Nomogram to predict liver surgery-specific complications for hepatocellular carcinoma: A multicenter study

Abstract

BackgroundEarly identification of patients at risk for surgical complications enables surgeons to make better treatment decisions and optimize resource utilization. We propose to develop a nomogram for predicting the risk of moderate-to-severe liver surgery-specific complications after hepatectomy in hepatocellular carcinoma (HCC) patients. MethodsWe retrospectively enrolled HCC patients who underwent radical hepatectomy at four medical centers from January 2014 to January 2019 in southwestern China, randomly (7:3) divided into training and validation cohorts. We used least absolute shrinkage and selection operator (LASSO) logistic regression to build a nomogram model. ResultsThe nomogram model contained 6 variables: diabetes mellitus (yes vs. no, OR: 2.32, 95% CI: 1.16–4.64, P = 0.02), major hepatectomy (yes vs. no, OR: 2.65, 95% CI: 1.64–4.27, P < 0.001), platelets (PLT, ≥100 × 103/μl vs. <100 × 103/μl, OR: 0.53, 95% CI: 0.33–0.87, P = 0.01), prothrombin time (PT, >13 s vs. ≤13 s, OR: 1.78, 95% CI: 1.04–3.05, P = 0.04), albumin-indocyanine green evaluation grade (ALICE grade, grade B vs. grade A, OR: 2.06, 95% CI: 1.17–3.61, P = 0.01), and prognostic nutrient index (PNI, >48 vs. ≤48, OR: 0.55, 95% CI: 0.33–0.92, P = 0.02). The concordance index (C-index) and area under the receiver operating characteristic curve (AUC) were 0.751 (95% CI, 0.703–0.799) and 0.743 (95% CI, 0.653–0.833) for the training and validation cohorts, respectively. Decision curve analysis (DCA) showed that the nomogram had good clinical value. ConclusionWe provide good preoperative predictors for the risk of moderate-to-high FABIB score complications in patients with HBV-related HCC posthepatectomy.