Zyprexa (olanzapine) became rapidly popular after its introduction in the United States in 1996. Its structural similarity to Clozaril (clozapine) led to hopes that it would be as effective for treatment-resistant cases as Clozaril but without the risk of agranulocytosis. It quickly proved itself as an effective agent for both schizophrenia and bipolar mania. In sales, it surpassed Risperdal, which had been released two years earlier.

Unfortunately for Lilly, several significant side effect issues have emerged, and over the last year controlled studies have been giving clinicians second thoughts about prescribing Zyprexa at all. For example, a recent article in Archives of General Psychiatry included this rather astonishing caveat: “The potential cardiovascular consequences of olanzapine therapy, and its association with the metabolic syndrome, warrant serious consideration of its riskbenefit ratio by treating physicians.” (1)

“Zyprexa causes one pound of weight gain per week for the first 10 weeks, and 15-25 pounds in a year.”

What is going on here? Is this is a sober assessment of the real risks of prescribing Zyprexa? Or is it an alarmist over-reaction based on one research group’s ax-grinding? In this article I will apply the TCR magnifying glass to the data, claims, and counterclaims, and end up with something useful for you to apply to your practice. Wish me luck.

Weight. Does Zyprexa cause weight gain?
There’s little controversy here. Even Lilly reps agree that Zyprexa causes more weight gain than its first-line atypical peers (by “first line atypicals” I mean all those atypicals other than Clozaril, which is considered secondline because of its agranulocytosis risk). The most widely cited study documenting this effect is a meta-analysis written by Allison and colleagues (2), which includes a nice chart comparing all the commonly used neuroleptics with regard to weight gain. Piecing together this report and others, TCR’s rule of thumb is that Zyprexa will, on the average, cause one pound of weight gain per week for the first 10 weeks, and that this rather horrifying slope tapers off rapidly thereafter. The average one year weight gain on Zyprexa is 15 to 25 pounds.

However, this is probably overstating its weight liability, since this data is based largely on the treatment of schizophrenics, who are at higher risk for weight gain anyway due to lifestyle factors. In addition, these days most of us are so clued in to the possibility of weight gain with psychotropics that we take various proactive measures to minimize it. These include encouraging diet and exercise and using various psychopharmacologic nostrums to prevent weight gain. In the case of Zyprexa, these include high dose H2 blockers (shown effective in the Axid 300mg bid range (3)) and amantadine 100mg BID (4). On the other hand, who wants to deal with the potential for weight gain if you can avoid it by choosing a different agent? Few clinicians like to prescribe a second medication from day one just to prevent side effects caused by the first one.

The TCR Bottom Line on weight gain: It’s a huge liability for Zyprexa.

Diabetes. Does Zyprexa cause diabetes?
I asked Medical Affairs at Eli Lilly and Co. this very question, and they sent me a thick packet saying, essentially, that it is no more likely to do this than Risperdal. This packet was based primarily on results from their clinical trials for FDA registration, and as we all know, such trials select for relatively healthy patients who are unlikely to be representative of a real world clinical practice.

The diabetes issue reared its head because of a flurry of case reports published since the mid 1990s associating initiation of Zyprexa with new onset Type 2 diabetes, usually within 6 months of starting the med (5). While case reports can never prove causality, three factors should give us pause:
1. The sheer number of reports implicating Zyprexa has been impressive (237 cases since 1994);
2. The other atypicals have rarely been fingered in case reports; and
3. Many of the Zyprexa reports have been dramatic, including 15 reported deaths due to diabetic complications. Nonetheless, case reports are the poorest evidence available upon which to make clinical decision.

Recently, the first high quality controlled data was published (6), unfortunately confirming the implication of these case reports. University of Maryland researchers (with no obvious pharmaceutical funding) examined a huge general practitioner data base from England and Wales. They identified close to 20,000 patients diagnosed with schizophrenia between 1987 and 2000, and compared the risk of developing diabetes among patients on Zyprexa, Risperdal, and conventional neuroleptics. Their main finding: Zyprexa conferred a 4.2-fold increased risk of diabetes versus conventionals, while Risperdal led to a non-significant increased risk of 1.6.

