Objective evaluation of lameness in dogs is performed for research purposes and clinical evaluation of patients. Historically, force plate analysis has been used to evaluate kinetic parameters such as peak vertical force and vertical impulse. Gait analysis by use of this method has been investigated in clinically normal animals at a walk and trot and while jumping1–4 and is used to objectively measure outcomes following treatment for musculoskeletal conditions.5–8
Some disadvantages of force plate analysis include a limited ability to measure consecutive foot-falls9 and the need for dogs to make multiple passes to collect data for each limb.10 Furthermore, force plate analysis cannot be performed in dogs that are minimally weight bearing (eg, prior to or following surgical intervention). Although some of these limitations have been addressed through the use of multiple force plates, PSWs have become a popular alternative that allows for collection of temporospatial (eg, stride length and gait velocity) and GRF measurements.11,12 Comparisons of force plate analysis and PSW analysis for detecting lameness in dogs indicate that measurements obtained by these 2 methods are well correlated.10,13 Reports14,15 detailing the use of PSWs to clinically evaluate the effect of treatment in dogs with cruciate ligament disease and hip dysplasia have also been published.
The manufacturer of a commonly used PSW recommends covering the sensors with a suitably heavy cover pad to protect the system during canine gait analysis.16 To the authors’ knowledge, the effect of cover type on temporospatial and GRF measurements of dogs obtained during gait analysis with a PSW has not been previously evaluated. Because multi-institutional studies are becoming more commonplace in veterinary medicine, it is important to ensure that data are collected in a standardized manner. The use of different cover types for PSWs among study sites or during longitudinal studies may limit the comparability of outcome measurements. Furthermore, if a PSW is used to monitor a patient's clinical outcome, the use of different cover types between visits could result in erroneous conclusions. Therefore, the aim of the study reported here was to compare the effects of 3 cover types on temporospatial and GRF measurements of dogs during gait analysis with a PSW. We hypothesized that there would be no differences in these measurements among the 3 cover types.
Materials and Methods
Dogs
The institution's animal care and use committee approved this study, and client- and staff-owned dogs were enrolled with written owner consent. A physical examination, including an orthopedic examination, was performed on all dogs by a board-certified veterinary surgeon (NRK), and results were used to categorize dogs as nonlame or lame. A modified lameness scale from the American Association of Equine Practitioners was used by the surgeon to subjectively grade dogs as 0 (no lameness), 1 (lameness difficult to observe and not consistently apparent with any gait), 2 (lameness difficult to discern at a walk or trot and more apparent with circling or stairs), 3 (lameness consistently present at a trot), 4 (intermittent non–weight-bearing lameness obvious at a walk), or 5 (non–weight bearing).17 Dogs were excluded from the study if they had evidence of neurologic disease on physical examination or were graded as 1 or 5 on the lameness scale to ensure that evidence of lameness was clear and to exclude dogs with non–weight-bearing lameness that would limit the ability to conduct gait analyses. Dogs graded as 2, 3, or 4 on the lameness scale and that had evidence of orthopedic disease on examination were categorized as lame. All other dogs were considered nonlame. The age, sex, breed, and BW of each enrolled dog were recorded.
Gait analysis
A crossover study design was used in which dogs were evaluated on a PSWa to which the 3 cover types (a corrugated 0.32-cm-thick vinyl matb [cover A], a 0.64-cm-thick polyvinyl chloride yoga matc [cover B], and a 0.32-cm-thick polyvinyl chloride yoga matd [cover C]) were applied in a random order that was determined by use of a computer-generated randomization schedule.e Dogs were allowed to acclimate to the room until they appeared comfortable (minimum of 5 minutes) prior to walking on the PSW.
Dogs were first allowed to walk on the PSW with each new cover type prior to data collection and were given a minimum of 20 minutes to rest between cover changes. The PSW was calibrated for each dog prior to start of data collection and following each cover change. Ten video-recorded trials were acquired per dog and per cover type, for a total of 30 trials/dog. All dogs were walked on the PSW at a relaxed, steady pace with the handler on the right side. Each dog had the same handler (either owner or study personnel) for all 3 cover types. A trial was considered valid if the dog walked with a gait pattern that consisted of 3 feet on the walkway at any given time, did not pull on the leash or overtly turn the head from midline, and maintained a gait velocity of 0.8 to 1.3 m/s and acceleration within 0.1 m/s2.
