Gait analysis plays an intrinsic role in veterinary orthopedics as a means to diagnose and monitor lameness, thereby allowing detection of disease and comparison of treatments.1 Visual assessment remains the mainstay of gait analysis in clinical settings but is flawed by subjectivity and lack of sensitivity to detect subtle changes.2 Objective gait analysis in veterinary medicine has traditionally focused on kinetics (ie, ground reaction forces).3,4 The use of force platforms has been found to be accurate for evaluation of limb function in dogs as well as for monitoring changes in forces due to disease or treatment.5–11 Studies12,13 have documented the impact of factors inducing inherent variations in ground reaction forces, such as trial repetition, size of the dog, velocity, acceleration, and handler. Therefore, ground reaction forces are typically measured with a single force platform with standardization of handler, velocity, and dog size. These conditions limit the application of kinetic gait analysis in clinical practice. In addition, the technique requires that a single foot contact the plate, thereby recording the forces applied on a single limb at each pass. At low to moderate speeds, dogs tend to maintain > 1 foot in contact with the ground, resulting in overlap of feet on the force platform.14 Multiple passes are required to collect valid data for multiple footfalls of 1 foot or obtain measurements for multiple limbs. Therefore, kinetic studies15 dealing with symmetry indices or compensatory mechanisms rely on the assumption that consecutive strides are identical. Moreover, stride length constraints complicate data acquisition from small dogs.13 Overall, these limitations have prompted the development of alternative tools for objective gait assessment. Among these, treadmills instrumented with force platforms have recently been used for continuous measurement of ground reaction forces generated by all limbs of dogs during consecutive foot strikes.16 The main advantages of this method, compared with overground studies, are the limited space required and the ability to control the dogs’ speed, thereby reducing variability in measurements. However, dogs must be acclimated to locomotion on a treadmill and the resulting gait may differ from a natural overground walk or trot.17,18 These limitations, combined with the cost of the equipment, hinder the application of instrumented treadmills for canine gait analysis in clinical practices.
Pressure walkway systems have recently become available to measure spatiotemporal gait characteristics of humans and other animals. Several studies19–26 have validated their use to study normal gait as well as gait changes due to orthopedic or neurologic conditions. Although these pressure-sensing walkways do not measure mediolateral or craniocaudal forces during locomotion, they allow simultaneous recording of data for all limbs of dogs of various sizes over consecutive foot strikes during a single pass. They also measure actual foot velocity (without photoelectric cells), static weight distribution, and distribution of forces across canine foot pads.27 Among these pressure-sensing walkways, 1 portable gait analysis systema for automated measurement of spatiotemporal gait variables, pressure distribution, and symmetry indices between limbs28 is commonly used in human orthopedic and neurologic studies29,30; however, several publications24,25,31,32 support its validity for use with dogs, especially dogs of medium to large breeds. The veterinary medical literature is more limited with regard to its use with small-breed dogs. One brief report32 describes symmetry indices in young clinically normal small dogs and in old small dogs with osteoarthritis. A more recent publication23 combined data collected from different pressure-sensing walkways, including the portable gait analysis system, to study the variation of spatiotemporal variables in dogs < 46 inches in height with neurologic disease, mostly Dachshunds. The primary objective of the study reported here was to characterize the gait of small-breed dogs walked on the portable gait analysis system. Although automated data collection and portability make the portable gait analysis system especially attractive for use in the clinical setting and in multicenter trials, the sources of variation in data recorded with this system remain unclear. The secondary objective of the study reported here was to determine the impact of handlers and leash side on the variables recorded with the portable gait analysis system. We hypothesized that leash side, but not handlers, would have a significant impact on spatiotemporal gait variables in small-breed dogs.
Materials and Methods
Animals—Five healthy adult small-breed dogs (< 11.4 kg [25 lb]) were enrolled in the study, which was approved by our institutional animal care and use committee. The dogs had no history of lameness and were accustomed to being walked on a leash. Additional criteria for inclusion included a normal body score (ie, 4 to 6/9) and no detectable abnormality on orthopedic and neurologic examinations. Age and body weight of each dog were recorded. With each dog in a standing position, the length of the RF was measured from the floor to the highest point of the shoulders (withers)23; the length of the RH was measured in standing position, from the floor to the dorsal aspect of the lumbar vertebrae, at the level of the hips; the width of the shoulders was measured as the distance between the lateral aspects of the scapulae; and the width of the hips was measured as the distance between the greater trochanters. Absolute values for gait analysis were obtained for both the right and left limb, but only those for the right limbs are reported; however, symmetry indices inherently include the left limb data. We elected to focus on the right limbs because the left and right limbs of a given dog are not independent and there was no reason to believe that conclusions would differ if we also analyzed left limb data because all study dogs were clinically normal.
