Antenatal detection of Edwards (Trisomy 18) and Patau (Trisomy 13) syndrome: England and Wales 2005-2012

Abstract

Objectives

Pregnancies with Edwards or Patau syndrome are often detected through screening for Down’s syndrome. We aimed to evaluate the impact of screening for Down’s syndrome on the prevalence of live births and antenatal diagnoses of Edwards and Patau syndrome.

Setting

England and Wales, 2005 to 2012.

Methods

Data from the National Down Syndrome Cytogenetic Register, which contains information on nearly all ante- or postnatally diagnosed cases of Edwards or Patau syndrome in which a karyotype was confirmed, were analysed.

Results

From 2005 to 2012, 3,941 diagnoses of Edwards syndrome and 1,567 diagnoses of Patau syndrome were recorded (prevalence of 7.0 and 2.8 per 10,000 births respectively). Only 11% (95% confidence interval [CI]: 10–12) of diagnoses of Edwards syndrome and 13% (95% CI: 11–14) of Patau syndrome were live births, resulting in live birth prevalences of 0.8 (95% CI: 0.7–0.8) and 0.4 (95% CI: 0.3–0.4) per 10,000 live births respectively. About 90% of pregnancies with Edwards or Patau syndrome were diagnosed antenatally, and this proportion remained constant over time. The proportion of diagnoses detected before 15 weeks increased from 50% in 2005 to 53% in 2012 for Edwards syndrome, and from 41% in 2005 to 63% in 2012 for Patau syndrome.

Conclusions

Almost 700 women per year had a pregnancy with Edwards or Patau syndrome. Over 90% of these pregnancies were detected antenatally, with the increased use of first trimester screening for Down’s syndrome resulting in the reduction in the mean gestational age at diagnosis of these syndromes.

Introduction

Edwards syndrome (trisomy 18) and Patau syndrome (trisomy 13) are the second and third most common autosomal trisomies in live births (Down’s syndrome, trisomy 21, is the most common), with live birth prevalence in the absence of antenatal screening and selective termination of 2.3 per 10,000 births for Edwards syndrome and 1.4 per 10,000 births for Patau syndrome. ¹ Both syndromes affect multiple systems, and have a high risk of fetal death ² and a median postnatal survival of two weeks, with only 1 in 12 surviving for one year or more. ³

Both first and second trimester screening tests for Down’s syndrome also identify some pregnancies with an increased risk of Edwards or Patau syndrome. The fetal anomaly scan between 18⁺⁰ and 20⁺⁶ weeks’ gestation will, in addition, detect structural anomalies associated with these syndromes. In 2010 the NHS Fetal Anomaly Screening Programme in England assigned a target detection rate of 90% for both Edwards and Patau syndromes for the fetal anomaly scan. Antenatal diagnoses of both syndromes can be obtained by karyotyping cells from amniocentesis or chorionic villus sampling.

This study used data from the National Down Syndrome Cytogenetic Register (NDSCR) to determine the proportions of pregnancies with these syndromes that are antenatally detected in England and Wales, and to evaluate whether the increased use of the first trimester combined screening test for Down’s syndrome from 2005 to 2012 has influenced the antenatal detection of these syndromes.

Methods

The data

Since 1 January 2004, all 20 clinical cytogenetic laboratories in England and Wales have completed a form for the NDSCR for each diagnosis of Edwards syndrome and Patau syndrome, and their variants. ⁴ Almost every baby with clinical features suggesting Edwards or Patau syndrome, as well as any antenatal diagnostic sample from a pregnancy suspected to have either syndrome, receives a cytogenetic examination, as the definitive test for the syndromes is detection of a whole or part of an extra chromosome 18 (Edwards syndrome) or chromosome 13 (Patau syndrome). The cases include live births, fetal deaths comprising stillbirths (24+ weeks) and late miscarriages (20–23 weeks) and terminations of pregnancy for fetal anomaly. Earlier fetal deaths (before 20 weeks) are rarely karyotyped, and those that were registered are unlikely to be representative of those in the population.

