Assisted reproductive technology: Pregnancy and maternal outcomes

INTRODUCTION — Assisted reproductive technology (ART), which encompasses in vitro fertilization (IVF) and intracytoplasmic sperm injection (ICSI), has become increasingly successful. As a result, indications for use of ART have expanded and concerns about the outcome of these pregnancies have accompanied their increasing prevalence. Whereas most individuals undergoing IVF and their resulting offspring are healthy, ART has been associated with increased adverse pregnancy and maternal outcomes. However, research in this field is complicated by the need to distinguish between the effects of ART on outcomes in offspring versus multiple other confounding or mediating factors, including selection bias related to maternal age and other conditions linked to infertility, technological changes in the performance of ART, number of fetuses produced, and changes in obstetric and neonatal care. In general, the best outcomes following ART interventions, including IVF, occur with singleton pregnancies that started as singleton pregnancies.

The impact of ART on maternal and pregnancy outcomes will be discussed here. In-depth discussions of IVF and ICSI procedures can be found separately.

●(See "In vitro fertilization: Overview of clinical issues and questions".)

●(See "Intracytoplasmic sperm injection".)

In this topic, when discussing study results, we will use the terms "woman/en" or "patient(s)" as they are used in the studies presented. We encourage the reader to consider the specific counseling and treatment needs of transgender and gender diverse individuals.

BACKGROUND

Terminology — ART includes "all interventions that include the in vitro handling of both human oocytes and sperm or of embryos for the purpose of reproduction. This includes, but is not limited to, IVF and embryo transfer (ET), intracytoplasmic sperm injection (ICSI), embryo biopsy, preimplantation genetic testing (PGT), assisted hatching, gamete intrafallopian transfer (GIFT), zygote intrafallopian transfer (ZIFT), gamete and embryo cryopreservation, sperm, oocyte and embryo donation, and gestational carrier cycles. Thus, ART does not, and ART-only registries do not, include assisted insemination using sperm from either a woman’s partner or a sperm donor” [1]. Additional terminology for fertility and related therapies are presented separately. (See "In vitro fertilization: Overview of clinical issues and questions", section on 'Terminology'.)

Global impact of ART — The first pregnancy after in vitro fertilization (IVF) of a human egg and the first birth of an IVF baby were reported in 1976 and 1978, respectively [2,3]. Since then, there have been more than eight million pregnancies worldwide [4], and another 500,000 deliveries are being added annually by IVF and its modifications [5]. Live birth rates after IVF and ICSI are presented in detail in related content.

●(See "In vitro fertilization: Overview of clinical issues and questions", section on 'What are the pregnancy and live birth rates?'.)

●(See "Intracytoplasmic sperm injection", section on 'Pregnancy'.)

Pregnancy rates after ART are influenced by a variety of factors, which are discussed in detail separately. (See "In vitro fertilization: Overview of clinical issues and questions", section on 'What factors impact IVF success?'.)

IMPORTANCE OF SINGLETON GESTATION — In general, the best pregnancy outcomes following ART, including IVF, occur with singleton pregnancies that started as singleton pregnancies (rather than those that started as multiple gestations and underwent either spontaneous or assisted reduction to a singleton [6]). As such, fertility programs should prioritize the use of clinical pathways that maximize the rate of singleton gestation and minimize the rate of multiple gestation [7].

PREGNANCY OUTCOMES

General summary — Both short- and long-term maternal and infant outcomes have been studied and the findings are generally reassuring [8,9]. Much of the risk from ART comes from the increased rates of multiple gestation. Among singleton gestations, IVF has been associated with an increased risk of pregnancy complications such as placenta previa and abruption, gestational diabetes, preeclampsia, and cesarean delivery in nonrandomized studies; however, the absolute increase in risk has generally been small, and most such pregnancies have normal outcomes [10-17]. Thus, most professional organizations advocate for elective single-embryo transfer to minimize the risks of multiple gestation and optimize pregnancy outcomes [18,19]. Delivery of singleton IVF pregnancies during the 39^th week of gestation may be beneficial, particularly for nulliparous individuals, but further study is warranted for pregnancies conceived using ART [20,21]. (See "Induction of labor with oxytocin".)

Factors to consider when assessing ART outcomes include:

●Fresh versus frozen embryo transfer – Transfer of fresh or frozen embryos appears to impact maternal outcomes. Freeze-all cycles, in which all suitable embryos are frozen for transfer at a future time, are associated with reduced risk of ovarian hyperstimulation syndrome but increased risk of maternal hypertensive disorders of pregnancy and large-for-gestational age birth weight [22].

●Impact of multiple gestation – While conception by IVF is associated with an increased incidence of some obstetric and perinatal complications, most risks are related to the increased incidence of multiple gestations [23]. Nevertheless, singleton pregnancies conceived with ART may be at slightly higher risk compared with spontaneously conceived singleton gestations.

•(See 'Multiple gestation incidence and risks' below.)

•(See "Twin pregnancy: Overview".)

•(See "Neonatal complications of multiple births".)

