Lag screw cutout risk factors in geriatric intertrochanteric femur fractures

Lag screw cutout remains the most frequent mechanical failure after proximal femoral nailing (PFN) for geriatric intertrochanteric femur fractures and is strongly influenced by bone quality, reduction, and implant positioning. This study aimed to evaluate these domains comprehensively and to identify independent predictors of cutout.

In this retrospective cohort study, patients aged ≥ 65 years who underwent PFN (single implant design, APFN^®) for AO/OTA 31.A1-A3 intertrochanteric fractures between January 2017 and June 2025 were reviewed. Patients with follow-up < 6 months, non-APFN fixation, insufficient radiographs, cut-through, pathological fractures, major trauma, severe prefracture dependency, or secondary surgery due to infection/major complications were excluded. Demographics and Charlson comorbidity index were recorded. Preoperative and early postoperative plain AP/lateral radiographs were assessed by three blinded observers (two orthopaedic surgeons and one radiologist) who were not involved in the initial planning or surgery to prevent potential bias, by consensus. Bone quality/morphology (Singh index, Dorr type, fracture type), reduction quality (neck-shaft angle-NSA, Baumgaertner criteria, reduction pattern), and fixation quality (tip-apex distance-TAD, calcar referenced tip-apex distance-CalTAD, lag screw position/Cleveland zones) were analyzed. Receiver operating characteristic (ROC) analysis determined clinically relevant cutoffs, and binary logistic regression analysis identified independent predictors.

A total of 371 patients (mean age 78.8 ± 6.95 years; 66% female) met inclusion criteria. Cutout occurred in 48 patients (12.9%) at a mean of 6.29 ± 5.52 weeks. The cutout group demonstrated significantly poorer bone quality (Singh index 2.56 ± 0.51 vs. 3.70 ± 0.67; p < 0.001), greater varus fixation (NSA 126.6° vs. 131.7°; p < 0.001), higher TAD and CalTAD (both p < 0.001), and a higher proportion of center-anterior zone lag screw position (p < 0.001), which conferred increased risk (Odds ratio-OR 6.26; p < 0.001), whereas center-center zone was protective (OR 0.32; p < 0.001). CalTAD showed the highest discriminative performance for cutout (AUC 0.757; cutoff 26 mm; sensitivity 63.8%, specificity 75%). In logistic regression analysis, only Singh index (OR 0.009; 95% CI 0.001-0.060; p < 0.001) and NSA (OR 0.830; 95% CI 0.710-0.971; p = 0.020) remained independent predictors.

In geriatric intertrochanteric femur fractures treated with PFN, osteoporosis severity and varus fixation are the strongest independent predictors of cutout complication. Meticulous avoidance of varus reduction and anterior lag screw placement combined with rigorous assessment of bone quality may decrease cutout risk.