Criteria for Selection of Cases

Medical records of dogs examined at the Norwood Park Animal Hospital between June 2001 and June 2005 were reviewed to identify dogs that underwent TPLO because of CCL rupture. Cases were excluded from the study if medical records were incomplete or if radiographs had not been obtained 8 weeks after surgery.

Procedures

Medical record review—Information obtained from the medical records of cases included in the study consisted of breed, sex, age at the time of surgery, body weight, affected limb, recommended angle of rotation of the proximal tibial segment, actual angle of rotation of the proximal tibial segment, attending surgeon, TPLO plate size, and TPLO saw blade size. All procedures had been performed by 1 of 2 authors (CRH or DLH) in accordance with recommendations of the inventors of the procedure.¹³ In all dogs, a meniscal release had been performed through a limited caudomedial arthrotomy.

Determination of DJD score—Mediolateral and caudocranial radiographic projections of affected stifle joints obtained before and 8 weeks after surgery were examined by 2 of the authors (CRH and DLH), and DJD scores were assigned. A modification of a previously reported system³ was used to assign scores. Investigators worked independently when assigning scores and were blinded to patient identities and the other investigator's assigned scores. Radiographs were evaluated by each investigator in a randomized order, with the order for preoperative radiographs different from the order for postoperative radiographs. Because implants were placed during surgery, investigators were aware of whether they were examining preoperative or postoperative radiographs. However, they were unable to match preoperative and postoperative radiographs for individual dogs.

Scores for 30 factors were used to calculate the DJD score. Each factor was evaluated independently and graded on an integer numeric scale from 0 to 3, with a score of 0 indicating the factor was not present, a score of 1 indicating mild changes were present, a score of 2 indicating moderate changes were present, and a score of 3 indicating severe changes were present. Scores for each of the 30 factors were summed to obtain the DJD score; potential DJD scores, therefore, ranged from 0 to 90. Individual scores from the 2 investigators were averaged to obtain DJD scores used in statistical analyses.

Measurement of TSA—Mediolateral radiographic projections of affected tibias obtained before, immediately after, and 8 weeks after surgery were used to measure TSA. Dogs were anesthetized when radiographs were obtained. The stifle and tarsal joints were included on radiographic views, and the limb was positioned such that the femoral condyles were superimposed on the mediolateral projection.

The TSA was determined by drawing a line on the radiograph to connect the midpoint of the tibial intercondylar tubercles to the center of the talus, establishing the tibial axis. A second line was then drawn to connect the craniomedial margin of the tibial plateau to the caudomedial margin of the tibial plateau. The angle between this second line and a line drawn perpendicular to the tibial axis was measured as the TSA.¹³ All measurements were made by a single investigator (CH) with a protractor.

Statistical analysis—Commercially available software^a was used for data management and statistical analysis. Histograms were used to evaluate distributions of noncategorical variables for statistical normality. Scatterplots, Pearson correlations, and paired t tests were used to assess the degree of agreement between the 2 investigators with respect to DJD scores. The Friedman test was used to compare DJD scores and scores for the 30 individual factors used to calculate the DJD score between time periods (ie, before and 8 weeks after surgery). The Kruskal-Wallis test and Mann-Whitney test were used to compare independent groups with respect to DJD score. The paired t test was used to compare TSAs immediately after and 8 weeks after surgery. One-way ANOVA followed by the Tukey multiple comparison procedure was used to compare independent groups with respect to TSA; the assumption of equal variances was examined by use of the Levene test. Scatterplots and Pearson or Spearman correlations were used to evaluate associations between noncategorical variables. The χ² test of association was used to compare independent groups with respect to percentages. To avoid violations of the assumption of statistical independence, for dogs that underwent bilateral surgery, data for the first stifle joint treated were used for all statistical analyses. Data for the second stifle joints were then analyzed separately as a smaller sample. For all analyses, a value of P < 0.05 was considered significant. No 1-sided statistical tests were used.

