a curved awl is inserted into the center of the greater trochanter or the trochanteric fossa to introduce an opening into the IM canal. The identified entry point is confirmed on fluoroscopy. A longer 2.5 to 3 mm diameter rod (i.e. guide wire) is then advanced into the IM canal through the opening to the fracture level. The fourth step is intra-operative fracture reduction, a process that realigns the fracture fragments in order to allow passage of a guide wire into the distal fragment. Successful advancement/docking of the long guide wire within the distal fragment is confirmed via fluoroscopy. The nail length is then measured using a probe surrounding the guide wire. Nail placement constitutes the fourth phase. In order to adapt the bone to the nail and maximize the contact area between the bone and nail, reaming is performed [75] while maintaining the obtained reduction. The reaming step begins with a small diameter reamer and progresses incrementally to a size 1 to 2 mm bigger than the selected nail diameter. Definitive IM nail insertion comprises the final step of this phase. The fifth phase secures the IM nail within the bone through proximal and distal locking steps, so as to control length and rotation of the surgically stabilized bone. The sixth phase is focused on the clinical and radiographic assessment of both leg length and rotation. The surgery is completed in the seventh phase, ending with skin closure. The sequence of phases and their associated steps for all eight surgeries are shown in Figure 13.

, with the associated steps below stepsbelow associated the , with

Hierarchical decomposition of IM nailing. The seven phases of IM nailing grey boxes inlight are IM phases represented of seven nailing. of The IM decomposition Hierarchical

13 inboxes white Figure

Consistent with the interviews, the observations also identified the entry point selection and reduction steps in the guide wire insertion phase as the most challenging and frustrating steps of the procedure. This conclusion was qualitatively drawn by the observer from surgeons’ facial expressions, use of harsh language and brief discussion after the surgeries. The specific activities associated with these two steps are presented in Figure 14.

Figure 14 Activities associated with the a) Entry point selection and b) Reduction steps in the guide wire insertion phase 2.4.2.1 Entry point selection

As shown in Figure 14 a, a series of activities must be performed iteratively for successful drilling of a K-wire inside the IM canal. In order to better understand these activities, a brief description of this step follows. When the K-wire is initially placed near the desired entry point (either greater trochanter or trochanteric fossa insertion site), an anterior-posterior (AP) image is taken. This is used to confirm the K-wire entry point in the coronal plane. If the entry point is suboptimal, the K-wire is adjusted and another AP image is taken. Once the K-wire entry point is confirmed on AP fluoroscopy, the C-arm is rotated to capture a lateral (or oblique-lateral) image to confirm the K-wire entry point in the sagittal plane. Lateral imaging and adjustments continue until the K-wire entry point in the sagittal plane is satisfactory. Once the optimal K-wire trajectory (entry point and orientation) is reached and confirmed with biplanar fluoroscopic images, the K-wire drilling is initiated to an initial provisional depth of ~1-2mm. Upon reconfirmation of entry point on an AP image, the K-wire can be drilled to its full ~2-3cm depth,

and reconfirmed on lateral images. Following insertion of the K-wire, a larger ~1cm cannulated drill bit widens the entry into the IM canal. This then enables the insertion of the long guide wire used in subsequent surgical steps.

2.4.2.2 Reduction

As shown in Figure 14 b, the reduction step contains multiple activities. Once the long guide wire is advanced to the fracture level, AP images are taken to verify the alignment of two/multiple fragments in the coronal plane. If the fracture is mal-aligned, several techniques can be utilized to facilitate control of the displaced fractured pieces in the sagittal plane in order to perform the reduction. For indirect bone manipulation, a sterile drape or towel wrap can be placed around the femur and used to pull the proximal fragment down to line up with the distal fragment. Alternative approaches to re-align the fractured fragments employ manual elevation of the distal fragment by use of a crutch or lowering of the proximal femur through external pressure from a mallet. Preliminary reduction may also be followed by either the direct manipulation of the fractured femur via Schanz traction pins (Figure 15 a) inserted into the fractured fragments or by utilizing a bone hook device (Figure 15 b). These techniques impart more direct control of the fracture fragments and enable more secure stabilization of an anatomic alignment. The use of such hardware and manipulation of fracture fragments, however, must be performed carefully to minimize soft tissue trauma and prevent damage to neurovascular structures.

The steps utilized in the eight surgeries were overall found to be quite similar with minor differences in the techniques utilized in activities during the guide wire insertion phase. The times for the individual phases are listed in Table 4 (surgeons often did not perform the closure, as such the time spent in this phase is not reported). The large range in times found for some phases can be better understood by highlighting challenges encountered in specific cases. The average time for the patient preparation phase was 24.88 min (SD = 14.26). In case 1, setting up the fracture table and the fluoroscopy C-arm and in case 6 cleaning open wounds away from the closed fracture site took a relatively longer time. Accessing the surgical site phase took 4.75 min (SD = 4.30) on average. In both cases 3 and 4, readjustments of the patients’ femur and torso were conducted that resulted in longer operative time. The average time for the long guide wire insertion phase was 26.88 min (SD = 8.69). For this phase, times for the entry point selection, reduction steps and overall fluoroscopy use are presented in Table 5.

Table 4 Time in minutes for individual phases associated with the IM nailing procedure

Phases Patient Accessing the Guide Wire Nail Locking Final Case Preparation Surgical Site Insertion Placement Assessment 1 40 3 26 16 51 1 2 32 2 42 60 38 3 3 14 13 25 15 14 4 4 21 10 22 16 37 1 5 16 4 33 16 32 1 6 50 2 18 17 29 1 7 10 2 16 66 28 1 8 16 2 33 22 54 1 Mean 24.88 4.75 26.88 28.50 35.38 1.63 SD 14.26 4.30 8.69 21.46 12.89 1.19

Table 5 Time spent during the entry point selection and reduction steps and overall fluoroscopy time for the entire surgical procedure in minutes

Steps Fluoroscopy Seniority Case Entry Point Reduction Challenges Time Level2 Selection 1 - 5 3 Senior Time to set up fracture table C-arm set up 2 1.7 18 5-131 Junior Entry point selection Reduction 1Accidental guide wire removal Muscular patient 3 5.5 4 13 Junior Reduction 4 1.8 10 10 Junior Entry point selection Reduction 5 2.6 3 17 Senior Reduction 6 3.5 9 2 Junior Entry point selection Open wound 7 1.6 2 5-461 Senior Reduction 1Fracture opened 8 3.9 9 2-71 Senior Time to set up fracture table 1Accidental guide wire removal

1The reduction step in this phase was conducted twice with the reason highlighted in the challenges.

2Seniority level was defined based on surgeons’ number of years in practice. A surgeon with less than 10 years in practice was categorized as “Junior”. A surgeon with more than 10 years in practice was categorized as “Senior”.

In case 2, the patient was muscular, which made the entry point selection challenging. The average time for the nail placement phase was 28.50 min (SD = 21.46). In cases 2 and 8, the guide wire was accidentally removed in the reaming step (which occasionally can occur unintentionally), which prolonged the nail placement phase. In case 7, due to unsatisfactory position of the nail, the fracture site was opened and additional reduction was required, which resulted in extended operative time for this step. The locking phase took the longest average time; use of the fracture table (in cases 1 and 8) increased the difficulty of fine leg adjustments.

2.5 Discussion

Initial interviews highlighted two specific steps, entry point selection and reduction, as the most challenging and frustrating to the surgical team. During the interviews each surgeon identified adapted techniques which they employ to tackle the challenges associated with these two steps. The described techniques were similar with slight differences based on surgeon preferences and the order in which they utilize the adapted procedures. During the observations, these identified challenges were encountered and adapted techniques utilized to overcome them. While the adapted techniques employed ultimately overcame the entry point and reduction challenges, they were associated with high levels of frustration. Opportunities to improve the surgical procedure lie in understanding the challenges associated with these steps as existing techniques do not offer an optimal solution. As such, activities associated with entry point selection and reduction steps were studied in depth.

2.5.1 Entry point selection analysis

While the series of entry point selection activities appear relatively straightforward, they sometimes proved frustrating and time consuming (i.e. case 2 which required 18 minutes for this step). When modifications to the K-wire entry point are performed to achieve alignment in the coronal plane, correct positioning is often found to be lost in the sagittal plane. Therefore, multiple cycles of AP and lateral imaging may be required to confirm the optimal entry point positioning. In order to better understand the challenge associated with the entry point selection step, it is important to differentiate between the entry point location and orientation, both of which are critical for the overall outcome of the procedure. Two independent studies by Roberts et al. and Kanawati et al. have shown that an entry point in the posterior one-third of the greater trochanter may result in cortical impingement which can lead to major complications [76,77].

Similarly, if the orientation of the entry point is not in line with femoral shaft, subsequent reaming may lead to weakening of the medial cortex and the potential for a blow-out fracture complication [78].

Once adequate images with respect to entry point location and K-wire orientation are acquired in the AP direction (Figure 16 a), errors in sagittal placement must be addressed. Based on the analysis of the observed activities, if the sagittal entry point location is correct but the lateral orientation is incorrect (Figure 16 b), the surgeon must alter the anterior-posterior angle of entry of the K-wire about the identified entry location (ensuring no displacement of the K-wire tip from the entry site). If the sagittal entry point location is incorrect but the orientation is correct (Figure 16 c), the surgeon must adjust the anterior-posterior translation along the sagittal plane without any change in sagittal angulation of the K-wire. If the sagittal entry point location and orientation are both incorrect (Figure 16 d) then the surgeon must first readjust the entry location in the sagittal plane. Once the correct new entry location is verified with lateral fluoroscopy, the surgeon must then readjust the K-wire orientation in the sagittal plane to ensure parallelism in access to the IM canal (Figure 16 e); following this step, the surgeon needs to recheck the AP view to ensure that both the coronal plane entry point and wire orientation are acceptable.

Many experienced surgeons are able to maintain the movement of their hand in one plane while they are adjusting the location and/or orientation of the K-wire in the corresponding perpendicular plane. However, this can be challenging, particularly for new or inexperienced surgeons when considering that even small hand or body movements can disrupt positioning/orientation and some motion may be required to maintain the sterile field during C- arm rotation. As well limited access to the surgical site due to the minimally invasive nature of this surgery and the presence of muscle/fat forces acting against the K-wire can present difficulties in maintaining in plane position.

Currently, commercially available devices do exist which are designed to facilitate entry point selection. These devices generally consist of a cannulated tissue protector (Figure 17 a) and a

trocar with multiple parallel holes (Figure 17 b). This allows surgeons to insert multiple K-wires simultaneously when the initial K-wire placement is suboptimal. Such devices provide a finite number of alternative locations for the entry point, but do not allow for modification of the entry point orientation (orientation is fixed as per the initial K-wire inserted). As such, these devices do not address challenges with the entry point orientation vis-à-vis desired K-wire trajectory into the femoral IM canal. Another alternative utilized to open the IM canal is the manual advancement of a curved cannulated sharp tipped awl (Figure 17 c). Similar to the entry point selection needed for K-wire drilling, positioning and orientation of the awl must be confirmed on fluoroscopic AP and lateral images, with similar associated challenges.

Similar to entry point selection, challenges in the reduction step are encountered as the alignment in the coronal plane is difficult to maintain while aligning the fragments in the sagittal plane. However, unlike the entry point selection step, there is no direct access to the fracture fragments. Manipulation of the fragments using force applied on the leg through a mallet or crutch may not necessarily yield translation of the targeted fragment in the intended direction. Due to

connectivity between the fragments (i.e. through soft tissue attachments), manipulation of one fragment may result in the unintended movement of other fragments.

One technique utilized to facilitate the reduction step is to slightly bend the guide wire tip to facilitate its entry into the distal fragment if perfect alignment has not been achieved. Rotation of the guide wire then makes picking up the displaced distal fragments easier. The same technique can be applied using a straight cannulated rod with a bent tip (reducer). The guide wire is inserted through the reducer, and the rotation of the reducer facilitates the reduction. The advantage of the reducer over the bending the tip technique is its flexural rigidity. The reducer gives a joy-stick-like control over the proximal section which decreases unintentional movement when manipulating the distal fragments. However, these techniques may still require multiple cycles of AP and lateral imaging (as described in Figure 14 b).

Navigation systems have also been used to facilitate IM nailing. The use of fluoroscopy-based navigation has been reported to reduce radiation exposure compared to conventional procedures, and improve surgical hand-eye coordination by reducing the need for mental correlation of the acquired fluoroscopic images with the procedure [60,79]. However, the complexity and time consuming nature of utilizing navigation (including the attachment of optical trackers and registration), high financial costs, logistical issues and training demands have been major drawbacks limiting the clinical uptake of this technology. While surgical steps and activities associated with these navigation systems have been studied, the surgical processes have not been analyzed to elucidate the root sources of challenges or frustration [60,80,81]. Rather, entire steps or activities in IM nailing have been replaced with the use of a navigation system, which in turn involves its own adaptation challenges.

2.5.3 Common challenges and limitations

In both entry point selection and reduction steps, multiple 3D geometries (bone fragments, K- wire and guide wire) need to be aligned using 2D AP and lateral images. The reconstruction process requires the location and orientation of geometries to be aligned in both sagittal and coronal planes simultaneously. This can be challenging when considering the difficulty associated with the maintenance of location and/or orientation of geometries in the one plane while adjusting the position in the perpendicular plane. Reported and observed medical staff frustration during the entry point selection and reduction steps may also be due to the variability of the guide wire insertion phase, in terms of difficulties experienced and the time required to achieve alignment.

The observational findings in this study are specific to a teaching hospital where residents and fellows are generally present to assist in the operating room. Although in the interviews surgeons from community hospital identified same steps to be challenging, other factors may cause frustration in community operating room with fewer assistants. A further limitation may have been potential inaccuracies associated with hand-written recordings, rather than digitally acquired data, during the observations. Finally, this paper focused on surgeons as the operator; the role of imaging technicians, scrub nurses, residents/fellows and anesthesiologists and their challenges with specific steps or activities were not considered.

2.6 Conclusion

Significant challenges may arise during IM nailing that impede the surgical workflow and elevate surgical frustration levels. Entry point selection and reduction were identified as the most challenging steps. Activities associated with both steps attempt to achieve 3D alignment using 2D fluoroscopic images. Maintenance of the correct alignment (location and/or orientation) in one plane while adjusting the position in the perpendicular plane is critical in these steps. If the alignment is lost, the cycle must be repeated, causing an increase in time and frustration. Maintenance of location and/or in plane orientation in entry point selection may be facilitated with the utilization of surgical instrumentation/technique which restrict out of plane movements.

The ability to consistently maintain the alignment obtained in one plane while adjusting positioning in the perpendicular plane may reduce the need for repetition in the entry point

activity cycle. Similarly, the reduction step can be facilitated by utilizing a device/technique that maintains the alignment in one plane while adjusting positioning in the perpendicular plane or more complex solutions which incorporate real time navigation. These findings provide new insights on the causation of surgical challenges in IM nailing and may be used as a platform for future development of improved techniques and surgical instrumentation.

Chapter 3: Femoral Access Starting Tool – FAST 3.1 Abstract

Purpose: The surgical protocol associated with closed intramedullary (IM) nailing is well defined. Yet, challenges commonly arise during the entry point selection step in which 2D fluoroscopic imaging is used to guide 3D alignment of the guidewire with respect to the IM canal. Specifically, the loss of K-wire alignment achieved under fluoroscopy in one plane may be lost while adjusting the alignment in the perpendicular plane. This can impede the surgical workflow through unpredictable repetition of surgical activities and lead to frustration in the operating room. The primary aim of this study is to develop an innovative surgical instrument to facilitate entry point selection for femoral IM nailing which enables lateral reorientation and repositioning while maintaining anteroposterior alignment.

Methods: A user-centered design approach was utilized in the development of the surgical tool. Multiple design iterations were conceived to ensure device ease-of-use and functionality leading to an initial prototype device. Formative usability testing was conducted with four participants using the initial prototype on synthetic femora. Based on identified design issues, modifications were made and a redesigned prototype was prepared for evaluation on cadaveric femora. Three surgeons identified the entry point for femoral IM nailing in paired cadaveric specimens, using the device on one side and a free-hand approach on the contralateral side under operating room conditions. The number of acquired images, drilling attempts and the operative time were recorded for each procedure and compared using paired t-tests. Surgeons were interviewed after the procedure to qualitatively assess the new device and identify areas for further improvement.

Results: The user-centred design led to the development of the Femoral Access Starting Tool (FAST) consisting of a fixed curved frame and a rotatable multicannulated arm to assist in K- wire insertion during IM nailing. In use, the device is initially placed at the approximate IM nailing entry point location on the femoral greater trochanter under fluoroscopic image guidance. FAST enables maintenance of the AP alignment during adjustment of the K-wire position and orientation under lateral imaging. All surgeons successfully used the device to insert the K-wires into their desired entry points during the synthetic bone testing. The cadaveric testing (n=6 per group) showed a trend toward improvement of the K-wire orientation alignment when FAST was

utilized. It also highlighted the need for a comprehensive device-use protocol and training program, and suggested potential improvements to the device design.

Conclusions: Appropriate use of FAST (with solid fixation) allows adjustment of K-wire positioning during lateral imaging while maintaining AP alignment. This may reduce the number of required AP and lateral fluoroscopic images, drilling attempts, and operative time and improve orientation within the IM canal. Ultimately, FAST has the potential to improve predictability in the IM nailing workflow, reducing radiation exposure, medical staff frustration and the incidence of malunion.

3.2 Introduction Closed Intramedullary (IM) nailing is the standard of care for adult femoral shaft fractures with union rates approaching 97% [28]. This minimally invasive surgery consists of seven surgical phases: patient preparation, accessing the surgical site, guide wire insertion, nail placement, locking, final assessment and closure [82]. For a successful IM nailing surgery, the choice of correct entry point in the guide wire insertion phase plays a significant role. The trochanteric fossa and greater trochanter are two common starting points for antegrade femoral nailing. Numerous studies in the literature investigated the optimal anatomical location of the entry point based on: comparisons of trochanteric fossa versus greater trochanter clinical outcomes [83], nails with different bend radii [77,84,85], fracture component length and reamed diameter [85], and curvature of the femur [86]. An incorrect entry point in the anterior-posterior or mediolateral direction has been reported to cause complications such as varus malalignment [84,87], anterior penetration of the distal femur [77], iatrogenic fracture comminution [83,84,88,89] and bursting of proximal femoral component [85].

The surgical activities for entry point selection are well defined. The surgeon may adjust the positioning of the patient to obtain a better access to the surgical site. The K-wire is then placed on the desired entry point and an Anterior-Posterior (AP) image is taken to confirm the K-wire alignment with respect to the proximal medullary canal. Consequently, entry point adjustments and multiple AP image acquisitions may be necessary until satisfactory K-wire positioning is obtained. Finally, the location and orientation of the K-wire is evaluated by taking lateral images. These surgical activities can be challenging in obese/muscular patients [82], cases with the presence of complex fracture patterns [90] or when the surgeon is inexperienced [82]. The root sources of these challenges were described in a recent study of the IM nailing surgical process as the lack of connectivity between sequential fluoroscopic 2D alignment [82]. This lack of connectivity occurs during the C-arm rotation for obtaining lateral or AP images. Once the K- wire AP alignment is obtained, C-arm is rotated to take a lateral image. The AP alignment can be lost if the surgeon hand movement is not maintained in the coronal plane while modifying the K- wire location and/or orientation in the sagittal plane [82]. The alignment may also be lost due to muscle/fat forces acting against the K-wire or simple involuntary hand movement to maintain the sterile field of view during the C-arm rotation. If the alignment is lost, the surgical activities must be repeated which results in an increase in operative time, radiation exposure and often medical

team frustration. This clinical problem motivated the development of a novel and simple entry point device that addresses the following clinical need: a way to maintain in plane AP positioning during the acquisition of lateral images and subsequent guide wire reorientation required in the IM workflow.

This study presents a user-centred design approach for development of the Femoral Antegrade Staring Tool (FAST). FAST is designed to 1) facilitate the surgical access to the desired entry point and 2) maintain the AP alignment of the K-wire while allowing K-wire adjustments in the sagittal plane.

The following sections outline the user-centred design approach in understanding the device requirements, device design and prototyping. A series of prototypes were developed and formative usability testing was conducted. The usability testing results were utilized to develop the final prototype.

3.3 Methods

3.3.1 User-centred design The design, manufacturing and testing of the device were performed at a level 1 trauma teaching hospital. An initial design was modeled and manufactured based on information gathered from IM nailing surgical observations (8) and semi-structured interviews with orthopaedic surgeons performing IM nailing [82]. A team, consisting of an experienced staff surgeon, a junior orthopaedic surgeon, and a mechanical engineer was then consulted. The team comments were addressed and the new design was manufactured. This iterative process was repeated multiple times until the team was satisfied with the device ease-of-use and functionality. The prototype of the new design was tested on a synthetic bone models surrounded with foam to simulate soft tissue. Four surgeons conducted the surgery under standard operating room conditions and provided feedback. The above process allowed the inventors to finalize the device design and establish the device use protocol.

3.3.2 Device requirements The recommended incision size for this minimally invasive surgery varies between 3 - 7 cm based on the type of the nail. The incision is longitudinal and centered at the intersection of the line drawn laterally on the skin along femoral shaft and the radiological shadow of the K-wire

placed over the axis of femoral shaft [29]. Hence, the part of the device that goes inside the body must fit inside this opening to eliminate the need for a bigger incision. In addition, this part must be radiolucent to accommodate K-wire position verification on AP and lateral X-ray images.

The K-wire placement for subsequent nail insertion requires the skin incision, the entry point of the nail and the medullary canal to be in a straight line [29]. Access to the surgical site can be challenging in obese or extremely muscular patients as the patient repositioning or leg adduction may not be sufficient to obtain an incision below the iliac crest. Therefore, the device must facilitate access to the surgical site, specifically in cases with obese or extremely muscular patients.

The number of assistants in the operating room depends on the type of hospital/clinic. In teaching hospitals often residents, fellows and assistants are available during the surgery whereas in community hospitals the surgeon may only be accompanied by a single assistant. The insertion of the K-wire is performed by the surgeon with a single hand while an assistant may manipulate the leg based on the surgeon’s preference. Therefore, the device must preferably be used with a single hand so that an extra assistance is not required for the device use. In addition, since the surgeon’s hands come in contact with the device directly, the device must be sterilizable.

