The clinical value of invasive prenatal diagnosis in fetuses with isolated aberrant right subclavian artery: a retrospective study

Abstract

Background

Aberrant right subclavian artery is a common anatomical variant of the embryonic aortic arch, with a prevalence ranging from 0.4 to 2.0%. Although frequently associated with vascular rings or congenital cardiac defects, prenatal assessment primarily relies on the three-vessel and trachea view in ultrasonography. Currently, there is no consensus regarding whether isolated ARSA necessitates invasive diagnostic procedures. This study aimed to evaluate the necessity of routine invasive prenatal diagnosis for fetuses with sonographically isolated ARSA. By conducting a long-term postnatal follow-up of a large cohort and utilizing Bayesian analysis for risk assessment, we sought to provide empirical data to support clinical decision-making.

Methods

The fetuses diagnosed with ARSA via prenatal ultrasound at Hefei Maternal and Child Health Care Hospital from January 2019 to December 2022 were retrospectively analyzed. They were divided into isolated ARSA and non-isolated ARSA groups based on the presence or absence of other ultrasound abnormalities. Within each of these two groups, the fetuses were further categorized into diagnostic and undiagnosed subgroups based on whether they underwent invasive prenatal diagnosis. The study explored the baseline characteristics, genetic testing results, pregnancy outcomes, infant feeding and developmental status, and the results of neonatal color Doppler ultrasound re-examinations in these two groups.

Results

A total of 540 cases of ARSA fetuses were identified, including 449 cases (83.1%) of isolated ARSA and 91 cases (16.9%) of non-isolated ARSA. There were no statistically significant differences in baseline characteristics such as age, pre-pregnancy BMI, and history of diabetes between the two groups (P > 0.05). However, the proportion of non-invasive prenatal testing (NIPT) applications and the pregnancy termination rate were significantly higher in the non-isolated group compared to the isolated group (P < 0.05). Pregnancy outcomes revealed that there were 496 live births (91.6%), while 44 cases (8.1%) chose to terminate their pregnancies due to chromosomal abnormalities and/or severe structural abnormalities. Among the 90 fetuses that underwent invasive prenatal diagnosis, the overall detection rate of chromosomal abnormalities was 11.1%. The detection rates for isolated and non-isolated ARSA were 9.1% (6/66) and 16.7% (4/24), respectively, with no statistically significant difference between the two groups (P > 0.05).The follow-up results of live births showed that 25 (5.0%) of 496 cases had abnormal phenotypes. Among 446 live births with isolated ARSA, 10 cases (2.2%) were found to have abnormal manifestations, with 1.6% (1/66 cases, diagnosed as 21-trisomy mosaicism) in the invasive diagnosis group and 2.4% (9/380 cases) in the undiagnosed group. The difference between the two groups was not statistically significant (P > 0.05). In contrast, the abnormal phenotype rate of live births with non-isolated ARSA was nearly 30.0%. Bayesian risk assessment indicated that the overall posterior risk of abnormal phenotype for isolated ARSA was 2.46% (95% HDI: 1.195%-4.080%), and whether or not invasive diagnosis was performed did not alter this risk. Among the 407 live births that did not undergo invasive diagnosis, 17 cases (4.2%) exhibited abnormalities during follow-up, among whom, genetic testing identified pathogenic variants in two neonates.

Conclusions

The positive predictive value for postnatal aberrant clinical symptoms in fetuses with sonographically isolated ARSA is low (2.24%). In the absence of additional ultrasound markers or significant risk factors, routine invasive prenatal diagnosis is not recommended. Comprehensive genetic counseling should be prioritized to facilitate informed and autonomous decision-making by pregnant women and their families.

Aberrant right subclavian artery (ARSA) is recognized as one of the most prevalent anomalies during aortic arch bifurcation. The prevalence rate of ARSA in fetuses with normal chromosomes is approximately 0.4% to 2.0% [1], and it often coexists with other congenital heart diseases (CHD). Multiple studies have indicated that ARSA is more prevalent among individuals with chromosomal abnormalities [2–4], and the presence of ARSA increases the risk of trisomy 21 syndrome by 3.94-fold [5]. In addition, ARSA is also a phenotypic manifestation of numerous chromosome microdeletion and microduplication syndromes, such as 1q21 deletion, 8p23.1 deletion, 22q11.2 deletion, etc. [6, 7]. However, some studies have pointed out that the risk of chromosomal abnormalities in fetuses with isolated ARSA is low and lacks clinical significance, and should not be used as an indication for invasive prenatal diagnosis [8–11]. Therefore, for such a high incidence of ARSA, especially the isolated type of ARSA, whether invasive prenatal diagnosis is necessary in the context of pregnancy management and prenatal counseling is of particular importance.The purpose of this study is to explore the necessity of invasive prenatal diagnosis for ARSA, especially the isolated type, by retrospectively analyzing 540 cases of fetuses with ARSA detected by color Doppler ultrasound in our hospital. This analysis includes the results of genetic testing, postnatal feeding, growth and development, as well as the long-term follow-up of color Doppler ultrasound results (with a median follow-up time of 35 months). This study complements the lack of long-term prognostic data for isolated ARSA cases that did not undergo invasive prenatal diagnosis in the existing literature, and conducted a quantitative risk assessment using Bayesian analysis, providing more powerful empirical support for prenatal counseling of isolated ARSA.

