вход по аккаунту


Cerebral mapping of apraxia in Alzheimer's disease by positron emission tomography.

код для вставкиСкачать
Cerebral Mapping of Apraxia
in Alzheimer's Disease by Positron
Emission Tomography
Norman L. Foster, MD, Thohas N . Chase, MD, Nicholas J. Patronas, MD, Marjorie M. Gillespie, RN,
and Paul Fedio, PhD
The ability to mimic skilled movements or to pantomime them in response to spoken command was compared with
psychometric performance and with regional glucose utilization as estimated by {"F]fluorodeoxyglucose positron
emission tomography in 17 right-handed patients with Alzheimer's disease and 6 age-matched normal subjects.
Apraxia scores, both on tests to command and to imitation, were significantly lower in the Alzheimer patients.
Imitation scores correlated best with performance on tests of visual-spatial ability and with cortical metabolism in the
right parietal lobe; command scores related most closely with the results of tests reflecting verbal proficiency and with
cortical metabolism in the left inferior hemisphere, especially frontally. Apraxia to command and imitation may thus
reflect neuronal dysfunction in distinct cerebral regions in patients with Alzheimer's disease.
Foster NL, Chase TN, Patronas NJ, Gillespie MM, Fedio P: Cerebral mapping of apraxia in Alzheimer's
disease by positron emission tomography. Ann Neurol 19:139-143, 1986
Apraxia, the inability to perform purposive or skilled
movements in the absence of paralysis, occurs frequently in patients with Alzheimer's disease and often
contributes substantially to their disability C1, 77. Despite extensive study, cerebral mechanisms responsible
for apraxia remain incompletely understood. In Alzheimer's disease, analysis of this deficit is particularly
difficult because the neuronal injury is both progressive and diffuse. Positron emission tomography (PET)
following C'8F]fluorodeoxyglucose (FDG) adrninistration permits cerebral neuronal activity to be directly
related with psychomotor function 128, 307. The cortical distribution of areas in which glucose metabolism
has been found to correlate most closely with performance on tests of language and visual-spatial ability is
in general conformance with classic localizing notions
C5, 6, 12). Accordingly, in a search for regions having
the highest Correlations between glucose utilization
and the ability to carry out various motor tasks, the
cortical sites most closely related to the symptoms of
apraxia might be indicated. In addition, examination of
other cognitive functions that occur in association with
apraxia may shed light on the localization and mechanism of this deficit in Alzheimer's disease.
From the Intramural Research Program, National Institute of
Neurological and Communicative Disorders and Stroke, Bethesda,
MD 20205.
Received Dec 7 , 1984, and in revised form May 1, 1985. Accepted
for publication July 12, 1985.
Seventeen right-handed patients (12 males, 5 females; ages
48 to 72 years, mean 61) with clinically diagnosed Alzheimer's disease and 6 age-matched normal subjects (3 males, 3
females; ages 52 to 66 years, mean 58) were studied. Many
of these persons have been the subject of previous reports
[6, 12, 131. Each Alzheimer patient had a history of gradually progressive intellectual impairment without focal motor
or primary sensory deficits or other known causes of dementia. None had clinical o r laboratory evidence of cerebrovascular disease (Hachinski's ischemic scores {29] of 4 or less)
or other noteworthy illness and they were all mildly to somewhat severely impaired. Electroencephalographic recordings
were generally free of localized abnormalities and computed
tomographic (CT) scans revealed generalized cerebral atrophy. All patients and control subjects were alert, cooperative, and medication-free at the time of testing. Each
received a complete neurological and general medical evaluation as well as an extensive psychometric battery of tests. O n
the basis of findings from their initial clinical examination, 4
patients were considered to have disproportionately severe
visual-spatial deficits and 5 to have predominant verbal dysfunction.
