1 INTRODUCTION

Birth defects are the leading causes of death in babies under 1 year of age.1 Neural tube defects (NTDs) have a birth incidence of approximately 1 in 1000 in American Caucasians and are the second most common groups of birth defect after congenital heart defects. NTDs affect an estimated 300,000 pregnancies worldwide each year,2 of which many are lost due to miscarriage or termination of pregnancy.

NTDs occur due to a defect in neurulation in the embryonic period, this results in the fetal brain and/or spinal cord not developing as normal. This study focused on three of the most common open NTDs: spina bifida, anencephaly and encephalocele. NTDs may be accompanied by other chromosomal, genetic and developmental abnormalities; additional anomalies increase the risk of permanent disability and potential mortality.3

The aetiology of NTDs is a complex interaction of genetic and environmental factors. Modifiable risk factors include maternal diabetes, obesity, exposure to cigarette smoke, medications (e.g., methotrexate); and nutritional factors (folate deficiency). The mechanisms of how these factors affect the process of neural tube closure remain largely unknown. A deeper understanding of the aetiology of NTDs will make it possible to find further strategies of prevention.2

The relationship between the risk of NTDs and folate deficiency has been widely researched and is proven to significantly increase the risk of NTDs. The use of folic acid supplements in pregnancy has been explored and folic acid deficiency is now arguably the strongest established risk factor, associated with a two- to eight-fold increased risk of NTD.4-6 Despite strong recommendations and campaigns to raise awareness, a substantial proportion of women of reproductive age are still unaware of the benefits of periconceptional folic acid supplementation.

Due to these difficulties numerous countries have introduced mandatory fortification of flour and grains with folic acid.7 This population wide approach is effective and has reduced the prevalence of NTDs worldwide. This fortification of staple foods has also helped battle the limitations of previous campaigns, such as incomplete outreach by reducing the socioeconomic inequalities in the prevalence of NTDs.8 Despite knowledge regarding prevention, NTDs remain highly prevalent in the UK and worldwide and continue to be a great public health burden globally.9

2 METHODS AND MATERIALS

This is an observational prospective study looking at terminations of pregnancy in England and Wales. This study is in collaboration with Kings College London and the Department of Health & Social Care (DHSC), England and Wales. The overall trends in terminations were analysed, whilst specifically focusing on those cases terminated due to a NTD, under Ground E, which states that – there is substantial risk that if the child were born it would suffer from physical or mental abnormalities as to be seriously handicapped.

2.1 The study population

The study period covered terminations that took place in England and Wales between 2007 and 2017. Cases for non-residents to England and Wales (including Northern Ireland and Scotland) were excluded from analysis.

In England and Wales abortions are carried out in accordance with the Abortion Act 1967, one of the requirements stated by this act is the completion of a Health and Social Act 4 (HSA4) form. The HSA4 form is a notification of the abortion and must be completed by the doctor and sent to the Chief Medical Officer within 14 days.

The information presented in this study is based on the HSA4 forms submitted by clinics and hospitals to the DHSC. The forms are in standard use throughout England and Wales, where recording and declaring abortion cases is a mandatory requirement. The DHSC receives and collates these HSA4 notification forms, from where, the data is extracted, stored and analysed. The results are published annually and are available in public domain.10

2.2 Inclusion criteria All cases of Ground E terminations are subdivided into the main conditions coded using the International Classification of Diseases (ICD10). Three diagnostic codes were included in this study:

Q00 – Anencephaly

Q01 – Encephalocele

Q05 – Spina bifida

2.3 Variables and categorization of data

On the HSA4 forms, many patient variables are recorded relating to both maternal characteristics and treatment details. The variables analysed in this study were: fetal diagnosis, gestational age, method of termination, fetocide, maternal age, maternal ethnicity, marital status and obstetric history. These variables were analysed, with an aim to determine the trends of termination of pregnancy with regard to above parameters.

2.4 Statistical analysis

All statistical analyses were performed using the SPSS (version 25.0). A Chi-squared analysis was used to evaluate the strength of association between either maternal ethnicity, maternal age or gestational age and the NTD diagnosis (anencephaly, encephalocele or spina bifida) of their affected pregnancy. The ethnic group, ‘not known/not stated’ was excluded in analysis which left five ethnic groups. For gestational age all four groups were used and for maternal age all three groups.

