Folic acid, body mass index and cardiovascular malformations

Dear Editor,

We have read the study titled “Predictors of postnatal complications and congenital heart diseases in babies of mothers who have prenatal diabetes and who develop diabetes during pregnancy” conducted by Demirpençe et al. (1) and published in the edition of 2014 December of the Turkish Archives of Pediatrics with interest. In this research study, a significant subject was emphasized by trying to evaluate postnatal complications, congenital malformations and the variables which might affect presence of cardiovascular anomalies in babies of mothers who had prenatal diabetes and developed diabetes during pregnancy.

However, we thought that there were some important methodological problems, when we examined the study in detail. We found that the authors did not consider some important variables which had previously been shown to be related with cardiovascular malformations in different studies (including folic acid and body mass index) and they did not state this as a limitation of their study.

The effect of use of folic acid in pregancy in preventing neural tube defects is well known. Current studies have shown that use of folic acid in pregnancy is also effective in preventing cardiovascular anomalies (2). In the study of Demirpençe et al. (1), there is no information about the usage of folic acid in the pregnant subjects included in the study. The fact that use of folic acid was not considered in such a study which evaluated pregnancy outcomes and especially malformations is a significant defect in terms of the method of the study.

In a meta-analysis performed by Stothard et al. (3), it was shown that obesity in the mother significantly increased the risk of occurence of cardiovascular anomaly in addition to increasing the risk of many congenital anomalies (OR, 1,30; 95% CI, 1,12–1,51). In our country, 30,4% of the women aged 15 years and older are overweight and 20,9% are obese (4). The study group of Demirpençe et al. (1) consisted of diabetic mothers and the relation between obesity and diabetes is well-known (5). Considering these two points, the fact that the pregnant women were not evaluated in terms of obesity and that this was not included in logistic regression analysis is again a significant defect.

We think that the results of this study may be misleading, because the two variables mentioned above and other variables (for example, smoking) which should have definetely been considered in terms of cardiovascular malformations were not included in the statistical analysis. For example, the result that “ use of insulin increased the possibility of congenital heart disease 1.4–5.5-fold” which was obtained by the investigators and which was not discussed sufficiently is in fact a result which predicts that insulin is teratogenic. However, this may be related with underlying diabetes and other factors which were not considered by the investigators and some of which have been mentioned above. Insulin does not cross the placenta and is the first-line treatment during treatment in a candidate mother with diabetes.

In conlusion, confounding factors should be known very well and should be definetely evaluated in such studies in which pregnancy outcomes and especially congenital anomalies are studied. The results which have not been evaluated because of different factors should be stated as study limitations. Keeping the accuracy and preciseness of studies conducted at the highest level possible is very important in terms of developing and pursuing beneficial clinical practices for patients.