Return to American Society of Consultant Pharmacists                                    Print This
The Impact of Antipsychotic-Related ADEs: Consultant Pharmacists’ Perspective

Selecting an Antipsychotic Agent: Pertinent Clinical Evidence

Lesley M. Blake, MD, Director of the Division of Geriatric Psychiatry at Northwestern University Medical School, gave an evidence-based comparison of the safety profiles of the four most commonly prescribed atypical antispychotic agents: olanzapine, clozapine, quetiapine and risperidone. In vitro evidence indicates that olanzapine has the highest affinity among the four for the muscarinic cholinergic receptor followed by clozapine. Quetiapine and risperidone, in contrast, have little affinity and, therefore, produce little cholinergic blockade (Richelson E et al. Life Sci 2000;68:29). This translates clinically to significantly less cumulative anticholinergic drug toxicity and less anticholinergic symptomatology with either quetiapine or risperidone.

The significance of this difference is demonstrated by the findings of Flacker and colleagues, who divided a population of hospitalized geriatric patients taking antipsychotic agents into quintiles based on serum anticholinergic activity, and then correlated the quintiles with the incidence of delirium. At the highest activity level, approximately 60% of patients were reported to have symptoms of delirium, one of the central anticholinergic effects, contrasted with approximately 9% in the quintile of lowest serum anticholinergic activity (Flacker JM et al. Am J Geriatr Psychiatry 1998; 6(1):31). Although the cause of delirium is uncertain, one plausible hypothesis is that the addition of an anticholinergic side effect to a nervous system that already has abnormal cholinergic neurotransmission, including all patients with dementia, greatly increases risk for delirium. Dysregulation of other neurotransmitters (i.e., GABA, dopamine) has also been implicated (Tune LE et al. Am J Psychiatry 1992;149:1393).

Clarification of the role of cholinergic blockade in individuals who probably have dementia requires that prescribing habits be changed. It has been demonstrated in one population, for example, that older patients with probable dementia (33.0%) were significantly (p=0.001) more likely to use anticholinergic agents than matched controls (23.4%), and that 26.1% of patients used multiple anticholinergic agents (Roe CM et al. J Am Geriatr Soc 2002;50: 836). In an older study, 13 of the 25 medications most commonly prescribed for elderly individuals, many of them life-sustaining drugs, had anticholinergic effects (Tune 1992). They are listed in Table 1. In addition, all typical antipsychotic agents have these effects. Because antipsychotic agents are frequently indicated for the treatment of behavioral and/or psychotic symptoms of dementia, the atypical antipsychotics with the least potential for cholinergic blockade, risperidone or quetiapine, should be prescribed first (Stoppe GS et al. Drugs and Aging 1999;14:41).

After the age of 65 years, the incidence of falls is three times as high in nursing homes and hospitals as among community-dwelling individuals; but in all settings, there is a consistent association between the use of psychotropic medications and falls (AGS Panel on Falls Prevention. J Am Geriatr Soc 2001;49; 664).

Placebo-controlled trials of olanzapine, quetiapine and risperidone in the setting of dementia provide useful data for assessing their relative contributions to falls. In a risperidone trial, Katz and colleagues compared three doses of the trial agent (0.5 mg, 1.0 mg and 2.0 mg) with placebo in over 600 long-term-care patients with agitated dementia. They reported fewer falls in the group taking 0.5 mg/day of risperidone than in the placebo group (18.0% vs. 22.3%) and a significantly fewer in the risperidone 1.0 m/day (12.7% vs. 22.3%). Although the highest dose was associated with more frequent falls than placebo, the difference was not statistically significant (Katz IR et al. IGGP Proc, 2002). Street and colleagues conducted a 6-week, randomized, double-blind, fixed-dose trial of olanzapine 5 mg, 10 mg and 15 mg in 206 patients with Alzheimer’s dementia and psychosis and/or behavioral symptoms. The incidence of falls was not one of the outcomes analyzed but abnormal gait, one of the principal contributors to falls, was observed to be significantly more frequent in the 5 mg/day and 15 m/day cohorts than in the control group (Street JS et al. Arch Gen Psychiatry 2000;57:968). In a 10-week, randomized, double-blind, flexible-dose study comparing quetiapine with haloperidol and placebo in 284 nursing-home patients, Tariot and colleagues observed no significant difference in falls in the three trial arms (Tariot PN et al. Proc Am Soc Geriatr Psychiatry, 2002).

Although the differences in trial designs, particularly the inclusion or exclusion of falls as a clinical outcome, make the comparisons imperfect, these three studies suggest that risperidone is more efficacious at limiting the incidence of falls than both of the other atypical antipsychotics: olanzapine and quetiapine. A comparison of the Katz and Tariot studies (Figure 1) appears to confirm that of the two atypical antipsychotic agents least capable of cholinergic blockade, risperidone at a dose of 1.0 mg/day may be considered the agent of choice in the older nursing-home population with dementia.

