The present study reveals that the prevalence of anencephaly, encephalocele, spina bifida, microcephaly, hydrocephalus, holoprosencephaly, hydranencephaly, diaphragmatic hernia, and bilateral renal agenesis at birth was significantly higher in public institutions in Buenos Aires City. This finding aligns with a prior study conducted on a smaller sample and over a more limited timeframe (Bronberg et al. 2020). Additionally, another study utilizing nationwide data also demonstrated higher prevalence rates in public institutions (Bronberg et al. 2021).
The association between the frequency of specific CAs and low socioeconomic status has been investigated in previous research. In Argentina, Pawluk and colleagues examined the correlation between 25 specific CAs and adverse social determinants. Their case-control study revealed a significant association between cleft lip with or without cleft palate, ventricular septal defect, and poverty indicators (Pawluk et al. 2014). However, they did not find a similar relationship with other CAs, as observed in our research. One potential explanation for this difference could be that Pawluk and colleagues used different socioeconomic status indicators in their analysis.
Studies from other countries have also noted a higher prevalence of CAs in vulnerable populations. For instance, a case-control study conducted in the United States compared the frequency of CAs in pregnant African American women born in the United States versus those born abroad (Hoyt et al. 2020). The study found that those born in the United States, who typically belonged to lower-income groups, exhibited a higher prevalence of CAs. In a meta-analysis conducted by Yu and colleagues, which examined three parameters of maternal socioeconomic status (education, family income, and maternal occupation), a significant association was found between low socioeconomic status and the occurrence of congenital heart disease (Yu et al. 2014). Additionally, Canfield and colleagues studied variations in specific CAs according to ethnicity in the United States and observed a higher occurrence of severe CAs with potential prenatal diagnosis in African American and foreign white pregnant women (Canfield et al. 2006).
In our study, we observed significantly higher prevalences of all CAs affecting the central nervous system (anencephaly, encephalocele, spina bifida, holoprosencephaly, microcephaly, hydrocephalus, and hydranencephaly) in public hospitals. These severe anomalies, which are associated with high morbidity, mortality and disability, have heterogeneous etiology, so their increased prevalence in the public sector cannot be attributed to a single causal factor. Anencephaly, encephalocele, and spina bifida are neural tube defects, and their occurrence has been associated with folic acid deficiency. In Argentina, the prevalence of these defects has decreased following the mandatory fortification of wheat flour with folic acid implemented in 2002 (Lopez-Camelo et al. 2010; Sargiotto et al. 2015; Bidondo et al. 2015). According to data from the National Nutrition Survey (ENNyS 2007), individuals from lower socioeconomic backgrounds tend to have higher folate levels, possibly due to increased consumption of wheat flour-derived foods. Therefore, one would expect the prevalence of neural tube defects to be lower in this population group; however, our results indicate the opposite.
Holoprosencephaly arises from incomplete division of the forebrain, and in most cases, it has a genetic basis. There is no evidence suggesting that environmental factors associated with socioeconomic status could account for differences in prevalence across different healthcare sectors. Microcephaly, hydrocephalus, and hydranencephaly are heterogeneous neurological anomalies influenced by both genetic and environmental factors. We cannot dismiss the possibility that variations in prevalence could be attributed to environmental factors such as congenital infections, which have been associated with lower socioeconomic status in previous studies (Cannon et al. 2010; Torgerson and Mastroiacovo 2013).
Studies conducted in countries where abortion is legal have observed that fetal anomalies affecting the nervous system are among those most associated with elective abortion (Pryde et al. 1992; Schechtman et al. 2002; Johnson et al. 2012). A global consortium of surveillance programs on CA revealed that during the years 2007–2009, no cases of anencephaly were detected in live births in certain regions such as Cuba, Wales, Tuscany, and the northern Netherlands. However, records indicate an increase in the prevalence of this anomaly in products of elective terminations, indicating that such CAs still occur. In many cases, timely prenatal diagnosis and access to safe elective terminations are available (WHO-ICBDSR 2022).
These background findings suggest that the higher prevalence of CA affecting the central nervous system in public hospitals could be partly explained by a higher proportion of prenatal diagnosis and greater access to termination among the population in the non-public healthcare subsector.
Bilateral renal agenesis and diaphragmatic hernia are both conditions associated with high lethality that also exhibited higher prevalence in public institutions. Regarding diaphragmatic hernia, its etiology is heterogeneous and no environmental risk factors have been identified to date. Although the etiology of agenesis remains unidentified in most cases, pregestational diabetes is known to increase the risk (Davis et al. 2010). The diagnosis of diabetes necessitates actions that require self-care, significant adherence to treatment, and engagement in medical care, which may pose greater challenges for women from socially disadvantaged backgrounds. However, the majority of cases of renal agenesis are not linked to diabetes. Once again, disparities in access to prenatal detection and elective termination of affected pregnancies could partly elucidate the lower prevalence of diaphragmatic hernia and bilateral renal agenesis in non-public institutions.
