Behavioral and Psychological Symptoms of Dementia: Treatment with Antipsychotics

In November 1906 the German psychiatrist Alois Alzheimer delivered a lecture at the 37th Conference of South-West German Psychiatrists in Tubingen reporting the case of Auguste D. (Alzheimer 1906). According to Dr. Alzheimer’ s description, the first noticeable symptom of illness shown by this 51-year-old woman was a strong feeling of jealously toward her husband. At times, believing that people were out to murder her, she started to scream loudly. At times, she seemed to have auditory hallu­cinations. Very soon, she showed rapidly worsening memory loss. She was disoriented in her flat and wandered aimlessly from room to room. After four years of illness, Auguste D. died. Dr. Alzheimer performed an autopsy and found in her brain tissue a large amount of senile plaques and neurofibrillary tangles (NFT) which are known to be the peculiar pathological lesions of Alzheimer’ s disease (AD). More than 100 years ago, Dr. Alzheimer described a case of AD with behavioral and psycho­logical symptoms of dementia (BPSD) such as delusions, hallucinations, agitation, and wandering.

The term BPSD has been introduced in 1996 in a consensus statement of the International Psychogeriatric Association and describes a wide spectrum of noncognitive symptoms of dementia including agitation, verbal and physical aggression, psychotic symptoms (delusions and hal­lucinations), oppositional behavior, socially inappropriate behavior, apa­thy, anxiety, sleep disturbances, and wandering (Finkel 2002). It has been
estimated that up to 90% of patients with AD may show at least one of these symptoms during the course of the disease (Benoit et al. 2005; Chan et al. 2003; Lyketsos et al. 2002). They may be manifest at any stage of AD although they have been less frequently reported in the early stages of the disease (Bozeat et al. 2000; Cummings et al. 1996). BPSD are also present in dementia syndromes other than AD, such as dementia with Lewy bodies or frontotemporal lobar degeneration, where they may appear also in the early stages.

BPSD have a tremendous impact on patients’ and families’ quality of life (Fitten 2006; Levy et al. 1996). Families struggle to manage patients with BPSD. Caregivers of patients with BPSD are at high risk of distress, depression and burn-out (Benoit et al. 2006). These symptoms are a pri­mary reason for patients’ institutionalization (Cohen et al. 1993). Patients with BPSD show accelerated cognitive deterioration; they are at high risk of functional decline, disability, hospitalization, emergency room visits and death (Scarmeas et al. 2005). Finally, BPSD have been associated with an enormous increase in costs of care (Herrmann et al. 2006; Alzheimer’s Association 2009). In our aging society, where the number of AD cases is growing fast and approaching 35 million worldwide, BPSD represent a major public health issue and a disruptive condition for families and soci­eties (Alzheimer Disease International 2009).

Although BPSD are generally considered treatable symptoms, their management represents a serious challenge for physicians and caregivers. Several therapeutic options are available and they include nonpharmaco- logical and pharmacological strategies. According to clinical guidelines, nonpharmacological interventions must represent the first-choice strat­egy of treatment for BPSD (American Geriatrics Society and American Association for Geriatric Psychiatry 2003; Alexopoulos et al. 2005; Expert Consensus Panel for Dementia 2005). Nonpharmacological treatments for dementia include a variety of interventions targeting the patient, the family, and the environment. They include music therapy, massage/touch therapy, physical exercise, aromatherapy, light therapy, environmental manipulation, reminiscence therapy, behavior management, multisensory stimulation, and validation therapy. To date, evidence supporting the use of such techniques in clinical practice is limited and mostly derives from clinical studies with small sample size and weak study designs (Hulme et al. 2010; Hersch and Falzgraf 2007). Moreover, most nonpharmacological interventions have been tested among community-dwelling patients and require specific caregiver training to be applied. Nonetheless, it is strongly recommended to consider a pharmacological approach to BPSD only after nonpharmacological strategies have failed to control these symptoms.

