Prenatal Caffeine Exposure Unrelated to Child IQ, Behavior

Tara Haelle

December 01, 2015

A mother's caffeine consumption during pregnancy does not appear to have any clinically significant effect on her child's IQ or problem behaviors, according to a prospective cohort study published online November 18 in the American Journal of Epidemiology.

"This study found that over a range of serum paraxanthine concentrations encompassing the majority of pregnant women, there was little evidence of an adverse association between maternal serum paraxanthine level and the cognitive and behavioral status of the child at ages 4 and 7 years," write Mark A. Klebanoff, MD, and Sarah A. Keim, PhD, from the Research Institute at Nationwide Children's Hospital, Columbus, Ohio.

Although several different associations appeared at certain paraxanthine concentration levels, no consistent pattern emerged, the authors write. "Taken as a whole, we consider our results to be reassuring for pregnant women who consume typical amounts of caffeine."

The researchers compared mothers' caffeine intake during pregnancy with IQ and problem behaviors at ages 4 and 7 years in the resulting children in 2197 pairs of mothers and children. The study participants were part of the control group of a case-control study running from 1959 to 1974, investigating any links between caffeine metabolites and miscarriage. Caffeine intake was determined from measurement of paraxanthine, the primary metabolite of caffeine, in maternal serum before 20 weeks' gestation, and at or after 26 weeks' gestation in the mothers.

The 90th percentile for paraxanthine concentration in the study population was 1405 μg/L in early pregnancy and 1526 μg/L in the third trimester. The 25th percentile for paraxanthine was 97 μg/L in early pregnancy and 187 μg/L in the third trimester. The 50th and 75th percentiles were 383 and 870 μg/L, respectively, for early pregnancy, and 522 and 941 μg/L, respectively, for later pregnancy. (The authors estimate that 1845 μg/L paraxanthine, which is approximately the 95th percentile, would equal caffeine intake of 6 cups of coffee per day.)

After adjustment for the child's sex and the mother's age, race, IQ, education level, smoking status, prepregnancy weight, and gestational age at blood draw, only two of 14 associations reached statistical significance. First, paraxanthine levels before 20 weeks' gestation showed a linear association with internalizing behaviors in children at 4 years of age. Odds of internalizing behaviors increased 30% with each additional 500 μg/L paraxanthine (odds ratio, 1.3; 95% confidence interval, 1.1 - 1.5). No significant associations with hyperactivity or oppositional or other externalizing behaviors existed.

Second, an inverted-J-shaped association emerged for child IQ at 7 years of age and paraxanthine concentration at 26 or more weeks' gestation. The difference in IQ peaked at 0.65 points, corresponding to paraxanthine levels of 750 μg/L compared with no detectable paraxanthine (overall P = .05; P for nonlinearity = .04), and then began to decline.

In addition, "at very high third-trimester paraxanthine concentrations, we observed several nonlinear associations suggestive of lower IQ at age 7 and elevated internalizing behavior at age 4 and externalizing behavior at age 7 (the latter two after additional adjustment for maternal IQ)," the authors write. They also identified an inverse association between early pregnancy paraxanthine levels and hyperactivity behavior at age 4 years, but it was in these areas where the direction of the associations varied.

"For example, the association between paraxanthine at both time points and internalizing behavior at age 4 was in the opposite direction from the association between paraxanthine and internalizing behavior at age 7," they explain.

One noteworthy limitation of the study was the lack of data on maternal alcohol intake, as a positive correlation between alcohol and caffeine intake has been seen in other studies; however, any confounding from alcohol "is unlikely to account for the largely reassuring results we observed," the authors say.

The authors have disclosed no relevant financial relationships.

Am J Epidemiol. Published online November 18, 2015. Abstract


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.