Introduction: Osteonecrosis of the femoral head (ONFH) has a close association with corticosteroid therapy. As corticosteroids are accepted to be metabolized mainly by CYP3A4 in the liver, low constitutive levels of the enzyme might lead to an excessive response to corticosteroids and lead to adverse events including bone necrosis. This clinical study was designed to elucidate this hypothesis and to present potential modalities to avoid corticosteroid-associated ONFH by tailoring the steroid dose according to individual metabolic capacities of corticosteroid.

Materials and Methods: Twenty-two steroid-associated ONFH patients, 27 alcohol-related ONFH patients, and 65 general population controls were enrolled in this study. To estimate functional level of hepatic CYP3A4 level, a midazolam (MDZ) clearance test was carried out in respective subjects. The results from the tests were compared between those groups.

Results: The distribution profile of the MDZ clearance in steroid-associated ONFH patients were shifted to the left, indicating lower hepatic CYP3A4 activity in those patients when compared with the general population. By using an unconditional logistic regression model, patients with low (< 9.7) MDZ clearance due to low hepatic CYP3A4 activity were at 9.5 times greater risk for corticosteroid-induced ONFH compared with those with high (9.7+) MDZ clearance (OR 9.5 [95% CI 2.79–32.2], p< 0.001). The hepatic CYP3A4 activity was not associated with prevalence of alcohol-associated ONFH.

Discussion: A significantly low constitutive hepatic CYP3A4 function in corticosteroid-associated ONFH patients was found. The corticosteroid-associated ONFH might result from excessive responsiveness to corticosteroids in those patients due to prolonged exposure of bone to high levels of corticosteroids because of low functional level of the steroid metabolizing enzymes. The steroid-associated ONFH might be avoided by tailoring the corticosteroid dose in accordance with the functional level of hepatic CYP3A4.

Introduction: Based on epidemiological results, steroid (glucocorticoid) hormone is accepted as a major causative agent of osteonecrosis, though its pathomechanism is not elucidated. However, not all patients who receive high doses of steroid develop osteonecrosis. This fact suggests risk factor(s) for steroid-induced osteonecrosis. In order to identify such risk factors, the association of CYP3A6 (a major enzyme metabolizing glucocorticoid in rabbit) level with the incidence and extent of osteonecrosis in a rabbit model was determined.

Materials and Methods: In this rabbit model, the CYP3A6 level was modulated by an inhibitory (Itraconazole, 150mg p.o. twice a day for 3 weeks) or inducing agent (rifampicin, 100mg/kg i.p. in first 3 days). Three weeks after modulation of CYP3A6, steroid-induced osteonecrosis was generated by i.m. injection of methyl-prednisolone (20mg/kg BW). Three weeks later, the animals were sacrificed and bilateral femurs were excised and examined histologically for bone and marrow necrosis.

Results: In control animals without modulation of CYP3A6, focal and/or extensive necrosis was noted in 5 of 7 animals. In Itraconazole-treated animals, all of 5 animals revealed extensive necrosis in femoral bone marrow. In rifampicin- treated animals, incidence of necrosis was similar to that of controls but necrotic foci were significantly smaller than those in controls.

Discussion: These experimental results indicated that low level of steroid-metabolizing enzyme CYP3A6 (CYP3A4 in humans) at the time of steroid treatment might be a risk factor for extension of steroid-induced bone necrosis, and that induction of CYP3A6 might prevent steroid-induced bone necrosis.

The benefit of mobile-bearing mechanism in total knee arthroplasty (TKA) has been controversial. The aim of this paper is to analyse the rotation of polyethylene (PE) and its effect on the range of motion (ROM) in mobile-bearing, posterior stabilised TKA (LPS-Flex™, Zimmer Co. Ltd.).

Thirty-four consecutive PS-Flex™ TKAs (28 patients) were analysed. Three tantalum markers were inserted in the PE and the tibia (one for medial side and two for lateral side; total six markers). The rotation of the PE and the tibia was analysed in fluoroscopically-controlled radiograph taken at one years in full extension, 90 degrees flexion and passive maximum flexion. The markers in the tibia were identifiable in 19 knees and the analysis was based on these knees.

The tibia rotated internally relative to the femur by 7.1± 5.2 degrees (mean ± SD, range −1 to 20 degrees). The amount of tibial rotation has no correlation to ROM. The rotation of the PE relative to the femur was unpredictable showing three knees with external rotation and four knees without rotation (Average; 4.0 ± 4.5 degrees internal rotation). The rotation of the PE on the tibia was 4.2± 5.2 degrees and seven knees (37 %) showed no rotation and 12 knees (63 %) showed less than 5 degrees rotation. There was a positive correlation between the amount of PE rotation on the tibia and ROM, which approached to statistical significance (p = 0.0684).

This study has demonstrated that the rotation of the PE on the tibial tray is generally small (< 5 degrees). Because not tibial internal rotation but PE movement on the tibia correlated to ROM, the essential benefit of the mobile-bearing mechanism could be to compensate the rotational mismatch between the components rather than to reproduce normal knee kinematics.