Deficiency of the CrCL is the leading cause of lameness and one of the most common orthopedic conditions diagnosed in stifle joints of dogs.1,2 A prevalence across all breeds of 2.55% has been reported on the basis of 31,698 CrCL injuries in 1,243,681 dogs examined between 1964 and 2003.3 Injuries to the CrCL lead to stifle joint instability, inflammation, signs of pain, and osteoarthritis.4–7 Surgical and medical treatment of CrCL deficiency in the United States in 2003 was estimated to cost > $1 billion.8 Surgical intervention is often used to stabilize the CrCL-deficient stifle joint, but on the basis of long-term data, no surgical procedure can successfully prevent the progression of osteoarthritis or can be claimed to be superior over other procedures.4 Although in small dogs CrCL deficiency has been managed successfully without surgical intervention, large-breed dogs are less likely to regain normal limb function and therefore usually require surgical intervention.9,10 However, surgical stabilization techniques used for large-breed dogs may be more technically demanding and invasive than those used for small dogs.9,11 Complication rates for tibial plateau leveling osteotomy (19% to 31%) and tibial tuberosity advancement (32% to 59%) are substantial.12 Associated reoperation rates range from 5% to 9% for tibial plateau leveling osteotomy and 11% for tibial tuberosity advancement.12 The wide variety of stifle joint stabilization techniques suggests the need for an improved understanding of the factors that contribute to CrCL deficiency, allowing for enhanced evidence-based approaches to surgical interventions.13 No single factor has been conclusively linked to CrCL deficiency.1,7 Rather, CrCL deficiency is thought to result from progressive degradation that is likely influenced by several risk factors, including age, breed, sex, reproductive status, TPA, patellar ligament angle, body weight, obesity, orthopedic comorbidities, gait dynamics, joint conformation and alignment, muscle dysfunction, degenerative synovial enzymes, and impaired cellular synthesis.3,6,7,14–16
The use of computer models permits simulation of controlled scenarios aimed at isolating key biomechanical factors potentially associated with CrCL deficiency.17 A previous mathematical model of the canine pelvic limb during stance was used to investigate the biomechanics associated with alteration of the tibial plateau slope.18 Use of that model predicted increased CaCL load in the CrCL-deficient stifle joint, which further increased when TPA was decreased (as would occur with a tibial plateau leveling osteotomy).18 Increased CaCL loads occurring after increased tibial plateau slope rotation were also reported in an in vitro study.19 To our knowledge, in silico studies (ie, performed via computer simulation) investigating the effects of variation of morphological parameters on ligament loads, tibial translation, and tibial rotation in the canine stifle joint have not been published. The purpose of the study reported here was to investigate the influence of key morphological parameters on stifle joint ligament loads, cranial tibial translation, and internal tibial rotation by use of a 3-D canine pelvic limb computer simulation model.20 We hypothesized that varying stifle joint morphological parameters via computer simulation would influence ligament loads, tibial translation, and tibial rotation in an intact or CrCL-deficient stifle joint.
Materials and Methods
Model design—A 3-D canine pelvic limb computer model of the stance phase of walking gait was developed and described in detail.20 Computer-aided design and modeling softwarea was used to represent the pelvic limb of a 5-year-old neutered male Golden Retriever (weight, 33 kg) during walking gait. Stifle joint ligaments were represented as tension-only force elements, and pelvic limb muscles were represented as force vectors.20 Medial and lateral menisci were represented in the model with contours established from CT imaging. Meniscal articulation with the femoral condyles was characterized with a contact function, which defined frictional properties and disallowed penetration of the condyles into the meniscal surface.20 For the baseline model, 3-D motion-capture kinematics were measured by use of reflective markers placed on bony landmarks of the pelvis and left pelvic limb.20 Ground reaction forces and moments were measured with a force platform placed level with the ground.20 A representative walking trial in which only the left pelvic limb was in contact with the force platform during the stance phase of gait was used for model input.20 Ground reaction forces were used to conduct an inverse dynamics analysis of the left pelvic limb to determine reaction joint forces and moments.20 Finally, optimized pelvic limb muscle forces were determined from the stifle joint reaction moments by use of the minimization of maximal muscle stress strategy.20 Motion-capture kinematics, ground reaction forces, and optimized muscle forces were then applied as input to the baseline canine pelvic limb model. These same model inputs were used for simulations with altered morphological parameters.
Model simulation and parametric analysis—Computer model simulation of the canine pelvic limb across the stance phase of gait was developed with rigid body motion software.b Model outcome measures included stifle joint ligament loads in the CrCL, CaCL, MCL, and LCL and tibial translation and tibial rotation. Outcome measures were determined for the baseline model across the stance phase at discrete 10% intervals through model simulation for the intact and CrCL-deficient stifle joint.20 Individual parameters were altered from the baseline value and then maintained across the entire stance phase.
