Weight Gain and Bipolar Disorder Treatment
                   (revised 11/2007)

This is a difficult problem to be sure, one that almost everyone getting medication treatment has to consider in some way, because:  

Weight gain and medications

Yes, it happens.  

There are some medications which have become famous for this:  Zyprexa, Depakote, lithium are all guilty some of the time (not always; it doesn't happen to everybody).  And now that we're paying more attention to this problem, it's becoming clear that other medications can do it:  all of the new-generation antipsychotics (Zyprexa, Risperidone, Seroquel at least; Geodon very little, aripiprazole more than Geodon but less than the rest) and many of the old-generation medications like Haldol can cause weight gain.  Even some antidepressants that were never really suspect in this way are now known to cause weight gain (slower than the medications listed above, though, and perhaps less often), such as Prozac and Paxil. 

Health consequences of medication-induced weight gain (briefly)

Everybody knows that excess weight gain is a health problem, with impacts on heart risk through cholesterol levels and blood pressure as well as some increase in cancer risk.  But medication-induced weight gain may be even more a problem, in my opinion, because of something called "metabolic syndrome".  It has not been established with certainty, but right now (December 2002) it looks like at least Depakote and Zyprexa can cause this syndrome. 

Depakote has been implicated in causing Polycystic Ovarian Syndrome (PCOS), which is a variation on metabolic syndrome (see the PCOS page for the data on that). 

Zyprexa has been implicated in causing diabetes, which is basically a severe version of metabolic syndrome.  Anyone who wants to argue that point will have to explain the man in the recent Depakote vs. Zyprexa trial who started with a blood sugar of 84 and died during the trial, on Zyprexa, with a blood sugar of 843mhinfosource .  If that sounds a bit too strident, scroll to "The Medical Implications" in a well known bipolar newsletter, where you'll see the same point put more softly, and more references.  

Does exercise work?  How can you raise the odds of being successful?

When people talk about medication-induced weight gain, you almost always hear about "diet and exercise", and usually not with much conviction.  Face it, this is hard enough for people who don't have mood problems, let alone people who have cyclic phases of depression that include being extremely hopeless and unmotivated -- not at all conducive to sticking with either diet or exercise.  However, the importance of this step should not be forgotten.  Exercise clearly has antidepressant effects, for example, that along with all its other known health benefits make it a "no-brainer".  Not that this is so obvious everybody does it, of course. 

So let's have a discussion about exercise.  It takes a bit of space, so it was moved to its own page:  Exercise, not the usual rap

Weight gain causing mood symptoms "metabolically"? 

There is some evidence, not much, just a sliver of it, to suggest that weight gain itself, probably through metabolic syndrome, can cause very significant mood and anxiety that can look like bipolar II.  In many cases it could be some mild version of bipolar II just being made worse by the new hormonal environment.  But in some cases I think it could be just the weight gain alone that was the real trigger of symptoms.  And in these cases I think that treating the metabolic syndrome itself could reverse the problem.  Now, some of you are probably ready now to go read more about metabolic syndrome.  Follow the link above. 

Ways to cope with medication-induced weight gain

1.  Start with exercise, start with exercise, start with exercise, get up and walk 10 minutes saying to yourself "start with exercise", and so forth as above.   It just has to be said, as the starting place, before we turn to other approaches that have known risks.  None of these other approaches comes close to the positive effects on overall health risk offered by exercise.  Obviously the trick is to find a way to do it on a regular basis, and that is in the same league with quitting cigarettes!  By my informal calculations, it's harder!  Read Exercise: not the usual rap (link) if you haven't been there yet.  

2. Second, there's trying to rely on medications that are not clearly associated with weight gain.  The only mood stabilizers in such a class are oxcarbazepine (Trileptal), which may not be quite as strong as some others but can be helpful; and lamotrigine, which is slow to get going but has strong antidepressant properties as well as probable mood stabilizing effects. 

