CORE Metadata, citation and similar papers at core.ac.uk

Provided by Helsingin yliopiston digitaalinen arkisto

DISSERTATIONES SCHOLAE DOCTORALIS AD SANITATEM INVESTIGANDAM UNIVERSITATIS HELSINKIENSIS 42/2014

SARI MÖLSÄ

Long-Term Outcome in Dogs After Surgical Repair of Cranial Cruciate Ligament Disease

DEPARTMENT OF EQUINE AND SMALL ANIMAL MEDICINE FACULTY OF VETERINARY MEDICINE DOCTORAL PROGRAMME IN CLINICAL VETERINARY MEDICINE UNIVERSITY OF HELSINKI

cover.indd 1 19.11.2014 7:34:29 Department of Equine and Small Animal Medicine Faculty of Veterinary Medicine University of Helsinki Finland

LONG-TERM OUTCOME IN DOGS AFTER SURGICAL REPAIR OF CRANIAL CRUCIATE LIGAMENT DISEASE

Sari Mölsä

ACADEMIC DISSERTATION

To be presented, with the permission of the Faculty of Veterinary Medicine, University of Helsinki, for public examination in Auditorium XII, University Main Building, on the 12th of December 2014 at 12 noon.

Helsinki 2014 Supervised by Professor Outi Vapaavuori Department of Equine and Small Animal Medicine Faculty of Veterinary Medicine University of Helsinki, Finland

Associate Professor Anna Hielm-Björkman Department of Equine and Small Animal Medicine Faculty of Veterinary Medicine University of Helsinki, Finland

Reviewed by Professor Michael Conzemius Department of Veterinary Clinical Sciences College of Veterinary Medicine University of Minnesota, USA

Professor Kenneth Johnson Faculty of Veterinary Science University of Sydney, Australia

Opponent Professor Eithne Comerford Institute of Ageing and Chronic Disease School of Veterinary Science Faculty of Health and Life Sciences University of Liverpool, UK

Published in Dissertationes Scholae Doctoralis Ad Sanitatem Investigandam Universitatis Helsinkiensis

ISBN 978-951-51-0484-7 (pbk.) ISBN 978-951-51-0485-4 (PDF)

ISSN 2342-3161 (Print) ISSN 2342-317X (Online) http://ethesis.helsinki.fi

Layout: Tinde Päivärinta/PSWFolders Oy

Hansaprint Helsinki 2014 To my family Abstract

Cranial cruciate ligament (CCL) disease is one of the most common causes of lameness in dogs. Surgical treatment is recommended to stabilize the stifl e joint, alleviate pain, and delay the progression of osteoarthritis (OA). A variety of surgical techniques has been introduced and can be broken down into the more traditional intracapsular ligament replacement and extracapsular suture techniques and the newer neutralizing dynamic osteotomy techniques. Although an enormous amount of literature is available concerning this disease, surprisingly few studies have assessed surgical outcome with objective evaluation methods and comparison between groups. Previous studies have demonstrated signifi cant improvement of limb function aft er CCL surgery. However, OA has been shown to progress in most patients, and this may cause deterioration in the function of the surgically treated limb over time. Long-term follow-up studies that include objective evaluation methods designed for assessing orthopedic outcome are needed. Th is thesis aimed to evaluate the long-term surgical outcome and signs of chronic pain aft er CCL surgery. Also, outcomes between the intracapsular, extracapsular and osteotomy techniques were compared. A multimodal approach, including an owner questionnaire and assessment of chronic pain using the validated Helsinki Chronic Pain Index (HCPI), orthopedic and radiographic examinations, force plate analysis, and as a new evaluation method, a physiotherapeutic examination was included. To provide reference values, clinically healthy Rottweilers and Labrador Retrievers were examined. A long-term retrospective follow-up with a mean time interval of at least 2.7 years between the surgery and evaluation was conducted. Of the 206 surgically treated dogs that were evaluated by their owners, 31.1% had a HCPI value ≥ 12, indicating pain. Of the 47 dogs evaluated by a veterinarian 31.9% showed a pain response to fl exion/extension of the surgically treated joint. In addition to the unilaterally treated CCL rupture, contralateral stifl e joint pathology and other orthopedic problems were frequently diagnosed during the evaluation. When symmetry of weight bearing was evaluated in dogs with no other orthopedic fi ndings (n=21), approximately 30% of dogs had decreased dynamic and static weight bearing in the surgically treated limbs. Also, goniometric angles of the surgically treated limbs remained inferior to those of healthy limbs, and impairment of active range of motion was frequently observed. Th e owner assessments revealed no signifi cant diff erences between the surgical techniques in long-term outcome. However, diff erences were seen in clinical evaluation of dynamic and static weight bearing between intracapsular and osteotomy technique groups. Although the retrospective study design and low sample size have to be acknowledged and may cause bias to the results, osteotomy techniques may off er long-term limb function that is superior to that achieved with the intracapsular technique. Th e low number of dogs treated with the extracapsular technique did not allow comparison of dynamic or static weight bearing to other technique groups. List of original publications

Th is thesis is based on the following publications:

I Mölsä SH, Hielm-Björkman AK, Laitinen-Vapaavuori OM. Force platform analysis in clinically healthy Rottweilers: comparison with Labrador Retrievers. Vet Surg 2010;39:701- 709.

II Mölsä SH, Hielm-Björkman AK, Laitinen-Vapaavuori OM. Use of an owner questionnaire to evaluate long-term surgical outcome and chronic pain aft er cranial cruciate ligament repair in dogs: 253 cases (2004-2006). J Am Vet Med Assoc 2013;243:689-695.

III Hyytiäinen HK, Mölsä SH, Junnila JT, Laitinen-Vapaavuori OM, Hielm-Björkman AK. Use of bathroom scales in measuring asymmetry of hindlimb static weight bearing in dogs with osteoarthritis. Vet Comp Orthop Traumatol 2012;25:390-396.

IV Mölsä SH, Hyytiäinen HK, Hielm-Björkman AK, Laitinen-Vapaavuori OM. Long-term functional outcome aft er surgical repair of cranial cruciate ligament disease in dogs (accepted for publication in BMC Vet Res 30.10.2014).

Th ese publications are referred to in the text by their Roman numerals.

Th e original publications have been reprinted with the kind permission of their copyright holders. In addition, some unpublished material has been presented. Acknowledgments

Th is study was carried out at the Department of Equine and Small Animal Medicine, Faculty of Veterinary Medicine, University of Helsinki, Finland. Th e patient material was gathered in cooperation with the Veterinary Teaching Hospital of the University of Helsinki and the private orthopedic referral clinics of Espoon eläinsairaala, Finnin Ratsutila ja Eläinlääkintä, HauMau, Malmin eläinklinikka Apex, and Mevet.

Many thanks to the former Aniwel Graduate School and to the new Doctoral Programme in Clinical Veterinary Medicine, for funding and supporting this work. Also many thanks to the Department of Equine and Small Animal Medicine and to the Veterinary Teaching Hospital for providing the facilities that made this thesis possible.

I am sincerely grateful to my supervisors, Professor Outi Vapaavuori and Associate Professor Anna Hielm-Björkman, for believing in me. Your support was outstanding and your constructive criticism and generous help in editing and fi nishing my work were invaluable.

Special thanks to Heli Hyytiäinen, my soulmate and coauthor. Without you, this study would not have been possible. I’ll never forget our “anatomic measurement” journeys across Finland. Also, thank you for tolerating my constant phone calls regarding any and all subjects between heaven and earth.

I thank former students, nowadays veterinary licenciates, Tiina Koivunen, Marjo Ruusunen, Tiina Seikkula, and Ari Suhonen for their enormous help in this work. I also thank Pia Björkenheim for her assistance in clinical evaluation of the dogs.

Many thanks to orthopedic surgeons Per Axelson, Esa Eskelinen, Pauli Keränen, Heikki Putro, Jan Räihä, and Kai Skutnabb for providing me with the patient material for this study. And, of course, special thanks to all dogs and their owners for participating in the study. It has been a pleasure to work with all of you.

Helena Korpelainen, Heli Rita, and Jouni Junnila are specially acknowledged for their assistance with statistical analyses. Carol Ann Pelli is warmly thanked for editing the language of this thesis and one of the original articles. I also thank Tinde Päivärinta for the layout of this thesis.

I am grateful to Professor Michael Conzemius and Professor Kenneth Johnson for reviewing the manuscript of this thesis and for their valuable comments, and to Professor Eithne Comerford for accepting the position of opponent at the public examination.

I thank all of my friends and colleagues at the Faculty of Veterinary Medicine and the Veterinary Teaching Hospital for their support.

Finally, my heartfelt gratitude is owed to my family. Th ank you Timo for staying at my side and supporting me during these trying years. Special thanks to my amazing children, Saana, Kalle, and Otso, for being there and reminding me of what is most important in life. And, of course, my deepest gratitude goes to my mother, Leena, for always being there for me and believing in me. Contents

Abstract List of original publications Acknowledgments Contents Abbreviations 1 Introduction ...... 1 2 Review of the literature ...... 2 2.1 Anatomy and function of the stifl e joint ...... 2 2.1.1 Cranial and caudal cruciate ligaments ...... 2 2.1.2 Menisci ...... 3 2.2 Pathogenesis of cranial cruciate ligament disease ...... 3 2.3 Diagnostics of cranial cruciate ligament disease ...... 5 2.3.1 Orthopedic examination ...... 5 2.3.2 Diagnostic imaging ...... 5 2.3.3 Arthroscopy ...... 5 2.4 Conservative treatment of cranial cruciate ligament disease ...... 5 2.5 Surgical treatment of cranial cruciate ligament disease ...... 6 2.5.1 Intracapsular reconstruction techniques ...... 6 2.5.2 Extracapsular stabilization ...... 7 2.5.3 Osteotomy techniques ...... 9 2.5.4 Complications aft er surgical treatment ...... 12 2.5.5 Treatment of menisci ...... 13 2.6 Outcome assessment aft er cranial cruciate ligament surgery ...... 14 2.6.1 Owner evaluation ...... 14 2.6.2 Orthopedic examination ...... 15 2.6.3 Kinetic and kinematic evaluation ...... 16 2.6.4 Diagnostic imaging ...... 18 3 Aims of the thesis ...... 20 4 Materials and methods ...... 21 4.1 Animals...... 21 4.1.1 Control dogs (I, III, IV)...... 21 4.1.2 Surgically treated dogs (II, IV) ...... 21 4.2 Medical records (II, IV) ...... 21 4.3 Owner evaluation and Helsinki Chronic Pain Index (II) ...... 22 4.4 Clinical evaluation (I,III,IV) ...... 22 4.4.1 Orthopedic examination (I,IV) ...... 22 4.4.2 Radiographic examination (IV) ...... 23 4.4.3 Anatomic measurements (I) ...... 23 4.4.4 Force plate analysis (I,IV) ...... 23 4.4.5 Physiotherapeutic examination (III, IV) ...... 24 4.5 Statistical analyses ...... 26 4.5.1 Study I ...... 26 4.5.2 Study II ...... 26 4.5.3 Study III ...... 27 4.5.4 Study IV ...... 27 5 Results ...... 28 5.1 Animals...... 28 5.1.1 Control dogs (I, III, IV)...... 28 5.1.2 Surgically treated dogs (II, IV) ...... 28 5.2 Owner evaluation and Helsinki Chronic Pain Index (II) ...... 29 5.3 Clinical evaluation (I,III,IV) ...... 33 5.3.1 Orthopedic and radiographic examination (I, IV) ...... 33 5.3.2 Anatomic measurements (I) ...... 34 5.3.3 Force plate analysis (I,IV) ...... 34 5.3.4 Physiotherapeutic examination (III, IV) ...... 38 6 Discussion ...... 40 6.1 Challenges of outcome measures ...... 40 6.2 Chronic pain in surgically treated dogs ...... 41 6.3 Long-term limb function aft er cranial cruciate ligament surgery ...... 42 6.4 Comparison of surgical techniques ...... 43 6.5 Strengths and limitations of the study ...... 44 7 Conclusions ...... 46 References ...... 47 Abbreviations

ANCOVA analysis of covariance ANOVA analysis of variance AROM active range of motion ASI asymmetry index AvFS average falling slope AvRS average rising slope Break imp breaking impulse Break peak breaking peak force Break time breaking time BrOAD Bristol Osteoarthritis in Dogs CaCL caudal cruciate ligament CBPI Canine Brief Pain Inventory CCL cranial cruciate ligament COI Canine Orthopedic Index CPL contralateral pelvic limb CTL contralateral thoracic limb CTWO cranial tibial wedge osteotomy CVWO chevron wedge osteotomy Degr degree Diff SWB% diff erence in quantitatively measured SWB% between hind limbs DPVF distribution percentage per limb of peak vertical force Ext extracapsular technique Funct functional limb length g gravitational acceleration h functional limb length HCPI Helsinki Chronic Pain Index Int intracapsular technique ITL ipsilateral thoracic limb L left MDS multifactorial descriptive scale meSWB manually evaluated static weight bearing MRI magnetic resonance imaging NA not analyzed NE not evaluated NRS numerical rating scale NSAID nonsteroidal anti-infl ammatory drug OA osteoarthritis Ost osteotomy technique PL pelvic limb PROM passive range of motion Prop imp propelling impulse Prop peak propelling peak force Prop time propelling time PTIO proximal tibial intra-articular osteotomy PVF peak vertical force PVF time time to peak vertical force qmSWB quantitatively evaluated static weight bearing R right ROM range of motion SD standard deviation SDS simple descriptive scale SI symmetry index STPL surgically treated pelvic limb SWB static weight bearing SWB% quantitatively measured static weight bearing as percentage of body weight TC thigh circumference TL thoracic limb TPLO tibial plateau leveling osteotomy TTA tibial tuberosity advancement TTO triple tibial osteotomy v velocity VAS visual analog scale VI vertical impulse Vrel relative velocity

Introduction 1 IntroducƟ on

Cranial cruciate ligament (CCL) rupture is the leading cause of hind limb lameness and one of the most common orthopedic conditions in stifl e joints of dogs (Whitehair et al. 1993, Johnson et al. 1994, Duval et al. 1999). It results in instability of the stifl e joint and leads to development of progressive osteoarthritis (OA) (Elkins et al. 1991, Innes and Barr 1998a). Th e cause for the disease remains unclear and is multifactorial. Th e disease is oft en bilateral, and many patients suff er from concurrent meniscal tears (Bennett and May 1991, Doverspike et al. 1993). Surgical treatment is currently recommended to stabilize the stifl e joint, alleviate the pain, and delay the progression of OA. A variety of techniques has been developed for repair of CCL rupture, and new technique modifi cations are constantly emerging. Surgical techniques can be broadly classifi ed into the traditional intracapsular ligament replacement and extracapsular suture techniques and the newer neutralizing dynamic osteotomy techniques (Piermattei et al. 2006). Th e large number of techniques indicates that no single technique is universally accepted, and the surgical technique is oft en chosen based on the surgeon’s personal preference and clinical experience. Th e use of intracapsular autograft techniques has strongly decreased in CCL repair during the last decade as the newer osteotomy techniques have gained popularity (Leighton 1999, Comerford et al. 2013, Duerr et al. 2014). However, concurrently, the surgical procedures have become technically more demanding and invasive, with increasing potential for surgical complications and a rising cost of procedures (Pacchiana et al. 2003, Hoff mann et al. 2006, Stauff er et al. 2006). Th e evaluation of surgical outcome of veterinary orthopedic patients is challenging. Surgical outcome aft er CCL repair has been examined by using owner assessment, veterinary evaluation, radiography, and force plate analysis, and a vast number of studies have been published. Th e prevailing general opinion is that dogs recover more quickly and reach better surgical outcome when treated with osteotomy than with traditional techniques. However, clinical studies have failed to consistently prove the superiority of any single surgical technique over another. Recently, an eff ort has been made to design and conduct prospective studies with objective outcome assessment and comparative aspects (Conzemius et al. 2005, Au et al. 2010, Böddeker et al. 2012, Gordon-Evans et al. 2013, Nelson et al. 2013). However, most reports evaluating surgical outcome consist of a description of a single surgical technique, postoperative complications, and a subjective evaluation of surgical outcome without any comparison to healthy animals or animals treated with other techniques. In the majority of dogs, limb function has been shown to improve signifi cantly aft er CCL surgery. However, regardless of the surgical technique chosen, OA of the stifl e joint oft en progresses (Elkins et al. 1991, Hoff man et al. 2006, Au et al. 2010). Long-term follow-up (>1 year) studies, including objective outcome evaluation methods, are needed to determine the impact of advancing OA on a dog’s clinical condition over time. Th e primary focus of this thesis was to evaluate the overall long-term surgical outcome and signs of chronic pain aft er CCL surgery. Th e secondary aim was to compare outcomes between the intracapsular, extracapsular, and osteotomy techniques. Th e assessment included force plate analysis and the validated Helsinki Chronic Pain Index (HCPI) as objective outcome evaluation methods and, as a new aspect, a physiotherapeutic evaluation conducted by a professional veterinary physiotherapist. Reference values were obtained from healthy Rottweilers and Labradors.

1 Review of the literature 2 Review of the literature

2.1 Anatomy and funcƟ on of the sƟ fl e joint Th e stifl e joint is a complex condylar synovial joint (Figure 1). Th e primary types of motion in the stifl e joint are fl exion-extension and rotation. Th e roller-like convex condyles of the femur articulate with the fl attened condyles of the tibia and form the primary weight-bearing femorotibial part of the joint. At the cranial surface of the femur, the wide, smooth, grooved femoral trochlea articulates with the patella, forming the femoropatellar part of the joint. Th e role of the femoropatellar joint is to improve the effi ciency of the extensor mechanism by increasing the moment arm of the quadriceps muscles. Th e proximal tibiofi bular joint is the third component of the stifl e joint. In addition to the patella, three smaller sesamoid bones are situated in the stifl e region. Th e lateral and medial fabellae are located in the lateral and medial heads of the origin of the gastrocnemius muscle and the popliteal sesamoid bone within the tendon of origin of the popliteus muscle (Evans 1993, Carpenter and Cooper 2000).

2.1.1 Cranial and caudal cruciate ligaments Th e primary ligamentous support to the stifl e joint is provided by four femorotibial ligaments: two collateral ligaments and two cruciate ligaments. Th e cruciate ligaments are intra-articular but covered by synovium, and thus, extrasynovial (Evans 1993). Th ey are composed of a core region that contains collagen fi brils and fi broblasts and an outer epiligamentous region of synovial intima (Hayashi et al. 2003). Th e cruciate ligaments are named based on their tibial attachments. Th e cranial cruciate ligament (CCL) originates from the caudomedial aspect of the lateral femoral condyle and the caudolateral part of the intercondyloid fossa of the femur and runs diagonally in a craniomedial direction across the intercondyloid fossa, spiraling outward axially to attach to the cranial intercondyloid area of the tibia. Th e CCL is further divided into a larger caudolateral and a smaller craniomedial part. Th e caudal cruciate ligament (CaCL) originates from the lateral surface of the medial femoral condyle and runs caudodistally, spiraling inward to attach to the medial edge of the popliteal notch of the tibia (Heff ron and Campbell 1978, Evans 1993). Th e main function of the CCL is to limit the cranial tibial translation with respect to the femur; the CaCL is the primary restraint against the caudal tibial translation. Th e CCL and CaCL prevent hyperextension and limit internal rotation of the stifl e joint. Together with collateral ligaments, they participate in limiting the varus and valgus angulations of the stifl e joint (Arnoczky and Marshall 1977, Monahan et al. 1984). Also, by modulating major muscle groups involving the stifl e, a complex system of mechano- and proprioreceptors and refl ex arcs takes part in limiting excessive joint motion (Arcand et al. 2000, Solomonow et al. 1987).

2 Review of the literature

Figure 1 Anatomy of the stifl e joint (Evans 1993).

2.1.2 Menisci Disks of fi brocartilage, the medial and lateral menisci, are located between the femoral and tibial condyles. Menisci are crescent-shaped bands of cartilage that are triangular in cross- section. Th ey adapt to femoral and tibial articulating surfaces, thus improving joint congruity (Evans 1993). Menisci contribute to load bearing, load transmission, and energy absorption in the joint, decrease friction between the joint tissues, and provide rotational and varus-valgus stability to the joint (Walker and Erkman 1975, Krause et al. 1976, Levy et al. 1982, Pozzi et al. 2008). Ligaments and soft tissue attachments to the tibia and femur hold the menisci in place. Th e medial meniscus is more fi rmly attached to the tibia than the lateral meniscus (Evans 1993, Carpenter and Cooper 2000).