Apraxia was assessed by observing the response to both
spoken command and visual demonstration. First, the subject was verbally requested to perform various motor acts
Address reprint requests to Dr Chase, Experimental Therapeutics
Branch, National Institute of Neurological and Communicative Disorders and Stroke, Bldg 10, Rm 5C103, Bethesda, MD 20205.
without benefit of any physical object. Then, in a different
order, the examiner pantomimed each task and the subject
was instructed to imitate what the examiner had done. Finally, whenever appropriate, objects were made available to
perform the task and the response to verbal command again
assessed. Commands and instructions were repeated if necessary. A total of 75 tasks were evaluated 141. Responses were
scored on the following basis: 1 (totally incorrect), 2 (partially correct), or 3 (fully correct). Substitution of a body part
as the object was noted and rated as partially correct. Nine of
the tasks were only imitated since they were nonrepresentational or required a complex sequence that could not be
elicited by giving a single specific verbal command. The maximum possible score to verbal command was 198 while that
to imitation was 225. Response with either hand was accepted, although previous experience suggested that there
was little difference in performance between the two hands
or two feet. Some complex tasks required coordinated
movement of both right and left extremities for an appropriate response. Responses were not timed and testing was divided into several periods so that attention could be maintained. The average time for test completion was about two
PET scanning (ECAT I1 tomograph; ORTEC, Oak Ridge,
TN) at 10-mm increments parallel to the canthomeatal (eyeto-ear) plane was performed in consenting individuals 30
minutes after the rapid injection of approximately 5 mCi of
FDG through an intravenous catheter in one arm. Following
isotope administration, serial “arterialized” venous blood
samples were obtained from the opposite arm for determination of FDG and glucose concentrations 127). Throughout
the procedure, all subjects were kept at rest in a quiet environment, with eyes patched and ears plugged to minimize
the effects of external stimuli. Calculation of local metabolic
rates for glucose was based on rate constants determined
from normal individuals using a modification of the Sokoloff
operational equation [3, 191. Image reconstruction was performed on the ECAT computer employing the standard ellipse method of attenuation correction and the mediumresolution convolution filter. The resulting spatial resolution
in the imaging plane was 1.7 cm full-width at half maximum
and 2.0 cm in the axial plane.
By means of the ECAT “region of interest” program, peak
metabolic rates were obtained for 63 contiguous nonoverlapping regions in both the right and left cerebral cortex [13).
On each horizontal scan, a rectangular box, approximately 3
mm along the X (transverse) axis and 14 mm along the Y
(anteroposterior) axis, was centered on a line equidistant
from the anterior and posterior poles of the brain. As the
rectangle was moved laterally along this line across the cerebral cortex, the peak mean ECAT number for all pixels enclosed by the rectangle was used to compute the metabolic
rate for that cortical area. After readings from both right and
left hemisphere cortex were obtained, the rectangle was advanced in 14-mm increments, anteriorly and posteriorly, to
yield metabolic values for the nine adjoining cortical regions
in both the right and left hemispheres at each of seven scan
levels. Metabolic values were excluded from analysis when
the rectangle appeared to fall beyond the cortical limits; cortex at the extreme frontal and occipital poles was thus generally not analyzed. Concern about partial volume effects also
140 Annals of Neurology Vol 19 No 2 February 1986
Fig 1. Cortical distribution of statistically sign$cant correlations between local rates of glucose utilization and apraxia t o
imitation test scores. (Dots indicate individual uncorrected
sign$cance levels ranging from p < 0.0001 to 0.0007; median,
p < 0.0002.)
Fig 2. Cortical distribution of statistically signif cant correlations between local rates of glucose utilization and apraxia t o
verbal command test scores. (Dots indicate individual uncorrected
significance levels ranging from p < 0.0001 to < 0.0004; median, p < 0.0002.)
precluded evaluation of the inferior temporal surfaces and
the medial (interhemispheric) cortex. The position in the
axial plane of each horizontal slice was determined by use of
characteristic landmarks E14, 251. The location of these regions on the lateral surface was confirmed by comparison
with C T scans and standard atlases.
Spearman rank order correlation coefficients were used to
compare apraxia scores with rates of local glucose utilization
as well as with performance on standard psychometric tests.
An informal cluster analysis technique was then applied to
these results in order to map the cortical distribution of the
closest correlations between regional rates of glucose utilization and apraxia test scores (Figs 1, 2). For this purpose,
probability values describing all such correlations were
ranked in descending order and plotted seriatim on cortical
manikins until contiguity with the index (highest p value)
region was lost, or all regions having correlations at p <
0.0007 had been plotted. All such regions, if considered
individually, are significant (p < 0.05) following application
of Bonferroni’s correction. The statistical significance of a
cluster of ad joining regions, on the other hand, substantially
exceeds this level.
