CT-defined sarcopenia as a predictor of early mortality following hip fracture: a retrospective cohort study

Abstract

Sarcopenia is increasingly recognised as a marker of physiological vulnerability. Computed tomography permits objective quantification of skeletal muscle mass, yet the prognostic significance of CT-defined sarcopenia after hip fracture remains incompletely characterised. This study aimed to determine sarcopenia prevalence in patients with hip fractures and evaluate associations with early postoperative outcomes. This single-centre retrospective cohort study included patients with hip fracture during 2015 to 2025 who had abdominal computed tomography imaging, including the third lumbar vertebra (L3), within one year before fracture. Skeletal muscle area at L3 was quantified by manual segmentation on the Picture Archiving and Communication System (PACS) according to a predefined protocol. The cohort comprised 486 patients: 250 sarcopenic and 236 non-sarcopenic. Thirty-day and ninety-day mortality were the primary outcomes, analysed using adjusted logistic regression and 90-day time-to-event analysis. Length of hospital stay and discharge destination were secondary outcomes. Sarcopenia was independently associated with increased mortality at 30 days (adjusted OR 3.88, 95% CI 1.83 to 8.22) and 90 days (adjusted OR 3.75, 95% CI 2.22 to 6.35). Cox regression similarly demonstrated increased 90-day mortality hazard (adjusted HR 3.19, 95% CI 2.01 to 5.06). No significant associations were observed between sarcopenia and length of hospital stay (median 13.27 vs. 13.25 days; p = 0.930) or discharge destination (χ2= 1.63, p = 0.202). CT-defined sarcopenia was independently associated with early mortality, supporting its potential role as an opportunistic marker of postoperative risk.