As in humans, osteoarthritis in dogs is associated with various clinical signs, including signs of pain, restrictions in joint mobility, and difficulties in undertaking activities of daily life such as stair climbing. Bearing this in mind, therapeutic care of these patients has 2 main objectives: alleviation of pain and improvement of joint function. In addition to classic, conservative pain management (eg, administration of NSAIDs), methods of physical therapy have become an important part of treatment of osteoarthritis. Many treatment options available for use in humans, such as electrotherapy, therapeutic ultrasonography, hydrotherapy, and movement therapy, are already frequently applied in animals.1 Nowadays, a combination of conservative pain modulation and physical therapy is considered a state-of-the-art treatment protocol.2,3 The purpose of physical therapy is to aid pain reduction and promote joint and limb function improvement. Given the large number of available physical therapy regimens, an individualized treatment program has to be applied to each patient on the basis of specific needs.
In the field of veterinary physical therapy, the selection of therapeutic procedures is often based on individual experience of the therapist or medical knowledge acquired from the study of humans. This is because controlled studies to investigate the efficacy of each procedure are lacking and because the knowledge of joint and muscle biomechanics in veterinary species is limited. Studies investigating the clinical outcome of physical therapy in nonhuman animals are especially rare. The therapeutic effects of electrotherapy as a method of physical therapy in dogs with osteoarthritis and rupture of the cranial cruciate ligament have been reported.4,a In another study,5 the GRFs of dogs treated with hydrotherapy after a lateral retinacular stabilization surgery because of rupture of the cranial cruciate ligament were significantly greater than GRFs in exercise-restricted dogs that did not undergo physical therapy. The effects of early intensive postoperative physical therapy on limb function after tibial plateau leveling osteotomy in dogs have been investigated.6 The dogs in that study were treated with physical therapy consisting of passive ROM and movement exercises and, after removal of sutures, training on an underwater treadmill. Compared with a control group of dogs that did not receive intensive postoperative physical therapy, treated dogs had significantly greater extension and flexion of the osteotomized femorotibial joint. A study7 of the effects of rehabilitation treatment options after fibrocartilaginous embolism in 75 dogs revealed that hydrotherapy performed immediately after the initial diagnostic and clinical evaluation had a major positive influence on the recovery rate. Controlled physical therapy can have beneficial effects in dogs with neurologic diseases. Dogs with suspected degenerative myelopathy that received intensive physical therapy had significantly longer survival times, compared with the survival times of dogs that received moderate or no physical therapy.8 The effects of standardized movements (passive stifle joint bending, straight limb raising, and dural stretch exercise) during physical therapy exercises for mobilization of lumbar spinal nerves and the dura mater in dogs have been studied.9 The movement standardization was done by use of a goniometric control, and the movement therapy had measurable effects on nerve roots L4 to L7 and on the dura mater in the T13 and L1 segments.
Another problem that should not be underestimated is the fact that most studies investigating canine biomechanics have been limited to assessments of healthy dogs during walking10 and trotting11–13 or were performed to evaluate surgical and medical treatments.14–18 To the authors’ knowledge, there is little information regarding joint and muscle function in various physical activities, such as walking on inclined or declined slopes or walking over low-level obstacles (eg, cavaletti). Only 4 studies19–21,b have been conducted to assess differences in movement patterns during performance of active exercises in dogs. Swimming causes a significantly greater flexion of the hip, stifle, and tarsal joints, compared with flexion during walking.19 That finding, coupled with the fact that body mass changes in relation to water depth, confirms the assumption that purposeful movement on an underwater treadmill leads to an improvement of the ROM with a lower loading of the joints. By use of kinematic and kinetic motion analyses of the active exercise of wheelbarrowing, it is known that the duration of stance and swing as well as step length decreases and the PFz increases, compared with findings during normal walking. Kinematic analysis revealed significantly greater extension of the shoulder and carpal joints and an increase in elbow joint flexion; conversely, flexion of the shoulder joint and extension of the elbow and carpal joints decreased.
Similar analyses were performed during so-called dancing (ie, movement forward and backward while the dog's forelimbs are physically raised so that it stands solely on its hind limbs). During dancing, the PFz and the vertical impulse decreased. Compared with findings during normal walking, forward dancing was associated with decreased flexion and lower ROM of the hip and tarsal joints; reverse dancing resulted in increased extension of the hip joint and flexion of the stifle joint. These data19,b were crucial in the development of adequate physical therapy programs for dogs with hip dysplasia because they provided therapists with useful information about the changes of the musculoskeletal system during special exercises, which should be considered if a therapy program is planned.
In another study20 of movement patterns and physiotherapeutic exercises, the 2-D kinematic analysis of walking and sit-to-stand motion in dogs was validated. Because that study was focused more on the technical aspects, there were no conclusions regarding the biomechanics of the sit-to-stand motion, compared with findings of the aforementioned studies. Peak vertical GRF acting on the forelimbs of dogs during landing after jumping over an obstacle has been investigated.21 The GRFs during landing after jumping differed significantly in relation to the jump height. Furthermore, factors (eg, body weight, breed, and sex) that influenced that force were identified.