Normal subjects performed all tasks nearly flawlessly:
their scores ranged from 223 to 225 on imitation testing and only one incompletely correct response was
recorded on testing to command (Table 1). No use of
body part as object or completely incorrect responses
were observed. In contrast, Alzheimer patient scores
Table I. Apraxia Scores for Alzheimer
Patients and Normal Controls"
Table 2. Correlations between Apraxia
Scores and Psychometric Functiona
Test Group
198 ? 0.2
224 k 0.6
164 5 6.3b
197 2 6.Zb
General tests
WAIS full scale IQ
Wechsler memory quotient
Verbal tests
WAIS verbal IQ
Boston naming test
Mattis verbal fluency
Visual-spatial tests:
WAIS performance IQ
Rey-Osterreith copy test
Benton copy (form C)
"Mean f SEM scores for 17 patients and 6 controls, out of a possible 198 for command and 225 to imitation.
"p < 0.0001 for difference from normals by Student's t test.
ranged from 91 to 194 to command and 118 to 223 to
imitation. When these scores are combined, Alzheimer patients could clearly be separated from normals.
Even the best patient's performance was marred by 10
errors. Combined scores for patients ranged from 27 1
to 413 (mean, 351 ? 9 SEM), while in normals these
scores ranged from 421 to 423 (mean, 422 -t 0.4).
The Alzheimer patients varied considerably with respect to their performance on apraxia tests to imitation
in comparison with their performance on tests to command: the 4 patients with predominant visual-spatial
involvement averaged 163
15 to imitation and 172
k 8 to command, while the 5 patients with mainly
language difficulty averaged 2 16 +- 3 to imitation and
139 ? 14 to command. (The p value for the difference
between the two patient groups for imitation scores
was <0.01; the p value was not significant for command scores.)
Performance to command generally improved
somewhat when relevant objects were provided; average patient scores increased from 52 f 3 to 60
1 (p
< 0.005) in the 22 tasks in which this type of testing
was appropriate. Only 1 patient's performance worsened with use of the object; although his pantomime
to spoken command appeared normal, when given an
object, he repeatedly held it incorrectly. This patient
also imitated very poorly.
Apraxia scores did not appear to correlate with patient age ( r = 0.03 and -0.25 for apraxia scores to
command and imitation, respectively) or degree of
overall dementia (Table 2). On the other hand, performance to spoken command correlated closely with
scores on tests highly dependent on verbal skill, while
the ability to imitate correlated closely with performance on tests of visual-spatial skill (see Table 2).
The cortical distribution of the most highly significant and positive correlations between regional metabolic rates and apraxia performance scores was strikingly different between the tests to imitation and
those to command. Overall ability to imitate correlated
best with right posterior parietal regions (see Fig l),
while correct responses to command correlated most
strongly with left inferolateral regions (see Fig 2).
These patterns were not significantly different when
patients with predominant language or predominant
- 0.36
"Values are the Spearman rank correlation coefficients for 17 Alzheimer patients.
"p < 0.01; 'p < 0.001.
Wechsler Adult Intelligence Scale
visual-spatial deficits were excluded. While most individual tasks tended to conform with these two general
patterns, some tests (for example, nonrepresentational
tasks and transitive movements) seemed to be less localizing than others.
The initial description of apraxia 1231 and much of our
current understanding of this disorder are based on the
close examination of a relatively few subjects in whom
a lesion was relatively well delineated and various
elementary functions appeared intact [lS, 18, 311.
Such cases are unusual, however, and it is difficult to
determine whether they reflect exceptional circumstances or if the findings can be generalued to others.
Alternatively, several large patient groups have recently been studied f9, 11, 171, although this approach
tends to confound precise localization. Moreover,
knowledge of the exact extent of structural damage
may sometimes be misleading. Depending on the
cause of the lesion, fibers of passage may or may not be
involved along with local neurons. Moreover, it now
appears that damage to one part of the cerebral cortex
can impair function in other parts of the same hemisphere [81, the ipsilateral basal ganglia 12, 8, 22, 261,
the contralateral cortex C22, 231, and even the contralateral cerebellum E2, 8, 231. It is thus understandable
that classification schemes and localization patterns for
apraxia have remained controversial.
