eLetters to:

ARTICLE: Adolfo Correa, Lorenzo Botto, Yecai Liu, Joseph Mulinare, and J. David Erickson Do Multivitamin Supplements Attenuate the Risk for Diabetes-Associated Birth Defects? Pediatrics 2003; 111: 1146-1151 [Abstract] [Full text] [PDF]

eLetters: Submit a response to this article

eLetters published:

MV vs folic acid supplementation

John A. Cranton   (27 May 2003)

No attenuation on the risk for diabetes-associated birth defects by folic acid fortification

Eduardo E Castilla, Maria da G. Dutra, Jorge S. Lopez Camelo   (2 December 2003)

Re: No attenuation on the risk for diabetes-associated birth defects by folic acid fortification

Adolfo Correa, Lorenzo Botto, Joe Mulinare, and J David Erickson   (29 December 2003)

MV vs folic acid supplementation 27 May 2003
John A. Cranton, RN University of Washington Medical Center Send letter to journal: Re: MV vs folic acid supplementation jcranton{at}u.washington.edu John A. Cranton	Correa et al., Excellent work. Although we have made great strides with folic acid supplementation, it has long been my contention that MV supplementation (under the direction of a physician or nurse) of all women of child bearing age should be recommended. It is not surprising that much of the research studying the use of one or two vitamins is conflicting. Most vitamins (folic acid, B12, B6; E & C) work together with a complexity that is still not fully understood. Although many would argue that Correa et al.’s conclusions/results can’t be generalized to the child bearing population as a whole, I disagree. The unclear mechanisms that increase risk of birth defects among offspring of pregnant diabetic women likely share many of the mechanisms among all pregnant women, at least until proven otherwise. I am currently working on a paper discussing the use of B vitamins for cardiovascular disease prevention in adults; however, I would not call that a competing interest. Sincerely, John Cranton, RN, BSN
No attenuation on the risk for diabetes-associated birth defects by folic acid fortification 2 December 2003
Eduardo E Castilla, Research ECLAMC: Latin American Collaborative Study of Congenital Malformations, Maria da G. Dutra, Jorge S. Lopez Camelo Send letter to journal: Re: No attenuation on the risk for diabetes-associated birth defects by folic acid fortification castilla{at}centroin.com.br Eduardo E Castilla, et al.	Prompted by Correa et al. paper (Pediatrics 2003; 111: 1146–1151) [1], we searched the data base of ECLAMC (Latin American Collaborative Study of Congenital Malformations)[2] to see if folic acid fortification in Chile induced any detectable effect on the maternal diabetes-congenital anomaly interaction. Even thought we are dealing with folic acid instead of multivitamins, fortification instead of supplementation, and recent time periods instead of remote ones, we hypothesized that extra folic acid supply in the diet could reduce the association strength between maternal diabetes, and allegedly related congenital anomalies in the offspring. Our data base includes consecutive births occurred in maternity hospitals in Chile, providing data during the 1999-2003 period. Based on the beginning of fortification on January 2000, two periods were defined: 1999-2000 for unfortified births, and 2001-2003 for fortified births. All registered malformed newborn infants were included in the study as well as one-to-one matched healthy controls, defined as the like sexed liveborn occurring next to the malformed infant at the same maternity hospital. In a second step the 20 malformation types reported by Correa et al., 2003 [1] as being associated with maternal diabetes were selected. The prevalence of reported maternal diabetes was 3.44% (448/13040), somewhat higher among cases (260/6510: 3.99%) than controls (188/6520: 2.88%) (chi-square, df:1: 3.45; p=0.06), as expected from known association as well as from maternal memory bias. The strength of the association between maternal diabetes and congenital anomalies of any type, as measured by odds ratio, did not differ (Interaction Breslow-Day test (chi-square, df:1: 2.42, p=0.112) between the folic acid fortified (cases: 194/4266; controls: 128/4266, OR=1.54, 95% CI: 1.22-1.94, p<0.001), as compared with the previously unfortified period (cases: 66/2194; controls: 60/2194, OR= 1.10, 95% CI: 0.76-1.59, p=0.587). For the 20 selected anomaly types this comparison resulted in non significant differences between both periods (Interaction Breslow-Day test chi-square, df:1: 0.24, p=0.112): fortified (cases: 45/995; controls: 128/4266, OR=1.53, 95% CI:1.07-2.19, p=0.015), and unfortified (cases: 15/439; controls: 60/2194, OR= 1.29, 95% CI: 0.70-2.37, p=0.379). Likewise, for the 20 selected anomaly types in their isolated forms (not associated with any other unrelated anomaly in the same infant), this comparison resulted in non significant differences between both periods (Interaction Breslow-Day test chi-square, df:1: 0.23, p=0.631): fortified (cases: 31/719 controls: 128/4266, OR=1.45, 95% CI: 0.96-2.21, p=0.064), and unfortified (cases: 11/344; controls: 60/2194, OR= 1.20 0.59-2.40, p=0.570). On the contrary, the diabetes risk for the three congenital malformations sub-samples were statistically significant or nearly significant in fortified mother´s, showing an opposite effect from that found by Correa at. al. (2003). Since these findings could be due to unconsidered confusion factors, such as maternal age, socioeconomical level, and/or ethnicity, further analysis will be performed in the future as the folic acid fortified sub- sample increases with time. [1]. Correa A, Botto L, Liu Y, Mulinare J, Erickson D. Do multivitamin supplements attenuate the risk for diabetes-associated birth defects ?. Pediatrics. 2003;111:1146-51. [2]. Castilla EE, Lopez-Camelo JS. The surveillance of birth defects in South-America:-I-The search for time clusters: Epidemics. Adv Mutagen Res 1990; 2:191-210.
