Comparative Effectiveness of Total Hip Arthroplasty and Hemiarthroplasty for Femoral Neck Fracture

Abstract

Background: The optimal treatment of older patients with a displaced femoral neck fracture remains a controversial topic. This study aimed to compare clinical outcomes across a matched group of patients with a femoral neck fracture treated with either hemiarthroplasty or total hip arthroplasty (THA).Methods: Routinely collected health-care databases were linked to create a population-based cohort of 49,597 patients ‡60 years old from Ontario, Canada, who underwent hemiarthroplasty or THA for a femoral neck fracture between 2002 and 2017. A propensity-score-matched cohort was created using relevant and available predictors of treatment assignment and outcomes of interest. Clinical outcomes consisting of hip dislocation, revision surgery, hospital readmission, and death were compared in the matched cohort using survival analysis.Results: Over 99% of THA patients (4,612) were adequately matched 1:1 to hemiarthroplasty patients (total matched cohort = 9,224). Patients treated with THA were at higher risk for hip dislocation at 30 days and 1 and 2 years postoperatively (2-year risk, 1.8% for THA versus 0.8% for hemiarthroplasty; p < 0.001). There was no difference in the short-term (30-day) or long-term (up to 10-year) risk of revision surgery between treatment groups. There was no significant difference in the risk of 30-day hospital readmission between groups. The risk of death at 1 year and 2 years postoperatively was lower for patients treated with THA.Conclusions: For patients with a hip fracture, shared decision-making should involve discussion of the potential higher risk of short-term hip dislocation after THA compared with hemiarthroplasty. The risk of revision surgery was similar between treatment groups at up to 10 years of follow-up.

Hemiarthroplasty Versus Total Hip Arthroplasty for Femoral Neck Fracture in Elderly Patients

“…These risks of dislocation are consistent with data from the more recent RCTs showing rates of <5%. Similarly, 2 population-based studies from Canada found dislocation rates of <2% for THA and HA in the treatment of femoral neck fracture 18,39 . Both of these studies demonstrated that THA was associated with an increased risk of dislocation despite the low rates-similar to our own findings 18,39 .…”

“…Similarly, 2 population-based studies from Canada found dislocation rates of <2% for THA and HA in the treatment of femoral neck fracture 18,39 . Both of these studies demonstrated that THA was associated with an increased risk of dislocation despite the low rates-similar to our own findings 18,39 . In contrast, a population-based analysis from France, in which dual-mobility implants were utilized in 18% of the procedures in the THA group, demonstrated that THA was associated with a reduced risk of dislocation, despite having found a dislocation rate of 5.9% in the THA group 19 .…”

“…following THA. A population-based analysis from Canada found no difference in revision rates between the procedures at early or long-term follow-up 18 .…”

“…Meta-analyses of RCTs have shown a lower revision risk at long-term follow-up 31,32 but higher rates of early revision 31 following THA. A population-based analysis from Canada found no difference in revision rates between the procedures at early or long-term follow-up 18 .…”

“…To detect differences in these rare events, an analysis of large data sets may be helpful. Several recent population-based analyses from outside of the U.S. have investigated the risk of these early adverse events, with conflicting results 18,19 . To our knowledge, no studies of nationally representative data sets that correct for selection bias have been performed on THA versus HA outcomes following femoral neck fractures in the U.S.…”