Avascular necrosis (AVN) is a pathological process associated with many medical conditions, including human immunodeficiency virus (HIV) infection. Whether or not the use of protease inhibitors (PIs) confers additional risk for AVN to HIV-infected patients is controversial. Previous epidemiological studies showed an increased risk of AVN among PI users, but these studies did not have enough power to achieve statistical significance. A meta-analysis of case-control studies reporting the odds ratios (ORs) of AVN among HIV-infected patients who were exposed to PIs compared with non-exposed patients was conducted. Pooled ORs and 95% confidence intervals (CIs) were calculated using a fixed-effect Mantel-Haenszel analysis. Four case-control studies were identified and included for data analysis. The meta-analysis demonstrated an increased odds of AVN in participants exposed to PIs, with an OR of 2.09 (95% CI 1.01-4.31; P=0.05). The statistical heterogeneity of this meta-analysis was determined not to be important, with an I(2) of 0%. The meta-analysis revealed a statistically significant increased odds of AVN among PI-exposed, HIV-infected patients. Physician should be aware of this association as it may help guide potential therapeutic options, particularly for patients with other classic risk factors for AVN.