Adverse pregnancy outcomes associated with first‐trimester exposure to angiotensin‐converting enzyme inhibitors or angiotensin II receptor blockers: A systematic review and meta‐analysis

Abstract This study aimed to determine the effects of prenatal exposure to angiotensin‐converting enzyme inhibitors (ACEIs) or angiotensin II receptor blockers (ARBs), particularly when exposure is limited to the first trimester of pregnancy, on adverse maternal and neonatal outcomes. A systematic search was performed on four databases, that is, PubMed, Scopus, Web of Science, and Cochrane Library, to identify relevant articles published up to December 31, 2019. Included studies were limited to original investigations assessing the association between prenatal exposure to ACEIs/ARBs and adverse pregnancy outcomes. Odds ratios were used as a summary effect measure. Pooled‐effect estimates of each outcome were calculated by the random‐effects meta‐analysis. The main outcomes included overall and specific congenital malformations, low birth weight, miscarriage, elective termination of pregnancy, stillbirth, and preterm delivery. Of 19 included articles involving a total of 4 163 753 pregnant women, 13 studies reported an increased risk of, at least, one adverse pregnancy outcome in pregnant women who were exposed to ACEIs/ARBs. Meta‐analysis revealed a significant association between overall congenital malformations and first trimester‐only exposure to ACEIs/ARBs (OR = 1.94, 95% CI = 1.71‐2.21, P < .0001). Cardiovascular malformations, miscarriage, and stillbirth also provided a significant relation with ACEI/ARB exposure. In conclusion, prenatal exposure to ACEIs/ARBs in the first trimester of pregnancy was found to be associated with an increased risk of adverse pregnancy outcomes. Women of reproductive age should be aware of the potential teratogenic risks of these drugs if they become pregnant.


| INTRODUC TI ON
Angiotensin-converting enzyme inhibitors (ACEIs) and angiotensin II receptor blockers (ARBs), an alternative for ACEI-intolerant patients, are commonly used for the treatment of cardiovascular disease whereby the renin-angiotensin-aldosterone system (RAAS) is involved in its pathophysiology. 1,2 ACEIs/ARBs modulate the RAAS by either inhibiting an enzyme responsible for the conversion of angiotensin I to angiotensin II or by antagonizing the effects of angiotensin II at its receptors. As such, ACEIs/ARBs are beneficial to enhanced natriuresis, reduced afterload, and deferral of cardiovascular remodeling, making them useful for various cardiovascular conditions, such as hypertension, heart failure, and postmyocardial infarction. [3][4][5] ACEIs/ARBs are, thus, one of the most widely prescribed drug classes, with hundreds of thousands of patients worldwide who are exposed to each year. 6,7 In 1980s-1990s, there were a series of cases reported to the US Food and Drug Administration (FDA) indicating that ACEIs/ARBs are teratogens when being used in the second and third trimesters of pregnancy. 8 Evidence suggests a relationship between neonatal adverse outcomes (ie, oligohydramnios and other adverse outcomes secondary to impaired fetal kidney development) and ACEI/ARB exposure. 9,10 A "black box" warning issued by the US FDA in 1992 has raised awareness of the teratogenic potential of ACEIs/ARBs, and the second and third trimesters of pregnancy are considered a contraindication to the use of ACEIs/ARBs accordingly. 11 In the 2013 Report of the American College of Obstetricians and Gynecologists' Task Force on Hypertension in Pregnancy, it has been recommended not to use ACEIs/ARBs in women of reproductive age if there is no compelling reason. 12 Despite such a feature in the labeling of the potentially teratogenic medications, several cases of fetal exposure to ACEIs/ARBs have been reported thereafter. [13][14][15] ACEI/ARB exposure during pregnancy is still highly prevalent in many settings. 16,17 Up to the present time, it still remains unclear whether ACEIs/ ARBs are teratogenic if exposure to these drugs is only limited to the first trimester of pregnancy. [18][19][20] Several epidemiologic studies report inconsistent results on the teratogenic effects of first-trimester ACEI/ARB exposure in humans. 17,21,22 Given the increasing incidence of hypertension and conditions in which ACEIs/ARBs are often indicated, systematic investigations on the potential teratogenic consequences of ACEI/ARB exposure during early pregnancy are highly needed to provide more concrete guidance for the use of ACEIs/ARBs in women of reproductive age. 23,24 Should ACEI/ARB exposure during the first trimester of pregnancy be considered nonteratogenic, female patients of childbearing potential could be safely prescribed either an ACEI or an ARB, provided that they are advised of the risks involved and can switch the drug to other alternatives within a few weeks after conception. On the other hand, if exposure to ACEIs/ARBs during the first trimester of pregnancy is associated with an increased risk of congenital malformations or adverse maternal outcomes, the use of ACEIs/ARBs in women of reproductive age should be discouraged, particularly given the availability of alternative medications to treat their conditions. 25 The objective of the present study was to determine the effects of prenatal exposure to ACEIs/ARBs, particularly when exposure is limited to the first trimester of pregnancy, on adverse maternal and neonatal birth outcomes by means of systematic review and meta-analysis.

