Abstract
Objective
To investigate the relationships between hindlimb joints angles under the passive movement of hip joints within a relatively wide range of motion in anesthetized Beagles.
Methods
9 hindlimbs of 5 clinically normal adult Beagles were included from September 2022 through August 2023. The tested hindlimb was positioned horizontally, with the dog under general anesthesia and muscle relaxation. Unforced angle (UA; when the paw was unforced), forced angle range (FAR; when the paw was under constant pushing/pulling force), and UA range (UAR; when the forces applied to the paw had been withdrawn) of the stifle and tarsal joints were evaluated using radiographic analysis while the hip joint angle was passively adjusted from 70° to 170°. Linear regression analysis was used to assess the relationships between hip, stifle, and tarsal joint angles.
Results
Significantly strong linear relationships (adjusted R2 > 0.7) were found for hip-stifle and hip-tarsal in UA, UAR, and when the paw was pushed with the hip joint angle exceeding 120° (FAR-push > 120°). The FAR of stifle and tarsal joints changed with the hip joint angle. Additionally, significantly strong linear relationships (adjusted R2 > 0.85) were found for stifle-tarsal in UA, UAR, and when a constant push force was applied on the paw (FAR-push).
Conclusions
This research revealed the linear relationships between canine hindlimb joint angles in the sagittal plane under specific circumstances as well as the narrowing of the FAR of stifle and tarsal joints when hip joint angle exceeds 120°.
Clinical Relevance
The findings on linkage mechanism–related relationships in hindlimb joints angles from this research may contribute to a better understanding of orthopedic and rehabilitation practices.
In dogs, especially in chondrodystrophic breeds, thoracolumbar intervertebral disk disease is the most common cause of acute paraparesis and paraplegia.1,2 For these canine patients, recovery of ambulation is considered important for both their quality of life and their owners.3,4 Thus, physical rehabilitation, which aims to maintain the normal range of motion (ROM) in the joints, moderate muscle atrophy, and improve neurological recovery,5 is usually regarded as an essential part of postoperative management. Intensive rehabilitation, as a branch of physical rehabilitation, including treadmill walking, swimming, balance exercises, and other advanced gait or proprioception exercises, with its effect in small animal patients with spinal cord injury being confirmed by several researchers, suggests that encouraging autonomic movement is more beneficial to muscle mass regaining6 and locomotor recovery7–11 as well as neurological recovery.12
The concept of a canine exoskeleton that can support functional ambulation while also stimulating plasticity of motor pathways has been put forward by the Canine Exoskeleton for Rehabilitation Team from Colorado State University since 2015. They were determined to design an exoskeleton for large-breed dogs by mimicking repeatable hindlimb motion, but until 2022 they were still struggling to improve the brace and control systems.13 A canine exoskeleton that aims to realize complete motion imitation is believed to have limitations on volume, weight, and battery capacity.14 We supposed this might be the reason for the delay in the Colorado State University team’s progress. Thus, we planned to design a simplified exoskeleton focused only on maintaining load bearing and repeating the basic forward swing of hindlimbs, which could be lighter in weight and easier to put into use. In humans, it is confirmed that the positions of the hip and stifle joints affect the ROM of the ankle joint.15–17 Using the linkage between joint angles in the design of long-leg braces for human patients, the swing phase can be achieved through the reciprocating motion of the hip joint, and the stance phase can be maintained by stabilization of the stifle and tarsal joints.18 We were curious about whether similar linkage between joint angles also exists in canine hindlimbs. If so, it would be the first step for the design of a simplified canine exoskeleton.
In dogs in motion, a phenomenon where the femur and metatarsus remain relatively parallel during most of the stance phase was described and defined as “pantograph behavior” (or “pantograph leg”) by Fischer and Lilje.19 This phenomenon was explained by the action of biarticular muscles spanning both stifle and tarsal joints,19 whereas the effect of muscle tone generated by active muscle contraction remained undiscussed as far as we know. Since the simplified exoskeleton is designed for lower motor neuron hindlimb paraparetic or paralyzed canine patients, whose muscle tone is defective or absent, it is necessary to know whether the pantograph relationship still exists without muscle contraction. To our knowledge, except for the investigation of the relationship between the ROM of stifle and tarsal joints in the canine cadaveric limb where only gastrocnemius muscle remained (conducted by Sugiyama et al,20) little research has been conducted regarding the relationships between canine hindlimb joints without the effect of muscle contraction, especially in vivo. Additionally, according to our knowledge, most existing studies21,22 on the relationships between hindlimb joint angles in dogs are based on motion analysis in awake dogs, in which condition the hip joint movement is limited to the active ROM (AROM).
