Epidemiological studies that deal with congenital malformations are vital as they allow not only for quantification but also for service planning. Moreover, the detection of clusters/outlier values may help to provide clues regarding aetiology/prevention. Specific conditions are mandatory for an epidemiological study dealing with congenital malformations to have validity and therefore utility. The catchment areas must be clearly defined with clearly identified referral routes for patients with suspected conditions, the methodology must be objective and reproducible, and there must be accurate record-keeping with precise registration.
The term “congenital heart disease” refers to structural abnormalities of the heart or great vessels that are actually or potentially of functional significance. Reference Mitchell, Korones and Berendes1 This is an umbrella term that encompasses a wide variety of lesions such that this is are commonest group of congenital malformations and has an overall birth prevalence of almost 10/10,000 live births. Reference Mitchell, Korones and Berendes1
Malta is a small island in the central Mediterranean with a population of approximately half a million. Diagnostic and therapeutic interventions are introduced swiftly in CHD, Reference Grech and Savona-Ventura2 and a declining age in diagnosis and surgery has been shown to decrease mortality. Reference Grech3,Reference Grech and Elliott4
While antenatal screening for congenital anomalies is universal in Malta via an anomaly scan at week 19–24 of pregnancy, Reference Myatt, Gatt, Cordina, Grech and Attard-Montalto5 termination of pregnancy remains illegal and those seeking termination must go abroad. Reference Myatt, Gatt, Cordina, Grech and Attard-Montalto5,Reference Gatt, England, Grech and Calleja6 Postnatally, paediatric echocardiography for diagnostic purposes is readily available with no waiting times at the only regional (state) hospital in Malta where almost all deliveries are carried out. Services are free at point of care (state funded from taxation similar to the United Kingdom National Health Service model).
An earlier study in this country conducted for the birth years 1990–1994 had shown that there was an excess of lesions causing right ventricular outflow tract obstruction, Reference Grech7 especially tetralogy of Fallot. Reference Grech8 Furthermore, despite the relatively small population size, there were no other excesses of conditions associated with CHD. Reference Grech and Gatt9
This study was carried out in order to ascertain whether there are any unusual patterns in CHD prevalence in this somewhat unique population by comparing a longer term of Malta data with equivalent European data using data available from the EUROCAT database (European Surveillance of Congenital Anomalies). This was deemed an appropriate comparison as the definitions of EUROCAT are standardised as far as possible across all participating registries. 10
The Malta Congenital Anomalies Registry is a full participating member of EUROCAT. The Malta Registry has full access to all public hospital paediatric echocardiogram reports thus having a good coverage and ascertainment of CHDs diagnosed in Malta.
Materials and methods
Anonymised data were obtained from the EUROCAT interactive database. EUROCAT is a high-quality network of population-based congenital anomaly registries across Europe for the monitoring, surveillance, and research of congenital anomalies. This network was founded in 1979 and now comprises 43 member registries from 23 countries covering more than 25% of European births. Reference Boyd, Haeusler, Barisic, Loane, Garne and Dolk11 Since EUROCAT prevalence data are anonymised and publicly available, ethics approval was not sought and data protection was not applicable.
Data were available as live births, fetal deaths from 20 weeks gestation and terminations of pregnancy for fetal anomaly following prenatal diagnosis (TOPFA). Data were available by individual cardiac defect lesions for the countries and registries shown in Table 1, and for the conditions shown in Table 2 and data were divided into Maltese and non-Maltese births. The “severe” subgroup is as defined by EUROCAT and includes single ventricle, hypoplastic right heart, hypoplastic left heart, tricuspid valve atresia, Ebstein anomaly, common arterial truncus, double outlet right ventricle, double outlet left ventricle, complete and corrected transposition of the great arteries, atrioventricular septal defect, tetra- and pentalogy of Fallot, pulmonary valve atresia, tricuspid valve stenosis, aortic valve atresia/stenosis, mitral valve atresia/stenosis, coarctation of aorta, aortic atresia/interrupted aortic arch, total and partial anomalous pulmonary venous return, Ivemark atrial isomerism, aortopulmonary window, cor triatriatum, subaortic valve stenosis, supravalvular aortic stenosis, and malformations of the coronary arteries. Reference Dolk, Loane and Garne12 Some of these conditions may present later in life, especially in milder forms (e.g. partial anomalous pulmonary venous drainage), albeit still needing intervention.
Country/registry, years, and births supplied.
N and birth prevalence (per 10,000 births) for CHD lesions obtained from EUROCAT, for Malta and for the rest of EUROCAT (Data taken from EUROCAT as at 26/04/2023).
FD = fetal deaths/still births from 20 weeks gestation; LB = live births; TOPFA = termination of pregnancy for fetal anomaly following prenatal diagnosis; TGA = transposition of the great arteries.
* As defined by EUROCAT. Reference Dolk, Loane and Garne12
Confidence intervals for rates were calculated using the equations of Fleiss. Reference Fleiss13 Chi tests were performed using a bespoke Excel sheet. Reference Grech14 A p value ≤ 0.05 was taken to represent a statistically significant result.
Results
This study analysed data available between 1993 and 2020. There were a total of 22,833,032 births in this study, 121,697 of which were from Malta. Rates (per 10,000 births) and p values for birth prevalence lesions for Malta versus for the rest of EUROCAT are shown in Table 1.
Using EUROCAT data, the overall prevalence of severe CHD for all EUROCAT registries 1993-2020 is reported as 19.62/ 10,000 births (with a wide variation of reported rates ranging from 33.30/10,000 in Wales to 7.95/10,000 in Southern Portugal). The prevalence rate for Malta being reported as 32.38/10,000 births (taken from EUROCAT website as at 26/04/2023).
