Psoas muscle attenuation on preoperative CT predicts 30-day complications after lateral lumbar interbody fusion

Introduction: Computed tomography (CT)-derived mean psoas muscle attenuation (P-HU) reflects myosteatosis and has been associated with frailty-related outcomes. Whether it predicts early postoperative complications after lateral lumbar interbody fusion (LLIF) remains unclear. Research question: Does preoperative P-HU predict 30-day postoperative complications, prolonged hospitalization, and approach-related hip flexion weakness after LLIF? Material and methods: We retrospectively analyzed 60 adults undergoing LLIF. Mean bilateral P-HU was measured at mid-L3 on routine preoperative CT. The primary endpoint was any 30-day postoperative complication classified according to Clavien-Dindo (CD ≥ I); clinically relevant complications (CD ≥ II) were evaluated in sensitivity analyses. Multivariable Firth logistic regression adjusted for age, ASA classification, and body mass index. Interobserver reliability was assessed in a subset of 36 patients. Results: Sixteen patients (26.7%) developed 30-day postoperative complications. P-HU was significantly lower in patients with complications than in those without (35.1 ± 6.0 vs 44.4 ± 7.3 HU; p < 0.001). Each 10-HU increase in P-HU was independently associated with lower odds of complications (OR 0.20; p = 0.002). Discrimination was high (AUC 0.87). Higher P-HU was also associated with shorter hospital stay but not with postoperative hip flexion weakness. Discussion and conclusion: Lower preoperative P-HU was independently associated with postoperative complications and prolonged hospitalization after LLIF, but not with approach-related hip flexion weakness. Opportunistic CT-based muscle assessment may represent a simple imaging biomarker for perioperative risk stratification.