Therefore, the first purpose of this study was to investigate the related changes in stifle and tarsal joint angles under the passive movement of the hip joint without the effect of muscle tone in Beagles. Considering that the linkage function of biarticular muscles may still exist without muscle tone, it is hypothesized that the angles of the stifle and tarsal joints exhibit linear relationships with the hip joint angle under some circumstances. The second purpose of this study was to explore the relationships between hindlimb joint angles when the hip joint is passively moved within a wider ROM than AROM.
Methods
Five sound, intact adult Beagles aged between 12 and 84 months23 owned by Nippon Veterinary Life and Science University were selected for this study. All dogs underwent complete physical, orthopedic, and neurologic examinations, as well as CT scans, in advance. No orthopedic or neurologic abnormalities were detected. The study period was from September 2022 through August 2023. The radiographic data for each dog were collected from either 1 day (2 of 5 dogs) or several separate days (2 of 5 dogs from 2 separate days and 1 of 5 dogs from 4 separate days). On each experimental day, age, body weight, and body condition score (BCS) were recorded. The 9-point BCS scale was used to assess BCS.24 For the 3 dogs undergoing several experimental days, the experimental period was 51, 54, and 90 days separately. For them, body weight and BCS were recorded as the mean value of each experimental day, and age was the median value of that on the first and last experimental day. The study was approved by the IACUC of Nippon Veterinary and Life Science University (approval No: 2022S-21; 2023K-41).
Animal preparation and position
The tested dog was under general anesthesia induced by IV propofol (PropoFlo 28; 2 to 6 mg/kg) and maintained with isoflurane (1.5% to 2.5%). Then, an initial IV bolus of rocuronium (ESLAX; 0.7 mg/kg) was administered, followed by a continuous rate infusion of 0.5 to 0.75 mg/kg/h to maintain muscle relaxation. A ventilator (Spiritus Anesthesia Ventilator; ACOMA) was connected to the breathing circuit to monitor the effect of muscle relaxant that suppresses spontaneous respiration and to assist with ventilation simultaneously. Physiological parameters, including heart rate, blood pressure, body temperature, oxygen saturation, and end-tidal carbon dioxide, were monitored by a multi-item monitor (Bio-Scope AM140; FUKUDA M·E) to ensure the safety of the tested dog during anesthesia.
After the anesthesia stabilized, the dog was dressed in a pair of hand-made Lycra pants with the tightness adjusted, then positioned in lateral recumbency on a custom-made operating table (100 X 50 X 5 cm, with a 3-cm-high gap beneath the table to allow the placement and removal of the imaging plate). The tested hindlimb was elevated on a shelf at a suitable height to make it horizontal (Figure 1). A mobile x-ray machine (IMC-125; TOSHIBA) was positioned beside the operating table, with the x-ray beams directed perpendicularly to the table. Three columns were used to immobilize the pelvis, ensuring that the 2 greater trochanters were superimposed in the lateral x-ray. The hip joint was then flexed and extended to ensure that neither the Lycra pants nor the columns interfered with the movement of the tested hindlimb. A plastic ball (diameter, 18 mm) was attached to the Lycra pants to mark the location of greater trochanter of the femur. The untested hindlimb was positioned away from the tested 1 and marked with a staple stuck to the distal tibia region by surgical tape.
Description of animal position and data acquisition. A—The tested hindlimb of an anesthetized dog ready for the experiment. The dog has already been dressed in Lycra pants, with a plastic ball attached to mark the location of the greater trochanter of the femur. The pelvis was immobilized with 3 columns, and the tested limb was positioned horizontally by elevating it on a shelf. B—Diagram of the anatomical landmarks for measuring joint angles based on the lateral radiographic view of the pelvis and tested hindlimb. Point A: the cranial aspect of the ilial wing. Point B: the greater trochanter of the femur. Point C: the lateral femoral condyle. Point D: the lateral malleolus of the fibula. Point E: the distolateral aspect of the fifth metatarsal bone. The hip joint is represented by points ABC, the stifle joint by BCD, and the tarsal joint by CDE. C—Positioning the tested limb when a constant push force was applied on the paw (FAR-push). The wooden clamp was used to adjust and fix the position of the femur by clamping the medial and lateral condyles of femur. The pushing force was conducted through the real-time dynamometer connecting with the paw. A dynamometer-attached laser was used to ensure that the orientations of the force were directed toward the greater trochanter by pointing at the greater trochanter–indicating plastic ball. An x-ray was taken when the stifle and tarsal joints reached maximal flexion. D—Lateral radiographic view of the pelvis and tested hindlimb for FAR-push. The tested hindlimb is indicated by the connected dynamometer.