This process was repeated for each parameter. Relative tibial translation was defined as follows:
where FT is the fixed point tibial translation (the distance between the tibial tuberosity position relative to a fixed point on the femur along the craniocaudal axis), with deficient and intact denoting the CrCL status. Relative tibial rotation was defined as follows:
where R is the internal-external rotation, as defined.21
Stifle joint parameters—Stifle joint parameters that were altered included PLIL, TPA, and FCD. Patellar ligament insertion location was defined as the point located along the tibial segment craniocaudal axis at a specified distance from the tibial tuberosity (Figure 1). The tibial craniocaudal axis was defined on the basis of the cross product between a vector containing the medial and lateral malleoli and a vector containing the proximal and distal points of the tibial crest.22 Fourteen simulations were conducted to evaluate the influence of PLIL on outcomes: baseline (unchanged PLIL) and PLIL changed by a distance of −6, −3, 3, 6, 9, and 12 mm from the baseline position in an intact and CrCL-deficient stifle joint (negative values were associated with the caudal direction, and positive values were associated with the cranial direction).
Illustration of PLIL change of +12 mm in a canine pelvic limb computer simulation model. The PLIL was defined as the point located along the tibial segment craniocaudal axis (arrow) at a specified distance from the tibial tuberosity (TT). The complex surface geometry is represented with surface contour elements.
Citation: American Journal of Veterinary Research 75, 1; 10.2460/ajvr.75.1.26
Illustration of PLIL change of +12 mm in a canine pelvic limb computer simulation model. The PLIL was defined as the point located along the tibial segment craniocaudal axis (arrow) at a specified distance from the tibial tuberosity (TT). The complex surface geometry is represented with surface contour elements.
Citation: American Journal of Veterinary Research 75, 1; 10.2460/ajvr.75.1.26
Illustration of PLIL change of +12 mm in a canine pelvic limb computer simulation model. The PLIL was defined as the point located along the tibial segment craniocaudal axis (arrow) at a specified distance from the tibial tuberosity (TT). The complex surface geometry is represented with surface contour elements.
Citation: American Journal of Veterinary Research 75, 1; 10.2460/ajvr.75.1.26
Tibial plateau angle was defined as the angle in the sagittal plane formed by the tibial plateau relative to a line perpendicular to the functional axis of the tibia.15 The medial tibial plateau was determined by the line in the sagittal plane joining its most cranial and caudal margins, and the functional axis of the tibia was defined as the line joining the intercondylar eminence and the center of the talus (Figure 2).15 The baseline TPA was 22°. Tibial plateau angle was varied by rotating the articulating surface about a pivot point located on the caudal aspect of the medial tibial condyle. Because of changes in geometry associated with TPA variation, ligament zero-strain lengths (free ligament lengths) were altered to maintain consistent ligament prestrain (ie, free ligament length and TPA may not be independent parameters). Therefore, the free ligament lengths were determined as described20 and accordingly adjusted for each TPA simulated scenario. The patellar ligament length and PLIL were unchanged for all scenarios evaluated. Nine TPA scenarios were simulated (14°, 16°, 18°, 20°, 22° [baseline], 24°, 26°, 28°, and 30°) for an intact and CrCL-deficient stifle joint.
Illustration of the TPA in a canine pelvic limb computer simulation model. A—The TPA was defined as the angle between the medial tibial plateau (solid line) and the line (dot-dash line) perpendicular to the functional axis of the tibia (dotted line). The TPA was altered by pivoting the medial tibial plateau about the caudal aspect of the medial tibial condyle (arrow). The complex surface geometry is represented with surface contour elements. B—The baseline tibia is indicated as a dashed line, and the adjusted tibiae are indicated as solid lines.
Citation: American Journal of Veterinary Research 75, 1; 10.2460/ajvr.75.1.26
Illustration of the TPA in a canine pelvic limb computer simulation model. A—The TPA was defined as the angle between the medial tibial plateau (solid line) and the line (dot-dash line) perpendicular to the functional axis of the tibia (dotted line). The TPA was altered by pivoting the medial tibial plateau about the caudal aspect of the medial tibial condyle (arrow). The complex surface geometry is represented with surface contour elements. B—The baseline tibia is indicated as a dashed line, and the adjusted tibiae are indicated as solid lines.
Citation: American Journal of Veterinary Research 75, 1; 10.2460/ajvr.75.1.26
Illustration of the TPA in a canine pelvic limb computer simulation model. A—The TPA was defined as the angle between the medial tibial plateau (solid line) and the line (dot-dash line) perpendicular to the functional axis of the tibia (dotted line). The TPA was altered by pivoting the medial tibial plateau about the caudal aspect of the medial tibial condyle (arrow). The complex surface geometry is represented with surface contour elements. B—The baseline tibia is indicated as a dashed line, and the adjusted tibiae are indicated as solid lines.
