Accepted Manuscript Title: Kinematic adaptations to tripedal locomotion in dogs Author: B. Goldner, A. Fuchs, I. Nolte, N. Schilling PII: DOI: Reference:
S1090-0233(15)00100-8 http://dx.doi.org/doi:10.1016/j.tvjl.2015.03.003 YTVJL 4441
To appear in:
The Veterinary Journal
Accepted date:
2-3-2015
Please cite this article as: B. Goldner, A. Fuchs, I. Nolte, N. Schilling, Kinematic adaptations to tripedal locomotion in dogs, The Veterinary Journal (2015), http://dx.doi.org/doi:10.1016/j.tvjl.2015.03.003. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
1
Kinematic adaptations to tripedal locomotion in dogs
2 B. Goldner a, A. Fuchs a, I. Nolte a, N. Schilling b*
3 4 5
a
6 7 8
University of Veterinary Medicine Hannover, Foundation, Small Animal Clinic, Hannover, Germany
b
Friedrich-Schiller-University, Institute of Systematic Zoology and Evolutionary Biology, Jena, Germany
9 10 11
* Corresponding author. Tel.: +49 175 5257195
12
E-mail address: [email protected] (N. Schilling)
13
Highlights
14
-
Significant kinematic differences affect both stance and swing phases
15
-
The limb the most affected is the remaining pelvic limb
16
-
Kinematic compensation of pelvic limb loss results in changes in all limb joints
17
-
Greater retroversion of the thoracic limbs facilitates pelvic limb unloading
18
-
Proximal limb segments show greater changes than distal ones
19 20 21 22
1
Page 1 of 21
23
Abstract
24
Limb amputation often represents the only treatment option for canine patients with certain
25
diseases or injuries of the appendicular system. Previous studies have investigated adaptations
26
to tripedal locomotion in dogs but there is a lack of understanding of biomechanical
27
compensatory mechanisms. This study evaluated the kinematic differences between
28
quadrupedal and tripedal locomotion in nine healthy dogs running on a treadmill. The loss of
29
the right pelvic limb was simulated using an Ehmer sling. Kinematic gait analysis included
30
spatio-temporal comparisons of limb, joint and segment angles of the remaining pelvic and
31
both thoracic limbs. The following key parameters were compared between quadrupedal and
32
tripedal conditions: angles at touch-down and lift-off, minimum and maximum joint angles,
33
plus range of motion.
34 35
Significant differences in angular excursion were identified in several joints of each limb
36
during both stance and swing phase. The most pronounced differences concerned the
37
remaining pelvic limb, followed by the contralateral thoracic limb and, to a lesser degree, the
38
ipsilateral thoracic limb. The thoracic limbs were, in general, more retracted, consistent with
39
previous observations of bodyweight re-distribution in amputees. Proximal limb segments
40
showed more distinct changes than distal ones. Particularly, the persistently greater
41
anteversion of the pelvis probably affects the axial system. Overall, tripedal locomotion
42
requires concerted kinematic adjustments of both the appendicular and the axial systems, and
43
consequently preventive, therapeutic and rehabilitative care of canine amputees should
44
involve the whole musculoskeletal apparatus.
45 46 47
Keywords: Hindlimb amputation; Kinematics; Angular excursion; Tripod 2
Page 2 of 21
48
Introduction
49
Limb amputation constitutes one care option for canine patients with certain traumatic,
50
neoplastic, neurological, congenital or chronic infectious diseases of the appendicular system.
51
Although surveys indicate owner satisfaction with the patient’s outcome (Withrow and
52
Hirsch, 1979; Carberry and Harvey, 1987; Kirpensteijn et al., 1999), attitudes to amputation
53
are still mixed among both owners and veterinary practitioners. A better understanding of the
54
biomechanical consequences of moving on three legs as a quadruped may facilitate informed
55
patient-oriented decision-making and improve the post-surgical care of these dogs. For
56
example, the level of expectation of joint disease because of substantial changes in range of
57
motion (ROM) might be aided by knowledge of specific kinematic adaptations to tripedal
58
locomotion. Similarly, identifying compensatory strategies to tripedalism may help to refine
59
rehabilitative approaches and develop new after-care concepts for amputees.
60 61
Few studies have investigated tripedal locomotion in dogs and identified kinetic and
62
kinematic differences between amputee and control animals (Kirpensteijn et al., 2000; Hogy
63
et al., 2013; Jarvis et al., 2013; Fuchs et al., 2014). Pelvic limb amputees, for example,
64
support more bodyweight with their thoracic limbs (Kirpensteijn et al., 2000). Additionally,
65
stride and stance durations decrease while relative stance duration increases when dogs
66
ambulate tripedally (Fuchs et al., 2014). Changes in limb joint angular excursions are less
67
well documented; it remains uncertain whether all joints differ in their motion patterns or if
68
some joints show more pronounced alterations than others, although there is evidence that in
69
the pelvic limb changes may be more pronounced in distal joints, at least during stance phase
70
(Hogy et al., 2013). However it is unknown whether kinematic alterations also occur during
71
the swing phase. 3
Page 3 of 21
72 73
To better understand how dogs cope with ambulating on three legs and identify gait
74
differences between quadrupedal and tripedal locomotion throughout the step cycle, we used
75
kinematic gait analysis and studied segment, joint and overall limb excursions before and
76
after the amputation of a pelvic limb was simulated.
