International Journal of Neuroscience
ISSN: 0020-7454 (Print) 1543-5245 (Online) Journal homepage: http://www.tandfonline.com/loi/ines20
The Relationship Between Disability and Memory Dysfunction in Multiple Sclerosis Howard R. Kessler, Ronald A. Cohen, Kevin Lauer & Donald F. Kausch To cite this article: Howard R. Kessler, Ronald A. Cohen, Kevin Lauer & Donald F. Kausch (1991) The Relationship Between Disability and Memory Dysfunction in Multiple Sclerosis, International Journal of Neuroscience, 62:1-2, 17-34 To link to this article: http://dx.doi.org/10.3109/00207459108999754
Published online: 07 Jul 2009.
Submit your article to this journal
Article views: 18
View related articles
Citing articles: 5 View citing articles
Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=ines20 Download by: [Tufts University]
Date: 08 December 2015, At: 18:17
0 1992 Gordon and Breach Science Publishers S.A. Printed in the United States of America
Inrern. J . Neuroscience, 1992, Vol. 62, pp. 11-34 Reprints available directly from the publisher Photocopying permitted by license only
THE RELATIONSHIP BETWEEN DISABILITY AND MEMORY DYSFUNCTION IN MULTIPLE SCLEROSIS HOWARD R. KESSLER Department of Psychology, Cushing Hospital, Framingham, MA 01701, U . S . A .
Downloaded by [Tufts University] at 18:17 08 December 2015
RONALD A. COHEN Department of Neurology, University of Massachusetts Medical Center, Worcester, MA, U . S . A .
KEVIN LAUER and DONALD F. KAUSCH Department of Psychology, Bowling Green State University, Bowling Green, O H , U . S . A . (Received April 19, 1991) We examined the relationship between memory impairment and functional disability in multiple sclerosis. Tests of memory, sensorimotor ability, and functional capacity were administered to fifty-six subjects with chronic-progressive or remitting-relapsing MS. Sensorimotor impairment, functional disability, and chronicity predicted impairment on various measures of memory acquisition, while age and type of diagnosis did not. After accounting for the effects of initial acquisition, delayed-recall performance was weakly-associated with disability. We suggest that: (1) Functional disability is associated with memory loss in MS; (2) MS-forgetting is caused by defective acquisition, rather by a deficit in consolidation or storage; (3) Level of disease activity, rather than type of MS diagnosis, determines the degree of memory impairment; and (4) MS disability needs to be evaluated multidimensionally, to account for both neurologic and functional impairment.
Keywords: multiple scierosis; memory: disability; learning; motor; information-processing.
Multiple sclerosis (MS) produces relatively circumscribed neuropsychological deficits, with impairment limited primarily to memory, executive functions, and the modulation of affect (Rao, 1986). In addition, MS produces considerable physical and functional disability. While MS patients possess a grossly intact immediate recall span (Minden, Moes, Orav, Kaplan and Reich, 1990; Heaton, Nelson, Thompson, Burks and Franklin, 1985; Grant, McDonald, Trimble, Smith and Reed, 1984; Rao, Hammeke, McQuillen, Khatri and Lloyd, 1984; Marsh, 1980; Staples and Lincoln, 1979), the disease exerts an effect on recent memory function, regarding both initial acquisition and later retrieval of information from long-term storage. Deficient performance in word-list learning (Jambor, 1969; Rao, et al., 1984; Minden, et al., 1990), narrative recall (Minden, et al., 1990), the Brown-Peterson task (Grant, et al., 1984), visuographic memory tasks (Minden, et al., 1990), and spatial memory tasks (Rao, et al., 1984) suggests the presence of deficient encoding, secondary in part to the efRequests for reprints to: H . R. Kessler, Ph.D., Department of Psychology, Cushing Hospital, P.O. Box 9008, Dudley Road, Framingham, MA 01701. 17
Downloaded by [Tufts University] at 18:17 08 December 2015
18
H.R. KESSLER et al.
fects of proactive interference. Defective retrieval operations appear to be manifested on a variety of delayed-recall and retrieval tasks, including the Wechsler Memory Scale stimuli (Grant, et al., 1984; Rao, et al., 1984; Staples and Lincoln, 1979) and on the Selective Reminding Test (Elpern, Gunderson, Kattah and Kirsch, 1984). MS patients may also be susceptible to retroactive interference and suffer from rapid forgetting (Rao, et al., 1984). Jambor (1969) initially postulated that the memory impairments in MS may be associated with plaques in the periventricular white matter. Rao, et al. (1984) later implicated more specifically the frontal-limbic system connections. Various recent studies have suggested that cognitive deficits in MS are closely related to structural, subcortical abnormalities, as revealed in CT or MRI scans and rCBF (Rao, Glatt, Hammeke, McQuillen, Khatri, Rhodes and Pollard, 1985; Brooks, Leenders, Head, Marshall, Legg and Jones, 1984; Franklin, Heaton, Nelson, Filley and Seibert, 1988; Reischies, Baum, Brau, Hedde and Schwindt, 1988; Rao, Leo, Haughton, St. Aubin-Faubert and Bernardin, 1989; Huber, Paulson, Shuttleworth, Chakeres, Clapp, Pakalnis, Weiss, and Rammohan, 1987; Franklin, Nelson, Filley and Heaton, 1989). Rao, et al. (1985) found initially that deficits in memory function were related to atrophy surrounding the area of the third ventricle, as observed on CT scan. Subsequent studies have replicated these findings. Memory dysfunction in MS has also been associated with observed attentional and information-processing deficits. Van den Burg, van Zomeren, Minderhoud, Prange and Meijer (1987) have suggested that slowed information processing speed, secondary to subcortical demyelination, may account for many of these cognitive deficiencies. Cohen and Fisher (1989) demonstrated that MS patients exhibit primary deficits in attention, information processing capacity and psychomotor speed, and related these deficits to fatigue. It has further been suggested that fatigue frequently interferes with functional status in MS (Caruso, LaRocca, Foley, Robbins and Smith, 1991). There is an apparent relationship among the presence of structural abnormality, memory dysfunction and fatigue in MS. Further, fatigue appears related to functional disability. Yet, there remains little agreement on the relationship between disability status and memory dysfunction in MS. The lack of any substantive relationship has been demonstrated utilizing both the Kurtzke (Minden, et al., 1990; Heaton, et al., 1985; Marsh, 1980) and Poser (Peyser, Edwards, Poser and Filskov, 1980) scales. This is likely because they are relatively brief measures geared toward ambulation capacity or specific neurologic deficits, rather than higher cortical functions. The limited usefulness of these scales is amply demonstrated in Baumhefner, Tourtellotte, Syndulko, Waluch, Ellison, Meyers, Cohen, Osborne, and Shapsak (1990) by the lack of correlation between scores on the Kurtzke Extended Disability Status Scale and MRI abnormalities in the brains of MS patients. Further, findings on the general neurological examination do not correlate with cognitive dysfunction in MS (Heaton, et al., 1985), and the prevalence of deficits on the neurological examination is far exceeded by the frequency of cognitive dysfunction in most studies (see Kahana, Leibowitz and Alter, 1971; Peyser, et al., 1980). This is not surprising, since false negatives on the general neurological examination are common in other degenerative disorders, for example Alzheimer’s Disease (Alexander and Geschwind, 1984). Indirect evidence suggests that disability may be related to cognitive dysfunction in MS. Heaton, et al. (1985) found that patients with chronic-progressive MS performed significantly worse than those with remitting-relapsing MS across a wide range of measures. As a result of this finding, one might infer a relationship between