Citation: American Journal of Veterinary Research 75, 1; 10.2460/ajvr.75.1.26
The FCD was defined as the line connecting the cranial and caudal margins of the femoral condyle articulating surface.23 Femoral condyle diameter was altered by proportionally scaling the femoral articulating surface in the craniocaudal and proximodistal directions (Figure 3). The femoral condyle mediolateral dimension was held constant for all evaluated scenarios. Five FCD scenarios were evaluated (−20%, −10%, 0% [baseline], 10%, and 20%) for intact and CrCL-deficient stifle joints. As the FCD increased or decreased, the femoral condyle surface in contact with the meniscus similarly increased or decreased. The femoropatellar contact surface remained unchanged. The articulating surface was oriented craniocaudally and proximodistally with the FCD midpoint defined by Ocal et al23 so that the FCD midpoint was coincident in the baseline articulating surface and the scaled articulating surface. The femoral condyle articulating surface angular orientation about the FCD midpoint in the sagittal plane with respect to the femoral axis remained unchanged as the FCD was increased or decreased. Changes in FCD geometry would anatomically lead to corresponding differences in free ligament lengths to maintain consistent ligament prestrain (ie, free ligament length and FCD may not be independent parameters). Therefore, the free ligament lengths were adjusted for each FCD scenario as described.20 Patellar ligament length and insertion location were unchanged across scenarios.
Illustration of the FCD in a canine pelvic limb computer simulation model. A—The FCD was defined as the line between the cranial and caudal margins of the femoral articulating surface on the medial femoral condyle (arrows).23 The FCD midpoint (circle) was used to locate the articulating surface of the femoral condyles. The angular orientation (α) of the femoral condyle articulating surface with respect to the femoral axis (solid line) remained unchanged. The complex surface geometry is represented with surface contour elements. B—The baseline femoral articulating surface (dashed line) and the altered femoral articulating surfaces (solid lines) are illustrated.
Citation: American Journal of Veterinary Research 75, 1; 10.2460/ajvr.75.1.26
Illustration of the FCD in a canine pelvic limb computer simulation model. A—The FCD was defined as the line between the cranial and caudal margins of the femoral articulating surface on the medial femoral condyle (arrows).23 The FCD midpoint (circle) was used to locate the articulating surface of the femoral condyles. The angular orientation (α) of the femoral condyle articulating surface with respect to the femoral axis (solid line) remained unchanged. The complex surface geometry is represented with surface contour elements. B—The baseline femoral articulating surface (dashed line) and the altered femoral articulating surfaces (solid lines) are illustrated.
Citation: American Journal of Veterinary Research 75, 1; 10.2460/ajvr.75.1.26
Illustration of the FCD in a canine pelvic limb computer simulation model. A—The FCD was defined as the line between the cranial and caudal margins of the femoral articulating surface on the medial femoral condyle (arrows).23 The FCD midpoint (circle) was used to locate the articulating surface of the femoral condyles. The angular orientation (α) of the femoral condyle articulating surface with respect to the femoral axis (solid line) remained unchanged. The complex surface geometry is represented with surface contour elements. B—The baseline femoral articulating surface (dashed line) and the altered femoral articulating surfaces (solid lines) are illustrated.
Citation: American Journal of Veterinary Research 75, 1; 10.2460/ajvr.75.1.26
Comparative analysis—Peak model outcome measures were compared for intact and CrCL-deficient stifle joints for all examined morphological parameters. Model sensitivity for each parameter was reported as the sensitivity index: the percentage change in outcome measure divided by the percentage change in input parameter. For instance, a 5% increase in outcome measure caused by a 10% increase in input parameter represented a sensitivity index of 0.5. Higher sensitivity index values equate to greater sensitivity to that parameter.
Results
Parametric analysis—Eighteen scenarios that differed from baseline were analyzed for each 10% discrete stance phase for intact and CrCL-deficient stifle joints, totaling 324 simulations, not including the baseline simulations. Ligament loads normalized by body weight, RTT, and RTR were determined for each discrete stance phase interval for the intact and CrCL-deficient stifle joints.
Ligament loads—Ligament loads were evaluated for each parameter varied in the intact and CrCL-deficient stifle joints. Peak ligament loads across the stance phase were compared across each parameter varied (Figure 4). The maximum and minimum ligament loads across the range of all varied parameters were compared with the peak baseline ligament loads in the intact and CrCL-deficient stifle joints (Figure 5).
Peak ligament loads as a percentage of body weight (%BW) determined with a canine pelvic limb computer simulation model of intact and CrCL-deficient stifle joints compared with baseline (dotted vertical line) for variation in PLIL, TPA, and FCD.
Citation: American Journal of Veterinary Research 75, 1; 10.2460/ajvr.75.1.26
Peak ligament loads as a percentage of body weight (%BW) determined with a canine pelvic limb computer simulation model of intact and CrCL-deficient stifle joints compared with baseline (dotted vertical line) for variation in PLIL, TPA, and FCD.
Citation: American Journal of Veterinary Research 75, 1; 10.2460/ajvr.75.1.26
Peak ligament loads as a percentage of body weight (%BW) determined with a canine pelvic limb computer simulation model of intact and CrCL-deficient stifle joints compared with baseline (dotted vertical line) for variation in PLIL, TPA, and FCD.
Citation: American Journal of Veterinary Research 75, 1; 10.2460/ajvr.75.1.26
Range of increase or decrease in peak ligament load change from baseline across the stance phase determined with a canine pelvic limb computer simulation model for variations in PLIL, TPA, and FCD in intact (I, unshaded) and CrCL-deficient (D, shaded) stifle joints.