The following temporospatial variables were measured: gait velocity for each foot (gait distance [total distance the dog walked on the mat]/gait time [time of first contact of the first left or right front stance to the time of first contact of the last left or right front stance]), number of gait cycles per minute, and gait cycle time (time between foot contacts). The following GRF variables were measured for each limb: maximum peak pressure, maximum force during stance, and vertical impulse (mean impulse of all foot strikes). Maximum force and vertical impulse were measured as absolute (kilograms and kilograms × seconds, respectively) and normalized values (%BW and %BW × seconds, respectively). For maximum force, whether there were multiple stances, the mean maximum force value was calculated. The first 5 valid trials/cover were included in the data analysis.
Statistical analysis
Statistical analysis was performed by use of commercially available software.f Wilcoxon rank sum tests were used to compare age and BW between the lame and nonlame groups. Three analytic approaches were used to assess the effect of cover type on gait measurements for lame and nonlame dogs groups separately. Pairwise correlations were used to measure the linear agreement among cover types for each of the measured GRFs. To protect against a linear agreement but a constant error, the amount of translational error was assessed by the use of paired 2-tailed t tests to determine if the mean difference in GRF variables between each cover type pair differed from zero. Next, nonparametric matched-pairs comparisons were performed between each cover type for gait velocity, number of gait cycles per minute, gait cycle time, and gait time. Finally, nonparametric pairwise tests were used to compare GRF measurements by cover type for lame and nonlame groups separately. A P value < 0.05 was considered significant.
Results
Thirty-five adult dogs (25 nonlame and 10 lame) were enrolled in the study. Dogs included mixed-breed dogs (13 nonlame and 4 lame), Labrador Retrievers (3 nonlame and 3 lame), Golden Retrievers (2 nonlame and 1 lame), Border Collies (2 nonlame), Standard Poodles (1 nonlame and 1 lame), 1 nonlame Doberman Pinscher, 1 nonlame Australian Shepherd, 1 nonlame Spinone Italiano, 1 nonlame German Shorthaired Pointer, and 1 lame Siberian Husky. The nonlame group consisted of 12 neutered males, 11 spayed females, and 2 sexually intact females, and the lame group of 6 neutered males and 4 spayed females. The mean ± SD BW of all dogs was 27.1 ± 7.7 kg (range, 16.5 to 48.2 kg), and the mean ± SD age was 5.1 ± 2.8 years (range, 1 to 10.6 years). All dogs completed a minimum of 5 valid trials/cover type tested. There was no significant difference between the nonlame and lame groups in age (P = 0.15) or BW (P < 0.18). Dogs categorized as lame had either forelimb or hind limb lameness; none had concurrent forelimb and hind limb lameness.
No significant differences were identified in temporospatial measurements (gait velocity, number of gait cycles per minute, gait cycle time, gait distance, and gait time) among cover types for both the nonlame and lame dogs (Table 1). There were significant differences in maximum peak pressure for each foot between all cover types for both nonlame (P ≤ 0.001) and lame (P ≤ 0.02) dogs (Table 2). For both the nonlame and lame dogs, significant differences in maximum force measurements (both absolute and normalized) for each foot were detected between covers A and C (P ≤ 0.001 and P ≤ 0.004, respectively) and covers B and C (P ≤ 0.001 and P ≤ 0.002, respectively) but not between covers A and B (P ≥ 0.75). For both the nonlame and lame dogs, significant differences in vertical impulse measurements (both absolute and normalized) for each foot were identified between covers B and C (P ≤ 0.001) and for all feet except the right hind between covers A and C (P ≤ 0.001).