Subjects and methods

Subjects and grouping

The study subjects and grouping included 540 fetuses with Aberrant Right Subclavian Artery (ARSA) detected during the first trimester ultrasound screening (11–13^+ 6 weeks of gestation), second trimester fetal structural ultrasound screening (20–24^+ 6 weeks of gestation), and systematic fetal structural ultrasound before invasive prenatal diagnosis (16–18^+ 6 weeks of gestation) at our hospital during the period from January 1, 2019, to December 11, 2022. The decision to perform invasive prenatal diagnosis in 90 cases was made by pregnant women on the basis of full genetic counseling. Clinically, the decision to perform invasive prenatal diagnosis is primarily based on maternal advanced age (≥ 35 years), high risk of non-invasive prenatal testing (NIPT), related family history, and other comprehensive factors. Therefore, “whether to accept invasive diagnosis” belongs to indication-driven selection in the real world, rather than random allocation. Inclusion criteria for invasive prenatal diagnosis: (1) singleton pregnancy; (2) The prenatal ultrasound report clearly documented ARSA signs. Exclusion criteria: (1) multiple pregnancy; (2) Complicated with other serious physical or organ diseases; (3) Those with consciousness or communication barriers who cannot provide effective informed consent; (4) Patients with contraindications to invasive procedures (such as active infection, limited puncture path, threatened abortion, or premature birth).Clinical data were extracted from prenatal ultrasound records, genetic testing reports, pregnancy outcomes, and postnatal telephone follow-ups. Cases with incomplete data or those lost to follow-up were excluded from the analysis.Grouping StrategySubjects were categorized into two groups based on ultrasound findings: an isolated ARSA group and a non-isolated ARSA group. Non-isolated ARSA was defined as the presence of concurrent CHD, extracardiac structural anomalies, and/or other ultrasound soft markers. Those who did not undergo invasive prenatal diagnosis could choose whether to undergo NIPT after providing informed consent.

Ethics statement

This study was approved by the biomedical ethics committee of Anhui Medical University (Approval No. 83230473). All procedures were conducted in strict accordance with the declaration of Helsinki and relevant institutional guidelines. Written informed consent was obtained from all participants.

Ultrasound examination

Voluson E8 and E10 ultrasound systems (RAB4-8-D probe; GE Healthcare, Milwaukee, Wisconsin, USA) were used. At least two physicians with ≥ 5 years of experience in prenatal ultrasound diagnosis conducted evaluations on the three-vessel-trachea view and the coronary view of the descending aorta [8]: if the right subclavian artery originated from the descending aorta, coursed posterior to the trachea in a “C” shape, and extended towards the right shoulder and subclavian region, it was diagnosed as ARSA (Fig. 1). Figure 2: The ultrasound image of the three-vessel-trachea view of a normal right subclavian artery (RSA) shows that the right subclavian artery arises normally from the aortic arch without a “C” shape.

Fig. 1.

a Three-vessel and trachea (3VT), DAO: Descending Aorta, ARSA: Aberrant Right Subclavian Artery, MPA: Main Pulmonary Artery, SVC: Superior Vena Cava, T: Trachea, ARCH: Aortic Arch. b Descending Aorta View, DAO: Descending Aorta, ARSA: Aberrant Right Subclavian Artery

Fig. 2.

a Proximal vascular branches of the aortic arch, RSA: Right Subclavian Artery, RCCA: Right Common Carotid Artery, LCCA: Left Common Carotid Artery. b Aortic arch long axis, LSA: Left Subclavian Artery, DAO: Descending Aorta

Genetic analysis

After genetic counseling, pregnant women determined whether to pursue invasive prenatal diagnosis based on informed consent. Chorionic villus sampling (CVS) or amniocentesis was chosen based on gestational age to get fetal samples for genetic analysis.

Karyotype analysis

Fetal specimens underwent routine cell culture, slide preparation, and G-banding. Twenty metaphases were counted under a microscope, and no fewer than 5 karyotypes were analyzed. Karyotypes were described following the International System for Human Cytogenomic Nomenclature (ISCN) 2024 standards [12].

Chromosomal Microarray Analysis

Evaluation was conducted utilizing the Affymetrix CytoScan 750 K array. Copy Number Variations (CNVs) were evaluated by contrasting them with internal and public databases (ClinGen, DGV, DECIPHER, ISCA, OMIM, UCSC Genome Browser). Copy number variations (CNVs) were categorized and analyzed in accordance with the standards set out by the American College of Medical Genetics and Genomics (ACMG) [13].