●Impact of advancing age – Additionally, the effects of advanced maternal age need to be considered since the average age of females undergoing IVF is higher and increasing age is associated with increased pregnancy complications. Paternal age has also been associated with impaired outcomes, though to a lesser extent compared with maternal age.

•(See "Effects of advanced maternal age on pregnancy".)

•(See "Effect of advanced paternal age on fertility and pregnancy".)

●Impact of ART versus subfertility – Population level data demonstrate increased adverse obstetric outcomes among IVF pregnancies compared with regularly fertile individuals. In certain instances, the increased perinatal risks are retained even when compared with a subfertile population who becomes pregnant without ART [24]. When compared with subfertile individuals who do not use ART, ART-exposed births have been associated with increased risk of congenital anomalies and preterm birth [25-27].

In a Finnish population analysis that estimated differences in birth outcomes for singleton children conceived with ART compared with spontaneously conceived children both in the general population and within the same family, risk of adverse birth outcomes (lower birth weight and preterm birth) were increased when ART-conceived children were compared with the general population but not when compared with siblings [25]. One potential study limitation is that couples who have both spontaneous and ART-conceived might differ from those who require ART for all children [28].

The precise reasons for this increase in adverse outcomes are not clear, but potential candidates include maternal and paternal characteristics, underlying medical conditions associated with subfertility and infertility, sperm factors, the use of fertility medications, laboratory conditions during embryo culture, culture medium, cryopreservation and thawing, prenatal genetic testing and embryo biopsy, differences in obstetric management, increased proportion of multiple gestations and vanishing twins, or a combination of these factors.

•(See "In vitro fertilization: Overview of clinical issues and questions", section on 'What factors impact IVF success?'.)

Early pregnancy loss — Early spontaneous pregnancy loss is common in pregnancies conceived naturally and with ART, and the rate of loss is similar for both methods of conception [29-31]. The rate of second-trimester loss does not appear to be impacted by ART. Overall, 23 percent of all women will experience at least one loss during their reproductive lifespan (all methods of conception) [32].

●Spontaneous pregnancy loss – As is the case with pregnancy rates, spontaneous pregnancy loss rates are strongly influenced by the maternal/oocyte age, with lesser effects from the underlying cause of infertility [32]. The spontaneous pregnancy loss rate after ART (IVF with or without ICSI) is the same as for the general population of the United States when fresh fertilized embryos are used and adjustments are made for age and multiple gestation [33,34]. (See "Pregnancy loss (miscarriage): Terminology, risk factors, and etiology", section on 'Risk factors'.)

●Impact of ART and preimplantation genetic testing-aneuploidy (PGT-A) on pregnancy loss – The addition of PGT-A with IVF assumes that the pregnancy loss rate may be mitigated by screening out aneuploid embryos prior to transfer. However, there exists ongoing controversy as to whether IVF with PGT-A compared with IVF without PGT-A appreciably decreases pregnancy loss. Ultimately, strong evidence supporting a lower pregnancy loss rate from PGT-A compared with standard methods has not yet been demonstrated.

•Overall impact – A 2020 meta-analysis including 13 trials involving 2794 women concluded that there are insufficient data to demonstrate a difference in miscarriage rates between IVF with PGT-A and without PGT-A [35]. A trial comparing IVF/PGT-A with conventional IVF in subfertile women with three or more good-quality blastocysts reported similar live birth rates for the two groups (77.2 versus 81.8 percent, absolute difference -4.6 percentage points, 95% CI -9.2-0.0) [36].

•Embryo mosaicism – In a study of over 9000 embryos transferred between 2017 and 2023, embryos identified as mosaic at the time of PGT-A had a significantly increased rate of spontaneous abortion compared with euploid embryos (22.2 versus 8.9 percent) [37]. Embryos with mosaicism of whole chromosomes, rather than segmental mosaicism, experienced the highest rates of pregnancy loss (27.6 percent).

•Impact of population – It is suggested that certain groups with a higher baseline aneuploidy rate, such as women >35 years or those with poor ovarian response, may have lower aneuploidy rates with use of PGT-A. A retrospective study of women undergoing IVF with poor ovarian response (four or fewer oocytes per retrieval) demonstrated lower pregnancy loss rates in the PGT-A group versus non-PGT-A (5.9 versus 40 percent) [38].

●Impact of spontaneous reduction of multiple gestations – It has been hypothesized that the slightly increased risk of adverse outcomes in singleton gestations conceived with ART is due to the high proportion of singletons that result from vanishing twins or triplets [39-43]. (See 'Proportion of singleton births' below.)

A study of the 2005 Society for Assisted Reproductive Technology (SART) database including 21,535 singleton deliveries conceived by ART reported 8 percent originated from a twin or triplet gestation [42]. Compared with pregnancies with one fetal heartbeat on early ultrasound examination, early loss of sonographically identified additional heartbeats significantly increased the odds of preterm birth and low birth weight (LBW) for gestational age, and the risk increased in proportion to the total number of fetal heartbeats lost [44]. However, studies have not consistently observed an adverse effect from early spontaneous loss of one twin [45,46]. One possible mechanism may be placental alterations in pregnancies with a vanishing fetus [46]. Increasing use of single embryo transfer reduces the risk of multifetal pregnancy [47] and the subsequent risk of spontaneous pregnancy reduction.

