To determine the influence of pregnancy on insulin sensitivity in patients with type 1 diabetes mellitus in more detail, a hyperinsulinemic euglycemic clamp study was performed in six pregnant type 1 diabetic women and eight nonpregnant women with type 1 diabetes mellitus. All of the pregnant women were studied three times: in early pregnancy (mean, week 13), late pregnancy (mean, week 34), and within a week after delivery. Insulin was infused in a constant rate of 1.0 mU/kg X min, which resulted in steady state serum free insulin levels (I) of 44 +/- 3 (+/- SEM), 56.6 +/- 6, and 55 +/- 8 microU/ml in the pregnant diabetic women and 52 +/- 4 microU/ml in the nonpregnant women. Mean glucose disposal (M) was 5.6 +/- 0.3 mg/kg X min early in pregnancy and 3.4 +/- 0.5 mg/kg X min late in pregnancy (P less than 0.02). However, in the early postpartum period, M was again higher (7.2 +/- 0.7 mg/kg X min; P less than 0.02) and similar to values in early pregnancy and nonpregnant diabetic women (7.2 +/- 0.6 mg/kg X min). When tissue sensitivity to insulin was expressed as the M to I ratio, similar results were obtained (nonpregnant women, early stage of gestation, and postpartum vs. late stage of gestation: 0.13 +/- 0.01, 0.13 +/- 0.01, and 0.15 +/- 0.03 mg/kg X min per microU/ml vs. 0.06 +/- 0.1 mg/kg X min per microU/ml; P less than 0.03 in all). There tended to be an inverse relationship between serum levels of human placental lactogen and the M to I ratio during pregnancy (r = -0.74; P = 0.09). However, we found no association between changes in the impairment of insulin action and serum estradiol, progesterone, or cortisol levels. In conclusion, pregnant type 1 diabetic women have insulin resistance in peripheral tissues in the late stage of gestation. Insulin sensitivity returns to values found in nonpregnant diabetic women within the first week after delivery.