Stifle joint disease is the most common cause of pelvic limb lameness in dogs, and the 2 most common stifle abnormalities include CrCL rupture and MPL.1 Congenital derangements and developmental disorders can lead to MPL in young dogs.2 The MPL creates malalignment of the extensor apparatus, internal rotation of the tibia, and a resultant increase in biomechanical stress on the CrCL, potentially leading to CrCL degeneration and rupture.2–4 This increase in stress can be exaggerated by the increase in TPA that has been observed in some dogs with CrCL rupture.5,6 Various surgical procedures have been developed to restore stability in CrCL-deficient stifle joints, and many are designed to passively control cranial tibial thrust in relation to the distal portion of the femur.7 Specifically, the TPLO procedure is intended to neutralize femorotibial joint reaction forces and cranial translation of the proximal portion of the tibia by reducing the TPA while minimizing caudal tibial thrust and caudal subluxation.2 The modified TPLO procedure, described by Lagenbach and Marcellin-Little,8 involves internal rotation of the proximal tibial fragment in relation to the main tibial fragment, thereby transposing the tibial crest and realigning the quadriceps mechanism. Those authors suggested that this modified technique addresses both MPL and CrCL rupture by displacing the osteotomized tibial plateau in relation to the principle tibial fragment.8 However, they did not mention whether an attempt was made to keep the patella reduced during preoperative radiographic TPA measurement. Patellar luxation during radiograph acquisition could potentially lead to inaccurate TPA measurements.
To the authors' knowledge, the effect of patellar luxation on radiographic TPA has not been evaluated in dogs. At our hospital, small-breed dogs with CrCL rupture and concurrent grade 2 or 3 MPL are commonly treated by TPLO in addition to retinacular release, imbrication, and wedge or block trochleoplasty to address the MPL. However, we were uncertain whether the unquantified internal tibial rotation that occurs with grades 2 and 3 MPL affects measurement of the TPA. Therefore, the purpose of the study reported here was to determine whether a luxated patella would need to be manually reduced during preoperative radiographic examination of small-breed dogs with grade 2 or 3 MPL to allow accurate TPA measurements. We hypothesized that there would be no significant difference in radiographic TPA measurements obtained with the patella in a luxated or reduced position and, therefore, patella status (luxated or reduced) would be inconsequential when measuring TPA in small-breed dogs.
Materials and Methods
Animals
From August 2014 through October 2015, dogs evaluated at MissionVet Specialty and Emergency for MPL were considered for inclusion in the study. Dogs were included if they weighed < 15 kg (33 lb), had grade 2 or 3 MPL in 1 or both stifle joints, and the patella could be maintained both in and out of the femoral trochlea for radiographic examination. The study protocol was reviewed by an institutional animal care and use committee, which waived the need for their oversight, but all owners consented to their dog's participation.
Radiography
For each enrolled dog, mediolateral digital radiographic images of the stifle joint were acquireda with the patella of the affected limb or limbs luxated and again with the patella manually reduced, yielding 2 images/joint. Patellar status was confirmed by palpation before and after each image was acquired. Stifle joints were positioned with superimposition of the femoral condyles, and every attempt was made to obtain condylar superimposition during image acquisition. However, perfect superimposition was not always possible owing to the inherent distal femoral varus and internal tibial rotation associated with MPL. In these situations, we believed the tibial condyles were visible enough to allow accurate TPA measurement. For each joint, a metallic marker was used to denote the image in which the patella was manually luxated.
TPA measurement
Acquired radiographic images were assigned a number between 1 and 50 from a random number table by the same investigator (MJO), who kept a record of the dogs and joints to which these numbers were assigned. To maintain blinding of observers to patella status, each image was digitally cropped to eliminate the metallic marker and patella to avoid observer bias. Three observers (2 diplomates of the American College of Veterinary Surgeons [THW and JDM; observers 1 and 2, respectively] and 1 surgical intern [MJO; observer 3]) were instructed to measure the TPA in each image by following published guidelines.9 Observers were able to magnify each digital image and then create a line between 2 anatomic points and calculate the angle of intersection to the nearest degree. The TPA was measured as the angle between a perpendicular line to the mechanical axis of the tibia and tibial plateau slope (Figures 1 and 2). Once each observer had measured the TPA in all images in the first measurement session, the digitally cropped images were assigned a different identification number between 1 and 50 and then observers performed the same measurements again in a second session > 48 hours later, yielding a total of 50 measurements/observer/session. Observers were unaware of each other's TPA measurements and of their own previous measurements at the time of the second session.