Downloaded by [Tufts University] at 18:17 08 December 2015
DISABILITY IN MULTIPLE SCLEROSIS
19
disability status and cognitive performance, given the relatively more chronic and severe effects of chronic-progressive MS on the central nervous system. This finding has since been replicated by Beatty, Goodkin, Monson and Beatty (1989) and by Minden, et al. (1990). Hence, one might infer that the lack of correlation between disability status and cognitive dysfunction is due more to the inadequacy of the disability estimates, rather than by the lack of a substantive relationship. In response to the variability in the MS literature on cognitive dysfunction, and also to the difficulty in delineating the relationship between functional impairment and cognitive dysfunction, Peyser, Rao, LaRocca and Kaplan ( 1 990) have proposed guidelines for neuropsychological research. These criteria include a variety of parameters for characterizing patient samples, including demographics, diagnosis by category, and chronicity. In addition, they call for the use of measures of physical handicap, as well as data concerning psychiatric status and medication usage. This study has investigated the relationship between disability status, as reflected in functional deficits and motor performance, and memory and learning in a MS sample. Purportedly, deficits in activities of daily living (ADLs) should be more closely related to deficits in memory function, as global ADL capacity is a closer approximation of overall functional status (or physical handicap) than any of the aforementioned measures of physical disability. Many of the variables set forth by Peyser, et al. (1990) were examined as well.
METHOD Subjects Fifty-six MS patients from the Multiple Sclerosis Clinic of the Medical College of Ohio at Toledo (MCOT) and from local MS support groups were studied. All subjects expressed interest via telephone contact in participating in the study. All were under the care of a qualified physician on an outpatient basis. The sample contained 41 females and 15 males, ranging in age from 23.9 to 60.1 years (mean = 43.5, S.D. = 9.1) and in education from 8 to 20 years (mean = 12.9., S.D. = 2.3). Three subjects were Black, the remaining 53 Caucasian. Fortysix were classified as chronic-progressive and 10 as remitting-relapsing, while all were diagnosed as having clinically-probable or definite MS using the criteria of Poser, et al. (1983). The sample had a mean of 13.0 years (S.D. = 8.4) since initial onset of MS symptoms and a mean of 9.9 years since initial diagnosis. Seven subjects were taking adrenocorticoids at the time of testing, while 11 had received ACTH treatment within the prior year. Materials
Various standardized tests were used to assess memory, general cognitive ability, and physical and functional status. These are described below. WAIS-R Digit Span subtest This test was administered in the standard fashion. The lengths of the longest correctly-produced digit string for both forward and backward span were used for scoring, rather than using scaled-scores (Wechsler, 1981).
Modified Knox Cubes This test, developed by Kimura (1960), consists of 5 1-inch blocks arranged consecutively (approximately 1 inch apart) on a strip of wood. The subject is required to imitate block-tapping sequences (each 5 items long) for both immediate and 5-second delayed-recall conditions. There are 10 trials per condition.
20
H.R. KESSLER et al.
The task is scored for the total number of sequences correctly-reproduced under each of the two conditions.
Downloaded by [Tufts University] at 18:17 08 December 2015
WMS Logical Memory subtest We utilized the two stories contained in the original WMS (Wechsler and Stone, 1946), which contain a total of 46 separate “memories.” They were administered in the standard manner, which requires that the subject recall as much of each story as possible, after presentation by the examiner. The two stories are administered separately, and were scored using the standardized criteria developed by Schwartz and Ivnik (1980). WMS Associate Learning subtest This was also drawn from the original WMS, and comprises 6 “easy” and 4 “hard” word pairs, which are presented for recall across 3 trials. Incorrect responses are corrected during each trial, in order to facilitate additional learning. The score used was the total across the three trials for “easy” and “hard” items combined. WMS Visual Reproduction subtest This is a visuographic recall test of geometric figures. There are 3 stimulus cards, and the subject is required to reproduce each after a 10-second exposure. Scoring is subdivided into a total of 14 possible points. The standardized criteria developed by Schwartz and Ivnik (1980) were used for scoring purposes. All three WMS subtests were initially administered in the standard manner. This was followed by a delayed-recall of all stimuli after 30 minutes, and again after 2 weeks. Califarnia Verbal Learning Test (CVLT) This test, developed by Delis, Kramer, Kaplan, and Ober (1987), comprises two 1Qitem word lists, each including 4 items within each of 4 semantic categories. The first list (Monday) includes items of clothing, fruits, tools, and spices and herbs; the second list (Tuesday) includes fish, utensils, fruits and spices and herbs. The Tuesday items in the latter two categories differ from those on the Monday list. The Monday list is initially presented 5 times, with immediate recall following each presentation. The Tuesday list is then presented once for immediate recall, followed by both short-delayed free- and cued-recall trials of the Monday list. Free- and cued-recall are once again required after a 20-minute delay period, after which a recognition test is presented. Trials are scored for total number correct, and the test manual provides a wide variety of memory indices for analysis. In addition to the standardized administration of the CVLT, delayed freeand cued-recall and recognition testing were administered again after a one-week delay period. Recurring Figures Test Developed by Kimura (1963), this consists of 140 cards, each containing a line drawing of either a geometric or an irregular nonsense figure. Twenty target designs are shown for 3 seconds each during a study trial. This is followed by 5 recognition testing trials, each comprising 20 cards, including 8 of the original targets (4 geometric, 4 nonsense) interspersed within each trial. The subject is tested for accurate recognition of the target stimuli, using yes-no responding. There are 2 additional blocks of 20 cards each, which are used for recognition after a specified delay period (one week in the present study). Scoring is for total number of true positives minus number of false positive, responses.