Physical therapy involving active exercises is particularly important in the treatment of osteoarthritis; the main goal of movement therapy is to establish specific exercises that will strengthen the locomotory musculature and improve the biomechanical function of joints. To evaluate the usefulness of active exercises in the treatment of osteoarthritis, the biomechanics of joints during different movement patterns must be investigated first. Therefore, the purpose of the study reported here was to assess fore- and hind limb joint kinematics in dogs during walking on an inclined slope (uphill), on a declined slope (downhill), or over low obstacles on a horizontal surface (obstacle exercise) and compare findings with data acquired during unimpeded walking on a horizontal surface. It has been suggested that the joint kinematics of the fore- and hind limb during execution of these types of exercises differ considerably from findings during normal walking. Our hypothesis was that flexion and extension of joints as well as the temporospatial characteristics of the joint angulations would undergo major changes during uphill and downhill walking and obstacle exercise, compared with joint kinematics during normal unimpeded walking.
Materials and Methods
Dogs—Eight client-owned dogs (1 Labrador Retriever, 3 Golden Retrievers, 1 Large Münsterländer, 1 Australian Shepherd, 1 Border Collie, and 1 Nova Scotia Duck Tolling Retriever) were included in the study. Mean ± SD age of the dogs was 3.4 ± 2.0 years; mean body mass was 23.6 ± 4.6 kg at the time of measurement. Before the dogs were incorporated in the study, thorough clinical and neurologic examinations22 and orthopedic examinations23 were performed on each dog.
No clinical or orthopedic abnormalities were detected in any dog. The study was discussed and approved by the institutional ethics committee, and the dogs’ owners provided informed consent.
Equipment and GRF measurement procedure—To substantiate the clinical findings and to exclude lameness objectively, kinetic measurements were performed on a treadmill that was specifically developed for companion animals by the University of Sport Medicine, Cologne, Germany.24,25 Vertical GRFs were determined at 300 Hz by use of 4 integrated piezoelectric force plates.c Measurements were started as soon as each dog established a smooth and well-coordinated gait. The measurement velocity was 1.2 m/s, and the duration of recording was 120 seconds for all dogs. For kinetic analysis, 5 valid steps were used. Steps were assessed as valid when each force plate was just touched by the associated extremity. All data were subsequently normalized in relation to body mass (%/kg of body mass). For evaluation of the kinetic data, an appropriate software productd was used. The PFz of each limb was evaluated. Weight distribution between the contralateral limb pairs was calculated for each variable by use of an equation as follows:
SIPFz = ([PFzL – PFzR]/[PFzL + PFzR]) × 100
where SIPFz is the percentage difference in weight distribution between limbs in each pair, subscript L denotes the left limb, and subscript R denotes the right limb. Dogs were considered to be free of lameness when SIPFz was < 5%.
Equipment and kinematic measurement procedure—Digitalization of each specific exercise was done by recording the movement of 10 reflecting markers (each 1 cm in diameter) that were positioned in accurately defined anatomic locations on the left forelimb and left hind limb. To prevent concealment of the dorsal markers by hair, the body of each dog was fitted with a tight-fitting covering created by use of a tube bandage.e Holes were cut in the tube bandage for the fore- and hind limbs to be put through, after which the tube bandage was pulled over the dog's head and down to surround its body. This seamlessly knitted tube was made from white bleached cotton (67%) and viscose (33%). The covering extended over the dorsal aspect of the scapulae, the back, and the pelvis. Two markers (on the dorsal aspect of the left scapula and left cranial dorsal iliac spine) were directly attached to the covering by use of double-sided tape and adhesive. In those areas not covered by bandage material, 8 additional markers were directly fixed to the haired skin by use of double-sided tape and nonirritant adhesive. In total, 10 markers were affixed to the left side of each dog as follows: on the dorsal border of the scapular spine, between the acromion and the greater tubercle of the humerus, on the lateral epicondyle of the humerus, on the styloid process of the ulna, on the distal aspect of the fifth metacarpal bone of the left forelimb, on the cranial dorsal iliac spine, on the greater trochanter of the femur, on the stifle joint between the lateral epicondyle of the femur and fibular head, on the lateral malleolus, and on the distal aspect of the fifth metatarsal bone.
