Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Evaluation of a computer-assisted technique for distal locking of femoral intramedullary nails Beadon, Katherine E. 2007

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_2007-0322.pdf [ 18.68MB ]
Metadata
JSON: 831-1.0080777.json
JSON-LD: 831-1.0080777-ld.json
RDF/XML (Pretty): 831-1.0080777-rdf.xml
RDF/JSON: 831-1.0080777-rdf.json
Turtle: 831-1.0080777-turtle.txt
N-Triples: 831-1.0080777-rdf-ntriples.txt
Original Record: 831-1.0080777-source.json
Full Text
831-1.0080777-fulltext.txt
Citation
831-1.0080777.ris

Full Text

E V A L U A T I O N OF A COMPUTER-ASSISTED TECHNIQUE FOR DISTAL L O C K I N G OF F E M O R A L I N T R A M E D U L L A R Y NAILS  by  KATHERINE E. B E A D O N B.Sc.(Physics), The University of British Columbia, 2005  A THESIS SUBMITTED IN P A R T I A L F U L F I L L M E N T OF THE REQUIREMENTS FOR THE D E G R E E OF M A S T E R OF APPLIED SCIENCE  in  THE F A C U L T Y OF G R A D U A T E STUDIES  (Mechanical Engineering)  THE UNIVERSITY OF BRITISH C O L U M B I A July 2007 © Katherine E. Beadon, 2007  Abstract The objective of this work was to design the devices and methods necessary for a computerassisted distal locking technique for intramedullary nailing of the femur using an electromagnetic tracking system. The long-term goal of this work is to increase the efficiency and decrease the radiation exposure to the surgical team during the intramedullary nailing procedure. The accuracy of the electromagnetic system was determined to be acceptable for our application, even in the presence of the titanium implant. A distal locking tool was designed that locks into the distal locking holes from within the intramedullary nail with high repeatability. The stresses predicted from a finite element analysis of the removal process predict maximum stresses of l/500 of the yield stress of the material. The use of this tool th  produces wear particles too few in number and too small in size to induce osteolysis in a patient. We conclude that the distal locking device is very likely to be successful in a surgical setting. A clinical trial was designed in order to test this hypothesis. The clinical trial has been delayed due to the establishment of new requirements at our hospital governing the introduction of custom-made research devices into a surgical setting. As a result, we have developed a set of guidelines for future researchers to facilitate completing all the required testing to obtain approval to perform a clinical trial. We are currently in the final stages of satisfying these requirements and anticipate that our clinical trial will begin in early fall.  Table of Contents Abstract Table of Contents List of Tables List of Figures List of Abbreviations Acknowledgments  .  ii iii v vi x xi  1  Introduction and Literature Review 1.0 History of Intramedullary Nailing of the Femur 1.1 Intramedullary Nailing Technique 1.1.1 Nail Insertion 1.1.2 Proximal Locking 1.1.3 Distal Locking 1.2 Disadvantages of the Current Technique 1.3 Alternative Techniques for Distal Locking 1.4 Purpose of This Study  1 3 4 4 5 6 7 7 12  2  Materials and Methods 2.1 Aurora Electromagnetic Tracking System 2.2 Synthes Universal Femoral Intramedullary Nail 2.3 Effect of the Intramedullary Nail 2.4 Distal Locking Tool 2.4.1 Design 2.4.2 Insertion Repeatability 2.5 Mechanical Testing 2.5.1 Experimental Moment to Remove D L T 2.5.2 Maximum Moment 2.5.3 Stress Analysis 2.5.4 Particulate Formation Results 3.1 Effect of the Intramedullary Nail 3.2 Insertion Repeatability 3.3 Mechanical Testing 3.3.1 D L T Removal Moment 3.3.2 Stress Analysis 3.3.3 Particulate Formation  14 14 16 17 20 20 23 27 28 29 29 30 33 33 36 37 37 38 39  4  Discussion 4.1 Effects of the Intramedullary Nail 4.2 Distal Locking Tool Accuracy 4.3 Ease and Safety of Distal Locking Tool Design 4.4 Optimizing the Design 4.5 Particulate Formation  42 43 46 52 52 54  5  Implementation of a Clinical Trial for the D L I M N project 5.1 Materials and Methods  57 58  3  iii  5.1.1 Required Instrumentation 5.1.2 Data Collection 5.1.3 Data Analysis 5.2 Current Status 5.2.1 Safety Testing for the D L I M N trial 5.2.2 Sterilization Method and Validation for the D L I M N Trial 5.2.3 Final Approval 5.3 Expected Contributions 6  :  Conclusion 6.1 Design and Testing of the Distal Locking Tool 6.2 Evaluation of the Method and Apparatus  59 61 62 63 63 64 67 67 69 69 70  References  71  Appendix Appendix Appendix Appendix  77 80 81  Appendix Appendix Appendix Appendix Appendix Appendix Appendix Appendix Appendix  1: 2: 3: 4:  Design Evolution of the Distal Locking Tool A B S Material Specifications Theoretical Force to Remove D L T and Factor of Safety Calculations Planning and Implementation of a Clinical Trial in a Research Setting 5: Clinical Research Ethics Board Application: Protocol for D L I M N Trial 6: Clinical Research Ethics Board Application: Consent Form for D L I M N Trial 7: Clinical Research Ethics Board Application: Online Application Form 8: Approval to Conduct Research at Vancouver Coastal Health 9: Guideline for Obtaining Approval to Conduct Research at Vancouver Acute 10: Radiation Stability of Polymer Materials 11: Aurora Approvals and Classifications 12: Cleaning and Sterilization Protocol for MIS Study 13: Results for Sterility Testing of the Reference Clip  83 107 113 117 134 137 148 150 151 154  iv  List of Tables Table 3.1: Summary of Results for Accuracy Testing Table 3.2: Moment Required to Remove the Distal Locking Tool Table 4.1: Standard Deviations for Sensor Accuracy and Insertion Repeatability for all 6 DOF Table 5.1: Radiation Doses Applied to the DLTs. Four tools were radiated at each radiation dose tested. The doses tested were 25, 35, and 45 kGy  34 37 46 67  List of Figures Figure 1.1 Rigidly Locked Femoral Intramedullary Nail. Proximal and distal locking of intramedullary nails allows for early mobilization without the risk of angular or femoral length malalignment, (source: Trigen® IM Nail System Surgical Technique) Figure 1.2: Nail Insertion. Once the intramedullary canal has been opened, a small guide wire is fed down the canal through all bone fragments under fluoroscopic guidance, (source: Synthes® Universal Nail System Technique Guide) Figure 1.3: Proximal Locking. The insertion handle attached to the proximal end of the I M N is fitted with an alignment guide for proximal locking (left). Both the drilling and screw insertion are done by placing the screw through the alignment guide and into the bone (right), (source: Synthes® Universal Nail System Technique Guide) Figure 1.4: Distal Locking Under Fluoroscopic Guidance. The fluoroscope must first be positioned such that the locking holes appear as perfect circles (B) and not ellipses (A). The drill bit is then aligned such that its tip appears in the centre of the circle (C) and then so that the drill bit appears only in the centre of the circle, (source: Synthes® Universal Nail System Technique Guide) Figure 1.5: Mechanical Guide for Distal Locking. The use of this mechanical guide requires the surgeon to drill a 6-10 mm working channel in order to palpate the distal holes using a guide hook. This method has not come into widespread use due to the potential for complications associated with the creation of a working channel, (source: Krettek 1998) Figure 1.6: Inflatable Nail. Insertion of the Fixion Inflatable Nail obviates the need for distal locking and therefore reduces surgical time and radiation exposure. This nail has not come into widespread use due to complications associated with inflation and potential lack of torsional stability, (source: www.disc-otech.com) Figure 1.7: Virtual Fluoroscopy Setup. In order to reduce radiation exposure during D L I M N , the fluoroscope (A) and surgical tools (B) are tracked using an optical tracking system (not shown). This allows the D L I M N procedure to be accomplished with as few as two fluoroscopic shots that are approximately perpendicular, (source: Suhm 2004) Figure 1.8: Magnetic Probe System. A permanent magnet is inserted into the central canal of an IMN. Local changes in magnetic field in the area of the locking holes are sensed by an external magnetic sensor fitted with an L E D display and a drill sleeve, (source: Szakelyhidi 2002) Figure 1.9: Testing of a Novel Surgical Method: After design and lab-based testing is complete, it is necessary to test the method and apparatus against the current surgical approach and to complete a surgical task simulation to ensure that the method is accurate enough to be tested in a clinical setting using a randomized controlled trial Figure 2.1: Aurora Electromagnetic Tracking System. The system is made up of a control unit, a field generator, a sensor interface unit and a 6 DOF sensor with a diameter of 2.5 mm. The sensor is attached to lead wires and a connector  4  5  6  6  8  9  10  11  12  vi  plug which mates with the system interface unit.(source: Northern Digital, www.ndigital.com) Figure 2.2: Aurora Measurement Volume. The measurement volume of the Aurora system is a cube with side lengths of 500 mm. The origin of the system lies on the front face of the field generator with the measurement volume projecting outwards, (source: Aurora User Manual 2005) Figure 2.3. Synthes Universal Femoral Nail. A l l of the testing was performed using Synthes Universal Femoral Nails Figure 2.4: Calibration Phantom. The calibration phantom was placed approximately 250 mm in front of the Field Generator. Data was collected while the calibration plug was in each of the alignment holes. The distance between the calibration plug and the reference plug was calculated and compared to the known locations of alignment holes, (source: Kirsch 2005) Figure 2.5: Calibration Plug. The sensor was fixed within the IMN using set screws (C). The IMN was rigidly fixed in the calibration plug by press fitting it into a cylinder attached to the calibration plug (B) and inserting a rod through the distal locking hole (A) Figure 2.6: Distal Locking Tool. The custom designed Distal Locking Tool is placed inside of the IMN once the nail has been inserted into the femur, (source: modified from Trigen® IM Nail System Surgical Technique) Figure 2.7: Design of the Distal Locking Tool. The locking arms are designed to expand when the distal locking tool is inserted and lock into the distal locking holes. The diameter of the distal locking tool is 3.4 mm. The sensor is offset from the locking plugs by 50 mm Figure 2.8: Distal Locking Tool Plug. The distal ends (left) of the DLT are designed with plugs to engage the locking holes of the IMN. The distal surface of the plugs are cylindrically shaped so that they do not slide past the locking holes. The proximal (right) surface of the plugs are spherically shaped in order to facilitate removal of the DLT Figure 2.9: C-shaped canal. The D L T is fitted with a C-shaped canal to allow the 6DOF sensor to be sealed within the DLT Figure 2.10: Digitizing Probe. The probe reports the tip location in 6 DOF. The tip of the probe is chamfered Figure 2.11: Digitizing the Locking Holes. The locking holes were digitized by placing the chamfered tip of the probe into the holes until the probe made contact with the rim of the locking hole Figure 2.12: Nail Frame Construction 1. Four points on the IMN were digitized in order to create the nail frame, the medial and lateral surfaces of the most distal locking hole (pi and p2) and the medial and lateral surfaces of the more proximal distal locking hole (p3 and p4) Figure 2.13: Nail Frame Construction 2. The centers of the locking holes were defined as the midpoints between the two sides of the locking holes Figure 2.14: Torque Necessary to Remove DLT. The IMN was fixed in the calibration box and the DLT was inserted. The proximal end of the DLT was threaded through a cylindrical attachment and supported at the proximal end. A rope was wrapped around the cylinder and threaded through the base of the calibration box. Weights with increasing mass were attached to the rope until the DLT was released from the locking holes  15  15 16  17  19  21  22  22 23 24  24  25 25  28  vn  Figure 2.15: F E M mesh for the distal locking tool. The DLT was meshed with 289,540 tetrahedral elements and 279,954 degrees of freedom Figure 3.1: Accuracy of the Aurora sensor with and without the IMN. The accuracy of the Aurora sensor in standard conditions was 0.41± 0.12 mm and 0.38± 0.28°. With the IMN, the accuracy was 0.52 ± 0.16 mm and 0.43 ± 0.20°. Precision error of the readings increased in the presence of the IMN Figure 3.2: Cumulative Distribution of Accuracy Errors. Results for positional readings (left) taken without the IMN (-) and with the I M N (--) show that a distortion of approximately 0.1 mm in induced by the presence of the nail. This difference is statistically significant (p = 0.006) but not clinically relevant. Orientation errors (right) were not significantly different ( p = 0.41) Figure 3.3: Probability Distribution for Accuracy Errors. Positional errors (left) without the IMN (-) and with the IMN (--) show very small 95% confidence intervals (0.06 and 0.05 respectively). Orientation errors (right) have 95% confidence intervals of 0.12° and 0.16° without and with the I M N Figure 3.4: Position Accuracy Error does not change as a function of distance from the sensor to the field generator Figure 3.5: Position precision error increases with increasing sensor distance from the field generator Figure 3.6: Mean Deviations for Translations. Boxplots showing the mean, +/- standard deviation, min and max for each of the cardinal directions. Translational errors were very low with standard deviations of 0.03, 0.12 and 0.20 mm in the x (longitudinal axis of the nail), y (anteroposterior axis), and z (mediolateral axis) directions Figure 3.7: Mean Deviations for Rotations. Boxplots showing the mean, +/- standard deviation, min and max for rotations about each of the cardinal directions. Rotational errors were low with standard deviations of 1.59°, 0.13° and 0.08° about the x (longitudinal axis of the nail), y (anteroposterior axis), and z (mediolateral axis) axes respectively Figure 3.8: Finite Element Model of the Distal Locking Tool. The maximum von Mises stress reported from the model was 8.5-10 Pa and was located at the junction of the locking arms and the locking plugs Figure 3.9: Number of Particles Produced. The number of particles produced was reduced by 77% for devices that were filed (trials 3-4) compared to devices that were not filed (trials 1-2) Figure 3.10: Particulate Samples. Very small (A), small (B), medium (C), and large (D) particles had maximum edge lengths of <10 urn, 10-20 um, 20-50 um and >50 urn respectively Figure 4.1: Maximum Error Tolerance for Distal Hole Targeting. The maximum error tolerance for targeting distal locking holes has been previously reported as 0.75 mm and 8°. (source Szakelyhidi 2002) Figure 4.2: Orthopaedic Fracture Table. Surgical tables designed for orthopaedic fracture fixation are designed to be radiolucent. The composite table top and abductor bars decrease the amount of steel near the sensing area thereby decreasing the interference for electromagnetic tracking, (source: www.steris.com) Figure 4.3: Accuracy (shown here as Trueness) of the Aurora 2 system is optimal at d < 400mm. Precision of the Aurora readings also decreases for distances over 400 mm. (source: Kirsch 2005-1)  30  33  34  34 35 35  36  36  4  38  39  40  42  44  45 vm  Figure 4.4: The total distance between the sensor and locking plugs is 50mm. This distance is made up of the locking arms (shown above) and the transition zone. Decreasing this total distance may decrease the error associated with predicting the locking hole axis from the sensor location Figure 4.5 Tracking accuracy decreases with decreased distance between the sensor and a CT table (source: Krucker 2006) Figure 5.1: Testing Involved in Proof of Concept for Electromagnetic Targeting in D L I M N . Accuracy has been demonstrated in a lab-based setting. It is now necessary to test the systems performance in a simulated surgical task and compared to the conventional technique. When this is complete, the system can be used for targeting in the OR Figure 5.2: Reference Clip. The reference clip will be attached to the drill bit during alignment for distal hole targeting using a fluoroscope Figure 5.3: Distal Targeting Device. The drill bit is fastened into the radiolucent targeting device using a set screw (A). The distal holes are the located by aligning the drill bit to the centre of the locking hole under fluoroscopic guidance, (source: Zimmer M / D N ® Femoral Interlocking and Recon Nail Intramedullary Fixation Surgical Technique) Figure 5.4: Reference Clip. The reference clip contains a small channel (A) into which the drill bit (E) is placed. The clip is then tightened down by turning a lever (B). The clip is also fitted with screw holes (C). These holes are used to attach the sensor (D) to the clip  48 51  57 59  60  61  IX  List of Abbreviations ABS  Acrylonitrile Butadiene Styrene  BME  Biomedical Engineering  CAOS  Computer Assisted Orthopaedic Surgery  CAS  Computer Assisted Surgery  CREB  Clinical Research Ethics Board  DLIMN  Distal Locking for Intramedullary Nailing  DLT  Distal Locking Tool  DOF  Degree of Freedom  EMTS  Electromagnetic Tracking System  FG  Field Generator  IMN  Intramedullary Nail  MDDSC  Medical Device Development Safety Committee  MDR  Medical Devices Regulations  MoH  (British Columbia) Ministry of Health  PMMA  Polymethyl methacrylate, Bone cement  RMS  Root Mean Square  SPD  Sterile Processing Department  THA  Total Hip Arthroplasty  UBC  University of British Columbia  UHMWPE  Ultra High Molecular Weight Polyethylene  VCHRI  Vancouver Coastal Health Research Institute  Acknowledgments I would like to thank all those who have supported me throughout this project. Thank you to my supervisor, Dr. Antony Hodgson for sharing his knowledge and providing encouragement. Thanks to Drs. Pierre Guy and Peter O'Brien for the idea that lead to this thesis topic and for providing valuable insight into all things surgical. Special thanks to Dr. Guy for making time in his busy schedule to coach me through the M D interview process. Many thanks to Jeff Stanley, Saibal Chakraburtty and their colleagues at Northern Digital Inc. for their financial support, guidance with the Aurora system and for agreeing to allow me to work with them in Waterloo, despite my being a security risk... I would also like to thank everyone at V C H R I who took part in the development of the new hospital guidelines. Thank you to Dr. Elizabeth Bryce, Darren Kopetsky, Stephania Manusha, Mona Sharman, Gord McConnell, Debbie Hendricks and Murray Dore for working tirelessly to arrive at a solution that worked for everyone instead of simply not allowing my work. Thank you to Alex English and Andrea Geere for initiating me into the world of sterilization of medical devices and for fielding my innumerable questions about processes and procedures. Thanks to Caron Fournier, Perri Berthelet and Nancy Henderson for their technical assistance in various aspects of the project. Special thanks to all my colleagues at the Neuromotor Control Lab. Sayra, Jill, Jeremy and all the others who have passed through here in the last several years, you have made our lab a productive and enjoyable place to work. Thanks for the random discussions about nearly everything, helping me with my project, listening to me rant and for escaping the lab for much needed breaks! Thank you also to team spondy for teaching me the skills necessary to be an effective researcher and for keeping me company throughout all the long hours spent watching the Instron. Thank you to Matt and Josh for making Friday afternoon anatomy labs something to look forward to! Many thanks to all of my friends who were so supportive throughout my degree and especially in the last several weeks. Thanks for the late night sushi deliveries, the last minute editing, and for listening to me whine on occasion. Thanks to all the brostellers for an incredible community in which to live and play; the weekends in the mountains and on the water have made these last few years some of my best. Thank you to my family for teaching me to work hard and play hard, for instilling in me the self-confidence and self-reliance that have guided me through my choices so far, and for your support as I begin my next challenge. Lastly, thanks to Brad for being here for the last six years despite my neuroses and indecision. Thanks for understanding me. This work was supported by the Natural Sciences and Engineering Research Council of Canada (NSERC).  xi  1 Introduction and Literature Review Over the past two decades, computer-assisted orthopaedic surgical (CAOS) techniques have been developed for a variety of traditional orthopaedic procedures. Incorporating real-time feedback during surgery can allow for better precision, better limb and implant alignment, better fracture reduction and less radiation exposure (Kanlic 2006). New surgical tools and techniques have also allowed many traditionally open procedures to be completed less invasively allowing for faster recovery, decreased post-operative pain, and improved functional outcomes (Kanlic 2006). This general approach of combining intra-operative tracking of surgical tools with pre or intraoperative images has been successfully used in a wide variety of applications ranging from neurosurgery to gynecology.  One of the most common procedures performed by orthopaedic surgeons is fracture treatment. A British study spanning ten years found that long bone fractures (eg, of the femur, tibia/fibula and humerus) are very common: the femur is the fifth most common fracture site, fractures of the tibia and fibula ranked third most common and those of the humerus ranked seventh (van Staa 2001). Fractures of the shaft of long bones are often treated by insertion of an intramedullary nail which has been shown to be an extremely successful procedure(Winquist 1984). Nonetheless, femur fracture cases represent the single most prevalent and the third most expensive type of malpractice suit brought against orthopaedic surgeons. Tibial and humeral fractures rank second and tenth respectively on the list of most common malpractice suits (AAOS 2001). It is therefore not surprising that a wide variety of surgical improvements have been proposed for safe and accurate intramedullary nailing (Pennig 1997, Krettek 1999, Slomczykowski 2001, Szakelyhidi 2002, Hazan 2003, Suhm 2004, Ozturk 2005, Hollstein 2006). 1  During intramedullary nailing of the femur, the items of interest are the bone fragments, the surgical tools, and the intramedullary nail (IMN) itself. Some researchers have used optical systems to track bone fragments during fracture reduction (Hart 2007, Khoury 2007, Weil 2007) but optical systems cannot accurately track the position of the intramedullary nail due to the deformation associated with nail insertion (Pennig 1997, Slomczykowski 1998, Krettek 1998). This makes distal locking of intramedullary nails a time consuming and technically demanding procedure. The traditional approach relies on fluoroscopic guidance. This method exposes the surgical team to a significant amount of cumulative radiation over the course of their careers.  In the last several decades, surgeons have become increasingly specialized and consequently perform procedures that expose them to potentially harmful radiation more often. The accumulation of radiation exposure over the course of an orthopaedic surgeon's career is a rising concern and has received significant attention in the literature of late (Mehlman 1997, Herscovici 2000, Madan 2002, Hafez 2005, Singer 2005). A l l of these studies emphasize the need to minimize fluoroscopic time and to avoid being in the direct path of the x-ray beam. Hafez argues that previous studies have significantly underestimated the actual radiation exposure received at surgeons' fingertips. The authors found that radiation doses recorded at the fingertips were as much as 75 times higher than doses recorded at the base of the fingers. This is likely due to the fingers being positioned directly in the x-ray beam during screening. This conflicts with the conclusions drawn in older studies (Miller 1983, Levin 1987, Riley 1989, Goldstone 1993, Noordeen 1993, Sanders 1993, Smith 1996, Jones 1998, Muller 1998) that the radiation exposure to the hands of orthopaedic surgeons is below acceptable limits. The risks of ionizing radiation in relation to initiating cancers are well known and the incidence of such cases is rising  2  (Gundestrup 1999). It is therefore pertinent to explore surgical solutions that minimize the use of ionizing radiation.  The long-term goals of this research are to develop the tools and methodology necessary to accomplish distal locking of intramedullary nails (DLIMN) with a more time-efficient, accurate and radiation-free technique. The scope of this project was the development and evaluation of the tools necessary for targeting distal locking holes from within the intramedullary nail (IMN) using an electromagnetic-based tracking system.  1.0 History of Intramedullary Nailing of the Femur Beginning as early as the 16 century, intramedullary fixation for the treatment of long bone th  fractures has been explored (Farill 1952). Modern intramedullary nailing was spearheaded by a 1931 report by Smith-Petersen on the use of stainless-steel nails in femoral neck fracture treatment. This technique was adapted by Kutscher in the early 1940's for the treatment of femoral shaft fractures (Bong 2007). These early surgeries involved the use of head-worn fluoroscopy to visualize the bone. The advent of penicillin in the late 40's led to a resurgence of an open surgical technique in order to avoid the detrimental effects of the high levels of the radiation emitted by the radiographic techniques of the time. With the development of better radiological image intensification in the 60's and 70's, closed intramedullary procedures could once again be performed with reduced risks to patients and surgeons (Bong 2007).  Winquist (1984) reviewed 520 intramedullary nailing procedures performed between 1968 and 1979 and found union rates of over 99% with very few complications. Intramedullary nailing remains the standard of care for the treatment of femoral fractures today.  3  1.1 Intramedullary Nailing Technique Modern treatment of femoral fractures involves the insertion of a metal nail into the intramedullary canal (Figure 1.1). The implant used is usually a hollow tube, 10-15 mm in diameter that is placed into the central canal of the bone and locked into place with screws drilled through the bone and implant at both the proximal and distal ends. The surgical procedure can be divided into three main steps: nail insertion, distal locking, and proximal locking. Locking the IMN both proximally and distally allows for early mobilization without the risk of limb length discrepancies or angular malalignment (Dobbs 2006).  Figure 1.1 Rigidly Locked Femoral Intramedullary Nail. Proximal and distal locking o f intramedullary nails allows for early mobilization without the risk o f angular or femoral length malalignment, (source: T r i g e n ® I M N a i l System Surgical Technique)  The order in which locking is performed is at the discretion of the surgeon. Once the nail is locked distally, the distal femoral bone fragment can be manipulated to obtain optimum alignment before locking the IMN proximally. Distal locking is therefore often completed prior to proximal locking. 1.1.1  Nail Insertion  After the correct entry point is identified, the intramedullary canal is opened and a guide wire is inserted through the femur to the distal metaphysis (Figure 1.2).  4  Figure 1.2: Nail Insertion. Once the intramedullary canal has been opened, a small guide wire is fed down the canal through all bone fragments under fluoroscopic guidance, (source: Synthes® Universal Nail System Technique Guide)  Both entry point localization and guide wire insertion are done under fluoroscopic guidance. The hollow implant is then threaded onto the guide wire and hammered into place inside the femur assisted by the use of an insertion handle. Once the IMN is in place, final placement is verified using the fluoroscope. 1.1.2  Proximal Locking  In order to insert the proximal locking screws into the IMN, a mechanical guide is fixed to the insertion handle. Both the drill and the locking screws are placed through the guide and into the bone and implant (Figure 1.3).  5  Figure 1.3: Proximal Locking. The insertion handle attached to the proximal end o f the I M N is fitted with an alignment guide for proximal locking (left). Both the drilling and screw insertion are done by placing the screw through the alignment guide and into the bone (right), (source: S y n t h e s ® Universal N a i l System Technique Guide)  1.1.3  Distal Locking  Inserting the IMN often leads to unpredictable deformation of the implant (Pennig 1997, Krettek 1998). This makes it difficult for surgeons to identify the position of the distal locking holes externally and has led to the abandonment of proximally mounted distal aiming jigs (Whatling 2005). The conventional solution involves the use of a fluoroscope to visualize the holes. The surgeon must adjust the positioning of the fluoroscope until the holes appear as perfect circles (Figure 1.4).  Figure 1.4: Distal Locking Under Fluoroscopic Guidance. The fluoroscope must first be positioned such that the locking holes appear as perfect circles ( B ) and not ellipses ( A ) . The drill bit is then aligned such that its tip appears in the centre o f the circle ( C ) and then so that the drill bit appears only in the centre o f the circle, (source: S y n t h e s ® Universal N a i l System Technique G u i d e )  6  Then a radiolucent drill guide must be placed in the correct orientation and drilled under x-ray guidance. Once the first screw is in place, the alignment method must be repeated for the second locking screw.  1.2  Disadvantages of the Current Technique  This method has two major disadvantages: it is not time efficient and it exposes the surgical team to a significant amount of radiation. In an ex vivo study, Krettek found that distal locking of the implant took 4.8 minutes and accounted for over 23% of the total surgical time (Krettek 1999). A study by Suhm revealed that distal locking in vivo was accomplished in a mean time of 27.4 minutes (Suhm 2004). Okcu reported a total surgical time of 141.6± 20.2 minutes for intramedullary nailing of the femur in vivo (Okcu 2003). Distal locking in vivo may therefore account for approximately 20 % of total surgical time.  Distal locking ex vivo required 0.62 minutes of fluoroscopic screening time and accounted for over 53% of the total fluoroscopic screening time of 1.15 minutes for one senior orthopaedic surgeon (Krettek 1999). In vivo total fluoroscopic screening times as low as 0.52 minutes have been reported for senior orthopaedic surgeons (Madan 2002) but are more commonly reported to be in the range of 4 to 7 minutes (Blattert 2004, Mueller 1998). The wide range of screening times is likely due to the complexity of reducing comminuted fractures and the range of experience of orthopaedic surgeons (Blattert 2004, Madan 2002, Hafez 2005).  1.3  Alternative Techniques for Distal Locking  Several alternative locking techniques have been developed in order to address the aforementioned problems. These alternatives include nail mounted mechanical guides, self-  7  locking nails, computer-assisted fluoroscope navigation and magnetic-based targeting systems. Unfortunately, none of these methods have come into widespread use. First generation nail-mounted mechanical guides were unsuccessful due to the difficulty of quantifying nail deformation as it enters the intramedullary canal (Pennig 1997, Krettek 1998, Slomczykowski 1998). Krettek found mean mediolateral and anteroposterior translations of 18.1 and 3.1 mm respectively after insertion of intramedullary nails into cadaveric femora. They also found nail deformations of up to 10° about the longitudinal axis (Krettek 1998). A more recent mechanical guide requires surgeons to drill a 6-10 mm working channel into the cortex of the bone near the screw hole so that screw holes may be palpated and located directly using a guidehook (Figure 1.5) (Krettek 1999). This method is unpopular due to the additional time required and the risks associated with drilling a working channel, which can cause stress concentrations that weaken the bone (Whatling 2005).  