Working dogs often wear service vests. The function of these vests includes use in transportation of supplies and equipment, cooling of the dog, protection of the dog, or providing a harness for lifting of the dog. Although the uses of canine vests are diverse, limited information is available on how the design of these vests affects the general body movement and, more specifically, the movement of the trunk of a dog. There is also a lack of information describing how movement of the truncal region impacts the mechanical energy required of dogs during various gaits. Furthermore, few data exist regarding the relationship of canine truncal kinematics to gait patterns. A better comprehension of the relationship between truncal rotation and gait patterns is required if the overall impact of service vests on the kinematics of dogs is to be understood.
One model to quantify movement of the back of dogs involves creating vectors between vertebrae and then calculating rotation of one vector relative to another during a gait.1,2 However, movement of the vertebrae does not necessarily reflect trunk rotation. The relationship between trunk rotation and gait has been measured in horses. Results of these studies3,4 suggest that truncal rotation can have a major impact on animal gait. Investigators in 1 study3 used inertial sensors to track a horse's center of mass, which then defined the pattern of trunk rotation during a gait; their findings suggested a relationship between truncal rotation and forelimb stance phase. Investigators in another study4 used a cinematography technique to monitor the position of a horse's center of mass during turning and to quantify trunk rotation relative to the vertebral column.
The overall objective of the study reported here was to develop a 3-D kinematic model that could be used to measure truncal motion in dogs and assess changes in truncal motion in dogs when wearing each of 2 service vests. To achieve this objective, the first goal was to identify anatomic landmarks that could be used to establish vectors that would be acceptable positional reference points for quantifying the rotation of a dog's trunk during a gait cycle. Acceptable landmarks were defined as those that produced a vector or a set of vectors with magnitudes that remained nearly constant during the entire gait cycle. These acceptable landmarks would provide a vector system needed to create an LCS necessary to quantify rotation of a dog's trunk. Four hypotheses were tested for the first goal:
• Hypothesis 1: relative to the left iliac crest, the magnitude of vectors connecting vertebrae T1 and T13 will not remain constant as a dog walks and trots on a treadmill.
• Hypothesis 2: relative to the right iliac crest, the magnitude of vectors connecting vertebrae T1 and T13 will not remain constant as a dog walks and trots on a treadmill.
• Hypothesis 3: relative to the dorsal aspect of the left scapula, the magnitude of vectors connecting vertebrae T1 and T13 will not remain constant as a dog walks and trots on a treadmill.
• Hypothesis 4: relative to the dorsal aspect of the right scapula, the magnitude of vectors connecting vertebrae T1 and T13 will not remain constant as a dog walks and trots on a treadmill.
The second goal of the study reported here was to use the data from the first goal to assess changes in truncal motion in dogs when wearing each of 2 service vests. For this second goal, hypothesis 5 was tested: rotation of the trunk for a dog wearing a service vest will not be equal to the rotation of the trunk for a dog not wearing a service vest.
We anticipated that the obtained data would provide insights into the methods used to assess the impact that wearing a service vest will have on gait patterns of dogs. Data collection would be conducted over multiple days for each dog, which would lead to the final hypothesis (hypothesis 6): for each dog, rotation of the trunk will remain the same throughout all days of testing.
Materials and Methods
Animals
Five client-owned adult mixed-breed dogs were used in the study. All dogs were of nonchondrodystrophic breeds, were between 1 and 4 years of age, had a body weight between 20 and 30 kg, and had a body condition score of 5 (scale, 1 to 9). No dog had previously worn a vest of any type. Physical examination of the dogs revealed no abnormalities, and there were no signs of lameness. Informed consent was obtained from the owners for participation of their dogs in the study. The study protocol was reviewed and approved by the Hospital Clinical Research Committee.
Study design
Kinematic data were collected for each dog on 3 days within a 10-day period. All testing was conducted in the same research laboratory, and the same person handled the dog during all testing sessions. Dogs were allowed a brief period to acclimate to a treadmill, and reflective markers were then placed on each dog. Kinematic data were collected while dogs walked on the treadmill and subsequently trotted on the treadmill. Dogs were then acclimated to wearing a service vest, and data were collected while dogs walked and trotted on the treadmill with each dog wearing a custom vest and then with each dog wearing an adjustable vest. Each dog was returned to the laboratory 2 more times, and the same testing protocol was repeated.
Acclimation and testing protocol
On the day of a test session, the owner brought the dog to the laboratory (only 1 dog was tested each day). Dogs were not tethered to the treadmill at any time. Dogs were allowed to acclimate to the treadmill,a which involved the handler walking beside the dog while it walked on the treadmill. Acclimation occurred in < 15 minutes. Once a dog appeared to be comfortable walking on the treadmill, half-spherical reflective markers (8 mm in diameter) were placed at various anatomic locations on the dog to identify various body segments. The thorax was identified by markers placed at T1, T13, and the xiphoid process. The pelvis was identifed by markers placed on the right iliac crest, left iliac crest, right ischium, and left ischium (Figure 1).
Schematic depiction of the anatomic locations of markers used to track body segments while a dog walked and trotted on a treadmill when not wearing a vest (A) and when wearing a custom or adjustable vest (B). When a dog was wearing a vest, markers were placed directly on the vest at the anatomic locations of T1, T13, and the xiphoid process. Because the vests obstructed some of the reflective markers, the concept of virtual markers was used for markers L1 to L4 and R1 to R4. Notice that a marker was placed at the fifth metacarpal bone on the right side of each dog; this marker was used to quantify the completion of a gait cycle.
