Long-term risk of reoperation after modular hemiarthroplasty: Any differences between uni- or bipolar design?

BACKGROUND: It is unclear whether unipolar (UHA) or bipolar (BHA) hemiarthroplasty should be the preferred treatment of femoral neck fracture (FNF). AIM: We investigated the reoperation rate at 13 years post-fracture after BHA and UHA as treatment of FNF, including a subgroup analysis of individuals who survived 5 years or more, and described the reasons for reoperation after BHA and UHA respectively. METHODS: In an observational cohort study on prospectively collected national register data, 16,216 BHA and 22,186 UHA were available for matching. A propensity score for treatment with bipolar HA was estimated using logistic regression. Matching was done using the 1:1 nearest neighbor matching without replacement. Of the 16,216 BHA patients, 12,280 were matched to a UHA control. A subgroup analysis based on the matched sample excluded individuals who died within 5 years and comprised 3,637 individuals with BHA and 3,537 with UHA. Kaplan-Meier survival analysis was used. RESULTS: In the Kaplan-Meier analysis, 92% of the BHA group was free from reoperation at 13 years (95% CI 0.91–0.93), compared to 92% in the UHA group (CI 0.89–0.94). BHA was associated with more reoperations until 3 years. Reoperation due to infection was most common after BHA, n = 212 (1.7%) compared to n = 141 (1.1%) after UHA. Dislocation led to reoperation in 192 of the BHA cases (1.6%) and in 157 of the UHA cases (1.3%). Acetabular erosion/pain occurred in 0.1% and 0.4%. Amongst those surviving ≥ 5 years, 93% of the BHA group was free from reoperation (CI 0.92–0.94) at 13 years, 92% after UHA (CI 0.90–0.94). BHA had more reoperations during the 1st year only. The causes for reoperations showed similar rates except for acetabular erosion/pain. Here the BHA group had 2 cases (0.1%), the UHA had 39 (1.1%). CONCLUSION: With a modular hemiarthroplasty relatively few patients need a reoperation. During the first years, there is a higher reoperation rate after BHA compared to UHA. Thereafter, no differences are seen. In patients who survive ≥ 5 years after the fracture there are more reoperations due to acetabular erosion after UHA, but crude numbers are extremely low, and the total reoperation rate is not affected.