DISABILITY IN MULTIPLE SCLEROSIS
21
Facial Recognition Test This was developed by Milner (1968), and includes a set of 12 target pictures presented in a 4 by 3 array for a 45-second study period. This is followed by a 90-second, verbally-filled delay. Subjects are then asked to identify the original target pictures from a 5 by 5 array of faces, including the 12 targets and 13 distractors. Scoring is the number of targets correctly identified.
Downloaded by [Tufts University] at 18:17 08 December 2015
WAIS-R Similarities and Picture Completion subtests These were administered and scored in the standardized manner. The age-scaled scores were used for analysis (Wechsler, 1981). Finger-Tapping Test Adapted from Halstead (1947), this is a lever and mechanical counter mounted on a wooden board. The subject is instructed to tap using his/her index finger, as fast as possible for a 10-second interval. This is done using both hands. Scoring is the mean of 5 10-second trials for each hand. Performance for only the dominant hand was used for analysis. Grooved Pegboard This is a metallic board, 4 inches square, with 5 rows of 5 holes, each with a randomly positioned slot. The pegs must be rotated properly in order to be inserted accurately. Pegs must be inserted as rapidly as possible, one row at a time from top to bottom. Each hand is tested separately, and is scored for the time required to complete the task. Only the performance for the dominant hand was used for analysis. Because some subjects were too impaired to complete the task (and hence could not yield a scorable result) scoring was instead based upon an extension of the Russell, Neuringer and Goldstein (1970) norms to a 7-point scale. Benton Visual Form Discrimination Test (BVFD) This test of visuospatial capacity was developed by Benton, Hamsher, Varney and Spreen (1983). It consists of 2 sample and 16 test items, each comprising a target stimulus and an array of 4 alternatives. Each target has 2 “Major” figures and l “Minor” peripheral figure. The 4 alternatives include the target stimulus, a stimulus with an error involving the “Minor” peripheral figure, one with the rotation of one of the “Major” figures, and one with the distortion of one of the “Major” figures. Each type of alternative is represented 4 times in each of 4 alternative spaces, numbered 1 to 4. The test can be administered for either memory or simple matching. The latter technique was used in this study. Correct responses receive 2 points, while peripheral errors receive 1 point. The score is the total number of points across 16 trials. Activities of Daily Living Scale This measure was adapted from a study by Staples and Lincoln (1979) on MS disability and memory function. By measuring areas far beyond mobility and cortical signs, this scale taps a broader range of functional impairment relative to measures used in many other studies (notably the Kurtzke and Poser scales, for example). Further, impairment on selected items from this scale correlated with aspects of memory impairment in the original study. The ADL scale consists of 42 items, representing the following functional domains: Mobility (10 items), Communication (4 items), Personal Care (5 items), Domestic Activities (18 items), Education (1 item), Employment (1 item), and Social Activities (3 items). Each item is scored on a 1 to 3 scale, with lower scores representing less impairment. Scores can be totaled by subtest or across all items. Because of their ambiguity, the Employment and Education items were excluded from analysis. The test was completed by each subject, and separately by a significant other.
22
H.R. KESSLER et al.
Procedure Testing was accomplished across 3 sessions, each separated by one week. This was done in order to minimize the effects of interference and fatigue, and to allow for recall testing across 1 and 2-week intervals. Subjects were asked the approximate time of day at which they felt most capable to perform, and each session was held at the same scheduled time. The testing procedure is outlined in Table 1.
RESULTS
Downloaded by [Tufts University] at 18:17 08 December 2015
Memory Performance Related to Disease and Demographic Variables
Intercorrelations among demographic and disease-related-variables, and performance on major memory test indices, are reported in Tables 2 and 3. The subjects’ own responses were used for analyzing the disability score, since there was a very high Spearman rank-order correlation coefficient between the two separate reports ( r = .95, p < .001). In addition, because chronicity (time since initial symptom onset) correlated highly with years since diagnosis (Pearson Product-Moment Correlation coefficient = .90, p < .Owl),only chronicity was used for analysis. Age (Pearson r ) and diagnosis (point-biserial correlations) are not included, as each failed to correlate with a single memory index. Sex (point-biserial correlations) failed to correlate significantly with any visual memory tests, but did correlate with selected verbal measures, related primarily to initial acquisition of stimuli on list-learning tasks (CVLT, Associate Learning). Education correlated with multiple memory indices, including measures of both verbal and nonverbal recent recall (Logical Memory, Facial Recognition), as well as with a measure of verbal immediate memory span (Digits Backward). Chronicity correlated with a limited number of measures, but was related to multiple memory processes: It correlated with both initial acquisition and delayed recall on both verbal and visual measures (CVLT, Recurring Figures), as well as with visual immediate memory (Knox Cubes). The more direct measures of disability status correlated with memory indices with far greater frequency: Disability (ADL Scale) correlated with indices across all modalities and stages of acquisition and recall, with the exception of verbal immediate memory span (Digit Span); finger-tapping correlated with selected indices across all memory tests, although the pattern was somewhat variable. Specifically, it correlated better with the easier immediate memory span tests (Digits Forward, Knox CubesImmediate) than with the more difficult span measures (Digits Backward, Knox Cubes-Delayed); it also correlated more widely with verbal than visual measures, notably where it did not correlate with the long-delayed visual recall measures (Visual Reproduction 2-week recall, Recurring Figures 1-week recall). Finally, performance on the Grooved Pegboard correlated significantly with every memory index examined. Memory and Degree of Disability