guide hook  Figure 1.5: Mechanical Guide for Distal Locking. The use o f this mechanical guide requires the surgeon to drill a 6-10 mm working channel in order to palpate the distal holes using a guide hook. This method has not come into widespread use due to the potential for complications associated with the creation o f a working channel, (source: Krettek 1998)  Self-locking nails have obvious advantages in decreasing both surgery time and radiation exposure. However, the two different types of self-locking nails proposed to date - self-engaging bolts and inflatable nails - have been criticized for their lack of torsional stability and their  potential to induce fracture in compromised bone (Ozturk 2005). These implants are still in the prototype stage but researchers and surgeons have not expressed confidence that this solution will be widely accepted (Whatling 2005). During Fix ion" Nail Insertion  *~imcraan Handle  .  1 Pump  Bone Afte*r Pixjori* N a i l E x p a n s i o n  *—  ^»  •»-—  1 i \ l >h' Nail  Figure 1.6: Inflatable Nail. Insertion o f the Fixion Inflatable N a i l obviates the need for distal locking and therefore reduces surgical time and radiation exposure. This nail has not come into widespread use due to complications associated with inflation and potential lack o f torsional stability, (source: www.disc-o-tech.com)  More recently, virtual fluoroscopy has been developed in which two or more fluoroscopic images are acquired while the position of the patient and C-arm are tracked, typically with an optical position measurement system. The optical tracker also measures the position of surgical tools such as a drill which allows current position of the drill to be displayed on the previouslyacquired x-ray images without requiring additional exposures. This enables distal nail locking to be accomplished with as few as two single-shot x-ray exposures. On the positive side, this method has succeeded in reducing both x-ray exposure time (7.3 vs 108 seconds) and the time necessary for distal locking (13.7 vs 19.9 minutes) (Suhm 2004). However, using the system generally increases overall surgical duration due to the time required for equipment setup (approximately 40 minutes) and tool registration processes (Suhm 2004). Although there are over 500 hospitals worldwide currently using virtual fluoroscopy for intramedullary nailing or other orthopaedic procedures, the relatively high equipment costs ($100,000-250,000) and the additional overall surgical time have limited their widespread use (Hazan 2003).  9  Figure 1.7: Virtual Fluoroscopy Setup. In order to reduce radiation exposure during D L I M N , the fluoroscope ( A ) and surgical tools ( B ) are tracked using an optical tracking system (not shown). This allows the D L I M N procedure to be accomplished with as few as two fluoroscopic shots that are approximately perpendicular, (source: Suhm 2004)  More recently, a novel approach has been proposed which involves the use of a magnet and magnetic field sensor. This radiation-free technique relies on the fact that the implants used for tibial and femoral fracture contain a small open space within the centre of the nail. A permanent magnet on a probe is inserted into this space and is sensed by a magnetic sensor integrated into a drill-mounted guiding system. This system allows the drill to be aligned with the axis of the locking holes by sensing local changes in the magnetic field (Hollstien 2006). A prototype system has recently been licensed to DePuy (Depuy Orthopaedics Inc., Warsaw IN) but has not been marketed commercially. We believe that this concept is promising, but largely limited to this single procedure. It would therefore be difficult to promote this approach to hospitals and surgeons alike.  Figure 1.8: Magnetic Probe System. A permanent magnet is inserted into the central canal of an IMN. Local changes in magnetic field in the area of the locking holes are sensed by an external magnetic sensor fitted with an LED display and a drill sleeve, (source: Szakelyhidi 2002)  In contrast, navigation equipment for a wide variety of computer-assisted orthopaedic procedures is becoming relatively common and its market share will likely continue to increase. We therefore propose to develop a radiation-free targeting process based on a more general-purpose measurement tool that can be used in a variety of surgical procedures. In order to determine the location and orientation of the holes for the distal locking of an intramedullary nail, we require a sensor which can be placed within the nail itself and be detected with sufficient accuracy by an instrument outside the bone.  The use of the Aurora Electromagnetic Tracking System (Northern Digital Inc, Waterloo, ON) has been explored for a variety of surgical procedures including biopsy, radiotherapy, spinal injections, shunt placement and angiography (Krucker 2006, Kirchstein 2005, Wood 2005, Glossop 2002, Levy 2007). Numerous studies examining the distortions caused by common operating room (OR) equipment have been completed with varying results (Schicho 2005, Glossop 2002, Elfring 2007, Wood 2005). The release of a second generation Aurora system designed to minimize the error associated with distortion from metallic objects and the recent development of six degree of freedom sensors with diameters of less than 2.5 mm make this system a likely candidate for intramedullary nail distal locking. A system that can be used for a  wide variety of computer-assisted surgical procedures has distinct economic advantages over a system developed solely for the distal locking procedure.  1.4  Purpose of This Study  The long term goal of this research is to develop and evaluate a computer-assisted technique to achieve distal locking of intramedullary nails using an electromagnetic tracking system.  The first step in this research is to design the method and apparatus necessary to complete our technique. A series of evaluation steps must then be completed before the proposed technique can be used to complete distal locking in a clinical setting (Figure 1.9).  Design of the Method and Apparatus  Accurate in Labbased Testing  Accurate Compared to Conventional Technique  Accurate in Surgical Task Simulation  Randomized Control Trial  Figure 1.9: Testing of a Novel Surgical Method: After design and lab-based testing is complete, it is necessary to test the method and apparatus against the current surgical approach and to complete a surgical task simulation to ensure that the method is accurate enough to be tested in a clinical setting using a randomized controlled trial.  Lab-based testing of the apparatus includes evaluation of the accuracy and repeatability of the tracking method, the ease of use of the apparatus and all safety concerns surrounding the necessary apparatus. Once this has been completed, the proposed technique must be introduced into a surgical setting to evaluate its performance with all the complications that may present in such a setting. Our targeting approach will not be used to complete distal locking at this stage, but data will be collected and compared to the data from the traditional fluoroscopic technique. This will allow the accuracy of our proposed technique to be compared to that of the traditional fluoroscopic method. Once this has been established, surgeons will use the proposed method and apparatus to complete distal locking ex vivo. This surgical task simulation will familiarize the  12  surgeons with all aspects of the method and give a further measure of the accuracy of the technique in a risk-free setting. The final evaluation of the proposed method will be to compare our method to the standard fluoroscopic method using a randomized control trial.  In this project, we have completed the design of the method and apparatus. We have also completed the lab-based testing and have designed the study that will be used in the following step to compare the accuracy of the two distal locking methods.  The main research questions for the lab-based evaluation of the apparatus were:  1. Is the Aurora prototype six degree of freedom (6DOF) sensor accurate and repeatable enough to be used in distal locking of intramedullary nails? Does the presence of a titanium intramedullary nail adversely affect the readings of the 6DOF sensor?  2.  Can the Distal Locking Tool accurately and reproducibly locate the distal locking holes of the intramedullary nail?  3.  Can the distal locking tool be removed from the intramedullary nail easily and without damaging the tool or producing any unwanted debris?  Chapters 5 summarizes the next step in the testing of our device and protocol: the planning of a clinical trial to test the performance of our approach to intramedullary nailing compared to the current technique. The planned apparatus, method, and data analysis for our initial clinical trial are presented in this chapter.  13  2 Materials and Methods In order to complete distal locking of intramedullary nails without the use of fluoroscopy, a navigation technique must be developed that does not require line of sight, can predict the position and orientation of the locking hole axis with high accuracy and repeatability, and can function well in the OR environment. Most surgical navigation is optically-based (Elfring 2007, Ewers 2005) and is not suitable for techniques where line of sight is not available. Magneticbased systems have been developed for use in the OR, but traditionally these systems have had sensors that are too large for our purposes or that could only measure five degrees of freedom (DOF). Recently, Northern Digital (Northern Digital Inc, Waterloo, ON) has developed a prototype 6 DOF miniature electromagnetic sensor for their Aurora Electromagnetic Tracking System. The development of this sensor allows us to incorporate it into the IMN to track the location of the implant in real-time following implantation into the femur. In order to locate the distal locking holes from within the IMN using the 6DOF mini sensors, we designed a device known as a Distal Locking Tool (DLT) to engage with the locking holes and hold the sensor in a fixed position relative to these holes. After calibration, we can then track the position of the locking holes. In this chapter, I describe the testing of the 6 DOF mini-sensors and the design and subsequent testing of the Distal Locking Tool (DLT).  2.1  Aurora Electromagnetic Tracking System  The Aurora Electromagnetic Tracking System (EMTS) was used in this study. This commercially available system consists of a field generator, system control unit, a sensor interface unit and sensors. (Figure 2.1) This system calculates the position of a sensor that measures a magnetic alternating field. Fields arising due to eddy currents created in electrically  14  conductive materials (interference currents) are calculated based on the alternating field and accounted for in the reported readings. (Seiler 2004).  Figure 2.1: Aurora Electromagnetic Tracking System. The system is made up of a control unit, afieldgenerator, a sensor interface unit and a 6 DOF sensor with a diameter of 2.5 mm. The sensor is attached to lead wires and a connector plug which mates with the system interface unit.(source: Northern Digital, www.ndigital.com)  The Aurora system calculates the position and orientation of a tool and returns a transformation containing three rotations (in Euler xyz format) and three translations as well as an indicator value. The indicator value is a dimensionless number which measures the quality of the reading; it ranges from 0 (excellent) to 9.9 (poor). The measurements from the two magnetic coils within the sensor are compared to the known positions of the coils relative to each other and deviations from the expected values are assigned increased indicator values. Large indicator values can indicate magnetic field disturbances or faulty tool definition files. Generally a value of less than 1 is considered acceptable (Aurora User Manual 2005).  Figure 2.2: Aurora Measurement Volume. The measurement volume of the Aurora system is a cube with side lengths of 500 mm. The origin of the system lies on the front face of the field generator with the measurement volume projecting outwards, (source: Aurora User Manual 2005)  15  The measurement volume of the Aurora system is a cube 500 mm on edge (Figure 2.2). The measurement rate of the system when five or fewer sensor coils are being tracked is 45 Hertz (Hz). If six or more sensor coils are simultaneously tracked, the measurement rate is 20 Hz.  Electromagnetic tracking systems allow for navigation without line-of-sight (Kirsch 2005). The first generation Aurora E M T S has been used for several medical applications including biopsy, angiography and radiotherapy (Krucker 2006, Wood 2005, Kirchstein 2005). In 2007, the first commercial system combining the Aurora EMTS with pre or intra-operative C T and ultrasound images was released (PercuNav System, Traxtal Inc., Toronto, Ontario).  The development of prototype 6 DOF miniature sensors make the Aurora a plausible candidate for our envisioned computer-assisted distal locking technique.  2.2  Synthes Universal Femoral Intramedullary Nail  The implant used in this study is a Synthes Universal Femoral I M N (Figure 2.3). This implant is primarily made of titanium (87%) with small amounts of niobium (7%) and aluminum (6%) (Synthes ® Universal Nail System Technique Guide).  Figure 2.3. Synthes Universal Femoral Nail. All of the testing was performed using Synthes Universal Femoral Nails.  The Synthes I M N is a cylinder with an inner diameter of 4.5mm. The outer diameter ranges from 10-15 mm. The implant has a slight bow with a radius of curvature of 1.5 m in order to match the natural anatomy of the human femur. Over 50% of femoral shaft fracture fixations completed at Vancouver General Hospital use Synthes nails or other IMNs with the very similar designs (IM 16  Nails by Year and Diagnosis- 2003-2005, correspondence with Raman Johal, Orthopaedic Trauma Research Coordinator, VCHRI). The similarities in inner diameter, shape and location of the locking holes of all of the commonly used nails (M/DN Femoral Nails, Zimmer and Gamma Femoral Nails, Stryker) will allow us to generalize the technique and testing described for this particular nail to many other manufacturers' products.  2.3  Effect of the Intramedullary Nail  In order to determine the exact effects on sensor accuracy of enclosing the Aurora sensor in a titanium rod, a standard characterization method used at Northern Digital (Kirsch 2005-1) was completed using a bare sensor and then while the sensor was enclosed in an IMN. A l l testing was completed at Northern Digital Inc. (Waterloo, ON). The sensor was affixed to a calibration plug and data was collected from the sensor while it was in each of the alignment holes in a calibration phantom. The calibration phantom is a hemisphere of radius 85 mm with 50 alignment holes designed to cover the sensors workspace. The calibration phantom was placed approximately 250 mm from the field generator. Sensor in Calibration Plug  Figure 2.4: Calibration Phantom. The calibration phantom was placed approximately 250 mm in front of the Field Generator. Data was collected while the calibration plug was in each of the alignment holes. The distance between the calibration plug and the reference plug was calculated and compared to the known locations of alignment holes, (source: Kirsch 2005)  In order to determine the accuracy and precision of the sensor, the calibration plug was placed within each of the 50 alignment holes. Throughout the calibration procedure, readings were also  17  collected from a reference plug which remained stationary in a single alignment hole. In order to determine the accuracy of the sensor, the transformation from the calibration plug to the reference plug was calculated and compared to the known locations of the alignment holes. The absolute locations of the alignment holes were defined using a coordinate measurement machine (CMM) which is accurate to less than 10 microns. The accuracy error is reported for both position and orientation. Position accuracy error was defined as the difference between the radial position vector measured by the sensor and the radial position vector that was pre-registered from the C M M .  The consolidated rms position accuracy error value was then determined.  *".RMS  =  J2.  N  0i-€i)  /N  s  -1  Orientation accuracy error was defined as the angle between the orientation vectors measured by the sensor and pre-registered using the C M M .  Si = arccos(n. •«,„.) r  The consolidated rms orientation accuracy error was then determined. A',  <5RMS =  During the characterization procedure, 30 frames of data were collected in each of the alignment holes. The precision of the sensor was calculated by determining the jitter of the 30 frames of data collected for each hole. The mean position and orientation of each alignment hole was determined and the deviations from the mean locations were calculated. The mean position  18  precision and orientation precision was then determined from the 1500 values calculated (50 holes x 30 frames in each hole).  This characterization procedure was then repeated with the sensor enclosed in the intramedullary nail. The sensor was rigidly fixed in the intramedullary nail using set screws (Figure 2.5). A custom calibration plug was made with a cylinder to house the intramedullary nail. The nail was press fit into the cylinder and then held in place by inserting a small plastic rod through the locking hole (Figure 2.5).  Figure 2.5: Calibration Plug. The sensor was fixed within the IMN using set screws (C). The IMN was rigidly fixed in the calibration plug by pressfittingit into a cylinder attached to the calibration plug (B) and inserting a rod through the distal locking hole (A).  The previously described characterization was then completed to determine the accuracy and precision of the sensor when inside an intramedullary nail. In order to determine the offset of the sensor from the calibration plug, data was collected in 43 of the 50 alignment holes and a best-fit was performed using the known locations of the alignment holes (as determined using the C M M ) . The offset was then applied to the readings prior to determining the vector between the calibration plug and the reference plug. Data could not be collected from the remaining seven alignment holes because the proximal end of the intramedullary nail interfered with the field generator.  19  The probability distribution of position accuracy error and orientation accuracy error were plotted in order to confirm a normal distribution. A t-test was then performed to determine whether there was any significant difference between the errors reported with and without the IMN present. Because the alignment holes closest to the field generator could not be targeted due to I M N making contact with the filed generator and because the sensor was offset from the calibration plug by several centimeters while inside the IMN, the mean distance between the field generator and the sensor was different with and without the IMN. To determine whether the increased distance between the field generator and the sensor was responsible for any decrease in accuracy or precision, the accuracy error and precision error were plotted as a function of distance between the sensor and the field generator.  2.4 2.4.1  Distal Locking Tool Design  In order to orient the E M sensor within the IMN, the D L T was developed. The D L T is designed to slip into the IMN after it has been inserted into the IM canal.  20  Femoral Intramedullary Nail  Figure 2.6: Distal Locking Tool. The custom designed Distal Locking Tool is placed inside of the IMN once the nail has been inserted into the femur, (source: modified from Trigen® IM Nail System Surgical Technique)  The D L T must be placed in the IMN after it has been inserted into the bone because during I M N insertion the implant follows a pre-inserted guide wire to ensure that it remains centered in the I M canal. The guide wire's diameter is approximately equal to the inner diameter of the insertion handle (which is attached to the proximal end of the IMN and removed after the distal locking process is complete) and it is therefore impossible to insert a tool into the I M N while the guide wire is inserted. In order to be inserted after the removal of the guide wire, the D L T must be small enough to fit through the I M N insertion handle. This device has an inner diameter of 3.5 mm.  Once through the insertion handle, the DLT must expand to the inner diameter of the I M N (4.5 mm) so that as soon as it encounters the transverse locking holes of the I M N , it expands even further into the locking holes (Figure 2.7). This allows the DLT to be reliably locked into place relative to the locking holes. In order to achieve this lock, many iterations of the distal locking tool were explored (See Appendix 1: Evolution of Distal Locking Tool Design). In our current design, the distal end of the D L T has an elevated plug that matches the diameter of the locking  holes (Figure 2.7). locking plugs sensor location h = 10 mm 0 3.4 mm  d = 50 mm  +  V  locking arms = 25 mm  Figure 2.7: Design of the Distal Locking Tool. The locking arms are designed to expand when the distal locking tool is inserted and lock into the distal locking holes. The diameter of the distal locking tool is 3.4 mm. The sensor is offsetfromthe locking plugs by 50 mm.  The distal ends of the plugs are cylindrically shaped to prevent the tool from slipping past the locking holes. The proximal ends of the plugs are spherical in order to allow the tool to be removed by simultaneously applying tension and torsion to the proximal end of the D L T (Figure 2.8).  Figure 2.8: Distal Locking Tool Plug. The distal ends (left) of the DLT are designed with plugs to engage the locking holes of the IMN. The distal surface of the plugs are cylindrically shaped so that they do not slide past the locking holes. The proximal (right) surface of the plugs are spherically shaped in order to facilitate removal of the DLT.  22  Sensor Location  Figure 2.9: C-shaped canal. The D L T is fitted with a C-shaped canal to allow the 6DOF sensor to be sealed within the DLT.  The D L T is fitted with a c-shaped canal into which the 6 DOF mini sensor is placed (Figure 2.9). The sensor is laid into this C-shaped canal of the DLT and secured using medical grade adhesive silicone (Silastic Type-A, Dow Corning Inc., Midland, MI).  In order to achieve the desired geometry and dimensions, the DLT was created by fused deposition modeling on a Stratasys Rapid Prototyping Machine (Stratasys Inc., Eden Prairie, MN) using ABS-P400 (Appendix 2: A B S Material Specifications). This bio-inert (Raya 1993) material is non-magnetic and therefore does not affect the functioning of the sensor.  2.4.2  Insertion Repeatability  The repeatability of locking the sensor into the same place within the I M N each time the DLT is inserted was determined by doing multiple insertions while the IMN was rigidly fixed to the benchtop. The D L T was inserted into the distal locking holes and 50 frames of data were collected from the sensor. The DLT was then completely removed from the IMN and reinserted for a total of 50 trials.  The data reported from the system includes the position and Euler rotation angles of the sensor relative to the global reference frame. In order to relate these transforms to a more meaningful 23  frame, a nail frame was created. The nail frame was created by digitizing the locations of the distal holes while the I M N was affixed to the tabletop. A n Aurora Probe (Traxtal Inc., Toronto, Ontario) (Figure 2.10) was inserted into the locking hole as shown in Figure 2.11.  Figure 2.10: Digitizing Probe. The probe reports the tip location in 6 DOF. The tip of the probe is chamfered.  Figure 2.11: Digitizing the Locking Holes. The locking holes were digitized by placing the chamfered tip of the probe into the holes until the probe made contact with the rim of the locking hole.  The tip of the probe is chamfered and it was inserted until the neck of the probe made contact with the locking holes. The locations of the outer surface of both distal locking holes on both sides of the nail were digitized (Figure 2.12).  24  Figure 2.12: Nail Frame Construction 1. Four points on the IMN were digitized in order to create the nail frame, the medial and lateral surfaces of the most distal locking hole (pi and p2) and the medial and lateral surfaces of the more proximal distal locking hole (p3 and p4).  The centre of each locking hole was determined by finding the midpoint between the points collected (Figure 2.13): M i = (p2-pl)/2  M = (p3-p4)/2 2  Figure 2.13: Nail Frame Construction 2. The centers of the locking holes were defined as the midpoints between the two sides of the locking holes.  The origin of the nail frame was defined to be M . The nail frame x-axis was defined on the line 2  between the two midpoints: 25  X = M2-M1 X * = TH  This vector was then normalized:  rl  The z-axis was defined by creating a vector from the midpoint of the more proximal locking hole to p4:  z = 7^7  z = p4-M2,  rl  The y-axis was then defined as the cross product of these two unit vectors: A  A  A  In order to ensure that all three axes are orthogonal, the final z-axis was found by taking the cross product of the x and y axes: A  A  A  z = xxy The rotation matrix from the transform from the global frame (O) to the nail frame (N) was then created with the unit vectors of the three axes.:  N0 =  R  The homogeneous Transform was then created: ^NO  0  ^NO  1  where RNO is the rotation matrix above and d>io is the displacement from  origin of nail frame to origin of the global frame expressed in nail frame coordinates. As previously mentioned, the origin was defined as M2. This origin is reported by the system in global coordinates and must therefore be translated: dNo  =  - RNO*doi  The data reported by the Aurora system during the insertion trials included the positions (Tx, Ty, Tz) and orientations (Rx, Ry, Rz) of the sensor. 50 frames of data were collected for each of the 26  50 insertion trials. This data was imported into Matlab and the 6 variables were translated into a homogenous transform from the global frame (O) to the sensor frame (S). This was defined as Tosi. From this variable, the mean transform from the global frame to the sensor frame was defined (T $). 0  Both of these variables were transformed into the nail frame by applying the transform from the global frame into the nail frame: TNS  TNO  =  ' Tos  TNSJ  =  TNO  ' Tosi  In order to determine the deviation of each measurement from the mean location, the transform from the mean location (TNS) to each measurement (TNSD was calculated: TNNJ  =  TNS ' (TNSO '  The positions (Tx, Ty, Tz) and orientations (Rx, Ry, Rz) of the deviations from the mean were then extracted from the homogeneous transform (TNNJ) and the standard deviations of these deviations were determined.  2.5  Mechanical Testing  After the accuracy and repeatability of the Aurora sensor and distal locking tool were verified, several basic mechanical tests were conducted. The purpose of these tests was not to define the actual values of various mechanical properties of our tool, but to gain insight into whether or not any particular responses of the tool should be of concern to us. Our testing was designed to determine whether the distal locking tool could be easily removed from the intramedullary nail, to determine whether the stresses produced during that removal were near the yield stress of the material, and to determine whether distal locking tool insertion and removal would cause particulate formation that may be harmful to patients.  27  2.5.1  Experimental Moment to Remove DLT  In order to remove the distal locking tool from the intramedullary nail the proximal end of the distal locking tool must be twisted and pulled away from the proximal end of the nail. We were interested in how difficult this removal process is. In order to quantify ease of removal, a simplified release mechanism which incorporates only twist was modeled. The I M N was fixed to a calibration box to keep it from moving or rotating and the D L T was inserted. The D L T was supported at the proximal end by a support structure and a cylinder was attached to the D L T between the proximal end of the IMN and the support structure. A rope was rigidly fixed to the cylinder and then wrapped around the cylinder multiple times. This rope then descended straight down through a hole in the calibration box (Figure 2.14). A starting mass of 60 g-which was shown in preliminary tests not to release the DLT- was applied and then mass was added in 5 g increments until the D L T released from the locking holes (as evidenced by an audible noise and a large drop of the weight).  Figure 2.14: Torque Necessary to Remove DLT. The IMN was fixed in the calibration box and the DLT was inserted. The proximal end of the DLT was threaded through a cylindrical attachment and supported at the proximal end. A rope was wrapped around the cylinder and threaded through the base of the calibration box. Weights with increasing mass were attached to the rope until the DLT was releasedfromthe locking holes.  The total mass required to release the D L T was recorded. This process was completed three times per distal locking tool and tested on 5 separate tools. 28  The diameter of the cylinder was measured and the moment applied to the distal locking tool was calculated: M = rxF where r is the radius of the cylinder and F is the force produced by the mass shown in 2.14.  2.5.2  Maximum Moment  In order to determine how easily the distal locking tool can be removed, we compared the moment required to release the tool to the maximum moment that two researchers could produce. It was determined in preliminary testing that a rotation of the proximal end of the distal locking tool of approximately 360° was necessary in order to cause release from the distal locking holes. Moments were produced by placing the distal locking tool in the set-up shown in Figure 2.14 and then twisting the cylinder 360° using only the thumb and forefinger. Mass was added to the string in 100 g increments until it was no longer possible to move the cylinder 360°. The largest mass that was moved through the required angle was used to determine the maximum moment that can be produced in this set-up. The percentage of the maximum moment that is required to release the distal locking tool was then calculated.  2.5.3  Stress Analysis  The actual mechanism of release of the distal locking tool from the distal locking holes likely involves a complex combination of bending and twisting of the locking arms. In order to obtain a simple measure of the stress produced in the distal locking tool during removal, the theoretical force necessary to compress the locking arms was determined and the stress produced in the locking plugs was calculated (See Appendix 3: Theoretical Force to Remove the DLT). This was accomplished by modeling the locking arms as simple beams with constant cross sections. In 29  order to gain some insight into the stress that would be produced in a model with more realistic geometry and determine whether this stress is anywhere near the yield stress of the material, the C A D file used for manufacture of the distal locking tool was imported into finite element analysis software (COMSOL 3.3). The boundary conditions were set such that the distal locking plugs were fixed while the rest of the D L T was free to move. The tool was meshed with 289,540 tetrahedral elements and contained 279,954 degrees of freedom (Figure 2.15).  Figure 2.15: F E M mesh for the distal locking tool. The DLT was meshed with 289,540 tetrahedral elements and 279,954 degrees of freedom.  