вход по аккаунту


Cortical and subcortical dementia.

код для вставкиСкачать
Epilepsy and Intelligence
Stephen Shafer, MD, MPH
The prospective study by Bourgeois and colleagues El] is a
valuable clarification of the literature on epilepsy and intelligence. The sensible final statement of the abstract, however,
does not follow from the analysis. The statement reads, “The
findings suggest that . . . total seizure control should not be
achieved at the price of repeated episodes of drug toxicity.”
The authors avoid any explicit causal inference but do suggest
that “drug toxicity” leads to I Q decline.
To distinguish the effect of excessive drug levels from the
effect of “difficult-to-control” seizures, the investigators compare the significance levels yielded by t tests on two different
groups. I Q decline in the difficult-to-control group is not
significant, whereas the I Q drop in the “toxic-range” group is.
Therefore, the argument goes, toxicity is “more closely related” to I Q decline than is difficulty in seizure control.
The actual IQ score declines in the two groups are almost
the same (103.2 - 97.1 = 6.1, versus 100.5 - 95 = 5.5).
The toxic-range group has more subjects and, at both time
points, less within-group variation; hence, the second test has
greater statistical power. The finding of significance in one
test and lack of it in another does not tell us here that one
variable weighs more than the other.
Moreover, the comparison of two single-variable tests does
not allow control of the relationship between them. In the
causal chain one would suspect that difficulty in controlling
seizures promotes both more frequent antiepilepsy drug
level testing and higher risk of so-called “toxic” levels. Therefore, to assess any independent effect of toxic levels, the
analysis should have controlled statistically for difficulty in
clinically controlling seizures. This was not done. The relationship between difficulty in control and toxic-range blood
levels, which is probably very close, has not been satisfactorily disentangled.
Certainly repeated episodes or sustained periods of clinical
toxicity are undesirable, and certainly these are related (but
not equivalent) to having one or more antiepilepsy drug
levels above the therapeutic range. My objection is not to the
principle that excess drug effect may be too much cost for too
little benefit, but to the way this conclusion is drawn in this
Departments of Public Health and Neurology
Gertrude H . Sergievsky Center
Columbia University
630 W 168th St
New York, NY 10032
1. Bourgeois BFD, Prensky AL, Palkes HS, et al: Intelligence in
epilepsy: a prospective study in children. Ann Neurol 14:438444, 1983
John Rimmer, PhD,” and Arthur L. Prensky, MDT
We thank Dr Stephen Shafer for his comments and for pointing out an ambiguity in our presentation. Indeed, as he points
out, the I Q score decline in the two groups is not a good basis
for concluding that toxicity is more important than seizure
control. The conclusion was based not on the difference in
the significance level, but on the discriminant analysis. Discriminant analysis is a method for attempting to control statistically that which may be impossible to control clinically, i.e.,
related variables. In the discriminant analysis toxicity was always the best predictor of declining IQ, and ease of control
the seventh-best predictor. When toxicity was statistically
controlled, in the discriminant analysis the variable “ease of
seizure control” dropped to twentieth in relative predictive
power (at the 0.50 level of significance). These results were
the basis of our conclusion that toxicity was more important
than ease of control in the decline in I Q scores.
One is often confronted in a paper with the problem of not
presenting an overly technical analysis, yet still providing
sufficient detail to provide clarity. It appears in this instance
we may have erred on the side of insufficient detail, and we
want to thank D r Shafer for bringing this omission to our
”School of Social Work
Washington University
TDepartment of Neurology and
The Edward Mallinckrodt Department o f Pediatrics
Washington University School o f Medicine and the
St Louis Children’s Hospital
500 S Kingshigbway Blvd
St Louis. MO 631 78
Cortical and
Subcortical Dementia
Morris Freedman, MD, FRCP(C)
Mayeux and colleagues [4]suggest that the concept of subcortical dementia is misleading. I agree with the authors that
the label may be unsatisfactory from the standpoint of
anatomy; I am not convinced by their evidence, however,
that the concept of subcortical dementia as a clinical entity is
In the original descriptions by Albert and co-workers [ 1)
and McHugh and Folstein [6], the absence of aphasia was a
major feature distinguishing the pattern of cognitive impairment in subcortical dementia from that seen in so-called cortical dementia. Mayeux and associates [41, in contrast, found
no difference in neuropsychological profile, including language, between patients with Alzheimer’s disease (DAT)
(i.e., cortical dementia) and patients with Huntington’s disease (HD) and Parkinson’s disease (PD) (i.e., subcortical dementia). One problem with their study is that they produced
a bias by matching patients from different groups on the basis
of scores on a modified Mini-Mental Srdte examination 12,
51. This test is heavily weighted with items sensitive to language deficits, because it is almost entirely verbally mediated.
Statistically significant differences in language function would
therefore not be detected, because the groups were essentially equated for language ability.
Another methodological bias was also present in their
study. Because the group with DAT was more severely impaired intellectually than the groups with PD and H D , the
patients could be matched on the modified Mini-Mental
State examination only after the most severely impaired patients with DAT and the least impaired patients with P D and
HD were omitted from analysis. Neuropsychological profiles
were therefore compared for the patients with the mildest
“cortical dementia” and the most severe “subcortical dementia.” It is important to consider comparisons at comparable
stages of disease. A recent study on cognitive deficits in the
early stages of PD supports this point; Lees and Smith [3)
found specific neuropsychological changes consistent with
the subcortical dementia pattern. Such patients were excluded from the study of Mayeux and colleagues 141.
In addition, although the authors measured language by a
formal battery, they did not state whether the impaired language scores reflected aphasia or other cognitive deficits.
