may occur (1 4). Objective measurement techniques for gait analysis include force platforms, baropodometric systems, kinematic

Original Research Kinematic analysis of Labrador Retrievers and Rottweilers trotting on a treadmill F. S. Agostinho 1 ; S. C. Rahal 1 ; N. S. M. L. Miqueleto 1 ; M. R. Verdugo 1 ; L. R. Inamassu 1 ; A. O. El-Warrak 2 1 Department of Veterinary Surgery and Anesthesiology, School of Veterinary Medicine and Animal Science, Univ. Estadual Paulista (UNESP), Botucatu, SP, Brazil; 2 Department of Clinical Sciences, Faculty of Veterinary Medicine, University of Montreal, Canada Schattauer 2011 1 Keywords Gait analysis, dog, kinematic Summary Objectives: The purpose of the study was to evaluate kinematic patterns in clinically normal Labrador and Rottweiler dogs trotting on a treadmill at a constant velocity. Methods: Ten Labrador Retrievers aged from 2.2 to 5.1 years, and 10 Rottweilers aged from two to 5.9 years were used. A three-dimensional capture system was used to perform analysis of joint kinematics. Kinematic data were collected by use of a triple-camera system. The kinematic study was performed first on the right side of the dog, and then on the left side. Data were analysed by use of a motion-analysis program. Flexion and extension joint angles, angular velocity and angular acceleration were determined for the shoulder, elbow, carpal, hip, stifle, and tarsal joints. Results: Within each group, the differences Correspondence to: Sheila C. Rahal, MS, PhD Department of Veterinary Surgery and Anesthesiology Univ Estadual Paulista (UNESP) School of Veterinary Medicine and Animal Science Botucatu, PO Box 560 Rubião Júnior s/n, Botucatu (SP) Brazil 18618 000 Phone: +55 14 3811 6054 Fax: +55 14 3815 2343 E-mail: sheilacr@fmvz.unesp.br Introduction Subjective gait evaluations are commonly used in diagnosing clinical lameness in dogs. However, human perceptive skills are not as efficient and precise as objective gait evaluations for measuring gait patterns, and inter- and intra-observer variability between the right and left limbs in all variables were not significant. Significant differences occurred between Labradors and Rottweilers in the following categories: angular displacement and minimum angular acceleration of the stifle (Rottweiler >Labrador); angular displacement and maximum angular velocity of the tarsus (Rottweiler >Labrador); minimum angular velocity of the shoulder (Labrador >Rottweiler); angular displacement, maximum angular acceleration, maximum angular velocity, and minimum angular velocity of the elbow (Labrador>Rottweiler); and maximum angle and maximum angular velocity of the carpus (Labrador>Rottweiler). Clinical significance: Both breeds had similar kinematic patterns, but there were magnitude differences, especially of the elbow and stifle joints. Therefore, each breed should have a specific database. Vet Comp Orthop Traumatol 2011; 24: doi:10.3415/vcot-10-03-0039 Received: March 15, 2010 Accepted: January 5, 2011 Pre-published online: February 16, 2011 may occur (1 4). Objective measurement techniques for gait analysis include force platforms, baropodometric systems, kinematic systems, electromyography, and electrogoniometry (1 3, 5, 6). Force plate and computer-assisted kinematic are the most often used tools for the investigation of the kinetic and kinematic analyses in veterinary medicine (2, 5). In several studies, kinematic analysis has been performed with force plate analysis or pressure sensing walkway systems where the characteristics of limb movement are correlated with ground-reaction force measurements (5, 7 10). Non-invasive kinematic analysis identifies movement irrespective of the influence of weight and force or, more specifically, it describes the geometry of motion in terms of displacement, velocity and acceleration (4, 6). There are different capabilities of capture systems, but a three-dimensional (3D) analysis enables the acquisition of more accurate data when compared with a two-dimensional (2D) analysis (4, 6, 11, 12). Kinematic