The device must maintain the AP positioning of the K-wire while allowing the adjustments in the location and/or orientation of the K-wire in the sagittal plane. In addition, the presence of muscle/fat forces acting against the device when the C-arm is rotated must be considered in the device design.

The tolerance of the optimal entry point location is reported to be 2 to 3 mm [87]. Precise sagittal plane localization is recommended in multiple studies [5,7,91]. Hence, the device must be able to offer a precise localization in the sagittal plane.

The IM nailing surgical process must be considered to avoid impeding the workflow. The device insertion step is after the surgical palpation of the greater trochanter. The device also must accommodate the subsequent insertion of a reamer on top of the inserted K-wire.

Finally, the surgeon can be left or right handed and the fracture can be present in the left or right femur. Hence, the device ease of use, functionalities and operating instructions should accommodate these possibilities.

3.4 Existing technology While the majority of IM nailing entry point selection is done using a free-hand technique and intraoperative fluoroscopy, Computer-Assisted Surgery (CAS) systems for IM nailing have been developed [92] to address limitations of this approach. The limited field of view and two- dimensional nature of fluoroscopic images served as a motivation for development of CAS systems [36,92]. A stated advantage of these systems is the decrease in complications associated with the alignment of bone fragments and surgical tools [36]. Although CAS systems have been utilized to help with fracture reduction [79,92,93] and prevent malrotation [92,94,95], few studies investigated advantages of CAS systems in entry point selection [61,81]. The study conducted by Crookshank et al. [61] concluded that computer navigation-based entry points are either similar or less accurate compared to fluoroscopy alone in a “best-case scenario”. This counterintuitive result was explained to stem from overtrust and bias for navigation-based calculations. It may also be that navigation systems offer an optimal tool and anatomy configuration, however, they do not consistently maintain the desired configuration during the respected surgical steps [36]. A robotic system was developed by Westphal et al. [81] for long bone fracture reduction which could also facilitate entry point selection. In the study, two groups, with five femurs were tested by the robotic system and the free-hand technique. Although, no precision or accuracy study was conducted, the robotic system showed an slightly better results with respect to number of drill realignments and mean number of X-ray images. The common drawback of CAS have been reported as, prolonged setup time (as high as 62 min[94]), increased operation time, increase risk of fracture, bleeding and nerve damage as a result of percutaneous placement of pins and reference bases, and financial cost (investment, maintenance, additional staff, prolonged operation time, training) [94].

There exist current clinically available devices intended to facilitate the free-hand entry point selection. The trocar with a honeycomb design offers multiple parallel holes, enabling advancement of an initial entry point through parallel holes based on an initial reference K-wire. However, the orientation of the device is fixed to the first inserted K-wire, preventing any modification of the K-wire orientation in the coronal plane.

3.5 Device design features The proposed design consists of a fixed curved frame (Figure 18 a,b - 1), a rotatable arm (Figure 18 a,b - 2), a locking mechanism (Figure 18 a,b - 3) and a cover (Figure 18 a,b - 4). The device design is uniplanar allowing surgical operation on the left or right leg without any device configuration modification. The following sections elaborate the main design features of the device.

a b

2 2 1 1 2 1

4 4

4 3 3

Figure 18 a) The final prototype and b) the SolidWorks model of FAST showing: the fixed curved frame (1), the rotatable arm (2), (3) the locking mechanism, and the cover (4) 3.5.1 Fixed curved frame. The fixed curved frame is made of stainless steel, cut to the desired shape using the water jet technology. The curved shape facilitates insertion of the device on the greater trochanter, especially helpful in obese or extremely muscular patients as the distal end of the device curves away from the patient body. This provides the surgeon with an ergonomic adjustment of the K- wire positioning compared to the collinear access configuration required in the free-hand technique. The bend angle and the location where the bend radius starts were obtained by prototyping the rotatable arm in a range of angles and curvature starting points. The increase in bend angle increases the required impact necessary to hammer the device onto the greater trochanter area. Based on the preference of the surgeons during the user-centred design process a 20 degree angled rotatable arm was chosen.

The frame in the proximal end houses two insertion pins, a radiolucent tip and a radiolucent connecting plate. The radiolucent tip and the connecting plate allow K-wire visualization during

lateral imaging and is made of Polyether Ether Ketone (PEEK). Alternatively, the tip and two insertion pins can be incorporated into one part made out of stainlessness steel with an empty pocket design to allow visualization. The tip also houses a compact design to connect to the proximal end of the rotatable arm as shown in Figure 19. This configuration allows only one degree of freedom for rotation along the pin (Figure 19 – 2). The sole rotational degree of freedom ensures maintenance of the K-wire AP alignment. The axis of rotation is along the pin that is located adjacent to the proximal end of the tip. The close distance between the axis of the rotation and the device proximal end is designed to maintain the K-wire insertion location adjacent to the device tip. In the distal end of the device the locking mechanism for the rotatable arm adjustment is mounted. The distal end also provides a threaded hole for handle attachment.

4 1

3 4 5

3.5.2 Rotatable Arm

The rotatable multicannulated arm is connected proximally to the fixed curved frame and distally to the rotating shaft (Figure 20). The pin joint connections on both ends of the the rotatable arm offer a sturdy configuration.

2 4 1 3

The curved shape of the device, similar to the fixed curved frame, facilitates insertion and maintenance of the device at the desired entry location agaist the patient’s muscle/fat forces. The rotatble arm was rapid prototyped with Sterolitogrpahy (SLA) technology and is made of ClearVueTM plastic which is a USP class VI certified material. This material is sterilizable and translucent.

The petal design in Figure 21 shows K-wire guide channels extending from the distal to proximal end of the device. Each channel in this design is located precisely below or above other channels, to ensure the maintance of the AP alignment once one of the three columns (medial, middle, lateral) is selected on the AP view. The channel diameters can be modified to accommodate different diameter K-wires. The current prototype channels are 2.1 mm in diameter with a 2.9 mm spacing from each other. This configuration satisfies the 3 mm threshhold of entry point

accurancy [87]. In addition, five alternatives are available for the surgeon to obtain the desired entry point in the middle column of the rotating arm to improve the options for sagittal placement accuracy [5,7,91].

Figure 21 The zoomed view of the petal design. Each entry point entrance is chamfered to facilitate K-wire insertion. The entry points are stacked on top of one another in 3 columns and 5 rows. Each entry point is 2.1 mm in diameter and is 2.9 mm away from neighboring points. One column is selected during AP imaging and based on the lateral view one entry point from the selected column is chosen 3.5.3 Locking mechanism The locking mechanism was designed with one-handed operation in mind. This unique design consists of two gears, a trigger, a set screw, a shoulder screw and a spring as shown in Figure 22. The set screw couples the connecting shaft rotation to the outer gear. The gap on the outer gear enables axial movement of the gear with respect to the connecting shaft.

3 5 4 6 2 5 3 6 7 4 1 2 2 1

In the default configuration the shoulder screw is tightened inside the connecting shaft while pressing the outer gear to engage with the fixed gear via a spring. In this configuration, the rotatable arm is locked into position. Once the trigger is engaged with the index finger of the hand holding the device by the handle, the outer gear is disengaged and the gear moves axially. This allows the rotation of the rotatable arm, subsequent rotation of the connection shaft and ultimately rotation of the outer gear. The rotatable arm can be intuitively adjusted with the hand holding the drill and the connected K-wire similar to the free-hand approach. Upon the release of the trigger, the device returns to its default locked position and the rotatable arm is locked securely at the desired position. This one-handed intuitive design offers a robust locking mechanism and ensures maintenance of the rotatable arm orientation in the sagittal plane.

3.5.4 Cover The cover is made of stainless steel, slides on top of the rotatble arm and engages with the fixed curved frame at its distal end. The main functionality of the cover is to transfer the hammering impact force through the main frame to the device pins allowing the advancement of the pins into the bone. The secondary function of the cover is to protect the rotatble arm from accidental impact.

3.6 Device use protocol

Following GT palpation, the initial device entry point placement is conducted freehand at the greater trochanter under fluoroscopic image guidance (Figure 23 - a). While the device concept allows the device utilization for all common entry point locations, the current device is designed for the trochanteric fossa (piriformis fossa) entry point. Using a traditional approach, this entry point location is more challenging to access and requires K-wire alignment with the IM canal. FAST uses a greater trochanteric attachment site but directs the k-wire entry point to occur at the trochanteric fossa (piriformis fossa). A GT entry point would require modifications to the pin lengths and orientations.

While the rotatable arm allows for multiple lateral entry points, the device should not be placed too anteriorly or posteriorly. This can be checked by making sure both pins are in contact with the greater trochanter area by inserting the index finger or taking a lateral image.

An AP image is taken to align the rotatable arm with the intramedullary canal of the femur (Figure 23 - b). The device location and orientation are adjusted until satisfactory AP alignment is obtained. Once the optimal alignment is observed on the AP image, the device is hammered and temporarily fixed on the greater trochanter. Another AP image is taken to confirm placement of the device on the greater trochanter. The cover can then be removed allowing the K-wire insertion through the rotatable arm. a b

Figure 23 a) Inserting FAST on the greater trochanter area and b) aligning the orientation with the IM canal under image guidance

Based on the AP image, one of the three columns (medial, middle, lateral) of the rotatable arm that aligns best with the medullary canal is selected. Subsequently, a K-wire is placed into one of the entry points in the selected column (Figure 24).

Figure 24 The device is securely hammered into the bone and a cannula is selected for optimal AP entry point location

A lateral (or oblique-lateral) image is taken to identify the correct 3D trajectory for accessing the intramedullary canal in the sagittal plane (Figure 25). While the device is held in one hand and the drill with the K-wire attached in the other hand, the trigger can be activated to allow the rotatable arm in-plane rotation. The rotatable arm can be easily adjusted intuitively by the hand holding the drill under fluoroscopic guidance. If necessary, other entry points within the selected column can be used. Upon obtaining satisfactory K-wire location and orientation, the trigger is released to lock the position of the rotatable arm. The K-wire is subsequently drilled into the intramedullary canal through the rotatable arm. Once the K-wire is inserted 3-4 cm, the K-wire is released from the drill and the device is removed, leaving the K-wire in the correct position and orientation.

Figure 25 Adjusting the rotatable arm to obtain optimal lateral entry point orientation, cannula adjustment for position and K-wire advancement 3.7 Testing

3.7.1 Formative usability testing

According to the FDA, “A hazard is a potential source of harm. Hazards arise in the use of medical devices due to the inherent risk of medical treatment, from device failure (or malfunction), and from device use” [96]. Usability testing helps to identify hazards that may have escaped detection during simulation and initial prototyping [97]. In addition, usability testing can lead to an optimized device that is easy to learn, use and manufacture. There are various types of usability testing based on the different stages of the product development cycle. Formative usability testing involves the evaluation of an evolving design, with the focus on design verification and identifying opportunities for improvements.

Formative usability testing was conducted with four participants with varying levels of experience. The aim of this testing was to answer the following questions: 1) Can surgeons successfully insert the K-wire in their desired entry points using the tool 2) What are the possible unforeseeable issues associated with the device use and what are the associated mitigation plans? 3) What are the strengths and weaknesses of the device design? An experienced surgeon, two junior staff surgeons and an orthopaedic resident participated in the study. Synthetic bones were used and each femur was surrounded by foam to simulate soft tissue (Figure 26) of an obese patient [98]. The testing was conducted in an operating room with the distal section of the femur attached to the surgical table via multiple C-clamps. A C-arm was utilized for intra-operative imaging. Each surgeon was given instruction on how to use the device and asked to think-aloud while performing the surgery. Each surgeon performed the experiment once and was interviewed after the surgery. Note, the device at the time of the testing did not offer a trigger for the locking mechanism (as shown in Figure 27).

Figure 26 Synthetic bone set-up with surrounding foam to simulate soft tissue

2 1 3 3

All surgeons successfully inserted the K-wire in their desired entry point. Five design issues were identified and mitigation plans applied to the design (Table 6).

Table 6 List of identified design issues and corresponding mitigation plans

Design Issues Mitigation Handle loosening Incorporated a spring washer to fix the handle and prevent loosening Length of pins requiring a time consuming Shortened pin lengths by 1/3 - sufficient hammering process purchase in bone still obtained Locking mechanism difficult to use (lateral Redesigned the locking mechanism by movement of the outer gear to disengage from incorporating a trigger mechanism the inner gear with two fingers was not ergonomic + the outer gear was slippery) Device Weight Lowered the device weight by incorporating holes Weakness in the plate connecting the device tip Machined the tip and the connecting plate out to the proximal end of the rotatable arm of PEEK (higher stiffness and ductility) (Figure 28)

Figure 28 The device tip showing the broken connecting plate which connects the tip and the rotating arm

All four surgeons showed interest in using the device clinically after necessary approval is obtained. The most important strength of the device was the maintenance of the AP alignment during the lateral imaging.

3.7.2 Cadaveric Testing

3.7.2.1 Redesigned Device Prototype Upon addressing the design issues identified during the pilot testing, a second prototype was manufactured as shown in Figure 29. The new locking mechanism was designed with a trigger such that a direct manipulation of the gear1 to disengage from the fixed gear2 was not necessary. Once the trigger was engaged, the gear1 was pushed medially allowing the rotation by two fingers. The default positon of the trigger allowed the gear1 to rest on the fixed gear2 resulting in a locked position secured by a spring pushing the gear1 laterally as shown by the dotted arrow in Figure 29. A new tip and a connecting plate were also made out of PEEK by CNC machining. As a backup the tip along with the pins were also cut using water jet technology out of stainless steel.

5 4

1 2

b c

Figure 29 (a) The new design featured a new locking mechanism (1: gear1, 2: gear2 and 3: trigger), 4: tip and 5: connecting plate. (b) The backup device with identical configuration as (a) except the tip and the pins which are made of stainless steel. The empty pocket design allows visualization of the K-wire under fluoroscopy as shown in (c) 3.7.2.2 Experimental Setup Follow up cadaveric testing was conducted by three surgeons who participated in the synthetic bone study with varied levels of experience. The testing was performed on cadaveric femora under standard OR conditions (Figure 30). The fresh frozen cadavers were left at room temperature to thaw for 48 hours prior to the testing. A dummy torso was made to block direct access to the surgical site simulating a more realistic operative environment when the patient is obese or extremely muscular (Figure 30 - a).

a b

Five paired femurs were tested (10 total) with each surgeon performing on two pairs (one pair was tested twice). Based on a randomized block design, both the side (left or right leg) and order in which the device or freehand technique was utilized were randomized. Each leg was adducted and strapped to a custom made extension radiolucent frame. Upon surgeons’ request assistants were available to adjust the leg position and hold the leg while hammering the device or inserting the K-wire. The reused pair was randomly selected; the specimen previously utilized for the free- hand approach was prepared to evaluate the device and the contralateral side was used in the testing of the free-hand approach. The surgeon was instructed to either use the pre-existing incision or to introduce a new incision site as per the surgeon’s preference.

The number of acquired fluoroscopic images and the required time to complete each surgical step were documented. The operative time recording and fluoroscopic images count were started when the surgical dissection was completed and the surgeon requested the first AP image acquisition with a K-wire (free-hand approach) or the device inserted. The time recording was stopped when the surgeon was satisfied with the K-wire placement in both AP and lateral views. The number of K-wire drilling attempts was also recorded to estimate the risk of femoral neck weakening [99]. Each attempt was defined as insertion of a K-wire at a specific location and

orientation followed by AP and/or lateral imaging to confirm the K-wire positioning. For example two attempts mean a K-wire was inserted, upon checking AP or lateral views the surgeon was not satisfied with either the K-wire orientation or location, the K-wire was pulled out and a new optimal location and/or orientation was selected and achieved.

Prior to the testing commencement, surgeons were given instruction about the device use and were given time to familiarize themselves with the device. A short instruction on how to use the device was prepared and read to each surgeon to review the surgical steps associated with the device use as shown in Table 7. The surgeons were asked to think-aloud and all procedures were video recorded. Upon completion of each testing, a brief semi-structured interview was conducted about the device ease-of-use, device functionality and the differences they noticed between the two approaches.

The orientation of the K-wires in AP and lateral images with respect to the IM canal was measured and compared using the paired t-test. Each image was imported to SolidWorks and two straight and parallel lines were drawn to mark the IM canal borders. Next, the center of these two lines was selected and a third straight line representing the IM canal axis was drawn. Finally, by marking the K-wire’s edges another axis located at the center between the two was drawn to represent the orientation of the K-wire. The marking of the K-wire’s edges and its axis were drawn at an approximate K-wire insertion site. However, in some cases, due to the presence of tissue/fat, the proximal section of the K-wire outside the femur was bent all the way to the site of insertion. In such cases, the edges of the K-wire were marked more distally inside the femur. Finally, the angle between the two axes was measured (Figure 31).

Figure 31 This image illustrates the angle between the IM canal and the K-wire in the AP view

Table 7 List of surgical steps read to each surgeon prior to the testing

 Place the device on the femoral GT  Alter the device position and orientation until the rotatable arm is aligned with the intramedullary canal on the AP image  Check to see if the device is not placed too anteriorly or posteriorly  Hammer the device until the device is fully secured into the bone  Remove the device cover  Place a K-wire into one of the three columns of the rotatable arm channels until the K- wire is aligned with the intramedullary canal  Take a lateral image, rotate the knob if necessary to change the orientation of the rotatable arm until satisfactory trajectory is achieved  If needed, you can change the position of the K-wire by selecting a different channel within the same column  Advance the K-wire  Obtain a confirmation AP image  Release the K-wire from the drill  Remove the device

3.7.2.3 Results The quantitative results for the cadaveric testing are shown in Table 8 in the order in which surgeons performed each case. The greatest improvement in terms of minimizing the number of acquired images and the required time was observed in obese specimens. Resident performance was most improved by the device use, yielding times and numbers of images similar to the highly experienced surgeon. Below is a detailed description of the cases.

In case 1, the senior surgeon initially placed the device too anteriorly, hence when a lateral image was acquired the most posterior channel still was too anterior to be an optimal option for a K- wire insertion. The device was removed, C-arm was rotated back to AP and the device was placed on the bone again. This time the device was not securely hammered into the bone resulting in a loose bone/device attachment. As a result, when the C-arm was rotated to obtain a lateral image the device became loose and the AP alignment was lost. Hence, a third attempt for securing the device into the bone was necessary. This time the device was fixed adequately and when moved to the lateral view a K-wire was inserted. The surgeon chose to remove the K-wire and manipulate the rotatable arm to obtain a better trajectory. This resulted in 4 K-wire drilling attempts versus only one attempt using the free hand approach in the first thin cadaver pair. The operative time required to perform the K-wire insertion task was almost four times more than the free-hand approach due to the number of repetitions in device placement. Consequently, the total number of image acquisitions was higher as well, due to the number of required images to visualize the device and the K-wire in each device placement.

In case 2, the surgeon’s first attempt to place the K-wire using the device was successful. However, using the free-hand approach 3 K-wire drilling attempts were made to achieve the satisfactory K-wire positioning in the second obese cadaver pair. The operative time was shorter and total number of acquired images was smaller utilizing the device compared to those of the free-hand approach.

In case 3, similar to case 1, the device was not docked adequately by the junior surgeon. This forced the surgeon to repeat the surgical steps leading to higher operative time and number of acquired images. In the second attempt, the device was secured and a K-wire was placed into the intramedullary canal. The K-wire placement was not satisfactory in the lateral plane, hence the K-wire was removed, the rotatable arm was manipulated and a different channel was selected.

The operative time lasted almost twice as long using the device compared to the free-hand approach.

In case 4, the device was successfully inserted into the bone in the first attempt. The rotatable arm was manipulated a few times leading to 3 K-wire drilling attempts to obtain a satisfactory K- wire positioning. In the free hand approach, manipulating the leg and obtaining a direct access line in an obese cadaver were challenges faced leading to a relatively higher operative time and more image acquisitions.

In both cases 5 and 6, the resident firmly hammered the device into the bone. In case 5 however, the device was initially placed too medially which was quickly corrected. In both cases, the operative time was shorter and the number of acquired images was smaller when the device was utilized.

NA NA

Fixed Fixed Fixed Fixed

Loose Loose

Attachment

Device/Bone Device/Bone

2 3 2 3

NA NA

Weight

1(Slim)

Cadaver Cadaver

Relative Relative

3(Obese)

4 1 2 3 3 5 3

1.41

FAST

1 3 4 2 8

Number of of Number

5.33 4.89

hand

Free

Drilling Attempts Drilling

he device/bone attachment was not sufficiently secured. sufficiently not was device/boneattachment he

5.5

12.1 4.07 4.17 4.43 7.26 4.14

13.27

FAST

Time(Min)

3.33 5.53 7.55 7.03 6.51

hand

Free

18.02 19.57 10.24

49 23 58 30 20 19

33.17 16.46

FAST

Images

Number of of Number

19 52 49 36 68 54

hand

Free

46.33 16.86

1 2 3 4 5 6

NA NA

Case Case

Number

Number of acquired fluoroscopic images and required time in minutes to complete the surgical task for three surgeons with surgeonswith forsurgicalthe three task complete to in minutes time required imagesand fluoroscopicof acquired Number

Junior

Senior

Level of of Level

Average

Resident

Standard Standard

Deviation

experience

Table the t cases where show rows highlighted of varied The experience. levels

The K-wire orientation misalignment in the AP and lateral views when the free-hand or the device approach was utilized are presented in Table 9. The paired-sample t-test was used to compare the means between two groups (SPSS). A trend was found in which AP misalignment was smaller using FAST in comparison to the free-hand approach (p=0.07). However, no significant difference was observed with respect to lateral misalignment.

Table 9 Observed AP and Lateral K-wire misalignment with respect to IM canal and the results for comparing the device with the free-hand approach. The difference is calculated as the Free-hand angle minus the device angle, a positive value representing an improved orientation outcome using the device.

AP Lateral Specimens Device (°) Free-Hand (°) Difference Device (°) Free-hand (°) Difference 1 0.77 10.64 9.87 3.04 7.22 4.18 2 1.04 4.36 3.32 4.89 15.97 11.08 3 1.00 3.00 2.00 0.41 16.04 15.63 4 1.34 11.14 9.8 3.68 0.28 -3.40 5 3.3 5.51 2.21 16.73 1.93 -14.8 6 7.7 6.26 -1.44 9.70 6.70 -3.00 Mean 2.25 6.81 4.29 6.40 8.02 1.615 Standard 2.70 3.34 4.58 5.90 6.73 11.03 Deviation t-value 2.297 0.359 p-value 0.070 0.735

The results for semi-structured interviews after the surgeries are summarized and presented in Table 10.

