LAUSR

PURPOSE Influence of surgical quality (as evaluated in the post-surgical radiographic control) on mortality, complications and recovery of walking ability in patients older than 64 years with hip fracture. PATIENTS AND METHODS Retrospective observational study of a single-center, consecutive cohort of 1425 patients operated from December/2012 to March/2018. Average age was 85.66±7.13 (65-108) years and 966 (67.86%) were female. Extracapsular fractures (811 cases, 56.91%) were fixed with trochanteric nails (Affixus-DePuy / PFNA-Synthes) (795 cases) or sliding-hip-screw-plates (Biomet) (16 patients). Intracapsular fractures (614, 43.09%) were treated with cannulated screws (Smith-Nephew) (134 cases) or with arthroplasty (472 bipolar and 8 total hip, Furlong-JRI or Exeter-Stryker). Postoperative radiographs were used to determine surgical quality (reduction, compression, screw position, tip-apex, stem position, Voss, intraoperative fracture). Patients were followed clinically and radiographically in outpatient clinic after 1, 3, 6, and 12 months. STATISTICAL ANALYSIS Bivariate analysis (Pearson, Fischer, Mann-Whitney, Wilcoxon) was applied to study statistically significant relations, and Relative Risks (RR) were calculated. RESULTS 196 patients (13.75%) died along follow-up, and poor reduction was found to be a risk factor for mortality both in trochanteric nail group (p=0.0495) (RR 1.510, 1.01-2.26) and in cannulated screw group (p=0.0023) (RR 6.48, 2.40-17.53); this is a previously non-published risk factor. Surgical complications included 33 infections (2.36%), 12 non-unions (1.27%), 12 cut-outs (1.47%), 7 broken nails (0.86%), 21 hemiarthroplasty dislocations (4.37%), and 8 ischemic necrosis of the femoral head (5.97%). The significant risk factors were: poor reduction for trochanteric nail breakage (p=0.041) (RR 4.47, 1.13-17.67), poor reduction for non-union in cannulated screws (p=0.035) (RR 10.91, 1.70-70.09), as well as "tip-apex distance" over 25mm in trochanteric fractures for "cut-out" type of fixation failure (p=0.0159) (RR 3.84, 1.19-12.40). Previous walking ability was restored in 564 patients (39.58%) and worsened in 581 (40.77%), but statistical relationships were not found, although follow-up data were inadequate in 212 cases (14.88%). CONCLUSIONS Appropriate reduction of hip fractures is a significant step to reduce mortality, nail breakage and non-union. Adequate position of sliding screw (avoiding tip-apex distances over 25mm) is important to reduce "cut-out" in trochanteric fractures.