- Email: [email protected]
Extrapyramidal Side Effects in Patients With Alzheimer's Disease Treated With Low-Dose Neuroleptic Medication
BRIEF REPORT
Extrapyramidal Side Effects in Patients With Alzheimer’s Disease Treated With Low-Dose Neuroleptic Medication Michael P. Caligiuri, Ph.D. Enid Rockwell, M.D. Dilip V. Jeste, M.D. The authors examined whether the presence of extrapyramidal side effects (EPS), measured before neuroleptic treatment was initiated, could be used to predict the development and severity of neuroleptic-induced parkinsonism (NIP) in Alzheimer’s disease (AD). Twenty-four newly medicated probable AD patients were assessed with a battery of measures of extrapyramidal motor function. Assessments were made before neuroleptic therapy, and 3 and 9 months after treatment. Posttreatment clinical findings revealed that 66.7% of the AD patients developed NIP. Patients who developed NIP exhibited more severe pretreatment bradykinesia on instrument-derived measures. These findings suggest that a substantial proportion of AD patients treated with neuroleptics develop significant EPS and that the risk for EPS can be estimated before intervention by use of instruments measuring motor function. (Am J Geriatr Psychiatry 1998; 6:75–82)
N
euroleptics are currently the most effective form of management of behavioral and psychiatric disturbances in Alzheimer’s disease (AD).1 Schneider et al.2 reviewed 33 published studies on the therapeutic outcome of neuroleptics in agitated dementia and concluded that
neuroleptic treatment offers modest benefit in the management of agitation. Despite this somewhat optimistic position, the use of neuroleptics with AD patients remains a controversial issue because of the significant side-effect profile associated with these medications.3 Using lower
Received September 16, 1996; revised January 20, 1997; accepted January 28, 1997. From the Department of Psychiatry, University of California, San Diego, and the Department of Veteran Affairs Medical Center, San Diego, California. Address correspondence to Dr. Caligiuri, Geriatric Psychiatry Clinical Research Center (116A1), VA Medical Center, 3350 La Jolla Village Drive, San Diego, CA 92161. e-mail: [email protected]. Copyright 䉷 1998 American Association for Geriatric Psychiatry THE AMERICAN JOURNAL OF GERIATRIC PSYCHIATRY
75
Side Effects of Neuroleptics in AD
dosages or low-potency neuroleptics does not necessarily reduce the risk of debilitating extrapyramidal side effects.4–6 Preexisting parkinsonism has been shown to negatively influence a patient’s ability to tolerate neuroleptic treatment. McKeith et al.7 observed that over half of their patients with senile dementia of Lewy-body type who were given neuroleptics in standard dosage showed acute and often irreversible adverse reactions indicative of a neuroleptic sensitivity syndrome. Survival analysis showed that patients who experience adverse reactions to neuroleptics had increased mortality in the year after neuroleptics were initiated, compared with patients with mild or no neuroleptic sensitivity. Although neuroleptic intervention for behavioral disturbances in AD cannot always be avoided, it may be helpful to understand the risk for developing disabling motor side effects. The purpose of the present study is to examine whether certain demographic variables or motor disturbances, measured before neuroleptic treatment was initiated, could be used to predict the development and severity of neuroleptic-induced parkinsonism (NIP) in AD patients treated with very low dose neuroleptics.
METHODS Subjects A group of 24 community-dwelling AD patients were recruited from the University of California at San Diego (UCSD) Geropsychiatry Clinical Research Center for inclusion into the study. Patients were identified by their primary physician and referred to our center as part of an ongoing research effort to study adverse effects of antipsychotics in late life. Diagnoses of probable AD were made by staff neurologists and psychiatrists according to NINCDS/ADRDA8 criteria. Patients were 76
excluded from study for the following reasons: 1) viral, metabolic, or traumatic causes of dementia; 2) history of alcoholism, stroke, or other neurological illness; 3) evidence of non-parkinsonian movement disorder; 4) diagnosis of idiopathic Parkinson’s disease; and/or 5) severe musculoskeletal disorders that would impair mobility. The mean (ⳲSD) age of the AD patients was 75.1Ⳳ8.3 years. The purpose of the study was explained to the patient and Significant Other or caregiver, and informed consent was obtained before neuroleptic treatment was initiated. Neuroleptic Treatment Patients were enrolled in the study before neuroleptic medication was initiated. Neuroleptics were administered for psychiatric disturbances, including psychosis and agitation, at the discretion of the patient’s primary physician. The physician determined the appropriate neuroleptic type (potency) and dosage. Sixteen patients were treated with haloperidol, and the remaining eight patients were treated with thioridazine. None of the patients received concomitant anticholinergic medication. The mean initial dosage, expressed in chlorpromazine-equivalents, was 26.8 Ⳳ17.8 mg/day. Each patient’s neuroleptic status was evaluated by Center staff psychiatrists monthly. The mean dosage at the 9-month assessment was 25.9Ⳳ18.2 mg/ day. The neuroleptic dose was increased in 1 patient, decreased in 2 patients, and remained constant for the remaining 21 patients over the 9-month study period. Baseline and Follow-Up Assessments Patients being considered for neuroleptic treatment who met inclusion criteria were seen for an initial examination consisting of neurological screening to rule out potential vascular or metabolic etiologies for their dementia, a brief cognitive assessment, and clinical and quantitative VOLUME 6 • NUMBER 1 • WINTER 1998
Caligiuri et al.
