Inferior Screw Referenced Calcar Tip Apex Distance as the Most Accurate Predictor of Mechanical Cut Out in Dual-Screw...

Background and Objectives: Screw cut-out is the most common mechanical complication after intertrochanteric fracture fixation with proximal femoral nails (PFNs). While the traditional tip-apex distance (TAD) is widely used, the calcar-referenced TAD (CalTAD) may better represent inferomedial cortical support. This study aimed to identify radiographic predictors of cut-out in dual-screw PFN fixations and establish a clinically relevant threshold for inferior-screw-based CalTAD. Materials and Methods: A retrospective cohort of patients treated with a dual cephalic screw PFN between 2017 and 2024 was analyzed. The implant uses two equal-diameter screws. Radiographic parameters included TAD, inferior-screw CalTAD, reduction quality, lateral wall thickness (LWT), collodiaphyseal angle (CDA), and Cleveland zone positioning. Logistic regression analyses were used to identify independent predictors of mechanical failure. Results: Both TAD and CalTAD values were significantly higher in patients who experienced screw cut-out. ROC analysis identified an inferior-screw-referenced CalTAD cutoff with strong predictive accuracy (AUC = 0.84). Optimal screw positioning, particularly avoiding superior placement on AP radiographs, was associated with reduced cut-out risk, while anterior positioning on the lateral view demonstrated only a borderline effect. Reduction quality showed borderline significance in univariate testing but remained independently predictive in multivariate modeling, while LWT and CDA were not significantly different between groups. Conclusions: Ensuring the inferior lag screw is positioned close to the calcar and achieving a low CalTAD, together with proper Cleveland zone alignment, appear to be key technical goals for minimizing mechanical cut-out in dual-screw PFN fixations. These findings support the use of inferior-screw-referenced CalTAD as a reliable and reproducible parameter for surgical optimization.