In order to delineate the relative contribution of each disease and performance variable to prediction of memory performance better, the memory acquisition measures were regressed on the following predictor variables: Type of MS diagnosis; chronicity; disability score; finger tapping; and Grooved Pegboard. (The delayed recall measures were not used in these analyses, as they correlated very highly with the acquisition measures; analyses to be cited later will serve to elucidate this point.) An examination of Table 4 reveals that the following measures received statistically-
DISABILITY IN MULTIPLE SCLEROSIS
23
TABLE 1 Testing Procedure
Downloaded by [Tufts University] at 18:17 08 December 2015
~~
~
~
Session I A. Informed Consent B. WMS subtests 1 . Logical Memory 2. Visual Reproduction 3. Associate Learning C. 30 minutes of interpolated activity I . Picture Completion 2. Similarities 3. Finger Tapping 4. Grooved Pegboard D. Delayed recall 1 . Logical Memory 2. Visual Reproduction 3. Associate Learning E. ADL Scale given to subject, to be returned at following session. Session I1 A. CVLT 1 . 5 “Monday” list trials 2. “Tuesday” list trial 3. Short-delay, free recall, “Monday” list 4. Short-delay, cued recall, “Monday” list B . 20 minutes of interpolated activity 1. Recurring Figures Test 2. Other visual measures (not reported) C. CVLT Long-delay testing 1. Free recall of “Monday” list 2. Cued recall of “Monday” list 3. Recognition testing D. Modified Knox Cubes Test E. ADL Scale collected from subject Session 111 A. Facial Recognition Test B. Two-week delay testing of WMS subtests 1 . Logical Memory 2. Visual Reproduction 3. Associate Learning C. One-week delay testing of Recurring Figures Test D. One-week delay testing of CVLT 1. Free recall of ‘Monday” list 2. Cued recall of “Monday” list 3. Recognition testing E. Visual Form Discrimination Test
significant predictions: Knox Cubes, both the Immediate and Delay conditions; CVLT Monday Total recall; Logical Memory; Associate Learning; Visual Reproduction; and the Recurring Figures. Performance on the Grooved Pegboard contributed to prediction of performance on all of these tests, the exception being the Recurring Figures. Chronicity (Knox Cubes-Immediate, Recumng Figures) and Disability (Knox Cubes-Immediate, CVLT, Visual Reproduction, Recurring Figures) were less powerful and/or influential predictors. Finger-tapping and type of diagnosis did not contribute significantly (tapping contributed to Facial Recognition prediction, but it was not statistically-significant) to any prediction. It is not surprising that diagnosis did not serve as a useful predictor, as it did not correlate with any of the measures. A
Chronicity
-.27 -.25 -.27 -.23 -.26 .05 N.S. N.S. N.S. .05 -.53 -.51 -.52 -.34 -.50 .02 ,0001 ,0001 .0001 ,0001 .38 .47 .46 Tapping .29 .48 ,001 ,001 .01 .05 ,001 .57 .49 .49 Pegboard .33 .45 .02 .001 .OOOl .OOOl .OW1 CVLT = California Verbal Learning Test MONl = Monday, Trial 1 MON5 = Monday, Trial 5 MONT = Monday, Total recall SDF = Short-delayed free recall LDF = Long-delayed free recall WEEK = One-week delayed free recall
-.15 N.S. -.33 .02 .31 .05 .37 .01
VR = Visual Reproduction IMM = Immediate recall DEL = 30-minute delayed-recall 2WK = Two-week delayed recall
AL = Associate Learning IMM = Immediate recall DEL = 30-minute delayed-recall 2WK = Two week delayed-recall
-.21 -.17 -.14 -.24 -.11 -.29 N.S. N.S. N.S. N.S. N.S. .05 -.19 -.27 -.16 -.38 -.48 p.47 N.S. .05 N.S. .01 ,001 .001 .20 .25 .29 .32 .29 .33 .05 .02 N.S. N.S. .05 .02 .31 .37 .30 .34 .44 .51 .05 .01 .05 .02 .001 .OOOI LM = Logical Memory IMM-Immediate recall DEL = 30-minute delayed-recall 2WK = Two week delayed-recall -.I7 -.12 N.S. N.S. -.20 -.15 N.S. N.S. .21 .29 .05 N.S. .34 .32 .02 .01
-.38 -.22 -.39 .01 N.S. .01 -.34 -.I3 -.28 -.43 .01 N.S. .05 ,001 .29 .24 .32 .42 .05 N.S. .02 .01 .36 .30 .28 .42 .01 .05 .05 .37 = Digits Forward = Digits Backwards
MKC = Modified Knox Cubes IMM = Immediate Condition DEL = Delayed condition
MFR = Milner Facial Recognition
KIM = Kimura Recurring Figures IMM = Immediate condition DEL = Delayed condition
.01
-.42
.01
-.30 .05 -.16 N.S. .25 N.S. .37
VR VR KIM KIM MKC MKC DEL 2WK IMM DEL MFR IMM DEL
-.21 -.26 -.24 N.S. N.S. N.S. p . 4 4 -.43 -.34 ,001 ,001 .02 .37 .33 .26 .01 .02 N.S. .45 .40 .36 ,001 .O I .01 DIGFOR DIGBACK
CVLT CVLT CVLT CVLT CVLT CVLT LM LM LM AL AL AL DIG DIG VR MONl MON5 MONT SDF LDF WEEK 1MM DEL 2WK IMM DEL 2WK FOR BACK IMM
Intercorrelations Between Memory Indices and Predictor Variables
TABLE 2
Downloaded by [Tufts University] at 18:17 08 December 2015
P
N
DISABILITY IN MULTIPLE SCLEROSIS
25
TABLE 3 Intercorrelations Among Disease-Related Variables Chronicity Disability Pegboard
Downloaded by [Tufts University] at 18:17 08 December 2015
Tapping
Disability
Pegboard
.07
n.s. -.39 p < .01 - .28 p < .os
- .60
p < ,0001
-
- .6S
p < .0001
.63 p
ISSN: 0020-7454 (Print) 1543-5245 (Online) Journal homepage: http://www.tandfonline.com/loi/ines20
The Relationship Between Disability and Memory Dysfunction in Multiple Sclerosis Howard R. Kessler, Ronald A. Cohen, Kevin Lauer & Donald F. Kausch To cite this article: Howard R. Kessler, Ronald A. Cohen, Kevin Lauer & Donald F. Kausch (1991) The Relationship Between Disability and Memory Dysfunction in Multiple Sclerosis, International Journal of Neuroscience, 62:1-2, 17-34 To link to this article: http://dx.doi.org/10.3109/00207459108999754
Published online: 07 Jul 2009.