| ME THODS
This study conformed to the preferred reporting items for systematic reviews and meta-analyses (PRISMA) guidelines. 26 The study protocol was prospectively registered at the PROSPERO international prospective register of systemic reviews in health and social care (CRD42019140107).  Relevant studies were selected based on the following criteria:

| Search strategy and eligibility criteria
(a) a study involved pregnant women; (b) there was ACEI/ARB exposure during pregnancy; and (c) either adverse maternal outcomes or neonatal birth outcomes, or both, were reported. Included studies were limited to original investigation performed on humans. No restriction was made with respect to study design or subjects' underlying conditions. Studies lacking a control group (eg, case reports, case series, or expert opinion) and review articles (including systematic reviews) were excluded. All studies deemed suitable were retrieved and reviewed independently by two authors to determine study eligibility. Study selection was carried out by two authors independently; disagreements were resolved through discussion and consensus.

| Data extraction and quality assessment
Two authors independently extracted data from original full-text articles using a standardized data collection form. The data extracted included (a) first author, (b) publication year, (c) study design, (d) study setting/location, (e) study period, (f) stage of pregnancy, (g) number of participants, (h) exposure (ie, ACEIs or ARBs), (i) control (ie, exposure to other antihypertensive drugs or nonexposure), and (j) outcome of interest (ie, overall and specific congenital malformations, low birth weight (LBW) (birth weight < 2500 g), miscarriage or spontaneous abortion, elective termination of pregnancy (ETOP), stillbirth, and preterm delivery). In cases where data were missing in an original publication or required clarification, attempts were made by e-mail contact with the corresponding author.
The quality of included studies was evaluated using the standardized Good Research for Comparative Effectiveness (GRACE) checklist for observational studies and a revised tool for assessing the risk of bias in randomized trials (RoB 2) for randomized-controlled trials. 27,28

| Data analyses
The exposure of interest was maternal exposure to ACEIs/ARBs during any trimesters of pregnancy or during the first trimester only, and the outcome of interest was adverse pregnancy outcomes, including both maternal and neonatal outcomes. The first trimesteronly exposure was defined as any use of ACEIs/ARBs from the last menstrual period to the third month of pregnancy. An exposure cohort was defined as a group of pregnant women who were exposed to ACEIs/ARBs (ACEI/ARB group), while a control cohort was those who were exposed to other antihypertensive medications (OAH group) or those with no exposure to any antihypertensive drugs (nonexposure group). Extracted relevant data were tabulated in a 2 × 2 contingency table. Odds ratios (ORs) were used as a summary measure for meta-analysis of dichotomous outcomes. Risk ratios (RRs) were calculated from ORs using the following formula for ease of interpretation: RR = OR/ [1 -ACR × (1 -OR)]; given that the assumed comparator risk (ACR) is the risk that the outcome of interest occurred in the control group.
Pooled-effect estimates of each outcome of interest were calculated by the Mantel-Haenszel random-effects meta-analysis.
A cumulative meta-analysis was conducted to determine whether each study added to the pool affected the overall estimate changes.
Statistical heterogeneity among included studies was assessed using the Cochran's Q test and the percentage of total variability across studies due to heterogeneity (I 2 value). Subgroup analyses were conducted to determine the impact of ACEI/ARB exposure on adverse maternal and neonatal birth outcomes when exposure was limited to the first trimester of pregnancy.
Potential bias from small-study effects (eg, publication bias) was assessed through visual examination of funnel plots displaying the log OR of individual studies on the horizontal axis and its standard error on the vertical axis. 29 A rank correlation test and a linear regression test were applied to identify any potential publication bias in a meta-analysis with 10 or more included studies. 30,31 Sensitivity analyses on the impact of study design, drug classes, and exclusion of a single study from meta-analysis as well as the impact of fixed-effect or random-effects models on summary measures were performed.
All tests were two-tailed; P < .05 was considered statistically

