Abstract
We aimed to examine outcomes between displaced femoral neck fracture (FNF) patients managed with total hip arthroplasty (THA) or hemi-arthroplasty (HA) via the anterolateral vs. posterior approach.
We used data from the HEALTH trial (1,441 patients aged ≥50 with displaced FNFs randomized to THA vs. HA). We calculated each patient's propensity to undergo arthroplasty via the posterior approach, and matched them to 1 control (anterolateral approach) based on age (±5 years), and propensity score. We used Chi-Square/Fisher-Exact tests to compare dichotomous outcomes, and repeated measures ANOVA to examine differences in patient-reported outcomes (via the WOMAC subscores) from baseline to one-year postoperative. We used logistic regression to identify independent predictors of reoperation for instability in the posterior group.
We identified 1,306 patients for this sub-analysis, 876 (67.1%) who received arthroplasty via an anterolateral approach, and 430 (32.9%) a posterior approach. The unadjusted rate of reoperation was significantly higher in the posterior group (10.7% vs. 7.1%).
Following propensity score matching, we retained 790 patients (395 per group), with no between-group differences in patient, fracture, or implant characteristics. The matched cohort had a higher rate of comorbidities, and were less likely to be employed vs. the unmatched cohort. The rate of treatment for dislocation remained higher in the posterior group (6.1% vs. 2.0%) following matching.
Repeated measures ANOVA revealed significantly better WOMAC pain, stiffness, function, and total scores in the posterior group. Between-group differences at 12-months were: pain - 0.59 (0.03–1.15); stiffness - 0.62 (0.35–0.87); function - 2.99 (0.12–5.86); total - 3.90 (0.24–7.56). We identified THA (vs. HA, odds ratio 2.05 [1.05–4.01]) as the only independent predictor of treatment for dislocation in the posterior group.
Our analyses revealed that compromised patients with displaced FNFs who undergo arthroplasty via the posterior approach may report better symptoms at one-year vs. the anterolateral approach, despite a higher odds of reoperation for instability.