Follow-up

Neonatal ultrasound follow-up was arranged at 1, 3, and 6 months after birth, and the follow-up period extended up to September 2025. The follow-up duration for all cases ranged from 33 to 36 months, with a median follow-up time of 35 months. According to the guidelines for infant nutrition and feeding assessment services issued by the National Health Commission of China [14], we conducted telephone follow-ups and defined “abnormal clinical phenotype” as: (1) intracardiac and extracardiac abnormalities confirmed by imaging after delivery; (2) chromosomal abnormalities or syndromes confirmed by genetic testing after delivery; (3) recurrent choking or swallowing obstruction, i.e., frequent feeding-related choking or dysphagia (≥ 3 times per week) for at least 4 consecutive weeks, and/or requiring medical evaluation or intervention. Cases of simple weight deviation (lean or fat) or transient physiological cough that spontaneously resolved and did not require treatment during follow-up were not classified as the main adverse clinical outcomes. For children with recurrent choking or swallowing obstruction, pediatricians and sonographers assessed the presence of esophageal/tracheal compression signs through a combination of outpatient physical examination and ultrasound examination. For children with complex or persistent symptoms, further imaging examination and/or genetic testing were recommended when necessary. All outcome judgments were independently reviewed by at least two investigators according to uniform standards. Cases with inconsistent opinions were subject to arbitration by a third doctor holding a senior professional title to minimize the risk of information bias and misclassification.

Statistical methods

SPSS 27.0 statistical software was used for data analysis. Measurement data conforming to normal distribution are presented as mean ± standard deviation, with independent sample t-tests for intergroup comparisons. Count data are shown as [n (%)], with χ² tests for intergroup comparisons. When theoretical frequencies are < 5, the Fisher exact probability method is applied. The odds ratio (OR) and its 95% confidence interval (CI) were used to assess effect size. Differences are considered statistically significant when P < 0.05.For isolated ARSA cases, we defined “the proportion of abnormal clinical phenotypes after birth” as the ratio of confirmed abnormal cases among all isolated ARSA newborns completing follow-up, expressed using the positive predictive value (PPV): PPV = true positives (confirmed abnormal cases) / (true positives + false positives), where specifically false positives denote isolated ARSA cases without postnatal clinical abnormalities. It should be emphasized that this PPV reflects the “risk level of isolated ARSA detected on imaging suggesting postnatal abnormalities” in our study cohort, rather than the traditional PPV derived from single diagnostic tests.

Bayesian analytical approach

In this study, the R language (HDInterval package) was used to perform a Bayesian assessment of the risk of abnormal phenotypes in live-born infants with isolated ARSA, and n represents the corresponding total number of infants. Under the condition that the true probability of anomaly p is given, we assume a binomial sampling model y∼binomial (n, p). Based on previous literature [4, 10] and our clinical experience (suggesting that the risk of abnormal phenotype in isolated ARSA is about 1%), we set a weakly informative Beta (α,β) prior distribution for p, and specified a weakly informative Beta (0.5, 49.5) (effective sample size 50), ensuring the posterior distribution is primarily data-driven. We calculated the posterior mean and 95% highest density interval for the overall, invasive, and non-invasive diagnostic groups. In addition, sensitivity analysis was conducted by adjusting the prior mean (0.5%-2.0%) and effective sample size (50–200) to verify the robustness of the estimates.

Results

This study identified 540 fetuses with ARSA by prenatal ultrasonography. All cases have completed follow-up with no instances of loss to follow-up. The average mother age was 30.35 ± 3.91 years, and the gestational age at diagnosis was 20 ± 1.6 weeks.

Pregnancy outcomes and classification

Among 540 fetuses, 449 (83.1%) had isolated ARSA and 91 (16.9%) had non-isolated ARSA. A total of 44 cases (8.1%) in the two groups chose to terminate pregnancy due to chromosomal abnormalities and/or severe structural abnormalities (in the 41 cases of termination of pregnancy in the non-isolated ARSA group, 1 case was confirmed to be abnormal by invasive diagnosis and then induced labor; the remaining 40 cases did not undergo invasive diagnosis, mainly based on clinical decisions made due to severe structural abnormalities, high risk of NIPT, or other high-risk factors). The remaining 496 (91.6%) were live births. There was no statistically significant difference between the two groups in age, pre-pregnancy BMI, diabetes history, and other baseline characteristics (P > 0.05). The proportion of cases choosing NIPT among those without invasive prenatal diagnosis showed a statistically significant difference between the two groups (P < 0.05). The pregnancy termination rate in the non-isolated group was significantly higher than that in the isolated group (p < 0.05). Table 1.

Table 1.

