Future liver remnant function as a predictor of postoperative morbidity following liver resection for hepatocellular carcinoma - A risk factor analysis.

BACKGROUND

Advances in anaesthesia and surgical technique have considerably reduced mortality in hepatocellular carcinoma (HCC) patients undergoing liver resection. However, extended resections in patients with liver cirrhosis still represent a challenge. The aim of this study was to investigate the predictive value of volume/function analysis for the prediction of morbidity in HCC patients following liver resection.

METHODS

Between 2001 and 2014, a total of 261 patients who underwent open hepatectomy for HCC were enrolled in this study. Future liver remnant volume (FLRV) and future liver remnant function (FLRF) based on LiMAx testing were obtained retrospectively. Uni- and multivariable analyses were performed to identify predictors for postoperative ascites, post-hepatectomy haemorrhage (PHH), and wound healing disorders (WHD) within the total cohort and in a subgroup of cirrhotic patients.

RESULTS

The most commonly observed complication was ascites (57.1%), followed by liver failure (25.3%), PHH (19.5%), and WHD (19.2%). FLRF was a major predictor of postoperative ascites (AUC 0.776; OR 0.987, p = 0.001), PHH (AUC 0.717; OR 0.984, p = 0.001), and WHD (AUC 0.660; OR 0.994, p = 0.032) in total cohort. Multivariable analysis of the cirrhosis subgroup showed FLRF to be an independent predictor of ascites (AUC 0.814; OR 0.989, p = 0.021), PHH (AUC 0.677; OR 0.991, p = 0.040), and WHD (AUC 0.615; OR 0.989, p = 0.033).

CONCLUSIONS

FLRF is a major predictor of postoperative ascites, haemorrhage, and wound healing disorders in cirrhotic and non-cirrhotic patients whereas FLRV failed to show significant correlations. Preoperative calculation of FLRF may augment surgical decision-making in high-risk patients and thereby improve perioperative outcome.