835 RELATIONSHIP BETWEEN DEXTROSTIX, PLASMA AND WHOLE BLOOD GLUCOSE IN THE RANGE 0 - 100 MG/DL

Abstract

Two laboratory measurements frequently made in the newborn period are Dextrostix (Ames) and hematocrit; yet the relationship between Dextrostix, plasma glucose and whole blood glucose values has not been described. Umbilical vein and peripheral vein blood samples were studied for whole blood, plasma and red cell glucose (YSI Model 23A) and were compared to values obtained using the Dextrostix read by eye. Dextrostix predicted whole blood glucose (R = .83, slope of regression = 1.1, not significantly different from 1.0). Plasma glucose was underestimated by the Dextrostix (R = .64, slope = .70 in the range measured in vivo (30-100 mg.dl−1). Measurements in vitro were carried out using umbilical vein blood during incubation with and without added glucose to determine the relationship between plasma and whole blood glucose at the lower glucose levels. No statistically significant difference was observed between plasma and whole blood glucose at these low levels. Dextrostix reflected both of these levels. No clinically significant error would have been made using the Dextrostix. In vivo, red cell glucose fell more quickly than plasma glucose at values greater than 100. Below that level, plasma and red cell glucose fell at similar rates. However, preliminary data from infants with polycythemia/hyperviscosity reveal red cell glucose values lower than would be predicted by the plasma glucose value alone.