Quantitative and Qualitative Analyses of Clock Drawings in Alzheimer's and Huntington's Disease
Quantitative and Qualitative Analyses of Clock Drawings in Alzheimer's and Huntington's Disease
ISABELLE ROULEAU
DAVID P. SALMON
AND
This study was supported by funds from the Medical Research Service of the Department
of Veterans Affairs, by NIA Grants AG-05131 and AG-08204 to the University of California
at San Diego, and by FRSQ (Fonds de la recherche en Sante du Quebec) Postdoctoral
Fellowship to Dr. Rouleau. All correspondence and reprint requests should be addressed
to Dr. Nelson Butters, Psychology Service (116B), Dept. of Veterans Affairs Medical Center,
3350 La Jolla Village Drive, La Jolla, CA 92161.
70
0278-2626192 $3.00
Copyright 0 1992 by Academic Press, Inc.
All rights of reproduction in any form reserved.
VISUOCONSTRUCTIVE IMPAIRMENT 71
marked improvement in performance. These results indicate that while both DAT
and HD patients have significant visuoconstructive difficulties even in the early
stages of their disorders, the specific cognitive processes underlying their quan-
titative impairments are quite different. It is possible that the DAT patients’
conceptual errors are yet another indicator of the deterioration of their semantic
knowledge. (2‘ 1992 Academic Preu. Inc
INTRODUCTION
Studies comparing the neuropsychological profiles of Alzheimer’s
(DAT) and Huntington’s disease (HD) patients have demonstrated that
various forms of dementia may differ in terms of the quantitative and
qualitative features of their language, attention, and memory impairments
(for review, see Butters, Salmon, & Heindel, 1990b). Alzheimer’s pa-
tients, even in the early stages of their disorder, make frequent seman-
tically based naming errors and are more impaired on category than on
letter fluency tasks (Bayles & Tomoeda, 1983; Butters, Granholm,
Salmon, Grant, & Wolfe, 1987; Hodges, Salmon & Butters, 1991; Huff,
Corkin & Growdon, 1986). This pattern of deficits has been viewed as
an indicator of a deterioration in the structure of these patients’ semantic
knowledge (Chertkow & Bub, 1990; Martin, 1987; Martin & Fedio, 1983).
In contrast, patients with HD make relatively few naming errors, most
of which represent perceptual (rather than semantic) confusions, and are
equally impaired on both category and letter fluency tasks (Butters,
Salmon, Heindel, & Granholm, 1988b; Hodges et al., 1991). This profile
of impairments has been interpreted as representing HD patients’ defi-
ciencies in initiating retrieval strategies and in processing visuoperceptual
materials (Butters et al., 1990b; Hodges et al., 1991). Studies that have
compared DAT and HD patients’ attentional abilities have noted greater
deficits in HD than in DAT patients (Brandt, Folstein, & Folstein, 1988;
Folstein, Brandt, & Folstein, 1990; Starkstein et al., 1988; Troster, Jacobs,
Butters, Cullum, & Salmon, 1989).
Differences between DAT and HD patients are apparent on both ex-
plicit and implicit memory tasks. Patients with DAT seem unable to store
new information (Moss & Albert, 1988), evidence rapid rates of forgetting
(Butters et al., 1988a; Hart, Kwentus, Taylor, & Harkins, 1987; Moss,
Albert, Butters, & Payne, 1986; Welsh, Butters, Hughes, Mohs, & Hey-
man, 1991), are very prone to intrusion errors (Butters et al., 1987; Jacobs,
Salmon, Troster, & Butters, 1990), and have severe retrograde amnesias
even in the early stages of the disease (Beatty, Salmon, Butters, Heindel,
& Granholm, 1988; Sagar, Cohen, Sullivan, Corkin & Growdon, 1988).
On implicit memory tasks DAT patients have been reported to be im-
paired on lexical, semantic, and perceptual priming tasks but to perform
normally in the acquisition of motor skills (for review, see Butters, Heindel
& Salmon, 1990a). On the other hand, HD patients’ explicit memory
72 ROULEAU ET AL.
TABLE 1
AGE, EDUCATION,AND DEMENTIA RATING SCALE SCORE(DRS) (MEAN -+ SD) for the
DAT, HD, AND NC SUBJECTS
DAT HD NC
gressive dementia (Henderson et al., 1989; Kirk & Kertez, 1991; Sun-
derland et al., 1989). More specifically, DAT and HD patients, as well
as age-matched intact control subjects, are asked to draw a clock to
command and to set the hands to a specified time. To compare drawing-
to-command with copying ability, the patients and control subjects are
also asked to copy the drawing of a clock with the hands already set to
a specified time. A modified version of Sunderland et al.3 (1989) scoring
criteria and error categories are used in the analyses of all drawings.