Discussion specific to multifetal pregnancy reduction is presented elsewhere. (See "Multifetal pregnancy reduction and selective termination".)

●Ectopic pregnancy – One to 2 percent of IVF pregnancies will result in an ectopic pregnancy [48,49]. Ectopic risk among ART pregnancies varies most according to reproductive health characteristics of the female carrying the pregnancy (eg, higher in those with tubal factor infertility or smoking status) [48]. Factors associated with lower risk of ectopic pregnancy include frozen embryo transfer and blastocyst stage transfer [49,50].

•Frozen embryo transfer – A retrospective cohort analysis utilizing the SART database reported a lower ectopic pregnancy rate with frozen embryo transfer compared with fresh (odds ratio [OR] 0.35, 95% CI 0.29-0.42) [49].

•Blastocyst stage transfer – Blastocyst stage transfer is associated with a 67 percent lower risk of ectopic pregnancy compared with a day 3 cleavage stage transfer (relative risk [RR] 0.67, 95% CI 0.54-0.87) [50]. (See "Ectopic pregnancy: Epidemiology, risk factors, and anatomic sites", section on 'Infertility and related factors'.)

Heterotopic pregnancy is far more common in pregnancies conceived by ART than spontaneous conceptions (1 of 100 versus 1 of 30,000). The increase in risk parallels the number of embryos transferred. (See "Ectopic pregnancy: Epidemiology, risk factors, and anatomic sites", section on 'Heterotopic pregnancy'.)

Multiple gestation incidence and risks

Proportion of singleton births — As ART technology has evolved, increasing implantation rates have resulted in increased overall ART birth rates as well as increased rates of multiple gestation. The rates of singleton and multiple births following ART vary around the world [5,51,52]. While the overall rates of multiple births are higher for ART births than natural births, the percent of multiple births from ART has been declining since 2009 because fewer embryos are being transferred per cycle and centers are increasing the use of elective single embryo transfer. In the United States in 2016, 32 percent of ART infants were part of a multiple birth compared with 3.4 percent of naturally conceived infants [52]. By 2019, more than 90 percent of all pregnancies from ART in the United States were singletons [53].

Strategies to reduce the risks of multiple gestation and multifetal pregnancy reduction are discussed separately.

●(See "Strategies to control the rate of high order multiple gestation".)

●(See "Multifetal pregnancy reduction and selective termination".)

Twin pregnancy incidence and outcomes

●Incidence – The increased rate of twin gestation with ART is a result of the number of embryos transferred by clinicians and not a function of ART itself. In countries with strict rules for transferring only one embryo, the twinning rate is lower than in countries without such policies. As examples, the rate of twins following IVF is approximately 7 percent for Australian and New Zealand (single embryo transfer rate of 68 percent) compared with more than 25 percent for North America and the Middle East (single embryo transfer rates of 19 and 11 percent, respectively) [54].

Internationally, there is a wide range in the twin delivery rate with IVF using fresh or frozen embryos, respectively, ranging from 5.8 and 4.7 percent in Sweden, 23.6 and 16 percent in Spain, 27 and 35.8 percent in Greece, 28.6 and 15 percent in Germany, and 35.8 and 21.3 percent in Bulgaria in 2011 [55].

●Comparison of outcomes for ART and spontaneous twin pregnancies – Evidence is mixed as to whether twins conceived by IVF have an equivalent or higher rate of adverse obstetric and perinatal outcomes [39,56]. One challenge is the choice of comparator population as individuals who undergo ART may differ from the general population and the procedures themselves may impact outcome. Examples include:

•General population comparison - A longitudinal cohort study including 10,352 individuals with twin pregnancies based in Massachusetts demonstrated an increased rate of adverse pregnancy outcomes for subfertile individuals and individuals with IVF pregnancies compared to fertile controls [57].

•ART does not impact preterm birth rate in twin gestations – The lack of a strong association between IVF and preterm birth in twins in multiple studies [39,58-61] may be due to confounding factors, such as the much higher proportion of monozygotic twins (which are at increased risk of preterm birth and LBW) among spontaneously conceived twin gestations than among IVF twin gestations [62]. It is possible that the profound effect that twinning has on pregnancy outcome overshadows any additional effect that IVF may exert. (See "Spontaneous preterm birth: Overview of risk factors and prognosis", section on 'Multifetal gestation in the current pregnancy'.)

Monozygotic multiples — Although most multiple gestations are multi-zygotic and result from transfer of multiple embryos, the frequency of monozygotic multiples is also increased with ART. The baseline risk of spontaneously conceived monozygotic twinning is 0.4 percent. By comparison, a retrospective study reported a monozygotic rate of 2.3 percent with frozen embryo transfer compared with 3.1 percent for fresh embryo transfers after IVF [63]. Although the exact mechanism is not known, the increase in monozygotic twinning has been attributed to the in vitro culture environment, extended duration of culture (ie, day 5 to 6 embryos), and use of assisted hatching [64,65].