Ten digital camerasf (each with a resolution of 1.3 × 106 pixels at full resolution [1,280 × 1,024]) were situated in a circle around the area of measurement (4.25 × 1.10 m). A treadmill for horses was built into the floor in the middle of the circle of cameras. Each camera had 237 light-emitting diodes attached annularly around the lens for brighter and better light uniformity. Sample frequency of each camera was 120 Hz. For each session, the system was calibrated with a calibration frame of known dimensions. Data were captured and analyzed by use of a personal computerg and a motion-analysis program.h
Assessments of each dog during uphill and downhill walking and walking over low obstacles (cavaletti) or unimpeded on a horizontal surface were performed. All assessments were conducted consecutively in 1 experimental session and in the same order for each dog (unimpeded walking on a horizontal surface, uphill walking, downhill walking, and walking over low obstacles). Between the different walking regimens, the dogs were allowed to rest for 10 minutes. The treadmill used in the study could attain an inclination of 11%. This slope was used for the uphill and downhill walking exercises. For the obstacle exercise, 5 height-adjustable cavaletti (timber jumps) were used. The cavaletti heights were adjusted to the height of the carpal joint. The distance between each cavaletti bar was set accordingly to the distance between the fore- and hind limb of each dog. Prior to measurement, each exercise was performed 3 times to accustom the dogs to the experimental schedule. Between the period of acclimatization and the actual measurement, there was a 10-minute rest period. The dog handler walked diagonally left in front of the dog to prevent concealing the markers. After acclimatization, each exercise was carried out at least 6 times, starting with assessment of unimpeded walking on a horizontal surface. The mean velocity of the dog was determined by use of the marker on the distal aspect of the fifth metacarpal bone of the left forelimb. Mean ± SD velocity for each activity varied from 1.06 ± 0.21 m/s for normal walking, 1.07 ± 0.17 m/s for uphill walking, 1.10 ± 0.16 m/s for downhill walking, and 0.89 ± 0.11 m/s for the obstacle exercise. The evaluation of data was carried out with a motion-analysis programh for angle calculation and sequencing programsi,j for analyzing movement cycles. Motion cycle sequencing was done on the basis of the horizontal velocity of the marker on the distal aspect of the fifth metacarpal bone; motion cycles were separated by detecting the zero positions of the velocity.26
The 3-D angles of the shoulder, elbow, carpal, hip, stifle (femorotibial), and tarsal joints were calculated for each time frame of 5 motion cycles in each dog. For this purpose, all angles were defined on the basis of 3 markers—1 marker placed distal to the joint (marker 1), another marker near the rotation center of the joint (marker 2), and a third marker placed proximal to the joint (marker 3). For example, the 3 markers used for the carpal joint were those on the distal aspect of the fifth metacarpal bone (marker 1), the styloid process of the ulna (marker 2), and the lateral epicondyle of the humerus (marker 3). Vector 1 was from marker 1 to marker 2, and vector 2 was from marker 2 to marker 3. The angle for each joint was calculated by use of the inverse tangent function.
Each motion cycle was adjusted to a neutral position to differentiate flexion and extension. For this purpose, the overall mean of the individual motion cycle was calculated and then subtracted from each time frame. By calculating the mean of the 5 motion cycles evaluated for each dog, an angle-time curve was determined for each dog.
Variables evaluated during the motion cycle for each joint were maximum flexion (°), maximum extension (°), angle velocity (radians/s), and angle acceleration (radians/s2). Range of motion (°) was calculated as maximum extension minus maximum flexion.
Statistical analysis—All data were tested for normal distribution by use of the Kolmogorov-Smirnoff test.k Data are reported as arithmetic mean ± SD. To detect differences between the evaluated variables during unimpeded walking and uphill walking, downhill walking, or obstacle exercise, an ANOVA for repeated measurements with a Bonferroni post hoc test was used. Values of P < 0.05 were considered significant.
Results
All data were normally distributed. The mean SIPFz was 1.1 ± 0.6% for the forelimbs and 1.7 ± 1.2% for the hind limbs. There were no significant differences in SIPFz among walking exercises.
In the study dogs, only minor differences in joint kinematics were evident during uphill walking, compared with findings during unimpeded walking on a horizontal surface (Table 1). During uphill walking, flexion in the hip joint was significantly (P = 0.01) greater, whereas there was a significantly (P = 0.01) negative change in extension of the stifle joint. With regard to ROM of any of the evaluated joints, no alteration was detected. Retrograde acceleration in the carpal (P = 0.02) and elbow joints (P = 0.02) was significantly decreased, compared with findings during normal walking. The elbow joint also had significantly (P = 0.02) lower maximum forward acceleration during uphill walking; this variable was not significantly changed in the carpal joint.
Maximum flexion, maximum extension, ROM, and maximum forward and retrograde velocities and accelerations for the left shoulder, elbow, carpal, hip, stifle, and tarsal joints in 8 nonlame dogs during uphill walking, downhill walking, and walking over low-level obstacles or unimpeded walking on a horizontal surface (normal walking).