Citation: American Journal of Veterinary Research 75, 1; 10.2460/ajvr.75.1.26
Range of increase or decrease in peak ligament load change from baseline across the stance phase determined with a canine pelvic limb computer simulation model for variations in PLIL, TPA, and FCD in intact (I, unshaded) and CrCL-deficient (D, shaded) stifle joints.
Citation: American Journal of Veterinary Research 75, 1; 10.2460/ajvr.75.1.26
Range of increase or decrease in peak ligament load change from baseline across the stance phase determined with a canine pelvic limb computer simulation model for variations in PLIL, TPA, and FCD in intact (I, unshaded) and CrCL-deficient (D, shaded) stifle joints.
Citation: American Journal of Veterinary Research 75, 1; 10.2460/ajvr.75.1.26
RTT and RTR—Tibial translation and rotation were evaluated for each parameter varied. Peak tibial translation and rotation across the stance phase were compared across each parameter (Figure 6). The peak RTT and RTR across the stance phase associated with each parameter varied were compared with the peak baseline RTT and RTR, respectively (Figure 7).
Peak RTT and RTR determined with a canine pelvic limb computer simulation model compared with baseline (vertical dotted line) for variation in PLIL, TPA, and FCD.
Citation: American Journal of Veterinary Research 75, 1; 10.2460/ajvr.75.1.26
Peak RTT and RTR determined with a canine pelvic limb computer simulation model compared with baseline (vertical dotted line) for variation in PLIL, TPA, and FCD.
Citation: American Journal of Veterinary Research 75, 1; 10.2460/ajvr.75.1.26
Peak RTT and RTR determined with a canine pelvic limb computer simulation model compared with baseline (vertical dotted line) for variation in PLIL, TPA, and FCD.
Citation: American Journal of Veterinary Research 75, 1; 10.2460/ajvr.75.1.26
Range of increase or decrease in peak RTT and RTR from baseline across the stance phase determined with a canine pelvic limb computer simulation model for variations in PLIL, TPA, and FCD.
Citation: American Journal of Veterinary Research 75, 1; 10.2460/ajvr.75.1.26
Range of increase or decrease in peak RTT and RTR from baseline across the stance phase determined with a canine pelvic limb computer simulation model for variations in PLIL, TPA, and FCD.
Citation: American Journal of Veterinary Research 75, 1; 10.2460/ajvr.75.1.26
Range of increase or decrease in peak RTT and RTR from baseline across the stance phase determined with a canine pelvic limb computer simulation model for variations in PLIL, TPA, and FCD.
Citation: American Journal of Veterinary Research 75, 1; 10.2460/ajvr.75.1.26
Sensitivity indices—Peak sensitivity indices were compared to assess the influence of changes in morphological parameters on model outcome measures (Table 1). In the CrCL-intact stifle, CrCL, CaCL, LCL, and MCL loads were most sensitive to TPA. In the CrCL-deficient stifle, CaCL and LCL loads were most sensitive to FCD and MCL load was most sensitive to TPA. Relative tibial translation and rotation were most sensitive to TPA. While instances of peak sensitivity indices associated with a specific parameter are of interest, assessing an outcome's sensitivity across a range of parameters may provide a more generalized evaluation of naturally occurring anatomic variation. The mean peak (absolute value) sensitivity index was used to compare the collective influence of each input parameter across the range of parameters evaluated on each outcome measure for the CrCL-intact and CrCL-deficient stifle joints (Figures 8 and 9). Using this approach, CrCL, CaCL, LCL, and MCL loads were most sensitive to changes in TPA in the CrCL-intact stifle. In the CrCL-deficient stifle, TPA had the greatest influence on CaCL and MCL loads, whereas LCL load was most sensitive to changes in FCD. Relative tibial translation and rotation were sensitive to PLIL and TPA, respectively.
Ligament load mean peak (absolute value) sensitivity indices for PLIL, TPA, and FCD determined with a canine pelvic limb computer simulation model for intact and CrCL-deficient stifle joints. Each axis increment represents a 5-unit increase in sensitivity index in an intact stifle joint and a 1-unit increase in sensitivity index in a CrCL-deficient stifle joint.
Citation: American Journal of Veterinary Research 75, 1; 10.2460/ajvr.75.1.26
Ligament load mean peak (absolute value) sensitivity indices for PLIL, TPA, and FCD determined with a canine pelvic limb computer simulation model for intact and CrCL-deficient stifle joints. Each axis increment represents a 5-unit increase in sensitivity index in an intact stifle joint and a 1-unit increase in sensitivity index in a CrCL-deficient stifle joint.
Citation: American Journal of Veterinary Research 75, 1; 10.2460/ajvr.75.1.26
Ligament load mean peak (absolute value) sensitivity indices for PLIL, TPA, and FCD determined with a canine pelvic limb computer simulation model for intact and CrCL-deficient stifle joints. Each axis increment represents a 5-unit increase in sensitivity index in an intact stifle joint and a 1-unit increase in sensitivity index in a CrCL-deficient stifle joint.
