Influence of blood sample oxygen tension on blood glucose concentration measured using an enzyme-electrode method.

Abstract

To determine the accuracy of a bedside glucometer with an enzyme-electrode sensor based on enzyme oxidation by glucose oxidase. Prospective, cross-sectional clinical study. Operating room in a public hospital. Fifty-four patients undergoing surgical procedures for a derivation (n = 17) and a validation (n= 37) study. Arterial blood samples were obtained via a 20-gauge cannula inserted into each patient's radial artery. Glucose measurements and arterial blood gas analyses were concurrently performed, using 48 blood samples for the derivation study and 45 blood samples for the validation study of this technique. Blood glucose concentrations were measured with both a bedside glucometer using an enzyme-electrode method and a laboratory glucometer based on the colorimetric method. The bedside glucometer consistently underestimated the glucose concentrations and the underestimation was related to the sample oxygen tension but not to hematocrit, plasma protein, creatinine, uric acid, or bilirubin. The present investigation used the following correction formula: (corrected glucose value) = (glucose concentration obtained by a bedside glucometer) + 0.1 x (sample oxygen tension) + 16. The corrected data were in agreement with the laboratory-determined glucose values (i.e., the mean difference and precision were 0.4 and 7.1 mg/dL, respectively). A validation study confirmed the generalization of the present correction formula which facilitates a more accurate estimation of blood glucose concentrations. Blood glucose values measured using a bedside glucometer in this study were influenced by the sample oxygen tension. We used a corrective equation which improved the accuracy of estimating blood glucose values to a clinically acceptable range.