Increased Nuchal Translucency in Trisomy 13Fetuses at 10–14 Weeks of Gestation

Rosalinde J.M. Snijders, Neil J. Sebire, Roshini Nayar, Athena Souka, and Kypros H. Nicolaides*Harris Birthright Research Centre for Fetal Medicine, King’s College Hospital Medical School, London, UnitedKingdom

In a multicenter screening study for trisomy21 involving ultrasonographic measure-ment of fetal nuchal translucency thickness(NT) at 10–14 weeks of gestation, 100,311singleton pregnancies with a live fetus wereexamined. There were 46 cases of trisomy13, and in 33 (72%) of these, the NT wasabove the 95th centile. The estimated riskfor trisomy 21, based on maternal age-related risk for this chromosomal abnormal-ity and fetal NT, was above 1 in 300 in 37(80.1%) of the trisomy 13 fetuses. The fetalcrown-rump length was significantly re-duced, but the fetal heart rate was in-creased, being above the 95th centile in 64%of cases. Additionally, 24% of trisomy 13 fe-tuses had holoprosencephaly and 10% hadexomphalos. This study has demonstratedthat at 10–14 weeks of gestation, about 80%of fetuses with trisomy 13 can be identifiedin a screening program for trisomy 21, basedon a combination of maternal age and fetalNT. Am. J. Med. Genet. 86:205–207, 1999.© 1999 Wiley-Liss, Inc.

KEY WORDS: trisomy 13; prenatal diagno-sis; first trimester; ultra-sound; nuchal translucency

INTRODUCTION

At 10–14 weeks of gestation, 70% of fetuses with tri-somy 21 have increased nuchal translucency thickness(NT), and in more than 80% of affected pregnancies theestimated risk based on a combination of maternal ageand fetal NT is more than 1 in 300, which correspondswith a false-positive rate of 5% [Snijders et al., 1996a].The respective rates for trisomy 18 are about 75% and80% [Sherod et al., 1997]. In trisomy 21, the crown-

rump length (CRL) is normal and the fetal heart rate(FHR) is increased, being above the 95th centile in 21%[Kuhn et al., 1995; Hyett et al., 1996]. In trisomy 18,FHR is decreased and CRL is also decreased, beingbelow the 5th centile in 24% [Sherod et al., 1997]. Inaddition, about 30% of trisomy 18 fetuses have an ex-omphalos [Sherod et al., 1997]. This multicenter studyexamines the early sonographic findings in trisomy 13.

SUBJECTS AND METHODS

This is an ongoing ultrasound screening study at 10–14 weeks of gestation involving 22 centers in andaround London. The scans were carried out by one of306 sonographers who had received the Fetal MedicineFoundation certificate of competence in the theory andpractice of the 10–14-week scan. Fetal NT was mea-sured, as previously described, by transabdominal ul-trasound examination, unless visualization was poor,in which case vaginal sonography was carried out[Nicolaides et al., 1992].

Demographic details and ultrasound findings, in-cluding number of fetuses, CRL, NT, any obvious fetalabnormalities, and, since 1994, FHR, were entered intoa computer database at the time of scanning. Karyo-type results and details on pregnancy outcome wereadded as soon as these became available. Pregnancyoutcome was obtained from the maternity units or thepatients themselves.

A computer search was made to identify all singletonpregnancies with live fetuses at the 10–14-week scanand of crown-rump length of 38–84 mm that had anestimated date of delivery before June 1, 1997. Thedatabase was then searched for all cases with triso-my 13.

In the trisomy 13 fetuses, the distribution of risks fortrisomy 21 by maternal age and NT was determined,and the sensitivity of a cutoff risk of 1 in 300 (corre-sponding to 5% of the population) was calculated [Pan-dya et al., 1995]. In addition, the NT for CRL, the CRLfor gestation (in those with regular 26–30-day men-strual cycles and a certain date for last menstrual pe-riod), and the FHR for CRL were expressed as eithermultiples of the normal median (MoM), or the numberof standard deviations by which the individual valuesdiffered from the appropriate normal mean (deltavalue). Student’s t-test was used to examine the sig-nificance of the difference in MoMs or delta values be-

Contract grant sponsor: Fetal Medicine Foundation; Contractgrant number: 1037116.

*Correspondence to: Professor K.H. Nicolaides, Harris Birth-right Research Centre for Fetal Medicine, King’s College HospitalMedical School, Denmark Hill, London SE5 8RX, UK.

Received 10 April 1997; Accepted 21 October 1997

American Journal of Medical Genetics 86:205–207 (1999)

© 1999 Wiley-Liss, Inc.

tween the trisomy 13 and normal pregnancies. Regres-sion analysis was used to examine the possible signifi-cance of associations between NT, CRL, and FHR. Inthose trisomy 13 fetuses with holoprosencephaly or ex-omphalos, the NT was compared to those trisomic fe-tuses without these defects.

