Women who took antipsychotic drugs during pregnancy - including those drugs thought to have less potential for unwanted metabolic effects - were at increased risk for developing diabetes and giving birth to small infants, researchers said.

Analysis of Swedish national birth and health registry data showed that, relative to mothers-to-be not taking antipsychotic drugs, the risk of gestational diabetes was nearly doubled, according to Robert Bodén, MD, PhD, of Uppsala University in Uppsala, Sweden, and colleagues.

The researchers also found that certain antipsychotics were associated with unusually large heads in newborns, after adjusting for maternal characteristics such as smoking status, they reported in the July issue of Archives of General Psychiatry.

“Pregnant women treated with antipsychotics should be closely monitored for gestational diabetes and deviating fetal growth,” Bodén and colleagues concluded.

In their report, they noted that research on antipsychotics during pregnancy has been “lacking or weak,” with little to go on even from observational studies. (In the U.S., antipsychotics are generally in pregnancy class C, indicating adverse fetal effects in animal studies but no solid evidence in human studies.)

But some of the side effects of newer antipsychotics - including weight gain, insulin resistance, and hyperlipidemia - would reasonably raise concern about risk of gestational diabetes in pregnant women as well as birth defects, Bodén and colleagues suggested.

They examined registry data on nearly 360,000 Swedish women giving birth from July 2005 to December 2009.

Antipsychotic drug use in pregnancy was uncommon, with only 507 women recorded as having taken them.

Among them were 169 who took the two such drugs most commonly associated with adverse metabolic effects, olanzapine (Zyprexa) and clozapine (Clozaril). The other 338 took other antipsychotics. Outcomes associated with the two groups of agents were analyzed separately.

After adjusting for maternal age, country of origin, smoking, height, cohabitation status, and number of previous pregnancies, antipsychotics in both groups were associated with about double the risk of gestational diabetes compared with no antipsychotic use. Odds ratios were as follows:

  Olanzapine and/or clozapine: 1.94 (95% CI 0.97 to 3.91)
  Other antipsychotics: 1.77 (95% CI 1.04 to 3.03)

Both odds ratios shrank and became nonsignificant when body mass index in early pregnancy was included in the adjustments.

Bodén and colleagues also looked for relationships between antipsychotics and birth outcomes.

Newborns considered small for gestational age in weight and length appeared to be more common with both groups of antipsychotics before the maternal adjustments were taken in the statistical analysis, with odds ratios ranging from 1.7 to more than 2.6. But after the maternal adjustments, the associations were attenuated substantially and were no longer significant.

But there was an odd finding with regard to head circumference. In the “other” antipsychotics group, there was a significant association with small head size for gestational age prior to adjustments for maternal factors that remained a strong trend even after the adjustments were taken (OR 1.64, 95% CI 0.97 to 2.77).

Among newborns born to mothers taking olanzapine and/or clozapine, on the other hand, there was an association with oversized heads that remained significant in the adjusted model (OR 3.02, 95% CI 1.60 to 5.71).

Bodén and colleagues explained that babies born large for gestational age have previously been associated with gestational diabetes. Because the fetus is not insulin resistant, the high levels of sugar to which it is exposed prompt high levels of insulin secretion, which, in turn, produces macrosomia and increased fetal size.

The divergent findings on head size between the two groups of antipsychotics suggest that direct pharmacologic effects may also play a role, the researchers suggested. But ultimately they could not explain why olanzapine and clozapine, as opposed to the other group, would produce babies with oversized heads. “We do not know the potential mechanism underlying this observation,” they wrote.

They did suggest, however, that maternal smoking probably accounted for much of the apparent pre-adjustment risk of low birth weight and length associated with antipsychotics.

The study was funded by the Lennander’s Foundation and Gillbergska Foundation.

Study authors declared they had no relevant financial interests.

Primary source: Archives of General Pyschiatry
Source reference: Boden R, et al “Antipsychotics during pregnancy: relation to fetal and maternal metabolic effects” Arch Gen Psychiatry 2012; 69: 715-721.