Sunday, March 26, 2006
There are two papers that report on results from the STAR*D (Sequenced Treatment Alternatives to Relieve Depression) project. One study shows what happens when patients are switched from one antidepressant that is not working for them, to a different one. The other shows what happens when a second drug is added to the first one.
The significance of these studies stems from the fact that it is a common decision point in the application of psychopharmacology, to have a patient who has not responded to the first drug tried. The question then comes up: Is it better to stop drug A, and try drug B (switch); or, to stay on drug A, and add drug B (add)? Although the studies do not provide a really definitive answer to the question, they at least provide some useful information.
One of the studies examined the question of what happens when the drugs are switched (Bupropion-SR, Sertraline, or Venlafaxine-XR after Failure of SSRIs for Depression). Briefly, the result is that about 25% of the patients achieve remission. It did not matter which f the three drugs was chosen:
Results Remission rates as assessed by the HRSD-17 and the QIDS-SR-16, respectively, were 21.3 percent and 25.5 percent for sustained-release bupropion, 17.6 percent and 26.6 percent for sertraline, and 24.8 percent and 25.0 percent for extended-release venlafaxine. QIDS-SR-16 response rates were 26.1 percent for sustained-release bupropion, 26.7 percent for sertraline, and 28.2 percent for extended-release venlafaxine. These treatments did not differ significantly with respect to outcomes, tolerability, or adverse events.A few observations are in order. First, conventional practice has been to switch from one family of drugs to another family, when employing the switch strategy. In this study, all of the patients had not responded to an SSRI, citalopram. It did not matter if they were switched to another SSRI, or a similar class (SNRI), or a completely different class (buproprion). Note that the "conventional practice" was never based upon empirical evidence; up until now, there simply wasn't much empirical evidence to go on. Second, the study does not tell us which patients would be better off with one drug over another. We are still waiting for that study.
The second study examined what happens when a second drug is added (Medication Augmentation after the Failure of SSRIs for Depression). Briefly, the study took a group of patients similar to those who enrolled in the other study; they all had not improved sufficiently on citalopram. Briefly, about 30% attained remission:
Results The sustained-release bupropion group and the buspirone group had similar rates of HRSD-17 remission (29.7 percent and 30.1 percent, respectively), QIDS-SR-16 remission (39.0 percent and 32.9 percent), and QIDS-SR-16 response (31.8 percent and 26.9 percent). Sustained-release bupropion, however, was associated with a greater reduction (from baseline to the end of this study) in QIDS-SR-16 scores than was buspirone (25.3 percent vs. 17.1 percent, P<0.04), a lower QIDS-SR-16 score at the end of this study (8.0 vs. 9.1, P<0.02), and a lower dropout rate due to intolerance (12.5 percent vs. 20.6 percent, P<0.009).Note that the bupropion group did somewhat better. This is interesting. In common practice, it is much more common for psychiatrist who employ the add strategy to add bupropion, even though there previously was no specific empirical guidance for this:
Background Although clinicians frequently add a second medication to an initial, ineffective antidepressant drug, no randomized controlled trial has compared the efficacy of this approach.The common practice was guided by intuition, more or less. In this case, the intuition appears to have been correct.
The third psychopharmacology paper in the NEJM issue is a "Perspective" piece on the cardiovascular safety of stimulants used in the treatment of ADHD. That paper is available only as a (180KB) PDF download. Their conclusion:
Although the committee recognized that there are important potential benefits of these drugs for certain highly dysfunctional children, we rejected the notion that the administration of potent sympathomimetic agents to millions of Americans is appropriate. We sought to emphasize more selective and restricted use, while increasing awareness of potential hazards. We argued that the FDA should act soon and decisively.This sounds a bit harsh to me, but I would still take it seriously. It may be that we need to start systematically monitoring blood pressure in all patients who are getting these drugs. The number of deaths reported was extremely small, given the millions of patients who receive these drugs. Even allowing for vast underreporting, it still means the risk is small. But since the potential consequences are great, and the monitoring is so easy, it would make sense to do it.
(Note: The Rest of the Story/Corpus Callosum has moved. Visit the new site here.)
E-mail a link that points to this post: