Prevalence
Although TOP is a very common procedure in England and Wales, analysis of DHSC data in this study showed that only 0.5% of all abortions were preceded by feticide, a total of 9310 feticides over the study period. Overall, there was a decrease in the prevalence of feticides across the study period, primarily due to the decrease in 2020, likely due to the Covid-19 pandemic. This affected service provision across several services in the NHS. There was probably a reduced capacity to offer feticide, secondary to understaffing. Prior to 2020, there was an overall increase in feticide from 2012 to 2019. This increase suggests that more women are having terminations later in the second and third trimesters, as feticide indicates a TOP at 22 + weeks gestation.
Method of feticide
The most common method of feticide in this study was intracardiac KCl. 67.2% of feticides in England and Wales were performed using the administration of intracardiac KCl. This is in line with the RCOG guidelines, which recommend intracardiac KCl to ensure asystole of the fetal heart[2]. Intracardiac administration of KCl is not always possible, for example, when the fetal heart is not accessible due to the position or movement of the fetus. However, this proportion may still be underestimated as cases where only KCl was recorded as the agent, but no administration route would have likely come under ‘other methods’.
Only 389 feticides (5.4%) were carried out using lidocaine through any administrative route. Although this is a small proportion, it is the second most common named method used for feticide. The RCOG guidelines state that feticide may be performed by injection of up to 30ml of 1% lidocaine through the intracardiac or umbilical route, with the caveat that it does not consistently result in fetal asystole[2]. Studies have supported this finding of a less than 100% success rate with intracardiac or umbilical lidocaine[12–14]. However, potential safety benefits of umbilical lidocaine over umbilical KCl have been suggested[14]. This may partially explain the decision to use lidocaine over KCl in these cases, despite the current guidelines.
The method used in 4.0% of feticides was recorded as either cordotomy, intraamniotic KCl/urea, umbilical KCl/urea, extra amniotic KCl/urea, or laser. No more than 10 feticides were carried out using a laser during the whole study period. This method is, therefore, very rarely used for feticide in a singleton pregnancy, compared to multiple pregnancies where interstitial laser ablation was used for 6% of selective abortions in one study[15]. Interestingly, these methods are currently being used, although rarely, despite very little evidence for their use in feticide and that there are other options available with an evidence base.
A significant proportion (22.6%) of feticide methods were coded as Other. Due to limitations with the precision of coding for feticide methods at the DHSC, many feticide cases have fallen into this category. This category includes any method not mentioned in the categories mentioned above, including digoxin. Digoxin is the only other method mentioned in the RCOG guidelines, which state that feticide may also be performed using an injection of up to 1mg of digoxin via the intraamniotic or intrafetal route. However, as with lidocaine, it is warned that it does not consistently induce fetal asystole[2]. This finding has been replicated in studies, primarily with regard to intraamniotic digoxin[3, 16, 17]. A potential explanation why digoxin is still chosen in some cases over intracardiac KCl is that intraamniotic digoxin requires less technical skill than intracardiac KCl[8].
Method of termination
More medical than surgical terminations were carried out following feticide, accounting for 58.0% of all feticides. This percentage is lower than the proportion of total abortions that are medical, which was 85% in 2020. Although medical abortions are typically preferred for TOP after 24 weeks occurring under Ground E, a significant proportion of feticides were carried out under Ground C and before 24 weeks in this study, which may explain the difference.
Statutory grounds
In England and Wales, almost all feticides were carried out under Ground E or C. 55.8% were carried out under Ground E and 43.6% under Ground C. Out of all abortions in 2020, 98.1% were performed under Ground C and 1.5% under Ground E. This difference is likely a result of when feticide takes place. All cases of feticides took place after 20 weeks, with the majority (86.4%) taking place after 22 weeks. In contrast, out of all abortions in 2020, only 1% were performed at 20 weeks or over.
