Ds. Louder et al., DETECTING HYPOGLYCEMIA IN VENOUS-BLOOD - COMPARISON OF 2 NEONATAL BEDSIDE GLUCOSE MONITORS, Laboratory medicine, 27(3), 1996, pp. 184-187
We evaluated the performance of the One Touch II Hospital in neonatal
mode (Lifescan, Militipas Calif) and Accu-Chek II (Boehringer Mannheim
, Indianapolis) in screening for neonatal hypoglycemia. We used blood
with low-to-normal glucose concentrations and a wide range of hematocr
its to mimic infant blood. Single-donor adult blood was collected in h
eparinized tubes and allowed to metabolize overnight. Red blood cells
were separated and reconstituted With the plasma, normal saline, and 1
0% dextrose to 5 hematocrits and 6 approximate plasma glucose concentr
ations. All 30 combinations of hematocrit and glucose concentration we
re tested. The One Touch II Hospital glucose monitor showed the best a
ccuracy of numeric results (32%-78% vs 11%-54%), precision (pooled sta
ndard deviation, 2.4 mg/dL vs 4.6-13.9 mg/dL), and correlation (Pearso
n's r, 0.95-0.98 vs 0.82-0.86). Sensitivity and specificity for hypogl
ycemic glucose levels were statistically similar.