Alcohol, Dementia, and Alzheimer\'s Disease: Comparison of Neuropsychological Profiles

Journal of Geriatric Psychiatry and Neurology http://jgp.sagepub.com/

Alcohol, Dementia, and Alzheimer's Disease: Comparison of Neuropsychological Profiles Judith Saxton, Cynthia A. Munro, Meryl A. Butters, Carol Schramke and Melissa A. McNeil J Geriatr Psychiatry Neurol 2000 13: 141 DOI: 10.1177/089198870001300308 The online version of this article can be found at: http://jgp.sagepub.com/content/13/3/141

Published by: http://www.sagepublications.com

Additional services and information for Journal of Geriatric Psychiatry and Neurology can be found at: Email Alerts: http://jgp.sagepub.com/cgi/alerts Subscriptions: http://jgp.sagepub.com/subscriptions Reprints: http://www.sagepub.com/journalsReprints.nav Permissions: http://www.sagepub.com/journalsPermissions.nav Citations: http://jgp.sagepub.com/content/13/3/141.refs.html

>> Version of Record - Jan 1, 2000 What is This?

Downloaded from jgp.sagepub.com at UNIV OF PITTSBURGH on June 11, 2013

Alcohol, Dementia, and Alzheimer’s Disease: Comparison of Neuropsychological Profiles Judith Saxton, PhD, Cynthia A. Munro, PhD, Meryl A. Butters, PhD, Carol Schramke, PhD, and Melissa A. McNeil, MD

ABSTRACT Thirty-nine detoxified elderly alcoholics (mean age = 65.85) completed a comprehensive assessment designed to identify individuals meeting DSM-IV criteria for alcohol-related dementia. Ten subjects meeting criteria (mean age = 69.8; mean Mini-Mental State Examination [MMSE] = 25.1) were compared to the 29 nondemented alcoholics (mean age = 64.5; mean MMSE = 27.8), 9 patients with Alzheimer’s disease (mean age = 73.4; mean MMSE = 22.3), and 15control subjects (mean age = 70.8; mean MMSE = 28). Comparison of neuropsychological test scores revealed several statistically significant differences. Furthermore, the overall pattern of test performance between the two demented groups was different. Alzheimer’s patients were more impaired on confrontation naming, recognition memory, animal fluency, and orientation. Alcohol dementia subjects were more impaired than controls on initial letter fluency, fine motor control, and free recall. However, alcohol dementia subjects did not differ from controls on tests of verbal recognition memory. This study suggests that it is possible to clinically differentiate the cognitive deficits of alcohol-related dementia from typical Alzheimer’s disease. However, the results are preliminary and are based on small sample sizes so sliould be interpreted with caution. (J Geriatr Psychiatry Neurol2000; 13:141-149).

Alcohol-related dementia is essentially a descriptive term used clinically to refer to older alcoholic subjects who manifest a slowly progressive decline in intellectual and cognitive abilities without the profound amnestic disorder of Wernicke-Korsakoff syndrome (WKS).14 Alcoholrelated dementia is clinically recognized in the Diagnostic and Statistic Manual of Mental Disorders, Fourth Edition (DSM-IW5under the term alcohol-induced persisting dementia. It has been estimated that between 22% and 29% of dementia cases may be alcohol Among patients presenting for treatment of alcoholrelated disorders, it has been suggested that up to 23% have dementia of some type.g In addition, an increased prevalence of dementia has been reported in older, compared to younger, drinkers.lOJ1The existence of the syn-

From the Department of Psychiatry, University of Pittsburgh (Drs. Saxton, AIunro and Butters), Department ofNeurology and Psychiatry (Dr. S c h r a m k e ) , Allegheny G e n e r a l H o s p i t a l a n d Veteran’s Administration AIedical Center (Dr. AIcNeil), Pittsburgh, Pennsylvania.

This work was supported by grant numberAA10981 from the National Institute on Alcohol Abuse and Alcoholism to Judith Saxton.

