Abstract
OBJECTIVE: Evidence suggests that prenatal alcohol exposure (PAE) may adversely impact insulin production and signaling but there is limited information on the range of these effects and their future health consequences. METHOD: A prospective cohort of predominantly African-American individuals identified while in utero and followed into adulthood were used to evaluate differences in various indicators of diabetes, including fasting plasma glucose, hemoglobin A1c (HbA1c), and insulin levels. The homeostatic model assessment of insulin resistance (HOMA-IR) was also computed. Body mass index (BMI) was calculated and normal defined as < 25 kg/m(2) . Participants were categorized as having PAE (n = 39) if their mothers drank at least 1 ounce of absolute alcohol per week or more during the 1(st) trimester of pregnancy and as Controls (n = 22) if their mothers reported abstaining from alcohol consumption during pregnancy. RESULTS: Mean age of the sample was 36.0 ± 1.5 years. Indices of glucose metabolism, including fasting plasma glucose and hemoglobin A1c levels, did not vary by group status but insulin levels and HOMA-IR values varied by group status and BMI level. PAE individuals with a normal BMI had lower insulin levels than Controls. However, in PAE subjects, there was a steeper increase in insulin levels relative to their BMI than in Control subjects. A cluster of 5 PAE cases had low levels of insulin and 4 of the 5 had severe cognitive impairment. CONCLUSIONS: The bidirectional effects on insulin level and insulin resistance associated with PAE may indicate differential rates of diabetes disease impact or differential PAE impact in the brain and peripheral areas involved in insulin production and signaling. These alterations may contribute to the metabolic disease risk associated with PAE.