David Baxter PhD
Late Founder
Measuring the Effects of Antidepressants versus Untreated Depression on Pregnancy Outcomes
by Ruta Nonacs, MD PhD, WomensMentalHealth.org
July 10, 2012
Multiple studies have demonstrated an increased risk of poor neonatal adaptation associated with exposure to selective serotonin reuptake inhibitor (SSRI) antidepressants in late pregnancy. Results in the literature consistently indicate that about 25%-30% of infants exposed to SSRIs late in pregnancy manifest symptoms of poor neonatal adaption, including jitteriness, restlessness, increased muscle tone, rapid breathing. These symptoms are transient, resolving spontaneously with no specific medical intervention.
Some, but not all studies, have also linked SSRI exposure with increased risk of preterm birth.
Few studies have attempted to parse out the relative contributions of different factors on neonatal outcomes. Pregnancy outcomes in women taking antidepressants are most often compared to outcomes in healthy, non-depressed women with no medication exposure. These sorts of comparisons make it impossible to differentiate between the effects of the medication and the effects of the underlying disorder. Depressive symptoms and anxiety, absent medication exposure, increases the risk of adverse outcomes, such as preterm birth, lower birth weight, and lower Apgar score. Depression in pregnancy may also be associated with other factors that may negatively affect outcomes, including smoking, substance abuse, insufficient prenatal care, and increased levels of psychosocial stressors.
A recent study conducted in the Netherlands as part of the Generation R Study, a population-based prospective cohort study from early fetal life, attempts to distinguish between the effects of SSRI exposure and the effects of untreated depression on neonatal outcomes. 7696 pregnant women were recruited during the first trimester. During each trimester, SSRI use was assessed using questionnaires and verified by pharmacy records. 188 women took SSRIs only before pregnancy; these women were excluded from analyses because spillover effects could not be ruled out.
During each trimester, fetal ultrasonography was performed to determine head and body growth. Prematurity was defined as birth before 37.0 weeks of gestation and low birth weight was defined as a birth weight less than 2500g.
Depressive symptoms were assessed with the Brief Symptom Inventory (BSI) at, on average, 20.6 weeks of gestation. “Clinically relevant depressive symptoms” were defined as a score greater than 0.75 on the 6-item Depression Scale of the BSI.
Based on the information about depressive symptoms and SSRI use, the subjects were classified into 3 groups:
Group 1: Women not using SSRIs, low depressive symptoms (n= 7027 [91.3%]), the control group.
Group 2: Women not using SSRIs, with depressive symptoms (n=570 [7.4%]).
Group 3: Women using SSRIs (n=99 [1.3%]).
Infants born to mothers with prenatal depressive symptoms showed reduced body growth and head growth. The risk of preterm birth in the depressed group (6.3%) was similar to that observed in controls (5.1%). Prenatal SSRI exposure was not associated with reduced body growth but was associated with reduced fetal head growth. The SSRI-exposed children were at higher risk for preterm birth (odds ratio = 2.14) compared to controls. The women using SSRIs during pregnancy (group 3) had significantly lower symptom scores on the BSI than those who were depressed but did not use SSRIs (group 2).
From this data, the researchers concluded that both depression and SSRI exposure may negatively affect outcomes. They noted, however, that the effects may be different. While SSRI exposure increased the risk of preterm birth, exposure to depression did not. Exposure to depression had a negative effect on overall growth; SSRI exposure affected only head growth.
This is the first study to demonstrate a specific effect of SSRI exposure which differs from the effect depressive symptoms on the developing fetus. However, the authors note that the findings are not conclusive and acknowledge certain methodological limitations. In this study, the three groups are not identical. In the depressed and the SSRI-exposed women, smoking, alcohol use and cannabis use were more common. These women were also more likely to be in the lowest socioeconomic group. These are all factors which have been shown to affect fetal growth. Although the investigators attempted to adjust for these risk factors, they note that some “unmeasured residual confounding” could still be present.
Other limitations (not addressed by the authors) include the use of the Brief Symptom Inventory (BSI) to diagnose depression and the decision to assess for depression at only one time point. The BSI is a tool used to measure a spectrum of psychological symptoms, including depression. It is, however, not a diagnostic tool and reflects symptoms only during the week prior to assessment. Using the BSI on only one assessment may either under- or overestimate the prevalence of depression in this population. The overall rate of “clinically relevant depressive symptoms” in this study is lower than would be expected (8.6%); other studies have demonstrated rates of antenatal depression ranging from 10% to 15%.