The mean moment necessary to release the distal locking tool (as determined in the tests described in section 2.5.1) was applied to the proximal end of the DLT and the resulting von Mises stress was predicted. The maximum von Mises stress was compared to the yield stress of the material to determine whether failure by yielding is likely in this loading scenario.  2.5.4  Particulate Formation  It was observed early in testing that the insertion and removal of the DLT, particularly the first few times it is inserted, leaves behind small particles of A B S on the implant. To determine  30  whether or not the quantity and size of these particles were such that there might be a concern about future osteolysis i f some particles remained in the patient, a series of tests were performed. The I M N was thoroughly cleaned, dried, and inserted into the calibration box. A D L T that had never previously been inserted into an I M N was inserted and removed a total of three times. After the third removal, the distal locking hole and distal portion of the I M N inner canal were washed with 1 ml of distilled water using a small pipette. This wash was collected into a small vial held below the locking hole or distal end of the IMN. Any remaining water was forced into the vial using high pressured air. In order to quantify the size and number of particles left on the IMN, the solution was examined microscopically. The vial was vigorously shaken to ensure that there was a good mechanical mixture of the solution and then 0.25 ml was removed using a pipette. This solution was placed on a slide and covered with a coverslip. The slide was observed using the 40x objective lens and lOx ocular lens for a total magnification of 400 times. The slide was positioned such that the leftmost margin of the sample was visible. Both the number and the relative size of particles were recorded as the slide was incrementally advanced to the far sample margin. Particles were classified according to their longest edge length. Very small particles were less than 10 um, small particles were 10-20 um, medium particles were 20-50 um and large particles had edge lengths of over 50 um.  This process was repeated for a second row using the same slide. Using the microscope-mounted digital camera and the appropriate software (Fujifilm EXTouch and Fujifilm Photograb-300Z), representative photos were taken of each of the sizes of particles. In order to quantify their size, a micrometer slide was also photographed under the same conditions. The size of representative samples was determined by comparison with the micrometer slides. This method was repeated for a second DLT.  31  In order to test the effect of original surface roughness on particle formation, a second pair of DLTs were filed and sanded using a fine file and sandpaper prior to insertion. They were then washed thoroughly with water, allowed to dry, and the method described above was used to collect particles resulting from three insertions.  32  3 Results 3.1  Effect of the Intramedullary Nail  The results of the calibration testing with and without the IMN are shown in Figure 3.1. 0.8 0.7  • •  N o Nail  •1  W i t h Nail  0.6  o-  Position  Accuracy  Orientation  Position  Precision  Orientation  Figure 3.1: Accuracy of the Aurora sensor with and without the IMN. The accuracy of the Aurora sensor in standard conditions was 0.41± 0.12 mm and 0.38± 0.28°. With the IMN, the accuracy was 0.52 ± 0.16 mm and 0.43 ± 0.20°. Precision error of the readings increased in the presence of the IMN.  The mean position accuracy error with and without the I M N were 0.41± 0.12 mm and 0.52 ± 0.16 mm respectively. This difference was significant (p = 0.006). The orientation accuracy error reported for the two cases were 0.38 ± 0.28° and 0.43 ± 0.20°. The effect of the I M N on screw angle orientation trueness was not significant (p = 0.41).  The cumulative error distributions for distance and orientation trueness are shown in Figure 3.2. The probability distributions for the errors are shown in Figure 3.3.  33  ion  45 40 35  /  J  /J  /  /  /  /S  f  f  30  -  /^  GO 25  b  J  20  > TO15  1  c  E •  o  f I /  10 5'  / /  • //  /  •  0  3D Error ( m m ) S c r e w Angle Error (degrees) Figure 3.2: Cumulative Distribution of Accuracy Errors. Results for positional readings (left) taken without the IMN (-) and with the IMN (--) show that a distortion of approximately 0.1 mm in induced by the presence of the nail. This difference is statistically significant (p = 0.006) but not clinically relevant. Orientation errors (right) were not significantly different (p = 0.41).  3D Error (mm)  Screw Angle Error (degrees)  Figure 3.3: Probability Distribution for Accuracy Errors. Positional errors (left) without the IMN (-) and with the IMN (--) show 95% confidence intervals or 0.06 and 0.05 respectively. Orientation errors (right) have 95% confidence intervals of 0.12° and 0.16° without and with the IMN.  The 95% confidence intervals for position and orientation for the tests with and without the I M N are shown in Table 3.1. With the I M N in place, the 95% confidence levels are 0.57 mm and 0.55°. Table 3.1: Summary of Results for Accuracy Testing.  Positional 3D Error (mm)  Screw A x i s Angle Error (°)  Mean  Std  95% C.I.  W i t h o u t Nail  0.41  0.12  0.06  With Nail  0.52  0.16  0.05  W i t h o u t Nail  0.39  0.28  0.16  With Nail  0.43  0.20  0.12  The precision reported when no I M N is present (0.072 mm and 0.046°) was significantly different from the reading taken with the I M N (0.192 mm and 0.145°). 34  The position accuracy and precision are shown as a function of distance to the field generator Figures 3.4 and 3.5 respectively.  4 With IMN No IMN  3.5 3-h w 2.5 4 o n 2  i  o u 1.5 < c 5 1 (/> o Q_ 0.5  —-  0  100  -  200  300  400  Radial Distance From Sensor to Field Generator Figure 3.4: Position Accuracy E r r o r does not change as a function of distance from the sensor to the field generator.  0.16 With IMN  0.14  1 0.12 c o w  Without IMN  0.1  § 0.08  I  0 0 6  § 0.04 Q.  0.02  4" ^  00  100  200  Ai  300  400  Radial Distance From Sensor to Field Generator Figure 3.5: Position precision error increases with increasing sensor distance from the field generator.  3.2  Insertion Repeatability  Repeatability of distal locking tool insertions was well within the acceptable limits. The largest differences from the mean value were seen in the z-direction, with a standard deviation over 50 trials of 0.20 mm. The standard deviations in x and y directions were 0.03 and 0.12 mm respectively (Figure 3.5). The largest rotational deviations from the mean were in the x direction (1.59°). The rotational deviations in y and z had standard deviations of 0.13° and 0.08° respectively (Figure 3.6). 0.4 0.2 E~ £  0  c o  |-0.2 CD Q  -0.4 -0.6  Tx  Ty  Tz  Figure 3.6: Mean Deviations for Translations. Boxplots showing the mean, +/- standard deviation, min and max for each of the cardinal directions. Translational errors were very low with standard deviations of 0.03, 0.12 and 0.20 mm in the x (longitudinal axis of the nail), y (anteroposterior axis), and z (mediolateral axis) directions.  ~  2  |  0  O) CD TD '>  CD Q -2  -4  Rx  Ry  Rz  Figure 3.7: Mean Deviations for Rotations. Boxplots showing the mean, +/- standard deviation, min and max for rotations about each of the cardinal directions. Rotational errors had standard deviations of 1.59°, 0.13° and 0.08° about the x (longitudinal axis of the nail), y (anteroposterior axis), and z (mediolateral axis) axes respectively.  36  3.3  3.3.1  Mechanical Testing  DLT Removal Moment  The average moment necessary to remove the D L T from the intramedullary nail was 0.011 ± 0.001 N m (Table 3.2). The first tool tested had several tests applied to it prior to the final protocol being determined. The mean removal moment for this tool was 0.008 Nm. We observed over the course of all testing that the moment required to remove the D L T decreased with increasing numbers of insertions and removals. This D L T was not included in the analysis.  Table 3.2: Moment Required to Remove the  Tool  1  2  3  4  Mean  Trial 1 2 3 1 2 3 1 2 3 1 2 3  Moment (Nm) 0.010 0.009 0.009 0.011 0.011 0.011 0.011 0.011 0.010 0.011 0.011 0.011 0.011  The maximum moment that could be produced using the set-up for the removal tests ranged from 0.129 to 0.172 Nm. The moment necessary to remove the distal locking tool is approximately 7.5% of maximum moment that can be applied by the thumb and forefinger.  37  3.3.2  Stress Analysis  The maximum Von Mises stress predicted when a moment of 0.011 Nm was applied to the proximal end of the tool was 8.5 x 10 Pa. This was seen at the junction of the locking plugs and 4  the locking arms.  Figure 3.8: Finite Element Model of the Distal Locking Tool. The maximum von Mises stress reportedfromthe model was 8.5 TO Pa and was located at the junction of the locking arms and the locking plugs. 4  The yield stress of the A B S plastic used in the construction of the D L T is 41MPa. The stress predicted for the removal of the distal locking tool is approximately 1/500 of the yield stress.  38  3.3.3  Particulate Formation  Particles were classified according to their longest edge length. Very small particles were less than 10 um, small particles were 10-20 um, medium particles were 20-50 um and large particles had edge lengths of over 50 p.m (Figure 3.9, 3.10). The smallest particle recorded had an edge length of 1.2 um while the average particle size was 8.2 um. The distal locking tools that were not filed showed a mean of 18 very small, 12.5 small, 5 medium, and 0.25 large particles for a total of 36 ± 19 particles per count. For the devices that were filed prior to insertion, a mean of 3.75 very small, 2.25 small, 2 medium, and 0.25 large particles were observed. The mean total number of particles for a single count was 8.25 ± 2.87. The number of particles produced was reduced by 77% when the device was filed before use.  H< 10 (jm •  10-20  0 20-50 pm S >50 pm  1a  1b  2a  2b  3a  3b  4a  4b  Trial Figure 3.9: Number of Particles Produced. The number of particles produced was reduced by 77% for devices that were filed (trials 3-4) compared to devices that were not filed (trials 1-2).  39  Figure 3.10: Particulate Samples. Very small (A), small (B), medium (C), and large (D) particles had maximum edge lengths of <10 urn, 10-20 \im, 20-50 um and >50 um respectively.  The number of particles for each count was extrapolated to the entire solution collected from each insertion trial. The mean total number of particles seen for three insertions and removals of  40  an unfiled.distal locking tool was 8,580 ± 4,330. For the filed tool the mean total number of particles was 1,980 ± 7 6 0 .  41  4 Discussion In a previous study, the maximum error tolerance for locking screw alignment in intramedullary nailing was defined as 0.75 mm and 8° (Figure 4.1) (Szakelyhidi 2002).  0.75 mm maximum lateral tolerance  IMN  Screw  Figure 4.1: Maximum E r r o r Tolerance for Distal Hole Targeting. The maximum error  tolerance for targeting distal locking holes has been previously reported as 0.75 mm and 8°. (source Szakelyhidi 2002)  It is the experience of our clinical collaborators that the actual lateral tolerance is significantly larger than this. As long as the drill tip is within the 5 mm diameter locking hole (less then 2.5 mm from the hole centre), the drill bit will be guided into the hole with minimal damage to the nail. During the distal locking of the first hole, the intramedullary nail is still free to move relative to the bone, small shifts of the intramedullary nail are therefore possible. The upper bound of the 95% confidence limit for lateral errors found in our testing (0.57 mm) falls below this stringent cut-off described in this previous study (0.75 mm) and well below the clinically consequential level of 2.5 mm. The equivalent error for orientation was found to be 0.55°, well below the 8° tolerance.  42  4.1  Effects of the Intramedullary Nail  Enclosing the prototype 6 DOF sensor in a titanium intramedullary nail changes the accuracy by a detectable amount but likely not by a clinically consequential amount. The difference between radial position (3D rms) and screw angle errors for readings with and without the IMN were only 0.1 mm and 0.05°.  In a 2005 study of the susceptibility of the Aurora system to the presence of various metallic objects in the sensing volume, a titanium rod was among the materials tested. For these tests, the rod was placed in different locations relative to the sensor and field generator. The results obtained were very similar to those reported in our testing: the differences in mean error with and without the titanium rod were approximately 0.2 mm and 0.025° (Kirsch 2005-2). This study is consistent with our conclusion that the IMN will not degrade the readings of the Aurora sensor by any clinically significant amount.  Although the tracking of the Aurora system is not affected by the intramedullary nail, interactions with other intra-operative equipment such as surgical tables, imaging equipment and surgical tools may also affect the accuracy of the system. A report by the manufacturer of the Aurora system showed that sheets of stainless steel, designed to mimic surgical tables, introduced distortions of up to 5 mm and 5° at distances of 70 mm from the sensor when the first generation Aurora system (Aurora 1) was used (Kirsch 2005-2). However, when the distance between the sensor and the sheet was increased to 270 mm, the distortions decreased to 0.7 mm and 0.5°. Repeating the same test for the Aurora system used in our study (Second generationAurora 2) showed distortions of 0.7 mm and 0.8° when the steel sheet was 70 mm from the sensor and 0.3 mm and 0.5° when the sheet to sensor distance was 270 mm (Kirsch 2005-2). The surgical table is therefore unlikely to be a problem using the Aurora 2 system. 43  The use of specialized orthopaedic fracture tables can also reduce the probability of table-related distortion. Orthopaedic fracture tables are designed to be radiolucent and therefore have carbon fiber or composite table tops and radiolucent leg stabilizers (Steris Corporation, vvww.steris.com) (Figure 4.2). These tables further increase the distance between the electromagnetic sensors and the steel table frame, which would decrease the interference felt by the sensor.  Figure 4.2: Orthopaedic Fracture Table. Surgical tables designed for orthopaedic fracture fixation are designed to be radiolucent. The composite table top and abductor bars decrease the amount of steel near the sensing area thereby decreasing the interference for electromagnetic tracking, (source: www.steris.com)  It is well documented that the Aurora system's performance depends on the distance between the sensor and any sources of distortion, as discussed above and also on the distance between sensors and the field generator (Kirsch 2005-1). The Aurora 2 system has optimal accuracy when sensorfield generator distance is less than 400 mm (Figure 4.3).  44  Ill o  CO  1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 100  Trueness Aurora 2 Precision Auroras Trueness Aurora 1 Precision Aurora 1  200  12? -x-  300 400 500 600 Sensor Distance [mm]  700  Figure 4.3: Accuracy (shown here as Trueness) of the Aurora 2 system is optimal at d <  400mm. Precision of the Aurora readings also decreases for distances over 400 mm. (source: Kirsch 2005-1)  During surgery we will aim to keep the sensor-field generator distance below 400 mm and the sensor-surgical table distance over 270 mm in order to optimize the accuracy of the sensor.  The interference of imaging machines has been explored in several studies. Wood et al found that tracking directly inside a CT gantry caused some distortion but there was no significant distortion (defined as a shift of over 1 mm) if the image intensifier was over 100 mm away from the Aurora sensors (Wood 2005). There was also no significant effect associated with turning the x-ray tube on. This study was performed using an Aurora 1 system and we would therefore expect even better results using the Aurora 2. Schicho et al examined the effect of an ultrasonic scanhead on Aurora 1 readings. In this study, the scanhead was placed at distances of 10-60 mm from the sensor. The maximum distortion was found to be 2.6 mm (Schicho 2005). Schicho et al. also studied the effects of 2 surgical instruments: a dental drill and a steel dental hook. The maximum distortions caused by these instruments at distances of 10-60 mm were 1.2 and 2.1 mm respectively. Results for distances of over 100 mm were not obtained.  45  To simulate surgical tools, Kirsch introduced aluminum and ferromagnetic stainless steel into the sensing volume (Kirsch 2005-2). In this study, distortion increased significantly as the rod was moved closer to the sensor. The aluminum rod showed distortions of up to 2 mm for distances of fewer than 25 mm and distortions of approximately 0.25 mm for distances of 25 to 200 mm. The stainless steel rods caused distortions of up to 10 mm at distances of less than 25 mm and distortions ranging from 1 to 3.5 mm for distances of 25 to 200mm. The rods used in this study were fairly large (200 mm length and 15 mm diameter) and are therefore unlike tools we would expect to encounter in surgery, but the results of the Kirsch study emphasize the need to test our device in a live surgical environment, incorporating all possible sources of distortion, prior to judging the absolute error associated with our distal locking method.  4.2  Distal Locking Tool Accuracy  Our custom-built distal locking tool can reproducibly target the distal locking holes from within the IMN. The largest translational standard deviation was 0.20 mm and the largest rotational standard deviation was 1.59°.  To estimate the variability associated with identifying the locations of the locking holes we need to combine the variability arising from both the sensor's position in the measurement volume and the variability due to targeting the locking holes using the DLT. Table 4.1: Standard Deviations for Sensor Accuracy and Insertion Repeatability for all 6 DOF Rz Rx Ry Tx Tz Ty (mm) (mm) (mm) (°) O O 0.25 0.32 Sensor Std 0.37 0.27 0.25 0.40 0.13 0.08 Insertion Std 0.12 0.20 1.59 0.03  46  It is important to note that the deviations for the sensor accuracy tests are reported in the global reference frame whereas the insertion repeatability deviations are reported in the nail frame. If we consider the worst case scenario for translational error, the nail frame z-axis (std = 0.20 mm) will coincide with the global frame x-axis (std = 0.40 mm). By summing the variances we find that the standard deviation is approximately 0.45 mm in that direction. Similarly, if the nail frame x-axis were aligned with the global z-axis, the standard deviation for the rotation about this axis would be approximately 1.62°.  For the clinical trial which will be described in detail in chapter 5, we will be comparing the readings from the distal locking tool to a second Aurora sensor attached to a tool used to determine the drilling axis for distal locking (drilling tool). This second sensor would add an additional error to the overall reading. If we once again assume the worst possible alignment in terms of error, the standard deviation for the combination of these three sources of error in one dimension would be approximately 0.6 mm or 1.65°.  If we extend this worst case scenario to three dimensions, rms translational standard deviation would be approximately 0.89 mm and 1.7°. These deviations are still below our clinically consequential tolerances for targeting the distal locking holes. Tracking the DLT relative to the drilling tool is therefore likely to be accurate enough for our purposes.  The minimum translational deviation (0.03 mm) that we observed was along the longitudinal (x) axis of the nail (roughly equal to the longitudinal axis of the femur). Due to the design of the DLT, there is very little freedom to move on this axis. The deviations in the y and z axes (0.12 and 0.20 mm) were larger.  47  The D L T was designed to fit through the insertion handle used during surgery (inner diameter 3.5 mm). The inner diameter of the I M N is 4.2 mm, which means that the DLT was not tightly restrained laterally during this test; there was therefore freedom to move along the anteroposterior (y) and mediolateral (z) axes. Although the presence of the insertion handle will largely restrict the possible y and x axis movements of the proximal end of the DLT, there could still be error associated with the bending or twisting of the DLT between the sensor and the locking holes. Currently, the locking arms are 25 mm long, as shown in Figure 4.5. This design was adopted in order to allow for a locking arm that will expand upon intersection with the locking holes but is flexible enough to be compressed easily to facilitate insertion and removal from the implant. There is an additional 25 mm between the end of the locking holes and the beginning of the C-channel where the sensor lies. This transition zone was included in the design in order to ensure adequate torsional strength for transmitting the moment used to remove the DLT. In the current design, the total distance between the locking holes and sensor is 50 mm.  Figure 4.4: The total distance between the sensor and locking plugs is 50mm. This distance  is made up of the locking arms (shown above) and the transition zone. Decreasing this total distance may decrease the error associated with predicting the locking hole axis from the sensor location.  As will be discussed later, the tool design can likely be optimized to reduce the separation between sensor and locking holes; the stresses associated with insertion and removal of the DLT are well below critical failure values. The D L T ' s design could therefore be altered by decreasing the length of the transition zone and increasing the torsional stiffness of the locking arms. Both 48  of these design changes would serve to limit the errors associated with rotation and deviation of the tool relative to the locking holes. The torsional stiffness could possibly be increased by modifying the geometry of the locking arms or possibly by using a stiffer material.  Deflections of the DLT between the sensor and the locking tips could potentially be exacerbated by the presence of debris in the inner canal of the implant. Fluids and bony debris are normally forced up into the IMN during insertion into the femur. We don't anticipate that this will be a major problem because during insertion, a 3.5 mm guide wire is present. When the guide wire is removed, many bony particles are also removed. When the DLT (diameter 3.4 mm) is then inserted, will not fill the inner diameter of the nail (4.5 mm). Fluids and small bone particles will therefore be able to pass around the outside of the tool. If it does become problematic to advance to DLT to the distal locking holes, it might be necessary to clean out the canal prior to DLT insertion by using pressurized water or a probing tool, for example. The difficulties associated with this potential challenge will be evaluated during our preliminary clinical trial, the planning of which will be discussed in Chapter 5.  Several other studies using the Aurora system have reported accuracy errors significantly larger than the 0.63 mm error (95% C.L.) found in our study (Levy 2007, Krucker 2006, Wood 2005, Schicho 2005, Glossop 2002, Elfring 2007). Errors reported in these studies range from 1.28 mm to 5.8 mm. However, there are several differences between these previous studies and our own. Firstly, with the exception of Elfring, all the studies mentioned above used the first generation Aurora. The second generation Aurora system was developed in response to the high error associated with distortion from metallic objects and was released in September of 2005. Improved accuracy in a surgical setting could therefore be expected if the other studies were repeated using the Aurora 2. Elfring's results showed mean distortions of 2 mm caused by a 49  surgical environment. However, they did not clarify what type of surgical instruments were introduced into the field nor at what distances the objects were placed relative to the field generator or sensors.  Secondly, as previously mentioned, the use of radiolucent surgical tables decreases the interference caused by the surgical table. Wood tested both angiography and CT tables and found mean errors of 1 mm and 2.5 mm respectively (Wood 2005). Intramedullary nailing of the femur is routinely performed on a radiolucent surgical table to allow for fluoroscopic guidance. We can therefore expected results similar to, or better than Wood's study.  Thirdly, another key variable between various electromagnetic tracking studies is how errors are reported. FCrucker reports on the accuracy of the Aurora-2 combined with pre-operative CT and intra-operative ultrasound for guidance in soft-tissue biopsy and radio-frequency ablation. The study was designed as a bystander study using CT tracking for intra-operative guidance and comparing it to the electromagnetic/ultrasonic tracking. All 20 procedures were carried out successfully and without any clinical complications. The mean tracking accuracy for the Aurora system was found to be 5.8 ± 2.6 mm. The researchers noted a strong correlation between the distance from the CT surgical table and decreased accuracy, (see Figure 4.5)  50  2  I  ~M-  -j-  - j -  i  Distance from table (mm)  Figure 4.5 Tracking accuracy decreases with decreased distance between the sensor and a C T table (source: Krucker 2006)  The authors suspect that this distortion may be the largest contributor to error in their clinical evaluation. The use of the second generation Aurora system may therefore lead to significantly better accuracy results. The authors also cite patient motion as another important source of error. In their guidance procedure, a significant difference in tracking accuracy was found between tracking in the lungs and in the liver. Respiratory motion as well as gross patient motion can be compensated for using skin trackers but needs to be further investigated (Krucker 2006).  The maximum accuracy errors of 5.8 mm reported in this study are over our suspected clinical tolerance however, the D L I M N procedure does not share the two most significant sources of error in this study since intramedullary nailing would take place on a radiolucent table and patient motion would be irrelevant because the sensor would be located within the patient and not registered to any pre-acquired images. Computer-assisted navigation in D L I M N is therefore very likely to produce acceptable clinical results.  51  4.3  Ease and Safety of Distal Locking Tool Design  The moment required to remove the DLT from the nail was 0.011 ± 0.001 Nm, approximately 7% of the maximum moment that can be applied using the thumb and index finger to rotate the DLT. When the moment required to remove the tool was applied to the proximal end of the tool in a finite element analysis, the resulting Von Mises stress was predicted to be highest at the junction of the locking plugs and locking arms, with a value of 8.5 -10 Pa. This is approximately 4  l/500 of the yield stress of the ABS used, so we conclude that the moment required to remove th  the tool from the IMN will not cause the DLT to fail by yielding. It should be noted that the location of maximum stress was predicted to be at the boundary between the portion of the tool that was rigidly fixed (locking plugs) and the portion that was free to move (locking arms). In reality, the locking plugs will not be held rigidly in place. The deformation caused by the applied moment will therefore be distributed throughout the locking plugs and arms. The predicted value of maximum stress from this model is therefore likely to overestimate the actual maximum stress seen during removal.  4.4  Optimizing the Design  As previously mentioned, the design could be further optimized by minimizing the separation between the locking plugs and the sensor. Since the stresses induced throughout the D L T are so far below the yield failure values (1/500 ), we conclude that much of the solid portion of the th  transmission zone could be removed without introducing critical stress values. Although it is difficult to determine the actual mechanism involved in releasing the DLT from the locking holes, it is likely that the moment applied induces a lateral force on the locking holes which, through interactions with the side of the holes, likely produces a combination of twisting and 52  bending of the locking arms that results in releasing them from the locking holes. Changing the bending and torsional stiffness of the locking arms would therefore control the moment necessary to release the tool. The theoretical force necessary to bend the locking arms sufficiently is approximately 2.6 N (See Appendix 3: Theoretical Force to Remove DLT). If this stress is applied through the surface of the locking plug, the stress induced would be approximately 15 • 10 Pa. This is similar to the 8.5 • 10 Pa stress predicted by the F E M . As 4  4  previously discussed, the repeatability error could be further reduced by increasing the torsional stiffness of the locking arms. It is important, however, that we do not compromise the ease of locking the D L T in place in an effort to increase torsional stiffness. Currently, removal of the D L T only requires 7% of the maximum producible moment; it is therefore reasonable to assume that small increases in bending stiffness of the locking arms (through material choice or arm length or cross-sectional geometry changes) would not have an adverse effect on the locking procedure. However, increasing the bending stiffness of the locking arms would increase the friction felt between the I M N and D L T during insertion. This may make it more difficult to target the distal locking holes intra-operatively. We conclude that the 50 mm separation between locking plugs and sensor could be reduced to 20 mm by decreasing the length of the transition zone and reducing the length of the locking arms without affecting the locking procedure but otherwise, we believe that the design is near-optimal. If further design developments are required, all of these parameters must be taken into account. The finite element model created for the evaluation of the D L T may be expanded to allow parametric analysis of any potential design alterations to be accomplished quickly and easily.  53  4.5  Particulate Formation  The presence of bio-inert particles in the tissues surrounding implants has long been associated with osteolysis and subsequent implant loosening so we must ensure that any particles left in the patient during use of the DLT are unlikely to cause harm. This phenomenon has been extensively studied in relation to Total Hip Arthroplasty (THA) (Revell 1997, Green 2000, Ingham 1999, Ingham 2005, Howie 1993, Schmalzreid 1994, Nashed 1995).  In the presence of wear particles, macrophages will attempt to phagocytose small particles (<10 um) or surround larger particles (>10 um) in order to degrade the foreign object. For bio-inert materials, the macrophages will be unable to degrade the particles (Ingham 1999). As a result, additional macrophages will be recruited, fuse, and create multi-nucleated giant cells. These giant cells form a barrier between the foreign particle and the surrounding tissue to avoid any potential threat (Ingham 1999). These activated macrophages initiate a foreign body reaction that consists of the release of cytokines and other inflammatory mediators which stimulate the recruitment of osteoclasts. This process ultimately leads to osteolysis and eventual loosening of the implant (Ingham 2005).  Although bone cement (PMMA) and metal particles (titanium, cobalt-chrome) have been shown to cause inflammatory reaction and osteolysis, the evidence is very strong that osteolysis is primarily related to the Ultra High Molecular Weight Polyethylene (UHMWPE) particle generation in T H A (Ingham 2005).  The extent of the inflammatory reaction and subsequent osteolysis due to wear particles depends on the number of particles present. Osteolysis is likely to occur in T H A when over H O  1 0  particles are present per gram of tissue (Revell 1997). Repeated insertion and removal of the unfiled distal locking tool resulted in a maximum of 11,640 particles, only 0.0001% of this threshold for a single gram of tissue. The filed DLTs produced a maximum of 2,520 particles or 0.00003%) of the critical number. It is therefore be very unlikely that the number of particles associated with the use of the D L T could lead to osteolysis.  