Clinical outcomes and care for spontaneous bacterial peritonitis: A national cohort study.

Abstract

BACKGROUND AND AIMS: Spontaneous bacterial peritonitis (SBP) leads to high rates of acute kidney injury (AKI), hepatorenal syndrome, and mortality. Population-based studies on contemporary SBP epidemiology are needed to inform care. In a large, national cohort of patients diagnosed with SBP and confirmed by ascitic fluid criteria, we characterized ascitic fluid characteristics, in-hospital and 12-month mortality, AKI, and recurrent SBP. APPROACH AND RESULTS: We investigated how individual and bundled quality measures for SBP associated with outcomes after multi-level adjustment for health-system, patient clinical factors, and quality measures. Individual and bundled quality metrics were inpatient antibiotics within 48 hours, i.v. albumin, repeat paracentesis within 48 hours, recognition of SBP, and prophylactic antibiotics upon discharge. Among 4330 patients with newly diagnosed SBP, in-hospital mortality was 15.5%, and 12-month mortality was 56.6%. The incidence of stage 1 AKI was 26.6%, 15.7% for stage 2, and 22.8% for stage 3. The cumulative incidence of recurrent SBP was 10.3%. Guideline-recommended albumin was the only individual metric associated with reduced in-hospital mortality (HR: 0.73, 95% CI: 0.59-0.91). Receipt of a higher number of metrics from the SBP bundle was associated with progressively lower 12-month post-discharge mortality: patients who received 3, 4, and 5 SBP bundle components had 20%, 38%, and 56% lower hazard of mortality, respectively, relative to those receiving 2 or fewer (all p <0.001). The SBP bundle was associated with a lower incidence of stage 3 versus stage 0-2 AKI (OR: 0.66, 95% CI: 0.51-0.86). CONCLUSIONS: Prospective implementation of evidence-based SBP bundles may improve care outcomes and mortality in SBP.