Reprint requests: Dr. Judith Saxton, Western Psychiatric Institute 8: Clinic, 3811 O H a r a Street, Room 1098, Pittsburgh, PA 15213.

drome of alcohol-related dementia as a distinct entity, however, is not universally accepted,12and it has been suggested that the majority of cases of alcohol-related dementia are, in fact, WKS, Alzheimer’s disease, or a dementia of vascular origin. The increased risk of Alzheimer’s disease in older individuals presents a dilemma for the clinician in the differential diagnosis of dementia in these older patients with a history of significant alcohol consumption. Cognitively impaired elderly individuals with a significant lifetime alcohol intake may be more likely to be diagnosed with alcohol-relateddementia when, in fact, a significant number of these elderly alcoholic subjects may have mixed Alzheimer’s disease and alcohol-relateddementia. Thus, elderly alcoholic subjects who experience a progressive cognitive decline secondary to Alzheimer’s disease may benefit from recently developed pharmacologic agents aimed a t slowing the cognitive progression of Alzheimer’s. However, if there is a tendency to diagnose the cognitive disorder experienced by these individuals as purely alcohol related, these medications may not be offered. One method of investigating the relation between alcohol use and dementia is to compare the neuropsychological tesl profiles of demented patients with and 141

Downloaded from jgp.sagepub.com at UNIV OF PITTSBURGH on June 11, 2013

142

Journal of Geriafric Psychiatry and Neurology I Vol. 13, Fall 2000

without significant alcohol abuse histories and contrast these with the profiles seen i n elderly alcoholic subjects with and without dementia. The cognitive profile associated with early Alzheimer’s disease has been well documented and includes deficits in orientation, naming, category fluency, and both free recall and recognition memory with milder impairments in visuoperceptual ability and initial-letter fluency. The cognitive profile associated with alcohol-related dementia has been the focus of few investigations. However, the literature suggests t h a t nondemented individuals with alcohol dependence histories demonstrate deficits in visuoperceptual abilities and executive tasks, with mild deficits in free recall and relatively well-preserved naming ability, category fluency, and recognition memory. This study had three goals: (1)to characterize the neuropsychological test profile associated with alcoholrelated dementia, (2) to compare the neuropsychological test profile of individuals meeting criteria for alcoholrelated dementia to patients with Alzheimer’s disease, and (3)to compare the neuropsychological test profile of alcohol-related dementia to t h a t seen in nondemented alcoholic subjects.

METHOD Procedure All subjects completed a structured clinical interview to determine if they met the DSM-IV5 criteria for alcohol dependence and/or alcohol-related dementia. Subjects also completed a brief neurologic examination and a comprehensive lifetime drug and alcohol use questionnaire, together with medical and psychiatric history questionnaires including a n assessment of depression and laboratory tests (see below for details) followed by a two-phase cognitive assessment. The first phase included a brief, standardized cognitive test battery used exclusively to help m a k e the clinical diagnosis of dementia. This assessment is described in full below. The second phase involved a comprehensive neuropsychological assessment designed to allow comparison of the neuropsycho-

Table 1.

N

logical test profiles across the groups. Data were confirmed by a significant other whenever possible and by review of the individual’s medical chart.