Submit your article to this journal
Article views: 18
View related articles
Citing articles: 5 View citing articles
Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=ines20 Download by: [Tufts University]
Date: 08 December 2015, At: 18:17
0 1992 Gordon and Breach Science Publishers S.A. Printed in the United States of America
Inrern. J . Neuroscience, 1992, Vol. 62, pp. 11-34 Reprints available directly from the publisher Photocopying permitted by license only
THE RELATIONSHIP BETWEEN DISABILITY AND MEMORY DYSFUNCTION IN MULTIPLE SCLEROSIS HOWARD R. KESSLER Department of Psychology, Cushing Hospital, Framingham, MA 01701, U . S . A .
Downloaded by [Tufts University] at 18:17 08 December 2015
RONALD A. COHEN Department of Neurology, University of Massachusetts Medical Center, Worcester, MA, U . S . A .
KEVIN LAUER and DONALD F. KAUSCH Department of Psychology, Bowling Green State University, Bowling Green, O H , U . S . A . (Received April 19, 1991) We examined the relationship between memory impairment and functional disability in multiple sclerosis. Tests of memory, sensorimotor ability, and functional capacity were administered to fifty-six subjects with chronic-progressive or remitting-relapsing MS. Sensorimotor impairment, functional disability, and chronicity predicted impairment on various measures of memory acquisition, while age and type of diagnosis did not. After accounting for the effects of initial acquisition, delayed-recall performance was weakly-associated with disability. We suggest that: (1) Functional disability is associated with memory loss in MS; (2) MS-forgetting is caused by defective acquisition, rather by a deficit in consolidation or storage; (3) Level of disease activity, rather than type of MS diagnosis, determines the degree of memory impairment; and (4) MS disability needs to be evaluated multidimensionally, to account for both neurologic and functional impairment.
Keywords: multiple scierosis; memory: disability; learning; motor; information-processing.
Multiple sclerosis (MS) produces relatively circumscribed neuropsychological deficits, with impairment limited primarily to memory, executive functions, and the modulation of affect (Rao, 1986). In addition, MS produces considerable physical and functional disability. While MS patients possess a grossly intact immediate recall span (Minden, Moes, Orav, Kaplan and Reich, 1990; Heaton, Nelson, Thompson, Burks and Franklin, 1985; Grant, McDonald, Trimble, Smith and Reed, 1984; Rao, Hammeke, McQuillen, Khatri and Lloyd, 1984; Marsh, 1980; Staples and Lincoln, 1979), the disease exerts an effect on recent memory function, regarding both initial acquisition and later retrieval of information from long-term storage. Deficient performance in word-list learning (Jambor, 1969; Rao, et al., 1984; Minden, et al., 1990), narrative recall (Minden, et al., 1990), the Brown-Peterson task (Grant, et al., 1984), visuographic memory tasks (Minden, et al., 1990), and spatial memory tasks (Rao, et al., 1984) suggests the presence of deficient encoding, secondary in part to the efRequests for reprints to: H . R. Kessler, Ph.D., Department of Psychology, Cushing Hospital, P.O. Box 9008, Dudley Road, Framingham, MA 01701. 17
Downloaded by [Tufts University] at 18:17 08 December 2015
18
H.R. KESSLER et al.
fects of proactive interference. Defective retrieval operations appear to be manifested on a variety of delayed-recall and retrieval tasks, including the Wechsler Memory Scale stimuli (Grant, et al., 1984; Rao, et al., 1984; Staples and Lincoln, 1979) and on the Selective Reminding Test (Elpern, Gunderson, Kattah and Kirsch, 1984). MS patients may also be susceptible to retroactive interference and suffer from rapid forgetting (Rao, et al., 1984). Jambor (1969) initially postulated that the memory impairments in MS may be associated with plaques in the periventricular white matter. Rao, et al. (1984) later implicated more specifically the frontal-limbic system connections. Various recent studies have suggested that cognitive deficits in MS are closely related to structural, subcortical abnormalities, as revealed in CT or MRI scans and rCBF (Rao, Glatt, Hammeke, McQuillen, Khatri, Rhodes and Pollard, 1985; Brooks, Leenders, Head, Marshall, Legg and Jones, 1984; Franklin, Heaton, Nelson, Filley and Seibert, 1988; Reischies, Baum, Brau, Hedde and Schwindt, 1988; Rao, Leo, Haughton, St. Aubin-Faubert and Bernardin, 1989; Huber, Paulson, Shuttleworth, Chakeres, Clapp, Pakalnis, Weiss, and Rammohan, 1987; Franklin, Nelson, Filley and Heaton, 1989). Rao, et al. (1985) found initially that deficits in memory function were related to atrophy surrounding the area of the third ventricle, as observed on CT scan. Subsequent studies have replicated these findings. Memory dysfunction in MS has also been associated with observed attentional and information-processing deficits. Van den Burg, van Zomeren, Minderhoud, Prange and Meijer (1987) have suggested that slowed information processing speed, secondary to subcortical demyelination, may account for many of these cognitive deficiencies. Cohen and Fisher (1989) demonstrated that MS patients exhibit primary deficits in attention, information processing capacity and psychomotor speed, and related these deficits to fatigue. It has further been suggested that fatigue frequently interferes with functional status in MS (Caruso, LaRocca, Foley, Robbins and Smith, 1991). There is an apparent relationship among the presence of structural abnormality, memory dysfunction and fatigue in MS. Further, fatigue appears related to functional disability. Yet, there remains little agreement on the relationship between disability status and memory dysfunction in MS. The lack of any substantive relationship has been demonstrated utilizing both the Kurtzke (Minden, et al., 1990; Heaton, et al., 1985; Marsh, 1980) and Poser (Peyser, Edwards, Poser and Filskov, 1980) scales. This is likely because they are relatively brief measures geared toward ambulation capacity or specific neurologic deficits, rather than higher cortical functions. The limited usefulness of these scales is amply demonstrated in Baumhefner, Tourtellotte, Syndulko, Waluch, Ellison, Meyers, Cohen, Osborne, and Shapsak (1990) by the lack of correlation between scores on the Kurtzke Extended Disability Status Scale and MRI abnormalities in the brains of MS patients. Further, findings on the general neurological examination do not correlate with cognitive dysfunction in MS (Heaton, et al., 1985), and the prevalence of deficits on the neurological examination is far exceeded by the frequency of cognitive dysfunction in most studies (see Kahana, Leibowitz and Alter, 1971; Peyser, et al., 1980). This is not surprising, since false negatives on the general neurological examination are common in other degenerative disorders, for example Alzheimer’s Disease (Alexander and Geschwind, 1984). Indirect evidence suggests that disability may be related to cognitive dysfunction in MS. Heaton, et al. (1985) found that patients with chronic-progressive MS performed significantly worse than those with remitting-relapsing MS across a wide range of measures. As a result of this finding, one might infer a relationship between