Handlers—Five healthy adult humans between 18 and 49 years of age were recruited for the study, with approval from our institutional review board. To participate in the study, the handlers had to be comfortable around dogs, have a normal body mass index, and have no history of orthopedic or neurologic disorders.
Equipment—A 4.3 × 1.2-m human pressure matb was placed on a flat surface to record the velocity and cadence of the human subjects (Figure 1). The mat contained a 16-level pressure-sensing pad and circuitry between a base layer and a top linoleum cover. Approximately 5 m of floor space was used on either end of the walkway to accustom the handler to the metronome prior to recording gait cycle. A camera was positioned at the end of the walkway system so that digital video files of each pass across the walkway system were automatically linked to the data files recorded with a computer and software.c
Photograph of pressure walkways (5.8 × 0.6-m portable dog pressure mat placed over a 4.3 × 1.2-m human pressure mat) used to record simultaneously gait characteristics of small-breed dogs and their handlers.
Citation: Journal of the American Veterinary Medical Association 246, 11; 10.2460/javma.246.11.1215
All dogs were walked over a 5.8 × 0.6-m portable canine pressure mata placed over the human pressure walkway (Figure 1). Each end of the mat consisted of an inactive transition surface, to allow acclimation of dogs to the surface. The central, active section of the mat measured 4.9 × 0.6 m and contained 18,432 encapsulated, 1-cm-wide, square sensors. The walkway system interfaced with a computer and software program for processing and storage of raw data. The sensors of the mat had 8 equal switching levels, each interpreted by the softwared as a change in pressure on the sensor. Another camera was positioned at the end of the walkway system linked to the canine mat to simultaneously record movement. Digital video files of each pass across the walkway system were automatically linked to the data file for footfall verification.
Experimental design—Dogs and handlers were allowed to acclimate to the portable walkway system and their surroundings by walking back and forth on the mat a minimum of 5 times prior to recording. All handlers walked to the beat of a metronome set at 100 beats/min throughout the study. Each dog was walked on a loose leash by each of the 5 handlers in 1 session, alternating leash side (ie, position of leash and handler relative to the dog; for example, right leash side indicated that the handler and the leash were on the dog's right side) with each pass. Each pass involved a dog walking the length of the portable walkway system in 1 direction and consisted of multiple gait cycles. A pass was considered valid if the dog walked on the pressure mat with a relaxed, steady gait, without pulling on the leash or obvious deviation of the head from the midline, and a minimum of 3 sensors were activated for each footprint. A minimum of 3 valid passes were obtained for each dog walked under each test condition (handler and leash side), requiring a minimum of 6 passes across the portable walkway system with each handler. All trials were recorded on the same day for a given dog.
Data analysis—Velocity and cadence of handlers were collected from the human pressure walkway system. Data collected to evaluate the gait of the dogs included velocity, cadence, ST, number of sensors activated, TPI, and LH and RH reach (distance that a hind limb is placed relative to the previous forelimb placement on the same side). Symmetry indices between left and right limbs were calculated for ST and TPI.
Mean, SD, and CV (ratio between SD and mean expressed as a percentage) were calculated for each variable. A mixed model for repeated-measures ANOVA and post hoc Tukey highly significant difference test were used to analyze differences in velocity between dogs and handlers throughout the study. The same tests were used to explore differences in each variable (velocity, TPI-RF, TPI-RH, ratio between the TPIs of the forelimbs [TPI-LF:TPI-RF], ratio between the TPIs of the hind limbs [TPI-LH:TPI-RH], ratio between the STs of the forelimbs [ST-LF:ST-RF], and ratio of the STs of the hind limbs [ST-LH:ST-RH]) within dogs and between handlers (to test the effect of handlers) and between trials obtained for a dog with the same handler holding the leash on the right versus left side (to test the effect of leash side). Multiple regressions were used to explore the percentage of variation attributed to leash or handler for each variable. Values of P ≤ 0.05 were considered significant. The effects of leash side (independent variable, left vs right) were further explored for the following dependent variables with an independent samples t test: ratio of the TPIs of left versus right limbs (TPI-L:TPI-R), number of sensors (ratio between left and right limbs [NS-L:NS-R], left and right forelimbs [NS-LF:NS-RF], and left and right hind limbs [NS-LH:NS-RH]), ratio of the ST of left versus right limbs (ST-L:ST-R), and reach (LH and RH).