gait analysis provides information on subject velocity, segmental velocities of each portion of the limb, joint angles (flexion and extension data), angular velocities, temporal data, frequency, and stride length (5). Sometimes, to facilitate statistical analysis of the movement, mathematical models for analysis of joint motion waveforms as polynomial equations and Fourier transformation have been used (2, 7, 13). Establishment of normative data in the healthy population is crucial to improve understanding and treatment of orthopaedic diseases (5, 14). specific kinematic studies adapted to breed conformation standards are necessary to explain how particular conformational features may affect the musculoskeletal function (11). However, differences in type of locomotion such as walking or trotting, number of gait cycles analysed, trot velocity, habituation on the treadmill, and the method of 2D or 3D analysis used may limit the use of results from previous studies (9, 12 19). The goal of this study was to evaluate Vet Comp Orthop Traumatol 3/2011

2 F. S. Agostinho et al.: Kinematic analysis trotting on a treadmill Material and methods This study followed the guidelines for the care and use of laboratory animals, and was approved by the Ethical Committee of the School of Veterinary Medicine and Animal Science, Unesp Botucatu (Brazil). Twenty large, clinically healthy dogs were used. There were 10 Labrador Retrievers (2 males, 8 females) weighing 33.3 39.4 kg (mean 35.7 kg), and aged from 2.2 to 5.1 years, and 10 Rottweilers (3 males, 7 females) weighing 37.4 44.8 kg (mean 40.2), and aged from two to 5.9 years. All dogs were judged to be healthy on the basis of results of complete physical and orthopaedic examinations, and radiographic evaluation of the coxofemoral and femorotibial joints. Data Collection Fig. 1 Graphs of flexion and extension angles, angular acceleration, and angular velocity of the carpal, elbow, shoulder, tarsal, stifle and hip joints in Labrador dogs. For all three columns, the x-axis represents the percent time elapsed (0-100%). For the 1st column (Angles), the y-axis is 'Degrees', the y-axis for the 2nd column (Angular Acceleration) is 'Degrees/sec/sec', and the y-axis for the 3rd column (Angular Velocity) is 'Degrees/sec'. In the last two columns, the horizontal line dividing the y-axis in most of the individual charts is the marker for the value '0'. kinematic patterns of forelimbs and hindlimbs in clinically normal Labrador and Rottweiler dogs trotting on a treadmill at a constant speed. The choice was based on high prevalence of orthopaedic diseases, such as hip dysplasia, cranial cruciate rupture, elbow dysplasia, and osteochondrosis in both of these breeds. We hypothesized that despite similarities in size, these two breeds of dogs would generate different kinematic data. In addition, breed-specific gait databases may be useful in comparing dogs with abnormal gaits. A 3D capture system was used to perform analysis of joint kinematics. Before beginning the kinematic analysis, the dogs were trained to trot on a custom-built canine treadmill for a mean period of two weeks. Each dog was trained every second day with three sessions per day, for approximately three to five minutes. In the first day, the dogs were initially accustomed to walking on the treadmill. Those dogs that became rapidly familiarized with the treadmill started trotting on the same day. Each dog was tagged with 11 retroflective spherical markers (1.8 cm in diameter) placed by the same investigator using double-sided adhesive tape. Markers were placed on the skin over the dorsal point of the iliac crest, lateral prominence of the ischial tuberosity, eminence of the greater trochanter of the femur, femorotibial joint between the lateral epicondyle of the femur and the fibular head, lateral malleolus of the distal tibia, distal lateral aspect of the fifth metatarsus, the point of the cranial angle of the scapula, acromium of the scapulohumeral joint, lateral epicondyle of the humerus, styloid ulnar process, and distal lateral aspect of the fifth metacarpal bone. Kinematic data were collected by use of a three-camera system a with a recording frequency of 120 Hz strategically placed on one side of the treadmill. The camera set- Vet Comp Orthop Traumatol 3/2011 Schattauer 2011

