- Top of page
Abstract The objectives of this retrospective, naturalistic study were to provide preliminary data on the effects of 6 months treatment with risperidone, olanzapine and quetiapine on behavioral disturbances, within a sample of outpatients with mild to moderate Alzheimer's disease, and on predictors of response. Between July 2005 and December 2005, data were collected from 58 consecutive outpatients with a DSM-IV-TR diagnosis of Alzheimer's disease with behavioral disturbances, who received a 6-month treatment with risperidone, olanzapine or quetiapine. Primary outcome measures were Neuropsychiatric Inventory (NPI) total score and its items forming the basic core of behavioral disturbances in Alzheimer's disease: delusions, hallucinations and agitation/aggressiveness. Secondary outcome measures were Mini-Mental State Examination (MMSE), Activities of Daily Living, Instrumental Activities of Daily Living and Clinical Insight Rating scale. Correlations between baseline MMSE score and improvements in behavioral disturbances were investigated. At 6 months mean NPI total score had fallen 43.5% in the risperidone group, 45.6% in the olanzapine group and 33.3% in the quetiapine group, with no significant between-group differences. Global cognitive function showed no significant change from baseline to end-point. Incidence of adverse events was low. A significant correlation was found between MMSE score and NPI total score and NPI item agitation decreases. Risperidone, olanzapine and quetiapine produced significant improvements in behavioral disturbances and were well tolerated. No significant differences emerged among treatments. The preliminary results also suggest that baseline cognitive function might influence treatment response.
- Top of page
Behavioral and psychological symptoms of dementia (BPSD) represent non-cognitive disturbances observed in patients with dementia, including affective symptoms, such as anxiety, depression, irritability, psychotic symptoms such as delusions and hallucinations, behavioral disorders, such as insomnia, wandering, aggression, agitation, eating and sexual disorders.1 These are currently recognized as important and intrinsic aspects of dementing illnesses, affecting 50–80% of patients with dementia, at some point during their illness.
BPSD are usually treated with first-generation antipsychotics (FGA). However, antipsychotic treatment is associated with a high risk of adverse events, even at modest doses.2–4 Second-generation antipsychotics (SGA) may offer an advance over FGA for the treatment of behavioral disturbances in patients with dementia in terms of a more favorable side-effect profile.5–8 A recent review of 12 randomized, controlled trials reported evidence of effectiveness compared with placebo for olanzapine and risperidone, limited evidence for quetiapine and aripiprazole, and none for clozapine and ziprasidone.8
Long-term studies of efficacy of SGA in elderly patients have been limited in number. Furthermore, there are only few head-to-head studies comparing risperidone and olanzapine, the most widely drugs investigated, in the treatment of behavioral disturbances in dementia, showing no statistically significant differences between the two treatments after 2 weeks,9 8 weeks10 or 10 weeks.11 Cognitive and side-effect profiles of these drugs in all aforementioned studies may differ substantially. The recent Clinical Antipsychotic Trials of Intervention Effectiveness–Alzheimer's Disease (CATIE-AD) study,12 a double-blind, placebo-controlled trial, lasting 36 weeks, showed that there was no large clinical benefit of treatment with olanzapine, quetiapine and risperidone as compared with placebo.
Several health-care agencies have recently paid attention to randomized controlled studies in which olanzapine and risperidone were associated with a greater risk of cerebrovascular adverse events (CVAE) than placebo in resident Alzheimer's disease (AD) patients.13 However, a recent study in 32 710 patients with dementia did not find any differences in cerebrovascular risk between the group receiving atypical antipsychotic therapy and the group receiving typical antipsychotic therapy.14 Furthermore, ‘Guidelines for the treatment of Alzheimer’s disease from the Italian Association of Psychogeriatrics' recommended that SGA should be preferred over traditional agents for use in behavioral disturbances in AD.15
This retrospective, naturalistic study was done to obtain preliminary data on the effects of a long-term treatment with risperidone, olanzapine and quetiapine on behavioral disturbances in AD patients and to explore predictors of response, by investigating correlations between baseline cognitive function and psychotic feature improvement. The effects on other variables such as cognition, clinical insight, activities of daily living and their tolerability were also assessed.
- Top of page
Data were analyzed from 58 patients (36 women and 22 men). All patients were on stable doses of ChE-I: 22 subjects received donepezil (mean dose ±SD, 7.73 ± 2.55 mg/day), 12 galantamine (mean dose ±SD, 14.67 ± 3.11 mg/day), and 24 rivastigmine (mean dose ±SD, 7.50 ± 3.43 mg/day). For the total patient group, the average mean age (±SD) was 75.14 ± 3.67 years. At the time of entry into the study, patients had a total MMSE score of 19.17 ± 3.53 and a total NPI score of 26.34 ± 6.68.