Additional information on zygosity can be found elsewhere. (See "Neonatal complications of multiple births", section on 'Terminology'.)

Low birth weight

Impact of gestation type — Use of ART appears to increase the risks of LBW, even for singleton gestations. (See "Fetal growth restriction (FGR) and small for gestational age (SGA) newborns".)

●Singleton pregnancies – Singleton IVF pregnancies, with or without ICSI, are at higher risk of preterm birth and LBW (≤2500 g) compared with spontaneously conceived pregnancies (table 1) [10,24,26,39,58,59,66-73]. Two large studies by the same group, whose findings are similar to those generally reported in the literature, illustrate these risks:

•Initial study – A population-based study including over 42,000 infants conceived with ART from 1996 to 1997 and 3 million births in the general population reported the following [39]:

-Term LBW – LBW was significantly more common among term singleton, but not twin, infants conceived by ART compared with the general population (for singletons 6.5 versus 2.5 percent, RR 2.6, 95% CI 2.4-2.7).

-Preterm LBW – Preterm LBW was significantly more common among singleton, but not twin, infants conceived by ART (for singletons 6.6 versus 4.7 percent, RR 1.4, 95% CI 1.3-1.5).

The increased risk persisted after adjustment for maternal age and parity, gestational age at delivery, multifetal reduction procedures, and cause of infertility. These data may also be interpreted as supporting the contention that an increase in adverse outcomes among singletons is due to the "vanishing twin" phenomenon described above. (See 'Early pregnancy loss' above.)

•Follow-up data – An extended follow-up study included over 62,000 singleton infants from ART procedures from 1996 to 2000 [74]. Over this interval, there was a decline in LBW deliveries but no significant change in the rate of preterm birth. Although the frequency of LBW declined, it was still higher than the expected rate adjusted for age, parity, and race/ethnicity. Clear reasons for the fall in LBW were not identified.

●Multiple gestation pregnancies – The risks of extremely LBW, very preterm birth, newborn intensive care unit (NICU) admission, and perinatal mortality are increased among ART-conceived dichorionic twins compared with spontaneously conceived twins. Overall, the maternal outcomes of ART- and spontaneously conceived dichorionic twins are similar [75].

Reduced risk with frozen embryo transfer — Compared with fresh embryo transfer, frozen embryo transfer appears to be associated with a decreased risk of small for gestational age (SGA) and LBW births as well decreased risk of preterm birth [76-80]. This suggests that the more natural endometrial preparation prior to frozen embryo transfer plays a role in these parameters, possibly by allowing for more natural placentation than that which occurs in stimulated cycles. In one of the studies [77], incidence of SGA, LBW, and preterm births was nevertheless higher in frozen embryo transfer singletons than in naturally conceived singletons. In frozen blastocyst cycles resulting in a live birth, trophectoderm biopsy has not been associated with differences in the birth weight at delivery [81,82].

Potential mechanisms — There are several potential explanations for the difference in birth weight between ART and naturally conceived pregnancies:

●Impact of ART procedure – It could be related to the ART procedure (drugs, manipulation of gametes and embryos, culture, effect of ovarian stimulation on endometrial receptivity), although a physiological mechanism explaining how the procedures and drugs used in ART increase the risk of LBW has not been determined.

●Treatment bias – Treatment biases likely exist in studies of IVF pregnancies as these pregnancies can be associated with increased medical interventions as well as generally higher maternal ages, lower parity, and higher socioeconomic status. As a result, IVF pregnancies are likely to undergo more intense monitoring and more frequent intervention, such as a higher rate of elective cesarean delivery (66.7 versus 33.3 percent in one study of 162 pregnancies) [83]. However, very LBW (<1500 g) is unlikely to be the result of elective preterm delivery, and this outcome is also more common among singletons conceived through ART [10,39,58].

●Impact of subfertility – Subfertility appears to have an adverse effect on pregnancy outcome, independent of its treatment. Several studies have reported that those with untreated subfertility who became pregnant had a greater frequency of adverse outcomes than the general population [24,84-88], and their frequency of complications was similar to that in subfertile individuals who underwent ART [89]. All of the studies were observational and many potential confounders were not considered in the analyses.

Additional support for this hypothesis comes from two population-based cohort studies. The first compared the pregnancy outcome of multiparous individuals who underwent ART with the pregnancy outcome of (1) the same women in a previous or subsequent naturally conceived pregnancy, and (2) the general obstetric population [90]. Multiparous women who underwent ART had infants of similar gestational age and birth weight in pregnancies before and after the procedure, but their infants delivered earlier and had lower birth weights than the general obstetric population. A similar population-based cohort study that also compared siblings conceived either spontaneously or through IVF reported that the maternal characteristic of subfertility was associated with lower birth weight, but the IVF procedure itself was not [91].

Hypertensive disorders — Use of ART has been associated with a nearly 50 percent increased risk of hypertensive disorders of pregnancy, including preeclampsia [56].