2.2 Pathogenesis of cranial cruciate ligament disease CCL disease causes instability of the stifl e joint due to cranial subluxation of the tibia in relation to the distal femur. Th e disease process is associated with synovitis, articular cartilage degeneration, periarticular osteophyte development, and capsular fi brosis, resulting in stifl e OA (Marshall and Olsson 1971, Elkins et al. 1991, Vasseur and Berry 1992, Innes and Barr 1998a, Innes et al. 2000). Th e menisci, especially the less mobile caudal horn of the medial meniscus may become entrapped under the sliding motion of the femoral condyle, resulting in secondary meniscal tears (Bennett and May 1991). Th e etiology of CCL disease is multifactorial and complex, and despite decades of research, the pathological mechanisms behind the disease remain poorly understood. In a vast majority of patients, CCL rupture is caused by a chronic degenerative process that leads to partial or complete tearing of the ligament, usually during the dog’s normal daily activities. In 40-50% of the dogs, the contralateral CCL ruptures within 5-17 months of the fi rst one (Doverspike et al. 1993, Buote et al. 2009). Th e patient’s abnormal conformation or gait, age-

3 Review of the literature related changes, obesity or lack of fi tness, and immune-mediated as well as genetic factors have been implicated as potential causes for ligament degeneration. CCL rupture can occur in a wide variety of dog breeds. Th e most commonly aff ected breeds are Rottweilers, Labrador Retrievers, Newfoundland and Saint Bernhard Dogs, Mastiff s, and Staff ordshire Bull Terriers (Whitehair et al. 1993, Duval et al. 1999), and therefore, a genetic background for the disease has been suspected. Heritability of CCL rupture was studied in Newfoundland Dogs, and a recessive mode of inheritance with 51% penetrance was proposed (Wilke et al. 2006). Th e strength of the CCL has been shown to deteriorate with age (Vasseur et al. 1985). Th ese changes and also clinical signs of CCL rupture develop earlier in breeds of larger size, whereas smaller dogs tend to be aff ected later in life (Vasseur 1984, Vasseur et al. 1985, Bennett et al. 1988, Whitehair et al. 1993). Neutered dogs, whether male or female, have a higher prevalence of CCL rupture than sexually intact dogs (Whitehair et al. 1993, Duval et al. 1999). Th e association of CCL rupture with conformational factors of pelvic limbs and stifl e joints has been evaluated in many studies. Bowlegged (genu varum) or “straight” stance conformations and medial patellar luxations have been traditionally regarded as risk factors for CCL rupture (Johnson and Johnson 1993, Moore and Read 1996, Gibbons et al. 2006). Th e narrowed conformation of the intercondylar notch has been diagnosed in dogs with CCL disease and also in Retrievers predisposed to CCL defi ciency, potentially resulting in impingement and wear of the CCL (Vasseur and Berry 1992, Aiken et al. 1995, Comerford et al. 2006, Lewis et al. 2008). High tibial plateau angle was proposed as a risk factor in a study by Morris and Lipowitz (2001), but other studies have not been able to confi rm this fi nding (Wilke et al. 2002, Reif and Probst 2003). Mean tibial plateau angles in clinically normal dogs have been reported to range from 18° to 24° (Morris and Lipowitz 2001, Wilke et al. 2002, Reif and Probst 2003). Th e role of exercise in the pathogenesis of CCL degeneration remains unclear. Obesity, immobilization, and poor physical condition have been suspected to predispose to CCL rupture by increasing the load to the ligaments and decreasing the condition of the thigh muscles involved in protecting the CCL from excessive strain (Solomonow et al. 1987, Johnson and Johnson 1993, Moore and Read 1996). On the other hand, overuse and fatigue changes due to repetitive overloading have been proposed to be associated with ligament rupture (Hayashi et al. 2003). Recently, also abnormal neuromuscular control of hamstring muscles was suggested to play a causal role in the progression of CCL disease (Hayes et al. 2013). Finally, the role of immunologic factors in CCL disease has been evaluated in many studies. However, controversy remains whether the infl ammatory changes are primary contributors or secondary consequences to CCL disease (Doom et al. 2008, Comerford et al. 2011). In a minority of patients, CCL rupture is purely traumatic and results from overloading the ligament, typically when the animal falls or jumps from high or when the limb becomes caught in a hole or fence. In adult dogs, this injury usually results in a midsubstance “mop end” tear of the CCL. In skeletally immature dogs, the attachment of ligament to bone may be materially stronger than the bone itself, resulting in an avulsion fracture of the ligament from its femoral or, more commonly, tibial attachment site (Moore and Read 1996, Piermattei et al. 2006).

4 Review of the literature 2.3 DiagnosƟ cs of cranial cruciate ligament disease 2.3.1 Orthopedic examinaƟ on CCL disease causes hind limb lameness, which may vary from non-weight-bearing lameness in acute, complete tears to mild exercise-induced lameness in more chronic or partial tears. Th e typical palpatory changes are joint eff usion, pain response to fl exion and extension of the joint, and, in more chronic cases, crepitation and periarticular fi brous tissue development or “medial buttress” in the medial side of the joint. Th e pain and disuse of the limb causes atrophy of the thigh muscles. Th e possible concurrent meniscal tear may induce a clicking sound when the joint is moved through its range of motion. Cranial drawer test and tibial compression test are pathognomonic clinical tests for complete CCL rupture. Th ey both induce abnormal cranial tibial subluxation in relation to the distal femur. In partial CCL tears, however, the abnormal movement may be more subtle or absent, depending on which part or parts of the CCL are torn (Piermattei et al. 2006, Kowaleski et al. 2012).

2.3.2 DiagnosƟ c imaging Even though the CCL rupture itself is usually not visible in stifl e radiographs, radiographic examination of the stifl e joints is recommended for all patients to verify the pathology of the joint and to rule out other concurrent diseases. Typical radiographic fi ndings comprise osteoarthritic changes in the joint. Depending on the chronicity of the disease, joint eff usion with typical loss of infrapatellar fat pad shadow and osteophyte and enthesiophyte formation within the cranial intercondyloid area of the tibia, the femoral trochlear ridges, the tibial condyles, the base and apex of the patella, and the intercondylar notch of the femur accompanied by subchondral sclerosis are seen (Marshall and Olsson 1971, Widmer et al. 1994). In addition to radiography, magnetic resonance imaging (MRI), ultrasonographic examination, and thermal imaging have been evaluated and found useful in diagnosing CCL ruptures and meniscal tears (Marino and Loughlin 2010). However, these imaging modalities usually require specialized instrumentation, advanced training, and practice to be successful.

2.3.3 Arthroscopy Arthroscopy is considered the golden standard of joint evaluation and allows a thorough inspection of the synovium, articular cartilage, cruciate ligaments, and menisci with magnifi cation. Arthroscopic examination can be used to diagnose early lesions of cruciate ligaments, to confi rm pathology in uncertain cases, to evaluate the menisci for secondary meniscal lesions, and to facilitate concurrent treatment of the fi ndings (Siemering 1978, Whitney 2003).

2.4 ConservaƟ ve treatment of cranial cruciate ligament disease Conservative nonsurgical treatment of CCL rupture includes weight loss, medication such as nonsteroidal anti-infl ammatory drugs (NSAIDs) or nutraceuticals, physiotherapy, and controlled exercise. Th e outcome of conservative treatment has not been extensively investigated. In two retrospective studies, nonsurgical management was reported to result in satisfactory function in the majority of small dogs (<15-20 kg), whereas lameness persisted in larger dogs, necessitating surgical treatment for successful outcome (Pond and Campbell 1972, Vasseur 1984). Recently, a

5 Review of the literature prospective study was conducted to evaluate the outcome of surgical and nonsurgical treatment for overweight dogs with CCL rupture. Although the dogs treated surgically in addition to conservative management had better outcomes than the dogs treated with nonsurgical methods alone, almost two-thirds of the dogs in the nonsurgical treatment group had a successful outcome one year aft er initiation of treatment (Wucherer et al. 2013).

2.5 Surgical treatment of cranial cruciate ligament disease Surgical treatment is advocated to stabilize the stifl e joint, alleviate pain, treat any concurrent meniscal pathology, and decelerate the development of OA. Regardless of the surgical technique chosen, the stifl e joint is usually evaluated via arthroscopy or arthrotomy to diagnose and treat possible concurrent meniscal pathology. In addition, most surgeons debride the remnants of the torn cruciate ligament with the intention of decreasing the infl ammatory process maintained by a remaining cruciate ligament stump. Primary repair of the torn CCL is seldom possible. However, avulsion fractures of the CCL attachment sites in young animals can be treated with internal fi xation (Piermattei et al. 2006, Kowaleski et al. 2012).

2.5.1 Intracapsular reconstrucƟ on techniques Intracapsular reconstruction techniques comprise the oldest group of CCL repair methods, and these techniques also have a long tradition in Finland. Intracapsular techniques are considered the standard for CCL repair in humans (Hospodar and Miller 2009). In intracapsular reconstruction, the torn CCL is replaced with autogenous tissue, other biological material such as allograft s, xenograft s, synthetic material, or a combination of synthetic and biological materials by mimicking the normal anatomy and attachment points of the CCL. In dogs, the most widely used replacement material has been fascia lata and/or patellar ligament (Figure 2). One of the fi rst intracapsular techniques was developed by the Finn Saki Paatsama in the 1950s. In the “Paatsama” technique, the CCL is replaced with a strip of fascia lata, which is left attached distally. Th e graft is brought through the bone canals that are drilled in the tibia and femur at the anatomic insertion and origin of the CCL, tightened, and attached with sutures along the patellar ligament (Paatsama 1952). In 1979, Arnoczky et al. described the “over-the-top” technique. In this technique, the replacement for CCL is harvested from the medial third of the patellar ligament, part of the patella and fascia lata. Th e graft is brought through the joint, over the top of the lateral condyle, and sutured. Signifi cant injury to the patella may occur when harvesting the graft , but with this technique the risk for unaccurate placement of drill canals and fraying of the graft to bony edges of the drill canals can be avoided (Arnoczky et al. 1979). Several variations and modifi cations to these original techniques have since been introduced (Hulse et al. 1980, Hulse et al. 1983, Denny and Barr 1984, Shires et al. 1984, Denny and Barr 1987, Piermattei et al. 2006). Th e goal of intracapsular reconstruction is that aft er phases of initial graft infl ammation and necrosis the graft will revascularize, remodel, and begin to act as native CCL; achievement of this goal has been shown experimentally (Arnoczky et al. 1982). Th e process may, however, be slow or incomplete, leading to elongation and failure of the graft , with recurrent instability of the joint (Butler et al. 1983, Hulse et al. 1983). Th e continuing degenerative process related to the naturally occurring CCL disease also may aff ect the graft ed tissue in a manner similar to native CCL. Inferior strength properties of the autograft relative to native CCL can also decrease the

6 Review of the literature chances of graft survival under load. Unlike in humans, none of the tissues around the stifl e joint in dogs approaches the strength of the native CCL (Johnson et al. 1989).

Figure 2 Intracapsular reconstruction using fascia lata and patellar tendon

Replacement of the CCL with allograft s or xenograft s has been evaluated in experimental studies, but these materials have not yet been used in clinical canine patients (Curtis et al. 1985, Arnoczky et al. 1986, McMaster 1994). Such disadvantages as tissue availability, potential disease transmission, negative impact of graft sterilization or radiation, and immune reactions have limited the development of allograft or xenograft tehniques (Kowaleski et al 2012). In studies evaluating synthetic materials, permanent prosthetic graft s have mostly failed due to mechanical failure in both humans and dogs (Gupta and Brinker 1969, Rubin et al. 1975, Denny and Goodship 1980, Legnani et al. 2010). In addition, the use of autograft s combined with synthetic material as an augmentation device and materials that have the ability to allow and promote tissue ingrowth have been evaluated in both humans and dogs (Park et al. 1985, Laitinen 1994, Legnani et al. 1010). Although some studies have shown promising early results, many techniques have been unsuccessful due to eventual failure of the graft and long-term results are needed. Current research appears to be focusing on tissue engineering and development of scaff olds that support tissue ingrowth for neoligamentization but gradually resorb to allow progressive loading of the new ligament.

2.5.2 Extracapsular stabilizaƟ on Extracapsular stabilization covers a variety of stabilization methods in which the craniocaudal instability is eliminated using extra-articular prostheses, usually sutures. Th ese include imbrication techniques, fabellotibial suture techniques, the TightRope CCL technique, and

7 Review of the literature fi bular head transposition. Th ese techniques rely on development of periarticular fi brous tissue to provide long-term stability because the primary implants tend to fail over time. Th e fi rst extracapsular stabilization procedures in the 1960s were done by tightening of the stifl e joint with Lembert sutures placed on the lateral retinacular fascia of the joint. DeAngelis and Lau (1970) modifi ed the technique by providing anchorage points to the suture: proximally, the dense connective tissue caudoproximal to the lateral fabella, and distally, the patellar tendon. Further modifi cations of the technique consisting of changes in anchorage points and number of sutures were described by Flo (1975), Gambardella et al. (1981), and Brinker et al. (1990). Smith and Torg presented the fi bular head transposition technique in 1985. In this technique, the fi bula was advanced cranially, thus altering the orientation of the lateral collateral ligament and eliminating the cranial drawer. Th e most commonly used extracapsular stabilization method today is the “fabellotibial suture” technique, again a modifi cation of the technique by DeAngelis and Lau (1970) (Figure 3). In this technique, the craniocaudal instability is eliminated by using heavy nonabsorbable suture material, usually nylon leader line, nylon fi shing line, or braided polyester sutures inserted and tensioned between the tibial tuberosity and the lateral fabella. To prevent slipping of the suture, it is placed in the fi brous origin of the lateral head of the gastrocnemius, slightly proximal to the fabella. Th e suture is passed through the hole predrilled in the tibial crest and under the patellar ligament from the medial to the lateral side, to complete the mattress suture. Alternatively, two holes can be made in the tibial tuberosity. Th e mattress suture is tensioned to neutralize the cranial drawer and fastened by knots or a tensioning device and a metallic crimp tube. In order to tighten the soft tissues and induce periarticular fi brosis, imbricating suture patterns are

Figure 3 Extracapsular stabilization using a fabellotibial suture (left ) or a suture and bone anchor (right).

8 Review of the literature used to close the fascia and the joint capsule in cases where the joint is inspected via arthrotomy (Kowaleski et al. 2012). Minimally invasive modifi cation of the fabellotibial suture technique has been described (Hoelzler et al. 2004, Biskup and Griff on 2014). Recent attention has been focused on isometry of the tibial and femoral suture anchorage points (i.e. maintaining a consistent distance between the points during the stifl e range of motion). Th e anchorage points used in the traditional fabellotibial suture technique were reported to be nonisometric, and this was shown to cause tightening and loosening of the suture during the stifl e range of motion (Hulse et al. 2010). Although determination of a truly isometric combination of anchorage points may not be possible, near-isometric locations for the anchorage have been proposed (Roe et al. 2008, Hulse et al. 2010), and technique modifi cations aimed at a more isometric suture insertion by using suture anchors or bone tunnels have been reported (Guénégo et al. 2007, Kunkel et al. 2009) (Figure 3). Moreover, the latest modifi cation of extracapsular techniques, the TightRope CCL technique, achieves stabilization by using a fl at, braided polyethylene polyester tape that is anchored to near-isometric locations via femoral and tibial bone tunnels and suture buttons (Cook et al. 2010a). Th e proposed advantages of the fabellotibial suture technique include technical ease of the procedure, minor surgical trauma, no requirement for specialized instrumentation, a shorter surgical time than in many other techniques, and thus, also a reasonable surgical cost. Inability to anchor or secure the suture loop adequately may predispose to slipping of the suture from the fabella, elongation or failure of the suture too early, resulting in recurrent instability (Tonks et al. 2011). With bone suture anchors or bone tunnels, a near-isometric anchorage is achieved and problems related to circumfabellar suture placement are avoided, although pullout of the anchors from the femoral bone has been reported (Guénégo et al. 2007). In addition to these, a minimally invasive approach is used in the TightRope technique (Cook et al. 2010a)

2.5.3 Osteotomy techniques Osteotomy techniques are based on an approach that is very diff erent from the traditional intracapsular and extracapsular stabilization methods. In osteotomy techniques, the aim of the surgical procedure is to achieve functional stability of the stifl e joint during weight bearing. Unlike in the traditional surgical techniques, the passive function of the CCL is not restored. Th e biomechanical rationale for the osteotomy techniques was introduced and modifi ed by Slocum in 1983 and 1993, respectively, and later criticized by Tepic in 2002. Slocum described the total tibiofemoral joint force of weight bearing to be oriented parallel to the axis of the tibia and to be divided into components of a cranial tibial thrust force oriented parallel to the tibial plateau and a joint compressive force oriented perpendicular to the tibial plateau. Due to the caudodistally directed slope of the tibial plateau, limb loading results in a cranial tibial shear force and the magnitude of this shear force is dependent on the steepness of the tibial plateau slope. Based on Slocum’s “Active Model of the Stifl e”, the joint stability in a normal joint is achieved by synergism between the stifl e joint fl exors and extensors. Th e cranial tibial thrust force is balanced by the pull of the stifl e fl exor muscles of the thigh and the passive restraints of the stifl e joint including the CCL and the caudal horn of the medial meniscus. However, in the CCL-defi cient joint the cranial tibial thrust force causes cranial tibial subluxation during limb loading. By reducing the tibial plateau angle surgically, this cranially oriented shear force can be eliminated (Slocum and Devine 1983, Slocum and Devine Slocum 1993).

9 Review of the literature

Based on Slocum’s fi ndings, the fi rst osteotomy technique, cranial tibial closing wedge osteotomy (CTWO), was described in 1984 (Slocum and Devine 1984). In the CTWO technique, the tibial plateau angle is reduced by using a cranially based wedge ostectomy of the proximal tibia. Th e second osteotomy technique, tibial plateau leveling osteotomy (TPLO) was introduced in 1993 (Slocum and Devine Slocum 1993) (Figure 4). In the TPLO, the slope of the tibial plateau is reduced by a crescent-shaped osteotomy of the proximal tibia using a biradial saw blade. Subsequent rotation of the proximal segment aims at achieving a postoperative tibial plateau angle of 5°. Th e amount of rotation is calculated based on the preoperative tibial plateau angle and the radius of the osteotomy. Internal fi xation with a TPLO plate and conventional or locking screws is used to support the new position of the proximal bone segment (Slocum and Devine Slocum 1993, Kowaleski et al. 2012).

Figure 4 Tibial plateau leveling osteotomy.

Th e TPLO alters the biomechanics of the tibiofemoral joint, but maintains the original position of the femoropatellar joint and the tibial tuberosity (Slocum and Devine Slocum 1993). Th e TPLO has been shown to neutralize the cranial tibial subluxation in CCL-defi cient stifl es in vitro (Warzee et al. 2001, Reif et al. 2002, Kim et al. 2009a), but the results were more inconsistent when the eff ect of TPLO on femorotibial subluxation was evaluated in vivo during standing (Kim et al. 2012). Secondary to altered biomechanics, the TPLO may cause increased strain to the extensor mechanism of the stifl e (Carey et al. 2005). Th e TPLO technique requires specialized instrumentation and is technically more demanding and a more invasive procedure than most of the other CCL repair methods, which is mirrored in the higher cost of the procedure. In 2002, Tepic et al. suggested that the compressive forces of weight bearing are not axially oriented, but are directed parallel to the patellar tendon. According to this, the cranial tibial thrust force depends on the angle between the tibial plateau and the patellar tendon. It was also suggested that if the patellar tendon is oriented perpendicular to the tibial plateau the cranial shear component of the joint force is eliminated (Tepic et al. 2002). Th is led to the development

10 Review of the literature of the tibial tuberosity advancement (TTA) technique (Montavon et al. 2002) (Figure 5). Th e aim of the TTA technique is to eliminate the cranial tibial thrust by reducing the angle between the patellar tendon and the tibial plateau to 90°. Normally, this angle is 105° when the stifl e is extended to 135°. A longitudinal osteotomy is made subjacent to the tibial tuberosity, and the tibial tuberosity is advanced cranially with a spacer-cage inserted proximally between the osteotomy. A tension-band bone plate is inserted to support the new position, and a bone graft is applied to the osteotomy space to accelerate bone union. Th e amount of advancement and the size of the cage needed are determined based on preoperative radiographic measurements (Montavon et al. 2002). An excessive tibial plateau angle may be a contraindication for TTA, with an anecdotally proposed cut-off point of 30°.

Figure 5 Tibial tuberosity advancement.

Th e TTA technique is generally considered less invasive and technically less demanding than the other tibial osteotomies. Like in TPLO, special instrumentation is needed and a number of surgical errors may occur. Th e TTA maintains the original position of the femorotibial articulation, thus having less eff ect on load transmission across the joint than TPLO (Kim et al. 2009b, Guerrero et al. 2011). However, a recent in vivo study showed that, similar to TPLO, TTA did not consistently restore femorotibial stability during standing, and two-thirds of the dogs had persistent subluxation aft er surgery (Skinner et al. 2013). TPLO and TTA are the most commonly used osteotomy techniques and their popularity has increased enormously during the last decade. In addition, several technique modifi cations developed to overcome certain limitations related to the above-mentioned techniques include a triple tibial osteotomy (TTO) (Bruce et al. 2007), a combination of TPLO and CTWO (Talaat et al. 2006), a proximal tibial intra-articular osteotomy (PTIO) (Damur et al. 2003, Jerram et al. 2005), and the chevron wedge osteotomy (CVWO) (Hildreth et al. 2006). In the TTO technique, the longitudinal partial osteotomy of the tibial tuberosity is combined with a partial wedge osteotomy of the tibia. Th e reduction of the wedge osteotomy decreases the tibial plateau

11 Review of the literature slope and simultaneously shift s the tibial tuberosity cranially (Bruce et al. 2007) (Figure 6). Th e proposed advantages of the TTO include minimal change in the orientation of the articulating surfaces of the tibiofemoral joint, no loss of limb length, and, unlike in TTA, a small osteotomy gap caudal to the tibial tuberosity. However, despite using recommended surgical calculations, unpredicted variation of the postoperative patellar tendon to the tibial plateau angle has been reported (Bruce et al. 2007). Recently, a proximal tibial epiphysiodesis was described for treatment of CCL rupture in young dogs with open growth plates either as a sole treatment or to augment primary repair of CCL avulsion fracture. Th e tibial osteotomy is not done, but a premature union of the cranial tibial epiphysis is surgically induced while the caudal physis continues to grow, resulting in a reduction of the tibial plateau angle (Vezzoni et al. 2008).