Consultant Pharmacist’s Perspective I

Harlan Martin, RPh, CCP, FASCP, President of Pharma-Care, Inc. of Clark, NJ, presented data from a study involving 730 eligible patients residing in 89 skilled nursing facilities. In order to eliminate acclimation to the environment as a confounding factor, residence in the nursing home for at least 90 days was required for inclusion. All patients had diagnoses of Alzheimer’s disease, vascular dementia, mixed or other dementia. Approximately half of patients had concurrent depression and approximately 13% had concurrent psychosis. Patients with diagnosed Parkinson’s disease, those who were receiving medication for Parkinson’s disease, patients with essential tremor, and patients using prn antipsychotic agents were excluded. The objective of the study was to compare the frequency of adverse events 3 months before and 3 months following the start of treatment with risperidone <2 mg/day (N=474) or olanzapine <10 mg/day (N=256). Data were collected from physicians’ and nurses’ progress notes, physicians’ order forms and psychotropic monitoring forms.

In order to establish a baseline for five selected adverse effects (constipation, weight change, agitation, anxiety and falls), the investigators consulted the labeling information for each drug. For risperidone, the incidences of agitation, anxiety and falls were reported at 22%, 12% and 16 to 25%, respectively. For olanzapine, the incidence of agitation was reported as 23% and that of anxiety as 9%. There were no fall data for olanzapine and no weight-change data for risperidone. The hypothesis was that the actual incidence of these antipsychotic-related ADEs in their trial population, at the doses of these medications used in these facilities, would be lower than those reported to the FDA.

No significant differences were observed between the two trial drugs in verbal target behavior (screaming/yelling, cursing/vulgarity, repetitive verbalization, threatening accusations and crying out) or physically aggressive target behaviors (self-injuries, throwing objects, inappropriate sexual behavior, pinching, kicking and hitting). A subanalysis is currently in progress on aggressive physical behaviors to determine if they correlate with falls. Although labeling information attributes a 6% incidence of weight gain to olanzapine, no significant change was observed in this study. There was, however, a statistically significant difference between the two drugs in the increased number of days on which laxatives were needed by the two populations. Among patients taking risperidone, the increase was 4.3%, while among olanzapine-treated patients the increase was 19.1% (p=0.0002). In the risperidone cohort, the number of patients requiring laxatives rose by 2% compared with 10% with olanzapine. The increased use of eye lubricants was observed as a secondary marker for anticholinergic side effects. Although their use was low in both patient cohorts, among those requiring treatment the increase in number of doses administered per day associated with risperidone was 7.9% compared with 29.7% for olanzapine. Because of the small number of patients requiring eye lubricants, however, this difference is not statistically significant.

The data on falls were analyzed in three ways. First, the entire population of 730 patients was divided into those with histories of falling and those with no histories of falling prior to treatment. Among those who had not fallen previously, the percentage of residents who fell while taking antipsychotic medications was 14.3%. The number of patients falling who had histories of falls was 24.8%. These data indicate that a history of falls is a predictor of future falls while under antipsychotic treatment. Second, all patients with osteoporosis, histories of falls, requiring restraints, or having fallen at least once in the 3 months preceding treatment were eliminated from the data pool. Data for the remaining patients confirmed that the risk of falls while under treatment with either of the trial drugs increases with age in approximately linear manner. Third, using the same exclusions, there were significant differences between the percentage of patients falling since the start of medication and the percentage having multiple falls. For those two indicators, risperidone was associated with increases of 6.9% and 3.0%, respectively, compared with 17.8% and 7.8% for olanzapine. Figure 2 presents these results and the p values indicating statistical significance.

Mr. Martin and his colleagues concluded from these data that the incidence of adverse effects associated with the use of low-dose risperidone and low-dose olanzapine in the nursing home is not as high as is indicated by the package labeling of either agent. Nevertheless, because there is some risk, selection of a medication for controlling behavioral symptoms in patients with dementia should include a consideration of potential anticholinergic effects. Even among the atypical antipsychotic agents, there appear to be significant differences in risk for falling. Based on the study outlined here, risperidone appears to be the preferential drug.

