INTRODUCTION

Morphology of the distal femur helps elucidate the causes of osteoarthritis,1-3 the outcome of osteotomy for osteoarthritis,4,5 and the optimal design and procedure for total knee arthroplasty (TKA).6,7 The ratio of anteroposterior width to mediolateral width of the distal femur differs between Caucasian and Japanese populations,8,9 such that mismatch of the prosthesis may occur.10 We aimed to compare the morphology of the distal femur between Caucasian and Japanese osteoarthritic women using lateral radiographs.

MATERIALS AND METHODS

Between April 1997 and December 1998, 30 Caucasian women aged 41 to 84 (mean, 67; standard deviation [SD], 9) years and 70 Japanese women aged 54 to 86 (mean, 70; SD, 8) years who underwent TKA for osteoarthritis were randomly selected. Their femorotibial angle (FTA) was measured using fulllength anteroposterior radiographs. The severity of osteoarthritis was evaluated according to the Kellgren and Lawrence classification.11 The anatomic axis and points of the distal femurs were compared using lateral radiographs.

The anatomic axis was the line connecting the midpoints of the shaft at 6 and 12 cm from the distal end (Fig. 1). Points A and B were tangents of the medial and lateral anterior condyles, respectively. Points C and D were tangents of the medial and lateral posterior condyles, respectively. Points E and F were midpoints of lines AB and CD, respectively. Points G and H were intersections of the anatomic axis and lines perpendicular from points E and F, respectively. The length of lines EG, FH, IJ (the distance from the anatomic axis to the anterior cortex), and KL (the distance from anterior to posterior cortices of the shaft) were measured. KL was defined as the anteroposterior (AP) width of the shaft, IJ as the half metaphyseal width of the shaft, EG as the length of the anterior condyle, FH as the length of the posterior condyle, and EG plus FH as the AP width of the condyle. When TKA was performed according to the anterior reference12 and independently cut,13 the mean length of the resected anteromedial and anterolateral femoral condyle was almost the same as the value of EG minus IJ, which was therefore defined as the length of the resected condyle, and IJ plus FH as the length of the remaining condyle.

Correlations between the AP width of the condyle/shaft and height/weight were measured using the Pearson correlation coefficient (r). Both race and height influenced the morphology of the distal femur. Therefore to adjust for the influence of height on morphology, each measurement was divided by the patient's height and the ratios were compared.

Measurements were made by 2 observers. One of them made measurement 3 times. Intra-observer and inter-observer reliability were assessed using the intraclass correlation coefficient. The 2-tailed Student's t test was used for comparison of the groups. A p value of <0.05 was considered significant.

RESULTS

Caucasian women were generally taller and heavier (p<0.001) and had higher body mass indices (p=0.03) than the Japanese women (Table). 70% of the Caucasian14 and 50% of the Japanese patients were obese. 28 Caucasian and 66 Japanese women had FTA of >175°. There were 13 (1+12) grade-3 and 87 (29+58) grade-4 osteoarthritis cases. The intra-observer and inter-observer intraclass correlations were 0.96 (95% confidence interval [CI], 0.92-0.98) and 0.89 (95% CI, 0.77-0.95), respectively.

Each morphologic measurement of the distal femur was significantly greater in the Caucasian women (Table). In both groups, AP width of the condyle correlated more with height (r=0.71) than weight (r=0.57) [Fig. 2]. The correlation coefficient between AP width of the femur and height was just 0.5.

The height-adjusted ratios of the half metaphyseal femur (p<0.001), the whole condyle (p=0.03), the anterior condyle (p<0.001), and the resected condyle (p=0.004) were significantly greater in the Caucasian women (Table). However, the height-adjusted ratio of the posterior condyle was significantly smaller in Caucasian women (p=0.02). The height-adjusted ratio of the remaining condyle was not significantly different between the 2 groups (p=0.74).

DISCUSSION

The size and shape of the distal femur differ between Caucasians and Japanese.6-9 Osteoarthritis and patellofemoral pain syndrome is more common in women.15-19 The AP width divided by the mediolateral width of the femur was greater in Caucasians than Japanese, but the differences in height, weight, and body mass index were not analysed.9 In Japanese people, the width of the femur (at the articular level) and the AP length of the medial condyle correlated better with height than weight.20 In our study, height correlated better with the AP width of the condyle and therefore each measurement was divided by the patient's height to adjust for the influence of height on morphology.

In our patients of equal height, the AP and metaphyseal widths of the femur and the anterior and resected condyles were longer in Caucasian women, but the posterior condyle was longer in Japanese women. As the width of the anterior flange of most prostheses is 8 to 10 mm, the resected condyle can be shorter than the anterior flange. The resected posterior condyle is usually equal to the posterior flange, according to the independent cut theory.13 Thus in Japanese patients after TKA, the AP width of the distal femur and hence the pressure on the patellofemoral joint may be excessive.

In normal knees, full flexion is accompanied by impingement between the posterior femur and the posterior horn of the medial meniscus.21 A significant correlation was noted between operative restoration of the posterior condylar offset (the depth of the posterior condyle) and the maximal flexion in the posterior cruciate ligament-retaining TKA.22 Increased flexion in the knee joint may require a longer posterior condyle to avoid impingement of the tibia on the posterior femur.

Further studies to compare the medial and lateral condyles independently are needed to elucidate the kinematics of knee joints and the optimal design and procedure for successful TKAs in each race.

ACKNOWLEDGEMENTS

We thank Dr Hiroaki Minehara, Dr Masaki Ueno, Dr Kouji Naruse, Dr Mamoru Fujita, Dr Jun Aikawa, Dr Motoi Miyabe, and Ms Yuko Oonuki for assistance in preparing the manuscript.

© 2008 Western Pacific Orthopaedic Association Provided by ProQuest LLC. All Rights Reserved.