CT-based skeletal muscle loss predicts long term prognosis in patients with distal cholangiocarcinoma undergone pancreaticoduodenectomy

Skeletal muscle index (SMI), as an effective indicator of nutritional status, plays an important role in the prognosis of malignancy. However, the impact of skeletal muscle changes on tumor prognosis has not been systematically elaborated. We aimed to explore the value of skeletal muscle changes in the prognosis of distal cholangiocarcinoma (DCC) patients undergone pancreaticoduodenectomy (PD). Patients who underwent PD for DCC between 2015 and 2023 were included in this study. Demographic, laboratory and follow-up information was obtained. The cross-sectional images of skeletal muscle area at the level of the third lumbar spine was obtained based on computed tomography (CT), and the SMI was calculated by skeletal muscle mass through height squared normalization. Skeletal muscle index and skeletal muscle loss (SML) were obtained before PD and three to six months after surgery. Patients were classified into two groups (High-SML and Low-SML) based on the optimal SML cut-off value. The univariate and multivariate Cox proportional hazards analysis was conducted to evaluate the influence of SML in predicting over survival (OS) and recurrence free survival (RFS) of DCC. Of the 112 patients with distal cholangiocarcinoma, 55 (49%) were diagnosed with low SMI preoperatively. The best cut-off values of SML were - 4.01% and - 5.99% for OS and RFS. In multivariate analysis, tumor size > 2.0 cm (hazard ratio (HR) = 1.90, P = 0.017), poor differentiation (hazard ratio (HR) = 2.80, P > 0.001), higher SML (SML ≤ - 4.01%) (hazard ratio (HR) = 3.60, P < 0.001), lymph metastasis (hazard ratio (HR) = 4.00, P < 0.001) and vascular invasion (hazard ratio (HR) = 2.10, P = 0.013) were independent risk factors forOS. Meanwhile, poor differentiation (hazard ratio (HR) = 1.90, P = 0.043), higher SML (SML ≤ -5.99%) (hazard ratio (HR) = 3.80, P < 0.001) and lymph metastasis (hazard ratio (HR) = 2.60, P = 0.003) was an independent risk factor forRFS. The models combining SML and clinical characteristics had excellent predictive performance for OS and RFS. The nutritional status marker SML are effective and convenient indicators for predicting the long-term prognosis of DCC after PD, and the SMLafter PD is notable. The combination of CT quantified SML and clinical features can help clinicians predict the long-term survival of DCC patients after PD.