When comparing Malta rates with all EU Registries average rates, Malta reports a significant excess of commoner, often comparatively non-severe CHDs including atrial septal defect, ventricular septal defect, pulmonary stenosis, patent ductus arteriosus, and aortic stenosis. For the more severe and rarer congenital heart conditions atrioventricular septal defect, Tetraology and Pentalogy of Fallot, Ebstein’s anomaly, total anomalous pulmonary venous drainage, aortic atresia/interruption, and coarctation of aorta were found to be significantly higher.
While the Malta rates compare higher than the overall EU average rate, most fall within the range reported from the different registries (Table1). The anomalies that fall outside the range of reported rates are atrial septal defects, pulmonary valve stenosis, and Ebstein’s anomaly. These are conditions known to be significantly affected by differences in practices and case ascertainment methods.
Discussion
It is known that there are wide variations in reported prevalence of CHD between registries which are explained by variations in the inclusion of the more minor anomalies, e.g. atrial septal defect, small ventricular septal defects, pulmonary stenosis, as well as differences in the access to and waiting time for echocardiography. Reference Dolk, Loane and Garne12,Reference Mamasoula, Addor and Carbonell15,Reference Garne16
Efficient and intense screening increases the ability to pick up some of the smaller defects which may resolve spontaneously soon after such as small atrial septal defects and ventricular septal defects and minor pulmonary stenosis, resulting in increased reported rates of such self-resolving defects.
In Malta access to echocardiography is free for all residents and waiting time is negligible. Furthermore, all public hospital echocardiography reports are available to the Malta Congenital Anomalies Registry enabling a detailed and comprehensive national coverage of all CHD diagnosed on the Islands. Reference Myatt, Gatt, Cordina, Grech and Attard-Montalto5
The higher reported rates of relatively minor conditions such as atrial septal defect, entricular septal defect, and patent ductus arteriosus may be due to quick referral and diagnosis before spontaneous closure. Rapid referral and diagnosis also apply to minor pulmonary valve flow acceleration which resolves over time. Reference Grech3 For the more severe conditions, there may be higher pickup of milder forms such as coarctation of the aorta and Ebstein’s anomaly which would tend to inflate Malta’s numbers. Reference Grech3
Malta is unique when compared to other EUROCAT Registers in that TOPFA is illegal. Reference Myatt, Gatt, Cordina, Grech and Attard-Montalto5 Thus, the apparently higher rate of other, more severe CHD may be partially explained by the fact that in EUROCAT registries where TOPFA is practiced, terminations for antenatally diagnosed syndromes may not have had their CHD diagnosed and these defects would not be registered. It is conceivable that syndromes, including trisomies which are inherently associated with atrioventricular septal defects and 22q11 microdeletions which are associated with truncus and aortic arch anomalies, would not be registered as terminated due to their cardiac lesion but due to the syndrome, and the CHD will not be present in the available EUROCAT congenital heart defects listings. This potential source of bias has been noted in previous studies. Reference Wyldes and Tonks17,Reference Garne, Urhoj and Bakker18
Another potential explanation for the higher reported rates may be the centralisation of services in the only two public hospitals, with deliveries followed by echocardiograms when necessary, typically prior to hospital discharge, as well as the access of the Malta Registry to these results. Reference Grech3
While significantly higher rates were found when comparing Malta rates with EUROCAT average rates, when considering the wide variation of rates between Registries Malta registered highest rates only for the minor conditions of atrial septal defect, pulmonary valve stenosis, and Ebstein’s anomaly (Table 1). Indeed, overall, for severe CHD as defined by EUCOCAT, the prevalence rate for Malta was 32.38/10,000 births, within the extant EUROCAT data range (7.95 in S. Portugal to 33.3/10,000 births in Wales) (taken from EUROCAT website as at 26/04/2023). Maltese rates would naturally fall were TOPFA legal especially since the antenatal diagnosis of congenital malformations is on par with that of other centres. Reference Myatt, Gatt, Cordina, Grech and Attard-Montalto5
An interrogation of the Maltese Paediatric Cardiology database showed that almost all children with univentricular hearts went down the univentricular palliation route to total cavopulmonary connection, except for hypoplastic left heart (no family opted to go down the Norwood route and the patients succumbed after a few days on ductal closure with palliative care) or with very complex isomerism sequences. Reference Grech and Pace19 Those that had surgery all had their interventions in tertiary centres in the United Kingdom, with declining perioperative mortality rates and overall excellent results. Reference Grech and Elliott4 However, this type of surgery carries a high risk for long-term morbidity, mortality, poor clinical outcomes, and eventually poor quality of life when patients reach the stage of “failing Fontan”, Reference Broda, Downing and John20 inevitably incurring significant costs to both Malta’s healthcare system as well as that of the reference surgical centres, with potential need for heart/heart-liver transplantation. Reference Sganga, Hollander and Vaikunth21
Conclusion
It is relatively easy to study rates of conditions in countries with efficient and centralised services and effective record-keeping; however, caution is still required in the interpretation of these rates. In spite of best efforts at harmonisation, wide variation in reported rates of CHD persists between registries especially for the less severe defects, the diagnosis of which is significantly influenced by differences in clinical practices and case ascertainment. Furthermore, non-termination results in higher congenital malformation rates at birth and indeed, a previous study for 1994–2013 had shown that these accounted for 36.7% of the neonatal deaths and the proportionate neonatal mortality attributed to congenital anomalies was the highest reported from Europe. Reference Gatt, England, Grech and Calleja6
Acknowledgements
None.
Financial support
This research received no specific grant from any funding agency, commercial or not-for-profit sectors.
Competing interests
None.