Comparison of baseline characteristics and pregnancy outcomes between pregnant women with isolated and Non-Isolated ARSA fetuses

	Isolated ARSA [n (%)]n(%)] (n = 449)	Non-isolated ARSA [n (%)]n(%)] (n = 91)	Total [n (%)]n(%)] (n = 540)	χ² / t value	Pvalue
Maternal age, x ± s (years)	30.38 ± 3.81	30.23 ± 4.39	30.35 ± 3.91	0.335 aa	0.739
< 35 years	398	76	474	1.394b	0.238
≥ 35 years	51	15	66
Diabetes
Pre-gestational	6	1	7	0.000	1.000
Gestational	80	15	95	0.024b	0.878
Pre-pregnancy BMI				0.951b	0.265
< 18.5	6	4	10
18.5 ≤ BMI ≤ 23.9	238	48	286
24.0 ≤ BMI ≤ 27.9	107	20	127
≥ 28.0	98	19	117
Invasive prenatal diagnosis	66	24	90	6.605b	0.010
No invasive prenatal diagnosis	383	67	450
Noninvasive genetic risk among cases without invasive diagnosis					<0.001
Not screened	66	30	96
Low risk	312	27	339
High risk	5	10	15
Pregnancy termination	3(0.7%) cc	41	44		<0.001
Live birth	446	50	496

a: Quantitative data are expressed as mean \pm standard deviation, and inter-group comparisons were performed using the independent samples t-test; b: Categorical data are expressed as n (%), and inter-group comparisons were performed using the χ² test or Fisher’s exact test. ARSA Aberrant right subclavian artery, NIPT Non-invasive prenatal testing, GDM Gestational Diabetes Mellitus, BMI Body Mass Index. c: The 3 cases of pregnancy termination in the isolated ARSA group without invasive diagnosis were due to high-risk NIPT results, where the families requested to discontinue the pregnancy

Invasive prenatal diagnosis and detection of chromosomal abnormalities

A total of 90 fetuses underwent invasive prenatal diagnosis (66 cases (73.3%) of isolated type and 24 cases (26.7%) of non-isolated type), and the overall detection rate of chromosomal abnormalities was 11.1% (10 / 90). The detection rates of chromosomal abnormalities in isolated and non-isolated ARSA were 9.1% (6 / 66) and 16.7% (4 / 24), respectively, with no significant difference between groups (P > 0.05). The origins of chromosomal abnormalities were not verified. See Table 2 for genetic results.

Table 2.

Genetic results (including normal and abnormal findings), neonatal outcomes, and follow-up results of ARSA fetuses undergoing invasive prenatal diagnosis

Isolated / Non-isolated	Indications for Invasive Prenatal Diagnosis (n = 16)	Abnormal Genetic Test Resultsn = 10)	Pathogenicityn)	Neonatal Follow-up Resultsn)	Neonatal Sex / Outcomen)
Isolated					Female / Live birth (3); Male / Live birth
	ARSA	18p11.32p11.31 deletion 1.64 Mb (1); 1p36.31p36.23 duplication 1.07 Mb (1); 4q35.1q35 duplication 2.94 Mb (1); 3p13 deletion 257 kb (1); 16p11.2 deletion 761.4 kb	LP (2); VUS (1); P	No tracheal/esophageal compression; no surgical intervention required during follow-up; current weight within normal range (5)
	ARSA	Trisomy 21 (Mosaic)	P	Mild cognitive impairment, short stature; color Doppler ultrasound showed only ARSA; no tracheal/esophageal compression; no surgical intervention required during follow-up; current weight indicates underweight (1)
Non-Isolated					Terminated (1); Female / Live birth (4); Male / Live birth
	ARSA combined with maternal serum screening Trisomy 18 risk 1/238, fetal cavum veli interpositi cyst (1)	4q35.2 deletion 1.58 Mb	P	Terminated
	ARSA combined with fetal tricuspid regurgitation	46,XX, 22p- (1); 12q24.21q24.22 duplication 786 Kb and 22q12.3 duplication 728 Kb (1)	LP (1); VUS	No tracheal/esophageal compression; no surgical intervention required during follow-up; current weight within normal range (2)
	ARSA combined with fetal nasal bone hypoplasia, ventricular septal defect, intracardiac echogenic focus (left ventricle) (1)	46,XY, t(4;11)(p15.3;q21)	LP	Color Doppler ultrasound confirmed ARSA combined with ventricular septal defect; no tracheal/esophageal compression; no surgical intervention required during follow-up; current weight within normal range (1)
	ARSA combined with maternal serum screening Trisomy 21 risk 1/1129, single umbilical artery, left hand polydactyly (1); fetal right cleft lip with alveolar cleft (1); fetal right duplex kidney (1); fetal left hand polydactyly (1); fetal cleft lip (1); fetal left fused kidney (1)	No abnormalities detected	-	ARSA combined with left hand polydactyly (2); cleft lip with alveolar cleft (1); right duplex kidney (1); cleft lip (1); left fused kidney (1); no tracheal/esophageal compression; no surgical intervention required during follow-up; current weights all within normal range

ARSA Aberrant right subclavian artery, P Pathogenic, LP Likely pathogenic, VUS Variant of uncertain significance, “-”: Indicates no genetic abnormalities detected or non-pathogenic; Count data are expressed as frequency (n)

Neonatal outcomes and aberrant phenotypes

Of the 496 live births, follow-up was performed in its entirety, predominantly through outpatient visits at our hospital or via telephone. During follow-up, 25 patients (5.0%) were identified with aberrant clinical characteristics.

