(Stroke. 1996;27:1290-1295.)
© 1996 American Heart Association, Inc.
Articles |
INSERM U360, Hopital de la Salpetriere (A. Auperin, C.B., A. Alperovitch), INSERM U258, Hopital Broussais (C.B.-K., P.D.), and the Service de Neurologie, Hopital Saint Antoine (P.-J.T.), Paris; and Centre EVA (I.R.), Nantes, France.
Correspondence to C. Berr, INSERM U360, Hopital de la Salpetriere, 47 Blvd de l'Hopital, 75651 Paris Cedex 13, France.
| Abstract |
|---|
|
|
|---|
Methods A cohort of 1279 subjects (men, 41%) aged 59 to 71 years was recruited from the electoral rolls of the city of Nantes (western France). Cognitive performances were evaluated with the Mini-Mental State Examination (MMSE) and seven neuropsychological tests assessing attention, psychomotor rapidity, verbal abilities, memory, and visuospatial perception. For each test, subjects were classified into three performance levels with a quartile distribution: 25% highest, 25% lowest, and 50% middle. The intima-media thickness of common carotid arteries and the presence of plaques in the carotid arteries were assessed with B-mode ultrasound examination.
Results Only 28% of men and 17% of women had carotid plaques inducing moderate stenosis of the lumen (<40%). After adjustment for possible confounders, odds ratios for poor cognitive performance associated with plaques were above 1 for all cognitive tests in men. This association was statistically significant for the MMSE and another test assessing attention skills. There was a slight association between increase of the common carotid intima-media thickness and poor cognitive scores in men with plaques. In women, no association was found between cognitive functions and presence of plaques or intima-media thickness.
Conclusions This study indicated a moderate association between atherosclerosis of the carotid arteries and poor cognitive functioning in men aged 59 to 71 years. In view of these moderate cross-sectional results, further studies are required to better assess the relationship between carotid atherosclerosis and cognitive impairment.
Key Words: aging atherosclerosis carotid arteries cognition epidemiology
| Introduction |
|---|
|
|
|---|
We examined cross-sectional relationships between cognitive functioning and carotid wall characteristics in a large population sample of individuals aged 59 to 71 years to determine whether carotid atherosclerotic lesions were associated with low cognitive performance.
| Subjects and Methods |
|---|
|
|
|---|
Carotid Examination
A high-resolution B-mode ultrasound examination of left and right carotid arteries in 1385 subjects was performed using an Aloka SSD-650 with a 7.5-MHz transducer. Acquisition, processing, and storage of B-mode images were computer-assisted using a software specially designed for longitudinal studies.9 The ultrasound examination involved scanning the common carotid arteries, the carotid bifurcations, and the first 2 cm of the internal carotid arteries. The IMT (distance between the media-adventitia interface and the lumen-intima interface) was automatically measured twice on longitudinal B-mode images of the far wall of each common carotid artery at plaque-free sites. The common carotid IMT was defined as the mean of four measurements. All segments of the carotid arteries were scanned longitudinally and transversely to assess the presence of plaques, and a localized echo-structure protruding into the lumen was considered to be a plaque if the distance between the media-adventitia interface and the internal side of the lesion was
1 mm. The total number of plaques was recorded. For each, maximum thickness was measured perpendicular to the vessel wall. When several plaques were present in the same carotid segment (ie, common carotid artery or bifurcation/origin of the internal carotid artery), plaque thickness measurement was made only for that with the greatest protrusion into the lumen. On the basis of transverse views, the degree of carotid stenosis was defined as (1-residual area/vessel area)x100%.
Interreader variations were analyzed in random subsamples of longitudinal images. Interreader agreement regarding the presence of plaques was excellent (
coefficient=.90). The correlation coefficient between repeated readings was .82 for IMT and .78 for plaque thickness. Because of technical reasons, the ultrasound examinations performed during the first weeks of the study were considered unreliable and were excluded (n=77).
Neuropsychological Testing
Subjects were evaluated with a neuropsychological battery that included seven tests assessing different cognitive functions and a global test (MMSE) of 18 items roughly assessing various cognitive skills with scores ranging from 0 to 30. An MMSE score <24 is considered indicative of poor cognitive skills.10 Visual attention and psychomotor rapidity were assessed with the TMT, part B, and the DSST from the Wechsler Intelligence ScaleRevised. Auditory attention was evaluated with the PASAT. The BVRT assessed visuospatial perception. The battery also included an assessment of immediate memory and learning strategies with the AVLT and of verbal fluency and search strategies with the WFT. General skills such as logical intelligence and reasoning were studied with the RPM. The MMSE, DSST, WFT, and RPM were administered to all participants. The other tests were administered to subjects who obtained an MMSE score
24. Six subjects with major sensorial inabilities were excluded.
Measurement of Covariates
The currently available literature suggests potential risk factors for poor cognitive functioning; these factors were measured and used in the analysis. Educational achievement was distributed in four categories based on years of schooling: 0 to 5, 6 to 9, 10 to 12, and >12. Subjects were also categorized as current smokers, former smokers, and nonsmokers. Weekly consumption of alcohol was determined, and three categories of average daily intake (in milliliters) were created: no alcohol consumption, 1 to 40 mL/d, and >40 mL/d.