Citation: American Journal of Veterinary Research 81, 3; 10.2460/ajvr.81.3.210
Schematic depiction of the anatomic locations of markers used to track body segments while a dog walked and trotted on a treadmill when not wearing a vest (A) and when wearing a custom or adjustable vest (B). When a dog was wearing a vest, markers were placed directly on the vest at the anatomic locations of T1, T13, and the xiphoid process. Because the vests obstructed some of the reflective markers, the concept of virtual markers was used for markers L1 to L4 and R1 to R4. Notice that a marker was placed at the fifth metacarpal bone on the right side of each dog; this marker was used to quantify the completion of a gait cycle.
Citation: American Journal of Veterinary Research 81, 3; 10.2460/ajvr.81.3.210
Schematic depiction of the anatomic locations of markers used to track body segments while a dog walked and trotted on a treadmill when not wearing a vest (A) and when wearing a custom or adjustable vest (B). When a dog was wearing a vest, markers were placed directly on the vest at the anatomic locations of T1, T13, and the xiphoid process. Because the vests obstructed some of the reflective markers, the concept of virtual markers was used for markers L1 to L4 and R1 to R4. Notice that a marker was placed at the fifth metacarpal bone on the right side of each dog; this marker was used to quantify the completion of a gait cycle.
Citation: American Journal of Veterinary Research 81, 3; 10.2460/ajvr.81.3.210
The left scapula was identified by markers placed at the dorsal aspect of the left scapula, spine of the left scapula, and the center of a semirigid cluster system (ie, a triangular structure with a reflective marker on each point5,6 placed on the left scapula). The right scapula was identified by markers placed at the dorsal aspect of the right scapula, spine of the right scapula, and center of the cluster system placed on the right scapula. A semirigid cluster marker system was used to minimize the impact of skin movement and to help mark the anatomic locations on each dog consistently, particularly in situations where it was difficult to locate 3 anatomic locations on the scapula. This cluster system was a triangular structure (each side was approx 5 cm in length) that was placed on the dog, with one marker on the acromion and a second marker on the spine of the scapula at a point dorsal to the acromion. The orientation of this triangular structure determined the position of the third marker. The center of the triangular structure defined a location on the scapula.
A single marker was placed at the right fifth metacarpal bone of the dog, and the farthest forward motion of this marker defined the start of the stance phase. End of the stance phase and start of the swing phase were defined as the point at which this marker changed from a backward motion to a forward motion. Motion data for 1 gait cycle were divided into a set of 100 frames. Therefore, the position of the fifth metacarpal bone could be expressed for every 1% of the gait cycle. Analysis of these data indicated that the stance phase during walking began at 0% of the gait cycle and ended at 70% of the gait cycle. Similarly, the swing phase during walking started at 70% of the gait cycle and ended at 100% of the gait cycle. For trotting, the stance phase began at 0% of the gait cycle and ended at 60% of the gait cycle, and the swing phase began at 60% of the gait cycle and ended at 100% of the gait cycle.
Speed of the treadmill was increased until the dog could maintain a walking gait at a speed between 0.8 and 1.2 m/s. For each dog, the speed for the walking gait was recorded and used for each collection period to eliminate velocity variation within a dog. Data collection was started once the dog appeared to be comfortable walking on the treadmill at its designated speed.
The position of each marker was monitored by use of 6 infrared camerasb integrated with a commercial software package.c Cameras were placed around the treadmill and tracked the location of the reflective markers within a calibrated control space. The treadmill was positioned within the calibrated control space such that the position of each of 27 markers on the dog could be determined as the dog ambulated on the treadmill.7,8 A triangulation algorithm within the software package calculated the 3-D location of the markers relative to a reference axis system defined within the calibrated control space. On the basis of an algorithm outlined in another study,9 a customized software routine was used to translate the positional data from the global axis system to the local reference axis system. Small deviations of positional data could exist as a result of skin movement or other factors. To minimize the problem of skin movement and marker dropout during the gait, the algorithm of that study9 used a least squares procedure involving a static trial whereby the test dog stands still and marker position is recorded prior to the data collection session.10
After data for the walking gait were collected, each dog was acclimated to trotting on the treadmill. Speed was increased until the dog could maintain a trotting gait with the treadmill speed between 1.7 and 2.2 m/s. For each dog, the treadmill speed for the trotting gait was recorded and used for each collection period to eliminate velocity variation within a dog. Data collection was started once the dog appeared to be comfortable trotting on the treadmill at its designated speed.
Once data collection was completed for a dog at the walking and trotting gaits, the dog was acclimated to the applicable service vest. Both vestsd used in the study (a custom vest and an adjustable vest) weighed 1 kg (Figure 2). The custom vest was a dog-specific vest that was fitted for each test dog. Each vest had been fabricated by use of morphometric measurements, including circumferential measurements of the body at the neck, cranial aspect of the shoulders, caudal aspect of the shoulders, and caudal aspects of the thoracic and abdominal cavities of each dog, for a total of 5 custom vests overall. A single adjustable vest was used on all 5 dogs. The adjustable vest had straps that could be used to adjust the vest to the specific dog being tested. The same person who collected measurements used for fabrication of the custom vests also fitted the adjustable vest (eg, adjusted the vest size) on each dog. Once the adjustable vest was fitted to a particular dog, no changes to the vest were made until all data were collected on that day. When a dog returned for testing on another day, the adjustable vest was refitted.
Photographs of a dog walking on a treadmill wearing no vest (A), wearing an adjustable vest (B), and wearing a custom vest (C). The same adjustable vest was used on all dogs; the straps were adjusted to fit each dog. A custom vest was fabricated for each dog, and the custom vest wrapped around the entire trunk of the dog.
