Outcomes Vary by Surgical Approach for Hemiarthroplasty after Low-energy Displaced Femoral Neck Fracture: A Study of the Arthroplasty for Hip Fracture Consortium

OBJECTIVES: To compare hemiarthroplasty (HA) outcomes for low-energy femoral neck fracture (FNF) among a direct anterior approach (DAA), direct lateral approach (DLA), and posterior approach (PA). METHODS: Design : Retrospective review Setting: Nine Level-1 Tertiary Academic Centers Patient Selection Criteria: All patients from January 2010 through December 2019 undergoing HA for a low-energy FNF (AO/OTA 31B1-3). Exclusion criteria were pathological fractures, concomitant acetabular fractures, and high-energy mechanisms. Outcome Measure and Comparisons: Multivariable regression models were constructed for outcomes among DAA and DLA compared to PA including mortality, revision, and infection. It was hypothesized anterior approaches would be independently associated with fewer postoperative complications including infection, dislocation, and mortality. RESULTS: A total of 956 consecutive patients with FNF treated with HA using DAA ( n =71; 7.4%; mean (SD) age 82.6 (8.5) years), DLA ( n =252; 26.4%; mean (SD) age 81.3 (10.8) years), or PA ( n =633; 66.2%; mean (SD) age 79.2 (10.9) years) were included. There were no differences in sex among groups ( p =0.83). The DAA was independently associated with higher periprosthetic joint infection (PJI) rates at 90 days (OR, 7.295; 95% CI, 2.35 to 22.61; p <0.001) and one-year (OR, 5.769; 95% CI, 1.93 to 17.22; p =0.002), as was the DLA at 90 days (OR, 2.952; 95% CI, 1.29 to 6.74; p =0.010) and one-year (OR, 3.047; 95% CI, 1.43 to 6.50; p =0.004). The DLA was associated with lower 90-day dislocation (OR, 0.129; 95% CI, 0.03 to 0.60; p= 0.009), one-year dislocation (OR, 0.175; 95% CI, 0.05 to 0.63; p =0.008), and greater independent ambulation at discharge (OR, 3.273; 95% CI, 2.00 to 5.37; p <0.001). CONCLUSIONS: Among hemiarthroplasty for low-energy femoral neck fracture, the lateral approach was associated with decreased dislocation postoperatively and greater independent ambulation at discharge, but with decreased ambulatory distance compared to a posterior approach. Furthermore, the anterior and lateral approaches were associated with increased infection versus a posterior approach. LEVEL OF EVIDENCE: Prognostic Level III. See Instructions for Authors for a complete description of levels of evidence.