Figure 6 Triple tibial osteotomy

2.5.4 ComplicaƟ ons aŌ er surgical treatment Scant information is available on complication types and rates associated with intracapsular autograft techniques. Although many reports have been written on diff erent technique modifi cations of extracapsular repair and diff erences in properties of suture materials and methods of securing sutures, only one report concentrating on complications aft er the fabellotibial suture technique has been published. In this report, complications occurred in 17.4% of procedures and included incisional problems, wound infection, peroneal nerve damage, implant-related problems, and late meniscal tears (Casale and McCarthy 2009). A second surgery was required in 7.2% of patients. Th e postoperative infection rate aft er fabellotibial suture stabilization reported in two studies was 3.9% and 4.2% of patients, respectively (Casale and McCarthy 2009, Frey et al. 2010). Two studies have reported complications of TightRope

12 Review of the literature technique in 17.8% and 29.2% of procedures, including implant failure, seroma, infection, and late meniscal tears (Cook et al. 2010a, Cristopher et al. 2013). In 8.9% and 12.5% of the procedures, respectively, the complications were major, requiring further treatment. Numerous intra- and postoperative complications have been described in dogs undergoing osteotomy procedures. Th e highest complication rates were reported in studies documenting initial experience with the techniques, and the number of complications has tended to decrease as more knowledge has been gained. Th e complications related to TPLO surgery have been thoroughly reported. Th e overall complication rates ranging between 15% and 28% have included incisional problems (edema, discharge, seroma, dehiscence), infection, tibial tuberosity fracture, implant failure, patellar tendonitis, osteomyelitis, late meniscal tears, fi bular fracture, and patellar fracture (Pacchiana et al. 2003, Priddy et al. 2003, Kergosien et al. 2004, Stauff er et al. 2006, Fitzpatrick and Solano 2010, Gatineau et al. 2011, Christopher et al. 2013). Aft er TTA surgery, reported complication rates have varied substantially, ranging from 10% to 59%. As in TPLOs, incisional problems, infection, tibial tuberosity fracture, implant failure, and late meniscal tears have been the most common (Hoff man et al. 2006, Lafaver et al. 2007, Steinberg et al. 2011, Wolf et al. 2012). Aft er TTO surgery, complications have been reported to occur in 23-36% of patients (Bruce et al. 2007, Moles et al. 2009). Although overall complication rates are commonly reported, diff erences between reports in description and classifi cation of the complications make direct comparison of the surgical techniques diffi cult. In the study by Christopher et al. (2013), the complications were classifi ed as catastrophic, major, and minor, according to the presently recommended criteria (Cook et al. 2010b). More major complications and late meniscal tears occurred aft er TTA surgery than aft er TPLO or TightRope techniques, and also aft er TPLO surgery than aft er the TightRope technique. However, as a study limitation it must be mentioned that only 18 of the 162 dogs were treated using the TTA technique. When individual complication types are evaluated, patellar tendonitis seems to occur less oft en aft er TTA than aft er TPLO surgery, probably due to the lower strain on the extensor mechanism (Tepic 2002). Incisional problems, such as postoperative edema, discharge, bruising, and seroma formation, appear more frequently aft er osteotomy procedures than aft er other techniques, likely due to the more invasive approach. Th e reported postoperative infection rates aft er osteotomy procedures have varied between 2.9% and 8.4% (Priddy et al. 2003, Hoff mann et al. 2006, Fitzpatrick and Solano 2010, Frey et al. 2010, Gatineau et al. 2011), and infection occurs more oft en aft er TPLO than aft er extracapsular stabilization (Frey et al. 2010).

2.5.5 Treatment of menisci Meniscal tears associated with the CCL rupture occur in 33-83% of CCL-defi cient dogs (Bennet and May 1991, Flo 1993, Ralphs and Whitney 2002, Ritzo et al. 2014). Most commonly reported types are the bucket-handle tear and the folded caudal horn of the medial meniscus (Bennet and May 1991). Treatment options comprise partial meniscectomy with removal of the damaged part, caudal hemimeniscectomy with removal of the entire caudal pole, total meniscectomy with removal of the entire meniscus, and primary repair (Kowaleski et al. 2012). Although meniscal damage usually results from joint instability associated with CCL rupture, meniscal tears may also occur as a late complication aft er stabilization procedures (Case et al. 2008). Late meniscal tears have been suggested to occur especially aft er osteotomy procedures in which passive laxity of the joint is not addressed. Originally introduced by Slocum

13 Review of the literature for use with the TPLO technique, a meniscal release procedure is advocated to decrease the risk of these late tears. Th e meniscus is sharply cut either at the caudal meniscotibial ligament or at the midbody of the medial meniscus to allow caudal displacement of the caudal horn of the medial meniscus, thus preventing it from being entrapped under the femoral condyle (Slocum and Devine Slocum 1998). However, recent in vitro studies have proven that the meniscal release procedure also has negative infl uences and actually has eff ects similar to caudal medial hemimeniscectomy with regard to meniscal load transmission function (Pozzi et al. 2008, Pozzi et al. 2010). Th e eff ects of meniscal release were also evaluated in vivo in CCL-intact stifl e joints, where the meniscal release procedure alone was reported to cause lameness, articular cartilage loss, further meniscal pathology, and osteoarthritic changes (Luther et al. 2009).

2.6 Outcome assessment aŌ er cranial cruciate ligament surgery Th e assessment of surgical outcome aft er CCL repair has traditionally been based on a veterinarian’s or an owner’s opinion of lameness and surgical outcome and measurement of radiographic OA in the surgically treated limb. With this kind of simple and oft en qualitative assessment, the outcome evaluation has been very subjective. Only recently, as general awareness of the need for quantitative and more objective outcome assessment has increased, such objective outcome methods as force plate analysis or validated owner-assessment tools have become more available and routinely included also in studies reporting outcome aft er CCL surgery.

2.6.1 Owner evaluaƟ on Although owner evaluation is subjective, owners have the advantage of being able to evaluate the dogs’ function during its daily activities, in its own unstressful environment over time. Th e owner’s opinion of the surgical outcome has long been ascertained with simple questions regarding surgical outcome, dog’s quality of life, and owner’s satisfaction with the surgical procedure and willingness to have it repeated if necessary. High success rates (>90% of owners evaluating the surgical outcome as excellent or good) have been reported, regardless of the surgical technique used. Similarly, >90% of owners are usually satisfi ed with the surgical procedure and would have it performed on another dog with CCL injury. Recently, more sophisticated and reliable owner-assessment methods have been developed for evaluation of OA and chronic pain in orthopedic patients (Innes and Barr 1998b, Hudson et al. 2004, Wiseman-Orr et al. 2006, Cimino Brown et al. 2007, Hercock et al. 2009, Hielm- Björkman et al. 2009, Cimino Brown 2014a, Cimino Brown 2014b, Cimino Brown 2014c). Bristol Osteoarthritis in Dogs (BrOAD) questionnaire (Innes and Barr 1998b) and Texas A&M Client Questionnaire (Hudson et al. 2004) are validated visual analog scale (VAS)-based owner questionnaires, and Canine Brief Pain Inventory (CBPI) is a validated numeric rating scale (NRS)-based questionnaire (Cimino Brown et al. 2007). Several studies have used these questionnaires also to evaluate surgical outcome aft er CCL repair. In three studies reporting results using the BrOAD questionnaire, signifi cant improvements in the levels of disability and inactivity stiff ness, in the eff ect of cold weather, and in the ability to jump were seen aft er repair with intracapsular, TPLO, and TTO techniques (Innes et al. 2000, Corr and Brown 2007, Renwick et al. 2009). Similarly, Boyd et al. (2007) reported signifi cant improvement in the summed scores of BrOAD questions in the postoperative evaluation aft er TPLO surgery. One of these studies also compared TPLO and CTWO techniques, and, unlike the TPLO, CTWO failed to produce

14 Review of the literature signifi cant improvements in the level of inactivity stiff ness and in the eff ect of cold weather (Corr and Brown 2007). On the other hand, another two studies found no signifi cant diff erences in the outcome between the TPLO and TightRope techniques when the results on the Texas A&M questionnaire were compared (Cook et al. 2010) or between the TPLO and extracapsular techniques when the CBPI questionnaire was used (Gordon-Evans et al. 2013). Th e Helsinki Chronic Pain Index (HCPI) is a validated pain index based on 11 MDS questions concerning a dog’s mood, behavior, and locomotion, and, unlike the previous indices, it also takes into account the emotional aspect of pain (Hielm-Björkman et al. 2003, Hielm- Björkman et al. 2009, Walton et al. 2013). Th e HCPI has been used in the evaluation of chronic pain in dogs with hip, elbow, and stifl e joint OA (Hielm-Björkman et al. 2003, Hielm-Björkman et al. 2009, Wernham et al. 2011, Eskelinen et al. 2012, Hielm-Björkman et al. 2012, Heikkilä et al. 2014). In the HCPI, the owner-reported answers in the 5-point descriptive scale are tied to a value (0-4), which when summed, yields a total index minimum score of 0 and maximum score of 44. Individual scores of 0 and 1 on each question indicate normal mood, behavior, and locomotion, and scores of 2, 3, and 4 indicate pain in ascending severity. Healthy dogs usually have an HCPI between 0 and 6, whereas values between 7 and 35 have been measured in dogs with chronic pain. However, healthy dogs may have index values up to 11, assuming that an individual score of 1 is considered normal. Th us, the index values between 6 and 11 are considered a gray zone where the dog may be either pain-free or in pain (Hielm-Björkman et al. 2003). Th e newest investment in improving the quality of owner-reported outcome assessment is a multifactorial descriptive scale (MDS)-based questionnaire, the Canine Orthopedic Index (COI), which was recently validated (Cimino Brown et al. 2014a, Cimino Brown et al. 2014b, Cimino Brown et al. 2014c) but has not yet been established for evaluation of CCL patients.

2.6.2 Orthopedic examinaƟ on An orthopedic examination has traditionally been included in CCL outcome studies. Typically, it has consisted of a visual lameness evaluation using a simple descriptive scale (SDS), VAS, or varying NRS-based scores. Although oft en used, unfortunately, the visual evaluation has been found to be insensitive for detecting lameness (Evans et al. 2005, Quinn et al. 2007). Quinn et al. (2007) reported that both NRS- and VAS-based scores had low agreement among observers unless the lameness was severe. Also, if the lameness was not severe, the correlation with the results of force plate analysis was generally low. In addition to lameness evaluation, varying descriptions of the stifl e joint palpatory fi ndings and cranial drawer test have usually been included in the orthopedic examination. Diff erences in grading scales, presentation methods, and follow-up times render the comparison of orthopedic examinations among studies diffi cult. In most studies, lameness has improved postoperatively and 73-94% of dogs have shown no visually detectable lameness at follow-up (DeAngelis and Lau 1970, Arnoczky et al. 1979, Gambardella et al. 1981, Denny and Barr 1987, Talaat et al. 2006, Guénégo et al. 2007, Lafaver et al. 2007, Moeller et al. 2010). Similar to owner assessments, usually >90% of veterinarians have described the surgical outcome as successful or satisfactory, using varying criteria. Recently, also the quantitative measurements of stifl e joint range of motion (ROM) and thigh muscle symmetry have been increasingly used (Jerram et al. 2005, Bruce et al. 2007, Jandi and Schulman 2007, Au et al. 2010, Kunkel et al. 2009, Moeller et al. 2010, Gordon-Evans et al. 2013, Oxley et al. 2013). Stifl e joint ROM and thigh circumference (TC) have improved in

15 Review of the literature almost all studies aft er surgery relative to the preoperative situation (Jerram et al. 2005, Bruce et al. 2006, Kunkel et al. 2009, Au et al. 2010, Gordon-Evans et al. 2013, Oxley et al. 2013). However, several studies have discovered that ROM and/or TC have started decreasing aft er more time has passed since surgery (Jerram et al. 2005, Bruce et al. 2007, Au et al. 2010, Gordon-Evans et al. 2013). Moreover, Moeller et al. (2010) reported that stifl e joint ROM and TC remained lower than in healthy control limbs in the long term. Surgical techniques have been compared in two studies aft er TPLO and extracapsular surgery and in one study aft er repair with TPLO and CTWO, all reporting equal outcomes, with no signifi cant diff erences in goniometric or TC results between the techniques (Au et al. 2010, Gordon-Evans et al. 2013, Oxley et al. 2013). According to my knowledge, the ROM or TC measurements have not been reported aft er repair with intracapsular, TightRope, or TTA techniques. Postoperative intensive physiotherapy has been shown to be benefi cial for ROM and muscle mass (Monk et al. 2006).

2.6.3 KineƟ c and kinemaƟ c evaluaƟ on Compared with subjective visual lameness evaluation, force plate analysis allows quantitative measurement of dynamic weight bearing and can detect mild lameness or gait abnormality that may not be apparent during visual evaluation (Evans et al. 2005). In kinetic evaluation, peak vertical forces (PVFs) and vertical impulses (VIs) are the most commonly used parameters. Numerous factors can aff ect the generation of ground reaction forces during gait analysis. Of overall variance related to ground reaction force measurements, 29-85% was shown to be attributable to trial repetitions, 14-69% to dogs, and 0-7% to handlers (Jevens et al. 1993). Ground reaction forces can be measured at both trot and walk, but subject velocity has a signifi cant eff ect on force plate values and must be limited to a narrow range when data are obtained (Riggs et al. 1993). Acceleration and deceleration during the trial must be controlled (Budsberg et al. 1999). Body weight of the dog aff ects the ground reaction forces, and to allow comparison of individual dogs or dog groups with diff erent body weights, forces and impulses are usually expressed as a percentage of body weight. Recently, also the eff ect of body conformation on ground reaction forces has been considered (Bertram et al. 2000, Voss et al. 2010, Voss et al. 2011). Th e best approach for outcome evaluation with force plate analysis is to compare the postoperative ground reaction forces with the pre-injury values. Unfortunately, this is usually impossible in clinical patients and can only be used in experimental studies. To compare the postoperative values with normal reference values, force plate data obtained from either healthy control dogs or the patient’s contralateral limb may be used. In comparison of the surgically treated limb with the contralateral limb, potential load distribution from the aff ected to the contralateral limb must be taken into account (Rumph et al. 1993). Alternatively, if ground reaction forces are compared with data obtained from healthy control dogs, factors like subject velocity, acceleration, and body weight should be carefully controlled.

PostoperaƟ ve limb funcƟ on aŌ er CCL surgery Force plate analysis has been used in several reports to assess postoperative limb function aft er CCL repair (Budsberg et al. 1988, Jevens et al. 1996, Ballagas et al. 2004, Conzemius et al. 2005, Voss et al. 2008, Böddeker et al. 2012, Gordon-Evans et al. 2013, Nelson et al. 2013, Skinner et al. 2013, Wurcherer et al. 2013). All studies using force plate evaluation report improvement of dynamic weight bearing aft er surgery, but the results are more variable when return to the

16 Review of the literature preinjury level is considered. In clinical patients with naturally occurring CCL disease, four studies have reported return of dynamic weight bearing to the level of healthy limbs (Budsberg et al. 1988, Böddeker et al. 2012, Nelson et al. 2013, Skinner et al. 2013). Budsberg et al. (1988) evaluated ground reaction forces in nine dogs before and aft er extracapsular repair with the modifi ed retinacular imbrication technique and did not fi nd any signifi cant diff erences between aff ected and clinically normal pelvic limbs 7-10 months aft er surgery. More recently, a force plate treadmill was used to evaluate 14 dogs aft er repair using the TPLO technique, and by 4 months aft er surgery no signifi cant diff erences could be seen between the aff ected and unaff ected limbs (Böddeker et al. 2012). Also in a study by Nelson et al. (2013) the calculated symmetry indices (SIs) for PVF, VI, and contact time in 15 TPLO-treated dogs were not diff erent from the control group by 6 months to 1 year postoperatively. However, only six dogs in the group completed the one-year follow-up. Finally, Skinner et al. (2013) evaluated clinical outcome in 26 TTA-treated dogs and found no signifi cant diff erence between the surgically treated and contralateral limbs with a mean follow-up of 18 months. Contrary to the previous studies, in several reports the ground reaction forces in surgically treated limbs or pelvic limb SIs have remained inferior to full function. Conzemius et al. (2005) evaluated 131 dogs that underwent surgery with intracapsular modifi ed over-the-top, extracapsular fabellotibial suture, and TPLO techniques and compared the results with healthy control dogs. By 6 months, only 15% of dogs in the intracapsular, 14.9% in the extracapsular, and 10.9% in the TPLO group were judged to have normal limb function. As well, in a study evaluating postoperative limb function 4-16 months aft er TTA surgery in 37 dogs, PVF and VI remained signifi cantly lower than in control group dogs, and in 64.3% of dogs the SI for PVF indicated residual lameness (Voss et al. 2008). In the study mentioned earlier, Böddeker et al. (2012) also reported that by 4 months aft er surgery, the 14 dogs treated with the extracapsular technique with a capsular-fascial imbrication method continued to have signifi cantly lower force plate values in the aff ected limb than in the unaff ected limb. Also Nelson et al. (2013) reported that aft er extracapsular fabellotibial suture repair the SIs in 16 dogs completing a one-year follow-up were lower than in control group dogs.

Comparison of surgical techniques Few studies directly compare surgical techniques by using force plate analysis. In only one study has the intracapsular technique been compared with other repair methods (Conzemius et al. 2005). In this study, the dogs treated with the intracapsular technique had inferior limb function to the extracapsular and TPLO groups 2 and 6 months postoperatively, but no signifi cant diff erences were found in ground reaction forces between the extracapsular and TPLO groups. Also in another study with a 2-year follow-up, no signifi cant diff erences were found in PVFs between extracapsular repair and TPLO when treatment in both groups included postoperative physical rehabilitation. In this study, the PVF was measured using a pressure sensitive walkway (Au et al. 2010). By contrast, the results of three recent studies indicate that TPLO leads to limb function that is superior to that produced by extracapsular surgery. Firstly, Böddeker et al. (2012) concluded that the dogs treated with TPLO had faster recovery and improved limb function relative to the extracapsular capsular-fascial imbrication technique group 4 months aft er surgery. SIs for PVFs were 5.83% in the TPLO group and 19.05% in the extracapsular group 4 months postoperatively, although the diff erence was not statistically signifi cant. Secondly, SIs for PVF and VI at trot were signifi cantly more symmetrical for the TPLO group than for the extracapsular

17 Review of the literature group at all evaluation points (2 weeks, 8 weeks, 6 months, 12 months) postoperatively (Nelson et al. 2013). Th irdly, Gordon-Evans et al. (2013) evaluated ground reaction forces at walk and trot in a well-designed study 6 and 12 months aft er surgery in 40 dogs treated with the extracapsular fabellotibial suture technique and in 40 TPLO dogs. Aft er 12 months, PVFs in aff ected limbs were 6% higher at walk and 11% higher at trot for dogs in the TPLO group relative to those in the extracapsular group; these diff erences were statistically signifi cant. In summary, TPLO seems to result in outcomes that are better than extracapsular repair and similar to extracapsular repair in approximately equal numbers of studies. Either the true diff erences between the techniques are too small to show signifi cance in all studies or the diff erence in factors related to patients, surgical techniques, postoperative treatments, or evaluation methods cause variance in results. Th e available research is insuffi cient to allow conclusions to be drawn about the surgical outcome aft er intracapsular repair or TTA surgery in relation to other techniques.

AlternaƟ ves to basic force plate analysis Force plate analysis has the disadvantage of measuring the dynamic weight bearing of only one limb at a time. In addition, force plate analysis is unable to evaluate swing phase of the stride and successive strides cannot be measured with a single force plate. As an alternative to basic force plate analysis, pressure sensitive walkways measuring several footsteps during one trial, equipment with force plates in series, and force plate treadmills have been developed and used also for evaluation of CCL-defi cient dogs, as mentioned above (Au et al. 2010, Böddeker et al. 2012, Krotscheck et al. 2014). In addition to force plate analysis, the use of kinematic evaluation of gait in CCL disease has increased (Marsolais et al. 2003, Tashman et al. 2004, Lee et al. 2007, Sanchez-Bustinduy et al. 2010, deMedeiros et al. 2011). Kinematic analysis allows assessment of the whole stride cycle, evaluation of fl exion and extension movements of joints and temporal and distance variables of the dog’s gait (McLaughlin 2001). When kinematic analysis was used to evaluate changes related to CCL disease, CCL-defi cient limbs were shown to have signifi cantly lower stride length and paw velocity than healthy limbs (Sanchez-Bustinduy et al. 2010). Aft er TPLO surgery, the stride length returned to normal by 12 weeks, but paw velocity remained lower than recorded in normal dogs (de Medeiros et al. 2011). In kinetic and kinematic evaluation, the focus is on assessment of dynamic weight bearing and movement of the dog. Static weight bearing (SWB) that is measured while the dog is standing, has been evaluated in healthy dogs and also used in assessment of orthopedic outcome and perioperative pain (Horstman et al. 2004, Lascelles et al. 2006, Phelps et al. 2007, Lascelles et al. 2010, Davila et al. 2013). However, according to my knowledge the SWB has not been used to evaluate surgical outcome aft er CCL repair.

