Introduction

Manic depression, also called bipolar disorder, afflicts some 4 million Americans. Characterized by severe mood swings between incredible elation and deep depression, the disorder has been difficult to treat. Many patients either do not respond to drug therapy or stop taking their medication because of unwanted side effects.

Now a new drug has entered the arena: Zyprexa (olanzapine), manufactured by Eli Lilly. Zyprexa had already been approved for the treatment of schizophrenia, and recently got the green light from the FDA for use in patients with manic depression.

Clinical Study Results

Studies evaluating the effectiveness of the drug in patients with acute mania showed that those who took Zyprexa (olanzapine) achieved a significantly greater improvement in their symptoms than patients who received a placebo. Other studies have shown that Zyprexa is as effective as lithium, another agent commonly used to treat manic depression.

What You Should Know

Like other drugs in its class, Zyprexa does have side effects, including dry mouth and weight gain. Drowsiness and dizziness have also been associated with Zyprexa, so patients should be cautioned about operating hazardous machinery, including automobiles, until they are reasonably certain that they will not experience these effects. Because there are other potential side effects, patients with manic depression who are interested in Zyprexa (olanzapine) should consult their physicians.

Zyprexa

Introduction

Bipolar disorder is a lifelong psychiatric disorder that features mood swings between mania and depression, and afflicts some 4 million Americans. Although lithium, valproate, carbamazepine, and antipsychotics have been prescribed to treat the disorder, many patients do not respond to these treatments or stop taking them due to unwanted side effects. Now there is another drug available for these patients: Eli Lilly received FDA approval to market Zyprexa (olanzapine) – a drug originally approved for schizophrenia – for the treatment of manic depression.

How It Works

Zyprexa is an antipsychotic drug. Although the exact mechanism of action is unknown, it is proposed that the drug’s activity is mediated through a combination of dopamine and serotonin type 2 (5-HT2) antagonism, as well as antagonism at other receptors with similar affinities.

Zyprexa (olanzapine): Clinical Study Results

A randomized, double-blind, placebo-controlled study was conducted over a 3-week period in patients with acute mania. Clinical response was compared between 70 patients who received 10 mg Zyprexa once daily and 69 patients who received placebo. After the first day of treatment, the Zyprexa dose was adjusted upward or downward, as clinically indicated, by one capsule, within the permitted range of one to four capsules. Clinical response was defined as a decrease of 50% or more in the Young Mania Rating Scale total score, from baseline to endpoint. At the end of the study period, significantly more patients taking Zyprexa (48.6%) showed clinical improvement than those taking placebo (24.2%). Although somnolence, dry mouth, dizziness and weight gain occurred more often with Zyprexa than with placebo, no patients discontinued treatment due to these side effects.

What the Patient Should Know

As with other antipsychotic drugs, Zyprexa (olanzapine) is associated with several side effects. Since orthostatic hypotension has been reported with Zyprexa, patients who experience this effect should be advised to avoid standing up too quickly. Because somnolence and dizziness have been associated with Zyprexa, patients should be cautioned about operating hazardous machinery, including automobiles, until they are reasonably certain that they will not experience these effects. Patients should not breast-feed during Zyprexa treatment and should inform their physicians if they become pregnant while on the treatment. They should also avoid alcohol and tell their physicians if they are taking other medications, since Zyprexa may interact with other drugs (for example, it may potentiate the effect of antihypertensive drugs and antagonize those of levodopa and dopamine agonists). Since Zyprexa may elevate transaminase levels, it should be used with caution in patients with hepatic impairment. Patients with a history of seizures should tell their physicians, since Zyprexa may increase the risk of seizures. Other side effects associated with Zyprexa (olanzapine) include restlessness, constipation, dry mouth, and weight gain.

Published in the September 1999 issue of the Canadian Journal of Psychiatry, the prospective, open-label study by Serdar Dursun et al. reported on 16 patients (11 male, five female; mean age 40±7 years) having a documented history of resistance to treatment. Dursun is associate professor and head of clinical pharmacology and experimental psychiatry in the department of psychiatry’s psychopharmacology research unit at Dalhousie University, Halifax, Nova Scotia, Canada.

Patients in the study were consecutive outpatient referrals who had shown poor response to antipsychotic treatment. All patients met the DSM-IV criteria for schizophrenia (mean duration of illness was 16±7 years) and scored at least 45 on the 18-item version of the Brief Psychiatric Rating Scale (BPRS) at baseline. Researchers defined treatment resistance by the following criteria: no significant symptomatic relief from schizophrenia symptoms, no period of good functioning and at least three previous trials with antipsychotics from at least two different chemical classes for a minimum of six weeks in the previous five years. The study excluded any patients with clinically significant organic disorders, or who had shown evidence of alcohol or substance abuse within the preceding six months.

