Characterization of Echocardiogram, Peritoneal Fluid, and Transjugular Liver Biopsy Assessment in Patients With Noncirrhotic Cardiogenic vs. Nephrogenic Ascites

Purpose: Patients with heart failure (HF) may present with cardiogenic ascites in the absence of cirrhosis. Likewise, patients with end-stage renal disease (ERSD) on hemodialysis (HD) may develop nephrogenic ascites in the absence of cirrhosis. In many cases, these patients are assumed to have underlying portal hypertension. However, onset of ascites in ESRD patients on HD could instead be cardiogenic in nature as a result of right-sided HF. Peritoneal fluid analyses, echocardiogram, and transjugular (TJ) liver biopsy with hepatic venous pressure gradient (HVPG) may more accurately define etiology and clinical disease status of noncirrhotic ascites in both these patient cohorts. Methods: A retrospective comparative analysis of two non-cirrhotic cohorts: (1) Patients with HF and ascites (cardiogenic ascites) and (2) Patients with ESRD and ascites was done from 2014 to 2021 at a high-volume academic hospital with a liver transplant program. The identified etiologies of cardiogenic ascites included ischemic, non-ischemic, pulmonary hypertension, and valvular heart disease in patients with histology consistent with passive hepatic congestion and corresponding clinical history/evaluation of HF and ascites. Patients with ESRD on HD who presented with new-onset ascites were also included. Both cohorts were assessed by three measures, including echocardiography, peritoneal fluid analysis, and TJ liver biopsies. Results: Of the 29 non-cirrhotic patients included in analysis, 10 had cardiogenic ascites and 19 had ESRD with ascites. Patients were confirmed to have no cirrhosis by histology; 55% exhibited only minimal hepatic fibrosis (stage < 2). Sinusoidal dilation was present in 21/29 (72%) of patients overall, including 90% in the cardiogenic group and 63% in ESRD (p=NS). On echocardiogram, a larger proportion of patients with cardiogenic ascites had reduced right ventricular systolic function (89% vs. 41%, p=0.04). Peritoneal fluid analysis revealed serum-ascites albumin gradient (SAAG) was higher among those with cardiogenic ascites (1.65, 0.6-2.4) vs. ESRD (0.9, 0.4- 4.0; p=0.005), such that cardiogenic ascites was more likely associated with high SAAG >1.1 (80%) while ESRD was more likely to have low SAAG (79%; p=0.004). Peritoneal fluid protein was higher in ESRD (4.3, 3.0-5.8 g/dL) vs. cardiogenic ascites (3.3, 2.0-4.6 g/dL; p=0.002). TJ hepatic venous pressure gradient (HVPG) was performed in 27/29 (93%) patients. Patients with cardiogenic ascites had significantly elevated free hepatic vein pressure (22, 11-54 mmHg) vs. (12, 2-29 mmHg; p=0.01) and wedged hepatic vein pressure (24, 11-57 mmHg) vs. (14, 3-32 mmHg; p=0.03) compared with ESRD. Conclusion: Although patients with cardiogenic ascites exhibited elevated protein, SAAG, and both free and wedged hepatic vein pressures by TJ assessment compared to the ESRD cohort, evidence of right-sided HF was still present on echocardiogram in multiple patients of the ESRD cohort. Therefore, further characterization of hemodynamic, histologic, and advanced echocardiographic features, especially in a larger cohort size, may help improve our understanding of stratifying the risks of HF severity/prognosis as well as the etiology of nephrogenic ascites and how frequently it actually occurs without any true underlying evidence of HF.