# Retrospective cohort study of the microbiology and clinical outcomes of spontaneous bacterial peritonitis (SBP) within a US academic health system

## TL;DR

This study evaluated SBP treatment guidelines in a US health system, finding that most cases did not require broad-spectrum antibiotics.

## Contribution

The study assesses local applicability of 2021 SBP guidelines through microbiology and clinical outcomes analysis in a US academic health system.

## Key findings

- 16% of culture-positive SBP isolates were nonsusceptible to third-generation cephalosporins.
- TGC monotherapy had similar clinical outcomes as broad-spectrum therapy in SBP patients.
- No significant differences in mortality or hospitalization duration were found between treatment groups.

## Abstract

To determine the local applicability of 2021 American Academy for the Study of Liver Diseases spontaneous bacterial peritonitis (SBP) treatment guidelines by evaluating the microbiology and clinical outcomes of SBP cases in an academic health system.

Retrospective cohort study.

Five-hospital academic health system.

Hospitalized adult patients with SBP.

This study involved 2 components. First, patients meeting inclusion criteria with peritoneal fluid cultures positive for a pathogen were included in the culture-positive group. Antibiotic susceptibilities were analyzed for these patients. Second, remaining culture-negative patients were randomly selected and sequentially evaluated until the culture-negative and culture-positive groups were approximately the same size. Clinical data for all patients were evaluated based on empiric antibiotics received.

Forty-nine patients with culture-positive SBP and 48 patients with culture-negative SBP were included. Eight (16%) positive cultures contained TGC-nonsusceptible organisms. Patients receiving empiric third-generation cephalosporin (TGC) monotherapy had similar clinical outcomes as patients receiving empiric broad-spectrum therapy, including similar 30-day mortality (36% vs 38%; P = 1.00), 90-day mortality (55% vs 55%; P = 1.00), and median duration of hospitalization (6.5 d vs 8 d; P = .25). ICU admission, recent hospitalization, and nosocomial infection were not associated with TGC-nonsusceptible pathogen isolation in a univariate logistic regression analysis.

Within our health system, 16% of isolates in culture-positive SBP patients were nonsusceptible to TGCs. No statistically significant difference was detected in clinical outcomes in patients receiving TGC or broader-spectrum antimicrobial therapy.