View/Download full
monograph in Adobe Acrobat format (505 kbytes)
View/Download this
4-page summary in Adobe Acrobat format (suitable for printing) Health Technology Assessment 1998; Vol.2: No.1 (Executive summary)
Executive summary
View/Download full
monograph in Adobe Acrobat format (505 kbytes)
View/Download this
4-page summary in Adobe Acrobat format (suitable for printing)
Nicholas J. Wald1
Anne Kennard1
Allan Hackshaw1
Ali McGuire2
1 Department of Environmental and Preventive Medicine, Wolfson Institute of
Preventive Medicine, St Bartholomew's and the Royal London School of Medicine and
Dentistry, Charterhouse Square, London EC1M 6BQ, UK
2School of Social Sciences, City University, London, UK
Over the past 15 years there have been notable advances in antenatal screening for Down's syndrome. First serum alpha-fetoprotein (AFP) and later human chorionic gonadotrophin (hCG) and unconjugated oestriol (uE3), together with maternal age, have been widely used in screening for Down's syndrome, with a detection rate of about 70% for a 5% false-positive rate. More recently inhibin A has been added as a fourth serum marker.
The literature on antenatal screening for Down's syndrome was reviewed.
Methods of screening need to be fully evaluated before being introduced into routine clinical practice. This includes choosing markers for which there is sufficient scientific evidence of efficacy, quantifying performance and establishing methods of monitoring performance. Screening services need to be well integrated and managed.
Screening performance varies according to the choice of markers used and whether ultrasound is used to estimate gestational age. When the latter is used in combination with maternal age, the detection rate for a 5% false-positive rate is estimated to be 59% for the double test (AFP and hCG), 69% for the triple test (AFP, hCG, uE3) and 7 6% for the quadruple test (AFP, hCG, uE3, inhibin A).
Urinary beta-core hCG has been shown to be raised in Down's syndrome pregnancies. Urinary total oestriol and free beta-hCG may also be of value but it would be premature to introduce them into screening practice.
Foetal cells can be identified in maternal circulation and techniques such as fluorescent in situ hybridisation can be used to identify Down's syndrome. However, this does not have the performance, simplicity or economy needed to replace existing methods.
Several demonstration projects using triple and double tests have been conducted, in which screening uptake was about 80% with screen positive rates of about 5-6%. Approximately 80% of women with positive results had an invasive diagnostic test, and about 90% of those found to have a pregnancy with Down's syndrome chose to have a termination.
There are a number of ultrasound markers of Down's syndrome at 15-22 weeks, of which nuchal fold thickness is the most discriminatory on its own, but not discriminatory enough for screening. The markers could be used in combination with the serum markers but no studies assessing this have been completed to date.
The serum markers pregnancy-associated plasma protein-A (PAPP-A) and free beta- hCG, combined with maternal age have an estimated detection rate of 62% for a 5% false-positive rate.
Nuchal translucency is a useful marker of Down's syndrome. There are differing estimates of screening performance and some are subject to bias. Further studies are n eeded to quantify the performance of this test alone and in combination with biochemical markers. There is also a need to compare the performance of such screening with screening at 15-22 weeks to determine which has the greater efficacy and which is the most cost- effective.
The standard method of antenatal diagnosis is amniocentesis at about 15 weeks of pregnancy followed by karyotyping of cultured cells from the amniotic fluid. The excess foetal loss attributed to amniocentesis is approximately 0.9%.
Before 15 weeks of pregnancy, transabdominal chorionic villus sampling (CVS), although less accurate than mid-trim ester amniocentesis, seems to be the diagnostic method of choice.
In general, serum screening is more cost-effective than screening based on maternal age alone at detection rates of about 50% or greater. As the number of screening markers increases, the cost per pregnancy screened increases but, if an extra marker is sufficiently discriminatory, the cost per Down's syndrome birth avoided may decline. For example , the estimated cost per pregnancy screened and the cost per Down's syndrome birth avoided is: £8.90 and £25,600 for the double test; £9.60 and £22,700 for the triple test, and £11.60 and £23,100 for the quadruple test.
Screening leads to women having an invasive diagnostic procedure that can result in foetal loss. As screening performance improves, the number of unaffected foetal losses per Down's syndrome birth avoided declines by 24%, from 0.59 (double test) to 0.45 (quadruple test).
Several studies have shown that the anxiety associated with screening is short lived and can be minimised by the provision of clear and simple information before screening, together with counselling for women with positive results.
Health professionals often do not have adequate knowledge of serum screening and therefore have difficulty in repo rting screening results to women.
Quality assurance and monitoring should be an integral part of a screening service. It is currently not possible to tell whether screening centres undertake epidemiological monitoring and service audit satisfactorily.
Serum screening for Down's syndrome has been widely introduced into practice and has enabled a substantially higher proportion of pregnancies to be identified without materially increasing the proportion of women requiring an invasive diagnostic procedure. Although the screening approach, using multiple markers concurrently, was novel, it has been introduced reasonably effectively using statistical methodology that has been accepted and empirically validated. There is also an active research programme being conducted alongside the clinical service. In spite of the achievements, a number of problems were identified - incomplete coverage of screening, inconsistent practice and a lack of over all direction . The introduct ion of alternati ve methods of screening has led to multiple stepwise screening in an uncoordin ated manner which is confusing to staff and patients. Some research findings have been introduce d into practice before being fully evaluated .
The evidence indicates that screening using the triple test with maternal age is more effective, safe and cost-effective than the double test. The performance of the quadruple test including inhibin A appears somewhat better.
There is substantial variation in screening services for Down's syndrome throughout the UK. This needs to be rectified. The authors recommend that policy makers should ensure overall direction, with a written policy, specified funding and line responsibility, while preserving local commitment.
The authors suggest the establishment of local screening units (covering 15,000 births per year - about three to four maternity units) which would have full responsibility for their service. These would each have a dedicated screening coordinator who would work together with a screening consultant.
Inequity of access to the service and the current multiple, stepwise uncoordinated screening of Down's syndrome should be addressed. The tendency to offer more than one method of screening to the same women at differe nt stages of pregnancy should be avoided.
There is evidence that better staff education and training is needed so that patients are adequately informed about screening and its implications.
Serum markers and nuchal translucency have been shown to be effective in screening for Down's syndrome in the first trimester. However, this needs further evaluation in carefully monitored pilot screening programmes before a decision is made to introduce first trimester screening into general routine practice.
Other research areas include the study of urinary markers and foetal cells in maternal blood
Wald NJ, Kennard A, McGuire A. Antenatal testing for Down's syndrome. Health Technol Assessment 1998; 2(1).
Series Editors:
Andrew Stevens, Ruairidh Milne, Ken Stein
Assistant Editor:
Jane Robertson