Davangere P. Devanand, MBBS, MD

In Alzheimer's disease (AD), psychiatric symptoms were thought to be secondary and their assessment was based on traditional psychiatric scales used in schizophrenia and bipolar disorder. However, several symptoms identified in these scales rarely occurred in AD but several other symptoms were common. In the early 1990s, Dr. Davangere Devanand developed a scale, the Columbia University Scale for Psychopathology in Alzheimer’s disease, and used it in a multi-center study, Predictors of Severity of Alzheimer's Disease, that was led by Dr. Yaakov Stern. In 235 patients with mild to moderate AD, 64.3% had one or more psychiatric symptoms at baseline and only 8.5% of patients were free of all psychiatric symptoms during three years of follow-up. Agitation was common and increased with disease progression, aggression was uncommon in mild stages but increased with disease severity, delusions were present in 20% and hallucinations were relatively uncommon. A number of interesting findings emerged from long-term follow-up of this and subsequent cohorts studied under the same NIA-funded project, including an increased mortality risk associated with the presence of psychosis in AD. Other studies showed that the most common precipitants to institutionalization were agitation, wandering, incontinence, and psychosis (i.e., three of the four precipitants were psychiatric symptoms).

The FDA has not approved any medication to treat agitation or psychosis in AD, though risperidone is approved for these indications in Canada and Germany. Nonetheless, antipsychotics have been used widely in patients with AD and other dementias in the U.S. Dr. Devanand and colleagues initially examined this issue in an NIMH-funded clinical trial comparing a typical antipsychotic, haloperidol, at doses of 0.5-0.75 mg, 2-3 mg, and placebo. Haloperidol 0.5-0.75 mg was ineffective, and 2 to 3 mg daily was effective and superior to placebo but was associated with significant Parkinsonian side effects. Subsequently, studies with the newer atypical antipsychotics risperidone and olanzapine also showed a narrow therapeutic window with very low doses being ineffective and moderate to high doses being associated with significant side effects. Recent large electronic medical records database studies from Canada and the U.S. show that high doses of antipsychotics are associated with increased mortality, and dosage may largely account for the increased mortality risk associated with antipsychotic use in patients with dementia.

In addition to dose, duration of treatment with antipsychotics in patients with dementia has been a concern for regulators. In the U.S., since 1987 there has been a regulation that nursing home patients receiving antipsychotics should have these medications discontinued after 3 to 4 months. Antipsychotic discontinuation had not been studied systematically, so Dr. Devanand and colleagues conducted an 8-site NIA-funded multicenter trial, the Antipsychotic Discontinuation in Alzheimer's Disease (ADAD) trial, in which 180 patients received open-label risperidone for 16 weeks followed by randomization of 110 responders to one of three arms for 32 weeks in two 16-week periods: risperidone-risperidone, risperidone-placebo, placebo-placebo. The results, published in New England Journal of Medicine, showed a 2-4 fold increased risk of relapse following discontinuation even 4 to 8 months following treatment response. In the latest paper from this ADAD trial, appearing in the American Journal of Psychiatry, Dr. Devanand and colleagues examined which types of symptoms, both at baseline and after the initial 16 weeks in the ADAD trial, were related to an increased risk of relapse following discontinuation. The presence of hallucinations at baseline, primarily auditory hallucinations, was associated with increased risk of relapse (hazard ratio 2.96) after discontinuation of risperidone. Among patients with baseline hallucinations, 13 of 17 (76.5%) who discontinued risperidone relapsed, compared with 10 of 26 (38.5%) who continued risperidone (p<0.02). This group difference remained significant for severe (77.8%) compared with mild (36%) hallucinations. NPI domain scores after the initial open-treatment phase were not associated with relapse. The findings suggest that for patients with hallucinations, particularly auditory hallucinations, antipsychotic discontinuation after initial treatment response should be approached cautiously because of high relapse risk.

Dr. Ted Huey, Dr. Gregory Pelton and Dr. Devanand have also obtained interesting pilot data with low dose lithium to treat agitation in AD, as well as frontotemporal dementia (FTD). Dr. Devanand is currently leading a randomized, double-blind, placebo-controlled NIA-funded trial of low dose lithium to treat agitation in AD in a four-site study in the U.S. and actively enrolling patients with AD who also have symptoms of agitation.

Davangere P. Devanand, MBBS, MD
Professor of Psychiatry (in Neurology and the Gertrude H. Sergievsky Center)
dpd3@cumc.columbia.edu