Subjects All subjects were at least 55 years of age. Thirty-nine recently detoxified elderly alcoholic subjects (35 males and 4 females) were recruited from substance abuse clinics at three major hospitals in the Greater Pittsburgh Metropolitan area (Veterans Administration Medical Center, Highland Drive; Veterans Administration Medical Center, University Drive; and St. Francis Medical Center). Subjects were recruited and informed consent obtained at their primary medical site. Subjects were given a breathalyzer test (Alco-Sensor IV)to ensure that they were sober at the time of testing. No subject reported a history of other substance use. Fifteen normal elderly control subjects were also recruited from the community (eight males, seven females). I n addition, nine individuals (three males, six females) meeting National Institute of Neurological a n d Communicative Disorders and StrokdAlzheimer’s Disease and Related Disorders Association (NINCDUADRDA) criteria13 for probable Alzheimer’s disease were recruited from the Alzheimer’s Disease Research Center (ADRC) at the University of Pittsburgh. Subjects were excluded if they had a history of other conditions known to affect cognitive functioning, such as stroke, WKS, coma, or significant psychiatric history (e.g., schizophrenia, bipolar disorder, or current major depression). Table 1 presents the demographic information for the four groups. Alcohol-Related D e m e n t i a Subjects Ten alcoholic subjects (nine males, one female) met DSMIV criteria for alcohol-related dementia. The mean age of the group was 69.8 years (SD = 9.3; range = 55-83 years), and the mean education level was 12.1 years (SD = 3.5; range = 8-20 years). These subjects had been drinking a n average of 47.2 years (SD = 8.5 years; range = 35-62 years) and had been abstinent at the time of testing for a mean of 128.8 weeks (SD = 161.0 weeks; range

Demographic Information Together with ANOVAs and Post Hoc Tukey Tests ( P c .05)

Male/female Age Education Mini-Mental State Examination National Adult Reading Test-Revised

A: Alzheimer‘s Disease

B: Alcohol Dementia

C: Alcohol Nondemented

D: Control

Mean (SD)

Mean (SD)

Mean (SD)

Mean (SD)

ANOVA

Tukey

9 316 73.4 (5.5) 11.9 (2.5) 22.3 (4.4) 101.7 (8.7)

10 9/1 69.8 (9.31 12.1 (3.5) 25.1 (3.0) 98.5 (9.7)

29 2613 64.5 (6.7) 12.2 (2.2) 27.8 (2.0) 100.3 (11.9)

15 8/7 70.8 (6.61 13.2(2.7) 28.0 (1.8) 106.4 (7.5)

NA NA F (3,591 = 5.67* F (3.59) = 0.69 F (3,591 = 13.04*** F (3.59) = 1.53

NA NA A = B. A > C,C < D Pz.560 A = B < C = D, A < C p = 215

A. B. C. and D refer 10 groups. = means same as, > means better than, c means worse than, X > Y, Z intermediate means Z lies between X and Y but is not significantly different from either, “ * P < .001, “ P c .01. * P < .05.

Downloaded from jgp.sagepub.com at UNIV OF PITTSBURGH on June 11, 2013

Alcohol, Dementia, and Alzheimer’s Disease I Saxton et al

= 4-504 weeks). Subjects reported a n average of 58.0 drinks per week (SD = 38.2; range = 8-126) during their

heaviest drinking period.

Nondemented Alcoholic Subjects Twenty-nine subjects (26 males, 3 females) met DSM-IV criteria for alcohol dependence but did not meet criteria for alcohol-related dementia, The mean age of the group was 64.5 years (SD = 6.1; range = 56-81 years), and the mean education level was 12.2 years (SD = 2.2; range = 9-18 years). These subjects had been drinking an average of 43.9 years (SD = 8.5 years; range = 26-60 years) and had been abstinent a t the time of testing for a mean of 39 weeks (SD = 52.4 weeks; range = 4-248 weeks). Subjects reported an average of 119.9 drinks (SD = 10.2; range = 15-189) per week during their heaviest drinking period. Alzheimer’s Disease Subjects Nine subjects (three males, six females) meeting NINCDUADRDAcriteria13for probable Alzheimer’s disease were recruited from the ADRC at the University of Pittsburgh. The diagnosis of Alzheimer’s disease a t the center is made following a comprehensive medical, psychiatric, neurologic, and neuropsychological assessment.14J5The mean age of the Alzheimer’s group was 73.4 years (SD = 5.5; range = 63-82 years), and the mean education level was 11.9 years (SD = 2.5; range = 8-17 years). None of the Alzheimer’s patients met DSM-IV criteria for alcohol abuse or dependence.

