PHILADELPHIA – reported at the annual meeting of the American Society for Reproductive Medicine.
In addition, women in a subgroup receiving fluoxetine experienced a lower live birth rate and greater incidence of pregnancy loss than women taking other SSRIs, but the results were not statistically significant, said Dr. Sjaarda of the Eunice Kennedy Shriver National Institute of Child Health and Human Development.
“[It] is biologically plausible that fluoxetine might have some different effects,” she noted. “It does have some different interaction with [cytochrome] P-450 enzyme activity, and this translates to it having a much longer half-life as well. It’s different in terms of drug metabolism and in its interaction with the hormone biosynthesis pathway.”
Most of the research on antidepressants and SSRIs in pregnancy has focused on the safety of the agents, rather than the effect on pregnancy for women trying to conceive, explained Dr. Sjaarda. Previous research also has shown inconsistent findings for fecundability in women of reproductive age taking SSRIs, and the risk of specific SSRI compounds on pregnancy loss is unclear.
The researchers performed a longitudinal exposure assessment of the Effects of Aspirin in Gestation and Reproduction (EAGeR) trial, which consisted of 1,228 women aged between 18 and 40 years trying to conceive. Patients were included if they had one to two prior pregnancy losses, up to two live births, and had been trying to become pregnant for six menstrual cycles; they were excluded if they had a severe history of mental illness. There were 1,035 women who had no preconception antidepressant exposure and 183 who did have preconception antidepressant exposure.
Patients provided longitudinal urine samples at various time points, including while trying to conceive and in early pregnancy, during the menses phase of each menstrual cycle and at their last menstrual cycle, and at 4 and 8 weeks’ gestation if they become pregnant. The urine samples were collected at home or in clinic; human chorionic gonadotropin levels were measured on the stored samples. The researchers defined pregnancy loss as any kind of loss measured after detecting human chorionic gonadotropin, and they used the patient’s medical record to determine live birth. The fecundability odds ratio was used to estimate the odds of conception in menstrual cycles.
Aypical and tricyclic antidepressants and SSRIs such as sertraline, fluoxetine, and citalopram/escitalopram were analyzed, as well as use of opioids, cannabinoids, and benzodiazepines. In total, 172 women used SSRIs, which represented 94% of the patient group analyzed, said Dr. Sjaarda. “This cohort really represents women who are successfully controlled with first-line agents.”
Patients in both the SSRI and no-antidepressant groups had similar baseline characteristics, but there were differences with regard to body mass index (26 kg/m2 vs. 28 kg/m2), employment status (77% vs. 67%), perceived stress (1.0 vs. 0.9), and opioid exposure (16% vs. 23%).
The researchers found use of any SSRI was associated with a 23% reduction in fecundability, with patients using fluoxetine, sertraline, and citalopram/escitalopram having a similar reduction in fecundability, compared with patients not using SSRIs.
Patients who received SSRIs also had approximately a 53% live birth rate overall. When analyzed by individual SSRI, however, there was a statistically significant reduction in the live birth rate for patients who were using fluoxetine, compared with patients using sertraline and citalopram/escitalopram. “This suggests that there was something besides just reduced fecundability going on with the fluoxetine-exposed women,” said Dr. Sjaarda.
When SSRI use was analyzed with regard to pregnancy loss, there was a generally null effect between women exposed to SSRIs overall versus those not exposed at the time before conception, at last menstrual period, and at 4 or 8 weeks’ gestation. But when grouped by specific SSRI, patients receiving fluoxetine had increased risk of pregnancy loss prior to conception, compared with patients not taking fluoxetine (34% vs. 24%; adjusted risk ratio, 1.41; 95% confidence interval, 0.94-2.12) , as well as at their last menstrual period (34% vs. 24%; adjusted RR, 1.48; 95% CI, 0.98-2.24) and at 4 weeks of pregnancy (31% vs. 22%; adjusted RR, 1.61; 95% CI, 0.94-2.78). “This was about a 40%-60% increase in pregnancy loss, even though the sample size is generally small when you divide it into these groups,” said Dr. Sjaarda.
Mental health care is an important public health and maternal health issue, and SSRIs as a drug class are essential for helping to appropriately manage mental health, noted Dr. Sjaarda.
Because “patients’ disease severities all vary and the reactions to different drugs vary, no one-size-fits-all recommendation can be made for people planning a pregnancy while using SSRIs,” concluded Dr. Sjaarda. “However, we’re hoping that women and their physicians can now consider these new data, which are based on objective and longitudinally measured exposure, as well as prospectively-assessed outcomes for these most common antidepressants and develop to a more informed and individualized plan for women who are trying to conceive and use SSRIs.”
Dr. Sjaarda reported no relevant conflicts of interest.
SOURCE: Sjaarda L et al. ASRM 2019, .