Effects of Intrapartum Maternal Glucose Infusion on the Normal Fetus and Newborn

Abstract

In one retrospective study, mothers in labor who were given intravenous infusions of glucose delivered newborns with lower average blood glucose levels than did mothers who were not given intravenous glucose. In a further investigation of the subject, the authors conducted a prospective study to determine the relationships between maternal glucose infusion and fetal and neonatal insulin and blood glucose levels. In addition, they studied the subsequent neonatal neuro behavior in an otherwise healthy and normal population. Fifty-six mother-newborn pairs were studied. All mothers received intravenous glucose before delivery in accordance with standard practice. The total glucose load and the rate of glucose infusion were determined, and venous blood was obtained from the mother at the time of delivery. Umbilical venous blood glucose and insulin levels and neonatal blood glucose levels were measured. The total amount of maternal glucose administered ranged from 3.5 to 82.5 g (median, 32.5 g). The rate of glucose infusion ranged from 35 to 57 g/h (median, 8 g/h). The median time interval between the start of glucose infusion and delivery was 4.1 hours (range, 0.7 to 1 7 h). Maternal blood glucose levels at delivery had a median of 110 mg/dl (range, 65 to 230 mg/dl). The total amount of glucose infused correlated significantly (P < 0.01) with maternal blood glucose levels. The median umbilical vein glucose level was 104 mg/ dl (range, 73 to 215 mg/dl). The correlation coefficient between maternal and umbilical vein glucose levels was r = 0.94 (P < 0.001). Umbilical vein insulin concentrations ranged from 3.1 to 202 μU/ml. The umbilical vein glucose levels correlated significantly (P < 0.001) with the total glucose load and the glucose infusion rate. The umbilical vein glucose also correlated significantly with the maternal blood glucose concentration (P < 0.001). The umbilical vein insulin level was significantly related to the rate of glucose infusion in the mother (P < 0.001), as well as to maternal blood glucose and to umbilical vein glucose (P < 0.01). Six of the 56 babies (11 per cent) had hypoglycemia at 1 hour of age. The presence of low blood glucose in the newborn significantly correlated (P < 0.05) with a maternal blood glucose level of 120 mg/dl or greater. Furthermore, low neonatal glucose levels correlated (P < 0.05) with a maternal glucose infusion rate of 20 g/h or greater. Neonatal hypoglycemia also correlated significantly (P < 0.01) with umbilical vein insulin levels greater than 40 μU/ml. Hypoglycemia in all babies promptly responded to appropriate therapy. Adjustment to auditory stimulus was significantly better (P < 0.01) in hypoglycemic neonates than in the nonhy-poglycemic neonates at 4 hours of age. There were no other significant differences associated with neonatal blood glucose levels, although in hypoglycemic infants there was a tendency to diminish Moro's response and to have poorer rooting at age 4 hours. There were no differences for any neurobehavioral measure at 24 hours of age. It is recommended that the normal parturient be given less than 20 g/h of intravenous glucose before delivery and have a blood glucose level less than 120 mg/dl at the time of delivery. Newborns delivered to mothers with hyperglycemia or excessive glucose infusion should be tested for hypoglycemia at 1 and 2 hours of age.