Incidence, distribution, disease spectrum, and genetic deficits of congenital heart defects in China: implementation of prenatal ultrasound screening identified 18,171 affected fetuses from 2,452,249 pregnancies

Background Congenital heart defects (CHDs) are the most common birth defects. Assessment of the incidence, distribution, disease spectrum, and genetic deficits of fetal CHDs in China is urgently needed. Methods A national echocardiography screening program for fetal CHDs was implemented in 92 prenatal screening–diagnostic centers in China. Findings A total of 18,171 fetal CHD cases were identified from 2,452,249 pregnancies, resulting in 7·4/1,000 as the national incidence rate of fetal CHD. The incidences of fetal CHD in the six geographical regions, the southern, central, eastern, southwestern, northern, and northwestern, were 7·647 (CI: 7·383–7·915), 7·839 (CI: 7·680–8·000), 7·647 (CI: 7·383–7·915), 7·562 (CI: 7·225–7·907), 5·618 (CI: 5·337–5·906), and 4·716 (CI: 4·341–5·108), respectively, per 1,000 pregnancies. Overall, ventricular septal defect was the most common fetal CHD, accounting for 17.04% of screened pregnancies nationwide, and tetralogy of Fallot, the most common anomaly in the major defect of fetal CHD, was the second most common, accounting for 9.72%. A total of 76.24% cases of fetal CHD were found to be an isolated intracardiac single defect. The remaining 23.76% of cases of fetal CHD had multiple heart defects. Among all extracardiac malformations, the central nervous system (CNS) was the most common tissue with extracardiac anomalies associated with CHD, accounting for 22.89% of fetal CHD cases. Chromosomal karyotyping identified trisomy 18 as the most common chromosomal abnormality in fetal CHD. We also documented that CHD-containing syndromes could be identified with a comprehensive approach integrating prenatal ultrasound, MRI, pathological autopsy, and cytogenetics and molecular genetics. Conclusion Implementation of prenatal echocardiography as a practically feasible platform to screen fetal CHD will reduce the financial and emotional burden of CHD, which may facilitate intrauterine and neonatal intervention of CHD. Supplementary Information The online version contains supplementary material available at 10.1186/s13578-023-01172-7.


Background
Congenital heart defects (CHDs), which result from incomplete or abnormal development of the fetal heart during the early stage of pregnancy, are the most common type of birth defect (https://www.cdc.gov/ncbddd/birthdefects/index.html),affecting 8 to 12 per 1,000 liveborn infants and accounting for nearly one-third of all major congenital anomalies worldwide [1,2].In the United States, about 40,000 (1%) births per year are affected with CHD (https://www.cdc.gov/ncbddd/heartdefects/data.html).Several studies of the regional incidence of CHD in China have shown that the incidence of CHDs was similar to that in the global report, [2][3][4][5] except that the incidence could be as high as 38.1 per 1,000 live births in the ethnic minority of "Uygur" at Xinjiang Uygur Autonomous Region (AR), which is geographically adjacent to Russia [6].CHDs may result from genetic deficits; however, unexplained CHDs could occur secondarily to noncoding genetic, epigenetic, and environmental factors, among others [7][8][9].Such environmental factors include abnormal development and dysfunction of the placenta in pathological pregnancies of intrauterine growth retardation, preterm birth, stillbirth, and preeclampsia [10][11][12][13][14]. Application of transabdominal echocardiography ultrasonographic screening to evaluate for CHDs in clinical practice has increased the detection rate of CHDs [15,16].Because of its ability to detect CHDs non-invasively in the early stage of pregnancy, prenatal ultrasound screening has been implemented, along with prenatal hemodynamic screening, as a routine clinical procedure in prenatal healthcare management worldwide [17].
Prenatal screening for CHDs, along with birth defects, has been implemented in China, along with the implementation of a surveillance system for public healthcare and prenatal healthcare [18].Thousands of prenatal ultrasound specialists have been trained in the nationally credentialed training centers and have been certified for qualification to conduct prenatal ultrasound screening, which ensures that a standardized procedure is applied in the multicenter practices.However, detailed information about the true incidence and disease spectrum of CHDs throughout China is lacking.Therefore, most reports of the incidence/ prevalence and disease spectrum of CHDs in China identified from newborns before 2015 ignored fetuses with CHD that were aborted.In this first Chinese national study of the incidence and disease spectrum of CHDs, 18,171 fetal CHDs were identified and documented via fetal echocardiography through prenatal ultrasound screening for CHDs from among 2,452,249 pregnancies.

