Abstract

Background

Patients with cirrhosis have high risk of bacterial infections and cirrhosis decompensation, resulting in admission to emergency department (ED). However, there are no criteria developed in the ED to identify patients with cirrhosis with bacterial infection and with high mortality risk.

Study objective

The objective of the study is to identify variables from ED arrival associated with bacterial infections and inhospital mortality.

Methods

This is a retrospective single-center study using a tertiary hospital's database to identify consecutive ED patients with decompensated cirrhosis. Clinical variables and laboratory results were obtained by chart review. Logistic regression models were built to determine variables independently associated with bacterial infection and mortality. Scores using these variables were designed.

Results

One hundred forty-nine patients were enrolled, most of them males (77.9%) with alcoholic cirrhosis (53%) and advanced liver disease (Child-Pugh C, 47.2%). Bacterial infections were diagnosed in 72 patients (48.3%), and 36 (24.2%) died during hospital stay. Variables independently associated with bacterial infection were lymphocytes less than or equal to 900/mm³ (odds ratio [OR], 3.85 [95% confidence interval {CI}, 1.47-10]; P = .006) and C-reactive protein greater than 59.4 mg/L (OR, 5.05 [95% CI, 1.93-13.2]; P = .001). Variables independently associated with mortality were creatinine greater than 1.5 mg/dL (OR, 4.35 [95% CI, 1.87-10.1]; P = .001) and international normalized ratio greater than 1.65 (OR, 3.71 [95% CI, 1.6-8.61]; P = .002). Scores designed to predict bacterial infection and mortality (Mortality in Cirrhosis Emergency Department Score) had an area under the receiver operating characteristic curve of 0.82 and 0.801, respectively. The Mortality in Cirrhosis Emergency Department Score performed better than Model for End-Stage Liver Disease score.

Conclusions

In this cohort of ED patients with decompensated cirrhosis, lymphopenia and elevated C-reactive protein were related to bacterial infections, and elevated creatinine and international normalized ratio were related to mortality. Scores built with these variables should be prospectively validated.