Radiology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published online before print August 26, 2005, 10.1148/radiol.2371041496
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
2371041496v1
237/1/251    most recent
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Murray, A. D.
Right arrow Articles by Whalley, L. J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Murray, A. D.
Right arrow Articles by Whalley, L. J.
(Radiology 2005;237:251-257.)
© RSNA, 2005

Neuroradiology

Brain White Matter Hyperintensities: Relative Importance of Vascular Risk Factors in Nondemented Elderly People1

Alison D. Murray, FRCR, FRCPE, Roger T. Staff, PhD, Susan D. Shenkin, MRCP, Ian J. Deary, PhD, FRCPE, John M. Starr, FRCPE and Lawrence J. Whalley, MD, FRCPE, FRCPsych

1 From the Departments of Radiology (A.D.M.), Biomedical Physics and Bioengineering (R.T.S.), and Mental Health (L.J.W.), University of Aberdeen, Foresterhill, Aberdeen AB25 2ZD, Scotland; and Departments of Psychology (I.J.D.) and Geriatric Medicine (S.D.S., J.M.S.), University of Edinburgh, Edinburgh, Scotland. Received August 30, 2004; revision requested November 5; revision received November 26; accepted January 3, 2005. L.J.W., A.D.M., I.J.D., and R.T.S. supported by a grant from the Chief Scientist Office of the Scottish Health Department. I.J.D. supported by a Royal Society-Wolfson Research Merit Award. L.J.W. supported by a Wellcome Trust Career Development Award. S.D.S. supported by a Medical Research Council Clinical Training Fellowship. Address correspondence to A.D.M. (e-mail: a.d.murray{at}abdn.ac.uk).


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 References
 
PURPOSE: To prospectively determine whether there is an association between brain white matter signal hyperintensities on magnetic resonance (MR) images and potential risk factors for cerebral ischemia in a well-characterized narrow age cohort of nondemented community-dwelling elderly people.

MATERIALS AND METHODS: The study population consisted of surviving members of the Aberdeen 1921 Birth Cohort, a subsample of participants in the 1932 Scottish Mental Survey who were born in 1921. With the permission of the local ethics committee and with informed written consent, 106 nondemented subjects (62 men, 44 women) aged 78–79 years underwent T2-weighted brain MR imaging. Brain MR images were scored semiquantitatively for deep white matter hyperintensities and periventricular hyperintensities. Vascular risk factors and clinical measures potentially associated with cerebral ischemia included hypertension, diabetes, cerebrovascular disease, smoking, body mass index grade, respiratory function levels (forced expiratory volume in 1 second [FEV1], forced vital capacity [FVC], and peak expiratory flow rate [PEFR]) normalized for subject's height, plasma lipid levels (cholesterol, triglycerides, high-density lipoprotein, and low-density lipoprotein), glycated hemoglobin level, and mean fasting blood glucose level. Pearson correlation coefficients were calculated for correlations between potential vascular risk factors and scores for deep white matter and periventricular hyperintensities, and stepwise multiple linear regression analysis was performed for factors with a statistically significant correlation.

RESULTS: Significant Pearson correlations with deep white matter hyperintensities were found for glycated hemoglobin level (r = 0.31), hypertension (r = 0.27), normalized FEV1 (r = –0.27), normalized FVC (r = –0.22), normalized PEFR (r = –0.27), low-density lipoprotein (r = 0.24), and cholesterol (r = 0.20), and with periventricular hyperintensities for glycated hemoglobin level (r = 0.28) and normalized PEFR (r = –0.23). Multiple linear regression analysis showed that glycated hemoglobin level and hypertension were predictive of 16.2% of the variance in deep white matter hyperintensities. When subjects with non–insulin-dependent (type 2) diabetes mellitus (n = 11) were excluded, hypertension and decreased normalized PEFR were predictive of 11.7% of the variance.

CONCLUSION: White matter hyperintensities are associated with elevated levels of glycated hemoglobin in nondemented community-dwelling elderly subjects. Hypertension and decreased normalized PEFR are the principal predictors of deep white matter hyperintensities in nondiabetic subjects.