Table 10 Summary of the qualitative results obtained from short semi-structured interview after the surgeries

Level of Device Device vs. free-hand approach (transcribed from video) experience ease-of-use issues - The locking - “There are pros and cons in using any device, the mechanism is not user- device is a bit bulkier, hence it is taking time to getting friendly; help was it docked, sometimes there is a bit of soft tissue, necessary during the sometimes there is a lot. The free hand was easier to manipulation manoeuvre.”

Senior - “Once the device is docked you can adjust the - Sometimes the position easier.” device handle depending on the - “When using the device, you need to use a smaller K- position covers the wire, so it flexes a bit more than what you like.” views - “Device curvature really helps with the k-wire placement in the obese cadaver.” - The locking - “The best thing was the ability to get the more of a mechanism is not down the pipe shot, otherwise one needs to bend the intuitive wire and advance the K-wire. It goes down the shaft better and obviously, you have the options of other Junior holes to use if your wire is not perfectly in the spot.”

- “You don’t have to completely remove the wire out from the body just like in the traditional approach and having to take two more shots.” - The locking - “I liked the petal design, I had to adjust the angle in mechanism was hard the lateral, so I could fully withdraw the pin, readjust to rotate if only one and place it back and still have the same start point; hand operation is Whereas in the previous surgery (free-hand) when I had intended for this task to adjust the trajectory, I had to restart by finding the start point.”

Resident - “Getting the new start point was perfect.”

- “I did restart one time because my start point was not correct on the AP, Once the start point and trajectory is set, then there is no problem adjusting my trajectory in the lateral. I could do so in a very straight forward way without losing the progress that I made getting it correct in the AP. I think the device works.”

3.7.2.4 Discussion

Instruction adherence was critical to performance. The device must be fully hammered into the bone. If the device is loose the AP alignment may be lost and the surgical steps must be repeated. While the resident firmly secured the device with the current pin designs, it is important to further investigate improvements to the pin design to facilitate securing the device into the bone.

The device alignment in the AP must be checked prior to advancement into the bone. Otherwise, if the rotatable arm is misaligned with respect to the intramedullary canal, the optimal K-wire positioning in the coronal plane cannot be obtained. It is also critical to check the approximate position of the device in the sagittal plane. While the multiple channel design offers various choices to achieve optimal sagittal placement, if the device is initially placed too anteriorly or posteriorly, the device must be removed and reinserted.

The performance of the surgeons with the device in the second pair of cadavers was superior with an exception of case 6. This may be due to the fact that, although randomly assigned, the second pair of cadavers were always more obese making AP alignment more difficult in the free- hand approach. It may also be explained by the learning curve: first-hand experience the surgeons obtained through using the device and understanding the critical importance of instruction adherence. Further investigation of the device functionality must be accompanied by test samples set up under operating room conditions prior to the cadaveric testing to account for this potential effect. This will allow surgeons to understand the strengths of the device and the challenges that can occur if the device use instructions are not strictly followed.

The mean K-wire orientation misalignment with respect to the axis of the IM canal demonstrated a trend towards improvement in the FAST group compared to the free-hand technique in the AP view. This suggests the importance of the device in maintaining the AP alignment when lateral positioning of the K-wire is performed. Generally, the surgeons obtained an AP image to align the K-wire with the intramedullary canal. Subsequently, in both approaches the lateral positioning of the K-wire was performed. Upon obtaining the desired orientation and location in the lateral plane the K-wire using the free-approach was partially inserted. Another AP image was then acquired to confirm the K-wire positioning followed by final insertion of the K-wire if the alignment and location was still satisfactory. The results from this testing showed that even though the K-wire positioning was not ideal when confirmation images were acquired (due to

unintended hand movements), the positioning was still within the surgeons’ acceptable range. However, the alignment in the AP plane may have been more accurate using FAST as a result of maintaining the initial AP alignment. The acceptable limits of K-wire orientation misalignment are not well defined in the orthopaedic literature. Misalignment of fracture fragments has been defined with respect to angular deformity as greater than 5 degrees in either the coronal or the sagittal planes [15]. However, the causation of this deformity is not limited to the starting point (K-wire positioning) as it may also be introduced or mitigated during nail insertion, reaming, nail passage or interlocking [15]. Misalignments of greater than 5 degrees in the K-wire orientation were seen in both the freehand and FAST approaches, however as such misalignments can be corrected during the later surgical steps (i.e. reaming, nail insertion…) these K-wire positions were qualitatively considered as acceptable to the surgeons during the testing.

The impact of entry point location (trochanteric fossa versus the GT) in IM nailing has been studied with respect to ease of access, functional outcome and soft tissue damage at the hip [15, 99–101]. Although the trochanteric fossa is difficult to reach, it is collinear with the femoral axis which facilitates the alignment of the nail within the femur. However, functional impairments related to the trochanteric fossa entry point may result due to dissection of the abductor and external rotator muscles [101]. Two independent cadaveric studies demonstrated injury to branches from the medial femoral circumflex artery, piriformis and obturator muscles when the trochanteric fossa entry point was utilized [41,100]. However, a follow up randomized controlled study did not show appreciable differences between the two surgical approaches with regards to soft tissue damage and subsequent functional impairment [101]. The GT entry point has received popularity during recent years mainly due to the improved ease of access to the surgical site [15]. This translates to lower radiation exposure and decreased operative time, particularly in obese patients [83]. The reported soft tissue damage associated with this entry site is variable as a result of utilization of different entry point locations within the GT region, choice of reamer and nail design. FAST is designed to take advantage of the ease of access to the GT area while allowing the insertion of a K-wire into the trochanteric fossa facilitating the alignment of the nail with the IM canal. The GT entry point may also be utilized with pin design modifications.

In terms of device functionality, while a portion of the complaints associated with the locking mechanism may be explained by the quality of rapid prototyped parts, the concept of manipulating a knob during the procedure was not well liked. As per the user-centered design

process, in the final design the knob was removed and replaced with a trigger mechanism. The trigger allows the intuitive manipulation of the rotatable arm with the hand holding the drill, yielding a more similar feel to the free-hand approach for achieving the correct orientation. Releasing the trigger fixes the rotatable arm which ensures maintenance of the alignment.

One of the limitations of this testing was the use of cadavers as there is an inherent absence of active muscle forces. While, the presence of muscle forces may make the entry point selection more challenging using both the freehand and FAST techniques, the device may prove useful in this scenario in maintaining alignment that is challenging in freehand positioning. The second limitation of the study was the way the leg adduction was performed. As each cadaver had only a hemi-pelvis, it was difficult to adduct the leg while maintaining the pelvic position. This made insertion of the K-wire in both the free-hand and device use approaches more difficult.

Although, the number of samples was small in this pilot study, when the device instructions were followed step by step, the positive effect of the device was apparent. This was particularly evident when the resident used the device. The lack of instruction adherence in some cases even elongated the entry point selection step operative time. Poor compliance with new techniques as a result of old habits in experienced surgeons has been reported in the literature [103] and may be the reason behind the poor device use outcomes in cases 1 and 3. The performance improvement in the second pairs (cases 2 and 4) may also suggest a lack of proper instruction/training materials and sufficient time for surgeons to familiarize themselves with the device functionality before use in this trial. Overall, despite the limitations, all three participants indicated that they felt the device prototype was useful and helpful in completing the surgical task.

Based on the surgical observations and obtained feedback from the surgeons during and after the cadaveric testing, the list of surgical activities for FAST usage (previously shown in Table 7) has been modified to improve the instructional materials. Based on the hierarchical decomposition defined in chapter 2.4.2, the steps for IM nailing remain the same, however the entry point selection activities are replaced with the list presented below.

Entry point selection activities (trochanteric (piriformis) fossa). The orientation of the C-arm is indicated for each step in the left column, images are taken in steps highlighted in red. C-arm Steps orientation

1- Place the device pins (Figure 32 - 1,2) on the region of the femoral GT as shown.

4 5 AP 3

1,2

Figure 32 Placement of the device pins on the GT for trochanteric fossa (piriformis fossa) entry point

2- Obtain AP images and adjust the location and orientation of the pins on the GT if necessary until the distal rotatable arm trajectory (dotted line) is aligned with the IM canal and the pins are lined up (revealing only 1 pin on AP images).

a b

Figure 33 a) Correct rotatable arm orientation prior to the device insertion versus (b) incorrect orientation of the rotatable arm and (c) misalignment of pins (with 2 pins visible on the AP image)

3- Prior to fixation, ensure that the device is not placed too anteriorly or posteriorly AP by palpation. Adjust the AP positioning if required.

Figure 34 The region shows an approximate insertion site for the device pins

4- Hammer the device into the GT until the pins of the device are fully seated into the bone to ensure a secure attachment as shown (loose device/bone attachment will result in failure of the device to maintain alignment in the AP plane)

a b

Figure 35 The device must be inserted fully into the bone as shown in (a), insufficient insertion may result in loose device/bone attachment, as shown in (b)

5- Obtain an AP image to confirm that the device is fully inserted AP

6- Remove the device cover

Figure 36 The device cover is shown with hatched lines and the dotted arrow shows its direction of removal

7- Place a long K-wire (1.8 mm diameter, 285 mm length) into one of the three columns of the rotatable arm channels (Figure 37 - a). Obtain an AP image to confirm the K-wire alignment with the IM canal (Figure 37 – b, dotted line). If necessary, change the channel and obtain another AP image until satisfactory lateral K-wire and IM canal alignment is obtained. a b

1 AP

1 2 3

Figure 37 a) The petal design with entry points stacked on top of one another in 3 columns and b) a K-wire (1) inserted through the column 2 which is aligned with the IM canal axis

8- Take a lateral image. If the K-wire trajectory is: a. Satisfactory: advance to step 9

Figure 38 A lateral image showing a portion of the K-wire (solid black line) visible under fluoroscopy with a satisfactory K-wire trajectory (dotted line)

b. Not satisfactory: i. Engage and hold the trigger (1) to unlock the rotatable arm (2)

2 1

Lateral

Figure 39 Pushing the trigger as shown in the direction of the red arrow will unlock the rotatable arm

ii. With the trigger engaged, the hand holding the drill can be used to change the orientation of the rotatable arm until a satisfactory trajectory is achieved

Figure 40 K-wire trajectory adjustment using the hand holding the drill to indirectly manipulate the rotatable arm

iii. Release the trigger to relock the rotatable arm

Figure 41 Release of the trigger will move it to its original default position due to the spring loaded outer knob and lock the rotatable arm position

9- If needed, the AP position of the K-wire can be altered by selecting a different channel (rows 1-5) within the same column

a 1 2 3 4 5

Lateral

b 1 2 3 4 5

10- Advance the K-wire up to 4 cm into the bone. This limit is based on a K-wire of 285 mm length. Longer wires may go further into the bone. Shorter wires cannot Lateral be used.

11- Obtain a final lateral image. If the placement of the K-wire is not ideal, return Lateral to step 8

12- Confirm the K-wire position with an AP image. AP

13- Release the K-wire from the drill. AP

14- Remove the device from the bone, leaving the secure K-wire in place. AP

3.8 Conclusion

This simple device represents a novel surgical tool for use in femoral IM nailing. In contrast to the current entry point selection activity cycle (which may include the acquisition of multiple AP and lateral images, patient repositioning and inaccurate drilling), appropriate use of the device ensures maintenance of the 2D alignment obtained in the AP plane while adjusting positioning in the perpendicular (lateral) plane. This may enable entry point selection and orientation to be achieved with a reduction in cyclic repositioning and acquisition of AP and lateral fluoroscopic images. Future work evaluating the device performance is required on a larger cohort of specimens and utilizing a femoral shaft fracture model.

Chapter 4: Flexible Fracture Reduction Steerable Telescoping Wire – FLEX FiRST Wire 4.1 Abstract

Purpose: Intramedullary nailing (IM) nailing is the standard of care for adult lower extremity long bone fracture stabilization. In displaced fractures, initial intra-operative reduction is necessary to insert a long guide wire required for IM reaming that will facilitate subsequent nail insertion. Despite widespread usage of IM nailing, a procedure most commonly performed under fluoroscopic image guidance, surgical challenges often arise in obtaining adequate provisional reduction necessary to allow the insertion of a conventional IM guide wire. The aim of this study is to design a novel device that can be navigated through a diaphyseal fracture without accurate initial alignment.

Methods: A novel surgical tool was developed through a user-centered design approach to enable advancement of a cannulated wire through a malreduced diaphyseal fracture. A combination of prototype experimentation, finite element analysis (FEA) and design of experiments (DOE) was conducted to yield a prototype design with selected geometric and material parameters to enable mechanical performance (curvature, advancement, strength). Pilot experimental testing was conducted using the developed prototype on a synthetic bone to evaluate the ability of the device to traverse the fracture gap based on the simulation of malreduced fracture configurations (N=3).

Results: The proposed device, FLEX FiRST (FLEXible Fracture Reduction Steerable Telescoping) wire, design consists of a controller, a rigid outer tube with a flexible tip and a maneuverable thin cannulated inner tube. Insertion is guided by the controller which enables navigation of the device tip through the mal-reduced fracture site and subsequent insertion of the inner tube. In order to minimize the radius of curvature achievable with the inner tube, a commercially available helical spring design with a minimum height, a maximum pitch and a minimum width was selected based on the FEA DOE results. Based on the required forces needed for manipulation, a stranded wire was selected for attachment to the device tips. In the pilot experimental testing, the bending radii achieved by the outer and inner tubes along with the linear advancement of the inner tube enabled device advancement into the distal fragment in each malaligned configuration with an overdistraction of 1mm.

Conclusions: This work demonstrates an initial proof of concept of the FLEX FiRST Wire that enables navigation of the device through femoral fracture fragments without accurate initial alignment. Future development will focus on optimizing the design for ease of use, and development of the second phase of this device to facilitate reduction of the fracture prior to guide wire insertion. Ultimately, this novel technology may provide a simple, easy to use solution to facilitate challenging long bone fracture reduction by removing the requirement for accurate initial alignment prior to wire insertion.

4.2 Introduction

Reduction has been identified as the most challenging steps in IM nailing [82]. Prior to IM nail insertion, fractured fragments must be aligned, and a guide wire inserted into the bone IM canal to guide subsequent reaming or nail insertion. Malalignment of bone fragments makes passing a guide wire across a fracture site a technically challenging step in femoral fracture treatment. Current standard practice utilizes manually intensive techniques for fracture reduction. Intra- operative tools and methods have also been developed to facilitate complex femoral fracture reconstruction with variable success. However, problems in reduction accuracy remain and have been shown to lead to high radiation exposure to both medical staff and patients [60,81,104]. Extended operative times due to the reduction step can also significantly endanger patient safety, particularly in those who may suffer from poly-trauma and/or acute respiratory issues.

The surgical activities associated with the reduction step are well defined. With the help of an assistant, manual techniques (explained in detail later in this chapter) may be applied proximal and/or distal to the fracture. Fractured fragments are manipulated under fluoroscopic guidance until the fragments appear to be aligned in the AP view. The obtained reduction will be held carefully and the reduction checked in the lateral view. If fragments are also aligned in the lateral view a guide wire (commonly with a bent tip) is inserted in the proximal femur though the fracture level to the distal end of the femur. Otherwise, the surgeon must visualize the fragment locations in three dimensions and make the necessary manipulations. Such manipulation is often challenging particularly if the patient is obese or extremely muscular and requires repeated AP and lateral fluoroscopic images. The root source of these challenges was described as a lack of connectivity between 3D alignment and sequential fluoroscopic 2D imaging [82]. This lack of connectivity is illustrated in Figure 43 a, b. Once the surgeon successfully aligns the fragments in the AP view (Figure 43 - a1), the fractured pieces may not be aligned in the lateral image (Figure 43 - a2). Hence, the surgeon tries to reduce the fracture in the lateral view. Successful lateral reduction may be obtained (Figure 43 - b2), however, when an AP image is taken the fragment movement may result in recreating misalignment in the AP plane (Figure 43 - b1). This guided manipulation of one segment in a 2D plane and unintended movement of the alignment out of plane in the absence of a 3D visualization technology creates significant challenges in the reduction step.

a b

3 3

2 1 2 1

Figure 43 Technical challenges associated with the reduction step are shown in a) where AP alignment is achieved, however, in the lateral view the segments are not aligned. b) Further attempts to reduce the fracture in lateral view (2) may still result in AP malalignment (1). The isometric view (3) in both views (a) and (b) are presented as a reference, visualizing the relative location of the proximal and distal fragments

This study presents the development of a fracture reduction device through a user-centered design approach to address these challenges. The following sections outline device requirements and current state of art aimed at facilitating the reduction step. The proposed design features and the results of a finite element analysis using design of experiments are presented. Finally, pilot experimental testing of the device is presented.

4.3 Device Requirements A user-centered design is essential to accommodate limitations of surgical room and medical staff needs. The design criteria must incorporate geometric, functional and material/manufacturing components as listed below:

4.3.1 Geometry  The outer diameter of the device should be less than the inner diameter of the intramedullary canal (<11mm [4], <10 mm [15])  The device must be cannulated with a sufficient inner diameter to encompass a guide wire (2.5 to 3 mm) or offer a mechanism for subsequent guide wire insertion. Guide wires are made of Stainless Steel or Titanium with 2.5-3 mm outer diameter and length of 600-1000 mm

4.3.2 Device Functionality  The amount of distraction between proximal and distal end must be maintained below 10 mm to prevent soft-tissue injury [105]

 The device must be capable of navigating through the proximal and distal fragments of a malreduced femoral diaphyseal fracture after provisional reduction  The device needs to be flexible and manoeuvrable  Two simultaneous bending radii must be formed to enable the device tip to reach and advance into the distal fracture fragment (Figure 44)  During insertion through the intramedullary canal the device exhibit sufficient axial rigidity to prevent buckling and allow advancement through the fracture site  The device must be easy to use and allow for rapid insertion, realignment and withdrawal  The device must be robust to prevent any potential breakage within the intramedullary canal or damage to the canal.

First Bend Radius

Second Bend Radius

Figure 44 The device configuration after navigation through the distal fragment with two independent bend radii

4.3.3 Materials and Manufacturing  The device tip must be radio opaque in order to track the device fluoroscopically as it advances through the bone canal and is guided into the distal fragment  The device must be sterilizable  The device manufacturing, build and maintenance costs need to be considered with respect to hospital or budgetary constraints (must provide overall cost saving)  If the device package comes in parts and/or if there are disposable parts in the package that need to be assembled, the assembly procedure must be rapid and easy to accomplish  Life expectancy for the device parts and assembly must be indicated

 Any device parts with limited life expectances need to be available off-the-shelf and/or include replacement with the initial device  Step-by-step maintenance instructions need to be provided within the device packaging  A device manual and procedure for hands on training are required for: 1- device assembly (if the package comes in parts) and 2- successful device use

4.4 Existing technology

4.4.1 Manual Techniques

Several techniques have been introduced by AO in order to facilitate control of displaced fractured long bones to enable rigid guide wire insertion for IM nailing [34]. Depending on the location of the fracture, manipulation of the proximal or distal fragment may be challenging due to the presence of unbalanced muscle forces. In a more proximal femoral fracture, the iliopsoas muscle flexes and externally rotates the upper fragment. The gluteus medius and gluteus minimus muscles also abduct the proximal fragment. The lower fragment is adducted and pulled proximally by hip adductors. In femoral shaft fractures the proximal fragment experiences similar flexion and external rotation. The adductors pull also medialize the distal fragment. In a more distal fragment, the gastrocnemius muscle pulls the distal fragment backwards and hip adductors place the fracture into a varus position. Based on the fracture location, the aim of manual techniques is to act against these unbalanced muscle forces to align fractured fragments [106].

For indirect bone manipulation, a wrap can be placed around the femur (Figure 45 - a) to pull the proximal fragment down to line it up with the distal fragment. Alternative approaches to realign the fractured fragments employ manual elevation of the distal fragment by use of a crutch (Figure 45 - b) or lowering of the proximal femur through external pressure from a mallet (Figure 45 - c).

a b c

Figure 45 Fracture reduction using a) a wrap b) a crutch to elevate the distal fragment c) a mallet to lower the proximal fragment by external pressure [34]

The use of an “F-Tool” has also been reported to facilitate obtaining near complete or complete reduction [107]. In use, one limb of the “F-Tool” is placed on one side of the fracture and the other limb on the opposite side. The long lever arm is then manipulated to reduce the fracture (Figure 46).

2 1

Figure 46 A fracture reduction device with an F shape (1) for manipulating the proximal (2) and the distal fragments (3) to obtain alignment

Preliminary reduction may also be followed by the direct manipulation of the fractured femur via Schanz pin inserted into the proximal, distal or both segments (Figure 47 - a) or by utilizing a bone hook (Figure 47 - b). These techniques impart more direct control of the fracture fragments and enable more secure stabilization of an anatomic alignment.

a b

Figure 47 Direct manipulation of the displaced fragments via a) a Schanz pin and b) a bone hook [34]

The use of such hardware and manipulation of fracture fragments, however, must be performed carefully to minimize soft tissue trauma and prevent damage to neurovascular structures. When required, these manual techniques lead to a more invasive approach, an extended operative time, higher levels of radiation exposure and increased blood loss during and after the procedure.

4.4.2 Navigation Systems

Numerous navigation systems have been developed and analyzed to facilitate intramedullary nailing [60,79,81,108,109]. The use of fluoroscopy-based navigation has been reported to accurately measure femoral anteversion/antetorsion, reduce radiation exposure while increasing the precision of distal locking, preoperative planning, and improving surgeon’s hand-eye coordination by lessening the need for mentally correlating the image with the procedure.

An intra-operative navigation system consists of one or more tracking frames inserted into the main fractured fragments. An optical tracker unit emits infrared signals to monitor the frames’ locations. Based on the received signals, a computer unit integrates intra-operative images with the models of tracked frames to provide real-time tracking of the fractured fragments. Additionally, the navigation system software may enhance the quality of fluoroscopy images using image analysis techniques and create 3D models of bone fragments in real-time.