Pretty damning stuff, right? Well, maybe not. After all, only a total of 9 out of 970 Zyprexa-treated patents developed diabetes at all, meaning that the actual risk of any given patient becoming diabetic is pretty low. Furthermore, we have no way of knowing how serious these cases were, since many Type 2 diabetics can be successfully treated with diet and exercise alone.

So, does Zyprexa cause diabetes? Probably, but the effect is mild, and the required monitoring is very easy (monthly or bimonthly random glucose tests over the first several months of therapy, and less thereafter). Nonetheless, as with weight gain, is this a worry that any psychiatrist wants to add to all the concerns that come with clinical work, when there are good alternatives that cause little, if any, hyperglycemia (Geodon, Abilify)? Probably not.

Hyperlipidemia. Does Zyprexa increase cholesterol and triglycerides?
Let’s skip the case reports, since we have two recently published controlled studies to help answer this. The first comes from the same group of Maryland researchers who culled the United Kingdom General Practitioner data base to link Zyprexa with diabetes. In this study (1) they found a 4.7-fold increased risk of hyperlipidemia on Zyprexa vs. being on no antipsychotics, and a 3-fold risk as compared to conventional agents. Risperdal, on the other hand, conferred no increased risk of hyperlipidemia whatsoever.

The second study was less rigorous, but in some ways more informative. Researchers at Oregon State Hospital did a retrospective chart review of 39 patients on Risperdal vs. 37 patients on Zyprexa (8). All patients had cholesterol, triglyceride, and glucose levels measured before drug initiation and again after one year on the medication. There were significantly greater increases in all metabolic parameters in the Zyprexa group. Triglyceride levels increased an average of 105 in the Zyprexa group vs. 32 for Risperdal; cholesterol increased 31 points for Zyprexa, 7 for Risperdal; and glucose increased 11 points for Zyprexa, 1 point for Risperdal.

So, in the end, what to say about Zyprexa, the latest whipping-boy of psychiatry? It’s effective stuff, but with the increased risk of weight gain, diabetes, and hyperlipidemia, it’s a hassle to prescribe. You might as well just prescribe Clozaril, which has all these side effects plus agranulocytosis, but at least it is the only neuroleptic to soar above the pack in terms of efficacy, and is now the only psychiatric drug FDA-approved for the prevention of suicide.

TCR  VERDICT: Zyprexa? A powerhouse, but oy, what a headache!

<sub>1.Koro CE, Fedder DO, L’Italien GJ. An assessment of the independent effects of olanzapine and risperidone exposure on the risk of hyperlipidemia in schizophrenic patients. Arch Gen Psychiatry. 2002;59:1021-1026.
2. Allison DB, Mentore JL, Heo M, et al. Antipsychotic-induced weight gain: a comprehensive research synthesis. Am J Psychiatry. 1999;156:1686-1696.
3. Sacchetti E, Guarneri L, Bravi D. H2 antagonist nizatidine may control olanzapine-associated weight gain in schizophrenic patients. Biol Psych. 2000;48:167-168.
4. Floris M, Lejeune J, Deberdt W. Effect of amantadine on weight gain during olanzapine treatment. European Neuropsychopharmacology. 2001;11:181-182.
5. Koller EA, and Doraiswamy PM. Olanzapineassociated diabetes mellitus. Pharmacotherapy. 2002;22:841-852.
6. Koro CE, Fedder DO, L’Italien GJ. Assessment of independent effect of olanzapine and risperidone on risk of diabetes among patients with schizophrenia: population based nested case control study. Br Med J. 2002;325:243.
7. Meyer JM. A retrospective comparison of weight, lipid, and glucose changes between risperidone- and olanzapine- treated inpatients: metabolic outcomes after 1 year. J Clin Psychiatry. 2002;63:425-433.</sub>