LAUSR

To the editor, We read with great interest the study by Musto et al. entitled “Recovery and outcomes of patients denied early liver transplantation for severe alcoholic hepatitis.” In the paper by Musto and colleagues, the authors analyzed data from a single center of patients with alcoholassociated liver disease undergoing transplant evaluation.[1] The authors demonstrated that among 95 patients who did not undergo liver transplantation (LT), 34 candidates experienced spontaneous recovery. Furthermore, younger age, lower peak Model for EndStage Liver Disease (MELD) score, and lower international normalized ratio were factors associated with spontaneous recovery. These observations illustrate potential problems with the criteria being used to allocate organs to patients with alcoholic hepatitis (AH). The inability to determine who is likely to recover spontaneously from severe AH may unnecessarily place certain patients at risk of morbidity, mortality, and costs associated with LT. Furthermore, other patients on the waiting list who are likely to benefit from transplantation would be denied access to donor organs. Many lines of investigation suggest that a diagnosis of acuteonchronic liver failure (ACLF) grade 3 using the European Association for the Study of Chronic Liver Failure (EASLCLIF) ACLF classification can accurately identify patients with AH at high risk of shortterm mortality, within 7 days from presentation.[2] In a large study of patients with severe AH, it was demonstrated that the cumulative incidence of death was strongly associated with ACLF grade, ranging from 10.4% for patients without ACLF to 72.4% for patients with ACLF grade 3.[3] Furthermore, the likelihood of response to corticosteroids ranged from 76.6% response rate in those without ACLF to 8.3% in patients with ACLF grade 3. On the other hand, the MELD score may underestimate mortality, particularly among those with ACLF grade 3.[4,5] It would, therefore, be interesting to know whether ACLF grade was associated with spontaneous recovery, lack of recovery, or deaths on the waiting list and how this compared with the other predictors determined by the authors. If there is an association, as we hypothesize, such findings will provide further support for adoption of ACLF classification in the management of these patients.