Antipsychotics in AD Patients: The CATIE-AD Study

Abstract & Commentary

By Norman R. Relkin, MD, PhD, Associate Professor, Clinical Neurology and Neuroscience, New York Presbyterian Hospital, Cornell Campus. Dr. Relkin is on the speaker's bureau for Pfizer, Eisai, and Athena Diagnostics, and does research for Pfizer and Merck.
This article originally appeared in the February issue of Neurology Alert. It was edited by Matthew E. Fink, MD, and peer reviewed by M. Flint Beal, MD. Dr. Fink is Vice Chairman, Professor of Critical Care Neurology, NewYork-Presbyterian Hospital, and Dr. Beal is Professor and Chairman, Department of Neurology, Cornell University Medical College. Drs. Fink and Beal report no financial relationships relevant to this field of study.

Synopsis: Because of adverse effects, special care should be used when prescribing the atypical antipsychotic drugs for patients with dementia.

Source: Schneider LS, et al. Effectiveness of atypical antipsychotic drugs in patients with Alzheimer's disease. N Engl J Med. 2006;355:1525-1538.

Agitation and aggression are among the most dangerous and disruptive symptoms encountered in patients with Alzheimer's disease (AD). Hallucinations and delusions can be extremely disturbing to patients and care providers alike. Psychotic symptoms are often treated with medications, such as quetiapine (Seroquel®), olanzapine (Zyprexa®) and risperidone (Risperdal®). However, these agents were not specifically tested or approved in the context of psychosis in Alzheimer's disease. A double-blind, placebo-controlled, multicenter study called CATIE-AD (Clinical Antipsychotic Trials of Intervention Effectiveness — Alzheimer's Disease) examined the use of current-generation antipsychotics to determine whether these treatments offered real benefits relative to placebo effects in AD patients.

A total of 421 AD patients with an average MMSE score of 15 participated at 42 sites. All subjects had symptoms of psychosis, agitation and/or aggression at time of enrollment. Subjects were assigned either to placebo or to one of 3 antipsychotics at mean doses of 5.5 mg/d (olanzapine), 1 mg/d (risperidone), or 56.5 mg/day (quetiapine). Schneider and colleagues acknowledged that the dose of quetiapine used in this study may have been less than optimal.

The CATIE-AD study used an unusual "adaptive design" that was praised in an accompanying editorial in the New England Journal of Medicine as an exemplary method for evaluating the effectiveness of prescription medications. Participating physicians were not told which antipsychotic agent each patient would receive, but were given the opportunity to change or discontinue the medication at any time. The primary outcome measure was the time until the treating physician made a change in the antipsychotic medication for any reason. The rationale for this design was that physicians most often change medications, owing to lack of efficacy or side effects, and that change was thought to provide less biased indication of real-world clinical utility than measures such as psychosis rating scales. The study also used the Clinical Global Impression of Change (CGIC), a physician-based assessment commonly used in AD pharmaceutical trials, as a 12-week secondary outcome measure.

The primary outcome (time to discontinuation for any reason) was not significantly different for placebo than any of the 3 antipsychotics. The time to discontinuation for perceived lack of efficacy was greater for risperidone and olanzapine than for placebo, whereas quetiapine was comparable to placebo. No significant difference relative to placebo was reported on the CGIC for any of the treatment arms. Based on CGIC scores, 21% of placebo-treated patients were judged better at 12 weeks, compared to 32% on olanzapine, 26% on quetiapine and 29% on risperidone.

The overall rate of serious adverse events was lowest in patients receiving risperidone, but this difference did not reach statistical significance. Extrapyramidal side effects, including parkinsonism, were highest in patients receiving olanzapine (12%) or risperidone (12%), and only occurred in 1-2% of patients on quetiapine or placebo. Sedation was most problematic for patients on olanzapine (24%) and quetiapine (22%), but was also seen in patients on risperidone (15%), as well as a small percentage of those on placebo (5%). Cognitive disturbances (6%) and increased psychosis (7%) were most problematic in the olanzapine treated group, whereas increased agitation was reported in those receiving quetiapine (12%) or placebo (10%).

Schneider et al's conclusion was that the adverse effects of these 3 antipsychotic medications "offset" the small treatment advantages they offered to patients with psychosis and AD. They did not conclude that antipsychotics were ineffective, nor did they recommend against their use in AD patients. They acknowledged that, in light of the significant potential for adverse events, it may be prudent to use these antipsychotics in patients who demonstrate good tolerance and observable benefits.


Psychotic behaviors pose a real threat to the well-being of patients with AD and to those around them. Treatment-unresponsive agitation and aggression are among the reasons frequently cited for institutionalization of AD patients. The use of antipsychotic medications, as well as other psychotropic medications and non-pharmacologic interventions can be life-saving in some cases and can make the difference between a good outcome and a worst-case scenario.

While several criticisms can be made about design of the CATIE study, there is one major limitation that warrants special mention. The appropriate care of an agitated, psychotic AD patient is not simply a matter of writing a prescription for an antipsychotic drug. It is an interactive process that requires multiple types of parallel interventions, including identification and treatment of exacerbating physical precipitants, adjustments in psychotropic and non-psychotropic medications, attempted normalization of the sleep-wake cycle and nutrition, maintenance of a supportive environment, and appropriate caregiver education. Ideally, these interventions are made by physicians and allied health professionals, with expertise in this area, working as a team, with the cooperation and support of the patient's caregivers and family. However innovative the design of CATIE-AD may have been, it did not control for these critical co-factors.

Currently available antipsychotic medications may not be the drug of choice in every case, but they can play an extremely important role in the management of some AD patients with psychosis. The CATIE-AD study has been misinterpreted in the lay press as proof that novel antipsychotics lack efficacy and are unacceptably toxic in dementia patients. A closer look at the trial's design and actual outcomes contradicts these claims. The study's authors have publicly stated that these agents should still be tried in AD patients with psychosis and do work well in some cases. This study should not be accepted as justification for denying these medications to dementia patients with psychosis. CATIE-AD reminds us that all antipsychotics are potentially dangerous medications and that even the newest generation of these agents must be used judiciously.