Research design and ethics issue
This project was a multi-institutional clinical research investigation by the Chinese Consortium for Prenatal Ultrasound Screening of Congenital Heart Defects, with fetal echocardiography to determine the national incidence, distribution, spectrum, and genetic defects of fetal CHD.The procedures and protocols implemented were reviewed and approved by the Ethics Committees of Maternal and Child Health Hospital of Hubei Province.The research studies were coordinated by the Department of Diagnostic Ultrasonography at the Maternal and Child Health Hospital of Hubei Province.Informed consent was obtained from each pregnant participant.Ultrasound data between the period of January 1, 2011 and December 31, 2013 were collected as part of clinical procedures for prenatal healthcare management.

Implementation of prenatal screening of CHD with standardized procedures
As part of the ongoing national mandatory program to conduct nationwide surveillance for prevention of structural CHDs, the Ministry of Health (MOH)-credentialed National Prenatal Ultrasound Diagnostic Training Center (NPUDTC) initiated a training program in 2006 (Figure S1).A standardized operating procedure (SOP) for fetal echocardiography, following the guidelines of the International Society for Obstetrics and Gynecology (ISUOG) (http://www.ISUOG.org)and the American Institute of Ultrasound in Medicine (AIUM) Clinical Standards Committee, [19,20] was provided to train prenatal ultrasound specialists.These trained specialists were capable of performing fetal ultrasound scan at the mid-trimester-optimally at between 18 and 22 weeks' gestational age-with ultrasound skills for the four chambers, outflow tract including left ventricular outflow tract and right ventricular outflow tract, and three-vessel trachea views to capture and to recognize all types of CHD.Prenatal ultrasound specialists nationwide who graduated from and were certified by the NPUDTC for prenatal ultrasound screening of fetal CHD participated in this clinical research project among 93 prenatal screeningdiagnostic centers (Table S1).

Prenatal screening for CHD
Instruments used for prenatal ultrasound screening in this study included Siemens Acuson Sequoia 512 and S2000 (Siemens Medical Solutions Inc., Mountain View, CA, United States), Voluson 730 Expert and Voluson E8 (GE Healthcare, Kretz Ultrasound, Zipf, Austria), Philips IU 22 (Philips Medical Systems, Bothell, WA, United States), and Samsung UGEO WS80 (Samsung Medison Co., Ltd., Korea).In each ultrasound screening center, at least one advanced ultrasound system was employed.The final imaging was reviewed and confirmed by two ultrasound experts.The raw data were then transferred from the ultrasound unit directly into an electronic database to avoid generation of any errors via manual data entry.Random audits were performed to verify data reliability and accuracy during data transfer.

Follow-up and confirmation
To verify the accuracy of prenatal ultrasound screening, positive cases determined by echocardiography were verified by postnatal follow-up with (1) clinical visits, (2) ultrasound and/or magnetic resonance imaging (MRI), (3) surgical observation, (4) pathological autopsy of aborted fetus, and (5) genetic studies.

Data collection
Demographic information about pregnancies was collected from hospital administrative records.Software with multiple drop-down menus for the purpose of data collection from multi-centers was developed at the NPUDTC.Interpretations of fetal heart malformations followed the ISUOG and AIUM Clinical Standards Committee guidelines and were coded on the basis of the 10th revision of the International Statistical Classification of Diseases and Related Health Problems (ICD10) [17,[19][20][21].CHD cases were grouped as CHDs with intracardiac malformation and CHDs with extra-cardiac malformation.The intra-cardiac malformations were further subgrouped into single abnormalities and multi-abnormalities.For multi-abnormalities, if more than one cardiac malformation was observed, the major anomaly was counted into the subgroup.

Statistical analysis
All statistical analyses were performed using SPSS 13.0 software (SPSS Inc, Chicago, IL, USA) and Arcgis10.2.An unpaired t-test was used to compare incidence parameters among northern, central, eastern, western, and southern China.Categorical data were compared by using Pearson chi-square analysis.The relationship between incidence parameter and gross domestic product was assessed by simple linear regression analysis.p value of less than 0.05 was considered significant.

Incidence of fetal CHD
A total of 18,171 fetuses affected with CHDs were identified from 2,452,249 pregnancies (Table S2) among 92 (67 tertiary, 22 secondary, and three primary) hospitals across 31 provinces/municipalities/ ARs (Table 1a).Collection of reliable data was not successful in the Tibet AR (labeled as Xizang in Fig. 1), due to incomplete prenatal screening.The overall nationwide incidence of fetal CHD, with the lack of the Tibetanese population, was 74•099/10,000 (18, ) per 10,000 pregnancies.The incidences of fetal CHD in the ocean coast provinces, including the regions in the south and east, and the province of Shandong but not the province of Fujian or Zhejiang, were relatively higher than those in inland areas (Fig. 1).In the non-ocean coast region, the incidence of CHD was below 9•0 per 1,000 pregnancies in all provinces except Shanxi, which has the highest rate of neuronal tube defects (NTDs) in China and in the world.[22].

