Dr. Damian Sendler Exposure to Second-Generation Antipsychotic Drugs During Pregnancy Is Linked to Birth Defects
Damian Sendler: It is estimated that between 2% and 3% of all pregnancies are affected by a major congenital malformation (MCM). Many studies have looked into the possibility of MCM as a result of maternal antipsychotic exposure in this context. A recent meta-analysis summarized these studies. Among the outcomes examined in this article were the risks for MCMs associated with first-trimester gestational exposure to second-generation antipsychotic (SGA) drugs, as well as the findings of two recent observational studies that were not included in the meta-analysis. This finding was also true for early pregnancy exposure and for SGA exposure alone in the meta-analysis of data from six observational studies on the topic of antipsychotic drug use during pregnancy. Retrospective analyses of a huge population-based cohort study conducted in Finland found no link between first-trimester gestational exposure to SGAs and an increased risk of MCM compared to either no exposure or exposure to first-generation antipsychotics; however, exploratory analyses found an association between olanzapine exposure and an increased risk of musculoskeletal malformations, specifically.
Damian Jacob Sendler: SGA exposure during the first trimester of pregnancy was not associated with an increased risk in a group of women with psychiatric disorders, according to data from a pregnancy registry in the United States. Neither quetiapine nor aripiprazole had an increased risk of MCM after first trimester use in studies that drew on the same registry. False positives from multiple hypothesis testing or residual confounding may have contributed to significant findings in unadjusted or exploratory analyses, according to this study. A shared decision-making process is therefore required to weigh the substantial benefits of antipsychotics against the unsubstantiated risks of MCMs.
Dr. Sendler: Exposure to psychotropic drugs during pregnancy is associated with a wide range of negative fetal and neurodevelopmental outcomes. Antidepressant drugs, particularly selective serotonin reuptake inhibitors (SSRIs), have been extensively studied on this subject and do not appear to increase the risk of major congenital malformations2 (MCMs). Many studies have looked at the risk of MCMs in the context of prenatal antipsychotic drug exposure over the last decade and a half. Maternal-child malformations (MCMs) are one of the most common and feared complications of pregnancy; they are structural abnormalities that can be found in the womb or soon after birth; they can be cosmetically disfiguring or pose a health risk. Sodium valproate is the most well-known teratogen in neuropsychiatry, with a prevalence of 2%–3% in the general population.
A recent systematic review and meta-analysis summarized studies on MCMs linked to antipsychotic exposure during pregnancy.
3 In light of the fact that second-generation antipsychotics (SGAs) have largely replaced first-generation antipsychotics (FGAs) around the world, this article examines the meta-analysis with a focus on findings associated with first-trimester exposure to SGA drugs. Both of these studies were observational and were not included in the meta-analysis, so this article also discusses them.
Studies of congenital malformations associated with antipsychotic drug exposure during pregnancy have been reviewed and analyzed by Wang et al3. Using electronic databases and other resources, these researchers found 13 studies that could be included in random-effects meta-analyses. Six of those studies were selected for systematic review.
This study found that exposure to antipsychotic drugs during pregnancy was not associated with a significantly increased risk of congenital malformations (6 studies; risk ratio [RR], 1.23; 95 percent confidence interval [CI], 0.96–1.58) in the main analysis." There was no increased risk associated with exposure to SGAs according to a more detailed analysis (three studies; relative risk (RR) of 1.35; 95 percent confidence interval (CI), 0.73–2.47). Final analysis found that antipsychotic exposure in the first or second trimester was not associated with an increased risk (4 studies; RR, 1.05; 95% CI, 0.96–1.15). Because organogenesis occurs in the first trimester, fetal exposure to a teratogen could cause an MCM, this analysis is critical.
Limitations of this meta-analysis include data sources that were not clearly differentiated between major and minor congenital malformations, data that were mixed across trimesters of exposure, drug combinations, and appropriateness of control groups. There were only a handful of studies in the meta-analyses that included more than 10,000 women, and this was the case for the meta-analysis as a whole. This study6 found no significant association between first-trimester exposure to antipsychotic drugs and the risk of congenital malformations for either FGA or SGA (RR, 0.90; 95 percent CI, 0.62–1.31) after controlling for confounding variables.
A retrospective, population-based cohort study of MCMs linked to gestational antipsychotic drug exposure was described by Ellfolk et al4 in their paper. Linked registers in Finland were used to gather the data for this study. Pregnancies exposed to known teratogens were excluded, as were singleton pregnancies that ended in live birth or stillbirth or that were terminated due to MCMs in the period 1996–2017. There were 3,478, 1,030, and 22,540 pregnancies that had not been exposed to antipsychotics during the first trimester of pregnancy, respectively, which were the groups of interest. To be considered exposed, a woman must have purchased antipsychotic medications between one month before conception and the end of the first trimester; to be considered nonexposed, she must have made no such purchases between three months before conception and the end of the first trimester. Because of the possibility that it was prescribed for morning sickness, prochlorperazine was excluded from the FGA group, making it more difficult to control for confounding by indication in the SGA vs FGA analyses.
The most commonly prescribed SGA was quetiapine (n = 2,618), followed by olanzapine (n = 413), risperidone (n = 242), aripiprazole (n = 220), and clozapine (n = 106) in that order. MCMs that were clearly linked to known genetic disorders were excluded from the study. In the SGA, FGA, and unexposed groups, there were 187 (5.4 percent), 60 (5.9 percent), and 978 (4.3 percent) MCMs, respectively.