143

transminase (serum glutamate pyruvate transaminase), aspartate transaminase (serum glutamic oxaloacetic transaminase), albumin: thyroid stimulating hormone (TSH),T4, and Bl,. Although all subjects completed liver function tests, not all subjects completed the B,,, TSH, and T4 tests as these were added during the course of the study.

Structured Clinical Interview for D S M - N (SCID.P The SCID is a well-validated semistructured clinical assessment for the diagnosis of psychiatric disorders. A modified and expanded version of the Substance Use Disorders Section of the SCID17was used to diagnose substance use disorders. Lifetime History of Drinking and Drug Usela The Lifetime History of Drinking interview captures distinct phases and changes in a person’s lifetime drinking patterns by asking specific questions about quantity and frequency of drinking. Quantity and frequency of alcohol use were determined for each separate drinking period over the lifetime of the subject using the timeline followback procedure described below. In addition, the number and length of each period of abstinence were recorded. Based on the results of a pilot study, we modified this assessment slightly for use with geriatric subjects.

Assessment Scales All subjects completed the following scales:

Alcohol Timeline F o l l o ~ b a c k ~ ~ ~ ~ ~ This assessment provides a detailed picture of a n individual’s drinking over a designated time period. Based on the results of a pilot study, we modified this assessment slightly for use with geriatric subjects. This system uses life events as anchor points to define the length of each drinking episode and period of abstinence. The Lifetime History of Drinking questionnaire described above captured the amount and frequency of drinking at each episode. The timeline followback procedure was used to identify the period of heaviest drinking and length of each period of abstinence.

Clinical Interview All subjects completed a structured clinical interview that included demographic information, income, family living arrangements, medical and psychiatric history, curr e n t medications, subjective evidence of memory problems, and ability to carry out everyday activities.

Geriatric Depression Scale (GDS)21-z3 The GDS is a self-rating scale for assessment of depressive symptoms in elderly individuals. The abbreviated scale of 15yedno questions was used in this study. Subjects receiving a score of 5 o r greater were also administered the Hamilton Depression Rating Scale.

Laboratory Tests A range of laboratory studies were conducted to evaluate presence or absence of those disorders associated with alcoholism (e.g., liver disease and those disorders often associated with cognitive decline in the elderly, such as B,, deficiency and thyroid dysfunction). The following blood studies were-completed: total bilirubin, alanine

Hamilton Depression Rating Scale (Hamilton DRS)z4iz5 The Hamilton DRS is a widely used and well-validated interviewer-rated clinical assessment of symptoms of depression. A score of 17 or above indicates significant depression. Current major depression mas a n exclusionary criterion. No subject scored at this level.

Normal Control Subjects Fifteen normal control subjects (eight males, seven females) were recruited from the community. The mean age of these subjects was 70.8 years (SD = 6.6; range = 59-82 years), and the mean education level mas 13.2 years (SD = 2.7; range = 8-17 years). None of the normal control subjects met DSM-N criteria for alcohol abuse or dependence.

Downloaded from jgp.sagepub.com at UNIV OF PITTSBURGH on June 11, 2013

144

Journal of Geriatric Psychiatry and Neurology I Vol. 13, Fall 2000

Brief Neurologic Examination Abrief examination was administered to identify (1)presence of peripheral neuropathy, (2) nystagmus (vertical or horizontal and end gaze), (3)gait ataxia (10-footwalk assessing presence of shuffling, staggering, wide-based gait, and tandem walking); (4)cerebellar ataxia, (5) presence of tremor, and (6) presence of visual field cuts. Diagnosis ofDenentia Following review of the clinical interviews and screening battery test scores, the diagnosis of alcohol-related dementia was made by two of the authors (JS and MB), who were blind to the patients’ performance on the comprehensive neuropsychological battery. Alcohol-related dementia is recognized in DSM-IV under the heading of Alcohol-Induced Persisting Dementia (code 291.2). Individuals meet criteria if the following factors are met: (A) Evidence of impairment in memory plus impairment in one other cognitive domain. (B) These deficits cause significant impairment in social or occupational functioning and represent a decline from a previous level. (C) The deficits do not occur as part of a delirium and exist beyond intoxication and withdrawal. (D) There is clinical evidence that the deficits are related to the persisting effects of alcohol use. Cognitive deficit was operationalized in this study as performance a t least 1.5 SD below the normal control mean for a given test on the dementia screening battery (see below).