In the current study, praxis was examined in patients
with Alzheimer's disease because it offers certain advantages for analysis and because it probably accounts
for most cases of apraxia 171. Alzheimer patients manifest a wide range of performance across many different
cognitive skills so that the relationships between these
functions can be closely examined. Furthermore, Alzheimer's disease is relatively stable over a period of
Foster et
Cerebral Mapping of Apraxia in Alzheimer's Disease 141
weeks to a few months, and cerebral edema, diaschisis,
or white matter involvement does not complicate attempts to localize cortical dysfunction. Whether the
present results can be extrapolated to other neurological disorders remains to be determined.
Liepmann’s C241 origind description of apraxia derived from a patient with a left hemisphere lesion who
had difficulty both with performance to command and
on imitation. Others have since considered the inability to imitate as a more severely damaged subset of the
inability to perform movements to command 17, 15,
16, 31). On the other hand, the present results are not
wholly inconsistent with earlier models of pathways
required for praxis. A close relationship between language ability and performance of skilled movements to
command has previously been recognized { l l , 16, 20,
21). In addition, it has been hypothesized that the
premotor region in the left hemisphere contains motor
engrams critical to the coordinated and successful performance of complex motor tasks [Is). This region is
remarkably close to the cortical area where performance to commands and glucose utilization were
found to be most highly correlated (see Fig 2). It is also
well understood that auditory comprehension is necessary for verbally initiated praxis; thus it is not surprising that the comprehension section of the Boston
Diagnostic Aphasia Examination correlates closely
with performance to command. However, lack of comprehension does not fully explain the apraxia to spoken command. If the command were not understood,
then either no response or a completely inappropriate
act would usually be expected. However, our patients
frequently responded with recognizable actions that
were nevertheless partially inaccurate. Performance to
command thus appears to be a distinct neurological
function, albeit difficult to assess when there is
difficulty with comprehension or some degree of
aphasia. Furthermore, it appears to deteriorate at its
own pace, independent of the severity of overall dementia.
The cortical regions and neural pathways required
for imitation of movement are not so clearly understood. So far no patient with a deficit limited to imitation has been studied fully, although some patients
may exist 19, 171. Some have believed that failure to
imitate movement or gesture simply reflects a more
severe manifestation of apraxia to command 17, 15, 16,
31). The few studies that have been focused on imitation have presented the spoken command along with
the pantomime 120, 211, included an object to assist
with the task 1201, studied only those who were already aphasic C27, 211, or included patients whose
right hemisphere lesions were not precisely localized
110, 11, 161. In one study, 20% of right braindamaged patients were found to have mild to striking
impairment of imitation, but nothing further is known
142 Annals of Neurology
Vol 19 No 2
February 1986
of the extent of their lesions [lo). The present results
(see Fig 1) suggest that only a portion of the right
posterior hemisphere is critical for praxis to imitation,
so that patients with specific involvement of this region
should be studied. It has been suggested that a motor
engram for skilled movements also exists in the right
hemisphere 115, 201, and our findings would suggest
that such an engram may exist and that it may play a
special role in imitation. Identification of a movement
did not necessarily ensure its correct performance.
Thus, retrieval of the name of an act that presumably
could elicit a motor engram in the left frontal cortex
was not sufficient for correct imitation. It seems reasonable that visual-spatial recognition and then reconstruction of the relationship of various body partssimilar to constructional praxis-would
be necessary
for correct imitation. Although visual agnosia could
also prevent recognition of the movements, it is unlikely that this accounts for the disability in our patients, as recognition of objects in the Boston naming
test did not correlate with imitation. Similarly, others
have found little correlation between imitation and
aphasia {lo}.