Isolated ARSAn = 446)

A total of 10 cases (2.2%) exhibited abnormal clinical manifestations. Among them, the abnormal rate in the invasive prenatal diagnosis group was 1.6% (1/66 cases), and its genetic test indicated Trisomy 21 (chimeric type). The abnormal rate in the non-diagnosis group was 2.4% (9/380 cases, follow-up results are detailed in Table 6), and there was no significant difference between the two groups (P > 0.05). Bayesian risk assessment revealed that the overall abnormal posterior risk of isolated ARSA was 2.46% (95% HDI: 1.195%-4.080%), and whether or not invasive prenatal diagnosis was performed did not change this risk (Tables 3 and 5). Consequently, the clinical positive predictive value (PPV) for postnatal anomalies in the iARSA cohort was 2.24% (10/446).To evaluate the stability of these findings, sensitivity analyses were performed by varying the prior mean (0.5% and 2.0%) and prior strength (effective sample size (ESS) ranging from 50 to 200). The resulting fluctuations in posterior estimates were negligible, confirming the robustness of the primary conclusions (Table 4, continued).

Table 3.

Results of bayesian risk assessment for abnormal manifestations in fetuses with isolated ARSA

	Abnormal Cases / Total n)n)	Crude Proportion	Posterior Mean	95% HDI
Invasive diagnostic group	1/66	1.515%	2.941%	0.077%-8.947
Non-invasive diagnostic group	9/380	2.368%	2.619%	1.197%-4.657
Total births	10/446	2.242%	2.460%	1.195%-4.080

The prior distribution was set as the literature-based abnormality rate (1%). The posterior mean and 95% highest density interval (HDI) were calculated using the Bayesian model

Table 5.

Comparison of postnatal abnormal clinical manifestations in fetuses with ARSA under different prenatal diagnostic strategies

	Postnatal clinical phenotype		Total live births	χ2 value	PP P value	Odds ratio	95% CI
	Presence of abnormalities nn n	Absence of abnormalities nn n
Isolated ARSA				0.187	0.666	0.634	(0.079,5.090)
Invasive prenatal diagnosis	1	65	66
No invasive prenatal diagnosis performed	9	371	380
Non-isolated ARSA				0.004	0.951	1.039	(0.309,3.495)
Invasive prenatal diagnosis	7	16	23
No invasive prenatal diagnosis performed	8	19	27

ARSAARSA ARSA aberrant right subclavian artery

(1) Comparisons between groups were performed using the test or Fisher’s exact test. Count data are expressed as [n (%)]. (2) ORs (95% CI) were derived from univariate logistic regression analysis. (3) Definition of abnormal clinical phenotypes: Refers to the presence of feeding or swallowing problems assessed via telephone follow-up in accordance with the Guidelines for Infant and Young Child Nutrition and Feeding Assessment Services issued by the National Health Commission, or structural abnormalities (excluding ARSA) detected by physical examination or ultrasound after birth

Table 4.

Bayesian sensitivity analysis: posterior risk estimates for abnormal postnatal clinical phenotypes in isolated ARSA

Prior setting	Invasive diagnostic group(n = 66,y = 1): Posterior mean (95% HDI)	No invasive diagnostic procedure(n = 380,y = 9): Posterior mean (95% HDI)	All isolated ARSA(n = 446,y = 10): Posterior mean (95% HDI)
Primary analysisa	2.94%	2.62%	2.46%
Prior mean 0.5%	2.51%	2.43%	2.28%
Prior mean 2.0%	3.38%	2.82%	2.65%
Effective sample size 50b	2.87%	2.59%	2.43%
Effective sample size 200c	3.02%	2.66%	2.49%

HDHD HD highest density interval

a:In the primary analysis, the prior mean was set to 1%. b: An ESS of 50 corresponds to a weaker (less informative) prior. c: An ESS of 200 corresponds to a stronger (more informative) prior

Non-isolated ARSA n = 50)n = 50)

A total of 15 cases (30.0%) had abnormal clinical manifestations. The abnormal rates of the invasive prenatal diagnosis group and the non-diagnosis group were 30.4% (7 / 23 cases, see Table 2 for follow-up results) and 29.6% (8 / 27 cases, see Table 6 for follow-up results), respectively. There was no significant difference between the groups (P > 0.05), Table 5.

Table 6.

