Behavioral disturbances associated with dementia (BDD) are a common remediable cause of excess morbidity. Behavior such as agitation, aggression, delusions, hallucinations, anxiety, depression, shouting, and wandering can lead to significant impairment in the quality of life for both patients and their caregivers, as well as an increased risk of institutionalization. A recent review on the pharmacotherapy of BDD noted that, compared with other agents, the neuroleptics have been best studied with relatively rigorous placebo-controlled trials.⁺¹ Efficacy is modest, but concerns regarding side effects, especially extrapyramidal symptoms (EPS), have often limited their use. As a result, treatment of BDD with atypical neuroleptics such as clozapine or risperidone is potentially advantageous in view of their tendency to cause considerably fewer extrapyramidal effects. Unfortunately, there are no placebo-controlled trials of atypical neuroleptics for the treatment of BDD and very little anecdotal evidence for their use in this indication. To provide some preliminary information on the safety and effectiveness of risperidone for BDD, we report on our collective experience to date.

All patients met DSM-IV criteria for dementia. Specific etiological diagnoses were made clinically by academic geropsychiatrists. Twenty-two patients were included in this series, averaging 77.6±6.64 years of age (range 60—88). Of these, 73% (16 of 22) were males. There were 14 patients with dementia of the Alzheimer's type, 6 with vascular dementia, and 2 with Lewy body dementia (meeting consensus guidelines for the clinical diagnosis of dementia with Lewy bodies⁺²). The most common target symptoms included agitation, aggression, hallucinations, and delusions. All patients except 2 (#4 and #8) had previous trials of psychotropics, including at least one unsuccessful trial of a typical neuroleptic (ineffective or not tolerated). In fact, 16 of these patients had experienced serious EPS when treated previously with typical neuroleptics. This was the clinicians' most common rationale for the institution of a trial of risperidone. The patients, the treatment, and outcome are described in +Table 1.

The average therapeutic dose of risperidone used was 1.55±1.36 mg per day (range 0.5 mg qod to 3 mg bid). Only 2 patients received low-dose anticholinergic medication (e.g., procyclidine 2.5 mg bid). Length of trials varied from 1 week to 10 months. Risperidone was discontinued in 5 patients (23%). In 3 patients risperidone was discontinued within 1 to 2 weeks because of side effects including lethargy, stooped posture, rigidity, and falls. Risperidone was discontinued in another 2 patients because efficacy was lacking or insufficient after 2 and 3 months, respectively. EPS were noted in 11 of 22 (50%), although symptoms were mild in 4 of these. Other side effects noted in the group as a whole included sedation and drooling. There were no reported cases of orthostatic hypotension. On Clinical Global Impression,⁺³ 6 patients (27%) were rated as very much improved, 5 (23%) were rated as much improved, 6 (27%) were rated as minimally improved, 2 (9%) were rated as unchanged, 1 (5%) was rated as minimally worse, and 2 (9%) were rated as very much worse.

The following are several examples described in more detail.

Conclusions based on the results of a case series must be modest in view of their inherent biases and lack of controls. This caution is especially important with studies on the pharmacotherapy of BDD, in which placebo-response rates can be as high as 40%.⁺⁶ With this caveat in mind, the results of the series suggest that risperidone in low doses was fairly well tolerated, with 50% experiencing significant improvement and 77% experiencing at least minimal improvement. Risperidone was discontinued in only 3 patients because of side effects, and although 50% experienced at least some EPS, these were often mild in nature. The latter is particularly important in view of this group's previous experience with significant EPS when treated with typical neuroleptics. EPS occurred in both patients with Lewy body dementia (#21, #22), patients with preexisting parkinsonism (#13, #17), and patients who had previously experienced severe EPS when treated with typical neuroleptics (#6, #9, #12, #14, #15). These symptoms were often dose-related, and they diminished with dose reduction in 3 patients (#5, #12, #21). EPS occurred within the first week of treatment in 3 patients (#9, #13, #14), and within 1 week of dosage escalation in 3 others (#12, #16, #21). In the remaining 5 patients who also developed EPS, these occurred within 1 month of treatment initiation. There were no cases of reported orthostatic hypotension, a potential problem noted previously with risperidone.⁺⁷ Because of the retrospective nature of this report, it is unclear whether orthostatic hypotension did not occur or simply was not assessed in this series.

Agitation manifested by restlessness, negativism, pacing, attempts to leave, and aggression seemed most responsive to risperidone. Delusions and hallucinations improved in some patients and persisted in others.

Only a small amount of anecdotal literature has been published on the use of risperidone for the treatment of BDD. In a small series of 11 geriatric inpatients, only 3 of whom had BDD, marked improvement was noted with doses of 1.5—6 mg per day (bid dosages).⁺⁷ These 3 patients experienced side effects, including orthostatic hypotension, dizziness, somnolence, and nonpostural hypotension. In an unpublished preliminary report of 9 elderly patients with BDD treated prospectively with open-label risperidone, 7 showed significant decreases in scores on a behavioral rating scale, with average daily doses of 2.5 mg. Two patients dropped out because the drug was not efficacious.⁺⁸ Risperidone has also been recommended as a treatment for behavioral disturbances in Lewy body dementia in view of these patients' sensitivity to neuroleptic-induced EPS. Lee et al.⁺⁹ noted improvement in psychosis and agitation in a case report of a 74-year-old patient with Lewy body dementia treated with 2 mg per day of risperidone. Allen et al.⁺¹⁰ described 3 patients with Lewy body dementia who were treated with risperidone 0.5 mg once or twice daily. All 3 patients experienced significant reductions in psychosis and agitation. Risperidone was well tolerated, with no decline in cognitive function noted. There have been two conflicting reports on the use of risperidone for the treatment of behavioral disturbances associated with Parkinson's disease. Rich et al.⁺¹¹ noted improvement in only 1 patient and "intolerable" exacerbations of parkinsonism in 5 of 6 patients treated with risperidone. Patients received 2—4 mg per day (average 2.2 mg) of risperidone in this series. In contrast, Meco et al.⁺¹² noted significant improvement in hallucinations and Brief Psychiatric Rating Scale scores in a group of 6 Parkinson's disease patients when they were treated with 0.25—1.25 mg per day. In this study there was no reported worsening of motor symptoms or cognition when patients were treated with these low dosages.

In summary, although there is little anecdotal evidence for the use of risperidone in the treatment of behavioral disturbances in dementia, the current series suggests that risperidone can be used safely with potentially good effectiveness. Previous clinical recommendations for the use of risperidone in elderly schizophrenic patients have included a maximum daily dose of 2 mg per day.⁺¹³ Based on the current study and previous reports, recommendations for the use of risperidone in elderly demented patients with behavioral disturbances would include starting at the lowest possible dose available (0.25—0.5 mg) with slow titration upward to a maximum dose of 1.0 mg bid. In patients with preexisting parkinsonism, Lewy body dementia, or known EPS sensitivity to typical neuroleptics, maximum dosages might be even lower (0.5—1 mg per day). Side effects including lethargy, gait instability, orthostatic hypotension, and EPS should be monitored on an ongoing basis.

Besides the benefits of lower potential to cause EPS, shared by other atypical neuroleptics, risperidone may theoretically be advantageous for use in BDD considering this population's susceptibility to anticholinergic side effects.⁺¹⁴ Risperidone has negligible effects on muscarinic receptors, whereas both olanzapine and clozapine have a high affinity for the m₁ muscarinic receptor.⁺¹⁵ Placebo-controlled trials of risperidone for the management of behavioral disturbance in dementia are urgently needed.