American Journal of Epidemiology Original Contribution Effect of Supplemental Folic Acid in Pregnancy on Childhood Asthma: a Prospective Birth Cohort Study

This study aimed to investigate the effect of the timing, dose, and source of folate during pregnancy on childhood asthma by using data from an Australian prospective birth cohort study (n ¼ 557) from 1998 to 2005. At 3.5 years and 5.5 years, 490 and 423 mothers and children participated in the study, respectively. Maternal folate intake from diet and supplements was assessed by food frequency questionnaire in early (<16 weeks) and late (30–34 weeks) pregnancy. The primary outcome was physician-diagnosed asthma, obtained by maternal-completed questionnaire.

tionship between the 5-minute Apgar score and the risk of neonatal death in a cohort of over 150,000 infants born over a 10-year period at Parkland Hospital in Dallas. Term infants with a 5-minute Apgar score of 0 to 3 had a mortality rate of 244/1000, whereas among those with a score of at least 7 the rate was only 0.2/1000. For any given Apgar score, fetal acidemia increased the risk of death. For example, the combination of a 5-minute Apgar score of 0 to 3, and an umbilical artery pH at birth of Ͻ7.0 more than doubled the risk of neonatal death (NEJM 2001;344:467 and Survey 2001;56:406).
The abstracted study of Laptook et al describes the relationship of the 10-minute Apgar score to either death or moderate/severe disability at 18 to 22 months of age. In this cohort, if the Apgar score at 10 minutes was 4 or lower, the risk of death or moderate/severe disability exceeded 50%. It was approximately 80% when the score was 0 to 2.
The 3 studies discussed above demonstrate a progression that makes intuitive sense: the longer the Apgar score is very depressed after birth, the greater the risk of death. The study of Laptook adds disability information (remembering that in general the Apgar score is not a reliable predictor of subsequent neurologic outcome).
Over half of infants with 10 minute Apgar scores of 0 to 2 had moderate or severe disability at 18 to 22 months, whereas only one-third of infants with scores of 3 to 5 were similarly disabled.
It bears pointing out that Laptook cohort is a selected one. All infants in the cohort met stringent, predefined criteria for asphyxia and exhibited moderate or severe encephalopathy. Thus, the findings should not be generalized to all infants with low 10 minute Apgar scores. Moreover, it is possible that the Laptook data underestimate the death and disability rate for asphyxiated, encephalopathic infants with very low 10-minute Apgar scores, as some of the sickest infants in this category may not have been enrolled in the randomized trial from which this cohort was drawn.
No doubt, a very depressed 10-minute Apgar score, especially in the setting of asphyxia and encephalopathy, is a dire situation. But it is not hopeless, and the risk of death and disability can be reduced by the use of whole body cooling (Shankaran S, et al. NEJM 2005;353:1574 andSurvey 2006;61:158). Such cooling has been shown to be effective only when initiated within 6 hours of life, so for those infants deemed candidates for salvage, every effort should be made to initiate the cooling as soon as possible.-DJR) supplementation during pregnancy was associated with wheezing and lower respiratory tract infections in the early life of exposed offspring.
This prospective cohort study investigated the effect of the timing, dose, and source of folate during pregnancy on the subsequent occurrence of childhood asthma in exposed offspring. Data was obtained from an Australian prospective birth cohort study from 1998 to 2005. The study participants included 490 mothers and children at 3.5 years of age and 423 mothers and children at 5.5 years of age. Maternal dietary folate intake and intake of folic acid supplements was assessed through interviews with mothers and a food frequency questionnaire in early (Ͻ16 weeks) and late (30-34 weeks) pregnancy. More than 75% of the original sample participated in the study. The mother's report of physiciandiagnosed asthma was the primary study outcome. The association between maternal folic acid intake and each asthma outcome (3.5 years, 5.5 years, persistent) was investigated with use of a Poisson regression model. Asthma was diagnosed in 11.6% of children at 3.5 years (n ϭ 57) and 11.8% of children at 5.5 years (n ϭ 50); persistent asthma (present at both 3.5 and 5.5 years) occurred in 7% (n ϭ 30). The ingestion of supplemental folic acid in late pregnancy was associated with an increased risk of asthma at 3.5 years (relative risk [RR], 1.26: 95% confidence interval [CI], 1.08-1.43) and persistent asthma (RR, 1.32; 95% CI, 1.03-1.69). The observed effect persisted after adjustment for potential confounding factors. Ingestion of folic acid supplements in prior early pregnancy were not associated with the occurrence of asthma. Similarly, dietary folate intake had no effect on the primary outcome at any stage of pregnancy in either age group.
These findings suggest that the use of supplemental folate during late pregnancy is associated with increased the risk of physician-diagnosed asthma in the offspring at 3.5 years and persistent asthma (at both 3.5 and 5.5 years).