Citation: American Journal of Veterinary Research 2025; 10.2460/ajvr.25.02.0067
Data acquisition
Unforced angle (UA) was defined as the stifle and tarsal joint angles collected when the paw was unforced. A wooden clamp, clamping the medial and lateral condyles of the femur, was used to adjust the tested limb’s hip joint angle from approximately 90° to 170°. After the initial position of the distal femur was determined by a goniometer, the hip joint was extended gradually by the clamp. Though the target interval was set at 10°, the actual interval during radiographing was determined by visual inspection to minimize the total anesthesia time. X-rays were taken each time. The joint angles were indicated by the anatomical landmarks as follows: A, the cranial aspect of the iliac wing; B, the greater trochanter of the femur; C, the lateral femoral condyle; D, the lateral malleolus of the fibula; and E, the distolateral aspect of the fifth metatarsal bone. The hip joint angle was represented as ∠ABC, the stifle joint angle as ∠BCD, and the tarsal joint angle as ∠CDE (Figure 1). All hip, stifle, and tarsal joint angle data were acquired by the measurement tools of a medical image viewer (XiaoSaiViewer, version 3.2.1; Sino United Medical Technology Co LTD).25
The UA range (UAR) was determined by the angle after withdrawing the push force applied to the paw (UAR-push) and the angle after withdrawing the pull force applied to the paw (UAR-pull). The withdrawn push force in UAR-Push made the stifle and tarsal joints reach the maximal flexion. Similarly, the withdrawn pull force in UAR-pull made the stifle and tarsal joints completely extended. The stifle and tarsal joints were subjectively determined to have reached maximal flexion and extension when no more obvious changes occurred with further pushing/pulling. A real-time dynamometer (AD-4932A-50N; AND) was used with its probe connected with the paw in the area among the digital pads and metacarpal pad. The orientations of the forces were directed toward or away from the greater trochanter, ensured by a dynamometer-attached laser pointing at the greater trochanter–indicating plastic ball when applying the pushing or pulling force (Figure 1). The hip joint angle was adjusted from 70° to 170° in approximate 10° intervals. In each position, the paw was pushed and pulled to achieve the maximal flexion and extension of the stifle and tarsal joints. X-rays were taken after withdrawing the pushing or pulling force each time. The hip joint angle, unforced flexion (UAR-push) angles of the stifle and tarsal joints, and unforced extension (UAR-pull) angles of the stifle and tarsal joints were measured and recorded. This part was designed as a supplement to UA to explore whether the initial position of the paw affects the relationships between hindlimb joint angles.
The forced angle range (FAR) was defined as the range between the angle when constant push force applied on the paw (maximal flexion angle; FAR-push angle) and the angle when constant pull force applied on the paw (maximal extension angle; FAR-pull angle). The process of UAR that adjusts the hip joint angle from 70° to 170° was repeated in approximate 10° intervals, with the greater trochanter–directed pushing/pulling force applied on the paw to achieve the maximal flexion and extension of the stifle and tarsal joints in each position. X-rays were taken during the application of pushing or pulling force each time. The hip joint angle, FAR-push angles of stifle and tarsal joints, and FAR-pull angles of stifle and tarsal joints were measured and recorded. All measurements were made by a single observer.
Statistical analysis
To examine the relationship between hip joint angle (independent variable) and other hindlimb joint angles (dependent variable), preanalysis was conducted in data-processing software (Excel; Microsoft Corp) to check the R2 values for each type of trendline: linear trendline, polynomial trendline, logarithmic trendline, power trendline, and exponential trendline. Through comparison, the R2 value of linear trendline was confirmed to be the highest.
As a result, the statistical analyses were performed using statistical computing and graphics language (RStudio, version 4.3.3; Posit Software PBC),26 employing linear regression analysis to evaluate the linear relationships between hip, stifle, and tarsal joint angles in all hindlimbs, with the α level set as 0.05.
Results
Nine hindlimbs from 5 sound adult Beagles (2 males and 3 females, all intact) were included in this study. The 5 Beagles had a median age of 48 months (range, 15 to 82.5) and a median body weight of 10.1 kg (range, 10 to 16), and all of the BCSs were within the range of 4/9 to 5/9. One dog only provided its right hindlimb data due to the limitations in the initial pilot study, for it would exceed the age limit of 84 months old if continue the experiment on its left hindlimb.
Unforced angle and UAR
With all adjusted R2 values > 0.7 and P values < .001, the linear relationships between hip and stifle in UA and UAR-push/pull were significantly strong. Similarly, significantly strong linear relationships in UA and UAR-push/pull were found between hip and tarsal (Table 1).