| Joint | Exercise | Maximum flexion (°) | Maximum extension (°) | ROM (°) | Maximum forward velocity (radians/s) | Maximum retrograde velocity (radians/s) | Maximum forward acceleration (radians/s2) | Maximum retrograde acceleration (radians/s2) |
|---|---|---|---|---|---|---|---|---|
| Shoulder | NW | 12.5 ± 2.0 | 17.5 ± 3.3 | 30.1 ± 3.7 | 1.51 ± 0.4 | 2.27 ± 0.3 | 0.46 ± 0.1 | 0.41 ± 0.1 |
| UW | 11.7 ± 1.8 | 18.4 ± 2.3 | 30.2 ± 3.2 | 1.41 ± 0.3 | 2.20 ± 0.4 | 0.35 ± 0.1 | 0.32 ± 0.05 | |
| DW | 15.1 ± 2.0 | 17.3 ± 5.2 | 32.9 ± 5.5 | 1.33 ± 0.3 | 2.21 ± 0.3 | 0.34 ± 0.1 | 0.39 ± 0.1 | |
| OE | 14.6 ± 2.7 | 15.2 ± 2.9 | 29.6 ± 4.5 | 1.23 ± 0.4 | 2.01 ± 0.3 | 0.30 ± 0.1* | 0.32 ± 0.1 | |
| Elbow | NW | 30.0 ± 3.2 | 22.5 ± 5.4 | 52.9 ± 7.0 | 4.56 ± 0.5 | 3.6 ± 0.7 | 1.07 ± 0.3 | 0.89 ± 0.1 |
| UW | 30.2 ± 3.2 | 23.2 ± 5.7 | 54.2 ± 7.4 | 3.94 ± 0.6 | 3.26 ± 0.7 | 0.79 ± 0.2* | 0.70 ± 0.2* | |
| DW | 25.7 ± 3.6 | 17.1 ± 2.9 | 43.1 ± 5.8 | 3.39 ± 0.8 | 3.33 ± 0.7 | 0.70 ± 0.2* | 0.66 ± 0.2* | |
| OE | 35.7 ± 2.5† | 20.3 ± 4.4 | 57.0 ± 6.9 | 3.83 ± 0.4* | 4.23 ± 0.5 | 0.81 ± 0.2* | 0.57 ± 0.1* | |
| Carpal | NW | 48.5 ± 5.5 | 24.8 ± 4.8 | 73.9 ± 5.0 | 11.76 ± 1.1 | 7.99 ± 1.2 | 7.16 ± 2.0 | 2.54 ± 0.3 |
| UW | 45.6 ± 6.9 | 25.4 ± 5.0 | 69.3 ± 7.0 | 9.66 ± 1.1 | 6.69 ± 0.8 | 5.90 ± 1.7 | 1.66 ± 0.3* | |
| DW | 49.6 ± 9.2 | 26.0 ± 6.1 | 73.4 ± 10.5 | 11.27 ± 1.4 | 8.02 ± 1.3 | 6.50 ± 2.1 | 2.11 ± 0.4 | |
| OE | 55.0 ± 4.3† | 28.7 ± 4.4† | 81.8 ± 4.5 | 9.70 ± 1.4 | 6.74 ± 0.4 | 4.98 ± 1.8 | 1.58 ± 0.3* | |
| Hip | NW | 13.0 ± 3.3 | 14.8 ± 3.6 | 27.6 ± 6.7 | 1.54 ± 0.5 | 1.23 ± 0.3 | 0.34 ± 0.1 | 0.26 ± 0.05 |
| UW | 15.1 ± 3.7† | 15.1 ± 3.8 | 30.3 ± 7.1 | 1.59 ± 0.4 | 1.09 ± 0.3 | 0.21 ± 0.03 | 0.22 ± 0.05 | |
| DW | 11.0 ± 2.8* | 13.9 ± 3.6 | 24.9 ± 6.0* | 1.46 ± 0.4 | 1.00 ± 0.2 | 0.24 ± 0.05 | 0.17 ± 0.08* | |
| OE | 13.3 ± 3.4 | 14.2 ± 3.4 | 27.3 ± 6.3 | 1.49 ± 0.4 | 1.04 ± 0.3 | 0.22 ± 0.05 | 0.22 ± 0.07 | |
| Stifle | NW | 26.7 ± 5.1 | 18.5 ± 4.1 | 44.5 ± 8.1 | 2.49 ± 0.9 | 4.00 ± 0.6 | 0.81 ± 0.2 | 0.57 ± 0.1 |
| UW | 23.1 ± 2.3 | 13.1 ± 2.7* | 35.9 ± 3.8 | 2.16 ± 0.3 | 3.31 ± 0.6 | 0.55 ± 0.1 | 0.51 ± 0.1 | |
| DW | 25.1 ± 3.3 | 19.9 ± 3.5 | 44.7 ± 6.1 | 1.91 ± 0.3 | 3.30 ± 0.6 | 0.52 ± 0.2 | 0.50 ± 0.1 | |
| OE | 43.0 ± 7.6† | 22.5 ± 3.7† | 66.0 ± 10.1† | 3.96 ± 0.5† | 4. 77 ± 0.4 | 0.74 ± 0.1 | 0.76 ± 0.2 | |
| Tarsal | NW | 18.5 ± 4.5 | 19.9 ± 3.3 | 38.1 ± 7.1 | 4.41 ± 1.2 | 3.47 ± 1.0 | 1.35 ± 0.5 | 1.08 ± 0.4 |
| UW | 16.1 ± 2.0 | 21.9 ± 3.1 | 38.1 ± 4.0 | 3.51 ± 0.2 | 2.67 ± 0.4 | 0.82 ± 0.1 | 0.61 ± 0.06 | |
| DW | 17.1 ± 4.3 | 15.5 ± 2.5* | 32.6 ± 5.7 | 3.29 ± 0.4 | 2.56 ± 0.6 | 0.75 ± 0.2* | 0.60 ± 0.05 | |
| OE | 34.7 ± 5.6† | 19.5 ± 1.5 | 55.6 ± 7.4† | 4.76 ± 0.6 | 4.67 ± 1.0† | 0.99 ± 0.2 | 0.86 ± 0.1 |
Uphill and downhill walking were performed on a treadmill inclined at a gradient of 11%. For the obstacle exercise, 5 cavaletti were each adjusted to the height of the carpal joint and positioned on a horizontal surface at intervals set according to the distance between the forelimbs and hind limb of each dog. Data are reported as mean ± SD.
