Precision and accuracy: Comparison of point-of-care and laboratory glucose concentrations in cardiothoracic surgery patients

BACKGROUND:: There have been variable results on the practice of tight glycemic control, and studies have demonstrated that point-of-care (POC) glucometers have variable accuracy. OBJECTIVE:: The purpose of this study was to determine the difference between blood glucose concentrations obtained from POC glucometers as compared with laboratory results in cardiothoracic surgery (CTS) patients. METHODS:: This was a descriptive study on a convenience sample of 46 CTS patients. A single sample of arterial blood was collected and analyzed at the bedside with the POC glucometer and in the laboratory to obtain a serum glucose concentration and hematocrit (Hct). A paired t test was used to compare the mean differences along with Spearman ρ correlation to examine the relationship between difference scores and Hct. RESULTS:: The POC glucose was significantly higher than the laboratory result (t = 8.5, P < .001) with a mean of 12.3 (SD, 9.8) mg/dL. Spearman ρ correlation between the difference scores and Hct was -0.43, P = .003. Using a tercile split, groups with Hct of less than 26% (n = 16) and greater than 29% (n = 15) were identified. The unpaired t test on the mean difference scores of these 2 groups was t = -2.7, P < .01, with an overall mean difference 8.6 mg/dL (95% confidence interval, -15 to -2.2). The mean difference was 16.3 in the low-Hct group and 7.8 in the high-Hct group. CONCLUSIONS:: Point-of-care glucometer results differ significantly from laboratory glucose concentrations, with the difference widening as the Hct decreases. This raises the concern about using POC devices to provide tight glycemic control in CTS patients.