A few clinical studies have compared the performance of fluoroscopic navigation versus conventional operation for intramedullary nailing of femoral fractures [110,111]. Frank et al. utilized a navigation system for 19 patients and compared the results with 10 conventionally operated femurs. In the 19 cases, 1 case was not completed with navigation due to system error messages. The operative times for entry-point, reduction and distal locking were all significantly

longer when the navigation system was utilized. The total X-Ray exposure, however, was significantly lower compared with the conventional method in the reduction and distal locking steps.

The utilization of navigation in orthopaedic trauma applications is intended to simplify the fracture reduction work flow and create a “user friendly” environment. However, the complexity and time consuming nature of the navigation operation and handling (including attachment of optical trackers and registration), performance and the high financial, logistical and education demands have been major drawbacks limiting the clinical uptake of this technology.

4.4.3 Intramedullary Devices

Reducers have been reported to help manipulate proximal fracture fragments by giving the surgeon a “joy-stick” like control [106]. A reducer (also referred to as a “the finger”) is a curved cannulated rod made of stainless steel with a bent tip. Once advanced through the proximal fragment under image guidance, a guide wire can be inserted through the reducer. The bent tip of the reducer deflects the guide wire to facilitate passing into the distal fragment. The rigidity of the reducer also enables the surgeon to overcome muscle forces and extend the proximal fragment making alignment of fragments easier.

Intramedullary Bone Endoscopy (IBE) has also been proposed to assist in femoral fracture alignment [104]. One of the main advantages of this method is the very limited radiation exposure to patients and medical staff as a result of the direct visual control. In order to utilize this device, a preliminary reduction must be performed on an extension table. The rigid endoscope is then pushed to the end of the proximal fragment and visualization is used to identify the distal femoral fragment location. However, there is no description of any fracture reduction procedure associated with the use of this device. The device only replaces the need for an image intensifier with a direct visual control, but use of this device still requires manually intensive procedures to realign displaced fracture fragments.

4.4.4 Patents

Multiple fracture reduction devices and their method of use have been introduced in the patent literature. A description of two most relevant patents is given below.

In 1991, patent #5,002,543 was filed entitled “Intramedullary Fracture Reduction Device” which consisted of an inflexible elongated straight shaft with a steerable distal solid tip. For use, it requires preliminary fracture reduction (within the narrow device range-of-motion) and the force applied to the tip is used to further align the fragments. While this device may potentially reach the fracture fragments, its inflexible shaft and the solid tip with limited range of motion make it difficult to navigate through bone fragments. Additionally, the device only proposes a method to reach the fracture fragments, but offers no mechanism to realign (reduce) the fragments.

In 2015, a reduction device and a method for femoral shaft fractures (patent # CN104783874A) was proposed, which is placed around the leg on the skin and composed of series of air bags and a compressor. Based on X-ray imaging the desired airbag(s) is pressurized to reduce the fracture. This indirect manipulation of fractured fragments through pressurization of multiple airbags is intended to replace the need for a wrap, mallet or a crutch. While the amount of radial pressurization of the fractured leg is limited, the pressure may not be sufficient in obese or extremely muscular patients. Accurate fragment manipulation appears to be difficult to achieve considering connectivity between the fragments and subsequent unintended movement of other fragments.

4.5 Device Design features Presently, no clinically available product exists that satisfactorily addresses the physical reduction of challenging malaligned long bone fractures. As such, the aim of this study is to design a novel intramedullary fracture reduction device, the FLEX FiRST (FLEXible Fracture Reduction Steerable Telescoping) Wire, which facilitates guide wire insertion prior to IM nailing. The proposed device envisioned offers a two stage design process that will first allow surgeons to navigate the device through fracture fragments without accurate initial alignment. The second stage of the design consists of the reduction stage, which is beyond the scope of this thesis (see Chapter 5 future directions).

The device design for insertion through a malaligned fracture site, consists of a rigid outer tube with a flexible tip and a maneuverable thin cannulated inner tube. The outer tube and its tip are made of stainless steel and aluminum respectively. These pieces were machined separately and laser jet welding was utilized to make the pin joint as shown in Figure 48.

a b

c d

This design provides the axial rigidity necessary for device insertion into the bone canal while maintaining the device tip flexibility and resistance to lateral forces. The outer tube may also be used to manipulate the proximal segment and overcome muscle forces. This functionality is similar in nature to the method surgeons utilize gaining leverage via the reducer or sometimes the nail itself [1] to manipulate the proximal segment.

The inner flexible tube is made of T304 stainless steel, spiral wound to form a long helical spring tube with a protective cover (Figure 49). The protective cover of the inner tube is made of silicon rubber which is widely used in medical devices due to its resistance to high temperatures (for sterilization) and inert behavior. The cover maintains the integrity of the coils laterally and also increases the axial stiffness of the device by limiting the compression of the coils. The device is guided by a proximal joy-stick like controller which enables rotation (knob 1 – Figure 50 - a,b - 1) and bending of the outer tube tip (knob 2 – Figure 50 - a,b - 2). The controller

also allows inner tube axial advancement (knob 3 – Figure 50 - a,b - 3) as well as inner tube tip bending (knob 4 – Figure 50 - a,b - 4).

2 3

Figure 49 The final prototype of FLEX FiRST Wire showing: the controller (1), the rigid outer tube (2), (3) the outer tube tip, and the inner (4)

a b

3 1

4 2

3 5 4

The rotation of the outer tube must be coupled with the inner tube allowing both bend radii to be in the same plane. However, inner tube axial advancement and the bending of the tip must be entirely independent of one another. Such a mechanism was designed by machining a slotted shaft and coupling the knob 4 rotation with the bending mechanism as shown in Figure 51 a, b. In this, inner tube advancement is independent of the bending mechanism as the pin can freely move along the slotted shaft and if knob 4 is rotated the tip can be bent.

4 a 1 b

2 3 1

5 7 2

The controller unit consists of a simple mechanical wire based technology, which is robust, inexpensive, and easy to manufacture. In the final assembly the flexible inner tube is placed inside the outer tube and the control cables are attached to the controller. The control cable is connected to the circumference of inner tube as shown below (Figure 52), simulating a common endoscopic design configuration. For the outer tube the control cables are attached to both sides of the outer tube tip (Figure 53).

Figure 52 The control cable’s tip (1) is attached to the outer surface of the coil (2)

Figure 53 Cables (1) attachment on both sides of the outer tube tip (2) and supported with two 1.8 mm tubes (3)

The control cables are made of stainless steel and connect the tip of the outer and inner tubes to the control unit. In order to obtain bending motion of either the inner or the outer tube, the cables are pulled independently to bend the desired tip. This mechanism is similar to the bending mechanism implemented in traditional endoscopic designs and is commonly facilitated by a rotating disk as shown in Figure 50 - 5.

In order to traverse the fracture, the outer tube control cables can be engaged to bend the outer tube towards the distal femoral fragment by rotating knob 4. The rotation of knob 1 allows for movement in any desired direction. Upon formation of the first bend, the outer tube remains stationary to provide a guiding path for the inner tube. An identical mechanism is used on the inner tube to form the second bend by rotating knob 2. Axial advancement of the inner tube finally allows the inner tube to reach the distal end using knob 3 (Figure 54).

Outer Tube

Inner Tube

Figure 54 Inner tube advancement through the outer tube

Based on this workflow, the surgeon bends the tip of the outer tube to the desired orientation under fluoroscopic guidance in the AP view. The inner tube may also be bent and advanced such that alignment with the distal fragment is achieved. The surgeon then obtains a lateral image to determine if rotation via knob 1 is necessary for a lateral alignment of the tool and the distal end. If needed, such rotation introduces an error as shown in Figure 55 and depicted by the letter “e” in Figure 56. The amount of this error is directly proportional to the distance shown in Figure 56 with the letter “r” and inversely proportional to the rotation angle (α).

In order to maintain the AP alignment, automatic advancement or retraction of the inner tube must be applied. This length adjustment can be facilitated by use of a microcontroller to calculate the required length and applying the necessary correction via a motor. A Raspberry Pi, was utilized and programmed with Python to continuously calculate the required length adjustment and apply clockwise or counterclockwise wise rotation via a stepper motor for the retraction or advancement of the inner tube respectively. If the distance between the proximal and distal fragments is maintained within 2 cm and the amount of over-distraction at 1 cm, the maximum error introduced by rotation of the outer tube is less than 5 mm. (Note: the error correction mechanism was built but not implemented within the device to simplify the assembly for the proof-of-principle prototype).

a b

Figure 56 The diagram illustrates the error associated with the rotation of the device to be proportional to the distance X and the rotation angle α

The axial advancement is facilitated by knob 3 rotation. This knob is connected to a threaded shaft (Figure 51 - 5) that rotates inside the carriage threaded hole (Figure 51 - 6). Upon rotation of the threaded shaft, the carriage cannot rotate since its degrees of freedom are restricted with the presence of the pins (Figure 51 - 7). Therefore, rotation of the shaft results in advancement of the carriage and subsequent movement of the inner tube which is connected to the carriage.

The bending pattern and required forces for the outer tube tip manipulation are simple and easy to estimate. However, the inner tube helical spring comes in variety of physical and material properties. In order to understand the required bending forces to bend the helical spring to the desired radius of curvature and choose a control cable capable of withstanding the axial stresses a finite element analysis was performed.

4.6 Finite Element Analysis

In order to achieve the optimal bend radii, a thorough understanding of the helical spring stress distribution under specific load conditions are necessary. Multiple studies have developed analytical and numerical models to better understand the stress distribution in helical springs under axial and pure bending loads [112–115]. However, to date studies have not considered the optimization of bend radii under specific load conditions that represent a helical spring and cable assembly in traditional endoscopes. As such, the aim of this study is to understand the main

effects (geometric and physical properties of the tip and their interactions) that impact the radius of curvature governing steerability in traditional endoscopes, with direct applicability to the FLEX FiRST Wire design.

4.6.1 Methods

4.6.1.1 Experimental Design

A two-level five-factor design-of-experiments methodology was utilized to understand the main geometric and physical effects and their interactions on the attainable radius of curvature of an endoscopic device (outcome). The effects of five helical spring parameters (width, height, pitch, Young’s modulus (E) and applied force) were assessed at combination of high and low factor levels, resulting in 32 experimental groups (Table 11). The low and high values for width, height and pitch are based on commercially available endoscope designs. The boundaries for modulus of elasticity were chosen to include most typical metals chosen for biomedical applications from Aluminum (~70GPa) to Steel (~200GPa). The high factor level for force was calculated based on the maximum tensile force a typical 0.1 mm diameter steel cable can withstand before plastic deformation. The minimum force was set to 0.5 N such that the device tip in the most rigid configuration displaces at least 0.1 mm.

Factor Level Width (mm) Height (mm) Pitch (mm) E (GPa) Force (N)

Low 0.1 1 3 70 0.5

High 1 2.5 4 200 2

4.6.1.2 Finite Element Model Generation

A base finite element model was generated to represent a common endoscope design configuration. For this model, arbitrary values within a predefined range of physical properties (Table 11) were used to define the model geometry and material. In order to facilitate parametrization of these properties, a helical spring and a cable were created as separate parts within a finite element software program (ABAQUS, 6.12). The helical spring with a rectangular

cross section was modeled using linear brick elements (Figure 57 - a). The cable was modeled using linear 3-D truss elements such that once assembled the cable proximal end was connected to the coil proximal outer circumference (Figure 57 - b) and the cable distal end was placed inside the helical spring (Figure 57 - a). Physical properties were assigned to each part and interactions were modeled using a general contact algorithm.

An ENCASTRE boundary condition was used to fix the distal end of the helical spring. The cable was connected to the helical spring tip using a tie constraint. A concentrated force was then applied to the cable distal end to simulate the pulling force from the controller unit.

ABAQUS dynamic/explicit was utilized to analyze the generated model. The radius of curvature was calculated based on spatial displacement outputs identified for two nodes of interest. In this,

the first node was defined at the cable attachment site at the helical spring tip and the second node was identified by searching through the nodes displacements from the base of the helical spring. The first node with greater than zero displacement was marked as the second node. One of the limitations of this model is the absence of contact forces between the device outer shell and the IM canal.

Mesh convergence analysis was performed by refining the mesh size and calculating the radius of curvature and maximum deformation in each iteration. The model, initially containing 2988 nodes, 1733 elements and 8808 degrees of freedom, was refined until convergence was seen in the radius of curvature and maximum deformation (Figure 58). In order to minimize the computational cost for analyzing 32 models, the model with 2135 elements was utilized for the study which offers similar results (radius of the curvature difference ~ 5%, maximum deformation difference ~ 4%) as compared to the model with the most refined mesh. The selected model contains 3692 nodes, 2135 elements and 10920 degrees of freedom.

Radius of Curvature (mm) Maximum Deformation (mm)

0 1733 2135 3779 5465 Number of Elements

Figure 58 The mesh convergence analysis shows the change in the radius of curvature and the maximum deformation as a function of the number of elements

Automation of the above procedure was established to facilitate the analysis of multiple models. The ABAQUS Scripting Interface (ASI) was utilized to create parts, assign material properties, assemble parts, apply boundary conditions and force, request the nodal displacement outputs, request the deformed geometry image and solve the generated model. This data set was analyzed by a python script to identify the displacement associated with the two nodes of interest, calculate the radius of curvature, annotate images with their corresponding configuration parameters and organize the raw data to be used for design-of-experiments analysis (Figure 59).

Figure 59 Flowchart illustrating the automation procedure 4.6.1.3 Design-Of-Experiments

The two-level factorial design-of-experiments with five factors resulted in thirty-two configurations. The raw data containing each configuration along with its corresponding radius of curvature as the output response were imported to a commercial design-of-experiments- specific statistical package (Design-Ease). Based on the Box-Cox plot, a base 10 Log transformation of the output response was used to stabilize the variance. A Sum-of-Squares chart with a 2% weighted contribution threshold was utilized to determine the model inclusion criteria. A Shapiro-Wilk test was also conducted to confirm data normality.

In order to understand whether the model, effects and interactions were significant, Analysis of Variance (ANOVA) was conducted with a p value of <0.05 considered as significant. A Bonferroni correction was applied to account for the use of the same data in multiple independent estimates by dividing the alpha level by the number of estimates.

4.6.2 Results

All thirty-two models were solved successfully and the results are shown in Figure 60. The model, four of the input parameters (width, height, E, force) and two interactions (between pitch/width and pitch/height/E/width) were found to be statistically significant (P<0.0001). A Pareto chart summarizing the main effects and model errors is shown in Figure 61.

ined high and low low andinedhigh

measure (enabling tighter bends to be to bends be (enabling tighter measure

imize the radius of curvature outcome of curvature radius the imize

Automated analysis of 32 different endoscopic tip design configurations with pitch, height, E, applied force and width at def force height, width applied and E, pitch, with endoscopicconfigurations design ofdifferenttip 32 analysis Automated

60 min to theseof parameters enables resultant optimization model The

levels. achieved). Figure

Figure 61 A Pareto chart showing the main effects as hollow parts and excluded effects considered as model errors as solid bars

The largest contribution to the model was the width of the spring (effect=1.39, 58.33% contribution). The force and pitch/width interaction each had a negative influence, reducing the radius of curvature (allowing for tighter bends), whereas the height, E, width and pitch/height/E/ width interaction showed a positive influence. Pitch and all other interaction effects below the 2% contribution threshold were applied as model error. A summary of the statistical data as well as the final equation in terms of coded factors are presented in Table 12.

Table 12 Design-of-Experiments modeling results

Applied Source Coded Effect Contribution F Value P Value Transformation Factors (%) Base 10 Model NA NA NA 227.41 <0.0001 Logarithm Width E 1.39 58.33 810.45 <0.0001 Height B 0.77 17.82 247.54 <0.0001 Force D -0.63 12.19 169.30 <0.0001 Young’s C 0.37 4.26 59.23 <0.0001 Modulus (E) Pitch/Height/ ABCE 0.33 3.29 45.72 <0.0001 E/Width Pitch/Width AE -0.28 2.32 32.20 <0.0001 Final Equation Log (radius of Curvature) = 2.96 + 0.38*B + 0.19*C - 0.32*D + 0.69*E - 0.14*A*E + 0.16 *A*B*C*E

It is important to note that the negative AE effect shows that pitch and width work against each other’s effects to reduce the radius of curvature. This is also shown in a response surface modeled based on Width and Pitch interaction as shown in Figure 62.

Figure 62 Illustration showing the response surface based on pitch and width factors. An increase in pitch reduces the cable width effect on increasing the radius of curvature.

Using this model, geometric and physical properties can be optimized to meet specific endoscopic tip design criteria. A small pitch and width yield a radius of curvature optimal for acute bending radii. However, if due to design restrictions, a larger width is necessary, an increase in pitch may accommodate to yield a sufficiently small bending radius. A robust computational model was developed using a DOE approach that allows parametric optimization of endoscopic tip design parameters for radius of curvature minimization. Based on this model some of the identified disadvantages of traditional endoscopes can be addressed by altering design parameters for different applications. The challenges associated with device advancement have been reported in multiple studies [116–118] and served as a motivation to develop more complex robotic systems. These challenges may be addressed by understanding the tribological properties of the device use environment. If advancement of the device requires high pushing forces, the device axial stiffness must be increased by increasing the cable width and/or modulus of elasticity. Meanwhile, the acute bend radius can still be achieved by increasing the pitch, lowering the height and/or increasing the force applied to the cable forces. A study of FDA reports from 1985 to 2009 by Chen et al [119] shows that 21% of mechanism failures are due to

these cable malfunctions. In order to reduce this failure mechanism, once the required bend radius is identified along with coil design parameters, the amount of required force narrows down the selection of a suitable cable based on the DOE model.

4.7 Design Parameter Selection

In considering the device geometry, the rigid outer tube must fit inside the IM canal prior to reaming. The standard IM nailing implants are designed for patients with greater than 10 mm IM canal diameter [15]. Hence the rigid outer tube should fit inside an IM canal with 10 mm diameter. With a 1 mm gap to allow smooth insertion, the outer diameter of the device must be ≤ 9 mm. The control cables are attached to the circumference of the outer tube tip and stretched along the shaft to the controller. In order to reduce cable friction with the outer tube cover, the cables are housed inside two 1.8 mm diameter tubes (Figure 53 - 3) located on either side of the tip. Considering the dimensions of the two 1.8 mm tubes and the cover thickness, the inner diameter tube must be less than 5 mm. The closest imperial tube diameter is 3/16” (4.76 mm).

In order to design the helical spring for the inner tube, the overall diameter of the spring must be smaller than 4.7 mm. With such a small outer diameter, high axial stiffness can be obtained by selecting a material with high elastic modulus such as stainless steel. In order to minimize the radius of curvature based on the FEA, a minimum height of 0.04” and a maximum pitch of 0.078” were selected. The strength of the control cable was also increased by choosing a stranded wire to withstand the expected applied forces.

4.8 Experimental Evaluation

The initial evaluation of the prototype device was focused solely on functionality testing without integration of the compensation for linear advancement within the controller (manual control only of bending and linear advancement). This testing was performed utilizing a synthetic femur model (Sawbone) with a diaphyseal osteotomy. The ability of the device to navigate the malaligned femur was evaluated within a 6 degree-of freedom jig. The proximal segment was mounted on a platform which allowed translation and rotation in x, y and z-axes with 0.05 mm and 0.5° accuracy respectively. A clamp mounted on a ball joint was used to secure the distal fragment (Figure 63). The device was advanced distally through the proximal femur fragment to

the fracture level at which point the control cables were engaged to navigate the device into the distal fragment.

Figure 63. Experimental testing setup for consistent fracture displacement and rotation simulation shown in a) the lateral and b) the AP views

The ability of the device to traverse the fracture was tested in three configurations based on simulated displaced fragments due to muscle forces after proximal, mid-shaft and distal fractures. While the same specimen was used in all three simulated fractures, the orientation of the fragments was altered based on each simulated fracture.

In simulating a proximal fracture, the proximal fragment was flexed, externally rotated and abducted while the distal fragment was slightly adducted (Figure 64 a, b). In order to simulate mid-shaft fractures, the proximal fragment was flexed and the distal end was put in a slight

adduction (Figure 64 c, d). Finally, a distal fracture was simulated by extending the distal end and adducting the proximal end (Figure 64 e, f). For all cases, the provisional reduction was conducted such that the over-distraction and distance between shaft axes were approximately 1 cm and 2 cm respectively.

a b

c d

e f

Figure 64. Simulated proximal and distal displaced fragments in a proximal (a, b), mid-shaft (c, d), and distal (e, f) fractures

100

4.9 Results and Discussion The ability of the device to traverse the fracture was successfully achieved in all three configurations (Figure 65). The bending radii achieved by both the outer and inner tubes along with the linear advancement of the inner tube enabled device advancement into the distal fragment in each malaligned configuration. This preliminary testing demonstrates an initial proof of concept of the FLEX FiRST Wire stage one design that enables navigation of the device through femoral fracture fragments without accurate initial alignment. Upon the provisional reduction with the fragments within the operational range of the device, engaging control cables can guide the inner tube to be inserted inside the distal end of the femur.

a b

c d

Although the testing was not conducted under fluoroscopic guidance, this device may facilitate visualization of the fragments by giving the surgeon a reference for the location of the proximal fragment. The challenge associated with the reduction step as described in detail in section 4.1 is related to unintended movement of fragments which subsequently destroy the previously obtained alignment. Using the FLEX FiRST wire, the surgeon can maintain (prevent

101

unintentional movement) or manipulate the proximal end of the fracture. Hence, the provisional reduction may be facilitated by following the current surgical workflow while directly controlling the positioning of the proximal fragment. This functionality of the device is similar in nature to Schanz pin insertion, which is widely advised in the literature for direct manipulation of the fractured fragments in challenging cases [1,106,120]. If such comparison is proved accurate, one advantage of the FLEX FiRST wire may be the elimination of percutaneous pin insertion and subsequent tissue damage.

An important limitation of this study was the amount of provisional reduction required for the device use. A 1 cm over-distraction was applied in all three configurations and the distance between proximal and distal fragments axes viewed at a transverse plane at the fracture level was kept below 2 cm. Although the maximum allowable distraction is recommended to be 1 cm [105], in multiple trauma patients with existing soft tissue damage this value may not be considered safe. In addition, the overall geometry of the current device has been designed with consideration of the limitations of traditional machining. Micro or meso-machining would allow the manufacturing of parts, such as the outer tube tip, on a smaller scale with higher precision. Such modification could lower the over-distraction required between the proximal and distal fragments needed to utilize the device successfully.