77
78
Materials and methods
79
Animals
80
Three female and six male Beagles owned by the Small Animal Clinic (University of
81
Veterinary Medicine, Hannover) were enrolled in the study. The dogs had a body mass of
82
14.9 ± 0.9 kg (mean ± standard deviation, SD) and were 4.6 ± 1.2 years old. Inclusion criteria
83
were absence of orthopaedic abnormalities and lameness verified by clinical examination
84
(Fuchs et al., 2014).
85 86
All experiments were carried out in accordance with the German Animal Welfare
87
guidelines and approved by the Ethical Committee of the State of Lower Saxony, Germany
88
(12/0717; 4 December 2012).
89 90 91
Data collection The experimental design is described in detail in our companion study (Fuchs et al., 2014).
92
Briefly, dogs were trained to run on a treadmill using both tripedal and quadrupedal gait
93
patterns. To simulate the amputation of the right pelvic limb, an Ehmer sling was applied.
94
Data were collected at the speed at which each dog showed a smooth, regular gait pattern and
95
effortlessly matched the treadmill speed, particularly tripedally. Quadrupedally, the dogs 4
Page 4 of 21
96
exhibited a walking trot (Hildebrand, 1966) at this preferred speed. Unrestrained, steady state
97
locomotion was monitored using a leash customised with a force transducer (KMM30-
98
200NZ/D Inelta Sensor Systems; signal amplifier LAZ-DMS, 24 V, 1-10V). Locomotor speed
99
was the same during quadrupedal and tripedal trials for each individual dog and ranged
100
between 1.3 and 1.5 m/s for all dogs.
101 102
Kinematic data were collected using a horizontal four-belt treadmill instrumented with
103
force plates underneath each belt (Modell 4060-08, Bertec Corporation; force threshold 13 N,
104
sampling rate 1000 Hz). Retro-reflective markers (10 mm diameter) were placed adjacent to
105
defined anatomical landmarks using adhesive tape (Fig. 1A). Six infrared high-speed cameras
106
tracked the three-dimensional motions of the markers (MX3+, Vicon Motion Systems;
107
recording frequency 100 Hz). Prior to each recording session, the cameras were calibrated
108
using an L-shaped calibration device (Vicon Motion Systems). Additionally, a high-speed
109
video camera recorded the dogs from the lateral perspective (Basler Pilot, PiA 640-210gc).
110
Signals of the high-speed and infrared cameras as well as the leash force transducer and the
111
force plates were recorded synchronously in Vicon Nexus.
112 113 114
Data analysis Trials with 10 consecutive valid strides were selected for analysis. A trial was valid if the
115
leash force was below 0.2 N and single limb ground reaction forces were recorded (i.e. no
116
overstepping). The vertical force traces were used to define touch-down and lift-off of each
117
limb (Fuchs et al., 2014). Using customised kinematic models, the tracked markers were
118
labelled in Vicon Nexus and sagittal limb, segment and joint angles determined (Figs. 1B-D).
119
Before exporting the data to Microsoft Excel 2003 for further analysis, they were time-
120
normalised to the same stance and swing phase durations to facilitate comparison of the 5
Page 5 of 21
121
movements with reference to footfall events. As a result of the time-normalisation, each phase
122
covered 50% of the stride cycle.
123 124
To distinguish limb from segment movements, we used the terms protraction and retraction
125
for a limb and anteversion and retroversion for a segment, respectively, to refer to cranial and
126
caudal rotation. The following kinematic values were evaluated: segment and joint angles at
127
touch-down and lift-off as well as minimum and maximum values and range of motion
128
(ROM) during the stance and the swing phases (Appendix: Supplementary Tables 1 and 2).
129
Thereby, ROM is the difference between maximum and minimum values observed during a
130
given stride phase. Additionally, limb excursion was assessed by limb angle at touch-down
131
and at lift-off. Limb angle at mid-stance was determined as the angle between the vertical and
132
the line connecting the limb’s fulcrum and the most distal marker at mid-stance (Fig. 1D).
133
Positive values indicated that the limb was protracted; negative values indicated that the limb
134
was retracted.