Within a joint, value is significantly (P < 0.05) lower than the value determined during normal walking.
Within a joint, value is significantly (P < 0.05) higher than the value determined during normal walking.
DW = Downhill walking. NW = Normal walking. OE = Obstacle exercise. UW = Uphill walking.
Compared with findings during unimpeded walking on a horizontal surface, downhill walking caused a significant (P = 0.04) decrease in flexion of the hip joint, combined with a significantly (P = 0.02) lower ROM. With regard to joint extension, downhill walking affected (P = 0.01) only the tarsal joint; extension in that joint was comparatively decreased. Decreases in forward joint angle acceleration in the elbow (P = 0.05) and tarsal joints (P = 0.02) were detected. Retrograde joint angle acceleration in the elbow (P = 0.03) and the hip (P = 0.04) joints was significantly decreased when dogs were walking downhill versus walking unimpeded on a horizontal surface.
Low-level obstacle exercise caused most changes in the evaluated joint variables, compared with findings during unimpeded walking on a horizontal surface. Flexion was significantly increased in the elbow (P = 0.01), carpal (P = 0.04), stifle (P = 0.01), and tarsal (P = 0.01) joints. Extension was significantly increased in the carpal (P = 0.03) and stifle (P = 0.02) joints. Compared with normal walking, changes in ROM were limited to joints of the hind limb; ROMs of the tarsal and stifle joints were significantly (P = 0.01) greater during obstacle exercise.
Analysis of acceleration and velocity variables revealed complex biomechanical interactions. Compared with findings during unimpeded walking on a horizontal surface, the carpal joint had significantly (P = 0.01) lower retrograde acceleration during uphill walking and obstacle exercise. For the elbow joint, maximum forward velocity significantly (P = 0.04) decreased during obstacle exercise, as did maximum forward acceleration (P = 0.05) and maximum retrograde acceleration (P = 0.01) during uphill walking, downhill walking, and obstacle exercise. The shoulder joint had significantly (P = 0.01) lower maximum forward acceleration during obstacle exercise, compared with normal walking; this was the only significant change associated with the 3 activities. For the tarsal joint, maximum retrograde velocity during obstacle exercise was significantly (P = 0.03) greater than the corresponding value during normal walking; maximum forward acceleration during downhill walking was significantly (P = 0.02) decreased. During obstacle exercise, the maximum forward velocity of the stifle joint increased significantly (P = 0.01), compared with normal walking; there were no other significant changes in acceleration and velocity variables during obstacle exercise or during uphill and downhill walking. For the hip joint, the only acceleration or velocity variable that was altered during the 3 activities was maximum retrograde acceleration during downhill walking; the value was significantly (P = 0.04) decreased, compared with unimpeded walking on a horizontal surface.
Discussion
Results of the present study indicated that uphill walking, downhill walking, and walking over obstacles on a horizontal surface each had specific impacts on the evaluated kinematic joint variables in dogs. Obstacle exercise involving low-level cavalletti appears to have physiotherapeutic usefulness for increasing flexion of the carpal, elbow, tarsal, and stifle joints and increasing extension of the carpal and stifle joints. When walking over the obstacles on a horizontal surface, dogs were required to raise their extremities higher in the swing phase, compared with efforts required during unimpeded walking. However, increases in joint flexion can be contraindicated, especially within the first weeks after tibial plateau leveling osteotomy or tibial tuberosity advancement. These surgical techniques result in a positional change of either the tibial plateau or the tibial tuberosity, which alters the moment arm acting on the patellar tendon.2,b In dogs that have undergone tibial plateau leveling osteotomy or tibial tuberosity advancement, this change in tension dynamics of the tendon has to be considered to prevent excessive stress on the tendon during physical therapy.
The slight but also significant increase in extension of the stifle joint during obstacle exercise in the present study occurred at the end of the swing phase. This increased extension was caused by lifting of the limb extremity and a compensational effect of increased stifle joint flexion that occurred in the mid-swing phase. The biomechanical characteristics of extension of the carpal joint during obstacle exercise can be similarly explained.