During the initial testing conducted on the FLEX FiRST Wire the proximal and distal fragments and the device were directly visible. As such, this initial testing did not consider the limited 2D views for visualizing 3D geometries in the operating room nor the presence of muscle forces. Hence, the results of the testing are limited in their interpretation and may only be considered as a successful initial device proof-of-principle study. Further integration of the raspberry pi controller and testing in cadaveric models with the presence of soft tissue structures is needed to evaluate device performance. This work will also require evaluation of the range of provisional reduction under which the device can function.

The synthetic bone shown in Figure 63 - a appears to have a slight anticurvature, however, this curvature is not matching the human bone. The straight IM canal however, facilitated insertion of the device with a straight outer tube. In addition, in the design process, the canal was assumed to be ≥10 mm in diameter. Hence, in its current configuration, device use is not possible clinically in patients with small IM canals.

102

The ability of the device to traverse the fracture was conducted with no soft tissue or particles in between the two fragments. The advancement of the device through fracture hematoma at the site of fracture between proximal and distal fragments must be tested to understand whether the device offers enough axial and bending stiffness to penetrate through the hematoma to the distal fragment. The jagged edges of the fractured segments as well as intercalary fragments in segmental, comminuted fractures may impede advancement of the device. Further testing is required to simulated device advancement under different fracture patterns with intercalary fragments.

The FLEX FiRST Wire device at its current stage of development addresses only the first phase of the overall device design. Further development is required to enable the device to be capable of reducing the fracture prior to advancing the guide wire. In the next chapter, a proposal is presented in future work toward accomplishing this reduction step.

103

Chapter 5: Summary and Future Work 5.1 Summary of work

In chapter 1 a brief introduction to femoral fracture types, the standard of care for diaphyseal femoral fractures and surgical alternative techniques were discussed. A brief summary of all methodologies used in this thesis was outlined along with the motivation and objectives. Three specific aims identified in the objectives section were addressed in the subsequent chapters 2-4.

In chapter 2, the IM nailing surgical process was analyzed and a hierarchical decomposition model was presented to structure the IM nailing surgical procedure. Two surgical steps of entry point selection and reduction were identified as the most challenging and found to be associated with high levels of frustration in the operating room. The lack of connectivity between sequential 2D images for visualization of 3D objects was shown to be a root cause of observed challenges in these steps. This study was published in the international journal of computer assisted radiology and surgery.

In chapter 3, a user-centered design approach was described in the design, prototyping and testing of a novel surgical tool, FAST, to facilitate 2D - 3D alignment in entry point selection and orientation. A synthetic bone model was used to conduct formative usability testing on FAST. The design issues identified and surgeons’ comments were considered to prepare a next generation device used in cadaveric testing. FAST, when used according to its instructions, enabled a reduction in operative time, required number of fluoroscopic images and number of K- wire drillings compared to the free-hand approach. The cadaveric testing revealed the importance of instruction adherence, as insufficient initial stabilization of the device to the proximal femur led to poor outcome requiring additional time and imaging. All surgeons found the device useful, particularly in challenging cases dealing with the obese thighs. A provisional patent was submitted for FAST; the detailed patent description can be found in the appendix.

In chapter 4, the design of the FLEX FiRST Wire was introduced. The prototyping of the device was presented including an in depth Finite element study of the helical spring design. The initial prototype performance of the device to transverse the gap in femoral fractures was tested on a synthetic femoral fracture model.

104

5.2 Future Work (FAST)

5.2.1 Design Optimization

Based on the results of the cadaveric testing, several features can be explored to optimize the design and enhance the device ease-of-use, including: pin design, modifications to address poor initial device placement, visualization and the locking mechanism.

Improvements in the pin design may facilitate device placement reducing the need for extensive and time consuming hammering. Insufficient hammering with the current design configuration was found to result in loose bone/device attachment and subsequent AP alignment loss. In this, the diameter, length, number of pins, distance between the pins, as well as the tip profile may be altered to improve attachment. The current 2 mm diameter pins with 7 mm usable length were chosen to support the entire device weight when fully inserted into the bone. This is important as the surgeon may want to leave the device attached to the bone for a short period of time once it is fully inserted (as was observed in the testing) for C-arm manipulation or K-wire placement into the drill chuck. The use of two pins versus only one pin prevents unintended rotation of the device, however a pin with rectangular cross section may also prevent such rotation (Figure 66). Additionally, if the frame is made of a lighter material, the pin design can be altered based on the overall weight of the device and associated required bending stiffness of the pins.

Figure 66 A pin design with partial rectangular cross section to minimize unintended rotation

105

The use of static pins may not be the only method of achieving temporary device/bone attachment. In order to minimize possible soft tissue damage during insertion of the device into the proximal femur, pins may be deployed dynamically after the device is placed on the bone. Dynamic deployment of pins may also help the initial critical AP alignment prior to hammering. With the static pins, the surgeon aligns the rotatable arm, which is the pin length away from the bone with the intramedullary canal. This alignment may be improved with the use of dynamic pins. In addition, slight hand movement during the hammering step may also lower the AP alignment accuracy that may be improved by implementing a mechanism for dynamic pins insertion. Although the use of pins and the hammering of the device for attachment are very familiar to orthopaedic surgeons, the device may alternatively be screwed to the bone to maintain the alignment with the intramedullary canal. However, this type of attachment would likely prove more cumbersome and time consuming.

During the testing, the device was placed too anteriorly in one cadaver which made removal and repeated placement of the device inconvenient and time-consuming. While, this can be prevented by checking the approximate location of the device with respect to the greater trochanter area, it would be ideal if the device offers a faster and more convenient alternative to adjust the sagittal location of the rotatable arm. Two immediate solutions may be investigated to address this issue. The rotatable arm height could be increased, however, this may make the device insertion more challenging through a standard incision. In addition, the footprint of the pins on the bone must be considered in relation to the possible entry points the rotatable arm offers. The current device leaves small to no footprint of the pins on the bone after reaming and nail placement which makes the design ideal. Alternatively, the rotatable arm may be given a second degree of freedom in the sagittal plane to allow for a readjustment of the sagittal positioning (Figure 67). This change would require a redesign of the fixed curved frame, rotating shaft and the rotatable arm.

106

Figure 67 An alternative tip design to allow sagittal movement of the rotatable arm

Additional, minor design changes such as introducing a radiolucent handle may also have a significant effect in the device ease of use. The current handle as shown in Figure 68 leaves a shadow when lateral images are acquired. As such, a radiolucent handle may allow for better visualization of the K-wire and reduce the number of required fluoroscopic images.

Figure 68 The shadow of the handle is shown in this lateral image (the borders of the shadow is shown with a dotted line)

107

The current locking mechanism is designed around two 3D printed matching gears. As described in details in section 3.4.3, the trigger disengages the two gears allowing the rotation of the rotatable arm. After repeated use, some of the teeth on the plastic gears were worn down allowing limited rotatable arm movement in the locked position. Replacement of these gears with a more robust material, can improve the locking mechanism and provide a more consistent positioning of the rotatable arm.

5.2.2 Instruction Materials

A learning curve is defined graphically by plotting the performance in using a device or method against experience. It is defined by five stages: 1. commencement of training, 2 competency, 3. small improvements, 4. plateau and 5. a fall in performance (deteriorating eyesight, memory and cognition due to advancing age) [121]. Determination of the learning curve associated with device use is important in understanding the amount and the kind of training needed to optimize performance. Although, robotic systems are not comparable in terms of required technical competency with mechanical surgical tools, similar training modalities may be implemented on a smaller scale to create a robust training program for all new tools. A systematic review of different training modalities for robotic systems highlighted, didactic, skills laboratory, virtual reality-simulators, cadaveric testing, live case observation, proctoring and mentoring as the means to improve surgeons’ performance [122]. Considering the simplicity of the FAST device compared to robotic systems and its current development, didactic teaching, cadaveric testing (see following section 5.2.3) and proctoring would be applicable in construction of a robust FAST training program.

In the current didactic instruction, a list of surgical steps and a short device demo were provided Table 7. While, the steps are clear and device functionality was briefly discussed, the device limitations were not included in the training material. This may have contributed to the challenges surgeons experienced in cases 1 and 3 (improper attachment of the device to bone). An animation demonstrating the procedure, device functionality and incorrect device use scenarios could improve performance. Such an animation may include videos recorded during the pilot cadaveric testing to highlight challenges they may face if the device is used incorrectly.

Proctoring was not performed during the cadaveric testing of FAST. The teaching material at the time of testing was assumed sufficient and surgeons were left to perform the surgery on their

108

own (although they were able to ask questions during the procedure). In future cadaveric testing, if, for example, the device is not fully secured to the bone, the surgeons must be advised and continuation of the surgery with loose bone/device attachment must be prevented.

5.2.3 Cadaveric Testing

Following the integration of the suggested device improvements and the preparation of more robust instructional materials, additional cadaveric evaluation of FAST is required to demonstrate its performance in entry point selection/orientation in femoral shaft fractures. A similar matched cadaveric sample comparative study to free-hand technique is suggested. Based on the standard deviation of the operative time difference between the two groups in pilot testing, estimated power of ~80% and significance level of 0.05, a sample size calculation lead to 14 samples (28 femoral specimens) required for the next round of cadaveric testing.

The cadaveric anatomy should include the upper torso to simulate a realistic model blocking a direct access line to the desired entry point. In the distal end, the knee joint should be included to provide enough room for securing the cadavers. Consistent fractures should be simulated (i.e. via surgical osteotomy) to simulate a relevant clinical scenario. For example, proximal fractures often lead to surgical challenges in IM nailing. If such a fracture is modeled, the proximal segment may also be manually flexed, externally rotated and abducted for a better representation of iliopsoas and gluteal muscle actions. A strap can be placed around the femur and a cable/pulley mechanism may be implemented to pull the proximal fragment from the lesser trochanter area.

The three quantitative outcome measures of operative time, number of fluoroscopic images and number of drilling attempts should be recorded following identical variable definition as presented in Chapter 3. A repeatability study should also be conducted by asking each surgeon to repeat both approaches on the same cadaver. This can be accomplished by switching the sides used with respect to testing the device and freehand techniques. This would yield an assessment of the variability inherent in the device use. In addition, the assessment of reproducibility of the device use should be extended by selecting surgeons with varied levels of experience to evaluate the association between device performance and the end user’s level of expertise (resident, junior or senior level surgeon).

109

The length of incision, number of C-arm orientation changes (AP to L and vice versa) and final location of the K-wire can be measured. Each surgeon should be invited for a short interview after the surgery to understand their concerns, comments and evaluate their qualitative feedback.

The number of C-arm orientation changes could differentiate between repeated images taken to optimize the location of the K-wire in one plane when using the device versus AP to lateral and vice versa manipulation of the C-arm, frustrating the medical staff during the free hand approach. After testing, the femurs can be stripped of soft tissue and the exact location and orientation of the K-wires evaluated. A third party expert may be asked to evaluate the K-wire positioning and the accuracy and precision of the device versus the free-hand technique can be compared.

5.2.4 Translation

Consistent and successful use of the FAST device demonstrated through cadaveric testing would motivate consideration of in vivo study of the device use. An understanding of the device performance and associated risk can be generated from the cadaveric testing and used in application to the research ethics board for obtaining required permissions to test the device in humans. Preliminary testing may consider a single centre Sunnybrook cohort study, which can be performed prior to Health Canada approval (based on in house development of the device).

The FAST technology has been taken on by the Technology Transfer office at Sunnybrook Research Institute. In this a commercialization agent has prepared a summary of the device to reach potential medical companies in the orthopedic field for possible future collaborations. An initial meeting was held with Smith and Nephew in April 2016 and preliminary discussions are upcoming with Johnson & Johnson Innovations.

5.3 Future Work (FLEX FiRST Wire)

5.3.1 Design optimization – Insertion

Based on the initial prototype of the FLEX FiRST Wire, several features have been identified which can be implemented to optimize the design and enhance the device ease-of-use for stage one navigation of the device through fracture fragments without accurate initial alignment. These include the outer tube design, manufacturing, locking mechanism, modifications to address poor initial device placement, visualization and the locking mechanism.

110

Currently, the device outer tube is made of a straight stainless steel shaft. In order to facilitate insertion of the device into the IM canal, the anticurvature of the femur must be considered. The femoral anterior radius of curvature is 120 cm (±36 cm), however IM nails come with radii of curvature ranging from 186 to 300cm [123]. Hence, the outer shaft could be designed with a 120cm radius of curvature, or be manufactured with a range of curvatures based on current IM nail geometries.

With an exception of ball bearings and the threaded shaft responsible for axial movement of the inner tube, all other parts are 3D printed in the controller unit. The main advantage of the 3D printed parts is a significant decrease in the required manufacturing time for complex geometries. This allowed testing of the parts and alteration of design components until the desired functionality was achieved with multiple moving parts in the assembly. The overall performance of the device was inconsistent during the trials, since the 3D printed parts were made of plastic. Currently, with the final design of controller parts, the components can be machined instead of rapid prototyped for better performance.

The current design does not offer a locking mechanism for knob 1, 2 and 4. Hence, during testing it was difficult to maintain the desired bend radius while advancing the inner radius. A locking mechanism should be designed to facilitate the tube manipulations.

5.3.2 Design - Reduction

Following successful placement of the device through the distal fragment, reduction is required to achieve sufficient alignment to enable a K-wire to pass through the cannulated device and entre the distal fragment. Possible solutions are presented herein by which the FLEX FiRST Wire can be straightened to yield alignment.

The first idea uses internally generated forces capable of moving the fragments into alignment. In this, it is hypothesized that pressurization may be utilized to provide sufficient forces to reduce the fracture. One possibility is a balloon based inflatable system which can be attached to the outermost surface of the outer tube. Inflation of the balloon can be achieved through a pressurized saline infusion leading to a straightening of the device within the proximal and distal IM canals leading to their alignment. This approach would require sufficient purchase in the distal fragment to prevent dislodging of the device during inflation. Other design challenges to

111

be addressed in this approach include the dimensions and robustness of the balloon (as it must move through the IM canal and cannot be punctured by bone fragments) and the achievable forces in relation to the opposing physiologic forces (i.e. muscles).

With the experimental test set up as shown in Figure 69, loading conditions (including muscle forces) can be simulated to evaluate the ability of the balloon inflation for providing sufficient forces for fracture realignment. The initial prototype of the inflatable balloon can be evaluated first in simulated fractured femora (sawbones) and subsequently in cadaveric thighs incorporating simulated muscle loading. In the experimental set-up, the ability of the device to reduce the fracture can be evaluated within the 6 degree-of freedom jig, which allows consistent and repeatable fracture displacement and rotation to be achieved within a simulated physiologic environment. A pulley-weight system can be utilized to simulate the muscle forces. Based on intra-operative measurements, maximum lateral/medial and anterior/posterior resulting forces occur on the order of 280 N during fracture reduction [124]. Therefore, 280N should be applied to the proximal fragment through the supporting wrap. The balloon can then be pressurized with saline to reduce the fracture. The optimization of material selection for the balloon requires development of a finite element model and a comprehensive analysis.

If insufficient forces cannot be achieved utilizing the pressurization technique, alternative approaches can be investigated to improve fracture reduction. The end of telescoping tube could be equipped with a mesh that can be expanded (and retracted) once the device is within the distal fragment. The friction between the expandable mesh and the distal bone canal may assist in realignment of the fractured fragments as the device is retracted.

Another alternative would be to insert sensors at both ends of the telescoping tubes to provide sensory feedback guiding fragment alignment based on the linear positioning of the sensors. This would allow the fracture to be reduced using traditional methods, lessening the need for continuous fluoroscopic visualization. Directionality of motion may be indicated using varying the sensory clues (visual or auditory).

112

Proximal fragment Supporting wrap

Distal fragment

6 DOF jig

Pulley

Weight

Fixed support

Figure 69 Applying a force through a pulley system to the proximal fragment to simulate muscle forces 5.3.3 Translation

The translational timeline for the FLEX FiRST Wire remains a long way off. In this, prototyping including integration of the insertion and reduction components of the device will require an iterative process. The mechanical performance of the assembly as a whole must be evaluated based on synthetic and in-vitro cadaveric testing. Concurrent usability testing results can identify the requirements for a successful procedure and identify potential use-related hazards. If successful this work may ultimately lead to patent filing, in vivo testing and commercialization activities.

5.4 Applications

The surgical challenges associated with femoral IM nailing based on the lack of connectivity between sequential 2D imaging and 3D localization are found in the IM nailing of other long bones and in surgical stabilization of other skeletal structures (i.e. the pelvis). As such, the FAST device or its concept may be extended in the development of tools to facilitate entry point selection and orientation in other bones that currently rely on intraoperative fluoroscopic 2D imaging. Similarly, the FLEX FiRST Wire concept may be extended for use in the reduction of other long bones.

113

5.5 Contributions

This thesis contributed to the field of biomedical and clinical engineering in the following areas:

1- Advancing original knowledge in femoral IM nailing through surgical process analysis and identifying the root causation of common surgical challenges in this procedure based on a lack of connectivity between sequential 2D imaging and 3D localization

2- Development of a novel surgical device (FAST) to facilitate entry point selection in IM nailing

3- Creation of a computational model to optimize helical spring design parameters for endoscopic applications, including optimization of its parameters for the FLEX FiRST Wire.

4- Developing the first stage of a proof of principle fracture reduction device (FLEX FiRST Wire) that can navigate through a malaligned fracture.

5.6 Significance

A comprehensive analysis of the IM nailing surgical process through surgical observation and interviews allowed structuring of the procedure with hierarchical decomposition and highlighted challenging steps and activities. The obtained results may be broadly utilized to guide the future developments of surgical instrumentation and surgical process optimization by addressing the clinical needs in specific steps in the IM nailing procedure. The lack of connectivity between sequential fluoroscopic 2D alignment was identified as the underlying causation of surgical challenges in the entry point selection and reduction steps. Following a user-centered design approach, two devices were designed, prototyped and tested to address the identified challenges. This work increases our knowledge and capabilities with respect to the IM nailing surgical procedure which may ultimately lead to better predictability in the workflow and reducing OR time, radiation exposure, medical staff frustration, and incidence of malunion.

114

References

[1] R. Winquist, S.H. Jr, D. Clawson, Closed Intramedullary Nailing of Femoral Fractures, J. Bone Jt. …. 66 (1984) 529–539. http://jbjs.org/content/83/12/1912.short.

[2] S. Salminen, Femoral shaft fractures in adults: epidemiology, fracture patterns, nonunions, and fatigue fractures, 2005.

[3] H.A. Peterson, Epiphyseal growth plate fractures, 2007. doi:10.1007/978-3-540-33802-4.

[4] A.P. Uzel, J. Deloumeaux, J.L. Rouvillain, G.Y. Laflamme, A. Durandeau, P. Caix, Comparative study of femoral diaphyseal morphometry in two male populations, in France and a French West Indies island: An example of clinical relevance of comparative anatomy for orthopedic practice, Surg. Radiol. Anat. 33 (2011) 235–240. doi:10.1007/s00276-010-0722-5.

[5] G.M. Georgiadis, T. a Olexa, N. a Ebraheim, Entry sites for antegrade femoral nailing., Clin. Orthop. Relat. Res. (1996) 281–7. http://www.ncbi.nlm.nih.gov/pubmed/8804303.

[6] O.P. Lakhwani, P.S. Mittal, D.C. Naik, Piriformis fossa - An anatomical and orthopedics consideration, J. Clin. Diagnostic Res. 8 (2014) 96–97. doi:10.7860/JCDR/2014/6905.4118.

[7] S.P. Kale, N. Patil, S. Pilankar, A.R. Karkhanis, V. Bagaria, Correct anatomical location of entry point for antegrade femoral nailing, Injury. 37 (2006) 990–993. doi:10.1016/j.injury.2006.06.003.

[8] G. Küntscher, Practice of intramedullary nailing, Am. J. Surg. 116 (1968) 803. doi:10.1016/0002-9610(68)90374-7.

[9] T. Tamaki, A. Nimura, K. Oinuma, H. Shiratsuchi, S. Iida, K. Akita, An Anatomic Study of the Impressions on the Greater Trochanter: Bony Geometry Indicates the Alignment of the Short External Rotator Muscles, J. Arthroplasty. 29 (2014) 2473–2477. doi:10.1016/j.arth.2013.11.008.

[10] J. xin Zhao, X. yun Su, Z. Zhao, L. cheng Zhang, Z. Mao, H. Zhang, L. hai Zhang, P. fu

115

Tang, Predicting the optimal entry point for femoral antegrade nailing using a new measurement approach, Int. J. Comput. Assist. Radiol. Surg. 10 (2015) 1557–1565. doi:10.1007/s11548-015-1182-5.

[11] P.N. Streubel, A.H.W. Wong, W.M. Ricci, M.J. Gardner, Is there a standard trochanteric entry site for nailing of subtrochanteric femur fractures?, J. Orthop. Trauma. 25 (2011) 202–207. doi:10.1097/BOT.0b013e3181e93ce2.

[12] N.B. American Academy of Orthopaedic Surgeons (AAOS), Principles Of ALS Care Paperback, Jones & Bartlett Learning, 2009.

[13] X. Ding, J. Boney-montoya, B.M. Owen, A.L. Bookout, C. Coate, D.J. Mangelsdorf, S.A. Kliewer, Trends in subtrochanteric, diaphyseal, and distal femur fractures, 1984–2007, 16 (2013) 387–393. doi:10.1016/j.cmet.2012.08.002.

[14] N. Romeo, J. Deitch, Femur Injuries and Fractures, Medscape. (2011). http://emedicine.medscape.com/article/90779-overview#a0199 (accessed November 1, 2012).

[15] R.W. Bucholz, J.D. Heckman, Rockwood and Green’s Fractures in Adults, Wolters Kluwer, 2010.

[16] B.D. Browner, Skeletal trauma, Saunders, Philadelphia, PA, 1992.

[17] J. Bernstein, B. Monaghan, J.S. Sibler, W.G. De Long, Taxonomy and treatment - a classification of fracture classifications, J. Bone Jt. Surg. 79 (1997) 706–709. doi:10.1302/0301-620X.79B5.7177.

[18] C.L. Loizou, I. McNamara, K. Ahmed, G.A. Pryor, M.J. Parker, Classification of subtrochanteric femoral fractures, Injury. 41 (2010) 739–745. doi:10.1016/j.injury.2010.02.018.