In the present study, gastroschisis exhibited a prevalence 4.25 times higher in public institutions. The etiology of gastroschisis is still largely unknown, but several studies have identified its strong association with young maternal age (Castilla et al. 2008; Skarsgard et al. 2015; Baldaci et al. 2019). This consistent association suggests that gastroschisis could be caused by the exposure to environmental factors more frequent among adolescent mothers. Our findings likely reflect differences in the distribution of maternal age according to socioeconomic status. Data from Buenos Aires City from the National Health Statistics and Information department (DEIS 2022), revealed that the proportion of pregnant women under 19 years of age was 14.16% in public hospitals, compared to 2.88% in non-public hospitals. Therefore, the observed differences are likely attributable largely to variations in the age distribution of populations rather than to elective terminations.
On the contrary, given that advanced maternal age is the primary risk factor for Down syndrome. It was, one would expect that the prevalence was to be higher in non-public institutions, taking to account that. Official statistics from Buenos Aires City for the period 2011–2021 (DEIS 2022) indicated that the proportion of pregnant women aged 35 and above was 13.27% in public hospitals, compared to 30.77% in non-public hospitals (DEIS 2022). However, the present study revealed a higher prevalence of Down syndrome in public institutions, although the differences were not statistically significant. This result could be attributed to greater access among the population served by non-public hospitals to prenatal diagnosis and elective termination of pregnancy. Previous research has demonstrated that socioeconomic disparities in access to prenatal diagnosis have led to discrepancies in the prevalence of Down syndrome (Khoshnood et al. 2006). In our country, prenatal diagnosis for Down syndrome is routinely available in non-public institutions, whereas it is practically non-existent in the public healthcare subsector (Bidondo et al. 2020). This finding is consistent with other observations from the present study: 47.1% of cases born with Down syndrome in the non-public sector underwent prenatal diagnosis, whereas only 16.9% of those born in public hospitals received prenatal diagnosis.
We observed significant heterogeneity in the PDR of different CA. As anticipated, the PDR was higher for CAs that significantly impact normal fetal morphology and in cases involving multiple anomalies. While we noted a higher PDR for 13 out of 18 CAs in non-public institutions, the only statistically significant difference was observed for Down syndrome. Encephalocele was an exception, exhibiting a higher PDR in public hospitals. However, since fewer than 10 cases were registered in non-public institutions, this difference may be attributable to a bias resulting from the low number of cases rather than a genuine discrepancy.
The variations in PDR across different populations and their potential causes have been the subject of numerous studies investigating various social determinants such as health coverage, socioeconomic status, maternal residence (rural/urban), maternal race, and ethnicity. While some of these determinants may overlap within the same population, socioeconomic status appears to be the variable exerting the greatest impact (Peiris et al. 2009; Hill et al. 2015). For example, a study conducted by Kaur and colleagues explored potential barriers to prenatal diagnosis among pregnant women in the province of Alberta, Canada. Despite the Canadian healthcare system providing universal healthcare coverage, the study found that individuals in the lowest quintile of socioeconomic status had the lowest rates of prenatal diagnosis, and that when prenatal diagnosis was performed, it was often delayed (Kaur et al. 2022).
Pérez et al. investigated the impact of social vulnerability and the timing of prenatal care in a cohort of pregnant women diagnosed with congenital heart disease in fetuses from five hospitals in Boston. They found that pregnant women with higher vulnerability scores tended to undergo their first ultrasound later, and the diagnosis of heart disease was more likely to occur after 24 weeks of gestation. Additionally, lower rates of pregnancy termination were observed in this group. However, the authors noted that when prenatal diagnosis was conducted early in pregnancy, the proportion of elective abortions did not differ based on socioeconomic status (Pérez et al. 2022). This study highlighted that delays in diagnosis significantly limit women's ability to make informed decisions regarding the continuation or termination of their pregnancy.
Regarding trends, our study revealed an increasing prevalence of all CAs with potential prenatal diagnosis in both healthcare subsectors, with statistically significant increases observed in the non-public subsector. This trend could be attributed to greater access to prenatal diagnosis techniques and advancements in this field, resulting in improved detection rates over time.