F. S. Agostinho et al.: Kinematic analysis trotting on a treadmill 3 tings were: lens 8.5 mm, threshold level 25, strobe 100, gain 2, circularity 25. For each analysis, the system was calibrated with the procedure DynaCal b, where a rod with known dimensions (0.39 m) was driven by the volume to be calibrated for a specified period (60 s). Later, an L-frame was positioned on the treadmill in the capture area to generate the origin and the spatial coordinate system with the XYZ axes. A threedimensional testing space (3 m in length x 2.5 m in width x 2 m in height) was established on the treadmill. The kinematic study was performed with markers placed first on the right side of the dog, and then on the left side. For each analysis, the treadmill was repositioned and the system calibrated. The dogs were led on a leash by a handler positioned at the front of the treadmill. The treadmill was started at a low speed that was increased gradually until the dogs achieved a normal trotting gait. Treadmill speed was maintained between 2.1 and 2.2 m/s while the cameras recorded movement. A minimum of five valid trials of seven seconds duration were obtained from the right and left sides of each dog. In each trial, five completed strides were analysed, yielding a single mean value for each side of each dog. Specialised computer software was used to collect and process kinematic data. Data were analysed by use of a motion-analysis program c. The 11 individual markers were identified and labelled to construct a 3D stick-diagram representation of the dog. A stride was defined as the period from the beginning of the stance phase of the thoracic limb to the end of the swing phase of this limb. For the hindlimb, the beginning of stance phase was determined by the maximum extension of the stifle, and the beginning of swing phase was determined by the maximum extension of the tarsus at the end of the stance phase. For the forelimb, the beginning of stance phase was determined by the maximum extension of the a b c Vicon MX-3+: Vicon, Oxford Metrics Group, Oxford, UK DynaCal: Vicon, Oxford Metrics Group, Oxford, UK Vicon Motus: Vicon, Oxford Metrics Group, Oxford, UK Fig. 2 Graphs of flexion and extension angles, angular acceleration, and angular velocity of the carpal, elbow, shoulder, tarsal, stifle and hip joints in Rottweiler dogs. For all three columns, the x-axis represents the percent time elapsed (0-100%). For the 1st column (Angles), the y-axis is 'Degrees', the y-axis for the 2nd column (Angular Acceleration) is 'Degrees/sec/sec', and the y-axis for the 3rd column (Angular Velocity) is 'Degrees/sec'. In the last two columns, the horizontal line dividing the y-axis in most of the individual charts is the marker for the value '0'. elbow, and the beginning of swing phase was determined by the maximum extension of the carpus at the end of the stance phase. Flexion and extension joint angles (maximum, minimum, displacement), angular velocity (maximum, minimum), and angular acceleration (maximum, minimum) were determined for the shoulder, elbow, carpal, hip, stifle, and tarsal joints. The motion-analysis program automatically created tabulated digital data and the graphic presentation. In addition, the lengths of the limb segments were deter- Schattauer 2011 Vet Comp Orthop Traumatol 3/2011