Distribution of CHD
Subtypes of CHD in the major and minor anomalies and their presence in tertiary hospitals vs. in primary and secondary hospitals were documented (Table 1a).Overall, ventricular septal defect (VSD) was the most common fetal CHD, accounting for 17•04% of pregnancies screened nationwide.Tetralogy of Fallot (TOF), the most common anomaly in the major defect of fetal CHD, was the second most common after VSD, accounting for 9•72% of pregnancies.Following VSD and TOF, the incidences of CHD were atrioventricular septal defect (AVSD) (7   persistent truncus arteriosus (PTA) (2•71%).These CHDs comprised the top-10 most common fetal CHDs (Fig. 2).

Spectrum of all anomalies of CHD
The spectrum of CHD cases in China was determined to comprise 15 major and 21 minor anomalies (Table 1b).Among the 18,171 fetal CHD cases, 14,096 (76•24%) were isolated single defects within the heart, the top 5 most common of which were VSD (29•09%), TOF (14

Genetic study of chromosomal abnormalities with karyotyping and microarray
Genetic study for chromosomal abnormality was performed among 566 cases of fetal CHD, 487 of which were analyzed by karyotyping, and 79 by microarray to determine chromosomal copy number variations (CNVs) including DNA fragment deletion and/or repetition/ duplication or uniparental disomy (UPD).As presented in Fig. 4; Table 3 TOF was the most common CHD among the aborted fetuses, followed in descending order by AVSD, DORV, VSD, TGA, SV, PTA, HLHS, COA/IAA, and HRHS, as shown in Fig. 5.

Postnatal follow-up to verify the accuracy of prenatal ultrasound screening results
Prenatal ultrasound is a screening, rather than a diagnostic, procedure.In this study, postnatal follow-up was carried out for 20% (3,619) of 18,171 fetal CHD cases, including 1,965 cases by the postnatal diagnostic procedure(s) of clinical visits, imaging (chest X-ray, ultrasound, and/or MRI), and neonatal surgery; and 1,654 cases by pathological autopsy of aborted fetuses (Table 5).Our follow-up results showed 4•70% of cases with 100% agreement between prenatal screening results and postnatal verification, for the CHD subtypes of AA, CAF, dextrocardia, DOLV, ductus

Incidence and distribution
Applying prenatal ultrasound screening, we determined that the incidence of CHDs in China during 2011-2013 was 7•4099 per 1,000 (7•4099%) pregnancies.To our knowledge, this is the largest single cohort in the world, with 2,452,249 pregnancies reported so far.The incidence of Chinese fetal CHD cases in the current report is similar to that reported in Western countries (Table S3) [23][24][25][26].The incidence is higher in advanced, developed regions, such as in the southeastern provinces of Shandong, Jiangsu, and Guangdong, where CHD incidence is ≥ 9% (Fig. 1), and in provinces with a high rate of birth defects, such as 9•8% in Shanxi province [22].CHD incidence is much lowerbelow 6%-in the western provinces of Yunnan, Qinghai, Gansu, and Xinjian (Fig. 1).However, the higher incidence in coastal areas vs. the low rate in western developing regions appears not to be correlated with economic development (Figure S2).

Spectrum of CHDs: single vs. complex and intracardiac vs. extracardiac
A total of 36 subtypes of fetal CHD, including 15 major and 21 minor abnormalities, have been reported in this study.To our knowledge, this is the most complete collection of fetal CHD cases among Chinese populations.We determined that the most common single intracardiac anomaly was VSD, followed by TOF, AVSD, PLSVC, and TGA.Among the intracardiac multi-anomalies, DORV was the most common, followed by SV, COA/ IAA, TOF, and AVSD.We also documented that the CNS was found to be the most commonly involved extracardiac tissue, accounting for 20•89% of extracardiac defects, followed by the urogenital (15•21%), craniofacial (13•52%), digestive (8•60%), and skeletal (7•99%) systems.Fetal appendage malformations (21•04%), fetal hydrops (8•29%), and fetal abdominal-wall defects (4•45%) were also found to occur commonly with intracardiac anomaly.