Citation: American Journal of Veterinary Research 75, 1; 10.2460/ajvr.75.1.26
Relative tibial translation and RTR mean peak (absolute values) sensitivity indices for PLIL, TPA, and FCD determined with a canine pelvic limb computer simulation model for intact and CrCL-deficient stifle joints. Each axis increment represents a 2-unit increase in sensitivity index.
Citation: American Journal of Veterinary Research 75, 1; 10.2460/ajvr.75.1.26
Relative tibial translation and RTR mean peak (absolute values) sensitivity indices for PLIL, TPA, and FCD determined with a canine pelvic limb computer simulation model for intact and CrCL-deficient stifle joints. Each axis increment represents a 2-unit increase in sensitivity index.
Citation: American Journal of Veterinary Research 75, 1; 10.2460/ajvr.75.1.26
Relative tibial translation and RTR mean peak (absolute values) sensitivity indices for PLIL, TPA, and FCD determined with a canine pelvic limb computer simulation model for intact and CrCL-deficient stifle joints. Each axis increment represents a 2-unit increase in sensitivity index.
Citation: American Journal of Veterinary Research 75, 1; 10.2460/ajvr.75.1.26
Cranial cruciate ligament, CaCL, MCL, and LCL loads, and RTT and RTR peak sensitivity indices for PLIL change, TPA change, and FCD change in intact (I) and CrCL-deficient (D) stifle joints determined with a canine pelvic limb computer simulation model.
| CrCL load | CaCL load | LCL load | MCL load | |||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Parameter | I | D | I | D | I | D | I | D | RTT | RTR |
| PLIL change (mm) | ||||||||||
| −6 | 0.0 | 0.0 | 0.0 | 0.0 | 0.2 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 |
| −3 | 0.0 | 0.0 | 0.0 | 0.0 | 0.1 | 0.0 | 0.1 | 0.0 | 0.0 | 0.0 |
| 3 | 0.0 | 0.0 | 0.0 | 0.0 | −0.1 | 0.0 | 0.0 | 0.0 | 0.0 | 0.1 |
| 6 | 0.0 | 0.0 | 0.0 | 0.9 | −0.4 | 0.7 | 0.0 | 0.6 | 1.1 | 0.4 |
| 9 | −0.8 | 0.0 | −0.4 | 1.0 | −0.2 | 1.1 | −1.3 | 0.1 | 1.2 | 0.9 |
| 12 | −0.6 | 0.0 | −0.4 | 1.1 | −1.2 | 0.9 | −1.2 | 1.0 | 1.3 | 1.1 |
| TPA change (°) | ||||||||||
| 14 | −2.4 | 0.0 | −6.7 | 1.3 | −10.2 | −1.0 | −11.0 | −0.2 | 1.5* | −1.8 |
| 16 | −6.5 | 0.0 | −14.1* | 1.3 | −8.0 | 0.1 | −2.4 | −3.0 | 0.4 | 3.3 |
| 18 | −8.0 | 0.0 | −12.9 | 2.9 | −16.4 | 0.7 | −33.6* | −5.5 | 0.5 | 20.2* |
| 20 | −11.1* | 0.0 | −11.7 | 4.3 | −21.0* | 1.4 | −14.9 | −10.2 | 0.8 | 17.8 |
| 24 | 0.4 | 0.0 | −1.5 | −1.6 | 16.7 | −0.4 | 18.8 | −3.2 | 0.2 | 2.6 |
| 26 | 2.8 | 0.0 | 3.1 | 3.3 | 13.1 | 0.7 | 10.1 | 17.0* | 0.2 | 6.3 |
| 28 | 3.3 | 0.0 | 5.7 | 2.5 | 6.2 | 1.0 | 3.5 | 8.6 | 0.2 | 5.6 |
| 30 | 3.6 | 0.0 | 4.7 | 3.8 | 5.4 | 0.5 | 13.6 | 5.9 | 0.2 | 4.0 |
| FCD change (%) | ||||||||||
| −20 | −1.9 | 0.0 | −4.3 | −5.3* | 1.2 | 2.0 | −0.1 | −2.8 | −0.2 | −0.9 |
| −10 | −10.7 | 0.0 | −7.9 | 1.8 | −4.2 | −0.3 | −2.6 | 6.9 | 0.1 | 9.2 |
| 10 | 1.8 | 0.0 | −0.6 | −1.3 | 12.3 | 2.3* | 5.2 | 5.6 | −0.4 | −2.2 |
| 20 | 2.9 | 0.0 | 7.4 | 0.1 | 0.8 | 1.8 | 7.4 | 3.6 | −0.3 | 2.9 |
Peak (absolute value) sensitivity index.