The exclusion of 188 women who took SSRIs before (but not during) pregnancy may have also affected rates of depression in this study. Based on data indicating high rates of relapse among women who discontinue antidepressants proximate to conception, we might expect a substantial proportion of those women to be included in group 2 (women with depression who did not take SSRIs). Would the inclusion of these women affect the study findings?
While there are limitations to this study, it has considerable strengths, including its large size and its prospective design. Its findings are generally consistent with previous studies; based on this study and many others, we cannot conclude that treatment with an SSRI is risk-free. But this study also very clearly indicates that untreated depression carries its own risks. It must be emphasized that it is virtually impossible to do a “perfect” study which would accurately measure the risks of SSRI exposure. A scientifically rigorous – but ethically unfeasible – study would include a group of healthy, non-depressed women treated SSRIs. This is the only way to separate the impact of the drug from the disease it is used to treat.
The study included 63,395 women from the Norwegian Mother and Child Cohort Study. Information on medication use and medical, sociodemographic, and psychological factor was collected using two self-administered questionnaires at gestational weeks 17 and 30. Data on pregnancy outcome were retrieved from the Medical Birth Registry of Norway.
Of the 63,395 women, 699 (1.1%) reported using antidepressants during pregnancy, most frequently SSRIs (0.9%). Exposure to SSRIs during the first trimester was not associated with increased risk of congenital malformations (adjusted odds ratio [OR], 1.22) or cardiovascular malformations (adjusted OR, 1.51). Exposure to antidepressants during pregnancy was not associated with increased risk of preterm birth (adjusted OR, 1.21) or low birth weight (adjusted OR, 0.62).
Source: El Marroun H, Jaddoe VW, Hudziak JJ, et al. Maternal Use of Selective Serotonin Reuptake Inhibitors, Fetal Growth, and Risk of Adverse Birth Outcomes. Arch Gen Psychiatry 2012; 69(7):715-721.
by Ruta Nonacs, MD PhD, WomensMentalHealth.org
July 10, 2012
Multiple studies have demonstrated an increased risk of poor neonatal adaptation associated with exposure to selective serotonin reuptake inhibitor (SSRI) antidepressants in late pregnancy. Results in the literature consistently indicate that about 25%-30% of infants exposed to SSRIs late in pregnancy manifest symptoms of poor neonatal adaption, including jitteriness, restlessness, increased muscle tone, rapid breathing. These symptoms are transient, resolving spontaneously with no specific medical intervention.
Some, but not all studies, have also linked SSRI exposure with increased risk of preterm birth.
Few studies have attempted to parse out the relative contributions of different factors on neonatal outcomes. Pregnancy outcomes in women taking antidepressants are most often compared to outcomes in healthy, non-depressed women with no medication exposure. These sorts of comparisons make it impossible to differentiate between the effects of the medication and the effects of the underlying disorder. Depressive symptoms and anxiety, absent medication exposure, increases the risk of adverse outcomes, such as preterm birth, lower birth weight, and lower Apgar score. Depression in pregnancy may also be associated with other factors that may negatively affect outcomes, including smoking, substance abuse, insufficient prenatal care, and increased levels of psychosocial stressors.
A recent study conducted in the Netherlands as part of the Generation R Study, a population-based prospective cohort study from early fetal life, attempts to distinguish between the effects of SSRI exposure and the effects of untreated depression on neonatal outcomes. 7696 pregnant women were recruited during the first trimester. During each trimester, SSRI use was assessed using questionnaires and verified by pharmacy records. 188 women took SSRIs only before pregnancy; these women were excluded from analyses because spillover effects could not be ruled out.
During each trimester, fetal ultrasonography was performed to determine head and body growth. Prematurity was defined as birth before 37.0 weeks of gestation and low birth weight was defined as a birth weight less than 2500g.
Depressive symptoms were assessed with the Brief Symptom Inventory (BSI) at, on average, 20.6 weeks of gestation. “Clinically relevant depressive symptoms” were defined as a score greater than 0.75 on the 6-item Depression Scale of the BSI.
