Acute Kidney Injury in Alcohol-Associated Hepatitis: More than a Bystander

Alcohol-associated hepatitis (AH) is an acute inflammatory form of alcohol-associated liver disease (ALD) that encompasses significant morbidity and mortality, up to 30–40% at 90 days.¹ While the amount of alcohol intake that puts an individual at risk for AH is unknown, this condition is typically seen in a subset of patients who consume excessive amounts of alcohol.² Patients with AH typically present with new onset of jaundice, anorexia, fever, and abdominal pain. Laboratory testing reveals moderately elevated aminotransferases with an aspartate aminotransferase to alanine aminotransferase ratio of two or greater, elevated international normalized ratio (INR), and hyperbilirubinemia.³ Histopathologic findings of AH can include steatosis, hepatocellular injury (ballooning and Mallory-Denk bodies), neutrophilic or lymphocytic infiltration, and pericellular fibrosis. Liver biopsy is required for the definitive diagnosis of AH when confounding factors are present.⁴ Several noninvasive, clinical scoring models have been developed to predict the severity of AH, including Maddrey's discriminant function, the model for end-stage liver disease (MELD) score, the age, bilirubin, INR, and creatinine (ABIC) score, and the Glasgow Alcoholic Hepatitis Score (GAHS). Severe AH is further stratified based on cut-off values from these prognostic models^3,5. Several pharmacological therapies have been assessed for severe AH; however, corticosteroids are considered first-line. In 2015, the STOPAH randomized clinical trial demonstrated a nonsignificant reduction in 30-day mortality with corticosteroid use in patients with severe AH.⁶ This result was echoed by systematic reviews. A recent multinational, multicenter retrospective cohort study demonstrated that the use of corticosteroids reduces 30-day mortality in severe AH by 41% and are maximally beneficial in patients with a MELD score between 25 and 39.¹

Multiorgan failure can result in this population by virtue of liver failure, severe portal hypertension, and hemodynamic consequences seen in severe AH. Specifically, acute kidney injury (AKI) is one of the major determinants of clinical outcomes and occurs in approximately 30% of patients with severe AH.⁷ The pathophysiology behind AKI in this population is multifactorial; prerenal azotemia resulting from fluid losses and intravascular shift, reduced renal perfusion due to splanchnic vasodilation and/or hemodynamic instability, concomitant infection, and glomerulopathies due to metabolic abnormalities (Figure). Predictors of AKI in this population include presence of hepatic encephalopathy, systemic inflammatory response syndrome (SIRS), infection, and high MELD score.^8. Diagnosing AKI in this population is challenging as serum creatinine is inaccurate in estimating glomerular filtration rate (GFR) for multiple reasons. Patients with cirrhosis have reduced hepatic synthesis of creatinine, sarcopenia with resultant reduced muscular generation of creatinine, increased renal tubular secretion of creatinine and increased volume of distribution that may dilute serum creatinine. Several classification systems have been suggested to define AKI in patients with cirrhosis, including the AKIN criteria, RIFLE criteria, and the KDIGO criteria. These criteria all utilize an absolute change in serum creatinine compared to baseline and oliguria of <0.5 ml/kg/hour of urine volume for 6 h. More recently, the International Club of Ascites consensus suggested an updated diagnostic criteria for AKI in patients with cirrhosis. It was suggested to avoid utilizing the measurement of urine output as this can be confounded by use of diuretics, to consider measuring the baseline creatinine within the preceding 3 months, and suggesting an alternative calculation of serum creatinine if a baseline value is not available.⁹ The presence of AKI can hasten death in patients with severe AH, thus predictive models have been suggested to identify patients at high risk for inpatient AKI who may benefit from early treatment strategies. The AKI risk score it was developed and validated based on the four aforementioned predictors of AKI in patients with AH with data obtained from 773 patients with AH across four centers in the United States, Europe, and India.⁷ Another study evaluated 103 patients with biopsy proven AH admitted to the Liver Unit in a hospital in Barcelona. The authors found that the development of AKI markedly influenced the 3 month survival of patients (35% vs. 93% in patients with and without AKI, respectively).¹⁰

Figure 1.

Pathophysiology of acute kidney injury in alcohol-associated hepatitis.

Given that AKI in AH is correlated with poor survival, the use of renal replacement therapy for this population is topical and controversial. A recent retrospective cohort study demonstrated a 27% survival rate at 6-months for patients with AH and AKI who received renal replacement therapy (RRT). Only 13.3% of surviving patients required ongoing RRT at 6 months and a pre-RRT MELD-Na score of <35 was a significant predictor of lower 6-month mortality.¹¹ These results are favorable compared to existing literature which has demonstrated 6 month survival of 15% for patients with decompensated cirrhosis due to etiologies unrelated to alcohol. Interestingly, the etiology of AKI appeared to determine who would benefit from RRT and significantly predicted mortality. Patients had higher mortality if demonstrated features of acute tubular necrosis (ATN) or combined ATN and hepatorenal syndrome (HRS) when compared to patients with HRS alone. However, the impact of etiology of AKI should be further investigated in future studies utilizing kidney biomarkers or kidney biopsies before routinely denying this population RRT in this setting.¹¹ Overall, studies have demonstrated that RRT has a potential role in improving survival in patients with AKI and AH and should certainly be considered as a bridge to transplant or recovery in select patients.

In this issue of the Journal of Clinical and Experimental Hepatology, Khatua CR. et al. aimed to assess the impact of AKI on the survival of patients admitted with AH in the region of Cuttack, India;¹² 437 patients meeting inclusion and exclusion criteria who were hospitalized for AH at SCB Medical College in Cuttack, India, between October 2016 and December 2018 were selected for analysis. Of the patients with severe AH, 65% had AKI as per specified criteria on admission. Patients with severe AH and AKI had higher leukocyte count, total bilirubin, serum creatinine, serum urea, INR, MELD and MELD-Na, Child-Pugh score, Maddrey discriminant score, GAHS, and AIBC score (all p < 0.001) compared to those without AKI. These patients had a significantly longer length of hospital stay (6 vs. 4 days), increased incidence of death during hospitalization (25.4% vs. 4.6%), higher incidence of bacterial infections (63.7% vs. 42.6%) and decreased survival at both 28 days (55.2% vs. 80.6%) and 90 days (33.3% vs. 59.3%) compared to patients without AKI. Kaplan–Meier survival analysis also showed significant differences in survival with respect to AKI both at 28 days and at 90 days. This study corroborates current evidence that AKI is (a) predominant in patients presenting to hospital with severe AH, (b) is associated with high short- and long-term mortality, and (c) should be identified and intervened on early in the patient's clinical course.

In summary, there is a wealth of evidence to suggest that the presence of AKI in patients with AH significantly impacts clinical outcomes and mortality. These patients can be promptly identified by diagnostic criteria and risk stratification scores to consider targeted renal therapies that may result in improved prognosis. Future trials and high-quality observational studies should focus on examining the benefit of preventative interventions for AKI, validation of risk stratification scores, and possibly identifying biomarkers for early diagnosis of AKI to target this high risk population.