Assistant Professor in Maternal Fetal Medicine, Division of Maternal Fetal Medicine, Department of Gynecology & Obstetrics, Johns Hopkins University School of Medicine, Baltimore
SYNOPSIS: In this synthesis involving 15 meta-analytic studies, four studies demonstrated an association between paroxetine use during pregnancy and increased risk of major congenital anomalies (relative risk [RR], 1.18 [95% confidence interval (CI), 1.05, 1.32] to 1.29 [95% CI, 1.11-1.49]). For all selective serotonin reuptake inhibitors (SSRIs), the RR for major anomalies (1.10 [95% CI, 1.03, 1.16] to 1.27 [95% CI, 1.09, 1.47]) and cardiac defects (1.06 [95% CI, 0.94, 1.18] to 1.36 [95% CI, 0.61, 3.04]) were increased. This meta-analysis suggests an increased risk of cardiac and major anomalies with SSRI use, but the results should be interpreted with caution, since all included studies were meta-analyses of retrospective cohort studies.
SOURCE: Uguz F. Selective serotonin reuptake inhibitors and the risk of congenital anomalies: A systematic review of current meta-analyses. Expert Opin Drug Saf 2020;19:1595-1604.
The use of selective serotonin reuptake inhibitors (SSRIs) in obstetrics continues to increase.1 In the United States, approximately 8% to 13% of women take SSRIs during pregnancy and postpartum.2 SSRIs (including sertraline, fluoxetine, fluvoxamine, citalopram, escitalopram, and paroxetine) are effective therapies for the treatment of moderate and severe depression. Although there is limited evidence from randomized clinical trials and prospective cohort studies regarding the safety and efficacy of SSRI use during pregnancy, data from several retrospective and case-control studies suggest short- and long-term adverse effects with SSRI use, including the potential for persistent pulmonary hypertension of the newborn and congenital anomalies.3 However, because of challenges caused by confounding by indication and study design, data from retrospective cohort and case-control studies are difficult to interpret and compare across studies. Therefore, a systematic review and meta-analysis is a good way to summarize the evidence across studies.
This study was a synthesis of several meta-analytic studies that examined the association between SSRI use and congenital anomalies between 2010 and 2020.4 All meta-analytic studies that examined SSRI use in pregnant women and reported a congenital malformation were included.4 The outcomes evaluated included the prevalence of major anomalies involving any organ system, cardiac defects, and minor anomalies. Effect estimates were reported as pooled analyses of relative risks (RR) that measured the relationship between SSRIs (exposure) and outcomes (congenital anomalies).
Fifteen meta-analytic studies met the inclusion criteria and were included in the final synthesis of meta-analyses. Fourteen of these meta-analytic studies evaluated cardiovascular anomalies as their main outcome, while nine meta-analytic studies focused on major congenital anomalies involving all organ systems. Six, four, and two meta-analytic studies evaluated cardiac septal defects, anomalies in other systems, and minor anomalies, respectively.4 Paroxetine was the most frequently reported SSRI associated with major, cardiac, and minor congenital anomalies, and was evaluated in seven meta-analytic studies.4 The effects of fluoxetine were evaluated in six meta-analytic studies, while escitalopram and fluvoxamine were evaluated only in one meta-analysis. For all SSRIs, the RR for major anomalies was 1.10 (95% confidence interval [CI], 1.03, 1.16) to 1.27 (95% CI, 1.09, 1.47), and the RR for cardiac defects was 1.06 (95% CI, 0.94, 1.18) to 1.36 (95% CI, 0.61, 3.04).
The association between paroxetine use during pregnancy and the risk of major congenital anomalies was found to be statistically significant in four out of five meta-analyses (RR, 1.18 [95% CI, 1.05, 1.32] to 1.29 [95% CI, 1.11-1.49]) and the risk of cardiovascular defects were found to be significant in six out of seven meta-analyses (RR, 0.97 [95% CI, 0.75, 1.19] to 1.46 [95% CI, 1.17, 1.82]).4 Among specific cardiovascular anomalies, the risks of right ventricular outflow track defects (RR, 2.15 [95% CI, 1.04, 4.44] to 2.29 [95% CI, 1.06, 4.93]), bulbus cordis, and cardiac septal anomalies (RR, 1.42 [95% CI, 1.07-1.89]) were greatest. Four meta-analytic studies examined the association between sertraline use and the risk of congenital anomalies. Although two of these meta-analytic studies found an association between sertraline use and increased risk of cardiovascular anomalies (RR, 0.93 [95% CI, 0.70, 1.24] to 1.42 [95% CI, 1.12, 1.80]), the two other meta-analytic studies did not find any differences in cardiac anomalies between women who used sertraline while pregnant and those who did not. Four meta-analytic studies that evaluated the association between citalopram and congenital anomalies did not demonstrate any statistically significant differences with citalopram use and congenital cardiac anomalies (RR, 1.04 [95% CI, 0.92, 1.17] to 1.20 [95% CI, 1.09-1.31]).4 Although patients who used escitalopram had infants with abdominal wall defects (RR, 3.52 [95% CI, 1.56-7.93]) and gastroschisis (RR, 3.95 [95% CI, 1.46-10.68]), no congenital anomalies were found to be associated with fluvoxamine use.4
Meta-analysis is a statistical method for combining data from different studies (cohort, randomized trials, or case-controlled) with the aim of investigating the effect of interventions using a large sample size.5,6 Sometimes, synthesis of several meta-analyses can be conducted to understand the associations between an exposure and an outcome when data from an existing meta-analysis demonstrate conflicting results. The conclusions of such data synthesis from several meta-analytic studies can depend greatly on the quality of the included studies, the meta-analytic process (meta-evaluation), and the methods employed to synthesize data from the individual meta-analytic studies.6 Data from this study by Uguz demonstrated potential associations between several SSRIs and congenital anomalies.4 Previous data from meta-analyses suggest an association between SSRI use (paroxetine use specifically) during pregnancy and congenital heart defects. However, data from other systematic reviews/meta-analyses do not support this association. For example, a 2007 systematic review and meta-analysis of seven studies identified a summary odds ratio (OR) of 1.72 for cardiovascular malformations and first-trimester paroxetine exposure.7 This study also demonstrated that women taking SSRIs were more likely to have fetal echocardiogram during pregnancy than women not taking SSRIs, and the children of the mothers exposed to SSRIs were more likely to have echocardiograms during the first year of life. This raises the possibility that children exposed to SSRIs with transient septal defects were more likely to be diagnosed than unexposed children.