Results
The 5 dogs enrolled in this study ranged in age from 2 to 8 years. Mean ± SD weight for the 5 dogs was 6.8 ± 3.3 kg [14.96 ± 7.26 lb]; Table 1). All dogs had no abnormal findings on orthopedic examinations and were accustomed to walking on a leash. Each dog was assigned a number (1 through 5). Three female and 2 male handlers volunteered for the study. Handler age ranged from 26 to 46 years old (mean ± SD, 31.4 ± 8.9 years). Handler height ranged from 156.2 to 175.2 cm (mean ± SD, 164.8 ± 8.1 cm), and weight ranged from 49.4 to 76 kg (108.68 to 167.2 lb; mean ± SD, 66.6 ± 11.8 kg [146.53 ± 25.96 lb]). Mean ± SD cadence of handlers walking to a metronome set at 100 beats/min was 105 ± 3.6 steps/min.
Age, breed, sex, body weight, and anatomic variables of 5 healthy adult small-breed dogs used to characterize the gait of small-breed dogs walked on a pressure walkway at a metronome-set tempo and to determine the influence of handler and leash side on gait characteristics.
| Dog | ||||||
|---|---|---|---|---|---|---|
| Variable | 1 | 2 | 3 | 4 | 5 | Mean ± SD |
| Age (y) | 2 | 8 | 6 | 2 | 3 | NA |
| Breed | Maltese-Poodle cross | Dachshund-Corgi cross | Beagle | Dachshund | Chihuahua | NA |
| Sex | FS | MN | FS | MN | MN | NA |
| Weight (kg) | 4.81 | 8.09 | 10.63 | 8.09 | 2.18 | 6.76 ± 3.29 |
| RF length (cm) | 29.0 | 27.3 | 33.5 | 22.0 | 20.0 | 26.36 ± 5.44 |
| RH length (cm) | 30.0 | 30.3 | 34.0 | 24.5 | 22.5 | 28.26 ± 4.68 |
| Shoulder width (cm) | 13.0 | 16.5 | 15.0 | 17.0 | 11.0 | 14.50 ± 2.50 |
| Hip width (cm) | 15.0 | 17.8 | 16.5 | 17.5 | 12.0 | 15.76 ± 2.37 |
FS = Female, spayed. MN = Male, neutered. NA = Not applicable.
All dogs maintained similar (P = 0.7) velocities (1.3 ± 0.14 m/s) throughout the study, with a mean ± SD cadence of 177.6 ± 2.1 steps/min. A significant (P < 0.001) interaction between handlers and dogs was detected, and differences in velocities between handlers were identified in 3 dogs (dogs 2, 4, and 5). However, the maximum difference between the mean velocities within dogs and between handlers did not exceed 0.4 m/s. Mean ± SD and CV for spatiotemporal gait variables and symmetry indices for trials of all dogs under all conditions (leash side and handlers) were summarized (Table 2). The CVs of TPIs were > 30% regardless of the limb but decreased to < 20% when the same variables were expressed as symmetry indices. Data regarding the same variables were also collected with a minimum of 3 valid passes standardized by leash side and handler, within each dog (Table 3). These values represented the variation inherent to each dog walked several times under standardized conditions.
