Prenatal Folic Acid May Cut Subsequent Psychosis Risk in Teens

Pauline Anderson

July 03, 2018

Prenatal exposure to folic acid may reduce the risk for subsequent psychosis in teens, new research shows.

Dr Joshua Roffman

Fortifying foods with folic acid, an intervention that's safe, inexpensive, readily available, and is already recommended to prevent spina bifida, appears to have a protective effect on later brain development, author Joshua L. Roffman, MD, codirector of psychiatric neuroimaging at Massachusetts General Hospital and associate professor psychiatry at Harvard Medical School, Boston, told Medscape Medical News.

"We are increasingly understanding that risk for severe mental illness really begins in the womb, and as such, that may be an emerging focus for early intervention and prevention," Roffman said.

Effects of prenatal folic acid fortification exposure on brain development. Colored areas indicate regions of the cerebral cortex that are significantly thicker in youth who were exposed to folic acid fortification during pregnancy, compared to unexposed youth.


The study was published online July 3 in JAMA Psychiatry.

Impact on Brain Development

Folic acid reduces the risk for neural tube defects such as spina bifida, and women who may become pregnant are advised to take folic acid supplements. However, these defects can develop before a pregnancy is detected.

A program of fortifying food with folic acid was introduced to widen prevention efforts. In March 1996, the US government mandated that all food manufacturers fortify grain products such as bread, flour, corn meal, rice, and pasta with folic acid by January 1, 1998.

The intervention led to a rapid doubling of blood levels of folate among American women and a reduction in the incidence of spina bifida nationwide.

The United States is now among 81 countries that require folic acid fortification of grain products.

Evidence from epidemiologic studies suggests prenatal exposure to folate may influence postnatal brain development. Other studies have shown an association between periconceptional folic acid supplements and lower risk for language delay and autism.

However, a critical unanswered question is whether variation in fetal exposure to folate influences brain development in adolescence, a period associated with heightened risk for psychiatric disorders.

To address this question, the investigators took advantage of the program of folic acid fortification. They investigated the association of prenatal folic acid exposure with subsequent cortical thickness, as measured with MRI.

For this, they used a group of 292 persons aged 8 to 18 years who had undergone MRI at Massachusetts General Hospital (MGH). The group included those born just before, during, and just after the rollout of folic acid fortification.

No individual in the pre-rollout group was exposed to folic acid fortification at any time during gestation, whereas everyone in the post-rollout group was exposed during the entire pregnancy. Those born during the introduction of the program were intermediately exposed.

Investigators found exposure-associated increases in cortical thickness in the bilateral frontal and temporal regions (9.9% to 11.6%; corrected P < .001 to P = .03). In each of these areas, cortical thickness was higher in the fully exposed group compared with the nonexposed group. Intermediate effects were observed in the partially exposed group.

There were also group differences in the slope of age-associated decreases in cortical thickness. The authors note exposure-associated delayed age-associated thinning in temporal and parietal regions (corrected P = .002).

"We found this pattern where the onset of the thinning of the cortex in children who were exposed was delayed, as opposed to the onset of thinning in the children who were not exposed," said Roffman

Downward Trajectory

The onset of thinning in these unexposed individuals had already begun by age 8, he said. "They were already on a downward slope."

The MGH groups did not differ significantly by age at MRI scan, sex, scan indication, or insurance status.

This was a "convenience" sample, whereby researchers had access to hospital medical records and MRI scan results. There are numerous potential problems with this approach; for example, different scanners may have been used, and patients were being scanned for different reasons, said Roffman.

So the researchers turned to an age-matched subset of 861 individuals from the Philadelphia Neurodevelopmental Cohort (PNC), who had also undergone MRI scanning.

"For our replication set, we wanted to look at a cohort where all scans were acquired prospectively with exactly the same imaging protocol, and all of the participants were characterized clinically in a standardized way," said Roffman.

Luckily, the median date of birth for the persons in this group was close to the time of the fortification rollout. So just as with the MGH sample, "we were able to look at children born just before, during, and after the rollout," he added.

The researchers found a similar cortical thinning pattern in the PNC group.

"We saw that not only was thinning delayed in regions that were very analogous to what we saw in the MGH cohort, but the degree of delay was also similar," said Roffman.

Previous studies involving youths who had a psychosis such as schizophrenia or who were at high risk for psychotic illness found an early and accelerated process of cortical thinning.

"The direction of the effect in this new study, where it's a delayed onset of thinning, would seem to oppose that pattern," said Roffman. "That's what led us to the hypothesis that it could be protective against risk of psychosis."

An Indirect Link

To investigate this further, the researchers used the PNC cohort, in which clinical assessments characterized participants as either typically developing or exhibiting psychiatric symptoms.

Results showed that "flatter" thinning profiles in frontal, temporal, and parietal regions were associated with lower odds of psychosis spectrum symptoms (odds ratio [OR], 0.37 - 0.59; corrected P < .05).

