Introduction
Behavioral and psychological symptoms of dementia (BPSD), such as agitation, violence or psychosis, are among the most distressing manifestations of dementia. The management of these symptoms includes the identification of treatable physical and environmental precipitants, support and psycho-education for caregivers and psychosocial intervention. Nevertheless, pharmacotherapy is frequently used, especially in institutional settings (Schneider et al., Reference Schneider2006).
The widespread use of second-generation antipsychotics (SGAs) in elderly patients with dementia over the past decade has led to the accumulation of randomized clinical trial results that suggest an increased mortality compared with placebo. The Food and Drug Administration therefore instituted a black-box warning for all SGAs concerning an increased risk of death in elderly patients with dementia and a similar warning was subsequently added to the labels of first-generation antipsychotics (Food and Drug Administration, 2005). However, physicians have very limited options in treating dementia-related psychosis and behavioral symptoms. Despite the risks associated with antipsychotic drugs, there is insufficient evidence to suggest that other psychotropics represent a safe and well-tolerated, let alone more effective treatment choice for psychosis or agitation in dementia. It is recommended that physicians carefully determine the cause of psychosis and behavioral symptoms in dementia and discuss treatment options, including behavioral treatment strategies, with patients and families.
The role of serotonin in aggression in general and in aggression and psychosis in AD has been demonstrated, so there is a scientific rationale for evaluating serotonergic compounds (Sweet et al., Reference Sweet2001). There is some evidence to suggest that citalopram may be an option in treating patients suffering from psychosis or agitation in dementia. In two published randomized controlled trials, citalopram was more efficacious than placebo and as efficacious as, but better tolerated, than perphenazine or risperidone in patients with dementia hospitalized for the treatment of agitation or psychosis (Pollock et al., Reference Pollock2002; Reference Pollock2007). The rationale for studying escitalopram for this indication is the clinical response reported in geriatric patients treated with escitalopram (Gorwood et al., Reference Gorwood, Weiller, Lemming and Katona2007), who had better treatment persistence and fewer hospitalizations than patients treated with citalopram (Wu et al., Reference Wu, Greenberg, Yang, Yu and Erder2008). Furthermore, animal studies support the role of the allosteric modulation of the 5-HT transporter in the regulation of the recovery of 5-HT neuronal activity and long-lasting hippocampal cellular plasticity induced by escitalopram and not by citalopram (Mnie-Filali et al., Reference Mnie-Filali2007). Thus, an evaluation of the efficacy and tolerability of escitalopram for the treatment of BPSD is warranted.
Methods
This six-week, double-blind, randomized, fixed-dose study included 40 patients from a psychiatric inpatient setting in Israel (Abarbanel Mental Health Center) from April 2008 to April 2010. The study was approved by the local Institutional Review Board. Written informed consent was obtained from the patients or their legally authorized representatives. The trial registration is available at clinicaltrials.gov Identifier: NCT 01119638.
Under double-blind conditions patients were randomly equally (1:1) to receive treatment with risperidone or escitalopram over six weeks. Study medication was given for oral administration, supplied as tablets of identical appearance. Doses were adjusted according to the protocol as follows: (a) risperidone 0.5 mg once daily for the first week and then 1 mg once daily till study completion; (b) escitalopram 5 mg once daily for the first week and then 10 mg once daily until study completion.
Patients fulfilled criteria for dementia of the Alzheimer's type (according to the Diagnostic and Statistical Manual of Mental Disorders, 4th edition). In addition, patients had to have a Mini-Mental State Examination (MMSE) score between 5 and 24 (Folstein et al., Reference Folstein, Folstein and McHugh1975). To be eligible, patients had to be admitted to our psychiatric center (psychogeriatric ward) because of signs and symptoms of psychosis, aggression or agitation that were severe enough to warrant hospitalization. Eligible patients had to have suffered from delusions, hallucinations, aggression or agitation, which developed after the onset of dementia and which were severe enough to disrupt their functioning and, in the opinion of the study physicians, to justify treatment with antipsychotic drugs. Signs and symptoms of psychosis, aggression or agitation had to have occurred nearly daily during the week prior to enrollment. Patients were hospitalized for the duration of the study.