extrapyramidal motor assessment. Most patients were followed for several years after exposure to neuroleptics as part of an ongoing study of tardive dyskinesia; however, we report only the 3- and 9-month follow-up data in the present study so as to reduce potential confounding due to increasing illness duration and cognitive decline.9 Outcome Variable We used a modified version of the Simpson-Angus Rating Scale for Extrapyramidal Side Effects (SAS)10 to rate the severity of pretreatment parkinsonism and NIP. The standard scale consisted of 11 items rated using a four-point severity index, with greater emphasis on rigidity than bradykinesia. The standard scale was modified for three reasons. First, it does not contain a measure of bradykinesia, which is an important feature of parkinsonism associated with dementia. Second, the scale is biased strongly toward rigidity—this factor encompasses 4 of the 11 items. Third, the original scale does not consider asymmetry, which may prove to be an important feature in NIP. Our modification of this scale reduced the number of items to eight, consisting of ratings of gait, left and right upper-extremity tremor, left and right upper-extremity bradykinesia, left and right upper-extremity rigidity, and masked facies. The modified scale has been in use in our center by trained staff for over 7 years. Raters undergo periodic sessions to maintain high levels of interrater reliability. Parkinsonian signs were rated by use of a 5-point scale ranging from 0 (no impairment) to 4 (severe impairment), with 1, 2, and 3 representing minimal, mild, and moderate impairment, respectively. The modified scale (SASm) yielded a total score ranging from 0 to 32. Patients who exhibited a total score of at least 7 (mild-tomoderate impairment) were considered to exhibit NIP. Patients who exhibited an increase in two of the five items of at least THE AMERICAN JOURNAL OF GERIATRIC PSYCHIATRY
2 points or exhibited an increase in total SASm score of at least 200% above baseline after neuroleptic treatment were considered to have developed NIP. SASm ratings were made at baseline and at the 3- and 9month follow-up assessments. Predictor Variables Several independent variables were chosen as potential predictors of NIP. Total SASm score, as well as item scores for bradykinesia, rigidity, and tremor, served as the predictor variables. Neuroleptic dose, converted to standard chlorpromazine equivalents, and potency (haloperidol or thioridazine) served as the pharmacologic predictor variables. Age served as the sole demographic variable because estimates of illness duration were thought to be unreliable and were, therefore, not used. The Mini-Mental State Exam (MMSE)11 served as the cognitive status predictor variable. Three instrument-measured motor predictors were obtained for quantifying postural tremor, rigidity, and bradykinesia. The primary purpose of obtaining instrumental measures of motor function was to increase the sensitivity of detecting motor disturbances in patients before neuroleptic exposure. Recent studies in AD patients with and without clinically rated extrapyramidal involvement have demonstrated increased diagnostic sensitivity with instrument measures of motor function.12,13 Detailed descriptions of the instrumental procedures, their validation, reliability, and sensitivity, have been previously reported,14,15 and only a brief description is presented here. Using a strain gauge-instrumented force lever, we measured the spectral amplitude within the frequency range of 3–7 Hz to quantify upper-extremity (hand) tremor. For rigidity, we quantified activated stiffness by use of a device worn on the wrist that transduced force and rotation during passive movement. Stiffness slopes, derived 77
Side Effects of Neuroleptics in AD
from the force/rotation regression function were obtained during rest and during activation and were used to compute a stiffness index, defined as the ratio of activated stiffness to resting stiffness. For bradykinesia, we recorded the movement time during rapid alternating flexion and extension movements of the wrist by use of an electrogoniometer. Each of the three instrumental procedures yielded one score representing the more impaired upper extremity, regardless of hand dominance. These procedures have been found to yield reliable, valid, and sensitive indices of idiopathic extrapyramidal motor disturbances in Parkinson’s disease, AD, and neuroleptic-induced parkinsonism. Statistical Analyses Patients were assigned to subgroups on the basis of their posttreatment clinical motor ratings. Analysis of variance was used to examine group differences on demographic, pharmacologic, and pretreatment motor variables. Univariate logistic regression procedures were used to examine the strength of each variable in predicting clinical motor outcome.
RESULTS
sign (four with bradykinesia and one with masked facies). Thus, we found that 21% of the patients exhibited mild-to-moderate involvement of at least two of five cardinal parkinsonian signs, whereas 37.5% met more lenient criteria of mild-to-moderate involvement on a single sign, with seven of these nine patients exhibiting bradykinesia. Using a criterion of at least mild impairment, three patients (12.5%) exhibited rigidity, seven patients (29%) exhibited bradykinesia, one patient (4%) exhibited tremor, six (25%) exhibited masked facies, and one (4%) exhibited abnormal gait, based on observer ratings. Instrumental measures of parkinsonism. On the basis of a normative sample of 37 healthy older subjects with a mean age of 72.2Ⳳ7.3 years and on the basis of previously published normative data from a different group of 26 healthy older subjects with a mean age of 67.1Ⳳ10.1 years,13 we were able to dichotomize the AD patients into those with normal vs. those with abnormal instrumental motor function. Using the 95% confidence interval from these norms, we found that 14 patients (58%) exhibited abnormal scores on the instrumented rigidity measure; 13 patients (54%) performed abnormally on the bradykinesia measure, and three patients (12.5%) exhibited abnormal postural tremor.
Pretreatment Findings The mean (ⳲSD) score on the MMSE for the 24 patients at intake was 19.3Ⳳ 4.0. Fifteen patients (62.5%) had MMSE scores ⱕ20. Clinical ratings of parkinsonism. Pretreatment baseline clinical ratings of parkinsonism revealed five AD patients who exhibited mild-to-moderate parkinsonism characterized by bradykinesia and at least one other sign (two with rigidity, one with tremor, and two with masked facies). An additional four patients exhibited mild-tomoderate impairment of a single motor 78
Posttreatment Findings The presence and severity of NIP were based solely on the clinical ratings of parkinsonism as noted above. The mean SASm score recorded at baseline was 5.17Ⳳ3.48. There was a nonsignificant increase in SASm score for the 3-month posttreatment assessment, with a mean of 6.41Ⳳ4.40; (t[23]⳱1.33; P⬎0.1). There was a significant increase in SASm score for the 9month posttreatment assessment, with a mean of 11.52Ⳳ7.06 (t[23]⳱4.56; P⬍0.005). Mean severity ratings for rigidity, tremor, and bradykinesia are shown in VOLUME 6 • NUMBER 1 • WINTER 1998
Caligiuri et al.
FIGURE 1. Means and standard deviations (error bars) for clinical (SASm ) severity ratings of bradykinesia, rigidity, and tremor for 15 Alzheimer’s disease patients meeting criteria for neuroleptic-induced parkinsonism (NIP), at pretreatment and 3 and 9 months after intervention
7
Bradykinesia Rigidity Tremor
6
Severity Rating
5 4 3 2 1 0
Pretreatment
3 months
9 months
Assessment Interval Note: Asterisks indicate significance levels for differences between pre- and posttreatment ratings. (Maximum possible range of severity ratings: 0–8.) *P⬍0.05; **P⬍0.01; ***P⬍0.001.
