Impact of oral corticosteroid use for idiopathic osteonecrosis of the femoral head: a nationwide multicenter case-control study in Japan

Abstract

BackgroundPrevious studies on corticosteroid use and idiopathic osteonecrosis of the femoral head (ONF) have been conducted involving patients with systemic lupus erythematosus and renal transplantation, among other disorders, which make it possible only to compare high- and low-dose corticosteroids. The main objective here was to estimate the impact of oral corticosteroid use for idiopathic ONF compared to nonuse of corticosteroids. In addition, we investigated the relations between underlying medical conditions and idiopathic ONF. MethodsA nationwide multicenter case-control study was conducted in Japan. Cases were defined as patients who visited the cooperative orthopedic department after January 2002 and received a diagnosis of idiopathic ONF after the visit. For each case, up to five matched controls were selected from among patients without idiopathic ONF who consulted the same orthopedic department after the visit of the identified case. The matching conditions were sex, age (5-year age groups), and ethnicity. The subjects’ demographic and anthropometric data and histories of medications, diseases, alcohol consumption, and smoking were collected by a structured self-administered questionnaire. A conditional logistic regression model was used to obtain the odds ratios (ORs) and 95% confidence intervals (CIs). ResultsA total of 73 cases and 250 matched controls completed the questionnaire. After adjustment for the potential confounders, the multivariate OR of oral corticosteroid use compared to nonuse was 20.3 (95% CI 6.73–61.5). Furthermore, a positive association of liver disease with idiopathic ONF was observed with marginal significance (multivariate OR 3.53, 95% CI 0.88–14.1) regardless of oral corticosteroid use, alcohol consumption, and smoking. ConclusionsThe current findings first revealed the impact of oral corticosteroid use for idiopathic ONF compared to nonuse. However, because the regimen of corticosteroid therapy (dose and duration) could not be assessed, further study is required. An independent role of liver disease was also suggested.