Mediolateral radiographic image of the stifle joint of an 8-year-old spayed female Miniature Pinscher with grade 2 MPL of the right hind limb, with the patella in a manually reduced position. For the study, the radiograph was digitally cropped to eliminate the patella and, therefore, blind observers to the dog's patella status. The blue line represents the tibial plateau slope, and the yellow line represents the mechanical axis. The reported angle (115°) represents the TPA.
Citation: Journal of the American Veterinary Medical Association 253, 6; 10.2460/javma.253.6.746
Mediolateral radiographic image of the same stifle joint as in Figure 1, with the patella in a luxated position. For the study, the radiograph was digitally cropped to eliminate the patella and metallic marker (paperclip) used to indicate that the patella was manually luxated and, therefore, blind observers to the dog's patella status. The TPA measured with the patella luxated was consistently 1° greater than the TPA measured with the patella reduced. See Figure 1 for remainder of key.
Citation: Journal of the American Veterinary Medical Association 253, 6; 10.2460/javma.253.6.746
Statistical analysis
To determine the effect of patella status (luxated or reduced) on TPA measurements, mixed linear modeling was performed.b Patella status, observer identity, measurement session, and all 2-way interactions among these variables were included as fixed effects. Stifle joint identity, patella status within stifle joint, observer identity within stifle joint, and measurement session within stifle joint were included as random effects with a variance components covariance structure. The P values obtained in this modeling were adjusted to account for multiple comparisons.c
Intraobserver and interobserver variation in measurements were assessed via ANCOVA. For intraobserver variation, the coefficient of variation was calculated from the measurements of each observer for each stifle joint–patella status combination. Mixed linear modeling was then performed with the coefficient of variation as the dependent variable, patella status as the fixed effect, and stifle joint as a random effect. For interobserver variation, the coefficient of variation was calculated for each stifle joint–measurement session–patella status combination. A mixed model was then created similarly to the one for intraobserver agreement, with the addition of measurement session and its interaction with patella status as fixed effects.
For all statistical models, distributions of the conditional residuals were evaluated to confirm the appropriateness of the model. Values of P ≤ 0.05 were considered significant.
Results
Animals
Fifteen dogs (25 stifle joints) with grade 2 or 3 MPL were included in the study. Five dogs had unilateral luxation, and 10 had bilateral luxation. Seventeen stifle joints had grade 2 MPL, and 8 joints had grade 3 MPL. The most common breeds of dog were Yorkshire Terrier (n = 5), Shih Tzu (3), Chihuahua (2), and Bichon Frise (2). All dogs were of small breeds (< 15 kg), with body weights ranging from 2.2 to 12.9 kg (4.8 to 28.4 lb; mean, 5.5 kg [12.1 lb]).
TPA measurements
The TPA measurements made by all observers during the second measurement session were significantly (P = 0.01) greater than those made during the first session, for a mean difference of 0.7°. Least squares mean values of all TPA measurements made during the first measurement session were 24.1° (95% CI, 22.8° to 25.4°) for images in which the patella was luxated and 25.0° (95% CI, 23.7° to 26.3°) for images in which the patella had been manually reduced (mean difference between patella statuses, 0.9°). Values for the second measurement session were 25.0° (95% CI, 23.6° to 26.3°) and 25.6° (95% CI, 24.3° to 26.9°), respectively (mean difference between patella statuses, 0.6°). Mean differences between patella statuses did not differ significantly (P = 0.53) between sessions. No significant (P = 0.42) interaction was identified between measurement session and observer.