Exclusion criteria were: (a) a diagnosis of a primary psychotic disorder (e.g. schizophrenia) or delirium; (b) psychosis, agitation or aggression that could be better accounted for by another medical condition; (c) alcohol or substance abuse; and (d) previous treatment with the drugs under study, or contraindications to the study drugs.
A board-certified psychiatrist on the psychogeriatric ward who was blinded to medication status performed the Neuropsychiatric Inventory (NPI) assessments at baseline and weekly for the duration of the study (Cummings et al., Reference Cummings, Mega, Gray, Rosenberg-Thompson, Carusi and Gornbein1994).
Patients who met the selection criteria were assigned to double-blind treatment according to a computer-generated randomization list. The details of the randomization series were unknown to investigators and were contained in a set of sealed opaque envelopes. Sequentially enrolled patients were assigned the lowest randomization number available in blocks of four. All investigators, trial personnel and patients were blinded to treatment assignment for the duration of the study.
Other drugs
Participants were permitted to continue cholinesterase inhibitors or memantine if they had been taking them for at least 12 weeks and at the same dose for at least 4 weeks prior to screening. Other psychotropics (except for clonazepam, used as a “hypnotic” only at bedtime) had to be discontinued prior to randomization.
Sample size
A sample size of 17 patients in each group would have 80% power to detect a difference of 5 points in the total NPI score (assuming that the common SD is 4 to 6) using a t-test with a two-sided p-value <0.05 for statistical significance. Thus, we enrolled 20 patients in each group, expecting a 15% drop-out rate. All adverse events (including any change in concurrent illnesses or new illnesses) either observed by the investigator or reported spontaneously by the patient were recorded.
Statistical analysis
Patients who were randomized and had received at least one dose of the study medication were included in the intention-to-treat sample (ITT). The last observation-carried-forward (LOCF) method was used for imputing missing values. The primary outcome measure was the change from baseline in the Neuropsychiatric Inventory (NPI) total score (Cummings et al., Reference Cummings, Mega, Gray, Rosenberg-Thompson, Carusi and Gornbein1994). The primary efficacy analysis was an analysis of covariance (ANCOVA) of the change from baseline in NPI total score at Week 6 (ITT, LOCF), with treatment as the factor and the baseline NPI total score as a covariate. Variables are presented as mean and standard deviation (SD). All tests were two-tailed and p < 0.05 was considered statistically significant.
Results
Patients had a mean age of about 78 years and just over half were women. The mean MMSE score was about 14 and the mean NPI score was approximately 19 at baseline. The gender distribution, age and baseline MMSE scores of the 40 patients treated in this study did not differ significantly between the two treatment groups (Table 1). Two participants in each group continued existing treatment with a stable dose of a cholinesterase inhibitor. Two participants in the escitalopram group were on memantine during the study.
Patients’ demographic and clinical characteristics
Values are means (SD).
*p = 0.028, **p = 0.004 differences between baseline and week 6 mean NPI scores.
Overall, 67.5% (27/40) of the participants completed the 6-week trial; 75% (n = 16) in the escitalopram group and 55% (n = 11) in the risperidone group (Figure 1). Reasons for patient discontinuation were: two patients transferred to a nursing home and two patients withdrew consent (escitalopram); one patient was transferred to nursing home, two patients withdrew consent, two patients suffered from severe EPS and recurrent falls and four patients suffered from an acute physical illness necessitating transfer to a general hospital (myocardial infarction, pneumonia and two urosepsis) (risperidone). Thus, no adverse events were reported by escitalopram-treatment patients, and six risperidone-treated patients reported adverse events. The investigators considered the two EPs adverse events as due to the investigational medicine risperidone.
Patients recruited to the two treatment arms of the study and their outcomes.