The size and concentration of particles have also been shown to be determinants in bioreactivity. Micrometer and submicrometer sized particles are stimulatory (Archibeck 2000, Jacobs 2006). In particular, U H M W P E particles with a mean size of 0.24 urn have been shown to stimulate bone resorption in vitro at a concentration of approximately lOpmVcell. Larger particles (0.45 and 1.71 um) required a concentration of 100 um /cell to induce bone resorption (Green 2000). 3  Particles larger than this did not stimulate macrophage induced bone resorption.  It is reasonable to assume that the biological reaction to A B S particles is analogous to that of U H M W P E . In vivo and in vitro testing of A B S biocompatibility has shown the 'best results possible' (Raya 1993). Since both A B S and U H M W P E are bio-inert, the reactions of similar sizes and concentrations of particles are likely comparable. The smallest recorded particle for our testing was 1.2 um but the average was 8.2 urn. There was only a single particle seen that was under 2 urn in largest side length (1.2 um). The surface area of this particle was approximately 1 pm. If we assume the particle was approximately 0.5 um thick, this gives a particle volume of 0.5 um. In order to reach concentrations of 100 um /cell, 200 such particles would have to be 3  present in one cell. Since we only recorded one particle of this size and the since the particles will likely be dispersed over tens or hundreds of cell, it is unlikely that a concentration of 100 um would be achieved. The large size and small concentrations of particles that are produced  55  with use of the D L T are therefore very unlikely to cause any adverse biological reactions in the patients.  If it is necessary to further decrease the number of particles produced, this could be accomplished by modifying the ratio of acrylonitrile, butadiene and styrene in the A B S (Lee 2003) or by increasing the cross-linking of polymer molecules by applying radiation (Hemmerich 2000).  In summary, we have shown that our distal locking tool design is near optimal in its targeting of distal holes, ease of insertion and removal, and patient safety. We conclude that our computerassisted approach to distal locking will likely be successful in a surgical setting. To further validate our method, we will test our approach in a surgical setting in which we compare the targeting of distal holes using the customary fluoroscopic approach and electromagnetic tracking. The details of this trial will be discussed in chapter 5.  56  5 Implementation of a Clinical Trial for the DLIMN project In lab testing, we have demonstrated that the sensor and distal locking tool have sufficient accuracy when inserted down the nail to make our proposed electromagnetic targeting approach feasible. However, in order to move our technique towards clinical deployment, we need to determine whether it can reliably measure the hole locations under clinical conditions. This process involves both showing that our method is accurate compared to the current operative technique in a live surgical setting and determining the system's performance in a simulation of a surgical task (Figure 5.1).  Accurate in Surgical Task Simulation Accurate in Labbased Testing  Use in OR  Accurate Compared to Conventional Technique Figure 5.1: Testing Involved in Proof of Concept for Electromagnetic Targeting in DLIMN. Accuracy has been demonstrated in a lab-based setting. It is now necessary to test the systems performance in a simulated surgical task and compared to the conventional technique. When this is complete, the system can be used for targeting in the OR.  To test the system in a surgical task simulation, further software development is necessary. Our current software allows real-time tracking of the Aurora sensors but does not incorporate a targeting component to allow the alignment of the drill axis and the location of the distal locking holes to be visualized. We have therefore deferred this process until software development is complete. In the meantime, we will complete a bystander clinical study to evaluate the 57  performance of our system relative to the current technique. This trial will also give us preliminary insight into clinical practice issues which may need to be addressed in our final design and into the length of time that might be required for our computer-assisted procedure relative to the conventional technique.  We hypothesize that our CAS approach to distal locking is sufficiently accurate to allow for targeting of the distal locking holes without the use of fluroscopy and that it is efficient enough to decrease total surgical time.  5.1  Materials and Methods  In order to evaluate our system intraoperatively, the actual drilling path chosen for distal locking under fluoroscopic guidance will be compared to the axis computed using our technique. Using this method, we can quantify the success of our device and approach without putting the patient at any significant risk. In order to complete this comparison, both the locking holes of the intramedullary nail and the drill bit will be tracked during the drill alignment procedure. Tracking will be used solely for evaluation of our technique and will not be visible to the surgeon during surgery and will not be used to determine the drilling axis. This type of study is known as a bystander study.  Patients presenting at Vancouver General Hospital with diagnosed femoral shaft fractures will be invited to participate in this study. Consenting patients who meet the inclusion criteria will undergo standard IM Nailing of the femur. Our intent is to enroll approximately 10-15 patients at this stage as the purpose of this study is to obtain some pilot data which will allow us to better  58  estimate the variances needed for computing the number of patients required in a future randomized control trial.  5.1.1  Required Instrumentation  Tools will be tracked intraoperatively using the previously described Aurora Electromagnetic Tracking System (see section 2.1).  The I M N will be tracked using the distal locking device described in chapters 2 to 4. As previously described, the DLT will be inserted into the central canal of the intramedullary nail after nail insertion until it makes contact with the distal locking holes. Since the repeatability of distal locking is very good (see section 3.2), the D L T will be pre-calibrated so that the transformation between the sensor and the distal locking hole axis is known prior to surgery. The axis of the locking holes can therefore be computed intra-operatively by applying the predetermined transformation to real-time sensor tracking.  The actual drilling axis will be tracked using a drill-bit mounted reference clip. This tool is a 6 DOF Aurora sensor attached to an A B S instrument clip (Figure 5.2).  Figure 5.2: Reference Clip. The reference clip will be attached to the drill bit during alignment for distal hole targeting using a fluoroscope.  In the conventional I M N technique, the approximate location of the distal locking holes is determined under fluoroscopic guidance (see section 1.1.3) and an incision is made through the 59  soft tissues. The drill bit is then mounted into a radiolucent targeting device and secured into place by tightening a set screw (Figure 5.3).  Figure 5.3: Distal Targeting Device. The drill bit is fastened into the radiolucent targeting device using a set screw (A). The distal holes are the located by aligning the drill bit to the centre of the locking hole under fluoroscopic guidance, (source: Zimmer M/DN ® Femoral Interlocking and Recon Nail Intramedullary Fixation Surgical Technique)  The drill bit is inserted through the incision to the surface of the bone and alignment is determined using the method described in section 1.1.3. Once alignment is complete, the drill bit is anchored into the cortical bone by tapping it into place. The targeting device is then removed, the chuck and drill are attached, and drilling is completed.  The reference clip has a channel into which the drill bit can be inserted. The clip is then screwed down tightly to avoid any motion of the clip relative to the drill bit. Once the clip is attached, the direction of the long axis of the instrument can be defined by completing a pivot calibration before the alignment procedure (Cleary 2005). The intraoperative drilling axis can therefore be defined by applying the transformation determined during calibration to the real-time data from the reference clip (Figure 5.4).  60  A  A  C  Figure 5.4: Reference Clip. The reference clip contains a small channel (A) into which the drill bit (E) is placed. The clip is then tightened down by turning a lever (B). The clip is alsofittedwith screw holes (C). These holes are used to attach the sensor (D) to the clip.  5.1.2  Data Collection  To compare the efficiency of the two methods, the time necessary for insertion and removal of the distal locking tool, the time necessary for attachment, calibration and removal of the reference clip and the time necessary for fluoroscopic alignment will be recorded.  Data will be collected from the DLT and reference clip during alignment for both the more distal and more proximal distal locking holes. The more distal locking hole will be targeted first, since the D L T is locked into the more proximal hole. Data will be collected after alignment is complete, the reference clip will be removed from the drill bit and drilling will be completed. The reference clip will then be re-attached to the drill bit, recalibrated, and placed in the drilled hole axis. Distal screw insertion will then proceed in the standard manner. For the more proximal locking hole, data will be collected once alignment is complete. The distal locking tool will then be removed from the I M N , the reference clip will be removed from the drill bit, and drilling and screw insertion will proceed in the standard manner.  61  The surgeon will also be asked to classify the drilling by fluoroscopic guidance (as successful drilling and no contact with the nail, slight nail contact, severe nail contact, or unable to target hole) in order to determine the quality of the fluoroscopic alignment process.  5.1.3  Data Analysis  Our primary goal is to measure the difference between the locking hole axis as calculated using our approach and as aligned using the standard fluoroscopic technique. The locking hole axis will be defined as the long axis of the drill bit (z-axis). In order to determine the angular deviation between this axis and the hole axis calculated using the DLT, the angle between the orientation vector (w) for these two axes will be compared.  Scr exAxisAngle = arccos(w  Dtf  •u  DIT  )  The variations in the rotation and pitch will be determined by projecting the data onto the distal locking hole axis plane (x-y plane of the nail frame) and the transverse plane (x-z plane of the nail frame). This will be accomplished by taking the dot product of the two dimensional orientation vectors that define the drill bit and the 2D unit vectors that define the appropriate plane.  Rotation = arccos((w x, u y) /jW  m  • (/7 , h )) V  v  In order to compare the lateral deviations, an x-y plane will be defined on the z= 0 plane of the nail frame, the midpoint between the medial and lateral side of the more proximal distal locking  62  hole. The two dimensional radial distance between the pierce point of the hole axis as defined by the D L T and by the drill bit axis will be computed.  Our secondary goal is to estimate the amount of time needed to perform targeting using the CAS approach compared with the time needed for the conventional fluoroscopic approach. We propose to use a (non-parametric) ranked pairs Wilcoxon test to test for differences in the mean time needed.  5.2  Current Status  After obtaining a Health Canada Exemption for our device (Appendix 4:Planning and Implementation of a Clinical Trial in a Research Setting), our CREB application went before full board review and was approved (approved May 9 , 2007 H06-03500). th  In order to obtain V C H R I approval according to the new guidelines described in Appendix 4, section 3, a medical device development safety validation was conducted. Details of this trial are summarized below.  5.2.1  Safety Testing for the DLIMN trial  The equipment involved in our clinical trial includes the two tools described above (DLT and reference clip) as well as the Aurora system itself. The system is comprised of a system control unit, field generator, system interface unit, mounting arm and clamp (see section 2.1). These are commercially available products and all powered components are in compliance with the  63  appropriate electrical standards (See Appendix 11: Aurora Approvals and Classifications). This compliance was verified by the B M E department at V G H .  The distal locking tool and reference clip are not independently powered and do not need to be examined for electrical safety.  The reference clip was originally designed for use in a study exploring the kinematics of surgical movements during minimally invasive cholecystectomy. The reference clip was tested for mechanical integrity as a part of this study. There are no reprocessing steps involved with the reference clip; it is simply cleaned and then resterilized. We therefore did not complete any mechanical testing of the reference clip for this study.  The mechanical safety tests completed for the DLT are described in Chapter 2. The results of these tests were all positive. For our clinical trial, we plan to dispose of the used DLT but retrieve the 6 DOF sensors and re-use them for a maximum of three trials. Mechanical integrity testing of the removal and recalibration process is therefore required. We are currently completing these tests. The sensor can be removed from the DLT with very little difficulty and without using any sharp tools. We therefore do not expect any microscopic damage to the sensor casing. This has not yet been examined under microscope. These examinations will be performed and the results will be included in our report to the B M E department.  5.2.2  Sterilization Method and Validation for the DLIMN Trial  5.2.2.1 Reference Clip A n EtO sterilization method and the reference clip's response to EtO sterilization was tested for the previous cholecystectomy study and were therefore not explored in this project. The details 64  of this protocol can be found in Appendix 12: Cleaning and Sterilization Protocol for MIS Study. In order to satisfy the new guidelines, the EtO sterilization technique developed for the cholecystectomy study was applied to two clips and they were then subjected to the sterility test described in Appendix 4, section 3.3.1. The results from the sterility tests were not satisfactory, one of the tools was sterile but the second showed growth (See Appendix 13: Results of Sterility Testing for Reference Clip). The director of the third party microbiology lab that completed the sterility testing (Dr. Ian Geere, I.G. Micromed Environmental Inc.) expressed his concern that the contamination arose from their handling of the tool prior to sterility testing due to difficulties in placing the sensor leads into the test vessel and not due to problems with our sterilization technique. As discussed in Appendix 4, section 3.3.1, large items must be manipulated in order to place them in a bath of the solution used for sterility testing. They found the long sensor cord difficult to deal with when they first began testing. The trial will therefore be repeated with three additional reference clips and the results of these tests will be evaluated prior to using the reference clip in our clinical study.  5.2.2.2 Distal Locking Tool Since the DLT had not previously been sterilized, a sterilization method had to be developed and validated according to the guidelines presented in Appendix 4, section 3.  Because of the intolerance of both the sensor and ABS material to high temperatures, steam sterilization was not appropriate. In discussion with the SPD at V G H , there was concern expressed about the use of EtO for sterilization of this device. Although it is made of the same material as the reference clip, the small dimensions of the DLT and possible effects of EtO on ABS plastic (Rogers 2006) could pose a problem for EtO sterilization. These factors, as well as economic concerns, led us to choose electron-beam radiation for sterilizing the DLT.  65  Maximum Dose Determination  As discussed in Appendix 4, section 4.3, the dose required to sterilize a device produced under normal factory conditions is typically 25 kGy. We therefore sterilized devices at 25, 50, 75, 100, and 125 kGy to determine if any negative effects resulted from these doses. Post-sterilization, these devices were examined for any visual changes, tested by insertion and removal into the IMN, and tested for sensor accuracy.  A slight yellowing of the tool was noticed with the higher doses, but all performance was satisfactory after sterilization had been applied. The torque tests described in section 2.5.1 were slightly altered to accommodate the oversized sensors used for this trial. The DLTs could not be completely inserted into the I M N but were locked into the distal holes by insertion from the distal end of the IMN. The range of moments required for removal (0.006 to 0.0085 Nm) were similar to those reported in section 3.3.1 (mean 0.011 Nm). As described in this section, the moment necessary to remove a D L T decreases with increased number of insertions and removals. We conclude that the decreased moment necessary to remove the irradiated DLTs is likely due to the increased number of trials they had undergone.  Minimum Dose Determination  In order to determine the efficacy of sterilization, 12 devices were sterilized with varying doses (see Table 5.1). All sterilizing was performed by an FDA recognized e-beam sterilization facility (Iotron Industries Inc, Port Coquitlam, BC)  66  Table 5.1: Radiation Doses Applied to the DLTs. Four tools were radiated at each radiation dose tested. The doses tested were 25, 35, and 45 kGy Radiation Dose Tool# 1 to 4  25 kGy  5 to 8 9 to 12  35 kGy 45 kGy  After sterilization, the devices were sent to an accredited microbiology lab (1G Micromed, Richmond, BC) where the required sterility assurance tests will be completed. Because of the problems described in section 5.2.2.1, the sterility testing was delayed to allow the lab to do some preliminary testing on unsterilized tools. Once they have determined the optimal handling of our tools, sterility testing will be performed.  5.2.3  Final Approval  Pending positive results from the sterility testing of both tools and the mechanical integrity testing of the sensor after removal from the distal locking tool, we expect to receive approval from the Medical Device Development Committee. We have collected all other elements necessary for VCHR1 approval and have been in close contact with the V C H R I Clinical Trials Administrator throughout this process. As a result, we have been assured that our application for V C H R I approval will be expedited through the system. We therefore expect to be able to start our clinical trial in early September.  5.3  Expected Contributions  Success of the clinical trial will be initially be determined by determining the deviation between the hole axis calculated from the electromagnetic system and the using the fluoroscopic method. If deviations are within the clinically acceptable range (less than 2.5 mm lateral deviation and less than 8°angular deviation, see section 4.1) a simulation of the surgical task will be completed (when software development is complete) to train the surgeons in the methods that will be used  67  during actual OR use. Pending satisfactory results in this test, a randomized clinical trial will commence.  If the results of the bystander study are not within the acceptable range, one of two possible situations may be occurring: either the electromagnetic tracking technique may not be accurately predicting the hole axis or the fluoroscopic technique may not be reporting the actual locking hole axis. In order to clarify which of these cases is occurring, an in depth surgical simulation will be completed. By calculating the drilling axis and comparing it to the actual hole axis, we can quantify the performance of the system without the uncertainty of determining the location of the hole axis using fluoroscopy. This testing will also be performed in the presence of all surgical equipment (C-arm, surgical table and surgical tools) to determine if the error seen in the bystander study was the result of distortions to the electromagnetic field by these items.  Results of these tests will determine whether further modifications need to be made to the system prior to its use in the OR.  68  6 Conclusion  6.1  Design and Testing of the Distal Locking Tool  Evaluation of the Aurora EMTS has shown it to be sufficiently accurate to target the distal locking holes of an intramedullary nail. Enclosing the miniature six degree of freedom Aurora sensor in an intramedullary nail causes distortions of only 0.1 mm and 0.05°. The standard deviations of three-dimensional translational and screw angle axis readings from the Aurora sensor in the mid-to-high range of sensor to field generator distances were 0.16 mm and 0.20° respectively.  The insertion repeatability of the distal locking tool is high with maximal one-dimensional standard deviations of 0.20 mm and 1.59°. The three dimensional standard deviations of predicting the drilling axis using the D L T and a drill bit mounted tracker are 0.89 mm and 1.7°.  Removal of the D L T from the I M N requires only 7% of the maximal moment that can be produced using the thumb and index finger to twist the tool's proximal end. The stresses induced during this process are predicted to be approximately 1/500 of the yield stress of the material. th  Insertion and removal of the DLT resulted in the production of approximately 8,580 ± 4,330 particles for tools that were not filed prior to use. A mean of 1,980 ± 760 particles were produced when filing was performed prior to testing. The smallest particle recorded had an edge length of 1.2 um and the mean particle size was 8.2 pm. The particles produced are both too small in size and too few in numbers to cause osteolysis in vivo.  69  The distal locking tool can be safely and easily inserted and removed from the IMN. Accuracy testing indicates that targeting of distal locking holes using this method is likely to be successful.  6.2  Evaluation of the Method and Apparatus  The protocol for mechanical testing and sterilization validation necessary to ensure a safe and effective clinical trial according to all the applicable guidelines has been established. The evaluation of all the tools to be used in the testing of our approach has begun according to this protocol. Pending satisfactory results of these tests, a clinical trial will be completed to assess the relative accuracy of the electromagnetic system and the currently accepted fluoroscopic technique. If the results of this test are positive, the commencement of a full scale randomized controlled clinical trial will be planned. We hypothesize that our method is both accurate enough to target distal locking holes without the use of radiation and efficient enough to decrease the surgical time necessary for distal locking. If the results of the initial clinical trial are not satisfactory, further testing of our approach using surgical simulation will take place in order to refine the method.  The introduction of a radiation-free technique for distal locking will decrease the exposure of the orthopaedic team to potentially harmful radiation. Development of a more efficient targeting technique will decrease the surgical time necessary for the intramedullary nailing procedure.  70  References Archibeck MJ, Jacobs JJ, Roebuck K A , Giant TT. The Basic Science of periprosthetic osteolysis. JBJS (A) 2000; 82:1478-89 . A A O S On-Line Service February 2001 Bulletin. http://www2.aaos.org/aaos/archives/bulletin/feb01/fline4.htm Aurora User Manual. Revision 1.0 Northern Digital Inc. August 2005 B C Ministry of Health, Patient Safety Branch. Best Practice Guidelines for the Cleaning, Disinfection and Sterilization of Medical Devices in Health Authorities. March 2007 Blattert TR, Fill U A , Kunz E, Panzer W, Weckbach A , Regulla D. Skill Dependence of radiation exposure for the orthopaedic surgeon during interlocking nailing of lonbone fractures: a clinical study. Arch Orthop Trauma Surg 2004; B124:659-664 Bong M R , Koval, K J , Egol, K A . The history of intramedullary nailing. Bulletin of the hospital for joint diseases. 2006; 64:94. Bong M R , Kummer FJ, Koval KJ, Egol K A . Intramedullary nailing of the lower extremity: Biomechanics and biology. J Am Acad Orthop Surg 2007; 15(2):97-106. Cleary K , Zhang H , Glossop N , Levy E, Wood B, Banovac F. Electromagnetic Tracking for Image-Guided Abdominal Procedure: Overall System and Technical Issues. Proc IEEE EMBC 2005; Shanghai, China, Sep 1-4, pp.268 Dobbs M . Closed rigid locked intramedullary nailing. Journal of orthopaedic trauma 2006; 20:644. Elfring R, Klein R, Radermacher K. Error evaluation of an electro-magnetic tracking system. Proc 7 Ann CAOS 2007; Heidelberg, Germany, June 20-23, 2007: 285-287 ,h  Ewers R, Schicho K , Undt G, Wanschitz F, Truppe M , Seemann R, Wagner A. Basic research and 12 years of clinical experience in computer-assisted navigation technology: a review. Int J Maxillofac Surg 2005; 34:1-8 Fact Sheet on Ethylene Oxide Sterilization. Hazardous and Toxic Materials Office- Board of Public works. Last updated: January 8, 1996. Accessed: July 12, 2007 http://es.epa.gov/techinfo/facts/ca-htm/oxide-fs.html Farill J. Orthopaedics in Mexico. JBJS 1952; 34A(3): 506-512 Glossop N D , Cleary K , Banovac F. Needle tracking using the Aurora electromagnetic tracking sensor. Proced. 2 Computer Assisted Orthopaedic Surgery 2002; Santa Fe, N . M . June 19-22 nd  Goldstone K E . Wright IH, Cohen B. Radiation exposure to the hands of orthopaedic surgeons  during procedures under fluoroscopic X-ray control. Br J Radiol 1993; 66:899-901 Green TR, Fisher J, Matthews JB, Stone M H , Ingham E. Effect of size and dose on bone resorption activity of macrophages in vitro by clinically relevant ultra high molecular weight polyethylene particles. J Biomed Mater Res Appl Biomater 2000; 53:490-7. Gundestrup M , Storm H H . Radiation-induced acute myeloid leukaemia and other cancers in commercial jet cockpit crew: a population-based cohort study. Lancet 1999; 354:2029-2031 Hafez M A et al. Radiation Exposure to the Hands of Orthopedic Surgeons: are we underestimating the risk? Arch Orthop Trauma Surg 2005; 125:330-335 Hart R, Kovak T, deCordeiro J, Filan P. Fluoroscopy-based computer assistance used for reduction and internal fixation of long bone fractures. Proc 7 CAOS 2007; Heidelberg, Germany, June 20-23: 421-423 lh  Hazan E, Joskowicz L. Computer-Assisted Image-Guided Intramedullary Nailing of Femoral Shaft Fractures. Tech. Orthop. 2003; 18(2) Hemmerich K.1 Radiation Sterilization: Polymer Material Selection for Radiation-Sterilizaed Products. Medical Device and Diagnostic Industry 2000. hUp://www.devicelink.com/mddi/archive/00/02/006.html Herscovici D et al. The effects, Risks, and Guidelines for Radiation Use in Orthopedic Surgery. Clin, Orthop. Rel. Res. 2000; 375: 126-132 Hollstien, DS. Method and Apparatus for Distal Targeting of Locking Nails in Intramedullary Nails 2006; U.S. Patent 7 029 478 B2 Howie A W , Haynes DR, Rogers SD, McGee M A , Pearcey MJ. The response to particulate debris. Orthop Clin North A m 1993; 24:571-81. Ingham E, Fisher .1. Biological reactions to wear debris in total joint replacement. Proc Instn Mech Engrs J Eng Med 1999; 214(H):21 -37 Ingham E, Fisher J. The role of macrophages in osteolysis of total joint replacement. Biomalerials. 2005; 26:1271-1286. Jacobs JJ, Hallab NJ, Urban R M , Wimmer M A . Wear Particles JBJS 2006; 88(A)(S2):99-102 Jones DG, Stoddart J. Radiation use in the orthopaedic theatre: a prospective audit. Aust NZJ Surg 1998; 68:782-784 Kanlic E M , Delarosa F, Pirela-Cruz M . Computer assisted orthopaedic surgery ~ CAOS. Bosnian journal of basic medical sciences 2006;6:7. Khoury A , Moshieiff R, Beyth S, Joskowicz L, Finklestein J, Liebergall M . Evaluation of the  second generation of computer assisted orthopaedic fracture reduction. Proc 7' CAOS 2007; Heidelberg, Germany, June 20-23:392-394. Kirchstein U., Kronberg K., Hein A., Concept and Preliminary Experiments of Navigated Imaging with an Electromagnetic Position Measurement System, CURAC 2005; Berlin, September 22-24 , 2005 Kirsch SR (1). Accuracy Assessment of the Electromagnetic Tracking System Aurora. ND1 Europe GmbH 2005 Kirsch SR (2). Assessment of Metallic Distortions of the Electromagnetic Tracking System Aurora. NDI Europe GmbH 2005 Krettek C, Konemann B, Miclau T, Kobli R, Machreich T, Tscherne H . A Mechanical Distal Aiming Device for Distal Locking in Femoral Nails. Clin. Re. Rel. Res. 1999; 364:267-275 Krettek C, MannP J. Miclau T, Schandelmaier P, Linnemann, Tscherne H. Deformation of Femoral Nails with Intrameduallary Insertion. J. Orthop. Res. 1998; 16:572-575 Krucker J., X u S., Viswanathan A., Shen E., Glossop N . , Wood B. Clinical evaluation of electromagnetic tracking for biopsy and radiofrequency ablation guidance. Int J CARS2006; 1:169-171. Lee HB, Khang G, Lee JL. Polymeric Biomaterials Biomaterilas Principles and Applications 2003 CRC Press, Boca Raton, FL:55-77 Leventon W. Medical Device Sterilization: What Manufacturers Need to Know. Medical Device and Diagnostic Industry 2002. http://www.devicelink.com/mddi/archive/02/09/003.html Levin PE, Schoen RW Jr, Browner BD. Radiation exposure to the surgeon during closed interlocking intramedullary nailing. JBJS 1987; 69A:761-766 Levy EB, Tang J, Lindisch D, Glossop N , Banovac F, Cleary K. Implementation of an electromagnetic tracking system for accurate intrahepatic puncture needle guidance: Accuracy results in an in vitro model. Acad Radiol 2007;14: 377-354 Macatee G. Reprocessing of medical devices and patient care equipment. B C Ministry of Health communique number 2007-01. June 21, 2007 Madan S, Blakeway C. Radiation Exposure to surgeon and patient in intramedullary nailing of the lower limb. Int J Care Injured 2002; 33:723-727 Madan S., Blakeway C. Radiation Exposure to surgeon and patient in intramedullary nailing of the lower limb. Int J Care Injured 2002; 33:723-727 Medical Devices Regulations, SOR/98-282. Available online at: http://www.calii.org.ca/regu/sor98-282/whole.html. Regulation current to June 29 , 2007. lh  73  Mehlman CT, DiPasquale TG. Radiation exposure to the orthopaedic team during fluoroscopy: "how far is far enough?". J Orthop Trauma 1997;11:392-398 Miller M E , Davis M L , MacClean CR, Davis JG, Smith B L , Humphries JR. Radiation exposure and associated risks to operating-room personnel during the use of fluoroscopic guidance for selected orthopaedic procedures. J Bone Joint Surg 1983; 65A:1—4 Muller LP, Suffner J, Wenda K , Mohr W, Rommens P M . Radiation exposure to the hands and the thyroid of the surgeon during intramedullary nailing. Injury 1998; 29:461-468 Nashed RS, Becker D A , Gustilo RB. Are cementless acetabular components the cause of excess wear and osteolysis in total hip arthroplasty? Clin Orthop Relat Res 1995;317:19-28. Noordeen M H H , Shergill N , Twyman RS, Cobb JP, Briggs T. Hazards of ionising radiation to trauma surgeons: reducing the risk. Injury\992>; 24:562-564 Okcu G, Aktuglu. Antegrade nailing of femoral shaft fractures combined with neck or distal femur fractures. A retrospective review of 25 cases, with a follow-up of 36-150 months. Arch Orthop Trauma Surg 2003;123:544-550 Ozturk H , Oztemiir Z, Bulut O, Unsaldi T. Complication following intramedullary fixation with a fixion nail in a patient with osteogenesis imperfecta: A case report. Acta orthopaedica belgica 2005;71:227. Pennig D, Oppenheim W, Faccioli G, Rossi S. Intramedullary locked nailing of femur and tibia: insertion of distal locking screws without image intensifier. Injury 1997;28(4):323-326. Raya A . Biocompatibilidad de nuevo cemento osea. Universitat Autonoma de Barcelona; Ph D Thesis; 1993. Revell PA. Biological reaction to debris in relation to joint prostheses. Proc Inst Mech Eng J EngMed 1997; 211:187-97. Riley SA Radiation exposure from fluoroscopy during orthopaedic surgical procedures. Clin Orthop 1989;248:257-260. Rogers W. Steam: Uses and Challenges for Device Sterilization. Medical Device and Diagnostic Industry 2006. http://www.devicelink.com/mddi/archive/06/03/003.html Sanders R, Koval K J , DiPasquale T, Schmelling G, Stenzler S, Ross E. Exposure of the orthopaedic surgeon to radiation. JBJS 1993; 75A:326-330 Schischo K , Figl M , Donat M , Birkfellner W, Seeman R, Wagner A , Bergmann H , Ewers R. Stability of miniature electromagnetic tracking systems. Phys Med Biol 2004; 50:20892098  Schmalzried TP, Guttmann D, Grecula M J , Amstutz HC. The relationship between the design position and articular wear of acetabular components inserted without cement and the development of pelvic osteolysis. JBJS 1994; 76A:677-88. Seiler PG, Muench K, Kirsch R. Method for Determining the Position of a Sensor Element. 