Data collected on each case include the sampling date, place of and indications for referral, karyotype, maternal age, family history, maternal postcode of residence, pregnancy outcome and the date of the outcome, and enough information to detect duplicates. Most laboratories send a copy of this form to the referring physician, for confirmation and completion. Annual reports of the findings are produced, the most recent for the year 2012. ⁵

The data from the NDSCR have been compared with those from other congenital anomaly registers in England and Wales, and with those of the Office for National Statistics. These comparisons have shown that, since its inception in 1989, the register has captured data for an estimated 93% of all diagnosed births and pregnancy terminations for Down’s syndrome. ⁶ Data from 2004 were excluded, as it was the first year of data collection of Edwards and Patau syndrome, and there was likely to be under-reporting in that year. Since then it is likely that the register is similarly complete for Down's, Edwards and Patau syndromes.

Statistical Analysis

Poisson regression was used to analyse any changes in prevalence from 2005 to 2012. All comparisons between categorical variables were made using chi-squared tests, and between continuous variables using t-tests, as appropriate. The Poisson distribution was used to calculate 95% confidence intervals (CI) for prevalence, and the binomial distribution was used to calculate 95% CIs for proportions. Statistical analyses were carried out using STATA version 12.

Results

From 2005 to 2012, 3,942 diagnoses (live births, stillbirths [24+ weeks], late miscarriages [20–23 weeks] and terminations of pregnancy for fetal anomaly) of Edwards syndrome, and 1,567 of Patau syndrome were notified to the NDSCR; a prevalence of 7.0 (95% CI: 6.8 to 7.2) per 10,000 (live and still) births for Edwards syndrome and 2.8 (95% CI: 2.6 to 2.9) per 10,000 births for Patau syndrome. The prevalence of both syndromes remained constant between 2005 and 2012. The prevalence of Edwards syndrome was significantly higher amongst females than males: 7.4 (95% CI: 7.1 to 7.8) and 5.7 (95% CI: 5.4 to 6.0) per 10,000 births respectively. The prevalence of Patau syndrome did not differ between females and males: 2.5 (95% CI: 2.3 to 2.7) and 2.7 (95% CI: 2.5 to 2.9) per 10,000 births respectively.

The majority of cases of Edwards and Patau syndrome had an additional chromosome (97% and 90% respectively), as previously observed. ⁷ Just over 1% of cases of both Edwards and Patau syndrome were mosaic (some cells had an additional chromosome and others did not). However, there was a much larger proportion of cases with translocation Patau syndrome than translocation Edwards syndrome (8% vs. 1%) (each cell had an additional part of a chromosome, not the full additional chromosome).

There were 69 twins (1.8%; 95% CI: 1.4% to 2.2%) with Edwards syndrome (eight in whom both twins were affected) and 35 twins (2.2%; 95% CI: 1.6% to 3.1%) with Patau syndrome (four in whom both twins were affected), compared with the national twinning rate of 1.6% in 2012. ⁸

The prevalence of both syndromes increased significantly with maternal age, the mean age of mothers being 36.3 years for Edwards syndrome and 34.4 years for Patau syndrome, significantly higher than for all mothers in England and Wales in 2012 (29.8 years). ⁹

Table 1 shows that only 11% (95% CI: 10 to 12) of diagnoses of Edwards syndrome and 13% (95% CI: 11 to 14) of Patau syndrome were live births; live birth prevalences of 0.8 (95% CI: 0.7 to 0.8) per 10,000 live births for Edwards syndrome and 0.4 (95% CI: 0.3 to 0.4) per 10,000 live births for Patau syndrome. The live birth prevalences for both syndromes remained constant between 2005 and 2012.

Table 1.

Number of cases, percentage and 95% confidence intervals of Edwards and Patau syndrome according to birth outcome; England and Wales: 2005–2012.

Outcomes	Edwards syndrome		Patau syndrome
	Number	Percentage [95% CI]	Number	Percentage [95% CI]
Live birth	432	11 [10–12]	199	13 [11–14]
Stillbirth	212	5 [5–6]	44	3 [2–4]
Miscarriage	58	1 [1–2]	31	2 [1–3]
Termination	2,931	74 [73–76]	1,213	77 [75–79]
Antenatal diagnoses with missing outcomes†	309	8 [7–9]	80	5 [4–6]
Total	3,942	100	1,567	100

†

The outcomes are missing usually due to administrative reasons and it would be expected that at least 90% are terminations.

