Assisted reproductive technology (ART) has been used for more than 40 years. Since its introduction, it is estimated that 0.2%–6.1% of children in European countries have been conceived with the help of these techniques.1 The potential association between the use of ART and the development of congenital malformations (CMs) has been the object of extensive study. The prevalence of CMs is estimated at 2%–3% in the general population and has remained stable in recent years (https://eu-rd-platform.jrc.ec.europa.eu/eurocat_en2). Although most authors report higher proportions of CM in individuals born after ART,3–7 the evidence is still inconclusive, as there are also studies in which these differences disappear after adjusting outcomes for factors such as maternal age or ART-related variables.8,9 Still, the most recently published systematic review and meta-analysis10 did find an increased risk of CM in singleton births after ART (relative risk [RR], 1.41; 95% confidence interval [CI], 1.30–1.52), although this increased risk was not found in individuals fruit of multiple gestations (RR, 1.18; 95% CI, 0.98–1.42).

There are fewer studies on the subject of minor anomalies. Still, the use of ART has been described as a risk factor for infantile haemangioma,11 and great proportions of children of subfertile couples that used ART have been found to have minor anomalies considered clinically insignificant based on the classification of Merks et al.12 Between 50% and 54% presented this type of anomalies,13 most frequently infantile haemangioma (2%–5%), inguinal or umbilical hernia (1%–3%), hypospadias (1%–2%) or cryptorchidism (1%–2%).

Many authors have also attempted to determine whether outcomes differ between in vitro fertilization (IVF) and intracytoplasmic sperm injection (ICSI) and other ART-specific factors, but results have been inconclusive. Some authors have reported a higher incidence of CMs in association with ICSI14–16 and others a higher risk associated with IVF,17 while more recent studies have found no differences.18,19 There is also no clear evidence that the risk of CMs increases with frozen embryo transfer (FET),20 and further research is still needed.

In light of the above, the primary objective of our study was to assess whether there was an increased risk of major and minor CMs in a cohort of Spanish children aged up to 3 years born after ART, independently of maternal age, singleton/multiple order pregnancy, preterm birth and other possible confounders.

We also analysed ART-related factors that could be at play in the detected malformations.

We designed a prospective, observational, controlled cohort study. The study was approved by the Ethics Committee of the Hospital Clínico Universitario Virgen de la Arrixaca (HCUVA). Participation in the study required signed informed consent, which was procured in adherence with the principles of the Declaration of Helsinki.

Selection of cohort of clinical pregnancies achieved through ART

First, we selected a cohort of clinical pregnancies (evidence of a heartbeat on week 7–8 of gestation) in women that underwent IVF, ICSI or preimplantation genetic diagnosis (PGD) between May 2012 and May 2014 in the fertility unit of the HCUVA, a public tertiary care hospital that has partnership agreements with 5 private clinics. In the period under study, there were 243 clinical pregnancies, 26 miscarriages and 3 cases of intrauterine foetal demise. Fig. 1 presents the cohort selection process.

Figure 1.

Selection of participants.

(0.44MB).

In this phase, we collected the following data:

• -

  Epidemiological characteristics: country of origin, age at initiation of treatment and personal and family history of the parents, maternal body mass index, duration and causes of infertility.

• -

  Treatment-related variables: type of ART, type of ovarian stimulation and suppression used, complications, number of retrieved eggs and fertilized eggs, embryo quality, day of embryo transfer, cryopreservation, etc.

Clinical evaluation and follow-up of cases and controls

In this phase, we carried out the clinical evaluations: one at age 12–20 months and the other at age 26–40 months, all between September 2014 and June 2017.

In these evaluations, we collected data on the pregnancy, delivery and neonatal period (Fig. 2) and carried out a comprehensive physical examination, including a thorough morphological assessment. We classified major anomalies according to the International Classification of Diseases, 10th revision (ICD-10, https://www.mscbs.gob.es/estadEstudios/estadisticas/normalizacion/CIE10/Clasif_Inter_Enfer_CIE_10_rev_3_ed.diag.pdf21) and minor anomalies based on the classification proposed by Merks et al.12

Figure 2.

Collection of prenatal data.

(0.6MB).