With exception of the hip joint, assessments of all of the evaluated joints revealed complex changes of the temporospatial characteristics (ie, velocity and acceleration variables), particularly during obstacle exercise. For example, to maintain a normal gait pattern during obstacle exercise, forward velocity of the stifle joint increased. In our opinion, these changes occur because walking over obstacle rails is associated with a higher degree of advertence than unimpeded walking. Furthermore, there is greater demand on the dogs’ proprioceptive skills, which explains the considerable kinematic changes. On the basis of the results of the present study, obstacle exercise involving low-level cavalletti can be considered an active exercise that causes increased flexion of the carpal, elbow, tarsal, and stifle joints, with no evident effect on the shoulder or hip joint. In terms of joint extension, obstacle exercise of this nature appears to provide therapeutic benefit in the carpal and stifle joints. The complex biomechanical mechanisms identified in the present study support the suggestion that obstacle exercise can be beneficial for patients with orthopedic problems as well as for patients with neurologic disease, especially those that need improvement in voluntary motor control and greater accuracy in placement of the limbs.27
In the dogs of the present study, uphill walking caused an alteration of the joint angles of the hind limb; the forelimb was not affected by this activity. Compared with unimpeded walking on a horizontal surface, hip joint flexion is significantly increased when walking uphill; in addition, stifle joint extension is significantly decreased. This can be explained by the fact that walking on an inclined gradient requires the hip joint to be flexed more strongly during the physiologic movement, compared with walking on level ground. In contrast, to move forward on an upward slope, extension of the stifle joint must be decreased. Similar observations in dogs have been made by Gassel et al.l Relating these results to musculature, walking a dog up a gradient aids in strengthening the muscles that cause flexion of the hip joint. In our study, increased extension of joints during uphill walking was not evident, but it is obvious that a great amount of muscle strength is needed in the hip and stifle joints to propel a body up an incline.
Considering this, it can be assumed that uphill walking is useful for strengthening the hamstrings (biceps femoris, semitendinosus, and semimembranosus muscles), which are responsible for the forward propulsion. Nevertheless, further investigation of dogs walking up steeper gradients with additional assessments, such as electromyography, should be performed. Overall, uphill walking, as performed in our study, can be used to enhance the flexion of the hip joint in dogs.
In contrast to uphill walking, downhill walking caused significantly less hip joint flexion than unimpeded walking on a horizontal surface, which is not consistent with the hypothesis of Hamilton et al27 that flexion of the hind limb joints increases comparatively during downhill walking. However, generalizations should not be made because there is still a lack of data from studies involving different experimental protocols including varied declinations and walking velocities. Although kinematic changes in the forelimb during downhill walking, compared with uphill walking, would also be expected, none of the detected changes were sufficiently pronounced to be significant. With regard to the temporospatial variables evaluated (ie, maximum velocity and acceleration [both forward and retrograde]) in the present study, speed and acceleration were generally decreased during uphill or downhill walking, compared with findings during unimpeded walking on a horizontal surface. Uphill walking was associated with significant changes in the forelimb kinematics only, specifically in the carpal and elbow joints. Downhill walking was associated with significant changes in the elbow, tarsal, and hip joints. Unlike uphill walking, downhill walking significantly affected the fore- and hind limbs in the study dogs. Given the general reduction in speed and acceleration during either uphill or downhill walking, the authors suggest that both maneuvers involve a change of gait pattern. Contrary to normal gait on a horizontal surface, the automated movement alters into more intentional motion. That and the fact that the slope of 11% forms an obstacle are reasons for the evaluated biomechanical changes.
To substantiate our findings, a greater number of dogs and a reevaluation of measurements during exercises with different experimental setups (eg, different heights of and distances between obstacles and gradients for uphill and downhill walking) would be desirable. Use of varied experimental setups and comparisons between left and right limbs in individual dogs would allow better determination of clinical relevance.
Undoubtedly, investigation of a more homogeneous dog population would give a better understanding of the changes in the evaluated variables, but the dogs included in our study did have similar morphometric patterns. To increase accuracy of measurements, surgical fixation of the markers to each dog would avoid or reduce the geometric instability of the skin, but such a procedure is problematic in terms of animal welfare. Furthermore, it is also known that bone-fixated markers cause altered gait patterns.28 The use of a body covering to avoid concealment of the proximal markers by a dog's hair could also have influenced the data derived for the shoulder and hip joint angles; however, on comparison of the results of the present study with previously published data,29,30 we found comparable values for the angulations as well as for the SD values.
