Abdominal Pain After Transarterial Chemoembolization In Patients With Hepatocellular Carcinoma:Incidence And Predictors

Objective The age standardized rate of incidence and mortality of liver cancer in China ranks fifth and second among all cancers,respectively.As the standard treatment for intermediate stage hepatocellular carcinoma(HCC),the most common adverse effect after transcatheter arterial chemoembolization(TACE)is post-embolism syndrome(PES).There is great heterogeneity in the numerous published TACE-related studies,and domestic research on the prevention and treatment of PES is relatively rare;therefore,the main purpose of this study was to investigate the incidence of abdominal pain(one of the common symptoms of PES)in HCC patients after TACE and to explore its main risk factors.Methods A retrospective cohort study was conducted on 220 HCC patients and 420 TACE operations in a single center.All information related to patients,tumor,treatment,and pain was collected,including baseline,preoperative,intraoperative,and postoperative data.According to the occurrence of abdominal pain,binary logistic regression analysis and decision tree algorithm were used to determine the perioperative factors related to severe pain.Pain was defined as a right upper abdominal pain score ≥4 points,and received analgesic intervention within 72 hours after TACE.Finally,a concise predictive model was developed and internally validated to enable stratification of pain management in HCC patients.Results Of the 420 TACE procedures,256(70.0%)reported the occurrence of at least one PES symptom,of which the incidence of abdominal pain was the highest(49.0%),that is,57.3% of patients experienced abdominal pain requiring drug intervention within 72 hours after TACE.The prediction model based on logistic regression analysis determined the drug elution bead-TACE(DEB-TACE)[odds ratio(OR)=3.340;95% confidence interval(CI): 2.169-5.141],the number of tumors ≥3(OR=2.235;95% CI: 1.060-4.713),Right lobe embolism(OR=2.917;95% CI: 1.468-5.799)or bilateral lobe embolism(OR=2.310;95% CI: 1.109-4.813)and concomitant extrahepatic artery embolism(OR=2.654;95% CI: 1.227-5.739)were independent risk factors for severe abdominal pain.The decision tree algorithm screened three explanatory variables,including DEB-TACE with the highest level,history of hepatectomy and embolization of non-left lobe.Both the logistic regression model and the decision tree model had moderate classification prediction(sensitivity 74.3% and71.4%,respectively),with an area under the ROC curve(AUC)of 0.706(95% CI:0.658-0.755)for the former and 0.676(95% CI: 0.630-0.721)for the latter.Internal validation was performed by comparing the predicted outcome of abdominal pain occurring after TACE with the observed outcome,and the results showed that the accuracy of logistic regression model in predicting the occurrence of abdominal pain was 71.4%.Conclusions The higher incidence of abdominal pain after TACE affects patients’ quality of life(QOL)and disease prognosis,and more comprehensive analgesic intervention should be provided for patients with higher risk of pain.In this study,both logistic regression and decision tree models suggest that DEB-TACE and multiple tumor embolic sites were independent predictors of abdominal pain,and the simple prediction model would help provide reference for pain management after TACE.