There are a number of case series and open trials suggesting benefit of trazodone in doses from 50 mg to 400 mg/day (Rosenquist et al., 2000). Symptoms of irritability, anxiety, restlessness, and depressed affect have been reported to improve in some cases, along with disturbed sleep. The main side effects have included sedation, orthostatic hypotension, and occasional delirium. In a nine-week crossover study, Sultzer, Gray, Gunay, Berisford, and Mahler (1997) compared trazodone at a mean dose of 220 mg/day with haloperidol at a mean dose of 2.5 mg/day in 28 patients. Agitation improved equally in both groups with better tolerability in the trazodone group. More recently, Teri, Logsdon, et al. (2000) reported negative results for all active treatments in a multicenter outpatient study contrasting trazodone, haloperi-dol, placebo, and caregiver training. The magnitude of placebo response in this study was consistent with other published reports on the treatment of agitation in patients with AD. Mean doses for haloperidol, 1.8 mg, and for trazodone, 200 mg, appear adequate. The patients in this study had broadly defined agitation, ranging from mildly disruptive behaviors to physical violence and aggression. The heterogeneity may have affected the results and highlights the importance of clearer definition of study participants. Nevertheless, the marginal efficacy of the active treatments suggests a prompt reevaluation of patients who do not respond to one of these agents and consideration of other treatment approaches. The Expert Consensus Guideline (Alexopoulos et al., 1998) favors trazodone for sleep disturbance primarily, relegating it to second-or-third-line use for mild agitation. A typical starting dose is 25 mg/day, with maximum doses usually 100 to 250 mg/day.
There are mixed results in clinical trials employing selective serotonin re-uptake inhibitors for agitation in patients with dementia. A review of controlled studies of citalopram in patients with various dementia diagnoses was performed by Nyth and Gottfries (1990), suggesting some beneficial behavioral effects in nondepressed patients with AD but not VaD. This conclusion was supported by two open trials of this agent (Pollock et al., 1997;
Ragneskog, Eriksson, Karlsson, & Gottfries, 1996). Pollock et al. (2002) also reported on a controlled study, where perphenazine at a target dose of 0.1 mg/kg/day and citalopram 20 mg/day showed reduction in measures of agitation /aggression and/or psychosis in comparison with placebo during a 17-day treatment trial. More than 50% of subjects in each of the study arms withdrew from the study. The study patients were hospitalized because of the severity of their symptoms, which most likely explains the high attrition rate observed. The limited duration of the study reflects more what can be accomplished during a brief hospitalization versus outpatient settings, where the majority of these patients receive their treatment. As to other agents, no benefit has been reported yet for fluoxetine or fluvoxamine, and only anecdotes have been presented with respect to other serotonergic agents (reviewed in Loy et al., 1999). Use of serotonergic agents is fairly widespread despite limited data. This may be due in part to neurobiological data linking serotonergic dysfunction to behavior disturbance in dementia and the usually benign side effect profile of this class, which includes gastrointestinal symptoms, sedation or insomnia, sexual dysfunction, and, less commonly, paradoxical agitation.
Was this article helpful?