Citation: American Journal of Veterinary Research 81, 3; 10.2460/ajvr.81.3.210
Photographs of a dog walking on a treadmill wearing no vest (A), wearing an adjustable vest (B), and wearing a custom vest (C). The same adjustable vest was used on all dogs; the straps were adjusted to fit each dog. A custom vest was fabricated for each dog, and the custom vest wrapped around the entire trunk of the dog.
Citation: American Journal of Veterinary Research 81, 3; 10.2460/ajvr.81.3.210
Photographs of a dog walking on a treadmill wearing no vest (A), wearing an adjustable vest (B), and wearing a custom vest (C). The same adjustable vest was used on all dogs; the straps were adjusted to fit each dog. A custom vest was fabricated for each dog, and the custom vest wrapped around the entire trunk of the dog.
Citation: American Journal of Veterinary Research 81, 3; 10.2460/ajvr.81.3.210
The custom vest was placed on the dog, the dog was allowed to briefly acclimate to walking on the treadmill while wearing it, and data were then collected. Next, the dog was acclimated to the trotting gait, and data were collected. The adjustable vest was then placed on the dog, and after a brief acclimation period, the treadmill trials were repeated.
When the dogs wore a vest, markers were placed directly on the vest at the anatomic locations of T1, T13, and the xiphoid process. Because the vests obstructed some of the reflective markers, a virtual marker system was used.9 These virtual markers were designated as R1 and R2 for the right side and L1 and L2 for the left side. They were tracked by use of reflective markers placed at the various anatomic locations (R1, R2, R3, and R4 on the right side of the thorax and L1, L2, L3, and L4 on the left side of the thorax) used to identify the body segments for the kinematic data collection (Figure 1).
One repetition of a complete trial (dog wearing no vest, wearing the custom vest, and wearing the adjustable vest) required an entire day. Dogs were allowed to rest between treadmill cycles and were visually monitored for signs of fatigue. To obtain > 1 repetition required testing over multiple days; thus, each dog was returned to the laboratory on 3 days within a 10-day interval (first day of evaluation was designated as day 1), which yielded 3 repetitions for each dog wearing no vest, wearing the custom vest, and wearing the adjustable vest during walking and trotting on the treadmill for a total of 15 sets of kinematic data during walking and 15 sets during trotting.
At least three 15-second trials of gait data were collected for each dog for each condition (gait and vest).11,12 These 15-second trials equated to approximately 40 cycles for the walking gait and 60 cycles for the trotting gait. A minimum of 10 cycles for the walking gait and 15 cycles for the trotting gait in which there were no marker errors or marker dropouts were selected for analysis.13–15
Definition of truncal body segments
The markers located at T1, T13, and the xiphoid process were used to establish an orthogonal coordinate system (ie, the ATC system; Figure 3). The process used to create the ATC system was as follows. First, the x-axis of the ATC system, defined as a vector moving from the T1 marker to the T13 marker, was established by use of the following equation:
where 
designates the vector location of the marker on T1 and 
designates the vector location of the marker on T13. A temporary y-axis was established by use of the following equation:
where 
is the temporary y-axis and 
designates the vector location of the marker on the xiphoid process. The z-axis was established by use of the vector algebra function known as the cross product by use of the following equation:
wherein the z-axis is orthogonal to the x-axis and the cross product of the 2 vectors 
and 
produces the vector 
that is perpendicular to 
and 
. The final y-axis was established by use of the following equation
which thereby established the orthogonal ATC system for kinematic analysis of the motion of the trunk during a gait.
Schematic depiction of the ATC system. First, a rigid vector originating at T13 and ending at T1 defines the x-axis (solid line; left side of figure). Next, a rigid vector originating at the xiphoid process and ending at T1 defines the temporary y-axis (yt-axis; dashed line; left side of figure). The vector cross product of the x- and yt-axes determines the orientation of the z-axis, which is 90° perpendicular to both the yt- and x-axes. The orientation of the y-axis is the cross product of the z- and x-axes. The final orthogonal coordinate system is the x-, y-, and z-axes, which are located at T13 (right side of figure). Black circles represent the location of reflective markers.
Citation: American Journal of Veterinary Research 81, 3; 10.2460/ajvr.81.3.210
Schematic depiction of the ATC system. First, a rigid vector originating at T13 and ending at T1 defines the x-axis (solid line; left side of figure). Next, a rigid vector originating at the xiphoid process and ending at T1 defines the temporary y-axis (yt-axis; dashed line; left side of figure). The vector cross product of the x- and yt-axes determines the orientation of the z-axis, which is 90° perpendicular to both the yt- and x-axes. The orientation of the y-axis is the cross product of the z- and x-axes. The final orthogonal coordinate system is the x-, y-, and z-axes, which are located at T13 (right side of figure). Black circles represent the location of reflective markers.
Citation: American Journal of Veterinary Research 81, 3; 10.2460/ajvr.81.3.210
Schematic depiction of the ATC system. First, a rigid vector originating at T13 and ending at T1 defines the x-axis (solid line; left side of figure). Next, a rigid vector originating at the xiphoid process and ending at T1 defines the temporary y-axis (yt-axis; dashed line; left side of figure). The vector cross product of the x- and yt-axes determines the orientation of the z-axis, which is 90° perpendicular to both the yt- and x-axes. The orientation of the y-axis is the cross product of the z- and x-axes. The final orthogonal coordinate system is the x-, y-, and z-axes, which are located at T13 (right side of figure). Black circles represent the location of reflective markers.
