Correspondence
Increased Nuchal Translucency and Fetal Chromosomal Defects
N Engl J Med 1998; 338:1228-1230April 23, 1998
- Article
To the Editor:
The article by Taipale et al. (Dec. 4 issue)1 raises further questions regarding methodologic flaws common in the literature on nuchal-translucency screening for Down's syndrome during the first trimester. The principal role of such screening is to detect fetuses with Down's syndrome in the 60 to 70 percent of pregnant women with no other risk factors for aneuploidy — namely, maternal age of more than 35 years or a previous pregnancy with Down's syndrome. Measuring nuchal translucency in women who already have these risk factors constitutes screening of a high-risk group, the results of which may not be applicable to a low-risk population. Although Taipale et al. excluded women more than 40 years old, how many of the women had an increased risk of Down's syndrome based on their being more than 35 years old?
A major source of concern about nuchal-translucency screening is the risk of Down's syndrome in fetuses with normal nuchal measurements. In this study the fetal neck was not seen adequately in 138 fetuses, but they were all assumed to be normal. For an additional 50 fetuses, no outcome information was available. No information was given on how many of the 10,010 women lost their pregnancies or what follow-up was performed in these cases. Three percent of pregnancies of 10 to 13 weeks' duration can be expected to abort,2 a figure representing up to 300 fetuses in the study by Taipale et al. Can the authors be certain that there were no additional cases of Down's syndrome in this group of almost 500 pregnancies?
2 ReferencesFergal D. Malone, M.D.
Steven J. Ralston, M.D.
Mary E. D'Alton, M.D.
Tufts University School of Medicine, Boston, MA 021111
Taipale P ,Hiilesmaa V ,Salonen R ,Ylostalo P . Increased nuchal translucency as a marker for fetal chromosomal defects. N Engl J Med 1997;337:1654-1658
Full Text | Web of Science | Medline2
Pandya PP ,Snijders RJ ,Psara N ,Hilbert L ,Nicolaides KH . The prevalence of non-viable pregnancy at 10-13 weeks of gestation. Ultrasound Obstet Gynecol 1996;7:170-173
CrossRef | Web of Science | Medline
To the Editor:
Taipale et al. found increased nuchal translucency in 62 percent of the pregnancies affected by trisomies 13, 18, and 21, and 54 percent of the pregnancies affected by trisomy 21. We think their findings may be inflated as a result of verification bias.
The reference standard used in this study was either fetal karyotyping in early pregnancy or pregnancy outcome, depending on the nuchal thickness that was measured. This is worrisome, because the probability of the presence of trisomy 21 at the time of fetal karyotyping is twice the probability at term.1 Fetuses with increased nuchal translucency were always karyotyped, and therefore a chromosomal abnormality was always detected. In contrast, chromosomally abnormal fetuses with normal nuchal translucency were not always karyotyped. Approximately 50 percent of these fetuses were likely to have been aborted, and the others found to be chromosomally abnormal at birth. If for the 6 infants with Down's syndrome who had a nuchal-translucency thickness of less than 3 mm there were 6 who aborted between the time of the nuchal-translucency measurement and birth, then the detection rate for Down's syndrome would be 37 percent (7 of 19) instead of 54 percent (7 of 13). The detection rate for trisomies 13, 18, and 21 combined would decrease from 62 percent (13 of 21) to 45 percent (13 of 29).
The authors do not mention the percentage of fetuses with a nuchal-translucency thickness of less than 3 mm who were not karyotyped, nor whether this was due to early pregnancy loss. These data are necessary to the assessment of the possible effect of verification bias in their study.
1 ReferencesBen W. Mol, M.D.
St. Joseph Hospital, 7700 MB Veldhoven, the NetherlandsEva Pajkrt, M.D.
Catrina M. Bilardo, M.D., Ph.D.
Academic Medical Center, 1100 DE Amsterdam, the NetherlandsAuthor/Editor Response
The authors reply:
To the Editor: Malone et al. and Mol et al. correctly draw attention to our not having considered fetuses with Down's syndrome lost in spontaneous abortions. After 22 weeks of gestation, we karyotyped all the stillborn infants and the live-born infants who had dysmorphic features. Fetuses spontaneously aborted before 22 weeks (102 fetuses) were not karyotyped, but the number of fetuses with Down's syndrome can be estimated in two ways. First, a study of cytogenetic abnormalities among spontaneously aborted fetuses suggests that 1 of 66 has Down's syndrome at 11 to 22 weeks' gestation.1 This rate gives an estimate of about two cases in our study. Second, a study using cytogenetic-register data suggests that half the Down's syndrome pregnancies are lost between the time of chorionic-villus sampling (10 weeks) and term, and a quarter between the time of amniocentesis (16 weeks) and term.2 Little information is available, however, on the period of 13 to 14 weeks' gestation, when most screening for nuchal translucency was done, but we estimate that about one third of the pregnancies might be lost between that time and term. Thus, the four live-born babies with Down's syndrome at term in our study would imply two additional cases at 13 to 14 weeks. The two affected fetuses found by serum screening correspond to only 0.3 additional case at 13 to 14 weeks.
The best estimate, therefore, is that two fetuses with Down's syndrome were miscarried between 13 to 14 weeks' gestation and term. This adjusts our detection rate for Down's syndrome from 7 of 13 (54 percent) to 7 of 15 (47 percent). Mol et al. suggest a greater inflation, but they adjusted the birth prevalence to 10 weeks instead of the more appropriate 13 to 14 weeks. They also overadjusted for the two affected (and terminated) pregnancies in which the diagnosis was made by serum screening as if they had lasted until term.
It is not clear to what extent screening by increased nuchal translucency identifies fetuses destined to miscarry.3 Neither is this clear for screening by serum biochemistry.
Of our subjects, 9.8 percent were 35 to 39 years old. Malone et al. ask about the 138 fetuses whose necks were not adequately seen; the outcomes were known in 134, and none had abnormal karyotypes. The outcomes were not known in 50 pregnancies because either the women relocated abroad or there were clerical errors; these cases hardly bias our results.
3 ReferencesPekka Taipale, M.D.
Vilho Hiilesmaa, M.D.
Riitta Salonen, M.D.
Helsinki University Central Hospital, SF-00290 Helsinki, Finland1
Craver RD ,Kalousek DK . Cytogenetic abnormalities among spontaneously aborted previable fetuses. Am J Med Genet Suppl 1987;3:113-119
CrossRef | Medline2
Macintosh MC ,Wald NJ ,Chard T , et al. Selective miscarriage of Down's syndrome fetuses in women aged 35 years and older. Br J Obstet Gynaecol 1995;102:798-801
CrossRef | Medline3
Hackshaw AK ,Wald NJ ,Haddow JE . Down's syndrome screening with nuchal translucency. Lancet 1996;348:1740-1740
CrossRef | Web of Science | Medline
- Citing Articles (2)
Citing Articles
1
Kelley P. Anderson. (2005) Sudden Cardiac Death Unresponsive to Implantable Defibrillator Therapy: An Urgent Target for Clinicians, Industry and Government. Journal of Interventional Cardiac Electrophysiology 14:2, 71-78
CrossRef2
Theresa L. Stewart, Fergal D. Malone. (1999) First trimester screening for aneuploidy: Nuchal translucency sonography. Seminars in Perinatology 23:5, 369-381
CrossRef