Genetics and genomics: Cytogenetics vs. molecular genetics
It has long been known that CHDs may result from chromosomal abnormalities.CHD in Down syndrome (DS, trisomy 21) and in DiGeorge syndrome, characterized by microdeletion of the DiGeorge critical region at Table 2 Extracardiac defects associated with congenital heart defects 22q11.21,represents two traditional conditions [7,8].DS is by far the most common and the best-known disorder resulting from chromosomal aneuploidy and the most common cause of intellectual disability.CHD is the leading cause of mortality and morbidity during the first two years of life in the DS population [27].DiGeorge syndrome is caused by the deletion of a small segment of chromosome 22 [28].Advanced genetic studies have identified about 400 genes implicated in CHDs, encompassing transcription factors, cell signaling molecules, and structural proteins that are important for heart development [29].In our study, we initiated chromosomal karyotyping to determine the aneuploidy, followed by microarray assay for CNVs (Tables 3 and 4), to identify chromosomal abnormalities among the fetal CHD cases identified through prenatal echocardiography.In this study, four cases of TOF showed micro-deletion at 22q11.21, which is characteristic of DiGeorge syndrome.Micro-deletion at this DiGeorge's locus was also the most common abnormality in 53•84% of VSD cases.
Genome-wide association study (GWS) was also performed, as presented in Fig. 6.One case was CHARGE syndrome, which was originally screened as AVSD with prenatal fetal echocardiography, diagnosed with MRI, confirmed by autopsy, and studied with GWS, which identified a c.6482del at gene CHD7 that produced a mutant peptide.The second example was a point mutation, c.309 C > A, at gene LZTR1, identified from a case of fetal TOF and determined to be Noonan syndrome.

Clinical application and data reliability
The overall 88% agreement in this study between prenatal ultrasound results and the follow-up confirmation (Table 5) demonstrates the feasibility and capability of performing a nation-wide study with a standardized procedure.Our experience is that system-wide training in general prenatal ultrasound skills and knowledge with on-site initiation of prenatal ultrasound screening to recognize fetal cardiac anomalies is critical for success.Whether the affected fetus is born or is terminated, postnatal follow-up may determine the accuracy of screening outcome, especially at early stages.This follow-up may help in implementation of prenatal screening for CHD in under-developed regions.In developed areas, we encourage clinicians to consider the comprehensive approach to confirmation of results from prenatal screening for CHD, integrating prenatal ultrasound, MRI, pathological autopsy, postnatal cardiac ultrasound or imaging, and genetic studies (Fig. 6), to better understand the molecular pathogenesis of CHD.With this approach, we will be able to explore and implement early intervention to prevent fetal CHD and to reduce the global burden of treatment for CHDs.

Fig. 1
Fig.1Incidence of fetal CHD identified by prenatal echocardiography, and distribution across China.Six regions, the north, central, east, south, northwest, and southwest, presented with dark blue, violet, yellow, red, light green, and dark green, respectively.Data from prenatal ultrasound screening for fetal CHD cases that were collected from Tibet were not available during quality control.No prenatal ultrasound screening data were collected from Taiwan

Fig. 6
Fig.6 Comprehensive study of CHD.Prenatally identified CHD by fetal echocardiography (a1, b1, c1) was confirmed with MRI (b2, c2), pathological autopsy (a3, b3, c3), and genetic studies of microarray that showed microdeletion (a2) or genome wide sequencing (not shown).Three cases are presented to demonstrate the prenatally fetal-echocardiography-screening were verified and confirmed with MRI, autopsy, and genetic/genomic approach.Case 1 (a): DiGeorge syndrome was observed with DORV by prenatal ultrasound (a1) and verified in pathological autopsy (a3).Genetic study with a microarray (a2) determined a microdeletion of 2,825,77Kb at chromosome 22q11.21between 22.18920346-21746118, which fell into the DiGeorge critical region (DGCR) that embeds 69 genes.Case 2 (b); CHARGE syndrome was observed with AVSD by prenatal ultrasound (b1) and verified in pathological autopsy (a3).Genetic study with a microarray (a2), confirmed by a 7T MRI (as pointed by arrow in b2), verified by a single nucleotide deletion c.6482del at gene CHD7 (NM_017780.3), which resulted in a frameshift mutation p.His2161Leufs54 and produced a mutant C-terminal peptide of CHD7 protein.Case 3 (c): Noonan syndrome (type X) was identified with a TOF presented by VSD, PA, and ARCH in the echocardiography (c1), which was confirmed by a 7T MRI (c2) and pathological autopsy (c3).GWS determined a single nucleotide mutation c.309C>A at gene LZTR1 (NM_006767.3),resulting in a nonsense mutation of p.Cyc103* that truncated C-terminus of LZTR1 protein

Table 1a
National incidence of variant subtypes of congenital heart defects in China during 2011-2013 (CI 95%)

Table 1b
Spectrum and proportion of congenital heart defect subtypes

Table 4
Genetic study of congenital heart defects: Copy number variations of CHD