METHOD
Subjects
A total of 75 subjects participated in this study. They were equally divided into three
groups: 25 patients with DAT, 25 HD patients, and 25 elderly neurologically intact normal
control subjects. The DAT and HD patients were matched for overall level of dementia
with the Dementia Rating Scale (DRS) (Mattis, 1976). All subjects were participants in the
research activities of the Alzheimer’s Disease Research Center (ADRC) at the University
of California at San Diego. As part of this program, all patients received a yearly com-
prehensive neurological and neuropsychological assessment.
The diagnosis of probable DAT was made by two senior staff neurologists using the
criteria developed by the National Institute of Neurological and Communication Disorders
and Strokes (NINCDS) and the Alzheimer’s Disease and Related Disorders Association
(ADRDA) (McKhann et al., 1984). Laboratory tests (e.g., CT scan, urinalysis, SMA-12,
T4, folate levels) were performed to rule out various viral, metabolic, or traumatic causes
of dementia. To reduce the possibility of including multi-infarct dementias, patients with a
score of 5 or greater on the Hachinski ischemia scale (Hachinski et al., 1975) were excluded
from the DAT group.
The diagnosis of HD was made by a staff neurologist on the basis of positive family
history for the disease, the presence of involuntary choreiform movements and dementia.
The functional capacities were assessed with Shoulson and Fahn’s disability scale (1979).
Nine of the patients were judged to be at stage II, 12 at stage III, and 4 at stage IV. In
addition to being matched on the DRS, and DAT and HD patients were equated for years
of education.
The normal elderly control subjects (NC) were selected to match DAT patients for age
and for education. Any subject with a history of alcoholism, drug abuse, or serious neu-
rological or psychiatric condition was excluded from the study.
The demographic data for the three groups are shown in Table 1. A one-way analysis of
variance (ANOVA) of the mean age showed a highly significant group difference [F(2,
72) = 44.07, p < .OOl]; post hoc analysis revealed that HD patients were significantly
younger than both DAT patients [t(48) = 7.33, p < .OOl] and NC subjects [t(48) = 7.30,
74 ROULEAU ET AL.
p < .OOl]. There was no significant difference in age between DAT and NC subjects [t(48)
< 11. A one-way ANOVA showed no significant difference in the mean education levels
of the three groups [F(2, 70) < 11. There was no significant difference between DAT and
HD patients in the severity of dementia, as assessed by the DRS [t(47) < 11, and both
patient groups were significantly impaired compared to NC subjects (p < .OOl).
Rating Scale
The clocks were first scored using Sunderland et al.‘s lo-point rating scale from worst
(score = 1) to best (score = 10) representation of a clock (Sunderland et al., 1989). However,
this scale proved difficult to apply according to the criteria provided since it assumes that
the representation of the hands is first and solely affected (score 6 to lo), and additional
errors in the representation of numbers and the clockface occur later (score 1 to 5). Con-
sequently, some drawings received very low scores on this scale because of minor errors in
the representation of numbers (added or missing numbers) even though the hands were
adequately set. To circumvent this scoring problem, the clocks were reevaluated with a new
lo-point scale (see Table 2) designed to independently assess the accuracy of the represen-
tation of the clockface (maximum, 2 points), the layout of numbers (maximum, 4 points),
and the position of the hands (maximum, 4 points).
TABLE 2
REVISED SCALE USED FOR SCORING THE CLOCK DRAWINGS
(C) Deficit in spatial layout of numbers, without any specific pattern in spatial disorga-
nization (see Fig. 3d).
(D) Numbers written outside the clockface (see Figs. 3a and 3e).
(E) Numbers written counterclockwise (see Fig. 3f).
6. Perseverufion. In general, perseveration is defined as the continuation or the recurrence
of activity without an appropriate stimulus. In clock drawing, different types of perseverative
responses could be observed:
(A) Perseveration of hands: presence of more than two hands reflecting a failure to
terminate, when appropriate, the ongoing set of tracing the hands (see Figs. 4a and 4b).
(B) Perseveration of numbers: abnormal prolongation of numbers, such as writing beyond
“12” (See Fig. 4c) or inappropriate recurrence of the same numbers (see Fig. 4d).
The same three judges reviewed each clock independently, and an error was considered
present only if there was agreement between two of the judges.