RESULTS

During the study period (September 1992–December1996), ultrasound examination was performed in100,311 singleton pregnancies with a live fetus and afetal CRL of 38–84 mm. The median CRL was 58 mm,which corresponds to the gestational age of 12 weeks.First- or second-semester karyotyping was performedin 9272 cases. Second-trimester karyotyping was car-ried out in 3,581 cases. In addition, karyotyping wasperformed as part of postmortem examination in 49pregnancies that resulted in spontaneous abortion orperinatal death, and it was performed postnatally in 24cases where infants were found to have dysmorphicfeatures suggestive of chromosomal abnormality.

There were 46 cases of trisomy 13, and in all casesthe diagnosis was made prenatally. The fetal NT wasabove the 95th centile for CRL in 33 (72%) of the casesand above the 99th centile in 25 (54.4%) of the cases(Fig. 1); the median deviation was 2.1 mm above theexpected normal median for CRL (t 4 7.6, P < 0.0001).The estimated risk for trisomy 21 based on maternalage and fetal NT was above 1 in 300 in 37 cases(80.4%). Normal menstrual cycles were recorded in 37cases, and the median CRL for gestation was lowerthan normal (median difference 4 2 days, t 4 −2.8, P4 0.01), while it was below the 5th centile of normalrange in 8 (22%) cases. The FHR was recorded in thelast 25 cases and the median value was 2.3 SD abovethe normal mean for CRL (t 4 7.04, P < 0.001); in 16(64.0%) cases, FHR was above the 95th centile (Fig. 1).There was no significant association between deviationof NT and delta FHR (t 4 0.10, P 4 0.92).

Holoprosencephaly was detected prenatally in 11(23.9%) cases. The CRL was more than 45 mm in 41cases, and in 3 (7.3%) of these there was exomphalos.There was no significant difference in NT betweenthose with and without holoprosencephaly (t 4 −0.58,P 4 0.56) or exomphalos (t 4 1.5, P 4 0.12). In 2 of the9 cases where the risk for trisomy 21 was less than 1 in300, there was holoprosencephaly; therefore, screeningby maternal age, fetal NT, and holoprosencephalywould identify 85% (39 of 46) cases of trisomy 13.

DISCUSSION

This study has demonstrated that at 10–14 weeks ofgestation, about 70% of fetuses with trisomy 13 haveincreased NT, and about 80% of affected pregnanciescan be identified in a screening program for trisomy 21based on a combination of maternal age and fetal NT.Major structural defects, such as holoprosencephaly orexomphalos, are present in about 30% of cases. In ad-dition, trisomy 13 is characterized by fetal tachycardia,observed in about two thirds of cases.

The increased FHR may be a consequence of devel-opmental delay in the cardiovascular control mecha-nisms, since in normal fetuses, FHR decreases fromabout 170 bpm at 9 weeks to about 150 bpm at 14weeks, as a result of maturation of the parasympa-thetic nervous system [Wladimiroff and Seelen, 1972;Robinson and Shaw-Dunn, 1973; Wisser and Dirschedl,1994]. Alternatively, the tachycardia of trisomy 13 fe-tuses may be a manifestation of cardiac failure, whichmay also explain the increased NT in these fetuses.However, there was no significant association betweenNT and FHR.

The main clinical findings in trisomy 13, first de-scribed by Patau et al. [1960], are holoprosencephaly,facial cleft, and polydactyly [Jones, 1988]. With the in-troduction of routine antenatal ultrasound examina-tion, many anomalies were reported in affected fetuses

Fig. 1. Fetal crown-rump length (CRL) with gestation (left), fetal nuchal translucency thickness (NT) (middle), and heart rate (FHR) (right), withCRL in trisomy 13 fetuses plotted on the appropriate normal range.

206 Snijders et al.

in the second and third trimesters of pregnancy. In astudy of 54 trisomy 13 cases, about 40% had holopros-encephaly, posterior fossa cyst or enlarged cisternamagna, facial cleft, and cardiac defects, about 50% hadrenal defects and abnormal limbs, in 22% there wasnuchal edema, and 15% had exomphalos [Snijders etal., 1996b]. However, this was not a screening studythat included all cases of trisomy 13, but a study re-porting on the karyotype of fetuses with abnormalitiesdetected during routine ultrasound examination; con-sequently, the prevalence of associated defects wouldhave been overestimated.

In the present study of routine ultrasound scanningat 10–14 weeks of gestation, primarily for measure-ment of fetal NT, the major defect identified was holo-prosencephaly in 24% of cases. This lower prevalence,compared to the second-trimester (39%) studies, maybe because in the first-trimester study only the mostsevere types of alobar holoprosencephaly were identi-fied. Alternatively, the prevalence of holoprosen-cephaly in the second-trimester studies was increasedby preselection of the patients undergoing karyotyping.Apart from exomphalos, observed in 7% of our cases,the other defects associated with trisomy 13 have notbeen identified in the current study. However, suchdefects as polydactyly and cystic kidneys can now bedetected by ultrasound at 10–14 weeks [Sepulveda etal., 1997], and it is therefore likely that detailed scan-ning in patients with increased NT would lead to thediagnosis of not only the major defects but also themore subtle ones that are associated with trisomies.