To date, antipsychotic medications represent the most efficacious phar­macological option for the treatment of BPSD, although several psycho­tropic medications including cholinesterase inhibitors, benzodiazepines, antidepressants, N-methyl-D-aspartate receptor modulators, and anticon­vulsants have been suggested to be beneficial in controlling some specific symptoms (Madhusoodanan et al. 2007; Jeste et al. 2008).

The aim of this post is to describe the brain anatomical and bio­chemical abnormalities underlying BPSD and the main pharmacological properties of antipsychotics, to review the evidence regarding the efficacy and safety of antipsychotics in patients with dementia, and to provide the reader with practical recommendations for the daily management of BPSD.

Etiological processes underlying BPSD are yet to be fully understood. Multiple biologic and nonbiologic factors are believed to contribute to the development of BPSD. Psychological factors such as a patient’s premorbid personality and response to stress, social factors such as environmental changes and stressful events, caregiver factors such as caregiver distress and reaction to patient’s behavior interact with genetic aspects and neu- robiological abnormalities in determining the onset of BPSD, the pat­tern of symptoms, and their severity (Meins, Frey, and Thiesemann 2008; Zuidema et al. 2010).

Genetic studies have suggested that several chromosomal abnormali­ties may represent a risk factor for the development of BPSD. A mutation of the presenilin 1 gene on chromosome 14 has been linked with depres­sion and psychosis in AD (Harvey et al. 1998). Serotonin and dopamine are brain chemical neurotransmitters that bind specific cellular receptors, thus allowing communication among neuronal cells and nerve fibers. In particular, serotonin is believed to play a major role in modulating mood, emotion, and sexuality while dopamine is thought to mediate behavior, some aspects of cognition, sleep, attention, mood, and motor activity. The term genetic polymorphism indicates a specific variation of a gene. Polymorphisms of brain serotonin and dopamine receptor genes may predispose to the development of BPSD. In particular, visual and auditory hallucinations have been associated with polymorphisms of serotonin receptor genes (5HT2A 102-T/C and 5HT2c Cys23ser) (Holmes et al. 1998). Variation of the dopamine receptor DRD1 and DRD3 genes have been associated with an increased risk of developing psychotic and aggressive symptoms in AD patients (Sweet et al. 1998). Psychosis and aggressive behavior in combination have been associated with a genetic polymorphism of serotonin transporter (5-HTTPR II genotype) (Sweet et al. 2001). Finally, genetic variations of the brain-derived neurotrophic factor (BDNF) have been related to depression in AD patients (Borroni, Costanzi, and Padovani 2010).

It has been shown that the pathology of different brain regions is respon­sible for the development of different BPSD. It has been documented that relative to patients who do not develop psychosis, those exhibiting psy­chotic symptoms have a greater density of NFT in the brain neocortex (Farber et al. 2000). Higher NFT concentration has also been reported in the orbitofrontal cortex of AD patients with agitation (Tekin et al. 2001). Neurofunctional-imaging studies have documented that psychosis is associated with a decreased metabolism of the prefrontal, left frontal-tem­poral and right parietal areas (Lopez et al. 2001; Sultzer et al. 2003). Other psychotic symptoms such hallucination and delusional misidentification have been associated with low neuron count in the CA1 area of the hip­pocampus and in the dorsal raphe (Forstl et al. 1994).

Abnormalities of multiple neurotransmitter systems have been identi­fied in the brain of patients with dementia. The decrease of brain acetyl­choline levels is responsible for memory impairment, cognitive symptoms, and delirium in AD (van der Cammen et al. 2006). The decrease in cho­linergic activity may result in a relative increase of the dopaminergic and noradrenergic activity, which would lead to psychotic symptoms, behav­ioral disturbances, and aggression (Engelborghs et al. 2008; Herrmann, Lanctot, and Khan 2004; . Also, reduced levels of serotonin have been found in different brain regions of AD patients (Lanctot, Herrmann, and Mazzotta 2001). Most pharmacological treatments for BSPD include medi­cations able to either increase or decrease or modulate the activity of such neurotransmitters.

