Chromosomal aberrations in the foetus have been detected for more than 30 years now by karyotyping cells from chorionic villi or from amniotic fluid. This is an elaborate and time-consuming process, requiring highly skilled technicians. The vast majority of aberrations detected are numerical, loss or excess of entire chromosomes.

Fluorescence in situ hybridisation (FISH) using DNA probes specific for the most frequently involved chromosomes is a faster alternative to karyotyping. However, FISH still requires considerable hands on time of trained technicians, in particular counting the signals on a sufficient number of nuclei, a process not easily automated in the diagnostic laboratory.

What we need is a rapid, easily automated technique, which will allow a prescreening of samples for the most common chromosomal aberrations, after which full karyotyping can be reserved for those cases in which there is substantially increased risk: the foetus showing developmental anomalies on ultrasonography, including enlarged nuchal translucency, and an abnormal DNA test.

Several variants of quantitative PCR have been shown to rapidly and accurately detect copy number changes characteristic for chromosomal aberrations.1, 2, 3 The three approaches karyotyping, FISH, and quantitative PCR have been compared,4 and cost-effectiveness of various testing regimes have been assessed.5

Why then, are so many cytogenetecists reluctant to abandon full karyotyping, and to have it replaced by a DNA test? There may be several reasons, one of which is the fact that the practical efficacy of a screening test in terms of success rate, positive, and negative predictive value, cost, etc. can only be assessed on very large numbers of samples. The report by Mann et al6 fills part of this gap, since a very large number of samples have been tested in parallel using both quantitative PCR and karyotyping, showing the former to be a robust, reliable, and fast alternative to the latter.

The application of DNA techniques opens new possibilities that have as yet not been explored. So far only fragments of chromosome 13, 18, 21, X, and Y have been used. However, Rahil et al2 used PCR fragments within genes, hinting already at the possibility to detect mutations at the gene level. In a multiplex PCR, a very large number of fragments can be amplified simultaneously. Various detection techniques are being developed for such large numbers of PCR products using microarrays or beads. Thus, subtelomeric regions, microdeletion syndromes, frequently occurring marker chromosomes, etc can easily be included. In addition, we could use for detection of the X chromosome exons of the Duchenne gene, which are most often deleted in patients. Moreover, we could include a fragment specific for the deltaF508 mutation, responsible for cystic fibrosis, exon 7 of the SMN1 gene, deleted in patients with spinal muscular atrophy, etc.

One may argue that prospective parents have not asked for this, and therefore we should stick to those conditions we used to be able to detect by karyotyping. But, do we really know what the parents want to know, and have tested ? It is not unreasonable to assume that the parents wish to have a test done which will detect any condition which will cause prolonged and sustained suffering to their child, irrespective of whether it is a chromosomal aberration or a gene mutation.

If however, we do include gene tests such as deltaF508 and exon 7 of SMN1 in our test, we will detect also heterozygous carriers in a relatively large number of pregnancies, which will cause much unrest and anxiety that could have been avoided, had the test been performed in the parents before conception.

Thus, the increased prenatal diagnostic capabilities dictate radical change in obstetric care. Preferably, the prospective parents should be counselled in advance to allow for reflection on which diagnostic possibilities they wish to use. Since the majority of pregnancies today are planned, such preconception advice should be incorporated in primary care.

If prenatal diagnosis not only includes the most frequent aneuploidies, but also a large number of other conditions, the question arises whether we should stick to serum screening, nuchal translucency, and maternal age to select high-risk pregnancies for invasive procedures. Harris et al7 already belled the cat by contemplating eligibility to invasive prenatal testing of all pregnant women.

We can conclude that today, 30 years after its beginning, prenatal diagnosis is very much alive, and a great many questions on how to go about it are storming at us. In parallel with further developing our diagnostic capabilities we should pass on the information to prospective parents to allow for informed decision making and free choice. By much trial and little error, we should find a new standard of prenatal care which minimizes alarm and anxiety during pregnancy, and maximizes the benefits of improved diagnosisâ–ª