Zyprexa is famous for weight gain but started people really looking at this problem, and now it's clear that many of the new generation antipsychotics like Risperidone and Seroquel as well as Zyprexa can cause substantial weight gain (as well as the old generation as well, that's Haldol and Thorazine and the like), although several studies comparing them tend to come up with Zyprexa as the worst.Gianfrancesco, Caro, Hedenmalm  There is some excitement brewing about the brand-new "Abilify" cousin of these medications, which does not appear to cause weight gain.  Here's a very early report on its use in kids.  

3.  Thirdly, there's the stratgegy of using combinations of lower doses of medications to avoid side effects.  This strategy is particularly useful for Depakote, and probably lithium (where at least in my experience the weight gain problems are associated with higher doses; an animal study supports this Baptista).  You'll see this strategy reflected in my discussion of Treatment, in the section of this website on Bipolar II

4.  Fourth, there are the "antidote" medications.  The first we used a lot like this was Topiramate.  It definitely works to blunt appetite, often dramatically, leading to pretty substantial weight loss in nearly half the folks who take it, in my experience so far.  Too bad there's a side effect of confusion and memory problems, which affects about one person in 3 who tries this medication (that's a very high side effect rate, especially for something that can really interfere with a person's ability to function.  Ironically, it's also just subtle enough that often it takes quite a while for people to figure out that the medication is messing with their ability to think straight).  If it wasn't for that, we'd probably use this strategy much more often (Topiramate has been called "the California drug: it makes you thin and stupid", as well as a take-off on the trade name Topamax -- "stupimax", because of this problem.  You can tell I'm not a big fan of risking these kinds of side effects.) UPDATE 7/2005: A German research group compared topiramate to placebo, added to Zyprexa in patients who had gained weight on that medication.Nickel Weight loss with topiramate was over 10 pounds in 10 weeks. Unfortunately, for all the measurements they did, they didn't assess thinking ability (memory, word finding, for example) and just said they didn't see any cognitive problems. Two months prior, three cases of hypomania apparently induced by topiramate were reported,Kaplan and they do indeed look "caused by" to my eye (no abstract online; you need a helpful librarian like mine). These add to two other reports of mania with topiramate. Schlatter , Jochum  That matches my experience, in which I've seen topiramate seem to cause agitation rather like antidepressants. So I am still very mistrustful of topiramate as an antidote solution for medication-induced weight gain; and if used in bipolar disorder, it should be with some caution about worsening the patient's mood stability as well as his thinking abilities. 

Another antidote, for Zyprexa and Depakote, is the acid-production blockers (H2-blockers), e.g. Axid, Pepcid, and others.  One study that looked at this in some detail fount that it worked, at least somewhat:  instead of gaining 12 pounds in 16 weeks on Zyprexa, the patients on Axid only gained 6 pounds, and may have been leveling off.Cavazzoni  However, a later study looked at the same approach and found and it provided no benefit,Assuncao  Interestingly, and sadly: the first study, the one that found benefit from using an H-2 blocker, was produced with the financial support of the company that makes olanzapine.  The second study, finding no benefit, was produced with the financial support of a company that makes a competing medication, and began with  patients who are already gained a tremendous amount of weight.  No fair! That gives you some sense of how you have to watch out for how the research is done in this business.   Overall, this approach just has not been very popular among clinicians, which is another bit of evidence that it really doesn't help much (Zyprexa is such a effective medication, if something made it easier to use, less likely to cause weight gain, that would be a very big deal and would probably get quite a bit of attention).  In my view, if we're going to add an additional medication just to block another medication's side effects, it ought to work better than this.  Especially when there are others which may reverse weight gain, as follows.  