Before entering the study, patients had been taking either a conventional antipsychotic drug or risperidone (Risperdal). The initial dose of olanzapine was 5 mg daily. Based on severity of symptoms and patient tolerance, the dose was increased to as much as 40 mg daily, well above the manufacturer’s recommended maximum daily dose of 20 mg. Patients were followed for 16 weeks, and outcome measures were assessed against baseline scores at weeks 4, 8 and 16. In addition to the BPRS, response to treatment was also assessed using the Global Assessment Scale (GAS), and abnormal movements were measured using the Abnormal Involuntary Movement Scale (AIMS).

At week 4, all patients showed a decline in BPRS scores measured against baseline, and this improvement continued at weeks 8 and 16. Half of the patients responded to treatment, as defined by a 20% decrease in BPRS scores. The mean change in BPRS score for all patients was 18.2%±15.6%, and in the responder group the average drop in score was 43%±11.4%.

Scores on the GAS increased during the study, reaching a final average of 41.25, showing that significant functional impairment persisted in the patients. Abnormal involuntary movements, as measured by AIMS, increased significantly at week 4. At subsequent follow-ups, these scores did not vary significantly from baseline, although mean scores remained above baseline.

No patients dropped out of the study and no serious adverse events were reported, suggesting olanzapine is well-tolerated at high doses. The worsening of involuntary movements shown at week 4 was thought to have resulted from switching medications or from withdrawal effects from the discontinuation of conventional antipsychotics.

Researchers said their findings are limited by several factors, including nonsystematic evaluation of overall medication tolerability, the absence of a control group and a potential for investigator bias in the unblinded study.

The researchers concluded that the study results add to a growing body of evidence that atypical antipsychotic medications may be useful for patients with schizophrenia showing treatment refractoriness. They said that cost-effectiveness and tolerability of olanzapine in high doses requires proper assessment and recommended that treatment-resistant patients be given an eight-week trial of the drug at regular doses before increasing the dose.

Conservative estimates suggest that as many as 30% of patients treated for schizophrenia are refractory to treatment.

Most initial published data on atypical antipsychotic drugs in the elderly are clinical trials in nondemented patients with schizophrenia or Parkinson’s disease. In the last two to three years, controlled trials evaluating risperidone (Risperdal) and olanzapine (Zyprexa) in patients with dementia have been published. There is now evidence that these two atypical antipsychotic drugs offer efficacy in this patient population with fewer adverse effect concerns than the typical antipsychotic drugs.

Risperidone (Risperdal). In 1999, Katz et al. published the first large multicenter, double-blind, placebo-controlled study of risperidone in treating psychosis and behavioral disturbances in an elderly demented population. Among the 625 patients, 73% had a diagnosis of Alzheimer’s disease; average age was 83 years; and their mean baseline Mini-Mental State Examination (MMSE) score was 6.6+6.3, indicative of the most severe stages of dementia. In this 12-week trial, patients received either placebo or risperidone 0.5 mg, 1 mg or 2 mg daily. At endpoint, significantly greater reductions in Behavioral Pathology in Alzheimer’s Disease Rating Scale (BEHAVE-AD) total scores, as well as the psychosis and aggressiveness subscale scores were seen in patients receiving daily doses of 1 mg (p=0.005) and 2 mg (p=0.001) of risperidone compared to those receiving placebo. The 0.5 mg daily dose of risperidone was superior to placebo at week 12 in reducing BEHAVE-AD aggression scores (p=0.02). Improvement based upon total BEHAVE-AD scores for risperidone 1 mg or 2 mg was 56% in patients under 85 years old and 72% in older patients, while placebo response rate was 51% and 54%, respectively. While risperidone was clearly efficacious, the high placebo response rate indicates what Schneider (1999) described as “the waxing and waning and evanescence of disruptive behavior” in this patient population. The most common adverse effects from risperidone were motor symptoms, dose-related sedation and mild peripheral edema. Extrapyramidal symptoms did not differ significantly between placebo and risperidone 0.5 mg or 1 mg. Risperidone 2 mg daily produced significantly more parkinsonism and hypokinesia than placebo.