2.6.4 DiagnosƟ c imaging Radiographic evaluation of OA changes in stifl e joints has across the years been a basic part of the outcome assessment aft er CCL surgery. Varying scales from a simple 4-grade classifi cation (deRooster and vanBree 1999) to complex scoring systems (Elkins et al. 1991, Vasseur and Berry 1992, Innes et al. 2004) have been used. Although the OA changes provide evidence of pathology in the joint, the degree of these changes has been shown to correlate poorly with clinical function

18 Review of the literature of the limb, and this must be taken into account in assessment of clinical outcome (Gordon et al. 2003). Postoperative radiography can also be used to evaluate certain complications of CCL surgery such as postoperative patellar tendonitis and tibial fractures (Kergosien et al. 2004, Carey et al. 2005, Pettitt et al. 2014). Although one of the aims of CCL surgery has been to decelerate or halt the progression of degenerative joint disease, OA changes have been shown to advance aft er repair with both intracapsular and extracapsular techniques (Elkins et al. 1991, Vasseur and Berry 1992, Innes et al. 2004, Au et al. 2010). Th e newer dynamic stabilization techniques were fi rst proposed to resolve these problems and stop the progression of OA (Slocum and Devine Slocum 1998), but later studies have proven that degenerative changes progress in most dogs also aft er surgery with osteotomy techniques (Rayward et al. 2004, Jerram et al. 2005, Lineberger et al. 2005, Hoff mann et al. 2006, Boyd et al. 2007, Hurley et al. 2007, Au et al. 2010, Morgan et al. 2010). Of other diagnostic imaging modalities, MRI was reported to be more sensitive than radiography in assessing onset and progression of osteophytosis aft er experimental transection of CCL (D’Anjou et al. 2008). However, neither MRI nor computed tomography (CT) has been used to evaluate surgical outcome aft er CCL repair in clinical patients. Ultrasonographic examination has been used to evaluate postoperative signs of patellar ligament tendonitis aft er TTA and TPLO procedures (Mattern et al. 2006, Kühn et al. 2011, Pettitt et al. 2014).

19 Aims of the thesis 3 Aims of the thesis

Th e primary goal of this study was to evaluate the long-term (›1 year) surgical outcome aft er CCL repair in dogs. Th e secondary goal was to compare surgical techniques used to repair CCL rupture.

Specifi c aims of the study were as follows:

1. To report and compare normal reference data from force plate analysis in clinically healthy Rottweilers and Labrador Retrievers.

2. To describe a new method for measuring static weight bearing in pelvic limbs and to report the normal variance for the diff erence of weight bearing between pelvic limbs in clinically healthy Rottweilers and Labrador Retrievers.

3. To use an owner evaluation, including the validated Helsinki Chronic Pain Index, to assess chronic pain and long-term outcome aft er CCL surgery.

4. To use orthopedic, physiotherapeutic, radiographic, and force plate examinations for evaluation of long-term surgical outcome and function of the surgically treated limb aft er CCL surgery.

20 Materials and Methods 4 Materials and methods

4.1 Animals 4.1.1 Control dogs (I, III, IV) Clinically healthy adult Rottweilers and Labrador Retrievers, voluntarily presented by their owners, were used for force plate evaluation in Study I, for measuring pelvic limb static weight bearing in Study III, and as a control population (control group) for force plate and physiotherapeutic examination in Study IV. Th e inclusion criteria were age 1-8 years, no known orthopedic problems or abnormal fi ndings in orthopedic examination, and radiographic screening results free of elbow and hip dysplasia according to the Fédération Cynologique Internationale screening protocol (grade 0/0 for elbow dysplasia and grade A or B for hip dysplasia).

4.1.2 Surgically treated dogs (II, IV) Th e case selection for Study II was done in cooperation with the Veterinary Teaching Hospital of the University of Helsinki and 5 private orthopedic referral clinics in Finland (Espoon eläinsairaala, Finnin Ratsutila ja Eläinlääkintä, HauMau, Malmin eläinklinikka Apex and Mevet). Privately owned dogs with CCL rupture that had been surgically treated with intracapsular, extracapsular, TPLO, or TTA technique between January 2004 and December 2006 were included. Patients with other known concomitant stifl e problems, such as patellar luxation, collateral ligament damage, and septic or immune-mediated arthritis, at the time of the initial CCL surgery were excluded. Based on the results of Study II and another questionnaire study (unpublished data evaluating the outcome of TTO technique), owners were invited to bring their dogs for a clinical evaluation (study group in Study IV). Th e inclusion criteria were surgically treated unilateral CCL rupture, a minimum follow-up time interval of 1.5 years between surgery and evaluation, and a body weight ›17 kg. In addition to exclusion criteria set for Study II, also dogs with bilaterally treated CCL rupture or other known orthopedic or neurologic problems were excluded. Any NSAID, opioid, or corticosteroid pain medications and nutraceutical and fatty acid supplements were withdrawn at a minimum of 7 days, long-term corticosteroids 30 days, and pentosan polysulfate 90 days before the evaluation. In Studies I and II, institutional guidelines for research on animals were followed, and Studies III and IV were approved by the Ethics Committee of Viikki campus, University of Helsinki. In addition, a written consent was obtained from all owners.

4.2 Medical records (II, IV) With permission from the owners, the medical records of the surgically treated dogs were reviewed to determine the surgical date, technique used, surgically treated limb, information about meniscal tear, and possible complications or revision surgery needed.

21 Materials and methods 4.3 Owner evaluaƟ on and Helsinki Chronic Pain Index (II) A questionnaire was sent to the owners of surgically treated dogs that fulfi lled the inclusion criteria for the study. In addition to signalment, CCL surgery information, including the time of the operation/operations and the type of the rupture (uni- or bilateral), the duration of the clinical signs before surgery (< 1 month, 1 to < 2 months, 2 to < 3 months, 3 – 4 months, > 4 months), and the duration of postoperative lameness (< 1 month, 1 to < 2 months, 2 to < 3 months, 3 – 6 months, > 6 months, or dog is still lame), was collected. If the dog was bilaterally surgically treated, the owner was asked to answer based on the cruciate surgery performed between 2004 and 2006, and if both stifl e joints underwent surgery during this period, based on the latter surgery. Owners were also asked whether the dog had received postoperative physiotherapy. Th e Helsinki Chronic Pain Index (HCPI) was used to evaluate possible signs of chronic pain at follow-up (time interval between surgery and the questionnaire). It contained 11 questions on the dog’s mood, lameness, and willingness to move, play, and jump. Owners were asked to select the answer that best described their dog on a 5-point descriptive scale. Th e answers were later tied to a value (0 to 4), which when summed yielded a total index mimimum score of 0 and a maximum score of 44 (Hielm-Björkman et al. 2003, Hielm-Björkman et al. 2009). Owners were also asked to give their opinion on the surgical outcome (excellent, good, fair, or poor), and rate the frequency of lameness of the dog’s surgically treated limb (never, hardly ever, sometimes, oft en, or always) and the dog’s willingness to bear weight equally on the surgically treated and contralateral limb (always, very oft en, oft en, sometimes, or hardly ever) at follow-up. Th e questionnaire also inquired about the need for long-term administration of nonsteroidal anti-infl ammatory drugs (NSAIDs, e.g. carprofen and meloxicam) as well as whether the drug was administered for the surgically treated stifl e joint. Similar questions were asked about nutraceutical products (i.e. glucosamine and chondroitin sulfate).

4.4 Clinical evaluaƟ on (I,III,IV) Clinical evaluation in healthy dogs consisted of physical, orthopedic, and physiotherapeutic examinations, anatomic measurements, and force plate analysis (Studies I, III, and IV) and in surgically treated dogs physical, orthopedic, radiographic, and physiotherapeutic examinations and force plate analysis (Study IV). During the evaluation of the surgically treated dogs the members of the research team were unaware of which hind limb had been operated on and the surgical technique used.

4.4.1 Orthopedic examinaƟ on (I,IV) An orthopedic examination included lameness evaluation on a scale from 0 to 4 (no lameness, mild lameness or minor gait abnormality, moderate lameness or gait abnormality, severe weight-bearing lameness, non-weight-bearing lameness) (Mostafa et al. 2009) and palpation of the thoracic and pelvic limbs and spine as well as evaluation of conscious proprioception and withdrawal refl ex. Th e stifl e joints were palpated for pain (no, mild, moderate, severe), crepitation (no, mild, moderate, severe), periarticular swelling (no, mild, moderate, severe) and decrease in ROM (no, mild, moderate, severe), and patellar luxation, evaluated with the tibial

22 Materials and methods compression test and in sedation with the cranial drawer test. Th e spine was evaluated for pain and all other joints for pain, crepitation, swelling, decreased ROM, and instability. While sedated, an Ortolani test was used to evaluate hip joint laxity.

4.4.2 Radiographic examinaƟ on (IV) Th e stifl e, hip, and elbow joints bilaterally as well as the lumbar spine in surgically treated dogs were radiographed under sedation and evaluated. Th e amount of OA in stifl e joints was evaluated from mediolateral and craniocaudal views on a scale from 0 to 3 (no, mild, moderate, severe) (deRooster and vanBree 1999). In addition, the amount of OA in the hip joints was evaluated from an extended ventrodorsal view, in the elbow joints from 45° fl exed mediolateral and craniocaudal views, and in the lumbar spine from laterolateral and ventrodorsal views.

4.4.3 Anatomic measurements (I) Anatomic measurements were performed in healthy control dogs in order to normalize ground reaction forces according to conformational data. With the dog in standing position, distances from the ground to the highest point of the scapula, to the highest point of the iliac crest, and to the highest point of the olecranon process, as well as the length of the spine from the spinous process of the 2nd thoracic vertebra to the end of the last lumbar vertebra were measured. Also, humeral length (distance from greater tubercle to lateral epicondyle) and femoral length (distance from greater trochanter to lateral epicondyle) were measured while the dog was in lateral recumbency. Measurements were read by a second person, taken 3 times using a metric scale, and averaged. Functional limb length of a dog corresponding to the height of the center of the mass above the point of ground contact was calculated as a distance from the ground to the olecranon process plus one-third of the distance from the olecranon process to the highest point of the scapula (Bertram et al. 2000). Functional limb length and mean trotting velocity were used to calculate a dimensionless relative trotting velocity according to the equation

Vrel = v/(gh)1/2 ,

where v is the mean trotting velocity, h is the functional limb length, and g is the gravitational acceleration (9.81 m/s2).

4.4.4 Force plate analysis (I,IV) Ground reaction forces were measured using a piezoelectric force plate (Kistler force plate, type 9286, Kistler Instrumente AG, Winterthur, Switzerland) embedded in the center of a 14-m runway and a computer-based soft ware program (Aquire 7.3, Sharon Soft ware Inc., Dewitt, MI). Velocity and acceleration were measured by use of 3 photoelectric cells, positioned 1 m apart, and a start-interrupt timer system (Sharon Soft ware Inc.). Dogs were guided over the force plate by the owner or a member of the research team at a trotting speed of 2.10-2.50 m/s and an acceleration of – 0.5 to + 0.5 m/s2. A trial was considered valid when the thoracic limb was followed by the ipsilateral pelvic limb, both fully contacting the plate, and with the dog trotting next to the handler without pulling on the leash. Th e trial was discarded if the dog was distracted during the measurement, if the limb struck the edge of the plate, or if any portion of the contralateral paw hit the plate. Two members of the research team evaluated the trial to

23 Materials and methods confi rm which feet touched the plate. Aft er visual examination of the traces for consistency, a blinded assistant (person not otherwise participating in the study) chose 5 valid measurements for each side, and data from these trials were averaged.

Study I Peak vertical and craniocaudal forces and associated impulses, stance time, time to peak vertical force, breaking time, propelling time, and mean rising and falling slopes in thoracic and pelvic limbs were evaluated in Labrador Retrievers and Rottweilers. In the initial analysis and when the eff ects of relative velocity and anatomic measurements on force plate values were evaluated, the data of ground reaction forces and impulses normalized for the dog’s body weight were used. When the eff ect of body weight on force plate values was evaluated, raw data un-normalized for body weight were used to prevent double normalization of the values.

Study IV Peak vertical forces and vertical impulses in thoracic and pelvic limbs were evaluated as percentages of the dog’s body weight. In addition, distribution percentages per limb of peak vertical forces (DPVF) were calculated for study group dogs using the equation

DPVFSTPL = 100 PVFSTPL / (PVFSTPL + PVFCPL + PVFITL + PVFCTL),

where PVFSTPL is the PVFs recorded in the surgically treated limbs, PVFCPL in the contralateral pelvic limbs, PVFITL in the ipsilateral thoracic limbs, and PVFCTL in the contralateral thoracic limbs. Th e DPVFs were calculated similarly for contralateral pelvic limbs of study group dogs and for pelvic limbs of control group dogs. Th e symmetry index (SI) (Herzog et al. 1989) was calculated for the pelvic limb PVF and VI in surgically treated (study group) and control dogs using the equation

SI = 200[(XCPL/L – XSTPL/R) / (XCPL/L + XSTPL/R)],

where XCPL/L is the value recorded in the contralateral (study group) or left (control group) pelvic limb and XSTPL/R is the value recorded in the surgically treated (study group) or right (control group) pelvic limb. In surgically treated dogs, an SI value of 0 indicates perfect symmetry, a positive value decreased dynamic weight bearing in the surgically treated limb, and a negative value decreased dynamic weight bearing in the contralateral pelvic limb. Th e SIs were calculated similarly for control group dogs, but absolute values of the SIs were used. Th us, only the magnitude of SI was evaluated, and side (right or left pelvic limb) of deviation from perfect symmetry was disregarded.

4.4.5 PhysiotherapeuƟ c examinaƟ on (III, IV) A physiotherapeutic examination was done by a physiotherapist specialized in veterinary physiotherapy and consisted of 14 tests, as described previously (Hyytiäinen et al. 2013). In Study III, the use of bathroom scales in measuring SWB was described, and normal variance of the quantitatively measured static weight bearing (qmSWB) was reported. In Study IV, of all 14 tests, results of the visual evaluation of functional active range of motion (AROM), symmetry of

24 Materials and methods thrust between pelvic limbs, evaluation of muscle atrophy, qmSWB, and passive range of motion (PROM) in stifl e joints were presented in surgically treated dogs.

StaƟ c weight bearing (III, IV) Quantitative measurement of SWB in the pelvic limbs was done using two identical digital bathroom scales (Medica plus M-135, Truebell Vantaa, Finland), with the thoracic limbs placed on a custom-made platform of the same height as the scales. Th e measurement accuracy of the scale was 0.1 kg, with measurement range from 3 kg to 150 kg. Th e owner was instructed to hold the dog from the front, keeping it in a straight square-standing position and not to provide any manual support for the dog. Th e examiner kneeled behind the dog, placed the pelvic limbs symmetrically onto the scales and recorded measurements for both limbs. Th e results of four measurements on each limb were averaged and converted from kilograms to percentages proportional to the body weight (SWB%) and evaluated as mean ± SD values. In addition, diff erence in SWB between the pelvic limbs (Diff SWB%) was calculated using the following equation

Diff SWB% = 100 (SWBCPL/L (kg) – SWBSTPL/R (kg)) / body weight (kg),

where SWBCPL/L is the value recorded in the contralateral (study group) or left (control group) pelvic limb and SWBSTPL/R is the value recorded in the surgically treated (study group) or right (control group) pelvic limb. In surgically treated dogs, a Diff SWB% value of 0 indicates perfect symmetry, a positive value smaller SWB in the surgically treated pelvic limb, and a negative value smaller SWB in the contralateral pelvic limb. Th e absolute value of the Diff SWB% was used in the control group, and thus, the diff erence between right and left hind limb weight bearing was disregarded.

AcƟ ve range of moƟ on and thrust (IV) Th e dog was led over a 20-m distance and asked to “sit and sit-to-move”, and similarly “lie down and lie-to-move” 3 times. Th e position of pelvic limbs during sitting and lying was observed, and possible functional limitations or compensation of the pelvic limbs in sitting or lying position, such as external rotation or abduction of the limb, limited fl exion of stifl e or tarsal joints, or weakness/asymmetry of pelvic limbs in thrust from the ground, were noted.

Muscle atrophy (IV) Th e symmetry of the muscle bulk in pelvic limbs was evaluated manually by palpating and comparing simultaneously the width of the thigh muscles.

Passive range of moƟ on (IV) Th e passive range of motion in the stifl e joints was measured from unsedated dogs in lateral recumbency using a fl exible 180° universal goniometer with a 5° scale. Th ree measurements of each joint in maximal fl exion and extension from both pelvic limbs were taken and averaged. In addition to reporting the mean ± SD values, the PROM measurements were also evaluated as percentages of the contralateral limbs. Because fl exion angles were lower (higher in number) in

25 Materials and methods the surgically treated limbs, for calculation of percentages the fl exion angles were transformed by subtracting the angle from 180°.

4.5 StaƟ sƟ cal analyses Data analyses were performed with statistical soft ware programs. SPSS version 15.0 for Windows (SPSS Inc., Chigaco, IL) and SAS system version 9.1 (SAS Institute Inc., Cary, NC) were used in Study I, PASW version 18.0 for Windows (SPSS Inc., Chigaco, IL) and SAS system version 9.1 in Study II, SAS system version 9.2 in Study III, and SPSS version 21 for Windows in Study IV. Statistical signifi cance was set at the 5% level in all studies. Kolmogorov-Smirnov and Shapiro- Wilk tests served to determine whether data were normally distributed.

4.5.1 Study I Th e results were summarized as mean ± SD. Diff erences between left and right thoracic and left and right pelvic limbs were tested with the Wilcoxon rank test, and right and left limb values were combined. Pearson’s or Spearman’s correlation tests were used to analyze the relationship between weight and force plate values. Independent samples t-test or Mann-Whitney rank-sum test were used to detect diff erences in ground reaction forces between breeds. Linear regression analysis was applied to evaluate the eff ect of body weight and anatomic measurements on force platform values. Logarithmic transformations x´= log (x+100) were fi rst conducted to normalize the distribution of the variables. Th e eff ect of dog breed on force plate values was tested by ANCOVA, which allowed the removal of variance arising from relative velocity, anatomic measurements, and body weight of the dog.

4.5.2 Study II Explanatory variables included age, body weight, duration of clinical signs before surgery, surgical technique (intracapsular, extracapsular, or osteotomy), unilateral or bilateral surgery, meniscal integrity (damaged vs. intact), number of surgical procedures (1 vs. ≥ 2), and follow-up time, and outcome variables included HCPI, surgical outcome (excellent, good, fair, poor), duration of postoperative lameness, frequency of lameness at follow-up, and willingness to bear weight equally on the surgically treated and contralateral limbs at follow-up. Results from nominal and ordinal data were reported as frequencies and percentages. For continuous data, the results were summarized as mean ± SD and for HCPI, also as 95% confi dence interval. Th e Spearman rank test served to analyze the association between the HCPI and continuous or ordinal explanatory variables as well as other outcome variables. An unpaired t-test was used to compare diff erences in HCPI and age between the excellent and good outcome groups as well as between dogs with and without meniscal tears. Th e Mann-Whitney rank-sum and unpaired t-tests served to compare diff erences in explanatory and outcome variables between the dogs that underwent surgery at least twice and dogs that underwent surgery only once. Th e Kruskall- Wallis statistic and the Bonferroni-Holm correction method were used to compare diff erences for ordinal outcome variables between surgical techniques. An ANOVA and Bonferroni and Tukey post hoc analyses were used to compare diff erences for continuous variables between the surgical techniques. For comparison of HCPI between surgical techniques, ANCOVA allowed removal of the variance arising from the age of the dog.

26 Materials and methods 4.5.3 Study III Based on calculated mean ± SD diff erence in static weight bearing between the pelvic limbs in healthy control dogs, a cut-off value for normal Diff SWB% was determined (mean + SD). An unpaired t-test was used to compare diff erences in SWB% and Diff SWB% between the Rottweiler and Labrador breeds.

4.5.4 Study IV Results from nominal and ordinal data (surgical technique, condition of the meniscus, stifl e joint palpation and radiographic fi ndings, AROM, muscle atrophy), SI, and Diff SWB% were reported as frequencies and percentages. For continuous data (age at time of surgery, body weight at time of evaluation, follow-up, force plate data, SWB%, PROM) the results were summarized as mean ± SD. Other concurrent orthopedic problems were known to have a potential infl uence on results of ground reaction forces, AROM, SWB and muscle atrophy. In these evaluations the results were analyzed separately in dogs that, based on orthopedic and radiographic examinations, did not have concurrent orthopedic problems (subgroup 1). When PROMs of surgically treated limbs were compared with contralateral limbs, the results were analyzed in dogs that did not have concurrent contralateral stifl e or tarsal joint problems (subgroup 2). Concurrent problems in other joints, such as hip dysplasia or elbow OA were allowed, because these fi ndings were regarded to have no infl uence on results of PROM in stifl e or tarsal joints. When SIs were analyzed, the control group was used to determine the cut-off value for diff erentiation between normal and lame dogs by using mean + SD. Based on the cut-off values, the surgically treated dogs were assigned to be lame on the surgically treated limb, to have normal gait symmetry, or to be lame on the contralateral pelvic limb. Similarly, when Diff SWB%s were evaluated, the cut-off value determined in Study III was used to classify the surgically treated dogs to have symmetrical qmSWB, decreased qmSWB on the surgically treated limb, or decreased qmSWB on the contralateral pelvic limb. Th e Kruskall-Wallis statistic with pairwise comparisons was used to evaluate diff erences in age, follow-up, stifl e joint palpation fi ndings, and amount of radiographic OA between surgical techniques, and unpaired t-test or ANOVA was used to evaluate diff erences in body weight between study group and control group dogs or surgical techniques. Th e Mann-Whitney rank-sum test for not normally distributed and unpaired t-test for normally distributed force plate, SWB%, and PROM data were used to evaluate diff erences between surgically treated and contralateral limbs or control dog limbs. Similarly, the Kruskall-Wallis test with pairwise comparisons for not normally distributed and ANOVA with Bonferroni and Tukey post hoc analyses for normally distributed force plate, SWB%, and PROM data were used to evaluate diff erences between surgical techniques.

27 Results 5 Results

5.1 Animals 5.1.1 Control dogs (I, III, IV) A group of 9 Rottweilers and 12 Labrador Retrievers was used (Table 1). Rottweilers were signifi cantly heavier (43.6 ± 5.6 kg) than Labrador Retrievers (29.4 ± 2.8 kg) (P < 0.001). No signifi cant diff erence in age was present between the breeds.

