Plasminogen Activator Inhibitor-1 Levels in Non-alcoholic Fatty Liver Disease: A Systematic Review and Meta-Analysis.

Abstract

Several studies have investigated the role of multiple proteins in nonalcoholic fatty liver disease (NAFLD); one that has recently gained attention is plasminogen activator inhibitor-1 (PAI-1). However, studies evaluating PAI-1 levels in NAFLD demonstrated conflicting results. Our objective was to understand the role of PAI-1 in NAFLD more clearly by carrying out a systematic review and meta-analysis. We gathered evidence by performing a systematic search on PubMed, EMBASE, and Cochrane Library, through using a predefined search string. The included studies diagnosed NAFLD through either liver biopsy, ultrasonography, computed tomography (CT), magnetic resonance spectroscopy, or using one of the latter methods with blood parameters. Studies had to fulfill predefined inclusion and exclusion criteria. To assess the quality of the studies included, we used the NHLBI quality assessment tools. The main summary outcome was the mean difference (MD) in serum PAI-1 levels reported as ng/mL Results: 33 articles involving 10,840 subjects fulfilled our inclusion criteria and were systematically reviewed. 11 studies were included in our meta-analyses. We found a significant MD in PAI-1 levels in NAFLD patients vs. controls [17.147 (95%CI: 7.720-26.574)]. Moreover, subgroup analysis evaluating PAI-1 levels in biopsy- proven NAFLD vs. controls remained significant [24.086 (95%CI: 3.812-44.361)], as well as in CT-diagnosed NAFLD [15.523 (95%CI: 7.163-23.883)]. However, no significant MD in PAI-1 levels was found in ultrasound- diagnosed NAFLD patients vs. controls [10.394 (95%CI: -13.335-34.123)]. No significant MD in PAI-1 levels in NASH patients vs. controls was observed [26.835 (95%CI: -0.879-54.549)]. In summary, elevated serum PAI-1 levels are associated with adult NAFLD (biopsy-proven and CT-diagnosed). However, no significant difference was found in ultrasound-diagnosed NAFLD and NASH patients. Nonetheless, the included studies have methodological variance, dictating that the obtained results should be carefully interpreted.