Re: Selective serotonin reuptake inhibitors during pregnancy and risk of persistent pulmonary hypertension in the newborn: population based cohort study from the five Nordic countries
The study on selective serotonin reuptake inhibitors (SSRIs) by Kieler et al.  suggests that use of SSRIs in late pregnancy increase the risk of persistent pulmonary hypotension (PPH) more than two-fold - OR 2.1 (95% CI 1.5 to 3). However there remain other plausible explanations, particularly the likelihood of confounding by severity of depression, but also smoking and maternal weight, which have not been addressed adequately. The study included methodological limitations which should be taken into consideration before making a firm conclusion on the association between SSRIs and risk of PPH.
Women who continue to be dispensed SSRIs after 20 weeks of gestation will differ substantially on a range of characteristics other than exposure to SSRIs from women who are not exposed. That children of women with a previous admission to hospital for a psychiatric condition were identified to be at increased risk of persistent pulmonary hypertension in the absence of SSRI exposure, OR: 1.3 (95% CI 1.0 to 1.6), points to an association between outcome and the underlying severity of depression. Subsequently, the authors elected to combine previous admission and exposure to SSRIs, appearing to conclude that the risk of exposure to SSRIs is raised yet further in that group (OR: 3.1 (95% CI 1.9 to 4.9)), when of course other competing explanations exist. We wonder whether the association between SSRI and PPH remains after adjustment for previous psychiatric hospitalisations as an independent factor in the analysis?
Unfortunately, the tables and the statistic reported in the paper are not directly comparable. For example Table 1 describes the characteristics of those were dispensed SSRIs during pregnancy and those who were not dispensed, but does not describe the characteristics of the subset who were dispensed SSRIs late in pregnancy (e.g. from week 20) although inference is made in the paper regarding those women. Kieler et al. did adjust for some potential confounding variables, although it is not clear why smoking and BMI was not included in their main analyses. In this context a propensity score approach might have enabled a more efficient adjustment or matching for subject characteristics, and crucially indicate which women should be excluded from the analysis as they shared no characteristics with the women in the alternative risk stratum. Further, women in this sample are likely to be clustered within general practices, centres, regions or countries. The analyses should take this extra binomial variability into account.
Finally, as pointed out by Koren and Nordeng in the editorial , it cannot be assumed that SSRI’s dispensed in the latter half of pregnancy necessarily equate actual usage/exposure in pregnancy. Some women may get SSRIs dispensed for usage after delivery others may never use the medication. A stronger indication of usage (although not perfect) would be to include only women who have been dispensed two or more SSRIs within a short time interval prior to delivery.
Less than 12% of women on antidepressants treatment (prior to pregnancy start) continue treatment into the latter half of pregnancy.  Potentially, these women have no other alternative than to continue treatment. Wrongly concluding that SSRI exposure late in pregnancy for women with depression causes persistent pulmonary hypotension in their babies may cause substantial anxiety to mothers. We require further and more carefully designed studies before coming to such firm conclusions that SSRI in late pregnancy increases risk of persistent pulmonary hypertension.
1. Kieler, H., et al., Selective serotonin reuptake inhibitors during pregnancy and risk of persistent pulmonary hypertension in the newborn: population based cohort study from the five Nordic countries. BMJ (Clinical research ed.), 2011. 344: p. d8012.
2. Koren, G. and H. Nordeng, SSRIs and persistent pulmonary hypertension of the newborn. BMJ (Clinical research ed.), 2011. 344: p. d7642.
3. Petersen, I., et al., Pregnancy as a major determinant for discontinuation of antidepressants: an analysis of data from The Health Improvement Network. J Clin Psychiatry, 2011. 72(7): p. 979-85.
Competing interests: No competing interests