It has been eight years since child psychiatry first began to deal with controversy about SSRIs and their potential to prompt suicidal behavior in youth. The controversy was prevalent in the popular press from the time of the introduction of SSRIs, leading the FDA to state in 1991 that “there is no credible evidence of a causal link between the use of antidepressant drugs, including Prozac, and suicidality or violent behavior.”
However, this statement applied to SSRIs in general, not to SSRI use in children and adolescents. In April 2004, the Lancet published a meta-analysis of studies evaluating SSRIs versus placebo in children and adolescents. Looking at both published and previously suppressed (unpublished) studies, the authors concluded that the only antidepressant for which the potential benefits outweigh the risks was fluoxetine (Prozac) (Whittington CJ et al, Lancet 2004;363(9418):1341–1345).
For the other antidepressants, including paroxetine (Paxil), sertraline (Zoloft), citalopram (Celexa), and venlafaxine (Effexor), analysis of unpublished data seemed to show that they were relatively ineffective for depressive symptoms in kids, and that they caused an unacceptable rate of suicidal ideation and other serious adverse events. Their bottom line was that fluoxetine is the only reasonably safe and effective antidepressant for young patients.
This conclusion was rather controversial, with some readers accusing Whittington of everything from sloppy science to outright bias, while others were impressed with the analysis and with the difference that including unpublished studies could make.
In response to the concerns about antidepressants and suicidality, FDA researchers did a meta-analysis of 24 randomized, placebo-controlled trials of nine antidepressant drugs throughout the 1990s, which included all data, published or unpublished. They concluded that the use of antidepressant drugs in pediatric patients is associated with a “modestly increased” risk of suicidality (Hammad TA et al, Arch Gen Psychiatry 2006;63(3):332–339).
Unlike Whittington et al, the FDA scientists opted not to single out any particular drugs as being safer than others, citing many possible explanations for differences in risk. Nonetheless, it was clear from the study that some medications had a lower risk ratio for suicidal ideation (citalopram (Celexa) and fluoxetine at 1.37 and 1.53 respectively), while others were much higher (eg, venlafaxine at 8.84).
The FDA’s findings eventually led to a required black box warning about suicidality in children due to antidepressants in October of 2006 (the FDA had analyzed their results in 2004 but the paper took two years to go through the journal peer review process). After the black box warning, rates of prescriptions for antidepressants for children decreased, although debate quickly arose about precisely when this drop occurred and whether it led to negative consequences.
One study argued that the regulatory warnings (starting with FDA case reports of suicide issued in late 2003) and ensuing changes in prescribing resulted in increases in suicide rates in children and adolescents, citing a 14% increase in youth suicides in 2004, the largest increase in the period from 1988 to 2004. The authors argued that an inverse relationship between prescription of SSRIs and rate of suicide in youths exists in both the U.S. and the Netherlands (Gibbons et al, Am J Psychiatry 2007;164(9):1356–1363). In truth, however, prescribing decreases did not occur until 2005, and data published by the CDC for 2005 actually showed a decrease in suicide rate by about 3%.
Despite the heated academic debates about suicidality, most researchers agreed that the evidence for efficacy of antidepressants in children was, to say the least, underwhelming. Our confidence in antidepressant efficacy has improved over the past several years due to a series of studies.
The first of these, the NIMH-funded Treatment for Adolescents with Depression Study (TADS), endorsed the effectiveness of fluoxetine, though suicidal adolescents were specifically excluded from that study (March JS et al, JAMA 2004;292(7):807–820). The Treatment of Adolescent Suicide Attempters (TASA) study, reported in three papers in the October 2009 issue of JAACAP (Vol 48, Issue 10), addressed the question of efficacy around adolescents whose depression included suicidal ideation. (To read an in depth review of these studies, see CCPR, September 2010.) While these studies were not strictly comparable (TADS was double blind in design, while TASA was not), TASA lent credibility to the use of SSRIs as well as a specific form of CBT in adolescents with suicidal behavior prior to treatment, and TADS pointed towards fluoxetine.
Most recently, a large randomized, controlled study, Treatment of SSRI-Resistant Depression in Adolescents (TORDIA), examined depressed adolescents who had failed treatment with an initial SSRI. In SSRI-resistant adolescent depression, this study found that CBT plus a switch to either a different SSRI or venlafaxine (150 mg to 225 mg) showed a higher response rate (54.8%) than a medication change alone (40.5%)(Brent D et al, JAMA 2008;299(8):901–913). Of interest, the data from the TORDIA study replicated the FDA study finding that venlafaxine was associated with an increased risk of self harm events compared to those treated with an SSRI (Brent D et al, Am J Psychiatry 2009;166(4):418–426).
You may be more confused now than you were before starting this article! Is there a bottom line here? Not really, but the most recent data does something to enhance our confidence that a variety of SSRIs and SNRIs—and not just fluoxetine—can treat depression in kids. All antidepressants may or may not be equal in their risk of triggering suicidal ideation—though venlafaxine may be more risky than the others.