The effects of observer and patella status on TPA measurements were not independent, as indicated by the significant (P = 0.01) interaction between these 2 variables that was identified in the statistical model. Least squares mean values for TPA measurements for observer 1 were 24.7° (95% CI, 23.2 to 26.1°) for luxated patellae and 24.8° (95% CI, 23.4° to 26.2°) for reduced patellae (mean difference, 0.2°; P = 1.00); for observer 2 were 23.7° (95% CI, 22.2° to 25.1°) and 25.4° (95% CI, 23.9° to 26.8°), respectively (mean difference, 1.7°; P = 0.009); and for observer 3 were 25.3° (95% CI, 23.9° to 26.7°) and 25.7° (95% CI, 24.3° to 27.1°), respectively (mean difference, 0.4°; P = 0.96). Measurements of TPA made on luxated patella images by observer 2 were a mean of 1.6° lower than those made by observer 3 (P = 0.04). No other significant (P > 0.44) differences in TPA measurements on luxated or reduced patella images were identified among the 3 observers.
Observer variation
No significant differences in intraobserver variation were detected for any observer between measurements of TPA on reduced patella images and those made on luxated patella images (P > 0.21). Similarly, no significant effect of patella status, measurement session, and the interaction between these 2 variables was observed on interobserver variation (P > 0.32).
Discussion
In the present study involving small-breed dogs with grade 2 or 3 MPL, the most common breeds were Yorkshire Terrier and Chihuahua, supporting previously reported findings.5,10,11 Our results suggested that radiographic planning for TPLO may be clinically unaffected by patella position in small-breed dogs with a grade 2 or grade 3 MPL, given that the mean difference in TPA measurements between luxated and nonluxated patella images was ≤ 1.7°.
Developmental disorders of the femur and tibia such as coxa va ra, diminished anteversion angle, genu varum, hypoplasia of the medial femoral condyle, shallow trochlear sulcus, and medial displacement of the tibial tuberosity are some of the congenital derangements leading to advanced skeletal changes and MPL in young dogs.2,5 In healthy dogs, contraction of the quadriceps femoris muscle with the patella in axial alignment leads to extension of the stifle joint. As the stifle joint flexes, the lateral femoral condyle displaces caudally in relation to the tibia as the lateral collateral ligament relaxes. Meanwhile, the tautness of the medial collateral ligament remains relatively unchanged, causing internal tibial rotation.6,12 The intact CrCL limits this internal rotation of the tibia. Chronic patellar luxation has been suggested to precipitate the progression of skeletal deformities, accelerating pathological changes in the stifle joint and increasing the grade of patellar luxation.5,13 Indeed, 18 of 37 (49%) dogs with grade 4 MPL had concomitant CrCL rupture in a previous study.5
Cranial cruciate ligament rupture is the most common cause of lameness in skeletally mature dogs.14 Although no definitive cause for CrCL rupture has been identified, such rupture can occur secondary to MPL, given that 1 of the 3 major roles of the CrCL is to limit internal rotation of the tibia.15 In healthy dogs, the mean ± SD TPA angle is reportedly 24 ± 3.19°.16 An abnormal TPA may subject the CrCL to abnormal loading, which when combined with the malalignment and internal rotation of the proximal portion of the tibia, predisposes dogs to CrCL rupture.5 Results of biomechanical analysis of corrective TPLO in dogs indicate that rotation of the tibial plateau fragment to a TPA of approximately 6.5° neutralizes this cranial tibial thrust without causing excessive strain on the caudal cruciate ligament.2,6 To the authors' knowledge, no studies have been conducted to determine the clinically acceptable amount of variation in TPA measurements in regard to preoperative planning. However, 1 study17 revealed no significant difference in ground reaction forces in dogs with a postoperative TPA between 0° and 14° ≥ 4 months after TPLO surgery.