 

Consultant Pharmacist’s Perspective II

A second consultant pharmacist’s perspective on antipsychotic-related ADEs was offered by I. Barton Frenchman, RPh, CCP, FASCP, President of Pharm Rx Consultants, Inc. in Union, NJ. His point of departure was guideline F329 (the antipsychotic stress form) of the CMS survey, which defines unnecessary drugs. An unnecessary drug is one used in an excessive dose or for excessive duration without adequate indications for use, or one used with inadequate monitoring and with adverse consequences. In addition, the use of antipsychotic drugs requires a diagnosis per CMS guidelines of dementia with associated psychotic or agitated behavior, or other conditions with specific psychotic symptoms such as schizophrenia, schizoaffective disorder, Tourette’s syndrome, bipolar disorder, organic delusional syndrome and a few others. When using antipsychotic agents, the patient must be monitored for ADEs such as tardive dyskinesia, Parkinson’s disease, cognitive impairment, orthostatic hypotension, akathisia and behavioral changes. In order to justify the use of antipsychotic agents for behavioral control, psychotic behaviors must be quantitatively and objectively documented, and they must be persistent, non-preventable by other means, and dangerous to the patient and/or to other people.

With these standards in mind, Mr. Frenchman and colleagues conducted a retrospective chart audit of 7,000 patients in 65 facilities in New Jersey and Pennsylvania. From this population, they identified 289 individuals with psychiatric disorders including dementia (approximately 60%) and who were treated with either olanzapine or risperidone with or without selective serotonin reuptake inhibitors (SSRIs). SSRIs were included because their side effects may be similar to those of antipsychotic agents. Moreover, paroxetine has extensive anticholinergic side effects, and both paroxetine and fluoxetine increase blood levels of risperidone and olanzapine. Patients taking benzodiazepines for behavior were excluded. Data were collected over a period of 2 months from nursing, primary physician and psychiatrist notes as well as from MDS 2.0 ambulation status reports and behavior monitoring forms. Patient behaviors were tracked from the time of admission to the nursing home or from the time of initiation of antipsychotic treatment if records permitted. The incidence and number of falls were also recorded as were comorbid conditions, the most frequent of which were hypertension, osteoarthritis and diabetes.

The vast majority of patients (88%) were recorded as having a single inappropriate behavior, the most common of which were psychotic behaviors of hitting, paranoia and delusions. In the overall population, the mean starting doses of risperidone (N=106) and olanzapine (N=100) were 1.40 mg/day and 5.96 mg/day, respectively, and the ending doses were 1.23 mg/day and 6.35 mg/day, respectively, indicating that mean dosage of risperidone decreased slightly and that of olanzapine increased slightly over time. Mean doses of both antipsychotic agents increased slightly over time in combination with SSRIs (N=35 and 48). According to the CMS guidelines, there should be two dose decreases per year for all patients with diagnoses of dementia. In this study, however, segregation of data from individuals with dementia revealed a decline in mean dose for risperidone (1.13 mg/day to 0.81 mg/day start to finish) compared with a slight increase in mean olanzapine dose (4.56 mg/day to 4.78 mg/day). Patients taking risperidone plus an SSRI had a small mean reduction of risperidone dose (0.75 mg/day to 0.68 mg/day) compared with an increase in the group taking olanzapine plus an SSRI (3.35 mg/day to 4.97 mg/day).

Among the 12 behavioral variables tracked (four treatment cohorts times three ending statuses — improved, same, worse), the only one that was statistically significant was improvement in behavioral outcomes associated with risperidone monotherapy (p=0.0014 relative to olanzapine monotherapy). The addition of SSRI therapy to either olanzapine or risperidone was without significant effect on outcomes. Furthermore, the percentage of residents having at least one adverse event in the course of treatment was highest for both olanzapine monotherapy and combination therapy: olanzapine (39%), risperidone (4.7%), olanzapine plus SSRI (27.1%), and risperidone plus SSRI (11.4%). Although the numbers were too small for definitive comparison, extrapyramidal symptoms occurred at a consistent rate in the four cohorts, but tardive dyskinesia was more frequent in the olanzapine groups. The most pronounced differences were higher frequencies of dry eyes in both olanzapine groups and dry mouth, constipation, agitation, sedation, dysphagia and dizziness associated with olanzapine monotherapy.

Compared with olanzapine, the odds ratio of experiencing an adverse event was statistically significant for both risperidone monotherapy (p=0.00002) and risperidone plus SSRI (p=0.0269). The odds ratios for experiencing a fall compared with olanzapine were favorable for all of the other three treatment groups but best with risperidone monotherapy (0.2221). When falls per month were calculated by dividing total falls reported by total months of observation, both risperidone (p=0.0005) and risperidone plus SSRI (p=0.0474) were significantly lower than olanzapine
treatment.

These data confirm the findings in other studies that risperidone is associated with significantly more behavioral improvement, fewer treatment-related ADEs, and fewer falls than is olanzapine.

Return to American Society of Consultant Pharmacists                                    Print This

All contents Copyright © 1999 - 2003 Medical Association Communications