"The analysis here basically demonstrated that the delayed thinning predicts a significantly lower risk of psychosis spectrum symptoms, an intermediate reduction in risk for low-grade psychosis symptoms, although that reduction was not statistically significant, and no difference in risk with regard to other kinds of psychopathology," said Roffman.

However, he cautioned that the link between cortical thinning and psychosis was uncovered in a way that was "a bit indirect."

"For the relation to psychosis piece, we were limited by the way that information was ascertained, which made it impossible for us to look directly at the relationship between exposure and psychosis, mostly because of an age difference in the exposure group," he said.

Roffman noted that the results cannot directly link folic acid exposure to reduced schizophrenia risk, because the typical age of onset for that disorder is in the early 20s.

"But since such symptoms in youth are on the same continuum as schizophrenia, the results hold some promise for schizophrenia prevention," he said.

To compare the results with a cohort not at all exposed to prenatal folic acid fortification, the researchers used another age-matched cohort, this one from the National Institutes of Health MRI Study of Normal Brain Development. This group included healthy young people who were recruited nationwide and who had undergone MRI scans up to three times at various ages.

These persons were all born prior to folic acid fortification, so represented "a negative control," said Roffman. The final sample included 217 individuals and 383 MRI scans.

"We wanted to make sure that we did not see evidence of delayed thinning in the areas where this was observed in the MGH and PNC groups," he said. "And that's what we found."

Mechanism Unclear

It is unclear how folic acid may prevent cortical thinning. Rothman noted that folic acid contributes to methylation, an important biochemical process that plays a key role in DNA synthesis, DNA repair, and gene expression.

"Our idea, which needs to be verified, is that exposure to higher levels of folic acid alters the methylation milieu within the womb," and that this exposure leads to "potentially long-lasting effects on methylation at certain important points along the genome," he said.

But researchers still need to understand, "at a much more detailed molecular and cellular level," which genes might be differentially methylated as a result of folate exposure, he said.

These efforts, he added, might uncover important leads to new targets for intervention.

For this study, the researchers limited the analysis to cortical thickness. "We had the most confidence in that particular brain imaging marker, as it had been related in the past with risk for psychosis," Roffman said.

They plan to investigate other brain factors that might be involved in schizophrenia pathophysiology, including brain volume and subcortical regions, such as the thalamus and striatum, said Roffman.

One of the limitations of the study was that it assessed a population-level intervention and did not collect information on folate intake on an individual basis. Another study "where we follow everything prospectively" would be very important, said Roffman.

"The critical thing is how much folate is in the system — from diet, fortification, and prenatal supplements," he said.

In the wake of these new findings, the message for women considering pregnancy is that it's important to take folic acid, not only to prevent spinal bifida "but also potentially for this brain health benefit," said Roffman.

This finding of a brain health effect might spur more countries to consider fortifying the food supply with folic acid. Although 81 countries currently do so, "that leaves more than half the world's population uncovered by fortification," said Roffman.

Caution Warranted

In an accompanying editorial, Tomáš Paus, MD, PhD, Tanenbaum Chair in Population Neuroscience and senior scientist, the Rotman Research Institute, and professor of psychology and psychiatry, University of Toronto, Canada, praised the authors for taking advantage of the "natural experiment" of folate food fortification.

In an interview with Medscape Medical News, Paus said this was a "great idea," but the study included a relatively small number of participants.

"They had a few hundred participants, which sounds like a lot, but if you want to get robust data, you really need large numbers," he said.

The authors also used retrospective data that had been collected for a very different purpose. Because of this, there was a lot of variation in MRI scanning over time, said Paus. "And, of course, time here is the key variable."

Although the authors tried to take that into account statistically, "I think one has to be cautious," said Paus.

Another limitation of the study was that it did not include data on cortical surface measurements, said Paus. He noted that in a large community-based sample in the Netherlands (Generation R), which used plasma levels of folate for comparisons, the offspring of mothers with high vs low levels of folic acid during pregnancy showed greater prenatal head growth and larger total brain volumes at 6 to 8 years of age.

Paus agreed that exposure to folic acid in utero could induce molecular effects, such as DNA methylation, that persist into adolescence. He noted that such a phenomenon has been observed in association with maternal cigarette smoking during pregnancy.

"But this folate methylation hypothesis needs to be tested empirically in large datasets, ideally in conjunction with relevant brain phenotypes," he said.

The researchers' suggestion that prenatal exposure to folate fortification reduces schizophrenia risk is "a huge stretch," said Paus. "I really don't think that fortification of food by folate is a preventive measure for psychosis."

The study does not show that accelerated thinning of the cortex, by itself, causes schizophrenia, added Paus.

"It's possible that among many other things that are happening differently in individuals who are more likely to develop schizophrenia, cortical thinning is one of those things," he said.

The authors and Dr Paus have disclosed no relevant financial relationships.

JAMA Psychiatry. Published online July 3, 2018. Full text, Editorial

For more Medscape Psychiatry news, join us on Facebook and Twitter.


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.