Investigating the influence of possible liver fibrosis on mortality in patients with coronary artery disease - a post-hoc ISCHEMIA trial analysis

Abstract Background There is a plethora of studies demonstrating that comorbidities (e.g., diabetes and renal dysfunction) directly influence outcomes in patients undergoing PCI or CABG for coronary artery disease (CAD). Yet, liver dysfunction is not really included in current risk assessment beyond the presence of liver cirrhosis. Retrospective single center evidence now suggests that the presence of liver fibrosis may be an indicator of elevated risk. Using the FIB-4 score, where increasing values correlate with increaseing degrees of histological liver fibrosis, liver fibrosis could already be detected with all required laboratory values (AST, ALT, platelets) within normal range. In addition, the study suggests a beneficial effect of off-pump CABG when FIB-4 is elevated. This finding contradicts current randomized evidence, negating any difference between off and on-pump CABG. Since randomized trials always suffer from strong selection of low risk patients, it is conceivable that trial patients may primarily have low FIB-4 scores. Purpose To address the distribution of FIB-4 scores in a selected trial population with CAD (from the ISCHEMIA trial) and relate the degree of liver fibrosis reflected by the score to mortality. Methods The FIB-4 score was calculated for all patients. According to the standard classification, a value of greater than 1.2 reflects a 90% chance of the presence of histological liver fibrosis. The imapct of FIB-4 on late all-cause mortality was assessed as a continuous and categorical variable using maximally selected rank statistics determined FIB-4 cutoff. Cox regression model was peformed including the adjustment for the main CAD risk factors. Results Among 3735 included patients, 58.8% had a FIB-4 value above 1.2. Patients with higher FIB-4 score (cutoff= 1.64) were slighly older, presented higher prevalence of male sex, prior PCI and CABG, and prior history of atrial fibrillation and cerebrovascular disease. FIB-4 score was directly associated with higher risk of mortality when addressed as a continouos variable and as a categorical variable (Fig. 1). This relationship was evident in the overall population and in all subsets (PCI, CABG and no revascularization; Fig. 2). Conclusions The use of Fib-4 to determine the degree of liver fibrosis may be powerful to predict mortality in CAD populations which may not show any evident clinical signs of liver dysfunction. The Fib-4 distribution in this trial population, considered as selected and low risk, warrants investigating randomized trial populations having compared on- and off-pump CABG.Figure 1.Mortality: overall population.Figure 2.Mortality: by therapy subsets.