A 13-week study compared risperidone and haloperidol (flexible doses of 0.5 mg to 4 mg; 1 mg mean daily doses for both drugs) to placebo in 344 patients with dementia. At week 12, a reduction of 30% in the BEHAVE-AD total score was observed in 72%, 69% and 61% of patients receiving risperidone, haloperidol and placebo, respectively, and in 54%, 63% and 47% of patients by intention-to-treat analysis (no statistically significant difference). At endpoint, risperidone-treated patients did have significantly greater reductions in the aggressiveness scores compared to placebo. The most common adverse effects for both haloperidol and risperidone were sedation and falls, with haloperidol causing significantly more extrapyramidal effects.

Olanzapine (Zyprexa). In 2000, Street et al. published a large double-blind, randomized, placebo-controlled trial evaluating olanzapine in patients with Alzheimer’s disease. In this six-week trial, 206 patients were given placebo or a fixed olanzapine daily dose of either 5 mg, 10 mg or 15 mg. Mean baseline MMSE scores were 6.7+6.4, indicating severe dementia. Significantly greater improvements in agitation/aggression and delusions/hallucinations were observed in patients treated with olanza-pine 5 mg or 10 mg compared to placebo and olanzapine 15 mg. The most improvement was seen in patients treated with 5 mg. The total MMSE score did not change significantly over the course of the clinical trial, although there was a modest improvement in the 5 mg dose group and a modest decline in the 10 mg and 15 mg dose groups. Somnolence was the most common dose-related adverse effect, occurring in 25% to 36% of olanzapine-treated patients compared to 6% with placebo. Gait disturbance occurred in patients receiving 5 mg or 15 mg (20% and 17%, respectively), compared to 2% with placebo. There was no significant cognitive impairment, increase in extrapyramidal effects or central anticholinergic effects at any olanzapine dose compared to placebo. Vital signs and laboratory and electrocardiogram measures were unchanged in each dose group compared to placebo.

Discussion

The modest efficacy of conventional antipsychotic drugs in patients with dementia, coupled with their high potential for extrapyramidal effects, tardive dyskinesia, anticholinergic effects and worsening cognition has meant that many patients’ behavioral symptoms and psychosis have been inadequately treated. The atypical antipsychotic drugs, notably risperidone and olanzapine, now have evidence supporting their efficacy and lower potential for adverse effects. Daily doses of no more than 1 mg to 2 mg of risperidone or 5 mg to 10 mg olanzapine can provide significant therapeutic benefit without the risk of significant adverse effects compared to conventional antipsychotic drugs.

Question: What’s the latest on quetiapine and slit lamp examinations?

Dr. Petty: As many of you will know, quetiapine was associated with the development of cataracts in dogs during the clinical studies and there have also been case reports of humans getting cataracts. My own view is that probably it’s no higher than you would anticipate in a population that chronically abuses alcohol and cigarettes, both of which are known to predispose the development of cataracts. The legal situation at the moment has actually been slightly softened in that you can also do direct ophthalmoscopy so long as you are familiar with how to do it in order to pick up early cataracts. In general, the company is still recommending slit lamp examinations on patients on quetiapine at initiation of treatment and every six months thereafter.

Question: Why isn’t AIDS or HIV infection addressed on the risks of depression?

Dr. Gorman: Actually, for a very good reason, and we did a lot of research about this. As far as we can tell, and there’re many studies of this now. Believe it or not, the actual risk or incidence of major depression in people who are HIV-positive is not greater than people who are HIV-negative when the studies are properly controlled. There really is no evidence that there’s an increase in major depressive disorder as a diagnosis in HIV-positive people compared to HIV-negative people. On the other hand, now with the protease inhibitors available, there are some implications for which antidepressant medications to select when you do want to treat an HIV-positive or AIDS patient with antidepressants. I won’t go through all the drug-drug interactions, but its very important when you’re going to initiate antidepressant treatment for somebody on a protease inhibitor to be sure you know exactly which one they’re on and which antidepressant you’re going to pick, because there are some potentially important interactions there.

Question: What do you think about using Parlodel (bromocriptine) for bringing down prolactin levels?

Dr. Petty: Not much. It is something that you can do. There are two drugs that you can use in this indication, one is amantadine, the other is bromocriptine. The difficulty particularly with bromocriptine is that you can exasperate psychosis with it. It’s not common, but it can occur. I also always feel a bit uneasy about using one drug to chase another one if I can possibly avoid it. So I prefer to use one of the prolactin-sparing agents if at all possible. But it can be done. Certainly, I have done it on occasion when it’s been necessary to, as in the old days when we didn’t have the new prolactin-sparing agents. But it’s not what I would recommend. Most of us try to be therapeutic minimalists and use as little as medication as possible with our patients. I would attempt to use prolactin-sparing agents in preference, but you can use bromocriptine and amantadine to bring down prolactin levels.