Based o n the original formulations of subcortical dementia,
one would predict that the patients with DAT were aphasic,
whereas those with PD and HD made errors resulting from
nonlinguistic cognitive factors.
Behavioral Neurology
Mount Sinai Hospital
600 Univerjity Ave
Suite 433
Toronto, Ont, Canada MSG 1 x 5
1. Albert ML, Feldman RG, Willis AL The subcortical dementia of
progressive supranuclear palsy. J Neurol Neurosurg Psychiatry
37~121-130, 1974
2. Folstein MF, Folstein SE, McHugh PR: “Mini-Mental State”: a
practical method for grading the cognitive state of patients for the
clinician. J Psychiatr Res 12:189-198, 1975
3. Lees AJ, Smith E: Cognitive deficits in the early stages of Parkinson’s disease. Brain 106:257-270, 1983
4. Mayeux R, Stern Y, Rosen J, Benson DF: Is “subcortical dementia” a recognizable clinical entity? Ann Neurol 14:278-283, 1983
5. Mayeux R, Stern Y, Rosen J, LeventhalJ: Depression, intellectual
impairment, and Parkinson’s disease. Neurology (NY) 3 1:645650, 1981
6. McHugh PR, Folstein MF: Psychiatric syndromes of Huntington’s chorea: a clinical and phenomenologic study. In Benson DF,
Blumer D (eds): Psychiatric Aspects of Neurologic Disease. New
York, Grune 81 Stratton, 1975, pp 267-286
Y. Agid, MD, PhD, and M. Ruberg, PhD
In their article “Is Subcortical Dementia a Recognizable Clinical Entity?” [S), Mayeux and colleagues conclude that the
concept of subcortical dementia may be misleading. Indeed,
in all three types of dementia considered (Alzheimer’s
[DAT), Parkinson’s [PD), and Huntington’s [HD] diseases),
both cortical and subcortical lesions are observed (the word
lesion here being used to denote cell degeneration, not the
histological signs thereof). The known subcortical lesions
found in subjects with both P D and DAT (cholinergic cells in
the substantia innominata [I 1, 121, noradrenergic cells in the
locus ceruleus { 1, 7), serotoninergic cells in the raphe nuclei)
all involve subcortico-cortical pathways. The functional inci-
dence of these subcortical lesions is thus ‘‘cortical’’ in all cases
[ 5 , 9, 103. Somatostatin-containing neurons intrinsic to the
cortex are also damaged in P D [b]and DAT 141. In H D , cell
loss has been observed in the cerebral cortex but not in
subcortical regions 12, 31.
It seems reasonable to assume that clinical dementia results
from these lesions and depends on the brain structures involved, the nature of the neuronal system destroyed, the
degree of degeneration, and the efficacy of compensatory
mechanisms. On the basis of the performance of patients with
HD, PD, and DAT on the Mini-Mental State examination,
it was concluded that the clinical aspects of dementia in these
three diseases are similar. It must be asked, however,
whether this test has the subtlety necessary for differential
analysis of these dementia syndromes, should differences exist-as intuitively they seem to do-if only because of the
problematic absence of apraxia, agnosia, and aphasia in PD
and H D , considered typical examples of subcortical dementia.
Although this type of analysis has to be carried out, it
seems illusory to hope that the question of the difference
between subcortical and cortical intellectual impairment can
be solved purely on clinical grounds, as the results of Mayeux
and colleagues show. Rather, identification and quantitative
assessment of the brain lesions in relation to the symptoms of
intellectual deterioration may lead to understanding of the
pathophysiological mechanisms of dementia. It is important
to stress that this experimental approach is not directly related to the histopathological stigmata, which are dependent,
rather, on the causes of the respective diseases. These, unfortunately, are as yet unknown.
Luboratoire de Midecine Exphimentale
CH U Pitii-Salpetriire
91, Bd de I’Hopital
75634 Paris Cedex 13, France
1. Bondareff W, Mountjoy CQ, Roth M: Selective loss of neurons
of origin of adrenergic projection to cerebral cortex (nucleus
locus coeruleus) in senile dementia. Lancet 1:783-784, 1981
2. Bruyn GW, Bots GTAM, Dom R: Huntington’s chorea: current
neuropathological status. Adv Neurol 23:83-94, 1979
3. Clark AN, Parhad IM, Folstein SF, et al: The nucleus basalis in
Huntington’s disease. Neurology (Cleveland) 33: 1262-1267,
4. Davies P, Katzman R, Terry RD: Reduced somatostatin-like
immunoreactivity in cerebral cortex from cases of Alzheimer’s
disease and Alzheimer senile dementia. Nature 288:279-280,
5. Dubois B, Ruberg M, Javoy-kid F, et al: A subcortical choiinergic system is affected in Parkinson’s disease. Brain Res (in
6. Epelbaum J, Ruberg M, Moyse E, et al: Somatostatin and dementia in Parkinson’s disease. Brain Res 278:376-379, 1983
7. Greenfield JG, Bosanquet FD: The brainstem lesions in parkinsonism. J Neurol Neurosurg Psychiatry 16:213-226, 1953
8. Mayeux R, Stern Y ,Rosen J, Benson DF: Is “subcortical dementia” a recognizable clinical entity? Ann Neurol 14:278-283,
9. Rossot MN: Dementia. Lancet 2:1200-1204, 1982
10. Scatton B, Javoy-Agid F, Rouquier L, et al: Reduction of cortical
doparnine, noradrenaline, serotonin and their metabolites in Parkinson’s disease. Brain Res 275:321-328, 1983
Без категории
Размер файла
245 Кб
subcortical, dementia, cortical
Пожаловаться на содержимое документа