Figure 1 for the baseline, 3-month, and 9month assessments. As shown by the figure, severity ratings for bradykinesia and rigidity increased steadily over time. The increase in tremor severity between Months 3 and 9 was not as robust as were the increases in rigidity and bradykinesia. Rigidity was the most common motor sign that worsened after neuroleptic treatment (i.e., increased at least 2 points on the rating scale for a given item); it was observed in 13 patients (54.2%). Worsening of bradykinesia was observed in nine patients (37.5%). Worsening of tremor or gait disturbances was observed in eight patients (33.3%). Last, only five patients (20.8%) developed masked facies after neuroleptic treatment. Seven patients (29%) met criteria for NIP at the 3-month posttreatment interval, with four of the seven patients meeting criteria 9 months after treatment, suggesting THE AMERICAN JOURNAL OF GERIATRIC PSYCHIATRY
the presence of persistent parkinsonism. In addition to these four patients, nine others met criteria for NIP at the 9-month posttreatment interval and were considered to exhibit tardive parkinsonism. Thus, 16 patients (66.7%) met criteria for NIP at some time during the 9-month follow-up period. Eight patients did not meet criteria for NIP at any posttreatment assessment. Table 1 shows descriptive statistics for the demographic, pharmacologic, and pretreatment motor variables for each of the four outcome groups: no NIP, acute NIP, tardive NIP, and persistent NIP. Only the pretreatment instrumental measure of bradykinesia distinguished among the patient groups (F[3,17]⳱3.46; P⬍0.05). Post-hoc analyses revealed that patients who developed acute NIP were significantly slower at baseline than patients who developed tardive NIP or who did not develop NIP. 79
Side Effects of Neuroleptics in AD
TABLE 1.
Age, pretreatment cognitive status, neuroleptic dose, and pretreatment instrumental motor variables for AD patients with various posttreatment outcomes, meansⴣstandard deviation
n Age, years MMSE Dosea SASm Instrumental: Rigidityb Instrumental: Tremorc Instrumental: Bradykinesiad
No NIP
Acute
Tardive
Persistent
Pe
8 76.2Ⳳ5.5 20.7Ⳳ3.0 22.6Ⳳ13.4 5.3Ⳳ4.1 35.1Ⳳ53.4 1.27Ⳳ7.7 320Ⳳ101
3 81.0Ⳳ11.2 17.3Ⳳ2.1 16.5Ⳳ6.9 7.0Ⳳ1.0 15.7Ⳳ13.5 8.3Ⳳ5.8 611Ⳳ192
9 73.6Ⳳ10.4 19.2Ⳳ4.4 29.2Ⳳ21.7 5.4Ⳳ3.7 43.3Ⳳ83.7 ⳮ0.9Ⳳ7.4 341Ⳳ150
4 72.0Ⳳ4.9 18.0Ⳳ6.1 32.2Ⳳ12.9 3.0Ⳳ2.5 22.9Ⳳ5.8 ⳮ1.7Ⳳ14.0 382Ⳳ76
⬎0.10 ⬎0.10 ⬎0.10 ⬎0.10 ⬎0.10 ⬎0.10 0.03
Note: NIP⳱neuroleptic-induced parkinsonism; MMSE⳱Mini-Mental State Exam (pretreatment cognitive status); SAS⳱Simpson-Angus Rating Scale for Extrapyramidal Side Effects. a mg chlorpromazine-equivalent per day. b % activated stiffness. c postural tremor, in decibels. d movement time, in milliseconds. e by ANOVA.
Univariate logistic regression analyses were conducted to examine the predictive utility of patients’ age, dementia severity, neuroleptic dosage, and pretreatment instrumental measures of motor function in predicting the severity of NIP. For these analyses, patients were dichotomized into two groups: those who met clinical criteria for NIP and those who did not. Separate analyses were conducted for the 3-month and 9-month results. For the 3-month outcome data, only the pretreatment instrument measure of bradykinesia reached statistical significance as an independent predictor of NIP outcome (F⳱19.04; P⬍0.001), correctly classifying 86% of the patients. For the 9-month outcome data, only neuroleptic potency reached statistical significance as an independent predictor of NIP outcome (F⳱4.54; P⬍0.05), correctly classifying 71% of the patients.
DISCUSSION In summary, there were three key findings of the present study. First, five of 24 AD patients (21%) exhibited signs of parkinsonism on the basis of clinical criteria before neuroleptic exposure. This number 80
increased to as high as 14 (58%) on the basis of instrument measures of parkinsonism. Second, when NIP occurred, its onset was delayed in the majority of patients. This observation may be related to the low dose of neuroleptics, to pharmacokinetic properties of the neuroleptic, or to the time required for pharmacodynamic changes responsible for extrapyramidal symptoms. Third, we found that pretreatment instrumental, but not clinical, measurement of bradykinesia was a predictor of posttreatment NIP. Approximately half of the AD patients exhibited impairment on at least one of the three instrumentally derived measures of parkinsonism at baseline. Clinical severity ratings revealed the presence of mild parkinsonian motor pathology in less than one-third of the patients. Bradykinesia was the most common overt clinical sign, whereas rigidity appeared more common on the basis of instrumentation. These findings are consistent with recent instrument motor studies of AD patients.12,13 A summary of the literature on extrapyramidal motor signs in patients meeting NINCDS/ADRDA criteria for probable AD16 indicated an overall prevalence of 36.5% for the presence of any parkinsonian VOLUME 6 • NUMBER 1 • WINTER 1998
Caligiuri et al.