| RE SULTS
Of 3427 potentially relevant records identified through the sys-  (Table S1). Relevant studies were conducted in North America (n = 9), Europe (n = 9), or Australia (n = 2). Data syntheses involved a total of 4 163 753 pregnant women, with 7075 exposed to ACEIs/ARBs, 25 379 to other antihypertensive drugs, and 3 782 450 nonexposed individuals. Around two thirds of studies included in qualitative analysis (13/19) reported an increased risk of, at least, one adverse pregnancy outcome of interest in pregnant women with ACEI/ARB exposure (Table S2).
Meta-analysis of 17 included studies found a significant association between overall congenital malformations and prenatal exposure to ACEIs/ARBs (OR = 2.16, 95% CI = 1.72-2.71, P < .0001, calculated RR = 2.06; Table 2). A cumulative meta-analysis demonstrated that the addition of subsequent studies had little effect on the OR, but simply narrowed the 95% CI ( Figure S1). The significant relationship still existed when analysis was limited to studies with the first trimester-only exposure (OR = 1.94, 95% CI = 1.71-2.21, P < .0001, calculated RR = 1.91; Figure 2). The cumulative meta-analysis displaying results accumulated over successive studies is shown in Figure S2. Other outcome measures that enabled analysis included LBW, miscarriage, ETOP, stillbirth, and preterm delivery, all of which were significantly associated with prenatal exposure to ACEIs/ARBs (Table 2). Miscarriage, ETOP, and stillbirth were also significantly related to ACEI/ARB exposure in the only first trimester of pregnancy Abbreviations: ACEIs, angiotensin-converting enzyme inhibitors; ARBs, angiotensin II receptor blockers; n.a., not available; OAH, other antihypertensive medications. 1 No longer than gestational week 20. 2 Data derived from three randomized, placebo-controlled trials (ie, DIRECT-Prevent 1, DIRECT-Protect 1, and DIRECT-Protect 2).
most outcomes of interest were still existent when the analysis was limited to studies with the first trimester-only exposure (Table S4).  (Table S5). Prenatal exposure to ACEIs, but not ARBs, was found to be significantly associated with overall congenital malformations, LBW, miscarriage, ETOP, and preterm delivery.

| D ISCUSS I ON
To the best of our knowledge, this systematic review and meta-analysis includes the largest dataset in the literature for the purpose of examining the associations between prenatal exposure to ACEIs/ARBs and adverse pregnancy outcomes, including both adverse maternal outcomes and neonatal birth defects. The first trimester-only exposure to ACEIs/ARBs, previously presumably thought to be safe, 22  Abbreviations: ACEIs, angiotensin-converting enzyme inhibitors; ARBs, angiotensin II receptor blockers; CI, confidence interval; CNS, central nervous system; CVS, cardiovascular system; ETOP, elective termination of pregnancy; LBW, low birth weight; OR, odds ratio.