4 F. S. Agostinho et al.: Kinematic analysis trotting on a treadmill Joint Tarsal Stifle Hip Carpal Elbow Shoulder Maximum angle ( ) Minimum angle ( ) Angular displacement ( ) Labrador 156.09 ± 5.47 106.48 ± 7.45 49.60 ± 5.94 a 0.23 0.22 Rottweiler 158.49 ± 4.70 103.51 ± 5.03 54.98 ± 4.84 b Labrador 144.41 ± 5.14 91.92 ± 4.98 52.48 ± 4.04 a 0.048* 0.08 Rottweiler 149.73 ± 6.52 87.36 ± 6.80 62.37 ± 6.53 b Labrador 123.56 ± 8.27 96.64 ± 8.15 29.91 ± 3.38 0.60 0.41 Rottweiler 130.41 ± 7.07 97.36 ± 5.69 32.64 ± 4.31 Labrador 173.23 ± 1.66 a 101.44 ± 10.56 71.79 ± 10.55 0.01 0.08 Rottweiler 170.07 ± 2.76 b 108.57 ± 9.69 61.49 ± 9.91 Labrador 154.28 ± 9.64 90.52 ± 11.66 63.77 ± 4.83 a 0.16 0.34 Rottweiler 148.85 ± 9.15 93.99 ± 10.19 54.86 ± 5.16 b Labrador 152.06 ± 7.18 117.40 ± 7.66 34.65 ± 4.17 0.21 0.35 Rottweiler 152.89 ± 6.64 118.79 ± 9.54 34.10 ± 5.21 0.002 0.005 0.88 0.02* 0.0002 Key: A * indicates that there was no statistical significance after Bonferroni correction. Values followed by different letters as superscripts within each column are significantly different. 0.78 Table 1 Comparison of the maximum angle (º), minimum angle (º), and angular displacement (º) of the forelimbs and hindlimbs of Labrador Retrievers and Rottweiler dogs. mined by measuring the distance between two markers. Statistical analysis Analysis of the kinematic data was by a repeated-measures analysis of variance with one crossover factor (side) and one nest factor (breed) (two-way ANOVA). The sequential Bonferroni adjustment procedure was applied for our contrasts. With two levels for the repeated factor (left or right), only one type of correlation was relevant, therefore a compound symmetry structure was used in our model. Student's t-test for independent samples was performed to compare the lengths of the limb segments. Differences were considered significant at p <0.05. Table 2 Comparison of the maximum angular acceleration (º/s 2 ) and minimum angular acceleration (º/s 2 ) of the forelimbs and hindlimbs of Labrador Retrievers and Rottweiler dogs. Joint Tarsal Stifle Hip Carpal Elbow Maximum angular acceleration ( /s 2 ) Minimum angular acceleration ( /s 2 ) Labrador 40164.48 ± 10232.80-40631.18 ± 13570.12 0.51 Rottweiler 37193.06 ± 11745.72-44718.67 ± 10617.50 0.41 Labrador 20347.91 ± 4068.01-24123.63 ± 3783.62 a 0.39 0.008 Rottweiler 21704.39 ± 4454.65-29915.58 ± 4046.20 b Labrador 23660.82 ± 12739.25-27906.85 ± 13090.91 0.19 Rottweiler 25812.65 ± 14861.61-24373.00 ± 10721.48 Labrador 52380.78 ± 24960.49-105405.50 ± 52406.02 0.87 Rottweiler 51151.53 ± 32414.61-99905.55 ± 62856.84 Labrador 24481.66 ± 8318.85 a -35707.23 ± 11507.27 0.006 Rottweiler 16389.03 ± 4355.63 b -25242.95 ± 9761.13 0.68 0.85 0.03* Labrador 17297.68 ± 4337.99-20963.94 ± 5241.29 Shoulder 0.06 0.10 Rottweiler 13285.95 ± 6360.40-17604.68 ± 5869.80 Key: A * indicates that there was no statistical significance after Bonferroni correction. Values followed by different letters as superscripts within each column are significantly different. Results Adverse events were not observed. Within each group, the differences between the right and left limbs in all kinematic variables were not significant. The graphs of flexion and extension angles, angular velocity, and angular acceleration showed similar shape pattern in both groups ( Fig. 1 and 2). The differences between Labradors and Rottweilers in some kinematic parameters were significant ( Tables 1, 2 and 3): angular displacement and minimum angular acceleration of the stifle (Rottweiler >Labrador); angular displacement and maximum angular velocity of the tarsus (Rottweiler >Labrador); minimum angular velocity of the shoulder (Labrador >Rottweiler); angular displacement, maximum angular acceleration, maximum angular velocity, and minimum angular velocity of the elbow (Labrador >Rottweiler); maximum angle and maximum angular velocity of the carpus (Labrador >Rottweiler). Vet Comp Orthop Traumatol 3/2011 Schattauer 2011