motor sign. In that review, the estimated prevalence for parkinsonian tremor, rigidity, and bradykinesia was 9.5%, 30.3%, and 29.5%, respectively. The present results, based on clinical assessments, are consistent with these figures for bradykinesia and tremor; however, we found somewhat lower prevalence of rigidity among our patients. The relatively mild degree of dementia among our patients may explain the lower prevalence of rigidity, compared with the frequency found in the literature as a whole. Clinically significant NIP developed in AD patients treated with very low doses of neuroleptic, less than 25 mg of chlorpromazine-equivalent per day. The NIP symptoms primarily involved rigidity; next frequent were bradykinesia, tremor, gait disturbance, and masked facies. Our relatively high frequency of NIP among AD patients (66.7%) is consistent with McKeith et al.,7 who noted the presence of preexisting parkinsonism in their patients. The pretreatment instrument measure of bradykinesia distinguished patients who developed NIP from those who did not at comparable dosages. This study is the first to use instrumented procedures to predict clinically relevant neuroleptic-induced side effects. The principal advantage of the instrumentation appears to be its sensitivity to subclinical impairment. We found that the presence of instrument-derived parkinsonian motor signs, especially bradykinesia, may serve as a marker of increased vulnerability to adverse reactions of neuroleptic medication. This finding is consistent with previous prospective clinical studies17,18 suggesting that significant neuroleptic-induced rigidity and bradykinesia were common among AD patients with preexisting extrapyramidal motor involvement. The precise mechanism responsible for neuroleptic sensitivity in AD patients remains unclear; however, postmortem evidence19,20 suggests that the presence of Lewy body formation, gliosis of pigmented THE AMERICAN JOURNAL OF GERIATRIC PSYCHIATRY
nuclei, and neuronal loss may be causal factors for some AD patients. Perry et al.21 reported that Lewy body-variant AD patients had significantly lower neuron counts in the substantia nigra, relative to age-matched control subjects. Neuronal loss within the substantia nigra would compromise dopaminergic neurotransmission and predispose an individual to significant functional impairment after even low doses of neuroleptics. Unlike the compensation of schizophrenic patients undergoing neuroleptic treatment, compensation by dopamine-receptor up-regulation may be inadequate in AD because of the neuronal loss associated with the disorder. Insofar as the present study involved a limited patient sample size, the results must be considered preliminary, and they require replication from a larger sample of patients. Despite this limitation, the present findings underscore an important clinical issue pertaining to the use of neuroleptics in older patients with psychiatric or behavioral disturbances. Not all patients exhibit intolerance to the neuroleptics, as demonstrated by this and previous studies. These patients may be successfully managed with a low-dose pharmacologic strategy. Our long-term objective is to identify markers signaling the clinician that a particular patient may develop an unfavorable reaction to neuroleptics. The ability to identify AD patients with lower side-effect thresholds before treatment is initiated has important clinical ramifications. Neuroleptics may not be the treatment of choice for these patients; in these cases, alternative management strategies could be sought. Patients at greater risk for severe NIP could be monitored more closely than patients with less risk because the presence of tremor or rigidity may result in medication noncompliance.
The authors acknowledge the clinical staff of the Geriatric Psychiatry Clinical 81
Side Effects of Neuroleptics in AD
Research Center for their assistance in patient recruitment and Rebecca Vaughan for her assistance in data analysis. This research was supported in part
by a grant from Alzheimer Disease and Related Disorders, Inc. (PRG-93-025), and by USPHS grants R29-MH45959, R01MH45131, and R01-AG14291.
References 1. Wragg RE, Jeste DV: Neuroleptics and alternative treatments: management of behavioral symptoms and psychosis in Alzheimer’s disease and related conditions. Psychiatr Clin North Am 1988; 11:195–213 2. Schneider LS, Pollock VE, Lyness SA: A meta-analysis of controlled trials of neuroleptic treatment in dementia. J Am Geriatr Soc 1990; 38:553–563 3. Tune LE, Steele C, Cooper T: Neuroleptic drugs in the management of behavioral symptoms of Alzheimer’s disease. Psychiatr Clin North Am 1991; 14:353–373 4. Petrie WM, Ban TA, Berney S, et al: Loxapine in psychogeriatrics: a placebo- and standard-controlled clinical investigation. J Clin Psychopharmacol 1982; 2:122–126 5. Steele C, Lucas MJ, Tune L: Haloperidol vs. thioridazine in the treatment of behavioral symptoms in senile dementia of the Alzheimer’s type: preliminary findings. J Clin Psychiatry 1986; 47:310–312 6. Devanand DP, Sackeim HA, Brown RP, et al: A pilot study of haloperidol treatment of psychosis and behavioral disturbance in Alzheimer’s disease. Arch Neurol 1989; 46:854–857 7. McKeith I, Fairbairn A, Perry R, et al: Neuroleptic sensitivity in patients with senile dementia of Lewy body type. BMJ 1992; 305:673–678 8. McKann G, Drachman D, Folstein M, et al: Clinical diagnosis of Alzheimer’s disease: report of the NINCDS-ADRDA Work Group under the auspices of the Department of Health and Human Services Task Force on Alzheimer’s Disease. Neurology 1984; 34:939–944 9. Stern Y, Mayeux R, Sano M, et al: Predictors of disease course in patients with probable Alzheimer’s disease. Neurology 1987; 37:1649–1653 10. Simpson GM, Angus JW: A rating scale for extrapyramidal side effects. Acta Psychiatr Scand 1970; 45:11–12
82