Re: No attenuation on the risk for diabetes-associated birth defects by folic acid fortification 29 December 2003
Adolfo Correa, medical epidemiologist CDC, Lorenzo Botto, Joe Mulinare, and J David Erickson Send letter to journal: Re: Re: No attenuation on the risk for diabetes-associated birth defects by folic acid fortification acorrea{at}cdc.gov Adolfo Correa, et al.	We read with interest the comments by Castilla et al (2003)[1] reporting a recent evaluation of the possible effect modification by folic acid fortification of flour in Chile on the association between maternal diabetes and congenital anomalies. Because recent serosurveys of a sample of women in Chile have documented that such a fortification program has resulted in measurable increases in serum and red blood folate levels in the population (Hertrampf et al., 2003) [2], it would be reasonable to expect a reduction in the risk for birth defects preventable by folic acid in the general population. In particular, if folic acid happens to be the explanation for the attenuation of the risk of diabetes-associated nonsyndromic malformations associated with periconceptional use of multivitamins that we reported in a recent issue of the Journal (Correa et al., 2003)[3], then it would be reasonable to expect a reduction in the risk for diabetes-associated malformations resulting from folic acid fortification in Chile. However, Castilla et observed a 29% increase in the risk for the 20 selected anomaly types in the presence of maternal diabetes during the period before fortification (OR=1.29, 95% CI 0.70- 2.37), and a 53% increase in the risk for the same anomaly types with diabetes after folic acid fortification (OR=1.53, 95% CI: 1.07-2.19). The statistical test for interaction, however, showed no significant difference in the effect of diabetes on the risk for the selected anomaly types before and after fortification. The different findings in our study and that of Castilla et al. might be due in part to one or more of the following possibilities. Use of multivitamins in our study might have been a proxy measure for the level of control of diabetes and not a direct effect of the micronutrient contained in the pills. In such a case, fortification of the food supply with folic acid per se in any population would have no effect on the risk of congenital anomalies associated with diabetes. The excess risk for congenital anomalies associated with diabetes was much greater in our study population than in the study population detailed by Castilla et al. Maternal diabetes was associated with a 4-fold increase in the risk for selected congenital anomalies (in the absence of multivitamin use) in our study but only with a 10% increase in the risk for the same congenital anomalies (before the period of folic acid fortification) in the study by Castilla et al. If this 10% excess risk reflects the true effect of diabetes in Chile, then it would be difficult to detect any change of such a small excess risk caused by folic acid fortification. The small excess risk for congenital anomalies from maternal diabetes observed in the study population in Chile raises some interesting questions: 1)Why is it that the relative risk of congenital anomalies associated with maternal diabetes is so low when maternal diabetes is known to be a potent teratogen? 2) Was the prevalence of pre-gestational diabetes lower in the study by Castilla et al. than in our study population? 3) Was maternal pre-gestational diabetes better controlled in the study by Castilla et al. than in our study? 4) Was there a larger proportion of gestational diabetes among participants with diabetes in the study by Castilla et al. than in our study? Affirmative responses to questions 2)-4) would explain in part some of the observed differences between the two studies. The attenuation of the diabetes-associated risk for malformations that we observed might have been due to other micronutrients in the multivitamin supplements (e.g., antioxidants). In such a case, folic acid fortification per se would have no effect on the risk of diabetes- associated congenital anomalies. Last but not least, the number of pregnant women who had diabetes and had exposure to folic acid during the periconceptional period in our study was too small to detect a small excess risk for birth defects in the presence of both periconceptional use of multivitamins and maternal diabetes as was noted by Castilla et al. There is a need for additional studies to elucidate our findings and to help clarify the differences between our findings and those of Castilla et al. Large population-based case-control studies that attempt to characterize the type of diabetes, prevalence of diabetes, level of diabetes control among study subjects, periconceptional use of vitamin supplements among participants, and the type of vitamin supplements would be particularly valuable in this regard. Sincerely, Adolfo Correa, MD, PhD Lorenzo Botto, MD Joe Mulinare, MD, MPH J. Dave Erickson, DDS, PhD National Center on Birth Defects and Developmental Disabilities, Centers for Disease Control and Prevention, Atlanta, GA [1] Castilla EE, Gutra MG, and Lopez Camelo JS. No attenuation on the risk for diabetes-associated birth defects by folic acid fortification. Pediatrics 2003 [2] Hertrampf E, Fortes F, Erickson JD, Cayazzo M, Freire W, Bailey LB, Howson C, Kauwell GPA, Pfeiffer C. Consumption of folic acid- fortified bread improves folate status in women of reproductive age in Chile. J Nutr 2003;133:3166-3169. [3] Correa A, Botto L, Liu Y, Mulinare J, and Erickson JD. Do multivitamins attenuate the risk of diabetes-associated birth defects? Pediatrics 2003; 111:1146-1151.