Spatiotemporal gait characteristics obtained by use of a portable walkway system for 5 healthy adult small-breed dogs (described in Table 1) each walked on a pressure walkway at a metronome-set tempo by each of 5 handlers with a leash on the right and left sides (n = 150 valid trials).
| Variable | Mean ± SD | 95% CI | CV |
|---|---|---|---|
| TPI | |||
| RF (%) | 29.78 ± 10.76 | 28.05 to 31.52 | 0.36 |
| RH (%) | 17.54 ± 5.47 | 16.66 to 18.42 | 0.31 |
| Ratio between forelimbs (TPI-LF:TPI-RF) | 0.99 ± 0.13 | 0.97 to 1.01 | 0.13 |
| Ratio between hind limbs (TPI-LH:TPI-RH) | 1.08 ± 0.18 | 1.05 to 1.11 | 0.16 |
| ST | |||
| 1RF (s) | 0.19 ± 0.04 | 0.18 to 0.20 | 0.21 |
| RH (s) | 0.17 ± 0.04 | 0.16 to 0.17 | 0.24 |
| Ratio between forelimbs (ST-LF:ST-RF) | 1.01 ± 0.08 | 1.00 to 1.02 | 0.07 |
| Ratio between hind limbs (ST-LH:ST-RH) | 0.98 ± 0.13 | 0.96 to 1.00 | 0.13 |
| Stride length (ratio between L and R limbs) | 1.00 ± 0.14 | 1.00 to 1.00 | 0.14 |
| No. of sensors | |||
| Ratio between forelimbs (NS-LF:NS-RF) | 1.00 ± 0.11 | 0.98 to 1.02 | 0.11 |
| Ratio between hind limbs (NS-LH:NS-RH) | 1.08 ± 0.15 | 1.05 to 1.10 | 0.14 |
| Reach | |||
| LH (cm) | −5.75 ± 4.39 | −6.46 to −5.04 | −0.76 |
| RH (cm) | −5.44 ± 4.30 | −6.14 to −4.75 | −0.79 |
Five healthy adult humans (18 to 49 years of age) with normal body mass indices and no history of orthopedic or neurologic disorders were used as handlers. A human pressure mat was placed on a flat surface to record the velocity and motion cadence of the human subjects. A camera was positioned at the end of the walkway system so that digital video files of each pass across the walkway system were automatically linked to the recorded data files. All dogs were walked over a portable canine pressure mat placed over the human pressure walkway. Another camera was positioned at the end of the walkway system linked to the canine mat to simultaneously record movement. Digital video files of each pass across the walkway system were automatically linked to the data file for footfall verification. Dogs and handlers were allowed to acclimate to the portable walkway system and their surroundings by walking back and forth on the mat a minimum of 5 times prior to recording. All handlers walked to a standardized metronome tempo set at 100 beats/min throughout the study. Each dog was walked on a loose leash by each of the 5 handlers in 1 session, alternating leash side (ie, position of leash and handler relative to the dog; for example, right leash side indicated that the handler and the leash were on the dog's right side) with each pass. Each pass involved a dog walking the length of the portable walkway system in one direction and consisted of multiple gait cycles. A pass was considered valid if the dog walked on the pressure mat with a relaxed, steady gait, without pulling on the leash or obvious deviation of the head from the midline, and a minimum of 3 sensors were activated for each footprint. A minimum of 3 valid passes were obtained for each dog walked under each test condition (handler and leash side), requiring a minimum of 6 passes across the portable walkway system with each handler. All trials were recorded on the same day for each given dog. Absolute values for gait analysis were obtained for both right and left limbs, but only those for the right limbs are reported; however, symmetry indices inherently include the left limb data.
Coefficient of variation for spatiotemporal gait characteristics obtained on a minimum of 3 valid passes over a portable walkway system in 5 healthy adult small-breed dogs (described in Table 1) under standardized conditions (same handler and leash side).
| Variable | CV |
|---|---|
| TPI | |
| RF | 0.1 |
| RH | 0.1 |
| Ratio between forelimbs (TPI-LF:TPI-RF) | 0.1 |
| Ratio between hind limbs (TPI-LH:TPI-RH) | 0.15 |
| ST | |
| RF | 0.10 |
| RH | 0.10 |
| Ratio between forelimbs (ST-LF:ST-RF) | 0.05 |
| Ratio between hind limbs (ST-LH:ST-RH) | 0.09 |
| Stride length (ratio between L and R limbs) | 0.01 |
| No. of sensors (NS) | |
| Ratio between forelimbs (NS-LF:NS-RF) | 0.07 |
| Ratio between hind limbs (NS-LH:NH-RH) | 0.12 |
| Reach | |
| LH | −0.35 |
| RH | 0.48 |
See Table 2 for key.