11. Folstein MF, Folstein SE, McHugh PR: Mini-Mental State: a practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res 1975; 12:189–198 12. Kischka U, Mandir AS, Ghika J, et al: Electrophysiologic detection of extrapyramidal motor signs in Alzheimer’s disease. Neurology 1983; 43:500–505 13. Caligiuri MP, Lohr JB, Panton D, et al: A comparison of motor abnormalities associated with latelife psychosis and dementia. Schizophr Bull 1993; 19:747–754 14. Caligiuri MP, Lohr JB: Worsening of tremor in patients with levodopa-induced dyskinesia. Clin Neuropharmacol 1993; 16:244–250 15. Caligiuri MP: Portable device for quantifying parkinsonian rigidity. Mov Disord 1994; 9:57–63 16. Ellis R, Caligiuri MP, Galasko D, et al: Extrapyramidal signs in Alzheimer’s disease: a review. Alzheimer Dis Assoc Disord 1996; 10:103–114 17. Hamilton LD, Bennett JL: The use of trifluoperazine in geriatric patients with chronic brain syndrome. J Am Geriatr Soc 1962; 10:140–147 18. Girling DM, Berrios GE: Extrapyramidal signs, primitive reflexes and frontal lobe function in senile dementia of the Alzheimer type. Br J Psychiatry 1990; 157:888–893 19. Ditter SM, Mirra SS: Neuropathologic and clinical features of Parkinson’s disease in Alzheimer’s disease patients. Neurology 1987; 37:754–760 20. Hansen LA, Salmon D, Galasko D, et al: The Lewy body variant of Alzheimer’s disease: a clinical and pathological entity. Neurology 1990; 40:1–8 21. Perry EK, McKeith IG, Thompson P, et al: Topography, extent and clinical relevance of neurochemical deficits in dementia of Lewy body, Parkinson, and Alzheimer type. Ann NY Acad Sci 1991; 640:197–202
VOLUME 6 • NUMBER 1 • WINTER 1998
Extrapyramidal Side Effects in Patients With Alzheimer’s Disease Treated With Low-Dose Neuroleptic Medication Michael P. Caligiuri, Ph.D. Enid Rockwell, M.D. Dilip V. Jeste, M.D. The authors examined whether the presence of extrapyramidal side effects (EPS), measured before neuroleptic treatment was initiated, could be used to predict the development and severity of neuroleptic-induced parkinsonism (NIP) in Alzheimer’s disease (AD). Twenty-four newly medicated probable AD patients were assessed with a battery of measures of extrapyramidal motor function. Assessments were made before neuroleptic therapy, and 3 and 9 months after treatment. Posttreatment clinical findings revealed that 66.7% of the AD patients developed NIP. Patients who developed NIP exhibited more severe pretreatment bradykinesia on instrument-derived measures. These findings suggest that a substantial proportion of AD patients treated with neuroleptics develop significant EPS and that the risk for EPS can be estimated before intervention by use of instruments measuring motor function. (Am J Geriatr Psychiatry 1998; 6:75–82)
N
euroleptics are currently the most effective form of management of behavioral and psychiatric disturbances in Alzheimer’s disease (AD).1 Schneider et al.2 reviewed 33 published studies on the therapeutic outcome of neuroleptics in agitated dementia and concluded that
neuroleptic treatment offers modest benefit in the management of agitation. Despite this somewhat optimistic position, the use of neuroleptics with AD patients remains a controversial issue because of the significant side-effect profile associated with these medications.3 Using lower
Received September 16, 1996; revised January 20, 1997; accepted January 28, 1997. From the Department of Psychiatry, University of California, San Diego, and the Department of Veteran Affairs Medical Center, San Diego, California. Address correspondence to Dr. Caligiuri, Geriatric Psychiatry Clinical Research Center (116A1), VA Medical Center, 3350 La Jolla Village Drive, San Diego, CA 92161. e-mail: [email protected]. Copyright 䉷 1998 American Association for Geriatric Psychiatry THE AMERICAN JOURNAL OF GERIATRIC PSYCHIATRY
75
Side Effects of Neuroleptics in AD
dosages or low-potency neuroleptics does not necessarily reduce the risk of debilitating extrapyramidal side effects.4–6 Preexisting parkinsonism has been shown to negatively influence a patient’s ability to tolerate neuroleptic treatment. McKeith et al.7 observed that over half of their patients with senile dementia of Lewy-body type who were given neuroleptics in standard dosage showed acute and often irreversible adverse reactions indicative of a neuroleptic sensitivity syndrome. Survival analysis showed that patients who experience adverse reactions to neuroleptics had increased mortality in the year after neuroleptics were initiated, compared with patients with mild or no neuroleptic sensitivity. Although neuroleptic intervention for behavioral disturbances in AD cannot always be avoided, it may be helpful to understand the risk for developing disabling motor side effects. The purpose of the present study is to examine whether certain demographic variables or motor disturbances, measured before neuroleptic treatment was initiated, could be used to predict the development and severity of neuroleptic-induced parkinsonism (NIP) in AD patients treated with very low dose neuroleptics.
METHODS Subjects A group of 24 community-dwelling AD patients were recruited from the University of California at San Diego (UCSD) Geropsychiatry Clinical Research Center for inclusion into the study. Patients were identified by their primary physician and referred to our center as part of an ongoing research effort to study adverse effects of antipsychotics in late life. Diagnoses of probable AD were made by staff neurologists and psychiatrists according to NINCDS/ADRDA8 criteria. Patients were 76
excluded from study for the following reasons: 1) viral, metabolic, or traumatic causes of dementia; 2) history of alcoholism, stroke, or other neurological illness; 3) evidence of non-parkinsonian movement disorder; 4) diagnosis of idiopathic Parkinson’s disease; and/or 5) severe musculoskeletal disorders that would impair mobility. The mean (ⳲSD) age of the AD patients was 75.1Ⳳ8.3 years. The purpose of the study was explained to the patient and Significant Other or caregiver, and informed consent was obtained before neuroleptic treatment was initiated. Neuroleptic Treatment Patients were enrolled in the study before neuroleptic medication was initiated. Neuroleptics were administered for psychiatric disturbances, including psychosis and agitation, at the discretion of the patient’s primary physician. The physician determined the appropriate neuroleptic type (potency) and dosage. Sixteen patients were treated with haloperidol, and the remaining eight patients were treated with thioridazine. None of the patients received concomitant anticholinergic medication. The mean initial dosage, expressed in chlorpromazine-equivalents, was 26.8 Ⳳ17.8 mg/day. Each patient’s neuroleptic status was evaluated by Center staff psychiatrists monthly. The mean dosage at the 9-month assessment was 25.9Ⳳ18.2 mg/ day. The neuroleptic dose was increased in 1 patient, decreased in 2 patients, and remained constant for the remaining 21 patients over the 9-month study period. Baseline and Follow-Up Assessments Patients being considered for neuroleptic treatment who met inclusion criteria were seen for an initial examination consisting of neurological screening to rule out potential vascular or metabolic etiologies for their dementia, a brief cognitive assessment, and clinical and quantitative VOLUME 6 • NUMBER 1 • WINTER 1998
Caligiuri et al.