Dementia Screening Battery This test battery was developed specifically to determine whether subjects were experiencing cognitive impairment sufficient to meet DSM-IV criteria for alcohol-related dementia. The battery included measures from the dementia assessment of the Consortium to Establish a Registry of Alzheimer’s Disease (CERADIZ6 together with other tests sensitive to dementia: 1. MMSEZ7-a brief 30-item assessment of mental status. 2. National Adult Reading Test-Revised (NARTR)2830-a brief method of assessing premorbid verbal IQ. 3. CERAD battery9 (a) Word List Learning test (WLLTblO-word list learning task with three trials and delayed recall. (b) Constructional praxis-copy and recall of a circle, diamond, intersecting rectangles, and a cube. (c) Verbal fluency-number of articles of clothing named in 1 minute.

(d) 15-item Boston Naming Test (BNT) drawn from the 60-item BNT (see below for description of test). 4. Trailmaking Test.3l The score is the time to complete each of the two parts. Part Aassesses psychomotor speedrattention and Part B assesses psychomotor speedlattention and working memory. 5. Clock drawing. Subject is asked to draw a clock and set the time a t 10 after 11. Best possible score is 9.

Neuropsychological Test Battery All subjects completed a comprehensive neuropsychological test battery. The battery focused on a range of cognitive functions including memory, language, visuospatial ability, executive functioning, and motor functioning. The following tests were selected from a larger battery: 1. Wechsler Adult Intelligence Scale-Revised Information s u b t e ~ t ~ ~ measure -a of semantic knowledge. 2. California Verbal Learning Test (CVLT)33-a 16-item word list learning task with five learning trials, delayed free recall, and recognition memory (i.e., ability to discriminate target from nontarget words). ability to 3. BNT 60-item ~ersion~~-measures name line drawings. 4. Fluency: (a) Category-number of animals named in 1 minute and grocery items named in 1minute. (b) Initial letter-number of words beginning with f, a, and s named in 1minute each. 5. Token Test? Subject is asked to perform simple manipulations with colored tokens, which provides a measure of verbal comprehension. 6. Rey Ostenieth Complex Figure (ROCF)36,37copy, immediate recall, and delayed recall of a complex figure. mea7. Benton Visual Form Di~crimination~~-a sure of visual perception that involves matching figures with no memory requirement. recog8. Recognition Memory Te~t~~-forced-choice nition memory for words and faces. Sub9. Wisconsin Card Sorting Test (WSCT).4°,41 ject sorts cards according to various concepts. The task assesses ability to form abstract concepts, maintain, and shift set. 10. Grooved Pegb~ard~~-assesses biIatera1 fine motor control.

RESULTS Analyses of variance (ANOVAs) with post hoc Tukey tests were conducted to compare the four groups. No sig-