[19] Rommens, Pol Maria, and Hessmann, Martin H., eds. Intramedullary Nailing : A Comprehensive Guide. London, GB: Springer, 2015. ProQuest ebrary. Web. 20 July 2016. Copyright © 2015. Springer. All rights reserved., (2016).

116

[20] T.E. Shuler, G.S. Gruen, O. DiTano, B.L. Riemer, Ipsilateral proximal and shaft femoral fractures: Spectrum of injury involving the femoral neck, Injury. 28 (1997) 293–297. doi:10.1016/S0020-1383(96)00199-4.

[21] T. Meling, K. Harboe, C.H. Enoksen, M. Aarflot, A.J. Arthursson, K. Søreide, How reliable and accurate is the AO/OTA comprehensive classification for adult long-bone fractures?, J. Trauma Acute Care Surg. 73 (2012) 224–31. doi:10.1097/TA.0b013e31824cf0ab.

[22] M.E. Müller, Müller AO Classifi cation of Fractures—Long Bones, AO Found. (1987).

[23] J. Manninger, U. Bosch, P. Cserháti, K. Fekete, G. Kazár, eds., Internal fixation of femoral neck fractures, Springer Vienna, Vienna, 2007. doi:10.1007/978-3-211-68585-3.

[24] S.D. Heinrich, The operative stabilization of pediatric diaphyseal femur fractures with flexible intramedullary nails, Oper. Tech. Orthop. 5 (1995) 115–125. doi:10.1016/S1048- 6666(95)80003-4.

[25] M.S. Kocher, E.L. Sink, R.D. Blasier, S.J. Luhmann, C.T. Mehlman, D.M. Scher, T. Matheney, J.O. Sanders, W.C. Watters, M.J. Goldberg, M.W. Keith, R.H. Haralson, C.M. Turkelson, J.L. Wies, P. Sluka, K. Hitchcock, Treatment of pediatric diaphyseal femur fractures., J. Am. Acad. Orthop. Surg. 17 (2009) 718–725. doi:17/11/718 [pii].

[26] a L. Akinyoola, O.O. Orekha, F.O. Taiwo, a O. Odunsi, Outcome of non-operative management of femoral shaft fractures in children., Afr. J. Paediatr. Surg. 8 (2011) 34–39. doi:10.4103/0189-6725.78666.

[27] J.M. Flynn, The Treatment of Low-Energy Femoral Shaft Fractures: A Prospective Study Comparing the “Walking Spica” with the Traditional Spica Cast, J. Bone Jt. Surg. 93 (2011) 2196. doi:10.2106/JBJS.J.01165.

[28] G.W. Wood, Intramedullary nailing of femoral and tibial shaft fractures, J. Orthop. Sci. 11 (2006) 657–669. doi:10.1007/s00776-006-1061-6.

[29] S. Hoppenfeld, P. DeBoer, R. Buckley, Surgical Exposures in Orthopaedics: The Anatomic Approach, 4th Edition, 2009.

117

[30] X. Duan, T. Li, A.Q. Mohammed, Z. Xiang, Reamed intramedullary nailing versus unreamed intramedullary nailing for shaft fracture of femur: A systematic literature review, Arch. Orthop. Trauma Surg. 131 (2011) 1445–1452. doi:10.1007/s00402-011- 1311-8.

[31] F. Högel, U.V. Gerlach, N.P. Südkamp, C.A. Müller, Pulmonary fat embolism after reamed and unreamed nailing of femoral fractures, Injury. 41 (2010) 1317–1322. doi:10.1016/j.injury.2010.08.019.

[32] P.R.T. Kuzyk, R. Li, R. Zdero, J.E. Davies, E.H. Schemitsch, The effect of intramedullary reaming on a diaphyseal bone defect of the tibia., J. Trauma. 70 (2011) 1248–56. doi:10.1097/TA.0b013e3181e985bd.

[33] P. Messmer, T. Gross, Femur - Shaft fractures - Intramedullary Nailing with the Antegrade Femoral Nail (AFN), AO Found. (n.d.). http://www.aovideo.ch/published/player.2.aspx?id=40097eem0198.

[34] M.S. Peter V Giannoudis, Hans Christoph Pape, Femur shaft 32-A3 CRIF midshaft, AO Found. (2007). https://www2.aofoundation.org/wps/portal/surgery/?showPage=redfix&bone=Femur&seg ment=Shaft&classification=32- A3&treatment=&method=CRIF+(closed+reduction+internal+fixation)&implantstype=Ant egrade+nailing+(midshaft/distal+shaft)&redfix_url=.

[35] J.C. Coetzee, E.J. Van Der Merwe, Exposure of surgeons-in-training to radiation during intramedullary fixation of femoral shaft fractures, South African Med. J. 81 (1992) 312– 314.

[36] E.J. Hazan, L. Joskowicz, Computer-Assisted Image-Guided Intramedullary Nailing of Femoral Shaft Fractures, Tech. Orthop. 18 (2003) 191–200. doi:10.1097/00013611- 200306000-00008.

[37] R.H. Clarke, J.W. Stather, Implementation of the 1990 recommendations of ICRP in the countries of the European Community, (1993) 151–161.

118

[38] S. Madan, C. Blakeway, Radiation exposure to surgeon and patient in intramedullary nailing of the lower limb, Injury. 33 (2002) 723–727. doi:10.1016/S0020-1383(02)00042- 6.

[39] L.P. Müller, J. Suffner, K. Wenda, W. Mohr, P.M. Rommens, Radiation exposure to the hands and the thyroid of the surgeon during intramedullary nailing, Injury. 29 (1998) 461– 468. doi:10.1016/S0020-1383(98)00088-6.

[40] B. Linke, C. Ansari Moein, O. Bösl, M.H.J. Verhofstad, C. van der Werken, K. Schwieger, K. Ito, Lateral insertion points in antegrade femoral nailing and their influence on femoral bone strains., J. Orthop. Trauma. 22 (2008) 716–722. doi:10.1097/BOT.0b013e318189369e.

[41] C.M. Ansari Moein, M.H.J. Verhofstad, R.L.A.W. Bleys, C. Van Der Werken, Soft tissue injury related to choice of entry point in antegrade femoral nailing: Piriform fossa or greater trochanter tip, Injury. 36 (2005) 1337–1342. doi:10.1016/j.injury.2004.07.052.

[42] J.R. Hsu, D.J. Stinner, D. a Brown, Splitting of the proximal femur with a new femoral nail., J. Orthop. Trauma. 24 (2010) e40–e43. doi:10.1097/BOT.0b013e3181a53790.

[43] C. Sfeir, L. Ho, B. Doll, A. K, J. Hollinger, Fracture Repair, Bone Regen. Repair Biol. Clin. Appl. (2005) 21–44. doi:10.1385/1-59259-863-3:021.

[44] K. Ito, S.M. Perren, Biology of fracture healing, in: AO Princ. Fract. Manag., 2013: pp. 1689–1699. doi:10.1017/CBO9781107415324.004.

[45] L. Claes, P. Augat, G. Suger, H. Wilke, Claes_Influence of Size and Stability of the Osteotomy Gap on the Success of Fracture Healing.pdf, (1997) 0–7.

[46] M. Panagiotis, Classification of non-union, Injury. 36 (2005) 30–37. doi:10.1016/j.injury.2005.10.008.

[47] W.M. Ricci, C. Bellabarba, B. Evanoff, D. Herscovici, T. DiPasquale, R. Sanders, Retrograde versus antegrade nailing of femoral shaft fractures., J. Orthop. Trauma. 15 (2001) 161–169. doi:10.1097/BOT.0b013e31816c0255.

119

[48] Canadian Orthopaedic Trauma Society, Nonunion following intramedullary nailing of the femur with and without reaming. Results of a multicenter randomized clinical trial., J. Bone Joint Surg. Am. 85-A (2003) 2093–6. http://www.ncbi.nlm.nih.gov/pubmed/14630836.

[49] C. Bellabarba, W.M. Ricci, B.R. Bolhofner, Results of indirect reduction and plating of femoral shaft nonunions after intramedullary nailing., J. Orthop. Trauma. 15 (2001) 254– 63. doi:10.1097/00005131-200105000-00004.

[50] O. Karaman, E. Ayhan, H. Kesmezacar, A. Seker, M.C. Unlu, O. Aydingoz, Rotational malalignment after closed intramedullary nailing of femoral shaft fractures and its influence on daily life, Eur. J. Orthop. Surg. Traumatol. 24 (2014) 1243–1247. doi:10.1007/s00590-013-1289-8.

[51] T. Hartwig, The Fixion Nail : New Kind of Intramedullary Nailing, (2002) 58791.

[52] H.J. Siegel, W. Sessions, M.A. Casillas Jr, Stabilization of pathologic long bone fractures with the fixion expandable nail, Orthopedics. 31 (2008) 143–148. http://www.embase.com/search/results?subaction=viewrecord&from=export&id=L35144 9620\nhttp://sfx.library.uu.nl/utrecht?sid=EMBASE&issn=01477447&id=doi:&atitle=Sta bilization+of+pathologic+long+bone+fractures+with+the+fixion+expandable+nail&stitle =Orthopedic.

[53] L. Lepore, S. Lepore, N. Maffulli, Intramedullary nailing of the femur with an inflatable self-locking nail: Comparison with locked nailing, J. Orthop. Sci. 8 (2003) 796–801. doi:10.1007/s00776-003-0709-8.

[54] Z.-T. Zhou, Y.-C. Song, X.-Z. Zhou, H.-B. Zhou, Z.-P. Luo, Q.-R. Dong, Femoral Midshaft Fractures: Expandable Versus Locked Nailing, Orthopedics. 38 (2015) e314– e318. doi:10.3928/01477447-20150402-60.

[55] Y. Palatnik, S.R. Rozbruch, Femoral reconstruction using external fixation., Adv. Orthop. 2011 (2011) 10. doi:10.4061/2011/967186.

[56] S. Kishan, S. Sabharwal, F. Behrens, M. Reilly, M. Sirkin, External Fixation of the Femur:

120

Basic Concepts, Tech. Orthop. 17 (2002) 239–244. doi:10.1097/00013611-200206000- 00013.

[57] M.S. Peter V Giannoudis, Hans Christoph Pape, Femur shaft 32-A1 Temporary external fixator (midshaft/distal shaft), (2007). https://www2.aofoundation.org/wps/portal/!ut/p/a1/jY9LC4MwEIR_jVd3tfRBbzm00gdIs Q_NpUQao2CzIcYK_fW1nu1jbjvMt8wAhxS4Fo9KCVeRFvX75rNrGCNGmwR3cXw MkCWrIDqvEfEw7QPZlwCb_8fjBzH8xW- Bq5ryoWrGdD5ZKOBWFtJK67e2t0vnTLP00MOu60JfUEGtvg0DfbKqt03joSHrxp- U1DhIR1kw91P63MvLC (accessed January 1, 2016).

[58] M.S. Peter V Giannoudis, Hans Christoph Pape, Femur shaft 32-C2 CRIF midshaft, AO Found. (2007). https://www2.aofoundation.org/wps/portal/surgery?showPage=redfix&bone=Femur&seg ment=Shaft&classification=32-C2&treatment=&method=CRIF (closed reduction internal fixation)&implantstype=Bridge plating (midshaft)&approach=&redfix_url=1285238484615&Language=e (accessed January 1, 2016).

[59] Y. Zenios, Makover, Biodesign, 2012. doi:10.1007/s13398-014-0173-7.2.

[60] E.J. Hazan, L. Joskowicz, Computer-Assisted Image-Guided Intramedullary Nailing of Femoral Shaft Fractures, Tech. Orthop. 18 (2003) 191–200. doi:10.1097/00013611- 200306000-00008.

[61] M.C. Crookshank, M.R. Edwards, M. Sellan, C.M. Whyne, E.H. Schemitsch, Can fluoroscopy-based computer navigation improve entry point selection for intramedullary nailing of femur fractures?, Clin. Orthop. Relat. Res. 472 (2014) 2720–2727. doi:10.1007/s11999-013-2878-x.

[62] P. Liebmann, S. Bohn, T. Neumuth, Design and validation of a robust surgical workflow management system, M2Cai2011. (2011).

[63] F. Lalys, P. Jannin, Surgical process modelling: A review, Int. J. Comput. Assist. Radiol. Surg. 9 (2014) 495–511. doi:10.1007/s11548-013-0940-5.

121

[64] S.-A. Ahmadi, T. Sielhorst, R. Stauder, M. Horn, H. Feussner, N. Navab, Recovery of surgical workflow without explicit models., Med. Image Comput. Comput. Assist. Interv. 9 (2006) 420–428. doi:10.1007/11866565_52.

[65] M. Raimbault, P. Jannin, X. Morandi, L. Riffaud, B. Gibaud, Models of surgical procedures for multimodal image-guided neurosurgery, Stud. Health Technol. Inform. 95 (2003) 50–55. doi:10.3233/978-1-60750-939-4-50.

[66] D. Neumuth, F. Loebe, H. Herre, T. Neumuth, Modeling surgical processes: A four-level translational approach, Artif. Intell. Med. 51 (2011) 147–161. doi:10.1016/j.artmed.2010.12.003.

[67] T. Neumuth, P. Jannin, J. Schlomberg, J. Meixensberger, P. Wiedemann, O. Burgert, Analysis of surgical intervention populations using generic surgical process models, Int. J. Comput. Assist. Radiol. Surg. 6 (2011) 59–71. doi:10.1007/s11548-010-0475-y.

[68] L. Wang, J. Landes, S. Weidert, T. Blum, A. Von Der Heide, E. Euler, N. Navab, First animal cadaver study for interlocking of intramedullary nails under camera augmented mobile C-arm: A surgical workflow based preclinical evaluation, Lect. Notes Comput. Sci. (Including Subser. Lect. Notes Artif. Intell. Lect. Notes Bioinformatics). 6135 LNCS (2010) 56–66. doi:10.1007/978-3-642-13711-2_6.

[69] L. Riffaud, T. Neumuth, X. Morandi, C. Trantakis, J. Meixensberger, O. Burgert, B. Trelhu, P. Jannin, Recording of surgical processes: A study comparing senior and junior neurosurgeons during lumbar disc herniation surgery, Neurosurgery. 67 (2010). doi:10.1227/NEU.0b013e3181f741d7.

[70] G. Forestier, F. Lalys, L. Riffaud, B. Trelhu, P. Jannin, Classification of surgical processes using dynamic time warping, J. Biomed. Inform. 45 (2012) 255–264. doi:10.1016/j.jbi.2011.11.002.

[71] J. a. Casaletto, V. Rajaratnam, Surgical Process Re-Engineering: Carpal Tunnel Decompression — a Model, Hand Surg. 09 (2004) 19–27. doi:10.1142/S0218810404002066.

122

[72] M. Sandelowski, Focus on Qualitative Methods Sample Size in Qualitative, Res. Nurs. Heal. 18 (1995) 179–183. doi:10.1002/nur.4770180211.

[73] I.T. Coyne, Sampling in qualitative research. Purposeful and theoretical sampling; merging or clear boundaries?, J. Adv. Nurs. 26 (1997) 623–630. doi:http://dx.doi.org/10.1046/j.1365-2648.1997.t01-25-00999.x.

[74] G. Miller, E. Galanter, K. Pribram, Plans and the structure of behavior, 1960.

[75] L.D. Anderson, MANUAL OF INTERNAL FIXATION . TECHNIQUES RECOMMENDED BY THE AO-ASIF GROUP . ED . 3 ., Surgery. (2010).

[76] J.W. Roberts, L.A. Libet, P.R. Wolinsky, Who is in danger? Impingement and penetration of the anterior cortex of the distal femur during intramedullary nailing of proximal femur fractures, J. Trauma Acute Care Surg. 73 (2012) 249–254. doi:10.1097/TA.0b013e318256a0b6.

[77] A.J. Kanawati, B. Jang, R. McGee, J. Sungaran, The influence of entry point and radius of curvature on femoral intramedullary nail position in the distal femur, J. Orthop. 11 (2014) 68–71. doi:10.1016/j.jor.2014.04.010.

[78] M.S. Peter V Giannoudis, Hans Christoph Pape, Femur Shaft, (2007). https://www2.aofoundation.org/wps/portal/surgery?showPage=approach&contentUrl=srg/ 32/04-Approaches/32-A10- ante_nail_piri.jsp&bone=Femur&segment=Shaft&approach=Antegrade nailing %28piriformis%29&Language=en.

[79] Y.A. Weil, M. Liebergall, R. Mosheiff, D.L. Helfet, A.D. Pearle, Long bone fracture reduction using a fluoroscopy-based navigation system : A feasibility and accuracy study, 12 (2007) 295–302. doi:10.1080/10929080701657974.

[80] B. Diotte, P. Fallavollita, L. Wang, S. Weidert, E. Euler, P. Thaller, Multi-Modal Intra- Operative Navigation During Distal Locking of Intramedullary Nails, 34 (2015) 487–495.

[81] R. Westphal, S. Winkelbach, F. Wahl, T. Gosling, M. Oszwald, T. Hufner, C. Krettek, Robot-assisted Long Bone Fracture Reduction, Int. J. Rob. Res. 28 (2009) 1259–1278.

123

doi:10.1177/0278364909101189.

[82] H. Ebrahimi, A. Yee, C. Whyne, Surgical process analysis identifies lack of connectivity between sequential fluoroscopic 2D alignment as a critical impediment in femoral intramedullary nailing, Int. J. Comput. Assist. Radiol. Surg. 11 (2016) 297–305. doi:10.1007/s11548-015-1262-6.

[83] W.M. Ricci, J. Schwappach, M. Tucker, K. Coupe, A. Brandt, R. Sanders, R. Leighton, Trochanteric versus piriformis entry portal for the treatment of femoral shaft fractures., J. Orthop. Trauma. 20 (2006) 663–7. doi:10.1097/01.bot.0000248472.53154.14.

[84] T. Gausepohl, D. Pennig, J. Koebke, S. Harnoss, Antegrade femoral nailing: An anatomical determination of the correct entry point, Injury. 33 (2002) 701–705. doi:10.1016/S0020-1383(02)00158-4.

[85] K.D. Johnson, A.F. Tencer, M.C. Sherman, Biomechanical factors affecting fracture stability and femoral bursting in closed intramedullary nailing of femoral shaft fractures, with illustrative case presentations., J. Orthop. Trauma. 1 (1987) 1–11. http://eutils.ncbi.nlm.nih.gov/entrez/eutils/elink.fcgi?dbfrom=pubmed&id=3506582&retm ode=ref&cmd=prlinks\npapers2://publication/uuid/1AE94E31-CE01-4D9C-B3FD- C633E6F22594.

[86] M.C. Harper, W.L. Carson, Curvature of the femur and the proximal entry point for an intramedullary rod., Clin. Orthop. Relat. Res. (1987) 155–161. http://ovidsp.ovid.com/ovidweb.cgi?T=JS&PAGE=reference&D=med2&NEWS=N&AN =3594986.

[87] R.F. Ostrum, A. Marcantonio, R. Marburger, A critical analysis of the eccentric starting point for trochanteric intramedullary femoral nailing., J. Orthop. Trauma. 19 (2005) 681– 686. doi:10.1097/01.bot.0000184145.75201.1b.

[88] J. Castellanos, L. Garcia-Nuño, J.M. Cavanilles-Walker, J. Roca, Iatrogenic femoral neck fracture during closed nailing of the femoral shaft fracture, Eur. J. Trauma Emerg. Surg. 35 (2009) 479–481. doi:10.1007/s00068-009-8151-z.

124

[89] J. Christie, C. Court-Brown, A.W. Kinninmonth, C.R. Howie, Intramedullary locking nails in the management of femoral shaft fractures, J. Bone Jt. Surg. - Br. Vol. 70 (1988) 206–210.

[90] P. Weinrauch, Ultrasound assisted trochanteric entry point development for antegrade femoral nailing procedures, (2015) 7–10.

[91] S.D. Miller, B. Burkart, E. Damson, N. Shrive, R.C. Bray, The effect of the entry hole for an intramedullary nail on the strength of the proximal femur., J. Bone Joint Surg. Br. 75 (1993) 202–206.

[92] R. Hofstetter, M. Slomczykowski, C. Krettek, G. K??ppen, M. Sati, L.P. Nolte, Computer- assisted fluoroscopy-based reduction of femoral fractures and antetorsion correction, Comput. Aided Surg. 5 (2000) 311–325. doi:10.1002/1097-0150(2000)5:5<311::AID- IGS1>3.0.CO;2-J.

[93] T. Gosling, R. Westphal, T. Hufner, J. Faulstich, M. Kfuri, F. Wahl, C. Krettek, Robot- assisted fracture reduction: A preliminary study in the femur shaft, Med. Biol. Eng. Comput. 43 (2005) 115–120. doi:10.1007/BF02345131.

[94] T. Gösling, M. Oszwald, D. Kendoff, M. Citak, C. Krettek, T. Hufner, Computer-assisted antetorsion control prevents malrotation in femoral nailing: An experimental study and preliminary clinical case series, Arch. Orthop. Trauma Surg. 129 (2009) 1521–1526. doi:10.1007/s00402-009-0871-3.

[95] D. Kendoff, M. Citak, M.J. Gardner, T. Gösling, C. Krettek, T. Hüfner, Navigated femoral nailing using noninvasive registration of the contralateral intact femur to restore anteversion. Technique and clinical use., J. Orthop. Trauma. 21 (2007) 725–30. doi:10.1097/BOT.0b013e31812f69a9.

[96] R. Kaye, J. Crowley, Medical Device Use Safety: Incorporating Human Factors Engineering into Risk Management, 1999.

[97] M.E. Wiklund, J. Kendler, A.S. Yale, Usability Testing of Medical Devices, CRC Press, 2010.

125

[98] J. Beckmann, J. Goetz, L. Perlick, C. Luring, J. Grifka, M. Tingart, Fluoroscopically based drilling of the femoral head in a model of adiposis, Comput. Aided Surg. Off. J. Int. Soc. Comput. Aided Surg. 13 (2008) 55–61. doi:790244645 [pii]\n10.3109/10929080701882564.

[99] F.A. Khan, M.A. Ikram, A.A. Badr, H. Al-Khawashki, Femoral neck fracture: a complication of femoral nailing, Injury. 26 (1995) 319–321. doi:10.1016/0020- 1383(95)00049-F.