Question: Does a decrease in REM sleep occur with antidepressant treatment? Is it a problem and if so, how is it treated?

Dr. Gorman: That is a very interesting, complicated issue. You may know there was a study comparing nefazodone to fluoxetine in terms of their effects on sleep. And that was interpreted by the people who wrote it to mean that nefazodone improves sleep quality, but fluoxetine decreases sleep quality. If you actually look at the data, there were very subtle changes on EEG-measured sleep. If you actually asked the patients did they notice any differences, they didn’t. And all of the patients in both groups said that their sleep got better. So there’s a general finding that regardless of what the effect actually is on polysomnographic measures of sleep, they really don’t make that much difference in terms of response or overall outcome. Second of all, there’s another factor when you give SSRIs because the neurons in the median raphe area of the brain that are responsible for manufacturing serotonin are also part of the dream-inducing mechanism of the brain. For example, in cats, if you stimulate that part of the brain, the cat goes into REM sleep automatically. Some people call it the dream machine. So what you’ll often see, paradoxically, is that patients on SSRIs will say after a few weeks that they’re having very vivid dreams and more dreams. And if you’re doing a psychoanalytic treatment with the patient on SSRIs, it’s really great because you get some of the most wonderful dreams. I think that’s why Peter Kramer likes Prozac so much because he’s really a psychoanalyst and enjoys listening to all those increases in dreams. So, in general, we treat insomnia in patients but we don’t treat changes in REM sleep, because they don’t seem to have any real clinical significance as far as we now know.

Question: Should you start withdrawing the conventional drug as soon as you start the new one?

Dr. Petty: I honestly think the best thing to do is to have people on both drugs simultaneously, then gradually reduce the older drug over a period of time. I mentioned as I was talking earlier on, three to four weeks for the older convention agents, probably six to eight weeks for risperidone, and four months for clozapine. The reason being that there is this discontinuation syndrome, which I talked about earlier on, which is very real. It’s not as well described as the SSRI discontinuation syndrome, but it is for real. The very concrete question of how fast do you then reduce the older one? I just take it down in straight increments. So if somebody’s on 40 milligrams of haloperidol, I will reduce it by 10 milligrams a week. This seems to be a safe way to do it and this is what most of the experts in psychosis and most of the pharmacologists are now recommending. As I said earlier on, the silly thing is to simply stop the other drug because you will run into trouble and so will your patient very rapidly, and it’s very foolish to do that. So if you just gradually just have people on two drugs at once.

Question: What do you think about polypharmacy and using two drugs at once?

Dr. Petty: I think that we sometimes have gone a bit too far in trying to be purists and say, only one medicine. I’d like to just think about something for a moment. If you saw a patient in the 1950s with tuberculosis and you said, “What treatment are you on?” and they just said, “It’s Ambutol, you’d say, “And what else?” These days if you had somebody with AIDs, and you said, “What treatment are you on?” and they said “AZT.” You’d say, “And what else?” Because we learned this lesson many years ago, that you often need more than one medicine. Except in psychiatry we’re trying to be purists. We’re all thinking all the time, “Only use one drug.” I prefer to use only one drug, but there are situations where you need to use more than one. And certainly in another question that I’ll come to later on. In the acute situation with psychosis, I see nothing wrong at all with using an atypical and for the first few days using one of the older medications that have more sedative potential. That’s absolutely right and proper. There are many combinations that we have used. You just have to be very smart about understanding drug-drug interactions. And I think certainly my view is that possibly we need to do the same in depression. I wonder what you think, Dr. Gorman?

Dr. Gorman: First of all, I think we should abolish the term, polypharmacy. If the cardiologists and nephrologists can use two drugs then certainly we have to. Because the heart and the kidney are very primitive organs compared to the brain. And the idea that we could really make an impact on an organ that has 100 billion synapses or something like that with one agent is really naпve. There’s data now in bipolar illness that it’s actually quite rare that anyone treats a bipolar patient with only one drug. Most bipolar patients, at least at some points in their lives, are also taking antipsychotic drugs or antidepressant besides taking lithium or valproate. And that’s usually absolutely appropriate and the right thing to do.

Question: Please advise on SSRI increases in prolactin levels.