135 136
Statistical analysis
137
Because of the small sample size (n=9), non-parametric Wilcoxon-signed rank tests for
138
paired observations using the comparison-wise error rate were applied to detect kinematic
139
differences between quadrupedal and tripedal locomotion (significant at P
S1090-0233(15)00100-8 http://dx.doi.org/doi:10.1016/j.tvjl.2015.03.003 YTVJL 4441
To appear in:
The Veterinary Journal
Accepted date:
2-3-2015
Please cite this article as: B. Goldner, A. Fuchs, I. Nolte, N. Schilling, Kinematic adaptations to tripedal locomotion in dogs, The Veterinary Journal (2015), http://dx.doi.org/doi:10.1016/j.tvjl.2015.03.003. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
1
Kinematic adaptations to tripedal locomotion in dogs
2 B. Goldner a, A. Fuchs a, I. Nolte a, N. Schilling b*
3 4 5
a
6 7 8
University of Veterinary Medicine Hannover, Foundation, Small Animal Clinic, Hannover, Germany
b
Friedrich-Schiller-University, Institute of Systematic Zoology and Evolutionary Biology, Jena, Germany
9 10 11
* Corresponding author. Tel.: +49 175 5257195
12
E-mail address: [email protected] (N. Schilling)
13
Highlights
14
-
Significant kinematic differences affect both stance and swing phases
15
-
The limb the most affected is the remaining pelvic limb
16
-
Kinematic compensation of pelvic limb loss results in changes in all limb joints
17
-
Greater retroversion of the thoracic limbs facilitates pelvic limb unloading
18
-
Proximal limb segments show greater changes than distal ones
19 20 21 22
1
Page 1 of 21
23
Abstract
24
Limb amputation often represents the only treatment option for canine patients with certain
25
diseases or injuries of the appendicular system. Previous studies have investigated adaptations
26
to tripedal locomotion in dogs but there is a lack of understanding of biomechanical
27
compensatory mechanisms. This study evaluated the kinematic differences between
28
quadrupedal and tripedal locomotion in nine healthy dogs running on a treadmill. The loss of
29
the right pelvic limb was simulated using an Ehmer sling. Kinematic gait analysis included
30
spatio-temporal comparisons of limb, joint and segment angles of the remaining pelvic and
31
both thoracic limbs. The following key parameters were compared between quadrupedal and
32
tripedal conditions: angles at touch-down and lift-off, minimum and maximum joint angles,
33
plus range of motion.
34 35
Significant differences in angular excursion were identified in several joints of each limb
36
during both stance and swing phase. The most pronounced differences concerned the
37
remaining pelvic limb, followed by the contralateral thoracic limb and, to a lesser degree, the
38
ipsilateral thoracic limb. The thoracic limbs were, in general, more retracted, consistent with
39
previous observations of bodyweight re-distribution in amputees. Proximal limb segments
40
showed more distinct changes than distal ones. Particularly, the persistently greater
41
anteversion of the pelvis probably affects the axial system. Overall, tripedal locomotion
42
requires concerted kinematic adjustments of both the appendicular and the axial systems, and
43
consequently preventive, therapeutic and rehabilitative care of canine amputees should
44
involve the whole musculoskeletal apparatus.
45 46 47
Keywords: Hindlimb amputation; Kinematics; Angular excursion; Tripod 2
Page 2 of 21
48
Introduction
49
Limb amputation constitutes one care option for canine patients with certain traumatic,
50
neoplastic, neurological, congenital or chronic infectious diseases of the appendicular system.
51
Although surveys indicate owner satisfaction with the patient’s outcome (Withrow and
52
Hirsch, 1979; Carberry and Harvey, 1987; Kirpensteijn et al., 1999), attitudes to amputation
53
are still mixed among both owners and veterinary practitioners. A better understanding of the
54
biomechanical consequences of moving on three legs as a quadruped may facilitate informed
55
patient-oriented decision-making and improve the post-surgical care of these dogs. For
56
example, the level of expectation of joint disease because of substantial changes in range of
57
motion (ROM) might be aided by knowledge of specific kinematic adaptations to tripedal
58
locomotion. Similarly, identifying compensatory strategies to tripedalism may help to refine
59
rehabilitative approaches and develop new after-care concepts for amputees.
60 61
Few studies have investigated tripedal locomotion in dogs and identified kinetic and
62
kinematic differences between amputee and control animals (Kirpensteijn et al., 2000; Hogy
63
et al., 2013; Jarvis et al., 2013; Fuchs et al., 2014). Pelvic limb amputees, for example,
64
support more bodyweight with their thoracic limbs (Kirpensteijn et al., 2000). Additionally,
65
stride and stance durations decrease while relative stance duration increases when dogs
66
ambulate tripedally (Fuchs et al., 2014). Changes in limb joint angular excursions are less
67
well documented; it remains uncertain whether all joints differ in their motion patterns or if
68
some joints show more pronounced alterations than others, although there is evidence that in
69
the pelvic limb changes may be more pronounced in distal joints, at least during stance phase
70
(Hogy et al., 2013). However it is unknown whether kinematic alterations also occur during
71
the swing phase. 3
Page 3 of 21
72 73
To better understand how dogs cope with ambulating on three legs and identify gait
74
differences between quadrupedal and tripedal locomotion throughout the step cycle, we used
75
kinematic gait analysis and studied segment, joint and overall limb excursions before and
76
after the amputation of a pelvic limb was simulated.