Citation: American Journal of Veterinary Research 81, 3; 10.2460/ajvr.81.3.210
In accordance with the procedures described in another report,9 rotation of the ATC system was assessed relative to an LCS. One aspect of the present study was to assess the best anatomic positions for establishing the LCS. The 4 vectors used to define the LCS were as follows:
• V1 = the vector traveling from the spinous process of T1 to the spinous process of T13 relative to the left iliac crest.
• V2 = the vector traveling from the spinous process of T1 to the spinous process of T13 relative to the right iliac crest.
• V3 = the vector traveling from the spinous process of T1 to the spinous process of T13 relative to the dorsal aspect of the left scapula.
• V4 = the vector traveling from the spinous process of T1 to the spinous process of T13 relative to the dorsal aspect of the right scapula.
Data analysis established that the best LCS was defined by V3 and V4 (Figure 4). The procedure for defining this LCS was as follows. First, the z-axis of the LCS was started midway between the dorsal aspect of the left scapula and dorsal aspect of the right scapula and ended at the right iliac crest by use of the following equation:
where 
represents vector, DRS indicates the dorsal aspect of the right scapula, and DLS indicates the dorsal aspect of the left scapula. A temporary x-axis was established by use of the following equation:
where 
designates the vector location of the marker on T1 and 
designates the vector location of the marker on T13. The y-axis was established by use of the vector algebra function known as the following cross product:
where the y-axis is orthogonal to the z-axis. The cross product of the 2 vectors 
and 
produces the vector 
that is perpendicular to 
and 
. The final x-axis was established by use of 
cross 
, thereby establishing the orthogonal LCS for kinematic analysis of the motion data. The LCS was then translated to T1.
Schematic depiction of the LCS. First, a rigid vector originating at the dorsal aspect of the left scapula and ending at the dorsal aspect of the right scapula defines the z-axis. Next, a rigid vector originating at T13 and ending at T1 defines the temporary x-axis (xt-axis; left side of figure). The vector cross product of the z- and xt- axes establishes the direction of the y-axis (left side of figure). Finally, the direction of the x-axis is determined as the cross product of the y- and z-axes (right side of figure). This orthogonal axis system at T13 then was translated to T1. Notice that the z-axis is oriented 90° perpendicular to the x- and y-axes. See Figure 3 for remainder of key.
Citation: American Journal of Veterinary Research 81, 3; 10.2460/ajvr.81.3.210
Schematic depiction of the LCS. First, a rigid vector originating at the dorsal aspect of the left scapula and ending at the dorsal aspect of the right scapula defines the z-axis. Next, a rigid vector originating at T13 and ending at T1 defines the temporary x-axis (xt-axis; left side of figure). The vector cross product of the z- and xt- axes establishes the direction of the y-axis (left side of figure). Finally, the direction of the x-axis is determined as the cross product of the y- and z-axes (right side of figure). This orthogonal axis system at T13 then was translated to T1. Notice that the z-axis is oriented 90° perpendicular to the x- and y-axes. See Figure 3 for remainder of key.
Citation: American Journal of Veterinary Research 81, 3; 10.2460/ajvr.81.3.210
Schematic depiction of the LCS. First, a rigid vector originating at the dorsal aspect of the left scapula and ending at the dorsal aspect of the right scapula defines the z-axis. Next, a rigid vector originating at T13 and ending at T1 defines the temporary x-axis (xt-axis; left side of figure). The vector cross product of the z- and xt- axes establishes the direction of the y-axis (left side of figure). Finally, the direction of the x-axis is determined as the cross product of the y- and z-axes (right side of figure). This orthogonal axis system at T13 then was translated to T1. Notice that the z-axis is oriented 90° perpendicular to the x- and y-axes. See Figure 3 for remainder of key.
Citation: American Journal of Veterinary Research 81, 3; 10.2460/ajvr.81.3.210
Data analysis
Tracking data were analyzed to determine the landmarks that best established vectors needed for the LCS. This LCS would serve as the reference frame to quantify truncal rotation. The 3-D global position (global coordinate reference axes) of markers was collected (Figure 1), and the positional data were used to calculate the aforementioned vectors. Coordinates of these vectors were calculated for every 1% of each gait cycle, which provided 100 vector coordinates for the complete gait cycle. Data were collected over 15 gait cycles, which provided 1,500 vector magnitudes to assess hypotheses 1, 2, 3, and 4. Position of the vectors was relative to 1 of 4 landmarks (left iliac crest, right iliac crest, dorsal aspect of the left scapula, and dorsal aspect of the right scapula), depending on the hypothesis tested.
Truncal ROM
For each dog and each day of testing, the angle of rotation of each ATC axis (x-, y-, and z-axis) relative to its corresponding LCS axis was determined for 1% of the gait cycle by use of Euler rotation matrices.16 For each axis, the minimum and maximum angles of rotation over the entire gait cycle were calculated whereby the difference between the 2 angles defined the trunk body segment ROM. The mean ROM for each axis was calculated for each day that a dog ambulated on the treadmill. Rotation of the trunk body segment was also described by calculating the rotation angle of the ATC axis for each percentage of the gait cycle.
Statistical analysis
Hypotheses 1 to 4 were tested by calculating the coefficient of variation of the 100 vector magnitudes collected over the entire gait cycle. This approach quantified differences in these 100 magnitudes and provided a way to gauge uncertainty in motion analysis techniques.17 Hypothesis 5 involved assessment of the impact of each vest; it was tested by comparing the truncal ROM of the 3 ATC axes for each dog for each of the 3 test days. Hypothesis 6 involved assessment of the interday impact of data collection on ROM measurements. The data collected for each of the 5 dogs when not wearing a vest were assessed for each of the 3 test days. Repeated-measures ANOVA was used for the comparisons pertaining to hypotheses 5 and 6. Values of P ≤ 0.05 were considered significant.