SGA exposure was not linked to an increased risk of MCMs compared to either FGA exposure (OR, 0.82; 95% CI, 0.56–1.20) or no exposure (OR, 0.92; 95% CI, 0.72–1.19) in analyses that adjusted for a large number of confounding variables. There was an increased risk of musculoskeletal malformations (OR, 3.71; 95 percent CI 1.35–10.01) only when olanzapine was included in exploratory analyses compared to pregnancies that were not exposed to the drug (OR, 2.12; 95 percent CI, 1.19–3.76). As a result, there was a wide range of musculoskeletal abnormalities.
Damian Jacob Markiewicz Sendler: There were a lot of exposed pregnancies in this study, which is a plus. Additionally, the study included MCMs linked to miscarriage and stillbirth. Because the data were collected retrospectively, there is no guarantee that women who purchased medications actually took them; if they did not, the findings of the study would be biased toward the null hypothesis. Univariate analyses were used to identify potential confounds, which is a two-step procedure that is generally considered to be unreliable and can lead to overfitting.
Damian Sendler
The Massachusetts General Hospital National Pregnancy Registry for Atypical Antipsychotics, established in 2008, has been updated with new, prospectively collected data. Pregnant women between the ages of 18 and 45 with a diagnosable mental illness were divided into two groups: those who had been exposed to oral or long-acting SGAs during the first trimester of pregnancy (the exposed group) and those who had not (the comparison group). Telephone interviews with these women took place when they were enrolled in the study, at 7 months pregnant, and 3 months after giving birth. Additionally, data on possible confounding variables such as alcohol consumption and illicit drug use were gathered during these interviews. Finally, data on the pregnancy and its outcomes were gathered. In addition to the information gathered through interviews, medical records were also examined. A trained dysmorphologist who was not aware of medication exposure evaluated the MCMs that were identified.
640 SGA-exposed infants and 704 comparison group infants were studied by the authors5. The mothers were an average of 33 years old. If you've ever had a child, you've probably noticed that the mother has an increased BMI, a higher smoking rate, a higher risk of bipolar disorder, an under-representation of major depression and anxiety, and a lower likelihood of using selective serotonin reuptake inhibitors.
Only 2% of the women in the study had ever used FGAs. Quetiapine, aripiprazole, and lurasidone were the three most commonly prescribed SGAs. Antidepressants, anticonvulsants, and antianxiety medications were also prescribed to many women in both groups.
2.50 percent of SGA patients had MCMs, while 1.99 percent of comparison patients had them. The crude odds ratio (OR) was 1.26 (95 percent confidence interval, 0.61–2.61). No one of the nearly two dozen confounds examined in analyses that looked at just one confound at a time was found to be significantly associated with MCMs. As a result of a stepwise selection process and a sample size of only 30 MCMs, the model was vulnerable to overfitting in the final analyses that took into account five potential confounds, with an OR of 1.48 (95 percent CI, 0.63–3.51). There was no clear pattern of malformations in the MCMs of the two groups, though the number of MCMs in the SGA group was 16 and the comparison group was 14.
Damian Jacob Sendler
The study's strengths were its ability to collect data on a wide range of confounding variables, the inclusion of a comparison group that included patients with psychiatric disorders, and the evaluation of the outcome data that was confirmed by medical records and by a dysmorphologist who was blinded to the exposure status of the patients. Observational studies also suffer from detection bias because infants of exposed pregnancies are more likely to be examined or studied for malformations. All the women in this study had a diagnosable mental illness, making detection bias less likely. Only 4.5% of the exposed group had schizophrenia, and the sample size was too small to include all relevant confounders in the regression model because of the study's inability to adjust for diagnosis and severity of illness.
Damien Sendler: Because crude risks decreased after adjusting for confounding variables4, the evidence suggests that risks may be due to confounding by indication, i.e., to inadequately measured, unmeasured, and unknown confounds associated with the disorder for which the antipsychotics are prescribed. It's reassuring that studies of first-trimester exposure to SGAs have not yet identified a clear or consistent risk of MCMs. Individual SGAs, on the other hand, appear to have no effect on the risk of MCM. The large studies of Huybrechts et al.6 and Ellfolk et al.4 found an increased risk after first trimester exposure to risperidone and olanzapine, respectively, but neither study was able to correct for false positive discoveries associated with multiple hypothesis testing and to rule out residual confounding. Neither study, however, supported the findings of the previous study. In addition, these studies would have been subjected to the detection bias in infants, which was mentioned earlier in this section. Finally, it's worth noting that no research has been done on the effects of dosing.
Anxiety, insomnia, and tic disorders can also be treated with antipsychotic drugs in monotherapy or as an augmentation therapy for schizophrenia, mania, depression, and obsessive compulsive disorder. Because of the risk of MCMs and other adverse outcomes associated with mood stabilizers, SGAs may be preferred over mood stabilizers (especially valproate). As a result of treatment discontinuation, there is a high risk of relapse and subsequent impairment in multiple domains, including family, social and occupational functioning in the majority of the conditions for which SGAs are used Antipsychotic drugs (especially SGAs) must therefore be weighed against the small gestational risks, discussed in this article and another article in this column1, that are more likely to be caused by confounding by indication than by the drugs themselves, in order to weigh the substantial benefits during pregnancy of necessary initiation or continuation. The patient and her family, on the other hand, should be involved in the decision-making process. It's not going to be an easy task.