Interleukin-22 is associated with progression to acute-on-chronic liver failure and short-term mortality in cirrhosis

Background and aims Emerging data show a high potential of prognostic biomarkers in decompensated cirrhosis. Previously, an association between IL-22 and adverse outcomes of decompensated cirrhosis has been shown, but these findings are still to be validated. We therefore analyzed associations between IL-22 plasma levels and clinical endpoints of patients with different patterns of decompensated cirrhosis. Methods A post-hoc analysis of consecutive, hospitalized patients with liver cirrhosis was performed. Baseline IL-22 plasma concentrations were quantified and correlated with disease stage and clinical endpoints observed over a 90-day follow-up period. Results A total of 254 patients were included in the present study. Plasma IL-22 concentrations in patients with compensated cirrhosis were comparable to those of healthy controls. In contrast, patients experiencing decompensation showed significantly higher IL-22 levels (compensated vs. decompensated, p = 0.004; healthy vs. decompensated, p = 0.014), with further increases noted in those with ACLF (healthy vs. ACLF, p < 0.0001). Of note, patients classified as pre-ACLF demonstrated significantly higher IL-22 concentrations in comparison to chronic, acute stable, and acute unstable decompensated cirrhosis, with a cohort-derived threshold of 11.94 pg/mL showing the highest Youden index for pre-ACLF (AUC = 0.75, 95% CI: 0.67–0.83; p < 0.0001; sensitivity 82%, specificity 63%). Furthermore, IL-22 plasma levels were significantly higher in patients who died within 30 or 90 days ( p = 0.0018 and p = 0.0002). In subgroup analyses this association was confined to patients with ACLF at 30 days ( p = 0.043) and to patients without ACLF at 90 days ( p = 0.011), but it was not observed within the pre-ACLF subgroup. Conclusions IL-22 is associated with adverse outcomes and mortality in liver cirrhosis. Moreover, IL-22 concentrations discriminated pre-ACLF from the other courses of acute decompensation. As all thresholds were derived within the present cohort and were not externally validated, these findings are exploratory and need further analysis.