Discussion
The computer simulation model used in this study allowed for assessment of 3 canine stifle joint morphological parameters that contribute to intact and CrCL-deficient stifle joint biomechanics. Analyses focused on altering PLIL and TPA, 2 factors that influence the stability of the stifle joint,19,24,25 and on FCD, a factor that potentially influences the stability of the stifle joint.23 Peak values across the stance phase for model outcomes were compared with baseline values for the range of parameters evaluated. To compare outcome measures across parameters, the sensitivity index was defined. The sensitivity index is an appropriate measure to assess outcome sensitivity across parameters because it normalizes the change in outcome measure to the change in input parameter.
Cranially displacing the insertion site of the patellar ligament relative to the tibial tuberosity serves as a biomechanical rationale for stifle joint stabilization techniques such as tibial tuberosity advancement, the modified Maquet technique, and, to a lesser extent, triple tibial osteotomy.26,27 Similarly, the model used in the present study predicted that the PLIL was a primary contributor to stifle joint biomechanics in a CrCL-deficient stifle joint. Increasing the cranial distance between the PLIL relative to the tibial tuberosity had little effect on ligament loads in the CrCL-intact stifle joint. However, ligament loads were reduced in the CrCL-deficient stifle joint when the PLIL was cranially displaced by 6, 9, and 12 mm. Furthermore, RTT was eliminated when the PLIL was cranially displaced by 6, 9, and 12 mm. These results support the biomechanical rationale behind tibial tuberosity advancement, where the patellar ligament force vector alteration neutralizes cranial tibial thrust during stance. Sensitivity indices indicated that outcomes were less sensitive to PLIL than TPA and FCD for the range of parameters evaluated. However, RTT was sensitive to PLIL.
The TPA is one of the key anatomic parameters thought to lead to CrCL deficiency because of its contribution to cranial tibial thrust.19 A TPA of 18° has been reported as less likely to be associated with CrCL deficiency, whereas a TPA > 22° has been reported as more likely to be associated with CrCL deficiency.28 Tibial plateau angles may range from 13° to 34° in dogs, with 23° to 25° being the most common.29 Furthermore, a mean TPA of 26.5° was reported in an in vitro study19 that assessed the tibial plateau leveling osteotomy procedure. Tibial plateau angle in the present model was 22° as defined by the angle in the sagittal plane between the tibial plateau and a line perpendicular to the tibial functional axis. In this model, TPA affected all outcome measures in both an intact and CrCL-deficient stifle joint for the range of parameters evaluated. In an intact stifle joint, ligament loads increased for both increased and decreased TPA. The baseline TPA corresponded with a local minimum, indicating that the natural TPA may represent an optimized TPA for the canine subject used for model development. Cranial cruciate ligament–intact stifle joint outcomes were most sensitive to TPA. In a CrCL-deficient stifle joint, decreased TPA had less effect on outcome measures, compared with a CrCL-intact stifle joint. Increased TPA, however, resulted in substantially increased ligament loads. These increased loads suggested an increase in cranial tibial thrust opposed primarily by the stifle joint ligaments. Furthermore, increased RTT occurred earlier in the stance phase, compared with the baseline model, for both increased and decreased TPA. Baseline RTT occurred at midstance, and RTT following TPA change occurred as early as 20% stance. Although these findings may help explain why RTT occurred earlier in the stance phase in an in vivo study,30 compared with the baseline computer model,20 the TPA in the in vivo study is unknown. In a CrCL-deficient stifle joint, MCL load, RTT, and RTR were also sensitive to TPA.
We expect that FCD relative to limb length varies by breed. Additionally, we have anecdotally noted differences in the size and shape of the distal portion of the femur in Newfoundlands, Rottweilers, and Labrador Retrievers, compared with Dachshunds. Furthermore, Newfoundlands, Rottweilers, and Labrador Retrievers are predisposed to CrCL injury, whereas Dachshunds are less prone to CrCL injury.3 Therefore, it is possible that the size and shape of the femoral condyles influence the biomechanics of the stifle joint and the likelihood of CrCL deficiency. Femoral condyle diameter, however, has not been as extensively studied as the TPA or PLIL for its role in stifle joint stability. Both the distal portion of the femur and the proximal portion of the tibia form the stifle joint, and therefore both contribute to stifle joint biomechanics. For example, the femoral anteversion angle, a measure of the angle of the femoral neck in relation to the femoral shaft, is associated with CrCL deficiency.31 Ocal et al23 also quantified several femoral measurements associated with femoral shape, and FCD was defined in the sagittal plane as the line connecting the caudal and cranial margins of the femoral articulating surface. Femoral condyle diameter was scaled in the model used in the present study with the midpoint remaining coincident across FCD variation. Variation of FCD by ± 20% in this model encompassed the mean FCD and 1 SD reported by Ocal et al23 for 6 breeds similar in size to the dog modeled in this study.