Results
The coefficient of variation for the magnitude of the vectors used to determine the best LCS was determined (Table 1). Data analyses revealed that the vectors connecting markers on the dorsal aspect of the scapula to T1 and T13 had small changes over the entire gait cycle as the dogs walked or trotted on the treadmill. Thus, vectors connecting the dorsal aspect of the right and left scapulae, T1, and T13 were used to establish the LCS (Figure 4) needed to analyze truncal rotation.
Coefficients of variation (CVs; %) for the vectors connecting the 4 landmarks to the spinous process of T1 and T13 in a representative dog for a series of gait cycles as the dog walked and trotted on a treadmill.
| CV of the magnitude of a vector connecting T13 relative to | CV of the magnitude of a vector connecting T1 relative to | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| Gait | Vest | Left iliac crest | Right iliac crest | Left scapula | Right scapula | Left iliac crest | Right iliac crest | Left scapula | Right scapula |
| Walking | No vest | 4 | 4 | 1 | 3 | 4 | 5 | 3 | 2 |
| Custom vest | 5 | 6 | 1 | 5 | 6 | 7 | 4 | 3 | |
| Adjustable vest | 4 | 5 | 1 | 3 | 4 | 5 | 3 | 3 | |
| Trotting | No vest | 5 | 5 | 1 | 6 | 5 | 6 | 4 | 4 |
| Custom vest | 4 | 5 | 1 | 3 | 4 | 5 | 3 | 2 | |
| Adjustable vest | 5 | 5 | 1 | 2 | 5 | 6 | 3 | 1 | |
The magnitude of each vector was calculated for each 1% of a gait cycle, which provided 100 data points for each vector for each gait cycle. Kinematic data were collected over 15 gait cycles, which provided 1,500 magnitude values for each vector for each landmark, and the CV was calculated for the 1,500 magnitude values for each vector for each landmark. The vector system with the smallest CV was used to establish reference vectors for quantifying the trunk rotation during the gait of dogs.
Mean rotational angles for each of the study dogs over the testing periods were determined (Figures 5 and 6). As anticipated, the magnitude of the rotation of each ATC axis was greater when the dogs trotted on the treadmill. The ROM of the trunk rotation during a gait cycle did not differ significantly regardless of the day of testing for both the walking and trotting gaits.
Data for rotation of the trunk relative to the x-axis (A), y-axis (B), and z-axis (C) during the gait cycle for 5 dogs when not wearing a vest and walking on a treadmill. Values represent the mean rotation angle for all 5 dogs for all 3 testing days (solid line) and related 95% confdence intervals (dashed lines). The start of the stance phase (0% of the gait cycle) was defined as the farthest forward motion of a marker attached at the right fifth metacarpal bone. The stance phase ended when the position of the right fifth metacarpal bone changed from a backward direction to a forward direction. End of the stance phase was at 70% of the gait cycle. The swing phase started at 70% and ended at 100% of the gait cycle. Notice that the scale on the y-axis differs among panels.
Citation: American Journal of Veterinary Research 81, 3; 10.2460/ajvr.81.3.210
Data for rotation of the trunk relative to the x-axis (A), y-axis (B), and z-axis (C) during the gait cycle for 5 dogs when not wearing a vest and walking on a treadmill. Values represent the mean rotation angle for all 5 dogs for all 3 testing days (solid line) and related 95% confdence intervals (dashed lines). The start of the stance phase (0% of the gait cycle) was defined as the farthest forward motion of a marker attached at the right fifth metacarpal bone. The stance phase ended when the position of the right fifth metacarpal bone changed from a backward direction to a forward direction. End of the stance phase was at 70% of the gait cycle. The swing phase started at 70% and ended at 100% of the gait cycle. Notice that the scale on the y-axis differs among panels.
Citation: American Journal of Veterinary Research 81, 3; 10.2460/ajvr.81.3.210
Data for rotation of the trunk relative to the x-axis (A), y-axis (B), and z-axis (C) during the gait cycle for 5 dogs when not wearing a vest and walking on a treadmill. Values represent the mean rotation angle for all 5 dogs for all 3 testing days (solid line) and related 95% confdence intervals (dashed lines). The start of the stance phase (0% of the gait cycle) was defined as the farthest forward motion of a marker attached at the right fifth metacarpal bone. The stance phase ended when the position of the right fifth metacarpal bone changed from a backward direction to a forward direction. End of the stance phase was at 70% of the gait cycle. The swing phase started at 70% and ended at 100% of the gait cycle. Notice that the scale on the y-axis differs among panels.
Citation: American Journal of Veterinary Research 81, 3; 10.2460/ajvr.81.3.210
Data for rotation of the trunk relative to the x-axis (A), y-axis (B), and z-axis (C) during the gait cycle for 5 dogs when not wearing a vest and trotting on a treadmill. End of the stance phase was at 60% of the gait cycle. The swing phase started at 60% and ended at 100% of the gait cycle. Positive values for the angle of rotation indicate a clockwise rotation about the axis, and negative values indicate a counterclockwise rotation about the axis. Notice that the scale on the y-axis differs among panels. See Figure 5 for remainder of key.
Citation: American Journal of Veterinary Research 81, 3; 10.2460/ajvr.81.3.210
Data for rotation of the trunk relative to the x-axis (A), y-axis (B), and z-axis (C) during the gait cycle for 5 dogs when not wearing a vest and trotting on a treadmill. End of the stance phase was at 60% of the gait cycle. The swing phase started at 60% and ended at 100% of the gait cycle. Positive values for the angle of rotation indicate a clockwise rotation about the axis, and negative values indicate a counterclockwise rotation about the axis. Notice that the scale on the y-axis differs among panels. See Figure 5 for remainder of key.