extrapyramidal motor assessment. Most patients were followed for several years after exposure to neuroleptics as part of an ongoing study of tardive dyskinesia; however, we report only the 3- and 9-month follow-up data in the present study so as to reduce potential confounding due to increasing illness duration and cognitive decline.9 Outcome Variable We used a modified version of the Simpson-Angus Rating Scale for Extrapyramidal Side Effects (SAS)10 to rate the severity of pretreatment parkinsonism and NIP. The standard scale consisted of 11 items rated using a four-point severity index, with greater emphasis on rigidity than bradykinesia. The standard scale was modified for three reasons. First, it does not contain a measure of bradykinesia, which is an important feature of parkinsonism associated with dementia. Second, the scale is biased strongly toward rigidity—this factor encompasses 4 of the 11 items. Third, the original scale does not consider asymmetry, which may prove to be an important feature in NIP. Our modification of this scale reduced the number of items to eight, consisting of ratings of gait, left and right upper-extremity tremor, left and right upper-extremity bradykinesia, left and right upper-extremity rigidity, and masked facies. The modified scale has been in use in our center by trained staff for over 7 years. Raters undergo periodic sessions to maintain high levels of interrater reliability. Parkinsonian signs were rated by use of a 5-point scale ranging from 0 (no impairment) to 4 (severe impairment), with 1, 2, and 3 representing minimal, mild, and moderate impairment, respectively. The modified scale (SASm) yielded a total score ranging from 0 to 32. Patients who exhibited a total score of at least 7 (mild-tomoderate impairment) were considered to exhibit NIP. Patients who exhibited an increase in two of the five items of at least THE AMERICAN JOURNAL OF GERIATRIC PSYCHIATRY
2 points or exhibited an increase in total SASm score of at least 200% above baseline after neuroleptic treatment were considered to have developed NIP. SASm ratings were made at baseline and at the 3- and 9month follow-up assessments. Predictor Variables Several independent variables were chosen as potential predictors of NIP. Total SASm score, as well as item scores for bradykinesia, rigidity, and tremor, served as the predictor variables. Neuroleptic dose, converted to standard chlorpromazine equivalents, and potency (haloperidol or thioridazine) served as the pharmacologic predictor variables. Age served as the sole demographic variable because estimates of illness duration were thought to be unreliable and were, therefore, not used. The Mini-Mental State Exam (MMSE)11 served as the cognitive status predictor variable. Three instrument-measured motor predictors were obtained for quantifying postural tremor, rigidity, and bradykinesia. The primary purpose of obtaining instrumental measures of motor function was to increase the sensitivity of detecting motor disturbances in patients before neuroleptic exposure. Recent studies in AD patients with and without clinically rated extrapyramidal involvement have demonstrated increased diagnostic sensitivity with instrument measures of motor function.12,13 Detailed descriptions of the instrumental procedures, their validation, reliability, and sensitivity, have been previously reported,14,15 and only a brief description is presented here. Using a strain gauge-instrumented force lever, we measured the spectral amplitude within the frequency range of 3–7 Hz to quantify upper-extremity (hand) tremor. For rigidity, we quantified activated stiffness by use of a device worn on the wrist that transduced force and rotation during passive movement. Stiffness slopes, derived 77
Side Effects of Neuroleptics in AD
from the force/rotation regression function were obtained during rest and during activation and were used to compute a stiffness index, defined as the ratio of activated stiffness to resting stiffness. For bradykinesia, we recorded the movement time during rapid alternating flexion and extension movements of the wrist by use of an electrogoniometer. Each of the three instrumental procedures yielded one score representing the more impaired upper extremity, regardless of hand dominance. These procedures have been found to yield reliable, valid, and sensitive indices of idiopathic extrapyramidal motor disturbances in Parkinson’s disease, AD, and neuroleptic-induced parkinsonism. Statistical Analyses Patients were assigned to subgroups on the basis of their posttreatment clinical motor ratings. Analysis of variance was used to examine group differences on demographic, pharmacologic, and pretreatment motor variables. Univariate logistic regression procedures were used to examine the strength of each variable in predicting clinical motor outcome.
RESULTS
sign (four with bradykinesia and one with masked facies). Thus, we found that 21% of the patients exhibited mild-to-moderate involvement of at least two of five cardinal parkinsonian signs, whereas 37.5% met more lenient criteria of mild-to-moderate involvement on a single sign, with seven of these nine patients exhibiting bradykinesia. Using a criterion of at least mild impairment, three patients (12.5%) exhibited rigidity, seven patients (29%) exhibited bradykinesia, one patient (4%) exhibited tremor, six (25%) exhibited masked facies, and one (4%) exhibited abnormal gait, based on observer ratings. Instrumental measures of parkinsonism. On the basis of a normative sample of 37 healthy older subjects with a mean age of 72.2Ⳳ7.3 years and on the basis of previously published normative data from a different group of 26 healthy older subjects with a mean age of 67.1Ⳳ10.1 years,13 we were able to dichotomize the AD patients into those with normal vs. those with abnormal instrumental motor function. Using the 95% confidence interval from these norms, we found that 14 patients (58%) exhibited abnormal scores on the instrumented rigidity measure; 13 patients (54%) performed abnormally on the bradykinesia measure, and three patients (12.5%) exhibited abnormal postural tremor.
Pretreatment Findings The mean (ⳲSD) score on the MMSE for the 24 patients at intake was 19.3Ⳳ 4.0. Fifteen patients (62.5%) had MMSE scores ⱕ20. Clinical ratings of parkinsonism. Pretreatment baseline clinical ratings of parkinsonism revealed five AD patients who exhibited mild-to-moderate parkinsonism characterized by bradykinesia and at least one other sign (two with rigidity, one with tremor, and two with masked facies). An additional four patients exhibited mild-tomoderate impairment of a single motor 78
Posttreatment Findings The presence and severity of NIP were based solely on the clinical ratings of parkinsonism as noted above. The mean SASm score recorded at baseline was 5.17Ⳳ3.48. There was a nonsignificant increase in SASm score for the 3-month posttreatment assessment, with a mean of 6.41Ⳳ4.40; (t[23]⳱1.33; P⬎0.1). There was a significant increase in SASm score for the 9month posttreatment assessment, with a mean of 11.52Ⳳ7.06 (t[23]⳱4.56; P⬍0.005). Mean severity ratings for rigidity, tremor, and bradykinesia are shown in VOLUME 6 • NUMBER 1 • WINTER 1998
Caligiuri et al.
FIGURE 1. Means and standard deviations (error bars) for clinical (SASm ) severity ratings of bradykinesia, rigidity, and tremor for 15 Alzheimer’s disease patients meeting criteria for neuroleptic-induced parkinsonism (NIP), at pretreatment and 3 and 9 months after intervention
7
Bradykinesia Rigidity Tremor
6
Severity Rating
5 4 3 2 1 0
Pretreatment
3 months
9 months
Assessment Interval Note: Asterisks indicate significance levels for differences between pre- and posttreatment ratings. (Maximum possible range of severity ratings: 0–8.) *P⬍0.05; **P⬍0.01; ***P⬍0.001.