●Overall risk and risk factors – A meta-analysis of 15 studies comparing 12,923 IVF/ICSI singleton pregnancies with pregnancies from unassisted conception reported a nearly 50 percent increased risk of hypertensive disorders of pregnancy with IVF/ICSI (risk ratio 1.49, 95% CI 1.39-1.59) [56]. Frozen embryo transfer appears to increase risk compared with unassisted conception (ie, natural) and fresh embryo transfer [92]. In a study comparing outcomes of fresh and frozen embryo transfer cycles with unassisted conception, frozen embryo transfer was associated with increased risk of hypertensive disorders of pregnancy and preeclampsia in both population-level and within-sibling estimates while fresh embryo was not (with the exception of a possible small increased risk of preeclampsia with fresh embryo transfer at the population level only) [93].

●Protective effect of corpus luteum – ART cycles that retain a corpus luteum appear to be somewhat protected from hypertensive disorders of pregnancy. Studies of frozen embryo transfers comparing natural cycle frozen embryo transfer (with a corpus luteum) with traditional medicated frozen embryo transfer (ie, programmed, no corpus luteum) have reported lower rates of preeclampsia and preeclampsia with severe features with natural cycle frozen embryo transfer [94,95]. A randomized controlled trial is underway to address this observation [96].

●Impact of donor oocytes – Use of donor oocytes is associated with increased rates of preeclampsia and hypertensive disorders of pregnancy. Two- to four-fold increased risks of hypertensive disorders have been reported [97,98]. The increased risk may result from donor-oocyte pregnancies being a different immunological challenge compared with autologous oocyte pregnancies.

●Impact of endometrial thickness – A thin endometrium (<7.5 mm) prior to embryo transfer has been associated with an increased incidence of hypertensive disorders of pregnancy, decreased birthweight, and small for gestational age infants [99]. A thick endometrium (>14 mm) has not been associated with adverse outcomes.    

Limitations of the above studies include lack of information on the ART protocols and inability to adjust for biologic factors associated with infertility.

Spontaneous preterm birth — The risk of spontaneous preterm birth appears to be elevated for ART pregnancies compared with those conceived spontaneously but the magnitude of risk is impacted by the control population selected for comparison (eg, individuals without infertility or with subfertility) (table 1) [10,24,100,101]. Changes in obstetric care and reductions in multifetal gestation have likely contributed to the decrease of risk documented over time.

●Data from 2008 and beyond – A study comparing outcomes of US pregnancies conceived between 2008 and 2016 with ART (n = 106,248) or spontaneously (n = 34,167,246) reported a 26 percent increased risk of preterm birth after controlling for multiple risk factors, including maternal age and demographics, multiple gestation, and medical risk factors (odds ratio 1.26, 95% CI 1.20-1.32) [101]. The study did not differentiate between spontaneous and medically indicated preterm birth nor did it control for the presence of subfertility [24].

●Studies prior to 2002 – Meta-analyses comprising several thousand IVF and approximately two million naturally conceived singleton births matched for maternal age and parity found IVF pregnancies had nearly double the risk of preterm birth (OR 1.95-1.98) (table 1) [10,69]. The increased risk remained when only preterm births after spontaneous labor were considered [10,100]. Although there was significant heterogeneity among studies, subgroup analysis of studies with similar designs consistently yielded ORs similar to the overall summary OR.

●Impact of PGT-A/trophectoderm biopsy – A retrospective study comparing outcomes of frozen embryo transfers that resulted in singleton births reported a 20 percent increased rate of preterm birth for embryos that underwent PGT versus those that did not (adjusted odds ratio [aOR] 1.20, 95% CI 1.09-1.33) [102].

Additional discussions of preterm birth are presented elsewhere.

●(See "Spontaneous preterm birth: Pathogenesis".)

●(See "Spontaneous preterm birth: Overview of risk factors and prognosis".)

Abnormal placenta development and placentation

●Incidence – Placental disorders, including placenta previa and placenta accreta spectrum (PAS), as well as placental pathology appear to be increased in pregnancies conceived with IVF.

•Abnormal placenta pathology – A study comparing placental pathology from term singleton live births conceived with fresh embryo transfer (n = 511) with those conceived without ART (n = 121) reported increased risk of specific anatomic and vascular abnormalities in the ART-conceived placentas [103]. The difference was mainly driven by pregnancies conceived with intracytoplasmic sperm injection (ICSI). At least one study reported similar number of abnormal placental lesions between placentas resulting from fresh or frozen embryo transfer.

•Placenta accreta spectrum – In a retrospective cohort study of over 28,000 births between 2013 and 2018 in a Massachusetts tertiary care hospital reported higher incidence of PAS for IVF compared with non-IVF pregnancies (2.2 versus 0.3 percent, respectively) despite having had fewer prior cesarean deliveries (22.6 versus 64.2 percent, respectively) [104]. After adjusting for maternal age, nulliparity, and year of delivery, IVF pregnancies had more than five times the risk of PAS compared with non-IVF pregnancies (RR 5.5, 95% CI 3.4-8.7). (See "Placenta accreta spectrum: Clinical features, diagnosis, and potential consequences".)

