Association of Cirrhosis and Other Patient and Procedural Characteristics With Postembolization Syndrome After Bland Hepatic Artery Embolization for Hepatic Malignancy

BACKGROUND. Patients who undergo bland hepatic artery embolization (HAE) for the treatment of hepatic malignancy may undergo routine overnight postprocedure hospitalization to monitor for postembolization syndrome (PES) given the potential for ischemic injury from HAE to lead to rapid onset of PES. In our experience, PES after HAE is more frequent in patients without cirrhosis. OBJECTIVE. The purpose of this study was to investigate the utility of cirrhosis and other patient and procedural characteristics in predicting the development of PES after bland HAE performed for the treatment of hepatic malignancy. METHODS. This retrospective study included 167 patients (122 men and 45 women; mean age, 63.5 ± 13.1 [SD] years) who underwent a total of 248 bland HAE procedures to treat primary or secondary hepatic malignancy. All patients were hospitalized for 24 hours of observation after HAE to monitor for and manage PES symptoms. PES severity was graded using the Southwest Oncology Group's toxicity coding scale. Patient and procedural characteristics were recorded. Associations with the development of PES were explored. A risk model to predict the risk of PES was constructed using independent predictors of PES in multivariable analysis. RESULTS. PES developed after 51.2% (127/248) of procedures; 23 cases were mild, 50 were moderate, and 54 were severe. PES developed in 32.1% (45/140) of patients with cirrhosis versus 75.9% (82/108) of patients without cirrhosis, whereas severe PES developed in 10.0% (14/140) versus 37.0% (40/108) of such patients, respectively. In multivariable analysis (which controlled for primary versus secondary malignancy, comorbidities, pre-procedure laboratory values, size and multiplicity of treated lesions, lobar vs segmental embolization, embolized artery, and embolic material used), independent predictors of lower likelihood of PES were older age (OR = 0.95 [95% CI, 0.92-0.99]), cirrhosis (OR = 0.26 [95% CI, 0.11-0.64]), and primary hepatic malignancy (OR = 0.34 [95% CI, 0.13-0.93]); the only independent predictor of a higher likelihood of PES was embolization of 50% or more of liver volume (OR = 4.29 [95% CI, 1.89-10.18]). A risk model using these factors had sensitivity of 75.6% and specificity of 76.0% for predicting PES. CONCLUSION. Cirrhosis was associated with a decreased risk of PES after bland HAE performed for the treatment of hepatic malignancy. A risk model combining cirrhosis and other factors had good performance in predicting the risk of PES. CLINICAL IMPACT. These findings may be applied to the selection of patients for early discharge after bland HAE, to avoid the need for overnight inpatient monitoring.