Many studies on the health problems found in individuals born after ART have differentiated between singleton and multiple pregnancies and stratified risk based on maternal age, but to our knowledge, this is the first one to carry out a prospective follow-up in a group of children born after ART and a group of matched controls comparable in terms of the frequency of multiple pregnancy, preterm birth and maternal age, in which we also achieved a high participation rate (approximately 90%). The only relevant factor that we could not avoid to be able to essentially attribute outcomes to the use of ART was subfertility (failure to conceive after 1 year of unprotected regular sexual intercourse), which was only identified in 7.8% of couples with NC children. However, since ART is associated with difficulty conceiving (except, generally, cases of PGD), and this is the only option available for many couples to be able to conceive, we think that for the purpose of counselling, it would be more appropriate to consider the risks associated with subfertility and with ART as a whole.

Chief among the findings of the study is that, despite the higher incidence of major CMs in the ART group (7.8%), the comparison with the control group did not evince significant differences, contrary to the conclusions of the most recent meta-analysis on the subject.10 The high incidence of CMs in both groups was probably due to the characteristics of the populations under study (frequencies of multiple gestations and preterm birth of 24%–25%) and the study design: prospective study with follow-up through age 3 years, which allowed the detection of malformations not diagnosed at birth. This suggests that the criteria used to select the control group should be contemplated as potential key factors associated with the morbidity described in the literature focused on this population, as opposed to ART itself. Couples must receive adequate information, and age should be taken into account in the interpretation of success rates, in addition to its potential effects on health during gestation and the future health of the offspring. Our findings also support the growing practice of single-embryo transfer24 to reduce the frequency of MG and the associated frequency of preterm birth, aspects that couples must also be informed about.

We did find an increased risk of certain minor anomalies, such as PLs, MCALS and capillary malformations in the ART group, an association that had not been described in the previous literature. As we already noted, ART had been identified as a risk factor for infantile haemangioma in the past, but in our study, while we found a greater incidence of this anomaly in the ART group (4.76% vs 3.4%), the difference was not statistically significant after adjusting for sex, preterm birth, multiple gestation and other possible confounders. The identified association between ART and minor cutaneous abnormalities, while not clinically relevant per se, is of great interest, as the presence of PLs, particularly MCALS (found in 10% of cases in our series), has been found to be associated to other developmental disorders, epilepsy25 and some forms of cancer, such as acute lymphoblastic leukaemia, in the previous literature.26 We ought to note, however, that the association with cancer is currently under debate. A recent study conducted in Brazil did not find a higher prevalence of MCALS in patients with solid tumours.27 Thus, further research is required to confirm these associations, and long-term follow-up of this and other cohorts is necessary to establish potential long-term risks.

We also found a recurrent craniofacial pattern that had not been described in the past and associated with abnormal neurodevelopment. If we considered both disorders to be manifestations of deviations from normal development in the early stages of embryogenesis, as proposed by Seggers et al.,28 these findings suggest the possibility of mild changes in embryonic development associated with the use of ART that could give rise to minor anomalies at the facial level and more serious problems in a small percentage of cases. The association with male sex and younger maternal age is also interesting, as it suggests a potential role of the hormonal milieu. Assuming that younger women respond more strongly and have higher levels of estradiol, and given the higher frequency of the identified craniofacial pattern in male offspring, oestrogen levels may play a role in its development.29

It is essential that clinicians, especially those involved in the evaluation of patients with multiple CMs or neurodevelopmental disorders, be aware of this phenotype, as craniofacial features are often the starting point that eventually leads to diagnosis. At times, they may mask or hinder diagnosis or lead to performance of additional diagnostic tests.

Our study found a high prevalence of major CMs in both groups, without statistically significant differences between them. The incidence of minor anomalies was similar to the incidence described in individuals born after ART and the subfertile population, and we also did not find significant differences between the ART or NC groups for these anomalies.

This study is the first to describe an increased risk of PLs, MCALS and capillary malformation in individuals born after ART, independently of the order of the pregnancy and sex. We found an association between PLs and PGD, despite the small number of children conceived after PGD. We also unexpectedly identified a recurrent craniofacial pattern in children born after ART that had not been previously described. Further research is required to confirm these associations.

None of the ART-specific factors under study were associated with these anomalies, but we found an association with multiple gestation, which reinforces the important role of this factor in abnormal embryonic development. Couples need to be informed of these outcomes before initiation of treatment, emphasising the need to transfer only one embryo whenever possible.

The authors have no conflicts of interest to declare.