Preoperative radiographic TPA measurements are used to determine the magnitude of intraoperative rotation of the proximal tibial fragment in dogs with CrCL rupture, and the mean intraobserver variation in such measurements is reportedly 1.5°.18 Because surgeons generally use a straight osteotome, ruler, or caliper to measure the amount of rotation along a radial osteotomy, some minor measuring variation is typically introduced during the surgical procedure. However, to the authors' knowledge, the amount of variation introduced during the physical measurement of osteotomy rotation has not yet been determined. Additionally, loss of the kerf is not considered in rotation calculations and could possibly affect the degree of absolute rotation achieved. Given information provided in a published TPLO rotation chartd and assuming use of a 30-mm saw blade, a 2° difference in TPA could be expected to occur between planned and actual tibial plateau correction when a 1-mm error is made in the osteotomy rotation. In the present study, the mean difference between TPA measurements of stifle joints with luxated and nonluxated patellae was 0.9° in the first measurement session and 0.6° in the second session.
Perfect superimposition of the tibial condyles would result in a reliable measurement of TPA regardless of patella status. However, we believe it can be difficult to assess tibial condylar superimposition on mediolateral radiographs of stifle joints in small-breed dogs, particularly small-breed dogs with stifle joint disease, proximal tibial deformities, and osteoarthritic changes. Our goal, then, was to determine whether the easier-to-see femoral condyles (vs tibial condyles) could be used instead of the typical mediolateral stifle joint radiograph in dogs with MPL. However, we acknowledge that perfect superimposition may not always be feasible because the femoral condyles may not be aligned symmetrically with the femoral diaphysis owing to the distal femoral varus associated with MPL.2 Our hypothesis, that no significant difference would exist in TPA measurements between images in which the patella was luxated versus reduced, was rejected because, although a significant difference of 1.7° was observed for 1 observer, no significant difference was identified for the other 2 observers. Regardless, we believe that this small difference was not clinically important and, as such, internal rotation caused by grades 2 and 3 patellar luxation in small-breed dogs had no clinical effect on radiographic measurement of TPA when the femoral condyles were used as a guide for mediolateral radiographs of the stifle joint.
Although not the primary objective of the study reported here, intraobserver and interobserver variation were assessed. Consistent with previous reports,18–20,e no significant observer effect of patellar luxation on TPA measurements was identified, suggesting that all observers could consistently reproduce such radiographic measurements.
A limitation to the present study was the small sample size and inclusion of only small-breed dogs with grade 2 or 3 MPL. Inclusion of large-breed dogs and dogs with grades 1 and 4 MPL would have increased the sample size, thereby increasing the power of the study to detect significant differences and resulting in greater variation in TPA measurements, and yet sufficient statistical power existed to detect differences that were not large enough to be clinically relevant. Inclusion of dogs with grades 1 and 4 MPL would have also improved the generalizability of the results; however, such dogs were excluded because the patellae cannot be kept luxated (grade 1) or reduced (grade 4), precluding comparisons by patella status for such dogs. Because no dogs with grade 4 MPL were included in the study, and such dogs could be expected to have more severe femoral and tibial deformities than dogs with grade 2 or 3 MPL, the results reported here may not pertain to dogs with grade 4 MPL. Additional studies involving advanced imaging techniques such as CT may be needed to confirm our findings in dogs with severe MPL. Large-breed dogs were intentionally excluded to control for the additional interdog and observer variation that might have been otherwise introduced.
Given the results of the study reported here, we concluded that radiographic measurements of TPA were not clinically altered by the tibial rotation caused by different patella positions. In small-breed dogs with a grade 2 or 3 MPL, radiographic planning for TPLO would likely be clinically unaffected by patella position when superimposition of the femoral condyles is used as a guide for obtaining a mediolateral stifle joint radiograph.
Acknowledgments
The authors declare that there were no conflicts of interest.
ABBREVIATIONS
| CI | Confidence interval |
| CrCL | Cranial cruciate ligament |
| MPL | Medial patellar luxation |
| TPA | Tibial plateau angle |
| TPLO | Tibial plateau leveling osteotomy |
Footnotes
Sound Smart DR, Sound, Carlsbad, Calif.
PROC MIXED, SAS for Windows, version 9.4, SAS Institute Inc, Cary, NC.