5.1.2 Surgically treated dogs (II, IV) During the study period, 507 dogs were surgically treated. Owners of 272 dogs (53.6%) answered the questionnaire, and of returned questionnaires, 253 were eligible for evaluation in Study II (Table 1). Th e most common breeds were Labrador Retriever (14.2%), mixed breed (9.9%), Rottweiler (8.3%), Bichon Frisé (6.3%), German Shepherd Dog (3.2%), Golden Retriever (2.8%), Newfoundland Dog (2.8%), Bernese Mountain Dog (2.4%), and Staff ordshire Bull Terrier (2.4%). A modifi cation of the original intracapsular technique (Arnozcky et al. 1979) was used in 34.8% of dogs, two variations of an original modifi cation of the extracapsular technique (Flo 1975) with a nylon leader line or polyester sutures in 34.4% of dogs, and osteotomy techniques in 24.9% of dogs, including TPLOs (13.4%) and TTAs (11.5%), both performed as described elsewhere (Slocum and Devine Slocum 1993, Lafaver et al. 2007) (Table 1). In addition, in 5.9% of dogs the limb underwent surgery ≥ 2 times with a single or multiple techniques. Th e diagnosis was confi rmed and the joint visually inspected via arthrotomy in all included dogs. Th e condition of the meniscus was reported for 126 dogs (49.8%), with damage in 42% and an intact meniscus in 58%. A signifi cant diff erence was found in body weight between dogs that underwent surgery via the diff erent techniques (P < 0.001). Th e dogs in the extracapsular group were older at the time of surgery and at the time of evaluation than the dogs in the intracapsular and osteotomy groups (P < 0.005). In addition, the follow-up time was shorter in the osteotomy group than in the extracapsular and intracapsular groups (P < 0.001).

28 Results

Table 1 Description of the evaluated animals.

Control Surgically treated dogs dogs (I,III,IV) Owner evaluation (II) Clinical evaluation (IV) All Int Ext Ost All Int Ext Ost Number 253 88 87 63 47 19 7 21 21 of dogs (220) (77) (72) (57) Age at time of 5.9 5.8 7.3 4.6 4.2 3.5 4.1 4.9 - surgery (years) ± ± ± ± ± ± ± ± 3.3 3.0 3.3 2.9 2.7 2.0 2.4 3.2 Age at follow-up 8.2 8.2 9.8 6.5 6.9 6.8 6.4 7.0 3.2 (years) ± ± ± ± ± ± ± ± ± 3.2 2.9 3.1 2.5 2.7 2.4 2.7 3.0 1.6 Weight at 30.0 31.4 18.9 42.5 38.2 39.3 33.0 39.1 35.5 follow-up (kg) ± ± ± ± ± ± ± ± ± 17.9 16.7 13.6 14.4 9.5 10.0 3.7 10.1 8.2 Follow-up 2.7 2.9 2.9 2.2 2.8 3.3 3.6 2.2 - (years) ± ± ± ± ± ± ± ± 0.8 0.8 0.8 0.7 0.9 0.8 0.4 0.6 Age at time of surgery and at time of evaluation, body weight, and follow-up of the dogs are reported as mean ± SD. () number of dogs alive at time of evaluation. Int, intracapsular; Ext, extracapsular; Ost, osteotomy.

For clinical evaluation in Study IV, 47 study group dogs were assessed (Table 1). Th e breeds were Labrador Retriever (15), Rottweiler (6), mixed breed (4), Golden Retriever (3), Bernese Mountain Dog (2), Doberman (2), Newfoundland Dog (2), Nova Scotia Duck Tolling Retriever (2), and one each of Beauceron, Black Russian Terrier, Bordeaux Dog, Bullmastiff , Collie, Dalmatian, Giant Schnauzer, Karelian Bear Dog, Short-Haired German Pointer, South African Boerboel, and Wheaten Terrier. An intracapsular technique was used in 19 of these dogs (40.4%), an extracapsular technique in 7 dogs (14.9%), and an osteotomy technique, including 9 TPLOs, 7 TTAs, and 5 TTOs, performed as described in Bruce et al. (2007), in 21 dogs (44.7%). Th e condition of the meniscus was reported for 23 dogs (48.9%), with damage in 10 and an intact meniscus in 13 dogs. In contrast to Study II, there was no signifi cant diff erence in age or body weight of the dogs between the surgical technique groups. Th e mean follow-up was, however, signifi cantly shorter in the osteotomy group than in the intracapsular or extracapsular groups (P < 0.005) (Table 1). Th e dogs in the study group were signifi cantly older than the healthy contol dogs (P < 0.001). No signifi cant diff erence was found in body weight between the study group and healthy control group dogs.

5.2 Owner evaluaƟ on and Helsinki Chronic Pain Index (II) Owner assessment of duration of postoperative lameness was < 3 months for most dogs (77.9%). Postoperative rehabilitation administered by a physiotherapist was used in 15.3% of dogs. Owners reported the surgical outcome in 226 dogs and felt it was excellent in 54%, good in 42.9%, fair

29 Results in 0%, and poor in 3.1%. Owners’ long-term assessments of lameness and weight bearing of the surgically treated limb are reported in Table 2. Of 213 and 196 dogs, 12.2% were receiving NSAIDs and 25.0% nutraceuticals (i.e. glucosamine and chondroitin sulfate), respectively, for the surgically treated stifl e joint.

Table 2 Owners’ long-term assessment of lameness and weight bearing of the surgically treated limb.

Owner assessment of lameness of surgically treated limb at follow-up Hardly Never Sometimes Oft en Always ever Lameness 45.1% 32.6% 13.5% 5.1% 3.7% (n=215) (97) (70) (29) (11) (8)

Owner assessment of weight bearing at follow-up Hardly Always Very oft en Oft en Sometimes ever

Willingness to bear weight equally on 50.9% 25.5% 9.0% 6.1% 8.5% treated and (108) (54) (19) (13) (18) contralateral limb (n=212)

Results expressed as percentage (number) of dogs.

Th e HCPI (mean ± SD, 8.9 ± 6.3; range, 0 to 24) could be calculated for 206 dogs, and 31.1% of these had an HCPI ≥ 12. Dogs that either had died, had recently undergone surgery on the contralateral stifl e joint, or had missing answers to HCPI questions were excluded from HCPI determination. Th e dog’s age at the time of the questionnaire and HCPI were positively associated (r = 0.252; P < 0.001). Th e HCPI in dogs with good outcomes (11.8 ± 5.4; 95% CI, 10.6 to 12.9) was signifi cantly (P<0.001) higher than in dogs with excellent outcomes (6.1 ± 5.7; 95% CI, 5.1 to 7.2) (Figure 7). However, no signifi cant diff erence in age was found between surgical outcome groups. Th e body weight of the dog, duration of clinical signs before surgery, follow-up time, and presence or absence of a meniscal tear had no signifi cant association with the HCPI. Th e duration of postoperative lameness and HCPI were positively correlated (r = 0.210; P = 0.002).

30 Results

Figure 7 Aft er surgical treatment of CCL rupture in dogs; box-and-whisker plots of the association between owner-assessed HCPI and owner-assessed “excellent” and “good” long-term outcomes.

In 15 dogs, the cruciate ligament-defi cient limb underwent surgery more than once. In these dogs, the duration of postoperative lameness was signifi cantly (P = 0.018) longer, and willingness to bear weight equally on the surgically treated and contralateral limb at follow-up was signifi cantly (P = 0.009) lower than in dogs that underwent surgery only once. Signalment or other outcome variables did not diff er signifi cantly between dogs that underwent surgery more than once and dogs that underwent surgery only once. When surgical outcome was compared with the surgical techniques used, the dogs that underwent surgery with ≥ 2 techniques on the same stifl e joint were excluded (n=11). Th e owner’s assessment of duration of postoperative lameness in dogs treated with osteotomy techniques was signifi cantly (P = 0.014; Bonferroni-Holm test signifi cance at P < 0.016) shorter than in dogs treated with the intracapsular technique (Figure 8). No signifi cant diff erence was found between surgical techniques in frequency of lameness or willingness to bear weight on the surgically treated limb at follow-up (Figures 9 and 10). Th e HCPI in dogs treated with osteotomy techniques was signifi cantly (P = 0.030) lower than in dogs treated with the extracapsular technique. However, aft er the diff erence in age between the dogs in the osteotomy and extracapsular groups was controlled with an ANCOVA, no signifi cant diff erence in HCPI between these dogs was found. Apart from a positive association between the dog’s age and HCPI, there was no association between signalment variables or follow-up time and the outcome variables used to compare surgical techniques.

31 Results

Figure 8 Composite bar graph of owner assessment of length of postoperative lameness in the surgically treated limb (percentage) aft er treatment with intracapsular (n=87), extracapsular (88), and osteotomy (61) techniques.

Figure 9 Composite bar graph of owner assessment of frequency of lameness in the surgically treated limb (percentage) at follow-up aft er treatment with intracapsular (n=78), extracapsular (n=70), and osteotomy (n=56) techniques.

32 Results

Figure 10 Composite bar graph of owner assessment of willingess to bear weight equally on the surgically treated and contralateral limb (percentage) at follow-up aft er treatment with intracapsular (n=77), extracapsular (n=70), and osteotomy (n=54) techniques.

5.3 Clinical evaluaƟ on (I,III,IV) 5.3.1 Orthopedic and radiographic examinaƟ on (I, IV) Th e 21 healthy control dogs did not show any signs of orthopedic disease or lameness. Of surgically treated dogs, grade 1/4 lameness of the surgically treated limb was seen in 8 dogs (3 intracapsular, 1 extracapsular, 4 osteotomy). One dog treated with the intracapsular technique had grade 2/4 lameness. Th e gait could not be evaluated in one dog. Orthopedic and radiographic fi ndings in the surgically treated stifl e joints are presented in Table 3. No signifi cant diff erences were found between surgical techniques in pain response to stifl e fl exion/extension. eTh amount of stifl e joint crepitation was signifi cantly lower in the osteotomy group than in the extracapsular (P = 0.030) or intracapsular (P = 0.028) groups, and the amount of periarticular swelling was signifi cantly lower in the osteotomy group than in the extracapsular group (P = 0.047). Th e drawer sign was tested in 39 dogs. Of 12 dogs that presented a positive drawer sign, 5 had been treated with intracapsular and 7 with osteotomy techniques. Th e tibial compression test was negative in all stifl e joints. eTh severity of radiographic OA was lower in the osteotomy group than in the intracapsular group (P = 0.013). Based on orthopedic and radiographic examinations, 21 surgically treated dogs were included in subgroup 1 and 33 dogs in subgroup 2.

33 Results

Table 3 Stifl e joint fi ndings in the surgically treated limb at follow-up; orthopedic and radiographic examination.

Severity of stifl e joint fi nding

No fi ndings Mild Moderate Severe

Pain response to stifl e 68.1% 25.5% 6.4% - fl exion/extension (n=47) (32) (12) (3) Periarticular swelling 36.2% 53.2% 10.6% - (n=47) (17) (25) (5) Crepitation 6.4% 34.0% 51.1% 8.5% (n=47) (3) (16) (24) (4) Cranial drawer sign 69.2% 15.4% 12.8% 2.6% (n=39) (27) (6) (5) (1) Radiographic OA 2.2% 19.6% 45.6% 32.6% (n=46) (1) (9) (21) (15) Results are expressed as percentages (number) of dogs. n, number of dogs evaluated.

5.3.2 Anatomic measurements (I) Anatomic measurements were available for 20 healthy control dogs. Mean ± SD distance from the ground to the highest point of the scapular spine and the highest point of the iliac crest for Rottweilers were 0.610 ± 0.036 and 0.607 ± 0.033 m and for Labrador Retrievers 0.537 ± 0.034 and 0.536 ± 0.030 m, respectively. Mean ± SD humeral length and femoral length for Rottweilers were 0.192 ± 0.011 and 0.195 ± 0.013 m and for Labrador Retrievers 0.173 ± 0.013 and 0.172 ± 0.016 m, respectively. Mean ± SD length of the spine for Rottweilers was 0.468 ± 0.021 m and for Labrador Retrievers 0.396 ± 0.343 m. Functional limb length for Rottweilers was 0.433 ± 0.026 m and for Labrador Retrievers 0.377 ± 0.025 m. All anatomic measurements were signifi cantly larger in Rottweilers (P<0.001).

5.3.3 Force plate analysis (I,IV) Study I Initial force plate data and diff erences between the Rottweilers and Labrador Retrievers are presented in Table 4. Mean ± SD values for velocity and acceleration were 2.28 ± 0.05 m/s and 0.02 ± 0.14 m/s2 for Rottweilers and 2.25 ± 0.05 m/s and -0.09 ± 0.12 m/s2 for Labrador Retrievers, respectively. No signifi cant diff erence existed between breeds in trotting velocity (P= 0.055) but there was a signifi cant diff erence between breeds in acceleration (P = 0.013). When breeds were combined, body weight correlated with all force platform values that were diff erent between breeds. Based on the anatomic measurements, calculated relative velocities in Rottweilers and Labrador Retrievers were 1.11 ± 0.04 and 1.17 ± 0.04, respectively, with signifi cant diff erence between breeds (P<0.001). When anatomic variables were analyzed, functional limb length was found to have the most signifi cant eff ect on force plate values and was chosen to be used in the ANCOVA as the anatomic variable. ANCOVA was applied to evaluate diff erences in force plate values between breeds when the eff ect of relative velocity and functional limb length, separately and together, 34 Results was eliminated from initial force plate data (Table 5). When the variation arising from relative velocity, functional limb length, or both was removed, diff erences between breeds in thoracic limb peak vertical forces as well as rising and falling slopes and in pelvic limb vertical impulses remained, but were eliminated in thoracic limb vertical impulses. For craniocaudal forces and impulses, the results were inconsistent. Finally, when ANCOVA was used to remove the eff ect of body weight from raw force plate data, all diff erences between the breeds were eliminated.

Table 4 Initial force plate data of healthy Rottweilers and Labradors

Gait variable Limb Rottweilers Labradors P-value

TL 274.9 ± 17.1 234.3 ± 11.5 <0.001 Stance time (ms) PL 238.1 ± 16.4 212.6 ± 15.5 <0.001 TL 117.0 ± 6.1 125.3 ± 10.7* 0.005 PVF (100*N/N) PL 71.9 ± 2.8 72.5 ± 4.8 0.657 TL 17.1 ± 1.3 15.7 ± 0.8 <0.001 VI (100*N-s/N) PL 9.3 ± 0.6 8.3 ± 0.6 <0.001 TL 131.3 ± 9.3 116.1 ± 7.6 <0.001 PVF time (ms) PL 102.0 ± 6.8 92.2 ± 8.1 <0.001

TL 0.90 ± 0.09 1.08 ± 0.16* <0.001 AvRS (100*N/N/ms) PL 0.71 ± 0.06 0.78 ± 0.11 0.009 TL -0.82 ± 0.06 -1.05 ± 0.12 <0.001 AvFS (100*N/N/ms) PL -0.54 ± 0.06 -0.60 ± 0.08 0.006 TL -16.5 ± 2.6 -18.2 ± 2.6* 0.034 Break peak (100*N/N) PL -6.6 ± 2.3 -6.3 ± 1.7 0.583 TL -1.50 ± 0.30 -1.43 ± 0.19 0.341 Break imp (100*N-s/N) PL -0.27 ± 0.13 -0.25 ± 0.11 0.712 TL 79.5 ± 9.8 76.5 ± 9.9 0.477 Break time (ms) PL 27.5 ± 5.5 27.7 ± 8.8 0.611 TL 8.3 ± 1.4 9.5 ± 1.5 0.007 Prop peak (100*N/N) PL 10.8 ± 1.9 11.3 ± 1.5 0.384 TL 0.62 ± 0.10 0.59 ± 0.11 0.366 Prop imp (100*N-s/N) PL 0.87 ± 0.17 0.81 ± 0.16 0.277 TL 204.1 ± 18.6 177.7 ± 9.0 <0.001 Prop time (ms) PL 149.2 ± 8.7 134.6 ± 9.9 <0.001

*Signifi cant (P<0.05) diff erence present between right and left limb values. Forces, impulses, and slopes are reported as percentages of body weight. PVF, peak vertical force; VI, vertical impulse; PVF fi me, time to peak vertical force; AvRS, average vertical rising slope; AvFS, average vertical falling slope, Break peak, breaking peak force; Break imp, breaking impulse; Break time, breaking time; Prop peak, propelling peak force; Prop imp, propelling impulse; Prop time, propelling time; TL, thoracic limb; PL, pelvic limb.

35 Results

Table 5 Diff erences between Rottweilers and Labradors in force plate data before and aft er analyses of covariance.

Initial Vrel + Gait variable Limb Vrel Funct Weight data Funct

TL <0.001 <0.001 <0.001 <0.001 0.592 Stance time PL <0.001 0.001 0.008 0.007 0.186

TL 0.003 0.012 0.003 0.002 0.492 PVF PL 0.657 0.477 0.422 0.494 0.387

TL <0.001 0.051 0.548 0.497 0.833 VI PL <0.001 <0.001 0.002 0.001 0.812

TL <0.001 0.001 0.009 0.006 0.358 PVF time PL <0.001 0.023 0.074 0.050 0.690

TL <0.001 0.004 0.004 0.003 0.371 AvRS PL 0.009 0.248 0.417 0.337 0.955

TL <0.001 <0.001 <0.001 <0.001 0.482 AvFS PL 0.006 0.088 0.162 0.151 0.053

TL 0.034 0.206 0.069 0.038 0.196 Break peak PL 0.583 0.018 0.024 0.041 0.884

TL 0.341 0.417 0.880 0.723 0.127 Break imp PL 0.712 0.037 0.028 0.044 0.981

TL 0.477 0.919 0.848 0.945 0.116 Break time PL 0.611 0.826 0.607 0.651 0.983

TL 0.007 0.081 0.253 0.280 0.823 Prop peak PL 0.384 0.686 0.998 0.795 0.235

TL 0.366 0.289 0.254 0.258 0.689 Prop imp PL 0.277 0.146 0.369 0.453 0.550

TL <0.001 0.001 0.018 0.012 0.166 Prop time PL <0.001 0.003 0.012 0.007 0.558

Diff erences between breeds when initial data was used and when ANCOVA was used to eliminate the variance arising from relative velocity (Vrel), functional limb length (Funct), relative velocity and functional limb length (Vrel+Funct), and body weight (Weight). Th e eff ect of relative velocity and functional limb length was tested from initial data, where forces, impulses, and slopes were expressed as percentage of body weight, and the eff ect of body weight was tested from un-normalized raw data. Th e level of signifi cance was set at P < 0.05, and P-values signifi cantly diff erent between breeds are indicated in boldface. For abbreviations, see Table 4.

36 Results

Study IV Th e results of force plate analysis were available for 41 dogs (study group). Body weight, trotting velocity, or acceleration did not diff er signifi cantly between the study group or subgroup 1 dogs and the control group dogs, between the surgically treated and the contralateral limbs, or between the surgical technique groups. When force plate results for the study group were evaluated, no signifi cant diff erences emerged when comparing the surgically treated limb with the contralateral pelvic limb or the pelvic limbs of control dogs. In subgroup 1, the surgically treated limbs had a signifi cantly lower PVF (P = 0.040) and DPVF (P = 0.005) than the contralateral pelvic limbs, but no signifi cant diff erences were found when surgically treated limbs were compared with control dog limbs (Table 6). Th e determined SI cut-off values were 7.1% for PVF and 8.1% for VI. To improve the specifi city of the cut-off value, they were rounded up to 8 % and 9%, respectively. Based on these cut-off values, the SI for PVF was <8% in 13 dogs (61.9%), indicating normal gait symmetry, and ≥ 8% in 7 dogs (33.3%), indicating lameness. In one dog, the SI indicated lameness of the contralateral pelvic limb (≤ -8%). Similarly, the SI for VI was <9% in 15 dogs (71.4%), indicating normal gait symmetry, and ≥ 9% in 6 dogs (28.6%), indicating lameness. Of the 7 dogs in which one or both SIs indicated lameness, 3 had been treated with intracapsular, 2 with extracapsular, and 2 with osteotomy techniques. When the force plate results of the diff erent surgical technique groups were evaluated, no signifi cant diff erences were found in study group between intracapsular, extracapsular, and osteotomy techniques or in study group or subgroup 1 when surgically treated limbs in each technique group were compared with pelvic limbs in control dogs (Table 6). Subgroup 1 dogs that had been treated with the intracapsular technique had signifi cantly lower DPVF in the surgically treated limb than the dogs treated with osteotomy techniques (P = 0.044). Also the dogs treated with the intracapsular technique had signifi cantly lower VI (P = 0.035) and DPVF (P = 0.018) in the surgically treated limb than in their contralateral pelvic limb (Table 6). No signifi cant diff erences were seen when limbs treated with the osteotomy technique were compared with their contralateral pelvic limb. Th e number of dogs treated with the extracapsular technique (n=3) was too low to be analyzed.

37 Results

Table 6 Mean ± SD results of force plate analysis and static weight bearing in subgroup 1 dogs, separately in intracapsular, extracapsular, and osteotomy surgical technique groups, and in control dogs.

