Admission Blood Glucose in the Emergency Department is Associated with Increased In-Hospital Mortality in Nontraumatic Critically Ill Patients.

Abstract

Abnormal admission blood glucose was reported as a useful predictor of outcome in critically ill patients. To identify patients at higher risk, this study aimed to evaluate the relationship between admission blood glucose levels and patient mortality during the management of nontraumatic critically ill patients in the emergency department (ED). In this prospective, single-center observational study in a German university ED, all adult patients admitted to the resuscitation room of the ED were included between September 1, 2014 and August 31, 2015. Directly after resuscitation room admission, blood samples for admission blood glucose were taken, and adult patients were divided into groups according to predefined cut-offs between the admission blood glucose. Study endpoint was in-hospital mortality. During the study period, 532 patients were admitted to the resuscitation room. The data of 523 patients (98.3%) were available for analysis. The overall in-hospital mortality was 34.2%. In comparison with an in-hospital mortality of 25.2% at an admission blood glucose of 101-136 mg/dL (n=107), admission blood glucose of ≤ 100 mg/dL (n=25, odds ratio [OR] 6.30, 95% confidence interval [CI] 2.44-16.23, p < 0.001), 272-361 mg/dL (n=63, OR 2.53, 95% CI 1.31-4.90, p=0.007), and ≥ 362 mg/dL (n=44, OR 2.96, 95% CI 1.42-6.18, p=0.004) were associated with a higher mortality. Abnormal admission blood glucose is associated with a high in-hospital mortality. Admission blood glucose is an inexpensive and rapidly available laboratory parameter that may predict mortality and help to identify critically ill patients at risk in a general nontraumatic critically ill ED patient cohort. The breakpoint for in-hospital mortality may be an admission blood glucose ≤ 100 and ≥ 272 mg/dL.