Downloaded by [Tufts University] at 18:17 08 December 2015
DISABILITY IN MULTIPLE SCLEROSIS
19
disability status and cognitive performance, given the relatively more chronic and severe effects of chronic-progressive MS on the central nervous system. This finding has since been replicated by Beatty, Goodkin, Monson and Beatty (1989) and by Minden, et al. (1990). Hence, one might infer that the lack of correlation between disability status and cognitive dysfunction is due more to the inadequacy of the disability estimates, rather than by the lack of a substantive relationship. In response to the variability in the MS literature on cognitive dysfunction, and also to the difficulty in delineating the relationship between functional impairment and cognitive dysfunction, Peyser, Rao, LaRocca and Kaplan ( 1 990) have proposed guidelines for neuropsychological research. These criteria include a variety of parameters for characterizing patient samples, including demographics, diagnosis by category, and chronicity. In addition, they call for the use of measures of physical handicap, as well as data concerning psychiatric status and medication usage. This study has investigated the relationship between disability status, as reflected in functional deficits and motor performance, and memory and learning in a MS sample. Purportedly, deficits in activities of daily living (ADLs) should be more closely related to deficits in memory function, as global ADL capacity is a closer approximation of overall functional status (or physical handicap) than any of the aforementioned measures of physical disability. Many of the variables set forth by Peyser, et al. (1990) were examined as well.
METHOD Subjects Fifty-six MS patients from the Multiple Sclerosis Clinic of the Medical College of Ohio at Toledo (MCOT) and from local MS support groups were studied. All subjects expressed interest via telephone contact in participating in the study. All were under the care of a qualified physician on an outpatient basis. The sample contained 41 females and 15 males, ranging in age from 23.9 to 60.1 years (mean = 43.5, S.D. = 9.1) and in education from 8 to 20 years (mean = 12.9., S.D. = 2.3). Three subjects were Black, the remaining 53 Caucasian. Fortysix were classified as chronic-progressive and 10 as remitting-relapsing, while all were diagnosed as having clinically-probable or definite MS using the criteria of Poser, et al. (1983). The sample had a mean of 13.0 years (S.D. = 8.4) since initial onset of MS symptoms and a mean of 9.9 years since initial diagnosis. Seven subjects were taking adrenocorticoids at the time of testing, while 11 had received ACTH treatment within the prior year. Materials
Various standardized tests were used to assess memory, general cognitive ability, and physical and functional status. These are described below. WAIS-R Digit Span subtest This test was administered in the standard fashion. The lengths of the longest correctly-produced digit string for both forward and backward span were used for scoring, rather than using scaled-scores (Wechsler, 1981).
Modified Knox Cubes This test, developed by Kimura (1960), consists of 5 1-inch blocks arranged consecutively (approximately 1 inch apart) on a strip of wood. The subject is required to imitate block-tapping sequences (each 5 items long) for both immediate and 5-second delayed-recall conditions. There are 10 trials per condition.
20
H.R. KESSLER et al.
The task is scored for the total number of sequences correctly-reproduced under each of the two conditions.
Downloaded by [Tufts University] at 18:17 08 December 2015
WMS Logical Memory subtest We utilized the two stories contained in the original WMS (Wechsler and Stone, 1946), which contain a total of 46 separate “memories.” They were administered in the standard manner, which requires that the subject recall as much of each story as possible, after presentation by the examiner. The two stories are administered separately, and were scored using the standardized criteria developed by Schwartz and Ivnik (1980). WMS Associate Learning subtest This was also drawn from the original WMS, and comprises 6 “easy” and 4 “hard” word pairs, which are presented for recall across 3 trials. Incorrect responses are corrected during each trial, in order to facilitate additional learning. The score used was the total across the three trials for “easy” and “hard” items combined. WMS Visual Reproduction subtest This is a visuographic recall test of geometric figures. There are 3 stimulus cards, and the subject is required to reproduce each after a 10-second exposure. Scoring is subdivided into a total of 14 possible points. The standardized criteria developed by Schwartz and Ivnik (1980) were used for scoring purposes. All three WMS subtests were initially administered in the standard manner. This was followed by a delayed-recall of all stimuli after 30 minutes, and again after 2 weeks. Califarnia Verbal Learning Test (CVLT) This test, developed by Delis, Kramer, Kaplan, and Ober (1987), comprises two 1Qitem word lists, each including 4 items within each of 4 semantic categories. The first list (Monday) includes items of clothing, fruits, tools, and spices and herbs; the second list (Tuesday) includes fish, utensils, fruits and spices and herbs. The Tuesday items in the latter two categories differ from those on the Monday list. The Monday list is initially presented 5 times, with immediate recall following each presentation. The Tuesday list is then presented once for immediate recall, followed by both short-delayed free- and cued-recall trials of the Monday list. Free- and cued-recall are once again required after a 20-minute delay period, after which a recognition test is presented. Trials are scored for total number correct, and the test manual provides a wide variety of memory indices for analysis. In addition to the standardized administration of the CVLT, delayed freeand cued-recall and recognition testing were administered again after a one-week delay period. Recurring Figures Test Developed by Kimura (1963), this consists of 140 cards, each containing a line drawing of either a geometric or an irregular nonsense figure. Twenty target designs are shown for 3 seconds each during a study trial. This is followed by 5 recognition testing trials, each comprising 20 cards, including 8 of the original targets (4 geometric, 4 nonsense) interspersed within each trial. The subject is tested for accurate recognition of the target stimuli, using yes-no responding. There are 2 additional blocks of 20 cards each, which are used for recognition after a specified delay period (one week in the present study). Scoring is for total number of true positives minus number of false positive, responses.
