Recurrence of discordant congenital heart defects in families.

BACKGROUND Variation within a single gene might produce different congenital heart defects (CHDs) within a family, which could explain the previously reported familial aggregation of discordant CHDs. We investigated whether certain groups of discordant CHDs are more common in families than others. METHODS AND RESULTS Using Danish national population and health registers, we identified CHDs among all singletons born in Denmark during 1977-2005 and their first-degree relatives. In a cohort of 1 711 641 persons, 16 777 had CHDs, which we classified into 14 phenotypes. We estimated relative risks of discordant CHDs by history of specific CHDs in first-degree relatives. The relative risk of any dissimilar CHD given the specified CHD in first-degree relatives was as follows: heterotaxia, 2.00 (95% CI, 0.96 to 4.17); conotruncal defects, 2.78 (95% CI, 2.12 to 3.66); atrioventricular septal defects, 2.25 (95% CI, 1.39 to 3.66); anomalous pulmonary venous return, 1.76 (95% CI, 0.66 to 4.64); left- and right-ventricular outflow tract obstruction, 2.55 (95% CI, 1.87 to 3.48) and 3.09 (95% CI, 2.03 to 4.71), respectively; isolated atrial septal defects, 2.76 (95% CI, 2.11 to 3.61); isolated ventricular septal defects, 2.27 (95% CI, 1.75 to 2.94); persistent ductus arteriosus, 1.92 (95% CI, 1.32 to 2.79); other specified CHDs, 3.29 (95% CI, 2.51 to 4.32); and unspecified CHDs, 2.30 (95% CI, 1.76 to 3.00). Relative risks for all pairwise combinations of discordant CHD phenotypes gave no indications that certain constellations of CHDs cluster more in families than others. CONCLUSIONS We documented strong familial aggregation of discordant CHD phenotypes. However, we observed no excess clustering of specific CHD phenotypes among the first-degree relatives.