While the PDR was higher in the non-public sector, the increasing temporal trend observed in the public sector suggests a narrowing of the gap in access to diagnosis between both subsectors. This trend indicates that improvements in access to prenatal diagnosis may be occurring more rapidly in the public sector, potentially reducing disparities in healthcare access over time.
CAs associated with disruptive events, such as limb reduction defects, Moebius syndrome, and amniotic band sequence, exhibited a significantly higher prevalence in public institutions. These anomalies may be linked to attempts to terminate pregnancy in unsafe conditions when the abortion is unsuccessful (Pöhls et al. 2000).
Misoprostol is an analogue of prostaglandin E1 that was initially developed for the prevention and treatment of gastric and duodenal ulcers but later on was repurposed for inducing pregnancy termination (Clark et al. 2007; Shannon and Winikoff 2004). It became widely used as an abortifacient since the late 1980s, even in some Latin American countries where abortion was illegal. The standardization of its use demonstrated safety and led to a reduction in maternal morbidity and mortality by replacing previous, more invasive, unsafe, and ineffective termination methods. This marked progress improved access to termination and minimized health risks. However, its use in the context of illegality had multiple consequences. Firstly, there was a lack of clinical practice guidelines for counseling and monitoring by the health system. Moreover, the criminalization of abortion restricted access to adequate formulations (dosage, combination with other drugs, routes of administration, etc.). Consequently, in conditions where abortion was illegal, the effectiveness of misoprostol as an abortifacient may have been lower than expected, leading to some pregnancies continuing despite its use.
Previous Various studies have demonstrated an association between prenatal exposure to misoprostol and the occurrence of vascular disruptive defects (Castilla and Orioli, 1994; Gonzalez et al. 1998; Pastuszak et al. 1998; Vargas et al. 2000; Dal Pizzol et al. 2008; Barbero et al. 2011; Vauzelle et al. 2013). For instance, in a study conducted by Vargas et al. in Brazil, 93 cases with disruptive anomalies and 279 controls were evaluated. The researchers observed a highly significant association between prenatal exposure to misoprostol and total disruptive CA (odds ratio [OR]: 22.0; 95% confidence interval [CI]: 7.3–81.3). Notably, all cases with exposure to misoprostol corresponded to failed attempts at termination. Furthermore, the study found highly significant associations for specific disruptive CA, including Moebius syndrome (OR: 49; 95% CI: 7.07–1,907) and distal transverse defects of limbs (OR: 24; 95% CI: 3.00–99.1) (Vargas et al. 2000).
Moebius syndrome is a disruptive CA characterized by paralysis of the abducens and facial cranial nerves, often accompanied by involvement of other cranial nerves and additional congenital defects. The etiology of Moebius syndrome is heterogeneous and not yet fully understood; it is postulated to result from abnormal brainstem development, which may be due to intrauterine hypoxia, exposure to teratogens (such as misoprostol), or genetically caused rhombencephalic vascular anomalies. However, the majority of cases occur as sporadic events not associated with a defined genetic cause (Bell et al., 2019). In a case series study of Moebius syndrome conducted in a pediatric hospital in Argentina, it was noted that in 7 out of 30 cases, mothers reported using misoprostol as an abortifacient (López et al. 2015).
Comparatively, the prevalence of Moebius syndrome detected in a study conducted in the Netherlands, where abortion has been legal since the 1980s, was 0.21 per 10,000 births, which is approximately seven times lower than the prevalence observed in our research (1.46 per 10,000) (Verzijl et al. 2003). Notably, all Moebius syndrome cases in our study occurred in public institutions, with no cases reported in non-public institutions. This significant disparity could be attributed to differences in access to safe abortion services, influenced by the socioeconomic status of the population.
Limb reduction defects and amniotic bands also exhibited a higher prevalence in public institutions, which could be associated with similar factors as those mentioned for Moebius syndrome.
Limitations
Our study had some limitations. Socioeconomic status was determined based on the hospital subsector of birth, which may not necessarily be associated with the individual socioeconomic status of the cases. However, a recent study conducted in Argentina, involving approximately 5800 households, found that the population living in poverty (measured by income) primarily sought healthcare in public institutions, unlike the higher-income population, which predominantly sought care in non-public institutions (Paternó Manavella et al. 2022). Secondly, differences in the observed prevalences of prenatally detectable CA may be associated, as previously mentioned, with different etiological risk factors between health subsectors and not necessarily with differential access to prenatal diagnosis and termination. Additionally, populations in the two healthcare subsectors may have differing attitudes towards elective abortion decisions. Finally, differences in disruptive CA may be associated, as mentioned, with other causal factors besides unsafe abortion methods.