F. S. Agostinho et al.: Kinematic analysis trotting on a treadmill 5 The differences between lengths of the hindlimb segments within the same group or between breeds were not significant ( Table 4). Differences between breeds in segment 3 (from lateral epicondyle of the humerus to acromium of the scapulohumeral joint) of the forelimbs were significant ( Table 5). Discussion Table 3 Comparison of maximum angular velocity (º/s) and minimum angular velocity (º/s) of the forelimbs and hindlimbs of Labrador Retrievers and Rottweiler dogs. Joint Maximum angular Velocity ( /s) Tarsal Stifle Hip Carpal Elbow Labrador Rottweiler Minimum angular velocity ( /s) 620.26 ± 101.49 a -770.66 ± 128.03 0.04 718.44 ± 100.78 b -736.13 ± 126.03 Labrador 635.63 ± 64.66-454.13 ± 77.04 0.24 Rottweiler 662.37 ± 109.00-499.69 ± 70.67 Labrador 281.16 ± 73.82-479.94 ± 86.77 0.49 Rottweiler 306.96 ± 106.87-404.94 ± 71.61 Labrador 1069.13 ± 228.59 a -1350.04 ± 313.94 0.004 Rottweiler 794.36 ± 178.96 b -1099.48 ± 350.09 Labrador 848.60 ± 119.90 a -676.65 ± 85.04 a 0.001 Rottweiler 662.12 ± 118.59 b -513.89 ± 66.52 b 0.15 0.046* 0.08 <0.0001 Labrador 403.27 ± 51.44-540.70 ± 80.08 a Shoulder 0.10 <0.0001 Rottweiler 364.42 ± 91.56-372.06 ± 81.47 b Key: A * indicates that there was no statistical significance after Bonferroni correction. Values followed by different letters as superscripts within each column are significantly different. Table 4 Comparison of hindlimb segments (cm) between Labrador Retrievers and Rottweiler dogs. Segments 1 From the distal lateral aspect of the fifth metatarsus to the lateral malleolus of the distal tibia 2 From the lateral malleolus of the distal tibia to the lateral epicondyle of the femur. 3 From the lateral epicondyle of the femur to the eminence of the greater trochanter of the femur. Key: A * indicates no significant difference. Length (mean ± SD) Labrador 8.66 ± 0.92 Rottweiler 9.12 ± 0.93 Labrador 21.55 ± 1.01 Rottweiler 21.65 ± 1.74 Labrador 20.51 ± 1.03 Rottweiler 21.01 ± 1.8 0.52 s 0.28* 0.87* 0.45* In order to facilitate gait analysis, the dogs in our study were observed at the trot. This gait pattern is considered to be a symmetrical gait in which the movements of one side of the body mirror the movements of the opposite side (2, 13). In a two-dimensional kinematic analysis using Labrador Retrievers and Labrador cross-breed dogs at the trot, the linear parameters (stride length, stride frequency, stride time, and linear velocity) and the angular parameters (angular displacement and velocity angular) for the right and left sides were similar (19). Also, in the present study, the kinematic patterns of the right and left sides were similar for each breed. This suggests that intra-dog analysis or use of one limb to characterize patterns during the trot in healthy dog would be permissible (4, 13). However, in a study using inverse dynamics analysis of gait, it was observed that during the trot, the dog has a mechanically dominant side (20). In our study, the waveform curves of flexion and extension angles, angular velocity, and angular acceleration showed similar patterns in both groups. However, significant differences were observed in some kinematic parameters between Labradors and Rottweilers. In a study comparing trotting gaits in Labrador Retrievers and Greyhounds, although both breeds moved in a dynamically similar manner, the Greyhounds used fewer, longer strides than the Labrador Retrievers. This was mainly attributable to differences in body size of the dogs, emphasizing the importance of breed standardization (21). Our results showed larger angular displacement for the stifle and tarsal joints in Rottweilers than Labradors, but the elbow joint had larger displacement in Labradors. Since the angular displacement is the difference between maximum angle and minimum angle, the Rottweilers had greater range-ofmotion of the stifle and lesser of the elbow at the trot compared to Labradors. At the trot, the hip joint extends slowly throughout the stance phase, and in the swing phase it follows with rapid flexion. The stifle joint extends at the end of stance phase corresponding to propulsion and the tarsal joint is similar to the stifle joint, but with higher magnitude of extension in late stance (2). In another study, gross differences were observed in the angular and mechanical patterns of the metatarsophalangeal joints between Greyhounds and Labradors, and the stifle and tarsal joints showed larger displacement in the Greyhounds (16). Angular acceleration may change in either magnitude or direction (6). For instance, dogs with hip dysplasia showed changes in coxofemoral joint articular acceleration in both stance (extension) and swing (flexion) phases of the stride, compared with controls (22). In the present study, the minimum angular acceleration of stifle was significantly higher in Rottweilers than Labrador Retrievers, but the maximum angular acceleration of elbow was significantly higher in Labrador Schattauer 2011 Vet Comp Orthop Traumatol 3/2011

