The pathogenesis of spontaneous abortion is very complex, among which chromosomal abnormality is considered to be the main cause [18, 22]. Abortion brings great physical and mental burden to pregnant women, especially those who have multiple spontaneous abortions, and their families have an urgent need for diagnosis of the causes of abortion. Analysis of the possible causes of fetal chromosomal abnormalities is important to provide guidance in assessing the risk of recurrent miscarriage and in choosing subsequent fertility strategies. Previous studies have shown that more than 80% of miscarriages occur within 12 weeks of gestation, and chromosomal numerical abnormality is the most important cause of early abortion, accounting for about 50% [4, 5, 7, 23, 24]. In this study, early abortion accounted for 88.2%, and the abnormal rate was 67.9% in the early abortion, which was mainly due to chromosomal numerical abnormality (50%), consistent with previous research results. The results of this study showed that the abnormal rate of miscarried POCs was 65.1%, which confirmed that chromosomal abnormalities were indeed the main cause of spontaneous abortion. Trisomy abnormalities were the main chromosomal abnormalities, in which chromosome 16 and 22 were the most common, and the incidence of X monomer was the highest in the haplotype, which was consistent with the previous reports [10, 25, 26].
The results of cytogenetic analysis for sporadic and recurrent miscarriages are inconsistent. Some of the studies suggested that there was no difference in the rate of abnormal chromosomal karyotype between sporadic and recurrent miscarriages [23, 27–29]. However, Ogasawara et al.  and Sullivan et al.  described decreased rates of chromosomal abnormalities in recurrent abortion. From our data, in overall, the chromosomal abnormality rate was no difference between sporadic and recurrent miscarriages (66.7% vs. 63.2%, P = 0.678). There was also no correlation between the rate of chromosomal abnormalities and the frequency of miscarriages. However, the incidence of chromosomal structural abnormalities was significantly higher in sporadic miscarriage than recurrent miscarriage (29.2% vs. 5.3%, P = 0.002). Most samples of sporadic miscarriages were from ART patients, which was consistent with the proportion of chromosomal abnormality in ART group.
Does ART increase the incidence of chromosomal abnormalities in embryos? A number of studies had compared and analyzed the incidence of chromosomal abnormalities in villi of aborted fetus during ART and SP, and found that there was no statistically significant difference in the abnormal rate between the two groups [32–35]. But this does not mean that ART technology has no impact on embryo chromosomal abnormalities. The results of this study suggested that although there was no significant difference in the rate of chromosomal abnormalities between the ART group and the SP group, there were significant differences in the types of embryo abnormalities between the two groups. The incidence of chromosomal structural abnormalities was significantly higher in the ART group than that in the SP group. The SP group was predominated by chromosomal numerical abnormalities. On this basis, we further analyzed the possible influencing factors of ART conception, such as fertilization and embryo transfer strategies. Although the ICSI fertilization has been reported to cause a higher incidence of aneuploidy than IVF , most studies have shown that ICSI may not contribute to an increase in the aneuploidy rate in miscarried POCs [25, 35, 37, 38]. In our study, the fertilization method (IVF and ICSI) was also found to be not significantly associated with embryonic chromosomal abnormality in the POCs. However, there was a significant difference in the ratio of chromosomal structural abnormalities between the two groups, with the majority of microduplication (71.4%) in the IVF subgroup and the majority of microdeletion (80%) in the ICSI subgroup. The incidence of visible fetal heart before abortion was higher in the IVF subgroup than in the ICSI subgroup. This hints that the effects of ICSI fertilization may appear earlier on the development of embryos.
A recently retrospective study included the miscarried tissues of 720 patients underwent IVF/ICSI found that frozen embryo transfer was associated with decreased frequencies of embryonic chromosomal abnormalities in miscarried POCs . Our result is consistent with it. The rate of chromosomal abnormality in fresh embryo transfer group was significantly higher than that in frozen embryo transfer group (92.3% vs.50%), and the types of embryo abnormality were also significantly different. The variants of fresh embryos were mainly microduplication, while the variants of frozen embryos were mainly microdeletion. On one hand, it may be related to the relatively better endometrial receptivity during frozen embryo transfer cycle . Since endometrial exposure to excessive ovarian stimulation could lead to an alteration in endometrial gene profile expression and histological and structural abnormalities [40–43], an efficacious embryo selection by the endometrium in frozen embryo transfer cycles may reduce the possibility of poor-quality embryo implantation, thus, reducing the chance of miscarriage with chromosomal abnormalities. On the other hand, embryo cryopreservation may temper the epigenetic alterations induced by assisted reproductive technologies . In addition, the self-repair of embryos after freezing and thawing may also be the reason for the decrease of chromosomal abnormality rate in frozen embryo transfer group. Further research is needed to unveil the underlying mechanisms involved in different embryo transfer cycles.
Age is a generally acknowledged factor that affects aneuploidy in the embryo or miscarriage of the conceptus . Several studies have demonstrated that pregnancy loss in women over 35 years of age is associated with a higher chromosomal aneuploid rate [23, 45, 46]. Fan et al.  have analyzed a total of 1010 miscarriage specimens and found that the group with advanced maternal age displayed a significantly higher frequency of overall chromosomal variants and aneuploidy whatever spontaneous or recurrent miscarriage cases. Kushnir et al.  found that the incidence of chromosomal numeric abnormalities in both IVF and ICSI aborted embryos increased with the age of the mother. In this study, there was no significant difference in the rate of chromosomal abnormalities between patients aged < 35 years and ≥ 35 years. However, there were significant differences in the types of fetal abnormalities between the ART group and the SP group < 35 years. The rates of numeric and structural abnormalities were 28.1% and 31.6% respectively in the ART group, and 57.5% and 10% in the SP group, respectively. Embryo abnormalities in patients aged ≥ 35 years were mainly chromosomal numeric abnormalities which was consistent with Fan et al. , and no significant difference was found between ART group and SP group. This might be related to the decreased ovarian function and egg quality in these patients, which leads to the abnormal separation and replication of chromosomes in gametes or fertilized eggs during early cleavage .
There are several limitations in this study. First, due to the small sample size, we could not continue the age-stratified analysis in fertilization method, embryo transfer strategy. Second, this study was a retrospective design, thus, potential bias factors cannot be fully identified and addressed.