Antipsychotics, also called neuroleptics, are the mainstay of pharma­cological treatment for BPSD. A list of the most commonly used anti­psychotic medications described by chemical structure and drug name is reported in Table 1 .

The so-called conventional, typical, or first-generation antipsychotics have been used since the 1950s for the treatment of schizophrenia. Accord­ing to their chemical structure, conventional antipsychotics are classified in phenotiazines, butyrophenones, and thioxanthenes. All conventional agents share high affinity for the D2 dopamine receptor. Although their mechanism of action has not been fully clarified, they are believed to exert their action by blocking the D2 receptors in the mesolimbic pathway (Xib- eras et al. 2001). It has been documented that the efficacy of these drugs is strictly correlated with the occupancy rate of D2 receptors. D2 occupancy rate has been shown to predict the clinical response to haloperidol.

Table 1

Antipsychotic Medications

Chlorpromazine, Promazine,

Levomepromazine, Acepromazine, Triflupromazine, Fluphenazine, Perphenazine, Prochlorperazine, Trifluoperazine, Acetophenazine, Periciazine, Thioridazine, Mesoridazine

Haloperidol, Droperidol, Trifluperidol, Bromperidol, Benperidol, MelperoneSertindole, Molindone, ZiprasidoneFlupentixol, Clopenthixol, Chloprothixene, Thiothixene, ZuclopenthixolSulpiride, Remoxipride, Amisulpride, LevosulpirideClozapine, Loxapine, Olanzapine,Risperidone, Iloperidone, Paliperidone

A thresh­old of 65% occupancy rate provides a good separation between responders and nonresponders to treatment (Kapur et al. 2000). Therapeutic action of conventional antipsychotics is often achieved together with the onset of extrapyramidal side effects (EPS) and tardive dyskinesia. EPS include dys­tonia (uncontrollable muscle contractions that can cause painful twisting of parts of the body, especially the neck), akathisia (a disabling form of internal or external restlessness that can lead to the complete inability to sit still and to the constant urge to be moving), parkinsonism (a set of symptoms that resemble symptoms of Parkinson’s disease, e.g., tremor, stiffness of trunk, arms, or legs, difficulty in starting movement, gait and balance distur­bances). Tardive dyskinesia is usually a side effect of long-term treatment with antipsychotics and consists of repetitive, involuntary and purposeless body or facial movements such as tongue protrusion, eye blinking, move­ments of fingers, arms, or legs. These cumbersome side effects as well as other side effects such as hyperprolactinemia result from the occupancy of D2 receptors in the basal ganglia (Nordstrom et al. 1993; Farde et al. 1992).

Second-generation antipsychotic drugs have been described as “atypi­cal” because they were found to exert antipsychotic effect with significant lower propensity to cause EPS and hyperprolactinemia. These advantages can be explained examining the pharmacodynamic characteristics of atyp­ical antipsychotics. These compounds, while showing affinity for D2 dop­amine receptors, target multiple receptor systems in the brain including D1, D3, D4, D5 dopamine receptors, 5HT1, 5HT2, 5HT3, 5HT6, 5HT7 sero­tonin receptors, alpha 1 and 2 adrenergic receptors, H1 histamine recep­tors, M1 acetylcholine receptors, and glutamate receptors. In particular, reduced EPS observed with second-generation antipsychotic drugs have been explained with their high 5HT2/D2 occupancy ratio (Meltzer, Mat- subara, and Lee 1989). According to a recent hypothesis, the fast dissocia­tion of atypical antipsychotics from the D2 receptor would be the potential mechanism by which these drugs have an antipsychotic effect without causing EPS or prolactin elevation (Kapur and Seeman 2001).

