Is there a relationship between assisted reproductive technology and maternal outcomes? A systematic review of cohort studies

Abstract Background: Pregnancy with assisted reproductive technology (ART) is accompanied by fetal and maternal outcomes. Objective: This systematic review aimed to assess the relationship between ART and maternal outcomes. Materials and Methods: In this systematic review, the electronic databases, including PubMed, MEDLINE, Web of Science, Scopus, Science Direct, Cochrane Library, Google Scholar, Magiran, Irandoc, and Scientific Information Database were searched for maternal outcomes reported from 2010-2021. The Newcastle-Ottawa Scale for cohort studies was used to assess the methodological quality of studies. Results: A total of 3362 studies were identified by searching the databases. After screening abstracts and full-text reviews, 19 studies assessing the singleton pregnancy-related complications of in vitro fertilization/intracytoplasmic sperm injection were included in the study. The results demonstrated that singleton pregnancies conceived through ART had higher risks of pregnancy-related complications and adverse maternal outcomes, such as vaginal bleeding, cesarean section, hypertension induced by pregnancy, pre-eclampsia, placenta previa, and premature membrane rupture than those conceived naturally. Conclusion: In conclusion, an increased risk of adverse obstetric outcomes was observed in singleton pregnancies conceived by ART. Therefore, obstetricians should consider these pregnancies as high-risk cases and should pay special attention to their pregnancy process.


Introduction
Infertility can be defined as the failure of a couple to conceive after at least one year of having regular, unprotected intercourse (1).Nowadays, 10-20% childbearing age women are affected by infertility worldwide, and more than 15% of couples suffer from infertility worldwide (2).Similarly, several studies indicated that the prevalence of infertility in Iran has increased since 2009, with a reported rate of 20% in 2019 (3).According to the results of studies, the infertility rate in Iran is higher than the global average (4).Assisted reproductive technologies (ART) can be defined as a technique by which fertilized embryos are handled in vitro to induce pregnancy (5).An ever-increasing population of patients seeks to conceive using ART (6).According to the studies conducted, about 1.5 million in vitro fertilization (IVF) is implemented annually.Approximately 8% of all infants born in the United States result from ART, and the total number of infants born worldwide exceeds 8 million (7)(8)(9)(10).Moreover, 1-3% of births in Western countries and 1-5% in other countries, such as China and Japan, result from IVF (11).On the other hand, several studies indicated that the adverse consequences of ART could adversely influence the health of families and, consequently, the society (12).
From the beginning of the first pregnancies resulting from ART, the health of both mothers and children conceived by ART is considered a public concern.Monitoring the data for children born after such procedures is promising, and the perinatal consequences of children conceived by ART have improved over the years; however, there is still evidence of adverse maternal and fetal consequences (6).ART-conceived women are at higher risk of pregnancy compared to those who conceived spontaneously.These consequences include ovarian hyperstimulation, preeclampsia, ectopic pregnancy, diabetes (13)(14)(15)(16), postpartum hemorrhage (17), and twin/multiple pregnancies (18).Studies have also shown that women who have conceived using ART suffer from pregnancy-specific anxiety, lower quality of life, similar or lower symptoms of depression, and poor self-esteem (19)(20)(21).On the contrary, some studies have reported that the ART procedures associated with IVF and intracytoplasmic sperm injection (ICSI) (22) are not responsible for adverse health-related outcomes (23).
The use of ART, nowadays, is increasing, and similar to any medical intervention, it too involves a potential risk.Therefore, it is necessary to study the maternal outcomes in cases using ARTs for pregnancy.However, the literature review showed that there are contradictory studies in this regard, so this study aimed to systematically evaluate the relationship between ART and maternal outcomes.

Design
(F.H. and M.J.).The research process was primarily carried out based on systematic research using Persian and English keywords such as ["pregnancy" OR "pregnant women" OR "maternal outcomes" OR "pregnancy outcomes" OR "obstetric outcomes" OR "perinatal outcomes" OR "adverse maternal outcomes"] AND ["In vitro fertilization" OR "Intra cytoplasmic sperm injection" OR "assisted reproductive technology" OR "assisted reproductive technique"] AND ["cohort studies" OR "retrospective cohort" OR "prospective cohort" OR "analytical studies"].
The included studies' reference lists were manually searched to identify more relevant articles from 2010-2021.The search strategies in the mentioned databases are presented in table I.

