There are a couple of brighter spots in the otherwise glum conclusions of this study. In contrast to the decline in cognitive functions with placebo observed over the 12-week study among placebo-treated patients, cognition as measured by the Severe Impairment Battery and the Standardized Mini-Mental State Examination improved a little with donepezil(Drug information on donepezil), supporting similar claims for an effect on cognitive impairment in patients with severe dementia who have AD reported earlier by another group.13
Given widespread concerns about the adverse effects of the cholinesterase inhibitors, it is also worth highlighting this study's finding of a low level of significant GI adverse effects. Nausea, vomiting, and anorexia occurred in 2.3% of donepezil recipients and in 0.8% of placebo recipients, which suggests the infrequent clinical significance of that effect in this population. A clinical adverse effect's impact is sometimes expressed as the number needed to harm (NNH), and the relevant calculation here would yield an NNH of 67, meaning that 67 of these patients would have had to be treated with donepezil instead of placebo in order to yield one serious case of nausea, vomiting, or anorexia not attributable to background placebo effect. Furthermore, in light of concerns regarding pharmaceuticals' effects on mortality, it is reassuring that no excess deaths occurred in the cholinesterase inhibitor group in this study.
Finally, the response of 13.7% of subjects to a minimal caregiver psychosocial intervention is consistent with the work of Mittelman's group2 and others, and should encourage our promotion of psychoeducation and intervention in patients' support systems. Perhaps the inclusion of the psychosocial intervention and elimination of psychosocial responders from the medication phase resulted in a study cohort of individuals with more resistant agitation.
My suspicion, consistent with my own clinical experience, is that patients with severe and established agitation are not as capable as less severely affected individuals of benefiting from treatment for agitation with cholinesterase inhibitors. These medications, and perhaps memantine(Drug information on memantine) (Namenda) as well, may be more relevant as preventive interventions to defer the worsening of mild agitation or its emergence in patients not agitated at baseline.Alternative medications
Previous studies of NCBS in patients with dementia have offered few medication alternatives to the antipsychotics and cholinesterase inhibitors. Use of citalopram(Drug information on citalopram) (Celexa)14 or carbamazepine(Drug information on carbamazepine) (Carbatrol, Tegretol, others)15 has been advocated. The evidence bases for divalproex sodium(Drug information on divalproex sodium) (Depakote)16 and trazodone (Desyrel)17 are mixed. Case reports or small series extol the potential but largely untested values of opiates,18 cannabinoids,19 and electroconvulsive therapy (ECT)20 in treating agitation and other NCBS. These approaches are in need of more extensive evaluation.What to do?
How you will answer your voice mail, therefore, depends to some degree on your level of optimism, your level of risk averseness, and your belief in a set of current treatment options that is of only limited effectiveness in many cases. The first approach, it is agreed, should be medical evaluation and behavioral intervention. Caregiver education and the use of distraction, redirection, and environmental manipulations designed to guide rather than teach acceptable behaviors may be sufficient.
If medications are needed and enough time is available to permit use of an algorithmic sequence of treatments, a cholinesterase inhibitor should be considered—not solely for its effectiveness but rather for the relatively low level of risk. Next, several of the atypical antipsychotics remain reasonable choices when used in patients whose vascular risk factors do not outweigh their behavioral treatment needs. The atypical antipsychotics have not been proved to control agitation over an extended period of time and should be used at the lowest effective doses and for the shortest interval necessary, with sufficient psychoeducation and disclosure of risks to caregivers and relatives.
If these medications offer insufficient relief, it appears a toss-up whether to proceed to a conventional antipsychotic, an SSRI antidepressant, a less thoroughly investigated medication, or even ECT. The future, which holds just beyond current reach a set of agents with alternative mechanisms for attacking the pathophysiology of AD, may provide additional and more effective ways to address agitation.