Predictive value of aberrant right subclavian artery for fetal chromosomal in women of advanced maternal age

Background: In entire population, aberrant right subclavian artery ( ARSA ) was in closely association with chromosomal abnormalities. The risk of fetal chromosomal abnormalities increased with the maternal age exponentially. While, the situation in advanced maternal age ( AMA ) group is uncertain. This study aimed to establish the incidence of ARSA in Chinese AMA and non-AMA women and to determine the frequency of aneuploidy among AMA and non-AMA women with ARSA. Methods: The retrospective study included 13,690 singleton pregnancies which were divided into AMA and non-AMA groups. Integrated obstetric ultrasonic screening, biochemical screening, noninvasive prenatal screening and fetal karyotype analysis were analyzed. Results: 1. The overall incidence of ARSA was 0.69 % with no difference in both groups. 2. The prevalence rate of chromosomal abnormalities in AMA group ( 37 / 2,860 ) was much higher than that in non-AMA group. The risk of chromosomal abnormalities significantly increased with both ARSA detected and additional ultrasound findings. 3. With ARSA detected, the incidence increased to 20.00 % and 10.00 % in AMA and non-AMA cases, respectively. 4. Additionally, a case with chimeric Turner syndrome ( 45X / 46XX ) was found with isolated ARSA in AMA pregnancy. Conclusion: There is a high prevalence of chromosomal abnormalities in AMA fetuses. Both isolated and nonisolated ARSA would increase the risk of chromosomal abnormalities. Moreover, when ARSA is found in AMA ones, it confers a sharp increase in the incidence of chromosomal abnormalities.


Introduction
Advanced maternal age ( AMA) is defined as a woman who conceiving and delivering babies at the age of 35 years or older [1][2][3][4]. According to data from Office for National Statistics, in 2013, 20 % of births in England and Wales were to women aged 35 years or over and the average age of mothers has dramatically increased with years [1]. In china, the proportion of increased gradually from 10.1 % in 2011 to 20.5 % in 2016, and it would grow to new highs especially after " Two-child Policy " carried out [2].
As we known, maternal age was in closely association with a range of pregnancy complications, such as preeclampsia, stillbirth and fetal anomalies [1][2][3]. The risk of fetal chromosomal abnormalities increased with the maternal age exponentially. For example, the overall incidence of Down's syndrome is one in every 800 births while it roughly climbs to 1.44 in every 100 births in AMA women [4].
Even so, according to Chen et al. [3]. there is no need for AMA pregnant women to accept invasive prenatal diagnosis directly. But what if there are other associated fetal structural abnormalities in AMA women, like aberrant right subclavian artery ( ARSA)?
During pregnancy, ARSA may easily be detected by prenatal ultrasonography whatever the trimester of pregnanc [5]. The incidence rate of ARSA as an isolated abnormality in healthy populations is known to be about 1 % to 2 % [6]. Even so, Chaoui et al [7]. reported the prevalence of ARSA in fetuses with Down's syndrome for the first time and recommended that an ARSA would act as a new soft marker for trisomy 21 risk assessment. Prevalence of ARSA was 1.02 % in euploid fetuses, whereas 23.6 % in Down's syndrome fetuses [8]. Therefore, ARSA appears to be a fairly reliable ultrasound clue for fetal chromosomal abnormalities, especially congenital cardiac defects and aneuploid abnormality [5][6][7][8].
Most studies indicated that isolated ARSA had no much clinical significance, and there was no need to serve an invasive prenatal chromosomal testing [7][8][9][10]. In Italy, an invasive procedure was offered to all patients with intermediate risk and a retrotracheal ARSA [8]. In the study of Fehmi et al. [10], they suggested fetus with ARSA and aneuploidy relevant ultrasonic soft features, AMA and abnormal biochemical screening to undergo amniocentesis. They concluded that in fetuses with ARSA, karyotyping could be offered to detect Down's syndrome if any high risk factors present. However, in most existing studies, the predictive value of from ARSA in AMA and non-AMA women was not compared. In AMA women, whether ARSA is an effective predictor for fetal chromosomal abnormalities? Is it necessary for those women to take an invasive routinely screened? Hence, the main purposes of this study were to: ( 1) determine the incidence of both isolated and nonisolated ARSA in Chinese AMA and non-AMA women. ( 2) assess the correlation between fetal chromosomal abnormalities and ARSA with or without additional ultrasound findings in AMA pregnancy. There were 13,690 singleton pregnancies with complete materials who underwent fetal ultrasonic   prenatal screening including grade I, grade II, grade III and