Dr. Gorman: I actually think there’s not enough research in that area. If you give a serotonin receptor-stimulating drug like fenfloramine, which we can’t use anymore, you do see from an interesting pharmacological mechanism an increase in prolactin level, acutely. So stimulating the postsynaptic serotonin receptor results in an immediate increase in prolactin. However, in SSRI treatment, many of those same receptors actually get down-regulated over time. In general, symptoms secondary to actual prolactin increase are not a major problem and certainly not anyway near the order of magnitude that they are with antipsychotic drugs like the typical drugs and to some extent risperidone. Nevertheless, there are patients who get breast tenderness, galactorrhea from some of the SSRIs. And it’s not even clear if they’re all the same in that regard, some may be different from others. I think that one thing that would actually be important is to study that prospectively more than it’s ever been done.

Question: Halodol seems to be superior to Zyprexa for treating agitation. Please comment on treatment of agitation with atypicals.

Dr. Petty: Yes, I think one of the difficulties that we’ve had with the currently available atypicals is that none of them is sufficiently sedative if you’ve got a very agitated patient in the acute phase. On my own unit at Penn, what we have been doing routinely is that we have clinical pathways. I have now been using olanzapine as my first-line agent. The residents actually call it Vitamin O. But when a patient is agitated, we also give them chlorpromazine for the first few days. I prefer chlorpromazine for a number of reasons, I’ve been using it for more than 20 years. But again, as Dr. Gorman was saying just now. I see absolutely nothing wrong in that situation. We use it for five days. I prefer not using benzodiazepines in acutely psychotic patients because of the relatively high risk of paradoxical effects with them. So this is what we tend to do, actually go and use chlorpromazine together with olanzapine, but we’re always very ready to take it off again after a few days. And that seems to work very effectively and certainly we’ve been doing that for the last 2 Ѕ years. It’s worked with us just fine. I should just mention that there are injectible versions of the atypicals, certainly of olanzapine and ziprazodone in clinical development at the moment and they look like they may actually be pretty effective. The second part of this question actually relates to that and says, “Loxitane seems to have less EPS. Is it an atypical?” Well, some of you who wrote the APA would know that there was a symposium, chaired by Bill Glasser, asking that very question. I would have to say that from my read on the data, it really is not. But again, I don’t really like the term ‘atypical’ because as the new debate is that maybe loxitane the old studies were using too high a dose and if you used a low dose then you don’t get EPS, but you also don’t get much clinical effect. So I really do not think that it’s an atypical.

Question: Dr. Gorman repeatedly uses the term ‘clinician.’ What does that term include?

Dr. Gorman: I usually use the word ‘clinician’ in this context to distinguish between people who actually see patients in real-life settings and people who are seeing patients only in the setting of research clinical trials. There really are legitimate concerns these days not just in psychiatry, but in many branches of medicine. It’s not that we don’t need clinical trials, but they only measure what we call efficacy. They can only tell us in a relatively short period of time if a highly select group of patients who meet all the inclusion/exclusion criteria have a better response to an experimental drug than they do to a placebo or reference drug. And that’s all those trials can tell us. There’s another whole world out there that we call effectiveness which is now that we know this drug is better than placebo in the rare patient with panic disorder who isn’t also depressed and doesn’t drink, etc., etc. Are these drugs any good? Do they really work? Do people actually take them? Can they take them for any length of time? Are they too expensive? Do they require more or less monitoring? I think people have seen the interesting situation with Seroquel, which seems to be a pretty good antipsychotic drug, but the fact that you need to get an eye examination, it does not change its placebo-controlled response in clinical trials, but makes a tremendous difference in terms of whether clinicians can actually use that drug. Because it’s not always easy to accomplish to get that examination accomplished with patients who are in mental health centers and things like that. So that’s the distinction we’re making. We’re really saying that drugs need to be evaluated in real-life clinical settings, not just in the research settings.

Question: Considering the antidepressant effects of novel antipsychotics is a risk of medication induced switch to manic phrase in patients with schizoaffective disorder or bipolar?

Dr. Petty: This has been something that has been of great concern to many of us for some time now. And certainly thought originally that I had patients I had made more manic when I started using olanzapine. Because initially when it came on the market and we had it, I was using it extensively on all sorts of patients to get more hands-on experience, as one should do, I think, with a new medicine, and certainly if you’re in an academic center. I have been less convinced about that as time has gone by, and certainly, I was instrumental at asking Lilly to re-look at some of their own data. Because initially when they were looking at the mood changes when they were initiating treatment, they were actually looking at them after quite long periods of time. They’ve now looked and are doing monitoring of studies every four to six hours over the first few days. And it certainly is the case that some patients do get a little bit of agitation, but not full-blown mania. I think what I was talking about before and I was probably incorrect and I was looking at agitation rather than mania. This incidentally has also been described with risperidone and clozapine. So all of these drugs can cause agitation. I know I mentioned earlier on the anxiolytic activity of risperidone, but it certainly can happen. I don’t believe it’s the case we need to worry about a lot, just be aware that it can happen.