Simulate option in an LSMESTIMATE statement, SAS for Windows, version 9.4, SAS Institute Inc, Cary, NC.
TPLO rotation chart, Slocum Enterprises Inc, Eugene, Ore.
Reif U. Influence of limb positioning and interobserver variation on the measurement of the tibial plateau angle (abstr). Vet Comp Orthop Traumatol 2001;14:A10.
References
Johnson JA, Austin C, Breur GJ. Incidence of canine appendicular musculoskeletal disorders in 16 veterinary teaching hospitals from 1980 through 1989. Vet Comp Orthop Traumatol 1994;7:56–69.
Kowaleski MP, Boudrieau RJ, Pozzi A. Stifle joint. In: Tobias KM, Johnston SA, eds. Veterinary surgery: small animal. St Louis: Elsevier, 2012;906–998.
DeAngelis M, Hohn RB. Evaluation of surgical correction of canine patellar luxation in 142 cases. J Am Vet Med Assoc 1970;156:587–594.
DeCamp CE, Riggs CM, Olivier NB, et al. Kinematic evaluation of gait in dogs with cranial cruciate ligament rupture. Am J Vet Res 1996;57:120–126.
Campbell CA, Horstman CL, Mason DR, et al. Severity of patellar luxation and frequency of concomitant cranial cruciate ligament rupture in dogs: 162 cases (2004–2007). J Am Vet Med Assoc 2010;236:887–891.
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.
Abel SB, Hammer DL, Shott S. Use of the proximal portion of the tibia for measurement of the tibial plateau angle in dogs. Am J Vet Res 2003;64:1117–1123.
Langenbach A, Marcellin-Little DJ. Management of concurrent patellar luxation and cranial cruciate ligament rupture using modified tibial plateau leveling. J Small Anim Pract 2010;51:97–103.
Dismukes DI, Tomlinson JL, Fox DB, et al. Radiographic measurement of canine tibial angles in the sagittal plane. Vet Surg 2008;37:300–305.
LaFond E, Breur GJ, Austin CC, et al. Breed susceptibility for developmental orthopedic diseases in dogs. J Am Anim Hosp Assoc 2002;38:467–477.
Hayes AG, Boudrieau RJ, Hungerford LL. Frequency and distribution of medial and lateral patellar luxation in dogs: 124 cases (1982–1992). J Am Vet Med Assoc 1994;205:716–720.
Vasseur PB, Arnoczky SP. Collateral ligaments of the canine stifle joint: anatomic and functional analysis. Am J Vet Res 1981;42:1133–1137.
Roush JK. Canine patellar luxation. Vet Clin North Am Small Anim Pract 1993;23:855–868.
Haynes KH, Biskup J, Freeman A, et al. Effect of tibial plateau angle on cranial cruciate ligament strain: ex vivo study in the dog. Vet Surg 2015;44:46–49.
Kim SE, Pozzi A, Kowaleski MP, et al. Tibial osteotomies for cranial cruciate ligament insufficiency in dogs. Vet Surg 2008;37:111–125.
Sabanci SS, Ocal MK. Lateral and medial tibial plateau angles in normal dogs. Vet Comp Orthop Traumatol 2014;27:135–140.
Robinson DA, Mason DR, Evans R, et al. The effect of tibial plateau angle on ground reaction forces 4–17 months after tibial plateau leveling osteotomy in Labrador Retrievers. Vet Surg 2006;35:294–299.
Fettig AA, Rand WM, Sato AF, et al. Observer variability of tibial plateau slope measurement in 40 dogs with cranial cruciate ligament–deficient stifle joints. Vet Surg 2003;32:471–478.
Aulakh KS, Harper TAM, Lanz OI, et al. Effect of stifle angle on the magnitude of the tibial plateau angle measurement in dogs with intact and transected cranial cruciate ligament. A cadaveric study. Vet Comp Orthop Traumatol 2011;24:272–278.
Caylor KB, Zumpano CA, Evans LM, et al. Intra- and interobserver measurement variability of tibial plateau slope from lateral radiographs in dogs. J Am Anim Hosp Assoc 2001;37:263–268.
