Role of predisposition, injury, response and organ failure in the prognosis of patients with acute-on-chronic liver failure: a prospective cohort study

Abstract

Introduction: Acute deterioration of cirrhosis is associated with high mortality rates particularly in the patients who develop organ failure (OF), a condition that is referred to as acute-on-chronic liver failure (ACLF), which is currently not completely defined. This study aimed to determine the role of predisposing factors, the nature of the precipitating illness and inflammatory response in the progression to OF according to the PIRO (predisposition, injury, response, organ failure) concept to define the risk of in-hospital mortality.

Methods: A total of 477 patients admitted with acute deterioration of cirrhosis following a defined precipitant over a 5.5-year period were prospectively studied. Baseline clinical, demographic and biochemical data were recorded for all patients and extended serial data from the group that progressed to OF were analysed to define the role of PIRO in determining in-hospital mortality.

Results: One hundred and fifty-nine (33%) patients developed OF, of whom 93 patients died (58%) compared with 25/318 (8%) deaths in the non-OF group (P < 0.0001). Progression to OF was associated with more severe underlying liver disease and inflammation. In the OF group, previous hospitalisation (P of PIRO); severity of inflammation and lack of its resolution (R of PIRO); and severity of organ failure (O of PIRO) were associated with significantly greater risk of death. In the patients who recovered from OF, mortality at three years was almost universal.

Conclusions: The results of this prospective study shows that the occurrence of OF alters the natural history of cirrhosis. A classification based on the PIRO concept may allow categorization of patients into distinct pathophysiologic and prognostic groups and allow a multidimensional definition of ACLF.

Figure 1

Study flow and summary of outcomes.

Figure 2

Survival analyses of included patients. (a) Thirty-day mortality the patients with and without organ failure (log-rank test: P < 0.001). The analyses started from the day of onset of organ failure. (b) Thirty-day mortality of patients with organ failure divided according to whether they had a previous hospital admission with decompensated cirrhosis within the previous six months (log-rank test: P < 0.001). (c) Long-term outcome of patients with acute deterioration of cirrhosis that did not develop organ failure compared with the patients with organ failure. The organ failure group is further subdivided into those who required hospital admission with an episode of decompensation within the previous six months and those that did not (log-rank test: P < 0.001).

Figure 3

Changes in C-reactive protein (CRP) (mean (SD)) over the first seven days in the patients that survived compared with the patients who died. The data show that there was a significant reduction in CRP in the survivors (ANOVA: P < 0.001) whereas the CRP increased significantly in the non-survivors (ANOVA: P < 0.05).