Figure 1 for the baseline, 3-month, and 9month assessments. As shown by the figure, severity ratings for bradykinesia and rigidity increased steadily over time. The increase in tremor severity between Months 3 and 9 was not as robust as were the increases in rigidity and bradykinesia. Rigidity was the most common motor sign that worsened after neuroleptic treatment (i.e., increased at least 2 points on the rating scale for a given item); it was observed in 13 patients (54.2%). Worsening of bradykinesia was observed in nine patients (37.5%). Worsening of tremor or gait disturbances was observed in eight patients (33.3%). Last, only five patients (20.8%) developed masked facies after neuroleptic treatment. Seven patients (29%) met criteria for NIP at the 3-month posttreatment interval, with four of the seven patients meeting criteria 9 months after treatment, suggesting THE AMERICAN JOURNAL OF GERIATRIC PSYCHIATRY
the presence of persistent parkinsonism. In addition to these four patients, nine others met criteria for NIP at the 9-month posttreatment interval and were considered to exhibit tardive parkinsonism. Thus, 16 patients (66.7%) met criteria for NIP at some time during the 9-month follow-up period. Eight patients did not meet criteria for NIP at any posttreatment assessment. Table 1 shows descriptive statistics for the demographic, pharmacologic, and pretreatment motor variables for each of the four outcome groups: no NIP, acute NIP, tardive NIP, and persistent NIP. Only the pretreatment instrumental measure of bradykinesia distinguished among the patient groups (F[3,17]⳱3.46; P⬍0.05). Post-hoc analyses revealed that patients who developed acute NIP were significantly slower at baseline than patients who developed tardive NIP or who did not develop NIP. 79
Side Effects of Neuroleptics in AD
TABLE 1.
Age, pretreatment cognitive status, neuroleptic dose, and pretreatment instrumental motor variables for AD patients with various posttreatment outcomes, meansⴣstandard deviation
n Age, years MMSE Dosea SASm Instrumental: Rigidityb Instrumental: Tremorc Instrumental: Bradykinesiad
No NIP
Acute
Tardive
Persistent
Pe
8 76.2Ⳳ5.5 20.7Ⳳ3.0 22.6Ⳳ13.4 5.3Ⳳ4.1 35.1Ⳳ53.4 1.27Ⳳ7.7 320Ⳳ101
3 81.0Ⳳ11.2 17.3Ⳳ2.1 16.5Ⳳ6.9 7.0Ⳳ1.0 15.7Ⳳ13.5 8.3Ⳳ5.8 611Ⳳ192
9 73.6Ⳳ10.4 19.2Ⳳ4.4 29.2Ⳳ21.7 5.4Ⳳ3.7 43.3Ⳳ83.7 ⳮ0.9Ⳳ7.4 341Ⳳ150
4 72.0Ⳳ4.9 18.0Ⳳ6.1 32.2Ⳳ12.9 3.0Ⳳ2.5 22.9Ⳳ5.8 ⳮ1.7Ⳳ14.0 382Ⳳ76
⬎0.10 ⬎0.10 ⬎0.10 ⬎0.10 ⬎0.10 ⬎0.10 0.03
Note: NIP⳱neuroleptic-induced parkinsonism; MMSE⳱Mini-Mental State Exam (pretreatment cognitive status); SAS⳱Simpson-Angus Rating Scale for Extrapyramidal Side Effects. a mg chlorpromazine-equivalent per day. b % activated stiffness. c postural tremor, in decibels. d movement time, in milliseconds. e by ANOVA.
Univariate logistic regression analyses were conducted to examine the predictive utility of patients’ age, dementia severity, neuroleptic dosage, and pretreatment instrumental measures of motor function in predicting the severity of NIP. For these analyses, patients were dichotomized into two groups: those who met clinical criteria for NIP and those who did not. Separate analyses were conducted for the 3-month and 9-month results. For the 3-month outcome data, only the pretreatment instrument measure of bradykinesia reached statistical significance as an independent predictor of NIP outcome (F⳱19.04; P⬍0.001), correctly classifying 86% of the patients. For the 9-month outcome data, only neuroleptic potency reached statistical significance as an independent predictor of NIP outcome (F⳱4.54; P⬍0.05), correctly classifying 71% of the patients.
DISCUSSION In summary, there were three key findings of the present study. First, five of 24 AD patients (21%) exhibited signs of parkinsonism on the basis of clinical criteria before neuroleptic exposure. This number 80
increased to as high as 14 (58%) on the basis of instrument measures of parkinsonism. Second, when NIP occurred, its onset was delayed in the majority of patients. This observation may be related to the low dose of neuroleptics, to pharmacokinetic properties of the neuroleptic, or to the time required for pharmacodynamic changes responsible for extrapyramidal symptoms. Third, we found that pretreatment instrumental, but not clinical, measurement of bradykinesia was a predictor of posttreatment NIP. Approximately half of the AD patients exhibited impairment on at least one of the three instrumentally derived measures of parkinsonism at baseline. Clinical severity ratings revealed the presence of mild parkinsonian motor pathology in less than one-third of the patients. Bradykinesia was the most common overt clinical sign, whereas rigidity appeared more common on the basis of instrumentation. These findings are consistent with recent instrument motor studies of AD patients.12,13 A summary of the literature on extrapyramidal motor signs in patients meeting NINCDS/ADRDA criteria for probable AD16 indicated an overall prevalence of 36.5% for the presence of any parkinsonian VOLUME 6 • NUMBER 1 • WINTER 1998
Caligiuri et al.
