GLUCOSE CONTENT OF TRACHEAL ASPIRATES - IMPLICATIONS FOR THE DETECTION OF TUBE-FEEDING ASPIRATION

Objective: To determine, using a sensitive glucose assay, whether moni toring of tracheal aspirate glucose concentration could serve as a mar ker of aspiration of enteral feedings. Design: Prospective, controlled trial. Setting: Intensive care units of a tertiary care hospital. Pat ients: Fifteen enterally fed and 15 nonenterally fed, tracheally intub ated patients who had normal lung fields on a routine chest radiograph . Interventions: Patients with endotracheal tubes undergoing routine t racheal suctioning had tracheal secretions collected three times per d ay with a minimum of 4 hrs between samples for up to 5 days. Daily che st radiographs were reviewed for evidence of the development of pneumo nitis using defined criteria. Measurements and Main Results: Glucose c oncentrations in five commonly used commercial feeding formulas, as we ll as in the medications patients were receiving enterally or as an or al wash, were measured. Tracheal secretion glucose concentrations were 66 +/- 54 (SD) mg/dL (3.7 +/- 3.0 mmol/L) and 105 +/- 70 mg/dL (5.8 /- 3.9 mmol/L) in the enterally fed and nonenterally fed patients, res pectively (p = NS). Of the medications administered, the majority cont ained negligible glucose, but ten had >3 mg/dose of glucose. However, there was no correlation between administration of these medications a nd the tracheal glucose concentrations. Tracheal glucose concentration s were similar in patients who received medications containing glucose and patients who received either no medications or medications with n egligible glucose content. A small but significant correlation between blood glucose and tracheal secretion glucose concentrations (r(2) = . 15, p<.05) was observed. None of the patients developed aspiration pne umonitis. Glucose concentrations in widely used commercial formulas (4 4 to 202 mg/dL; 2.4 to 11.2 mmol/L) overlapped considerably with gluco se concentrations in tracheal secretions in the absence of aspiration and were for the most part within 2 so of mean values in tracheal secr etions. Conclusions: Tracheal secretions contain high glucose concentr ations, both in enterally fed patients without evidence of aspiration pneumonitis and in nonenterally fed patients. The concentration of glu cose in tracheal secretions appears to be determined, in part, by ambi ent extracellular glucose concentrations. We conclude that measurement of glucose in tracheal secretions is unlikely to be useful in monitor ing for tube feeding aspiration in tracheally intubated, enterally fed patients.