Inclusion and exclusion criteria
Studies eligible for inclusion in the study if they: 1) had a prospective or retrospective cohort design; 2) compared maternal outcomes of ART singleton pregnancies with those conceived naturally; 3) used IVF and/or ICSI as the exposure of interest; 4) reported maternal outcomes (or data to calculate them); and 5) were published in English and Persian languages.Review papers, non-peer-reviewed local and/or federal government reports, conference abstracts, and presentations were excluded from the study.Potential studies were evaluated to avoid the duplication of the case series.

Data collection and extraction
The following data were extracted from the included studies: author's name, year of publication, study design (retrospective or prospective), age of participants, gestational age based on day or week, the number of participants in each group, type of ART, reports of adverse maternal outcomes and complications during pregnancy, and the main results reported by studies. 2 researchers (F.H. and M.J.) conducted independent reviews of the titles and abstracts of the included studies.The full texts of the included studies were considered for further evaluation based on the study inclusion and exclusion criteria.The 2 authors would receive consultation from the 3 rd author (M.A.) regarding any discrepancies between them.

Methodological quality assessment
Based on the Newcastle-Ottawa Scale (NOS) principles, the quality of the studies was evaluated (26).Each quality item was awarded a star as a quick visual assessment.The number of stars determined the study quality, with 9 being given to the highest quality.Studies receiving 7 stars were considered as higher methodological quality.

NOS for cohort studies is a widely known scale
for evaluating the quality and potential for bias in observational studies (27,28).The NOS can be utilized for cross-sectional, case-control, and cohort studies (29).The NOS assesses 3 quality parameters (selection, comparability, and outcome) which are divided into 8 specific elements, differing slightly when considering cross-sectional, case-control, and cohort studies.Each scale component is graded from a point, except for the comparability parameter, which gets up to 2 points.Therefore, the maximum for each study is 9, and studies with a grade of < 5 are identified as having a high risk of bias (29,30).

Search results
A total of 3362 articles were identified by using initial search criteria.1602 out of 3362 studies were excluded due to duplication.During the title and abstract screening, 1385 records were removed.Review papers, such as narrative and systematic reviews (n = 85), studies in which singleton data could not be extracted (n = 95), lacked a control group of natural conception (n = 160), and those having limited information for outcomes (n = 16) were excluded from the study.Finally, 19 studies were included in this systematic review (Figure 1).

Main results
In this systematic review, the adverse maternal outcomes of singleton pregnancy were reported by reviewing 19 studies.The main results are shown in table II.

Vaginal blood loss during pregnancy
This maternal outcome was assessed in 5 studies (32,38,39,41,45).All these 5 studies showed that the vaginal bleeding (V.B.) or blood loss in the IVF group was considerably higher than in CG.
The study by Koudstaal et al. showed that women with placenta previa and preterm contraction in the IVF group had higher bleeding in the second and third trimesters compared to the CG (2.3 vs. 0.3, p = 0.05) (39).According to the results of a study pregnancy complications (including spontaneous abortion, gestational diabetes mellitus (GDM), and cesarean delivery) significantly increased the risk of excess VB in ART pregnancies compared with CG (21.4% vs. 12.9%; OR = 1.67, 95% CI: 1.18-2.37)(32).The results of Schieve et al. indicated a higher risk of VB and uterine bleeding among women who had undergone the IVF compared to the CG (R.R. = 3.2, [1.5-6.8],p < 0.001) (45).
Another study reported that the first-trimester bleeding in the IVF group was higher than that in the CG (OR = 1.68; 95% CI: 1.0-2.86,p < 0.05) (41).Also, a study results, revealed that blood loss rate during delivery were 662.1 ± 6.8 mL in the CG and 998.2 ± 18.9 mL in the ART group, while the atonic bleeding in the ART group was significantly higher than that in CG (p = 0.006) (38).

