Trends in prenatal screening for and diagnosis of Down's syndrome: England and Wales, 1989-97

BMJ 1998; 317 doi: (Published 03 October 1998) Cite this as: BMJ 1998;317:922
  1. David Mutton, senior research fellow (d.e.mutton{at},
  2. Roy G Ide, data manager,
  3. Eva Alberman, professor emeritus
  1. National Down Syndrome Cytogenetic Register, Wolfson Institute of Preventive Medicine, Department of Environmental and Preventive Medicine, St Bartholomew's and Royal London Medical Colleges, London EC1M 6BQ
  1. Correspondence to: Mr Mutton
  • Accepted 16 April 1998

The National Down Syndrome Cytogenetic Register holds data on 10651 cases of prenatally or postnatally diagnosed Down's syndrome occurring in England and Wales from 1 January 1989 to 31 December 1997 (93% of all eligible cases) (J Morris, personal communication). We used these data to investigate the trends in prenatal diagnosis of Down's syndrome between 1989 and 1997.

Subjects, methods, and results

All clinical cytogenetic laboratories in England and Wales notify the register anonymously of trisomy 21 or related karyotypes, together with the date, place of and indications for referral, parental age, and family history. Most send a copy of the notification to the referring physician for confirmation and completion.1 The outcome of the pregnancy is requested for all cases diagnosed prenatally (but may not be known for several months) and is known for 91% of registered prenatal diagnoses overall.

In 1989 registrations numbered 1081; in 1993, 1144; and in 1997, 1336 (table). In contrast, births nationally dropped from 688 000 in 1989 to 642 000 in 1997.2 The proportion diagnosed prenatally at all maternal ages rose from 30% to 53% between 1989 and 1997. In 1989 only 9% of affected pregnancies were diagnosed prenatally in mothers under 35, rising to 45% in 1997. For mothers aged 35 and over prenatal diagnosis of affected pregnancies rose from 60% in 1989 to 72.2% in 1996 (table). However, the percentage diagnosed prenatally levelled out from 1994 onwards.

Numbers (percentages) of prenatal and postnatal diagnoses of trisomy 21 by maternal age and reason for referral for cytogenetic diagnosis*

View this table:

Maternal age was the first indication for 78% of all prenatal diagnostic tests in 1989 but only 16% in 1997. Maternal serum screening results as an indication increased from 6% in 1989 to 37% in 1996 and fetal ultrasound findings from 13% to 43%. Around a fifth (8/43) of ultrasound referrals for diagnosis concerned the nuchal zone in 1989, but this was 55% (191/350) in 1997.

Amniocentesis accounted for 78% (246/316) of prenatal diagnoses in 1989 and 61% (431/701) at a mean gestational age of 17.2 weeks in 1997. Between 1989 and 1992, 37% of all prenatal diagnoses followed amniocentesis before 16 weeks' gestation; between 1993 and 1996 this fell to 31%. Prenatal diagnosis by chorionic villus sampling or placental biopsy decreased from 18% (58/316) (mean gestation 11.5 weeks) in 1989 to 10% in 1992 following reports of limb defects after biopsy before week 12, increasing to 37% in 1997 at a mean gestation of 13.7 weeks.

From 1989 to 1997, 92% of cases diagnosed prenatally and with known outcome were terminated, 1.9% were miscarried, 1.5% were recorded as a stillbirth or neonatal death, and nearly 5% were live births surviving to discharge from hospital. In some of these an offer to terminate the pregnancy had been declined. When diagnosis occurred after 23 weeks, usually on the basis of ultrasound findings, only a quarter of affected pregnancies were terminated, while 23% were miscarried, stillborn, or died in the neonatal period.


A previous report provided evidence of the validity of these data.3 This showed that after adjustment for the number of terminations and the fetal losses that would have occurred naturally, the maternal age specific risks of Down's syndrome remained similar to those before the availability of prenatal diagnosis. The rise in registrations is due to the demographic trend towards starting families later in life; the high proportion of older mothers from the postwar births bulge4; and the increase in early prenatal diagnoses of trisomy 21 in fetuses leading to terminations of pregnancy that would previously have aborted spontaneously.5 All of these would tend to stabilise the annual number of affected births, in spite of the increase in the numbers diagnosed prenatally.


We thank Christina Kelekun and Karen Wells for data entry and verification. The heads of the cytogenetic laboratories of England and Wales and the Association of Clinical Cytogeneticists provide continued support and data.

Contributors: EA was the grant holder and director of research and contributed in discussion and to the drafting of the manuscript; she is guarantor for the study. RGI was responsible for the development and efficiency of the coding and data entry for the register from 1989 to 1997. DM was project coordinator and cytogeneticist for the register with responsibility for analysing the data and writing the manuscript.


  • Funding The register was initially funded by the Medical Research Council; since 1994 it has been funded by North and South Thames NHS Research and Development.

  • Conflict of interest None


View Abstract