SSRIs and persistent pulmonary hypertension in the newborn (PPHN)

Last week’s BMJ scare (click here) associating persistent pulmonary hypertension in the newborn (PPHN) with a popular class of antidepressant, selective serotoin reuptake inhibitors (SSRI’s) taken in the second half of pregnancy, was even less justified than usual. It certainly did not justify the press interviews with the Scandinavian authors, and the headlines which flew around the world.

The accompanying editorial (click here) urged caution because previous studies had given conflicting results, neither exposures nor diagnoses were confirmed and the relative risk was modest (odds ratio 2.1; 95% CI 1.5 – 3). Confounding, far from being excluded, was present; women admitted for psychiatric reasons who did not take antidepressants had an odds ratio of 1.3 (95% CI 1.1-1.7).  The editorialists also made the point that even if genuine, the absolute risk was tiny, and when other causes such as cardiac disease or meconium aspiration are excluded, PPHN is usually self-limiting. They might have added that the authors did not pre-specify their analysis plan, choice of adjustment factors or definitions of exposure timing, so readers can never be sure that no data dredging occurred.

Obstetricians already limit antidepressants in pregnancy as far as possible, and pregnant women are well aware of the dangers in general. The BMJ knew the evidence didn’t stack up. Why else publish the debunking editorial?

So what purpose was served by publication in a general journal? Surely the accompanying hype wasn’t just for the glorification of the BMJ and the authors?

Jim Thornton