Dear Dr. Phelps,

I was wondering why atypical antipsychotics have become popular in treating bipolar disorder? Do doctors feel they have benefits beyond lithium and anticonvulsants, or does it have more to do with marketing?

Thanks very much.


Dear Mr. C --

This is an important question. Thank you for asking. 

Obviously, what would be really nice would be a "head-to-head" study comparing lithium, and perhaps valproate (Depakote) versus one or several of the atypical antipsychotics. In other words, could we see some direct evidence of which is more effective? Otherwise, we have to compare apples and oranges: a randomized trial versus placebo for one medication is not necessarily directly comparable to a randomized trial versus placebo for another medication. For example, the the patient population under study might have been different: from a different socioeconomic status, or a different degree of severity of illness.

Interestingly, one of our commonly used medications, Lamictal, has been subjected to a head-to-head trial at least three times. In one set of studies, it was compared to lithium; and in another, to an atypical antipsychotic, olanzapine (Zyprexa, which in this study was combined with fluoxetine; both are made by Eli Lilly, and the combination pill is called Symbyax). Lamictal was as effective as lithium in the treatment of bipolar depression (in a recent Scandinavian study, not yet published), and nearly as effective as Symbyax, with substantially fewer side effects. Importantly, in the latter study, unbeknownst to most psychiatrists, Lamictal was also as rapidly effective as Symbyax, which is quite surprising given the slow start required for Lamictal dosing.

Ahem, so back to the question: are atypical antipsychotics actually better than lithium, say? Better than Depakote? If not, why have they become so popular? Here are a few answers, basically off the top of my head:

1. In some instances, an atypical actually is better. For example, no other oral medication is as fast as Zyprexa (even Depakote "loading", when the doses pushed out very quickly to get the blood levels up fast, is still nowhere near as rapid). It can work within 30-45 minutes, or at least getting a foothold in that time. Similarly, nothing is as reliable for helping people sleep asquetiapine (Seroquel).

2. However, these potential advantages must be weighed against the risks of the atypicals. Usually people are worried about weight gain with these medications, but unfortunately that risk is shared by many of the alternatives: lithium can also cause weight gain, though much less often, and much less dramatically compared to olanzapine (quetiapine is also a problem in this respect, but less so than olanzapine; it might end up very roughly in the same ballpark as lithium -- I would love to see a comparison there).

Sometimes I think doctors forget that the atypicals carry a risk which none of the alternatives share, namely "tardive dyskinesia". It's amazing how often patients don't know about this risk, even when they are taking an atypical antipsychotic. I too have been guilty, I'm sure, on occasion, a failing to adequately warn patients about this movement disorder which comes on after a long duration of exposure to this medication (usually years, usually in the context of high- dose use). Because the risk is not immediate, I think it is often glossed over when weighing risks and benefits. That might be okay, as long as one remembers to come back and seriously consider that risk once symptoms are controlled and long-term medication exposure is a potential.

3. So is it marketing? There is no question that the marketing is powerful and pervasive. Many doctors like to think that they are immune, including me. One of the strategies I use to avoid thinking narrowly when considering medication alternatives is to brainstorm as many relevant options as I can think of while sitting with the patient and then, "thinking out loud", compare the rationale in favor of each versus the risks and potential side effects. I hope that in this fashion we can move beyond the marketing effect, as long as I am rigorous with that risk/benefit comparison. We are still stuck with the fact that companies influence much of the research, and it has been shown very clearly that they "spin" the results (and even the design, frequently).

So I would have to admit that it takes a determined effort to try to keep the playing field level.

4. Lastly, here is a frightening thought: is it possible that atypicals are just simpler to use? Compare lithium: frequent blood tests are required when this medication is started to make sure it is used safely. Compare carbamazepine: this requires a relatively complex ramping up to the full dose, and even more blood tests than lithium. Compare lamotrigine (Lamictal): this too requires a careful start. Only ziprasidone (Geodon) is as tricky to get started; interestingly, it is one of the least used. Depakote is roughly in the same ballpark as olanzapine or quetiapine or aripiprazole, but you are right: it seems to have fallen out of favor, right around won the manufacturer really backed off on their marketing.

This is only the beginning of an answer to this important question. 

Dr. Phelps



Published January, 2008