This study was conducted to test the hypothesis that aerobic conditioning prevents exercise-induced hypoglycemia and preserves the capacity to utilize carbohydrates and to produce lactate during heavy exercise in late gestation. The effects of closely monitored cycle ergometer conditioning (heart rate = 143 +/- 2 beats/min, 25 min/day, 3 days/week) during the second and third trimesters were studied in 18 previously sedentary women (exercised group, EG). A nonexercising pregnant control group (CG, n = 9) was also studied. Data collection times for both groups were as follows: start of the second trimester (Entry), ends of the second (TM2) and third (TM3) trimesters (post-training), and 4-6 months postpartum (nonpregnant control). Respiratory gas exchange was studied and venous blood samples were obtained before, during, and after a graded cycle ergometer test that was terminated at a peak heart rate of 170 beats/min. Measurements included plasma glucose, insulin, free fatty acids, the respiratory exchange ratio at peak exercise, and peak postexercise lactate concentration. A significant aerobic conditioning effect in the EG was confirmed by a 17% increase in O2 pulse at peak exercise between Entry and TM3. As expected, values for free fatty acids in the CG rose with advancing gestational age. The CG showed a clear trend for a rise in plasma insulin with advancing gestational age, under all experimental conditions. Also, peak exercise respiratory exchange ratio and peak postexercise lactate concentration were significantly reduced in late gestation, and plasma glucose decreased significantly during and following the end of TM3 testing. Effects of pregnancy to reduce peak postexercise lactate and to reduce plasma glucose during and after exercise at the end of the third trimester were significantly attenuated in the EG. These effects were attributed to attenuation of pregnancy-induced insulin resistance (as reflected by insulin/glucose ratio) by physical conditioning. These findings support our original experimental hypothesis that aerobic conditioning prevents exercise-induced hypoglycemia and preserves the ability to utilize carbohydrate and produce lactate during heavy exercise in late gestation.