Assessing Risk for Mortality in Trauma: The Value of Admission Hemoglobin A1c

Abstract

Recent efforts have been made to identify admission characteristics of trauma patients that are associated with increased risk of mortality. Contemporary literature has established an increased risk of mortality with admission hyperglycemia. However, the effects of longstanding hyperglycemia, as surrogated by hemoglobin A1c (HbA1c), has not been studied. A prospective trauma database was retrospectively reviewed identifying patients with collected HbA1c at admission. Three cohorts were defined by HbA1c: normal (N), <5.7; prediabetic (PD) 5.7-6.5; and diabetic (D) >6.5. Regression models were used to evaluate the risk of increased hospital and intensive care unit (ICU) length of stay (LOS), ventilator days, and mortality. Relative risk (RR) and 95% CI are provided as measures of significance. A total of 2978 patients were included in the analysis (N: n = 1895, PD: n = 744, and D: n = 339). The D cohort was more likely to be older, female, obese, suffered blunt trauma, and triaged at the highest activation acuity level (P < .0001). Mean injury severity score (ISS) was similar between groups. The D group was more likely to have longer ICU-LOS (RR 1.5; 95% CI 1.10-2.07) and ventilator days (RR 1.52; 95% CI 1.03-2.26) than the N group. Relative to the N group, the risk of mortality was 50% higher in the PD (RR 1.49; 95% CI 1.17-1.90) and in the D cohorts (RR 1.50; 95% CI 1.03-2.18). Trauma patients with an elevated admission HbA1c have a significantly higher risk of mortality regardless of their history of diabetes. These data add to the body of literature that documents the untoward effect of hyperglycemia on the trauma patient.