It should be noted that the cortical localization patterns obtained in these studies at best reflect only the
dysfunctional areas most proximately related to clinical
test performance. They do not provide information
about all cerebral regions and interconnections required for normal praxis. For example, although the
primary sensory and motor cortex are certainly required to accomplish the tasks tested, the lack of
highly significant correlations in these regions suggests
that they are not the areas most closely related to the
praxis failure in Alzheimer’s disease. On the other
hand, regional increases in glucose utilization during
the performance of complex learned motor paradigms
should give a clearer picture of the extent of cortical
activation required for accurate replication of skilled
No striking differences in performance were seen in
any of the various types of movements or gestures
tested. There did not appear to be sparing of facial or
axial movements, as has been reported with some
types of lesions [I 5). Moreover, there was relatively
little difference between simple and more complex
tasks or those which required a sequence of movements. These results may reflect the testing of relatively mildly impaired subjects. At this stage, most
patients were still able to spontaneously perform tasks
that they had difficulty mimicking or performing on
command. At least in Aizheimer patients, there seems
to be little basis for classifying acts beyond whether
they are familiar and meaningful to the patient. Although the tasks we used had been divided into twelve
categories based on their presumed ideational content,
complexity, and parts of the body used 11s), the pat-
terns observed in these groups varied little. Thus we,
as others 1111, have found essentially two types of
apraxia: apraxia to spoken command and apraxia to
imitation. Each seems to localize to a separate area of
brain and progress independently. Classic terminology
such as ideational and ideomotor apraxia is perhaps more
useful when describing disconnection syndromes or
more specific deficits. In Alzheimer's disease there appears to be a progressive deterioration in praxis that
does not conform to these categories, nor is it limited
to one side of the body.
Nonrepresentational movements that do not use
either a previously learned or useful sequence of
movements failed to locahze to any cortical area. As
such movements do not call upon a previously learned
motor engram, this finding suggests that the areas of
high correlation represent the location of motor engrams: in the left hemisphere a verbally initiated motor
engram and in the right hemisphere a visually initiated
motor engram. For those apraxia tasks yielding a localizing pattern, this pattern was consistent: imitation
related to glucose utilization in the right posterior
hemisphere, and command related to glucose utilization in the left inferior hemisphere, including, to a
lesser or greater degree, Broca's and Wernicke's areas.
Apraxia is a characteristic feature of Alzheimer's disease. Its severity cannot be predicted by the degree of
general intellectual decline or memory loss, but like
other cognitive functions appears to reflect discrete
regional brain damage. In assessing individual cases,
the degree of aphasia, visual agnosia, and comprehensional loss must also be taken into consideration. A
better understanding of these deficits and their contribution to the progressive disability occurring in Alzheimer's disease should help in the assessment of new
therapeutic approaches to this disorder.
Presented in part at the 107th Annual Meeting of the American
Neurological Association, Washington, DC, October 1982.
1. Allison RS: The Senile Brain: A Clinical Study. Baltimore, Williams & Wilkins, 1962, pp 202-220
2. Baron JC, Bousser MG, Comar D, et ak Noninvasive tomographic study of cerebral blood flow and oxygen metabolism in
vivo. Eur Neurol 20:273-284, 1981
3. Brooks RA: Alternative formula for glucose utilization using
labeled deoxyglucose. J Nucl Med 23:538-539, 1982
4. Brown JW: Aphasia, Apraxia and Agnosia-Clinical
