Comparison of dual mobility total hip arthroplasty and bipolar arthroplasty for femoral neck fractures: A retrospective case-control study of 199 hips.

BACKGROUND: The choice between performing total hip arthroplasty (THA) or hemiarthroplasty (HA) is not straightforward in older patients with femoral neck fracture, particularly when co-morbidities are factored in. This led us to carry out a case-control study to determine (1) the rate of mechanical complications for these two types of implants, and (2) the rate of medical complications and mortality.

HYPOTHESIS: THA with dual mobility cup (DM) will result in fewer mechanical complications than HA.

PATIENTS AND METHODS: This was a single-center, retrospective case-control study. Between 2010 and 2015, all patients with a femoral neck fracture treated by HA or DM THA were included. The primary outcome was the occurrence of any type of surgical complication. The Charlson Co-morbidity Index (CCI) and the independence during Activities of Daily Living (ADL) score were calculated for every patient. Two subgroups of patients were made based on whether they met frailty criteria. The effect of covariates on 1-year mortality was controlled using Cox's proportional hazards regression model.

RESULTS: The cohort consisted of 101 HA and 98 THA procedures in 193 patients (139 women, 54 men) with a mean age of 80.6years (range, 76-101). The mean follow-up was 24.2months (range, 0-83) with a median of 14.5months. Fifteen of the HA hips (15%) had surgical complication, of which 10 were posterior dislocations (10%). Ten patients in the HA cohort had a serious medical complication (10%). Ten of the THA hips (10%) had a mechanical complication, including three posterior dislocations (3%) and four infections (4%). Nine patients in the THA cohort had a medical complication (9%). There were significantly fewer posterior dislocations in the THA hips (p = 0.05). In the subgroup analysis, the 117 patients (58%) who met the frailty criteria had a significantly lower dislocation rate after undergoing THA (p = 0.048). After adjusting on age, ADL and CCI score, the dislocation rate no longer differed significantly between the two groups (p = 0.1). The dislocation rate was lower in the THA hips only in the "frail" patients (Odds ratio = 0.137, 95% CI: [0.003-0.97] (p = 0.04)). There was no difference in the dislocation rate in the "non-frail" patients. The overall 1-year mortality was 85% [95% CI: 78-94%]. It was 78% [95% CI: 69-86%] for the HA hips and 88% [95% CI: 82-95%] for the THA hips (p = 0.01). After factoring in the impact of age, CCI and ADL, the differences in the 1-year mortality between HA and THA were no longer present (p = 0.42). Thus, there is no increased risk of mortality in THA patients.

DISCUSSION: When the CCI and independence level are taken into consideration, the frailest patients can undergo DM THA to reduce the dislocation risk, without increasing the mortality rate at 1year. Patients who are not frail will benefit equally from undergoing HA or THA.

LEVEL OF EVIDENCE: III, case-control study.