Introduction

China has a large population and ranks among the countries with a high prevalence of birth defects. In China, there are significant disparities among different geographical regions, with the prevalence of birth defects ranging from 0.715 to 19.184%1,2,3. In addition to traditional chromosomal aneuploidy abnormalities, chromosome microdeletion or microduplication syndromes have emerged as a significant category of birth defects. The diagnostic rate of chromosomal microarray analysis (CMA) for chromosomal abnormalities is higher than that of traditional G-banding karyotype analysis. The prenatal diagnosis of microdeletion and microduplication syndrome through CMA has significant implications for the prevention of birth defects.

The indications for invasive prenatal diagnosis mainly include abnormal ultrasound findings, high-risk maternal serum screening (MSS), abnormal non-invasive prenatal testing (NIPT) results, advanced maternal age (AMA), intrauterine growth restriction, history of adverse pregnancy outcomes, maternal request, in vitro fertilization, drug use or exposure to toxic substances during pregnancy, consanguineous marriage, and parental anxiety. Since 2009, the United States, Europe, Canada, and other developed countries have recommended CMA as the first-tier diagnostic test for fetuses, both for patients with fetal ultrasound anomalies and for patients who choose to undergo prenatal testing despite normal ultrasound of the fetus4,5,6. To date, many studies evaluating the performance of CMA in prenatal diagnosis have been published7,8,9. The incidence of pathogenic CNVs in fetuses with abnormal ultrasound results can be further classified based on the specific organ system involved and the number of observed abnormalities. The most frequently affected organ systems related to abnormal CMA results are the cardiovascular system, skeletal system, genitourinary system and central nervous system10,11,19,20,21, and the findings of our study (3.78%) fell within this range. In the AMA-only group, the abnormal MSS results group, and the abnormal NIPT results group, the detection rates of CNVs (P/LP) by CMA were 2.35%, 2.52%, and 8.55%, respectively. These results were consistent with previously reported detection rates in the literature ranging from 0.84% to 5.8%, 1.6% to 8.0%, and 0.76% to 35.3%, respectively16,19,20,21,22. However, in the group with abnormal ultrasound results, the detection rate of CNVs (P/LP) by CMA was 3.55%, which was lower than that reported in previous studies (4.5–7.9%)16,19,21, particularly within the subgroup of single-system structural abnormalities (3.35% vs. 3.66–10.9%)23. The presence of clinically significant CNVs may be observed when prenatal ultrasound indicates thickened nuchal translucency or enhanced renal echo. Fetuses with ultrasound soft marker abnormalities exhibited varying incidence rates of CNVs (P/LP). Our research revealed a small but not statistically significant increase in the likelihood of clinically relevant CNVs in fetuses with one or more ultrasound soft markers (3.1% vs. 4.29%, p = 0.853), consistent with previous investigations24. These findings suggested that CMA should be performed in pregnant women when two or more ultrasound soft markers are detected by ultrasound.

Among the subgroup of fetuses with ultrasound structural abnormalities, the detection rate of CNVs (P/LP) was the highest in the group of fetuses with structural anomalies combined with soft marker abnormalities (8.2%), followed by those with multiple system structural abnormalities (7.41%). The detection rate of CNVs (P/LP) in fetuses with a single-system malformation is similar to that previously reported (3.1–7.9%)25, with skeletal (9.52%), genitourinary (2.2%), central nervous (9.09%) and cardiovascular system abnormalities (3.43%) being the most commonly associated with CNVs (P/LP). These results showed that CMA should be recommended when fetal ultrasound reveals multiple structural abnormalities, especially in skeletal, genitourinary, central nervous, and cardiovascular system abnormalities. In this study, Two fetuses with Emanuel syndrome were identified, and CMA results showed 11q23.3q25 region duplication and 22q11.1q11.21 region duplication. Prenatal ultrasound revealed a single umbilical cord and underdeveloped cerebellar vermis in one fetus, while the other exhibited a posterior fossa anomaly. These are the most common defects of ES and might be diagnosed in early pregnancy26. The clinical phenotype of the 22q11.2 proximal deletion exhibits significant heterogeneity, primarily including heart defects, palatal abnormalities, developmental retardation, and immunodeficiency. In this study, Two fetuses with a 22q11.2 proximal deletion had cardiac defects: one exhibited right foot varus and left ventricular punctate strong echoes, and the other presented with a right aortic arch. 22q11.2 deletion is frequently observed in fetuses with heart defects diagnosed prenatally27. Therefore, CMA can be used as a detection method for prenatal echocardiography abnormalities.

