GCSF may improve the prognosis of severe liver disease by promoting liver regeneration and immune restoration. Our Aim was to investigate its controversial efficacy in decompensated cirrhosis, acute alcoholic hepatitis (AAH), or acute-on-chronic liver failure (ACLF) through meta-analysis.

Meta-analysis of proportions (random effect model) including 19 RCTs (1287 patients from 16 Asian and 3 European studies including 487 ACLF, 231 AAH and 569 cirrhotic patients) evaluating survival at day-28, day-90, 6 months, one year, and/or occurrence of sepsis as major outcomes.

In patients with decompensated cirrhosis, G-CSF administration was associated with a reduction in the weight-adjusted risk of mortality of 9% at day-90 (OR=0.33; 95%CI: 0.18–0.58; p = 0.0002), 16% at 6 months (OR=0.31; 95%CI: 0.15–0.62; p = 0.0009), 26% at one year (OR=0.21; 95%CI:0.12–0.38, p<0.0001) and a weight-adjusted 28% risk reduction for sepsis (OR=0.28; 95%CI: 0.16–0.49; p<0.0001). Only Asian studies were positive. In AAH, G-CSF was associated with an 18% reduction in weight-adjusted mortality risk at day-28 (OR=0.31; 95%CI:0.11–0.83, p = 0.021), 32% at day-90 (OR=0.20; 95%CI:0.09–0.46, p<0.0001) and a weight-adjusted 42% risk reduction for sepsis (OR=0.17; 95%CI: 0.08–0.38; p<0.0001). Only Asian studies, in which corticosteroids were not given systematically in case of severe AAH, were positive. In patients with ACLF, the results on mortality at day-28 were heterogeneous, and GCSF had no beneficial effect on sepsis or survival at day-90.

G-CSF may be effective in patients with decompensated cirrhosis or AAH by reducing the occurrence of sepsis and mortality. Further meta-analyses of individual data, or new, powerful and methodologically flawless therapeutic trials, are warranted to confirm these results, which harbor wide divergences between Asian and European RCTs.