International reports of withdrawal symptoms in 93 newborns whose mothers had taken selective serotonin reuptake inhibitors during pregnancy raise concerns about a possible causal relationship between such symptoms and drugs in this class, particularly paroxetine, according to authors of a study that identified these cases.
Nearly two-thirds (64) of these cases were seen in babies whose mothers had taken paroxetine (Paxil), which the authors concluded should not be used in pregnancy, “or, if used, should be given at the lowest effective dose.”
The use of other SSRIs “should be carefully monitored and new cases promptly communicated to the pharmacovigilance systems,” wrote Emilio Sanz, M.D., professor of clinical pharmacology at the University of La Laguna (Spain), and associates (Lancet 2005;365:482–7).
When asked to comment on the study, two experts on drug therapy during pregnancy disagreed with the authors' conclusions, which they said fail to balance the risks and benefits of these drugs in pregnant women with depression.
Gideon Koren, M.D., director of the Motherisk Program, a teratogen information service at the Hospital for Sick Children, Toronto, said that while the identification of these cases in an international database was commendable, he took issue with the conclusion that paroxetine should not be used in pregnancy. This recommendation is not based on an appropriate risk-benefit analysis, he said, and it does not take into account the increased risk of maternal morbidity associated with untreated maternal depression, which is the strongest predictor of postpartum depression.
Moreover, the authors fail to take into account a study published last year, which found that in a large Swedish database, the association between paroxetine and these symptoms was no greater than with other SSRIs, added Dr. Koren, who said he has no financial ties to manufacturers of antidepressants.
He noted that neonatal withdrawal symptoms are self-limited and that the syndrome has “a very benign course,” which also was not discussed by the authors. He and his associates at Motherisk have conducted many prospective case-control studies on the effects of different drugs in pregnancy. One of the studies, published in 2002, found a significantly higher rate of neonatal withdrawal symptoms in newborns exposed to paroxetine in the third trimester, compared with unexposed controls.
Lee Cohen, M.D., director of the perinatal psychiatry program at Massachusetts General Hospital, Boston, emphasized that while an appropriate level of vigilance is warranted in neonates who have been exposed to SSRIs in the third trimester, the cases in the study represent spontaneous reports, not controlled data.
They are “not a clap of thunder” but represent another data set that is starting to suggest that there is some association between SSRI exposure and risk for perinatal syndrome, Dr. Cohen said in an interview.
What complicates the situation is that use of these drugs in the general population and in pregnant women is significant, but the incidence of these symptoms is probably extremely small. There are no controlled data available that can be used to reliably estimate the prevalence of these symptoms in pregnant women on antidepressants, added Dr. Cohen, who is a consultant to manufacturers of several antidepressants.
What concerns him most, Dr. Cohen said, is that the study could not only lead to a reduction in antidepressant use during the peripartum period, but could affect a woman's willingness to take medication she may need at other points during pregnancy. The study also could affect the clinicians' willingness to prescribe therapy when needed during pregnancy.
The study, published last month, involved a search for reports of cases in the WHO adverse drug reaction database, where spontaneous reports of suspected adverse drug reactions are sent from centers in 81 countries. The first case, which was associated with fluoxetine, was reported in 1995. As of November 2003, 93 suspected cases of SSRI-associated neonatal withdrawal syndrome had been reported. In 73 of those cases, no concomitant medications were reported or the concomitant medications were thought to be unrelated to the symptoms.
For 10 of the remaining 20 cases, an association with the SSRIs was considered “doubtful,” because of the concomitant use of medications that included antipsychotics or other drugs for which an association with neonatal withdrawal symptoms have not been clearly established. Another 10 were considered as “probably not” associated with SSRIs, because concomitant medications included drugs like opioids or tricyclics, according to the authors.
The most common neurological symptoms reported were nervousness, abnormal crying, tremors, and hypertonia. Other symptoms included digestive symptoms (vomiting, feeding disorders, or diarrhea), and respiratory symptoms (including two cases of respiratory depression). There were 13 cases of neonatal convulsions–11 listed as neonatal convulsions and 2 as grand mal convulsions.