Influence of handlers—An interaction was detected between handlers and the TPI of the RF, with up to 8% change in TPI-RF of the same limb being induced by handlers. No significant difference was identified for TPI-LF:TPI-RF, TPI-RH, or TPI-LH:TPI-RH and ST-LH:ST-RH when dogs were walked by different handlers. Percentage of variation in each spatiotemporal variable due to handlers was summarized (Table 4).
Percentage of variation induced by handlers and leash side and the P value calculated for the model for the spatiotemporal variables obtained by use of a portable walkway system in 5 healthy adult small-breed dogs (described in Table 1).
| Variable | Variation due to handler (%) | Variation due to leash side (%) | P value |
|---|---|---|---|
| TPI | |||
| RF | 1.7 | 0.13 | 0.766 |
| RH | 0.59 | 0.23 | 0.858 |
| Ratio between forelimbs (TPI-LF:TPI-RF) | 2.04 | 11.63 | < 0.001 |
| Ratio between hind limbs (TPI-LH:TPI-RH) | 1.53 | 0.48 | 0.52 |
| ST | |||
| RF | 5.60 | 0.52 | 0.104 |
| RH | 3.73 | 1.37 | 0.115 |
| Ratio between forelimbs (ST-LF:ST-RF) | 1.81 | 2.25 | 0.469 |
| Ratio between hind limbs (ST-LH:ST-RH) | 1.76 | 0.14 | 0.6 |
| Stride length (ratio between L and R limbs) | 3.23 | 0.29 | 0.248 |
| No. of sensors | |||
| Ratio between forelimbs (NS-LF:NS-RF) | 2.39 | 14.44 | < 0.001 |
| Ratio between hind limbs (NS-LH:NS-RH) | 1.14 | 0.50 | 0.721 |
| Reach | |||
| LH | 0.19 | 0.05 | 0.992 |
| RH | 0.66 | 0.03 | 0.980 |
See Table 2 for key.
Influence of leash side—Significant interactions were identified between leash side and several variables involving the forelimbs. The most obvious effect involved the pressure distribution between the forelimbs, where approximately 12% of variation in TPI-LF:TPI-RF was attributed to leash side (Table 4). The mean of this ratio was 0.95 when the leash was on the right side of the dogs, compared with a mean of 1.03 when the leash was on the left side of the dogs (P < 0.001; Table 5). Similar differences were identified for the number of sensors activated by each limb: mean ± SD NS-LF:NS-RF was 0.95 ± 0.1 when the leash was on the right side of the dogs and 1.04 ± 0.1 when the leash was on the left side of the dogs (P < 0.001). Leash side did not influence any of the variables characterizing the hind limbs.
Mean ± SD spatiotemporal variables obtained by use of a portable walkway system for 5 healthy adult small-breed dogs (described in Table 1) evaluated with the leash on the left and right side and difference (left – right) in variables between sides to determine the influence of leash side.
| Leash side | |||
|---|---|---|---|
| Variable | Left | Right | Difference (absolute value [%]) |
| TPI | |||
| RF | 29.34 ± 9.71 | 30.15 ± 11.60 | −0.81 (–3) |
| RH | 17.81 ± 4.95 | 17.31 ± 5.89 | 0.50 (3) |
| TPI-L:TPI-R | 0.99 ± 0.1 | 1.04 ± 0.1 | −0.05 (–5)* |
| TPI-LF:TPI-RF | 0.94 ± 0.1 | 1.03 ± 0.1 | −0.09 (–9)* |
| TPI-LH:TPI-RH | 1.07 ± 0.1 | 1.09 ± 0.2 | −0.02 (–2) |
| ST | |||
| RF | 0.19 ± 0.04 | 0.19 ± 0.04 | 0.00 (0) |
| RH | 0.16 ± 0.04 | 0.17 ± 0.04 | −0.01 (–6) |
| ST-LF:ST-RF | 1.02 ± 0.1 | 1.00 ± 0.1 | 0.02 (2) |
| ST-LH:ST-RH | 0.98 ± 0.1 | 0.99 ± 0.1 | −0.01 (–1) |
| Stride length (ratio between L and R limbs) | 1.00 ± 0.01 | 1.00 ± 0.01 | 0.00 (0) |
| No. of sensors | |||
| NS-LF:NS-RF | 0.95 ± 0.1 | 1.04 ± 0.1 | −0.09 (–9)* |
| NS-LH:NS-RH | 1.06 ± 0.1 | 1.09 ± 0.2 | −0.03 (–3) |
Differences are indicated as absolute values and percentages of the ratio (ie, [right – left]/left).
For a given variable, the difference between leash sides was significantly (P < 0.05) different from 0.
See Table 2 for key.
Discussion
The present study was conducted to characterize the gait of small-breed dogs walked on a pressure walkway by handlers moving at a metronome-set tempo and to determine the influence of handler and leash side on gait characteristics. The main study findings were that symmetry indices varied less than ST and TPI of a single limb, that changing handlers may have influenced the TPI of a given forelimb but did not impact symmetry indices, and that changing leash side influenced symmetry indices of the dogs’ forelimbs but not their hind limbs.
The spatiotemporal variables, pressure, and symmetry indices for the small dogs in our study were similar to those published for large dogs,24 although, to our knowledge, this is the first report of CIs for values determined for healthy dogs. The CV of TPI and ST on individual forelimbs and hind limbs in all passes obtained in the study (n = 150/limb) exceeded 30% and 20%, respectively. These values corroborated those reported for 56 Labrador Retrievers walked by the same handler, with 3 valid passes/dog.24 In the present study, CVs decreased to approximately 10% when calculated on the basis of 3 valid passes obtained with the same handler and same leash side, representing intertrial variation inherent to individual dogs. This degree of variation was similar to that observed in 21 clinically normal dogs < 46 inches in height, based on 3 valid passes/dog, with the same handler.23 On the basis of the present study findings, the variability of TPI and ST for individual limbs seemed to be influenced by the number of passes obtained per dog or number of dogs tested. In contrast, symmetry indices varied by approximately 10% under standardized conditions but their CV remained < 20% when all passes (n = 150/limb) were included in the analysis. The use of symmetry indices, rather than absolute measurement on individual limbs, appears advantageous, especially in studies with a large number of trials.
Kinetic studies in dogs traditionally rely on the same handler to walk or trot dogs for all trials on overground force plates or pressure walkways, to reduce variation. However, this approach limits the application of gait analysis in large clinical settings where multiple staff members are trained to collect data. Similarly, multicenter trials inherently require that data be collected by multiple investigators. In our study, criteria for inclusion of our handlers were selected to ensure that they would have a normal gait and could safely walk dogs on a leash. Age range was controlled to avoid the impact of aging on the human gait.33 A metronome was used to standardize the cadence of handlers’ motion because of its accessibility and ease of use in practice. The rate of the metronome was selected on the basis of preliminary experiments to ensure that the velocity of small dogs would be consistent with a walk and that handlers could maintain a natural gait. The results confirmed that handlers maintained a motion cadence similar to the metronome rate and that this approach was effective in controlling the velocity of dogs.
The only influence of handlers detected in the present study was an effect on the TPI of the forelimb. A significant difference among handlers was identified, with a magnitude of handler-induced variation reaching up to 8% in a given dog. Although the percentage of variation in pressure index or peak vertical force indicative of a relevant alteration in gait remains unclear, this magnitude of change has been considered as meaningful in clinical studies.34,35 This threshold implies that, depending on the trial, the gait of an individual dog may be falsely considered as altered, when in fact, the variation reflects a change in handler. This finding differed from that reported by Jevens et al,12 where the authors concluded that multiple handlers could be included in a study without appreciable influence on data. In that study,12 handlers contributed 0% to 7% of the variance observed, which is similar to 1.7% of variation in TPI attributable to handlers in the present study (Table 4). The difference in conclusions most likely stems from the statistical method, considering that we compared data obtained within dogs but with different handlers. In contrast, none of the symmetry indices were significantly influenced by the handlers in our study. The use of these indices is recommended in studies of the forelimbs in which the handler cannot remain constant throughout all trials. The evaluation of symmetry between limbs has previously been recommended over the comparison of absolute gait variables on individual limbs because of their ease of calculation and interpretation.25
The position of the leash relative to the dog influenced data collected for the forelimbs but not data collected for the hind limbs. Leash side contributed to approximately 14% and 12% of the variation in symmetry indices of TPI and number of activated sensors (representative of the area in contact with the paw) between the forelimbs, respectively. The changes observed were consistent with a shift in weight toward the forelimb opposite to the leash, despite a lack of visual evidence of tension on the leash. This shift was not detected when individual TPIs were compared between leash sides. This discrepancy is most likely due to the fact that symmetry indices reflect the combined effects of decreased loading of a limb and compensatory changes on the contralateral limb. In that regard, these indices may be more sensitive for detection of subtle gait alteration than are measurements obtained for single limbs. The magnitude of the difference between symmetry indices of TPI and number of activated sensors collected on the forelimbs with the leash on the left versus the right side was 9%. This value could be considered as clinically relevant on the basis of previous studies34,35 and could lead to a false diagnosis of gait alteration, if leash side was kept constant among trials. Including an equal number of left-sided leash-led and right-sided leash-led trials, along with evaluation of absolute variables, may mitigate the risk of a false diagnosis when studying gait alterations of the forelimbs in small-breed dogs. To our knowledge, the impact of leash side on canine gait characteristics has not been reported before and may vary with dog size and gait status.
The scope of our study was limited to small-breed clinically normal dogs that were walked by a controlled population of handlers. The study results may not apply to humans and dogs that did not meet our criteria for inclusion. In addition, all trials were recorded on the same day for each dog, eliminating interday variation. Trial repetitions and interweek data collection have been identified as the main sources of variation in ground reaction forces, when speed, handler, gait, weight, and dog breeds were controlled.36 The influence of trial repetition was evaluated in the present study to serve as a reference for the magnitude of changes induced by handlers and leash sides. However, interday variation was eliminated from the study design to maximize the detection of effects due to handlers and leash sides. This factor would be expected to compound the effects of handlers and leash sides and further increase measurement variability in longitudinal studies.
To our knowledge, the present study is the first to provide CIs for gait characteristics of small-breed dogs walked by handlers moving at a metronome-set tempo on a pressure walkway. Symmetry indices seemed to vary less than variables obtained for individual limbs and may be advantageous in studies with large number of trials. Among the data for dogs evaluated in the present study, data collected for hind limbs were not influenced by handlers or leash side. Symmetry indices were not affected by changes in handlers between 18 and 49 years of age, and use of those indices are therefore recommended in studies of dogs’ forelimbs requiring multiple handlers. The study findings revealed that small-breed dogs tended to shift weight toward the forelimb opposite to the leash, thereby inducing changes in symmetry indices of the forelimbs. Studies focusing on the forelimbs should aim for equal distribution of left-sided leash-led and right-sided leash-led trials and include variables obtained for individual dog limbs. Further studies are warranted to confirm these findings in lame dogs.
ABBREVIATIONS
| CI | Confidence interval |
| CV | Coefficient of variation |
| LF | Left forelimb |
| LH | Left hind limb |
| RF | Right forelimb |
| RH | Right hind limb |
| ST | Stance time |
| TPI | Total pressure index |
GAITRite, platinum version 16-foot Portable Walkway, CIR Systems Inc, Sparta, NJ.
Zeno Walkway, ProtoKinetics, Havertown, Pa.
ProtoKinetics Movement Analysis Software (PKMAS), ProtoKinetics, Havertown, Pa.
GAITFour software, version 4.4, CIR Systems Inc, Sparta, NJ.
References
1. McLaughlin RM. Kinetic and kinematic gait analysis in dogs. Vet Clin North Am Small Anim Pract 2001; 31:193–201.
2. 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.
3. Gillette RL, Angle TC. Recent developments in canine locomotor analysis: a review. Vet J 2008; 178:165–176.
4. Griffon DJ. Canine gait analysis: a decade of computer assisted technology. Vet J 2008; 178:159–160.
5. Budsberg SC. Long-term temporal evaluation of ground reaction forces during development of experimentally induced osteoarthritis in dogs. Am J Vet Res 2001; 62:1207–1211.
6. 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.
7. 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.
8. 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.
9. Roush JK, McLaughlin RM Jr. Effects of subject stance time and velocity on ground reaction forces in clinically normal Greyhounds at the walk. Am J Vet Res 1994; 55:1672–1676.
