• E A Bartkus, M Daya, J R Hedges, and J Jui.
• Department of Emergency Medicine, Oregon Health Sciences University, Portland 97201.
• Prehosp Disaster Med. 1993 Jul 1;8(3):217-27.

IntroductionCurrent prehospital protocols for the management of patients with altered mental status include the empiric administration of hypertonic glucose, naloxone, and thiamine. The injudicious use of 50% dextrose (D50W) may result in hyperosmolarity, a worsening of hypokalemia, and unwarranted additional health-care costs for the patient. The administration of D50W also may worsen the neurological outcome of patients with local or generalized ischemia.ObjectiveTo evaluate the ExacTech blood glucose meter's ability to estimate blood glucose levels accurately and rapidly.MethodsEmergency medical technicians (EMTs) from selected advanced life support (ALS) units in the Portland, Ore., metropolitan area participated in a prospective clinical trial of the ExacTech blood glucose meter. A convenience sample was drawn from emergency medical services (EMS) patients with suspected diabetic emergencies, altered mental status, and other neurological deficits. Venous blood samples were drawn from these populations at the same time as the ExacTech readings were obtained. The venous blood was submitted to the receiving hospitals for laboratory analysis of blood glucose levels, and a comparison was made between the results of the two methods.ResultsA total of 80 matched sets of data were obtained from 1 April 1990 through 6 May 1991. The hospital blood glucose values ranged from 8 to 1233 mg/dl. Sixteen (20%) of the patients were hypoglycemic (&.lt.60 mg/dl) and 23 (28.8%) were hyperglycemic ( greater than 180 mg/dl). The ExacTech device sensitivity and specificity for hypoglycemia using venous samples were 94.6% and 89.2%, respectively. For hyperglycemia, these same parameters were 87.5% and 97.1%. Pearson's r over the range of the instrument (40-450 mg/dl) was 0.8656 (p less than .001). If the prehospital "definition" of hypoglycemia (for threshold-to-treat) is raised to 65 mg/dl, the device has 100% sensitivity in the sample population.ConclusionThe device functioned accurately and consistently in the prehospital environment over a wide range of temperatures, and in the hands of many different individuals.

19 keywords...

hide…