Table 2 shows that 90% of Edwards syndrome and 89% of Patau syndrome cases were diagnosed antenatally, with the diagnoses occurring earlier for Edwards syndrome; 49% (95% CI: 47% to 51%) occurred before 15 weeks’ gestation for Edwards syndrome, compared with 45% (95% CI: 42% to 47%) for Patau syndrome. The proportions of diagnoses occurring before 15 weeks’ gestation significantly increased from 50% in 2005 to 53% in 2012 for Edwards syndrome (p = 0.02) and from 41% in 2005 to 63% in 2012 for Patau syndrome (p < 0.001).

Table 2.

Number of cases, percentage and 95% confidence intervals of Edwards and Patau syndrome according to timing of diagnosis; England and Wales: 2005–2012.

Timing of diagnosis	Edwards syndrome		Patau syndrome
	Number	Percentage [95% CI]	Number	Percentage [95% CI]
Antenatal	3,560	90 [89–91]	1,398	89 [88–91]
<15 weeks	1,932	49 [47–51]	702	45 [42–47]
15–17 weeks	421	11 [10–12]	158	10 [9–12]
18+weeks	992	25 [24–27]	444	28 [26–31]
Not known	215	5 [5–6]	94	6 [5–7]
Postnatal	382	10 [9–11]	169	11 [9–12]
Total	3,942	100	1,567	100

After an antenatal diagnosis of Edwards syndrome, 10% of women decided to continue the pregnancy. Of these, 63% (95% CI: 58 to 69) of fetuses were stillbirths and miscarriages. This was similar for Patau syndrome, with 8% of women continuing with the pregnancy, in which 42% (95% CI: 33 to 51) of the fetuses were stillbirths and miscarriages. These fetal loss rates are consistent with those reported by Morris et al. ¹⁰

Table 3 shows that 63% of mothers with fetuses with Edwards syndrome and 57% with Patau syndrome decided to have a karyotype due to first trimester screening. A higher proportion of babies with Patau syndrome proceeded to a diagnostic test as a result of an ultrasound scan at around 18⁺⁰ to 20⁺⁶ weeks than those with Edwards syndrome (30% v 24%).

Table 3.

Number of cases, percentage and 95% confidence intervals of Edwards and Patau syndrome according to reason for antenatal karyotyping; England and Wales: 2005–2012.

Reason for karyotyping	Edwards syndrome		Patau syndrome
	Number	Percentage [95% CI]	Number	Percentage [95% CI]
First trimester screening	2,245	63 [61–65]	795	57 [54–59]
Second trimester screening	289	8 [7–9]	111	8 [7–9]
Ultrasound	868	24 [23–26]	415	30 [27–32]
Age	38	1 [1–1]	17	1 [1–2]
Other reasons/No information	120	3 [3–4]	60	4 [3–5]
Total	3,560	100	1,398	100

Of the 382 postnatal diagnoses for Edwards syndrome (Table 2), 81 were recorded as declining screening, 98 were screened, and there were no data for 203. It is therefore known that at least 21% (81/382) were not diagnosed as screening had been declined. Similarly, for 169 postnatal diagnoses of Patau syndrome, 35 declined screening, 40 were screened, and there was no information for 94; at least 21% were not diagnosed as screening had been declined. When including only those women who accepted screening, the proportion antenatally diagnosed is 92% (3,559/(3,941-81)) for Edwards syndrome and 91% (1,398/(1,567-35)) for Patau syndrome, both of which fall above the NHS Fetal Anomaly Screening Programme target detection rate of 90%.

Figures 1 and 2 show that babies with Edwards and Patau syndrome tended to be born earlier than unaffected babies (median gestational age is 37 weeks), and were much lighter than unaffected babies, even when adjusted for gestational age, with over 75% of babies with Edwards syndrome and 22% of babies with Patau syndrome lying below the 3^rd centile. The mean birth weight at 37 weeks’ gestation was 1900 g (1st centile) for Edwards syndrome, and 2600 g (23^rd centile) for Patau Syndrome.

Figure 1.

The distribution of gestational age at birth in babies with Edwards (a) and Patau (b) syndrome (bars) (2005–2012) compared with unaffected babies (line) (Office for National Statistics 2011).