All statistical tests were two-sided with a 95% confidence interval for significance (p-value of <0.05). As several hypotheses were tested Bonferroni correction was used to compensate for the multiple comparisons.

3 RESULTS 3.1 Overall trend

From 2007 to 2017 there were 2,145,607 terminations recorded in England and Wales. 2,080,844 were for residents of England and Wales and the remaining 64,763 for non-residents. Only resident data from 2007 to 2017 is used in the following analysis (see Figure 1).

Terminations under Ground E from 2007 to 2017

In 2017, there were 192,000 terminations for residents in England and Wales. An increase of 3.5% since 2016 (185,596 cases) and a 3.4% decrease since 2007 (198,499 cases). This is the highest number of terminations recorded since 2008 when there were 195,296 terminations for residents of England and Wales. Other than the peak in 2007 there has been a gradual overall increase in total number of terminations over the past few decades.

3.2 Ground E

Ground E terminations account for around 2% of all terminations in England and Wales, of which around 20% are due to anomaly with the Nervous System.

The number of terminations under Ground E rose steadily with 1938 cases in 2007 and 3314 in 2017 – an increase of 71.0% (see Figure 1). The number of termination of pregnancies for NTD diagnosis also progressively increased over this time period, from 308 cases in 2007 to 517 in 2017 – a 67.9% rise (see Figure 2). This rise reflects the overall increase in Ground E terminations and so the proportion of Ground E cases with a NTD diagnosis has remained relatively constant – 15.9% of cases in 2007% and 15.6% in 2017. This proportion fluctuated over the decade but remained relatively constant.

Termination for Neural tube defects (NTDs) from 2007 to 2017

Over the same time period there were 2297 cases of anencephaly (the most common), 1771 cases of spina bifida and 407 cases of encephalocele. Over the time period each of these three diagnoses showed increasing trends independently: anencephaly rates rose by 87.1%, encephalocele by 52.9% and spina bifida by 49.6% (see Figure 3).

Prevalence of Anencephaly, Encephalocele and Spina Bifida

3.3 Gestation

In recent years around 80% of terminations are performed <10 weeks gestation; 10% between 10 and 12 weeks; 8% between 13 and 19 weeks and only 2% at ≥20 weeks gestation.

Terminations under Ground E or for NTDs are generally performed at a later gestation, most commonly between 13 and 19 weeks (see Table 1). A much greater number are performed at ≥20 weeks gestation and many fewer at <13 weeks gestation compared with all terminations.

TABLE 1. Gestational age at termination for Ground E and NTD cases Gestation (weeks) Ground E NTDs Cases % Cases % ≤9 217 0.75 17 0.38 10–12 3789 13.13 1238 27.98 13–19 15,261 52.87 1701 38.44 ≥20 9599 33.25 1469 33.20

Over the time period, records show that earlier terminations for NTD cases are increasing. In 2007, only 58 cases (19.0%) were terminated at 10–12 weeks compared with 136 (26.8%) cases in 2017. However, no subsequent decline in cases performed ≥20 weeks gestation is seen.

The prevalence of each NTD diagnosis differed at each gestational age (see Figure 4):

Type of Neural tube defect (NTD) at different gestations

A termination following the diagnosis of anencephaly was significantly higher at 10–12 weeks (1114 cases – 48.5%) and significantly lower at ≥20 (157 cases – 6.8%) (p < 0.0001), compared with all other gestational age groups.

A termination following the diagnosis of encephalocele was significantly higher at 13–19 weeks (227 cases – 55.8%) compared with 10–12 weeks (73 cases – 17.9%) or ≥20 weeks (105 cases – 25.8%) (p < 0.0001); encephalocele was also significantly lower at 10–12 weeks compared to 13–19 weeks (p < 0.0001).

A termination following the diagnosis of spina bifida was significantly higher at ≥20-weeks (1214 cases – 68.5%) and significantly lower at 10–12-weeks (67 cases – 3.8%) (p < 0.0001) compared with other gestational age groups.