10. Rumph PF, Lander JE, Kincaid SA, et al. Ground reaction force profiles from force platform gait analyses of clinically normal mesomorphic dogs at the trot. Am J Vet Res 1994; 55:756–761.
11. Rumph PF, Kincaid SA, Visco DM, et al. Redistribution of vertical ground reaction in dogs with experimentally induced chronic hind limb lameness. Vet Surg 1995; 24:384–389.
12. 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.
13. 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.
14. Lee DV, Bertram JE, Todhunter RJ, et al. Force overlap in trotting dogs: a Fourier technique for reconstructing individual limb ground reaction force. Vet Comp Orthop Traumatol 2002; 15:223–227.
15. 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.
16. Ragetly CA, Griffon DJ, Hsu MK, et al. Kinetic and kinematic analysis of the right hind limb during trotting on a treadmill in Labrador Retrievers presumed predisposed or not predisposed to cranial cruciate ligament disease. Am J Vet Res 2012; 73:1171–1177.
17. Torres BT, Moëns NM, Al-Nadaf S, et al. Comparison of overground and treadmill-based gaits of dogs. Am J Vet Res 2013; 74:535–541.
18. Fanchon L, Valette JP, Sanaa M, et al. The measurement of ground reaction force in dogs trotting on a treadmill: an investigation of habituation. Vet Comp Orthop Traumatol 2006; 19:81–86.
19. 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.
20. 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.
21. Webster KE, Wittwer JE, Feller JA. Validity of the GAITRite walkway system for the measurement of averaged and individual step parameters of gait. Gait Posture 2005; 22:317–321.
22. Youdas JW, Hollman JH, Aalbers MJ, et al. Agreement between the GAITRite walkway system and a stopwatch-footfall count method for measurement of temporal and spatial gait parameters. Arch Phys Med Rehabil 2006; 87:1648–1652.
23. Gordon-Evans WJ, Evans RB, Conzemius MG. Accuracy of spatiotemporal variables in gait analysis of neurologic dogs. J Neurotrauma 2009; 26:1055–1060.
24. 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.
25. LeQuang T, Maitre P, Roger T, et al. Is a pressure walkway system able to highlight a lameness in dog? J Anim Vet Adv 2009; 8:1936–1944.
26. Chien SL, Lin SZ, Liang CC, et al. The efficacy of quantitative gait analysis by the GAITRite system in evaluation of parkinsonian bradykinesia. Parkinsonism Relat Disord 2006; 12:438–442.
27. Besancon MF, Conzemius MG, Evans RB, et al. Distribution of vertical forces in the pads of Greyhounds and Labrador Retrievers during walking. Am J Vet Res 2004; 65:1497–1501.
28. Bilney B, Morris M, Webster K. Concurrent related validity of the GAITRite walkway system for quantification of the spatial and temporal parameters of gait. Gait Posture 2003; 17:68–74.
29. Dusing SC, Thorpe DE. A normative sample of temporal and spatial gait parameters in children using the GAITRite electronic walkway. Gait Posture 2007; 25:135–139.
30. Parmar V, Shyam Kumar AJ, Harper WM. Reliability of the GAITRite walkway system for the quantification of temporospatial parameters of gait in young and older people (lett). Gait Posture 2006; 23:523–525.
31. Viguier E, LeQuang T, Maitre P, et al. The validity and reliability of the GAITRite system's measurement of the walking dog. Comput Methods Biomech Biomed Engin 2007; 10:113–114.
32. LeQuang T, Maitre P, Roger T, et al. The GAITRite system for evaluation of the spatial and temporal parameters of normal dogs at a walk. Comput Methods Biomech Biomed Engin 2007; 10:109–110.
33. Autenrieth CS, Karrasch S, Heier M, et al. Decline in gait performance detected by an electronic walkway system in 907 older adults of the population-based KORA-Age study. Gerontology 2013; 59:165–173.
34. 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.
35. Gordon-Evans WJ, Griffon 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; 243:675–680.
36. Nordquist B, Fischer J, Kim SY, et al. Effects of trial repetition, limb side, intraday and inter-week variation on vertical and craniocaudal ground reaction forces in clinically normal Labrador Retrievers. Vet Comp Orthop Traumatol 2011; 24:435–444.
