Predictors of Hepatic Decompensation after Yttrium90 Transarterial Radioembolization—Optimizing Patient Selection

Abstract

Objective Yttrium 90 (Y90) transarterial radioembolization (TARE) is effective for unresectable hepatocellular carcinoma (HCC) or to bridge/downstage before transplant; however, optimal patient selection is not well-described. This study aims to identify factors that increase risk of liver decompensation resulting in hospital admissions after TARE. Methods Patients who received Y90 as their first treatment during 2012 to 2022 were identified from a prospectively collected database of 1675 HCC patients. Clinically significant hepatic decompensation was defined as total bilirubin more than or equal to 3 mg/dL or any increase in Model for End-stage Liver Disease (MELD) score resulting in readmission within 60 days or death. Results Of 137 patients, 7 (5.1%) developed hepatic decompensation requiring admission within 30 days and an additional 8 (10.9%) within 60 days. Two of these patients (1.4%) died and two (1.4%) required urgent transplant within 2 months. Preprocedure albumin less than 3.5 gm/dL (p = 0.0207), international normalized ratio more than 1.2 (p = 0.017), ascites (p = 0.036), elevated MELD (p = 0.012), and Child-Pugh (p = 0.007) scores were significant predictors of decompensation, while creatinine and sodium were not. Patients with Child-Pugh B score were three to four times more likely to decompensate (28 vs. 8%) compared to Child-Pugh A. For every unit increase in Child-Pugh score more than 6, odds of decompensation increased by a factor of 2.15. Conclusion Y90 TARE is safe and effective; however, 10.9% patients require readmission for worsened liver function. Because ascites is a significant factor in predicting decompensation and all patients require adequate renal function to receive Y90 TARE, Child-Pugh score may be more useful than MELD for patient selection. Further risk stratification may be required for those with a Child-Pugh score more than or equal to 7.