Atypical Antipsychotics: A Practical Review

Donna A. Wirshing, MD, William C. Wirshing, MD, Stephen R. Marder, MD, C. Scott Saunders, MD, Elizabeth H. Rossotto, PhD, Stephen M. Erhart, MD


Medscape Psychiatry & Mental Health eJournal. 1997;2(5) 

In This Article


Olanzapine is a thienobenzodiazepine antipsychotic that is a chemical analogue of clozapine with substantial affinity for 5-HT2a, 5-HT2c, 5-HT3, alpha1, D2, D4, M1, and H1 receptors (Table I). Among the newer antipsychotics, it most closely resembles clozapine in its structure and, to a lesser degree, receptor activity. Olanzapine has an elimination half-life of approximately 31 hours, and it is mainly glucuronidated after minor metabolism by the cytochrome P450 enzymatic system; thus, drug-drug interactions are unlikely to occur with this medication.

The largest study of olanzapine was an international double-blind comparison of olanzapine to haloperidol in 1996.[37] Olanzapine produced slightly greater improvement in total scores on the BPRS (P<0.02) and in negative symptoms on the Positive and Negative Symptoms Scale (PANSS) (P=0.03) than did haloperidol,[38] but it showed only a trend toward superiority over haloperidol on positive-symptom improvement (P=0.06).

Its effectiveness was also evaluated in a double-blind comparison of 3 dose ranges of olanzapine (2.5-7.5, 7.5-12.5, and 12.5-17.5mg/day); haloperidol (10-20mg/day); and placebo in 431 patients with schizophrenia.[39] The middle and highest doses of olanzapine and haloperidol were more effective than placebo on total BPRS as well as on positive symptoms. Both the low-dose and high-dose ranges of olanzapine (but not the middle range or any haloperidol range) were more effective than placebo on the Scale for the Assessment of Negative Symptoms (SANS).[40] These findings suggest that olanzapine is effective in the range of 2.5 to 17.5mg, but that there are advantages to using higher doses. No data are currently available on the safety and efficacy of doses higher than 25mg.

Olanzapine's efficacy in treatment-refractory individuals needs further study. The major efficacy study (N=1996)[37] included a large proportion of patients who had previous clozapine treatment (approximately 10%) and were judged to have failed their most recent neuroleptic trial (75%), suggesting that olanzapine may have utility in refractory individuals. In a recently completed and reported (but not published) study by Conley and colleagues,[41] 89 subjects densely unresponsive to haloperidol were enrolled in a protocol comparing olanzapine (25mg/day) to chlorpromazine (1200mg/day). Both groups improved to a clinically trivial degree (about 7%), with neither group demonstrating superiority over the other. As with the treatment-responsive study described above, olanzapine did show some superiority, albeit mild, in improving negative symptoms. Taken together, these studies suggest that olanzapine is, in fact, superior to conventional medications in certain subjects with schizophrenia, but the most densely unresponsive individuals are not the best population to target for treatment at this time.

Tollefson and associates[42] conducted a retrospective cohort analysis of 83 "first-episode" patients from the above described study[37] of 1966 patients. In spite of the very broad definition of "first episode" (<=5 years of illness), olanzapine demonstrated statistically significant superiority both in terms of tolerability and symptomatic improvement. Because first-episode subjects are known to be comparatively more intolerant of the neurotoxic consequences of conventional drugs, these results can be cautiously interpreted to mean that treatment-intolerant populations would be very reasonable targets for olanzapine treatment trials.

Olanzapine's utility in other psychotic illnesses is currently being explored. An open-label study of 15 psychotic Parkinson's disease patients reported that olanzapine was effective without worsening EPS. There are several studies under way utilizing the medication in bipolar illness, and geriatric and childhood disorders (Table II). In a reported but unpublished trial,[43] subjects with dementia of the Alzheimer's type and associated psychosis and behavioral disturbance were randomly assigned to either placebo or 1 to 8mg/day of olanzapine. This industry-sponsored trial enrolled 238 subjects and demonstrated that olanzapine was very well tolerated (virtually comparable to placebo across the board) in this highly sensitive geriatric population. Unfortunately, olanzapine was not different from placebo in the major target symptoms. These data suggest that higher-than-anticipated doses may be necessary in the geriatric subject.

Olanzapine is relatively free of serious side effects. The most common side effects are weight gain, mild sedation, and dizziness (Table III). In the study reported by Beasley and colleagues,[44] patients who received olanzapine did not show any increase in parkinsonism or akathisia. This finding suggests that olanzapine produces minimal EPS at doses up to 20mg. It induces only mild and transient, albeit dose-dependent, increases in prolactin secretion. It may therefore be an alternative for women who have irregular menses or galactorrhea due to the prolactin elevation from conventional drugs (provided the expected weight gain is managed prudently). It does cause an increase in liver transaminases, especially at the higher doses. This is usually, but not always, transient and of unknown long-term clinical pertinence. Much has been made of the in vitro binding affinity for olanzapine at the M1 receptor. These animal-model-derived in vitro data suggest that olanzapine would demonstrate substantial clinical anticholinergic toxicities in humans. In actual clinical practice, however, olanzapine causes very little constipation, xerostomia, accommodation disturbances, or urinary retention. This discrepancy underscores the known limitation of both animal-derived and in vitro (vs in vivo) data with respect to binding-affinity characteristics.

These data confirm that olanzapine is an effective antipsychotic with very low EPS liability and is relatively free of other adverse experiences. It is, in fact, so well tolerated that many clinicians have been tempted (and some have succumbed) to push the dose beyond those studied in a systematic fashion before marketing. Given the paucity of available safety data on doses over 25mg, extreme caution should be exercised when using higher doses. In such a dose range, more sedation, greater hepatotoxicity, more akathisia, and more weight gain would be expected.

The current dosing recommendation for olanzapine is that 10mg is a therapeutic dose and that no titration is necessary for the majority of patients. However, not all patients can tolerate 10mg of olanzapine. An elderly patient, for example, may find 10mg too sedating, and patients with extreme sensitivity to EPS may actually experience some EPS (usually akathisia) at this dose. We have treated several patients who have done well on olanzapine 5mg/day (including 1 clozapine nonresponder). Similarly, careful examination of the large (1996 patients) study[37] reveals that patients were judged clinically improved on all 4 dose levels of the medication (Fig. 1), and the most commonly prescribed dose was 20mg (31%). The average olanzapine dose in a trial comparing olanzapine to risperidone was 16.9mg.[45] Examination of all of the available controlled studies involving multiple doses of olanzapine shows that, in large part, higher doses are superior. The lack of a plateau in the dose-response curves elaborated to date means that we have not found the maximum effective dose. It may ultimately be determined that higher doses of olanzapine (eg, 30-50mg/day) are superior to the dose ranges tested to date. However, for the reasons elaborated above, such dose ranges cannot be recommended at this time.

Modal daily dosages of olanzapine and haloperidol in a double-blind study of 1996 patients.[37]

Taken in aggregate, these data clearly demonstrate that olanzapine has utility at doses well below and well above 10mg. A prudent dosing strategy is to begin at 5mg (except in the very fragile elderly, where 2.5mg should be the starting dose) and titrate up in 5-mg increments every 4 to 6 days. The titration should be terminated once clinical efficacy can be demonstrated or significant toxicity has emerged. While this strategy means that patients who ultimately require the higher dosages will be undertreated for some period of time, it does maximize safety, minimize toxicity, and perhaps, ultimately, engender the greatest compliance in patients.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: