Birth Prevalence of Congenital Heart Disease

To the Editor: Congenital heart defects occur in 1 of every 100–150 newborns.1 Periconceptional use of folic acid may reduce the incidence of some of these defects, including conotruncal heart defects.2,3 In November 1998, the Canadian government instituted mandatory fortification of flour and enriched pasta with folic acid.4 This is estimated to increase the daily intake of folic acid among women 18–34 years of age by approximately 50%.5 We assessed whether the birth prevalence of conotruncal heart defects subsequently decreased in the Canadian province of Quebec. Using echocardiography databases and medical records of Quebec's 4 pediatric cardiology referral centers, we identified all children born alive from 1 January 1993 to 31 December 1996 and from 1 January 1999 to 31 December 2002 with one of the following diagnoses: tetralogy of Fallot, pulmonary atresia and ventricular septal defect, double-outlet right ventricle, truncus arteriosus, interrupted aortic arch, pulmonary atresia and intact ventricular septum, transposition of the great arteries, and hypoplastic left heart syndrome. The first 5 are conotruncal defects. Since fortification was optional between December 1996 and November 1998, this period was not included. We used segmented Poisson regression to investigate changes in birth prevalence. We considered 2 groups: conotruncal defects and other congenital heart defects. Year was included as a linear factor. We used a segmented model allowing for 2 regression slopes, one before and one after 1998, described as follows: where μij is the mean number of congenital heart defects of type j during year i. The variable type j takes the value 0 for conotruncal defects and 1 for other congenital heart defects. The years considered are 1993–1996, and 1999–2002. The indicator I (year >1998) is a function that takes the value 1 if year >1998, and 0 if year <1998. The model for conotruncal defects before 1998 is: and after 1998 is: Similarly, the model for other congenital heart defects before 1998 is: and after 1998 is: From these equations, the change in the slope between the 2 periods for the birth prevalence of conotruncal defects is β4 and the change in the slope for the prevalence of other congenital heart defects is β4 + β5. Data were analyzed using SAS software, version 9.1 (SAS Institute, Inc., Cary, NC). The Table shows the birth prevalence of congenital heart defects during the 2 time periods. There was a marked decline in the slope of prevalence of conotruncal defects between the 2 periods (−0.24; 95% confidence interval = −0.42 to −0.06) with less evidence for a decrease in the slope for other congenital heart defects (−0.15; −0.35 to 0.04) (Figure).TABLE: Prevalence (per 10,000 live births) of 8 Major Congenital Heart Defects in the Province of Quebec Before and After Mandatory Fortification of Flour and Enriched Pasta With Folic AcidFIGURE.: Birth prevalence of conotruncal defects (CTDs) and other congenital heart defects (NON-CTDs) in the province of Quebec during the 2 studied time periods, with their corresponding slopes.The evidence for periconceptional use of folic acid in primary prevention of congenital heart defects is inconsistent.6 This ecologic study of selected congenital heart defects in Quebec following food fortification with folic acid, suggests that the birth prevalence of conotruncal defects has decreased, and perhaps of other congenital heart defects as well. Potential ascertainment bias was minimized by having the same investigator collecting the data in all 4 centers. Increased use of folic acid supplements by pregnant women between the 2 periods could have contributed to this decline, and thus overestimate the effect of fortification. However, surveys of Quebec women have shown no substantial change in the intake of folate-containing multivitamin supplements before conception between the 2 periods.7,8 Selective pregnancy termination may have had an impact on the results, although the prevalence of hypoplastic left heart syndrome (frequently associated with termination of pregnancy) did not decrease between the 2 time periods, suggesting that prenatal diagnosis did not have a substantial impact on the results. Our findings add further support to the body of literature implicating folic acid deficiency in the pathogenesis of congenital heart defects. ACKNOWLEDGMENT We thank Jean-Luc Bigras, Christine Houde, Marc Bellavance, and Gregor Andelfinger for their contributions to this study. Milan Prsa Tania Saroli Andrew S. Mackie Adrian B. Dancea Division of Cardiology Department of Paediatrics Montreal Children's Hospital McGill University Health Centre Montreal, Quebec, Canada, [email protected] José A. Correa Masoud Asgharian Department of Mathematics and Statistics McGill University Montreal, Quebec, Canada