Cesarean delivery
Cesarean delivery as an essential maternal outcome has been reported in 12 studies (31-33, 37, 39-41, 43, 45-48).Among numbered studies, 10 studies demonstrated that the rate of cesarean International Journal of Reproductive BioMedicine delivery in the ART group was higher than that in the CG (31-33, 37, 39, 41, 43, 45-48); however, no significant differences were observed in 2 studies regarding the risk of cesarean delivery in ART group compared to CG (37,40).
In a study which assessed the risk of prematurity in singleton pregnancy using ART, showed that a significant difference was observed concerning the prevalence rate of emergency and elective cesarean delivery between control and ART groups (28.4% vs. 14.2%, p < 0.01) and (13.7 vs. 6.3%, p = 0.04), respectively (31).
A study also demonstrated that the prevalence of elective cesarean delivery in the IVF group was significantly higher than in the CG (8.8 vs. 4.2, p = 0.03) (39).According to the results of a study the rate of cesarean delivery in the ART group was higher than that in CG (OR = 1.33, 95% CI: 0.095-1.87,p = 0.012) (41) (p = 0.001), the cesarean delivery rate was found to be significantly higher in ART group compared to CG (47).
The results of a study showed that the rate of cesarean delivery on maternal requests with no medical indication was significantly increased in the ART group compared to CG (OR = 1.03, 95% CI: 0.75-1.41,p < 0.05) (32) (48) were other included studies, which indicated that the rate of cesarean delivery in the ART group was significantly higher than that in CG.
In a study, the results indicated no significant difference was observed between the ART and the CGs in terms of cesarean delivery (77.4% vs. 75.0%,p = 0.497) (37).Also, the study by Isaksson et al. showed no significant difference between the 2 groups for cesarean delivery (40).

Preterm delivery
9 studies (22,31,32,(37)(38)(39)(44)(45)(46) assessed preterm delivery as an adverse maternal outcome among the ART groups.In 7 studies, the prevalence of preterm delivery was higher in ART such as IVF or ICSI groups compared to CG.For example, Koudstaal et al. showed that in the IVF group, pregnancies were more likely to be terminated preterm compared to CG (15 vs. 5.9, p < 0.001) (39).Perri et al., who investigated the association between singleton ART pregnancies and the risk of prematurity indicated that preterm delivery in the ART group (20%) was significantly higher than that in CG (4%) (p = 0.001) (31).Also, Katalinic et al. revealed that preterm birth in the ICSI group was higher than that in CG (3.9% vs. 3.5%, p < 0.01) (22) 95% CI: 1.32-8.13,p = 0.01) also indicated that rate of preterm labor in ART group was significantly higher than that in CG (46).
The studies by Isaksson et al., showed that the rate of HIP in CG was significantly lower than that in the IVF group (p < 0.05) (40), and da Silva et al. (46), demonstrated that a significant difference was observed between the 2 groups concerning the prevalence rate of HIP (11.1% in ART vs. 25 (36) showed that the rate of the HIP in the ART group was significantly higher than that in CG (p < 0.001).

GDM
GDM as a maternal outcome was assessed in 13 studies (32, 34-37, 40-42, 44-48).In 7 studies, no increased risk of GDM was seen among the ART group compared to CG.  46) (p = 0.20) demonstrated that no significant differences were observed between 2 groups in terms of the prevalence of GDM.In addition, the studies by Szymusik et al. (44) (p = 0.48) and Wang et al. (37) showed that GDM in the ART group was not higher than that in CG (p = 0.996).47) reported weak evidence for the higher risk of GDM in the ART group compared to CG (OR = 1.25, 95% CI: 0.96-1.63,p = 0.045).
According to the results of a study demonstrated that the risk of pre-eclampsia was increased in the ART group as compared with CG (OR = 2.15, 95% CI: 1.33-3.46),and also no significant difference rate was observed in the 2 groups in terms of eclampsia/hemolysis, elevated liver enzyme levels and low platelet levels (OR = 0.40, 95% CI: 0.14-1.09)(48).
A study reported that the late onset of pre-eclampsia in the ART group was higher than that in CG, but no significant difference was observed in the early onset pre-eclampsia of the ART group (7.0% vs. 4.1%, aOR = 1.17, 95% CI: 0.83-1.63,p = 0.34) (38).
In a study which investigated the pregnancy course and outcome after ICSI, showed that placenta previa was higher in the ICSI group (R.R. = 5.68, 95% CI: 3.59-9.01)(44).Zhu et al.