Summary data for the linear formula of hip-stifle, hip-tarsal in unforced angle (UA), UA after withdrawing the push force (UAR-push), UA after withdrawing the pull force (UAR-pull), and angle under the constant push force when the hip joint angle is > 120° (FAR-push > 120°).
| Hip-stifle: y = ax + b | Hip-tarsal: z = cx + d | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| a | b | R2adj | P value | n | c | d | R2adj | P value | n | |
| UA | 0.72 | 25 | 0.87 | < .001 | 163 | 0.45 | 73 | 0.74 | < .001 | 163 |
| UAR-push | 0.74 | 15 | 0.87 | < .001 | 123 | 0.45 | 68 | 0.83 | < .001 | 122 |
| UAR-pull | 0.72 | 30 | 0.93 | < .001 | 120 | 0.34 | 91 | 0.74 | < .001 | 119 |
| FAR-push > 120° | 1.9 | −200 | 0.75 | < .001 | 127 | 1.4 | −120 | 0.74 | < .001 | 127 |
a,b,c,d = Parameters for linear formula. R2adj = Adjusted R2 value. x = Hip. y = Stifle. z = Tarsal.
The related changes of stifle UA and UAR with change in hip angle were plotted together in Figure 2, as were the related changes of tarsal UA and UAR. The strong linearity of the fitting line was visualized, with all adjusted R2 values for the linear formulas exceeding 0.7, among which the adjusted R2 values for hip-tarsal were smaller than those for hip-stifle. It was also observed that most parts of the UA fitting line were within the UAR for both hip-stifle and hip-tarsal.
A and B—The unforced angle (UA) and UA range (UAR) of the stifle (A) and ankle (B) with changes in hip angle collected from 9 hindlimbs from 5 Beagles. Each dot represents the angles of 1 radiographic picture. Black dots represent the UA, blue dots represent the UAR after withdrawing the pull force (UAR-pull), and yellow dots represent the UAR after withdrawing the push force (UAR-push). Linear regression analysis was conducted on the dots of UA, UAR-pull, and UAR-push separately. The matching colored line represents the fit line, and shading represents the CIs (α level, 0.05).
Citation: American Journal of Veterinary Research 2025; 10.2460/ajvr.25.02.0067
Forced angle range
The FAR of stifle and tarsal with changes in hip angle for each hindlimb were plotted separately in Figure 3. For both stifle and tarsal, the dots of push were divided into push < 120° (hip angle < 120°) and push > 120° (hip angle > 120°).
A and B—The forced angle range (FAR) of the stifle (A) and tarsal (B) with changes in hip angle in each hindlimb: left and right hindlimbs of dog DAIFUKU (DF-L, DF-R), right hindlimb of dog POTTER (PT-R), left and right hindlimbs of dog RYO (RY-L, RY-R), left and right hindlimbs of dog SAKURA (SK-L, SK-R), and left and right hindlimbs of dog WAMU (WM-L, WM-R). Each dot represents the angles of 1 radiographic picture. Blue dots represent angles under the constant pull force (maximal extension; FAR-pull), yellow dots represent angles under the constant push force when the hip angle is < 120° (FAR-push < 120°), and red dots represent angles under the constant push force when the hip angle is > 120° (FAR-push > 120°). C through F—The general variation tendencies for hip-stifle (C) and hip-tarsal (D) in FAR-pull and hip-stifle (E) and hip-tarsal (F) in FAR-push collected from 9 hindlimbs. Each dot represents the angles of 1 radiographic picture. Each colored curve represents the fitting curve of the corresponding hindlimb as shown to the right of panels D and F. Shading represents the CIs (α level, 0.05).
Citation: American Journal of Veterinary Research 2025; 10.2460/ajvr.25.02.0067
The general variation tendencies for hip-stifle and hip-tarsal in FAR-pull and FAR-push were also plotted in Figure 3. In the general variation tendency in FAR-pull, the pull angles of stifle and tarsal were maintained in a relatively high range in most cases. In the general variation tendency in FAR-push, as the hip angle increased, the change in stifle angle was initially mild. When the hip angle was > 120°, the slope of the fitting line increased abruptly, demonstrating a significantly strong linear relationship with hip angle (adjusted R2 = 0.75; P < .001). The general variation tendency of the tarsal was similar, showing a significantly strong linear relationship with hip angle (adjusted R2 = 0.74; P < .001) when the hip angle was > 120° (Table 1).
Although the patterns of either FAR-pull or FAR-push < 120° in stifle and tarsal could hardly be concluded in a specific formula, a proximate range for the FAR of stifle and tarsal with different hip joint angles was portrayed, with a notably narrow range observed when the hip angle was > 120°.
Stifle-tarsal
The linear relationships between stifle and tarsal angles in UA and FAR-push for each limb and for all limbs are presented in Figure 4.
A and B—The relationships between stifle and tarsal joint angles in FAR-push (A) and UA (B) for each limb. Each dot represents the angles of 1 radiographic picture. C—The general relationship between stifle and tarsal joint angles in FAR-push (C) collected from 9 hindlimbs. Linear regression analysis was conducted. The green line represents the fit line, and shading represents the CIs (α level, 0.05). D—The general relationships between stifle and tarsal angles in UA and UAR (D) collected from 9 hindlimbs. Black dots represent UA, blue dots represent UAR-pull, and yellow dots represent UAR-push. Linear regression analysis was conducted on the dots of UA, UAR-pull, and UAR-push separately. The matching colored line represents the fit line, and shading represents the CIs (α level, 0.05).