Downloaded from jgp.sagepub.com at UNIV OF PITTSBURGH on June 11, 2013

Alcohol, Dementia, and Alzheimer's Disease / Saxton et

nificant differences were observed between demented and nondemented alcoholics in nutritional status, blood studies, mean number of abstinence periods (demented = 3.63; nondemented = 3.69), or mean number of head injuries. Three (33.3%) demented alcoholic subjects reported a head injury involving loss of consciousness greater than 1hour and 12 (41.4%)nondemented reported head injury. Statistically significant differences were observed among the groups for age (Alzheimer's disease subjects were older, P < .05)(see Table l), but no differences among the groups were found for education level (F= 2.75, P >.05). There mere no differences in estimated premorbid verbal I&scores of the four groups (based on performance on the NART-R) and no difference in the MMSE scores of the two demented groups (Alzheimer's patients and alcohol-related dementia patients). Table 2 presents the performance of the four groups on the dementia screening battery. Significant differences reflect the selection criteria for the four groups (i.e., DSM-IV criteria for diagnosis of memory requires evidence of memory impairment plus one other cognitive domain; see Method section above). Table 3 shows performance on the neuropsychological test battery. Comparison of the overall pattern of test performance indicates markedly different test profiles for the three clinical groups. Figure 1illustrates standardized scores of representative measures of several cognitive domains. To obtain the ,values used in this figure, the mean scores of each subject group were converted to z-scores relative to the control group. Thus, the mean score for each measure for the Alzheimer's disease group was assigned relative to only the control group. Likewise,

Table 2.

al

145

the scores for each alcohol group (i.e., demented and nondemented) were compared separately to the control group and were assigned z-scores relative to controls. Specific group comparisons are discussed below.

Alzheimer's Disease and Alcohol-Related Dementia There were no statistically significant differences in the MMSE scores or estimated premorbid verbal IQ scores of the two demented groups. Comparison of the neuropsychological test scores revealed statistically significant differences between the two groups on four tests (see Table 3). Patients with Alzheimer's disease performed worse than alcohol-related dementia subjects on the test of confrontation naming (BNT), the test of category fluency (animals), and two tests of recognition memory (the discriminability index from the CVLT and the Recognition Memory test for words). Impaired category fluency in Alzheimer's patients with no difference between Alzheimer's patients and alcohol-related dementia subjects on a test of initial letter fluency supports reports in the literature of a specific semantic impairment in Alzheimer's disease. Impaired performance on a test of confrontation naming in Alzheimer's patients with no difference on this task between alcohol subjects and control subjects supports the suggestion of significant naming impairments in patients with Alzheimer's disease and relative sparing of language deficits in alcoholic subjects. Finally, there was no difference between either of the alcohol groups and normal controls on tests of recognition memory. This finding lends support to reports in the literature that alcoholic subjects have normal recognition memory.

Dementia Screening Battery Together with ANOVAs and Post Hoe Tukey Tests (P< .05) A: Alzheimer's Disease

8: Alcohol Dementia

C: Alcohof Nondernented

D: Control

Mean (SD)

Mean (SD)

Mean (SD)

Mean (SDl

ANOVA

Tukey

22.33 (4.4) 101.71 (8.7)

25.00 (3.0) 98.54 (9.7)

27.83 (2.0) 100.33 (11.9)

28.00 (1.8) 106.37 (7.5)

F (3,591 = 13.04*** F (3.59) = 1.53

A =BcC=D P = .215

11.22 (3.4) 1.33 (2.1) 8.89 (2.7)

12.50 (3.4) 2.70 (1.5) 7.90 (1.8)

17.48 (3.3) 5.55 (1.8) 8.93 (1.8)

21.60 (4.3) 6.87 (1.5) 10.40 (2.0)

F (3.59) = 21.35*** F (3.59) = 26.15***

Delayed recall Category fluency

3.11 (3.0) 11.89 (2.2)

4.20 (2.5) 15.78 (5.9)

7.17 (2.8) 17.21 (4.5)

8.71 11.8) 19.67 (4.5)

F (3,58) = 11.75***

'15-item Boston Naming Test

12.00 (2.7)

13.60 (1.4)

14.17 (1.5)

14.87 (0.4)

F (3,59) = 6.84***

72.67 (33.2) 45.70 (9.7) 235.26 (98.3) 158.00 (88.3)

44.38 (24.1) 120.00 (79.2)

33.27 (8.2) 98.13 (24.9)

F (3,591 = 6.54** F (3,56) = 2.62**

6.93 (1.3)

7.73 (1.1)

F (3,59) = 4.04"

A