Comparison of the effects of 3 walkway cover types on temporospatial measurements obtained for 35 client- and staff-owned dogs (25 nonlame and 10 lame) during gait analysis with a PSW.
| Variable | Dog group | Cover A | Cover B | Cover C |
|---|---|---|---|---|
| Gait velocity (m/s)* | Nonlame | 1.06 ± 0.11 | 1.07 ± 0.08 | 1.05 ± 0.11 |
| Lame | 1.04 ± 0.11 | 1.07 ± 0.10 | 1.09 ± 0.08 | |
| No. of gait cycles/min | Nonlame | 81.76 ± 9.99 | 82.20 ± 10.66 | 81.00 ± 10.99 |
| Lame | 78.00 ± 11.31 | 80.00 ± 10.02 | 80.20 ± 10.02 | |
| Gait cycle time (s) | Nonlame | 0.75 ± 0.09 | 0.72 ± 0.16 | 0.76 ± 0.11 |
| Lame | 0.78 ± 0.10 | 0.76 ± 0.10 | 0.76 ± 0.09 | |
| Gait distance (m) | Nonlame | 2.39 ± 0.13 | 2.40 ± 0.14 | 2.39 ± 0.13 |
| Lame | 2.43 ± 0.15 | 2.36 ± 0.11 | 2.43 ± 0.15 | |
| Gait time (s) | Nonlame | 2.29 ± 0.31 | 2.27 ± 0.30 | 2.34 ± 0.33 |
| Lame | 2.37 ± 0.33 | 2.22 ± 0.19 | 2.24 ± 0.28 |
Data represent mean ± SD.
Gait velocity = Gait distance/gait time.
Comparison of the effects of 3 walkway cover types on GRF measurements obtained for each foot of the dogs in Table 1.
| Variable | Dog group | Limb | Cover A | Cover B | Cover C |
|---|---|---|---|---|---|
| Maximum peak pressure (psi) | Nonlame | Left forelimb | 38.4 ± 7.5a | 30.1 ± 7.2b | 34.6 ± 7.1c |
| Right forelimb | 37.8 ± 7.9a | 28.9 ± 8.2b | 33.2 ± 8.7c | ||
| Left hind limb | 28.4 ± 6.0a | 21.6 ± 6.1b | 24.9 ± 6.3c | ||
| Right hind limb | 29.3 ± 6.5a | 21.7 ± 6.1b | 25.3 ± 6.3c | ||
| Lame | Left forelimb | 40.9 ± 9.3a | 32.2 ± 7.8b | 35.9 ± 7.2c | |
| Right forelimb | 33.9 ± 10.7a | 29.9 ± 5.1b | 29.4 ± 9.8c | ||
| Left hind limb | 28.5 ± 4.8a | 21.1 ± 3.9b | 24.7 ± 3.9c | ||
| Right hind limb | 29.9 ± 5.1a | 22.8 ± 5.1b | 26.9 ± 5.3c | ||
| Normalized maximum force (%BW) | Nonlame | Left forelimb | 51.0 ± 9.1a | 52.6 ± 7.8a | 58.9 ± 5.5b |
| Right forelimb | 50.4 ± 9.0a | 50.8 ± 5.9a | 55.6 ± 4.4b | ||
| Left hind limb | 33.1 ± 6.8a | 35.7 ± 7.8a | 37.6 ± 5.4b | ||
| Right hind limb | 32.1 ± 5.5a | 33.1 ± 7.2a | 34.5 ± 4.6b | ||
| Lame | Left forelimb | 56.1 ± 4.7a | 55.7 ± 6.1a | 62.4 ± 7.9b | |
| Right forelimb | 52.6 ± 8.2a | 51.5 ± 10.4a | 57.4 ± 13.4b | ||
| Left hind limb | 29.8 ± 5.6a | 30.6 ± 4.9a | 34.5 ± 6.1b | ||
| Right hind limb | 31.9 ± 2.5a | 30.9 ± 2.7a | 34.9 ± 2.6b | ||
| Normalized vertical impulse (%BW × s) | Nonlame | Left forelimb | 18.4 ± 2.8a | 17.9 ± 3.9a | 20.9 ± 3.8b |
| Right forelimb | 18.0 ± 3.0a | 16.8 ± 3.7a | 19.4 ± 3.8b | ||
| Left hind limb | 10.9 ± 1.7a | 10.9 ± 2.7a | 11.9 ± 1.9b | ||
| Right hind limb | 10.8 ± 1.6a | 10.4 ± 2.6b | 11.0 ± 2.1a | ||
| Lame | Left forelimb | 21.0 ± 3.1a | 19.4 ± 2.5a | 21.9 ± 3.4b | |
| Right forelimb | 19.5 ± 5.1a | 17.3 ± 4.5a | 19.6 ± 5.2b | ||
| Left hind limb | 10.3 ± 2.3a | 9.8 ± 2.0a | 19.6 ± 5.2b | ||
| Right hind limb | 11.1 ± 1.7ab | 10.1 ± 1.6a | 11.3 ± 1.7b |
Data represent mean ± SD.