In this study, increased fetal NT was found in 76% oftrisomy 13 fetuses, whereas in the second trimesteronly 22% of affected fetuses had nuchal edema. Thereare three possible explanations for this apparent dis-crepancy. First, in the 10–14-week study, the primaryindication for karyotyping was increased NT, andtherefore there was preselection for this marker; how-ever, all cases of trisomy 13 in the sample were iden-tified either by antenatal karyotyping or postnatal ex-amination. Second, the nuchal abnormality in trisomy13, as with trisomy 21, may be a transient phenomenonthat resolves with increasing gestation. Thirdly, casesof trisomy 13 with increased fetal NT may die in utero;in a previous study, we estimated that the rate of in-trauterine lethality of trisomy 13 fetuses between 12weeks and term was about 80% [Snijders et al., 1995b].

Screening for trisomy 21 by a combination of mater-

nal age and fetal NT at 10–14 weeks of gestation hasnow been shown, in a multicenter study involving morethan 60,000 pregnancies, to identify more than 80% ofaffected fetuses for a false-positive rate of 5% [Nico-laides et al., 1997]. The findings of the present studyindicate that the same method of screening can alsoidentify about 80% of trisomy 13 pregnancies.

REFERENCES

Hyett JA, Noble PL, Snijders RJ, Montenegro N, Nicolaides KH. 1996.Fetal heart rate in trisomy 21 and other chromosomal abnormalities at10–14 weeks of gestation. Ultrasound Obstet Gynecol 7:239–244.

Jones KL. 1988. In: “Smiths Recognisable Patterns of Human Malforma-tion.” Philadelphia: W.B. Saunders. pp 20–25.

Kuhn P, Brizot ML, Pandya PP, Snijders RJ, Nicolaides KH. 1995. Crown-rump length in chromosomally abnormal fetuses at 10 to 13 weeks’gestation. Am J Obstet Gynecol 172:32–35.

Nicolaides KH, Azar G, Byrne D, Mansur C, Marks K. 1992. Fetal nuchaltranslucency: Ultrasound screening for chromosomal defects in firsttrimester of pregnancy. Br Med J [Clin Res] 304:867–869.

Nicolaides KH, Sebire NJ, Snijders RJM. 1997. Down’s syndrome screen-ing with nuchal translucency. Lancet 349:438.

Pandya PP, Snijders RJM, Johnson SJ, Brizot M, Nicolaides KH. 1995.Screening for fetal trisomies by maternal age and fetal nuchal trans-lucency thickness at 10 to 14 weeks of gestation. Br J Obstet Gynaecol102:957–962.

Patau K, Smith DW, Therman E, Inhorn SL, Wagner HP. 1960. Multiplecongenital anomaly caused by an extra autosome. Lancet 1:790–793.

Robinson HP, Shaw-Dunn J. 1973. Fetal heart rates as determined bysonar in early pregnancy. J Obstet Gynecol Br Commonw 90:805–809.

Sepulveda W, Sebire NJ, Souka A, Snijders RJM, Nicolaides KH. 1997.Diagnosis of the Meckel-Gruber syndrome at 11–14 weeks’ gestation.Am J Obstet Gynecol 176:316–319.

Sherod C, Sebire NJ, Soares W, Snijders RJM, Nicolaides KH. 1997. Fetaltrisomy 18 at 10–14 weeks of gestation. Ultrasound Obstet Gynecol 10:387–390.

Snijders RJ, Sebire NJ, Souka A, Santiago C, Nicolaides KH. 1995a. Fetalexomphalos and chromosomal defects: Relationship to maternal ageand gestation. Ultrasound Obstet Gynecol 6:250–255.

Snijders RJM, Sebire NJ, Nicolaides KH. 1995b. Maternal age and gesta-tional age specific risk for chromosomal defects. Fetal Diagn Ther 10:356–367.

Snijders RJM, Johnson S, Sebire NJ, Noble PL, Nicolaides KH. 1996a.First-trimester ultrasound screening for chromosomal defects. Ultra-sound Obstet Gynecol 7:216–226.

Snijders RJM, Farrias M, Von Kaisenberg C, Nicolaides KH. 1996b. Fetalabnormalities. In: Snijders RJM, Nicolaides KH, editors. Ultrasoundmarkers for fetal chromosomal defects. Casterton, UK: ParthenonPress.

Wisser J, Dirschedl P. 1994. Embryonic heart rate in dated human em-bryos. Early Hum Dev 37:107–115.

Wladimiroff JW, Seelen JC. 1972. Fetal heart action in early pregnancy.Development of fetal vagal function. Eur J Obstet Gynecol 2:55–63.

First-Trimester Trisomy 13 207