Pitfalls of prenatal diagnosis of 21-hydroxylase deficiency (21-OH def) congenital adrenal hyperplasia (CAH)

Abstract

Hormonal measurements and HLA genotyping of amniotic fluid at midgestation correctly predicted the postnatal dx of CAH in 26 of 29 fetuses at risk for CAH. Of these 26, 6 were predicted to have classical 21-OH def based on elevated amniotic fluid 17-hydroxyprogesterone (17-P) and Δ4-androstenedione (A). These 6 fetuses and their index cases were ultimately proven to have salt-wasting classical 21-OH def. Of 3 HLA typed, genotype was identical to the index case. Normal amniotic fluid 17-P and A in the remaining 20 predicted fetuses unaffected with classical CAH, and these patients have been clinically asymptomatic to date or biochemically proven not to be affected with classical or nonclassical CAH. Of the 20 fetuses, 6 were HLA typed and predicted to be homozygous unaffected or heterozygous. However, in 3 of the 29 fetuses, prenatal diagnosis was incorrect. In one, the fetus was predicted to have CAH based on HLA identity to the index case, but amniotic fluid 17-P and A were normal and the fetus was normal. The index case of this family did not have CAH but was a normal child. Thus, amniotic fluid hormone levels accurately predicted a normal fetus while HLA typing was not relevant in prenatal dx because the index case was unaffected. The second fetus was predicted to be a carrier on the basis of HLA typing and normal amniotic fluid 17-P and A. However, during infancy the female infant was shown to have nonclassical CAH and be HLA identical to the index case. The index case in this family, presumed to have classical CAH, was later diagnosed to have nonclassical CAH. In the third case, the fetus was predicted to be a heterozygote by HLA genotyping and to be unaffected by hormonal measurement. Postnatally, at age 2 7/12 yrs, the male child was shown to have non-salt-losing classical CAH and to be HLA identical to the index case who also has non-salt-losinq CAH. These data demonstrate that in salt-losing classical CAH, prenatal dx of the homozygous affected fetus is reliable by the measurement of amniotic fluid 17-P and A. In nonclassical or non-salt-losing classical CAH, amniotic fluid 17-P and A are in the normal range, thus prenatal dx is not possible by hormonal evaluation. Both hormonal and HLA studies of fetal amniotic fluid, as well as correct dx of the specific form of 21-OH def and HLA studies of the index case, are essential in diagnosing the specific form of 21-OH def CAH in the fetus.