DISABILITY IN MULTIPLE SCLEROSIS
21
Facial Recognition Test This was developed by Milner (1968), and includes a set of 12 target pictures presented in a 4 by 3 array for a 45-second study period. This is followed by a 90-second, verbally-filled delay. Subjects are then asked to identify the original target pictures from a 5 by 5 array of faces, including the 12 targets and 13 distractors. Scoring is the number of targets correctly identified.
Downloaded by [Tufts University] at 18:17 08 December 2015
WAIS-R Similarities and Picture Completion subtests These were administered and scored in the standardized manner. The age-scaled scores were used for analysis (Wechsler, 1981). Finger-Tapping Test Adapted from Halstead (1947), this is a lever and mechanical counter mounted on a wooden board. The subject is instructed to tap using his/her index finger, as fast as possible for a 10-second interval. This is done using both hands. Scoring is the mean of 5 10-second trials for each hand. Performance for only the dominant hand was used for analysis. Grooved Pegboard This is a metallic board, 4 inches square, with 5 rows of 5 holes, each with a randomly positioned slot. The pegs must be rotated properly in order to be inserted accurately. Pegs must be inserted as rapidly as possible, one row at a time from top to bottom. Each hand is tested separately, and is scored for the time required to complete the task. Only the performance for the dominant hand was used for analysis. Because some subjects were too impaired to complete the task (and hence could not yield a scorable result) scoring was instead based upon an extension of the Russell, Neuringer and Goldstein (1970) norms to a 7-point scale. Benton Visual Form Discrimination Test (BVFD) This test of visuospatial capacity was developed by Benton, Hamsher, Varney and Spreen (1983). It consists of 2 sample and 16 test items, each comprising a target stimulus and an array of 4 alternatives. Each target has 2 “Major” figures and l “Minor” peripheral figure. The 4 alternatives include the target stimulus, a stimulus with an error involving the “Minor” peripheral figure, one with the rotation of one of the “Major” figures, and one with the distortion of one of the “Major” figures. Each type of alternative is represented 4 times in each of 4 alternative spaces, numbered 1 to 4. The test can be administered for either memory or simple matching. The latter technique was used in this study. Correct responses receive 2 points, while peripheral errors receive 1 point. The score is the total number of points across 16 trials. Activities of Daily Living Scale This measure was adapted from a study by Staples and Lincoln (1979) on MS disability and memory function. By measuring areas far beyond mobility and cortical signs, this scale taps a broader range of functional impairment relative to measures used in many other studies (notably the Kurtzke and Poser scales, for example). Further, impairment on selected items from this scale correlated with aspects of memory impairment in the original study. The ADL scale consists of 42 items, representing the following functional domains: Mobility (10 items), Communication (4 items), Personal Care (5 items), Domestic Activities (18 items), Education (1 item), Employment (1 item), and Social Activities (3 items). Each item is scored on a 1 to 3 scale, with lower scores representing less impairment. Scores can be totaled by subtest or across all items. Because of their ambiguity, the Employment and Education items were excluded from analysis. The test was completed by each subject, and separately by a significant other.
22
H.R. KESSLER et al.
Procedure Testing was accomplished across 3 sessions, each separated by one week. This was done in order to minimize the effects of interference and fatigue, and to allow for recall testing across 1 and 2-week intervals. Subjects were asked the approximate time of day at which they felt most capable to perform, and each session was held at the same scheduled time. The testing procedure is outlined in Table 1.
RESULTS
Downloaded by [Tufts University] at 18:17 08 December 2015
Memory Performance Related to Disease and Demographic Variables
Intercorrelations among demographic and disease-related-variables, and performance on major memory test indices, are reported in Tables 2 and 3. The subjects’ own responses were used for analyzing the disability score, since there was a very high Spearman rank-order correlation coefficient between the two separate reports ( r = .95, p < .001). In addition, because chronicity (time since initial symptom onset) correlated highly with years since diagnosis (Pearson Product-Moment Correlation coefficient = .90, p < .Owl),only chronicity was used for analysis. Age (Pearson r ) and diagnosis (point-biserial correlations) are not included, as each failed to correlate with a single memory index. Sex (point-biserial correlations) failed to correlate significantly with any visual memory tests, but did correlate with selected verbal measures, related primarily to initial acquisition of stimuli on list-learning tasks (CVLT, Associate Learning). Education correlated with multiple memory indices, including measures of both verbal and nonverbal recent recall (Logical Memory, Facial Recognition), as well as with a measure of verbal immediate memory span (Digits Backward). Chronicity correlated with a limited number of measures, but was related to multiple memory processes: It correlated with both initial acquisition and delayed recall on both verbal and visual measures (CVLT, Recurring Figures), as well as with visual immediate memory (Knox Cubes). The more direct measures of disability status correlated with memory indices with far greater frequency: Disability (ADL Scale) correlated with indices across all modalities and stages of acquisition and recall, with the exception of verbal immediate memory span (Digit Span); finger-tapping correlated with selected indices across all memory tests, although the pattern was somewhat variable. Specifically, it correlated better with the easier immediate memory span tests (Digits Forward, Knox CubesImmediate) than with the more difficult span measures (Digits Backward, Knox Cubes-Delayed); it also correlated more widely with verbal than visual measures, notably where it did not correlate with the long-delayed visual recall measures (Visual Reproduction 2-week recall, Recurring Figures 1-week recall). Finally, performance on the Grooved Pegboard correlated significantly with every memory index examined. Memory and Degree of Disability