Control Subgroup 1 Intra Extra Osteo group (n=21) (n=7) (n=3) (n=11) (n=21) STPL CPL STPL CPL STPL CPL STPL CPL PL Force plate analysis PVF 70.6 75.3 67.2 75.1 73.2 75.2 72.2 NA NA (% N/N) ±7.0# ±7.3 ±7.6 ±6.7 ±6.8 ±8.9 ±4.0 VI 8.7 9.4 8.3 9.5 9.2 9.5 8.7 NA NA (%Ns/N) ±1.2# ±1.0 ±0.9# ±1.0 ±1.3 ±1.1 ±0.8 18.5 19.8 17.8 19.9 19.2 19.7 18.7 DPVF (%) NA NA ±1.3# ±1.4 ±1.5#* ±1.3 ±0.9 ±1.7 ±1.1 Static weight bearing 15.5 19.7 13.8 21.2 16.8 19.4 17.7 SWB (%) NA NA ±5.6# ±3.6 ±7.3# ±3.3 ±5.0 ±3.9 ±2.8 # Signifi cant (P < 0.05) diff erence present between the surgically treated and the contralateral pelvic limb.* Signifi cant (P < 0.05) diff erence present in the measured value between the limbs treated with intracapsular and osteotomy techniques. Forces and impulses are reported as percentages of body weight. STPL, surgically treated pelvic limb; CPL, contralateral pelvic limb; PL, pelvic limb, PVF, peak vertical force; VI, vertical impulse; DPVF, percentage distribution of peak vertical force; SWB(%), static weight bearing reported as percentages proportional to body weight; Intra, intracapsular; Extra, extracapsular; Osteo, osteotomy; NA: not analysed due to low number of dogs in the extracapsular group.

5.3.4 PhysiotherapeuƟ c examinaƟ on (III, IV) Study III Th e mean ± SD SWB% in pelvic limbs of healthy control dogs was 17.7 ± 2.8 %, and the mean ± SD Diff SWB% between pelvic limbs in healthy control dogs was 3.3 ± 2.7 %, resulting in a 6% cut-off value as the normal limit. No signifi cant diff erence existed in SWB% or Diff SWB% between the Rottweiler and Labrador breeds.

Study IV Th e qmSWB was not evaluated in 2 dogs due to lack of cooperation, and the PROM was not evaluated in 3 dogs for the same reason and in one dog because of a concurrent orthopedic problem in the tarsal joint. When results of qmSWB were evaluated, dogs in both the study group and subgroup 1 had a signifi cantly lower SWB% in the surgically treated limb than in the contralateral pelvic limb (P = 0.010 and P = 0.006, respectively) (Table 6). Based on the Diff SWB cut-off value (Study III), 13 of 21 dogs (61.9%) had symmetrical weight bearing, 6 (28.6%) had decreased weight bearing in the surgically treated limb, and 2 (9.5%) bore less weight on the contralateral pelvic limb. Of the dogs with decreased weight bearing in the surgically treated limb, 3 had been treated with intracapsular and 3 with osteotomy techniques.

38 Results

In the study group, no signifi cant diff erences emerged when SWB in the surgically treated limb was compared between surgical techniques or separately in each surgical technique group with either the contralateral limb or the control dogs’ limbs. Similarly, in subgroup 1 dogs, no signifi cant diff erences were found between surgical techniques. When each surgical technique group in subgroup 1 was evaluated separately and surgically treated limbs were compared with the contralateral and control dog limbs, no signifi cant diff erences were observed in the osteotomy group, but in the intracapsular group the SWB% in the surgically treated limb was signifi cantly lower than that in the contralateral limb (P= 0.030) (Table 6). Th e number of dogs in the extracapsular group (n=3) was too low to be analyzed. In visual evaluations of functional AROM, the most frequent fi ndings were in sitting position. In subgroup 1, a decreased fl exion of the tarsal joint was seen in 12 dogs (57.1%; 2 intracapsular, 10 osteotomy), a decreased fl exion of the stifl e joint in 10 dogs (47.6%; 2 intracapsular, 1 extracapsular, 7 osteotomy), and an abduction of the limb in 9 dogs (42.9%, 1 intracapsular, 1 extracapsular, 7 osteotomy). Except for one dog, the positive fi ndings were always localized in the surgically treated limbs. Th rust from the sitting position was weaker in the surgically treated limb in 14 dogs (66.7%; 3 intracapsular, 1 extracapsular, 10 osteotomy). When evaluating muscle atrophy in subgroup 1, altogether 16 dogs (76.2%) had decreased muscle mass of the surgically treated pelvic limb on palpation (5 intracapsular, 2 extracapsular, 9 osteotomy), while only 4 dogs had symmetrical muscle mass, and one dog had decreased muscle mass in the contralateral pelvic limb. When PROMs of surgically treated limbs were compared with those of contralateral limbs, stifl e extension angles (P < 0.001) were signifi cantly lower and stifl e fl exion angles (P < 0.001) were signifi cantly higher (loss of fl exion) in the surgically treated limb (Table 7). Similar diff erences also emerged in the intracapsular and osteotomy groups when each surgical technique was evaluated separately and the surgically treated limb was compared with the contralateral limb. In the extracapsular group, stifl e fl exion angles (P=0.045) were signifi cantly higher (loss of fl exion) in surgically treated limbs than in contralateral limbs. No signifi cant diff erences were found between the surgical techniques.

Table 7 Mean ± SD stifl e joint extension and fl exion angles and percentages of surgically treated limbs relative to contralateral limbs in subgroup 2 dogs, and separately in intracapsular, extracapsular, and osteotomy surgical technique groups.

Subgroup 2 Intra Extra Osteo (n=33) (n=11) (n=6) (n=16) STPL CPL STPL CPL STPL CPL STPL CPL Stifl e extension 149.2 160.3 148.2 163.0 144.7 155.3 151.7 160.3 Degr ±8.6* ±6.6 ±7.5* ±2.5 ±10.9 ±11.1 ±8.0* ±5.6 % 93.1 90.9 93.2 94.6 Stifl e fl exion 47.9 41.1 48.0 40.3 49.7 44.2 47.1 40.4 Degr ±7.2* ±6.5 ±5.4* ±8.5 ±5.2* ±2.0 ±8.9 ±6.0 % 95.1 94.5 95.9 95.2 *Signifi cant (P < 0.05) diff erence present between the surgically treated limb and the contralateral pelvic limb. STPL, surgically treated pelvic limb; CPL, contralateral pelvic limb; Degr, degrees. Intra, intracapsular; Extra, extracapsular; Osteo, osteotomy.

39 Discussion 6 Discussion

6.1 Challenges of outcome measures Force plate analysis has been considered the gold standard for objective assessment of limb function (Budsberg et al. 1987, Rumph et al. 1994). To enable the comparison of force plate values between individual patients or patient groups, ground reaction forces and impulses are usually expressed as a percentage of body weight and the subject velocity and acceleration are limited to a narrow range. In Study I, the ground reaction forces in healthy Rottweilers diff ered signifi cantly from force plate data obtained from healthy Labradors when standard normalization by linear scaling to body weight, given by the soft ware, was used. Th e body weight and anatomic measurements of the Rotteilers were also signifi cantly larger than in Labrador Retrievers. When breeds were combined, body weight correlated with all force plate values that were signifi cantly diff erent between breeds. A similar correlation of body weight with normalized ground reaction forces has been reported in studies using heterogeneous populations of dogs with large variation in body weight or breed (Budsberg et al. 1987, Voss et al. 2007). Although the absolute trotting velocity was controlled, the calculated relative velocity, infl uenced by the limb length, diff ered signifi cantly between the breeds in Study I, meaning that the smaller Labradors actually traveled at a faster relative velocity than taller Rottweilers. To evaluate the cause for the breed diff erences, ANCOVA was used to eliminate variance associated with relative velocity, functional limb length, and body weight. Based on the results, the ground reaction force diff erences between breeds were attributable to diff erences in conformation and the related changes to relative velocity and body weight of the dog. ANCOVA appeared to remove the eff ect of body weight more effi ciently than the stardard normalization. Recently, also other studies have focused on evaluating the eff ect of body conformation or body weight on ground reaction forces (Bertram et al. 2000, Voss et al. 2010, Voss et al. 2011, Krotscheck et al. 2014), and normalization methods to account for the conformational diff erences have been proposed (Voss et al. 2010). Th ese studies indicate that when comparing dog groups with diff ering body weights or sizes within and between research studies the standard methods of normalization for ground reaction data may be inadequate and this can cause bias in the study results. One option to overcome the problems related to comparison of heterogeneous populations of dogs is to use symmetry or asymmetry indices (SI or ASI). Several SI or ASI formulas have been introduced (Herzog et al. 1989, Budsberg et al. 1993). In Study IV, the SIs were calculated for healthy and surgically treated dogs and the values of the healthy dogs were used to determine a cut-off value for diff erentiation between normal gait symmetry and lameness. Also other reports using SIs to evaluate the outcome aft er CCL repair have been published (Voss et al. 2008, Böddeker et al. 2012, Nelson et al. 2012). In order to use these indices reliably, the contralateral limb should be free of orthopedic problems. In Study IV, a physiotherapeutic examination from the perspective of the professional veterinary physiotherapist was included in the evaluation methods. Th e evaluation concentrated on the stifl e joints, and, in addition to measurements of stifl e joint ROM and evaluation of muscle atrophy, also new methods, such as SWB, AROM during sitting and lying positions, and thrust from the ground, were included. Th e assessments of AROM and thrust from the ground were targeted to evaluate the functionality of the patient, in other words, the dog’s ability to perform

40 Discussion the activities of daily living (e.g. sit or lie down, get up from sitting or lying position, climb stairs, get into a car). Although dynamic weight bearing has been evaluated in many CCL studies, SWB measurements have not been reported before in assessment of surgical outcome aft er CCL repair. In Study III, a simple and cost-eff ective method for measurement of SWB was described, and normal variance for the diff erence between the pelvic limbs was set.

6.2 Chronic pain in surgically treated dogs Based on the owner evaluation in Study II, aft er a mean period of 2.7 years, approximately 31% of the surgically treated dogs had a HCPI value ≥ 12, indicating pain. Also approximately 32% of the clinically evaluated dogs in Study IV showed a pain response to extension/fl exion of the surgically treated stifl e joint. Th e signs of pain in these dogs may be related to the progression of osteoarthritis. Osteoarthritis can cause chronic pain and discomfort and have signifi cant impact on a patient’s quality of life (Lamont et al. 2000). Chronic pain might also be related to poor preoperative condition of the surgically treated stifl e joint or to surgical complication such as technical error during surgery, implant failure and/or migration, late meniscal tear, patellar luxation, septic arthritis, suture infection or bone/joint neoplasia. Innes et al. (2000) reported that patients with meniscal injury diagnosed during CCL repair had signifi cantly lower long- term levels of exercise and higher levels of disability and inactivity stiff ness than in patients without a meniscal injury. In study II, the HCPI had no signifi cant association with the duration of the clinical signs before surgery, or presence or absence of the meniscal tear in surgery, but the condition of the meniscus was reported only in 50 % of the patients. In 5.9 % of dogs the cruciate ligament-defi cient limb underwent surgery more than once with a single or multiple techniques. Based on owner evaluation, these dogs were less willing to bear weight on the surgically treated limb at follow-up than dogs that underwent surgery only once, but no signifi cant diff erences were found in HCPI values between the groups. Chronic pain has not been extensively studied in CCL-treated patients. Th e focus of the postoperative studies has been on evaluation of dynamic weight bearing, which indirectly may measure pain, but can also be aff ected by other factors such as limitations in range of motion or loss of muscle mass of the surgically treated limb. In addition, objective evaluation of pain is diffi cult and orthopedic outcome assessment has long lacked appropriate tools for pain assessment. Validated owner-reported evaluation methods targeted for evaluation of OA and chronic pain have been used in several CCL studies (Innes et al. 2000, Boyd et al. 2007, Corr and Brown 2007, Renwick et al. 2009, Gordon-Evans et al. 2013). However, these questionnaires have mainly focused on evaluating improvement of the dog’s condition postoperatively and comparing improvement between the surgical techniques, not evaluating whether the dog experiences pain or not. One option for evaluation of pain is to use VAS-based scores. A drawback to owner- reported VAS for pain is that if owners are not trained they have diffi culties in recognizing signs of pain, and thus, may underestimate it (Hielm-Björkman et al. 2011). Christopher et al. (2013) used owner-reported VAS to evaluate degree of pain in surgically treated dogs aft er > 1 year postoperatively. With a scale from 0 to 100, the mean VAS of pain ranged between 12 and 17, depending on the surgical technique used. When only presence or absence of pain was evaluated, 44-61% of dogs were judged by their owners to have at least some degree of long-term pain. Although owner assessments indicate that many CCL patients experience some degree of long-term pain aft er surgery, only 12% of the dogs in Study II received NSAIDs for the treatment

41 Discussion of pain arising from the surgically treated limb. Th is may refl ect the failure of the owners to recognize signs of pain in their dogs. Chronic pain may cause very subtle changes in demeanor, general mood, and willingness to move, which can be diffi cult to detect and understand as a pain response. It is therefore important that clients are informed about the long-term prognosis of CCL disease and are educated to recognize the signs of pain and to treat the pain accordingly. When the owner assessment of surgical outcome was evaluated, dogs with good outcomes had a mean HCPI indicative of chronic pain, whereas dogs with excellent outcomes had mean values at the level of healthy dogs. In studies evaluating the outcomes of CCL surgery, excellent and good outcomes are oft en combined, generally resulting in success rates of >90% (Elkins et al. 1991, Hoff mann et al. 2006, Corr and Brown 2007, Stein and Schmoekel 2008). Based on ndingsfi in Study II, this kind of generalization may give a false impression of a pain-free outcome and should not be used.

6.3 Long-term limb funcƟ on aŌ er cranial cruciate ligament surgery In Study IV, when evaluated at a mean of 2.8 years aft er the CCL surgery, the average dynamic and static weight bearing of the surgically treated limbs had returned to the level of healthy limbs. In force plate analysis, no signifi cant diff erences were found in ground reaction forces between the surgically treated pelvic limbs of CCL dogs and the control limbs of healthy dogs. Th e PVF of surgically treated limbs was signifi cantly and approximately 5% lower than the corresponding value for the contralateral pelvic limbs, but load distribution from injured to contralateral pelvic limb may artifi cially increase the diff erence between limbs (Rumph et al. 1993). Furthermore, the PVF of the surgically treated limbs was only 1.6% lower than the corresponding value for the control dog limbs, and mean VI was the same in surgically treated and control dog limbs. When static weight bearing was evaluated, a similar trend was observed. Although the SWB% of the surgically treated limbs was signifi cantly and approximately 4% lower than that of contralateral limbs, the diff erence between surgically treated and control dog limbs was only 2.2%. It can be assumed that, similar to dynamic weight bearing, load distribution to contralateral limbs occurs also when measuring SWB. When clinical importance of observed diff erences is evaluated, these result indicate that the mean diff erences between the surgically treated and healthy limbs were minor and without clinical relevance. Although the average dynamic and static weight bearing returned to the level of healthy limbs, in approximately 30% of dogs the SIs for PVF and VI indicated lameness and the Diff SWB% indicated decreased weight bearing in the surgically treated limb, thus refl ecting a clinically relevant problem in weight bearing in these dogs and also revealing considerable variation between the evaluated dogs. By using the cut-off values for SIs and Diff SWB%, more diff erences were revealed than those discovered when the mean ground reaction forces between the surgically treated and healthy limbs were compared. Interestingly, also compensations of AROM were seen in 40-50% of surgically treated limbs and weakness in thrust from the ground in two-thirds of limbs. Although subjective, these evaluations give clinically relevant information on how dogs function in their normal daily activities. When goniometric results of surgically treated limbs were evaluated, a statistically signifi cant loss of 5% in fl exion and 7% in extension of the stifl e joints was seen. Th e clinical relevance of these changes is unknown and can only be speculated. Jandi et al. (2007) reported

42 Discussion that loss of fl exion or extension ≥ 10 ° was associated with higher lameness scores, but in Study IV no correlation between goniometric results and dynamic or static weight bearing was found. Few studies have evaluated the long-term (>1 year) return of limb function to the preinjury level by using objective evaluation methods (Au et al. 2010, Moeller et al. 2010). Similar to the results of Study IV, Moeller et al. (2010) reported that stifl e joint extension and fl exion angles of surgically treated limbs remained signifi cantly decreased relative to contralateral joints at 1-5 years aft er TPLO surgery. Also in another study the stifl e joint ROM angles in surgically treated limbs were below the preoperative values 2 years aft er TPLO and extracapsular repair (Au et al. 2010). Several short- or medium-term studies have shown that aft er the initial improvement of ROM angles with surgery, stifl e joint ROM starts to decrease (Jerram et al. 2005, Bruce et al. 2007, Au et al. 2010, Gordon-Evans et al. 2013). Th e most probable cause for the decrease of ROM over time is the progression of OA, causing functional impairment due to pain and mechanical restriction of joint mobility related to fi brous tissue and new bone formation in the joint and its periarticular tissues. Another explanation might be that many dogs undergo postoperative rehabilitation, which has been shown to improve the ROM in the aff ected joint. Aft er rehabilitation is completed and without continuing rehabilitation exercises, the ROM may start decreasing. No long-term force plate studies evaluating return of limb function to the level of healthy limbs have been reported. However, force plate analysis results of the Study IV are in accordance with previous short- and medium-term studies, where the average dynamic weight bearing in the surgically treated limb was at the level of contralateral or control dog limbs 7 months aft er extracapsular repair (Budsberg et al. 1988), 4 and 12 months aft er TPLO surgery (Böddeker et al. 2012, Nelson et al. 2013), and 2-54 months aft er TTA surgery (Skinner et al. 2013). In summary, although force plate analysis provides objective and reliable information about dynamic weight bearing and limb function, it may be inadequate as the sole method for outcome assessment. Evaluation of functional outcome aft er CCL surgery should be based on several outcome assessment methods combining diff erent aspects of limb function.

6.4 Comparison of surgical techniques Intracapsular, extracapsular, and osteotomy techniques were compared. When owner evaluation with HCPI was used in Study II, no signifi cant diff erences emerged in long-term chronic pain between the techniques. Occurrence of chronic pain aft er intracapsular or extracapsular fabellotibial suture techniques has not been compared before with other techniques, but in a recent study comparing TPLO, TTA, and TightRope techniques no signifi cant diff erences were found in owner-reported VAS for long-term pain (Christopher et al. 2013). In Study II, owners’ perceptions of long-term frequency of lameness and willingness to bear weight on the surgically treated limb were similar between the diff erent technique groups. However, the owners’ opinion on duration of postoperative lameness was that it was shorter in dogs treated with osteotomy techniques than with the intracapsular technique. No signifi cant diff erence was found between the osteotomy and the extracapsular techniques. Although the owners’ ability to recollect the time of surgery and postoperative recovery must be interpreted cautiously, our fi ndings do support the general subjective opinion that weight bearing is seen earlier in dogs treated with osteotomy techniques. A comparison between osteotomy techniques and the intracapsular treatment such as the one presented in our study has not been done before, but two recent force

43 Discussion plate studies have suggested that TPLO leads to faster recovery than the extracapsular technique (Böddeker et al. 2012, Nelson et al. 2013). In Study IV, the dogs treated with the intracapsular technique had signifi cantly lower DPVF in the surgically treated limb than the dogs treated with the osteotomy techniques. No other signifi cant diff erences emerged between surgical technique groups when dynamic and static weight bearing or goniometric results were evaluated. However, when each technique group was evaluated separately, operated limbs of dogs in the intracapsular group had signifi cantly lower VI, DPVF, and SWB% values than contralateral pelvic limbs. In the osteotomy group, no signifi cant diff erences were seen in dynamic or static weight bearing between surgically treated and contralateral limbs. In goniometry, stifl e joint extension and fl exion angles were decreased relative to the contralateral limb in both the intracapsular and osteotomy groups. Unfortunately, the low number of dogs in the extracapsular group did not allow statistical comparison. In summary, although the evaluated groups were small and weight distribution from aff ected limb to contralateral limb must be taken into account, the long-term average limb function in dogs treated with osteotomy techniques may with a higher probability be indistinquishable from healthy limbs relative to dogs treated with the intracapsular technique. Parallel results were reported also in two previous short-term force plate studies. Th e limb function aft er intracapsular surgery remained inferior to limb function aft er TPLO and extracapsular surgery 6 months postoperatively (Conzemius et al. 2006), and in another experimental study limb function aft er intracapsular surgery remained inferior to preoperative function at a 20-week follow-up (Jevens et al. 1996). In Study IV, the OA severity was evaluated with radiography and stifl e joint palpation. Th e severity of radiographic OA in the surgically treated limb was lower in the osteotomy group than in the intracapsular group, and the dogs treated with osteotomy techniques had less crepitation in the surgically treated limb than dogs treated with intra- or extracapsular techniques. However, the preoperative radiographs were not available and potential initial diff erences between the surgical techniques may have aff ected these results. On the other hand, it could be speculated that the preoperative diff erences in radiographic OA between groups might level off over time. Th is has been reported in two studies, where the preoperative amount of OA was found to correlate negatively with the progression of OA postoperatively, meaning that dogs with severe pre- existing OA showed less potential for OA to worsen relative to dogs with only minor pre-existing OA (Lineberger et al. 2005, Hurley et al. 2007). As mentioned earlier, the clinical signifi cance of radiographic OA has been questioned (Gordon et al. 2003). Albeit of minor signifi cance, these fi ndings are in accordance with the results of dynamic and static weight bearing and are thus worth mentioning.