2004; U.S. Patent 6 836 745 B2 Singer G. Radiation Exposure to the Hands From Mini C-Arm Fluroscopy. J Hand Surg 2005; 30A: 795-797 Slomczykowski M , GoertzenDJ, Oxland TR, et al. Intramedullary deformation of unreamed femoral nails. Transactions of the 44 Annual Meeting of the Orthopaedic Research Society.. New Orleans, March 16-18, 1998:1133 th  Slomczykowski M,Hofstetter R, Sati M , Krettek C, Nolte LP. Novel Computer-Assisted Fluoroscopy System for Intraoperative Guidance: Feasibility Study for Distal Locking of Femoral Nails. J Orthop Trauma 2001; 15(2): 122-131 Smith GL, Wakeman R, Briggs TWR. Radiation exposure of orthopaedic trainees: quantifying the risk. JR Coll Surg Edinb 1996; 41:132-134. Sordellini PJ, Satter SE, Caputo V A . EtO Sterilization: Principles of Process Design. Medical Device and Diagnostic Industry 1998. http://www.devicelink.com/mddi/archive/98/12/006.html Stiehl JBB, Comparison of optical and electromagnetic tracking protocols in total knee replacement. CAOS 2007; Heidelberg, Germany. June 20-23 Suhm N , Messmer P, Zuna I, Jacob L , Regazzoni P. Fluoroscopic guidance versus surgical navigation for locking of intramedullary implants A prospective,controlled clinical study. Int. J. Care Injured2004;35:567-574 Szakelyhidi DC. Development of a magnetic targeting device applied to interlocking of distal intramedullary nail screw holes. Virginia Ploytechnic Institute and State University; M Sc Thesis; 2002. van Staa TP, Dennison E M , Leufkens H G M , Cooper C. Epidemiology of Fractures in England and Whales. Bone 2001; 29(6): 517-522 Weil Y , Gardner M J , Helfet D L , Pearle A D . Computer navigation allows for accurate reduction of femoral fractures. Proc 7 CAOS2007; Heidelberg, Germany, June 20-23:417-420 th  Whatling G M , Nokes L D M . Literature Review of current Techniques for the insertion of distal screws into intramellary locking nails. Injury, Int. J. Care Inj. 2005;37(2): 109-119 Winquist, R A Hansen, ST, Clawson D K . Closed intramedullary nailing of femoral fractures. A report of five hundred and twenty cases. JBJS 1984;66A:529.  Wood B.J., Zhang H., Durrani A., Glossop N . , Ranjan S., Lindisch D., Levy E., Banovac F., Borgert J., Krueger S., Kruecker J., Viswanathan A., Cleary K. Navigation with Electromagnetic Tracking for Interventional Radiology Procedures: A Feasibility Study../ Vase Interv Radiol 2005; 16:493-505  76  Appendix 1: Design Evolution of the Distal Locking Tool In order to target the distal locking holes of the intramedullary nail from within the central canal of the nail, a distal locking tool was necessary. The requirements for this tool were: 1. 2. 3. 4. 5.  It must It must It must It must It must  be able to fit through the insertion handle which has a diameter of 3.5 mm. stop when it encounters the distal locking holes. lock into the distal locking holes repeatably. fix the sensor's location relative the locking holes reliably. be easy to remove from the intramedullary nail.  In order to address all of these issues, we started by focusing on the distal end of the tool that would lock into the distal locking holes. In order to fit through the insertion handle (3.5 mm) and expand into the intramedullary nail (inner diameter 4.5 mm) so that it can lock into the locking holes, we decided to design an end bit with locking plugs that match the locking hole diameter (5 mm) and locking arms which can be compressed to allow the insertion of the tool. Both the torsional rigidity and the bending stiffness will affect the locking and removal process. These properties depend on the material chosen, the cross-sectional area of the locking arms and the length of the locking arms. The material used must be biocompatible and must not interfere with the electromagnetic sensor being used for targeting. We therefore chose to use a polymer. The combination of the small dimensions of our tool and the plastic material made it very difficult to machine. A rapid prototyping machine was therefore used to manufacture the tools. Of the materials available for this machine, A B S P400 was the most economical material that displayed all of the properties we were looking for. To ensure that the locking plug would fit securely into the locking holes, it was designed with a raised locking plug. This locking plug has a spherical proximal end to promote easy removal and a cylindrical distal end in order to stop the locking plug when it reaches the distal locking hole.  Figure 1: Distal Locking Tool Plug. The distal ends (left) of the DLT are designed with plugs to engage the locking holes of the IMN. The distal surface of the plugs are cylindrically shaped so that they do not slide past the  77  locking holes. The proximal (right) surface of the plugs are spherically shaped in order to facilitate removal of the DLT.  In order to test variations in cross section and length of the locking arms, we created many different variations of end bits. The first design had a constant cross section with straight locking arms. This design was manufactured with locking arm lengths of 15, 25, 35 and 45 mm.  Figure 2: Straight Locking Arms. This design was tested with arm lengths of 15, 25,35 and 45 mm (left to right). The base of these tools had diameter 3.5 mm.  The locking arms of these tools did not expand sufficiently when inserted in to the intramedullary nail. A new design that included a bend in the locking arms was made. This design was manufactured for locking arm lengths of 15, 25 and 35 mm.  Figure 3: Bend Design. Locking arms with a gradual bend were manufactured with arm lengths of 1.5, 2.5 and 3.3 cm.  This design was more successful but there was still too little rigidity of the locking arms. The smallest arm length was also found to be inappropriate due to the high forces necessary to compress the arms. The next design of the locking arms included a gradually decreasing crosssectional area. This allowed for an increase in rigidity while maintaining a maximum diameter (when the locking arms are compressed) of 3.5 mm. This design was manufactured for arm lengths of 25 and 35 mm.  Figure 4: Bend Design. Locking arms with a gradual bend were manufactured with arm lengths of 1.5, 2.5 and 3.3 cm  78  This design proved to lock very well into the distal locking holes. The 25 mm arm length was chosen as this length showed the best balance between ease of insertion and ease of removal. The tool was then extended so that it had a total length of 475 mm. The portion of the tool that extends towards the proximal end of the tool has a c-shaped channel built into it. The diameter of this channel is 2.5 mm and is designed to allow the Aurora sensor to be laid into it.  Figure 5: The distal locking tool is Fitted with a C-shaped channel to house the Aurora sensor (2.5  mm).  This gives a final design as shown below, with a 50 mm distance between locking plugs and sensor. By designing the locking arms with a relaxed separation of 10 mm, insertion and removal of the tool can be accomplished easily and repeatably.  locking plugs sensor location ,....„.....;.  • 10 mm  locking arms = 25 mm  Figure 6: Distal Locking Tool. The final design of the distal locking tool has a relaxed locking arm separation of  10 mm. The distance between the sensor location and the locking plugs is 50 mm.  79  Appendix 2: ABS Material Specifications  F D M T E C H N O L O G Y >., M A T E R I A L S  A true industrial thermoplastic. A B S is widely used throughout industry. When combined with the Fused Deposition Modeling (FDM) systems by Stratasys, this material is ideal for the rapid production of prototypes, tooling and the direct (tool-less) manufacturing of production parts.  Tensile Strength, Type t. 0.125  ASTM D638  3,2000 psi  Tensile Modulus, Type 1, 0.125  ASTM D638  236,000 psi  Tensile Elongation, Type 1, 0.125  ASTM D638  22 MPa 1,627 MPa 6%  Ftexural Strength  ASTM D790  6,000 psi  41 MPa  Ftexural Modulus  ASTM D790  266.000 psi  1334 MPa  IZOD Impact, un-notched  ASTM D256  4 tl-lb/in  IZOD Impact, notched  ASTM D256  2 tt-lb/in  Heat Deflection (HDT) Glass Transition (Tg) Coefficient of Thermal Expansion Melt Point  ASTM D648 DMA (SSYS)  205° F 2ie°F  96° C 104°C  5.60E-05 in/in/F Not Applicable'  Not Applicable'  Specific Gravity Vertical Buring Test Rockwell Hardness Dielectric S (kV/mm) Dielectric C (#60Mhz)  ASTM D792 UL 94 ASTM D785 IEC 60112 IEC 60250  APPEARANCE • White available on all FDM systems • Colors available on the FDM Maxum: • Black. Blue. Green. Grey (light). Grey (steel). Red and Yellow • Custom color program available • Colors available on FDM Prodigy Plus: • Black. Blue. Green, Red and Yellow • Custom color program available  1.05 HB R105 32 2.4 SYSTEM AVAILABILITY • FDM Maxum • FDM Titan Tl • FDM Vantage SE • FDM Vantage S • FDM Vantage / (when configured with ABS) • FDM Prodigy Bus  Th»fafaoMfenprmnUri »<• typtral v«/uw M w d W h, mfevnc* and t i » m « i « i P"'***"* "¥ *y <***" '«* tptKaHom o: qaHr control w o w End-on mitral :•->••••.>•-•* em bo mptcfd <+/•) by. but not Imxfd to, put eloign. Mhm OOMMMW M conciMns. «te Actual ntoM W» my w<A bu,id m M W K Tl  I g / U(  a  /s iw) s/ofc  2  Do 10 amorphous nmiro, nuttonil doos nor display a nWfwp p o M  For more information about Stratasys systems and materials, contact your representative +1 888.480.3548 or visit www.stratasys.com  6 2005 SUfttasy*. ItK. All nglrtt, t*s*fv»d Piottuct sf*ctfk*«iG09 aw MtyKI 1° clw«g» w*h<*ft notice. MF*BS 3*05  STRATASYS'  Appendix 3: Theoretical Force to Remove DLT and Factor of Safety Calculations In order to determine the theoretical force necessary to remove the D L T from the IMN, we determined the force necessary to compress the locking arms of the distal end of the D L T sufficiently to release it from the locking holes.  L =  mm  Figure 1: Distal Locking Arms. The deployed height of the distal locking arms is 5 mm. In order to remove the DLT from the IMN, the locking arms must be compressed this distance.  In order to do this, we approximated that the force necessary to compress the locking arms would be similar to the force necessary to deform a straight beam to the final expanded state. We also modeled the shape of the locking arms to be constant cross-sections and compared the results for a) rectangular cross-section and b) circular cross-section. The standard equation for deflection of a beam under cantilever loading was then used to determine the force required to deflect the beam by 5 mm.  F/ max  3  3EI  F=  My,  max  bh  3  1 .Model locking arm as a constant rectangular cross section: •  12  The force necessary to move the locking arms is: F = 3 . 2 N 81  2. Model locking arm as a constant circular cross section:  / = 64  •  The force necessary to move the locking arms is: F = 1.9 N  Factor of Safety Take the force necessary to deform the locking arms as the average of the rectangular and circular cases: F = 2.6N  Assuming that this force was only being applied to the portion of the D L T that is engaged by the distal locking holes, the working stress for this situation was determined. This working stress was compared to the yield stress for the A B S material to determine the factor of safety. Factor of Safety  yieldstress workingstress  a = — = 1.5-10 N / m A 5  2  Flexural Strength of A B S = 41MPa FoS =  4  l  M  P  a  OASMPa  ^275  82  Appendix 4: Planning and Implementation of a Clinical Trial in a Research Setting In the past, new medical devices developed at the University of British Columbia (UBC) for use in clinical trials at one of the Vancouver Coastal Health Authority (VCHA) sites have been required to obtain U B C Clinical Research Ethics Board and Vancouver Coastal Health Research Institute (VCHRI) approval. The major components of the later simply involved obtaining approval from the Biomedical Engineering Department (BME) that devices were electrically and mechanically safe and developing a sterilization method in collaboration with the hospital's Sterile Processing Department (SPD).  In March of 2007, the British Columbia Ministry of Health (MoH) released a Best Practice Guideline for Cleaning, Disinfection and Sterilization of Medical Devices in Health Authorities (MoH March 2007). In June of this year, the Ministry of Health released a further communique requiring all health authorities to conform with the aforementioned guidelines (Macatee 2007). Adoption of this policy invalidated the approval process that had previously been in place at the VCHRI. In order for development and testing of novel medical devices within our institution to continue, a new approval process was necessary. Ours was the first clinical trial to request approval after the introduction of this new policy and as a result, it was necessary for me to find a method of satisfying all the requirements set out in the M o H and to determine a new path through the V C H R I approval process. In order to facilitate VCHRI approval for all future researchers intending to bring new medical devices into a surgical environment, I have compiled all the necessary information to guide decision making and approval into this chapter. Figure 1 illustrates the main steps in obtaining approval for a clinical trial. Each of these requirements will be discussed in detail below.  Sterility A s s u r a n c e Development and testing of sterilization method  Biomedical Engineering • Electrical/ Mechanical Safety Testing  Medical Device Development Safety Committee Approval  Health Canada Exemption  CREB Approval  VCHRI Approval 1 Clinical Trial Figure 1: Approval Process. This chart displays all of the necessary steps to obtain approval of a clinical trial at VCHRI. The items in bold (left) are new requirements that have been created in response to new legislation governing the use of medical devices. Descriptions of each of these steps can be found in this chapter.  1. Clinical Research Ethics Board Approval The C R E B application involves filling in an online application, preparing a detailed protocol and preparing any other necessary documents such as a consent form for subjects. The steps necessary for this procedure are well documented by U B C ' s Office of Research Services and can be found at rise.ubc.ca. Examples of these documents in the form of our approved protocol, consent form, and online application can be found in Appendix 5, 6 and 7 for further reference.  One portion of the online application has particular significance to researchers who have developed a medical device for use in clinical trial. Section 7.8B reads as follows:  84  7.8B For clinical trials involving investigational drugs/devices or marketed drugs/devices outside their approved indications, indicate whether or not approval has been obtained from the appropriate federal regulatory agency for this purpose.  For medical devices, the appropriate federal regulatory agency is Health Canada. The Medical Devices Regulations (MDR 1998) govern the use of medical devices and their licensing.  Custom-made medical devices, medical devices for special access and medical devices for investigational testing involving human subjects do not require Health Canada device licenses. Medical devices for special access are defined as devices that can be used for emergency use or if conventional therapies have failed, are unavailable or are unsuitable. These devices require Health Canada approval and are regulated according to Part 2 of the Medical Devices Regulations (MDR 1998, Section 68). Part 2 of the M D R applies to custom made devices only if devices are to be imported or sold.  Devices for investigational testing involving human subjects also require Health Canada approval and are regulated by part 3 of the M D R but this approval is also only necessary for medical devices that are to be imported or sold (MDR 1998, Section 79).  Because medical devices that are developed by a given institution for testing in that institution do not fall under any of the aforementioned categories, Health Canada approval is not necessary. Researchers should contact the Regulatory and Scientific Section, Device Licensing Division of the Medical Devices Bureau of Health Canada via email and request an explicit Health Canada exemption. It is advisable that you clarify the details of the device with the Biomedical 85  Engineering (BME) department at the appropriate hospital site and request that they contact Health Canada.  Health Canada classifies medical devices according to the potential risk they pose to patients and healthcare workers (I- lowest risk, IV-highest risk). The details provided to the B M E department should include the risk classification of the device. All surgically invasive devices, defined as invasive devices that are intended to enter the body through an artificially created opening that provides access to the body structures and fluids, are a minimum of class II. If the device is designed to diagnose, monitor, or control elements of the central nervous system, cardiovascular system, or a fetus in utero or if the device is intended to be absorbed by the body or remain in the body for more than 30 days, the device will fall under a higher classification. A complete definition of the risk classification of medical devices can be found in Schedule One of the Medical Devices Regulations (MDR 1998).  2. Vancouver Coastal Research Institute Approval  In order to obtain permission to conduct a clinical trial at one of the V C H R I sites, there are several steps necessary. A detailed description of the application process can be found in Appendix 8: Approval to Conduct Research at Vancouver Coastal Health Authority and Appendix 9: Guideline for Obtaining Approval to Conduct Research at Vancouver Acute.  A l l forms and guidelines can also be found at http://www.vchri.ca/s/FormsLogos.asp.  It is important to note that the Principal Investigator must have a medical appointment at V C H , be a V C H employee, or have received a V C H Affiliated Investigator Appointment.  86  The new M o H policy as it pertains to clinical trials and our protocol to satisfy this policy are discussed below.  2.1 Best Practice Guidelines for the Cleaning, Disinfection and Sterilization of Medical Devices in Health Authorities  Unlike the Health Canada device license requirements, all equipment and devices, regardless of their source, must meet the standards set out in the Best Practices Guidelines (MoH March 2007). One of the major consequences of the communique for device developers is that it explicitly states that manufacturers are responsible for supplying written information regarding the safe and appropriate reprocessing of the medical equipment or device (MoH, section 5.4). It further specifies that disinfection and sterilization procedures must include information regarding the type and concentration of chemical products, the duration and temperature of exposure, and physical and chemical properties that could potentially have an impact on the efficacy of sterilization (MoH, section 8.1). Finally, the document emphasizes that the sterilization process must be validated thoroughly before being used in a clinical setting and at regular intervals thereafter (MoH, section 13.4). The recommendations made by the manufacturer must then be reviewed by infection control and sterile processing personnel. There is also a requirement that biomedical engineering verifies the safety of the device prior to its use in a clinical setting (MoH, section 2.3).  3. Guidelines for Medical Device Development and Testing for Use in Clinical Trials at Vancouver Coastal Health Acute Sites  In order to satisfy the new regulations of the aforementioned policy change and clarify the VCHRI requirements that are already in place, I have proposed a protocol for all researchers who 87  plan on completing a clinical trial at V C H A sites. In order to evaluate this protocol, a collaboration of experts from a variety of hospital units was necessary. These experts have formed what we refer to as the Medical Device Development Safety Committee (MDDSC). The committee members are listed in Table 1. Table 1: Members of the V G H Medical Device Development Safety Committee.  Name Dr. Elizabeth Bryce Darren Kopetsky Stephania Manusha Mona Sharman Gord McConnell Debbie Hendricks Murray Dore  Affiliation Regional Manager, Infection Control Regional Manager, Risk Management Regional Manager, Clinical Trials Administration Head, Sterile Processing Department Head, Biomedical Engineering Department Manager, Operating Rooms Biomedical Engineering Departement  The objective of this protocol is to provide a detailed description of the requirements for using custom-made medical devices within the Vancouver Coastal Health Authority. For the reasons described in section 1 of this appendix, this protocol does not apply to medical devices imported or sold to researchers for use at V C H A sites, as these devices require Health Canada approval.  This protocol does not replace the C R E B or V C H R I approval processes described earlier in this chapter. Instead, it provides a means of satisfying several of the V C H R I requirements all at once.  In order to obtain this stage of approval, researchers will be required to pass electrical and mechanical safety testing and develop and validate a sterilization protocol. Sterilization validation requires the participation of third-party processors and testing that lasts several weeks. Researchers are advised to plan their timelines accordingly.  The first step to obtain approval from the M D D S C is completion of aforementioned safety tests and the development of a cleaning, packaging and sterilization protocol. Following completion 88  of these tests and development of the sterilization protocol, a Medical Device Development Proposal must be submitted to the M D D S C . This submission should include the C R E B protocol, the results of safety testing, the sterilization method and any additional information relating to device design or function that the developer deems necessary (Figure 2). Details of all of these requirements will be discussed below.  Mechanical Safety Tests  Electrical Safety Tests  Proposed Sterilization Method  Additional Design/ Function Info  CREB Protocol  (if necessary)  Medical Device Development Proposal Figure 2: The Medical Device Development Proposal. This proposal should include all the required mechanical and electrical safety tests as well as the proposed sterilization method and C R E B protocol. A n y additional information regarding design o f function o f the device may also'be included.  3.1 Electrical Safety Requirements  Any equipment that will be used during the trial must meet the relevant Electrical Safety for Medical Devices standards (CSA 601 Series). Any components that are commercially available and designed for medical use will likely already be appropriately certified. Testing of custombuilt powered devices should be discussed with the Biomedical Engineering Department at the appropriate hospital site. CSA compliance testing can be arranged for new devices but researchers should expect an additional cost and plan for the time necessary to complete these tests. Compliance with electrical safety standards will be checked by the Biomedical Engineering Department.  89  3.2 Mechanical Integrity  The developer must show that the device remains intact and retains function after exposure to the environment and stresses it will be subjected to throughout sterilization and use cycles.  For multiple use items, any handling of the device that will occur between uses must be tested. The developer must treat the device exactly as it will be treated after it has been used; for example, if the device must be recalibrated, or if some portion of the device must be replaced, the device should be disassembled, cleaned, reassembled and calibrated as it would be prior to another trial. Functional tests as well as mechanical integrity testing must be performed. Mechanical integrity testing must include observation of the device under microscope to look for any nicks or cuts that may affect re-sterilization. It is not necessary to complete sterilization between tests for this stage of testing. The number of reprocessing cycles tested must exceed the allowed number of uses defined for the device.  3.3 Cleaning, Packaging and Sterilization Method  The choice of an appropriate sterilization method to meet the requirements of a particular device is an involved process. There are many different factors including material response to sterilization and the relative costs of both routine sterilization and the validation trial. A detailed guide to choosing a sterilization modality for a device will be presented in Section 4 of this Appendix.  90  Once the type of sterilization has been chosen, there are several tests that need to occur to show the efficacy of the proposed method. These are: maximum dose determination, minimum dose determination and, when radiation is used, dose mapping (Figure 3).  Max. Dose Determination  Min. Dose Determination  1  Dose  I  Mapping  1  J  Sterilization Validation Figure 3: Sterilization Validation Trial. The sterilization validation trial must consist of maximum and minimum dose determination. If radiation validation will be used, dose mapping is also necessary.  3.3.1 Minimum Dose Determination  The minimum sterilization dose necessary to ensure sterility must be determined. In order to complete this process, a third party microbiology lab must be employed to complete sterility testing. Four devices for each different level of sterilization (eg. radiation dose or exposure time for ethylene oxide) must be cleaned, sealed, and sterilized according to the proposed method. Depending on the type of sterilization chosen and the volume of devices required for the clinical trial, sterilization may be possible through the hospital SPD. Otherwise, it is the responsibility of the researcher to find an appropriate third-party sterilization company. This company should be F D A or equivalently approved to perform sterilization of medical devices. The name and any accreditations of the sterilization company must be included in the medical device development proposal.  91  After sterilization, the device must be sent to a microbiology lab for sterility testing. The testing that must be completed by the lab involves washing the device (post-sterilization) in a special solvent for a given amount of time, filtering the solvent and then incubating the filters in both aerobic (oxygen rich) and anaerobic (oxygen-free) environments for 14 days. If any contamination exists after sterilization, bacterial growth will occur in one of these two environments. If there is no growth, the device is sterile. The microbiology lab must provide the researcher with a report outlining the results of sterility testing.  It should be noted that the majority of sterility testing at small microbiology labs is performed on small devices or pharmaceuticals. The researcher is therefore encouraged to be detailed in the description of their device when choosing a lab to complete the testing in order to ensure that their facilities can accommodate the device. It may also be necessary to provide an unsterilized device for the lab to use in the determination of the optimal handling of the device for sterility testing.  The third-party laboratory must be comfortable and experienced with the necessary testing. They must also be appropriately accredited or certified (eg. Standards Council of Canada, Health Canada, FDA) to perform these tests. The suitability of the third-party will be judged by the committee during evaluation of the proposal.  Along with the details of the method of sterilization chosen, the researcher must include a description of their planned sterilization validation trial- including the name and certifications of all third parties- in the medical device development proposal.  92  3.3.2 Maximum Dose Determination  Testing must be performed in order to determine the maximum sterilization dose that can be applied to the device without compromising its function or safety.  Sterilization doses may fluctuate slightly from case to case or from one area of a device to another. It is imperative that the developer shows that their device can withstand the results of such fluctuations. Single use devices must be sterilized to a dose of twice the intended sterilization dose (eg. time for steam and EtO, radiation dose) to ensure function after sterilization. Re-usable devices should be subjected to a dose in excess of the total dose that can be applied to the device based on the number of reprocessing cycles allowed.  The mechanical integrity and function of the device must be tested after the maximum dose is applied. The methods that will be used to test mechanical integrity and function as well as the details of doses that will be applied during maximum dose determination must be included in the medical device development proposal and should be discussed with B M E in advance.  3.3.3 Dose Mapping for Irradiation Sterilization  Devices that will be sterilized using irradiation techniques must undergo a dose-mapping procedure. Dose mapping allows determination of the zones of minimum and maximum dose in the device package by placing dosimeters at strategic locations. This process allows for the dose being applied at various areas of the device to be correlated to the dose measured at a position outside the device packaging. This is necessary to ensure that the minimum dose is applied to the entire device. Dose mapping should be completed by the facility performing the irradiation of the  93  device. The researcher should ensure that the company contracted to complete sterilization is equipped to complete dose mapping as well.  3.3.4 Sterilization Challenge  During proposal evaluation, the M D D S C may require a sterilization challenge test to be performed in addition to the requisite sterility test previously described. This may be the case if the device is likely to be difficult to sterilize due to the presence of small lumen or crevasses that may not be penetrated during sterilization. The sterilization challenge involves inoculating the device with a known concentration of organisms, processing it with the proscribed cleaning and sterilization technique, and then performing the aforementioned sterility tests. Should it be required, the details of this test should be discussed with the appropriate members of the M D D S C (Infection Control and Sterile Processing representatives) in order to determine whether inoculation should be performed by the hospital representatives or by the microbiology lab.  3.4 Obtaining Final MDDSC Approval  Once the sterilization validation has been successfully completed, the researcher must submit a final report including the results of this test and the timeline for planned sterilization method audits. Provided there are no significant changes to the design, manufacture, packaging, or raw materials used for the device, only the minimum dose must be re-validated on a regular basis. The sterility testing previously described must be repeated after a given time period to ensure sterility is being maintained. Process auditing is required quarterly for Health Canada licensed devices (MDR 1998) but this timeline may not be appropriate for research devices due to their infrequency of use. The auditing process for research devices has not yet been finalized.  