Acute‐on‐chronic liver failure (ACLF) is characterized by acute decompensation (AD) of cirrhosis, organ failure(s), and high 28‐day mortality. We investigated whether assessments of patients at specific time points predicted their need for liver transplantation (LT) or the potential futility of their care. We assessed clinical courses of 388 patients who had ACLF at enrollment, from February through September 2011, or during early (28‐day) follow‐up of the prospective multicenter European Chronic Liver Failure (CLIF) ACLF in Cirrhosis study. We assessed ACLF grades at different time points to define disease resolution, improvement, worsening, or steady or fluctuating course. ACLF resolved or improved in 49.2%, had a steady or fluctuating course in 30.4%, and worsened in 20.4%. The 28‐day transplant‐free mortality was low‐to‐moderate (6%‐18%) in patients with nonsevere early course (final no ACLF or ACLF‐1) and high‐to‐very high (42%‐92%) in those with severe early course (final ACLF‐2 or ‐3) independently of initial grades. Independent predictors of course severity were CLIF Consortium ACLF score (CLIF‐C ACLFs) and presence of liver failure (total bilirubin ≥12 mg/dL) at ACLF diagnosis. Eighty‐one percent had their final ACLF grade at 1 week, resulting in accurate prediction of short‐ (28‐day) and mid‐term (90‐day) mortality by ACLF grade at 3‐7 days. Among patients that underwent early LT, 75% survived for at least 1 year. Among patients with ≥4 organ failures, or CLIF‐C ACLFs >64 at days 3‐7 days, and did not undergo LT, mortality was 100% by 28 days. Conclusions: Assessment of ACLF patients at 3‐7 days of the syndrome provides a tool to define the emergency of LT and a rational basis for intensive care discontinuation owing to futility. (Hepatology 2015;62:243‐252)

Acute decompensation (AD) of cirrhosis is defined by the acute development of ascites, gastrointestinal hemorrhage, hepatic encephalopathy, infection or any combination of these, requiring hospitalization. The presence of organ failure(s) in patients with AD defines acute-on-chronic liver failure (ACLF), while their absence defines AD. We designed the PREDICT study, a European, prospective, observational study, to characterize the clinical course of AD and predict ACLF. A total of 1071 patients with AD were enrolled to collect detailed pre-specified information on the 3-month period prior to enrollment, and clinical and laboratory data at enrollment. Patients were then closely followed-up for 3 months. The 12-month outcomes (liver transplantation, and death) were also recorded. Three groups of patients were identified: Pre-ACLF patients (n=218), who developed ACLF and had 3-month and 1-year mortality rates of 53.7% and 67.4%, respectively. Unstable decompensated cirrhosis (UDC) patients (n = 233) required ≥1 readmission but not developing ACLF and had 21.0% and 35.6% mortality rates. Stable decompensated cirrhosis (SDC) patients (n = 620) who were neither readmitted, nor developed ACLF and showed a 1-year mortality of only 9.5%. The 3 groups differed significantly in the grade and course of systemic inflammation (high-grade at enrollment with aggravation during follow-up in pre-ACLF; low-grade at enrollment with subsequent steady-course in UDC; and low-grade at enrollment with subsequent improvement in the SDC) and prevalence of surrogates of severe portal hypertension throughout the study (high in UDC versus low in pre-ACLF and SDC). Acute decompensation without ACLF is a heterogeneous condition with three different clinical courses and two major pathophysiological mechanisms: systemic inflammation and portal hypertension. Prediction of ACLF development remains a major future task.

Acute decompensation (AD) of cirrhosis is defined by the development of ascites, hepatic encephalopathy and/or variceal bleeding. Ascites is traditionally attributed to splanchnic arterial vasodilation and left ventricular dysfunction, hepatic encephalopathy to hyperammonaemia, and variceal haemorrhage to portal hypertension. Recent large-scale European observational studies have shown that systemic inflammation is a hallmark of AD. Here we present a working hypothesis, the systemic inflammation hypothesis, suggesting that systemic inflammation through an impairment of the functions of one or more of the major organ systems may be a common theme and act synergistically with the traditional mechanisms involved in the development of AD. Systemic inflammation may impair organ system function through mechanisms which are not mutually exclusive. The first mechanism is a nitric oxide-mediated accentuation of the preexisting splanchnic vasodilation, resulting in the overactivation of the endogenous vasoconstrictor systems which elicit intense vasoconstriction and hypoperfusion in certain vascular beds, in particular the renal circulation. Second, systemic inflammation may cause immune-mediated tissue damage, a process called immunopathology. Finally, systemic inflammation may induce important metabolic changes. Indeed, systemic inflammatory responses are energetically expensive processes, requiring reallocation of nutrients (glucose, amino acids and lipids) to fuel immune activation. Systemic inflammation also inhibits nutrient consumption in peripheral (non-immune) organs, an effect that may provide one mechanism of reallocation and prioritisation of metabolic fuels for inflammatory responses. However, the decrease in nutrient consumption in peripheral organs may result in decreased mitochondrial production of ATP (energy) and subsequently impaired organ function.