Based on the information about depressive symptoms and SSRI use, the subjects were classified into 3 groups:
Group 1: Women not using SSRIs, low depressive symptoms (n= 7027 [91.3%]), the control group.
Group 2: Women not using SSRIs, with depressive symptoms (n=570 [7.4%]).
Group 3: Women using SSRIs (n=99 [1.3%]).
Infants born to mothers with prenatal depressive symptoms showed reduced body growth and head growth. The risk of preterm birth in the depressed group (6.3%) was similar to that observed in controls (5.1%). Prenatal SSRI exposure was not associated with reduced body growth but was associated with reduced fetal head growth. The SSRI-exposed children were at higher risk for preterm birth (odds ratio = 2.14) compared to controls. The women using SSRIs during pregnancy (group 3) had significantly lower symptom scores on the BSI than those who were depressed but did not use SSRIs (group 2).
From this data, the researchers concluded that both depression and SSRI exposure may negatively affect outcomes. They noted, however, that the effects may be different. While SSRI exposure increased the risk of preterm birth, exposure to depression did not. Exposure to depression had a negative effect on overall growth; SSRI exposure affected only head growth.
This is the first study to demonstrate a specific effect of SSRI exposure which differs from the effect depressive symptoms on the developing fetus. However, the authors note that the findings are not conclusive and acknowledge certain methodological limitations. In this study, the three groups are not identical. In the depressed and the SSRI-exposed women, smoking, alcohol use and cannabis use were more common. These women were also more likely to be in the lowest socioeconomic group. These are all factors which have been shown to affect fetal growth. Although the investigators attempted to adjust for these risk factors, they note that some “unmeasured residual confounding” could still be present.
Other limitations (not addressed by the authors) include the use of the Brief Symptom Inventory (BSI) to diagnose depression and the decision to assess for depression at only one time point. The BSI is a tool used to measure a spectrum of psychological symptoms, including depression. It is, however, not a diagnostic tool and reflects symptoms only during the week prior to assessment. Using the BSI on only one assessment may either under- or overestimate the prevalence of depression in this population. The overall rate of “clinically relevant depressive symptoms” in this study is lower than would be expected (8.6%); other studies have demonstrated rates of antenatal depression ranging from 10% to 15%.
The exclusion of 188 women who took SSRIs before (but not during) pregnancy may have also affected rates of depression in this study. Based on data indicating high rates of relapse among women who discontinue antidepressants proximate to conception, we might expect a substantial proportion of those women to be included in group 2 (women with depression who did not take SSRIs). Would the inclusion of these women affect the study findings?
While there are limitations to this study, it has considerable strengths, including its large size and its prospective design. Its findings are generally consistent with previous studies; based on this study and many others, we cannot conclude that treatment with an SSRI is risk-free. But this study also very clearly indicates that untreated depression carries its own risks. It must be emphasized that it is virtually impossible to do a “perfect” study which would accurately measure the risks of SSRI exposure. A scientifically rigorous – but ethically unfeasible – study would include a group of healthy, non-depressed women treated SSRIs. This is the only way to separate the impact of the drug from the disease it is used to treat.
The study included 63,395 women from the Norwegian Mother and Child Cohort Study. Information on medication use and medical, sociodemographic, and psychological factor was collected using two self-administered questionnaires at gestational weeks 17 and 30. Data on pregnancy outcome were retrieved from the Medical Birth Registry of Norway.
Of the 63,395 women, 699 (1.1%) reported using antidepressants during pregnancy, most frequently SSRIs (0.9%). Exposure to SSRIs during the first trimester was not associated with increased risk of congenital malformations (adjusted odds ratio [OR], 1.22) or cardiovascular malformations (adjusted OR, 1.51). Exposure to antidepressants during pregnancy was not associated with increased risk of preterm birth (adjusted OR, 1.21) or low birth weight (adjusted OR, 0.62).
Source: El Marroun H, Jaddoe VW, Hudziak JJ, et al. Maternal Use of Selective Serotonin Reuptake Inhibitors, Fetal Growth, and Risk of Adverse Birth Outcomes. Arch Gen Psychiatry 2012; 69(7):715-721.