Results

A total of 295 dogs met the criteria for inclusion in the study, of which 78 (26.4%) had undergone bilateral surgery during the 4-year study period. Preoperative and immediate postoperative TSAs were obtained for all 295 dogs, and the 8-week TSA was measured in 240 dogs. Degenerative joint disease scores were measured before surgery in 286 dogs, 8 weeks after surgery in 236 dogs, and both before and 8 weeks after surgery in 233 dogs. Complete measurements could not always be obtained because of difficulties with some of the radiographs. Labrador Retrievers and Labrador Retriever mixes (n = 96 [32.5%]), Rottweilers (34 [11.5%]), mixed-breed dogs (32 [10.8%]), and German Shepherd Dogs and German Shepherd Dog mixes (23 [7.8%]) were most common. Dogs ranged from 1 to 13 years old at the time of surgery (mean ± SD, 5.7 ± 2.8 years). Body weight at the time of surgery ranged from 10.2 to 89.1 kg (22.4 to 196.0 lb; mean ± SD, 35.0 ± 12.0 kg [77 ± 26.4 lb]). The left stifle joint was affected in 146 dogs, and the right stifle joint was affected in 149 (for dogs that underwent bilateral surgery, only the initial procedure was considered). A 3.5-mm plate was used in 225 (76.3%) dogs, a 2.7-mm plate was used in 63 (21.4%), and a 2.0-mm plate was used in 7 (2.4%). A 30-mm saw blade was used in 24 (8.1%) dogs, a 24-mm saw blade was used in 192 (65.1%), an 18-mm saw blade was used in 71 (24.1%), and a 12-mm saw blade was used in 8 (2.7%).

Individual preoperative and 8-week postoperative DJD scores assigned by the 2 investigators were significantly (P < 0.001) correlated (Pearson correlation coefficient, 0.92 and 0.85, respectively). A small, but significant (P < 0.001), difference in mean preoperative DJD scores was found between investigators (mean ± SD, 14.7 ± 7.7 vs 16.2 ± 6.9). However, mean 8-week postoperative DJD scores were not significantly different between investigators. The average of individual DJD scores for the 2 investigators was used for all subsequent analyses.

A small, but significant (P < 0.001), increase was found in mean DJD score 8 weeks after surgery, compared with mean preoperative score (17.4 ± 4.2 vs 15.2 ± 6.9). Preoperative and 8-week postoperative DJD scores were positively correlated (r = 0.75; P < 0.001). Mean preoperative DJD score was significantly higher for dogs in which 3.5-mm plates were used, compared with mean scores for dogs in which 2.0-mm plates were used (P = 0.006) and dogs in which 2.7-mm plates were used (P < 0.001; Table 1). Similarly, mean 8-week postoperative DJD score was significantly higher for dogs in which 3.5-mm plates were used, compared with mean scores for dogs in which 2.0-mm plates were used (P = 0.001) and dogs in which 2.7-mm plates were used (P = 0.001). However, mean scores were not significantly different between dogs in which 2.0-mm plates were used and dogs in which 2.7-mm plates were used.

When the 30 factors used to determine DJD score were examined separately for changes between preoperative and postoperative values (Table 2), the largest mean increases were identified for medial soft tissue thickening, cranial apical patellar enthesiopathy, subchondral sclerosis of the proximomedial aspect of the tibia, subchondral sclerosis of the proximolateral aspect of the tibia, periarticular osteophytes of the medial femoral condyle, periarticular osteophytes of the cranioproximal aspect of the tibia, and periarticular osteophytes of the caudoproximal aspect of the tibia. The largest mean decreases were identified for tibial condylar remodeling and femoral supratrochlear lysis.

An inverse relationship was found between preoperative DJD score and the difference between postoperative and preoperative DJD scores (Spearman's rho, −0.74; P < 0.001; Figure 1). In particular, 50 of the 115 (43.5%) dogs with preoperative DJD scores ≥ 15 had lower DJD scores 8 weeks after surgery, whereas only 3 of the 118 (2.5%) dogs with preoperative DJD scores < 15 had lower DJD scores 8 weeks after surgery.

