Acute liver failure (ALF) is a severe complication of spontaneous ruptured hepatocellular carcinoma (SRHCC) that requires accurate prediction for effective treatment strategies. We aimed to develop a predictive nomogram to estimate the risk of ALF in patients with SRHCC undergoing treatment. Patients and Methods: We performed a retrospective analysis of historical data from 284 patients diagnosed with SRHCC at the First Hospital of Jilin University over the past decade. Variables were selected through univariate and multivariate logistic regression analyses, and a predictive nomogram was constructed. We evaluated its predictive accuracy against the Child-Pugh Score, R.MELD, and ALBI by assessing discrimination, calibration, and net clinical benefit. Results: Among the 284 patients, 65 developed ALF. The risk factors identified for model development included largest tumor size (LTS), platelet counts, prolonged prothrombin time, and elevated serum α-fetoprotein levels. The nomogram exhibited high accuracy in predicting ALF risk with a C-index of 0.91 (0.87-0.95). The Delong test showed a significant difference between the nomogram and the other three models (p<0.05). The calibration curve for the nomogram fit well, and the decision curve analysis revealed superior net benefit. The optimal cut-off point for the nomogram was determined to be 40, yielding sensitivity, specificity, positive predictive value, and negative predictive value of 83.10%, 87.20%, 65.90% and 94.60%, respectively.
Conclusion:The nomogram we developed provides an optimized tool for predicting ALF in SRHCC patients. Its application can help determine individual patient's risk of ALF, enabling more rational and personalized treatment strategies.