Twenty-two were treated with risperidone, 16 with olanzapine and 20 with quetiapine. No statistically significant differences between risperidone-, olanzapine- or quetiapine-treated patients with respect to age, age at onset, or length of illness before study entry were found. In the olanzapine group there were only female patients (Table 1).
Table 1. Subject characteristics
|Variable||Risperidone (n = 22)||Olanzapine (n = 16)||Quetiapine (n = 20)||Statistic||P|
|Age (years), mean ± SD||75.4 ± 3.96||75.6 ± 3.30||74.50 ± 3.71||0.476‡||0.634|
|Age at onset (years), mean ± SD||71.2 ± 4.04||72.6 ± 5.23||71.60 ± 3.73||0.534‡||0.589|
|Duration of illness (months), mean ± SD||43.8 ± 23.4||37.9 ± 23.0||30.40 ± 17.1||2.079‡||0.135|
|Education level (years), mean ± SD||6.73 ± 4.16||4.63 ± 1.25|| 7.20 ± 4.20||2.475‡||0.093|
| Family's home||36.4||75||20||11.453†||0.003|
|Dose (mg), mean ± SD||0.82 ± 0.45||5.62 ± 2.14||77.50 ± 35.3|| || |
Table 2 and Table 3 show the assessment data in the three treatment groups at baseline and at the end of the study. Baseline psychopathology was similar in all groups, further supporting the groups' comparability. Study patients were experiencing behavioral symptoms at baseline, as reflected by mean total NPI score >25.
Table 2. Mean changes in primary outcome measures
|NPI total score|
| Baseline||28.00 ± 5.44||25.50 ± 6.61||25.20 ± 7.85|
| End-point||15.82 ± 4.68||13.88 ± 4.66||16.80 ± 7.88|
| Baseline||3.64 ± 2.55||3.50 ± 2.06||3.60 ± 2.30|
| End-point||1.64 ± 1.92||1.00 ± 1.03||1.50 ± 1.61|
| Baseline||2.64 ± 2.10||3.25 ± 1.91||2.20 ± 2.09|
| End-point||0.55 ± 1.01||1.25 ± 1.44||0.90 ± 1.62|
| Baseline||2.18 ± 2.65||2.25 ± 2.96||2.70 ± 2.00|
| End-point||1.00 ± 1.57||1.13 ± 1.75||1.40 ± 1.85|
| Baseline||3.36 ± 2.06||2.75 ± 1.44||2.70 ± 2.68|
| End-point||2.36 ± 2.32||2.00 ± 1.26||2.00 ± 2.79|
| Baseline||4.18 ± 2.54||3.37 ± 1.26||3.40 ± 1.67|
| End-point||2.36 ± 1.65||2.13 ± 1.41||2.80 ± 1.58|
| Baseline||0.00 ± 0.00||0.13 ± 0.34||0.10 ± 0.31|
| End-point||0.00 ± 0.00||0.00 ± 0.00||0.00 ± 0.00|
| Baseline||3.00 ± 2.00||2.75 ± 1.77||2.50 ± 1.47|
| End-point||2.91 ± 2.41||2.13 ± 1.96||2.40 ± 1.46|
| Baseline||2.27 ± 2.76||2.00 ± 1.71||2.50 ± 1.73|
| End-point||1.45 ± 1.47||1.13 ± 1.31||2.20 ± 1.28|
| Baseline||2.09 ± 2.40||1.50 ± 1.79||1.30 ± 1.72|
| End-point||1.09 ± 1.27||1.50 ± 1.46||1.20 ± 1.58|
|NPI aberrant motor behavior|
| Baseline||2.18 ± 2.38||1.75 ± 1.53||1.60 ± 1.05|
| End-point||1.18 ± 1.50||1.63 ± 2.00||0.80 ± 1.10|
|NPI night-time behavior disturbances|
| Baseline||2.09 ± 2.40||2.25 ± 1.84||1.30 ± 1.84|
| End-point||1.09 ± 1.77||0.63 ± 0.88||0.80 ± 1.28|
|NPI changes in appetite and eating behavior|
| Baseline||0.41 ± 0.91||1.00 ± 1.71||0.85 ± 0.87|
| End-point||0.36 ± 0.66||0.00 ± 0.00||1.60 ± 4.68|
Table 3. Mean changes in secondary outcome measures
| Baseline||18.92 ± 3.51||19.32 ± 2.88||19.33 ± 4.14|
| End-point||19.33 ± 4.27||17.79 ± 2.83||21.26 ± 5.45|
| Baseline||4.00 ± 1.07||3.87 ± 1.96||3.60 ± 1.79|
| End-point||2.82 ± 1.74||2.00 ± 1.93||3.40 ± 1.90|
| Baseline||2.09 ± 1.92||3.25 ± 2.18||2.60 ± 1.79|
| End-point||1.45 ± 2.11||1.63 ± 1.36||2.40 ± 1.90|
| Baseline||1.27 ± 0.46||1.25 ± 0.68||1.30 ± 0.66|
| End-point||1.55 ± 0.51||1.38 ± 0.50||1.30 ± 0.47|
| Baseline||1.27 ± 0.46||1.13 ± 0.62||1.40 ± 0.68|
| End-point||1.55 ± 0.51||1.62 ± 0.50||1.60 ± 0.50|
| Baseline||1.27 ± 0.63||1.25 ± 0.68||1.50 ± 0.51|
| End-point||1.36 ± 0.66||1.43 ± 0.51||1.50 ± 0.69|
| Baseline||1.55 ± 0.51||1.62 ± 0.72||1.60 ± 0.50|
| End-point||1.82 ± 0.39||1.75 ± 0.45||1.80 ± 0.41|
