SSRIs

SSRIs: Implications for women’s health across the reproductive life span

Department of Obstetrics and Gynecology, Stanford University Medical Center, Stanford, California

Drs Friedman, Shahine, and Lathi have no financial relationships or commercial interests to disclose.

The highest prevalence of major depressive disorders occurs in reproductive-aged and perimenopausal women.¹ Although selective serotonin reuptake inhibitors (SSRIs) are commonly used to treat this large population, little is known about the links among SSRIs, gonadotropins, estrogen, fertility, pregnancy outcomes, and vasomotor symptoms. Further, there is evidence that the effects of serotonin depend on the surrounding hormonal milieu, which could explain the variability in individual responses to SSRIs.

This review explores the potential effects of SSRIs at different stages of the reproductive life span, with regard to the existing literature about the impact of SSRIs on fetal development, fertility, and thermoregulation.

Pregnancy: Considerations for SSRI use

Approximately 1 in 10 women will experience depression during pregnancy and/or the postpartum period.¹ The prevalence of SSRI use during pregnancy is estimated to be approximately 2% to 3%.² The available literature on SSRI exposure and birth defects is conflicting; however, the clear negative effects of untreated depression must be considered when making a treatment decision.

What is the potential for negative fetal effects?

Although previous investigators did not find an increased risk of major congenital malformations associated with SSRI use in pregnancy,^3-5 recent studies have raised concern about potential teratogenic effects. For example, reports of increased risk of cardiac malformations with first trimester exposure to paroxetine prompted a reclassification of paroxetine as a pregnancy Category D drug.^6,7 Two recent studies from large multicenter surveillance programs found overall differing results, but both did suggest associations between SSRIs and certain rare anomalies such as omphalocele.^8,9 It is important to note that although these studies demonstrated increased relative risk, the absolute risk for rare anomalies was quite low.

Less controversial is the association between SSRI exposure later in pregnancy and poor neonatal adaptation, primarily involving transient symptoms of the respiratory and central nervous systems, admission to the neonatal specialized care unit, and persistent pulmonary hypertension of the newborn.^10-16

The mechanisms by which SSRIs may negatively affect fetal development and perinatal outcomes are not known. Transient neonatal behavioral effects have been described as resulting from SSRI-induced withdrawal,¹⁷ although others have proposed that symptoms are more consistent with neonatal serotonin toxicity.¹⁸ It has also been suggested that SSRIs may affect perinatal outcomes through decreased placental perfusion.³ In vitro perfusion studies have supported this hypothesis in demonstrating that serotonin exerts a strong vasoconstrictive effect on human umbilical arteries.^19,20 More research is needed in this area to better understand the pathophysiology underlying the association between SSRIs and adverse perinatal outcomes.

Factoring in the risks of depression

The American College of Obstetricians and Gynecologists (ACOG) currently recommends that the potential risk of SSRI use in pregnancy be considered in the context of the risk of depression if the medication is discontinued.²¹ One study found that pregnant women who discontinued antidepressants had 5 times the risk of depression relapse during pregnancy, compared with women who chose to continue their medications.²² Moreover, studies have suggested an association between untreated maternal depression and increased risk of pregnancy and postnatal complications, such as fetal growth restriction, low birth weight, and newborn irritability.^23,24 Depressive symptoms during pregnancy have also been associated with significantly higher rates of cigarette and alcohol use.²⁵

Finally, the American Psychiatric Association lists other risks of untreated depression during pregnancy that must be considered, including suicide, long-term hospitalization, familial discord, and obtaining inadequate obstetrical care.²⁶

Resolving the treatment dilemma

One study sought to distinguish the effects of prenatal exposure to SSRIs from those of maternal depression on neonatal health. Even accounting for the effects of depression, infants born to mothers who had been treated with SSRIs had lower birth weights and higher rates of respiratory distress.²⁷

Thus, treatment with SSRIs in pregnancy must account for the severity of depression and must involve a discussion of risks and benefits with the obstetrician, patient, and mental health clinician. Patients must be assessed case-by-case; however, in many cases, pregnant women taking SSRIs are encouraged to continue their medication throughout the pregnancy in order to optimize maternal mental health.