Citation: American Journal of Veterinary Research 81, 3; 10.2460/ajvr.81.3.210
Data for rotation of the trunk relative to the x-axis (A), y-axis (B), and z-axis (C) during the gait cycle for 5 dogs when not wearing a vest and trotting on a treadmill. End of the stance phase was at 60% of the gait cycle. The swing phase started at 60% and ended at 100% of the gait cycle. Positive values for the angle of rotation indicate a clockwise rotation about the axis, and negative values indicate a counterclockwise rotation about the axis. Notice that the scale on the y-axis differs among panels. See Figure 5 for remainder of key.
Citation: American Journal of Veterinary Research 81, 3; 10.2460/ajvr.81.3.210
Truncal ROM was compared between dogs while not wearing a vest and while wearing each of the 2 vests when walking on the treadmill. Both vests significantly reduced trunk rotation about all 3 axes (Figure 7). Additionally, truncal ROMs about the y-and z-axes were significantly less for the custom vest than for the adjustable vest.
Mean ± SE rotation of the trunk relative to the x-axis (A and B), y-axis (C and D), and z-axis (E and F) for 5 dogs during walking (A, C, and E) and trotting (B, D, and F) on a treadmill while wearing no vest, wearing an adjustable vest, and wearing a custom vest. Data were collected from each dog on 3 days (D1, D2, and D3) during a 10-day period. Notice that the scale on the y-axis differs among panels.a–c Within a panel, values with different superscript letters differ significantly (P ≤ 0.05).
Citation: American Journal of Veterinary Research 81, 3; 10.2460/ajvr.81.3.210
Mean ± SE rotation of the trunk relative to the x-axis (A and B), y-axis (C and D), and z-axis (E and F) for 5 dogs during walking (A, C, and E) and trotting (B, D, and F) on a treadmill while wearing no vest, wearing an adjustable vest, and wearing a custom vest. Data were collected from each dog on 3 days (D1, D2, and D3) during a 10-day period. Notice that the scale on the y-axis differs among panels.a–c Within a panel, values with different superscript letters differ significantly (P ≤ 0.05).
Citation: American Journal of Veterinary Research 81, 3; 10.2460/ajvr.81.3.210
Mean ± SE rotation of the trunk relative to the x-axis (A and B), y-axis (C and D), and z-axis (E and F) for 5 dogs during walking (A, C, and E) and trotting (B, D, and F) on a treadmill while wearing no vest, wearing an adjustable vest, and wearing a custom vest. Data were collected from each dog on 3 days (D1, D2, and D3) during a 10-day period. Notice that the scale on the y-axis differs among panels.a–c Within a panel, values with different superscript letters differ significantly (P ≤ 0.05).
Citation: American Journal of Veterinary Research 81, 3; 10.2460/ajvr.81.3.210
When the dogs trotted on the treadmill, significantly more truncal ROM was observed about the x-and z-axes when they wore no vest than when they wore the adjustable or custom vest (Figure 7). Truncal ROM about the y-axis was significantly less for the custom vest than for the adjustable vest.
Discussion
Most canine gait studies focus on limb joint kinematics and reference joint movement to define the anatomic location of coordinate systems placed on adjacent body segments.9,17–27 However, information about trunk kinematics is scarce. In the study reported here, anatomic landmarks (dorsal aspect of the right and left scapulae, T1, and T13) were established for creating a thorax coordinate system in dogs. Interestingly, the anatomic landmarks used to establish the coordinate system in the present study were similar to ones used to assess trunk rotation in humans and recommended by the International Society of Biomechanics.28–30 It should be noted that studies conducted to analyze human trunk movement may use different anatomic landmarks, depending on the application, and may divide the trunk into a lower trunk segment and an upper trunk segment.
Several human kinematic studies have quantified the kinematic relationships between the thorax body segment and the scapula-shoulder body segment, particularly when investigating thoracic arm moments. Such studies include an attempt to clarify how trunk-scapula kinematics affect humeral elevation,31 an investigation of effects of mastectomy on upper limb movement,32 and examination of overhead sports (eg, baseball, volleyball, and tennis) on asymmetry of the trunk-scapula kinematics.33 Equine studies have revealed an impact of thoracic limb movement on thorax-scapula kinematics. Retraction of a horse's forelimb was found to cause muscles connecting the thorax and scapula to move the ribcage and thus move the trunk to the side.34 Similar muscle activation takes place during the canine gait cycle.35,36 Therefore, ribcage motion and trunk rotation should be comparable. These studies of humans and horses support the findings of the present study of dogs, which suggested that a coordinate system located on the scapula is the most appropriate reference site for assessing trunk rotation as a dog walks and trots on a treadmill.
In the present study, testing each dog on multiple days did not impact the study outcome, which aligns with findings published in the literature. Repeatable kinematic data for sheep37 and dogs38 could be obtained when replicating treadmill tests over multiple days. Similar findings have been reported for humans,39 although interday effects are less significant when a person is allowed to self-select the speed of the treadmill to match his or her usual gait. In the study reported here, the experimental design mimicked a self-selection process whereby treadmill speed (resulting in a velocity within the limits for walking and trotting) was set for each dog, instead of the use of the same treadmill speed for all dogs.