Clozapine is the prototype of these newer medications. This drug has demonstrated low affinity for both D1 and D2 dopamine receptors, along with high affinity for D4 dopamine receptor and serotonin receptors (5HT2 and 3) (Kapur, Zipursky, and Remington 1999). It also has an antigluta- matergic action (Lidsky et al. 1993), as well as alpha-2 receptor affinity and M1 cholinergic receptor blocking activity (Factor 2002). The blockade of dopamine receptors exerted by clozapine is evident especially in the mesolimbic pathway but not in the nigro-striatal system (Baldessarini and Frankenburg 1991). This selectivity may in part explain the low incidence of side effects. Nearly two decades ago, risperidone was introduced in the market. To date, it is the most frequently prescribed atypical antipsy­chotic. Its pharmacodynamic profile is characterized by a high 5HT2/D2 affinity ratio. However, a similar proportion of D2 receptors occupied by risperidone and haloperidol has been documented with a dose-dependent propensity of this atypical agent to cause EPS (Kapur et al. 1995). Olan­zapine was introduced shortly after risperidone, and it has a pharmaco­dynamic profile similar to that of clozapine (Factor 2002). Quetiapine was the fourth atypical antipsychotic marketed. It has a chemical structure similar to clozapine and a multireceptor activity but it acts selectively in the limbic system (Richelson 1996; Jibson and Tandon 1998). A 5HT/D2 occupancy ratio similar to that of other atypical antipsychotics has been documented for ziprasidone although its affinity for D2 receptors is high ( Fischman et al. 1996 ).

Very recently paliperidone, a derivate of risperidone and iloperidone, has been marketed. Their mechanism of action is yet to be established although both drugs have shown affinity for D2 dopamine receptors and 5HT2A serotonin receptors (Madhusoodanan and Zaveri 2010; Citrome 2009).

Relative to other drugs of this class, the atypical sertindole exerts a selective activity binding the D2, 5HT2 and alfa-1 adrenergic receptors. Also, its action is prevalent in the limbic system (Hertel 2006).

Atypical antipsychotics include sulpiride and amisulpride, which are compounds belonging to the chemical class of benzamides. These drugs are highly selective blockers of the D2 and D3 receptors with a preferential limbic activity. They have also been shown to modulate dopamine release by binding presynaptic receptors (Schoemaker et al. 1997).

Aripiprazole is a newer antipsychotic agent that has been identified as the “third-generation” antipsychotic. It is considered a stabilizer of the dopamine/serotonin system due to its peculiar pharmacological profile (Burris et al. 2002). In pharmacodynamics, a drug is defined as an antago­nist if it binds to a given receptor and produces a full inhibition of that receptor. A drug is defined as an agonist if it binds to a given receptor and produces a full activation of that receptor. A drug is defined as a partial agonist if it competes with the natural neurotransmitter for a given recep­tor but it is able to activate that receptor to a lesser degree than its natural neurotransmitter would do, thus causing an attenuated response. Aripip- razole acts as a partial agonist at the D2 dopamine receptors. It also binds serotonin receptors acting as an antagonist of the 5HT2A receptors and as partial agonist of the 5HT1A receptors.

None of the available antipsychotics to date has been approved by the U.S. Food and Drug Administration (FDA) for the treatment of BPSD. This is due to the fact that, despite a general clinical perception of effi­cacy reported by physicians, scientific evidence of efficacy is modest and derives from a limited number of randomized clinical trials (RCTs). None­theless, antipsychotic medications are widely used off-label as first-line agents for the pharmacological treatment of BPSD.