Subjects completed the CES-D scale, a 20-item self-report of depressive symptoms presented during the week preceding the baseline examination.11 CES-D scores >16 in men and >22 in women are considered the cutoff levels for depression screening in the French population.12
Two independent measurements of systolic and diastolic blood pressure were taken with a digital electronic tensiometer after a 10-minute rest. The mean of the two measures was used in the analysis.
Statistical Analysis
Analysis was performed separately in men and women, using the SAS software package. For each cognitive test, three levels of cognitive performance were defined according to quartiles of test-score distribution: the lowest quartile defined a poor level, the highest a high level, and the two intermediate quartiles a middle level.
Bivariate associations were assessed with the
2 test and variance analysis, and statistical testing of trends was performed when necessary. Nominal polychotomous logistic regressions13 (BMDP statistical software) were used to analyze the relationships between each cognitive score distribution of the eight neuropsychological tests and carotid characteristics (presence of plaques, IMT), adjusting for age, education, depressive symptomatology, systolic blood pressure, body mass index, smoking, and alcohol consumption. The cognitive performance was the explained variable. Because there were three cognitive performance levels, two odds ratios were computed for each carotid characteristic considered a potential risk factor for cognitive functioning; one contrasted poor versus high cognitive score categories and the other contrasted intermediate versus high cognitive score categories. For each odds ratio, the high cognitive level was the reference category.
| Results |
|---|
|
|
|---|
|
Mean MMSE score was 28.3±2.0 in men versus 27.9±2.3 in women; 3.4% of men and 5.6% of women had MMSE scores <24. All cognitive scores were strongly associated with educational level in both sexes.
Crude relationships between plaques and cognitive functioning are shown in the Figure
. In all cognitive tests, prevalence of carotid plaques was higher in men with poor cognitive performance than in those with high performance. A significant (P<.05) or borderline (P<.10) trend to an increasing prevalence of plaques from the high to the poor cognitive performance category was found in five of the eight neuropsychological tests: MMSE, DSST, TMT, PASAT, and AVLT. Thus, with the MMSE, the percentages of men with plaques in the poor, middle, and high score categories were 34.9%, 26.5%, and 25.0%, respectively. Moreover, in four tests, we found a significant trend (P<.05) to an increased number of plaques associated with decreasing cognitive performance (DSST and PASAT) or a borderline trend (P<.10; MMSE and TMT). In men with carotid plaques, there was no relationship between plaque thickness and cognitive functioning. Table 2
shows odds ratios and 95% confidence intervals with regard to carotid plaques in men, which were obtained from polychotomous logistic regression models. In all neuropsychological tests, the odds ratios related to carotid plaques for poor versus high cognitive performances were >1, but they were significant only with MMSE and DSST. The odds ratios for middle versus high cognitive performances were close to 1.0 in all neuropsychological tests but one (DSST), indicating that these two levels of performance were almost similar whether carotid plaques were present or not.
|
|
In women, no association, not even a trend, was found between cognitive performance and the presence or number of plaques (Figure
).
In men, we found no consistent relationship between common carotid IMT and cognitive functions (Table 3
). In only two tests (DSST and WFT) was a statistically significant trend observed, since the mean IMT increased with high to poor cognitive performance levels. Separate analyses were performed in men with and without plaques. In men with plaques, significant or borderline associations between higher IMT and lower test scores were found for four of the eight tests. In men without plaques, no association was found. In men with plaques, estimates of odds ratios obtained from logistic regression models, with adjustment for possible confounders, also showed a slight increase in the risk of poor cognitive scores associated with a 0.1-mm increase in the common carotid IMT (Table 4
).
|
|
In women, no relationship was observed between common carotid IMT and cognitive performance (Table 3
).
| Discussion |
|---|
|
|
|---|
Because the EVA cohort has a higher educational level and better cognitive scores than the French population of the same age,14 participation in the study was probably selective. Consequently, the contrast between high and poor cognitive levels defined according to quartiles of cognitive-score distribution was weak. Because the reproducibility of carotid measurements was good, the ultrasound measurement errors should only slightly have affected the relationship under study. The prevalence of plaques was low, and the mean IMT of the common carotid was below that reported for other elderly populations.15 16 17 With the number of subjects included in this study, the power of a two-tailed test (
=5%) was adequate in men and women (
80%) to detect an odds ratio
2 for poor cognitive performance with plaques.
