A comprehensive method for predicting fatal liver failure of patients with liver cancer resection.

ABSTRACT

There are many methods to assess liver function, but none of them has been verified as fully effective. The purpose of this study is to establish a comprehensive method evaluating perioperative liver reserve function (LRF) in patients with primary liver cancer (PLC).In this study, 310 PLC patients who underwent liver resection were included. The cohort was divided into a training set (n = 235) and a validation set (n = 75). The factors affecting postoperative liver dysfunction (POLD) during preoperative, intraoperative, and postoperative periods were confirmed by logistic regression analysis. The equation for calculating the preoperative liver functional evaluation index (PLFEI) was established; the cutoff value of PLFEI determined through analysis by receiver-operating characteristic curve was used to predict postoperative liver function.The data showed that body mass index, international normalized ratio, indocyanine green (ICG) retention rate at 15 minutes (ICGR15), ICG elimination rate, standard remnant liver volume (SRLV), operative bleeding volume (OBV), blood transfusion volume, and operative time were statistically different (all P < 0.05) between 2 groups of patients with and without POLD. The relationship among PLFEI, ICGR15, OBV, and SRLV is expressed as an equation of "PLFEI = 0.181 × ICGR15 + 0.001 × OBV - 0.008 × SRLV." The cutoff value of PLFEI to predict POLD was -2.16 whose sensitivity and specificity were 90.3% and 73.5%, respectively. However, when predicting fatal liver failure (FLF), the cutoff value of PLFEI was switched to -1.97 whose sensitivity and specificity were 100% and 68.8%, respectively.PLFEI will be a more comprehensive, sensitive, and accurate index assessing perioperative LRF in liver cancer patients who receive liver resection. And keeping PLFEI <-1.97 is a safety margin for preventing FLF in PLC patients who underwent liver resection.