Abstract
PURPOSE: It was found that 33.8-52.5% of patients experienced nausea and vomiting after Transcatheter Arterial Chemoembolization (TACE) for liver cancer, based on prior literature. But there are no models that predict this risk. In this study, we investigated the factors associated with nausea and vomiting after TACE and developed a predictive model to predict these adverse events. METHOD: The study will include 401 patients who will be randomly assigned to the training group and validation group in a 7:3 ratio. An analysis of logistic regression was used to identify predictors and build a risk prediction model. Model performance was evaluated using the Area Under Curve (AUC), Calibration Curve, and Decision Curve Analysis (DCA). RESULTS: This study ultimately included 401 patients for TACE, of whom 132(32.92%) patients experienced nausea and vomiting. Multivariate analysis identified five independent risk predictors: history of vomiting, prophylactic use of antiemetics, postoperative pain, platinum, and anthracycline. The training group had an AUC of 0.839, and the validation group had an AUC of 0.742. They had calibration curves with P = 0.208 and P = 0.482, respectively. The DCA curves showed that the model had good clinical benefit at threshold probabilities greater than 20%. CONCLUSION: A predictive model of nausea and vomiting after TACE has been developed, based on the individual risk factors, surgical factors and chemotherapy drug factors, with satisfactory predictive ability. This model can identify patients for TACE who are at high risk of nausea and vomiting. Our study provides an empirical basis for early detection, early diagnosis and early intervention of patients for TACE at high risk of nausea and vomiting.