The olanzapine manufacturer seems to like this idea much more than metformin (below).  My guess is that metformin treats a problem they don't want to admit Zyprexa probably causes -- diabetes.   Update 10/2006:  Whereas previously the only study of this approach was by the company who makes the medication, I've just come across a study by some independent researchers. Dr. Graham and her team showed that amantadine could stop Zyprexa-induced weight gain even after people had already had a weight increase on the medication. Those on placebo continued to gain, 9 pounds over the 8 week study.Graham

But one person on amantadine (out of 12) dropped out because he or she became psychotic. This is a recognized risk of using amantadine in this setting.

I have tried this approach in three patients.  It stopped weight gain in all three but none are still on it.  Tremor was the main reason for stopping.  For now, until further research is published, amantadine just doesn't seem worth the risk to me, so I am not routinely offering it. Indeed, I try to use some medication other than Zyprexa to avoid this situation.  But I must admit, it's extremely effective. I would continue to consider amantadine if for some reason my patient really needed to stay on Zyprexa -- especially if they'd never had an episode of psychosis before. 

Here is some basic information about amantadine, and more technical information from another website. 

My original comments about metformin, from around 2003, appear below. 

Update May 2007:  Though metformin still seems to make sense in theory, in practice trying to use it to treat or prevent weight gain has not been very successful.  It doesn't work miracles, that much is clear.  It might help some.  A recent study in VenezuelaBaptista randomly assigned 80 patients to metformin or placebo. These people had been taking Zyprexa for four months or more. The placebo group maintained a stable weight, while the metformin group lost an average of about 3 pounds over 12 weeks. Not a very big difference. Does this potential for benefit, perhaps 3 pounds, justify taking even a minimal risk? We probably need another study like this, or two, before we will be able to answer that question more easily.

While we're waiting, one more thought:  exercise has almost exactly the same pattern of potential benefits as metformin, and almost no risks, at least by comparison.  Anyone contemplating the metformin approach should definitely have their exercise component already underway!  I know, easier said than done.  But let's get it done, people; let's show up at work with a little "glow" (aka "sweat")!  Our country is going stark raving mad, waiting for people to get diabetes and then treating them with expensive medications, when all we need to do is walk.  Park a mile a way from work and take your umbrella!  Whoops, sorry, that was a rant, was't it.  Just came over me....] 

The rest of this section about metformin, written when this idea was first considered, follows next. 

A possible new entry in the "antidote" group is Glucophage (metformin).  We have much less experience with this medication in this role, but tons of experience with it overall -- it's been a standard medication for diabetes for years.  It does not lower blood sugar directly, so lowering your blood sugar "too far" using this medication is not a problem.  It directly reverses on of the possible ways that Depakote and Zyprexa cause weight gain, through something called "metabolic syndrome".  Too bad Glucophage is a little tricky to use:  it causes diarrhea in about 50% of people who take it, although if you start low enough and go up slowly enough you are much more likely to avoid this problem. 

However, there is one small "open trial" (no control group) which showed that adding Glucophage to Zyprexa, or Depakote, or Risperidone, was enough to cause 15 of the 19 kids who had gained weight on those agents to start losing weight.Morrison  In most of the kids it was a substantial and sustained weight loss.  And there's perhaps -- just perhaps -- one more reason to think that this "antidote" strategy with Glucophage might be a good thing.  

As you'll see in the "metabolic syndrome" story, there is a sliver of evidence (so far; this is just becoming an active area of research) that Glucophage could actually treat mood symptoms somewhat similar to what we see mood stabilizers do for mood.  If this was true, then we'd have here a medication that not only could prevent and reverse metabolic syndrome, it might actually be a "treatment" itself for mood symptoms.  So far it looks much easier to tolerate in the long run than topiramate.  I've even had a patient improve on it and then able to stop the Glucophage, while staying on Depakote at a lower dose than originally caused her weight gain, without a return of weight gain.  That would be a great trick if we can get this result frequently with this medication!  But, be aware that this approach is almost completely unresearched, so your doctor virtually certain to be unaware of all this.  Iif she/he is hesitant about considering this approach, because of the lack of research, be aware that such hesitation is generally wise position to be taking at this stage.