6.5 Strengths and limitaƟ ons of the study Th ese studies yielded objective evidence regarding a clinically important problem, the surgical outcome of CCL disease. A true long-term evaluation was conducted, with a time interval of 1.3 to 4.5 years between the surgery and the evaluations. Multiple approaches evaluating diff erent aspects of the surgical outcome were used. In addition to traditional subjective evaluation methods, such as visual lameness evaluation, palpatory fi ndings and measurement of radiographic OA, objective and validated assessment tools, including HCPI, force plate analysis, and goniometry, were used. As a new aspect, physiotherapeutic examination concentrating on

44 Discussion stifl e joints was included, and thus, also the functionality of the surgically treated limbs was taken into account. In addition, a healthy control group was included to obtain reference values. Th e main limitation of Studies II and IV was their retrospective nature. Th e patient material was very heterogeneous, with many possible confounding factors. As this was not a prospective study, variables such as conformation, body weight and age of the dog, duration of clinical signs before surgery, completeness of the CCL tear, meniscal pathology and treatment, and postoperative rehabilitation and medications were not standardized, and these may have aff ected the surgical outcome. In addition, it’s possible that patient material included in the study did not truly refl ect the original population of 507 dogs because factors such as outcome of surgery may have aff ected the owners’ willingness to answer the questionnaire, either positively or negatively. Also, at the time the dogs were treated, postoperative rehabilitation was not yet widely used. Th us, the number of dogs receiving postoperative physiotherapy, which is presently recommended for all CCL-defi cient dogs, was low. Although it would have been ideal to perform a prospective, randomized, controlled clinical trial with the eff ect of confounding factors minimized, the circumstances in these studies do refl ect a real-life situation, where age, weight, conformation, concurrent health issues of the patient, owner compliance, and postoperative treatment oft en vary in the patient material, and thus are representive of dogs typically predisposed to CCL injury. Th is material can well be used to evaluate general surgical outcome, but when comparing surgical techniques, the confounding factors may cause bias in the results. Although over 250 dogs were evaluated by the owners in Study II, the number of dogs in the clinical evaluation (Study IV) was unfortunately low. Th e body weight limitations set by the force plate and other concurrent orthopedic problems that were diagnosed reduced the number of dogs eligible for the force plate, SWB, and AROM evaluations. To detect statistical signifi cance in PVF between the surgically treated and control dogs using the data generated from this study, 400 dogs would have been needed (using α = 0.05 and 80% power). On the other hand, it may be speculated that even had statistical signifi cance been shown, a diff erence of 1.6% in PVF between the surgically treated and control dog limbs might not have been clinically signifi cant. Th e anatomic measurements were available only for healthy control dogs (Study I) and were not measured for the surgically treated dogs in Study IV. Th us, the possible eff ect of conformation on diff erences in force plate data between the control and study groups could not be accounted for and only standard normalization of forces and impulses to body weight was used. However, there were no signifi cant diff erences in mean body weight between control and study group dogs, and, although including many breeds, the most common breeds also in the study group were Labrador Retrievers and Rottweilers. As a compromise, the force plate results and measurements of SWB were compared with both the control dog and the contralateral limbs, and aft er taking into account the possible weight distribution, the results were in accordance with each other. In Studies I, III, and IV, the healthy control dogs were not radiographed. Orthopedic examination and previous screening for hip dysplasia were relied on to exclude any orthopedic problems. Neither the healthy control dogs nor the study group dogs were habituated in advance to the gait analysis routine, and only one force plate was used during the measurements. Th e SWB measurements were performed only on pelvic limbs, and possible compensation in weight bearing towards the thoracic limbs was ignored. In addition, the muscle mass symmetry was evaluated manually, and no quantitative measurement with a tape measure was included.

45 Conclusions 7 Conclusions

Based on these studies, the following conclusions can be drawn:

1. Th e ground reaction forces in Rottweilers diff ered signifi cantly from the data obtained for Labrador Retrievers when standard methods of normalization were used. Th e diff erences were attributable to a diff erence in conformation and body weight between the breeds.

2. A simple, cost-eff ective tool for measurement of static weight bearing in pelvic limbs was described. Based on the calculated mean ± SD diff erence between the pelvic limbs in healthy dogs, a cut-off value of 6% for diff erentiating between normal and decreased static weight bearing was set.

3. According to the Helsinki Chronic Pain Index, approximately 31% of 206 dogs had signs of chronic pain evaluated retrospectively at a mean of 2.7 years aft er CCL surgery. Th e owner assessments revealed no signifi cant diff erences between the surgical techniques in long-term outcome.

4. While I acknowledge the retrospective nature of study IV and its small sample size, some fi ndings are still worth reporting. At a mean of 2.8 years aft er CCL surgery, the average dynamic and static weight bearing in surgically treated limbs had returned to the level of healthy limbs. However, when symmetry of weight bearing was evaluated using cut-off values, approximately 30% of dogs had decreased dynamic and static weight bearing in the surgically treated limbs. Also, goniometric angles of the surgically treated limbs remained inferior to those of healthy limbs, and impairment of the active range of motion was frequently observed. Ground reaction forces may be inadequate as the sole method for evaluating functional outcome aft er CCL repair. Based on diff erences in dynamic and static weight bearing, osteotomy techniques may off er long- term limb function that is superior to that of the intracapsular technique.

46 References References

Aiken SW, KassPH, Toomps JP: Intercondylar notch width in dogs with and without cranial cruciate ligament injuries. Vet Comp Orthop Traumatol 1995;8:128-132 Arcand MA, Rhalmi S, Rivard CH: Quantifi cation of mechanoreceptors in the canine anterior cruciate ligament. Int Orthop 2000;24:272-275 Arnoczky SP, Marshall JL: Th e cruciate ligaments of the canine stifl e: an anatomical and functional analysis. Am J Vet Res 1977;38:1807-1814 Arnoczky SP, Tarvin GB, Marshall JL, et al: Th e over-the-top procedure: A technique for anterior cruciate ligament substitution in the dog. J Am Anim Hosp Assoc 1979;15:283-290 Arnoczky SP, Tarvin GB, Marshall JL: Anterior cruciate ligament replacement using patellar tendon: an evaluation of graft revascularization in the dog. J Bone Joint Surg 1982;64:217-224 Arnoczky SP, Warren RF; Ashlock MA: Replacement of the anterior cruciate ligament using a patellar tendon allograft : an experimental study. J Bone Joint Surg 1986;68:376-385 Au KK, Gordon-Evans WJ, Dunning D: Comparison of short- and long-term function and radiographic osteoarthrosis in dogs aft er postoperative physical rehabilitation and tibial plateau leveling osteotomy or lateral fabellar suture stabilization. Vet Surg 2010;39:173-180 Ballagas AJ, Montromery RD, Henderson RA, et al: Pre- and postoperative force plate analysis of dogs with experimentally transected cranial cruciate ligaments treated using tibial plateau levelling osteotomy. Vet Surg 2004;33:187-190 Bennet D, Tennant B, Lewis DG, et al: A reappraisal of anterior cruciate ligament disease in the dog. J Small Anim Pract 1988;29:275-297 Bennett DD, May C: Meniscal damage associated with cruciate disease in the dog. J Small Anim Pract 1991;32:111-117 Bertram JEA, Lee DV, Case HN, et al: Comparison of the trotting gaits of Labrador Retrievers and Greyhounds. Am J Vet Res 2000;61:832-838 Biskup JJ, Griff on DJ: Technical diffi culties during the training phase for Tightrope® and percutaneous lateral fabellar suture techniques for cranial cruciate ligament repair. Vet Surg 2014;43:347-354 Boyd DJ, Miller CW, Etue SM, et al: Radiographic and functional evaluation of dogs at least 1 year aft er tibial plateau levelling osteotomy. Can Vet J 2007;48:392-296 Brinker WO, Piermattei DL, Flo GL: Diagnosis and treatment of orthopaedic conditions of the hind limb. In Brinker WO, Piermattei DL, Flo GL, editors: Handbook of small animal orthopaedics and fracture treatment, ed 2, Philadelphia, 1990, WB Saunders Bruce WJ, Rose A, Tuke J, et al: Evaluation of the triple tibial osteotomy. A new technique for the management of the canine cruciate-defi cient stifl e. Vet Comp Orthop Traumatol 2007;20:159-168 Budsberg SC, Verstraete MS, Soutas-Little RW: Force plate analysis of the walking gait in healthy dogs. Am J Vet Res 1987;48:915-918 Budsberg SC, Verstraete MS, Soutas-Little RW, et al: Force plate analyses before and aft er stabilization of canine stifl es for cruciate injury. Am J Vet Res 1988;49:1522-1524

47 References

Cook JL, Luther JK, Beetem J, et al: Clinical comparison of a novel extracapsular stabilization procedure and tibial plateau leveling osteotomy for treatment of cranial cruciate ligament defi ciency in dogs. Vet Surg 2010a;39:315-323 Cook JL, Evans R, Conzemius MG, et al: Proposed defi nitions and criteria for reporting time frame, outcome, and complications for clinical orthopedic studies in veterinary medicine. Vet Surg 2010b;39:905-908 Corr SA, Brown C: A comparison of outcomes following tibial plateau levelling osteotomy and cranial tibial wedge osteotomy procedures. Vet Comp Orthop Traumatol 2007;20:312-319 Curtis RJ, Delee JC, Drez DJ: Reconstruction of the anterior cruciate ligament with freexe dried fascia lata allograft s in dogs. Am J Sports Med 1985;13:408-414 Damur DM, Tepic S, Montavon PM: Proximal tibial osteotomy for the repair of cranial cruciate- defi cient stifl e joints in dogs. Vet Comp Orthop Traumatol 2003;16:211-216 D’Anjou M-A, Moreau M, Troncy É, et al: Osteophytosis, subchondral bone sclerosis, joint eff usion, and soft tissue thickening in canine experimental stifl e osteoarthritis: Comparison between 1.5T magnetic resonance imaging and computed radiography. Vet Surg 2008; 37:166- 177 Davila D, Keeshen TP, Evans RB, et al: Comparison of the analgesic effi cacy of perioperative fi rocoxib and tramadol administration in dogs undergoing tibial plateau leveling osteotomy. J Am Vet Med Assoc 2013; 243:225-231 DeAngelis M, Lau RE: A lateral retinacular imbrication technique for the surgical correction of anterior cruciate ligament rupture in the dog. J Am Vet Med Assoc 1970;157:79-84 deMedeiros M, Sánchez Bustinduy M, Radke H, et al: Early kinematic outcome aft er treatment of cranial cruciate ligament rupture by tibial plateau levelling osteotomy in the dog. Vet Comp Orthop Traumatol 2011;24:178-184 Denny HR, Goodship AE: Replacement of the anterior cruciate ligament with carbon fi bre in the dog. J Small Anim Pract 1980;21:279-286 Denny HR, Barr AES: An evaluation of two ”over-the-top” techniques for anterior cruciate ligament replacement in the dog. J Small Anim Pract 1984;25:759-769 Denny HR, Barr ARS: A further evaluation of the “over-the-top” technique for anterior cruciate ligament replacement in the dog. J Small Anim Pract 1987;28:681-686 deRooster H, vanBree H: Use of compression stress radiography for the detection of partial tears of the canine cranial cruciate ligament. J Small Anim Pract 1999;40:573-576 Doom M, deBruin T, deRooster H, et al: Immunopathological mechanisms in dogs with rupture of the cranial cruciate ligament. Vet Immunol Immunopathol 2008;125:143-161 Doverspike M, Vasseur PB, Harb MF, et al: Contralateral cranial cruciate ligament rupture: Incidence in 114 dogs. J Am Anim Hosp Assoc 1993; 29:167-170 Duerr FM, Martin KW, Rishniw M, et al: Treatment of canine cranial cruciate ligament disease. A survey of ACVS Diplomates and primary care veterinarians. Vet Comp Orthop Traumatol 2014;6:1-6 Duval JM, Budsberg SC, Flo GL, et al: Breed, sex, and body weight as risk factors for rupture of the cranial cruciate ligament in young dogs. J Am Vet Med Assoc 1999;215:811-814

48 References

Budsberg SC, Jevens DJ, Brown J, et al: Evaluation of limb symmetry indices, using ground reaction forces in healthy dogs. Am J Vet Res 1993;54:1569-1574 Budsberg SC, Rytz U, Johnston SA: Eff ects of acceleration on ground reaction forces collected in healthy dogs at a trot. Vet Comp Orthop Traumatol 1999;12:15-19 Buote N, Fusco J, Radasch R: Age, tibial plateau angle, sex, and weight as risk factors for contralateral rupture of the cranial cruciate ligament in Labradors. Vet Surg 2009;38:481-489 Butler DL, Hulse DA, Kay MD, et al: Biomechanics of cranial cruciate reconstruction in the dog. Part II: mechanical properties. Vet Surg 1983;12:113-118 Böddeker J, Drüen S, Meyer-Lindenberg A, et al: Computer-assisted gait analysis of the dog: Comparison of two surgical techniques for the ruptured cranial cruciate ligament. Ve Comp Orthop Traumatol 2012;25:11-21 Carey K, Aiken SW, DiResta GR, et al: Radiographic and clinical changes of the patellar tendon aft er tibial plateau leveling osteotomy. Vet Comp Orthop Traumatol 2005;18:235-242 Carpenter DH, Cooper RC: Mini review of canine stifl e joint anatomy. Anat Histol Embryol 2000;29:321-329 Casale SA, McCarthy RJ: Complications associated with lateral fabellotibial suture surgery for cranial cruciate ligament injury in dogs: 363 cases (1997-2005). J Am Vet Med Assoc 2009;234:229-235 Case JB, Hulse D, Kerwin SC, et al: Meniscal injury following initial cranial cruciate ligament stabilization surgery in 26 dogs (29 stifl es). Vet Comp Orthop Traumatol 2008;21:365-367 Christopher SA, Beetem J, Cook JL: Comparison of long-term outcomes associated with three surgical techniques for treatment of cranial cruciate ligament disease in dogs. Vet Surg 2013;42:329-334 Cimino Brown D, Boston RC, Coyne JC, et al: Development and psychometric testing of an instrument designed to measure chronic pain in dogs with osteoarthritis. Am J Vet Res 2007;68:631-637 Cimino Brown D: Th e canine orthopedic index. Step 1: Devising the items. Vet Surg 2014a;43:232- 240 Cimino Brown D: Th e canine orthopedic index. Step 2: Psychometric testing. Vet Surg 2014b;43:241- 246 Cimino Brown D: Th e canine orthopedic index. Step 3: Responsiveness testing. Vet Surg 2014c;43:247- 254 Comerford EJ, Tarlton JF, Avery NC, et al: Distal femoral intercondylar notch dimensions and their relationship to composition and metabolism of the canine anterior cruciate ligament. Osteoarthritis and Cartilage 2006;14:273-278 Comerford EJ, Smith K, Hayashi K: Update on the aetiopathogenesis of canine cranial cruciate ligament disease. Vet Comp Orthop Traumatol 2011;24:91-98 Comerford E, Forster K, Gorton K, et al: Management of cranial cruciate ligament rupture in small dogs: A questionnaire study. Vet Comp Orthop Traumatol 2013;26:493-497 Conzemius MG, Evans RB, Besancon MF, et al: Eff ect of surgical technique on limb function aft er surgery for rupture of the cranial cruciate ligament in dogs. Vet Surg 2005;226:232-236

49 References

Elkins AD, Pechman R, Kearney MT, et al: A retrospective study evaluating the degree of degenerative joint disese in the stifl e joint of dogs following surgical repairof anterior cruciate ligament rupture. J Am Anim Hosp Assoc 1991;27:533-540 Evans HE: Arthrology. In Evans HE, editor: Miller’s anatomy of the dog, ed 3, Philadelphia, 1993, WB Saunders Evans R, Horstman C, Conzemius M: Accuracy and optimization of force platform gait analysis in Labradors with cranial cruciate disease evaluated at a walking gait. Vet Surg 2005;34:445- 449 Fitzpatrick N, Solano MA: Predictive variables for complications aft er TPLO with stifl e inspection by arthrotomy in 1000 consecutive dogs. Vet Surg 2010;39:460-474 Flo GL: Modifi cation of the lateral retinacular imbrications technique for stabilizing cruciate ligament injuries. J Am Anim Hosp Assoc 1975;11:570-576 Flo GL: Meniscal injuries. Vet Clin North Am 1993;23:831-843 Frey TN, Hoelzler MG, Scavelli TD, et al: Risk factors for surgical site infection-infl ammation in dogs undergoing surgery for rupture of the cranial cruciate ligament: 902 cases (2005-2006). J Am Vet Med Assoc 2010;236:88-94 Gambardella PC, Wallace LJ, Cassidy F: Lateral suture technique for management of anterior cruciate ligament rupture in dogs: A retrospective study. J Am Anim Hosp Assoc 1981;17:33- 38 Gatineau M, Dupuis J, Planté, et al: Retrospective study of 476 tibial plateau levelling osteotomy procedures. Rate of subsequent “pivot shift ”, meniscal tear and other complications. Vet Comp Orthop Traumatol 2011;24:333-341 Gibbons SE, Macias C, Tonzing MA, et al: Patellar luxation in 70 large breed dogs. J Small Anim Pract 2006;47:3-9 Gordon WJ, Conzemius MG, Riedesel E, et al: Th e relationship between limb function and radiographic osteoarthrosis in dogs with stifl e osteoarthrosis. Vet Surg 2003;32:451-454 Gordon-Evans WJ, Griff on DJ, Bubb C, et al: Comparison of lateral fabellar suture and tibial plateau leveling osteotomy techniques for treatment of dogs with cranial cruciate ligament disease. J Am Vet Med Assoc 2013;23:675-680 Guénégo L, Zahra A, Madelénat A, et al: Cranial cruciate ligament rupture in large and giant dogs. A retrospective evaluation of a modifi ed lateral extracapsular stabilization. Ve Comp Orthop Traumatol 2007;20:43-50 Guerrero TG, Pozzi A, Dunbar N, et al: Eff ect of tibial tuberosity advancement on the contact mechanics and the alignment of the patellofemoral and femorotibial joints. Vet Surg 2011;40:839-848 Gupta BN, Brinker WO: Anterior cruciate ligament prosthesis in the dog. J Am Vet Med Assoc 1969;154:1057-1061 Hayashi K, Frank JD, Dubinsky C, et al: Histologic changes in ruptured canine cranial cruciate ligament. Vet Surg 2003;32:269-277 Hayes GM, Granger N, Langley-Hobbs SJ, et al: Abnormal refl ex activation of hamstring muscles in dogs with cranial cruciate ligament rupture. Vet Journal 2013;196:345-350

50 References

Heff ron LE, Campbell JR: Morphology, histology and functional anatomy of the canine cranial cruciate ligament. Vet Rec 1978;102:280-283 Heikkilä HM, Hielm-Björkman AK, Morelius M, et al: Intra-articular botulinum toxin A for the treatment of osteoarthritic joint pain in dogs: A randomized, double-blinded, placebo- controlled clinical trial. Vet Journal 2014;200:162-169 Hercock CA, Pinchbeck G, Giejda A, et al: Validation of a client-based clinical metrology instrument for the evaluation of canine elbow osteoarthritis. J Small Anim Pract 2009;50:266- 271 Herzog W, Nigg BM, Read LJ, et al: Asymmetries in ground reaction force patterns in normal human gait. Med Sci Sport Excerc 1989;21:110-114 Hielm-Björkman AK, Kuusela E, Liman A, et al: Evaluation of methods for assessment of pain associated with chronic osteoarthritis in dogs. J Am Vet Med Assoc 2003;222:1552-1558 Hielm-Björkman AK, Rita H, Tulamo R-M: Psychometric testing of the Helsinki chronic pain index by completion of a questionnaire in Finnish by owners of dogs with chronic signs of pain caused by osteoarthritis. Am J Vet Res 2009;70:727-734 Hielm-Björkman AK, Kapatkin AS, Rita HJ: Reliability and validity of a visual analogue scale used by owners to measure chronic pain attributable to osteoarthritis in their dogs. Am J Vet Res 2011;72:601-607 Hielm-Björkman AK, Roine J, Elo K, et al: An un-commissioned randomized, placebo-controlled double-blind study to test the eff ect of deep sea fi sh oil as a pain reliever for dogs suff ering from canine OA. BMC Vet Res 2012;8:157 Hildreth BE, Marcellin-Little DJ, Roe SC, et al: In vitro evaluation of fi ve canine tibial plateau levelling methods. Am J Vet Res 2006;67:693-700 Hoelzler MG, Millis DL, Francis DA, et al: Results of arthroscopic versus open arthrotomy for surgical management of cranial cruciate ligament defi ciency in dogs. Vet Surg 2004;33:146-153 Hoff mann DE, Miller JM, Ober CP, et al: Tibial tuberosity advancement in 65 canine stifl es. Vet Comp Orthop Traumatol 2006;19:219-227 Horstman CL, Conzemius MG, Evans R, et al: Assessing the effi cacy of perioperative oral carprofen aft er cranial cruciate surgery using noninvasive, objective pressure platform gait analysis. Vet Surg 2004;33:286-292 Hospodar MSJ, Miller MD: Controversies in ACL reconstruction: Bone-patellar tendon-bone anterior cruciate ligament reconstruction remains the gold standard. Sports Med Arthrosc Rev 2009;17:242-246 Hudson JT, Slater MR, Taylor L, et al: Assessing repeatability and validity of a visual analoque scale questionnaire for use in assessing pain and lameness in dogs. Am J Vet Res 2004;65:1634- 1643 Hulse DA, Michaleson F, Johnson C, et al: A technique for reconstruction of the anterior cruciate ligament in the dog: preliminary report. Vet Surg 1980;9:135-140 Hulse DA, Butler DL, Kay MD, et al: Biomechanics of cranial cruciate reconstruction in the dog. Part I: in vitro laxity testing. Vet Surg 1983;12:109-112