3.4 Method of termination

Medical terminations were used for 20,974 (72.7%) of Ground E terminations and 3138 (70.9%) of NTD terminations, with the remaining cases being performed via a surgical method. This division has remained roughly the same over the time period.

The method of termination varied depending on the gestational week (see Figure 5). At 10–12 weeks gestation a surgical method was more likely to be performed; 685 cases (55.3%) whereas at later gestations the majority of cases were performed medically with: 1191 cases (70.0%) at 13–19 weeks and 1384 cases (94.2%) at ≥20 weeks.

Medical or Surgical termination by Gestation

3.5 Fetocide

Of all ground E terminations, 5561 cases (19.3%) included fetocide compared with 660 (14.9%) of terminations for NTDs. From 2007 to 2017 the use of fetocide in NTD terminations has remained at around 15% throughout. Fetocide is predominantly used for NTD terminations at later gestations. 597 cases (40.6%) of terminations at ≥20 weeks gestation used fetocide; compared with only 52 cases (3%) at 13–19 weeks and 11 cases (0.9%) at 10–12 weeks.

The use of fetocide varied greatly depending on the prenatal diagnosis (see Figure 6). 509 (76.5%) of the terminations using fetocide were for spina bifida cases; 28.8% of all spina bifida cases. This compares to only 65 terminations (16.0%) of all cases of encephalocele and 91 terminations (3.96%) of all cases of anencephaly.

Fetocide use in terminations of pregnancy

3.6 Maternal age

For both Ground E and NTD terminations, more than half the women were in the 20–34 years age category, this remained constant throughout the time period. Of the 4425 cases of NTD, 3231 (73.0%) mothers were aged 20–34; 904 (20.4%) were 35 and over and 290 (6.6%) were under 20.

The prevalence of each condition varied slightly with maternal age (see Figure 7). A diagnosis of anencephaly was more common in women under 20 (55.6% of cases) and a diagnosis of spina bifida was more common in women ≥35 years (41.8% of cases). Although the adjusted residuals in these analyses indicated a strong trend, none of these outcomes were statistically significant. The sample sizes of the three maternal age groups differed which most likely contributed to the lack of statistical significance.

Type of Neural tube defect (NTD) by maternal age

3.7 Ethnicity

The ethnicities of women undergoing a termination under Ground E and for NTD were similar (see Table 2). Results stayed fairly constant over the time period, however an increase was seen in Asian or Asian British cases, with 27 cases in 2007 (8.8%) and 57 cases in 2017 (11.2%).

TABLE 2. Maternal ethnicity for Ground E and NTD terminations Ethnicity Ground E NTDs Cases % Cases % White 23,049 79.85 3603 81.42 Asian or Asian British 2734 9.47 446 10.08 Black or Black British 1236 4.28 144 3.25 Chinese or Other 457 1.58 38 0.86 Mixed 346 1.20 44 0.99 Not known or not stated 1044 3.62 150 3.39

The proportion of anencephaly, encephalocele and spina bifida was similar across ethnicities; anencephaly always the most prevalent and encephalocele always the least (see Figure 8).

Percentage of Neural tube defect (NTD) diagnosis by Ethnicity

For anencephaly (Q00) there were no significant differences across ethnicities. When directly comparing just the White and Asian ethnic groups there is a trend to significance (χ2(1) = 3.27, p = 0.071) and an examination of the group means suggests that anencephaly is less common in the White population.

A diagnosis of encephalocele (Q01) was significantly lower (p = 0.001, χ2 test) in the White ethnic group (8.5%) than the Asian (12.3%) and Black (16.0%) ethnic groups. An association between ethnicity and termination of spina bifida (Q05) was observed, χ2(4). The adjusted residuals for the Chinese and Mixed ethnic groups indicated trends to possible effects.

A diagnosis of spina bifida (Q05) was significantly higher (p = 0.007) in the White ethnic group (40.9%) compared with the Asian ethnic group (33.3%). An association between ethnicity and Q05 was observed, χ2(4) = 14.18. The adjusted residuals suggested that spina bifida was more common in the White ethnic group than other ethnic groups. However, this was not statistically significant.