EDITORIAL COMMENT
(This is an interesting but somewhat confusing and counterintuitive study of the relationship between maternal antenatal folic acid nutriture and childhood asthma at ages 3.5 and 5.5 years. The investigators accessed data collected prospectively from 557 pregnant women enrolled before 16 weeks' gestation at 4 antenatal clinics in Adelaide, Australia. Participants were interviewed by a research nurse before 16 weeks and at 30 to 34 weeks' gestation, and again at 6, 9, and 12 months and 2 and 3.5 years after delivery. At 5.5 years, the women answered a postal questionnaire. The occurrence of childhood asthma was ascertained by asking the women at 3.5 years if their child had ever been diagnosed with asthma and if so, by whom, and at 5.5 years if their child had ever been diagnosed with asthma and had asthma at that time. Persistent asthma was defined as asthma at both 3.5 and 5.5 years. The investigators found that folic acid intake from either a supplement or a supplement plus dietary intake in late pregnancy (but not early pregnancy) was associated with a significantly increased risk of persistent asthma (adjusted RR, 1.32; 95% CI, 1.02-1.69). Folic acid was also associated with asthma at 3.5 years only (aRR 1.26;1.09-1.47) but not at 5.5 years only.
These data are counterintuitive for several reasons. First, the strongest association was between persistent asthma and late pregnancy folic acid supplement intake. Yet the median supplemental dose of folic acid ingested in late pregnancy by participants who were sampled at 5.5 years was 300 g/d, and the dietary folate intake was estimated to be 210.4 g/d. Because only 50% of dietary folate is bioavailable, this resulted in a total folic acid dose of approximately 400 g/d, the United States Department of Agriculture recommended daily dose. Since this dose was originally determined by estimating the minimal dose necessary to cure folate deficiency anemia, this is a conservative estimate of the daily folic acid requirement (Oakely GP. N Engl J Med 1998;338:1060). Second, asthma was not increased in all children. The children of women who had asthma themselves (a significant risk factor for asthma in their offspring) and took folic acid actually had a significantly decreased incidence of childhood asthma; for every 100 g increase in maternal dietary folate in early pregnancy, the RR of asthma at 3.5 years was decreased by 43% (RR interaction ϭ 0.57; P ϭ 0.04), and for every 1000 g increase in folic acid intake in late pregnancy, the incidence of asthma at 5.5 years decreased by 37% (RR interaction ϭ 0.63; P ϭ 0.03).
Although statistical significance is an important criterion for evaluating research findings, biological plausibility is also important. Many statistically significant findings are not clinically Physiology and Pathophysiology of Pregnancy, Labor, and the Puerperium significant, and some are not biologically plausible. For example, the incidence of persistent asthma at 5.5 years-the stated outcome of interest in this study-was 7%, and the incidences at 3.5 and 5.5 years were 11.6% and 11.8%, respectively. According to the International Study of Asthma and Allergies in Childhood-an enormous worldwide study that included 257,800 children evaluated at age 6 to 7 years and 463,801 children evaluated at age 13 to 14 years at 155 centers in 56 countries-the incidence of childhood asthma varies considerably around the world, from 4.1% to 32.1% at age 6 to 7 years and 2.1% to 32.2% at age 13 to 14 years (Eur Respir J 1998;12:315). However, data from the 4 participating centers in Australia, including a center in Adelaide, where Whitrow et al performed their study, indicate that the prevalence of asthma at age 6 is approximately 25% (26% in Adelaide), which is far higher than the 7% prevalence in 5.5 year olds noted in this study. Thus, although in this particular study group folic acid appeared to increase the incidence of asthma, the final prevalence was far lower than the prevalence reported both in Australia and in Adelaide specifically.
The authors cite another study that found an association between first trimester folate supplementation and an increased incidence of wheezing at 18 months of age. The data came from the Norwegian Mother and Child (MoBa) Cohort Study, which included 100,000 pregnancies enrolled up to 2008, and was obtained from questionnaires administered at 17 and 30 weeks' gestation and at 6 and 18 months after delivery (Haberg, et al. Arch Dis Child 2009;94:180). The RR of wheezing in children exposed to folic acid was significantly increased-but to a RR of only 1.06 (95% CI, 1.03-1.10). Interestingly, the MoBa study also yielded data indicating that the incidence of wheezing at age 18 months was 3.3% higher (95% CI, 1.2-5.3) in the offspring of women who were obese during pregnancy (Haberg SE. Paediatr Perinat Epidemiol 2009;23:352). Although both these findings are statistically significant, their clinical significance is debatable.
Considering the relatively low incidence of asthma at age 5.5 years associated with folic acid use, the not excessive folic acid intake reported by the study participants, and the fact that the children of women with asthma who took folic acid during pregnancy had a statistically and clinically significant decrease in the incidence of asthma, the findings of this study do not negate the many studies performed worldwide documenting the myriad benefits of antenatal folic acid supplementation. We should continue to recommend folic acid supplementation to pregnant women and especially to reproductive age women who are considering pregnancy.-KDW)