51 References

Hulse D, Hyman W, Beale B, et al: Determination of isometric points for placement of a lateral suture in treatment of the cranial cruciate ligament defi cient stifl e. Vet Comp Orthop Traumatol 2010;23:163-167 Hurley CR, Hammer DL, Shott S: Progression of radiographic evidence of osteoarthritis following tibial plateau leveling osteotomy in dogs with cranial cruciate ligament rupture: 295 cases (2001-2005). J Am Vet Med Assoc. 2007;230:1674-1679 Hyytiäinen HK, Mölsä SH, Junnila JT, et al: Use of bathroom scales in measuring asymmetry of hindlimb static weight bearing in dogs with osteoarthritis. Ve Comp Orthop Traumatol 2012;25:390-396 Innes JF, Barr ARS: Clinical natural history of the postsurgical cruciate defi cient canine stifl e joint: year 1. J Small Anim Pract 1998a;39:325-332 Innes JF, Barr ARS: Can owners assess outcome following treatment of canine cruciate ligament defi cinency? J Small Anim Pract 1998b;39:373-378 Innes JF, Bacon D, Lynch C, et al: Long-term outcome of surgery for dogs with cranial cruciate ligament defi ciency. Vet Rec 2000;147:325-327 Innes JF, Costello M, Barr FJ, et al: Radiographic progression of osteoarthritis of the canine stifl e joint: A prospective study. Vet Radiol Ultrasound 2004;45:143-148 Jandi AS, Schulman AJ: Incidence of motion loss of the stifl e joint in dogs with naturally occurring cranial cruciate ligament ruptures surgically treated with tibial plateau leveling osteotomy: Longitudinal clinical study of 412 dogs. Vet Surg 2007;36:114-121 Jerram RM, Walker AM, Warman CGA: Proximal tibial intraarticular ostectomy for treatment of canine cranial cruciate ligament injury. Vet Surg 2005;34:196-205 Jevens DJ, Hauptman JG, DeCamp CE, et al: Contributions to variance in force-plate analysis of gait in dogs. Am J Vet Res 1993;54:612-615 Jevens DJ, DeCamp CE, Hauptman J, et al: Use of force plate analysis of gait to compare two surgical techniques for treatment of cranial cruciate ligament rupture in dogs. Am J Vet Res 1996;57:389-393 Johnson JA, Austin C, Breur GJ: Incidence of canine appendicular musculoskeletal disorders in 16 veterinary teaching hospitals from 1980 to 1989. Vet Comp Orthop Traumatol 1994;7:56-59 Johnson JM, Johnson AL: Cranial cruciate ligament rupture: pathogenesis, diagnosis, and post- operative rehabilitation. Vet Clin North Am 1993;23:717-732 Johnson SG, Hulse DA, Hogan HA, et al: System behavior of commonly used cranial cruciate ligament reconstruction graft s. Vet Surg 1989;18:459-465 Kergosien DH, Barnhart MD, Kees CE, et al: Radiographic and clinical changes of the tibial tuberosity aft er tibial plateau leveling osteotomy. Vet Surg 2004;33:468-474 Kim SE, Pozzi A, Banks SA, et al: Eff ect of tibial plateau leveling osteotomy on femorotibial contact mechanics and stifl e kinematics. Vet Surg 2009a;38:23-32 Kim SE, Pozzi A, Banks SA et al: Eff ect of tibial tuberosity advancement on femorotibial contact mechanics and stifl e kinematics. Vet Surg 2009b;38:33-39

52 References

Kim SE, Lewis DD, Pozzi A: Eff ect of tibial plateau leveling osteotomy on femorotibial subluxation: In vivo analysis during standing. Vet Surg 2012;41:465-470 Kowaleski MP, Boudrieau RJ, Pozzi A: Stifl e joint. In Tobias KM, Johnston SA, editors: Veterinary surgery: small animal, St Louis, 2012, Elsevier Krause WR, Pope MH, Johnson RJ, et al: Mechanical changes in the knee aft er meniscectomy. J Bone Joint Surg 1976;58:599-603 Krotscheck U, Todhunter RJ, Nelson SA: Precision and accuracy of ground reaction force normalization in a heterogeneous population of dogs. Vet Surg 2014;43:437-445 Kühn K, Ohlerth S, Makara M, et al: Radiographic and ultrasonographic evaluation of the patellar ligament following tibial tuberosity advancement. Vet Radiol & Ultrasound 2011;52:466-471 Kunkel KAR, Basinger RR, Suber JT, et al: Evaluation of a transcondylar toggle system for stabilization of the cranial cruciate defi cient stifl e in small dogs and cats. Vet Surg 2009;38:975- 982 Lafaver S, Miller NA, Stubbs WP, et al: Tibial tuberosity advancement for stabilization of the canine cranial cruciate ligament-defi cient stifl e joint: Surgical technique, early results, and complications in 101 dogs. Vet Surg 2007;36:573-586 Laitinen O: Biodegradable polylactide implant as an augmentation device in anterior cruciate ligament repair or reconstruction. An experimental and clinical study, Helsinki, 1994 (thesis) Lamont LA, Tranquilli WJ, Grimm KA. Physiology of pain. Vet Clin North Am Small Anim Pract 2000;30:703-728 Lascelles BDX, Roe SC, Smith E, et al: Evaluation of a pressure walkway system for measurement of vertical limb forces in clinically normal dogs. Am J Vet Res 2006;67:277-282 Lascelles BDX, Freire M, Roe SC, et al: Evaluation of functional outcome aft er BFX® total hip replacement using a pressure sensitive walkway. Vet Surg 2010;39:71-77 Lee JY, Kim G, Kim J-H, Choi SH: Kinematic gait analysis of the hind limb aft er tibial plateau levelling osteotomy and cranial tibial wedge osteotomy in ten dogs. J Vet Med 2007;54:579-584 Legnani C, Ventura A, Terzaghi C, et al: Anterior cruciate ligament reconstruction with synthetic graft s. A review of literature. Int Orthop 2010;34:465-471 Leighton RL: Preferred method of repair of cranial cruciate ligament rupture in dogs: A survey of ACVS Diplomates specializing in canine orthopedics. Vet Surg 1999;28:194 Levy IM, Torzilli PA, Warren RF: Th e eff ect of medial meniscectomy on anterior-posterior motion of the knee. J Bone Joint Surg 1982;64(6),883-888 Lewis BA, Allen DA, Henrikson TD, et al: Computed tomographic evaluation of the canine intercondylar notch in normal and cruciate defi cient stifl es. Vet Comp Orthop Traumatol 2008;21:119-124¨

53 References

Lineberger JA, Allen DA, Wilson ER, et al. Comparison of radiographic arthritic changes associated with two variations of tibial plateau leveling osteotomy. A Retrospective clinical study. Vet Comp Orthop Traumatol 2005;18:13-17 Luther JK, Cook CR, Cook JL: Meniscal release in cruciate ligament intact stifl es causes lameness and medial compartment cartilage pathology in dogs 12 weeks postoperatively. Vet Surg 2009;38:520- 529 Marino DJ, Louhglin CA: Diagnostic imaging of the canine stifl e: A review. Vet Surg 2010; 39:284-295 Marshall JL, Olsson S: Instability of the knee. A long-term experimental study in dogs. J Bone Joint Surg 1971;53:1561-1570 Marsolais GS, McLean S, Derrick T, et al: Kinematic analysis of the hind limb during swimming and walking in healthy dogs and dogs with surgically corrected cranial cruciate ligament rupture. J Am Vet Med Assoc 2003;222:739-743 Mattern KL, Berry CR, Peck JN, et al: Radiographic and ultrasonographic evaluation of the patellar ligament following tibial plateau leveling osteotomy. Vet Radiol & Ultrasound 2006;47:185-191 McLaughlin RM: Kinetic and kinematic gait analysis in dogs. Vet Clin North Am Small Anim Pract 2001:31:193-201 McMaster WC: A histologic assessment of canine anterior cruciate substitution with bovine xenograft . Clin Orthop 1994;196:196-201 Moeller EM, Allen DA, Wilson ER, et al: Long-term outcomes of thigh circumference, stifl e range- of-motion, and lameness aft er unilateral tibial plateau levelling osteotomy. Vet Comp Orthop Traumatol 2010;23:37-42 Moles AD, Hill TP, Glyde M: Triple tibial osteotomy for treatment of the canine cranial cruciate ligament-defi cient stifl e joint. Surgical fi ndings and postoperative complications in 97 stifl es. Vet Comp Orthop Traumatol 2009;22:473-478 Monahan JJ, Grigg P, Pappas AM, et al: In vivo strain patterns in the four major canine knee ligaments. J Orthop Res 1984;2:408-418 Monk ML, Preston CA, McGowan CM: Eff ects of early intensive postoperative physiotherapy on limb function aft er tibial plateau leveling osteotomy in dogs with defi ciency of the cranial cruciate ligament. Am J Vet Res 2006;67:529-536 Montavon PM, Damur DM, Tepic S: Advancement of the tibial tuberosity for the treatment of cranial cruciate defi cient canine stifl e. Proceedings, 1st World Orthopaedic Veterinary Congress, Munich, Germany 2002:152 (Abstract) Moore KW, Read RA: Rupture of cranial cruciate ligament in dogs: part I. Compend Contin Educ Pract Vet 1996;18:223-234 Morgan JP, Voss K, Damur DM, et al: Correlation of radiographic changes aft er tibial tuberosity advancement in dogs with cranial cruciate-defi cient stifl es with functional outcome. Vet Surg 2010;39:425-432

54 References

Morris E, Lipowitz AJ: Comparison of tibial plateau angles in dogs with and without cranial cruciate ligament injuries. J Am Vet Med Assoc 2001;218:363-366 Mostafa AA, Griff on DJ, Th omas MW, et al: Morphometric characteristics of the pelvic limbs of Labrador Retrievers with and without cranial cruciate ligament defi ciency. Am J Vet Res 2009;70:498-507 Nelson SA, Krotscheck U, Rawlinson J, et al: Long-term functional outcome of tibial plateau leveling osteotomy versus extracapsular repair in a heterogeneous population of dogs. Vet Surg 2013;42:38- 50 Oxley B, Gemmill TJ, Renwick AR, et al: Comparison of complication rates and clinical outcome between tibial plateau leveling osteotomy and a modifi ed cranial closing wedge osteotomy for treatment of cranial cruciate ligament disease in dogs. Vet Surg 2013;42:739-750 Paatsama S: Ligament injuries in the canine stifl e joint: a clinical and experimental study, Helsinki, 1952 (thesis) Pacchiana PD, Morris E, Gillings SL, et al: Surgical and postoperative complications associated with tibial plateau leveling osteotomy in dogs with cranial cruciate ligament rupture: 397 cases (1998- 2001). J Am Vet Med Assoc 2003;222:184-193 Park JP, Grana WA, Chitwood JS: A high-strength Dacron augmentation for cruciate ligament reconstruction. A two-year canine study. Clin Orthop 1985:175-185 Pettitt R, Cripps P, Baker M, et al: Radiographic and ultrasonographic changes of the patellar ligament following tibial tuberosity advancement in 25 dogs. Vet Comp Orthop Traumatol 2014;3:216-221 Phelps HA, Ramos V, Shires PK, et al: Th e eff ect of measurement method on static weight distribution to all legs in dogs using the Quadruped Biofeedback System. Vet Comp Orthop Traumatol 2007;20:108-112 Piermattei DL, Flo GL, DeCamp CE: Th e stifl e joint. In Piermattei DL, Flo GL, DeCamp CE, editors: Brinker, Piermattei and Flo’s handbook of small animal orthopedics and fracture repair, ed 4, St Louis, 2006, Elsevier Pozzi A, Litsky AS, Field J, et al: Pressure distributions on the medial tibial plateau aft er medial meniscal surgery and tibia plateau levelling osteotomy in dogs. Vet Comp Orthop Traumatol 2008; 21:8-14 Pozzi A, Kim SE, Lewis DD: Eff ect of transection of the caudal menisco-tibial ligament on medial femorotibial contact mechanics. Vet Surg 2010;39:489-495 Priddy NH, Tomlinson JL, Dodam JR, et al: Complications with and owner assessment of the outcome of tibial plateau leveling osteotomy for treatment of cranial cruciate ligament rupture in dogs: 193 cases (1997-2001). J Am Vet Med Assoc 2003;222:1726-1732 Quinn MM, Keuler NS, Lu Y, et al: Evaluation of agreement between numerical rating scales, visual analogue scoring scales and force plate gait analysis in dogs. Vet Surg 2007;36:360-367 Ralphs SC, Whitney WO: Arthroscopic evaluation of menisci in dogs with cranial cruciate ligament injuries: 100 cases (1999-2000). J Am Vet Med Assoc 2002;221:1601-1604 Rayward RM, Th omson DG, Davies JV, et al: Progression of osteoarthritis following TPLO surgery: a prospective radiographic study of 40 dogs. J Small Anim Pract 2004;45:92-97

55 References

Reif U, Hulse DA, Hauptman JG: Eff ect of tibial plateau leveling on stability of the canine cranial cruciate-defi cient stifl e joint: an in vitro study. Vet Surg 2002;31:147-154 Reif U, Probst CW: Comparison of tibial plateau angles in normal and cranial cruciate defi cient stifl es of Labrador Retrievers. Vet Surg 2003;32:385-389 Renwick AIC, McKee WM, Emmerson TD, et al: Preliminary experiences of the triple tibial osteotomy procedure:tibial morphology and complications. J Small Anim Pract 2009;50:212- 221 Riggs CM, DeCamp CE, Soutas-Little RW: Eff ects of subject velocity on force plate-measured ground reaction forces in healthy Greyhounds at the trot. Am J Vet Res 1993;54:1523-1526 Ritzo ME, Ritzo BA, Siddens AD, et al: Incidence and type of meniscal injury and associated long-term clinical outcomes in dogs treated surgically for cranial cruciate ligament disease. Vet Surg 2014;43:952-958 Roe SC, Kue J, Gemma J: Isometry of potential suture attachment sites for the cranial cruciate ligament defi cient canine stifl e. Vet Comp Orthop Traumatol 2008;21:215-220 Rubin RM, Marshall JL, Wang J: Prevention of knee instability: Experimental model for prosthetic anterior cruciate ligament. Clin Orthop 1975;113:212-235 Rumph PF, Kincaid SA, Baird DK, et al: Vertical ground reaction force distribution during experimentally induced acute synovitis in dogs. Am J Vet Res 1993;54:365-369 Rumph PF, Lander JE, Kincaid SA, et al: Ground reaction force profi les from force platform gait analyses of clinically normal mesomorphic dogs at the trot. Am J Vet Res 1994;55:756-761 Sanchez-Bustinduy M, de Medeiros MA, Radke H, et al: Comparison of kinematic variables in defi ning lameness caused by naturally occurring rupture of the cranial cruciate ligament in dogs. Vet Surg 2010;39:523-530 Shires PK, Hulse DA, Liu W: Th e under-and-over fascial replacement technique for anterior cruciate ligament rupture in dogs: a retrospective study. J Am Anim Hosp Assoc 1984;20:69- Siemering GH: Arthroscopy of dogs. J Am Vet Med Assoc 1978;172:575-577 Skinner OT, Kim SE, Lewis DD, et al: In vivo femorotibial subluxation during weight-bearing and clinical outcome following tibial tuberosity advancement for cranial cruciate ligament insuffi ciency in dogs. Vet Journal 2013;196:86-91 Slocum B, Devine T: Cranial tibial thrust: A primary force in the canine stifl e. J Am Vet Med Assoc 1983;183:456-459 Slocum B, Devine T: Cranial tibial wedge osteotomy: a technique for eliminating cranial tibial thrust in cranial cruciate ligament repair. J Am Vet Med Assoc 1984;184:564-569 Slocum B, Devine Slocum T: Tibial plateau leveling osteotomy for repair of cranial cruciate ligament rupture in the canine. Vet Clin North Am Small Anim Pract 1993;23:777-795 Slocum B, Devine Slocum T: Meniscal release. In Bojrab MJ, editor: Current techniques in small animal surgery, ed 4, Baltimore, 1998, Williams and Wilkins Smith GK, Torg JS: Fibular head transposition for repair of cruciate-defi cient stifl e in the dog. J Am Vet Med Assoc. 1985;187:375-383

56 References

Solomonow M, Baratta R, Zhou BH, et al: Th e synergistic action of the anterior cruciate ligament and thigh muscles in maintaining joint stability. Am J Sports Med 1987;15:207-213 Stauff er KD, Tuttle TA, Elkins AD, et al: Complications associated with tibial plateau leveling osteotomies (2001-2003). J Am Anim Hosp Assoc 2006;42:44-50 Stein S, Schmoekel H: Short-term and eight to 12 months results of a tibial tuberosity advancement as treatment of canine cranial cruciate ligament damage. J Small Anim Pract 2008;49:398-404 Steinberg EJ, Prata RG, Palazzini K, et al: Tibial tuberosity advancement for treatment of CrCL injury: Complications and owner satisfaction. J Am Anim Hosp Assoc 2011;47:250-257 Talaat MB, Kowaleski MP, Boudrieau RJ: Combination tibial plateau leveling osteotomy and cranial closing wedge osteotomy of the tibia for the treatment of cranial cruciate ligament- defi cient stifl es with excessive tibial plateau angle. Vet Surg 2006;35:729-739 Tashman S, Anderst W, Kolowich P, et al: Kinematics of the ACL-defi cient canine stifl e during gait: serial changes over two years. J Orthop Res 2004;22:931-941 Tepic S, Damur DM, Montavon PM: Biomechanisc of the stifl e joint. Proceedings, 1st World Orthopaedic Veterinary Congress, Munich, Germany, 2002:189-190 (abstract) Tonks CA, Lewis DD, Pozzi A: A review of extra-articular prosthetic stabilization of the cranial cruciate ligament-defi cient stifl e. Vet Comp Orthop Traumatol 2011;24:167-177 Vasseur PB: Clinical results following nonoperative management for rupture of the cranial cruciate ligament in dogs. Vet Surg 1984;13:243-246 Vasseur PB, Pool RR, Arnoczky SP, et al: Correlative biomechanical and histologic study of the cranial cruciate ligament in dogs. Am J Vet Res 1985;46:1842-1853 Vasseur PB, Berry CR: Progression of stifl e osteoarthrosis following reconstruction of the cranial cruciate ligament in 21 dogs. J Am Anim Hosp Assoc 1992;28:129-135 Vezzoni A, Bohorquel Vanelli A, Modenato M, et al: Proximal tibial epiphysiodesis to reduce tibial plateau slope in young dogs with cranial cruciate ligament defi cient stifl e. Vet Comp Orthop Traumatol 2008;21:343-348 Voss K, Imhof J, Kaestner S: Force plate gait analysis at the walk and trot in dogs with low-grade hindlimb lameness. Vet Comp Orthop Traumatol 2007;20:299-304 Voss K, Damur DM, Guerrero T, et al: Force plate gait analysis to assess limb function aft er tibial tuberosity advancement in dogs with cranial cruciate ligament disease. Vet Comp Orthop Traumatol 2008;21:243-249 Voss K, Galeandro L, Wiestner T, et al: Relationship of body weight, body size, subject velocity, and vertical ground reaction forces in trotting dogs. Vet Surg 2010;39:863-869 Voss K, Wiestner T, Galeandro L, et al: Eff ect of dog breed and body conformation on vertical ground reaction forces, impulses, and stance times. Vet Comp Orthop Traumatol 2011;24:106- 112 Walker PS, Erkman MJ: Th e role of the menisci in force transmission across the knee. Clin Orthop 1975;184-192

57 References

Walton MB, Cowderoy E, Lascelles D, et al: Evaluation of construct and criterion validity for the “Liverpool osteoarthritis in dogs” (LOAD) clinical metrology instrument and comparison to two other instruments. PLoS ONE 2013;8(3):e58125 doi:10.1371/journal.pone.0058125 Warzee CC, Dejardin LM, Arnozcky SP, et al: Eff ect of tibial plateau leveling on cranial and caudal tibial thrusts in canine cranial cruciate-defi cient stifl es: an in vitro experimental study. Vet Surg 2001;30:278-286 Wernham BGJ, Trumpatori B, Hash J, et al: Dose reduction of meloxicam in dogs with osteoarthritis-associated pain and impaired mobility. J Vet Intern Med 2011;25:1298-1305 Whitehair JG, Vasseur PB, Willits NH: Epidemiology of cranial cruciate ligament rupture in dogs. J Am Vet Med Assoc 1993;203:1016-1019 Whitney WO: Arthroscopically assisted surgery of the stifl e joint. In Beale BS, Hulse DA, Schulz KS, Whitney WO, editors: Small animal arthroscopy, Philadelphia, 2003, Elsevier Widmer WR, Buckwalter KA, Braunstein EM, et al: Radiographic and magnetic resonance imaging of the stifl e joint in experimental osteoarthritis of dogs. Vet Radiol Ultrasound 1994;35:371 Wilke VL, Conzemius MG, Besancon MF, et al: Comparison of tibial plateau angle between clinically normal Greyhounds and Labrador Retrievers with and without rupture of the cranial cruciate ligament. J Am Vet Med Assoc 2002;221:1426-1429 Wilke VL, Conzemius MG, Kinghorn BP, et al: Inheritance of rupture of the cranial cruciate ligament in Newfoundlands. J Am Vet Med Assoc 2006;228:61-64 Wiseman-Orr ML, Scott EM, Reid J, et al: Validation of a structured questionnaire as a instrument to measure chronic pain in dogs on the basis of eff ects on health-related quality of life. Am J Vet Res 2006;67:1826-1836 Wolf RE, Scavelli TD, Hoelzler MG, et al: Surgical and postoperative complications associated with tibial tuberosity advancement for cranial cruciate ligament rupture in dogs: 458 cases (2007-2009). J Am Vet Med Assoc 2012;240:1481-1487 Wucherer KL, Conzemius MG, Evans R, et al: Short-term and long-term outcomes for overweight dogs with cranial cruciate ligament rupture treated surgically or nonsurgically. J Am Vet Med Assoc 2013;242:1364-1372

58