94  It is the responsibility of the device developer to ensure that the device is only used according to its approved protocol. The developer must train all of the appropriate staff in the methods necessary for device use, cleaning and sterilization.  Any changes to the design, use, or processing of the approved device must be approved by the most appropriately qualified member of the committee, who may request approval by the whole committee if they feel the issue requires broader input. A form to be completed for these purposes is planned but not complete at this time.  Vancouver Hospital will bear the legal responsibility for use of the device in a surgical setting and will indemnify the researchers once approval of the Medical Device Development Safety Committee has been granted.  Once the device has been approved by the committee, the developer should submit the standard Request for Approval to Conduct Research at Vancouver Coastal Health Authority. The Medical Device Development Report should be included with this submission.  95  Proposed Sterilization Method  Electrical Safety Tests  Mechanical Safety Tests  Additional Design/ Function Info  CREB Protocol  Medical Device Development Proposal Accepted Min. Dose Determination  Max. Dose Determination  M.  Timeline for Process Auditing  Sterilization Validation  Dose Mapping  I  MDDSC Approval  Application for * VCHRI Approval Figure 4: Final MDDSC Approval. Once sterilization validation results and a timeline for process auditing are submitted to and approved by the MDDSC, application for VCHRI approval should be submitted.  4. Choosing a Sterilization Method For engineers and other researchers developing novel devices, mechanical integrity and function testing are a normal part of the design process. Sterilization concerns, however, are not as routine 96  in our line of work. For this reason, I would like to share the information that I have collected in the design of our clinical trial and the creation of the protocol just described.  Medical devices are most commonly sterilized using one of three methods: gas, irradiation, or steam. Gas sterilization is performed by sealing a device in a chamber, removing the air from the chamber and infusing it with Ethylene Oxide (EtO). EtO kills any microorganism with which it comes into contact. After the period of gas exposure is complete, the EtO is removed from the chamber and repeatedly aerated to remove unwanted gas residue. Irradiation sterilization by gamma rays or electron beams destroys microorganisms by compromising their D N A and causes them to fragment. Steam sterilization relies on high temperatures to degrade the D N A and proteins of microorganisms (Fact Sheet on Ethylene Oxide Sterilant Alternatives 1997). Each of these methods has its advantages and disadvantages. This document aims to help the researcher chose which method may be most appropriate for their device.  The major factors that must be taken into account when choosing a sterilization method are whether or not the device can withstand the particular method and maintain function, whether you have access to the sterilization method, and the cost of sterilization and validation. A summary of the response of various plastics to the aforementioned sterilization techniques is presented in Table 2. A detailed look at each of the three sterilization modalities can be found below. Table 2: Response of Various Plastics to Steam, Radiation, and Ethylene Oxide Sterilization Methods. (source: Rogers 2006) ^_ Steam Response Radiation Response EtO Response Plastic Acetal  Good  No  Good  Acrylic  Poor  Good  Good  Acrylonitrile butadiene styrene  Varies  Good  Varies  High-density polyethylene  Good  Good  Good  97  Nylon  Varies  Good  Good  Polycarbonate  Varies  Good  Good  Polyester  Poor  Good  Good  Polyethylene  Poor  Good  Good  Polyglycolic acid  No  No  Good  Polymethyl pentene  Good  Poor  Good  Polypropylene  Good  Varies  Good  Polypropylene and polyethylene copolymer  Good  Good  Good  Polystyrene  Poor  Good  Good  Polysulfone  Good  Good  Good  Polyurethane  Poor  Good  Good  Polyvinyl chloride  Varies  Varies  Good  Polyvinylidene fluoride  Good  Good  Good  Silicone Teflon  Good Varies  Good No  Good Good  1  4.1 Ethylene Oxide Sterilization Considerations  It is possible to design an effective EtO sterilization process for almost every medical device, but the challenge lies in the large number of variations possible. Temperature, pressure, humidity, EtO concentration and gas dwell time are all critical parameters in an EtO sterilization cycle (Sordellini 1998).  Other factors that must be considered when contemplating EtO methods are physical configuration, material of the device, the response of the device to pressure changes, the natural bioburden (number of microorganisms typically present on a device prior to sterilization), the maximum heat tolerance, and any potential chemical reactions with water vapour or ethylene oxide (Sordellini 1998).  Complex devices can be difficult to sterilize using EtO due to the difficulty associated with gas penetration to parts of the device with limited accessibility. In order to overcome this problem, a 98  more stringent sterilization cycle can be employed. This may involve increasing the depth of the vacuum, increasing gas concentration or increasing the temperature in the chamber. However, each of these processes may have a negative effect on the device. EtO can diffuse several millimeters into surfaces but may have limited effectiveness if a product is very dense (Leventon 2002).  Another important factor in EtO sterilization is the propensity of the device material to absorb EtO. Plastics tend to absorb a much larger amount of EtO than metals (Leventon 2002). The more EtO absorbed, the more must be removed from the product during aeration, prior to use. Residual analysis is a key step in any sterilization validation trial involving EtO. A list of the response of various plastics to EtO sterilization is shown in Table 2.  4.2 EtO Sterilization Validation  The two tests that need to be performed during EtO sterilization validation are sterility assurance and residual analysis. Table 3 shows the cost, number of samples, and the time necessary to complete an EtO validation at a major medical device testing facility (Nelson Laboratories Inc., Salt Lake City, UT). The price quoted assumes no failures of the devices during the validation process. A n additional $250 is charged for each additional test that must be performed due to device failure. This price also does not include the cost of preparing an FDA report, which would cost an additional $275. Such a report would be necessary if the researcher plans on applying for Health Canada approval. This is not necessary for clinical trials being completed at VCHRI but would be necessary in the future if the device is to be used at any other institution.  99  Table 3: EtO Sterilization Validation Trial Cost # Samples Time Test (wks) Sterility  $3,300  •>  3  Residual Analysis  $275  3  1.5  Total  $3,575  6  4.5  4.3 R a d i a t i o n S t e r i l i z a t i o n C o n s i d e r a t i o n s  Gamma and electron-beam sterilization are among the most popular choices for sterilization of devices containing polymers. However, the sterilization process can cause changes to the polymer including embrittlement, stiffening, softening, colour change, production of an odor, changes to melt temperature, or changes in chemical resistance (Hemmerich 2000).  The changes to mechanical properties are a result of excitation or ionization of atoms (Hemmerich 2000). The two main mechanisms which can affect the mechanical properties of the device are chain scission and cross-linking. Chain scission is the random rupturing of bonds; this reduces the molecular weight, and consequently the strength, of the polymer. Cross-linking of adjacent long molecules results in large three-dimensional molecular networks. Initially this can cause an increase in tensile strength of the material. Cross-linking can also decrease impact strength and cause the polymer to become more brittle with increased dose (Hemmerich 2000).  In most polymers, a combination of both of these processes will occur upon irradiation. It can therefore be difficult to predict the effects of radiation on your device. Table 2 gives a general guide to the response of various plastics to radiation. For a more detailed description of the response of polymers to irradiation, please see Appendix 10: Radiation Stability of Polymer Materials.  100  Gamma radiation generally has more deleterious affect on materials than electron-beam radiation. The advent of materials designed to minimize radiation-induced cross-linking and chain scission in the last decade has reduced the effects of both types of radiation on polymers. These materials are typically infused with anti-rads which either combine readily with free radicals generated in the polymer during radiation or act as primary energy absorbers which prevent the radiation from interacting directly with the polymer. However, care should still be taken when selecting a sterilization method (Leventon 2002). Electron-beam radiation is compatible with about 70% of medical devices in production (Leventon 2002) but items containing batteries or electronic components may be adversely affected.  E-beam sterilization has a surface penetration of about 50% that of gamma irradiation, it is therefore possible that the beams will not reach every small crack. Shadowing of certain portions of the device may also occur if there is any stacking of the device within its container. In order to prevent any dangers associated with these problems, dosimeters are placed in hard to reach places to record the radiation dose. This dose mapping process ensures that the entire device is exposed to the minimum dose required for sterilization. As a reference, the minimum radiation dose for devices manufactured in a clean room is typically 15 kilo Gray (kGy). Devices manufactured in regular factory conditions usually require 25 kGy and devices that experience more handling during manufacture require 35 kGy (correspondence with Alex English, Quality Manager, Iotron Industries Canada, May 22 , 2007). nd  4.4 Radiation Sterilization Validation The tests necessary for sterilization validation of irradiation are threefold: sterility assurance, material response, and dose mapping. Table 4 outlines the time, number of samples, and costs for 101  completion of a radiation validation trial by a major US laboratory (Nelson Laboratories Inc., Salt Lake City, UT). These costs do not include the price of an F D A report nor the cost of producing the devices required. If desired, an FDA report would be an additional $275.  Table 4: Radiation Validation Completed by a well known American Laboratory. Test  Cost  # Samples  Timeline  Sterility  $1,400  30  7-8 wks  Material Response  $400  10  1 wk  Dose Mapping  $850  Total  $2,650  1 wk 9-10 wks  40  However, when we looked into finding local companies to complete the same trials, we were able to accomplish the testing with a smaller number of samples, in a shorter timeline, and for a lower cost. Table 5 shows the time, number of samples, and costs associated with completing the validation trial using qualified local companies (Iotron Industries Inc., Port Coquitlam, B C and IG Micromed, Richmond, BC).  Table 5: Radiation Validation Completed By Qualified Local Companies •  1  Test  1  Cost  1  # Samples  1  Timeline  Sterility  $600  12  2 wks  Material Response  $400  5  1 wk 1 wk  Dose Mapping  Total  1  $1,000  17  4 wks  4.5 Steam Sterilization Considerations  102  Steam sterilization is a popular sterilization method due to its simplicity and low cost. It produces virtually no waste and equipment costs are approximately one third that of EtO and one quarter that of radiation equipment. Steam sterilization is an effective method of killing microbial organisms as well as inactivating resistant nonpathogenic thermopile spores and extremely resistant prions that survive other sterilization methods (Rogers 2006).  That being said, steam sterilization is rarely used in the medical device industry. The temperatures necessary for this procedure range from 100°C to 134°C, as can be seen in Table 6. Surgical instruments, which are typically made of metal, can easily withstand these extreme temperatures. Most medical devices, however, do not retain function after being exposed to these conditions because of damage to plastic or electronic components. Table 2 shows the responses of various plastics to steam sterilization.  Table 0.6: Time required for steam sterilization at different temperatures.  Temperature 132°-134°C  Time (min) 3-4  121°C  8-30  115°C  35-45  lire  80-180  4.6 Steam Sterilization Validation  The tests required for validation of a steam sterilization method are sterility assurance and drytime verification. The cost of performing these tests at a large US laboratory (Nelson  103  Laboratories, Salt Lake City, UT) are shown in Table 7. Preparation of a formal F D A report would also cost an additional $275. Table 7: Steam Sterilization Validation Completed by a well known American Laboratory.  # Test  Cost  Samples  Timeline  Sterility  $1,850  3  3  Dry-Time Verification  $750  Total  $2,600  1 wk 3  4 wks  4.7 Sterilization Packaging IP*-  Packaging for sterilization by steam and EtO is available in the SPD of all V C H A sites. These are typically Wipak Medical Steriking products (Winpak Medical, Winnipeg, MB). These envelopes are usually made of a see-through plastic film and a layer of medical paper (Figure 5).  Figure 5: Packaging for  Some see-through films are made from polypropylene and may  sterilization. Envelopes  therefore not be suitable for radiation sterilization. It is important to  usually consist of a sheet of medical paper and a plastic film (source: Wipak Medical Inc., www.wipak.com)  ensure that your sterilization packaging is suitable for your chosen  sterilization method. Table 8 outlines the performance of various Wipak products.  Table 8: Suitability of Packaging to various types of Sterilization.  (source: http://www.wipak.com/medical/safety suitability.html)  Description Steriking See-through  Special Properties  Temperature Durability  Steam  EtO  Radiation  Hot Air  Hydrogen Plasma  138°C/280°F  YES  YES  NO  NO  OK*  Heat-sealable, Pouches & Rolls;  porous, made of Polyester / transparent, Polypropylene film and peelable Medical Paper  104  Steriking Tyvek See-  Heat-sealable, through Pouches &  porous, Rolls; made of  100°C/212°F  NO  YES  YES  NO  YES  134°C/273°F  YES  OK*  YES  NO  NO  134°C/273°F  YES  OK*  YES  NO  NO  200°C / 392°F  NO  NO  YES  YES  NO  100°C/212°F  NO  NO  YES  NO  NO  transparent, Polyester / Polyethylene peelable fdm and White Tyvek Steriking Crepe  Creping increases  Papers; made of  flexibility, special  medical pulp fibers  clean pulp Creping & plastic  Steriking Nonwoven  fibers increase  Sheets; made of pulp  smoothness,  and plastic fibers  special clean materials  Steriking Hot A i r  Heat-resistant, Rolls; tubing made of transparent Polyamide Steriking Cover Bags;  Heat-sealable,  made of Polyester /  transparent,  Polypropylene film  peelable  Researchers should contact SPD in order to determine what products are available and contact Wipak directly to inquire about alternatives.  4.8 Routine Sterilization of the Device  The final consideration one should make when deciding on a sterilization technique is the availability and cost of routine sterilization of the device. In a large urban centre such as Vancouver, we have accessible to us every possible modality. However, there remains a large 105  difference in price amongst the various techniques. Steam sterilization, which can be performed in the sterile processing department of any hospital, costs just pennies to operate. The only cost associated with this technique is the labour necessary to disassemble and reassemble the device pre and post sterilization. Ethylene oxide, which can also be accomplished by the SPD at major hospitals such as Vancouver General Hospital, costs approximately $150 per run, plus the cost of assembly/disassembly labour. Radiation sterilization is typically about $400 per run, but one run can accommodate multiple devices, depending on their size. Figure 6 shows a comparison of the cost of sterilization validation and routine sterilization of 5 devices. This figure illustrates a key trade off that must be considered in order to determine the optimum sterilization procedure for your device. $4,000  • Routine Cost/ 5 devices  Steam  EtO  Radiation (NL)  Radiation (local)  Figure 5: Sterilization Costs. Sterilization validation costs (black) and the cost of routine sterilization of 5 devices (gray) differ significantly from one sterilization technique to another.  106  Appendix 5: Clinical Research Ethics Board Application: Protocol for DLIMN Trial Protocol for Evaluation of a Computer-Assisted Technique for Distal Locking of Femoral Intramedullary Nails  Background Literature Review: Mid-bone fractures of long bones such as the femur and tibia are often repaired using intramedullary nails. The implant used is a hollow tube of 10-15 mm in diameter that is placed into the central canal of the bone. The implant is locked into place with screws that are drilled through the bone and implant at both the proximal and distal ends. Inserting the nail into the intramedullary canal often leads to unpredictable deformation of the implant. This makes it difficult for surgeons to identify the position of the distal locking holes. The conventional solution involves the use of a portable x-ray machine (fluoroscope) to visualize the holes. The surgeon must adjust the positioning of the fluoroscope until the holes appear as perfect circles. They must then place a radiolucent drill guide in the correct orientation and drill under x-ray guidance. This method has two major disadvantages: it is not time efficient and it exposes the surgical team to a significant amount of radiation. Distal locking of the implant accounts for over 23% (11 minutes) of the total surgical time and up to 53% of the total fluoroscopic screening time [Krettek 1999]. The accumulation of radiation exposure over the course of an orthopaedic surgeon's career is a rising concern and has received significant attention in the literature of late [Hafez 2005, Singer 2005, Madan 2002, Herscovici 2000]. A l l of these studies emphasize the need to minimize fluoroscopic time and to avoid being in the direct path of the x-ray beam. Hafez, in particular, has argued that previous studies have significantly underestimated the actual radiation exposure received at a surgeon's fingertips, especially for surgeons early in their training. Several alternative locking techniques have been developed in order to address these problems. These alternatives include nail mounted mechanical guides, self-locking nails, computer-assisted fluoroscope navigation and magnetic-based targeting systems. Unfortunately, none of these methods have come into widespread use. The first generation of nail-mounted mechanical guides was unsuccessful due to the difficulty of quantifying the deformation of the nail as it enters the intramedullary canal. A more recent mechanical guide requires the surgeon to drill a 6-10 mm working channel into the cortex of the bone near the pin hole locations so that the screw holes can be palpated and located directly using a guide-hook. [Krettek 1999] This method is unpopular due to the additional time required and the risks associated with drilling a working channel, which can cause stress concentrations which weaken the bone [Whatling 2005]. Self-locking nails have obvious advantages in decreasing both surgery time and radiation exposure. However, the two different types of self-locking nails proposed to date - self-engaging bolts and inflatable nails - have been heavily criticized for their lack of torsional stability. These implants are still in the prototype stage but researchers and surgeons have not expressed confidence that this solution will take off [Whatling 2005]. More recently, a technique known as virtual fluoroscopy has been developed in which two or more fluoroscopic images are acquired while the position of the patient is tracked, typically with an optical position measurement system. If the optical tracker also simultaneously measures the position of a surgical tool such as a drill, the current position of the drill can be displayed on the previously-acquired x-ray images without requiring additional exposures. This enables distal nail locking to be accomplished with as few as two single-shot x-ray exposures. On the positive 107  side, this method has succeeded in reducing both x-ray exposure time (7.3 vs 108 seconds) and the time necessary for distal locking (13.7 vs 19.9 minutes), but using the system generally requires more surgical time overall due to equipment setup (approximately 40 minutes) and tool registration processes [Suhm 2004]. Although there are over 500 hospitals worldwide currently using virtual fluoroscopy techniques for intramedullary nailing or other orthopaedic procedures, the relatively high equipment costs ($100,000-250,000) and the additional overall surgical time needed have limited their widespread use [Hazan 2003]. Very recently, a novel approach has been proposed which involves the use of a magnet and magnetic field sensor. This radiation-free technique relies on the fact that the implants used for tibial and femoral fracture contain a small open space within the centre of the nail. A permanent magnet on a probe is inserted into this space and is sensed by a magnetic sensor integrated into a drill-mounted guiding system. This system allows the drill to be aligned with the axis of the locking holes by sensing local changes in the magnetic field [Hollstien 2006]. A prototype system has recently been licensed to DePuy Products Inc. We believe that this concept is promising, but the use of a system based on this concept would be largely limited to this single procedure. It would therefore be difficult to promote to surgeons. In contrast, navigation equipment for a wide variety of computer-assisted orthopaedic procedures is becoming relatively common and its market share will likely continue to increase. We therefore propose to develop a radiation-free targeting process based on a more general-purpose measurement tool that can be used in a variety of surgical procedures. In order to determine the location and orientation of the holes for the Distal Locking of an Intramedullary Nail (DLIMN), we require a sensor which can be placed within the nail itself and be detected with sufficient accuracy by an instrument outside the bone. The recent development of 6 degree of freedom magnetic sensors for the Aurora Electromagnetic Tracking System (Northern Digital Inc, Waterloo, ON) makes this system a likely candidate for this intramedullary nail locking application. The 5 DOF version of the Aurora system is already commonly used in computerassisted surgical procedures, which makes it a more flexible choice than a system developed solely for this procedure. Purpose of Study: We have developed a computer-assisted surgical technique for distal locking of an intramedullary nail (DLIMN) used for femur fracture fixation which is intended to decrease exposure to radiation by reducing the need for fluoroscopic imaging. It can also potentially reduce the overall time of the operation. In lab testing, we have demonstrated that the sensor has sufficient accuracy when inserted down the nail to make our proposed approach feasible. However, in order to move our technique towards clinical deployment, we need to determine whether it can reliably measure the hole locations under clinical conditions, so the primary purpose of the proposed study is to measure in live surgeries the relative positions of the hole axis as estimated by our system and the reference screw axis as found using the conventional fluoroscopic technique. This will also give us preliminary insight into clinical practice issues which may need to be addressed in our final design and into the length of time that might be required for our computer-assisted procedure relative to the conventional technique. Hypotheses: 1. The proposed C A S approach is sufficiently accurate to allow the surgeon to target the locking screws without using fluoroscopy. 2. The proposed CAS approach is efficient enough to decrease total surgical time.  108  Justification: The current fluoroscopic-based targeting method for D L I M N exposes the surgical team to radiation which, when accumulated over a working lifetime, may reach harmful levels. The method also requires a high level of skill to properly adjust the positioning of the fluoroscope and align the drill appropriately based on the two-dimensional images provided by the fluoroscope. Our C A S approach is intended to eliminate the use of radiation during the targeting phase of locking the distal end of the nail and to provide an intuitive 3D targeting display which may significantly shorten the learning curve and decrease the time needed to perform this procedure, thereby decreasing the time during which the patient is under anaesthetic and increasing the efficiency of the operating room. Objectives: The primary outcome of this study will be an estimate of how closely the drill axis found using the proposed C A S technique approximates the axis actually drilled by surgeon using the standard fluoroscopic technique. The times required to find the axis using the standard fluoroscopic technique and to insert the navigation instrument will also be recorded in order to determine the potential impact on surgical time. Research Design: Patients presenting at Vancouver General Hospital with diagnosed femoral shaft fractures will be invited to participate in this study. Consenting patients who meet the inclusion criteria will undergo standard Intramedullary (IM) Nailing of the femur. Our intent is to enroll approximately 10-15 patients at this stage as the purpose of this study is to obtain some pilot data which will allow us to better estimate the variances needed for computing the number of patients required in a future randomized control trial. This surgical procedure consists three stages: 1. Opening of the Intramedullary Canal 2. IM Nail Insertion 3. Stabilization of IM Nail Using Locking Screws The first two stages of this procedure will be unaffected by our study. The third stage of the procedure is comprised of two steps: distal and proximal nail locking. It is at this point in the procedure that small deviations from the standard protocol will take place. After the I M Nail has been inserted, we will introduce two sensors into the OR. The sensors used will be Aurora Electromagnetic sensors (Northern Digital Inc, Waterloo, ON). The electromagnetic field produced by this system is not harmful to patients or the surgical team. It has been used previously in several experimental surgical procedures designed to improve navigation in biopsy, angiography and radiotherapy. [Krucker 2006, Wood 2005, Kirchstein 2005] One sensor will be inserted into the hollow canal down the centre of the IM Nail. In order to protect the sensor and fix it into place within the nail, an Insertion Tool (IT) has been designed. The IT is a cylindrical tube made of A B S i (P500), a special grade of A B S plastic that is resistant to chemicals that medical devices often come into contact with. The distal end of the IT is designed to fit snuggly into the distal locking holes of the I M Nail. The IT will be pre-calibrated so that the location of the locking holes relative to the location of the sensor is known presurgery. The second sensor will be fixed to the drill used for distal drilling. Since there are a variety of drills available to complete this stage of the surgical procedure, intra-operative calibration is necessary. Once the sensor is in place, 30 seconds of data will be collected while the drill is moved in a prescribed way. This will allow us to accommodate the surgeon's choice of drills. Both the IT and sensors can be sterilized using a standard ethylene oxide gas sterilization protocol. 109  Current Protocol  Experimental Protocol  1. OpenIM Canal  1. Open I M Canal - no changes  2. Reduce fracture / insert I M nail  2. Reduce fracture / insert I M nail - no changes  3. a. Distal Locking - align drill using fluoroscope - insert locking screw  3. a. Distal Locking -attach drill sensor and calibrate (2 min) -insert IT (1.5 min) - align drill using fluoroscope -collect data from both sensors (1 min) -remove IT (1 min) - insert locking screw -remove drill attachment (1 min) Total Extra Time = 7-8 minutes  b. Proximal Locking  b. Proximal Locking- no changes  Once both of the sensors are in place, the surgeon will proceed with aligning the drill using the standard fluoroscopic protocol. Once the drill is in the correct position, data will be collected from both the IT and drill sensors. The IT will then be removed from the I M Nail and drilling will proceed in the standard manner. The sensor on the drill will then be removed and the remainder of the procedure will be unaffected. Lab-based testing has shown that inserting and removing the sensors and calibrating the drill will add less than 8 minutes to the overall surgical time. Risks: The only risk associated with this study is potential failure of the Insertion Tool (eg, separation of the sensor from the insertion tube or breaking off of the locking tips). The forces involved in IT insertion and removal are very small and make this a very unlikely event. If failure of the IT does take place, it will be contained within the small inner canal of the I M Nail where it poses no immediate risk. The IT will be removed prior to inserting any locking screws. When it is removed, it will be carefully inspected for any damage. If any portion of the IT is missing, the I M Nail can and will be removed from the patient and the missing piece located. The I M Nail will then be re-inserted into the patient. This entire process will take less than 10 minutes and will not adversely affect the patient. If any unforeseen complications occur during surgery, the sensors can be quickly removed and the trial can be abandoned. Since the actual nail locking is performed using the standard procedure, there is no risk associated with instrumentation errors. Statistical Analysis: Our primary goal is to measure the difference between the drill axes as measured by the new magnetic sensor and as actually drilled using the standard fluoroscopic technique. We will measure the absolute angular deviation between these two axes across at least 10 patients and will report the mean absolute angular deviation and the 95% CI for this estimate. 