Details of abnormal clinical manifestations in ARSA neonates not undergoing invasive prenatal diagnosis

Isolated / Non-isolated	NIPT Risk n)n)	Neonatal Abnormal Clinical Manifestations n)n)	Prenatal Ultrasound Results n)n)	Postnatal Neonatal Ultrasound Results n)n)	Neonatal Sex / Outcome n)n)
Isolated
	Low risk	Mild choking during early feeding (transient physiological choking excluded), no aggravation under conservative observation by end of follow-up; current weight: underweight (2), overweight (1), normal range (2)	ARSA	ARSA only, no obvious compression observed	Male / Live birth (2); Female / Live birth
	Low risk	Swallowing syndrome, congenital choanal stenosis (left-side atresia? ), esophageal web? (with epiglottic dysfunction), etc.	ARSA	ARSA combined with tricuspid insufficiency, moderate pulmonary hypertension, and neonatal pericardial effusion	aa a Female / Live birth
	Not screened	Mild choking during early feeding (transient physiological choking excluded), no aggravation under conservative observation by end of follow-up; current weight within normal range (2)	ARSA	ARSA only, no obvious compression observed	Male / Live birth (1); Female / Live birth
	Not screened	No abnormal clinical manifestations; current weight within normal range (1)	ARSA	Ultrasound > 3 months after birth indicated ARSA combined with mild right renal pelvic separation; no aggravation under conservative observation by end of follow-up	Male / Live birth
Non-Isolated
	Low risk	No abnormal clinical manifestations, current weight within normal range (4); right clubfoot, current weight underweight (1); cleft lip, current weight underweight (1)	ARSA combined with fetal left aortic arch and left ductus arteriosus (2); fetal tricuspid regurgitation (mild-moderate) (1); fetal bilateral renal pelvic separation (1); fetal right clubfoot (1); cleft lip	Ultrasound indicated ARSA only (2); ARSA combined with left aortic arch and left ductus arteriosus (2); tricuspid regurgitation (mild-moderate) (1); bilateral renal pelvic separation (1); Ultrasound indicated ARSA only	Male / Live birth (2); Female / Live birth
	High risk	No abnormal clinical manifestations, current weight within normal range (1); Intellectual disability, epilepsy, microcephaly, movement disorder, current weight underweight (1)	ARSA combined with ventricular septal defect (muscular) (1); fetal dysgenesis of corpus callosum	ARSA combined with ventricular septal defect (1); agenesis of corpus callosum, bilateral ventriculomegaly (1)	Female / Live birth

ARSA Aberrant right subclavian artery, NIPT Non-invasive prenatal testing

This table outlines examples that did not receive invasive prenatal diagnosis yet displayed aberrant clinical characteristics after birth. 2. Prenatal ultrasound revealed ARSA in all instances; some individuals exhibited concomitant abnormalities (see the “Prenatal Ultrasound Results” column). Count data are represented as frequency (n). CHARGE syndrome (Coloboma, Heart problems, Atresia choanae, Growth retardation, Genital abnormalities, Ear abnormalities syndrome) has been validated through genetic testing as resulting from a pathogenic mutation in the CHD7 gene

.

Atypical clinical presentations in non-invasive diagnostic cases

Among the 407 infants who did not undergo invasive prenatal diagnosis, 17 cases (4.2%) were identified with adverse clinical phenotypes during postnatal follow-up. The majority of this cohort exhibited no atypical clinical symptoms.Within the subset of 17 children presenting with anomalies, postnatal genetic testing yielded a definitive diagnosis in two cases: one revealing a pathogenic variant in the CHD7 gene (c.481 C > T; p.Gln161Ter, Exon 2) and the other identifying a 3.94 Mb pathogenic sequence deletion at the 8q23.1 locus. Table 6 provides detailed characteristics of the abnormal clinical phenotypes observed in newborns with ARSA who did not receive invasive prenatal diagnosis.

Discussion

Previous studies have revealed the potential association between aberrant right subclavian artery (ARSA) and abnormal genetic outcomes, but most prior research has focused on prenatal imaging findings and chromosomal abnormality risk assessment, with insufficient data on postnatal feeding, development, and long-term systematic follow-up of isolated ARSA fetuses [15, 16]. This study provides postpartum follow-up data from a large cohort of ARSA fetuses, particularly supplementing long-term prognostic information for isolated ARSA cases lacking invasive diagnostic confirmation, thereby offering real-world evidence for prenatal counseling. Compared with existing literature, we quantified the postpartum abnormal phenotype risk of isolated ARSA using positive predictive value (PPV, 2.24%) and Bayesian posterior risk (2.46%; 95% HDI: 1.195%-4.080%), providing a clinically interpretable quantitative metric for clinical consultation.

