Strengths and limitations
The present study had several limitations. First, owing to limited information regarding the use of ARTs in the database, we could not adjust the data for the effect of ARTs. Several previous studies have shown that the incidence of malformations is higher in multiple pregnancies that are achieved using in vitro fertilization or intracytoplasmic sperm injection than in singletons. Second, most of the BDs associated with multiple pregnancies were isolated fetal abnormalities, because multiple abnormalities may lead to stillbirth or abortion.25 Because the database does not contain information regarding spontaneous abortion, artificial abortion, or stillbirth, this may have skewed the analysis of the BD data. Thus, the high incidence of BDs associated with multiple pregnancies may have been at least in part determined by selection bias, because clinicians are more careful with respect to twin pregnancies and may therefore have conducted more diagnostic tests to check for fetal defects. These limitations should be addressed in the design of future studies.