Spotlight on hospitalizations for cirrhosis

Three recent studies focused on patients hospitalized with cirrhosis.

A single-center study, published by the American Journal of Gastroenterology on Aug. 7, found that most readmissions of patients initially hospitalized with cirrhosis were not preventable. The prospective cohort study included 654 patients with cirrhosis hospitalized in June 2014 to March 2020. In the 30 days after discharge, 246 patients (38%) were readmitted, and 29 of these readmissions (12%) were judged to be preventable (the three reviewers agreed on preventability for 70% of readmissions). The most common reasons for readmission were hepatic encephalopathy (22%), GI bleeding (13%), acute kidney injury (13%), and ascites (6%). Preventable readmissions were often related to paracentesis timeliness, diuretic adjustment monitoring, and hepatic encephalopathy treatment, the study found. Preventable readmissions were also more common in patients who were from a racial or ethnic minority (odds ratio [OR], 5.80; 95% CI, 1.96 to 17.13), not married (OR, 2.88; 95% CI, 1.18 to 7.05), or previously admitted in the prior 30 days (OR, 3.45; 95% CI, 1.48 to 8.04).

“Although this study was not designed to create a predictive model for preventable readmission, these findings can provide important context for clinicians arranging transitional and post-discharge care,” said the study authors, who added that “top opportunities for improving outcomes are to facilitate timely outpatient paracentesis, coordinate close monitoring of diuretic effects, ensure an adequate supply of medications for [hepatic encephalopathy], and reinforce appropriate dosing of lactulose.”

Another study, published by the Journal of Hepatology on July 28, assessed etiologies and outcomes of acute kidney injury (AKI) in hospitalized patients with cirrhosis. The retrospective cohort study included 2,063 consecutive patients with AKI and cirrhosis treated at any of 11 hospital networks in 2019. The most common AKI etiology was prerenal AKI (44.3%), followed by acute tubular necrosis (ATN; 30.4%) and hepatorenal syndrome (HRS; 12.1%); 6.0% of patients had some other cause of AKI, and 7.2% couldn't be classified. The 90-day mortality rate was 36.5%; 8.1% of patients received a liver transplant. Patients with prerenal AKI had the lowest mortality rate (22.2% vs. 49.0% with HRS-AKI and 52.7% with ATN). In adjusted analyses, higher AKI stage and lack of complete response to treatment were associated with an increased risk of 90-day mortality (P<0.001 for both comparisons).

The study also looked at treatment, finding that 66.5% of patients were treated with IV albumin during their admission. Vasoconstrictors for HRS were used most often in HRS-AKI (77.1% vs. 21.5% of patients with prerenal AKI and 49.0% with ATN), while vasopressors for shock were used most commonly in ATN (52.7% vs. 14.2% in prerenal AKI and 33.6% in HRS-AKI). The study authors called the variation in use of therapies between etiologies “notable.” They described their study as the largest and first consecutive, fully adjudicated cohort study in this population. “These data establish a baseline for U.S. patients as we anticipate optimization and implementation of newly available vasoconstrictors in this population,” they wrote.

Finally, a small study, published by JMIR Human Factors on Aug. 15, tested a method for improving health literacy in patients with decompensated cirrhosis at hospitals and liver clinics. The Australian study used Lucy LiverBot, a beta prototype chatbot. The 20 participating patients were allowed to freely interact with the bot on a tablet under moderator observation. At baseline, 35% reported having difficulty reading and understanding written medical information. The participants actively engaged with the bot and identified it as a potential educational tool that could also act as a social companion to improve well-being. In qualitative interviews, participants suggested that the bot could provide improvements in mental health and well-being and that it would be most useful at the time of diagnosis. The bot requires additional study but “may have the potential to extend its disease specific content to directly address mental well-being” and “fill a gap in the provision of health care to this group of complex chronic disease patients,” the study authors said.