The empiric use of hypertonic dextrose in patients with altered mental status: A reappraisal

Abstract

Neither response to hypertonic glucose (D50W) nor presence of hypoglycemia can be reliably predicted by "typical" clinical findings (tachycardia, diaphoresis, and/or an available history of diabetes mellitus) in prehospital patients with altered mental status (AMS). Three hundred forty consecutive patients who received D50W for prehospital AMS as ordered by a university hospital paramedic base. Review of prehospital records and tape recordings for all subjects to determine presence or absence of tachycardia, diaphoresis, and/or available history of diabetes mellitus at the time of field presentation, as well as response to D50W; final diagnosis was determined from emergency department charts in the 301 patients for whom they were available. Twenty-five patients (7.4%) had a complete response to D50W: 20 had hypoglycemia, diagnosis for one patient was unknown, and four had other causes of AMS. Three patients with an ED diagnosis of hypoglycemia had partial or equivocal responses to D50W, and five had no response. Complete responders were more likely than other patients to have diaphoresis (40% vs 13%, P less than .001) and available history of diabetes (52% vs 12.1%, P less than .001) but not tachycardia (36% vs 35.2%, P = NS). Nine complete responders, including five with hypoglycemia, had none of the three clinical signs. Although patients with hypoglycemia who respond to D50W are diaphoretic and have an available history of diabetes more often than other patients with prehospital AMS, 25% of complete responders who are hypoglycemic would not receive D50W if it were used only in patients with "typical" clinical findings. Selective use of D50W for AMS is desirable because very few patients respond, but it is only feasible with concomitant field use of a rapid test of serum glucose.