View Full Size

Scatterplot of preoperative DJD score versus the difference between 8-week postoperative and preoperative scores in 233 dogs with CCL rupture that underwentTPLO.The solid line represents the regression line.

Citation: Journal of the American Veterinary Medical Association 230, 11; 10.2460/javma.230.11.1674

• Download Figure
• Download figure as PowerPoint slide

Overall, 53 of the 233 (22.7%) dogs had a lower DJD score 8 weeks after surgery than they had had before surgery. In 6 (2.6%) of the dogs, the decrease ranged from −8.5 to −19.5. To verify that these decreases were not a result of errors, radiographs for the 4 dogs with the largest decreases in DJD score (−10.5, −13.0, −17.0, and −19.5) were reexamined, and in all 4, results were verified. Examination of preoperative radiographs for these dogs revealed sharply detailed, radiodense, osteoarthritic changes, whereas examination of corresponding postoperative radiographs revealed that extensive remodeling of these lesions had occurred in the 8 weeks after surgery, with the result that lesions were less sharply detailed and less radiodense.

Mean TSA 8 weeks after surgery (7.8° ± 3.7°) was significantly (P < 0.001) higher than mean TSA immediately after surgery (6.4° ± 2.9°). Eight-week and immediate postoperative TSAs were positively correlated (r = 0.71; P < 0.001). Preoperative and immediate postoperative TSAs did not vary significantly among groups when dogs were grouped on the basis of plate size. However, mean TSA 8 weeks after surgery was significantly (P = 0.047) higher in dogs in which 2.0-mm plates were used (11.3° ± 3.9°) than in dogs in which 2.7-mm plates were used (7.4° ± 4.2°) and dogs in which 3.5-mm plates were used (7.8° ± 3.4°).

Analysis of data for the second stifle joints in the 78 dogs that underwent bilateral surgery yielded similar results.

Discussion

Results of the present study suggest that there was a small, but measurable, increase in the severity of radiographic changes attributed to osteoarthritis in the stifle joints of dogs that underwent TPLO because of CCL rupture. Specifically, we found a small, but significant, increase in DJD scores when radiographs obtained 8 weeks after surgery were compared with radiographs obtained immediately before surgery. Unexpectedly, we also found a small, but significant, increase in TSA when radiographs obtained 8 weeks after surgery were compared with radiographs obtained immediately postoperatively.

The mean increase in DJD scores in the present study was small, making it difficult to determine its clinical importance. Importantly, however, this change was identified after only 8 weeks, and it seems likely that even greater changes would have been identified if dogs had been followed up for a longer period. In support of this are findings of a recent study¹⁴ that involved evaluation of DJD scores in dogs with CCL rupture treated by means of extracapsular surgery or TPLO. In that study, an increase in DJD scores was found for dogs that underwent TPLO when radiographic evaluations were done at least 12 months after surgery. Similarly, another study¹⁵ found that the DJD score increased in dogs that underwent TPLO regardless of whether a limited or open arthrotomy was performed.

In the present study, we were able to identify significant increases in scores for several of the individual factors used to calculate DJD score. The significant increase in medial soft tissue thickening can be attributed to the medial approach and limited arthrotomy that were used. The increase in severity of cranial apical patellar enthesiopathy may have been a result of increased strain on the patellar tendon following TPLO. Increases in the severity of subchondral sclerosis of the proximomedial and proximolateral aspects of the tibia could have been a result of continued progression of osteoarthritis or a result of rotation of the proximal segment of the tibia, with the result that this portion of the bone was more perpendicular to the radiographic plane, giving the subchondral bone a false appearance of increased opacity. The increase in severity of periarticular osteophytes on the cranioproximal aspect of the tibia may have been a result of healing of the osteotomy or continued progression of osteoarthritis. Similarly, the increase in severity of periarticular osteophytes on the caudoproximal aspect of the tibia may have been a result of continued progression of osteoarthritis or a reaction to the caudal arthrotomy required for meniscal release.

