Nearly one half of patients with Alzheimer's disease may experience psychomotor agitation. Untreated psychosis and disruptive behaviors are distressing to both patients and caregivers, and may force institutionalization. A meta-analysis has suggested the moderate superiority of neuroleptic agents over placebo in the treatment of psychosis and agitation in patients with dementia. However, neuroleptics are poorly tolerated by patients with Alzheimer's disease. Devanand and associates performed a randomized, double-blind, placebo-controlled trial of haloperidol and placebo to determine the optimal dosage of haloperidol to treat patients who have Alzheimer's disease with psychosis and disruptive behaviors.

During the six-week trial, 71 outpatients with Alzheimer's disease were randomized to receive a “standard” dosage of haloperidol (2 to 3 mg per day), a “low” dosage of haloperidol (0.50 to 0.75 mg per day), or placebo. The standard dosage range was toward the high end of dosages used clinically but was also below the dosage of 5 mg per day that often results in severe extrapyramidal signs. In a subsequent six-week cross-over phase, patients who were taking haloperidol were switched to placebo, and those who were taking placebo were switched to haloperidol.

Analyses indicated that in all measures of efficacy, therapeutic effects were greater with the standard dosage of haloperidol (55 to 60 percent rate of response) than with either the low dosage of haloperidol (25 to 35 percent rate of response) or placebo (25 to 30 percent rate of response). Significant differences were seen in patients taking the standard dosage in the reduction of psychosis and psychomotor agitation, but not in physical aggression and hostility/suspiciousness.

Extrapyramidal signs tended to be greater with the standard dosage of haloperidol than with the low dosage or placebo. Twenty percent of the patients in the group taking the standard dosage developed moderate to severe extrapyramidal signs.

Forty-nine patients entered the cross-over phase. Patients who switched from placebo to standard dosage therapy showed nonsignificant superiority to the low dosage group in physical aggression, psychomotor agitation, hostility/suspiciousness and psychosis. The patients first treated with haloperidol who switched to treatment with placebo demonstrated a partial reversal of the improvement that was noted in the acute phase. Patients who switched to placebo after six weeks of standard-dose haloperidol displayed worsening symptoms, indicating that a longer period of treatment is necessary before discontinuation of medication should be attempted.

The authors conclude that the use of 2 to 3 mg per day of haloperidol leads to an acceptable trade-off between efficacy and side effects for psychotic or agitated patients with Alzheimer's disease, except in those who develop moderate to severe extrapyramidal signs. Dosages below 1 mg per day are ineffective. There appears to be a narrow therapeutic window for patients with Alzheimer's disease who are treated with haloperidol that may also apply to other neuroleptic agents. Since many patients who received 2 to 3 mg per day of haloperidol developed moderate to severe extrapyramidal signs, a starting dosage of 1 mg per day with gradual upward dosage adjustments is recommended. Close monitoring is needed to ensure the efficacy and safety of the treatment.