6 F. S. Agostinho et al.: Kinematic analysis trotting on a treadmill Table 5 Comparison of forelimb segments (cm) between Labrador Retrievers and Rottweiler dogs. Segments 1 From the distal lateral aspect of the fifth metacarpal bone to the styloid ulnar process. 2 From the styloid ulnar process to the lateral epicondyle of the humerus. 3 From the lateral epicondyle of the humerus to the acromium of the scapulohumeral joint. 4 Form the acromium of the scapulohumeral joint to the cranial angle of the scapula. Length (mean ± SD) s Labrador 13.21 ± 1.82 0.09* Rottweiler 14.69 ± 1.92 Labrador 16.82 ± 1.89 0.39* Rottweiler 17.4 ± 1.43 Labrador 17.01 ± 1.79 a 0.01 Rottweiler 19.09 ± 1.58 b Labrador 6.63 ± 0.59 0.34* Rottweiler 6.34 ± 0.74 Key: A * indicates that there was no statistical significance. Values followed by different letters as superscripts within each column are significantly different. than Rottweilers, suggesting that Rottweilers have a higher retardation of the extension of the stifle and a lower angular acceleration in elbow flexion. The parametric differences observed between the dogs may be associated with the body morphology that is considered a determinant of baseline kinematics (2). However, the angular acceleration depends on the size of the force, mass of the object, and the way in which the mass is distributed about the center of gravity (6). Joint angular velocity (the rate of flexion and extension of a joint) is related to the dynamics of muscle activation and force generation during walking (6). This information may be important in detecting subtle changes in joint motion during gait that are not detected by changes in joint angles (14). Paw velocity and stifle angular velocity were kinematic variables that had the largest differences when comparing affected and unaffected limbs in dogs with cranial cruciate rupture (23). In our study, most of the parameters of angular velocity of the elbow and shoulder, and maximum angular velocity of the carpal joint were higher in Labradors than Rottweilers, suggesting that the Labrador has a faster position change of these joints. This may be associated with conformational variation, since the measurement of the limb segments was higher in Rottweilers than Labradors in segment 3. In a study comparing ground reaction forces between clinically healthy Rottweilers and Labradors at the trot, the mean relative velocity was influenced by the limb length (24). Some limitations in our study were due to the limited number of cameras available. It was only possible to record one side at a time, and then recalibrate the system in each analysis to record the opposite side, which increased the time required to perform the examinations. Other anatomical measurements could have been included to establish the conformational data. Skin movement and soft tissue under the markers were also a potential source of error in the kinematic analysis. However, invasive procedures are unreasonable for clinical use. In conclusion, both breeds had similar kinematic patterns, but there were magnitude differences, especially of the elbow and stifle joints. Therefore, each breed should have a specific database. These databases may be used for comparisons with injured clinical populations, but more studies are necessary to determine the importance of these differences in a clinical situation. Acknowledgements We are grateful to FAPESP (The State of São Paulo Research Foundation) and CNPq (National Council for Scientific and Technological Development) for the financial support, Dr. Guy Beauchamp from the Faculty of Veterinary Medicine of the University of Montreal for the statistical analysis, and Wagner de Godoy who is an engineer at the gait laboratory from the Albert Einstein Hospital, Brazil, for the technical support. References 1. Anderson MA, Mann FA. Force plate analysis: a noninvasive tool for gait evaluation. Comp Cont Educ Pract 1994; 16: 857 867. 