Prediction of postoperative recurrence of hepatocellular carcinoma after radiofrequency ablation combining psychological and sleep quality using a nomogram model based on Cox regression

Abstract

Purpose This study aimed to investigate the risk factors for postoperative recurrence in patients with hepatocellular carcinoma (HCC) and develop a Cox regression-based nomogram model incorporating psychological factors and sleep quality to predict postoperative recurrence after radiofrequency ablation (RFA) for HCC. The model was further visualized for practical use. Methods A prospective follow-up was conducted at Shengzhou People’s Hospital (the First Affiliated Hospital of Zhejiang University Shengzhou Branch) from January 2013 to December 2021, including HCC patients who underwent RFA. Sleep quality and psychological status were assessed through questionnaires, and relevant baseline and tumor data were collected, including age, gender, Pathology, PT, INR, PLT, Alb, TBIL, AFP, DCP, PHT, ALBI grade, Cirrhosis, ascites, Maximum tumor diameter, and tumor number. Cox proportional hazards models were used to analyze the factors associated with postoperative recurrence, both in univariate and multivariate analysis. A nomogram prediction model was constructed, and its performance was evaluated using ROC curve, AUC, and calibration curve. Results The study included 70 patients with a mean age of 61.07 years (range: 23–87 years). The median time to recurrence was 13 months (range: 2–64 months), and 32 patients (45.70%) experienced recurrence during the follow-up period. Univariate analysis showed significant correlations between AFP, PHT, ALBI grade, DCP, Maximum tumor diameter, tumor number, cirrhosis, SAS, SDS, and postoperative recurrence in HCC patients (P < 0.05). Multivariate analysis confirmed that AFP, ALBI grade, Maximum tumor diameter, cirrhosis, SAS, and SDS were independent risk factors for postoperative recurrence (P < 0.05). The nomogram model based on these factors showed good predictive accuracy with a concordance index of 0.857 (95% CI: 0.798–0.916). The ROC curve analysis demonstrated that the nomogram model had a high predictive accuracy and clinical utility. The calibration curve showed good consistency between the predicted and actual recurrence rates. Additionally, the decision curve analysis indicated that the nomogram model had superior clinical value compared to individual variables. Conclusion AFP, ALBI grade, Maximum tumor diameter, cirrhosis, DCP, SAS, and SDS were identified as independent factors associated with postoperative recurrence in HCC patients undergoing RFA. The nomogram model incorporating these factors can provide better guidance for personalized clinical decision-making.