Since their approval in the 1950s for the treatment of schizophrenia, conventional antipsychotics have been systematically used for the treat­ment of BPSD. They are prescribed in spite of a substantial lack of scien­tific evidence supporting their use in dementia. Few RCTs investigating
the efficacy of conventional agents for the treatment of BPSD have been conducted between the 1960s and the late 1980s (Schneider, Pollock, and Lyness 1990; Barnes et al. 1982; Lonergan, Luxenberg, and Colford 2002). These studies have provided evidence on the effect of haloperidol and thioridazine while the effect of other conventional agents in dementia has been neglected by clinical research. Data from these early studies showed a modest advantage of conventional antipsychotics over placebo with a nearly 40% placebo response. Also, according to some of these studies, the observed superiority of conventional antipsychotics over placebo would be limited to symptoms of aggression. The validity of such find­ings is questionable due to the fact that these studies were characterized by small sample sizes and possible lack of power to detect any effect of antipsychotics.

Atypical antipsychotics have been approved by the FDA exclusively for the treatment of schizophrenia and marketed in the 1990s. Although not approved for BPSD, these drugs rapidly became the gold standard of care for BPSD shortly after their introduction in clinical practice. This was due to the fact that atypical antipsychotics had shown a substantial advantage over conventional medications: They were able to exert anti­psychotic effect without causing EPS and tardive dyskinesia (Gerlach 2000; Mossman and Lehrer 2000; Collaborative Working Group on Clinical Trial Evaluations 1998). Scientific societies of physicians and experts in the field developed clinical practice guidelines and consensus statements to promote the off-label use of atypical antipsychotics for the treatment of BPSD in spite of the limited number of RCTs documenting the efficacy and safety of these agents in dementia (American Geriatrics Society and American Association for Geriatric Psychiatry 2003) . In the late 1990s, atypical agents accounted for more than 80% of antipsychotic prescrip­tions in dementia (Glick et al. 2001; Liperoti et al. 2003). At this time, ris­peridone, olanzapine, quetiapine, and aripiprazole are the only atypical antipsychotics that have been investigated in RCTs conducted on patients with dementia (see Table 2 ).

Three placebo-controlled RCTs have shown that, compared to pla­cebo, risperidone may be beneficial on psychotic symptoms and aggres­sion at doses of 1 mg and 2 mg per day (Katz et al. 1999; De Deyn et al. 1999; Brodaty et al. 2003). These studies were conducted on patients with Alzheimer’s disease, vascular dementia, or mixed dementia on a 12-week time period.

Two placebo-controlled RCTs have suggested that olanzapine may im­prove behavioral symptoms, psychosis, and aggression at doses of 5 to 10 mg per day compared with placebo (De Deyn et al. 2004; Street et al. 2000).

Randomized Placebo-Controlled Clinical Trials of Atypical Antipsychotics in Patients with BPSD risperidone

vs.

quetiapine vs. placebo

mean dose 5.5 mg; risperidone mean dose 1 mg; quetiapine mean dose 56.5 mgrisperidone, both superior to quetiapine and placebo

These studies were conducted among patients with dementia for a 10-week and 6-week period of time, respectively. In contrast with these data, a recent study on patients with moderate to severe psychotic symptoms of dementia randomly assigned to receive a flexible dose of olanzapine (2.5-10.g per day), risperidone (0.5-2 mg per day) or placebo demon­strated similar improvement of BPSD in the three treatment groups with higher discontinuation rate due to adverse events in the olanzapine and risperidone groups (Deberdt et al. 2005).

More recently, the impact of stopping long-term antipsychotic treat­ment was investigated in a small sample of nursing home residents with BPSD who were receiving haloperidol, risperidone, or olanzapine (Ruths et al. 2008) . Study participants were randomized to either stop or con­tinue antipsychotic treatment. Findings from this study documented that BPSD remained stable or improved in nearly 50% of residents who dis­continued the treatment and in almost all those who continued to receive antipsychotics.

