Research Letter
Gastroenterology and Hepatology
May 21, 2025
Acute Kidney Injury and Hepatorenal Syndrome Among Hospitalized Patients With Chronic Liver Disease
Robert J. Wong, Jeyasankar Balasubramanian, Mary Panaccio, et al
JAMA Netw Open. 2025;8(5):e2511816. doi:10.1001/jamanetworkopen.2025.11816
Introduction
The prevalence of chronic liver disease (CLD) is increasing,1,2 with alcohol-related liver disease (ALD) a common etiology.3 Hospitalized patients with CLD often experience acute kidney injury (AKI) and hepatorenal syndrome (HRS),4 contributing significantly to health care resource utilization (HCRU).5,6 Trends in the incidence of AKI and HRS among hospitalized patients with CLD in the US and the conditions’ association with inpatient mortality, costs, and HCRU may help guide resource planning.
Methods
Ethics committee approval and informed patient consent were not required for this retrospective, longitudinal cohort study of Health Insurance Portability and Accountability Act–compliant, deidentified electronic health records because this was not human participants research as defined by 45 CFR §46.102. We adhered to the STROBE reporting guideline. Study details, including statistical tests, are in the eMethods and eFigure in Supplement 1. Briefly, this study used US hospital inpatient data from 2018 to 2023 for patients aged 18 years or older with CLD, including cirrhosis, derived from the Premier Healthcare Database. Patients with CLD and with CLD and AKI or HRS were identified using International Statistical Classification of Diseases, Tenth Revision, Clinical Modification (ICD-10-CM) codes (eTable in Supplement 1). Total charges were adjusted to 2024 US dollars. Data analysis was performed using Spark SQL version 3.3.0 (Apache Software Foundation) and Python version 3.9 (Python Software Foundation), with 2-tailed P values < .05 considered significant.
Results
Among 4 056 287 hospitalizations of patients with CLD (mean [SD] age, 58.3 [15.5] years; 55.4% male), representing 8.3% of hospitalizations during the study period, 44.9% had Medicare and 26.0% had Medicaid insurance. The diagnosis was AKI for 1 221 546 patients (30.1%; mean [SD] age, 62.6 [14.5] years; 59.7% male; 53.8% on Medicare) and HRS for 90 235 patients (2.2%; mean [SD] age, 58.9 [13.3] years; 60.7% male; 41.6% on Medicare). A substantial proportion of patients with AKI (27.9%) or HRS (59.6%) had alcohol-related disorders. Of patients with HRS, 95.5% had a diagnosis of liver cirrhosis. Over the study period, the proportion of patients hospitalized with CLD increased (7.4% to 9.0%); hospitalizations also increased among patients with CLD and AKI (27.1% to 31.2%; P < .001) or CLD and HRS (1.9% to 2.3%; P < .001). The proportion of patients hospitalized with CLD and AKI with ALD remained stable from 2018 (20.2%) to 2023 (19.8%), whereas the proportion with CLD and HRS and concurrent ALD increased from 52.8% in 2018 to a peak of 59.8% in 2021, declining to 56.5% in 2023. Overall, inpatient mortality was significantly higher among CLD hospitalizations with HRS vs AKI (26.4% vs 18.4%; P < .001) (Table). From 2018 to 2023, the mean (SD) total hospitalization charges per patient increased from $141 005 ($235 624) to $165 458 ($302 830) for individuals with CLD and AKI and $153 456 ($250 681) to $186 760 ($307 255) for those with HRS (Table). From 2018 to 2023, the projected annual incidence of inpatients diagnosed with CLD increased from 2 288 242 to 2 754 542 individuals; among patients hospitalized with CLD, projected inpatient numbers increased from 619 775 to 860 119 individuals for AKI and 42 930 to 63 381 individuals for HRS (Figure).
Discussion
Using US hospital inpatient data in this cohort study, we observed an increasing clinical burden associated with hospitalizations in patients with CLD and with CLD and AKI or HRS. This trend is concerning and is expected to increase the economic burden owing to substantial comorbidities, potential liver transplant, high inpatient mortality, and increasing hospitalization charges likely due to inflation. Study limitations include a potential for misclassification bias (especially between metabolic dysfunction–associated steatotic liver disease or steatohepatitis vs metabolic dysfunction with ALD vs ALD) because of limitations in data captured for alcohol use, and underdiagnosis of CLD owing to reliance on ICD-10-CM codes. Our findings highlight the need to optimize management of kidney-related complications in patients with CLD, offering opportunities to improve clinical outcomes and potentially halt or reduce the troubling trend of increasing HCRU.