RESULTS
The scores obtained for the clock drawings using both Sunderland et
al.% and the revised scale are presented in Table 3. The overall correlation
between the total scores (three judges) obtained with Sunderland et al.%
scale and with the revised scale was .89. The interrater reliability was
high in all conditions (i.e., all subjects pooled vs. separated by diagnostic
groups) varying from .86 to .97 with Sunderland et al.‘s scale and from
VISUOCONSTRUCTIVE IMPAIRMENT 77
.92 to .97 using the revised scale. The scores obtained with the revised
scale were retained for further analysis.
A two-way ANOVA (Group [DAT, HD, NC] x Condition [Command,
Copy]) with repeated measures across conditions yielded significant main
effects of Group [F(2, 69) = 9.52, p < .OOl] and Condition [F(l, 69) =
10.44, p < .005] and a significant Group x Condition interaction [F(2,
69) = 5.85, p < .005]. The results are presented in Fig. 5.
Post hoc comparisons revealed a significant difference between NC
subjects and DAT patients for both command [t(48) = 4.54, p < .OOl]
and copy [t(46) = 2.43, p < .05] conditions. The difference between NC
and HD subjects was also significant in both conditions [command: t(48) =
4.54, p < .OOl; copy : t(46) = 4.37, p < .OOl]. Although there was no
significant difference between DAT and HD patients in the command
[t(48) = < l] and in the copy conditions [t(46) < 11, further comparisons
(paired sample t test) between command and copy performances revealed
a significant effect for DAT patients [t(23) = 3.27, p < .005] but not for
78 ROULEAU ET AL.
B
0
doq&
a) pt 2176: DAT (command) b) pt 2075 HD (command) c)pt 612: HD (command)
HD patients [t(23) < l] or for NC subjects [t(23) < 11. That is, the DAT
patients performed significantly better under copy conditions than under
command conditions, whereas the NC and HD subjects showed no dif-
ferences between these drawing conditions.
were more likely than DAT and NC subjects to have moderate (see Fig.
la) and severe (see Fig. lb) graphic difficulties (see Table 4). A x2 test
of independence between Group (HD-DAT-NC) and Graphic difficulties
(present-absent) was highly significant for both command [x2(2 df> =
27.49, p < .OOl] and copy [x2(2 @) = 16.32, p < .OOl] conditions.
An attempt was made to determine whether the HD patients’ graphic
difficulties were related to their overall disability and motor dysfunction.
When HD patients with no or mild graphic difficulties were compared
with those patients who evidenced moderate or severe deficits, no sig-
nificant differences were noted on the Shoulson and Fahn (1979) disability
TABLE 3
SCORNS OBTAINED (MEAN -t SD) FOR THE DAT. HD, AND NC SUBJECTS FOR CLOCK
DRAWINGS USING SUNDERLAND ET AI..‘S AND REVISED SCALES
DAT HD NC
‘0 r
u g- 0 0
‘5
i:: +
6 -
5
M
O-0
NC
DA1
0-O HO
Command COPY
Condition
FIG. 5. Scores obtained in the command and copy conditions as a function of diagnostic
groups (NC, DAT, and HD).
scale [t(23) = 1.65, n.s.] and on measures of chorea [t(9) < l] and motor
impairment [t(9) < l] derived from the Chorea and Motor Impairment
scale (Folstein, Jensen, Leigh, & Folstein, 1983).
FIG. 6. Distribution of error types in HD and DAT patient groups in the command
condition.
VISUOCONSTRUCTIVE IMPAIRMENT 81
TABLE 4
NUMBER OF DAT, HD, AND NC SUBJECTS PRESENTING WITH GRAPHIC DIFFICULTIES IN THE
COMMAND AND COPY CONDITIONS
Command COPY
DAT HD NC DAT HD NC
Mild 6 8 3 6 5 7
Moderate 0 I 0 3 11 0
Severe 0 5 0 0 4 0
TABLE 5
DEMENTIA RATING SCALE SCORES (MEAN, SD) IN DAT PATIENTS AS A FUNCTION OF THE
PRESENCE OR ABSENCE OF CONCEPTUAL ERRORS
Presence of Absence of
conceptual errors conceptual errors
N = 13 N = 12
Note. Significant statistical differences between groups (**p < .Ol, *p < .05).