Study design
The information of all included pregnant women was retrieved from our computerized database.
Pregnancies was divided into AMA and non-AMA group based on the age of their due dates. Upon the discovery of an ARSA, relevant prenatal diagnostic information was classified and summarized including ultrasonography abnormity, soft markers, serum screening, noninvasive and invasive karyotype analysis, chromosomal microarray and so on. All the participants received the ARSA scanning. All ARSA fetuses and suspected cases should be diagnosed by two physicians ( with prenatal diagnosis qualification and rich experience) and confirmed by at least one follow-up review.
All ARSA cases were followed up until birth by neonatal conventional ultrasonography and echocardiography for postnatal confirmation. ARSA negative fetuses should also be followed up as often as possible to ensure the reliability of the diagnosis, especially to these fetuses at a relatively small gestational age. Prenatal consultation with noninvasive or invasive karyotype analysis was recommended for all fetuses with ARSA. Women who did not undergo invasive karyotype analysis, major aneuploidy abnormalities and karyotype abnormalities were excluded by noninvasive DNA tests, detailed prenatal examinations and neonatal follow-up, and they were considered to have a normal karyotype. All aborted fetuses received autopsies with informed consent on time. The flow chart of this study was shown in Figure 1.

Method of ARSA detection
After routine examination, the fetal heart mode was turned on, and the local amplification function was adjusted to make the section clearly displayed. The angle of ARSA and incident sound wave was ensured less than 30. Axial view ( the three vessel and trachea view), longitudinal view and coronal view were conventionally observed to screen ARSA [11]. When ARSA was displayed in two dimension mode, Doppler velocity was adjusted to ( 15 -30) cm/s to verify the diagnosis. ARSA departed from the origin of the descending aorta, namely the junction of the aortic arch and ductal arch, walked between the trachea and the vertebra, and extended towards the right shoulder. Anatomical and ultrasonic diagram of fetal ARSA were shown in Figure 2.

Statistical analysis
Statistical analysis was completed using SPSS software, Version 20.0 ( IBM Corporation, Somers, NY).
A chi-square test and Fisher's exact test were utilized to compare the incidences of chromosomal abnormalities and ARSA between groups. P < 0.05 was considered statistically significant.

Demographic data and general characteristics
During this period, a total of 13,690 singleton pregnancies ( ranging from 16 weeks + 0 days to 38 weeks + 5 days) were recruited in the study, including 10,830 non-AMA women with the average age of ( 27.1 ± 4.15) year old and 2,860 AMA women with the average age of ( 38.3 ± 3.48) year old.
The positive and negative likelihood and predictive value of ARSA for chromosomal abnormalities were gathered in Table 3. İn entire cohort, the positive and negative likelihood ratios of nonisolated ARSA for chromosomal abnormalities were 69.49 and 0.89, respectively. And those of isolated ARSA for chromosomal abnormalities were 15.47 and 0.90, respectively.

