POINT-OF-CARE GLUCOSE TESTING IN THE NEONATAL-INTENSIVE-CARE-UNIT IS FACILITATED BY THE USE OF THE AMES GLUCOMETER ELITE ELECTROCHEMICAL GLUCOSE METER

Objective: To evaluate the Ames Glucometer Elite glucose meter for use in point-of-care glucose testing in the neonatal intensive care unit. Methods: An important part of our quality control program involves a weekly comparison of glucose values obtained with each of the seven El ite analyzers and the Beckman CX7 analyzer located in the central labo ratory. Each ''split'' sample involves measurement of the glucose by u sing the Elite analyzer in a sample of blood obtained from a heel stic k at the bedside, followed by bleeding (''milking'') 150 to 200 mu l ( 4 to 5 drops) of blood into a heparinized microcontainer. This process should take no longer than 1 to 2 minutes, whereupon the microcontain er is placed on ice and sent to the laboratory. The values obtained we re compared by regression analysis. Imprecision of the Elite meter was estimated at four levels of blood glucose concentration and on a norm al-level quality control sample used for a period of 4 months. Results : Regression analysis between the glucose values obtained on the Elite meter and the CX7 meter revealed r = 0.93, p less than 0.0001, n = 18 8, Sy/x = 0.59 mmol/L, intercept = 0.47 +/- 0.14 mmol/L(1 SEM), and sl ope = 0.91 +/- 0.028 (1 SEM). When we switched to on-ice delivery of s plit samples to reduce metabolic activity during transport of the spec imens to the laboratory, scatter about the regression line was decreas ed and the Sy/x was reduced to 0.45 mmol/L. Before the on-ice delivery of split samples, 24% of the Elite analyzer's results differed from t hose of the CX7 analyzer by more than 15%, whereas only 8% differed fr om those of the CX7 meter by more than 15% after on-ice delivery of sp lit samples. Of 30 samples read as ''Lo'' by the Elite meter, 29 were less than 2.2 mmol/L on the CX7 meter, whereas only 1 was 2.2 mmol/L. The coefficients of variation taken as a measure of imprecision were l ess than 5% for the normal-level aqueous control and less than 5% for four heel-stick blood glucose levels. Conclusions: The Ames Glucometer Elite analyzer can be used with confidence in measuring heel-stick bl ood glucose concentrations at the bedside in the neo-natal intensive c are unit. Hypoglycemic blood samples are reliably detected. As with ad ults, meticulous technique should be followed to prevent filling defec ts, and all split samples should be analyzed promptly on the CX7 analy zer, with delivery to the laboratory on ice. Unlike previous generatio ns of glucometers, the Elite meter has been well accepted by the neona tal nursing staff.