2. DeCamp CE. Kinetic and kinematic gait analysis and the assessment of lameness in the dog. Vet Clin North Am Small Anim Pract 1997; 27: 825 840. 3. Waxman AS, Robinson DA, Evans RB, et al. Relationship between objective and subjective assessment of limb function in normal dogs with an experimentally induced lameness. Vet Surg 2008; 37: 241 246. 4. Gillette RL, Angle TC. Recent developments in canine locomotor analysis: A review. Vet J 2008; 178: 165 176. 5. McLaughlin RM. Kinetic and kinematic gait analysis in dogs. Vet Clin North Am Small Anim Pract 2001; 31: 193 201. 6. Whittle MW. Gait Analysis: An Introduction. 4 th ed. Edinburgh: Elsevier; 2007. pg. 255. 7. DeCamp CE, Riggs CM, Olivier B, et al. Kinematic evaluation of gait in dogs with cranial cruciate ligament rupture. Am J Vet Res 1996; 57: 120 126. 8. Bennett RL, DeCamp CE, Flo GL, et al. Kinematic gait analysis in dogs with hip dysplasia. Am J Vet Res 1996; 57: 966 971. 9. Hottinger HA, DeCamp CE, Olivier B, et al. Noninvasive kinematic analysis of the walk in healthy large-breed dogs. Am J Vet Res 1996; 57: 381 388. 10. Lascelles BD, 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: 227 282. 11. Gillette RL. Gait analysis. In: Canine Rehabilitation and Physical Therapy. Millis DL, Levine D, Taylor RA, editors. Philadelphia; W.B. Saunders; 2004. pg. 201 210. 12. Kim J, Rietdyk S, Breur GJ. Comparison of two-dimensional and three-dimensional systems for kinematic analysis of the sagittal motion of canine hind limbs during walking. Am J Vet Res 2008; 69: 1116 1122. 13. Schaefer SL, DeCamp CE, Hauptman JG, et al. Kinematic gait analysis of hind limb symmetry in dogs at the trot. Am J Vet Res 1998; 59: 680 695. Vet Comp Orthop Traumatol 3/2011 Schattauer 2011

F. S. Agostinho et al.: Kinematic analysis trotting on a treadmill 7 14. Owen MR, Richards J, Clements DN, et al. Kinematics of the elbow and stifle joints in greyhounds during treadmill trotting an investigation of familiarization. Vet Comp Orthop Traumatol 2004; 17: 141 145. 15. Clements DN, Owen MR, Carmichael S, et al. Kinematic analysis of the gait of 10 Labrador retrievers during treadmill locomotion. Vet Rec 2005; 156: 478 481. 16. Colborne GR, Innes JF, Comerford EJ, et al. Distribution of power across the hind limb joints in Labrador retrievers and greyhounds. Am J Vet Res 2005; 66: 1563 1571. 17. Colborne GR, Walker AM, Tattersall AJ, et al. Effect of trotting velocity on work patterns of the hind limbs of greyhounds. Am J Vet Res 2006; 67: 1293 1298. 18. DeCamp CE, Soutas-Little RW, Hauptman J, et al. Kinematic gait analysis of the trot in healthy greyhounds. Am J Vet Res 1993; 54: 627 634. 19. Gillette RL, Zebas CJ. A two-dimensional analysis of limb symmetry in the trot of Labrador retrievers. J Am Anim Hosp Assoc 1999; 35: 515 520. 20. Colborne GR. Are sound dogs mechanically symmetric at trot? No, actually. Vet Comp Orthop Traumatol 2008; 21: 294 301. 21. Bertram JE, Lee DV, Case HN, et al. Comparison of the trotting gaits of Labrador retrievers and Greyhounds. Am J Vet Res 2000; 61: 832 838. 22. Poy NSJ, DeCamp CE, Bennett RL, et al. Additional kinematic variables to describe differences in the trot between clinically normal dogs and dogs with hip dysplasia. Am J Vet Res 2000; 61: 974 978. 23. Sanchez-Bustinduy M, Medieros MA, Radke H, et al. Comparison of kinematic variables in defining lameness caused by naturally occurring rupture of the cranial cruciate ligament in dogs. Vet Surg 2010; 39: 523 530. 24. Mölsä SH, Hielm-Björkman AK, Laitinen-Vapaavuori OM. Force platform analysis in clinically healthy Rottweilers: comparison with Labrador Retrievers. Vet Surg 2010; 39: 701 707. Schattauer 2011 Vet Comp Orthop Traumatol 3/2011