Question: This is a two-part question that asks about the treatment of major depression with psychotic features, and also asks which group of antidepressants has lower potential for secondary mania.

Dr. Gorman: We still firmly believe that the treatment of psychotic depression should be combination therapy of an antidepressant and an antipsychotic drug. I accepted two papers in the American Journal of Psychiatry from Italy which both seemed to show that SSRIs alone were effective in the treatment of psychotic depression. A very difficult decision to accept those papers because the studies were well done. None of the reviewers believed them. But they all seemed to be very well done. We were nervous about not publishing them because we don’t want to suppress information that might lead to further studies. We were nervous about publishing them because we don’t want clinicians to think that’s it’s been absolutely proven and to withhold antipsychotic medication because after all, psychotic depression is probably the most lethal form of depression that we have and the most likely to result in suicide. And also these studies cannot be placebo-controlled because no one is going to take the risk of putting anybody on a placebo with psychotic depression anymore. I’m not entirely convinced by those studies either and there’s a lot of debate about whether the diagnoses were correct, so for now we think that it’s got to be two medications. Now the next part is “Are SSRIs as good as tricyclic antidepressants and even are atypical antipsychotics as good as typical antipsychotics for that particular disorder because to my knowledge, the only real studies of any consequence that have been done in the study of delusional or psychotic depression involve the combination mostly of Trilafon and amitriptyline, and they’re older studies. There are not any really good studies of the SSRI plus an atypical. Most people use that combination and it seems to work, but we don’t have any systematic data. Most patients with psychotic depression are going to get hospitalized and we recommend now combining an atypical antipsychotic with an SSRI in those patients, but you should know that the data for that are not entirely secure. In terms of which group of antidepressants are less likely to induce mania, again, that is not a properly studied area at all. There is an idea in the field that bupropion is less likely to produce mania than other antidepressant drugs. That’s probably not true. The study that we’ve seen recently, I don’t know if it’s published yet, actually compared I think it was paroxetine to bupropion for the treatment of bipolar patients in the depressed phase. And it actually did show that buproprion was less likely to induce mania than paroxetine, but that was because buproprion was less likely to make the depression any better. So obviously buproprion was not a very good choice in that situation. Tricyclics and MAO inhibitors clearly have the potential to induce mania. SSRIs clearly induce mania. I would say you should pick the drug that you think is best for the patient and watch for mania and treat that and not rely on any class being any better than any other class at the present time.

Question: How long should a patient with paranoid schizophrenia keep taking antipsychotics? For life or what?

Dr. Petty: If you have a first-episode illness, treatment should be continued and these are the APA guidelines, for one year. If they have had two or more, it is five years. If they’ve have had more than that, it is recommended it is going to be lifetime therapy. Nobody wants to commit somebody to a lifetime treatment and that’s why it is so important to get the diagnosis right at the beginning, and why we need to be so careful.

Question: Does the risk of Parkinson’s justify continued treatment?

Dr. Petty: The take-home message is that the risk of Parkinson’s is so much lower with quietiapine and olanzapine. It’s also lower with low doses of risperidone, but again if you hit 5 or 6 milligrams, it begins to rise quite rapidly. So it’s not so much of an issue as it was. So one episode–one year; two or more-five years; more than that-lifetime.

Question: What antidepressant is best for the depressed patient with AIDS or HIV-infection?

Dr. Gorman: There are few studies of this. One controlled study showed tricyclics worked very well and were surprisingly well-tolerated. Judy Rabkin’s group at Columbia had a nice controlled trial of fluoxetine and showed that it was very well-tolerated and very effective. And a group at the University of Houston, led by Francisco Fernandez, has shown over and over again that psychostimulants are very useful, although I don’t think those have been controlled trials. So a variety of antidepressants work. In terms of what’s the best, really it’s again, like any other depressed patient, you’re picking based on the actual presentation of the depression, and in this case, very carefully looking at potential drug-drug interactions, because there are important drug interactions between protease inhibitors and some antidepressants. But for example, if a patient with HIV infection is not sleeping, one would want to pick an antidepressant that might help them sleep. If their appetite’s not very good, one would perhaps want to pick one that increases appetite. We’re usually picking from among managing various side effects, but as far as we know, the only really good studies have shown that fluoxetine and desipramine are effective.

Question: Can you comment on atypicals, e.g., risperidone to augment severe refractory OCD?