Theoretical Aspects. Springfield, IL, Thomas, 1972, pp 157160
5. Chase TN, Fedio P, Foster NL, et al: Wechsler adult intelligence scale performance. Cortical localization by fluorodeoxyglucose F 18-positron emission tomography. Arch Neurol
1211244-1247, 1984
6. Chase TN, Foster NL, Fedio P, et al: Regional cortical dysfunction in Alzheimer's disease as detetmined by positron emission
tomography. Ann Neurol 15:S170-S174, 1984
7 De Ajuriaguerra J, Tissot R:The apraxias. In Vinken PJ, Bruyn
GW (eds): Disorders of Speech, Perception and Symbolic Behavior (vol 4, Handbook of Clinical Neurology). Amsterdam,
North Holland, 1969, pp 48-66
8. DeLaPaz RL,Patronas NJ, Brooks RA, et al: Positron emission
tomographic study of suppression of gray matter glucose utilization by brain tumors. AJNR 4:826-829, 1983
9. DeRenzi E, Faglioni P, Sorgato P: Modality-specific and supramodal mechanisms of apraxia. Brain 105:301-312, 1982
10. DeRenzi E, Motti F, Nichelli P: Imitating gestures-a quantitative approach to ideomotor apraxia. Arch Neurol 37:6-10,
11. DeRenzi E, Piecturo A, Vignoto LA: Ideational apraxia: a quantitative study. Neuropsychologia 6:41-52, 1968
12. Foster NL, Chase TN , Fedio P, et al: Alzheimer's disease: focal
cortical changes shown by positron emission tomography. Neurology (Cleveland) 33:961-965, 1983
13. Foster NL, Chase TN, Mansi L, et ak Cortical abnormalities in
Alzheimer's disease. Ann Neurol 16:649-654, 1984
14. Gado M, Hanaway J, Frank R Functional anatomy of the cerebral cortex by computed tomography. J Comput Assist Tomogr
3:l-19, 1979
15. Geschwind N: The apraxias: neural mechanisms of disorders of
learned movement. Am Sci 63:188-195, 1975
16. Goodglass H, Kaplan E: Disturbance of gesture and pantomime
in aphasia. Brain 86:703-720, 1963
17. Heilman KM: Ideational apraxia-a redefinition. Brain 962361864, 1973
18. Heilman KM, Rothi LJ, Valenstein E: Two forms of ideomotor
apraxia. Neurology (NY) 32:342-346, 1982
19. Huang SC, Phelps ME, Hoffman EJ, et al: Noninvasive determination of local cerebral metabolic rate of glucose in man. Am
J Physiol 238:E69-E82, 1980
20. Kertesz A, Ferro JM, Shean CM: Apraxia and aphasia: the functional-anatomical basis for their dissociation. Neurology (Cleveland) 34140-47, 1984
21. Kertest A, Hooper P: Praxis and language: the extent and variety of apraxia in aphasia. Neuropsychologia 20:275-286, 1982
22 Kuhl DE, Phelps ME, Kowell AP, er al: Effects of stroke on
local cerebral metabolism and perfusion: mapping by emission
computed tomography of "FDG and I3NH3. Ann Neurol
8:47-60, 1980
23. Lenzi GL, Frackowiak RSJ, Jones T. Cerebral oxygen metabolism and blood flow in human cerebral ischemic infarction. J
Cereb Blood Flow Metab 2:321-325, 1982
24. Liepmann H. Das Krankheitsbild der Apraxie (motorische
Asymbolie). Monatsschr Psychiatr Neurol 8: 15-44, 102-132,
182-197, 1900
25. Maziotta JC, Phelps ME, Miller J, et al: Tomographic mapping
of human cerebral metabolism: normal unstimdated state. Neurology (NY) 31:503-516, 1981
26. Metter JE, Wasterlain CG, Kuhl DE, et al: "FDG positron
emission computed tomography in a study of aphasia. Ann
Neurol 10:173-183, 1981
27. Phelps ME, H u n g SC, Hoffman EJ, et al: Tomographic measurement of local cerebral glucose metabolic rate in humans
validation of method.
with (F-18)2-fluoro-2-deoxy-~glucose:
Ann Neurol 6:371-388, 1979
28. Reivich M, Kuhl D, Wolf A, et al: Measurement of local cerebral glucose metabolism in man with '*F-2-fluor0-2-deoxy-Dglucose. Acta Neurol Scand 56(suppl 64):192-193, 1977
29. Rosen WG, Terry RD, F d d PA, et al: Pathological verification
of ischemic score in differentiation of dementias. Ann Neurol
7~486-488, 1980
30 Sokoloff L Relationships between functional activity and energy
metabolismin the CNS. Trans Am Soc Neurochem 11:171,1980
31. Watson RT, Heilman KM: Callosal apraxia. Brain 106:391403, 1983
Foster et al: Cerebral Mapping of Apraxia in Alzheimer's Disease
Без категории
Размер файла
577 Кб
mapping, tomography, emissions, disease, apraxia, positron, alzheimers, cerebral
Пожаловаться на содержимое документа