4 DISCUSSION

Termination of pregnancy is one of the most commonly performed medical procedures, however still raises much debate and ethical discussion. Over the last few decades, terminations have been legalised and accepted in many countries worldwide; only in 2018 was abortion legalised in Ireland. NTDs are a largely preventable group of congenital conditions which affect infants worldwide. Over the last 20 years, folic acid deficiency has been highlighted as a key risk factor in the prevalence of NTDs. Both supplementation and fortification have been successful at reducing the prevalence of NTDs worldwide, if taken at the correct dose and periconceptionally.5 Processes are now underway for flour fortification in the UK.

Globally there is a huge variation regarding the prevalence of NTDs. The average incidence is around 1 in 1000 births, however there is marked geographical variation. The highest rates occur in the UK and USA with significantly lower rates in Japan.11 In 2015 there were an estimated 260,100 births affected by NTDs worldwide, including stillbirths and terminations.12 This is most probably an underestimate, as even comprehensive systems often fail to record spontaneous miscarriages and early fetal losses.12 For the first time in England and Wales, we have collated and analysed national data and reported on the national trends in the last 11 years for terminations for NTDs.

Many studies record a considerable decline in the prevalence of NTD live births over the past three decades.13, 14 These changes are most likely due to advances in prenatal diagnosis, increase in terminations of affected pregnancies and folic acid supplementation.15-19

Due to radically improved and now widely accessible prenatal screening most NTDs in the UK are diagnosed prenatally.20 This improved identification is reflected by the increasing trends in NTD terminations with the corresponding decrease in live NTDs birth rates. Studies show that around 75% of NTD pregnancies are electively terminated, 20% live births and ∼5% stillborn or miscarried.12, 21

Over time terminations are being performed at earlier gestations. In 2000, 43% of terminations were performed <10 weeks gestation compared with 77% in 2017.10 In general, terminations under Ground E or for NTDs are performed at later gestations than for all terminations. However, the improved antenatal screening has meant that more recently terminations have been performed at progressively earlier gestations. Ultrasound remains the gold standard diagnostic tool for NTDs although MRI and 3D ultrasound are beginning to be used as a prerequisite before surgical repair. Surgery is currently the only treatment for open NTDs and is usually performed soon after birth. The recent development of in utero surgery can also be beneficial in selected cases. The level and size of the lesion in spina bifida can be accurately identified via scanning which can then help to gauge both the short- and long-term outcomes. This is essential for patient counselling as spina bifida is not a lethal condition and many patients can live into adulthood.21 The findings of this study reflect the literature showing an increasing prevalence of terminations at an earlier gestation (10–12 weeks) over the study period. However, still spina bifida is detected commonly at routine anomaly scan at 20 weeks, hence the termination is done later, necessitating the procedure of fetocide prior to termination. A recent study has identified a first trimester sonographic marker which is associated with open spina bifida.22 This ‘crash sign’ would enable early diagnosis and therefore decrease later gestation terminations and use of fetocide.

4.1 Method of termination

Since Mifegyne was legalised in 1991 in England and Wales, there has been a continuing upward trend in the ratio of medical to surgical terminations. In 2017, 66% of terminations were performed using a medical technique compared with 35% in 2007 and 4% in 1991. The method of termination offered depends on the availability of services locally, gestation and both clinician and women's preference.

Fetocide has been used in clinical practice in the UK since the early 1990s but still raises many ethical dilemmas. The use of fetocide is recommended by RCOG for a termination after 21 + 6 weeks,23 around 1% of all terminations in England and Wales (1895 in 2017). Fetocide is often only used if the abnormality is not lethal.

This study shows a clear differentiation between the type of NTD diagnosed and the use of fetocide. Fetocide is more likely to be used for a case of spina bifida compared with anencephaly because unlike anencephaly it is not a universally fatal condition; additionally there is often a later diagnosis of spina bifida, commonly at 20-week scan. In this study 63 cases (9.5% of fetocide terminations) were for cases <20 weeks gestation. This is against RCOG guidelines and the reasoning for these cases is unclear although may be due to selective terminations in multiple pregnancies.

4.2 Maternal age

In England and Wales, the termination rate is the highest for women aged between 20 and 24 years.