© RSNA, 2005


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 References
 
White matter lesions are frequently observed on brain images, particularly in elderly people. Originally described as "leukoaraiosis" on the basis of findings at computed tomography (CT) (1), brain white matter lesions are depicted with greater sensitivity at magnetic resonance (MR) imaging (2) as areas of high signal intensity on T2- and intermediate-weighted MR images. They are not usually visible on T1-weighted MR images, however, and they do not follow specific vascular territories; these characteristics help to distinguish them from infarcts (3). The term white matter hyperintensities applies both to patchy deep white matter areas and to smooth periventricular areas of high signal intensity. Most MR imaging visual rating scales distinguish between these two patterns (4), which are probably of different pathologic origin (5).

The prevalence of white matter hyperintensities increases with age, and age is a confounding variable in any population-based study in which risk factors are investigated (6,7). White matter hyperintensities are associated with impaired cognition, balance, and gait (6,8–10) and are likely to be multifactorial in origin. Hypertension and reduced respiratory function (as defined by decreases in forced expiratory volume in 1 second [FEV1] and in forced vital capacity [FVC]) are the factors most consistently associated with white matter hyperintensities (11–13). In previous evaluations of lifetime cognitive change, it was demonstrated that white matter hyperintensities may account for 13%–14% of the variance in cognitive ability in old age from values predicted on the basis of childhood mental ability scores, and that such variance is independent of childhood mental ability (14,15). White matter hyperintensities are acquired abnormalities that may reflect age-related disturbances in brain metabolism that contribute to the lifetime burden of brain aging. In view of the increasing proportion of elderly people in the population, it would be highly desirable to identify the relative importance of various risk factors for cerebral ischemic change that might be amenable to prevention or treatment. Thus, the aim of this study was to prospectively determine whether there is an association between white matter hyperintensities and potential risk factors for cerebral ischemia in a well-characterized narrow age cohort of nondemented community-dwelling elderly people.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 References
 
Subjects
Narrow age cohorts are particularly useful for examining processes that are associated with biologic aging (16). Our study sample was drawn from the Aberdeen 1921 Birth Cohort, a group of volunteers who took part in the Scottish Mental Survey of 1932 (17), the goal of which was to measure the mental abilities of almost all (n = 87 498) children born in 1921 who were attending Scottish schools on June 1, 1932. With the permission of the local ethics committee, 354 participants in the 1932 survey who were still living in Aberdeen, Scotland, in 1999 were invited to take part in a prospective longitudinal study of brain aging and health. A total of 285 subjects aged 78–79 years agreed to take part, and 235 were well enough to be recruited. From this sample, 144 subjects who were not hospitalized and were not living in a nursing home were randomly invited to undergo brain MR imaging. Inclusion criteria were the availability of childhood mental ability scores, current independent living in the community, and ability to give informed written consent. One hundred nine subjects fulfilled these inclusion criteria and consented to undergo MR imaging. Exclusion criteria were neurologic illness (such as Parkinson disease, multiple sclerosis, or stroke resulting in loss of independence); dementia (defined as a Mini-Mental State Examination score of less than 24); inability to give consent; and the usual contraindications to MR imaging, which were present in two subjects (one with claustrophobia and the other with shrapnel in the neck). A third subject was excluded because of movement during image acquisition. Thus, complete MR imaging data were obtained in 106 subjects aged 78–79 years, of whom 62 (58.5%) were men, 44 were women (41.5%), and all were white. The larger proportion of men reflects the greater ease of tracing male subjects, who do not usually change their names, as well as the fact that a greater number of women declined to undergo MR imaging.

Vascular Risk Factor Assessment
Vascular risk factors were assessed at interview, clinical examination, and blood testing. Interviews, clinical examination, and blood sampling were carried out by two trained research nurses, with the supervision of a senior psychiatrist (L.J.W.) who had 27 years of experience as a consultant psychiatrist. At the interview, 44 (41.5%) subjects were known to be hypertensive and were receiving antihypertensive treatment at the time of recruitment to the study. Six (5.6%) subjects were known to have type 2 diabetes mellitus. In two of the six, diabetes was controlled with a dietary regimen; in the other four, it was controlled with oral hypoglycemic agents. There were no subjects with insulin-dependent diabetes mellitus. Thirty subjects (28.3%) specified a history of possible transient ischemic attacks, including temporary dizziness or other visual, motor, or sensory symptoms, but none had a history of cerebrovascular events with resultant persistent neurologic disability. Eight (7.5%) of the subjects were current smokers, 64 (60.4%) were previous smokers, and 34 (32.1%) had never smoked. Previous smokers had stopped smoking a mean of 34 years ± 16.3 (standard deviation) (range, 6–66 years) prior to the study.

