Dementia associated with Lewy bodies accounts for up to 25 percent of dementia presentations in the elderly. This form of dementia is characterized by fluctuating cognitive impairment, attention deficit, delusions, depression, parkinsonism, sleep disturbance, and visual and auditory hallucinations. Many of these features would normally be treated with neuroleptic drugs, as in patients with Alzheimer's disease, but these agents can produce a fatal reaction in patients with this type of dementia. In clinical trials, patients with this form of dementia have responded better than patients with Alzheimer's disease to therapy with cholinesterase inhibitors. This finding may be explained by the relatively greater deficiency of choline acetyltransferase and the better preservation of postsynaptic muscarinic receptors in Lewy-body dementia. McKeith and colleagues studied the efficacy, tolerability and safety of a 20-week course of therapy with rivastigmine, a cholinesterase inhibitor, in patients selected by the clinical probability of dementia with Lewy bodies.

The 120 patients were selected by specialists at several European centers on the basis of a diagnosis of mild to moderate dementia (Mini-Mental State Examination score of at least 9) plus other clinical features suggesting Lewy-body-type dementia. Patients had to be able to give informed consent and participate in psychometric testing. Exclusions included the presence of severe extrapyramidal symptoms, asthma or sensitivity to the trial medications, current use of neuroleptics, anticholinergics, selegiline or similar drugs, or no responsible caretaker.

Sixty-one patients were randomly assigned to placebo and 59 to rivastigmine therapy for 20 weeks of active treatment and three weeks of observation without medication. Rivastigmine dosages started at 1.5 mg twice daily and increased gradually to a maximal dosage of 6 mg twice daily or until a well-tolerated maintenance dose was reached. The patients, caretakers and study personnel were unaware of the treatment allocations. A neuropsychiatric inventory, used to monitor 12 behavioral domains, and other measures of physical and psychologic functioning were made at baseline, at weeks 12 and 20, and three weeks after discontinuation of the drug.

The mean age of patients in both groups was 74 years, and most had one or more coexistent medical problems. The two groups were similar in other demographic and behavioral variables. About one half (56 percent) of the patients assigned to rivastigmine achieved the maximal dosage of 12 mg daily. Overall, the mean dose was 9.4 mg daily. Most patients (92 percent) taking rivastigmine reported adverse effects, but adverse effects were also reported by 75 percent of the placebo group. Most adverse effects were mild and tolerable, but nausea, vomiting, anorexia and somnolence were significantly more common and severe in patients taking rivastigmine. These patients also showed a reduction in body weight over time, but no other clinical or laboratory adverse effect was disproportionately more common in patients taking rivastigmine than in patients taking placebo.

Patients treated with rivastigmine showed significant improvement in functioning, had fewer delusions and hallucinations, and were less apathetic and anxious than those in the placebo group. An improvement of at least 30 percent from baseline was shown by 63 percent of rivastigmine-treated patients compared with 30 percent of patients in the placebo group. Tasks involving an attention component appeared to benefit most from treatment. More than one half of the treated patients lost hallucinations and psychotic symptoms. During the nontreatment follow-up period, the differences between the groups tended to diminish.

The authors conclude that rivastigmine was associated with significant improvement in clinical and behavioral symptoms of Lewybody dementia. If side effects can be tolerated, this drug could provide substantial benefit for patients.

editor's note: Managing patients with dementia or, more often, helping families support patients through the process, seems to be beset with paradoxes. Natural desire to do as much as possible to preserve functioning during these last precious years is tempered with the horror of adding drug side effects to the symptom burden. The prospect of false, but expensive, hope raised by some treatments can be an even worse burden. This study appears to demonstrate real benefit for the patients and their families, if the medication is tolerated and taken over a sustained period. The theoretic danger of increased parkinsonian features did not occur, and the relief of psychotic symptoms must have been particularly welcome. Interestingly, 30 percent of the placebo group also showed significant improvement, and more than 70 percent reported side effects. If longer and larger trials substantiate the benefit of this treatment, applying it to practice will require at least two significant contributions from family physicians. The first is better help in identifying the subtypes of dementia and hence the specific patients who could benefit from these new therapies. The second is good, old-fashioned coaching of patients about the benefits and systemic effects of medication so the placebo effect is recruited. Skillful family doctors know how to manage the situation so that the medication is prescribed and taken with a realistic optimism and expectation of benefit combined with prudent monitoring for adverse effects.—a.d.w.