Effect of new onset of jaundice on outcomes in patients with sepsis: A population-based propensity score-matched study

Abstract

Background Although jaundice have been associated with mortality in critically ill patients, yet no data from large studies demonstrates its effect in sepsis. The aim of the study is to investigate the impact of jaundice on outcomes in septic patients.Methods Propensity-matched analysis of cohort database from Medical Information Mart for Intensive Care III (MIMIC-III), a large database of septic patients at a tertiary care hospital in Boston, Massachusetts (June 2001 to October 2012). Individuals with preexisting jaundice or liver diseases at the time of admission were excluded from analysis. Jaundice was diagnosed in septic patients with total bilirubin levels >2mg/dL at any time during hospitalization. The multivariate Cox was employed to adjust for baseline and confounding parameters.Results A total of 2784 septic patients were enrolled in the study, including 456 patients in jaundice group, and 2328 in non-jaundice group. Before propensity score matching, multivariate Cox hazard analysis showed age [HR 1.029; 95% CI (1.009-1.049); P=0.005], malignancy [HR 3.244; 95% CI (1.729h-6.085); P<0.001], SOFA score [HR 1.179; 95% CI (1.054-1.318); P=0.004], serum total bilirubin at hospital discharge [HR 1.050; 95% CI (1.022-1.079); P<0.001] were the independent risk factors of mortality in sepsis. In 432 pairs after matching according to new presented jaundice status, jaundice group had higher in-hospital mortality (76±17.6 vs. 49±11.3; P=0.012) than non-jaundice group. In multivariate logistic regression model, the only independent risk for jaundice in sepsis was SOFA score [OR 1.314; 95% CI (1.248-1.385); P<0.001], whereas mechanical ventilation [OR 0.310; 95% CI (0.222-0.432); P<0.001], serum platelet [OR 0.998; 95% CI (0.997-1.00); P=0.015] and serum bicarbonate [OR 0.962; 95% CI (0.929-0.996); P=0.030] were jaundice’s independent protective factors.Conclusions New onset of jaundice is associated with higher risk rate of in-hospital mortality in sepsis. In addition, our study demonstrates that serum total bilirubin at hospital discharge is an independent determinant for mortality.