Efficiency of neonatal screening for congenital adrenal hyperplasia due to 21-hydroxylase deficiency in children born in mainland France between 1996 and 2003

Arch Pediatr Adolesc Med. 2012 Feb;166(2):113-20. doi: 10.1001/archpediatrics.2011.774.


Objective: To assess the efficiency of the French national screening program for 21-hydroxylase deficiency (21-OHD). Neonatal screening for congenital adrenal hyperplasia due to 21-OHD is mainly intended to prevent death due to salt wasting but remains controversial because of the number of false-positive results and the ease with which most female cases can be identified by virilized genitalia at birth.

Design: Population-based study.

Setting: National neonatal screening program, pediatric endocrinologists nationwide, and reference center for genotyping.

Participants: All neonates screened for 21-OHD in mainland France between January 1, 1996, and December 31, 2003.

Outcome measures: Screening efficiency indicators, disease severity, contribution of screening to early diagnosis, and disease-specific mortality before and during the study period.

Results: A total of 6,012,798 neonates were screened; results in 15,407 were considered positive for 21-OHD. Three hundred eighty-three cases were identified, giving a prevalence of 1 for every 15,699 births. The positive predictive value of screening was 2.3% (95% CI, 2.1%-2.6%), with a sensitivity of 93.5% (90.9%-95.9%) and a specificity of 99.7% (99.7%-99.7%). The false-positive rate was particularly high in preterm infants, for which the positive predictive value was 0.4% (95% CI, 0.2%-0.5%). Screening allowed clinical diagnosis in 162 of 383 cases (42.3%), with the others being detected clinically or through family history. There was a trend toward declining neonatal mortality due to 21-OHD.

Conclusions: In this large population-based study, the efficiency of routine 21-OHD screening was moderate in neonates born at term and very low in preterm neonates. We recommend the discontinuation of screening, as currently performed in France, in preterm neonates.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Adrenal Hyperplasia, Congenital / diagnosis*
  • Adrenal Hyperplasia, Congenital / epidemiology*
  • False Positive Reactions
  • Female
  • France / epidemiology
  • Humans
  • Infant, Newborn
  • Infant, Premature
  • Linear Models
  • Male
  • Neonatal Screening*
  • Predictive Value of Tests
  • Prevalence
  • Retrospective Studies
  • Sensitivity and Specificity