Preoperative morphometric differences in the distal femur are based on skeletal size in Japanese patients undergoing total knee arthroplasty

The objectives of this study were to measure the morphometric parameters of preoperative distal femurs to determine the differences by diagnosis and gender after accounting for skeletal size.

One-hundred and seventy-nine Japanese patients who underwent total knee arthroplasty (TKA) (25 males and 154 females) were assessed. The anteroposterior length (AP), mediolateral width (ML), aspect ratio (AR), surgical epicondylar axis (SEA) to posterior condylar axis (PCA) angle, and Whiteside to SEA angle were measured on preoperative computed tomography scans. The AP/ML, AR/ML, SEA/PCA, and Whiteside/PCA relationships were evaluated and compared by patient diagnosis and gender. The results were also compared with the sizes of 10 currently available TKA implants in Japan.

The mean AP, ML, AR, SEA/PCA angle, and Whiteside/PCA angle were 58.8 mm, 64.7 mm, 0.91, external rotation (ER) 3.5°, and ER 1.6°, respectively. AP and AR each were significantly correlated with ML (p < 0.001). AP, ML, and AR were not significantly different between patients diagnosed with osteoarthritis and rheumatoid arthritis. AP/ML and AR/ML were significantly correlated within each diagnosis (p < 0.001), but the analysis of covariance showed no significant differences between the diagnoses. AP and ML were significantly longer (p < 0.001) in males (63.6, 72.7 mm) than in females (58.1, 63.4 mm), while AR was smaller in males (0.88 vs. 0.92), with significant correlations for AP/ML (male: p < 0.010, female: p < 0.001) and AR/ML (male: p = 0.002, female: p < 0.001) in each gender. However, the analysis of covariance showed no significant differences between gender in the AP/ML and AR/MR correlations. The AP/ML ratio of our data was similar to the size variations of the 10 TKA implants, but the AR/ML ratio was quite different from almost all the implants.

No differences in preoperative femur morphometry were found between patients with different diagnoses, but the gender difference in AR was related to the difference in skeletal size between males and females.

Case series with no comparison groups, Level IV.