Short-term reductions in non-protein respiratory quotient and prealbumin can be associated with the long-term deterioration of liver function after transcatheter arterial chemoembolization in patients with hepatocellular carcinoma

Transcatheter arterial chemoembolization (TACE) is an effective treatment for hepatocellular carcinoma (HCC) that can cause deterioration of liver function. We aimed to make an early predictive model of long-term liver dysfunction after TACE. We performed a prospective cohort study involving 109 HCC patients who underwent TACE at Kobe University Hospital. Indirect calorimetry and blood biochemical examinations were performed before and 7 days after TACE. As an indicator of liver function, the Child’s score was evaluated before and 3 months after TACE. Patients with and without Child’s score deterioration were compared, and the independent risk factors for Child’s score deterioration were statistically examined. An early predictive model of Child’s score deterioration after TACE was developed using multivariate logistic regression. Multivariate analyses showed that the non-protein respiratory quotient (npRQ) and prealbumin (preAlb) ratios (7 days after/before TACE) were independent determinants of Child’s score deterioration (p = 0.039 and 0.020, respectively). Decreases of the npRQ and preAlb ratios were significantly related to increases of Child’s score 3 months after TACE (p = 0.007 and 0.002, respectively). The following predictive model of Child’s score deterioration was developed: exp(−6.383 × npRQ ratio − 3.038 × preAlb ratio + 7.755)/(1 + exp(−6.383 × npRQ ratio − 3.038 × preAlb ratio + 7.755)). The model discriminated well between patients with and without Child’s score deterioration (area under the receiver operating curve [ROC]; AUC 0.713; 95% confidence interval [CI] 0.613–0.812). The optimal cut-off point for the Child’s score was 0.449, and the sensitivity and specificity of the model were 57.1 and 79.1%, respectively. Reductions in npRQ and preAlb 7 days after TACE were associated with the long-term deterioration of liver function. With our model, we were able to identify high-risk patients.