psi = Pounds per square inch.
Within a row, values with different superscript letters differ significantly (P < 0.05).
Discussion
Traditionally, force plate analysis has been recognized as the gold standard for obtaining temporospatial and GRF measurements.18,19 Despite this, PSWs are being used with increasing frequency for gait analysis in veterinary medicine because of their ease of use and ability to provide more information than force plate systems, such as limb symmetry data and stride length. Studies10,11,20 in dogs and cats show that PSWs are a valid alternative to force plate systems.
To the authors’ knowledge, the walkway cover type used during data collection with a PSW is rarely provided in published reports on this topic. In the present study, most GRF measurements obtained with the PSW differed significantly among the 3 cover types for both nonlame and lame dogs. This is important to consider in a clinical setting when following patient progress on an individual basis and when performing clinical studies to assess treatment outcome.
The significant differences in GRFs identified by pairwise comparisons of cover types were inconsistent between nonlame and lame dogs for vertical impulse. Because vertical impulse is related to force, the reason for this inconsistent finding between nonlame and lame dogs is unclear, given the similarity in findings for the other GRF variables (ie, maximum pressure and maximum force).
To the authors’ knowledge, no published literature exists regarding the effect that walkway cover type or cover thickness might have on temporospatial or GRF measurements obtained with PSWs. It is reasonable to expect that the thickness or shock-absorbing properties of the cover might affect the measurement of these variables. In the present study, most pairwise comparisons of cover types revealed significant differences in GRF measurements for both nonlame and lame dogs. Although comparable maximum force values were obtained for both nonlame and lame dogs when covers A and B were used, maximum peak pressure measurements were not consistent among cover types.
Cover types for the present study were selected on the basis of manufacturer recommendations and discussions with other veterinary clinicians who use PSWs for clinical research. One limitation of this study was that only 3 cover types were evaluated. Many other cover types are available; however, it was not within the scope of the present study to evaluate all available cover types—only to determine whether cover type could affect temporospatial and GRF measurements. Another limitation was that dogs were only walked on the PSW with a cover present. A comparison of measurements obtained with and without a cover would permit determination of which cover type, if any, might be ideal for both protecting the PSW and providing the most accurate measurements. Such information would be useful, particularly when attempting to assess the true GRFs on a foot rather than making relative comparisons among feet.
Investigators should consider the potential effect of walkway cover type when attempting to make direct comparisons between studies involving PSWs. To ensure that data are collected in a standardized manner during PSW studies involving dogs, investigators should use the same cover type (if any) for the duration of the study and, in the case of multi-institutional studies, should use the same cover type (if any) at all locations. In addition, the same cover type should be used during follow-up visits to evaluate a dog's clinical outcome. Reports of future studies involving PSWs should specify the type of cover used to facilitate comparisons among studies.
Acknowledgments
Funded by Colorado State University's SMIRK Program and the Eldred Foundation.
The authors declare that there was no proprietary interest for this project.
Presented in abstract form at the Veterinary Orthopedic Society Conference 2016, Big Sky, Mont, March 2016.
The authors thank Dr. Kyle Martin for assistance in data collection.