The phenotype resulting from a susceptibility CNV is unpredictable due to incomplete penetrance and variable expressivity28,29. These features have not been systematically described although the clinical phenotypic features associated with postnatal recurrent microdeletion/duplication syndromes are well-defined, these features have not been systematically described in prenatal cases due to the limitations of prenatal identification. In our study, the most common recurrent microdeletion/duplication syndromes were 1q21.1 deletion syndrome (10 fetuses) and 15q11.2 deletion syndrome (10 fetuses). 22q11.2 duplication syndrome (9 fetuses), 16p13.11 duplication syndrome (9 fetuses), proximal 16p12.2 microdeletion syndrome (7 fetuses), 16p11.2 microdeletion syndrome (4 fetuses), and 16p11.2 duplication syndrome (3 fetuses), 16p13.11 deletion syndrome (2 fetuses) were also common in this study. Among these fetuses, only 7 (7/54, 12.2%) exhibited structural abnormalities and all pregnancies were terminated. Moreover, we observed that the proportion of normal fetuses after birth is greater in recurrent microdeletion/duplication syndromes with penetrance of less than 10%. For example, approximately 90% of fetuses with 15q11.2 deletion syndrome and 16p13.11 duplication syndrome have a normal outcome. Therefore, we recommend not informing couples about CNVs classified as loci with penetrance of less than 10%. These potential neurodevelopmental sites may have some structural abnormalities and may not have obvious clinical indications during pregnancy, which will increase parents' anxiety about fetal development. The precise ultimate phenotype remains unknown, posing significant challenges for genetic counseling.

The single nucleotide polymorphism array (SNP array) technology of CMA can detect not only CNVs, but also ROH, uniparental disomy (UPD), and low-level mosaics30.As chromosomes 6,7,11,14,15 and 20 are known to be associated with parental-specific expression genes, further testing is necessary to clarify the diagnosis and distinguish between ROH and UPD. In clinical practice, child-parent trio analysis through CMA is often required31. This study reported two cases of fragmented ROH involving the long arm of chromosome 15. Due to AMA and the absence of future childbearing plans, both couples abandoned further diagnostic tests and terminated their pregnancies, so we could not determine the pathogenicity of fragmented ROH involving the long arm of chromosome 15. A case of fragmented ROH on the short arm of chromosome 5 was reported. The karyotype of this fetus was 46, XY, del(5)(p13)[32]/46, XY[70], which led to mosaic loss in the critical region causing Cri du chat syndrome. Consequently, the couple decided to induce labor. When a critical region causing microdeletion syndrome exhibits ROH, it is imperative to complement this region with karyoty** analysis to ascertain its pathogenicity. The presence of multiple large regions of ROH on various chromosomes may indicate a potential consanguineous relationship between the tested individual's parents, either close or distant32. In this study, one fetus was found to have multiple large regions of ROH on various chromosomes due to consanguineous marriage between the parents. The couple ultimately decided to terminate the pregnancy. The presence of ROH on the whole of chromosome 4 was identified in our study. In the advanced stages of gestation, intrauterine growth restriction led to fetal demise within the uterus. This finding was consistent with previous studies that fetuses with ROH frequently exhibited the most prevalent prenatal manifestation of intrauterine growth restriction33. For 6 fetuses involving other chromosome fragmentary ROH regions, all six couples opted to continue their pregnancies, and subsequent assessments confirmed normal fetal development postpartum. Due to the complex pathogenesis of ROH, encompassing gene imprinting effects, homozygous recessive gene mutations, and low-level chromosomal mosaics, a comprehensive assessment of prognosis should be conducted by combining ultrasound findings, family verification, whole exome sequencing, and other relevant factors.