The mean NPI scores decreased significantly from baseline in both treatment groups over six weeks (escitalopram p = 0.02; risperidone p = 0.004). The treatment difference between groups in mean change from baseline was not statistically significant (p = 0.28), although the improvement was greater in the risperidone-treated group. In the escitalopram-treated group, the mean (SD) change from baseline to Week 6 of treatment was ‒4.7 (7.3). In the risperidone-treated group, the mean (SD) change from baseline to Week 6 of treatment was ‒7.7 (6.8).
Although not defined as an outcome measure we analyzed (post-hoc) the changes in the “delusions, hallucinations, agitation/aggression” cluster. The mean total score at baseline and at Week 6 (completion) for these three NPI items was 4.4 (SD = 1.8) reduced to 2.5 (SD = 1.7) with escitalopram and 4.7 (SD = 2.1) reduced to 1.8 (SD = 2.3) for risperidone.
Discussion
The major neuropsychiatric symptoms occurring as part of AD include aggression, disinhibition, agitation, restlessness, shouting, violence and other behavior likely to give rise to difficulties for the patient, caretakers and health-care professionals. It has been well established that aggression, agitation and psychosis occur in about 20% of people suffering from AD living in the community and in 40% to 60% of patients in care facilities (Margallo-Lana et al., Reference Margallo-Lana2001). Importantly, 90% of patients suffering from AD develop BPSD at some point during their illness, with relative persistence of agitation and psychosis (Ballard and Howard, Reference Ballard and Howard2006). Despite its frequency and adverse impact, BPSD remains a major treatment challenge with no established treatment approach that is both well tolerated and effective.
In the present study, we compared the effectiveness of escitalopram and risperidone in reducing aggression, agitation and psychosis in patients suffering from AD. A decrease in the NPI total scores was observed in both treatment groups. The decrease was greater in the risperidone group, but there were fewer withdrawals in the escitalopram group. It is of interest that risperidone treatment began to show an effect sooner, which is an advantage with such distressing and time-sensitive symptoms.
Limitations inherent in the small sample size, the recruitment of inpatients and the fact that this was a single-center study need be taken into account when generalizing these findings. The higher than expected withdrawal rate, and the higher SD in NPI scores, means that this study was not powered to detect a 5-point difference in NPI scores. Caution must also be used in generalizing these findings, as no placebo arm was employed.
Recently, the Cochrane Database Systemic Reviews published their meta-analysis focusing on antidepressants for agitation and psychosis in dementia (Seitz et al., Reference Seitz, Adunuri, Gill, Gruneir, Herrmann and Rochon2011). Nine randomized, controlled trials comprising a total of 692 individuals were included in the review. Five studies compared SSRIs to placebo. Although there was a significant difference between antidepressants and placebo on measures of agitation the results were heavily weighted by one large study. Patients' withdrawals due to adverse events were lower for citalopram. The authors concluded that SSRIs and trazodone appear to be tolerated reasonably well when compared to either placebo or antipsychotics. Our findings are in line with these conclusions.
Any behavioral effects of SGAs in patients suffering from AD must be weighed against the adverse events and risks. Thus, we need to ask: “Are there pharmacological alternatives to antipsychotics?” Citalopram was associated with an improvement in BPSD in a double-blind, placebo-controlled trial (Nyth and Gottfries, Reference Nyth and Gottfries1990) and in two more recent blinded studies by Pollock and colleagues (Pollock et al., Reference Pollock2002; Reference Pollock2007). We conclude that the present study with escitalopram needs replication in a much larger study, perhaps as second-line treatment after withdrawal from less well-tolerated treatment.
Conflict of interest
None.
Description of authors’ roles
Y. Barak wrote the protocol, supervised the data collection, and wrote the first draft of the paper. I. Plopski and S. Tadger collected the data and assisted with writing the article. D. Paleacu was responsible for designing the study and helped in writing the paper. All authors contributed to and have approved the final paper.
Acknowledgment
This study was supported by an unrestricted educational grant from H. Lundbeck A/S.