Over the last decade, termination rates have been decreasing for women under 25, particularly under 20, and rising for women aged 30 and over.10 This might be a reflection of the changing demographics in the general pregnant population.

There are a few studies investigating the relationship between maternal age and NTD affected pregnancy. Some studies show a U-shaped risk, indicating an increased risk for the younger (<20 years) and the older mothers (>40 years).24, 25 Studies have shown that older women were specifically at increased risk of spina bifida.26 Our study agrees with this existing literature, a higher proportion of terminations for women >35 years were for spina bifida (41.8%) compared with other age groups. There is limited data in this field; considering the well-established relationship between older mothers and risk of congenital anomalies (e.g., trisomies), more research in this area would be beneficial.

4.3 Ethnicity

Ethnicity has only been recorded on HSA4 forms since 2002 and is self-reported by ∼98% of women. Studies have shown that considerable variation exists in the relationship between NTD prevalence, ethnicity and type of NTD. However, there is significant evidence showing that Hispanic women exhibit the highest overall rate of NTDs compared to other ethnic groups6, 27 in the USA. There are many possible reasons for this variation. A systematic review in 2013 reported that non-Caucasian women in the UK had less knowledge about and were less likely to take periconceptional folic acid supplements than Caucasian women. Campaigns across all ethnicities are important to increase awareness about folic acid supplementation.28

In our study significant differences were seen across ethnic groups. Spina bifida was significantly more common in the White ethnic group whereas encephalocele was more common in the Asian and Black ethnic groups.

4.4 Strengths and limitations

Due to patient confidentiality it was not possible to get all the information from the raw data. This would have allowed more detailed statistical analysis as well as analysis of additional variables such as selective terminations and maternal region of habitants.

Although the DHSC uses a thorough process for monitoring the HSA4 forms there is no way of ascertaining the accuracy of the data via cross-checking the medical records and therefore the validity of the data submitted in the HSA4 form cannot be ascertained.

Folic acid deficiency is the main risk factor in NTD cases. The HSA4 forms do not directly record the use of folic acid supplements so it is not possible to comment on the impact that wider use of folic acid supplements may have. This is another area that requires further research and discussion.

This was a large study, using national level data of which there have been no previous comparable studies. The large number of cases and variables analysed allowed for more accurate conclusions to be drawn. Statistical analysis was implemented, showing significant results in many areas.

5 CONCLUSION

This study shows that both ground E terminations and terminations for NTDs are increasing among residents in England and Wales. Despite considerable research into aetiology and prevention for NTDs, the termination of pregnancy for NTD is still high. Although live births affected by NTDs are decreasing, terminations are increasing each year.

More research is required to assess whether there is a natural increase in the incidence of NTDs or there are other contributing factors. Increased accuracy of prenatal diagnostics through screening regimes and improved technology may impact prevalence. Additionally, there may be a link between changing societal attitude towards terminations of pregnancy.

As prenatal surgery continues to improve, investigating its effect on the rates of termination of pregnancy with NTD would be beneficial.

Continual improvement of preventative methods and of antenatal screening are essential to reduce late gestation terminations. Additionally, further studies to investigate the impact of ethnicity and maternal age on the prevalence and diagnosis of NTDs would be valuable. This study highlights the importance of proper prevention, screening, diagnosis and counselling for women with NTD affected pregnancies.

ACKNOWLEDGEMENTS

I would like to thank the continual advice and support of the Women's Health department at Kings College London. I would also like to thank the team at the Department of Health and Social Care for their help with data organisation and formatting. There was no funding involved in this study. The projected was completed as an iBSc project at King's College London by a medical student.

CONFLICT OF INTEREST

Authors have no conflict of interest to declare. There are no relevant financial, personal, political, intellectual or religious interests.

ETHICS STATEMENT

No ethics approval was required for this study. This was confirmed by the NHS Health Research Authority and R&D at Guy's and St Thomas' Hospital.

INFORMED CONSENT

As per Department of Health and Social care guidelines (March 2020); healthcare providers should make women aware that the contents of the statutory HSA4 form used to inform the CMO of abortions will be used for statistical purposes by the Department of Health and Social Care. The data published are anonymised.