[Example of teratogen detection using a birth defects registry: depakine and spina bifida]

Most of the human teratogens were discovered as a result of case reports and not primarily through epidemiological studies. We report the detection of a teratogenic effect based both clinical experience and epidemiological analysis. The possibility that anticonvulsant drugs might be teratogenic has been discussed in the literature for several decades. In 1982, we suggested a specific association between valproic acid and spina bifida. The observation was coming both from personal interviews with families of liveborn infants operated upon for spina bifida and from registry data: among 200,000 infants born in the Rhône-Alpes region between August 1979 and August 1982, 11 had spina bifida and were born to epileptic women, 9 of them treated with valproate. Further data confirmed these findings and led to an estimation of about 1-2% for the risk to have an infant with spina bifida after use of valproic acid. The detection of valproic acid teratogenicity raised from an unusual set of circumstances in the Rhône-Alpes region: the existence of a birth defects registry, a special interest for spina bifida, a questionnaire with a routine question on epilepsy, a high prescription rate of valproic acid (28% of women in childbearing age) and the registry status of member of the International Clearinghouse for Birth Defects Monitoring Systems. From a public health point of view, this finding has led doctors to consider treatment with valproic acid as a new indication for prenatal diagnosis. 20 new cases were prenatally detected in 10 years in the Central East region of France.