Year: 2017
Type: editorial
Abstract: Potential conflict of interest: Nothing to report. See Article on Page 1464 Alcoholic hepatitis (AH) develops in patients with excessive and recent alcohol consumption and may be associated with acute‐on chronic liver failure and death. How can we predict which patients with AH will live and which will die, and when, and from what? Stratification of AH based on severity has been examined with the generation of a number of prognostic models. A Maddrey discriminant function ≥32 and/or a Model for End‐stage Liver Disease (MELD) score >20 are associated with 20%‐50% 30‐day mortality rates; these are some of the best‐validated models.1 Preclinical and clinical studies have focused on novel therapies to improve this 30‐day short‐term mortality in severe AH (SAH). This is also the time point at which a survival benefit of corticosteroids can be demonstrated. By contrast, less emphasis has been given to long‐term outcomes in SAH, with a small retrospective study from the United Kingdom identifying sustained alcohol abstinence as an independent predictor of long‐term survival in SAH.2 Thus, a current gap in knowledge relates to the temporal relationship of prognostic factors to death in AH. In the current issue of hepatology, Louvet and colleagues report the impact of alcohol relapse on both short‐term and long‐term survival in SAH.3 All patients admitted with SAH were treated with corticosteroids and followed prospectively at 2 months, 3 months, 6 months, 9 months, 12 months, and every 6 months thereafter. Response to treatment was assessed by the Lille score, and corticosteroid treatment was discontinued in nonresponders at day 7 (Lille score >0.45); responders were treated for 28 days. A total of 398 patients were included, of whom 60% were steroid responders. During short‐term follow‐up (<6 months), 37% of patients died, with 90% of deaths related to liver insufficiency. The Lille model and MELD score, but not alcohol relapse, were predictive of short‐term mortality. However, improvement in liver function was greatest in abstinent patients who responded to steroid treatment. Beyond 6 months, approximately 40% of patients died, and mortality was related to liver failure in 66.7% of the cases. In this group, alcohol recidivism was associated with the worst long‐term outcomes. In addition to alcohol relapse, the Lille model independently predicted long‐term mortality, whereas the MELD score at baseline did not. Thus, short‐term survival is predicated by severity of AH and influenced by pharmacologic therapy directly targeting AH, while longer‐term survival is predicated by alcohol recidivism. This conclusion has important implications for the management of AH patients and the rational design of pharmacologic interventions with liver‐targeted therapies administered early and abstinence‐targeted therapies to follow. Why did Lille perform better than MELD for longer‐term prognosis? The Lille model, developed in 2007 by Louvet and colleagues, includes a set of objective variables: age, albumin level, serum creatinine, prothrombin time, and change in bilirubin level from day 0 to day 7 of steroid treatment.4 Besides being a dynamic model that assesses response to treatment, the inclusion of age and serum albumin level is likely responsible for the better prognostic performance of the Lille model over the MELD score at baseline. Older patients with poor synthetic liver function or malnutrition at presentation, based on low serum albumin, are more likely to do worse in the long term. Whether a delta‐MELD score pretreatment and posttreatment predicts long‐term survival in SAH remains unanswered. Are there other parameters that predict survival outside of MELD, Lille score, and abstinence? Although liver biopsy is not routinely performed in AH in North America, advanced fibrosis,5 the degree of neutrophil infiltration, bilirubinostasis, and presence of megamitochondria6 have been independently associated with 90‐day mortality in SAH. Nutrition is another important prognostic variable. While alcohol provides energy, these are “empty calories”; and patients with SAH typically present with sarcopenia and protein‐calorie malnutrition, which correlates with short‐term and long‐term mortality.7 While recent randomized controlled trials have failed to demonstrate a survival benefit with intensive enteral nutrition,8 poor calorie intake is associated with lower survival at 6 months compared to those with adequate oral or enteral nutrition. Finally, complications of underlying portal hypertension also likely influence prognosis at some time point. Further studies are needed to address these important variables and their temporal effects on outcomes. A limitation of the current study and one applicable to many studies focused on abstinence is how alcohol consumption is defined. The researchers assessed alcohol relapse by medical history and self‐report, with a recidivism rate (defined as >30 g/day) of 35% within 3 years of follow‐up. Although this number is generally consistent with alcohol relapse rates after liver transplantation, which range from 11% to 26%,9 the methodology likely underestimates recidivism. Additional monitoring tools for alcohol relapse, such as urine ethyl glucuronide and blood phosphatidylethanol,10 may provide greater accuracy, as will newer breath and skin biomarkers that are actively being validated. Regardless, the study highlights the need for intensive alcohol treatment, which can include counseling and enrollment in Alcoholics Anonymous, to be part of long‐term management of SAH for success in survival past 1 month. Additionally, the safety of currently approved pharmacological interventions for alcohol use disorder should be further investigated in patients with liver disease to determine if they are safe and effective in this population. Finally, the hepatology community will need to “dig in” to these often messy alcohol addiction issues in our medical encounters with these patients rather than limiting our assessments to liver function and liver‐based pharmacotherapy. Although this is a time challenge in our current “relative value unit–driven” medical system, only then will we impact longer‐term survival in our patients with AH. In summary, the work of Louvet et al. highlights the differences and causes between short‐term and long‐term outcomes in SAH. While acute management of SAH focuses on 30‐day pharmacological therapies and prevention of infections, long‐term management should concentrate on maintenance of alcohol abstinence and nutritional support. A combinatorial pharmacologic approach that uses drugs to promote liver recovery early, along with drugs for alcohol use disorder, is compelling. Responders to steroid therapy who remain abstinent have an excellent long‐term prognosis and can return to functioning in the community. Nonresponders to steroid therapy who are abstinent but deteriorate on follow‐up may be candidates for liver transplantation. However, nonresponders who relapse despite aggressive alcohol treatment have a poor prognosis, and palliative care may be the appropriate option for this group (Fig. 1). The current study provides evidence for such a tailored approach to our patients with AH.Figure 1: Long‐term management and survival of SAH according to steroid response.
Source: Hepatology
DOI: 10.1002/hep.29349
Institution: Mayo Clinic
Language: English
Subfield: Hepatology
Field: Medicine
Domain: Health Sciences
Open Access status: bronze
OpenAlex ID: W2734179535