In an intact stifle joint, peak cruciate ligament loads increased for both increased and decreased FCD, whereas the collateral ligament loads were relatively unchanged. The baseline FCD corresponded with a local minimum, indicating that the natural FCD may represent an optimized FCD for the canine subject used to develop the model. Femoral condyle diameter variation of −10% resulted in the maximum CrCL load. At −10% FCD, excessive internal rotation was present, which increased the peak CrCL load in the intact stifle joint. The sensitivity of peak CrCL load to FCD in the intact stifle joint suggested that the FCD may play a clinically relevant role in the likelihood of CrCL deficiency, which may present opportunities for surgical management strategies. Although all intact stifle joint ligament loads were more sensitive to FCD than PLIL, CaCL, LCL, and MCL loads were most sensitive to TPA across the range of parameters evaluated. Similarly, RTR was sensitive to FCD but was more sensitive to TPA.
In a CrCL-deficient stifle joint, increased FCD resulted in a slight increase in LCL and MCL loads. The peak CaCL load remained unchanged except for −20% FCD change for which substantial RTT occurred at 20% stance. This event resulted in the maximum CaCL load because of the combination of reduced FCD and the peak ground reaction force. Relative tibial translation was minimal for all other FCD changes at 20% stance. The model predicted that increased FCD variation slightly decreased RTT, and RTR patterns were not apparent. Finally, LCL load was most sensitive to FCD, and CaCL load was also sensitive to FCD for the range of parameters evaluated. In this study, the only femoral parameter investigated was FCD; additional femoral geometric parameters such as femoral anteversion angle and the orientation of the femoral condyles are in need of further investigation.
To our knowledge, this is the first 3-D computer model of the canine pelvic limb that used rigid body motion simulation to evaluate the sensitivity of ligament loads, tibial translation, and tibial rotation to morphological parameters for both an intact and CrCL-deficient stifle joint during stance.
Although individual parameters could be altered in the model, some may not be fully independent. Changing stifle joint conformation geometry (TPA and FCD) modified the distance between each stifle joint ligament origin and insertion from the baseline model. Therefore, the zero-strain lengths for each ligament during stance were correspondingly adjusted to ensure that the minimum ligament lengths during stance corresponded to 0% strain as in the baseline scenario. Furthermore, kinematics, ground reaction forces, and muscle forces used as input to the baseline model were maintained when simulations were conducted to evaluate altered morphology. This approach allowed for independent evaluation of the effects of change in PLIL, TPA, and FCD, compared with the baseline conditions. However, kinematics, ground reaction forces, and muscle forces may differ from baseline in dogs with naturally occurring variants of these morphological parameters; these potential differences were not investigated in this study.
Each parameter was varied to approximate clinically and morphologically relevant ranges. However, the range and the discrete values chosen within these ranges for each parameter may not have captured all model behaviors associated with parametric variation. This limitation was most apparent for parameters with substantial outcome measure variation from one discrete value to the next. Evaluation of intermediate parameter values may uncover trends not apparent in this study. The quasistatic nature of this model coupled with a discrete parametric sensitivity analysis established an approximation of canine stifle joint biomechanics for varied parameters. Furthermore, evaluation of a larger range of parameters may reveal model sensitivity to parameters that the present study did not indicate. Although this model indicated greater sensitivity to some parameters, these results may not be generalizable to all breeds or dogs of varying size or mass. The model was based on 1 canine subject, and it is expected that different canine subjects may be affected differently by PLIL, TPA, and FCD. Canine pelvic limb models representing a wide variety of breeds are needed to compare breed-specific parameters that may influence the likelihood of CrCL deficiency.
Computer models are well suited for parametric analyses. Model simulation confirmed that PLIL, TPA, and FCD affect intact and CrCL-deficient stifle joint biomechanical outcomes. To our knowledge, FCD has not been considered as a factor in CrCL deficiency and is in need of further investigation.
ABBREVIATIONS
| CaCL | Caudal cruciate ligament |
| CrCL | Cranial cruciate ligament |
| FCD | Femoral condyle diameter |
| LCL | Lateral collateral ligament |
| MCL | Medial collateral ligament |
| PLIL | Patellar ligament insertion location |
| RTR | Relative tibial rotation |
| RTT | Relative tibial translation |
| TPA | Tibial plateau angle |
SolidWorks, version 2010, SolidWorks Corp, Concord, Mass.
SolidWorks Motion, version 2010, Structural Research and Analysis Corp, Santa Monica, Calif.
References
1. Griffon DJ. A review of the pathogenesis of canine cranial cruciate ligament disease as a basis for future preventive strategies. Vet Surg 2010; 39:399–409.
2. Hayashi K, Manley PA, Muir P. Cranial cruciate ligament pathophysiology in dogs with cruciate disease: a review. J Am Anim Hosp Assoc 2004; 40:385–390.
3. Witsberger TH, Villamil JA, Schultz LG, et al. Prevalence of and risk factors for hip dysplasia and cranial cruciate ligament deficiency in dogs. J Am Vet Med Assoc 2008; 232:1818–1824.
4. Aragon CL, Budsberg SC. Applications of evidence-based medicine: cranial cruciate ligament injury repair in the dog. Vet Surg 2005; 34:93–98.
5. Arnoczky SP, Marshall JL. The cruciate ligaments of the canine stifle: an anatomical and functional analysis. Am J Vet Res 1977; 38:1807–1814.