Dr. Petty: There are some clinical trials looking at some atypicals on their own. Many experts have actually also been trying to use atypicals, all of them, in fact in the treatment of OCD. Olanzapine may be a shade better than risperidone, but as I say, this is anecdote, it has not been formally examined. It is certainly possible to use them in severe OCD. I have done it several times. I have one particularly difficult patient who’s actually been in inpatient care for six years, whose life has changed dramatically when we took him off the 80 milligrams a day of haloperidol he was on, in addition to fluvoxamine and all the rest. He’s now on eight milligrams, instead of 80, together with 20 of olanzapine and the symptoms are considerably better on that regimen.

Question: What happens if SSRIs don’t work?

Dr. Gorman: It is reasonable to try clomipramine if two SSRIs have not been effective, Sometimes a patient will respond to that who didn’t respond to an SSRI. In terms of augmentation, almost every augmentation attempt when studied has been a failure. Lithium has been shown not to be effective. Buspirone has not held up in controlled trials as an effective augmenter in OCD patients or SSRIs or clomipramine. And as Dr. Petty mentioned, the only thing that actually has been shown to work is the addition of antipsychotic medication. But really only in two situations. Antipsychotic medications enhance antidepressants in OCD patients who also have tics, or in OCD patients where the obsessional content is almost delusional. In those two cases, they work. In the tic situation, both pimazide and haloperidol have shown to be enhancing. In the delusional area, I think the atypicals have proven to be the best. Those are what have been shown, in addition to that, we have an interesting paradox which is that patients with schizophrenia treated with atypical antipsychotic agents, we sometimes see the new onset of OCD symptoms and that’s believed to be secondary to the blockade of the 5HT2 receptor. However, in OCD patients who have not had a complete response to clomipramine or an SSRI, we often see improvement when we add risperidone or olanzapine. No one knows exactly why that is, but the first thing we usually try now in a refractory OCD patient, after trying at least two SSRIs and clomipramine, is to add olanzapine or risperidone to their regimen and that seems to work for a surprisingly large number of patients.

Question: How often do you monitor your patients and how often do you see them?

Dr. Petty: As often as is necessary. We try and spend as much time a night, make a lot of use of other people who are treating them-ICMs and family, in particular, to try and see them. In the initial stages, try to see patients once a week. I am also quite allergic to infantilizing patients with psychotic illnesses. I think it is a serious mistake to treat them like children, because they usually respond by behaving like children. So I try to give them some autonomy. But also make quite sure they know where to come and what to do if they run into any kind of trouble.

In essence, the currently available atypicals are “5-HT₂/D₂” antagonists, although several of them (e.g., clozapine and olanzapine) affect a variety of other receptor areas. Clozapine appears to have affinity for both 5-HT_2C and 5-HT_2A receptors. Differing from both these types of agents is a group of drugs known as selective 5-HT_2A antagonists. Although they differ as to their specificity for 5-HT₂ receptor subtypes, this group of agents shares the property of having essentially no dopamine receptor blockade.

Setoperone, ritanserin, mianserin, ketanserin and the investigational agent M100907 are all in this class, however, pharmacodynamic differences exist among them; e.g., ritanserin antagonizes both 5-HT_2A and 5-HT_2C receptors, whereas M100907 is a pure 5-HT_2A antagonist. There is growing evidence that the antiserotonergic effects of these and currently available atypical agents contribute to improved efficacy and safety in the treatment of schizophrenia.

Answer. I appreciate how confusing some of these medication issues can be for parents and families of individuals with schizophrenia. Terms like “older” and “newer” can be confusing when applied to medications for psychosis or mood. For example, how “new” or “old” an antipsychotic is has more to do with marketing and approval processes than with how effective or safe it is.

Generally speaking, clinicians distinguish two broad groups of antipsychotic agents: older “typical” agents, also called neuroleptics, and the newer “atypical” agents. The latter group includes clozapine (Clozaril), the first atypical released; risperidone (Risperdal), which came out next; and then olanzapine (Zyprexa) and quetiapine (Seroquel). As a group, the atypicals are generally more effective for so-called negative symptoms of schizophrenia (apathy, social withdrawal), and are less likely to cause neuro-muscular problems like muscle spasms, stiffness, and abnormal movements (tardive dyskinesia).