In the present study, we found an inverse relationship between preoperative DJD score and the difference between postoperative and preoperative DJD scores. This means that in general, dogs with lower preoperative DJD scores had larger increases in DJD scores 8 weeks after surgery than did dogs with higher preoperative DJD scores. In part, this may represent nothing more than regression toward the mean. Alternatively, dogs with severe preexisting osteoarthritis may have less potential for osteoarthritis to worsen, compared with dogs with only minor preexisting osteoarthritis.¹⁵

Surprisingly, 53 of 233 (22.7%) dogs in the present study had lower DJD scores 8 weeks after surgery than they had had before surgery. Small decreases in DJD might be attributable to measurement error, but a review of radiographs for the 4 dogs with the largest decreases in the present study suggested that these differences were real and could be attributed to extensive remodeling after surgery. Thus, it appears that in some dogs, postoperative remodeling can substantially decrease the DJD score. Further research with even larger numbers of dogs is needed to determine the clinical importance of this phenomenon and identify patient and surgical characteristics that might account for it.

The finding that preoperative and postoperative DJD scores were significantly higher in dogs in which 3.5-mm plates were used than in dogs in which smaller plates were used suggests that larger dogs have more radiographic changes attributable to osteoarthritis before and 8 weeks after surgery.

The scoring system used in the present study had some important limitations, particularly when used to assess dogs that had undergone TPLO. In particular, the information needed to assess some of the factors used to calculate the DJD score may not have been evident after application of a TPLO plate. Plates may, for instance, obscure periarticular osteophytes involving the proximomedial portion of the tibia, and postoperative remodeling may falsely elevate scores assigned to this factor. Similarly, subchondral sclerosis of the proximomedial aspect of the tibia may be enhanced or decreased, depending on the tibia's remodeling response.

In addition, it is not clear to what extent an increase in DJD score is related to an increase in clinical signs of osteoarthritis. Factors used to calculate the DJD score were focused on radiographic changes in the region of the stifle joint that radiologists and surgeons have attributed to osteoarthritis. But it is not clear that a dog with a higher DJD score will necessarily have more severe clinical signs of osteoarthritis than a dog with a lower score. Intuitively, we believe that an increase in DJD score correlates with worsening of osteoarthritis, but it is unknown at this time how much of a change in DJD score is necessary to identify a corresponding change in clinical signs of osteoarthritis.

In the present study, we also found a small, but significant, increase in TSA when radiographs obtained immediately after surgery were compared with radiographs obtained 8 weeks later. Similarly, a recently published study¹⁶ also found a small, but significant, increase in TSA 4 to 8 weeks after surgery. The increase in TSA may be due to changes in the position of the anatomic landmarks that are used as a result of remodeling of the tibial plateau, as we found an increase in the severity of osteophytes on the cranioproximal and caudoproximal aspects of the tibia. Alternatively, the increase in TSA may have been due to movement at the osteotomy site.¹⁶ A similar phenomenon was identified in a previous study¹⁷ of antebrachial and crural fractures stabilized with a circular external fixator in which movement at the fracture line resulted in a change in limb angulation even though fracture stabilization was sufficient to allow for fracture healing.

Mean TSA 8 weeks after surgery was significantly higher in dogs in the present study in which 2.0-mm plates had been used than in dogs in which 2.7- or 3.5mm plates had been used. This difference, however, may have been a result of the small number of dogs in which 2.0-mm plates were used and the fact that in 2 of the dogs in this group, the proximal segment rocked backwards because of interference from proximal osteophytes. This resulted in widening of the cranial aspect of the osteotomy, substantially increasing the TSA.

Importantly, it is not known whether the measured increase in TSA is of any clinical importance. A previous study^b of peak vertical force and vertical impulse in dogs that had undergone TPLO 4 months earlier did not find any significant relationships between these factors and TSA. In addition, other investigators have found intraobserver variability of 1.5° and 1.7° in measurements of TSA.^18,19 Given the small magnitude of the mean increase in TSA in the present study, therefore, it might appear that this change was not clinically important. However, the follow-up period in the present study was short, and it is possible that small changes could have important clinical consequences over longer periods of time.