The pregnancy outcomes of all women were evaluated in this study. Although the majority of women with fetuses diagnosed with aneuploidy or CNV(P/LP) opted for pregnancy termination, 70 women whose fetuses with aneuploidy or CNVs (P/LP) decided to continue their pregnancy, and subsequent follow-up revealed that 65 of the newborns exhibited normal phenotypes. Notably, among these patients, 50% had fetuses with sex chromosome aneuploidy. The results of our investigation indicated that an increasing number of people are accepting of children with sex chromosome aneuploidy. In this study, 76 fetuses with CNVs (VUS) were found, for a detection rate of 1.85%. After follow-up, a total of 45 fetuses were delivered, and among these fetuses, 44 exhibited normal postnatal outcomes. This finding indicates that fetuses with CNVs (VUS) are most likely to have a good pregnancy outcome. Furthermore, reporting VUS in prenatal diagnosis may present challenges for genetic counseling, place pressure on pregnant women and their families, and lead to excessive termination of pregnancy. Therefore, in subsequent stages, further case accumulation and long-term follow-up are needed to comprehensively evaluate the prognosis. This study has significant implications for the development of future genetic counseling guidelines.

Due to the limited availability of detailed genotype–phenotype information for some fetuses, this retrospective study is limited in terms of data acquisition. For instance, some pregnant women have poor compliance or refuse to undergo testing due to financial constraints, while others may experience information loss during the referral process, resulting in the potential omission of high-risk indications for certain fetuses and ultimately introducing bias into the study findings. Furthermore, incomplete clinical examinations in fetuses with a young gestational age may lead to insufficient descriptions of certain clinical phenotypes. Ultimately, the absence of parental data presents a significant challenge in assessing the pathogenicity of certain CNVs, thereby complicating our clinical genetic counseling efforts.

The CMA test is applicable for all prenatal clinical indications because it enhances the detection of clinically significant CNVs. Clinicians should duly apprise pregnant women about the potential risk of undetectable CNVs through biochemical screening or most existing NIPT platforms, and emphasize the value of CMA in prenatal diagnosis for informed decision-making.

Materials and methods

Patients and clinical indications

All patients who underwent invasive prenatal diagnosis by CMA at Linyi Women and Children's Hospital between 2016 and 2022 in Shandong, China. Clinical samples (chorionic villi and amniotic fluid) were obtained by ultrasound-guided abdominal chorionic villus sampling (CVS) and amniocentesis. After receiving detailed genetic counseling before testing, each participant signed a written informed consent form. This study was approved by the Ethics Committee of Linyi Maternal and Child Hospital (No. KYL-YXLL-2022017). The research was conducted in accordance with the relevant guidelines and clinical norms.

CMA analysis

Fetal uncultured genomic DNA was extracted by a DNA extraction kit (TIANamp Micro DNA Kit, China). The whole genome CytoScan 750 K array (Thermo Fisher Scientific, USA) was used for CMA according to the manufacturer's instructions. The original data were analyzed by chromosome analysis software 4.2 (Thermo Fisher Scientific, USA) of genome version GRCh37/hg19. The results were classified according to genetic mode (familial or de novo), CNV length, genes involved, their classification, and literature information. The online databases consulted included PubMed (http://www.ncbi.nlm.nih.gov/PubMed), Decipher (https://www.deciphergenomics.org/), UCSC (https://genome.ucsc.edu/index.html), ClinGen (https://search.clinicalgenome.org/), Clinvar (https://www.ncbi.nlm.nih.gov/clinvar/), HGMD (https://www.hgmd.cf.ac.uk/ac/index.php) and DGV (http://dgv.tcag.ca/dgv/app/home). The results were classified as pathogenic (P), likely pathogenic (LP), variant of uncertain significance (VUS) and normal (including likely benign and benign) according to the 2019 ACMG guidelines. Benign (common polymorphism) and/or likely benign CNVs were detected, and the reported results were normal. Regions of homozygosity (ROH) involving chr 6, chr 7, chr 11, chr 14, chr 15, chr 20 with 5 Mb (at the end of the chromosome) or 10 Mb (not at the end of the chromosome) on one of the chromosomes were also reported, while no ROH was detected on the other chromosomes14,15.

Statistical analysis

SPSS 21.0 (Chicago, USA) was used for the statistical analysis. Classified variables are expressed as numbers and percentages, and the chi-square test was used for comparisons. A two-sided p-value < 0.05 was considered statistically significant. Linear regression models with gamma values were used to assess the changes in abnormal rates in different age groups.