Clinical evaluation of vascular risk factors comprised the following measurements: Body mass index was measured in kilograms per square meter and graded according to World Health Organization criteria, from grade 1 (seriously underweight, 10–16 kg/m2) to grade 6 (obese, 38–45 kg/m2), with normal body mass index defined as grade 3 (20–25 kg/m2). Systolic and diastolic blood pressure values were measured in millimeters of mercury by using a standard sphygmomanometer after the subject sat resting for 5 minutes in a warm and quiet room. The mean of three measured values was recorded. Respiratory function was estimated by using the highest value recorded during three attempts after one practice session with a spirometer (Microlab 3000; Micro Medical, Rochester, England). FEV1 and FVC were measured in liters, and peak expiratory flow rate (PEFR) was measured in liters per minute and was normalized for subject height in meters.

Laboratory analyses of venous blood included measurement of cholesterol, triglycerides, low-density lipoprotein, high-density lipoprotein, and glycated hemoglobin levels. Plasma glucose level was measured in millimoles per liter after the subject fasted overnight. All blood analyses were performed at the biochemistry laboratory that serves the regional teaching hospital.

MR Imaging Protocol and Scoring System
MR imaging was performed on a 1-T imager (Magnetom Impact; Siemens, Erlangen, Germany) by experienced radiologic technologists 8–47 weeks after the acquisition of clinical data for all subjects. The imaging protocol included a T2-weighted fast spin-echo MR sequence in the commissural plane with repetition time msec/echo time msec of 4000/96, total acquisition time of 1 minute 53 seconds, section thickness of 5 mm, and an intersection gap of 1.5 mm. A three-dimensional T1-weighted MR sequence also was applied with 11.4/4.4, a flip angle of 15°, field of view of 250 mm, slab thickness of 180 mm, effective section thickness of 1.41 mm, and total acquisition time of 6 minutes 7 seconds. T2-weighted MR images were scored according to a semiquantitative rating scale devised by Fazekas et al (5). Areas of signal hyperintensity were classified as either deep white matter (subcortical) or periventricular and were scored on a four-point scale, with scores defined as follows: for deep white matter areas of hyperintensity, as absent (grade 0), punctate (grade 1), nearly coalescent (grade 2), or confluent (grade 3); and for periventricular areas of hyperintensity, as absent (grade 0), pencil-thin lines (grade 1), caps or bands (grade 2), or confluent (grade 3). Examples of the different grades of deep white matter and periventricular hyperintensities are shown in the Figure. We previously demonstrated good interobserver reliability and excellent intraobserver reliability with use of this method by three observers (9). The mean of two ratings obtained 3 months apart by a neuroradiologist with 12 years of experience in the evaluation of brain MR images (A.D.M.) was entered into the analysis for both deep white matter and periventricular hyperintensities because this observer had previously been found to have the highest reliability.



View larger version (76K):
[in this window]
[in a new window]
[Download PPT slide]
  		T2-weighted transverse MR images (4000/96, section thickness of 5 mm, intersection gap of 1.5 mm) in three different subjects show the features typical for each score in the Fazekas grading system: A, grade 1, small punctate area of signal hyperintensity (arrow) in deep white matter, with a thin periventricular rim of hyperintensity; B, grade 2, large deep white matter hyperintensities in the left frontal lobe that appear to coalesce (arrow), with obvious bands of signal hyperintensity around the lateral ventricles; and C, grade 3, deep white matter hyperintensities and periventricular hyperintensities that are nearly confluent. Note that the scores for deep white matter and periventricular hyperintensities in the same subject need not be identical: It is possible for MR images to show deep white matter hyperintensities with a score of 1 and periventricular hyperintensities with a score of 2 in the same subject.