Discussion
In this retrospective study, 13,690 singleton pregnancies were evaluated. We found that the incidence of an ARSA was 0.87% in Chinese AMA and 0.65% in non-AMA women. The incidence of chromosomal abnormalities was much higher in AMA group and with ARSA detected it climbed as high as 20 % in AMA group and 10 % in non-AMA group. Among there, mostly detected was Down syndrome in both groups. Another chimeric Turner syndrome was found in an AMA woman with an isolated ARSA.
In most studies, the prevalence of a prenatal ARSA was studied in the entire population with a prevalence rate ranging from 0.4 % to 1.5 % [12,13]. Concordant with the previous studies, we found the incidence of ARSA in entire cohort was 0.69 %. And as a valuable complementarity, we confirmed there was no difference in the incidence of ARSA during AMA and non-AMA groups. On the timing of prenatal ultrasonic diagnosis of ARSA, Pico et al. [5] presented the mean gestational age for ARSA detecting was 19 weeks + 5 days, ranging from 11 weeks + 5 days to 34 weeks; SD = 4 days). In another study concerning the predictive value of ARSA for Down syndrome, authors successfully checked pregnant women at 16 weeks of gestation [10]. In our experience, ARSA can be detected as early as 12 weeks +4 days gestational age. However, due to the high omission diagnostic rate during the first trimester and the early mid-trimester, and in order to reduce the bias caused by the difference in technical level between examiners and ensure the accuracy, we set the threshold of 16 weeks of gestation for this study. The correlation between ARSA fetuses and chromosomal abnormalities such as Down syndrome was described by some scholars [10,14,15]. As we found whether isolated or nonisolated ARSA would increase the risk of chromosomal abnormalities. When it comes to the isolated ARSA, there a controversy should be proposed. Some authors insisted isolated 8 ARSA correlated strongly with trisomy 21 [16]. Reported by Paladini et al. [13], ARSA should be considered among the three most powerful ultrasound indicators of Down syndrome in the second trimester resembling nasal bone abnormality and increased nuchal fold. In their preliminary study, ARSA was found the only sonographic sign in 7.5% ( 8 / 106) relatively unbiased Down syndrome fetal, which was presented 1.5% in normal fetuses. While a weak association between isolated ARSA and chromosomal abnormalities were reported by other scholars [12]. Recommended by Pico et al. [5], these fetus contained an isolated ARSA required a comprehensive evaluation instead of an invasive karyotype analysis considering isolated ARSA is a condition rarely associated with a chromosomal abnormality.
When concerning the nonisolated ARSA, in virtue of prenatal detected the ARSA with other ultrasound signs, the risk for trisomy 21 increased by factor of 45 described by Fehmi et al. [10]. As in our study, the positive likelihood ratios of nonisolated ARSA for chromosomal abnormalities in the entire population and AMA group were as high as respectively 69.49 and 36.90. The study performed by Svirsky et al. [14] was in support of our results. They claimed ARSA with additional ultrasound findings constitutes a strong predictor for aneuploidy.
What is more, when ARSA is found in AMA ones, we got a sharp increase in the incidence of chromosomal abnormalities. The prevalence rate of chromosomal abnormalities with ARSA in AMA group was 20 % ( 5 / 25) which is much higher when comparing to that in non-AMA group 10 % ( 7 / 70). And the positive likelihood ratios of nonisolated ARSA for chromosomal abnormalities were 36.90 in AMA group and 14.51 in non-AMA group. These findings may contribute to prenatal counseling especially for advanced maternal age pregnant women.
Last but not least, a chimeric ( 45X / 46XX) case with isolated ARSA was found in AMA group in our study. Turner syndrome is well known to be closely associated with cardiovascular malformations with the frequency of 23% to 45%. As reported by Lee et al. [17], an aberrant right subclavian artery was one of the most common major vessel abnormalities in the Turner syndrome patients ( 3 / 20 patients, 15 %). The obstruction of lymphocinesia during embryogenesis may be the possible mechanism of cardiovascular defects in Turner syndrome. In our case of Turner syndrome, we found 9 the unique structural defect by prenatal ultrasound. The suspicious diagnosis clue of sex chromosome aneuploidies was presented by noninvasive prenatal screening with cell-free DNA. And it was eventually confirmed by chromosomal karyotype analysis with amniocentesis and SNP-Array chip inspection. The clinical effect of noninvasive prenatal screening in AMA pregnancy was discussed in a multicenter retrospective study [18]. Authors found the detection efficiency was satisfactory with almost 100% accurate results. Even for sex chromosome aneuploidy in AMA pregnancy, the positive predictive value was approximately 41.30%. Considering the high prevalence of chromosome abnormality including sex chromosome, scholars strongly suggested noninvasive prenatal screening for AMA pregnancy, especially over 40. Therefore, we believe that although only isolated ARSA was found in advanced ones, vigilance will still be needed and the noninvasive prenatal screening might benefit. It would be very conducive to further managing and genetic counseling for advanced maternal women.
In a word, for AMA pregnancy, we suggest that ARSA should be used as a soft indicator of chromosomal abnormalities in genetic ultrasound. Once ARSA is found, a comprehensive assessment is essential. Although only isolated ARSA was found in AMA ones, the noninvasive prenatal screening would help to some degrees.
There were also some limitations in our study. Firstly, a small percentage of fetuses in non-AMA group did not undergo invasive karyotype analysis. The major aneuploidy abnormalities and karyotype abnormalities were excluded by the negative results of noninvasive DNA tests, detailed prenatal examinations and neonatal follow-up. In clinical practice, many people are reluctant to accept invasive karyotype analysis considering the possible risks of invasive operations. In theory, an isolated ARSA was not a sufficient indication for karyotype analysis [5]. Reported by Ranzini et al. [12], all fetuses with ARSA and genetic anomalies totally had additional ultrasound findings. Thus, in similar studies ( respectively published in Fetal Diagn Ther and J Ultrasound Med) [5,6], the authors typically included fetuses have consistently classified fetuses with negative prenatal screening and postpartum follow-up as normal karyotypes. Therefore, we believe that the method adopted in this study is acceptable. Secondly, chromosomal microarray analysis was not analyzed in our current study. While it worth noting, quite a part of deformity might be neglected without chromosomal microarray analysis according to Maya et al. [15]. What's more, the incidence of ARSA in super-aged women ( 40) and its' predictive value for chromosome abnormality were not evaluated individually in our current study. That may probably have profound guiding significance in this group. We look forward to further discussion of these issues in future studies.

Conclusions
In conclusion, the incidence of an ARSA in Chinese AMA women resembled that in non-AMA women.
We have described a high prevalence of chromosomal abnormalities in AMA fetuses. Either isolated or nonisolated ARSA would increase the risk of chromosomal abnormalities. Moreover, when ARSA is found in AMA ones, it confers a sharp increase in the incidence of chromosomal abnormalities and the noninvasive prenatal screening may sever the purpose. In addition, it was worth further exploration in the incidence of ARSA in super-aged women ( ≥40) and its' predictive value for chromosome abnormality.

Supplementary Files
This is a list of supplementary files associated with this preprint. Click to download.