The association observed in men between carotid atherosclerosis and cognitive functioning was slight, but this is consistent with previous reports. In the Rotterdam study, Breteler et al7 8 found that, in individuals aged 75 to 94 years, plaques in the carotid bifurcation were significantly associated with low MMSE scores; this was not significant in individuals aged 55 to 74 years. In this cross-sectional analysis, peripheral arterial diseases and previous myocardial infarction assessed with electrocardiography were also associated with low MMSE scores. Two small longitudinal studies18 19 suggested that atherosclerotic diseases were a risk factor for cognitive decline.
In the present study, poor cognitive functioning was associated with plaques, whereas there was only a weak association with common carotid IMT in the subgroup of men with plaques. This may suggest that these two types of vascular lesions result from different pathophysiological mechanisms: atherosclerosis for plaques, and adaptive or aging-associated nonatheromatous process for intimal and medial thickening.20 21 22 Ultrasound studies, however, have indicated that intima-media thickening is associated with classic atherosclerosis risk factors such as age, hypertension, lipid alterations, or smoking.23 24 25 26 27 28 29 30 Thus, if a greater intima thickness is an early phase of the atherosclerotic process,31 it may be associated with no clinically observable disorders, especially for low values of the IMT such as those we measured. In the EVA study, age and systolic blood pressure were strongly correlated with both thicker intima-media and plaques, and an independent association between both types of carotid lesions was also found.32
Because atherosclerotic lesions led to low-grade carotid stenosis (<40% in all men but one), it is most unlikely that hemodynamic disturbances could explain cognitive impairment. In white matter lesions shown on MRI in elderly persons, pathological studies33 34 have described arteriolosclerosis, a degenerative disease of the smallest arteries. Recent studies found that white matter lesions were associated with increased carotid IMT, presence of plaques,35 36 and poor cognitive function.37 38 39 40 41 Since the association between carotid plaques and poor cognitive functioning persisted after adjustment for main vascular risk factors, other factors, such as angiotensin-converting enzyme, could affect both carotid arteries and intracerebral small arteries despite their histological difference.
In contrast to the results in men, we found no relationship between cognitive performances and carotid characteristics in women. The power of the study was not lower in women than in men. But it is well established that vascular diseases differ in men and women. First, the incidence of vascular morbidity is shifted to higher ages in women. The prevalence of carotid artery plaques is higher in men than in women of the same age,17 42 and the sex gap in morbidity due to coronary disorders tends to diminish only after 75 years of age.43 Second, the probabilities of vascular diseases attached to cardiovascular risk factors are not the same for men and women.17 43 The process that could explain the association observed in men between carotid atherosclerosis and cognitive functioning is unknown, and it could work differently in women. A gap in age or the absence of association in women are two possible hypotheses.
Further cross-sectional and longitudinal studies are needed to assess relationships between carotid artery disease, vascular risk factors, white matter lesions, and cognitive decline in the elderly.
| Selected Abbreviations and Acronyms |
|---|
|
| Acknowledgments |
|---|
| Footnotes |
|---|
Received November 24, 1995; revision received March 6, 1996; accepted April 12, 1996.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
P. J. Smith, J. A. Blumenthal, M. A. Babyak, B. M. Hoffman, P. M. Doraiswamy, R. Waugh, A. Hinderliter, and A. Sherwood Cerebrovascular Risk Factors, Vascular Disease, and Neuropsychological Outcomes in Adults With Major Depression Psychosom Med, July 1, 2007; 69(6): 578 - 586. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. Dufouil, F. Richard, N. Fievet, J. F. Dartigues, K. Ritchie, C. Tzourio, P. Amouyel, and A. Alperovitch APOE genotype, cholesterol level, lipid-lowering treatment, and dementia: The Three-City Study Neurology, May 10, 2005; 64(9): 1531 - 1538. [Abstract] [Full Text] [PDF] |
||||
![]() |
E. B. Mathiesen, K. Waterloo, O. Joakimsen, S. J. Bakke, E. A. Jacobsen, and K. H. Bonaa Reduced neuropsychological test performance in asymptomatic carotid stenosis: The Tromso Study Neurology, March 9, 2004; 62(5): 695 - 701. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. R. Waldstein, C. F. Tankard, K. J. Maier, J. R. Pelletier, J. Snow, A. W. Gardner, R. Macko, and L. I. Katzel Peripheral Arterial Disease and Cognitive Function Psychosom Med, September 1, 2003; 65(5): 757 - 763. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Stroke Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search Copyright © 1996 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited. |