Plasma glucagon-like peptide 1 was associated with hospital-acquired infections and long-term mortality in burn patients

Abstract

BackgroundAlthough glucagon-like peptide 1 levels have been closely associated with inflammation and mortality in septic patients, the clinical importance of glucagon-like peptide 1 on hospital-acquired infections and long-term mortality after burn injury remains unexplored. MethodsPlasma samples from 144 burn patients were collected on admission to determine total glucagon-like peptide 1, interleukin 6, and monocyte chemotactic protein-1 levels. Hospital-acquired infections were determined by positive microbial culture. One-year mortality was assessed by telephone interview. Factors associated with glucagon-like peptide 1 were determined by multivariable linear logistic regression. Predicting the clinical importance of glucagon-like peptide 1 on the development of hospital-acquired infections and mortality were determined by Cox proportional hazards models and further by receiver operating characteristic curve analysis. Kaplan-Meier analyses were performed to examine whether the mean glucagon-like peptide 1 level of the cohort could discriminate the hospital-acquired infections-free survival. ResultsMedian burn size was 41% (19%−70%) of total body surface area. Hospital-acquired infections developed in 36 (25%) patients after a mean of 10 ± 1 days after injury. Interleukin 6, monocyte chemotactic protein-1, and blood urea nitrogen levels and thrombin time were independently associated with increased glucagon-like peptide 1 levels. Levels of glucagon-like peptide 1 (median, interquartile range) were greater in patients who developed hospital-acquired infections than in those who did not (237 pmol/L, 76−524 vs 80 pmol/L, 51−158; P < .001) and in patients who died (536 pmol/L, interquartile range: 336−891 pmol vs 98 pmol/L, 47−189; P < .001). Although the glucagon-like peptide 1 level could not predict hospital-acquired infections-free survival in individual patients, it could predict 1-year mortality independently (P = .021). Moreover, a glucagon-like peptide 1 level of 200 pmol/L could discriminate hospital-acquired infections-free survival (P < .001). ConclusionAdmission glucagon-like peptide 1 level can discriminate hospital-acquired infections-free survival and predict long-term mortality in a group of patients with burn injury. Our data suggests that glucagon-like peptide 1 may be a predictive biomarker for hospital-acquired infections and mortality in burn patients.