The number of feticides under Ground E has gradually increased over the study period. This reflects either an increase in the prevalence of conditions or the risks such as advanced maternal age, assisted conception associated with congenital abnormalities and chromosomal abnormalities that can lead to Ground E abortions or an improvement in the diagnosis of these conditions. Prenatal screening and diagnosis have become increasingly available[18]. One study in an Austrian hospital demonstrated an increase in late TOPs of singleton pregnancies preceded by feticide across a study period of 15 years between 2004 to 2019. As abortion beyond 15 + 6 weeks is only allowed in cases of severe maternal complications or untreatable severe congenital malformations under Austrian law, all of these feticides were carried out under an equivalent of Ground E. They attributed this increase to improvements in prenatal diagnosis and detection of congenital abnormalities over time as a result of an improvement in ultrasound screening skills[19]. The continual improvement in prenatal diagnostic ability of congenital malformations is likely a contributing factor to the increase in feticides performed in the case of a Ground E diagnosis.
Ground E diagnosis
In this study, congenital malformations were reported as the primary diagnosis for 71.3% of Ground E feticides, and 18.5% were due to the presence of chromosomal abnormalities. This distribution is different from that of all abortions, where in 2018, 49% of Ground E abortions were primarily due to congenital malformations and 33% to chromosomal abnormalities. This difference is likely due to the earlier diagnosis of chromosomal abnormalities compared to congenital malformations. Screening for trisomies 13, 18 and 21 (Patau's, Edwards' or Down's syndromes) is offered to women between 10 and 14 weeks gestation. If there is a high risk of any of these abnormalities, diagnosis is made using either chorionic villus sampling (CVS) between 11–14 weeks or amniocentesis between 15–20 weeks[20]. Suppose any of these abnormalities are diagnosed, especially at earlier gestations, and a decision is made to terminate the pregnancy, the abortion is likely to take place before 22 weeks and does not require feticide.
In contrast, the majority of congenital malformations are detected at the 20-week fetal anomaly scan. By the time a decision is reached, the gestation will likely be greater than 22 weeks, and the ground E abortion will require feticide. This finding was also present in a recent study looking at abortions due to fetal anomalies in Sweden. They found that the underlying diagnoses in abortions between 12 and 16 weeks were predominantly chromosomal abnormalities, and at later gestational stages, congenital malformations were more common[21].
A third of feticides performed under Ground E was due to a diagnosis of a congenital malformation of the nervous system. This is likely due to the fact that brain development, compared to the majority of other organ development, continues through to the late second and third trimesters, especially in the formation and maturation of the cortex. This can lead to late development and detection of central nervous system (CNS) defects[19].
Gestation
Most feticides were carried out at 23 weeks. 13.5% of feticides were carried out before 22 weeks, at 20 or 21 weeks. The RCOG recommends that feticide should be routinely offered for TOPs for a fetal abnormality after 21 + 6 weeks due to the increasing likelihood of live birth following this gestational age[2]. This recommendation is the same for feticide before medical abortion under any grounds[22]. Although before 22 weeks, live birth following TOP is very uncommon, it is not impossible; thus, patients and staff are counselled and trained about this possibility. As the recommendation of feticide after 22 weeks is only guidance, clinicians may perform feticide at earlier gestations in the context of patient wishes and clinical judgement. Advice on feticide globally also varies, and in other countries, feticide may be performed at earlier gestations. For example, the WHO recommends inducing fetal demise before abortion after 20 weeks gestation[23]. Also, in the previously mentioned Austrian study of late TOPs following feticide, the minimum gestational age was 17 + 3 weeks. This suggests that feticide is used in Austria even earlier than 20 weeks[19].
Although most congenital malformations are diagnosed around 20 weeks, there are cases where TOP and feticide under Ground E take place at a gestational age beyond 24 weeks and in fewer cases beyond 30 weeks. These terminations at gestations past 24 weeks may be due to delays in the time between diagnosis and termination. However, studies have shown that most TOPs occur within 1–2 weeks of diagnosis[21, 24]. Therefore, this is unlikely to explain why 24.3% of feticides took place after 24 weeks, as if a diagnosis is made around 20 weeks, the TOP should take place in most cases within 4 weeks. Another possible reason is that some conditions are only detected when an ultrasound is done after 24 weeks, for example, in the context of slow fetal growth, polyhydramnios, or premature contractions[24]. If a diagnosis is made later, this explains why some feticides take place after 24 or even after 30 weeks gestation.