Another meta-analysis funded by GlaxoSmithKline (the pharmaceutical company that markets paroxetine) identified a summary OR for congenital defects at 1.24 for congenital cardiac defects.8 A 2016 meta-analytic study reported an increase in major fetal congenital malformations (OR, 1.23; 95% CI, 1.10, 1.38; n = 15 studies) and cardiac malformations (OR, 1.28; 95% CI, 1.11, 1.47; n = 18 studies) associated with paroxetine use during the first trimester. However, this finding lost statistical significance when the control group included women taking other SSRIs/antidepressant medications.9 Mothertobaby.org (a pregnancy network consisting of 15 affiliates at acclaimed universities and hospitals across the United States and Canada that conduct observational research to better understand the effects of medications and vaccines when taken during pregnancy) has suggested a risk for cardiac defects of 2% in pregnancies exposed to paroxetine (compared to the general population risk of 1%).
The inconsistencies in these SSRI data also can be explained by confounding by indication — a major challenge in virtually all retrospective studies of the association between SSRI use and depression. Although it is difficult to control for confounding factors when confounding by indication is thought to be present, several pharmacoepidemiological studies have demonstrated that confounding by indication can benefit from other analytic methods, including propensity score matching/analysis, use of instrumental variables, sensitivity analysis for unmeasured confounders, and G-estimation. One efficient way to control for the risk factors for depression from population-based data is the use of a chronic disease scoring system in the statistical analytic model. The chronic disease score can provide a reasonable measure of acute vs. chronic depression disease status and help accurately predict the association between SSRIs and adverse outcomes. None of these methods were used to control for confounding in this synthesis of several meta-analysis studies by Uguz.
Because of inconsistencies across available data and lack of robust well-designed studies from prospective cohort studies and randomized clinical trials, women taking SSRIs prior to pregnancy should discuss the risks vs. the benefits with their care providers. SSRIs can be continued during pregnancy when the benefit of their use to the pregnant women outweighs the risks, since untreated depression carries significant risks to the mother and the fetus. Paroxetine should not be used as a first-line therapy in women commencing antidepressant treatment during pregnancy. A fetal echocardiogram (in addition to the routine second trimester anomaly scan) is recommended for any pregnant woman exposed to paroxetine during pregnancy.
- Masarwa R, Bar-Oz B, Gorelik E, et al. Prenatal exposure to selective serotonin reuptake inhibitors and serotonin norepinephrine reuptake inhibitors and risk for persistent pulmonary hypertension of the newborn: A systematic review, meta-analysis, and network meta-analysis. Am J Obstet Gynecol 2019;220:57.e51-57.e13.
- Huybrechts KF, Palmsten K, Avorn J, et al. Antidepressant use in pregnancy and the risk of cardiac defects. N Engl J Med 2014;370:2397-2407.
- Bérard A, Sheehy O, Zhao JP, et al. SSRI and SNRI use during pregnancy and the risk of persistent pulmonary hypertension of the newborn. Br J Clin Pharmacol 2017;83:1126-1133.
- Uguz F. Selective serotonin reuptake inhibitors and the risk of congenital anomalies: A systematic review of current meta-analyses. Expert Opin Drug Saf 2020;19:1595-1604.
- Haidich AB. Meta-analysis in medical research. Hippokratia 2010;14:29-37.
- Nakagawa S, Noble DW, Senior AM, Lagisz M. Meta-evaluation of meta-analysis: Ten appraisal questions for biologists. BMC Biol 2017;15:18.
- Bar-Oz B, Einarson T, Einarson A, et al. Paroxetine and congenital malformations: Meta-analysis and consideration of potential confounding factors. Clin Ther 2007;29:918-926.
- Wurst KE, Poole C, Ephross SA, Olshan AF. First trimester paroxetine use and the prevalence of congenital, specifically cardiac, defects: A meta-analysis of epidemiological studies. Birth Defects Res A Clin Mol Teratol 2010;88:159-170.
- Bérard A, Iessa N, Chaabane S, et al. The risk of major cardiac malformations associated with paroxetine use during the first trimester of pregnancy: A systematic review and meta-analysis. Br J Clin Pharmacol 2016;81:589-604.