ARSA has often been regarded as an important soft marker for trisomy 21 syndrome. In this study, one case of trisomy 21 mosaicism was detected in the isolated ARSA group, sparking controversy over whether karyotyping should be routinely performed. However, it must be clarified that the significant association between ARSA and trisomy 21 syndrome is primarily based on unscreened populations or cases with concurrent other structural abnormalities. The latest large-scale study indicated that the predictive efficacy of isolated ARSA for trisomy 21 syndrome significantly decreases after excluding other ultrasound soft markers [5]. The study results revealed that the detection rate of chromosomal abnormalities in isolated ARSA cases was 9.1% (6 / 66 ), and the proportion of abnormal clinical phenotypes was only 1.6% (1/ 66 ). This finding aligns with the conclusion reported by Montero Carreras et al. (2024) regarding the favorable postpartum prognosis of isolated ARSA [4]. We further employed univariate logistic regression analysis to calculate the odds ratio (OR) of abnormal phenotypes between the group undergoing invasive prenatal diagnosis and the group not undergoing such diagnosis. The result was OR = 0.634 (95% CI: [0.079, 5.090], P = 0.666), consistent with the conclusion of Bayesian risk assessment. This indicates that for pregnant women with no other abnormalities detected on ultrasound and a low risk of NIPT, the presence of isolated ARSA does not elevate the fetal genetic risk to a level that necessitates the risks associated with amniocentesis. Therefore, the invasive prenatal diagnosis of isolated ARSA should be comprehensively evaluated in combination with other clinical indicators and risk factors, rather than being routinely considered as the first choice.

The findings of this study are highly consistent with previous literature. Several studies have shown that isolated ARSA has a low risk of chromosomal abnormalities and does not require invasive prenatal diagnosis [17–19]. For example, the study of Li et al. pointed out that the vast majority of isolated ARSA are actually benign vascular variants, rather than the direct consequence of pathogenic genetic variants [2]. One study integrated multiple cohort studies, and the calculated odds ratio (OR) was 1.2 (95% CI 0.8–1.8, I²=40%), indicating that although isolated ARSA is associated with the risk of trisomy 21, the overall detection gain is small, and the heterogeneity is mainly due to differences in sample size and detection technology [20]. The detection rate of chromosomal abnormalities in this study was 9.1%, which was in line with the previously reported range of 5.3% − 25% [11, 21]. Moreover, the proportion of abnormal clinical phenotypes after delivery was low (1.6% − 2.4%), and the sample size for invasive prenatal diagnosis of isolated ARSA was small. This finding was later verified by Bootstrap resampling and the Fisher exact test (P = 1.0), further confirming the absence of statistical difference and supporting the argument against recommending invasive prenatal diagnosis. Although some studies emphasize that isolated ARSA may suggest genetic abnormalities [2], these views are mostly based on small samples with selection bias.

Most studies see solitary ARSA as a benign anatomical variant, generally lacking substantial cardiovascular implications; nonetheless, a minority of individuals may exhibit dysphagia and/or respiratory distress due to constriction of the esophagus and/or trachea [19, 22]. In our subsequent evaluation, we meticulously differentiated between “specific anatomical/functional abnormalities” and “non-specific/transient symptoms.” This study noted a comparable occurrence: in the sample lacking invasive diagnostic, 7 infants demonstrated mild choking during initial feeding (excluding temporary physiological choking). Odacılar et al. indicated that the predominant portion of the so-called “dysphagia” resulting from isolated ARSA is mild and resolves spontaneously with maturation [21]. Moreover, whereas literature indicates that around 68% of patients with a right aortic arch exhibit analogous symptoms [23, 24], ultrasonography reassessment of these 7 children disclosed no structural abnormalities or evident compression aside from the ARSA. Their clinical presentations aligned more closely with physiological differences in general newborn development than with pathological dysphagia. From the initial instance of choking to the conclusion of follow-up, symptoms were stable; however, long-term prognosis necessitates additional monitoring. Prayer D et al. [25]observed that when symptoms manifest in ARSA, they are frequently misidentified as esophageal disorders, resulting in overlooked or erroneous diagnoses. For patients necessitating intervention, multislice spiral computed tomography (MSCT) is advised prior to treatment to thoroughly evaluate vascular variations. Among the 380 fetuses that did not receive invasive prenatal diagnostics, one case of CHARGE syndrome was overlooked, yielding a missed diagnosis rate of 0.26% (1/380). The prenatal ultrasound for this fetus revealed only isolated ARSA, while the NIPT results indicated a low risk. In conjunction with prevailing clinical practice, which does not consider ARSA as a direct indication for invasive prenatal diagnosis [26], no additional testing was conducted. Postnatally, the infant had swallowing syndrome, congenital choanal stenosis (suspected left-sided atresia), and an esophageal web (accompanied by epiglottic dysfunction). Genetic tests revealed a harmful mutation in the CHD7 gene, hence confirming CHARGE syndrome. CHARGE syndrome primarily results from point mutations or microdeletions/insertions in the CHD7 gene located at the 8q12 region [27, 28], rendering conventional invasive prenatal diagnostic procedures, such as karyotyping, unable for its detection. Corsten-Janssen et al. [29] indicated that arch vessel anomalies, including ARSA, may occur independently in patients with CHARGE syndrome; the potential of isolated arch vascular anomalies as a predictor for CHARGE syndrome requires additional investigation.Although this case of CHARGE syndrome was missed in diagnosis, it highlights the limitations of ultrasound examination and standard NIPT in detecting microdeletion/microduplication syndromes or single-gene disorders.