DISCUSSION
Although both DAT and HD patients are impaired in their drawings
of clocks, it is quite evident that the processes underlying their low quan-
titative scores are quite distinct. The DAT patients’ tendency to commit
conceptual errors contributed greatly to their especially poor performance
on the drawing-to-command condition. When they lacked a model to
guide their drawings, the DAT patients appeared unable to retrieve the
specific attributes that define a clock and consequently were very impaired
in their attempts to draw an accurate representation. Their marked im-
provement under the copy condition further emphasizes that their drawing
deficiencies are not due primarily to graphic, motor, or even visuoper-
VISUOCONSTRUCTIVE IMPAIRMENT 83
ceptual difficulties; rather, their deficit is likely due to a loss of the semantic
associations usually evoked by the word “clock.” The finding of an as-
sociation between the production of conceptual errors and lower scores
on the Initiation/Perseveration and Conceptualization sections of the DRS
also suggests that semantic impairments play a major role in DAT patients’
poor clock drawings. Thus, DAT patients’ well-documented impairments
on language tasks involving confrontation naming and fluency (Bayles &
Tomoeda, 1983; Butters et al., 1987; Chertkow & Bub, 1990; Hodges et
al., 1991; Huff et al., 1986; Martin & Fedio, 1983), on memory tests
requiring semantic encoding (Corkin, 1982; Granholm & Butters, 1988;
Salmon, Shimamura, Butters & Smith, 1988), and on construction tasks
involving the graphic representation of common objects may all reflect a
deterioration of the hierarchical structure of these patients’ semantic
knowledge. At the very least, the significant tendency of DAT patients
to manifest conceptual errors on a drawing-to-command task suggests that
such semantic deficits are ubiquitous even in the early and middle stages
of this disorder and may be evident on many neuropsychological tests.
Unlike the DAT patients, HD patients’ performance was characterized
by the presence of moderate and severe graphic deficits, a tendency to
draw small clocks, and planning deficits in the spatial layout of numbers.
The graphic deficits and small drawings are consistent with their marked
difficulties on tests of motor functions such as the Motor Steadiness Test
(Josiassen, Curry, and Mancall, 1983) and their low scores on most of
the WAIS-R Performance scale subtests (Butters et al., 1978; Fedio et
al., 1979; Folstein et al., 1990; Josiassen, Curry, Roemer, Debease, &
Mancall, 1982). Their planning deficits may be related to the frontostriatal
dysfunction commonly associated with HD (Brandt & Butters, 1986) and
are apparent on tasks such as the Wisconsin Card Sorting Test (Josiassen
et al., 1983) and the Picture Arrangement subtest of the WAIS-R (Butters
et al., 1978). As suggested by Kaplan and Delis (in press) a planning
deficit may also result from an inability to execute simultaneously two
sequential tasks (i.e., producing and writing the numbers in the right
order and spacing them equally around the clock), and may explain why
the HD patients often self-corrected their production.
Although visuospatial errors were noted in both patient groups, the
type of error differed: DAT patients tended to draw very large clocks,
to write the numbers outside the outline of the clock, and to commit
errors in the spatial layout of the numbers, whereas HD patients drew
small clocks and made a variety of visuospatial errors that could be at-
tributed to their inability to sequence or plan their motor responses. Some
evidence of spatial neglect was noted, albeit infrequently, in both patient
groups. However, the tendency to write all the numbers on the right side
of the clock could be an indication of a planning deficit rather than a true
neglect of the left hemispace.
84 ROULEAU ET AL.
REFERENCES
Ajuriaguerra, J., Muller, M., & Tissot, R. 1960. A propos de quelques probltmes poses
par I’apraxie dans les dtmences. Enckphale, 49, 275-401.
Bayles, K. A., & Tomoeda, C. K. 1983. Confrontation naming and dementia. Brain and
Language, 19, 98-114.
Beatty, W. W., Salmon, D. P., Butters, N., Heindel, W. C., & Granholm, E. L. 1988.
Retrograde amnesia in patient with Alzheimer’s or Huntington’s disease. Neurobiology
of Aging, 9, 181-186.
Brandt, J., & Butters, N. 1986. The neuropsychology of Huntington’s disease. Trends in
Neurosciences, 9, 118-120.
Brandt, J., Folstein, S. E., & Folstein, M. F. 1988. Differential cognitive impairment in
Alzheimer’s disease and Huntington’s disease. Annals of Neurology, 23, 555-561.
Brantjes, M., & Bouma, A. 1991. Qualitative analysis of the drawings of Alzheimer patients.
The Clinical Neuropsychologist, 5, 41-52.
Brouwers, P., Cox, C., Martin, A., Chase, T., & Fedio, P. 1984. Differential perceptual-
VISUOCONSTRUCTIVE IMPAIRMENT 85