Figure 2.

Proportion of live born cases of Edwards (a) and Patau (b) syndrome (bars) and unaffected cases (curve) according to birth weight centiles adjusted for gestational age; England and Wales: 2005–2012.

Discussion

This study shows that the prevalence of Edwards syndrome was 7.0 per 10,000 births, and the prevalence of Patau syndrome was 2.8 per 10,000 births in England and Wales between 2005 and 2012. During this time, 90% of cases of Edwards syndrome and Patau syndrome cases were diagnosed antenatally, and the proportions diagnosed before 15 weeks’ gestation significantly increased for both syndromes.

Strengths and limitations of the study

The major strength of this study is that it is the largest of its kind, covering the whole population of England and Wales, with a likely 93% ascertainment level.

The major weakness of this study is the missing data relating to pregnancy outcome (7%). The missing outcomes were largely due to administrative delays in receiving data on pregnancy outcomes after an antenatal diagnosis. However, these delays are unrelated to the outcome in cases that we have subsequently managed to trace, of which 90% were terminations of pregnancy.

The NDSCR also lacks information on anomalies associated with Edwards and Patau syndrome, which will affect the diagnosis, survival, and outcome. Approximately 45% of babies with Edwards syndrome and 35% of babies with Patau syndrome have a heart anomaly. Babies with Edwards syndrome frequently have trachea-oesophageal fistula, cleft lip, diaphragmatic hernia, and spina bifida, while babies with Patau syndrome frequently have oro-facial clefts, abdominal wall defects, limb defects, and nervous system anomalies. ¹¹

Comparison with other studies

The prevalence of Edwards and Patau syndrome were highly dependent on maternal age as, has been observed in many other studies.^12,13 This may partially explain why the total prevalence of Edwards and Patau syndrome in this study (7.0 and 2.8 per 10,000 births respectively) were significantly higher than other European registries that are members of the European Surveillance of Congenital Anomalies (EUROCAT) ¹⁴ (4.4 and 1.7 per 10,000 births). Only the Paris registry had a higher prevalence. A second explanation may be that England introduced first trimester screening at an earlier time than many other European countries, ¹⁵ and therefore more pregnancies are being detected during the first trimester rather than resulting in fetal losses later in pregnancy before being diagnosed. This will increase the total prevalence, but as many of these pregnancies result in fetal losses, it will not increase the live birth prevalence. As expected, the live birth prevalences were comparable for both Edwards (0.77 per 10,000 live births in this study and 0.79 per 10,000 live births in EUROCAT registries) and Patau syndrome (0.36 and 0.37), due to the high level of fetal loss and terminations of pregnancy for fetal loss for both of these syndromes. The high ascertainment of cases in the NDSCR data ⁶ may also partially explain the higher prevalence in England and Wales compared with the other European registries.

The early gestational age at birth and the low birth weight observed in this study are consistent with those observed by Boghossian et al in their 2012 study of 1,053 infants with Edwards syndrome aged 22–41 weeks, and 613 infants with Patau syndrome aged 22–40 weeks. ¹⁶ Due to the high levels of prenatal detection, the gestational age at birth should be interpreted with caution, as it may be influenced by clinicians deciding to induce the birth early. However, there is strong evidence that the median gestational age at birth is 38 weeks rather than 40 weeks gestation for babies with Down’s syndrome, and the same effect may be occurring in Patau and Edward syndrome. ¹⁵

The future

It is likely that cell-free fetal DNA screening for Down’s syndrome will become widely available in Europe during the next few years, and that such screening will also include screening for Patau and Edwards syndrome. The availability of a more accurate screening test may encourage more women to have screening and the detection of affected pregnancies to occur earlier. However, as over 90% of such affected pregnancies are already detected prenatally, it is unlikely that a large impact on the live birth rate will occur.

Conclusion

From 2005 to 2012 almost 700 women per year in England and Wales had a pregnancy with Edwards or Patau syndrome. Over 90% of these pregnancies were detected antenatally, with a continual decrease in the gestational age at diagnosis. At least a quarter of the 10% of cases not detected antenatally were not diagnosed because the parents did not wish to have an antenatal diagnosis, not because the screening tests missed cases.