Citation: American Journal of Veterinary Research 2025; 10.2460/ajvr.25.02.0067
As shown in Figure 4 and Table 2, the adjusted R2 value of the fitting line for stifle-tarsal in FAR-push was 0.96, with P < .001, indicating that the general linear relationships between stifle and tarsal in FAR-push were significantly strong.
Summary data for the linear formula of stifle-tarsal in UA, UAR-push, UAR-pull, and FAR-push.
| Stifle-tarsal: z = ey + f | |||||
|---|---|---|---|---|---|
| e | f | R2adj | P value | n | |
| UA | 0.64 | 55 | 0.92 | < .001 | 163 |
| UAR-push | 0.61 | 59 | 0.94 | < .001 | 122 |
| UAR-pull | 0.49 | 74 | 0.87 | < .001 | 119 |
| FAR-push | 0.71 | 26 | 0.96 | < .001 | 181 |
e,f = Parameters for linear formula. R2adj = Adjusted R2 value. y = Stifle. z = Tarsal.
Likewise, a significantly strong linear relationship was also found between stifle and tarsal in UA and UAR-push/pull, with all adjusted R2 values > 0.85 and P values < .001 (Figure 4; Table 2).
Discussion
This research revealed the linear relationships for hip-stifle and hip-tarsal in UA (hip ROM, 85° to 175°), UAR (hip ROM, 65° to 180°), and FAR-push > 120° (hip ROM, 120° to 180°). Therefore, the hypothesis that the angles of the stifle and tarsal joints exhibit linear relationships with hip joint angle under certain circumstances was preliminarily confirmed. In addition, the occasional findings of the strong linear relationship for stifle-tarsal in UA (hip ROM, 85° to 175°), UAR (hip ROM, 65° to 180°), and FAR-push (hip ROM, 60° to 180°) as well as the narrowing of the FAR of stifle and tarsal joints when the hip joint angle exceeds 120° further revealed the relationships between hindlimb joint angles when the hip joint is passively moved within a wider ROM than AROM.
Within the limited studies on the relationship between canine hindlimb joints, most focus on the linkage between the stifle and tarsal joints. The gastrocnemius, a biarticular muscle that spans both the stifle and tarsal joints, is considered to link the changes in the stifle and tarsal joint angles.19,20 The significantly strong linear relationship between the stifle and tarsal joints, whether unforced (UA) or under pushing force (FAR-push), found in our study further supports the view that the angle of tarsal joint could be linear related with the angle of stifle joint under the linkage of a biarticular muscle even if the muscle tone is absent. Regarding the hip and stifle joints, there are 2 groups of biarticular muscles: the rectus femoris cranially and the hamstrings caudally. In FAR-push > 120°, the flexion angle of the stifle joint showed a good linear relationship with the hip joint angle. It is possibly because the rectus femoris has reached its maximum length, so the stifle joint has to extend along with the hip joint. The hamstrings consist of the semimembranosus, semitendinosus, and biceps femoris.27 Anatomically, when the hip is flexed, these muscles will pull on the tibia, causing the stifle joint to bend as well. In our study, the linear relationship between the angles of the hip and stifle joints is well maintained by the inactive, unforced muscles regardless of the initial position of the hindlimb (see UA and UAR). This may be related to the linking function of the rectus femoris and hamstrings. The relationship between the angles of the tarsal and hip joints may be more complex. The linear relationship between the stifle and tarsal joints mediated by gastrocnemius has already been discussed. Although biarticular muscles, such as the gracilis, semitendinosus, and biceps femoris, span the hip and tarsal joints, we suppose that the mediating effect of the stifle joint plays a more dominant role than these biarticular muscles as the linear relationship between the stifle and tarsal joints in UA, UAR, and FAR-push is more significant than that of the hip and tarsal joints. In other words, it is assumed that the passive movement of the hip joint leads to a change in the stifle joint angle, and this change is then transmitted to the tarsal joint. However, since our study design is inadequate to exclude the effects of the gracilis, semitendinosus, and biceps femoris, further study is required to validate the assumption above.
