Serum concentrations of miR-122 were proposed as a marker for various inflammatory diseases, but the mechanisms driving alterations in miR-122 serum levels are unknown. We analysed miR-122 serum levels and hepatic miR-122 expression in mice after hepatic ischaemia and reperfusion (I/R) injury. These data were compared with data from mice after caecal pole ligation and puncture (CLP) procedure. To translate these data into the human, we analysed miR-122 serum concentrations in a cohort of 223 patients with critical illness and 57 patients with cirrhosis. We detected strongly elevated levels of miR-122 in mice after hepatic I/R injury. miR-122-concentrations correlated with the degree of liver damage according to AST/ALT and were associated with the presence of hepatic cell death detected by TUNEL staining. miR-122 levels were elevated in the cellular supernatants in an in vitro model of hepatocyte injury, supporting the hypothesis that the passive release of miR-122 represents a surrogate for hepatocyte death in liver injury. Moreover, miR-122 levels were almost normal in patients with cirrhosis without ongoing liver damage, but were elevated when liver injury was present. In contrast to previous assumptions, miR-122-concentrations were independent of the presence of infection/sepsis in mice or human patients. miR-122 levels did not correlate with disease severity or mortality in critically ill patients. In contrast, serum miR-122 levels strictly correlated with the presence of hepatic injury in these patients. In mice and humans, miR-122 levels represent an independent and potent marker of ongoing liver injury and hepatic cell death regardless of the underlying disease.