Another aspect of the present study that should be considered is the different measurement velocity. Kinetic evaluation was always carried out at 1.2 m/s, whereas the gait speed during performance of the exercises varied. This was a consequence of the exercise itself, which represented a handicap; the difference between the obstacle exercise (0.89 ± 0.11 m/s) and unimpeded walking on a horizontal surface (1.06 ± 0.21 m/s) appeared to be particularly marked. In terms of obstacle exercise, the reduction of speed was volitional. It was important that the dogs did not hop but stepped in a controlled manner over the low-level cavaletti. This was essential for achieving therapeutic effects and is the explanation for the decrease in speed during performance of the obstacle exercise. One might think that the measured alterations of the joint kinematics were caused more by the different gait velocities than by the exercises. With regard to measurement of GRFs, it has been shown that PFzs significantly increase when there is an increase in velocity of 0.6 m/s; changes in velocity > 0.3 m/s can be neglected because these can be seen as physiologic walking speed shifts that cause no alteration in results.31 We suggest that the effects of velocity changes in kinematics can be considered analogous to those in GRF measurement, although there are no published reports to confirm this, to our knowledge.
The results of the present study should also be interpreted in light of the fact that only nonlame dogs were evaluated. It is likely that dogs with various orthopedic disorders would have different joint biomechanics. For the assessment of clinical results and treatment effects associated with veterinary physical therapy, a long-term survey should be carried out with repeated measurements in treated dogs. Also, the measurement of muscle activity during performance of the exercises would provide a more comprehensive understanding of the physiologic processes involved.
On the basis of the results of the present study, it appears that each evaluated exercise, except for downhill walking, has a specific therapeutic value in veterinary physical therapy for dogs. Obstacle exercise over low-level cavalletti appears potentially useful for improving flexion of joints in the fore- and hind limbs; in addition, the ROM of joints can easily be increased but only in the hind limb. Uphill walking is an easy exercise to undertake and could be used to improve joint flexion, primarily in the hip joint. Dogs can be engaged in all of the exercises involved in the present study by a therapist or an owner without expensive special equipment. Undoubtedly, because not all evaluated variables were positively affected by every exercise, each exercise has its specific indications. Nevertheless, the results of the present study are a basis for guiding active movement protocols into an evidence-based field of veterinary medicine.
ABBREVIATIONS
| GRF | Ground reaction force |
| PFz | Peak vertical force |
| ROM | Range of motion |
| SIPFz | Symmetry indices of PFz |
Johnston KD, Levine D, Price MN, et al. The effects of TENS on osteoarthritic pain in the stifle of dogs (abstr), in Proceedings. 2nd Int Symp Phys Therap Rehab Vet Med 2002;199.
Millis DL, Schwartz P, Hicks DA, et al. Kinematic assessment of selected therapeutic exercises in dogs (abstr), in Proceedings. 3rd Int Symp Phys Therap Rehab Vet Med 2004;215.
Typ 9011 A, 25 × 50 cm, Kistler GmbH, Vienna, Austria.
SIMI Motion 3D, Simi Reality Motion Systems GmbH, Unter-schleissheim, Germany.
Tube Bandage, Lohmann Rauscher, Vienna, Austria.
Eagle Digital RealTime System, Motion Analyses Corp, Santa Rosa, Calif.
DELL Precision 690 PC, INTEL Xenon 3.20 Ghz CPU, 2 GB RAM, Dell Inc, Round Rock, Tex.
EVaRT, version 5.0.4, Motion Analyses Corp, Santa Rosa, Calif.
Excel, Microsoft Office 2007 for Windows, Microsoft Corp, Redmond, Wash.
MatLab, version 7.4.0.287 (R2007a), The Mathwork, Natick, Mass.
SPSS, version 14.0, SPSS Inc, Chicago, Ill.
Gassel AG, Millis DL, Schwartz P, et al. Kinematic gait analysis of the pelvic limb; comparison of overground versus incline walking in 10 dogs (abstr), in Proceedings. 32nd Annu Conf Vet Orthop Soc 2005;29.
References
- 1.↑
Levine D, Millis DL, Marcellin-Little DJ. Introduction to veterinary physical rehabilitation. Vet Clin North Am Small Anim Pract 2005;35:1247–1254.
- 2.↑
Davidson JR, Kerwin SC, Millis DL. Rehabilitation for the orthopedic patient. Vet Clin North Am Small Anim Pract 2005;35:1357–1388.
- 3.
Olby N, Halling KB, Glick TR. Rehabilitation for the neurologic patient. Vet Clin North Am Small Anim Pract 2005;35:1389–1409.
- 4.↑
Johnson JM, Johnson AL, Pijanowski GJ, et alRehabilitation of dogs with surgically treated cranial cruciate ligament-deficient stifles by use of electrical stimulation of muscles. Am J Vet Res 1997;58:1473–1478.
- 5.↑
Marsolais GS, Dvorak G, Conzemius MG. Effects of postoperative rehabilitation on limb function after cranial cruciate ligament repair in dogs. J Am Vet Med Assoc 2002;220:1325–1330.
- 6.↑
Monk ML, Preston CA, McGowan CM. Effect of early intensive postoperative physiotherapy on limb function after tibial plateau leveling osteotomy in dogs with deficiency of the cranial cruciate ligament. Am J Vet Res 2006;67:529–536.
- 7.↑
Gandini G, Cizinauskas S, Lang J, et alFibrocartilaginous embolism in 75 dogs: clinical findings and factors influencing the recovery rate. J Small Anim Pract 2003;44:76–80.