●Impact of frozen embryo transfer – There is a suggestion that frozen embryo transfer cycles are associated with decreased risk of placenta previa and abruptio placentae, suggesting that the endometrial environment at the time of implantation plays a role in the pathogenesis of these complications [76]. Absence of a corpus luteum cyst in these cycles appears to be a possible underlying mechanism [105,106]. (See "Placenta previa: Epidemiology, clinical features, diagnosis, morbidity and mortality".)

●Impact of multiple gestation on placentation – In a meta-analysis of 15 cohort studies comparing ART-conceived with naturally conceived dichorionic twins, women with ART dichorionic twins had nearly three times the risk of placenta previa as women with naturally conceived dichorionic twins (RR 2.99, 95% CI 1.5-5.9) [107]. When compared with the non-ART dichorionic twins, the ART neonates had an increased risk of preterm birth (RR 1.13, 95% CI 1.00-1.29), very preterm birth (RR 1.39, 95% CI 1.07-1.82), LBW (RR 1.11, 95% CI 1.00-1.23), and congenital malformations (RR 1.26, 95% CI 1.09-1.46). However, it is difficult to determine if the ART is the cause of increased risk of obstetric complications given all the variables involved. (See 'Pregnancy outcomes' above.)

Gestational diabetes — The risk of gestational diabetes is increased in singleton ART pregnancies, even after excluding patients with possible confounders such as polycystic ovary syndrome (PCOS) [108]. A meta-analysis including 13,399 ART patients reported a nearly 50 percent increased relative risk of gestational diabetes (RR 1.48, 95% CI 1.33-1.66) [56]. While some studies were not able to control for potential confounders of gestational diabetes, the data are highly suggestive of an increased risk among ART pregnancies.

MATERNAL OUTCOMES

Maternal mortality — Risk of maternal mortality rises with increasing maternal age. As age-related infertility is a common indication for patients using ART, this risk is discussed as part of the counseling process. More information is available in related content. (See "Effects of advanced maternal age on pregnancy", section on 'Maternal mortality'.)

Severe maternal morbidity — Use of ART, and specifically IVF, appear to increase the risk of severe maternal morbidity compared with spontaneously conceived pregnancies. Subfertile and IVF-treated females tended to be older and have more preexisting medical conditions, both of which increase the risk of maternal morbidity [73,101]. Data limitations include the use of cohort and retrospective data. Discussions of severe maternal morbidity are reviewed separately. (See "Severe maternal morbidity".)

By type of fertility treatment

•ART – In a retrospective cohort study of over one million US deliveries between 2008 and 2012, severe maternal morbidity was nearly twice as likely in pregnancies conceived with ART compared with non-ART pregnancies after controlling for maternal age, parity, comorbid conditions, history of prior cesarean delivery, and year of delivery [109]. When the same data set was analyzed for antenatal hospitalization, individuals with ART pregnancies had an increased risk of antenatal admission and longer hospitalizations compared with non-ART pregnancies [110].

•IVF – In a cohort study that used propensity score matching and controlled for multiple factors, including maternal age and multiple gestation, IVF was associated with a nearly 40 percent increased risk of severe maternal morbidity compared with spontaneous conception; the absolute risk of severe maternal morbidity was low (30.8 [IVF] versus 22.2 [spontaneous] per 1000 births) [111]. Postpartum hemorrhage, ICU admission, and sepsis were the most common indicators of severe morbidity. Less invasive fertility therapy, such as ovulation induction or intrauterine insemination, was not associated with an increased risk of either morbidity or mortality.

Stroke — Observational studies have suggested that using assisted reproductive technology is associated with an increased risk of postpartum stroke compared with natural conception [112-115]. The etiology of this association is unknown, and may reflect the underlying characteristics of patients with infertility (eg, increased likelihood of advanced age, comorbid conditions, and elevated body mass index) and the increasing use of frozen embryo transfer (FET) during IVF cycles (FET is associated with increased risk of hypertensive disorders of pregnancy, which are associated with increased risk of stroke [116-118]). Reassuringly, the absolute risk of stroke remains low.

●During delivery hospitalization – In a different retrospective review of over 19 million delivery hospitalizations, including over 200,000 pregnancies from ART, reported stroke rates were 27.0 versus 9.1 per 100,000 persons for those with and without prior ART, respectively [115].

●Up to 12 months from delivery – In an observational study of over 31 million US pregnant persons, the rate of hospitalization for stroke within 12 months of delivery was 37 per 100,000 people following assisted reproductive technology compared with 29 per 100,000 people after natural conception (adjusted hazard ratio 1.66, 95% CI 1.17-2.35) [113].  

Venous thrombosis — Individuals who conceive after IVF have been observed to have an increased risk of pulmonary and venous thromboembolism during pregnancy, especially during the first trimester, even in the absence of an overt diagnosis of ovarian hyperstimulation syndrome (OHSS) [119-121]. In one study, for example, the risk of pulmonary embolism in the first trimester in individuals after IVF versus in individuals with natural pregnancies was 3 in 10,000 versus 0.4 in 10,000 (HR 6.97, 95% CI 2.21-21.96) [119]. These findings do not warrant a change in current practice; the results need to be confirmed in large, carefully designed studies and the benefits of anticoagulation need to be balanced with the risks in this setting. Although the absolute risk of a serious thromboembolic event in the first trimester is low, it may be prudent to make patients aware of the symptoms of thromboembolism and inform them to call their provider if symptoms develop.

