There is evidence that hyperinsulinemia and insulin resistance play a role in the development of hypertension. Accordingly, in our ongoing longitudinal study of pregnancy-induced hypertension, we have measured fasting levels of insulin and glucose at 18 to 25 weeks gestation in 140 nulliparous African-American women followed prospectively to delivery. To test the hypothesis that hyperinsulinemia may be related to the development of preeclampsia, discriminant analysis of mean arterial pressure (MAP), fasting plasma insulin levels, insulin to glucose ratios, and left lateral forearm vascular resistance were examined as predictors of preeclampsia. Statistical analysis controlled for two factors known to be related to insulin levels, gestational age and pregestational body mass index. Gestational hypertensives were not different with regard to blood pressure and metabolic factors from normals and thus were placed in the control group. Women who subsequently developed preeclampsia had mean (+/-SE) fasting plasma insulin levels of 51.0 +/- 12.0 mu U/mL at 20 weeks and controls had values of 29.0 +/- 2.8. Only MAP [F(4,135) = 8.8, P < .01] and insulin [F(1,135) = 6.5, P < .05] were related to the development of preclampsia [F(4,135) = 4.39, R(2) = 11.5%]. The finding that elevated second-trimester insulin levels characterize the subsequent development of preeclampsia with control for increased MAP supports the hypothesis that hyperinsulinemia and associated insulin resistance may contribute to the pathogenesis of preeclampsia.