Strict normalization of blood glucose level improves outcome in critically ill patients. Accurate glucose measurement is an essential aspect of intensive insulin therapy. The various methods of testing (glucometer, blood gas analyzer and conventional laboratory analyzer) have not been studied simultaneously in patients in shock. This study is designed to prospectively evaluate different methods of glucose monitoring in critically ill patients. The specific aim is to evaluate the accuracy of bedside glucometer and blood gas analyzer in comparison with the clinical laboratory colorimetric method (gold standard).
ICU patients in shock, defined as a systolic blood pressure of <90 mm Hg despite adequate volume resuscitation or requiring vasopressor therapy, were considered for enrollment. Arterial, venous and capillary blood samples were obtained simultaneously. A total of 243 samples were obtained from 21 patients. Glucose determinations were made with the glucometer (ACCU-CHEK Comfort Curve, Roche) from each of the vascular compartments. Arterial and venous glucose levels were determined using a blood gas analyzer (Radiometer ABL 700). Arterial and venous blood samples were tested in the clinical laboratory using the colorimetric plasma glucose analyzer (VITRIOS).
Mean venous blood glucose level determined on the VITRIOS analyzer (gold standard) was 129.6 mg/dl with a range of 54 to 350 mg/dl. Capillary blood glucose tested on glucometer was higher than the gold standard by a mean of 20.95 mg/dl (16.99%). The difference changed minimally when arterial or venous samples tested on glucometer were compared to the gold standard. Blood gas analyzer on the other hand, was higher by a mean of 3.07 mg/dl (2.68%) when compared to the gold standard.
Blood glucose determination with a glucometer is associated with a risk of obtaining falsely elevated blood glucose level. Blood gas analyzer is significantly more accurate.
Using glucometers to monitor blood glucose levels in patients with shock is associated with the risk of obtaining falsely elevated results and thus placing the patient at risk for hypoglycemia.
Comparison of the glucometer and blood gas analyzer to the gold standard (clinical lab).ComparisonMean of difference (mg/dl)Standard deviationp ValueArterial sample on glucometer vs clinical lab17.662512.1435< 0.001Arterial sample on blood gas analyzer vs clinical lab2.214.5387< 0.001Venous sample on glucometer vs clinical lab22.17517.7184< 0.001Venous sample on blood gas analyzer vs clinical lab4.46255.5573< 0.001Capillary sample on glucometer vs venous sample in clinical lab20.112.5290< 0.001
Srinivas Chakravarthy, Grant monies (from sources other than industry) This study was supported by an award from The CHEST Foundation of the American College of Chest Physicians and Ortho Biotech Products, LP.; Grant monies (from industry related sources) The glucometer, chemistry strips and the reagents were provided by Roche.