●(See "Overview of the causes of venous thrombosis".)

●(See "Deep vein thrombosis in pregnancy: Clinical presentation and diagnosis".)

Risk of cancer — The use of fertility drugs and IVF does not appear to increase the risk of all cancers or breast, cervical, or ovarian cancer. However, definitive conclusions are limited by the observational and, often, retrospective nature of the data [122,123].

Ovarian cancer and borderline tumors — At present, there is no evidence that ART increases the risk of ovarian cancer beyond the baseline increase that is associated with infertility.

●Impact of infertility versus fertility therapy – The use of fertility drugs has been associated with ovarian neoplasia, but experts have generally concluded that infertility, by reducing the number of successful pregnancies, which in turn increases the number of times a patient ovulates, is an important risk factor for ovarian cancer and that infertility treatment does not independently increase the risk. The available literature on ovarian cancer risk associated with fertility drug treatment is reassuring but not definitive [123]. (See "Overview of ovulation induction", section on 'Cancer risks'.)

Whether IVF increases the risk of ovarian malignancy because of repeated ovarian punctures for egg retrieval or some other mechanism remains controversial, but data are generally reassuring [122,124-126]. In our practice, we counsel patients that there are several known factors which predispose individuals to the development of ovarian malignancies: infertility, the presence of endometriosis, and lack of childbearing, all of which are either an inherent part of the diagnosis of infertility or are associated with the diagnosis of infertility. The studies that have linked ovarian cancer to IVF or other fertility treatment have generally found that when these predisposing conditions are considered, that the treatment associated increase in risk of ovarian cancer disappears. As examples:

●Ovarian cancer – While the risk of ovarian cancer is increased in individuals undergoing ART compared with the general population, this increase is largely attributable to the diagnosis of infertility.

Examples of supporting data include:

•A 2020 cohort study comparing over 30,000 individuals who received ovarian stimulation for ART with nearly 10,000 subfertile individuals not treated with ART reported that ovarian cancer risk in the ART group was elevated compared with the general population (standardized incidence ratio 1.43, 95% CI 1.18-1.71) but similar when compared with the non-ART subfertile group (adjusted hazard ratio [HR] 1.02, 95% CI 0.70-1.50) [127]. The median follow-up duration was 24 years. Ovarian cancer risk decreased with increasing parity and with larger number of successful ART cycles.

•A 2019 meta-analysis reported an increased risk of invasive ovarian cancer for individuals exposed to clomiphene citrate compared with unexposed individuals and for nulliparous individuals using clomiphene citrate compared with multiparous individuals who used the drug, but the data were synthesized from cohort and case-control studies and felt to be of low quality, which limited definitive conclusions [123].

•A 2018 population-based cohort study that included over 255,000 individuals undergoing IVF reported an increased risk of ovarian cancer (standardized incidence ratio 1.39, 95% CI 1.25-1.53), but the absolute increased risk remained low (5 cases per 100,000 person years) [128].

Detailed discussions of ovarian cancer are available separately.

•(See "Overview of epithelial carcinoma of the ovary, fallopian tube, and peritoneum".)

•(See "Epithelial carcinoma of the ovary, fallopian tube, and peritoneum: Clinical features and diagnosis".)

●Borderline ovarian tumors – Observational studies suggest the risk of borderline ovarian tumors may be increased in individuals treated with ovarian-stimulating drugs (clomiphene, gonadotropin) for infertility, including the subgroup of these individuals undergoing IVF, but more study is needed and the influence of confounding factors needs to be determined before a causal relationship can be inferred [123,127,129]. In the above 2020 cohort study, the risk of borderline ovarian tumors was increased for individuals undergoing ART compared with the general population and the non-ART subfertility group [127]. However, no dose-response relationship was identified with increasing number of ART cycles, which raised concerns for the finding's accuracy. (See "Borderline ovarian tumors".)

Breast cancer — ART does not appear to increase the long-term risk of breast cancer, even in individuals with BRCA 1 and 2 mutations [130,131]. Thus, patients with a family history of breast cancer or genetic risk factor for breast cancer are not excluded from ART.

●Contemporary drug regimens – A meta-analysis of studies published between 1990 and 2020 including new breast cancer diagnoses in patients undergoing fertility therapy found no increased risk of breast cancer for individuals treated with ovarian stimulation drugs compared with unexposed individuals from the general population or those with infertility [131]. Study advantages included a mean follow-up duration of 27.1 years and use of contemporary pharmacologic regimens for ovarian stimulation. Study limitations included use of observational data with high heterogeneity and limited ability to adjust for confounders such as obesity. Despite these limitations, the overall findings are reassuring. Specific analyses by drug type included:

•Clomiphene citrate – For infertile individuals exposed to clomiphene citrate, the risk of breast cancer was not increased compared with unexposed individuals in the general population (pooled odds ratio [OR] 1.03, 95% CI 0.72-1.48, 10 studies) or compared with unexposed infertile individuals (pooled OR 1.04, 95% CI 0.70-1.54, nine studies).