6. Breshears LA, Cook JL, Stoker AM, et al. Detection and evaluation of matrix metalloproteinases involved in cruciate ligament disease in dogs using multiplex bead technology. Vet Surg 2010; 39:306–314.
7. Cook JL. Cranial cruciate ligament disease in dogs: biology versus biomechanics. Vet Surg 2010; 39:270–277.
8. Wilke VL, Robinson DA, Evans RB, et al. Estimate of the annual economic impact of treatment of cranial cruciate ligament injury in dogs in the United States. J Am Vet Med Assoc 2005; 227:1604–1607.
9. Conzemius MG, Evans RB, Besancon MF, et al. Effect of surgical technique on limb function after surgery for rupture of the cranial cruciate ligament in dogs. J Am Vet Med Assoc 2005; 226:232–236.
10. Vasseur PB, Pool RR, Arnoczky SP, et al. Correlative biomechanical and histologic study of the cranial cruciate ligament in dogs. Am J Vet Res 1985; 46:1842–1854.
11. Leighton RL. Preferred method of repair of cranial cruciate ligament rupture in dogs: a survey of ACVS diplomates specializing in canine orthopedics. American College of Veterinary Surgery (lett). Vet Surg 1999; 28:194.
12. Boudrieau RJ. Tibial plateau leveling osteotomy or tibial tuberosity advancement? Vet Surg 2009; 38:1–22.
13. Lopez MJ. Canine stifle disease: time for a paradigm shift? (edit) Vet Surg 2010; 39:269.
14. Breshears LA, Cook JL, Stoker AM, et al. The effect of uniaxial cyclic tensile load on gene expression in canine cranial cruciate ligamentocytes. Vet Surg 2010; 39:433–443.
15. Reif U, Probst CW. Comparison of tibial plateau angles in normal and cranial cruciate deficient stifles of Labrador Retrievers. Vet Surg 2003; 32:385–389.
16. Whitehair JG, Vasseur PB, Willits NH. Epidemiology of cranial cruciate ligament rupture in dogs. J Am Vet Med Assoc 1993; 203:1016–1019.
17. Bertozzi L, Stagni R, Fantozzi S, et al. Evaluation of a cruciate ligament model: sensitivity to the parameters during drawer test simulation. J Appl Biomech 2008; 24:234–243.
18. Shahar R, Milgram J. Biomechanics of tibial plateau leveling of the canine cruciate-deficient stifle joint: a theoretical model. Vet Surg 2006; 35:144–149.
19. Warzee CC, Dejardin LM, Arnoczky SP, et al. Effect of tibial plateau leveling on cranial and caudal tibial thrusts in canine cranial cruciate-deficient stifles: an in vitro experimental study. Vet Surg 2001; 30:278–286.
20. Brown NP, Bertocci GE, Marcellin-Little DJ. Development of a canine stifle computer model to evaluate cranial cruciate ligament deficiency. J Mech Med Biol 2013; 13:1350043–1350071.
21. Grood ES, Suntay WJ. A joint coordinate system for the clinical description of three-dimensional motions: application to the knee. J Biomech Eng 1983; 105:136–144.
22. Torres BT, Punke JP, Fu YC, et al. Comparison of canine stifle kinematic data collected with three different targeting models. Vet Surg 2010; 39:504–512.
23. Ocal MK, Sevil-Kilimci F, Yildirim IG. Geometry of the femoral condyles in dogs. Vet Res Commun 2012; 36:1–6.
24. Apelt D, Kowaleski MP, Boudrieau RJ. Effect of tibial tuberosity advancement on cranial tibial subluxation in canine cranial cruciate-deficient stifle joints: an in vitro experimental study. Vet Surg 2007; 36:170–177.
25. Reif U, Hulse DA, Hauptman JG. Effect of tibial plateau leveling on stability of the canine cranial cruciate-deficient stifle joint: an in vitro study. Vet Surg 2002; 31:147–154.
26. Etchepareborde S, Brunel L, Bollen G, et al. Preliminary experience of a modified Maquet technique for repair of cranial cruciate ligament rupture in dogs. Vet Comp Orthop Traumatol 2011; 24:223–227.
27. Kim SE, Pozzi A, Kowaleski MP, et al. Tibial osteotomies for cranial cruciate ligament insufficiency in dogs. Vet Surg 2008; 37:111–125.
28. Morris E, Lipowitz AJ. Comparison of tibial plateau angles in dogs with and without cranial cruciate ligament injuries. J Am Vet Med Assoc 2001; 218:363–366.
29. Comerford E. Stifle morphology. In: Muir P, ed. Advances in the canine cranial cruciate ligament. Hoboken, NJ: Wiley-Blackwell, 2010;68.
30. Tashman S, Anderst W, Kolowich P, et al. Kinematics of the ACL-deficient canine knee during gait: serial changes over two years. J Orthop Res 2004; 22:931–941.
31. Ragetly CA, Evans R, Mostafa AA, et al. Multivariate analysis of morphometric characteristics to evaluate risk factors for cranial cruciate ligament deficiency in Labrador Retrievers. Vet Surg 2011; 40:327–333.