Although the present study was focused more on establishing a technique to assess truncal rotation of dogs during a gait, it was interesting, but not surprising, that the custom vest restricted the trunk rotation more than did the adjustable vest. The custom vest had fabric that covered the entire trunk, whereas straps were used to hold the adjustable vest in place. Owing to its design and the fact it was fitted to each dog, the custom vest wrapped around the dog and could have reduced the desire of a dog to rotate the trunk or the propensity for the vest to slip relative to the trunk. The design of the adjustable vest appeared to be less restrictive on the ability of a dog to move its trunk because this vest fit primarily on the dorsal portion of the body and each strap could be tightened separately. We did not assess how the specific design features of each vest could impact truncal rotation. Additional studies need to be conducted to determine how the vests (and strap locations) could alter the reaction of dogs to them and change truncal movement or positioning. Furthermore, we did not assess how the designs could influence paw placement and balance during the gait cycle. Overall, more studies are needed to explain the clinical relevance of the data, and more markers on the limbs of the dogs could help make this clinical assessment.
One limitation to the study reported here was the small number of dogs that had little variation in conformation. The authors tried to choose a study group that was fairly homogenoneous and nonchondrodystrophic and that closely resembled dogs in a clinical setting. A second limitation was that the order of placement of the vests on the dogs was not randomized. The study did not include assessment of whether wearing the custom vest first influenced data acquisition for the adjustable vest. A third limitation was that the positions of the head and other body segments of each dog were not closely monitored during data collection. The handler only montitored that the dogs walked and trotted on the treadmill at appropriate speeds in a forward direction. A fourth limitation was the potential variation in data attributable to marker movement caused by movement of the skin. Marker movement due to movement of the skin has been well documented, and several methods are used to limit soft tissue movement.9,10,40–45 Use of markers placed over bony landmarks with minimal soft tissue is recommended and was the method used in the present study. Additionally, cluster markers were used, which can decrease marker variability.5,6,40,41 There will always be marker movement, but data comparison and analysis were performed such that data were computed within each dog, which potentially decreased variability.9 Furthermore, the same investigator placed all the markers to limit interexaminer variability. In addition, to limit intraexaminer variablitiy, markers were placed on well-defined anatomic structures with a dog standing in the same position and were not changed or replaced during a gait collection period.45
The first part of the study reported here focused on establishing a method for measuring truncal rotation and validating this method for repeatability over multiple testing days for multiple dogs. The kinematic data aquired for each dog over multiple days indicated that the proposed method for measuring truncal rotation was valid, with minimal differences among days for each dog. These data also indicated that the body segments could be considered as rigid structures during the kinematic data collection. The second part of the present study focused on use of this method to analyze changes in the rotation of body segments while dogs were wearing 2 types of vests (an adjustable vest and a custom-made vest), revealing measurable differences in gait for each of the 2 service vests, compared with values obtained when dogs were not wearing a vest.
Acknowledgments
The authors declare there were no conflicts of interest.
ABBREVIATIONS
| ATC | Anatomic thorax coordinate |
| LCS | Local coordinate system |
| ROM | Range of motion |
Footnotes
Large DogTread, PetZen Products, Ogden, Utah.
Eagle Industries, Virginia Beach, Va.
T-series, Vicon, Lake Forest, Calif.
Peak Motus, version 9, Vicon, Lake Forest, Calif.
References
1. Gradner G, Bockstahler B, Peham C, et al. Kinematic study of back movement in clinically sound Malinois dogs with consideration of the effect of radiographic changes in the lumbosacral junction. Vet Surg 2007;36:472–481.
2. Foss K, da Costa RC, Moore S. Three-dimensional kinematic gait analysis of Doberman Pinschers with and without cervical spondylomyelopathy. J Vet Intern Med 2013;27:112–119.
3. Pfau T, Witte TH, Wilson AM. Centre of mass movement and mechanical energy fluctuation during gallop locomotion in the Thoroughbred racehorse. J Exp Biol 2006;209:3742–3757.
4. Clayton HM, Sha DH. Head and body centre of mass movement in horses trotting on a circular path. Equine Vet J Suppl 2006;36:462–467.
5. Caron A, Caley A, Farrell M, et al. Kinematic gait analysis of the canine thoracic limb using a six degrees of freedom marker set. Vet Comp Orthop Traumatol 2014;27:461–469.
6. van Andel C, van Hutten K, Eversdijk M, et al. Recording scapular motion using an acromion marker cluster. Gait Posture 2009;29:123–128.
7. 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.
8. Pfister A, West AM, Bronner S, et al. Comparative abilities of Microsoft Kinect and Vicon 3D motion capture for gait analysis. J Med Eng Technol 2014;38:274–280.
9. Fu YC, Torres BT, Budsberg SC. Evaluation of a three-dimensional kinematic model for canine gait analysis. Am J Vet Res 2010;71:1118–1122.
10. Veldpaus FE, Woltring HJ, Dortmans LJ. A least-squares algorithm for the equiform transformation from spatial marker coordinates. J Biomech 1988;21:45–54.
11. Gustås P, Pettersson K, Honkavaara S, et al. Kinematic and temporospatial assessment of habituation of Labrador Retrievers to treadmill trotting. Vet J 2013;198:e114–e119.
12. Fanchon L, Grandjean D. Habituation of healthy dogs to treadmill trotting: repeatability assessment of vertical ground reaction force. Res Vet Sci 2009;87:135–139.
13. Abdelhadi J, Wefstaedt P, Nolte I, et al. Fore-aft ground force adaptations to induced forelimb lameness in walking and trotting dogs. PLoS One 2012;7:e52202.
14. Helms G, Behrens BA, Stolorz M, et al. Multi-body simulation of a canine hind limb: model development, experimental validation and calculation of ground reaction forces. Biomed Eng Online 2009;8:36.
15. Riley PO, Dicharry J, Franz JA, et al. A kinematics and kinetic comparison of overground and treadmill running. Med Sci Sports Exerc 2008;40:1093–1100.