The core of invasive prenatal diagnosis is to evaluate whether its residual genetic risk is significantly higher than the miscarriage risk (about 0.1%-0.3%) or the baseline risk of birth defects in the general population. This study found that although some isolated ARSA cases were associated with genetic abnormalities (such as 3p13 deletion, 18p11.32p11.31 deletion, etc.), postpartum follow-up showed that most individuals carrying these variants had no significant clinical phenotype. Research has shown that compared with the normal control group, isolated ARSA fetuses did not show significantly higher pathogenic CNV load [11]. In general, although the detection rate of genetic abnormalities in this study was 9.1%, with limited representativeness, which may be affected by the sample size, these cases did not show clinical abnormalities, and the overall pregnancy termination rate was very low (0.67%). Based on the above findings, after the completion of comprehensive ultrasound evaluation and genetic counseling, clinical screening can give priority to non-invasive prenatal testing (such as NIPT), thereby significantly reducing the risk of fetal loss caused by invasive procedures. In this study, the positive predictive value of postpartum abnormal clinical manifestations of isolated ARSA was low (2.24%). In order to further quantify the postpartum risk, Bayesian simulation estimation was performed, showing that the overall abnormal posterior risk of isolated ARSA was 2.46%, 95% HDI: 1.195%-4.080%.Sensitivity analyses using alternative weakly informative priors yielded similar posterior means and HDIs, suggesting that the main conclusions are robust to reasonable changes in prior assumptions. Therefore, these findings further support the favorable prognosis of isolated ARSA and supports the preference for non-invasive monitoring strategies in non-high-risk situations.Therefore, these results reinforce the generally favorable prognosis of isolated ARSA, supporting the prioritization of non-invasive monitoring strategies in low-risk pregnancies. However, residual uncertainty in risk stratification highlights the critical importance of long-term postnatal follow-up to definitively establish clinical outcomes.In the clinical management of isolated ARSA, we recommend focusing on the following aspects: First, a rigorous and systematic exclusion of concurrent anomalies is essential to ensure the accurate classification of isolated cases. Second, pregnant women opting for NIPT must be adequately counseled regarding its limitations in detecting pathogenic copy number variants (CNV) or monogenic disorders (e.g., CHARGE syndrome). Even in the setting of low-risk NIPT results, clinicians should maintain vigilance by integrating family history and performing detailed multi-system ultrasound screening to minimize the risk of missed diagnoses. Where clinically indicated, targeted genetic testing should be offered, and relevant clinical symptoms should be closely monitored postnatally.Looking forward, the integration of high-throughput sequencing technologies, such as whole exome sequencing (WES), holds promise not only for refining prognostic accuracy and recurrence risk assessment but also for elucidating the underlying genetic etiology of ARSA [30].

Despite the valuable data provided, this study has several limitations. First, this study is a single-center retrospective study, which may introduce selection bias, information bias, and unmeasured confounding factors. Second, the sample size for invasive prenatal diagnosis in this study was limited (90/540 cases, 16.7%), potentially reducing statistical power. Future research should increase the proportion of such cases to over 50% through multi-center cohort designs to enhance result robustness. Third, the follow-up period was limited to within 3 years of age, which may not fully evaluate long-term neurodevelopmental outcomes.

In conclusion, this study indicates that although isolated ARSA is associated with certain genetic abnormalities, its postpartum clinical outcomes are generally favorable. Therefore, the clinical benefits of performing invasive prenatal diagnosis are limited, and it may pose unnecessary risks to both the mother and fetus. In clinical practice, it is advisable to prioritize non-invasive screening techniques, such as non-invasive prenatal testing (NIPT), for fetuses with isolated ARSA, and to improve multidisciplinary postnatal follow-up through collaboration among pediatrics, cardiology, and genetics. These findings provide a foundational basis for clinical genetic counseling, aiding prospective families in making more informed reproductive choices.

Authors’ contributions

CB D and SY X contributed to data collection, statistical analysis, drafting and revision of the manuscript. XQ Land TT D contributed to organizing the survey and interpretation of the data. JY Q and XX C contributed to the study design, data collection and revision of the manuscript. All authors reviewed the manuscript.

Funding

This work was supported by the Anhui Provincial University Scientific Research Project [Grant Number: 2023AH050575].

Data availability

The datasets used in the current study are available from the corresponding author upon reasonable request.

Declarations

Ethics approval and consent to participate

This study was approved by the Ethics Committee of Hefei Maternity and Child Health Care Hospital. All procedures performed were in accordance with the principles of the Declaration of Helsinki. Informed consent was obtained from all participants.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.