乌鸦传媒

Women taking antidepressants during pregnancy had a small but statistically significant increased risk of their babies developing pulmonary persistent hypertension of the newborn (PPHN), according to the results of a population-based study.

Overall, 31.5 per 10,000 infants (95% CI 28.8-35.2) exposed to selective serotonin reuptake inhibitors (SSRIs) and 29.1 per 10,000 infants (CI 23.3-36.4) exposed to non-SSRIs in the last 90 days of pregnancy were diagnosed with PPHN compared with 20.8 per 10,000 infants (CI 20.4-21.3) diagnosed with PPHN who were not exposed to antidepressants, reported , of Brigham and Women's Hospital in Boston, and colleagues.

However, after adjusting for depression and stratification of high-propensity scores, odds ratios for SSRI use and PPHN dropped from 1.51 (95% CI 1.35-1.69) initially to 1.10 (CI 0.94-1.29), they wrote in the .

Non-SSRIs saw similar adjustments from the initial data, dropping from 1.40 (95% CI 1.12-1.75) to 1.02 (CI 0.77-1.35).

Prior research on the association between maternal antidepressant use and PPHN, and the ensuing conflicting results, was one of the reasons Huybrechts told 乌鸦传媒 that she wanted to conduct the current study. She also pointed out that a number of smaller studies may have had insufficient power to detect the increase in risk for PPHN.

"The most important takeaway in our study is that the absolute risk of PPHN is very small and the risk increase due to exposure to antidepressants during pregnancy, if it is present, is much smaller than previous studies have suggested," she said.

In this retrospective cohort study, researchers examined data on more than 3 million pregnant women from 2000 to 2010 in the Medicaid Analytic eXtract database for 46 states and Washington. Of these, 3.4% used antidepressants 90 days prior to delivery, with 2.7% using SSRIs and 0.7% using non-SSRIs. Compared with women unexposed to antidepressants, these women were more likely to older, white, use other psychotropic medications, have a chronic illness, be obese, be a smoker, and have great healthcare use.

Antidepressant use was defined as having filled a prescription for antidepressants after the 20th week of pregnancy. Women who took both SSRIs and non-SSRIs were excluded. PPHN was classified based on maternal and infant ICD-9 codes for persistent fetal circulation or primary pulmonary hypertension, and the positive predictive value of this outcome definition was 89.6%.

Researchers saw a slight increase in risk of PPHN when restricting to term deliveries and primary PPHN "in the absence of congenital cardiac malformations and lung hypoplasia" for both SSRI (high-dimensional adjusted OR 1.28, 95% CI 1.01-1.64) and non-SSRI use (high-dimensional aOR 1.14, CI: 0.74-1.74).

In sensitivity analyses, the authors found statistically significant associations between risk of PPHN and maternal diabetes (OR 2.93, 95% CI 2.72-3.15), obesity (OR 2.02, 95% CI 1.88-2.17), and black race (OR 1.30, CI: 1.24-1.36), but said that these were consistent with "well-established associations" in other studies.

They noted no evidence of increased association between antidepressant use and severe PPHN. A sensitivity analyses in which women were required to have filled a minimum of two prescriptions during the last 90 days before delivery did not result in stronger associations, the authors reported.

When study findings were added to that looked at the association between SSRI use late in pregnancy and PPHN, using a random effects model, the pooled OR went from 2.50 (95% CI 1.32-4.73) to 1.95 (9% CI 1.08-3.54) for their base-case estimate and to 2.03 (95% CI 1.21-3.41) for primary PPHN.

In a video commentary, 乌鸦传媒's , assistant professor at Yale University, remarked on the potential "overadjustment" (defined as adjusting for a factor that lies on the causal pathway between exposure and outcome of interest). He used premature birth as an example.

"For example, let's say that antidepressants increased the risk of premature birth and premature birth increased the risk of PPHN," he said. "By adjusting for prematurity, you would falsely conclude that the antidepressant wasn't linked to PPHN."

Huybrechts responded that it was important to adjust for a number of confounders to ensure accurate results, adding that other studies have restricted their analyses to full-term births, which is why any adjustments in this study were done as secondary analyses.

"We did restrict to depression diagnosis and adjusted for a number of other risk factors for PPHN, but we feel that was absolutely crucial in order to account for potential confounding variables," she said. Her group did acknowledge that stratifying by premature birth could create an association between antidepressant use and PPHN.

Other study limitations included its reliance on antidepressant prescription data, not actual use, and the fact that the Medicaid sample is younger and more racially diverse than the general population. In addition, some of the data on potential confounding factors, such as smoking, maternal body mass index, and diabetes severity was missing or incomplete, and thus those factors may not have been fully accounted for in the high-dimensional propensity score analysis.

"My personal opinion is that the signal of harm in this study is small enough to be potentially insignificant," Wilson stated.

The authors concluded that "clinicians and patients need to balance the potential small increase in the risk of PPHN, along with other risks that have been attributed to SSRI use during pregnancy, with the benefits attributable to these drugs in improving maternal health and well-being."

Comment