Provider
In England and Wales, most feticides took place in NHS hospital settings. The remainder occurred in NHS-approved independent clinics, with very few occurring in private hospital settings. The greater proportion of feticides taking place in hospitals compared to all abortions (60.3% vs 22%) is likely due to the more specialist skills needed to perform feticide. Additionally, after 24 weeks, TOPs and any associated feticide must take place in an NHS hospital, as independent or private clinics are registered to perform abortions for only up to 24 weeks[25].
Maternal age
The number of feticides was the highest for women aged 30–34 years. This is older than the age group for which the overall abortion rate is the highest, which is 21-year-olds. This may be due to the much higher proportion of feticides occurring under Ground E and at later gestations compared to all abortions. Although it is not the case in all patients, pregnancies prior to abortions and feticides in the second trimester under Ground E are more commonly wanted and often intended compared to those taking place under Ground C[21]. Unintended pregnancy is also associated with lower maternal age[26].
5.7% of feticides took place in women aged over 40, and 17.1% took place in those aged 35–39. It is well-established that the risk of chromosomal abnormalities increases with maternal age. However, there is limited and mixed evidence on an association of increased maternal age with non-chromosomal anomalies, which were the primary diagnosis in over 70% of feticides in this study. It has been suggested instead that the risk of non-chromosomal malformations decreases with increasing maternal age. One study found that women aged 34 years or less had an increased risk of a major fetal anomaly compared with those aged 35 and over, specifically concerning central nervous system (CNS), renal and abdominal wall malformations[27]. This, as well as the conception rate being by far the highest in women aged 25–34 in recent years[28], may explain the highest prevalence of feticide in women aged 25–34 years.
Ethnicity
The majority (78.6%) of women who had feticide reported their ethnicity as White. This is very similar to the proportion of women undergoing abortions who were of White ethnicity in 2020, which was 77%. However, 21.4% of women who underwent feticide were of non-White ethnicity, or the ethnicity was unknown. The data did not further break down this ethnicity variable, so it is impossible to draw conclusions about specific ethnic groups.
Obstetric history
There was no notable difference between the number of feticides in women who had previously given birth to a live born and/or stillborn and those who hadn't. This suggests that there is no to little association between parity and the likelihood of having a feticide. Only 17.5% of women undergoing feticide had a previous miscarriage and/or ectopic pregnancy. 17.6% of women undergoing feticide had had one or more previous abortions. This figure is 42% among all women undergoing abortions in 2020. This suggests women experiencing feticide were less likely to have had a prior abortion than all women undergoing abortions, implying the risk factors for feticide may be different to those for abortion.
Limitations
A fundamental limitation of this study is that it is retrospective. The data in this study was obtained from the DHSC and aggregated rather than available as patient-level data due to confidentiality concerns. A small sample size increased the risk of data becoming identifiable and meant that some data was grouped or given as ranges. The observational nature of the study also implies that causation could not be determined between factors affecting feticide use in England and Wales.
Additionally, this study was limited in its analysis of the methods of feticide. Not all methods are coded separately, and thus many fall under the category of Other, which was the second largest category of methods of feticide in this study. When HSA4 forms are submitted electronically by clinicians or data is uploaded by the DHSC from paper forms, not all methods of feticide have their category. For example, in the years of this study, there was no separate option for digoxin, so if this was mentioned on a form, it would be coded as Other. However, the DHSC intends on improving this in future years to increase the detail of data on feticide. For example, in 2021 digoxin was added as its own category. Future research on feticide using data from the DHSC should hopefully be able to analyse this information to a greater extent.
Further Research
As there are limitations with retrospective studies, a prospective study looking at feticide in England and Wales could provide additional insight. Despite clear recommendations from the RCOG on intracardiac KCl being the most reliable feticide method, only two-thirds of feticides in this study were confirmed as intracardiac KCl induced. Further qualitative research into clinicians' experience performing feticide, preferred methods, and why could provide valuable insight. This could also provide a further understanding of how the procedure is carried out to support both patients and clinicians involved in feticide. Furthermore, as this is a national study, it would be interesting for similar research to be done in other countries using national data and comparing how feticide is performed globally. This can provide further awareness into how feticide is carried out in different cultures, with different attitudes and laws towards abortion, specifically at later gestations. This would allow a more profound understanding for the best practice of feticide and enable services to provide the highest quality recommendations and care for patients undergoing this emotionally complex clinical procedure.