110  Our secondary goal is to estimate the amount of time needed to perform targeting using the CAS approach compared with the time needed for the conventional fluoroscopic approach. We propose to use a (non-parametric) ranked pairs Wilcoxon test to test for differences in the mean time needed. Anticipated Impact: This study will enable us to quantify differences between the drill axes as targeted by the conventional and proposed techniques. If the C A S technique reliably locates the target axis within a small range around the actual drill axis, we will be justified in moving forward with a randomized control trial in which surgeons actually use the C A S technique to target the distal locking screws. Measured differences in time needed to perform the targeting process (presuming that the C A S technique does take less time to perform) may also support further development of this technique. If the C A S technique ultimately proves successful, this intervention will be able to be performed with decreased radiation exposure to orthopaedic surgeons, healthcare workers, and patients and with improved utilization of scarce operating room resources through decreased operating time. References: Hafez M A et al. Radiation Exposure to the Hands of Orthopedic Surgeons: are we underestimating the risk? Arch Orthop Trauma Surg 2005; 125:330-335 Whatling G M , Nokes L D M . Literature Review of current Techniques for the insertion of distal screws into intramellary locking nails. Injury, Int. J. Care Inj. 2005; 1-11 Krettek C, Konemann B, Miclau T, Kobli R, Machreich T, Tscherne H. A Mechanical Distal Aiming Device for Distal Locking in Femoral Nails. Clin. Re. Rel. Res. 1999; 364:267-275 Suhm N , Messmer P, Zuna I, Jacob L, Regazzoni P. Fluoroscopic guidance versus surgical navigation for locking of intramedullary implants A prospective,controlled clinical study. Int. J. Care Injured 2004;35:567-574 Herscovici D et al. The effects, Risks, and Guidelines for Radiation Use in Orthopedic Surgery. Clin, Orthop. Rel. Res. 2000; 375: 126-132 Hollstien, DS. Method and Apparatus for Distal Targeting of Locking Nails in Intramedullary Nails 2006; U.S. Patent 7 029 478 B2 Hazan E, Joskowicz L. Computer-Assisted Image-Guided Intramedullary Nailing of Femoral Shaft Fractures. Tech. Orthop. 2003; 18(2) Slomczykowski M A , Hofstetter R, Sati M , Krettek C, Nolte LP. Novel Computer-Assisted Fluoroscopic System for Intraoperative Guidance: Feasibility Study for Distal Locking of Femoral Nails. J. Orthop. Trauma 2001; 15(2): 122-131 Krucker J., X u S., Viswanathan A., Shen E., Glossop N . , Wood B. Clinical evaluation of electromagnetic tracking for biopsy and radiofrequency ablation guidance. Int J CARS 2006; 1:169-171. Wood B.J., Zhang H., Durrani A., Glossop N . , Ranjan S., Lindisch D., Levy E., Banovac F., Borgert J., Krueger S., Kruecker J., Viswanathan A., Cleary K. Navigation with Electromagnetic Tracking for Interventional Radiology Procedures: A Feasibility Study. J Vase Interv Radiol 2005; 16:493-505 Kirchstein U . , Kronberg K., Hein A., Concept and Preliminary Experiments of Navigated Imaging with an Electromagnetic Position Measurement System, CURAC 2005; Berlin, September 22-24,2005 Singer G. Radiation Exposure to the Hands From Mini C-Arm Fluroscopy. J Hand Surg 2005; 30A: 795-797  111  Madan S., Blakeway C. Radiation Exposure to surgeon and patient in intramedullary nailing of the lower limb. Int J Care Injured 2002; 33:723-727 Blattert T.R., Fill U.A., Kunz E., Panzer W., Weckbach A., Regulla D. Skill Dependence of radiation exposure for the orthopaedic surgeon during interlocking nailing of lon-bone fractures: a clinical study. Arch Orthop Trauma Surg 2004; B124:659-664  112  Inclusion Criteria You are eligible to participate in this study if you satisfy each of the following conditions: • You are at Vancouver General Hospital with a femoral shaft fracture which surgeons feel will be best treated with a femoral nail • You are over 19 years of age •  You are competent to give informed consent in English  Exclusion Criteria You should not participate in this study if any of the following applies: • You are currently under 19 years of age • You are unable to give informed consent in English Study Procedures: A l l preparations and procedures for the operative treatment of your fracture will be the same whether you choose to participate or not. The only difference i f you choose to participate is that prior to inserting the locking screws, your surgeon will attach two position measurement devices to the surgical equipment. One will be attached to the drill and a second will be inserted into the hollow shaft of the nail. The surgeon will then prepare to insert the locking screws using the standard x-ray based method. Just before they insert the screws, we will measure where the nail and the drill are. We will then remove the measurement device inside the nail and will insert the screws in the usual manner. We will also remove the measurement device from the drill and the rest of the procedure will continue unaffected. The attachment, data collection and removal of the devices will add less than 8 minutes to the overall surgical time. The sensor is electromagnetic and is not harmful in any way. You will not be asked to complete any post-operative questionnaires or procedures outside of the standard care for your type of fracture. Reimbursement: You will not be paid for your participation. U B C , as a public institution, and the investigators may seek patent rights to the device being used in this study and may seek to commercialize it in future. Your participation in this study will not entitle you to any rights related to the device's commercialization, should it occur. Benefits: You will not benefit directly from the study. Regardless of whether or not you choose to participate, your fracture will be cared for by a surgeon using modern treatments and you will be followed by a team with a special interest in your type of fracture. Your participation may assist in the development of an improved surgical method that may be of value to future orthopaedic patients. Risks: The risks involved in this study are minimal. There is a very low chance that the tube or wire connected to the sensor will break while the sensor is still inside the nail. However, since the sensor is enclosed by the femoral nail, a break cannot cause any harm. The sensor will be removed before the nail is locked into place and will be inspected carefully for damage. If any portion is missing, the nail will be removed and the broken piece will be retrieved. The nail will then be re-inserted. 114  By signing below I consent to participate in this study.  Subject Signature  Date  Printed Name of the Subject signing above.  Witness Signature  Date  Printed Name of the Witness signing above.  Principal Investigator / Delegated Representative  Date  Printed Name of the Principal Investigator or Delegated Representative signing above.  Appendix 7: Clinical Research Ethics Board Application: Online Application Form H06-03500 Computer-Assisted DLIMN (Version 1.0) Principal Investigator: Antony J. Hodgson 1. Principal Investigator & Study Team - Human Ethics Application fView Forml 1.1. Principal Investigator Please select the Principal Investigator First „ , c -i (PI) for the study. Once you hit Last Department Email Select, you can enter the Pi's Name Name name, or enter thefirstfew letters . . UBC/Applied Antonv of his or her name and hit Go. Hodgson • Science/Mechanical ahodgson@mech.ubc.ca You can sort the returned list Engineering alphabetically by First name, Last name, or Organization by clicking the appropriate heading. T  1.2. Primary Contact Provide the name of ONE primary contact person in addition to the PI who Last will receive ALL correspondence, Name certificates of approval and , , Johal notifications from the REB for this study. This primary contact will have online access to read, amend, and track the application. T  1.3. Co-Investigators List all the Co-Investigators of the study. Last These members WILL have onlineName access which will allow them to read, amend and track the Beadon application. These members will be listed on the certificate of G approval (except BC Cancer uy Agency Research Ethics Board O'Brien certificates). If this research nen application is for a graduate degree, enter the graduate student's name in this section.  First Name „ Raman  Institution/Department Rank  First Name  Institution/Department  Last  Rank  UBC/Applied Katherine Science/Mechanical Engineering  Graduate Student  Pi rre lerre  UBC/Medicine, Faculty f/Orthopaedics  0  Assistant Professor  Peter J e er .  0  UBC/Medicine, Faculty f/Orthopaedics  Associate Professor  1.4. Additional Study Team Members - Online Access List the additional study team members who WILL have online access to Last Name read, amend, and track the application but WILL NOT be listed on the certificate of approval. 1.5. Additional Study Team  Research Coord/Nurse  First Name  First  Institution/Department  Institution/  Rank  Rank/Job Email 117  Members - No Online Access Click Add to list study team members who WILL NOT have online access to the application and will NOT be listed on the certificate of approval.  Name  Name  Department  Title  Address  1.6.1. All Graduate Students:  Yes  1.6.2. All Medical Residents:  N / A (no medical residents participating in this study)  Comments: 1.7. Project Title Enter the title of this research study as it will Evaluation of a Computer-Assisted Technique for Distal appear on the certificate. If applicable, include the protocol Locking of Femoral Intramedullary Nails (Protocol 2.0) number in brackets at the end of the title. 1.8. Project Nickname Enter a nickname for this study. What would you like this study to be known as to the Principal Investigator and study team?  Computer-Assisted D L I M N  NOTE, if this application was converted to RISe from our previous database, ORSIL, here is the previous ORSIL application number for your information. 2. Study Dates & Funding Information - Human Ethics Application [View Forml 2.1. A. Start date:  March 1,2007  2.1. B. End date:  March 1, 2008  2.2. Types of Funds Please select the applicable box(es) below to indicate the type(s) of funding you are receiving to conduct this Other (Enter details below) research. You must then complete Grant section 2.3 and/or 2.4 to enter the name of the source offunds to be listed on the certificate of approval. If you selected Other, specify the N S E R C Industrial Postgraduate Scholarship type of funding below. 2.3. Research Funding Application/Award Associated with the Study Submitted to the U B C Number UBC Office of Research Services Please click Add to identify the research funding  Title  Sponsor  118  application/award associated with this study. Selecting Add will list the sources of all research funding applications that have been submitted by the PI (and the person completing this application if different from the PI). If the research funding application/award associated with this study is not listed below, please enter those details in question 2.4. Title . , , „ . r. Advanced Tools for ^ . . . Computer-Assisted » , Orthopaedic Surgery  2.4. Research Funding Application/Award Associated with the Study not listed in question 2.3. Please click Add to enter the details for the research funding application/award Industrial Postgraduate associated with this study that is Scholarship not listed in question 2.3. 0  Sponsor Natural Sciences and . . , Engineering Research _ ., % , Council of Canada (NSERC) &  r  Natural Sciences and Engineering Research Council of Canada (NSERC)  „ . -. IDetailsl r  metailsl  2.5. Conflict of Interest Do any of the following statements apply to the Principal Investigator, CoInvestigators and/or their partners/immediate family members? Receive personal benefits in connection with this study. For example, the researcher or study team being paid by the funder for each subject enrolled, or for yes consultancy. Have a nonfinancial relationship with the sponsor (such as unpaid consultant, advisor, board member or other non-financial interest). Have direct financial involvement with the sponsor (source of funds) via ownership of stock, stock options, or membership on a Board. Hold patent rights or intellectual property rights linked in any way 119  to this study or its sponsor (source offunds). 3. Conflict of Interest - Human Ethics Application [View Forml 3.1. In the box below, please answer the following questions, if applicable. Include the corresponding letter (A, B, C, D, E) before each answer. A. Describe any personal benefits that the investigators and/or their partners/immediate family members will receive, connected to this research study. B. State how much money (in Canadian A. Katherine Beadon has received an N S E R C Industrial dollars) the investigator or research group is paid by the Postgraduate Scholarship in Partnership with Northern Digital funder for each subject enrolled. Inc. (Waterloo, ON). This scholarship is made up of a $6000 C. Explain what this amount contribution from Northern Digital and $15,000 from N S E R C covers with respect to the direct over the course of one year. costs associated with doing this B. No money paid for subjects enrolled. research. D. Explain if this C. N / A amount covers any indirect costs D. N / A associated with doing the E. No relationship with funding agency (NSERC) or Northern [research, for example, providing Digital Inc. \the funder with advice on study [design, conference expenses for {presenting results or any other [costs. E. Explain if investigators \and/or their partners/immediate family members have a nonfinancial relationship with the sponsor (such as unpaid consultant, advisor, board member or any other type of nonfinancial relationship or interest). 3.2. In the last three years, how much money (in Canadian dollars) has each investigator (or any company owned or managed by each investigator) received from the funder andfor what \purpose did they receive these funds? (e.g. consultancy)  The PI has received an annual N S E R C Discovery Grant, currently in the amount of $30,370 per year. In addition, he was the recipient of a Collaborative Health Research Projects grant (CHRP) in the amount of $137,000 over three years (2003-2006) and an Equipment Grant in the amount of $14,514 (2004). Katherine Beadon began receiving an N S E R C Industrial Postgraduate Scholarship in the amount of $21,000 per year in September 2006.  \3.3. Give details if any of the | investigators and/or their [partners/immediate family \members have direct financial  no involvement  120  involvement with the sponsor (source offunds) via ownership of stock, stock options, or membership on a Board. 3.4. Give details if any of the investigators and/or their partners/immediate family members hold patent rights or intellectual property rights linked in anyway to this study or its sponsor (source offunds).  Pending positive results of this study, the PI intends to submit a patent application for the computer-assisted technique whose feasibility is being examined. Any patent prosecution will be undertaken by the University Industrial Liaison Office. Northern Digital is not interested in pursuing patent rights; they are only interested in the potential marketing of their product that would arise from licensing the technology to Computer Assisted Surgery companies.  3.5. Are the UBC COI declarations for the Principal Investigator and Co-Investigators Yes, all COI declarations are current. who are also UBC Faculty up to date? Comments: 4. Study Review Type - Human Ethics Application [View Forml 4.1. UBC Research Ethics Board Indicate which UBC Research Ethics Board you are applying to Clinical Research Ethics Board - Clinical and the type of study you are applying for: 4.2. Institutions and Sites for Study A. Enter the locations for the institutions and sites where the research will be carried out under this Research Ethics Board approval (including specimens processed by pathology, special radiological procedures, specimens obtained in the Institution operating room, or tissue requestedfrom pathology). Click Vancouver Coastal Health (VCHRI/VCHA) Add and enter the appropriate letter to see the locations for the institutions and sites where the research will be carried out under this Research Ethics Board approval: B for BC Cancer Agency Cfor Children's and Women's Health Centre of BC P for Providence Health Care Ufor UBC Campus Vfor Vancouver Coastal Health (VCHRI/VCHA). ! Ifyou are NOT using any of these i  Site Vancouver General Hospital  121  sites select N/A from the list. B. Please enter any other locations where the research will be conducted under this Research Ethics Approval (e.g. private physician's office, community centre, school, classroom, subject's home, in the fieldprovide details). 4.3. A. If this proposal is closely linked to any other proposal previously/simultaneously submitted, enter the Research Ethics Board number of that proposal. B. If applicable, please describe the relationship between this proposal and the previously/simultaneously submitted proposal listed above. 4.4. A. External peer review details:  The overall proposal for developing advanced computerassisted surgical tools and processes has been reviewed by N S E R C , although the specific project proposed here was not envisioned at the time of application for the Pi's N S E R C Discovery Grant.  B. Internal (UBC or hospital) peer review details:  Approval by supervisory committee (graduate student research). Most recent assessment occurred in January 25, 2006 committee meeting. Research protocol approved by Drs Antony Hodgson (Mechanical Engineering), Peter Cripton (Mechanical Engineering), Peter O'Brien (Orthopaedics) and Pierre Guy (Orthopaedics).  4.5. If this research proposal has NOT received any independent scientific/methodological peer review, explain why no review has taken place. 4.6. After reviewing the minimal risk criteria on the right, does your application fall under minimal risk (and therefore is eligible to be considered for Expedited Review)?  no  5 Summary nf Study and Recruitment - Human Ethics Application [View Form] 5.1. Please summarize the research proposal.  1) Purpose We have developed a computer-assisted surgical technique for distal locking of an intramedullary nail (DLIMN) used for 122  femur fracture fixation which is intended to decrease exposure to radiation by reducing the need for fluoroscopic imaging. It can also potentially reduce the overall time of the operation. In lab testing, we have demonstrated that the sensor has sufficient accuracy when inserted down the nail to make our proposed approach feasible. However, in order to move our technique towards clinical deployment, we need to determine whether it can reliably measure the hole locations under clinical conditions, so the primary purpose of the proposed study is to measure in live surgeries the relative positions of the hole axis as estimated by our system and the reference screw axis as found using the conventional fluoroscopic technique. This will also give us preliminary insight into clinical practice issues which may need to be addressed in our final design and into the length of time that might be required for our computer-assisted procedure relative to the conventional technique. 2) Hypotheses a) The proposed CAS approach is sufficiently accurate to allow the surgeon to target the locking screws without using fluoroscopy. b) The proposed CAS approach is efficient enough to decrease total surgical time. 3) Justification jThe current fluoroscopic-based targeting method for D L I M N | exposes the surgical team to radiation which, when | accumulated over a working lifetime, may reach harmful levels. The method also requires a high level of skill to properly adjust the positioning of the fluoroscope and align the drill appropriately based on the two-dimensional images provided by the fluoroscope. Our C A S approach is intended to eliminate the use of radiation during the targeting phase of locking the distal end of the nail and to provide an intuitive 3D targeting display which may significantly shorten the learning curve and decrease the time needed to perform this procedure, thereby decreasing the time during which the patient is under anaesthetic and increasing the efficiency of the operating room. .4) Objectives The primary outcome of this study will be an estimate of how closely the drill axis found using the proposed CAS technique approximates the axis actually drilled by surgeon using the standard fluoroscopic technique. The times required to find the axis using the standard fluoroscopic technique and to 123  insert the navigation instrument will also be recorded in order | to determine the potential impact on surgical time. 5) Research Design Patients presenting at Vancouver General Hospital with diagnosed femoral shaft fractures will be invited to participate | in this study. Consenting patients who meet the inclusion criteria will undergo standard Intramedullary (IM) Nailing of | the femur. Our intent is to enroll approximately 10-15 patients at this stage as the purpose of this study is to obtain some pilot data which will allow us to better estimate the [variances needed for computing the number of patients required in a future randomized control trial. (This surgical procedure consists three stages: 11. Opening of the Intramedullary Canal j2. IM Nail Insertion 13. Stabilization of IM Nail Using Locking Screws The first two stages of this procedure will be unaffected by our study. The third stage of the procedure is comprised of |two steps: distal and proximal nail locking. It is at this point I in the procedure that small deviations from the standard [protocol will take place. After the IM Nail has been inserted, [we will introduce two sensors into the OR. The sensors used will be Aurora Electromagnetic sensors (Northern Digital Inc, Waterloo, ON). The electromagnetic field produced by this system is not harmful to patients or the surgical team. It has been used previously in several experimental surgical procedures designed to improve navigation in biopsy, ; angiography and radiotherapy. One sensor will be inserted into the hollow canal down the centre of the IM Nail. In order to protect the sensor and fix it into place within the nail, an Insertion Tool (IT) has been designed. The IT is a cylindrical tube made a special grade of I ABS plastic that is resistant to chemicals that medical devices often come into contact with. The distal end of the IT is designed to fit snuggly into the distal locking holes of the IM Nail. The second sensor will be fixed to the drill used for distal drilling. Once both of the sensors are in place, the surgeon will proceed with aligning the drill using the standard fluoroscopic protocol. Once the drill is in the correct position, data will be collected from both the IT and drill sensors. The IT will then be removed from the IM Nail and drilling will proceed in the standard manner. The sensor on the drill will then be removed [and the remainder of the procedure will continue unaffected. 124  Lab-based testing has shown that inserting and removing the sensors and calibrating the drill will add less than 8 minutes to the overall surgical time. Both the IT and sensors can be sterilized using a standard ethylene oxide gas sterilization protocol. 6) Statistical Analysis Our primary goal is to measure the difference between the drill axes as measured by the new magnetic sensor and as actually drilled using the standard fluoroscopic technique. We will measure the absolute angular deviation between these two axes across at least 10 patients and will report the mean absolute angular deviation and the 95% CI for this estimate. Our secondary goal is to estimate the amount of time needed to perform targeting using the CAS approach compared with the time needed for the conventional fluoroscopic approach. We propose to use a (non-parametric) ranked pairs Wilcoxon test to test for differences in the mean time needed. 5.2. Inclusion Criteria. Describe the subjects being selectedfor this study, and list the criteria for over 19 years old their inclusion. For research femoral fracture involving human pluripotent stem Competent to give Informed Consent in English cells, provide a detailed description of the stem cells being used in the research. \5.3. Exclusion Criteria. Describe Patients under 18 •which subjects will be excluded incompetent to consent from participation, and list the \criteriafor their exclusion. \5.4. Provide a detailed These fractures usually present to the emergency department description of the method of where the Orthopaedic Trauma Service is called to examine Irecruitment. For example, the patient and determine appropriate treatment plans for the \describe who will contact Iprospective subjects and by what patient. Once preliminary tests, history and physical exams means this will be done. Ensure are completed the patient (if deemed to be eligible for [that any letters of initial contact participation) will be informed of the study and invited to participate by a designate of the surgeon or a member of the [or other recruitment materials are attached to this submission on health care team within Orthopaedic Trauma. Page 9. 15.5. Describe how prospective \normal/control subjects will be [identified, contacted, and \recruited, if the method differs from the above.  N A - No normals to be recruited  125  A l l femoral shaft fractures for planned I M nailing will receive standard clinical care regardless of their choice to participate in the study or not. The only difference if they choose to participate is that prior to inserting the locking screws, the surgeon will insert a sensor into the hollow shaft of the nail and a second sensor will be attached to the drill. The surgeon will then prepare to insert the locking screws using the standard fluoroscopic method. Just before the screws are inserted, data on the location of the sensor within the nail as well as the location of the drill will be collected. The sensor will then be removed and the screws will be inserted in the usual manner. 5.6. Summary of Procedures  The difference between the drill axes as measured by the sensor within the nail and the actual drill axis using the fluoroscopic method will be determined. The time required to align the drill, time to insert the two sensors, and time to remove the sensors will be recorded to compare the operative efficiency of the two methods. The sensors are electromagnetic and are not harmful in any way. This system has been used in numerous experimental navigation techniques in operating rooms around the world. The sensors and sensor fixation devices can be sterilized using the standard gas sterilization protocol.  6. Subject Information and Consent Process - Human Ethics Application [View Forml 6.1. How much time will a subject The additional surgical time for insertion and removal of the be asked to dedicate to the navigation instrument will be less than 10 minutes. project beyond that needed for normal care? 6.2. If applicable, how much time will a normal/control volunteer n/a be asked to dedicate to the project?  6.3. Describe what is known about the risks (harms) of the {proposed research.  The only risk associated with this study is potential failure of the Insertion Tool (eg, separation of the sensor from the insertion tube or breaking off of the locking tips). The forces involved in IT insertion and removal are very small and make this a very unlikely event. If failure of the IT does take place, it will be contained within the small inner canal of the I M Nail where it poses no immediate risk. The IT will be removed prior to inserting any locking screws. 126  When it is removed, it will be carefully inspected for any damage. If any portion of the IT is missing, the IM Nail can and will be removed from the patient and the missing piece located. The IM Nail will then be re-inserted into the patient. This entire process will take less than 10 minutes and will not adversely affect the patient. If any unforeseen complications occur during surgery, the sensors can be quickly removed and the trial can be abandoned. Since the actual nail locking is performed using the standard procedure without reference to the system being evaluated, there is no risk associated with instrumentation errors. 6.4. Describe any potential benefits to the subject that could Patients participating in this study will have the satisfaction of arise from his or her contributing to an improved surgical technique. participation in the proposed research. 6.5. Describe any reimbursement for expenses (e.g. meals, parking, medications) or payments/giftsin-kind (e.g. honoraria, gifts, prizes, credits) to be offered to No reimbursements will be provided. the subjects. Provide full details of the amounts, payment schedules, and value of gifts-inkind.  16.6. Specify who will explain the • consent form and invite the subject to participate. Include details of where the consent will be obtained, and under what circumstances.  i i  Consent will be obtained by either (a) the surgeon or (b) surgeon's designate or (c) the Orthopaedic Trauma research assistant. These kinds of injuries present to the Emergency Room where the Orthopaedic Trauma Service is responsible for assessing injuries and providing for appropriate treatment. This type of fracture requires 'Emergent' surgery and therefore necessitates the patient being informed of the study after assessment and prior to surgery. c  I |  As stated in 6.6 the time between assessment and surgery is 16.7. How long after receiving the often less than 24 hours but, the time is also difficult to j consent form will the subject haveestimate due to wait times and available operative resources. \to decide whether or not to It can therefore be approximated that anywhere from 2 to 8 \participate? If this will be less \than twenty-four hours, provide hours could be the time available for the potential subject to consider participating or not in the study. an explanation.  127  Will  , . If not, If Yes, If not, will every , explain 6.8. Will every subject be he/she be sub"ect be competent to give fully informed competent nature of the consent able to give how . Details of the will assent consent on his/her own behalf? to give incompetence on assent to will be Please click Select to complete fully his/her participate? sought. the question and view further informed behalf? details. consent? rDetailsl Yes T  J  6.9. Describe any situation in which the renewal of consent for N/A this research might be appropriate, and how this would take place. 6.10. What provisions are planned for subjects, or those consenting on a subject's behalf, No provisions have been planned - all subjects must be to have special assistance, if capable of consenting on their own behalf. needed, during the consent process (e.g. consent forms in Braille, or in languages other than English). 6.11. Describe any restrictions regarding the disclosure of information to research subjects (during or at the end of the study) none that the sponsor has placed on investigators, including those related to the publication of results. 7. Number of Subjects and Drugs - Human Ethics Application For Clinical Study [View Forml \7.1. Is this a multi-centre study [(involves centres outside of thoseno [appliedfor under this Approval?) 7.2. A. How many subjects (including controls) will be enrolled in the entire study?  15  \B. How many subjects (including controls) will be enrolled at 15 institutions covered by this Research Ethics Approval? Of these, how many are controls?0 7.3. Enter the name of any  N A - no investigational drugs are involved in this study 128  Igeneric drug(s), investigational lor marketed, used outside of its I approved indication. 7.4. Enter the name of any |NA - drugs outside of standard of care are not included in the marketed drug(s) used within its | study. approved indication. 7.5. Enter the name of any Natural Health Products used:  None  7.6. Enter the name of any new investigational devices, or N / A No devices are being used outside their approved marketed devices used in experimental mode, that will be indication. used outside of their approved indication. 7.7. Enter the name of any positron-emitting radiopharmaceuticals (PERs). 7.8. A. Name the sponsor/institution/investigator responsible for filing a Clinical Trial Application (CTA) or Investigational Testing Authorization (ITA) with Health Canada or Other.  N A - None  N/A  B. For clinical trials involving investigational drugs/devices or marketed drugs/devices outside of their approved indications (including natural health Not Applicable products and PERs), indicate whether or not approval has been obtained from the appropriate federal regulatory agency for this purpose. \C. If approval has been obtained lor a request for approval is in [progress, click Add to enter the Name of Agency name of the regulatory agency and the date of approval (if received).  Date of Approval  7.9. Click Add to enter the Health Canada No Objection Letter (NOL) control number and date Health Canada N O L Control Number of approval if available for the initial application and subsequent NOLs for amendments. A copy of  Date of Approval  129  the NOL from Health Canada must also be attached in question 9.1 when it becomes available. 