F I G U R E 2
Forrest plot of overall congenital malformations in first trimester-only exposure to ACEI/ARB to be significantly associated with adverse pregnancy outcomes, including overall and CVS congenital malformations. The overall results of this study may raise concerns about the potential dangers of ACEI/ ARB use during early pregnancy.
The adverse pregnancy outcomes that occur following in utero exposure to ACEIs/ARBs may result either directly from the drugs or from underlying maternal illnesses. When the ACEI/ARB group was compared to the OAH group, the effect size was smaller than when it was compared to nonexposure. It is also possible that ACEIs/ ARBs may be prescribed more often than other antihypertensive drug classes in hypertensive patients with diabetes because of their proven efficacy against the progression of diabetic nephropathy. 51,52 A hypertensive or diabetic disorder in pregnancy may itself be associated with adverse pregnancy outcomes without drug specificity and, thus, may act as a confounder in some observational studies included in our analysis. [53][54][55] Moreover, patients with such underlying conditions tend to be older and may exhibit other comorbidities, including obesity, which may also be related to an elevated risk of adverse pregnancy outcomes. 56,57 Therefore, it should be kept in mind that there was a likelihood of the present meta-analyses being confounded by some of these factors, for which some included studies might not adequately control.
Assumed the observed adverse pregnancy outcomes ascribed mainly to the drugs, the increased teratogenic risk could be conceivably attributed to inhibition of RAAS, a system that plays a key role in the embryogenic and fetal development of several organs/systems. 9,58-60 Not only does fetal RAAS blockade syndrome occur following ACEI/ARB exposure during the second and third trimesters of pregnancy it also may occur in those who are exposed to ACEIs/ARBs at the beginning of pregnancy. 15,61 Although there are unknown biologic mechanisms underlying adverse birth outcomes, inhibition of angiogenesis has been postulated to be a possible mechanism for the CVS malformations. 62 Given limited knowledge on how ACEIs/ ARBs might interfere with embryonic development during the critical period for organogenesis, further research is warranted to gain a better understanding of underlying mechanisms whereby the drugs might result in adverse pregnancy outcomes. Moreover, differential effects of ACEI/ARB exposure in the first trimester as compared to the second and third trimesters need further investigations.
Although it remains uncertain whether the elevated risk of adverse pregnancy outcomes observed in our analysis is specific to ACEIs/ARBs or related to maternal underlying conditions, this systematic review and meta-analysis largely supports the current recom- to abruptly stop taking the drugs and offer alternatives. 66 The only one randomized-controlled trial included in our analysis suggested no significant association of adverse pregnancy outcomes with drug exposure when the patients discontinued an ARB within an estimated 8 weeks from the last menstrual period. 51 It is reasonable to postulate that very short-term cumulative exposure to ACEIs/ARBs during early pregnancy would be associated with better pregnancy outcomes; however, further investigation is required.
The results of this study should be interpreted with caution. First, asymmetric funnel plots, indicative of the evidence of small-study effects (eg, publication bias), were observed in the meta-analyses of most outcomes of interest. The formal tests for funnel plot asymmetry (either the Begg's rank correlation test or the Egger's linear regression test) are prone to type II errors (or false negative) in small meta-analyses and, thus, the possibility of small-study effects or publication bias cannot be ruled out. 67 Although search terms being used were broad without being limited to specific study designs, it is conceivable that our analysis might have missed some pertinent studies which are, for example, only available in other databases (eg, Embase) or even be unpublished. 68,69 Positive studies reporting a teratogenic effect of the drugs may be more likely to be published than studies with null results. 70 However, sensitivity analyses demonstrated no or little change on effect estimates, indicating the robustness of the results. Selective publications of studies may be of less concern to the validity of the present systematic review and meta-analysis. 71,72 Second, it has been widely acknowledged in the literature that several observational studies on pregnancy outcome after drug exposure during early pregnancy often ignore left truncation and competing risks, leading to biased crude rates of miscarriage. 73 Moreover, ETOP rates may reflect patients' anxiety, including misunderstanding of drug risk, rather than the toxic effects of a drug. As a result, the meta-analysis might misestimate the effects of prenatal exposure to ACEIs/ARBs, particularly when exposure is limited to the first trimester of pregnancy, on some F I G U R E 3 Forrest plot of CVS malformations in first trimesteronly exposure to ACEI/ARB compared with control and OAH outcomes of interest. Pharmacovigilance with regard to the exposure of newly pregnant women to their current medications will further provide more evidence on the association between ACEI/ARB use during the early stage of pregnancy and adverse pregnancy outcomes.
In conclusion, this comprehensive and quantitative analysis of the evidence available to date suggests an increased risk of adverse pregnancy outcomes, including congenital malformations, with prenatal exposure to ACEIs/ARBs, regardless of the trimester of pregnancy. Prescription of ACEIs/ARBs in women with the potential to become pregnant should be discouraged provided that there are alternative drugs with a more favorable risk/benefit profile to treat a condition. Large observational studies that are properly designed to adequately account for the role of confounders are necessary to confirm the results of this study. Further investigations are required to reveal possible pathogenic pathways leading to adverse pregnancy outcomes, particularly congenital birth defects, if confirmed, in those with first-trimester exposure to ACEIs/ARBs.

E TH I C A L A PPROVA L S TATE M E NT
This study is exempt from ethical review and received the certificate of exemption from the Research Ethics Committee of the Faculty of Medicine, Chiang Mai University.

PATI E NT CO N S E NT S TATE M E NT
Not applicable.

ACK N OWLED G M ENTS
This study was partially supported by Chiang Mai University.

CO N FLI C T O F I NTE R E S T
The authors declare that they have no conflict of interest.

AUTH O R S ' CO NTR I B UTI O N S
All authors contributed to the study conception and design. Data collection and analysis were performed by NB and NK. The first draft of the manuscript was written by NB and the manuscript was finalized by NK. All authors read and approved the final manuscript.

DATA AVA I L A B I L I T Y S TAT E M E N T
All data used to support the findings of this study are available from the corresponding author upon reasonable request.