16. Hamill J, Sebie WS, Kepple TM. Three-dimensional kinematics. In: Robertson G, Caldwell G, Hamill J, et al, eds. Research methods in biomechanics. 2nd ed. Champaign, Ill: Human Kinetics, 2013;35–39.
17. Chau T, Chau D, Casas M, et al. Investigating the stationarity of paediatric aspiration signals. IEEE Trans Neural Syst Rehabil Eng 2005;13:99–105.
18. Wachs K, Fischer MS, Schilling N. Three-dimensional movements of the pelvis and the lumbar intervertebral joints in walking and trotting dogs. Vet J 2016;210:46–55.
19. 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.
20. Hottinger HA, DeCamp CE, Olivier N, et al. Noninvasive kinematic analysis of the walk. Am J Vet Res 1996;57:381–388.
21. Poy NSJ, DeCamp CE, Bennet 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.
22. Torres BT, Fu Y-C, Sandberg GS, et al. Pelvic limb kinematics in the dog with and without a stifle orthosis. Vet Surg 2017;46:642–652.
23. Andrada E, Reinhardt L, Lucas K, et al. Three-dimensional inverse dynamics of the forelimb of Beagles at a walk and trot. Am J Vet Res 2017;78:804–817.
24. Rey J, Fischer MS, Bottcher P. Sagittal joint instability in the cranial cruciate ligament insufficient canine stifle. Caudal slippage of the femur and not cranial tibial subluxation. Tierarztl Prax Ausg K Kleintiere Heimtiere 2014;42:151–156.
25. Fischer MS, Lehmann SV, Andrada E. Three-dimensional kinematics of canine hind limbs: in vivo, biplanar, high-frequency fluoroscopic analysis of four breeds during walking and trotting. Sci Rep 2018;8:16982.
26. Kim SE, Jones SC, Lewis DD, et al. In-vivo three-dimensional knee kinematics during daily activities in dogs. J Orthop Res 2015;33:1603–1610.
27. Rohwedder T, Fischer M, Bottcher P. In vivo axial humeroulnar rotation in normal and dysplastic canine elbow joints. Tierarztl Prax Ausg K Klientiere Heimtiere 2018;46:83–89.
28. Park HJ, Sim T, Suh SW, et al. Analysis of coordination between thoracic and pelvic kinematic movements during gait in adolescents with idiopathic scoliosis. Eur Spine J 2016;25:385–393.
29. Leardini A, Biagi F, Belvedere C, et al. Quantitative comparison of current models for trunk motion in human movement analysis. Clin Biomech (Bristol, Avon) 2009;24:542–550.
30. Wu G, van der Helm FC, Veeger HD, et al. ISB recommendation on definitions of joint coordinate systems of various joints for the reporting of human joint motion—part II: shoulder, elbow, wrist and hand. J Biomech 2005;38:981–992.
31. Nagai K, Tateuchi H, Takashima S, et al. Effects of trunk rotation on scapular kinematics and muscle activity during humeral elevation. J Electromyogr Kinesiol 2013;23:679–687.
32. Crosbie J, Kilbreath SL, Dylke E, et al. Effects of mastectomy on shoulder and spinal kinematics during bilateral upper-limb movement. Phys Ther 2010;90:679–692.
33. Oyama S, Myers JB, Wassinger CA, et al. Asymmetric resting scapular posture in healthy overhead athletes. J Athl Train 2008;43:565–570.
34. Nauwelaerts S, Kaiser L, Malinowski R, et al. Effects of trunk deformation on trunk center of mass mechanical energy estimates in the moving horse, Equus caballus. J Biomech 2009;42:308–311.
35. Carrier DR, Deban SM, Fischbein T. Locomotor function of the pectoral girdlemuscular sling in trotting dogs. J Exp Biol 2006;209:2224–2237.
36. Carrier DR, Deban SM, Fischbein T. Locomotor function of forelimb protractor and retractor muscles of dogs: evidence of strut-like behavior at the shoulder. J Exp Biol 2008;211:150–162.
37. Tapper JE, Fukushima S, Azuma H, et al. Dynamic in vivo kinematics of the intact ovine stifle joint. J Orthop Res 2006;24:782–792.
38. Torres BT, Punke JP, Fu YC, et al. Comparison of canine stifle kinematic data collected with three different targeting models. Vet Surg 2010;39:504–512.
39. Ferber R, Davis IM, Williams DS, et al. A comparison of within-and between-day reliability of discrete 3D lower extremity variables in runners. J Orthop Res 2002;20:1139–1145.
40. Andriacchi TP, Alexander EJ, Toney MK, et al. A point cluster method for in vivo motion analysis: applied to a study of knee kinematics. J Biomech Eng 1998;120:743–749.
41. Alexander EJ, Andriacchi TP. Correcting for deformation in skin- based marker systems. J Biomech 2001;34:355–361.
42. Taylor WR, Ehrig RM, Duda GN, et al. On the influence of soft tissue coverage in the determination of bone kinematics using skin markers. J Orthop Res 2005;23:726–734.
43. Kim SY, Kim JY, Hayashi K. Skin movement during the kinematic analysis of the canine pelvic limb. Vet Comp Orthop Traumatol 2011;24:326–332.
44. Schwencke M, Smolders LA, Bergknut N, et al. Soft tissue artifact in canine kinematic gait analysis. Vet Surg 2012;41:829–837.
45. Kim SY, Torres BT, Sandberg GS, et al. Effect of limb positon at the time of skin marker application on sagittal plane kinematics of the dog. Vet Comp Orthop Traumatol 2017;30:438–443.
