Introduction
Pressure-sensitive walkway (PSW) systems have been growing in popularity as an effective and convenient way to objectively assess canine, feline, and equine gait patterns in clinical and research settings.1–8 Several PSW systems have been validated for assessment of temporospatial gait variables including, but not limited to, gait time, distance, velocity, cadence, and cycle time. Additionally, PSW systems are able to measure ground reaction forces (GRFs), including percentage of body weight placed on each limb, impulse (percentage of body weight times seconds), and maximum force exerted on each limb both as a percentage of body weight and as an absolute value. Together, temporospatial gait variables and GRFs allow for objective evaluation of potential or persistent lamenesses and gait abnormalities by tracking alterations in gait and stance patterns both instantaneously and over time. Both temporospatial gait variables and GRFs have been used to evaluate treatment effects in dogs, cats, and horses with various orthopedic diseases by comparing measurements obtained before and after treatment.1,2,8
Manufacturers of 1 commonly used PSW system have recommended placing a protective cover over the top of the walkway surface while it is in use to protect the walkway from damage.9 However, a recent study3 showed that temporospatial gait data and GRFs varied, with limited correlations between GRFs, when dogs were walked on a vinyl mat versus either of 2 polyvinyl chloride yoga mats of different thicknesses. Given these findings, it has been recommended that data not be compared between studies if the same cover is not used and that the same cover type be used at all participating institutions when performing multi-institutional studies. Although this previous study evaluated 3 commonly used cover types, it did not evaluate how data obtained with a protective cover in place would compare with data obtained when no cover was used, as is commonplace at some institutions.
Therefore, the objective of the study reported here was to determine whether use of a protective cover would affect temporospatial gait or GRF measurements obtained from dogs walking on a validated PSW. We hypothesized that measured temporospatial gait and GRF measurements would be significantly lower with the use of a cover than with no cover in place and that there would be only moderate correlations between the values obtained.
Materials and Methods
The study protocol was approved by the institutional animal care and use committee, and owners of all dogs enrolled in the study provided written informed consent. Healthy client-owned dogs that weighed between 25 and 40 kg were ≥ 2 years of age and free from any current or historical orthopedic or neurologic diseases were eligible for enrollment in the study. A physical examination, including orthopedic and neurologic examinations, was performed by a board-certified veterinary surgeon on all dogs considered for enrollment, and dogs with any evidence of orthopedic or neurologic disease were excluded. Additionally, dogs were excluded if any abnormalities were identified during gait analysis. Age, breed, body weight, and sex of participating dogs were recorded.
A crossover study design was used. Temporospatial gait measurements and GRFs were obtained with a PSW system (HRV Walkway 6 VersaTek System; Tekscan Inc) previously validated for clinical use. Dogs were walked across the PSW with no cover and with a protective cover (1/8-inch classic yoga mat; YogaAccessories.com) in place in random order, as determined by a coin toss, by a single handler. The PSW was calibrated in accordance with the manufacturer's instructions with a phantom and known weight.9 Separate calibration files were created with and without the cover in place, and calibration files were loaded for each dog on the basis of whether the cover was in place or not.
Prior to data collection, dogs were allowed to acclimate to the room where the PSW was located and were walked across the PSW multiple times until comfortable. Each dog completed multiple trials until results from a total of 5 valid trials with a cover and 5 valid trials without a cover over the PSW were collected. Trials were considered valid if dogs maintained a comfortable walk with a velocity of 0.8 to 1.3 m/s and acceleration < 0.5 m/s2, dogs were not pulling on the leash, the dog's head did not veer from midline, and they maintained a straight-line walk throughout the trial. Once the initial 5 valid trials were obtained, dogs were given a break of 10 minutes. The same process was then completed with the cover present or removed, as determined by the previous randomization. Data from the 5 valid trials were averaged by the PSW software, providing a single value for each condition that was used for statistical analyses.
Statistical analysis
All statistical analyses were performed with commercially available software (Excel version 16.43; Microsoft Corp). Data were tested for normality with the Shapiro-Wilk test, and paired Student t tests were used to compare values obtained with versus without the cover in place. Values of P ≤ 0.05 were considered significant. The Bland-Altman method was used to test for systematic differences (ie, bias) between measurements obtained with versus without the cover in place, and 95% limits of agreement were calculated for each variable.
Results
Six dogs were initially enrolled in the study; however, 1 dog was excluded from statistical analyses after gait assessment revealed an asymmetry between the forelimbs. The remaining 5 dogs consisted of 4 Golden Retrievers (3 spayed females and 1 neutered male) and 1 sexually intact male Gordon Setter. Mean ± SD weight of the dogs was 34.2 ± 4.3 kg (range, 27.5 to 38.4 kg), and mean age was 5.6 ± 3.0 years (range, 2 to 7 years).
