Research Article

Review of neonatal screening programme for phenylketonuria.

BMJ 1991; 303 doi: (Published 10 August 1991) Cite this as: BMJ 1991;303:333
  1. I Smith,
  2. B Cook,
  3. M Beasley
  1. Medical Research Council/Department of Health Phenylketonuria Register, Department of Child Health, London.


    OBJECTIVE--To review the neonatal screening programme during 1984-8. DESIGN--Analysis of data from screening laboratories and paediatricians. SUBJECTS--All live births in United Kingdom. MAIN OUTCOME MEASURES--Structure of programme; number of infants tested and number with phenylketonuria; number of infants missed; ages at testing and treatment. RESULTS--The proportion of infants tested approached 100%. The incidence of phenylketonuria was 11.7/100,000 births (445 subjects): 273 had classic phenylketonuria and three had defects of cofactor metabolism. One child with phenylketonuria was known to have been missed compared with three in 1979-83 and six in 1974-8. Seven subjects had been missed over the 15 years due to negative test results. All seven had been tested with the bacterial inhibition assay, although only 53% of infants had been so tested; the difference between the expected and observed proportion was significant (Fisher's exact test, p = 0.017). Eleven infants with classic phenylketonuria were not tested by 14 days of age and 23 (8%) did not start treatment until after 20 days, an improvement compared with 36 (15%) in 1979-83. There were, however, wide regional variations (0% to 27% treated after 20 days). CONCLUSION--The screening programme achieves high coverage and effectiveness, although some children are still missed. A national practice for screening may help reduce regional variations.