- 8.↑
Kathmann I, Cizinauskas S, Doherr MG, et alDaily controlled physiotherapy increases survival time in dogs with suspected degenerative myelopathy. J Vet Intern Med 2006;20:927–932.
- 9.↑
Gruenenfelder FI, Boos A, Mouwen M, et alEvaluation of the anatomic effect of physical therapy exercises for mobilization of lumbar spinal nerves and the dura mater in dogs. Am J Vet Res 2006;67:1773–1779.
- 10.↑
Hottinger HA, DeCamp CE, Olivier NB, et alNoninvasive kinematic analysis of the walk in healthy large-breed dogs. Am J Vet Res 1996;57:381–388.
- 11.
Owen MR, Richards J, Clements DN, et alKinematics of the elbow and stifle joints in greyhounds during treadmill trotting—an investigation of familiarisation. Vet Comp Orthop Traumatol 2004;17:141–145.
- 12.
Clements DN, Owen MR, Carmichael S, et alKinematic analysis of the gait of 10 Labrador retrievers during treadmill locomotion. Vet Rec 2005;156:478–481.
- 13.
Bertram JEA, Lee DV, Case HN, et alComparison of the trotting gaits of Labrador Retrievers and Greyhounds. Am J Vet Res 2000;61:832–838.
- 14.
Burton NJ, Dobney JA, Owen MR, et alJoint angle, moment and power compensations in dogs with fragmented medial coronoid process. Vet Comp Orthop Traumatol 2008;21:110–118.
- 15.
Kim SE, Pozzi A, Banks SA, et alEffect of tibial tuberosity advancement on femorotibial contact mechanics and stifle kinematics. Vet Surg 2009;38:33–39.
- 16.
Borer LR, Peel JE, Seewald W, et alEffect of carprofen, etodolac, meloxicam, or butorphanol in dogs with induced acute synovitis. Am J Vet Res 2003;64:1429–1437.
- 17.
Moreau M, Dupuis J, Bonneau NH, et alClinical evaluation of a nutraceutical, carprofen and meloxicam for the treatment of dogs with osteoarthritis. Vet Rec 2003;152:323–329.
- 18.
Cross AR, Budsberg SC, Keefe TJ. Kinetic gait analysis assessment of meloxicam efficacy in a sodium urate-induced synovitis model in dogs. Am J Vet Res 1997;58:626–631.
- 19.↑
Marsolais GS, McLean S, Derrick T, et alKinematic analysis of the hind limb during swimming and walking in healthy dogs and dogs with surgically corrected cranial cruciate ligament rupture. J Am Vet Med Assoc 2003;222:739–743.
- 20.↑
Feeney LC, Lin CF, Marcellin-Little DJ, et alValidation of two-dimensional kinematic analysis of walk and sit-to-stand motions in dogs. Am J Vet Res 2007;68:277–282.
- 21.↑
Yanoff SR, Hulse DA, Hogan HA, et alMeasurements of vertical ground reaction force in jumping dogs. Vet Comp Orthop Traumatol 1992;5:44–50.
- 22.↑
Baumgartner W. Klinische Propädeutik der inneren Krankheiten und Hautkrankheiten der Haus- und Heimtiere. 6th ed. Stuttgart, Germany: Parey, 2005;5–245.
- 24.
Off W, Matis U. Summary: gait analysis in dogs. Tierarztl Prax Ausg G Grosstiere 1997;25:8–14.
- 25.
Off W, Matis U. Gait analysis in dogs. Part 2: installation of a gait analysis laboratory and locomotor studies. Tierarztl Prax Ausg G Grosstiere 1997;25:303–311.
- 26.↑
Peham C, Scheidl M, Licka T. Limb locomotion—speed distribution analysis as a new method for stance phase detection. J Biomech 1999;32:1119–1124.
- 27.↑
Hamilton S, Millis DL, Taylor RT, et alTherapeutic exercises. In: Millis DL, Levine D, Taylor RT, eds. Canine rehabilitation and physical therapy. St Louis: Saunders, 2004;244–263.
- 28.↑
Haussler KK, Bertram JEA, Gellman K, et alSegmental in vivo vertebral kinematics at the walk, trot and canter: a preliminary study. Equine Vet J Suppl 2001;(33):160–164.
- 29.
Bockstahler BA, Henninger W, Müller M, et alInfluence of borderline hip dysplasia on joint kinematics of clinically sound Belgian Shepherd dogs. Am J Vet Res 2007;68:271–276.
- 30.
Bockstahler B, Müller M, Henninger W, et alKinetic and kinematic motion analysis of the forelimbs in sound Malinois dogs—sampling of basic values. Wien Tierärztl Monatschr 2008;95:127–138.
- 31.↑
Riggs CM, DeCamp CE, Soutas-Little RW, et alEffects of subject velocity on force plate-measured ground reaction forces in healthy greyhounds at the trot. Am J Vet Res 1993;54:1523–1526.