ABBREVIATIONS
| BW | Body weight |
| GRF | Ground reaction force |
| PSW | Pressure-sensitive walkway |
Footnotes
HRV Walkway 6 VersaTek System, Tekscan Animal Walkway System, South Boston, Mass.
V-Groove vinyl runner (2 × 20 ft), Commercial Mats and Rubber, Saratoga Springs, NY.
1/4-inch extra thick yoga mat, YogaAccessories.com, Richmond, Va.
1/8-inch classic yoga mat, YogaAccessories.com, Richmond, Va.
Randomness and Integrity Services Ltd, Dublin, Ireland. Available at www.random.org. Accessed March 4, 2015.
JMP, version 11.2.0, SAS Institute Inc, Cary, NC.
References
1. Budsberg SC, Verstraete MC, Soutas-Little RW. Force plate analysis of the walking gait in healthy dogs. Am J Vet Res 1987;48:915–918.
2. Budsberg SC, Verstraete MC, Brown J, et al. Vertical loading rates in clinically normal dogs at a trot. Am J Vet Res 1995;56:1275–1280.
3. Budsberg SC, Rytz U, Johnston SA. Effects of acceleration on ground reaction forces collected in healthy dogs at a trot. Vet Comp Orthop Traumatol 1999;12:15–19.
4. Yanoff SR, Hulse DA, Hogan HA. Measurements of vertical ground reaction force in jumping dogs. Vet Comp Orthop Traumatol 1992;5:44–50.
5. Budsberg SC, Verstraete MC, Soutas-Little RW, et al. Force plate analyses before and after stabilization of canine stifles for cruciate injury. Am J Vet Res 1988;49:1522–1524.
6. Vasseur PB, Johnson AL, Budsberg SC, et al. Randomized, controlled trial of the efficacy of carprofen, a nonsteroidal anti-inflammatory drug, in the treatment of osteoarthritis in dogs. J Am Vet Med Assoc 1995;206:807–811.
7. Kennedy S, Lee DV, Bertram JE, et al. Gait evaluation in hip osteoarthritis and normal dogs using a serial force plate system. Vet Comp Orthop Traumatol 2003;16:170–177.
8. Hoelzler MG, Millis DL, Francis DA, et al. Results of arthroscopic versus open arthrotomy for surgical management of cranial cruciate ligament deficiency in dogs. Vet Surg 2004;33:146–153.
9. 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.
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;67:277–282.
11. Kirtley C. Chapter 1: the temporal-spatial parameters. In: Clinical gait analysis. Edinburgh: Elsevier, 2006;15–37.
12. Light VA, Steiss JE, Montgomery RD, et al. Temporal-spatial gait analysis by use of a portable walkway system in healthy Labrador Retrievers at a walk. Am J Vet Res 2010;71:997–1002.
13. Besancon MF, Conzemius MG, Derrick TR, et al. Comparison of vertical forces in normal greyhounds between force platform and pressure walkway measurement systems. Vet Comp Orthop Traumatol 2003;16:153–157.
14. Horstman CL, Conzemius MG, Evans R, et al. Assessing the efficacy of perioperative oral carprofen after cranial cruciate surgery using noninvasive, objective pressure platform gait analysis. Vet Surg 2004;33:286–292.
15. Lascelles BD, Freire M, Roe SC, et al. Evaluation of functional outcome after BFX total hip replacement using a pressure sensitive walkway. Vet Surg 2010;39:71–77.
16. Tekscan Walkway user manual. Boston, Mass: Tekscan, 2010.
17. American Association of Equine Practitioners. Lameness exams: Evaluating the lame horse. Available at: aaep.org/horsehealth/lameness-exams-evaluating-lame-horse. Accessed Jul 15, 2016.
18. 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.
19. Gillette RL, Angle TC. Recent developments in canine locomotor analysis: a review. Vet J 2008;178:165–176.
20. Oosterlinck M, Pille F, Huppes T, et al. Comparison of pressure plate and force plate gait kinetics in sound Warmbloods at walk and trot. Vet J 2010;186:347–351.
