Blood glucose concentrations in prehospital trauma patients with traumatic shock: A retrospective analysis.

BACKGROUND

Deranged glucose metabolism after moderate to severe trauma with either high or low concentrations of blood glucose is associated with poorer outcome. Data on prehospital blood glucose concentrations and trauma are scarce.

OBJECTIVES

The primary aim was to describe the relationship between traumatic shock and prehospital blood glucose concentrations. The secondary aim was to determine the additional predictive value of prehospital blood glucose concentration for traumatic shock when compared with vital parameters alone.

DESIGN

Retrospective analysis of the predefined, observational database of a nationwide Helicopter Emergency Medical Service (34 bases).

SETTING

Emergency trauma patients treated by Helicopter Emergency Medical Service between 2005 and 2013 were investigated.

PATIENTS

All adult trauma patients (≥18 years) with recorded blood glucose concentrations were enrolled.

OUTCOMES

Primary outcome: upper and lower thresholds of blood glucose concentration more commonly associated with traumatic shock. Secondary outcome: additional predictive value of prehospital blood glucose concentrations when compared with vital parameters alone.

RESULTS

Of 51 936 trauma patients, 20 177 were included. In total, 220 (1.1%) patients died on scene. Hypoglycaemia (blood glucose concentration 2.8 mmol l or less) was observed in 132 (0.7%) patients, hyperglycaemia (blood glucose concentration exceeding 15 mmol l) was observed in 265 patients (1.3%). Blood glucose concentrations more than 10 mmol l (n = 1308 (6.5%)) and 2.8 mmol l or less were more common in patients with traumatic shock (P < 0.0001). The Youden index for traumatic shock ((sensitivity + specificity) - 1) was highest when blood glucose concentration was 3.35 mmol l (P < 0.001) for patients with low blood glucose concentrations and 7.75 mmol l (P < 0.001) for those with high blood glucose concentrations. In logistic regression analysis of patients with spontaneous circulation on scene, prehospital blood glucose concentrations (together with common vital parameters: Glasgow Coma Scale, heart rate, blood pressure, breathing frequency) significantly improved the prediction of traumatic shock in comparison with prediction by common vital parameters alone (P < 0.0001).

CONCLUSION

In adult trauma patients, low and high blood glucose concentrations were more common in patients with traumatic shock. Prehospital blood glucose concentration measurements in addition to common vital parameters may help identify patients at risk of traumatic shock.