The FAR of the stifle and tarsal joints in this study is considered primarily influenced by the linear traction rather than contraction of related muscles, tendons, and ligaments since the muscle tone was absent because of the use of muscle relaxant. Additionally, the interaction between the stifle and tarsal joints may also exert some influence. In both the stifle and tarsal joints, a marked narrowing of the FAR, resulting from the sudden increase in the slope of FAR-push, was observed when the hip extended beyond 120°, indicating that hip joint angle significantly affects the FAR of the distal joints. The potential involvement of the rectus femoris and the gastrocnemius in linking the hindlimb joints has been previously discussed. In FAR-push < 120°, the flexion angles of the stifle and tarsal joints remained within a specific range, although their variation patterns with hip joint extension were inconsistent. In some individuals, relatively large stifle and tarsal joint flexion angles were observed when the hip joint was fully flexed. This phenomenon may be explained by the impediment of further stifle joint flexion due to the increased mass of the proximal gastrocnemius muscle when all 3 joints were flexed as demonstrated in by Sabanci and Ocal.28
To the best of our knowledge, this is the first study to describe the relationships between hindlimb joint angles under passive movement of the hip joint within a wider range than AROM in Beagles under general anesthesia and muscle relaxation. These findings may contribute to a better understanding of orthopedic and rehabilitation practices. Furthermore, we believe that confirming the existence of linear relationships between hindlimb joints angles could be the first step for the design of a simplified canine exoskeleton.
Passive ROM (PROM) plays an important role in rehabilitation practice, and the PROM test is widely applied in orthopedic examination. In this study, the joint angles were determined based on the anatomical landmarks corresponding to the motion analysis experiment in the same series rather than the bone longitudinal axis commonly used in PROM measurements.29 Although the discrepancies in their values are inevitable, FAR and PROM still share some common patterns of variation. The findings about FAR in our study suggest that the PROM of the stifle and tarsal joints can be affected by the position of proximal joints.
Since PROM is defined by the maximal flexion and extension angles of a particular joint without consideration of the proximal joint’s position,28–34 misjudgment of the PROM of a distal joint may occur if the PROM of the proximal joint is restricted as seen in conditions such as osteoarthritis and stifle arthrodesis.
One of the core functions of the simplified canine exoskeleton is to maintain the stable extension of the stifle and tarsal joints while bearing load in the stance phase. The linear relationships of hip-stifle and hip-tarsal in FAR-push > 120° observed in our study demonstrated the hindlimb’s capability for load bearing when the hip joint is extended beyond 120°. However, according to the results of our motion analysis experiment in the same series—conducted on the same Beagles using the same landmarks and joint angle definitions—the hip joint angle at the beginning of the stance phase in both walking and trotting gaits was observed to be approximately 110°. This finding is also supported by other studies35 after calibration of joint angles. Therefore, additional extension of the stifle joint is required during the initial stance phase to ensure tarsal joint extension as the linear relationship of stifle-tarsal in FAR-push remained consistently strong. Moreover, the FAR of the stifle and tarsal joints provides a valuable reference for determining the safe passive ROM in the design of exoskeletons. Care should be taken to prevent iatrogenic injuries resulting from excessive flexion of the stifle and tarsal joints, particularly when the hip joint angle exceeds 120°.
The limitations of this study are mainly in 4 aspects. Firstly, the sample size in this study was relatively small due to the limited availability of Beagles. The current sample size is insufficient to derive a universal formula applicable to the Beagle breed. Thus, the general linear formula presented in this study can only be used to demonstrate the existence of the linear relationships under typical circumstances. Further research with a larger sample size would be recommended to develop a universal linear formula applicable to a specific breed or even all dogs. Alternatively, if the correlation between the formula parameters and specific anatomical parameters of each individual can be established, an individual-specific linear formula could potentially be derived from the corresponding anatomical data. Secondly, the Lycra pants may potentially influence the results. The original intention of using the Lycra pants was to verify the accuracy of the markers attached to the pants, which indicated the position of anatomical landmarks in the motion analysis experiments in the same series. Lycra pants are elastic and are considered to have a minimal impact on the dogs’ gait in the motion analysis experiment.36 Additionally, in our experiment, only a small amount of fabric was used to wrap around the thigh, with an initial adjustment of the tightness to minimize its impact on the movement of the hindlimb. Even so, since we have not conducted comparative experiments, the influence of Lycra pants on the results cannot be ruled out. Thirdly, the effect of gravity was not considered in the experimental design. Take the UA and UAR experiments as an example, the aim of which being to investigate the neutral interrelationship between the hindlimb joints angles without the effect of external forces, including gravity. However, considering the practical application of the exoskeleton where patients are in a standing posture, the effect of gravity should not be neglected, especially in the swing phase. Thus, further experiments on non–load-bearing hindlimbs in a vertically placed dog would be required to investigate the relationships between hindlimb joint angles under the effect of gravity alone. Finally, we have not recorded the magnitude of each push and pull force applied in the UAR and FAR experiments. Otherwise, further evaluation of the muscle and tendon elasticity under different combinations of hindlimb joint angles could have been achieved.
Acknowledgments
The authors thank Tamaki Watanabe, Masahiro Ishihara, Chihiro Nakada, and Seiya Suzuki for their assistance in the experimental stage. They also thank the cooperation of the experimental Beagles participating in this study.
