Abstract
BACKGROUND: Abnormal admission blood glucose levels were proved to have a mortality predictive value in polytraumatized patients, as reported by studies in developed countries. Reports from developing countries are scarce. OBJECTIVE: To evaluate the reliability of on-admission blood glucose levels in predicting mortality in polytraumatized patients presented to a North African (developing country) trauma center. The secondary objectives were to investigate other possible mortality predictors and if a cutoff value for each could be obtained. METHODS: In this prospective longitudinal study, over one year, we included adult (≥18 years) patients who were polytraumatized (ISS ≥17) and presented to our trauma center within six hours of the trauma incident. Various clinical, laboratory, and trauma scores were collected. Blood glucose levels were assessed from blood samples obtained directly after admission. Patients were divided into five groups based on the admission blood glucose levels. RESULTS: We included 202 patients, having a mean age of 44±13.9 (20 to 70) years, and 52% were females. The mortality rate was 10.9% (including all patients presented with blood glucose levels≥15 mmol/L). The following were significant mortality predictors, admission blood glucose (OR=3.31, 95% CI=1.902-5.763, p<0.001), serum lactate levels (OR=4.017, 95% CI=1.627-9.917, p=0.003), length of hospital stay (OR=1.18, 95% CI= 1.058-1.305, p=0.003), RTS score (OR=1.43, 95% CI=1.023-2.005, p=0.037), and TRISS score (OR=1.099, 95% CI=1.052-1.148, p<0.001). Admission blood glucose levels cutoff value of 15 mmol/L can significantly differentiate between survivors and non-survivors with sensitivity, specificity, PPV, and NPV of 86.4%, 100%, 100%, and 88%, respectively. CONCLUSION: Abnormal admission blood glucose with a cutoff value of 15mmol/L is a significant mortality predictor in polytraumatized patients from developing country trauma center, among other clinical, laboratory, and trauma scores parameters.