Risperidone, olanzapine and quetiapine treatments resulted in significant reduced scores on all primary outcome scales, but did not differ from each other. At the end of the study the mean total NPI score had fallen 43.5% in the risperidone group, 45.6% in the olanzapine group and 33.3% in the quetiapine group. Regarding the single items, all three drugs caused significant variations in six items: delusions, hallucinations, depression/dysphoria, agitation/aggression, aberrant motor behavior and night-time behavior disturbances, with no significant between-group differences. The baseline mean MMSE scores for all groups were between 18 and 20, indicating a mild–moderate cognitive impairment of AD patients. There were no significant changes in cognition over 6 months. Significant within-group improvements were observed in ADL and IADL. Some areas of insight such as awareness of cognitive deficits and perception of the progression of disease had a significant worsening.
We found that baseline cognitive function, assessed by MMSE, was significantly correlated with the relative decrease of NPI total score (r = 0.373, P < 0.004) and of NPI item agitation (r =−0.293, P < 0.025), while a trend was observed for decrease of NPI item delusions (r = 0.246, P < 0.062). No correlation was found between MMSE score and improvement of NPI item hallucinations (r =−0.086, P = 0.523).
The most commonly reported and observed side-effects related to medication are listed in Table 4. Concerning the prevalence of any specific side-effects, no significant differences between groups were detected.
Table 4. Most frequent adverse events in any treatment group
|Adverse event||Risperidone n (%)||Olanzapine n (%)||Quetiapine n (%)||Statistics†||P|
|Abnormal gait||2 (9.1)||2 (12.5)||1 (5)||0.645||0.724|
|Constipation||2 (9.1)||4 (25)||1 (5)||3.647||0.161|
|Dizziness||1 (4.5)||2 (12.5)||4 (20)||2.362||0.307|
|EPS||2 (9.1)||0 (0)||0 (0)||3.390||0.184|
|Falls/injuries||4 (18.2)||2 (12.5)||3 (15)||0.234||0.889|
|Somnolence||4 (18.2)||5 (31.2)||7 (35)||1.632||0.442|
|Urinary tract infections||2 (9.1)||2 (12.5)||3 (15)||0.349||0.840|
|Weight gain||7 (31.8)||5 (31.2)||6 (30)||0.017||0.992|
- Top of page
This retrospective, naturalistic study was performed in AD patients with behavioral disturbances to obtain preliminary data on the effects of a long-term treatment with three SGA, and to explore predictors of response, by investigating correlations between baseline cognitive function and psychotic feature improvement. Consistent and clinically significant improvements were noted for risperidone-, olanzapine- and quetiapine-treated groups on behavioral disturbances, demonstrated by significant reductions compared to baseline in NPI overall severity and in NPI items, delusions, hallucinations and agitation. No significant differences emerged among treatments. Baseline cognitive function was positively correlated with NPI total score improvement and negatively correlated with NPI item agitation decrease. This last finding could be explained by the observation that baseline higher cognitive performance was associated with lower agitation severity (data not shown). In contrast, MMSE showed no significant change from baseline, indicating that cognitive function was maintained as stable. Moreover, improvement in ADL attained significance from baseline, whereas some areas of clinical insight, particularly awareness of cognitive deficits and perception of the progression of the disease, showed a significant worsening.
