Preoperative nutritional score predicts underlying liver status and surgical risk of hepatocellular carcinoma

Background: Given the scarce evidence regarding the impact of preoperative nutritional status on surgical outcomes of patients with hepatocellular carcinoma, predictive powers of nutritional/inflammatory scores for short-term surgical outcomes in patients with hepatocellular carcinoma were investigated. Methods: Outcomes of 1272 patients with hepatocellular carcinoma were reviewed, and predictive powers of nine nutritional/inflammatory scores for short-term surgical outcomes were compared using the receiver-operating characteristic curve analysis. Clinical relevance of the best nutritional score was then studied in detail to clarify its utility as an alternative predictive measure for surgical risk of patients with hepatocellular carcinoma. Results: Receiver-operating characteristic curve analysis showed the controlling nutritional status score has the best performance in prediction of morbidity after hepatectomy for hepatocellular carcinoma (area under the curve, 0.593; 95% confidence interval: 0.552-0.635; p < 0.001), and multivariate analysis confirmed its correlation with the risk of any morbidity (odds ratio per +1 point, 1.17; 95% confidence interval: 1.08-1.27; p < 0.001) and major morbidity (odds ratio per +1 point, 1.14; 95% confidence interval: 0.99-1.27; p = 0.052). The undernutrition grade based on the controlling nutritional status score showed strong correlation with the degree of fibrosis in the liver (p < 0.001), platelet count (p < 0.001), and indocyanine green retention rate at 15 min (p < 0.001). In addition, the controlling nutritional status undernutrition grade well stratified the risk of postoperative morbidity especially in cirrhotic subpopulation (odds ratio, 1.17 per +1 point; 95% confidence interval: 1.05-1.29 for any morbidity and odds ratio, 1.20 per +1 point; 95% confidence interval: 1.03-1.40 for major morbidity). Conclusion: The controlling nutritional status score could be an alternative measure for underlying liver injury and the surgical risk of hepatocellular carcinoma.