In order to delineate the relative contribution of each disease and performance variable to prediction of memory performance better, the memory acquisition measures were regressed on the following predictor variables: Type of MS diagnosis; chronicity; disability score; finger tapping; and Grooved Pegboard. (The delayed recall measures were not used in these analyses, as they correlated very highly with the acquisition measures; analyses to be cited later will serve to elucidate this point.) An examination of Table 4 reveals that the following measures received statistically-
DISABILITY IN MULTIPLE SCLEROSIS
23
TABLE 1 Testing Procedure
Downloaded by [Tufts University] at 18:17 08 December 2015
~~
~
~
Session I A. Informed Consent B. WMS subtests 1 . Logical Memory 2. Visual Reproduction 3. Associate Learning C. 30 minutes of interpolated activity I . Picture Completion 2. Similarities 3. Finger Tapping 4. Grooved Pegboard D. Delayed recall 1 . Logical Memory 2. Visual Reproduction 3. Associate Learning E. ADL Scale given to subject, to be returned at following session. Session I1 A. CVLT 1 . 5 “Monday” list trials 2. “Tuesday” list trial 3. Short-delay, free recall, “Monday” list 4. Short-delay, cued recall, “Monday” list B . 20 minutes of interpolated activity 1. Recurring Figures Test 2. Other visual measures (not reported) C. CVLT Long-delay testing 1. Free recall of “Monday” list 2. Cued recall of “Monday” list 3. Recognition testing D. Modified Knox Cubes Test E. ADL Scale collected from subject Session 111 A. Facial Recognition Test B. Two-week delay testing of WMS subtests 1 . Logical Memory 2. Visual Reproduction 3. Associate Learning C. One-week delay testing of Recurring Figures Test D. One-week delay testing of CVLT 1. Free recall of ‘Monday” list 2. Cued recall of “Monday” list 3. Recognition testing E. Visual Form Discrimination Test
significant predictions: Knox Cubes, both the Immediate and Delay conditions; CVLT Monday Total recall; Logical Memory; Associate Learning; Visual Reproduction; and the Recurring Figures. Performance on the Grooved Pegboard contributed to prediction of performance on all of these tests, the exception being the Recurring Figures. Chronicity (Knox Cubes-Immediate, Recumng Figures) and Disability (Knox Cubes-Immediate, CVLT, Visual Reproduction, Recurring Figures) were less powerful and/or influential predictors. Finger-tapping and type of diagnosis did not contribute significantly (tapping contributed to Facial Recognition prediction, but it was not statistically-significant) to any prediction. It is not surprising that diagnosis did not serve as a useful predictor, as it did not correlate with any of the measures. A
Chronicity
-.27 -.25 -.27 -.23 -.26 .05 N.S. N.S. N.S. .05 -.53 -.51 -.52 -.34 -.50 .02 ,0001 ,0001 .0001 ,0001 .38 .47 .46 Tapping .29 .48 ,001 ,001 .01 .05 ,001 .57 .49 .49 Pegboard .33 .45 .02 .001 .OOOl .OOOl .OW1 CVLT = California Verbal Learning Test MONl = Monday, Trial 1 MON5 = Monday, Trial 5 MONT = Monday, Total recall SDF = Short-delayed free recall LDF = Long-delayed free recall WEEK = One-week delayed free recall
-.15 N.S. -.33 .02 .31 .05 .37 .01
VR = Visual Reproduction IMM = Immediate recall DEL = 30-minute delayed-recall 2WK = Two-week delayed recall
AL = Associate Learning IMM = Immediate recall DEL = 30-minute delayed-recall 2WK = Two week delayed-recall
-.21 -.17 -.14 -.24 -.11 -.29 N.S. N.S. N.S. N.S. N.S. .05 -.19 -.27 -.16 -.38 -.48 p.47 N.S. .05 N.S. .01 ,001 .001 .20 .25 .29 .32 .29 .33 .05 .02 N.S. N.S. .05 .02 .31 .37 .30 .34 .44 .51 .05 .01 .05 .02 .001 .OOOI LM = Logical Memory IMM-Immediate recall DEL = 30-minute delayed-recall 2WK = Two week delayed-recall -.I7 -.12 N.S. N.S. -.20 -.15 N.S. N.S. .21 .29 .05 N.S. .34 .32 .02 .01
-.38 -.22 -.39 .01 N.S. .01 -.34 -.I3 -.28 -.43 .01 N.S. .05 ,001 .29 .24 .32 .42 .05 N.S. .02 .01 .36 .30 .28 .42 .01 .05 .05 .37 = Digits Forward = Digits Backwards
MKC = Modified Knox Cubes IMM = Immediate Condition DEL = Delayed condition
MFR = Milner Facial Recognition
KIM = Kimura Recurring Figures IMM = Immediate condition DEL = Delayed condition
.01
-.42
.01
-.30 .05 -.16 N.S. .25 N.S. .37
VR VR KIM KIM MKC MKC DEL 2WK IMM DEL MFR IMM DEL
-.21 -.26 -.24 N.S. N.S. N.S. p . 4 4 -.43 -.34 ,001 ,001 .02 .37 .33 .26 .01 .02 N.S. .45 .40 .36 ,001 .O I .01 DIGFOR DIGBACK
CVLT CVLT CVLT CVLT CVLT CVLT LM LM LM AL AL AL DIG DIG VR MONl MON5 MONT SDF LDF WEEK 1MM DEL 2WK IMM DEL 2WK FOR BACK IMM
Intercorrelations Between Memory Indices and Predictor Variables
TABLE 2
Downloaded by [Tufts University] at 18:17 08 December 2015
P
N
DISABILITY IN MULTIPLE SCLEROSIS
25
TABLE 3 Intercorrelations Among Disease-Related Variables Chronicity Disability Pegboard
Downloaded by [Tufts University] at 18:17 08 December 2015
Tapping
Disability
Pegboard
.07
n.s. -.39 p < .01 - .28 p < .os
- .60
p < ,0001
-
- .6S
p < .0001
.63 p
