Effect of fetal hyperinsulinism on oral glucose tolerance test results in patients with gestational diabetes mellitus

OBJECTIVE: This study was undertaken to evaluate the impact of the fetoplac ental glucose steal phenomenon on the results of oral glucose tolerance tes ting in pregnancies complicated by gestational diabetes mellitus with fetal hyperinsulinism. STUDY DESIGN: This was an analysis of the cases of 34 patients with two con secutive abnormal oral glucose tolerance test results and amniotic fluid in sulin measurement before institution of insulin therapy. Patients were divi ded into groups on the basis of normal versus elevated amniotic fluid insul in concentrations. RESULTS: Oral glucose tolerance tests were done at a mean (+/-SD) of 24.9 /- 5.7 and 30.7 +/- 3.2 weeks' gestation, and amniotic fluid insulin measur ements were done at 31.1 +/- 3.2 weeks' gestation. In 13 women with gestati onal diabetes mellitus with normal amniotic fluid insulin concentration, ma ternal postload blood glucose levels at 1 hour increased by 12 mg/dL (168 v s 180 mg/dL; 9.3 vs 10.0 mmol/L; P = .0006) during the course of 6 weeks. I n contrast, in 21 women with gestational diabetes mellitus with elevated am niotic fluid insulin levels (>7 muU/mL; >42 pmol/L), 1-hour postload blood glucose levels decreased by 22 mg/dL (201 vs 179 mg/dL; 11.2 vs 9.9 mmol/L: P = .002) during the same period. The higher the amniotic fluid insulin le vel, the larger the decrease (R = 0.504; P = .02), Although low amniotic fl uid insulin levels were correlated significantly with 1-hour glucose levels of the first and second oral glucose tolerance tests, high insulin levels were no longer correlated with the second oral glucose tolerance test. CONCLUSION: Exaggerated fetal glucose siphoning may provide misleading oral glucose tolerance test results in pregnancies complicated by fetal hyperin sulinism by blunting maternal postload glucose peaks. Consequently, oral gl ucose tolerance test results in a pregnancy complicated by gestational diab etes mellitus with a fetus that already has hyperinsulinemia may erroneousl y be considered normal.