The risperidone and olanzapine doses were both within the ranges recommended by experts25 and used in previous studies.26 The quetiapine dose was only approximately three-quarters to one-third that used in three placebo-controlled trials included in a recent meta-analysis,26 but higher than that in the CATIE-AD study.12
The present findings were also consistent with head-to-head studies that have shown no statistically significant differences between the efficacy of risperidone and olanzapine in the treatment of BPSD.9–11,27 At the time of the present study there have been no trials comparing all three drugs on BPSD. Recently the CATIE-AD study reported no differences between active treatments and placebo because adverse events offset advantages in the efficacy of olanzapine and risperidone in the treatment of psychosis, aggression or agitation in AD patients.12 The present data on clinical improvements partially agree with those of the CATIE-AD study that noted no significant differences among the groups. Moreover, the present patients showed a significant improvement in NPI item depression/dysphoria at the end of the study. In the interpretation of the present data we should take into account that depressive symptoms are reduced by SGA treatment.28 However, this apparent antidepressant effect might also be related to the lower likelihood of dysphoric responses with SGA when compared with first-generation agents.29
In the present study the assessment of global cognitive functions showed that treatment with risperidone, olanzapine and quetiapine was not followed by detrimental effect on cognitive performances. This was in contrast to reports that indicated cognitive decline in patients with dementia taking SGA.26,27,30 Decrease of aggression and psychotic symptoms after 6 months of treatment may also contribute to lack of impairment in cognitive scores.
ADL and IADL showed an improvement over baseline in all three groups without significant between-group differences. One possible explanation of these findings is the improvement of delusions, hallucinations or agitation/aggression, in fact these symptoms have been linked to the disruption of functioning in AD patients.
When we started treatment of behavioral disturbances in AD patients we were aware of the possible increased risk of CVAE with risperidone and olanzapine.13 In the present study no CVAE were found. This finding might be related to the exclusion of patients with stroke risk factors (e.g. previous stroke, marked hypertension, atrial fibrillation etc.).
The present data from routine clinical practice in the outpatients setting confirmed the results of controlled clinical trials in which SGA were associated with a low incidence of EPS.5–8 In the interpretation of the present data we should take into account that all subjects were identified from a selected population of patients who received a 6-month treatment with risperidone, olanzapine or quetiapine, so the study design did not allow the possibility of dropouts due to adverse events.
The present study was subject to certain limitations, which are inherent to observational studies. By virtue of its retrospective nature the most obvious weakness was that it was non-blinded and non-randomized, creating the potential comparison of incomparable groups, with noticeable a priori differences between the three cohorts. Therefore, we recorded a range of baseline characteristics to control, if possible, for any fundamental differences in patients given risperidone, olanzapine or quetiapine. Another limitation was that the patients involved in the present study were taking concomitant medications such as ChE-I; therefore a possible combined effect of ChE-I should be excluded because they are also reported to be effective for some BPSD.31 However, treatment with ChE-I did not influence the present results directly because all AD patients were on stable doses of ChE-I before the initiation of treatment with risperidone, olanzapine or quetiapine.
Other notable limitations included the small sample size and the lack of a control group. Moreover, it is possible that observed improvement in symptoms could be partly due to environmental or other changes rather than to the antipsychotic treatment. However, it is difficult that a factor could have the same impact on observed improvement in all three groups.
Acknowledging these limitations, naturalistic studies such as this one should be considered as complementary to randomized controlled trials because they allow for the collection of information on what is really happening in the clinical setting without the artifacts of an experimental intervention.
In conclusion, the present study found an improvement of BPSD in outpatients with AD during long-term treatment with olanzapine, risperidone and quetiapine, without statistically significant between-group differences. These three SGA had similar side-effect profiles and did not significantly worsen cognition over 6 months. The preliminary results also suggest that baseline cognitive function might influence treatment response. Because the use of SGA may be restricted to patients who have few or no side-effects and for whom benefits can be discerned, it might be important to evaluate AD patients' severity of cognitive impairment before initiating treatment with SGA. These findings underline the importance of outpatient management with consideration of both risks and benefits in order to optimize a patient's care, with a greater impact on the family, economics, quality of life and resulting in the continuation of at-home care.