7.10. Does this research fall within the categories of pluripotent stem cell research that need to be submitted to the CIHR Stem Cell Oversight Committee (SCOC)?  no  7.11. Registration for Publication of Clinical Trials A. Does this no clinical study fall within the definition stated on the right (in the guidelines)? B. If Yes, click Add to enter the following information.  Has it been .„ registered? "  7.12. Is there a requirement for this research to comply with United States regulations for research ethics?  no  l™^^ r. • * Authorized Registry , used:  Y Clinical . , . . , .... Trial unique identifier:  E n t e r  om  8. Data Monitoring- Human Ethics Application For Clinical Study [View Forml 8.1. Describe the provisions made to break the code of a doubleNone as the subject and surgeon will not be blinded. blind study in an emergency situation, and indicate who has the code. 8.2. Describe data monitoring procedures while research is ongoing. Include details of N/A planned interim analyses, Data and Safety Monitoring Board, or other monitoring systems. 8.3. Describe the circumstances under which the study could be stopped early. Should this occur, describe what provisions would N / A be put in place to ensure that the subjects are fully informed of the reasons for stopping the study. Subjects will be assigned a unique study number which is not 8.4. Describe how the identity of related to any identifying numbers. These study ids will be the subjects will be protected both kept in a flat file not linked to any system containing during and after the research identifiers. A l l identifying data is maintained on a password study. protected PC accessible only to the PI and Research Staff. 130  Correlated data will not contain identifiers. 8.5. Explain who will have access The PI, Research Staff will have access to identifying data but the identifying data will not be available to anyone else. A l l to the data at each stage of analysis will be done on grouped non-identified data. A l l processing and analysis, and hardcopies of data are kept in a locked cabinet in a locked what steps will be taken to office. Electronic data is kept on a password protected PC in a safeguard the confidentiality of locked office. the data at each stage. 8.6. Describe what will happen to The data may be submitted for publication to scholarly the data at the end of the study, journals and may be presented at established educational rounds and used for educational purposes. The data will be and what plans there are for maintained for 5 years. future use of the data. 9. Documentation - Human Ethics Application TView Forml Name Version Date Protocol: Computer-Assisted . „ January 25, DLIMN 2007  9.1. Protocol Please view the examples on the right, then click Add to enter the required Protocol: Computer-Assisted - „ information and attach the DLIMN -° documents. 1  February 27, 2007  2  Name Consent Form: ComputerAssisted D L I M N  9.2. Consent Forms Please view the examples on the right, then click Add to enter the required CF: Computer-Assisted information and attach the forms. D L I M N  9.3. Assent Forms Please view the examples on the right, then click Name Add to enter the required information and attach the forms. 9.4. Investigator Brochures/Product Monographs (Clinical applications only) Please click Add to enter the required information and attach the documents.  Name  9.5. Advertisement to recruit subjects Please view the examples Name on the right, then click Add to enter the required information and attach the documents. 9.6. Questionnaire, questionnaire Name cover letter, tests, interview  r  \^„,i  0  Version Date ~ March 28, 2007 ft  1  n U  January 23, 2007  Version  Date  Version  Date  Version  Date  Version  Date  r  ,  ,  r  rviewl ^ ~ L  r y  i  .  .  131  scripts, etc. Please click Add to enter the required information and attach the documents. 9.7. Letter of initial contact Please click Add to enter the required information and attach the forms.  Name  9.8. A. Other documents: Please view the examples on the right, Name then click Add to enter the required information and attach the documents.  Version  Date  Version  Date  B. If a Web site is part of this study, enter the URL below. Since URL's may change over time or become non-existent, you must also attach a copy of the documentation contained on the web site to one of the sections above or provide an explanation. 10. Fee for Service - Human Ethics Application for Clinical Study [View Form] Please indicate which of the following methods ofpayment will be usedfor this application. Enter information stating when the fee will be sent: 11. Hospital Information - Human Ethics Application for Vancouver Coastal Health [View Forml 11.1 Have you already received approval from VCHA to conduct no this study? If Yes, please provide the VCHA/VCHRI approval number (e.g. V06-0000) 11.2. A. Does the Principal Investigator in question 1.1 have no a medical appointment with VCHRI/VCHA and a UBC faculty appointment? 11.2.B. Does the Principal \lnvestigator in question 1.1 have a medical appointment with no VCHRI/VCHA (but not a faculty appointment at UBC), or is the Principal Investigator an 132  employee of VCHA? Select the Browse button to attach the declaration form. 11.2. C. Does the Principal Investigator in question 1.1 have yes a UBC appointment? Select the Site Investigator at lastName VCHA if different from the Guy Principal Investigator in question 1.1.  firstName Pierre  Vancouver Acute (Vancouver Acute encompasses the 11.3. Select the VCHA Health following sites: Vancouver General Hospital, U B C Hospital, Service Delivery Area(s) that will GF Strong Rehabilitation Centre, Arthritis Research Centre of be involved in this study. Canada, Mary Pack Arthritis Centre) \ 12. Save Application - Human Ethics Application [View Forml  133  Appendix 8: Approval to Conduct Research at Vancouver Coastal Health  .Vancouver  C6astalHea.ltH' Research Institute APPROVAL TO CONDUCT RESEARCH AT V A N C O U V E R C O A S T A L H E A L T H A U T H O R I T Y (VCHA)  All research studies and clinical trials involving human subjects {"Research Projects") that involve the use of VCHA services, facilities, records, staff or patients must receive approval from the Vancouver Coastal Health Research Institute ("VCHRI") office in the relevant Health Sen-ice Delivery Area ("HSDA") aad must be approved by a University of British Columbia ("UBC") research ethics board ("REB"). There are four VCHA HSDAs: Vancouver Acute. Vancouver Community, Richmond Health Services, and Coastal. If a Research Project will be conducted within more than one VCHA HSDA, the researcher must obtain approvalfromthe VCHRI office in each of the relevant HSDAs (contact information for each VCHRI research office is listed at the end of this document). Once approval to conduct research has been granted by the applicable VCHRI office, the Research Project may begin at that site. The approval process ensures that all research involving humans conducted at VCHA is reviewed from an ethical, safety and resource use framework. DEFINITION OF RESEARCH A Research Project, regardless of how it is funded, will be considered under the terms of VCHA and UBC policy as being research invoking human subjects if: •  A human is subjected to procedures, the purpose of which go beyond the subject's need for prophylaxis, diagnosis or therapy; or  •  A human is subjected to procedures which are experimental but which do not necessarily go beyond the subject's need for prophylaxis, diagnosis, or therapy: or  •  Procedures are used in which an invasion of privacy may be involved, for example, by examination of records, by interviews, by observations, by administration of a questionnaire or test, or by audio or video recording:or  •  Human tissue, biological fluids, embryos or fetuses are being studied; or  Projects condt«:ted for quality assurance, continuous quality improvement, or program evaluation ma)! be excludedfromthe above definition of research: however, individuals are encouraged to consult with a UBC REB if they are unsure as to whether their proj ect constitutes quality improvement, program evaluation or research. Irrespective of the purpose of the Research Project, no work can be published without having received UBC REB approval before commencing the Research Project. Publication includes any  V«r.:o-aDs!*. i8 May 2007  dissemination of results, by any mechanism (conference poster or presentation, web site, journal paper) outside of VCHA. SUBMISSION CRITERIA TO CONDUCT RESEARCH A T V C H A Research Projects (meeting the above definition), which meet any one of the following criteria, must be submitted to VCHRI for review and approval: 1.  Research Projects that are conducted at any VCHA HSDA:  2. Research Projects where VCHA patieuts/clients/residents/staff are participants in the Research Project; 3.  Research Projects where VCHA medical staff or VCHA employees participate in the conduct of the Research Project (Note: hi this case, criteria kl and/or criteria U2 must also apply).  OBTAINING APPROVAL TO CONDUCT RESEARCH AT VCHA Three processes must occur before research is approved and may begin at VCHA: 1.  Submission to a UBC REB for review and approval;  2.  Submission to the VCHRI office of the relevant HSDA(s) for review and approval (forms and guidelines for the research approval process at each HSDA are located at httn://www.vclul.ca/s/ClltticalTilals-Foims.aso; and  3.  Execution of the study contract/agreement (if applicable).  Submission for VCHRI research review, the UBC REB review and for study contract/agreement review may be initiated in parallel. The study contract/agreement, if applicable, must be signed by the principal investigator. VCHA and UBC. Final authorization for research to begin will be given by the appropriate /VCHRI HSDA when the above three processes are satisfactorily completed.  Version Date: 18 May 2007  Applications for approval to couduct research at VCHA should be directed to die VCHRI office of the relevant HSDA:  HSDA  Address!  For Information or Assistance;  Contact Information:  Coastal (Coastal eucompasses hospitals, community health centres and residential care facilities in the following sites: North Vancouver, West Vancouver, Garibaldi. Sunshine Coast, Bella Bella, and Bella Coola)  Lions Gate Hospital c/o Employee Engagement 231 East 15* Street* North Vancouver. BC V 7 L 2L7  Dr. Cynthia Hamilton Assistant Director, VCHRI (Coastal)  Tel: (604)988-3131 Ext 4703  Richmond Health Services (Richmond Health Services encompasses the following networks: Richmond acute care, community care, primary health care, mental health and addiction sites.)  Quality Initiatives Room 3004 7000 Westminster Highway Rielwiond, B C V 6 X 1A2  Ms. Marion Wardley Administrative Assistant Quality Initiatives  Tel: (604) 244-5209 Email: marion.wnrdleviiBvch.cn  Dr. Ingrid Sochting Assistant Director. VCHRI (Riclunond Health Services)  Tel: (604) 278-97 H L o c : 4610 Email: inaiid.sochtinafftivch.cn  Vancouver Acute (Vancouver Acute encompasses the following sites: Vancouver General Hospital (including Mary Pack Arthritis Centre), G.F. Strong Rehabilitation Centre, and U B C Hospital)  Vancouver Coastal Health Research Institute Room 163 - 2647 Willow Street  Ms.WyloKayle Administrative Assistant, Cluneal Trials Administration  Tel: (604) 875-4111 Loc: 68368 Email: wvlo.kavleti8vcu.ca  Vancouver. B C VSZ 3P1  Ms. Stephania Manusha Regional Manager. Clinical Trials Aduumstvation  Tel: (604) 875-5649 Email: stenhanin.inanushaffivch.ca  Vancouver Community (Vancouver Community encompasses community health centres, mental health centres, additional sites and residential care facilities in Vancouver)  #200-520 West 6 Avenue Vancouver. B C VSZ 4H5  Ms. Nena Pineda Admuiistrative Assistant Vancouver Community  Tel: (604)714-3787 Email: nena.piuednffivch.cn  Ms. Val Munroe Assistant Director, VCHRI (Vancouver Community)  Tel: (604) 730-7610 Email: vnl.munroe^Hxb.ca  Version Dae: IS May 2007  th  Email: . cvutliin.hnmlltonffivch.ca  GUIDANCE NOTE #6: INTERNAL MAILING INSTRUCTIONS/ADDRESS Include the mailing address where all correspondence regarding the Vancouver Acute research study submission should be sent.  GUIDANCE NOTE #7: TYPE OF FUNDING SOURCE Include the type of funding the Principal Investigator has received to conduct the research study.  GUIDANCE NOTE #8: NAME OF FUNDING SOURCE Include the name of the funding source(s).  GUIDANCE NOTE #9: TYPE OF STUDY Indicate the type of research study.  GUIDANCE NOTE #10: VANCOUVER ACUTE SERVICES OR RESOURCES REQUIRED FOR THE RESEARCH STUDY If a research study impacts on Vancouver Acute services or resources, the appropriate department or clinical unit approval must be obtained. The departments and clinical units require information about the resources that are required (i.e., clinic staff nursing time, clinic space, access to clinic patients for recruitment, lab technician time). It is the responsibility of the departments and clinical units to determine if those services will have sufficient impact as to require recovery from the research study budget to offset Vancouver Acute operating costs. It is the responsibility of the departments and clinical units to provide investigators with the cost of those services. The department's or clinical unit's signature indicates a willingness to participate and support the research study. In this section, the Principal Investigator must indicate which departments or clinical units the research study will impact. Enter the name of the clinic or ward that will be impacted on the form.  PHARMACY The Pharmacy Department must review all research protocols that involve drugs (this includes both investigational drugs and drugs that have been approved for market). Study protocols are reviewed for: • The level of Pharmacy involvement needed and the implications for pharmacy operations. Pharmacy involvement is necessary for all studies that involve administration of study drugs to patients within the hospital.  139  • The appropriateness of the comparator treatment. Is the study drug being compared to the "standard of care" at our institution? Are any "extra" drugs needed to comply with the protocol that Pharmacy normally would not use, and if so, are they appropriate and will the sponsor supply them? • Determine if pre-printed Dr's orders, computerized order entry, medication administration need records and drug information sheets (for patients and/or staff) are needed. • Assess randomization and blinding procedures for potential problems and degree of pharmacy involvement. • Ensures facilities are available to store study drug as directed by the Protocol (i.e. monitored fridge or locked cupboard). • Ensure the study sponsor will cover any costs incurred for the study. • Ensure study medications intended for outpatient use are labeled appropriately and meet regulatory requirements.  The pharmacy department requires the following documentation for review: • Cover letter; • A copy o f the research study protocol; and • A copy of the "Request for Approval to Conduct Research at Vancouver Acute" form.  Once the above-listed documents have been received, reviewed and the "Request for Approval to Conduct Research at Vancouver Acute" form has been signed, the Pharmacy Department will contact the person listed o n the cover letter to pick-up the signed form.  RADIOLOGY If a research study involves the services or resources of the Radiology Department, the following documentation must be submitted to the Radiology Department for review: • Cover Letter (Include the number of research subjects that will be enrolled, the length of the research study, the services/resources that are being requested); • Copy o f the study protocol; • "Request for Approval to Conduct Research at Vancouver Acute" form.  V G H : The above documentation should be sent to the attention of Christina Romero at G940 Ground th  Floor, 899 West 12 Avenue. Please note that effective March 1, 2007, the Radiology Department at V G H will be charging a $100 fee for the review of all industry sponsored clinical trials. This review fee will not be charged on grant-funded studies. U B C H : The above documentation should be sent to the attention of the MRI supervisors, Karen Smith or Leslie Costley, MRI Supervisor. Please note that if a research study involves the use of PACS, the Radiology Department must sign the "Request for Approval to Conduct Research at Vancouver Acute" form. If the Principal Investigator already has access to PACS for clinical purposes, a signature from the Radiology Department must be obtained to access PACS for research purposes.  140  CLINICAL LABORATORY If a research study involves the services or resources of the Clinical Laboratory, the following documentation must be submitted to the Clinical Laboratory Department for review: • Cover Letter (include the number of subjects and tests that are applicable to research); • Copy of the study protocol; • "Request for Approval to Conduct Research at Vancouver Acute" form.  The above documentation should be sent to the attention of Romy Chan, Technical Support Coordinator, Laboratory Administration. O P E R A T I N G R O O M (OR) All research studies that take place in the Operating Room and/or impact the resources of the Operating Room must be reviewed and approved by the Operating Room prior to the start of the research study. The Principal Investigator must submit the following documentation to Ms. Debbie Hendricks, Manager, Equipment and Supplies, OR, or to Ms. Lorraine Ensminger, the Operating Room Patient Service Manager (PSM) at the Operating Room Administration Office, JPP North, Room 2304-2, for review: • • • •  Copy of the study protocol; "Request for Approval to Conduct Research at Vancouver Acute" form; "OR Research Form"; and "Specimen Collection for Research - Special Handling Instructions" form (applicable only to studies involving the collection of tissue).  The Operating Room will review the above documentation for all research studies that involve the OR, regardless if the procedure is considered standard of care. If it is determined that the Operating Room will be impacted by the research study, the Manager, Equipment and Supplies, Operating Room, or the Operating Room PSM will advise the Principal Investigator if any additional documentation is required. The Operating Room will sign the "Request for Approval to Conduct Research at Vancouver Acute" form once all Operating Room requirements have been met. If it is determined that the Operating Room will not be impacted by the research study, the Manager, Equipment and Supplies, Operating Room, or the Operating Room PSM will state on the "Request for Approval to Conduct Research at Vancouver Acute" form that there is "no involvement of the Operating Room". For further details, please refer to "Guidelines For the Review and Approval of a Research Study Impacting V C H Operating Rooms" posted on the VCHRI website. If tissue specimens will be collected for research purposes during a surgical procedure in a Vancouver Acute Operating Room, Anatomical Pathology and the Operating Room must review and approve the study. Please see the "Guidelines For the Review and Approval of a Research Study Impacting V C H Operating Rooms" posted on the VCHRI website for more specific details on required forms and procedures.  If blood specimens will be collected for research purposes during a surgical procedure in a Vancouver Acute Operating Room, Anesthesia must review and approve the study. Please submit the "Request for Approval to Conduct Research at Vancouver Acute" form along with a cover sheet outlining the requirements of the Anesthesiologist and submit this documentation to 2449 JPP2N,  910 West 10th Ave, c/o Dr. Raymer Grant for review and approval. ANATOMICAL PATHOLOGY If a research study involves the services or resources of Anatomical Pathology, the following documentation must be submitted to the Anatomical Pathology Department for review: • • • •  "Request for Approval to Conduct Research at Vancouver Acute" form; "Anatomic Pathology Laboratory Resource Utilization" form; The Study Protocol; and "Specimen Collection for Research - Special Handling Instructions" form (studies involving the collection of tissue).  Documentation should be submitted to Anatomical Pathology, c/o John Garratt, Technical Supervisor, Anatomical Pathology. Please note that if V C H Anatomical Pathology will process tissue specimens collected at V C H , the V C H Pathologist who will be responsible for processing the tissue specimens must be named as a coinvestigator on the research study (on the "Request for Approval to Conduct Research at Vancouver Acute" form and on the UBC ethics certificate). G U I D A N C E N O T E #11: A P P R O V A L S I G N A T U R E S R E Q U I R E D If the research study involves the resources or services of departments or clinical units located at Vancouver General Hospital or Mary Pack Arthritis Centre, print the name of the individual signing on behalf of the department or clinical unit under the " V G H Site - Name of Signatory" Column. If the research study involves the resources or services of departments or clinical units located at UBC Hospital, print the name of the individual signing on behalf of the department or clinical unit under the "UBC Hospital Site - Name of Signatory" Column. If the research study involves the resources or services of departments or clinical units located at GF Strong Rehabilitation Centre, please use the "Request for Approval to Conduct Research at GF Strong" form located on the VCHRI website. Obtain the signatures of the appropriate department head, patient service managers or other managers whose department or clinical unit may be impacted by the research study or whose services the Principal Investigator wishes to utilize.  G U I D A N C E N O T E #12: D E P A R T M E N T C O S T A N A L Y S I S Indicate if a department cost analysis is required for any departments or clinical units the research study may impact. G U I D A N C E N O T E #13: D E P A R T M E N T H E A D A T V A N C O V U E R A C U T E , IF D I F F E R E N T F R O M UBC ETHICS  142  Obtain the signature of the Principal Investigator/Site Investigator's V C H Department Head if that person is different from the person who has approved the UBC ethics application. Please print the name of the individual who is signing the form. The signature of the Principal Investigator's/Site Investigator's V C H Academic Department, School or Program Head indicates that the Principal Investigator/Site Investigator at V C H has the qualifications experience, and facilities to carry out the research study. If the Principal Investigator is a Department Head, the person who the Principal Investigator reports to, must sign the form. G U I D A N C E N O T E #14: DIVISION H E A D S I G N A T U R E Obtain the signature of the Division Head of the Principal Investigator /Site Investigator at V C H . Please print the name of the individual who is signing the form. If the Principal Investigator does not have a Division Head (i.e. research nurse, patient service manager, respiratory therapist, director etc.), a signature is not required. G U I D A N C E N O T E #15: S U P E R V I S O R / M A N A G E R  SIGNATURE  If the Principal Investigator is a V C H employee (research nurse, patient service manager, respiratory therapist, director etc.,) the Principal Investigator must obtain his or her supervisor's or manager's signature. Please print the name of the individual who is signing the form. G U I D A N C E N O T E #16: P R I N C I P A L I N V E S T I G A T O R  SIGNATURE  Obtain the signature of the Principal Investigator/Site Investigator at V C H . G U I D A N C E N O T E #17: S T U D Y P E R S O N N E L a) Indicate the Principal Investigator's affiliation with V C H . Please select only one category. b) Indicate if there are research personnel involved with the conduct of the research study that are not affiliated with V C H .  G U I D A N C E N O T E #18: P E R S O N A L H E A L T H I N F O R M A T I O N "Personal information" is defined in The Freedom of Information and Protection of Privacy Act (British Columbia) as any recorded information about an identifiable individual (excluding business contact information). Personal information can be recorded in any format including books, documents, maps, drawings, photographs, letters, vouchers, papers, and any other thing on which information is recorded or stored by graphic, electronic, mechanical or other means. Personal information includes information that can be linked back to or can identify a specific individual through association or inference. For example, generic information about an individual (e.g., ethnic origin) could be linked to one or more individuals if they lived in a small town with a limited number of people with that ethnic background. Examples of personal information include but are not limited to:  143  Appendix 10: Radiation Stability of Polymer Materials (source Hemmerich 2000) Material Polystyrene  Radiation Stability Excellent  Polyethylene-various densities  Good/Excellent  Polyamides (nylon) Polyimides Polysulfone Polyphenylene sulfide  Good Excellent Excellent Excellent  Polyvinyl chloride (PVC) Polyvinyl chloride/Polyvinyl acetate Polyvinylidene dichloride (Saran) Styrene/acylonitrile (SAN)  Good  Polycarbonate  Good/Excellent  Polypropene natural Polypropelyne Stabilized Fluoropolymers: Polytetrafluoroethylene (PTFE) Perfluoro alkoxy (PFA) Polychlorotrifluoroethylene (PCTFE) Polyvinyl floride (PVF) Polyvinylidene fluoride(PVDF) Ethylene-tetrafluoroethylene (ETFE) Fluorinated ethylene propylene (FEP)  Good Good Good/Excellent  Poor/Fair  Comments High density grades not as stable as low or medium density grades nylons 10,11,12,6.6 are more stable than 6. Film and fiber are less resistant Natural material is yellow Yellows, antioxidants and stabilizers prevent yellowing. High molecular weight organetin stabilizers improve stability; color-corrected radiation formulations are available Less resistant than PVC Less resistant than PVC Yellows. Mechanical properties not greatly affected: color corrected formulations are avialable Physical properties greatly reduced when irradiated. Radiation-stabilized grades utilizing high molecular weights and copolymerized and alloyed with polyethylene should be used in most radiation applications. High-dose rate E-beam processing may reduce oxidative degredation.  Poor Poor Good/Excellent Good/Excellent Good/Excellent Good Fair  Celluloses: Esters Cellulose acetate proprionate Cellulose acetate butyrate Cellulose paper, cardboard  Fair Fair Fair/Good Fair/Good  Polyacetals  Poor  ABS Acrylics (PMMA)  Good Fair/Good  Polyurethane Liquid Crystal Polymer (LCP)  Good/Excellent Excellent  Irridation causes embrittlement. Color changes have been noted (yellow to green) High-impact grades are not as radiation resistant as standard impact grades Aromatic discolors. Polyesters more stable than esters. Retains physical properties. Commercial LCP excellent. Natural LCP not stable.  148  Polyesters Thermosets:  Good/Excellent  Phenolics Epoxies  Excellent Excellent  Polyesters Allyl diglycol carbonate (polyester) Polyurethanes:  Excellent Excellent  Aliphatic Aromatic Elastomers: Urethane EPDM Natural Rubber  Excellent Good/Excellent  Nitrile  Excellent Excellent Good/Excellent Good/Excellent  Polychloroprene (neoprene)  Good  Silicone Styrene-butadiene  Good Good Poor Poor Poor  Polyacrylic Chlorosulfonated polythylene Butyl  PBT not as radiation stable as PET Includes the addition of mineral fibers All curing systems. Includes the addition of mineral or glass fibers Maintains excellent optical properties after irradiation Darkening can occur. Possible breakdown products could be derived.  Discolors Discolors. The addition of aromatic plasticizers renders the material more stable to irradiation Phenyl-methyl silicones are more stable than are methyl silicones. Platinum cure is superior to peroside cure. Full cure during manufacture can eliminate most postirradiation effects.  Friable, sheds particulate.  149  Appendix 11: Aurora Approvals and Classifications Approvals and Classifications  10  Approvals and Classifications  10.1  Electrical Safety Approvals Table 10-1 Electrical Safety Approvals  Standard  Title  U L 60601-1, First Edition (2003)  Medical Electrical Equipment, Part 1: General requirements for safety  CAN/CSA C22.2 No. 601.1 - M 9 0 (R2001)  Medical Electrical Equipment. Part 1: General Requirements For Safety  CAN/CSA C22.2 No. 601.1.S1-94 (R1999)  Supplement No. 1-94 to CAN/CSA C22.2 No. 601.1 - M90  CAN/CSA C22.2 No. 601.1B-90 (R2002)  Amendment 2 to CAN/CSA C22.2 No. 601.1 - M90  IEC 606'0l-l:1988+Al:1991+A2:1995 Medical Electrical Equipment. Part 1: General Requirements for safety w/ amendments  10.2  EMC/EMI Approvals Table 10-2 EMC/EMI Approvals  Standard  Title  FCC CFR47, Part 15, Subpart B  Class B Unintentional Radiators  CISPR.il/EN55011  Class B , Industrial, Scientific and Medical Equipment  EN6060l-l-2:2001 (2nd. Edition)  Medical Electrical Equipment. Part 1: General requirements for safety - Collateral standard: Electromagnetic Compatibility Requirements and tests  Appendix 12: Cleaning and Sterilization Protocol for MIS Study REFERENCE N O .  UBC HOSPITAL  PAGE:  PICTURE  S A Y R A - MIS STUDY'  XAHSLE  8393-OOC:  SHEATH - 8393-913 INSERT - 8393-281  REFERENCE SO. UBC HOSPITAL t>A08i 1 Of 2 SAYXA - MIS STUDY DEPARTMENT:  STERILE PROCESSING  LAST REVISION. REVIEWED BY 1  DISTRI B'JTI Hi: ASSEMBLY/DSCONTAM MASTER  DISASSEMBLE!  hold  APPROVED BYJ  foroep  A 3 flhown  turn knurled  locking collar  a n d pi: I 1 o f f  sheath  holding  sheath f i r m l y ,  r«5T.ov« i n s e r t  CLEAN?  JAN 2 0 0 7  b r u s h lumen o f  from  In d i r e c t i o n of  turn the  jaw i n s e r t  arrow and  sheath  sheath  pXOC—« a l l p a r t s t h r o u g h w a s h e r  TEST t  s h e a t h m « 8 t be t e s t e d w i t h  ASSEMBLE.  insert  3»« i n s e r t  di*in£«ctor  iiuml«tieci  tester  i n t o sheath and t u r n t o  lock  h o l d h a n d l e a* uliown  •lis*  alKiith Into  turn sheath u n t i l  handle it  clicks  into  place  s c r e w o n knurled c o l l a r test  the  {unction  knurled c o l l a r one  of lull  the turn  torcep,  then  loosen  the  i;o:»t:„.  152  REFERENCE N O . OEC HOSPITAL PASS: 2 of 2 SKtm 1  -  curved  - MIS STUDV 5%'olfe - s e e p i c t u r e  grasper  7, -  g r e y c o r d with t r a n s d u c e r  4A0314-02  1  MIS .hook.  Storz  25775 yp  I  cord  Kolf6  8106-035  i n designated  PACKAGE:  ZV  x 9'  w i t h ambetf l i d l a b e l e d  container  "S&yra"  SIX)  STERILIZE:  tOCATlQH:  blue perforated  -  Loac-ir  truck.  Ethylene Oxide (EtO) Gas Sterilization Note: Before sterilization, all instruments must be thoroughly cleaned and all organic material, blood, and cleaning solution completely removed 1. Place the instruments in the sterilization tray 2. All sterilization cycles must include a pre-conditioning cycle: Temperature: 130 ± 5° F (55 ± 5° C) Time: 30 minutes 3. Sterilization has been validated for the following parameters: Gas Mixture Temperature Relative Humidity Pressure Exposure Time EtO concentration  EtO: HCFC* 130 ± 5°F(55±5°C) 60 ± 20% 8 to 10 psi 120 minutes 600 ± 30 mg/L  * gas mixture composed of 10% ethylene oxide and 9 0 % chlorotetrafluoroethane (HCFC-124; percent by weight)  4. Aeration of the instruments can be accomplished in any aeration chamber at a temperature of 120 to 130° F (49-55° C) for 12 hours.  153  Maximum acceptable levels (per ANSI/AMMI7IS0 10993-7) of residues following EtO sterilization are as follows: ethylene oxide: ethylene chlorohydrin  < 20 mg < 12 mg  Appendix 13: Results for Sterility Testing of the Reference Clip Attn: Katie Beadon University of British Columbia 6250 Applied Science Lane Vancouver, BC V6T 1Z4 Ph: (604) 822-8785  05 July, 2007  Reference No: 139345.  These are the results of the samples received June 14.  Product Sampled: Two various samples were received in the laboratory for analysis.  Product:  Sterility  Sample #: 1 Clip  No growth after 14 days incubation  Sample #: 2 Wire  Growth  Sterility test using USP 30 Sterility Testing <71> - 1 4 days incubation.  154  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0080777/manifest

Comment

Related Items