However, the atypical agents are not without side effects. On rare occasion clozapine can cause a serious white blood cell problem called agranulocytosis, and it does tend to promote significant weight gain. On the other hand, many clinicians believe clozapine to be the most effective agent, old or new, for severe, refractory schizophrenia. Several studies do suggest clozapine is somewhat more effective than risperidone, though risperidone tends to have fewer side effects (such as weight gain). Olanzapine, as you note, is closely related to clozapine, but it is not yet known if it is as effective. Though olanzapine is generally better tolerated than clozapine and does not cause the white blood cell problem, it can also contribute to weight gain.

Regarding your son, many clinicians would argue that a failure to respond well to risperidone justifies a direct move to “the gold standard,” clozapine. Others (probably including me) might give olanzapine a try first, then go to clozapine. But this isn’t a matter of old vs. new as much as a matter of balancing risks and benefits on a case-by-case basis. The same applies to Paxil (which, compared to many agents on the market, is actually quite “new”). By the way, sometimes the older agents (such as haloperidol) are used to augment newer agents, as seems to be the case with your son.

To summarize, the changes in medication you have described aren’t “wrong,” but your son’s response is hard to predict. Clozapine certainly gives him a very good chance of recovery. The other atypical agent now available, quetiapine, has not really been compared in good studies with clozapine; my own impression is that quetiapine is a useful agent, but “no clozapine.” We are expecting yet another atypical agent, ziprasidone, in the next 6-10 months, and this agent appears to cause little if any weight gain. It may also have antidepressant properties, and could be useful for your son. But it’s not at all clear that it will out-perform clozapine.

Answer. I am not aware of published studies of olanzapine (Zyprexa) specifically for the treatment of depressed subjects. As you know, this agent is FDA-labeled for use in psychotic disorders, mainly schizophrenia. However, Lilly Research Lab does have data on file showing that olanzapine led to significant improvement on the Montgomery-Asberg Depression Scale (MADRS) in a mixed sample of patients (schizophrenic, schizoaffective and schizotypal). This improvement was statistically greater than that seen with the comparator drug, haloperidol.

In my own experience, I have noticed some benefit from Zyprexa (Olanzapine) in patients with mixed psychotic-depressive states (as is also seen with risperidone and clozapine, in some cases). Olanzapine’s effects in manic patients are unclear. Lilly has data showing reduction in manic scores, but I am aware of several cases in which olanzapine seems to have induced excitation/agitation in some patients. Clearly, we need those double-blind controlled studies!

Question. I have been trying to deal with my depression, anxiety and posstraumatic stress disorder for 10 years. About nine months out of the year, I feel low-grade depressed, but I am more or less functional. Two or three times a year, I get swamped with episodes of major depression, hopelessness, sleep panic attacks, unreasonable fears and intense suicidal ideation. I become essentially nonfunctional for about three or four weeks. Eventually, with the help of my psychotherapist, I get back to normal. Most of the time, psychotherapy is all I need. Is there a treatment I could use only during a severe depressive episode?

Answer. If I understand you correctly, it sounds as if you have at least two or three major depressive episodes per year, though they appear to be relatively brief. There also seems to be a PTSD component to your picture, but even during the nine months when you are not severely depressed, you experience low-grade depression. Although this may not be the response you want to hear, I would strongly consider retrial on an antidepressant, and staying on it indefinitely. Otherwise, the likelihood of recurrent bouts of depression, which are potentially life-threatening, is high. If, in addition, you periodically develop delusions or other features of psychosis, the periodic or ongoing use of an antipsychotic agent should also be considered. This does not mean, of course, that psychotherapy is any less important. Obviously, it has been critical for your well-being, and it should surely continue. Unless the depressive storms recur at a predictable time each year (e.g., every winter) it would not be effective to begin an antidepressant only when you feel severely depressed. That’s because it usually takes two to five weeks for an antidepressant to become fully effective. By that time, you would probably be through the episode.

Rather, I would consider staying on a relatively nonsedating antidepressant indefinitely or at least for the next three years, to see how you do. Wellbutrin is a good, nonsedating antidepressant, but it doesn’t have much effect on anxiety. However, it could be combined with a low dose of Depakote, which is quite good for anxiety and perhaps for some aspects of PTSD. Effexor, beginning with very low doses (25 mg/day), might also be a good option. Low-dose Risperdal (2 mg to 3 mg/day) might be useful if you tend to develop delusions with depression. This could be added as needed if you become delusional once or twice a year. However, if it happens more frequently, it might be more useful to stay on Risperdal in combination with, say, Effexor. Zyprexa is another agent that may help with both depression and psychosis, but it is quite sedating (not quite so sedating as clozapine, though). I would suggest discussing all these options with both your therapist and an experienced psychiatrist with good knowledge of medication issues. Good luck.