[100] C. Dora, M. Leunig, M. Beck, D. Rothenfluh, R. Ganz, Entry point soft tissue damage in antegrade femoral nailing: a cadaver study., J. Orthop. Trauma. 15 (2001) 488–493. doi:10.1097/00005131-200109000-00005.

[101] C.M. Moein Ansari, H.J. ten Duis, P.L. Oey, G.A.P. de Kort, W. van der Meulen, C. van der Werken, Intramedullary femoral nailing through the trochanteric fossa versus greater trochanter tip: A randomized controlled study with in-depth functional outcome results, Eur. J. Trauma Emerg. Surg. 37 (2011) 615–622. doi:10.1007/s00068-011-0086-5.

[102] C. Ansari Moein, H.-J. ten Duis, L. Oey, G. de Kort, W. van der Meulen, K. Vermeulen, C. van der Werken, Functional outcome after antegrade femoral nailing: a comparison of trochanteric fossa versus tip of greater trochanter entry point., J. Orthop. Trauma. 25 (2011) 196–201. doi:10.1097/BOT.0b013e3181eaa049.

[103] A. Duclos, J.L. Peix, C. Colin, J.L. Kraimps, F. Menegaux, F. Pattou, F. Sebag, S. Touzet, S. Bourdy, N. Voirin, J.C. Lifante, Influence of experience on performance of individual surgeons in thyroid surgery: Prospective cross sectional multicentre study, Clin. Otolaryngol. 37 (2012) 146. doi:10.1111/j.1749-4486.2012.02457.x.

[104] M. Oberst, P. Niemeyer, N. Suedkamp, Endoscopic controlled intramedullary fracture reduction. A case report about closed fracture reduction under visual control, Arch. Orthop. Trauma Surg. 127 (2007) 647–649. doi:10.1007/s00402-007-0289-8.

[105] H. Du, L. Hu, C. Li, C. He, L. Zhang, P. Tang, Preoperative trajectory planning for closed reduction of long-bone diaphyseal fracture using a computer-assisted reduction system, Int. J. Med. Robot. Comput. Assist. Surg. 11 (2015) 58–66. doi:10.1002/rcs.1573.

126

[106] J.T. Riehl, J.C. Widmaier, Techniques of obtaining and maintaining reduction during nailing of femur fractures., Orthopedics. 32 (2009) 581. doi:10.3928/01477447-20090624- 17.

[107] B. Mahaisavariya, P. Songcharoen, K. Riansuwan, Technique of closed unlocked femoral nailing using ultrasound guidance, Injury. 37 (2006) 1000–1003. doi:10.1016/j.injury.2006.01.002.

[108] N. Hawi, E.M. Suero, E. Liodakis, S. Decker, C. Krettek, M. Citak, Intra-operative assessment of femoral antetorsion using ISO-C 3D: A cadaver study, Injury. 45 (2014) 506–509. doi:10.1016/j.injury.2013.10.034.

[109] M. Citak, D. Kendoff, A.D. Pearle, P.F. O'Loughlin, C. Krettek, T. Hüfner, M. Citak, Navigated femoral anteversion measurements: General precision and registration options, Arch. Orthop. Trauma Surg. 129 (2009) 671–677. doi:10.1007/s00402-008-0804- 6.

[110] R. Mosheiff, Y. Weil, E. Peleg, M. Liebergall, Computerised navigation for closed reduction during femoral intramedullary nailing, Injury. 36 (2005) 866–870. doi:10.1016/j.injury.2004.12.036.

[111] J. Frank, B. Gritzbach, C. Winter, B. Maier, I. Marzi, Computer-assisted Femur Fracture Reduction, Eur. J. Trauma Emerg. Surg. 36 (2010) 151–156. doi:10.1007/s00068-009- 9155-4.

[112] S.G. Keller, A.P. Gordon, The Effect of Uniform Flexure Loading on Helixes, (2008) 1–6.

[113] S.G. Keller, a. P. Gordon, Equivalent stress and strain distribution in helical compression springs subjected to bending, J. Strain Anal. Eng. Des. 46 (2011) 405–415. doi:10.1177/0309324711410128.

[114] E.B. Wolansky, Lateral Deflection of Free-End Springs, Spring Manuf. Inst. (2005) 33– 34.

[115] E.B. Wolansky, Bending of helical springs, Spring Manuf. Inst. (2000) 76–80.

127

[116] D.K. Rex, M. Khashab, G.S. Raju, J. Pasricha, R. Kozarek, Insertability and safety of a shape-locking device for colonoscopy, Am. J. Gastroenterol. 100 (2005) 817–820. doi:10.1111/j.1572-0241.2005.40801.x.

[117] A.J. Loeve, D.H. Plettenburg, P. Breedveld, J. Dankelman, Endoscope shaft-rigidity control mechanism: “FORGUIDE”., IEEE Trans. Biomed. Eng. 59 (2012) 542–51. doi:10.1109/TBME.2011.2175730.

[118] T. Adebar, J. Greer, P. Laeseke, G. Hwang, A. Okamura, Methods for Improving the Curvature of Steerable Needles in Biological Tissue, IEEE Trans. Biomed. Eng. 9294 (2015) 1–1. doi:10.1109/TBME.2015.2484262.

[119] Y. Chen, S. Tanaka, I.W. Hunter, Disposable endoscope tip actuation design and robotic platform, 2010 Annu. Int. Conf. IEEE Eng. Med. Biol. Soc. EMBC’10. (2010) 2279– 2282. doi:10.1109/IEMBS.2010.5627677.

[120] C. Krettek, Percutaneous Reduction Techniques, (2016) 1–30.

[121] A.N. Hopper, M.H. Jamison, W.G. Lewis, Learning curves in surgical practice, Postgrad. Med. J. 83 (2007) 777–779. doi:10.1136/pgmj.2007.057190.

[122] H.W.R. Schreuder, R. Wolswijk, R.P. Zweemer, M.P. Schijven, R.H.M. Verheijen, Training and learning robotic surgery, time for a more structured approach: A systematic review, BJOG An Int. J. Obstet. Gynaecol. 119 (2012) 137–149. doi:10.1111/j.1471- 0528.2011.03139.x.

[123] K. a Egol, E.Y. Chang, J. Cvitkovic, F.J. Kummer, K.J. Koval, Mismatch of current intramedullary nails with the anterior bow of the femur., J. Orthop. Trauma. 18 (2004) 410–415. doi:10.1097/00005131-200502000-00014.

[124] T. Gösling, R. Westphal, J. Faülstich, K. Sommer, F. Wahl, C. Krettek, T. Hufner, Forces and torques during fracture reduction: Intraoperative measurements in the femur, J. Orthop. Res. 24 (2006) 333–338. doi:10.1002/jor.20045.

128

Appendices: Patent Submission

POSITIONING AND ALIGNMENT INSTRUMENT FOR INTRODUCING

SURGICAL DEVICES INTO BONE

BACKGROUND

5 Intramedullary (IM) nailing is a minimally invasive surgical procedure

typically performed under a general anesthetic and with fluoroscopic image

guidance. The surgical phases associated with this type of surgery are well

defined: patient preparation, access to the bone entry site, IM guide wire

insertion (including fracture reduction for guide wire placement), IM nail

10 placement, locking of the nail to control rotation and length, final clinical and

radiographic assessment of fracture reduction/restoration of

length/alignment/rotation, and surgical wound closure. Yet, despite widespread

usage of IM nailing, significant surgical challenges may arise. Such challenges

can significantly impede the surgical workflow, requiring additional operative time

15 and radiation exposure to both patients and medical staff. Fracture reduction and

proper localization for initial IM access are particularly challenging areas in the

workflow pathway. Lengthy delays in the procedure and unacceptable fracture

reduction or stabilization can also significantly endanger patient safety,

particularly in those who may suffer from polytrauma and/or acute respiratory

20 issues.

SUMMARY

A positioning and alignment instrument, and methods of use thereof, are

129

provided for facilitating the alignment and insertion of a device, such as a guide

wire, into bone. The instrument includes a handheld anchoring component and a

rotatable guidance component. During use, the anchoring component is

anchored into bone via anchoring protrusions, such that the position and

5 orientation of the anchoring component is fixed relative to the bone. The

guidance component, which is mechanically supported by the anchoring

component, includes a device guide channel for receiving the device and guiding

the device towards an insertion location adjacent to the distal end of the

anchoring component. The guidance component is rotatable relative to the

10 anchoring component about a rotation axis that is located adjacent to the distal

end of the anchoring component, such that the insertion location remains

adjacent to the distal end of the anchoring component under rotation.

Accordingly, in one aspect, there is provided a positioning and alignment

instrument for guiding insertion of a device into bone, the positioning and

15 alignment instrument comprising:

an anchoring component comprising a proximal portion and a distal

portion, wherein said proximal portion comprises a handle, and wherein one or

more anchoring protrusions extend from a distal end of said distal portion for

anchoring said anchoring component into the bone, such that a position and an

20 orientation of said anchoring component is fixed relative to the bone when said

anchoring component is anchored to the bone; and

a guidance component mechanically supported by said anchoring

component, said guidance component comprising a device guide channel for

130

receiving the device and guiding the device towards an insertion location

adjacent to the distal end of said anchoring component;

wherein said guidance component is rotatable relative to said anchoring

component about a rotation axis that is located adjacent to the distal end of said

5 anchoring component, such that the insertion location remains adjacent to the

distal end of said anchoring component under rotation of said guidance

component.

In another aspect, there is provided a method of employing fluoroscopy to

aligning a device during a medical procedure, the method comprising, after

10 having employed fluoroscopy, in a first direction, to anchor the positioning and 10

alignment instrument into bone:

obtaining fluoroscopy images of the positioning and alignment instrument

in a perpendicular direction; and

rotating the guidance component to a desired angle according to the

15 fluoroscopy images; and

thereby aligning the device guide channel for subsequent guidance and

insertion of the device into the bone.

A further understanding of the functional and advantageous aspects of the

disclosure can be realized by reference to the following detailed description and

20 drawings.

BRIEF DESCRIPTION OF THE DRAWINGS

Embodiments will now be described, by way of example only, with

131

reference to the drawings, in which:

FIGS. 1A and 1B show views of an example positioning and alignment

instrument.

FIG. 1C shows a detailed view of the distal region of the positioning and

5 alignment instrument.

FIG. 1D illustrates an alternative embodiment in which the guidance

component is rotatably supported by the anchoring component such that the

rotation axis lies distalward relative to the distal end 118 of the anchoring

component.

10 FIG. 1E shows a detailed assembly view of the rail that is provided in the

proximal portion of the guidance component.

FIG. 1F shows an alternative implementation of the orientation of the

handle.

FIG. 1G shows a detailed view of the distal region of the anchoring

15 component, illustrating an example embodiment in which the position of the

rotation pin is adjustable relative to the anchoring component.

FIGS. 2A-2C show the use of an example force coupling tool for applying

an impact force to drive the anchoring protrusions of the anchoring component

into bone.

20 FIGS. 3A-3E illustrate the steps and challenges in conventional Kirschner

wire (K-wire) positioning and alignment. FIG. 3A shows an initial anterior-

posterior (AP) image with a correct entry point location and orientation.

Perpendicular lateral images through the sagittal plane show the following: FIG.

132

3B: correct location, incorrect orientation, requiring AP rotation; FIG. 3C:

incorrect location, correct orientation, requiring AP translation; and FIG. 3D:

incorrect location and orientation of the entry point, requiring AP rotation and

translation to obtain correct lateral entry point location and orientation, shown in

5 FIG. 3E.

FIG. 4 is a flow chart illustrating an example method in which a position

and alignment instrument is employed a medical procedure involving the

introduction of a device into bone.

FIGS. 5A-5C show fluoroscopy images of the positioning and alignment

10 instrument during various steps of the method illustrated in FIG. 4.

FIG. 5D shows the two-dimensional array of device guide channels visible

at the proximal end of the rotatable guidance component.

FIG. 6 shows a fluoroscopy image showing the introduction of a Kirschner

wire into bone following the positioning and alignment of the Kirschner wire using

15 an example positioning and alignment instrument.

FIG. 7 shows the use of an example positioning and alignment instrument

during the insertion of a locking screw into the distal end of an intramedullary rod

during an intramedullary nailing procedure.

FIG. 8 shows an expanded view relative to FIG. 7, showing the orientation

20 of the C-arm relative to the example positioning and alignment instrument.

FIG. 9 shows an example alternative configuration of an anchoring

protrusion.

133

DETAILED DESCRIPTION

Various embodiments and aspects of the disclosure will be described with

reference to details discussed below. The following description and drawings are

illustrative of the disclosure and are not to be construed as limiting the disclosure.

5 Numerous specific details are described to provide a thorough understanding of

various embodiments of the present disclosure. However, in certain instances,

well-known or conventional details are not described in order to provide a

concise discussion of embodiments of the present disclosure.

As used herein, the terms “comprises” and “comprising” are to be

10 construed as being inclusive and open ended, and not exclusive. Specifically,

when used in the specification and claims, the terms “comprises” and

“comprising” and variations thereof mean the specified features, steps or

components are included. These terms are not to be interpreted to exclude the

presence of other features, steps or components.

15 As used herein, the term “exemplary” means “serving as an example,

instance, or illustration,” and should not be construed as preferred or

advantageous over other configurations disclosed herein.

As used herein, the terms “about” and “approximately” are meant to cover

variations that may exist in the upper and lower limits of the ranges of values,

20 such as variations in properties, parameters, and dimensions. Unless otherwise

specified, the terms “about” and “approximately” mean plus or minus 25 percent

or less.

It is to be understood that unless otherwise specified, any specified range

134

or group is as a shorthand way of referring to each and every member of a range

or group individually, as well as each and every possible sub-range or sub -group

encompassed therein and similarly with respect to any sub-ranges or sub-groups

therein. Unless otherwise specified, the present disclosure relates to and

5 explicitly incorporates each and every specific member and combination of sub-

ranges or sub-groups.

As used herein, the term "on the order of", when used in conjunction with

a quantity or parameter, refers to a range spanning approximately one tenth to

ten times the stated quantity or parameter.

10 Referring now to FIG. 1A, an example positioning and alignment

instrument 100 is shown, where the instrument 100 is configured for facilitating

the alignment and insertion of a device, such as a guide wire (not shown), into

bone. The instrument 100 includes a handheld anchoring component 110 and a

rotatable guidance component 120. During use, the anchoring component 110 is

15 anchored into bone via anchoring protrusions 115, such that the position and

orientation of the anchoring component 110 is fixed relative to the bone. The

distal surface 118 of the anchoring component 110, from which the anchoring

protrusions 115 extend, contacts the bone upon insertion of the anchoring

component 110 into the bone, thereby establishing a reference location at the

20 bone surface. In the example embodiment shown in FIGS. 1A and 1B, the distal

end 118 of the anchoring component 110 has a rectangular cross-section.

In the present example embodiment, the rotatable component 120 is a

multi-cannulated arm that includes a plurality of device guide channels 122,

135

where each device guide channel has a diameter suitable for receiving and

guiding a device (e.g. a guide wire) as the device is advanced towards the bone.

As shown in FIGS. 1A and 1B, each device guide channel 122 extends, through

the rotatable guidance component 120, from a proximal end 124 to a distal end

5 126. As described in detail below, the optional inclusion of a plurality of device

channel guides 122, arrayed in one or two dimensions, provides a surgeon with

the ability to select a suitable device guide channel for device insertion in order to

obtain a desired entry location relative to the anchoring location (the location in

the bone that is adjacent to the distal end of the anchoring component 110). The

10 diameters of the various device guide channels 122 may be identical or different,

and may vary in terms of diameter, pattern and number.

For applications involving the use of fluoroscopy, such as those described

below, at least a distal region of the rotatable guidance component 120 may be

radiolucent, such that the device guide channels 122 and/or a device inserted

15 within a given device guide channel 122 is observable. In such applications, it

may be preferable for the distal ends of both the rotatable guidance component

120 and the anchoring component 110 to be radiolucent. For example, the distal

region of the anchoring component 110 may include a radiolucent segment

110A. One of both of the rotatable guidance component 120 and the anchoring

20 component 110 may include a radiopaque material in order to identify one or

more features or locations of the positioning and alignment instrument.

The rotatable guidance component 120 is rotatable relative to the

anchoring component 110, in order to vary the angular orientation of device

136

channel guides 122 relative to the anchoring location, thereby enabling the

selection of a suitable orientation for insertion (entry) of the device into the bone

(e.g. a suitable angular orientation relative to the bone, or to internal anatomical

features or structures, or to an internal medical device). As shown in FIGS. 1A

5 and 1B, the rotatable guidance component 120 rotates about a rotation axis 130

that is located adjacent to the distal end 118 of the anchoring component 110. As

used herein, the phrase “adjacent to the distal end 118 of the anchoring

component” refers to a location that is at, or is proximal to, the distal end 118 of

the anchoring component 110, such as within 5 mm, within 4 mm, within 3 mm,

10 within 2 mm, or within 1 mm, of the distal end of the anchoring component. It will

be understood that a suitable maximum offset of the rotation axis 130 relative to

the distal end 118 of the anchoring component 110 may depend on clinical

application. For example, in the case of the insertion of a guidewire (e.g. a

Kirschner wire) during an intramedullary nailing procedure, a suitable maximum

15 offset may be 10 mm.

By rotatably securing the rotatable guidance component 120 to the

anchoring component 110 such that the rotation axis 130 lies adjacent to the

distal end 118 of the anchoring component 110, the entry location of the device

remains adjacent to the distal end of the anchoring component 110 over a wide

20 range of rotation angles of the rotatable component 120 about the rotation axis

130. This aspect of the rotatable guidance 120 component differs significant from

known devices in which an external rotatable component is rotatable about a

rotation axis that is configured to lie within the patient anatomy, at a location

137

corresponding to an internal anatomical feature.

In one example implementation, the rotation axis 130 may be located at

the distal end 118 of the anchoring component 110, such that the rotation axis

130 lies within the plane of the distal surface 118. In another example

5 embodiment, the rotation axis 130 may be located at a location that is proximal to

the distal end 118 of the rotatable guidance component 120 (such that the

rotation axis 130 passes through the distal region of the anchoring component.

For example, this may be achieved by providing, at a location that is adjacent to

the distal end of the anchoring component 118, a rotation pin 301 about which

10 the rotatable guidance component 120 is confined to rotate.

An example of such a configuration is shown in FIG. 1C, where a rotation

pin 301 is housed inside the anchoring component 110 to guide the rotatable

guidance component 120 via a connecting plate 302. The connecting plate 302 is

attached to the rotatable guidance component 120 and the anchoring component

15 110. The connection plate 302 is secured to the anchoring component 110 using

a screw 303 and rigidly attached to the rotatable guidance component 120 with

two or more screws 304.

In yet another example embodiment, the anchoring component 110 and

the rotatable guidance component 120 may be configured such that the rotation

20 axis 130 lies distalward relative to the distal end 118 of the anchoring component

110. For example, such a configuration is shown in FIG. 1D, where a supporting

mechanism 305 is attached to the rotatable guidance component 120 and the

anchoring component 110.

138

The positioning and alignment instrument 100 may include a rotation

actuation mechanism for actuating rotation of the rotatable guidance component

120. In the example implementation shown in FIGS. 1A and 1B, the rotation

actuation mechanism is located proximal to a handle 135, where the handle

5 forms or is attached or otherwise connected to a proximal portion of the

anchoring component 110.

The rotation actuation mechanism may be positioned such that it is

suitable for single-handed actuation by a user while holding the handle with a

single hand. FIGS. 1A and 1B illustrate an example rotation actuation

10 mechanism that includes a knob 150 that engages with a set screw 152 that is

provided in a rotatable connection shaft 154, such that when the knob 150 is

rotated, the connection shaft 154 is rotated.

The rotation of the knob 150 produces corresponding rotation of the

connection shaft 154, which in turn rotates linkage shaft 156, to which the

15 connection shaft 154 is connected. The linkage shaft 156 extends in a distalward

direction relative to the connection shaft 154, and a connection pin 158 extends

from the linkage shaft 156 at or near its distal end. The connection pin 158 is

received and secured with a screw 306 within a rail 160 provided in a proximal

portion of the rotatable guidance component 120 as shown in assembly view in

20 FIG. 1E. The rotatable guidance component 120, which is confined to rotate

about the rotation axis 130 as described above and illustrated in FIG. 1C, thus

rotates about the rotation axis 130 when the knob 150 is rotated, via the coupling

of the rotational motion through the connection shaft 154, the linkage shaft 156,

139

and the confined travel of the connection pin 158 within the rail 160. The proximal

location of the knob 150 relative to the handle enables single-handed rotational

actuation.

In some example embodiments, the rotatable guidance component 120

5 may be optionally and removably locked in a fixed angular configuration by a 5

rotation locking mechanism, such that the rotatable guidance component 120 is

only rotatable when the rotation locking mechanism is actuated by the operator.

An example rotation locking mechanism is illustrated in FIGS. 1A and 1B. When

the locking trigger 170 is not actuated by the operator, a first gear 162 formed in

10 or attached to the knob 150 is biased, by a spring 164 and a set screw 166, to

engage with a second gear 166 that is fixed relative to the anchoring component

110. When the operator applies a force to the trigger 170, the first gear 162 is

disengaged from contact with the second gear, thus permitting rotation of the

knob 150. This example locking mechanism is capable of single-handed

15 actuation.

It will be understood that the rotation mechanism and the rotation

actuation mechanism that are illustrated in FIGS. 1A and 1B are provided to

illustrate merely one example implementation, and that a wide variety of

alternative rotation mechanisms and rotation actuation mechanisms may be

20 employed without departing from the intended scope of the present disclosure.

Examples of alternative rotation mechanisms and configurations include,

but are not limited to a mechanism where the rotating knob 150 is placed

perpendicular to its current configuration and a gear mechanism is used to

140

transmit the rotational movement.

Examples of alternative rotation actuation mechanisms and configurations

include, but are not limited to, (i) a motorized mechanism where a push of a

button turns the motor in clockwise and/or counterclockwise directions which in

5 turn rotates the connection shaft 154, and (ii) a direct rotation of the rotational

guidance component 120.

Furthermore, examples of alternative rotation locking mechanisms include,

but are not limited to, (i) a ratcheting mechanism where the movement of the

rotatable knob 150 is constrained in clockwise and/or counterclockwise

10 directions, and (ii) a magnetic mechanism where the attraction of opposite

magnetic poles restricts the rotational movement of the connecting shaft 154.