77
78
Materials and methods
79
Animals
80
Three female and six male Beagles owned by the Small Animal Clinic (University of
81
Veterinary Medicine, Hannover) were enrolled in the study. The dogs had a body mass of
82
14.9 ± 0.9 kg (mean ± standard deviation, SD) and were 4.6 ± 1.2 years old. Inclusion criteria
83
were absence of orthopaedic abnormalities and lameness verified by clinical examination
84
(Fuchs et al., 2014).
85 86
All experiments were carried out in accordance with the German Animal Welfare
87
guidelines and approved by the Ethical Committee of the State of Lower Saxony, Germany
88
(12/0717; 4 December 2012).
89 90 91
Data collection The experimental design is described in detail in our companion study (Fuchs et al., 2014).
92
Briefly, dogs were trained to run on a treadmill using both tripedal and quadrupedal gait
93
patterns. To simulate the amputation of the right pelvic limb, an Ehmer sling was applied.
94
Data were collected at the speed at which each dog showed a smooth, regular gait pattern and
95
effortlessly matched the treadmill speed, particularly tripedally. Quadrupedally, the dogs 4
Page 4 of 21
96
exhibited a walking trot (Hildebrand, 1966) at this preferred speed. Unrestrained, steady state
97
locomotion was monitored using a leash customised with a force transducer (KMM30-
98
200NZ/D Inelta Sensor Systems; signal amplifier LAZ-DMS, 24 V, 1-10V). Locomotor speed
99
was the same during quadrupedal and tripedal trials for each individual dog and ranged
100
between 1.3 and 1.5 m/s for all dogs.
101 102
Kinematic data were collected using a horizontal four-belt treadmill instrumented with
103
force plates underneath each belt (Modell 4060-08, Bertec Corporation; force threshold 13 N,
104
sampling rate 1000 Hz). Retro-reflective markers (10 mm diameter) were placed adjacent to
105
defined anatomical landmarks using adhesive tape (Fig. 1A). Six infrared high-speed cameras
106
tracked the three-dimensional motions of the markers (MX3+, Vicon Motion Systems;
107
recording frequency 100 Hz). Prior to each recording session, the cameras were calibrated
108
using an L-shaped calibration device (Vicon Motion Systems). Additionally, a high-speed
109
video camera recorded the dogs from the lateral perspective (Basler Pilot, PiA 640-210gc).
110
Signals of the high-speed and infrared cameras as well as the leash force transducer and the
111
force plates were recorded synchronously in Vicon Nexus.
112 113 114
Data analysis Trials with 10 consecutive valid strides were selected for analysis. A trial was valid if the
115
leash force was below 0.2 N and single limb ground reaction forces were recorded (i.e. no
116
overstepping). The vertical force traces were used to define touch-down and lift-off of each
117
limb (Fuchs et al., 2014). Using customised kinematic models, the tracked markers were
118
labelled in Vicon Nexus and sagittal limb, segment and joint angles determined (Figs. 1B-D).
119
Before exporting the data to Microsoft Excel 2003 for further analysis, they were time-
120
normalised to the same stance and swing phase durations to facilitate comparison of the 5
Page 5 of 21
121
movements with reference to footfall events. As a result of the time-normalisation, each phase
122
covered 50% of the stride cycle.
123 124
To distinguish limb from segment movements, we used the terms protraction and retraction
125
for a limb and anteversion and retroversion for a segment, respectively, to refer to cranial and
126
caudal rotation. The following kinematic values were evaluated: segment and joint angles at
127
touch-down and lift-off as well as minimum and maximum values and range of motion
128
(ROM) during the stance and the swing phases (Appendix: Supplementary Tables 1 and 2).
129
Thereby, ROM is the difference between maximum and minimum values observed during a
130
given stride phase. Additionally, limb excursion was assessed by limb angle at touch-down
131
and at lift-off. Limb angle at mid-stance was determined as the angle between the vertical and
132
the line connecting the limb’s fulcrum and the most distal marker at mid-stance (Fig. 1D).
133
Positive values indicated that the limb was protracted; negative values indicated that the limb
134
was retracted.
135 136
Statistical analysis
137
Because of the small sample size (n=9), non-parametric Wilcoxon-signed rank tests for
138
paired observations using the comparison-wise error rate were applied to detect kinematic
139
differences between quadrupedal and tripedal locomotion (significant at P