motor sign. In that review, the estimated prevalence for parkinsonian tremor, rigidity, and bradykinesia was 9.5%, 30.3%, and 29.5%, respectively. The present results, based on clinical assessments, are consistent with these figures for bradykinesia and tremor; however, we found somewhat lower prevalence of rigidity among our patients. The relatively mild degree of dementia among our patients may explain the lower prevalence of rigidity, compared with the frequency found in the literature as a whole. Clinically significant NIP developed in AD patients treated with very low doses of neuroleptic, less than 25 mg of chlorpromazine-equivalent per day. The NIP symptoms primarily involved rigidity; next frequent were bradykinesia, tremor, gait disturbance, and masked facies. Our relatively high frequency of NIP among AD patients (66.7%) is consistent with McKeith et al.,7 who noted the presence of preexisting parkinsonism in their patients. The pretreatment instrument measure of bradykinesia distinguished patients who developed NIP from those who did not at comparable dosages. This study is the first to use instrumented procedures to predict clinically relevant neuroleptic-induced side effects. The principal advantage of the instrumentation appears to be its sensitivity to subclinical impairment. We found that the presence of instrument-derived parkinsonian motor signs, especially bradykinesia, may serve as a marker of increased vulnerability to adverse reactions of neuroleptic medication. This finding is consistent with previous prospective clinical studies17,18 suggesting that significant neuroleptic-induced rigidity and bradykinesia were common among AD patients with preexisting extrapyramidal motor involvement. The precise mechanism responsible for neuroleptic sensitivity in AD patients remains unclear; however, postmortem evidence19,20 suggests that the presence of Lewy body formation, gliosis of pigmented THE AMERICAN JOURNAL OF GERIATRIC PSYCHIATRY
nuclei, and neuronal loss may be causal factors for some AD patients. Perry et al.21 reported that Lewy body-variant AD patients had significantly lower neuron counts in the substantia nigra, relative to age-matched control subjects. Neuronal loss within the substantia nigra would compromise dopaminergic neurotransmission and predispose an individual to significant functional impairment after even low doses of neuroleptics. Unlike the compensation of schizophrenic patients undergoing neuroleptic treatment, compensation by dopamine-receptor up-regulation may be inadequate in AD because of the neuronal loss associated with the disorder. Insofar as the present study involved a limited patient sample size, the results must be considered preliminary, and they require replication from a larger sample of patients. Despite this limitation, the present findings underscore an important clinical issue pertaining to the use of neuroleptics in older patients with psychiatric or behavioral disturbances. Not all patients exhibit intolerance to the neuroleptics, as demonstrated by this and previous studies. These patients may be successfully managed with a low-dose pharmacologic strategy. Our long-term objective is to identify markers signaling the clinician that a particular patient may develop an unfavorable reaction to neuroleptics. The ability to identify AD patients with lower side-effect thresholds before treatment is initiated has important clinical ramifications. Neuroleptics may not be the treatment of choice for these patients; in these cases, alternative management strategies could be sought. Patients at greater risk for severe NIP could be monitored more closely than patients with less risk because the presence of tremor or rigidity may result in medication noncompliance.
The authors acknowledge the clinical staff of the Geriatric Psychiatry Clinical 81
Side Effects of Neuroleptics in AD
Research Center for their assistance in patient recruitment and Rebecca Vaughan for her assistance in data analysis. This research was supported in part
by a grant from Alzheimer Disease and Related Disorders, Inc. (PRG-93-025), and by USPHS grants R29-MH45959, R01MH45131, and R01-AG14291.
References 1. Wragg RE, Jeste DV: Neuroleptics and alternative treatments: management of behavioral symptoms and psychosis in Alzheimer’s disease and related conditions. Psychiatr Clin North Am 1988; 11:195–213 2. Schneider LS, Pollock VE, Lyness SA: A meta-analysis of controlled trials of neuroleptic treatment in dementia. J Am Geriatr Soc 1990; 38:553–563 3. Tune LE, Steele C, Cooper T: Neuroleptic drugs in the management of behavioral symptoms of Alzheimer’s disease. Psychiatr Clin North Am 1991; 14:353–373 4. Petrie WM, Ban TA, Berney S, et al: Loxapine in psychogeriatrics: a placebo- and standard-controlled clinical investigation. J Clin Psychopharmacol 1982; 2:122–126 5. Steele C, Lucas MJ, Tune L: Haloperidol vs. thioridazine in the treatment of behavioral symptoms in senile dementia of the Alzheimer’s type: preliminary findings. J Clin Psychiatry 1986; 47:310–312 6. Devanand DP, Sackeim HA, Brown RP, et al: A pilot study of haloperidol treatment of psychosis and behavioral disturbance in Alzheimer’s disease. Arch Neurol 1989; 46:854–857 7. McKeith I, Fairbairn A, Perry R, et al: Neuroleptic sensitivity in patients with senile dementia of Lewy body type. BMJ 1992; 305:673–678 8. McKann G, Drachman D, Folstein M, et al: Clinical diagnosis of Alzheimer’s disease: report of the NINCDS-ADRDA Work Group under the auspices of the Department of Health and Human Services Task Force on Alzheimer’s Disease. Neurology 1984; 34:939–944 9. Stern Y, Mayeux R, Sano M, et al: Predictors of disease course in patients with probable Alzheimer’s disease. Neurology 1987; 37:1649–1653 10. Simpson GM, Angus JW: A rating scale for extrapyramidal side effects. Acta Psychiatr Scand 1970; 45:11–12
82
11. Folstein MF, Folstein SE, McHugh PR: Mini-Mental State: a practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res 1975; 12:189–198 12. Kischka U, Mandir AS, Ghika J, et al: Electrophysiologic detection of extrapyramidal motor signs in Alzheimer’s disease. Neurology 1983; 43:500–505 13. Caligiuri MP, Lohr JB, Panton D, et al: A comparison of motor abnormalities associated with latelife psychosis and dementia. Schizophr Bull 1993; 19:747–754 14. Caligiuri MP, Lohr JB: Worsening of tremor in patients with levodopa-induced dyskinesia. Clin Neuropharmacol 1993; 16:244–250 15. Caligiuri MP: Portable device for quantifying parkinsonian rigidity. Mov Disord 1994; 9:57–63 16. Ellis R, Caligiuri MP, Galasko D, et al: Extrapyramidal signs in Alzheimer’s disease: a review. Alzheimer Dis Assoc Disord 1996; 10:103–114 17. Hamilton LD, Bennett JL: The use of trifluoperazine in geriatric patients with chronic brain syndrome. J Am Geriatr Soc 1962; 10:140–147 18. Girling DM, Berrios GE: Extrapyramidal signs, primitive reflexes and frontal lobe function in senile dementia of the Alzheimer type. Br J Psychiatry 1990; 157:888–893 19. Ditter SM, Mirra SS: Neuropathologic and clinical features of Parkinson’s disease in Alzheimer’s disease patients. Neurology 1987; 37:754–760 20. Hansen LA, Salmon D, Galasko D, et al: The Lewy body variant of Alzheimer’s disease: a clinical and pathological entity. Neurology 1990; 40:1–8 21. Perry EK, McKeith IG, Thompson P, et al: Topography, extent and clinical relevance of neurochemical deficits in dementia of Lewy body, Parkinson, and Alzheimer type. Ann NY Acad Sci 1991; 640:197–202
VOLUME 6 • NUMBER 1 • WINTER 1998