Quality assessment of the included studies
The quality of 19 studies were assessed using the NOS for cohort studies.The results showed that all the included studies had good methodological quality.The quality assessment of the included studies is presented in table III.b) The statistical test is not appropriate, not described, or incomplete
It is well known that ART pregnancies have an increased risk of maternal in comparison with naturally conceived pregnancies, whether singletons or multiples (49).In line with the present study, some studies have shown that most ART pregnancies are associated with higher maternal risks (50)(51)(52).However, recent advancement in ART caused some controversies, so that the singletons may be associated with higher obstetric risks due to how they are done.Researchers are becoming increasingly interested in this topic as single-embryo transfers become more common (53).
It is unclear why ART singleton pregnancies lead to an increase in adverse pregnancy outcomes (APOs) prevalence; however, some studies suggest that ART procedures, maternal infertility factors, or a combination of these factors can contribute to infertility (54).Researchers found that factors associated with ART itself, such as induce ovulation medications or maintain pregnancy early in the pregnancy cycle, the duration of culture, freezing and thawing embryos, polyspermic fertilization, and delayed oocyte fertilization, may change the hormonal environment at the time of implantation, failing to implant.Both gametes and embryos can be manipulated to produce APOs.Also, the higher rates of APOs observed in ART pregnancies may be attributable to closer monitoring of pregnancies made through ART than those conceived naturally (47,(54)(55)(56).In contrast, there are fewer studies suggesting that ART procedures, such as IVF and ICSI, are not responsible for APOs (40).
These complications can be seen in sub-fertile women who conceived without the aid of ART, but they experienced a higher risk of preterm birth (57)(58)(59)(60)(61), pregnancy-induced hypertension or pre-eclampsia (60)(61)(62), and GDM (60).In addition to the above reason, parents who use ART have higher socioeconomic status (63).Couples who have undergone IVF and/or ICSI, especially those with severe conditions requiring more invasive procedures and common monitoring sessions, have a stronger desire for experiencing healthy pregnancy (48).So, there are more reports of their complaints.Finally, there are other reasons for differences in outcomes prevalence among continents and countries, such as ethnic, socioeconomic, and environmental differences, medical insurance, screening programs, sample size, medical procedures, study design, data collection methods, etc., which may contribute to these differences.

Clinical implication
This study can be a useful resource for midwives and gynecologists to understand the most common maternal outcomes among couples using ART for pregnancy and pay considerable attention to these women during pregnancy.
Also, the findings of this systematic review showed the importance of conducting high-quality studies to improve the physical health of pregnant women.

Strength and limitations
This study had 2 limitations, including a lack of evaluation of the prevalence of maternal complications based on the type of pregnancy (singleton vs. twin pregnancy) and using only English and Persian full-text papers.The strengths In another 2 studies by Apantaku et al. (OR = 2.0, 95% CI: 0.7-5.8,p = 0.018) (43) and Jaques et al.

IdentificationFigure 1 .
Figure 1.Flow chart of study selection according to PRISMA guidelines.
1.33-3.45,p < 0.01) and PE mild (aOR = 1.57, 95% CI: 1.14-2.17,p = 0.01) in the ART group were higher than those in the CG International Journal of Reproductive BioMedicine ART and maternal outcomes

1 ) 1 ) 1 ) 2 )
Representativeness of the sample a) Truly representative of the average in the target population.*(all subjects or random sampling) b) Somewhat representative of the average in the target population.*(non-random sampling) 2) Sample size: a) Justified and satisfactory.*b) Not justified 3) Non-respondents: a) Comparability between respondents and non-respondents characteristics is established, and the response rate is satisfactory.*b) The response rate is unsatisfactory, or the comparability between respondents and non-respondents is unsatisfactory 4) Ascertainment of the exposure (Addis et al.): a) Validated measurement tool.**b) Non-validated measurement tool, but the tool is available or described.*Comparability: (Maximum 2 stars) The subjects in different outcome groups are comparable, based on the study design or analysis.Confounding factors are controlled a) The study controls for the most important factor (select one).*b) The study control for any additional factor.*Outcome: (Maximum 3 stars) Assessment of the outcome: a) Independent blind assessment.**b) Record linkage.**c) Self report.*Statistical test: a) The statistical test used to analyze the data is clearly described and appropriate, and the measurement of the association is presented, including confidence intervals and the probability level (p value).*

Table II .
The characteristics of the included studies

Table II .
Continued

Table II .
Continued

Table II .
Continued

Table III .
The methodological quality assessment of the included studies

Table III .
Continued