Disclosures
Dr. Murakami is a member of the AJVR Scientific Review Board, but was not involved in the editorial evaluation of or decision to accept this article for publication.
An AI tool (GPT-4o-mini) was used to check the grammar and vocabulary of this manuscript.
Funding
This work was supported by JSPS KAKENHI grant No. 23H05445.
ORCID
Sawako Murakami https://orcid.org/0000-0003-2558-595X
References
- 1.↑
Moore SA, Early PJ, Hettlich BF. Practice patterns in the management of acute intervertebral disc herniation in dogs. J Small Anim Pract. 2016;57(8):409–415. doi:10.1111/jsap.12496
- 2.↑
Rossi G, Stachel A, Lynch AM, Olby NJ. Intervertebral disc disease and aortic thromboembolism are the most common causes of acute paralysis in dogs and cats presenting to an emergency clinic. Vet Rec. 2020;187(10):e81. doi:10.1136/vr.105844
- 3.↑
Levine JM, Budke CM, Levine GJ, Kerwin SC, Hettlich BF, Slater MR. Owner-perceived, weighted quality-of-life assessments in dogs with spinal cord injuries. J Am Vet Med Assoc. 2008;233(6):931–935. doi:10.2460/javma.233.6.931
- 4.↑
Freeman PM, Holmes MA, Jeffery ND, Granger N. Time requirement and effect on owners of home-based management of dogs with severe chronic spinal cord injury. J Vet Behav Clin Appl Res. 2013;8(6):439–443. doi:10.1016/j.jveb.2013.06.001
- 5.↑
Olby N, Halling KB, Glick TR. Rehabilitation for the neurologic patient. Vet Clin North Am Small Anim Pract. 2005;35(6):1389–1409. doi:10.1016/j.cvsm.2005.08.004
- 6.↑
Gordon-Evans WJ, Johnson AL, Knap KE, Griffon DJ. The effect of body condition on postoperative recovery of Dachshunds with intervertebral disc disease treated with postoperative physical rehabilitation. Vet Surg. 2019;48(2):159–163. doi:10.1111/vsu.13142
- 7.↑
Gallucci A, Dragone L, Menchetti M, et al. Acquisition of involuntary spinal locomotion (spinal walking) in dogs with irreversible thoracolumbar spinal cord lesion: 81 dogs. J Vet Intern Med. 2017;31(2):492–497. doi:10.1111/jvim.14651
- 8.
Lovely RG, Gregor RJ, Roy RR, Edgerton VR. Effects of training on the recovery of full-weight-bearing stepping in the adult spinal cat. Exp Neurol. 1986;9(2):421–435. doi:10.1016/0014-4886(86)90094-4
- 9.
De Leon RD, Hodgson JA, Roy RR, Edgerton VR. Full weight-bearing hindlimb standing following stand training in the adult spinal cat. J Neurophysiol. 1998;80(1):83–91. doi:10.1152/jn.1998.80.1.83
- 10.
Boyce VS, Tumolo M, Fischer I, et al. Neurotrophic factors promote and enhance locomotor recovery in untrained spinalized cats. J Neurophysiol. 2007;98(4):1988–1996. doi:10.1152/jn.00391.2007
- 11.↑
Battistuzzo CR, Callister RJ, Callister R, Galea MP. A systematic review of exercise training to promote locomotor recovery in animal models of spinal cord injury. J Neurotrauma. 2012;29(8):1600–1613. doi:10.1089/neu.2011.2199
- 12.↑
Martins A, Gouveia D, Cardoso A, Viegas I, Gamboa Ó, Ferreira A. A comparison between body weight-supported treadmill training and conventional over-ground training in dogs with incomplete spinal cord injury. Front Vet Sci. 2021;8:597949. doi:10.3389/fvets.2021.597949
- 13.↑
Canine exoskeleton for rehabilitation. Colorado State University Class of 2023. Accessed June 24, 2024. https://projects-web.engr.colostate.edu/ece-sr-design/AY22/canine/canine-exoskeleton-2022-2023-ea2fa1.webflow.io/index.html
- 14.↑
Taghavi N, Luecke GR, Jeffery ND. A wearable body controlling device for application of functional electrical stimulation. Sensors (Basel). 2018;18(4):1251. doi:10.3390/s18041251
- 15.↑
Mitchell B, Bressel E, McNair PJ, Bressel ME. Effect of pelvic, hip, and knee position on ankle joint range of motion. Phys Ther Sport. 2008;9(4):202–208. doi:10.1016/j.ptsp.2008.08.002
- 16.
