Postoperative lateral-wall status and fixation failure after cephalomedullary nailing of extracapsular hip fractures: an ambispective cohort study

Background Extracapsular hip fractures cause substantial morbidity in older adults, and cephalomedullary nailing is widely used for fixation. Although lateral-wall disruption is an established concern in extramedullary fixation, the prognostic relevance of postoperative lateral-wall status after cephalomedullary nailing remains uncertain. We evaluated its association with fixation failure among patients who achieved satisfactory reduction. Methods An ambispective cohort of patients with acute extracapsular hip fractures treated with cephalomedullary nailing at Beijing Hospital (January 2010–October 2025) was analyzed. Only patients who achieved excellent or acceptable reduction were included. Postoperative lateral-wall status was classified using preoperative lateral-wall thickness and fracture morphology together with any new intraoperative lateral-wall fracture identified on postoperative imaging. Radiographic fixation failure was the primary endpoint; follow-up was at least 6 months unless failure was confirmed earlier. Failure-free survival and the association between postoperative lateral-wall status and fixation failure were evaluated using Kaplan–Meier analysis and prespecified parsimonious Firth penalized logistic regression models. Results Among 508 patients, 33 (6.5%) experienced fixation failure: 16 of 249 (6.4%) with complete and 17 of 259 (6.6%) with incomplete postoperative lateral-wall status. The log-rank test yielded p = 0.778. In the Firth model adjusted for age and tip–apex distance, no statistically significant association was detected between postoperative lateral-wall status and fixation failure (incomplete vs. complete: adjusted odds ratio, 1.04; 95% CI: 0.52–2.10; p = 0.915). Conclusion Among patients with AO/OTA 31-A extracapsular hip fractures who achieved satisfactory reduction, we did not detect a statistically significant association between postoperative lateral-wall status and fixation failure after cephalomedullary nailing. Because the analysis was restricted to satisfactorily reduced fractures and the number of failure events was limited, these findings should not be interpreted as evidence that lateral-wall morphology is unimportant in the broader fracture population.