Antipsychotic drugs linked to increased risk of death for some elderly Alzheimer's patients

Some newer antipsychotic medications may be associated with a small increased risk of death when used to treat elderly dementia patients, psychiatrists at Johns Hopkins warn.

In an editorial published in the Oct. 19 issue of the Journal of the American Medical Association, Peter V. Rabins, M.D., M.P.H., and Constantine G. Lyketsos, M.D., M.H.S., professors of psychiatry at The Johns Hopkins University School of Medicine, cautioned that clinicians should consider both the risks and benefits for patients suffering from dementia, such as Alzheimer's disease, and when possible delay prescribing so-called second-generation antipsychotic medications for patients who exhibit psychotic symptoms or aggression.

A study of the effects of these drugs, including aripiprazole, olanzapine, quetiapine and risperidone, in elderly patients with dementia, led by Lon S. Schneider, M.D., M.S., of the University of Southern California, Los Angeles, is featured in the same edition of JAMA.

The authors of that study reviewed all available published and unpublished randomized placebo-controlled, parallel-group, clinical trials of the second-generation antipsychotic drugs marketed in the United States to treat patients with dementia.

They concluded that the patients taking second-generation antipsychotic medications were 1.5 times more likely to die than patients taking a placebo.

"These results," said Rabins, "do not suggest that first-generation antipsychotic drugs like haloperidol and chlorpromazine, introduced in the 1950s, are safer alternatives to second-generation drugs."

He said first-generation antipsychotic drugs have their own set of adverse side effects, such as Parkinson's disease-like symptoms and low blood pressure.

"We do not believe the findings contraindicate the use of antipsychotics for patients with dementia who have psychotic symptoms and agitation, but rather that they change the risk-benefit analysis such that antipsychotics should be used only when the patient's symptoms present an identifiable risk to the patient or to others, when the distress caused by the symptoms is significant, or when alternative therapies have failed and symptom relief would be beneficial," Rabins said.

Rabins said antipsychotics should not be used when other treatments are available and the risk of harm or significant distress is low. He said a range of alternative treatments, including behavioral interventions and antidepressants, have proven to be effective in some cases.

In the study, Schneider and colleagues examined 15 trials, generally 10 to 12 weeks in duration, in which 3,353 patients were randomly assigned to take a study drug and 1,757 were randomized to placebo. Outcomes were assessed using standard methods to calculate risk differences.

Rabins cautioned that the short-term data used for Schneider's study might not be accurate for what happens over longer time periods because adverse drug reactions could be more prominent in the early stages of a course of medication treatment.

He noted the possibility that higher death and illness rates might exist among frail individuals exposed to other classes of drugs, but the absence of long-term data limits the ability to study this question.

"We look forward to international efforts to improve long-term monitoring for adverse events," Rabins said, "and for research into the important questions raised by this study."