Infertility: How do SSRIs change the reproductive hormonal milieu?

Although the rate of depression in the in vitro fertilization (IVF) population approaches 40%—almost double the rate found in a control population²⁸—only 4% of this population uses SSRIs.²⁹ There is a potential association between SSRIs and gonadotropin levels that may decrease pregnancy rates, although mechanisms by which SSRIs may affect fertility are largely unknown, and the few studies to date that address this issue have had conflicting results.

One retrospective study reported decreased pregnancy rates and fewer ongoing pregnancies among women taking SSRIs who were seeking to conceive via IVF.³⁰ We recently found, however, that although there was a higher IVF cycle cancellation rate due to poor response among SSRI users, neither pregnancy rates nor live birth rates were significantly affected by SSRI use.²⁹

SSRIs may lower pregnancy rates through interference with gonadotropin and hormone levels, although the exact mechanism is unclear. Interestingly, SSRI administration has been shown to decrease estrogen levels in ovariectomized rats on estrogen therapy.³¹ These rat models suggest that serotonin may decrease circulating estrogen by acting centrally to affect gonadotropin release. Alternatively, SSRIs may also increase steroid metabolism peripherally through the hepatic enzyme cytochrome P450.³²

In analyzing the effects of SSRIs on IVF outcomes, it is important to consider that depression itself—as opposed to SSRI treatment—may also contribute to decreased pregnancy rates. For example, prospective clinical studies have shown that depression, anxiety, and stress are associated with lower pregnancy rates in patients undergoing IVF.^33,34 To better understand the potential impact of SSRIs on IVF outcomes, prospective studies, ideally controlling for depressive symptoms, are warranted.

SSRIs and vasomotor symptoms: A complicated association

The potential association between SSRIs and gonadotropin levels that may decrease fertility may also affect vasomotor symptoms. A confounding issue is the widely variable vasomotor response of patients to treatment with SSRIs, which worsen symptoms in some women and improve them in others.

Studies in rat models support the notion that the effects of serotonin depend on the surrounding hormonal milieu. For example, estrogen has been shown to mediate 5-HT₂ (5-hydroxytryptamine; serotonin) receptor binding in the rat brain frontal cortex, suggesting that receptor responses to SSRIs may be influenced by one’s hormonal state.³⁵ Studies have also shown that serotonin (5-HT) can inhibit luteinizing hormone (LH) secretion in ovariectomized rats, but after exogenous estrogen administration, 5-HT can have a stimulating effect on LH secretion.³⁶

In general, research suggests that SSRIs may be a helpful treatment for postmenopausal women with hot flushes, but that pre- or perimenopausal women are more susceptible to SSRI-induced vasomotor instability. The following discussion reviews this research and the potential mechanisms for these responses.

Role of serotonin in vasomotor symptoms

The pathophysiology of menopausal vasomotor symptoms is not well understood. Most proposed theories involve estrogen withdrawal and serotonergic and noradrenergic pathways interacting within the thermoregulatory center of the hypothalamus. Studies have suggested that the effects of estrogen may be mediated through the serotonergic system. Following estrogen withdrawal during spontaneous or surgical menopause, blood 5-HT levels have been shown to decrease,³⁷ and return to levels within normal ranges with exogenous estrogen treatment.³⁸ Administration of estrogen in menopausal women has also been shown to increase urinary excretion of the main metabolite of 5-HT, 5-hydroxyindoleacetic acid (5-HIAA).³⁹

Berendsen has suggested a mechanism by which decreased serotonin levels may lead to hot flushes (FIGURE).⁴⁰ In this hypothesis, estrogen withdrawal leads to decreased serotonin levels, which in turn stimulates upregulation of 5-HT_2A receptors in the hypothalamus. Stimuli such as stress or alcohol increase the release of 5-HT moduline; this protein leads to increased release of 5-HT, which in turn stimulates the 5-HT_2A receptors. Evidence in rat models suggests that stimulation of this receptor subtype alters the body’s set point temperature.^41,42 This alteration may trigger autonomic responses to cool the body through increased skin temperature and sweating.

SSRIs as a cause of vasomotor symptoms?

Case reports of premenopausal women who developed night sweats while taking SSRIs have been published.^43,44 We recently reported on a case of a 35-year-old woman who was seeking treatment for infertility and was prescribed SSRIs for depression.⁴⁵ She subsequently experienced night sweats and elevated follicle-stimulating hormone (FSH) levels on fluoxetine, paroxetine, citalopram, and venlafaxine. Each time she stopped the medication, her symptoms resolved and her FSH levels returned to more normal limits. Studies in rat models lend support to this observed side effect by demonstrating that activation of the serotonergic system can lead to an increase in circulating FSH levels.^46,47

SSRIs as treatment for vasomotor symptoms: Mixed results

Although in some cases vasomotor symptoms are worsened by SSRI treatment, many women have experienced improvement in menopausal symptoms while taking SSRIs. Therefore, SSRIs have been proposed as an alternative to hormonal therapy for the treatment of vasomotor symptoms.

Several randomized, placebo-controlled trials have evaluated the efficacy of SSRIs for the treatment of vasomotor symptoms, and the results have varied significantly. Studies that found a difference between SSRIs and placebo have often shown only modest clinical benefit when analyzing improvement in hot flush frequency and severity. It has been suggested that this modest average clinical benefit truly encompasses strong individual responses.⁴⁸ For example, one group that found a statistically significant improvement in hot flush frequency and severity with SSRI use⁴⁹ conducted a post hoc analysis of their data to focus on the variable responses of individual participants.⁴⁸ This analysis found that 31% of women had a clinically significant reduction in symptoms, 32% had a mild reduction in symptoms, and 37% of women actually had an increase in hot flushes.

Interestingly, in this study perimenopausal women experienced the greatest increase in hot flushes. Of the women who had menstruated in the 3 months prior to study enrollment, 61% had an increase in hot flush symptoms and only 11% had a decrease. In comparison, of the women who had not menstruated within the year prior to study enrollment, 31% had a clinically significant reduction in symptoms, and only 33% had an increase in symptoms. These findings lend support to the case report we described of worsening hot flushes with SSRIs, suggesting that perimenopausal, as opposed to postmenopausal women, may be at greater risk of SSRI-induced thermoregulatory dysfunction.

Overall, studies have more consistently demonstrated efficacy in breast cancer survivors,^50-53 with only one randomized study failing to find SSRIs effective in treating vasomotor symptoms in this population.⁵⁴ In the general population, however, several randomized studies have failed to show a statistically significant difference in efficacy between SSRIs and placebo.^55-57 It has been suggested that the higher efficacy found in the breast cancer population may stem from the widespread use of the selective estrogen receptor modulator tamoxifen and the ability of SSRIs to increase the metabolism of tamoxifen to inactive metabolites.^56,58 Of note, there are no current recommendations against patients on tamoxifen taking SSRIs because of this proposed mechanism of action.

Conclusions

Several observations and studies suggest complex interactions between serotonin, estrogen, and gonadotropins, yet these relationships have not been clearly elucidated. Further, SSRIs have individual neurotransmitter effects because of factors such as polymorphisms within the serotonin transmitter gene, which may account for differences in benefits, side effects, and potential hormonal interactions.

The impact of SSRIs on fertility, pregnancy, and pregnancy outcomes is poorly understood. Research suggests that SSRIs may be a helpful treatment for postmenopausal women who have hot flushes. Pre- or perimenopausal women, however, may be more susceptible to SSRI-induced vasomotor instability, and clinicians should counsel women about the potential for worsening of symptoms. Because SSRIs are currently being prescribed to many reproductive-aged women, further research is needed to explore these relationships and to understand the implications of the use of SSRIs on the reproductive health of women.

Sexuality, Reproduction & Menopause ©2008 Quadrant HealthCom Inc.