Andrade RJ, Lacourpaille L, Freitas SR, McNair PJ, Nordez A. Effects of hip and head position on ankle range of motion, ankle passive torque, and passive gastrocnemius tension. Scand J Med Sci Sports. 2016;26(1):41–47. doi:10.1111/sms.12406
- 17.↑
Cho KH, Jeon Y, Lee H. Range of motion of the ankle according to pushing force, gender and knee position. Ann Rehabil Med. 2016;40(2):271–278. doi:10.5535/arm.2016.40.2.271
- 18.↑
Palermo AE, Maher JL, Baunsgaard CB, Nash MS. Clinician-focused overview of bionic exoskeleton use after spinal cord injury. Top Spinal Cord Inj Rehabil. 2017;23(3):234–244. doi:10.1310/sci2303-234
- 19.↑
Fischer MS, Lilje KE. Motion sequence. In: Dogs in Motion. 2nd ed. VDH Service GmbH.; 2014:88–97.
- 20.↑
Sugiyama T, Woodward A, Ryan SD. Effects of femorotibial and tibiotarsal joint ranges of motion on the gastrocnemius muscle-tendon working length in the canine cadaveric limb. Vet Comp Orthop Traumatol. 2021;34(6):411–418. doi:10.1055/s-0041-1735315
- 21.↑
Hyytiäinen HK, Mölsä SH, Junnila JT, Laitinen-Vapaavuori OM, Hielm-Björkman AK. Ranking of physiotherapeutic evaluation methods as outcome measures of stifle functionality in dogs. Acta Vet Scand. 2013;55(1):29. doi:10.1186/1751-0147-55-29
- 22.↑
Yoshikawa K, Kitazawa T, Sano T, Ino T, Miyasaka T. Kinematic characteristics of canine hindlimb movement during sit-to-stand and stand-to-sit motions. Res Vet Sci. 2023;162:104944. doi:10.1016/j.rvsc.2023.104944
- 23.↑
Lorke M, Willen M, Lucas K, et al. Comparative kinematic gait analysis in young and old Beagle dogs. J Vet Sci. 2017;18(4):521–530. doi:10.4142/jvs.2017.18.4.521
- 24.↑
Ward E, German A, Churchill J. The global pet obesity initiative position statement. June 3, 2019. Accessed April 1, 2025. https://static1.squarespace.com/static/6425ec5d33eaaa634113b2d4/t/653e5870610dff50a2fa1013/1698584719824/Global+pet+obesity+initiative+position+statement_2019.pdf
- 25.↑
XiaoSaiViewer. MacUpdater. Version 3.2.1. Accessed December 1, 2023. https://macupdater.net/app_updates/appinfo/com.xiaosaiviewer.xiaosai/index.html
- 27.↑
Wentink GH. The action of the hind limb musculature of the dog in walking. Acta Anat (Basel). 1976;96(1):70–80. doi:10.1159/000144662
- 28.↑
Sabanci SS, Ocal MK. Comparison of goniometric measurements of the stifle joint in seven breeds of normal dogs. Vet CompOrthop Traumatol. 2016;29(3):214–219. doi:10.3415/VCOT-15-05-0090
- 29.↑
Petazzoni M, Jaeger GH. Clinical goniometry. In: Petazzoni M, Jaeger GH, eds. Atlas of Clinical Goniometry and Radiographic Measurements of the Canine Pelvic Limb. 2nd ed. Merial SAS; 2008:22–31.
- 30.
Reusing M, Brocardo M, Weber S, Villanova J. Goniometric evaluation and passive range of joint motion in chondrodystrophic and non-chondrodystrophic dogs of different sizes. VCOT Open. 2020;3(2):e66–e71. doi:10.1055/s-0040-1713825
- 31.
Formenton MR, de Lima LG, Vassalo FG, Joaquim JGF, Rosseto LP, Fantoni DT. Goniometric assessment in French Bulldogs. Front Vet Sci. 2019;6:424. doi:10.3389/fvets.2019.00424
- 32.
Hady LL, Fosgate GT, Weh M. Comparison of range of motion in Labrador Retrievers and Border Collies. J Vet Med Anim Health. 2015;7(4):122–127. doi:10.5897/JVMAH2014.0298
- 33.
Jaegger G, Marcellin-Little DJ, Levine D. Reliability of goniometry in Labrador Retrievers. Am J Vet Res. 2002;63(7):979–986. doi:10.2460/ajvr.2002.63.979
- 34.↑
Thomas TM, Marcellin-Little DJ, Roe SC, Lascelles BD, Brosey BP. Comparison of measurements obtained by use of an electrogoniometer and a universal plastic goniometer for the assessment of joint motion in dogs. Am J Vet Res. 2006;67(12):1974–1979. doi:10.2460/ajvr.67.12.1974
- 35.↑
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(1):16982. doi:10.1038/s41598-018-34310-0
- 36.↑
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(1):112–119. doi:10.1111/jvim.12012
