Characterizing the Benefits of Serial Ascitic Fluid Testing in Ambulatory Patients with Cirrhosis: A National Cohort Study.

Abstract

Background and aims: Spontaneous bacterial peritonitis (SBP) is a life-threatening cirrhosis complication, yet guidance for sending fluid studies from outpatient paracentesis remains limited with unclear clinical impact. We aimed to characterize variation in ascitic fluid testing practices during outpatient paracentesis in cirrhosis and evaluate its association with short-term hospitalization and mortality.

Methods: We conducted a retrospective cohort study of 60,619 outpatient paracenteses among 4,467 patients with cirrhosis in the Veterans Health Administration from 2012-2023. Testing frequency was defined as the proportion of procedures in which diagnostic fluid studies for SBP were sent. Inverse probability treatment weighting and marginal proportional rates models were used to estimate associations with 30-day outcomes. Exploratory analyses evaluated interactions with Child-Turcotte-Pugh (CTP) score.

Results: Diagnostic fluid studies were performed in 32.7% of outpatient paracenteses, however testing frequency varied by region, provider type, and facility. Compared to intermediate-frequency testing, both always (100%) and never (0%) testing strategies demonstrated lower 7-day hospitalization risk. In final models, patients tested at every paracentesis had the lowest rates of hospitalization (RR 0.65 vs. 50% testing frequency, 95% CI 0.58-0.72) and 30-day mortality (p<0.001), with associations favoring universal testing particularly among patients with higher CTP scores. These patients were also more likely to receive outpatient SBP antibiotics.

Conclusion: Ascitic fluid testing practices for outpatient paracentesis vary substantially and are associated with differences in outcomes. Universal testing was associated with lower short-term hospitalization and mortality among patients with higher CTP scores, although these observational findings may reflect differences in patient selection and care delivery. Evaluation of standardized, risk-adapted testing protocols is warranted.