No significant differences were identified in temporospatial gait measurements obtained with versus without the cover in place (Table 1). For all variables, the bias was low and the 95% limits of agreement included 0.
Temporospatial gait measurements obtained for 5 dogs walked across a pressure-sensitive walkway with and without a cover in place over the walkway.
| Variable | Mean with cover | Mean with no cover | P valuea | Bias | Lower 95% LOAs | Upper 95% LOAs |
|---|---|---|---|---|---|---|
| Gait time (s) | 2.12 | 2.16 | 0.78 | −0.03 | −0.54 | 0.47 |
| Gait distance (m) | 2.35 | 2.40 | 0.62 | −0.04 | −0.37 | 0.29 |
| Gait velocity (m/s) | 1.12 | 1.12 | 0.95 | 0.00 | −0.13 | 0.13 |
| Gait cycle time (s) | 0.69 | 0.71 | 0.34 | −0.02 | −0.12 | 0.07 |
| Gait cadence (steps/min) | 91.4 | 86.8 | 0.24 | 3.83 | −9.66 | 17.32 |
Determined with a Student t test for paired values.
LOAs = Limits of agreement.
In contrast, significant differences were found between measurements obtained with versus without the cover in place for most GRFs, including peak pressure (kPa [N/m2]), absolute peak force (kg), peak force normalized to body weight (percentage of body weight), and impulse (percentage of body weight X seconds) for each of the 4 limbs, except the right hind limb (Table 2). In all instances, the bias was negative, indicating that measurements obtained with the cover in place were significantly lower than those obtained without a cover. No significant differences were found between measurements obtained with versus without the cover for the ratio of peak force of the forelimbs versus the hind limbs (P = 0.93), ratio of peak force of the left limbs versus the right limbs (P = 0.16), ratio of peak force of the left forelimb versus the right forelimb (P = 0.20), and ratio of peak force of the left hind limb versus the right hind limb (P = 0.55).
Ground reaction force measurements obtained for 5 dogs walked across a pressure-sensitive walkway with and without a cover in place over the walkway.
| Variable | Mean with cover | Mean with no cover | P valuea | Bias | Lower 95% LOAs | Upper 95% LOAs |
|---|---|---|---|---|---|---|
| Peak pressure (kPa) | ||||||
| Left forelimb | 329.2 | 588.2 | < 0.001 | −215.8 | −472.5 | 40.9 |
| Left hind limb | 185.0 | 375.6 | 0.01 | −158.8 | −362.9 | 45.3 |
| Right forelimb | 368.6 | 643.2 | < 0.001 | −228.8 | −515.1 | 57.4 |
| Right hind limb | 197.8 | 378.4 | 0.01 | −150.5 | −360.9 | 59.9 |
| Normalized peak force (% BW) | ||||||
| Left forelimb | 35.4 | 72.0 | < 0.001 | −30.5 | −67.3 | 6.3 |
| Left hind limb | 18.8 | 37.9 | 0.01 | −15.9 | −38.8 | 6.9 |
| Right forelimb | 38.3 | 71.3 | < 0.001 | −27.5 | −60.7 | 5.7 |
| Right hind limb | 18.3 | 35.7 | 0.02 | −14.6 | −36.8 | 7.7 |
| Absolute peak force (kg) | ||||||
| Left forelimb | 12.6 | 24.3 | < 0.001 | −9.8 | −20.7 | 1.1 |
| Left hind limb | 6.8 | 13.2 | 0.01 | −5.3 | −13.0 | 2.5 |
| Right forelimb | 13.7 | 24.1 | < 0.001 | −8.7 | −18.2 | 0.8 |
| Right hind limb | 6.6 | 12.4 | 0.02 | −4.8 | −12.3 | 2.7 |
| Impulse (% BW X s) | ||||||
| Left forelimb | 10.4 | 21.8 | < 0.001 | −9.5 | −21.0 | 1.6 |
| Left hind limb | 4.1 | 9.4 | 0.02 | −4.4 | −11.2 | 2.4 |
| Right forelimb | 10.7 | 20.7 | < 0.001 | −8.4 | −18.4 | 1.6 |
| Right hind limb | 4.2 | 3.2 | 0.34 | 0.8 | −2.7 | 4.2 |
| Peak force ratio | ||||||
| Forelimbs vs hind limbs | 2.14 | 2.16 | 0.93 | −0.01 | −0.62 | 0.60 |
| Left limbs vs right limbs | 0.97 | 1.04 | 0.16 | −0.06 | −0.22 | 0.11 |
| Left forelimb vs right forelimb | 0.97 | 1.03 | 0.20 | −0.06 | −0.25 | 0.13 |
| Left hind limb vs right hind limb | 1.05 | 1.09 | 0.55 | −0.03 | −0.27 | 0.21 |
BW = Body weight.