A small placebo-controlled RCT conducted on a sample of 40 patients with dementia and parkinsonism found that quetiapine 120 mg per day was not effective for controlling psychotic symptoms and agitation although well tolerated (Kurlan et al. 2007). More recently, the effect of quetiapine (median dose 200 mg per day) was not superior to that of pla­cebo in a small RCT conducted on patients with AD and BPSD for six weeks (Paleacu et al. 2008). Negative findings derived also from a late study that compared the effect of quetiapine, rivastigmine (an inhibitor of the enzyme acetylcholinesterase, which is used for treating cognitive symptoms of dementia), and placebo on agitation and cognition in a sam­ple of 93 institutionalized patients with dementia (Ballard et al. 2005). This study demonstrated no effect of quetiapine or rivastigmine on improv­ing agitation and an increased cognitive decline associated with the use of quetiapine. In contrast with these data, quetiapine at high daily dose (200 mg per day) has been found effective in a placebo-controlled RCT conducted on a sample of 333 institutionalized patients with dementia and agitation who were randomized to quetiapine 200 mg per day, quetia- pine 100 mg per day, or placebo for a 10-week period (Zhong et al. 2007). Finally, quetiapine and haloperidol appeared both superior to placebo and effective in controlling psychotic symptoms in a large study conducted on 284 patients with AD and BPSD randomized to flexible doses of quetia- pine (median daily dose 96.9 mg), haloperidol (median daily dose 1.9 mg), or placebo for 10 weeks (Tariot et al. 2006). Quetiapine resulted also better tolerated than haloperidol, which was associated with an increased risk of parkinsonism.

To date, a very limited number of RCTs have examined the effect of aripiprazole on BPSD. Aripiprazole at a mean dose of 10 mg per day appeared as beneficial as placebo on delusions and hallucinations and well tolerated in a 10-week study conducted on 208 outpatients with AD (De Deyn et al. 2005). More recently, the efficacy of aripiprazole has been documented in a large RCT conducted on 487 institutionalized patients with AD and BPSD for 10 weeks (Mintzer et al. 2007). In this study, com­pared with placebo, aripiprazole was effective at a dose of 10 mg per day for controlling psychotic symptoms and agitation. Finally, aripiprazole 15 mg per day was superior to placebo and effective for treating behavioral symptoms, agitation, anxiety, and depression in 10-week study on 256 institutionalized patients with BPSD. In this study, aripiprazole showed no effect on psychotic symptoms (Steim et al. 2008).

According to the CATIE-AD (Clinical Antipsychotic Trials of Interven­tion Effectiveness-Alzheimer’s Disease), a large multicenter, double-blind, placebo-controlled, effectiveness trial on outpatients with AD and psycho­sis, aggression or agitation, olanzapine (mean dose 5.5 mg per day) and risperidone (mean dose 1.0 mg per day) for the treatment of BSPD were equally beneficial and superior to placebo and quetiapine (mean dose 56.5 mg per day) (Schnedier et al. 2006). However, these benefits were evident only among those patients who tolerated these medications and did not discontinue them due to side effects. According to authors ; conclusions potential side effects associated with antipsychotic medications in demen­tia may outweigh possible benefits.

A comprehensive review of the available placebo-controlled RCTs has been conducted by Ballard and White for the Cochrane collaboration to determine the effectiveness of atypical antipsychotics for the treat­ment of psychiatric and behavioral symptoms in Alzheimer’ s disease (Ballard and Waite 2006). The authors analyzed 16 placebo-controlled RCTs among which only 6 studies were published in full in peer- reviewed journals at the time of completion of the review. According to the Cochrane authors, evidence suggests that both risperidone and olan­zapine may reduce aggression and risperidone may also reduce psycho­sis compared to placebo. However, an increased risk of extrapyramidal symptoms and adverse cerebrovascular events associated with atypi­cal antipsychotics would outweigh the modest effectiveness of these medications. For these reasons, authors conclude that the use of atypical antipsychotics in clinical practice would not be suitable and should be limited to those patients presenting with significant distress and risks associated with BPSD.