•Gonadotrophins – For infertile individuals exposed to gonadotropins, the risk of breast cancer was not increased compared with unexposed individuals in the general population (pooled OR 1.07, 95% CI 0.78-1.46, seven studies) or compared with unexposed infertile patients (pooled OR 1.13, 95% CI 0.75-1.71, seven studies).

●Historical drug regimens – In a retrospective study assessing the risk of breast cancer in over 19,000 individuals undergoing IVF between 1983 and 1995 who were followed for a median of 21 years, there were no differences in breast cancer risk among individuals treated with IVF compared with individuals in the general population or compared with individuals undergoing other fertility treatments (standardized incidence ratio 1.01, 95% CI 0.93-1.09 and HR 1.01, 95% CI 0.86-1.19, respectively) [132]. In addition, there were no differences in the incidence of breast cancer at age 55 between the IVF and other fertility treatment groups. This study is consistent with prior reviews that reported no increased risk of breast cancer [133-135] but expands on those reviews because of the much longer period of follow-up (mean duration of follow-up 21 years). Limitations of the data include that IVF protocols have evolved to use reduced amounts of gonadotropins, include regimens with gonadotropin-releasing hormone (GnRH) antagonists, and typically require fewer cycles. Therefore, it is not known how well these study results apply to contemporary IVF treatment. Additionally, there are limited data on individuals age >60 years. Despite these limitations, the results are encouraging that there is no large increase in breast cancer risk for individuals treated with IVF.

●Consideration of confounders – While reproductive hormones appear to have a relationship with breast cancer risk, this relationship is complicated because of multiple confounders. As an example, individuals in the above cohort study who underwent seven or more cycles of IVF had a significantly lower risk of breast cancer compared with individuals who underwent one to two cycles [132]. Possible explanations include that these individuals received more hCG (which may exert a protective effect [136,137]), had longer periods of down-regulation of hormone levels, or had some inherent characteristic that required more IVF cycles. The potential effects of endogenous and exogenous reproductive hormones, pregnancy and breastfeeding, and infertility and infertility therapy on breast cancer risk are presented separately. (See "Factors that modify breast cancer risk in women" and "Overview of ovulation induction", section on 'Cancer risks'.)

Long term risk of cardiovascular disease — While the risk of thromboembolism may be increased in pregnancy, the risk of long-term cardiovascular disease after IVF treatment does not appear to be increased. A meta-analysis of six observational studies including over 41,000 individuals reported no increased risk of a cardiac event among individuals who received fertility therapy and those who did not [114]. However, the small number of studies and significant heterogeneity of study designs and outcomes limits definitive conclusions.

●(See "Overview of atherosclerotic cardiovascular risk factors in females".)

●(See "Clinical features and diagnosis of coronary heart disease in women".)

RESOURCES FOR PATIENTS AND CLINICIANS

●European Society of Human Reproduction (ESHRE)

●Society for Assisted Reproductive Technology (SART) has information for both patients and providers, including an ART success predictor tool.

●Centers for Disease Control and Prevention

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Female infertility".)

SUMMARY AND RECOMMENDATIONS

●Summary – The best pregnancy outcomes following in vitro fertilization (IVF)/assisted reproductive technology (ART) occur with singleton pregnancies. ART programs should prioritize the use of clinical pathways that maximize the rate of singleton pregnancy and minimize the rate of multiple gestation. (See 'Importance of singleton gestation' above.)

●Terminology and background – ART includes all interventions that involve the in vitro handling of both human oocytes and sperm, or embryos, for the purpose of reproduction. ARt has resulted in more than eight million pregnancies worldwide. (See 'Background' above.)

●Pregnancy outcomes – Both short- and long-term maternal and infant outcomes after ART have been studied and the findings are generally reassuring; singleton pregnancies are associated with the lowest risks. Confounding factors include the impact of multiple gestation, advancing parental age, infertility itself, and fertility treatments. (See 'General summary' above.)

•Singleton gestation – For singleton pregnancies, conception with ART is associated with increased risk of low birth weight (LBW), spontaneous preterm birth, and preeclampsia, although these risks are not clearly independent of infertility itself. Frozen embryo transfer, rather than fresh embryo transfer, is associated with reduced risks of low birth weight and spontaneous preterm birth, although the magnitude of this difference and causation are not yet clear. (See 'Pregnancy outcomes' above.)

•Multiple gestation – For multiple gestation, factors related to pregnancy with multiple fetuses rather than factors related to infertility or ART are the principal cause of obstetric complications. (See 'Multiple gestation incidence and risks' above.)

●Maternal outcomes – The maternal morbidity of individuals achieving an ART pregnancy is generally comparable to naturally conceived pregnancies. There is a small increased risk of thromboembolic events and hypertensive disorders or pregnancy but not cardiovascular disease. The evidence is reassuring that rates of cancer are not elevated beyond the baseline risk for other infertile females. (See 'Maternal outcomes' above.)

ACKNOWLEDGMENT — The UpToDate editorial staff acknowledges Richard Paulson, MD, MS, who contributed to earlier versions of this topic review.