See Table 1 for remainder of key
Discussion
Results of the present study indicated that for this particular PSW system, GRF measurements obtained when dogs were walked over the PSW with a cover in place were significantly lower than the values obtained when dogs were walked over the PSW without a cover in place. On the other hand, temporospatial gait measurements obtained with versus without a cover in place did not differ significantly.
The lack of significant differences between temporospatial gait measurements, in combination with the relatively low bias for these measurements, indicates that use of a cover did not affect the PSW's ability to detect foot placements. These findings were in line with those of a previous study,3 which found no significant difference in temporospatial gait measurements obtained with 3 different walkway covers. Therefore, if temporospatial measurements are the only variables of interest, whether a cover is or is not used should not have any impact, and values can be compared across studies and institutions regardless of the type of protective cover used or whether a cover was used.
In contrast to our findings for temporospatial gait measurements, most GFR measurements obtained with a cover in place were significantly lower than values obtained without a cover. The use of a cover likely dampens the forces measured by the sensors, as the softness of the cover slows the impact of the foot against the sensors. In addition, use of a cover may result in slight changes in gait that could alter measured GFRs. Finally, the cover used in the present study was a rubber material that was less slick than the surface of the walkway, which may have also affected the gait. Importantly, changes in gait could lead to differences in braking-propulsion and medial-lateral forces; however, these forces could not be measured in the present study. Kinematic assessment could elucidate some of these changes in future studies.
It was not immediately clear why a significant difference in impulse of the right hind limb was not detected when measurements obtained with a cover were compared with measurements obtained without a cover. Although, to our knowledge, no study has examined the durability of various PSW covers over time or evaluated the role that mechanical breakdown of the cover may have of the measurements obtained, it seems possible that mechanical breakdown in a particular portion of the cover could alter measured forces for limbs that strike that particular portion of the cover. The cover used in the present study was visually inspected prior to each use and was determined to be free from any macroscopic breakdown or wear, but it is still possible that microscopic breakdown was present.
The present study had several limitations, with the small sample size being the most important. Thus, a follow-up study with a larger sample size would be beneficial as well as a study that includes dogs with wider variations in age, breed, and body weight. Still, the authors feel that the results of the study are clinically relevant. Another important limitation of the present study was that only a single cover type was used. This cover was selected because it has been used in previous studies.3,8 Given the variability in data collected with various cover types in a previous study,3 it seems likely that data obtained with other covers would also differ from data obtained without a cover.
Unfortunately, most published reports in which a PSW was used do not comment on whether a cover was used during data collection or what type of cover was used. The potential effect of a cover should be considered when comparing studies and even when comparing repeated measurements for individual dogs. To ensure standard data collection, institutions should determine a priori whether a cover will be used and should ensure that the same cover or same cover type be used for all studies and all follow-up visits. In the case of multi-institutional studies, the use of a cover and the type of cover should be consistent throughout all locations. Reports of future studies should specify whether a cover has been used and, if so, the type of cover used to facilitate comparisons among studies.
Acknowledgments
No third-party funding or support was received in connection with this study or the writing or publication of the manuscript. The authors declare that there were no conflicts of interest.
The authors thank Dr. Abigail Shoben for assistance with the statistical analysis.
References
- 1. ↑
Lascelles BD, Freire M, Roe SC, DePuy V, Smith E, Marcellin-Little D. Evaluation of functional outcome after BFX total hip replacement using a pressure sensitive walkway. Vet Surg. 2010;39(1):71–77.
- 2. ↑
Kalamaras AB, Aarnes TK, Moore SA, et al. Effects of perioperative saphenous and sciatic nerve blocks, lumbosacral epidural or morphine-lidocaine-ketamine infusion on postoperative pain and sedation in dogs undergoing tibial plateau leveling osteotomy. Vet Anaesth Analg. 2021;48(3):415–421.
- 3. ↑
Kieves NR, Juliette HL, Evans R, Duerr FM. Comparison of three walkway cover types for use during objective canine gait analysis with a pressure-sensitive walkway. Am J Vet Res. 2019;80(3):265–269.
- 4.
Kirtley C. The temporal-spatial parameters. In: Clinical Gait Analysis: Theory and Practice. Elsevier; 2006;15–37.
- 5.
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(2):277–282.
- 6.
Light VA, Steiss JE, Montgomery RD, Rumph PF, Wright JC. Temporal-spatial gait analysis by use of a portable walkway system in healthy Labrador Retrievers at a walk. Am J Vet Res. 2010;71(9):997–1002.
- 7.
Le Quang T, Maitre P, Colin A, Viguier E. Spatial, temporal and kinetic evaluation of normal cats at walk, using a pressure walkway. Comput Methods Biomech Biomed Engin. 2008;11(suppl 1):137–138.
- 8. ↑
Oosterlinck M, Pille F, Huppes T, Gasthuys F, Back W. Comparison of pressure plate and force plate gait kinetics in sound Warmbloods at walk and trot. Vet J. 2010;186(3):347–351.