FIG. 1F shows an alternative implementation of the anchoring component,

where the handle, rotation knob, and locking mechanism are positioned in a

different configuration than in the previous embodiments.

In some example embodiments, the guidance and alignment instrument 15

may be further configured to permit one or more additional degrees of freedom in

the alignment of the guidance component relative to the anchoring component.

For example, as illustrated in FIG. 1G, the position of the rotation pin 301 may be

adjustable relative to the anchoring component 110. Such an embodiment may

20 permit lateral adjustment of the location of the rotation axis once the anchoring

component 110 is anchored into bone. In other example embodiments, the

rotation pin 301 may be adjustable in other directions, such as along the

longitudinal direction of the anchoring component 110.

141

In order to anchor the anchoring component 110 into the bone, a force

must be applied with sufficient magnitude in order to cause the anchoring

protrusions to penetrate the bone surface and become embedded in the bone.

This may be achieved by applying a force, such as with a hammer or other

5 suitable tool, to a proximal surface of the anchoring component 110. In some

example embodiments, a suitable proximal surface may be accessible regardless

of the orientation of the rotational guidance component 120.

However, in the example implementation shown in FIGS. 1A and 1B, the

proximal surface of the anchoring component 110 is substantially occluded by

10 the rotational guidance component 120 when the rotational guidance component

120 is aligned with the anchoring component 110 (this aligned configuration is

shown in FIG. 1B). Therefore, it may be necessary to rotate the rotatable

guidance component 120 to a sufficiently large angle in order to provide access

to a suitable proximal surface for delivering an impact force thereto. An example

15 of such an orientation is shown in FIG. 1A, where the rotatable guidance

component is rotated to a sufficiently large angle to provide access to proximal

surfaces 172 and 174. This proximal region may be configured to present a

single surface that is suitable for receiving an impact force from a tool such as a

hammer.

20 In another example implementation, a force coupling tool may be

employed to temporarily and removably contact the anchoring component 110,

such that a force applied to a proximal surface of the force coupling tool is

coupled to the anchoring component 110 via contact therewith. An example of

142

such an embodiment is shown in FIGS. 2A-2C.

FIG. 2A shows an example embodiment in which the positioning and

alignment instrument is contacted with a force coupling tool 200. The example

force coupling tool slides overtop of the proximal portion of the anchoring

5 component 110 (and, in this example case, the proximal portion of the rotatable

guidance component 120). A distal surface 210 of the force coupling tool 200

abuts against the proximal surface 172 of the anchoring component 110 (a

similar abutment occurs on the opposite side of the device with proximal surface

174). This abutment provides contact such the application of a force (e.g. an

10 impact force) to the proximal surface 205 of the force coupling tool 200 is

coupled (e.g. communicated; transferred) to the anchoring component 110, and

is thereby also transferred to the anchoring protrusions 115. The proximal

surfaces 172 and 174 (or a single proximal surface) of the anchoring component

may be established by a separate component that is attached to the anchoring

15 component 110, or may be integrally formed as surfaces of the anchoring

component 110.

FIG. 2B shows the positioning and alignment instrument, having the force

coupling tool 200 provided thereon (e.g. contacted therewith or attached thereto),

with the anchoring protrusions 115 positioned at a suitable location for insertion

20 into the bone 10. Upon the delivery of a suitable force to the proximal surface

205 of the force coupling tool 200, as shown in FIG. 2C, the anchoring

protrusions 118 are delivered into and anchored within the bone 10, such that the

distal surface 118 of the anchoring component 110 lies adjacent to the bone

143

surface. The configuration shown in the present example embodiment, in which

the force coupling tool slides over proximal portions of both the anchoring

component 110 and the rotational guidance component 120, may be beneficial in

preventing rotational movement of the rotational guidance component during

5 anchoring and protecting various components (such as the rotational guidance

component 120) from impact or the application of undue stress.

As shown in FIGS. 1A, 1B, and in particular, FIG. 2C, the positioning and

alignment instrument may be shaped such that a proximal portion thereof

deflects, bends, curves, or otherwise angles outwardly relative to a longitudinal

10 axis (a device guidance axis) associated with a distal region of the rotatable

guidance component 120. This geometrical configuration may be useful,

beneficial or important in providing the surgeon suitable ergonomic manipulation

of the positioning and alignment device during use. For example, when the

position and alignment instrument shown in FIG. 2C is employed during an

15 intramedullary nailing procedure, the outward deflection of the proximal region of

the instrument enables the surgeon to apply an impact force to a proximal region

of the instrument without risk, or with reduced risk, of contacting the patient. The

deflected configuration also positions the handheld portion of the device further

away from the subject than in a collinear (straight) configuration, which may

20 provide for a more efficient and safe surgical procedure. In some example

embodiments, at least a portion of the positioning and alignment instrument is

deflected (angled) outwardly relative to the distal longitudinal axis by an angle

between 0 and 45 degrees. It will be understood that the shape of the instrument

144

may vary depending on clinical application and/or anatomical side of the subject.

For example, the instrument may have different shapes for operating on a left or

right limb (e.g. for optimal positioning of the handle).

Although the example positioning and alignment instruments described

5 herein may be employed for a wide variety of applications and medical

procedures, an example method of employing such an instrument during an

intramedullary nailing procedure is described below. It will be understood that the

methods below are merely provided as being illustrative of the application of the

example positioning and alignment instrument embodiments disclosed herein.

10 FIGS. 3A-3E illustrate the cumbersome and iterative nature of a

conventional intramedullary (IM) nailing procedure involving the alignment of a

Kirschner wire (K-wire), where the entry point location and orientation both have

a significant impact on the overall outcome of the IM nailing procedure. During a

conventional IM nailing procedure, fluoroscopy images are obtained in the AP

15 direction (through the coronal plane).

Once adequate images with respect to entry point location and K-wire

orientation are acquired in the AP direction (see FIG. 3A), errors in sagittal

placement must be addressed. If the sagittal entry point location is correct but the

lateral orientation is incorrect (FIG. 3B), the surgeon must alter the anterior–

20 posterior angle of entry of the K-wire about the identified entry location (ensuring

no displacement of the K-wire tip from the entry site). If the sagittal entry point

location is incorrect but the orientation is correct (FIG. 3C), the surgeon must

adjust the anterior–posterior translation along the sagittal plane without any

145

change in sagittal angulation of the K-wire. If the sagittal entry point location and

orientation are both incorrect (FIG. 3D), then the surgeon must first readjust the

entry location in the sagittal plane. Once the correct new entry location is verified

with lateral fluoroscopy, the surgeon must then readjust the K-wire orientation in

5 the sagittal plane to ensure parallelism in access to the IM canal (FIG. 3E);

following this step, the surgeon needs to recheck the AP view to ensure that both

the coronal plane entry point and wire orientation are acceptable. As can be

understood with reference to FIGS. 3A-3E, multiple cycles of anterior-posterior

(AP) and lateral imaging may be required to confirm the optimal entry point

10 positioning. This unpredictable repetition of activities, can be time-consuming,

frustrating, costly, and can impact patient outcomes.

Referring now to FIGS. 4, and 5A-5C an example method of aligning a

device during medical procedure involving guide wire insertion during a femoral

intramedullary (IM) nailing procedure is described. Unlike the conventional

15 iterative and repetitive method of K-wire positioning and alignment described

above, the example method described below employs the fixation of the

anchoring component and relative rotation of the rotatable guide component.

This fixation and controlled relative rotation provides a method that is

deterministic and thus avoids the iterative trial-and-error nature of the

20 conventional method.

FIG. 4 shows a flow chart illustrating the steps in the example fluoroscopy-

based method, and FIGS. 5A-C show images of an example positioning and

alignment instrument during steps of the procedure. It is noted that the following

146

method steps shown in FIG. 4 are performed after having inserted the guidance

instrument into the bone, and that the following steps do not involve the step of

the insertion of the device into the bone, and rather recite steps for aligning the

device for subsequent insertion into the bone. The illustrated method steps

5 therefore do not pertain to a surgical intervention per se.

Prior to performing the method steps shown in FIG. 4, the positioning and

alignment instrument is initially placed, under fluoroscopic image guidance, at the

approximate IM nailing entry point location, which, in the present non-limiting

example, resides at the piriformis fossa or greater trochanter. Anterior-posterior

10 (AP) images are employed to align the positioning and alignment instrument such

that at least one device guide channel of the guidance component is aligned with

the intramedullary canal of the femur, as shown in FIG. 5A. Once this initial two-

dimensional alignment is satisfactory, the instrument is subjected to a force (as

shown in FIG. 2C) to anchor the anchoring protrusions into the femoral head.

15 FIG. 5B shows an AP image showing the positioning and alignment instrument

with the anchoring component anchored to the bone.

As shown in FIG. 5D, the example positioning and guidance instrument

includes a two-dimensional array of device guide channels, including a plurality

of rows 400 (perpendicular to the rotation axis at the distal end of the rotational

guide component) and a plurality of columns 405 (parallel to the rotation axis at 20

the distal end of the rotational guide component). According to such an

instrument configuration, when multiple rows of device guide channels are visible

in the AP images, the AP image may be employed to select a guide channel row

147

that is best (optimally) aligned with the intramedullary canal, as shown at step

300 in FIG. 4. It will be understood that this step may optionally be performed in

cases in which the guidance component includes a plurality of rows of device

guide channels that are observable in the AP image (the optional nature of this

5 step is indicated by the dashed flow chart element 300 in FIG. 4). In various

example embodiments, the guidance component may only include a single

device guide channel, a single row of device guide channels, a single column of

device guide channels, or a two-dimensional array of rows and columns of device

guide channels.

10 As shown in step 305 of FIG. 4, one or more subsequent lateral (or

oblique-lateral) images (e.g. images in a perpendicular direction that includes the

sagittal plane) are then obtained to identify the correct three-dimensional

trajectory for accessing the intramedullary canal. The guidance component is

then rotated relative to the anchoring component in order to align the guidance

15 component with the intramedullary canal, as shown at step 310. Such a

configuration is shown in FIG. 6, where the rotational guidance component has

been rotationally aligned such that the K-wire is axially aligned with the

intramedullary canal upon insertion therein.

In the event that the position and alignment instrument includes multiple

20 columns of device guide channels (as in FIG. 5D), the device guide channels

visible in the lateral image may be employed, as shown in step 315 to select the

column (while maintain the selection of the row, if performed in step 300) that

provides the optimal alignment of the device guide channel with the

148

intramedullary canal.

As described above with reference to FIGS. 1A and 1B, the positioning

and alignment instrument may include a rotation actuation mechanism, and the

rotation actuation mechanism may be located proximal to, or adjacent to, a

5 handle portion of the anchoring component, such that the rotation actuation

mechanism can be actuated single-handedly while holding the handle.

Unlike the conventional method described with reference to FIGS. 3A-3E,

the present example method enables deterministic positioning and alignment of a

device without requiring an iterative trial-and-error based approach. This is

10 achieved by the fixation of the anchoring component during the initial collection of

images, and the subsequent rotation of the guidance component during the

acquisition of images from a perpendicular direction, where the guidance

component is rotated in an arc that is fixed relative to the anchoring component,

such that the initial alignment in the first direction is preserved and maintained

15 when aligning the remainder of the rotational guidance component in the second

direction.

Having obtained alignment of the guidance component, using the

aforementioned direct and deterministic method, the device may be subsequently

guided by the selected device guide channel (i.e. along the selected row and

20 column) for controlled insertion into the bone.

It will understood that the specific implementation shown in FIGS. 1A and

1B is provided to illustrate one example and non-limiting configuration of a

positioning and guidance instrument, and other configurations may be employed

149

for other clinical applications. For example, in other example implementations,

the device channel guides 122 need not extend to a distal location that is

adjacent to the distal end 118 of the anchoring component 110.

In the various examples embodiments described herein, the device

5 received within the device guide channel is a K-wire. However, it will be

understand that a K-wire is disclosed as a non-limiting example of a broad class

of devices. Accordingly, the term "device", as used herein, refers generally to any

number of implantable devices, materials and instruments suitable for bone

treatment and/or repair. For example, the device may be an implantable device,

10 an insertion tool, a drill bit, an injection needle, a catheter, or any other surgical

instrument.

While the positioning and alignment device illustrated in FIGS. 1A and 1B

employs two anchoring protrusions, it will be understood that this specific

configuration provides merely one example implementation, and that in general

15 the anchoring component may have one or more anchoring protrusions, provided

that the one or more anchoring protrusions are suitable for anchoring the

anchoring component in a fixed position and fixed orientation. For example, FIG.

9 illustrates an example implementation of a single anchoring component 500

having a rectangular elongate segment 505 that enforces a fixed position and

20 orientation when embedded in bone.

Although the preceding example embodiments involve intramedullary

nailing procedures, it will be understood that the embodiments of the present

disclosure may be applied to, or adapted to, a wide variety of surgical

150

procedures, in which tool guide channels are aligned to internal anatomical or

function features, or to features associated with embedded medical devices.

For example, as shown in FIGS. 7 and 8, the positioning and alignment

device may be employed to select a suitable position and orientation of a drill bit

5 that is employed to drill an initial hole to guide the subsequent insertion of a

locking screw into the distal region of an IM nail. Once the distal nail hole is seen

as a circle under fluoroscopy and an incision is made, the device can be inserted

into the bone. As shown in FIG. 7, the curvature of the rotatable guidance

component is beneficial to reducing both the amount of the position and

10 alignment instrument that lies within the path of the fluoroscopy beam and

potential radiation exposure to the surgeon’s hands. Once a correct guide

channel is selected, a flexible drill bit can be advanced to make an initial hole for

the subsequent insertion of a locking screw. This eliminates the need for a

radiolucent drill or repeated assessments of the drill bit if a free hand technique is

15 utilized. The positioning and alignment device of the aforementioned

embodiments could also be utilized, for example, in prophylactic femoral, tibial or

humeral nailing, or, for example, any other surgery that involves insertion of a

surgical instrument into bone under fluoroscopic guidance.

The specific embodiments described above have been shown by way of

20 example, and it should be understood that these embodiments may be

susceptible to various modifications and alternative forms. It should be further

understood that the claims are not intended to be limited to the particular forms

disclosed, but rather to cover all modifications, equivalents, and alternatives

151

falling within the spirit and scope of this disclosure.

152

THEREFORE WHAT IS CLAIMED IS:

1. A positioning and alignment instrument for guiding insertion of a device into bone, the positioning and alignment instrument comprising:

an anchoring component comprising a proximal portion and a distal portion, wherein said proximal portion comprises a handle, and wherein one or more anchoring protrusions extend from a distal end of said distal portion for anchoring said anchoring component into the bone, such that a position and an orientation of said anchoring component is fixed relative to the bone when said anchoring component is anchored to the bone; and

a guidance component mechanically supported by said anchoring component, said guidance component comprising a device guide channel for receiving the device and guiding the device towards an insertion location adjacent to the distal end of said anchoring component;

wherein said guidance component is rotatable relative to said anchoring component about a rotation axis that is located adjacent to the distal end of said anchoring component, such that the insertion location remains adjacent to the distal end of said anchoring component under rotation of said guidance component.

2. The positioning and alignment instrument according to claim 1 further comprising a rotation actuation mechanism for actuating rotation of said guidance component, wherein said rotation actuation mechanism is located proximal to

153

said handle such that said rotation actuation mechanism is capable of being actuated by a user while holding said handle with a single hand.

3. The positioning and alignment instrument according to claim 2 wherein said rotation actuation mechanism is configured to be actuated by a thumb while holding said handle with the single hand.

4. The positioning and alignment instrument according to claim 3 wherein said rotation actuation mechanism comprises a rotatable knob that is adjacent to said handle for actuation by the thumb.

5. The positioning and alignment instrument according to claim 2 or 4 further comprising a rotation locking mechanism for locking a rotation angle of said guidance component.

6. The positioning and alignment instrument according to claim 5 wherein said rotation locking mechanism is located proximal to said handle such that said rotation locking mechanism is capable of being actuated by the user while holding said handle with the single hand.

7. The positioning and alignment instrument according to any one of claims 1 to

6 wherein said guidance component is pivotally coupled to said anchoring at a pivot location that is adjacent to the distal end of said anchoring component and

154

adjacent to a distal end of said guidance component.

8. The positioning and alignment instrument according to claim 7 wherein said guidance component comprises a slot having a slot axis that is parallel to a distal portion of said device guide channel, wherein said positioning and alignment instrument further comprises a linkage having a first end that is pivotally coupled said proximal portion of said anchoring component, and a second end comprising a pin, wherein said pin is received in said slot.

9. The positioning and alignment instrument according to any one of claims 1 to

8 wherein said guidance component is rotatably supported relative to said anchoring component such that said rotation axis is within 5 mm of the distal end of said anchoring component.

10. The positioning and alignment instrument according to any one of claims 1 to

8 wherein said guidance component is rotatably supported relative to said anchoring component such that said rotation axis is within 2 mm of the distal end of said anchoring component.

11. The positioning and alignment instrument according to any one of claims 1 to

10 wherein a distal portion of said device guide channel is aligned along a device guidance axis, and wherein a proximal portion of said device guide channel deflects outwardly relative to said device guidance axis toward said handle.

155

12. The positioning and alignment instrument according to claim 11 wherein said proximal portion is deflected outwardly at an angle of 0 to 30 degrees relative to said device guidance axis.

13. The positioning and alignment instrument according to any one of claims 1 to

12 wherein said distal portion of said anchoring component has a rectangular cross-section, wherein a long axis of said rectangular cross-section is perpendicular to the rotation axis and wherein said one or more anchoring protrusions extend from the distal end of said anchoring component.

14. The positioning and alignment instrument according to any one of claims 1 to

13 wherein said proximal portion comprises a proximal impact receiving surface suitably oriented to receive an impact for driving said one or more anchoring protrusions into the bone.

15. The positioning and alignment instrument according to any one of claims 1 to

13 further comprising a force coupling tool comprises a distal surface suitable for contacting said anchoring component, wherein said force coupling tool comprises a proximal surface suitable for receiving an impact, such that when said force coupling tool is contacted with said anchoring component and an impact force is delivered to said proximal surface of said force coupling tool, the impact force is coupled through said force coupling tool to said anchoring component for driving

said anchoring protrusions into bone.

156

16. The positioning and alignment instrument according to any one of claims 1 to

15 wherein a distal portion of said guidance component is radiolucent.

17. The positioning and alignment instrument according to any one of claims 1 to

16 wherein said device guide channel is a first device guide channel, and wherein said positioning and alignment instrument further comprises one or more additional device guide channels, wherein said one or more additional device guide channels are laterally spaced relative to said first device guide channel in a direction that is perpendicular to the rotation axis.

18. The positioning and alignment instrument according to any one of claims 1 to

19. The positioning and alignment instrument according to any one of claims 1 to

18 wherein said guidance component comprises a two-dimensional array of guide channels, wherein the two-dimensional array of guide channels are arranged in two or more rows and two or more columns, wherein said rows arranged perpendicular to the rotation axis, and wherein said columns are arranged parallel to the rotation axis.

157

20. The positioning and alignment instrument according to any one of claims 1 to

19 wherein said device guide channel has a diameter suitable for guiding a

Kirschner wire.

21. A method of employing fluoroscopy to aligning a device during a medical procedure, the method comprising, after having employed fluoroscopy, in a first direction, to anchor the positioning and alignment instrument according to any one of claims 1 to 16 into bone:

obtaining fluoroscopy images of the positioning and alignment instrument in a perpendicular direction; and

rotating the guidance component to a desired angle according to the fluoroscopy images; and

thereby aligning the device guide channel for subsequent guidance and insertion of the device into the bone.

22. The method according to claim 21 wherein said device guide channel is a first device guide channel, and wherein the guidance component further comprises one or more additional device guide channels, wherein said one or more additional device guide channels are laterally spaced relative to said first device guide channel in a direction that is parallel to said rotation axis, and wherein the method further comprises:

prior to obtaining fluoroscopy images of the positioning and alignment instrument in the perpendicular direction, obtaining fluoroscopy images of the positioning and alignment instrument in the first direction, and selecting a suitable

158

device guide channel for guiding the device to a desired entry location.

23. The method according to claim 21 wherein said device guide channel is a first device guide channel, and wherein the guidance component further comprises one or more additional device guide channels, wherein said one or more additional device guide channels are laterally spaced relative to said first device guide channel in a direction that is perpendicular to said rotation axis, and wherein the method further comprises:

selecting a suitable device guide channel for guiding the device to a desired entry location.

24. The method according to claim 21 wherein the guidance component further comprises a two-dimensional array of guide channels, wherein the two- dimensional array of guide channels are arranged in two or more rows and two or more columns, wherein the rows arranged perpendicular to the rotation axis, and wherein the columns are arranged parallel to the rotation axis, wherein the method further comprises:

when obtaining fluoroscopy images of the positioning and alignment instrument in the perpendicular direction, selecting a suitable row for guiding the device to the desired entry location;

159

wherein the suitable row and suitable column identify a suitable device guide channel for guiding the device to the desired entry location.

25. The method according to any one of claims 21 to 24 wherein the device is a

Kirschner wire and the medical procedure involves the insertion and drilling of the

Kirschner wire into the intramedullary canal of the bone.

26. The method according to any one of claims 21 to 24 wherein the device is a drill bit and the medical procedure involves the insertion and drilling of the drill bit into the bone in order to generate a pilot hole for subsequent insertion of a locking screw into an intramedullary nailing.

27. The method according to any one of claims 21 to 26 wherein the perpendicular direction is an anterior-posterior direction.

160

ABSTRACT

A positioning and alignment instrument, and methods of use thereof, are provided for facilitating the alignment and insertion of a device, such as a guide wire, into bone. The instrument includes a handheld anchoring component and a rotatable guidance component. During use, the anchoring component is anchored into bone via anchoring protrusions, such that the position and orientation of the anchoring component is fixed relative to the bone. The guidance component, which is mechanically supported by the anchoring component, includes a device guide channel for receiving the device and guiding the device towards an insertion location adjacent to the distal end of the anchoring component. The guidance component is rotatable relative to the anchoring component about a rotation axis that is located adjacent to the distal end of the anchoring component, such that the insertion location remains adjacent to the distal end of the anchoring component under rotation.

161

162

163

164

165

166

167

168

169

170

171

172

173

174

175

176