BackgroundWell-designed studies of complications and readmission rates in patients undergoing total hip arthroplasty (THA) with osteonecrosis are lacking. Our objective was to examine if a diagnosis of osteonecrosis was associated with complications, mortality and readmission rates after THA.MethodsWe analyzed prospectively collected data from an integrated healthcare system’s Total Joint Replacement Registry of adults with osteonecrosis vs. osteoarthritis (OA) undergoing unilateral primary THA during 2001–2012, in an observational cohort study. We examined mortality (90-day), revision (ever), deep (1 year) and superficial (30-day) surgical site infection (SSI), venous thromboembolism (VTE, 90-day), and unplanned readmission (90-day). Age, gender, race, body mass index, American Society of Anesthesiologists class, and diabetes were evaluated as confounders. We used logistic or Cox regression to calculate odds or hazard ratios (OR, HR) with 95% confidence intervals (CI).ResultsOf the 47,523 primary THA cases, 45,252 (95.2%) had OA, and 2,271 (4.8%) had osteonecrosis. Compared to the OA, patients with osteonecrosis were younger (median age 55 vs. 67 years), and were less likely to be female (42.5% vs. 58.3%) or White (59.8% vs. 77.4%). Compared to the OA, the osteonecrosis cohort had higher crude incidence of 90-day mortality (0.7% vs. 0.3%), SSI (1.2% vs. 0.8%), unplanned readmission (9.6% vs. 5.2%) and revision (3.1% vs. 2.4%). After multivariable-adjustment, patients with osteonecrosis had a higher odds/hazard of mortality (OR: 2.48; 95% CI:1.31–4.72), SSI (OR: 1.67, 95%CI:1.11–2.51), unplanned 90-day readmissions (OR: 2.20; 95% CI:1.67–2.91) and a trend towards higher revision rate 1-year post-THA (HR: 1.32; 95% CI: 0.94–1.84), than OA patients.ConclusionsCompared to OA, a diagnosis of osteonecrosis was associated with worse outcomes post-THA. A detailed preoperative discussion including the risk of complications is needed for informed consent from patients with osteonecrosis.Electronic supplementary materialThe online version of this article (doi:10.1186/s12891-016-1385-0) contains supplementary material, which is available to authorized users.