VANCOUVER, B.C. – Despite agreeing to use birth control and receiving frequent counseling about pregnancy avoidance, a sizable share of teens with type 2 diabetes mellitus become pregnant, and these pregnancies often have poor outcomes, researchers reported at the World Diabetes Congress.
The analysis was based on 452 female participants in the national Treatment Options for Type 2 Diabetes in Adolescents and Youth (TODAY) study, the largest trial in youth with this form of diabetes to date and one designed to have good representation of various racial/ethnic groups.
Overall, 10% of the girls became pregnant during a period of up to 6.5 years. The majority were not using contraception and did not recall the counseling. More than one-fourth of the pregnancies ended in fetal loss or stillbirth. And one-fifth of live-born infants had major congenital anomalies.
“We need to better understand the reasons for pregnancy in youth with type 2 diabetes and why, despite counseling, they become pregnant,” commented first author Dr. Kristen J. Nadeau of the division of pediatric endocrinology, department of pediatrics, University of Colorado, Aurora.
“Best practices for metabolically unhealthy pregnancy prevention in type 2 diabetic teens also requires further study,” she added. “Long-acting contraception, we think, currently is the best method and likely the way to go in these girls who are not retaining the education and [adhering to] the behaviors that we are hoping for.”
“This study is critically important and horribly depressing,” commented session comoderator Dr. Robert E. Ratner, professor of medicine at Georgetown University and senior research scientist at the MedStar Health Research Institute, Washington, as well as chief scientific and medical officer of the American Diabetes Association, Alexandria, Va.
“It points out the difficulties of being a teenager and the even greater difficulties of being a teenager with a chronic disease,” he said in an interview, adding that there are no easy solutions.
“The combination of a teen pregnancy in someone with type 2 diabetes creates the perfect storm,” Dr. Nadeau noted. “Diabetes control is worse in adolescence than in any other time in the lifespan, and with increasing rates of type 2 diabetes in youth, increases in a teen pregnancy that are complicated by diabetes are anticipated,” she added.
At baseline, participants in TODAY were 10-17 years old, were overweight or obese, and had a diabetes duration of less than 2 years and a hemoglobin A1c level of less than 8%. They were randomized to three treatment arms (metformin alone, metformin plus rosiglitazone, or metformin plus an intensive lifestyle program).
Consent for the trial required the use of a birth control method, including abstinence; in addition, every 2-3 months, the girls received diabetes education and counseling to defer pregnancy until their HbA1c level fell below 6%. They also had regular pregnancy testing, and those with a positive result were taken off their trial medication and referred to a maternal-fetal medicine specialist.
The results reported at the meeting and simultaneously published in Diabetes Care (doi: 10.2337/dc15-1206) showed that 46 (10.2%) of the girls had 63 pregnancies. On average, they were 18 years old at the time of a first pregnancy.
Despite the counseling, only about 5% of those teens who became pregnant reported that they had been using contraception. Moreover, just 13% recalled the counseling.
The median body mass index closest to conception was 35.2 kg/m2, and the median HbA1c level was 7%. “Because of the fact that we were so heavily monitoring these girls, we had their HbA1c under better control than is typical. In our typical clinic, the mean is more like 8.5%-9%,” Dr. Nadeau noted.
Relative to peers who did not become pregnant, those who did were significantly older, were less likely to be living with both parents or their mother, and had a lower household income.
Seven of the 63 pregnancies were electively terminated. Of the 53 remaining pregnancies with data, 12 ended in pregnancy loss and 2 ended in a stillbirth.
Among the 39 live-born infants, 6 were preterm and 8 had major congenital anomalies. And among the 37 with known birth weight, 10 were either small or large for gestational age.
Girls randomized to metformin plus rosiglitazone had a higher rate of term normal births than peers in the other arms (P = .027). “Of note, none of the participants reported taking the rosiglitazone after the pregnancy was discovered, as was per the study protocol,” Dr. Nadeau commented. In contrast, neither maternal body mass index nor – surprisingly – HbA1c level was significantly associated with pregnancy outcome.