Atypical antipsychotic drugs slightly increase risk for death when used in people with dementia

Atypical antipsychotic drugs seem to confer a small increased risk for death when used in people with dementia, concludes a team of researchers from the Keck School of Medicine of the University of Southern California in a meta-analysis of 15 clinical trials published in the October 19 issue of the JAMA: The Journal of the American Medical Association.

Despite this risk, says Lon Schneider, M.D., professor of psychiatry, neurology, and gerontology at the Keck School and the USC Andrus School of Gerontology, physicians, families and patients need to keep in mind that psychosis itself is a very serious issue in dementia. "Aggression, hallucinations and delusions in dementia patients can also shorten a patient's life, and result in poor care and rapid deterioration," Schneider says. "It's a difficult problem with no easy answers."

Led by Schneider, the USC researchers analyzed the results of the 15 trials--nine of which are unpublished--to determine whether there was a correlation between the use of these second-generation drugs (the atypical antipsychotics) and an increased risk for death.

Within the 15 trials, four different atypical antipsychotics were assessed and compared with placebo: aripiprazole, olanzapine, quetiapine and risperidone. The trials looked at a total of 3353 patients who took the antipsychotics, as well as 1757 who had been given placebo, and lasted on average between 10 and 12 weeks.

In the final analysis, more patients taking an atypical antipsychotic died during their trial than did patients taking placebo. There were 118 deaths associated with atypical antipsychotics (3.5 percent of the drug arm of the trials) versus 40 deaths associated with placebo (2.3 percent of the placebo arm of the trials). The difference between those two groups was statistically significant, and resulted in an odds ratio of death on antipsychotics versus placebo of 1.54 to 1; in other words, the risk of death is 1.54 times greater for people taking antipsychotics than those who are not. In only three of the 15 studies was there no increased risk found for the antipsychotic group versus the placebo group.

When the specific drugs were examined individually, there was no significant difference between the risk conferred by one drug over another. And, the researchers note, this increase in risk "could not have been recognized by examining any individual trial. The events were too sparse and the trials too small to be able to meaningfully assess for a dose response that might make attribution even more compelling."

The results of this meta-analysis agree with an April 2005 Food and Drug Administration health advisory which stated that there was "an approximately a 1.6Ð1.7 fold increase in mortality" in studies of these drugs, with most due to either heart-related effects like heart failure or sudden death, or infections such as pneumonia.

There are, of course, caveats regarding these results. For instance, most of the studies were relatively short; arguments can be made that the risk might be either higher or lower if looked at over a longer term. And, as Schneider points out, this may not be a risk limited to the atypical antipsychotics. "These are mostly nursing home patients who are frail and medically ill," he points out. "If the same sorts of analyses were done with other drugs--antidepressants, anxiolytics, mood stabilizers, even cardiovascular or antihypertensive drugs--we might see something similar as well. Drugs have side effects, and even as they help, they harm."

For this reason, Schneider says he's concerned that these results will be taken out of context and that there will be an overreaction to the stated risk. "For instance," he says, "it would be wrong to say that because of this effect we should use other medications. It's only because the antipsychotics are the best studied that there were enough patients to be able to see this slight increase in risk of death. The truth is that evidence about the effectiveness of other medications is extremely thin, as is evidence of their safety. If we switch to these medications, we might be jumping from the frying pan into the fire without knowing it."

What should be done with this information, then? "Clinicians and families can use this information to better and more cautiously judge the need for antipsychotics in dementia patients," Schneider says. "It would be good for us all to be a little more cautious, and to stop using the medications if there doesn't seem to be a clinical effect."