 
Statistical Methods
All data analysis was performed by using statistical software (SPSS, version 12; SPSS, Chicago, Ill). Bivariate correlation analyses were performed with calculation of Pearson (including point biserial) correlation coefficients to identify vascular risk factors that had a significant (P < .05) association with deep white matter and periventricular hyperintensities and to examine associations among vascular risk factors. Spearman correlation coefficients ({rho}) also were calculated because some predictors did not meet criteria for parametric analysis. Vascular risk factors were entered into a stepwise multiple linear regression analysis, with deep white matter hyperintensity and periventricular hyperintensity used in turn as the dependent variable. The threshold of significance, which was the criterion for entry into the model, was set at .05. Cases were included pairwise for bivariate correlation analyses and listwise for multiple linear regression analysis. Because normoglycemia, impaired glycemic control, and type 2 diabetes represent a continuum in old age (18), diabetic subjects were included in these analyses initially. We then repeated the analyses after excluding 11 subjects with type 2 (non-insulin-dependent) diabetes.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 References
 
Clinical and Laboratory Measures of Vascular Risk Factors
The mean body mass index grade was normal (3.60 ± 0.74), with a range of 2–6. In addition to 44 subjects with a previous diagnosis of hypertension who were receiving antihypertensive medication, a further eight subjects had untreated hypertension, defined as a systolic blood pressure of more than 150 mm Hg and/or a diastolic blood pressure of more than 95 mm Hg; thus, 52 (49.5%) of 105 subjects were hypertensive. The mean systolic blood pressure was 144 mm Hg ± 20 (range, 100–201 mm Hg), and mean diastolic blood pressure was 75 mm Hg ± 11 (range, 48–117 mm Hg). Mean values for respiratory function parameters were as follows: FEV1, 1.91 L ± 0.59 (range, 0.49–3.17 L); FVC, 2.08 L ± 0.62 (range, 0.49–3.43 L); and PEFR, 324 L/min ± 109 (range, 115–597 L/min). Mean values for serum lipids were as follows: cholesterol, 5.4 mmol/L ± 1.3 (range, 2.8–8.8 mmol/L); triglycerides, 2.0 mmol/L ± 1.1 (range, 0.6–5.9 mmol/L); low-density lipoprotein, 3.2 mmol/L ± 1.0 (range, 1.0–5.7 mmol/L); and high-density lipoprotein, 1.3 mmol/L ± 0.4 (range, 0.6–3.2 mmol/L). The mean glycated hemoglobin level was 5.8% ± 0.9 (range, 4.4% to –11.1%). The mean fasting glucose level was 6.0 mmol/L ± 1.4 (range, 4.6–11.9 mmol/L). In addition to the six subjects with a diagnosis of type 2 diabetes mellitus, a further five subjects had type 2 diabetes mellitus (defined as a mean fasting glucose level of >7 mmol/L) that was not diagnosed prior to the study. Complete data about vascular risk factors were available for 94 (88.7%) of the 106 study subjects (Table 1).


View this table:
[in this window]
[in a new window]

  		TABLE 1. Vascular Risk Factors for White Matter Hyperintensities on MR Images in 106 Subjects

 
Correlations between Vascular Risk Factors and White Matter Changes
Pearson correlation coefficients for bivariate correlations of vascular risk factors with deep white matter and periventricular hyperintensities (Table 2) showed significant correlations with deep white matter hyperintensities for glycated hemoglobin level, normalized PEFR, normalized FEV1, presence of hypertension, low-density lipoprotein level, cholesterol level, and normalized FVC. Significant correlations with periventricular hyperintensities were found for glycated hemoglobin level and normalized PEFR. However, deep white matter hyperintensities and periventricular hyperintensities were highly correlated (r = 0.76, P < .001), and it is possible that the relationships we found between vascular risk factors and periventricular hyperintensities reflected this correlation. Bivariate correlation analyses with the exclusion of diabetic subjects showed significant correlations between deep white matter hyperintensities and hypertension, glycated hemoglobin level, normalized PEFR, normalized FEV1, and normalized FVC (Table 2). There were no significant correlations between periventricular hyperintensities and any risk factor in nondiabetic subjects. Spearman ({rho}) correlation coefficients showed similar statistically significant correlations between white matter hyperintensities and the same risk factors. (Values for Spearman correlation coefficients are available from the authors by request.)


View this table:
[in this window]
[in a new window]

  		TABLE 2. Results of Analyses for Correlation of Vascular Risk Factors with Deep White Matter and Periventricular Hyperintensities

 
Regression Models of Variance
The results of a stepwise multiple linear regression analysis of vascular risk factors, with deep white matter hyperintensities used as the dependent variable and with diabetic subjects included in the analysis, showed that the level of glycated hemoglobin and presence of hypertension together accounted for 16.2% of the variance in deep white matter hyperintensities. The glycated hemoglobin level alone was predictive of 11.1% of the variance. When type 2 diabetic subjects (n = 11) were excluded from the analysis, 11.7% of the variance in deep white matter hyperintensities was accounted for by hypertension and normalized PEFR, with hypertension alone accounting for 6.7% of the variance.

The results of a stepwise linear regression analysis with periventricular hyperintensities used as the dependent variable showed that glycated hemoglobin level and normalized PEFR together accounted for 11.8% of the variance in periventricular hyperintensities. When diabetic subjects (all of whom had type 2 diabetes mellitus) were excluded, only normalized PEFR was retained in the model, and it accounted for 4.7% of the variance in periventricular hyperintensities.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
References
 
In this well-characterized cohort of elderly people born in 1921, we found that an elevated level of glycated hemoglobin was the strongest predictor of deep white matter and periventricular hyperintensities. A history of hypertension also predicted deep white matter hyperintensities (14), and hypertension and normalized PEFR were the main predictors when type 2 diabetic subjects were excluded from the analysis. Normalized PEFR was the main predictor of periventricular hyperintensities in nondiabetic subjects. There were significant correlations between increased levels of cholesterol-rich lipids and other measures of respiratory function and deep white matter hyperintensities. The glycated hemoglobin level retained a significant correlation with deep white matter hyperintensities in nondiabetic subjects. The advantages of this study over others were the very narrow age range of the subjects and the availability of data for numerous clinical and laboratory variables that were potentially related to cerebral ischemia.

Investigators in the Cardiovascular Health Study showed that white matter hyperintensities are associated with older age, silent stroke, hypertension, lower FEV1, lower income, and female sex (6,7). Our findings are consistent with these previously identified associations of white matter hyperintensities with hypertension and reduced pulmonary function.

Diabetes and Glycated Hemoglobin Level
Despite the correlation between elevated levels of glycated hemoglobin and the presence of white matter hyperintensities, we did not find similar correlations for diabetes or mean fasting glucose level. Four of the six known diabetic subjects were undergoing therapy for diabetes and had normal mean fasting glucose and glycated hemoglobin levels. The fasting glucose level involves a single measurement, is dependent on insulin production at the time of testing, and does not reflect the effects of factors such as diet and exercise on longer-term glucose homeostasis. The level of glycated hemoglobin, on the other hand, reflects prevalent glucose homeostasis over the course of 3 months and is more likely to be relevant to vascular disease (19). We did not assess the duration of diabetes mellitus, as the number of subjects with a previous diagnosis of diabetes was small and disease duration was impossible to establish for subjects in whom diabetes was diagnosed during this study. Diabetes mellitus is a recognized risk factor for white matter hyperintensities (20–22) and dementia (23). An elevated level of glycated hemoglobin in subjects with type 2 diabetes is associated with earlier death (23–25), and, in nondiabetics, with silent cerebral infarction depicted at MR imaging (26). The mechanism is likely to be ischemia caused by microvascular occlusion and elevated erythrocyte aggregation (27). Increased blood-brain barrier permeability may also be important (28,29) and may allow the accumulation of end products of advanced glycation in the brain (30) as a result of an age-related decrease in insulin secretion and increase in the blood glucose level (18,31). Together with other metabolic perturbations due to aging (32), these processes may provide sufficient explanation of the positive association reported here between glycated hemoglobin levels and white matter hyperintensities.

Pulmonary Function
Reduced pulmonary function is associated with increased risk of stroke, cardiovascular disease, and all-cause mortality (33–35), regardless of smoking status. Reduced FEV1 and FVC were associated with cerebral infarction and white matter hyperintensities in both smokers and nonsmokers in the Atherosclerosis Risk in Communities Study (13). Proposed pathogenetic mechanisms of white matter hyperintensities due to reduced pulmonary function include oxidative stress, chronic inflammation, and impaired fibrinolysis (13). Our results support the association of white matter hyperintensities with reduced pulmonary function and indicate that this association is independent of current and previous smoking status. Both PEFR and FEV1 are reduced in obstructive airways disease, which therefore is also an important risk factor for white matter hyperintensities.

Hypertension
Both systolic and diastolic hypertension are known risk factors for white matter hyperintensities (6,8,11,14,36), and increased diastolic blood pressure is associated with the progression of deep white matter hyperintensities (37). Investigators in other studies reported a J-shaped association between diastolic blood pressure and white matter lesions in those with cardiovascular disease: That is, subjects with abnormally low or high levels of diastolic blood pressure have more white matter lesions than do subjects with diastolic blood pressure in the normal range (38). An excessive reduction in diastolic blood pressure is associated with the development of new, subclinical, MR imaging–depicted lesions in hypertensive subjects (39). In this study, we found hypertension to be a risk factor with a significant correlation with white matter hyperintensities, but we did not find a relationship between deep white matter hyperintensities or periventricular hyperintensities and systolic or diastolic blood pressure values. It is likely, in the context of acquired vascular brain damage, that mean blood pressure over the lifespan is more indicative than is a single abnormal measurement late in life. Our findings with regard to glycated hemoglobin level point to a similar effect, with a lifetime burden of hypertension and impaired glycemic control being more important than isolated abnormal measurements of blood pressure and fasting glucose, respectively. We did not assess the duration of hypertension, which is difficult to ascertain accurately in retrospect, and blood pressure may have been elevated for an indeterminate time before hypertension was diagnosed. A relatively high proportion of subjects in our study sample were hypertensive (49.5%), and the majority of these were treated with ß-blockers and/or thiazide diuretics. Investigators in a recent population-based cohort study in Swedish men reported a statistically significant interaction between treatment for hypertension (mainly with ß-blockers and thiazide diuretics) and increase in blood glucose level and proinsulin concentration (40). Thus, the association between hypertension and deep white matter hyperintensities may be partly mediated by the effect of antihypertensive medication on impaired glucose homeostasis.

Cholesterol
An increased plasma cholesterol level is a risk factor for cerebral infarction (41) and was associated with white matter hyperintensities in subjects older than 65 years in the Rotterdam Study (8). However, associations between cholesterol and cerebrovascular disease are not clear. Schmidt et al (36) found that subjects with white matter hyperintensities had lower serum concentrations of total cholesterol, compared with concentrations in subjects without white matter hyperintensities. In a study of risk factors for different types of stroke, Leppala et al (41) found that serum total cholesterol was inversely related to the risk of cerebral hemorrhage, whereas those with total cholesterol of more than 7.0 mmol/L had an increased risk of cerebral infarction. Schatz et al (42), in a study of subjects from the Honolulu Heart Program, found an association between low cholesterol concentrations over a 20-year period and greater all-cause mortality. We found positive correlations between both high cholesterol and low-density lipoprotein levels and deep white matter hyperintensities but not periventricular hyperintensities.

The main limitations of this study relate to sample size, missing data, possible bias, and the risk of a type I error. However, the statistically significant correlations were not randomly distributed and are consistent with those found in previous studies (6,7). The sample size was adequate to detect medium-sized effects, and we were able to account for a reasonable proportion of the variance in deep white matter hyperintensities. Mean fasting glucose values accounted for most of the missing data, a fact that reflects the difficulty we had in persuading elderly participants to maintain the overnight fast prior to testing. Subjects who participated in the larger Aberdeen 1921 Birth Cohort study had a significantly higher mean score for childhood mental ability, with a narrower standard deviation, than did nonparticipants (9). Subjects in this MR imaging study were no different in childhood mental ability or education from participants in the larger study. As superior cognitive ability is associated with longevity, any bias is likely to underestimate the degree to which our results can be applied to the general population aged 78–79 years, although not necessarily to a younger population (43).

A potential criticism of the MR imaging method we used is the lack of fluid-attenuated inversion recovery (FLAIR) images, which provide better depiction of white matter hyperintensities, particularly periventricular hyperintensities, than do T2-weighted images. The MR imaging system used does not provide full brain coverage with a FLAIR sequence, and it was the only system available during most of this study. The scanner and protocol were kept constant throughout. The Fazekas scoring system is relatively crude and, although use of a FLAIR sequence and thinner sections would have resulted in images with higher spatial resolution, these are unlikely to have altered white matter hyperintensity scores.

This study has shown that an elevated level of glycated hemoglobin and hypertension are predictors of deep white matter hyperintensities and that an elevated level of glycated hemoglobin and reduced normalized PEFR are predictive of periventricular hyperintensities. In nondiabetic subjects, hypertension and reduced normalized PEFR are stronger predictors of deep white matter hyperintensities, but there remains a significant correlation between glycated hemoglobin level and deep white matter hyperintensities. We suggest that impaired glycemic control may be as important a risk factor as is hypertension for deep white matter hyperintensities and morbidity associated with them. Our findings require confirmation through future longitudinal follow-up studies in this cohort and in a larger population. The potential then would exist for assessment of the effects of intervention (eg, with dietary modification or drug therapy to decrease the glycated hemoglobin level) on the prevalence of white matter hyperintensities and their associated clinical manifestations.


   ACKNOWLEDGMENTS
 
The authors thank the volunteers from the Aberdeen 1921 Birth Cohort, as well as the volunteers' relatives and friends, without whose help and commitment this study would not have been possible.


   FOOTNOTES
 

Abbreviations: FEV1 = forced expiratory volume in 1 second • FVC = forced vital capacity • PEFR = peak expiratory flow rate

Authors stated no financial relationship to disclose.

Author contributions: Guarantor of integrity of entire study, A.D.M.; study concepts/study design or data acquisition or data analysis/interpretation, all authors; manuscript drafting or manuscript revision for important intellectual content, all authors; approval of final version of submitted manuscript, all authors; literature research, A.D.M., J.M.S.; clinical studies, A.D.M., L.J.W.; statistical analysis, A.D.M., R.T.S., S.D.S., I.J.D.; and manuscript editing, A.D.M., S.D.S., L.J.W.


   References
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 References
 

  1. Hachinski VC, Potter P, Merskey H. Leuko-araiosis. Arch Neurol 1987; 44(1):21–23.[Abstract]
  2. Lopez OL, Becker JT, Jungreis CA, et al. Computed tomography but not magnetic resonance imaging identified periventricular white-matter lesions predict symptomatic cerebrovascular disease in probable Alzheimer's disease. Arch Neurol 1995; 52(7):659–664.[Abstract]
  3. Pantoni L, Simoni M, Pracucci G, Schmidt R, Barkhof F, Inzitari D. Visual rating scales for age-related white matter changes (leukoaraiosis): can the heterogeneity be reduced? Stroke 2002; 33(12):2827–2833.[Abstract/Free Full Text]
  4. Scheltens P, Erkinjunti T, Leys D, et al. White matter changes on CT and MRI: an overview of visual rating scales. European Task Force on Age-Related White Matter Changes. Eur Neurol 1998; 39(2):80–89.[CrossRef][Medline]
  5. Fazekas F, Chawluk JB, Alavi A, Hurtig HI, Zimmerman RA. MR signal abnormalities at 1.5 T in Alzheimer's dementia and normal aging. Am J Neuroradiol 1987;8:421–426.
  6. Longstreth WT Jr, Manolio TA, Arnold A, et al. Clinical correlates of white matter findings on cranial magnetic resonance imaging of 3301 elderly people: the Cardiovascular Health Study. Stroke 1996; 27(8):1274–1282.[Abstract/Free Full Text]
  7. Yue NC, Arnold AM, Longstreth WT Jr, et al. Sulcal, ventricular, and white matter changes at MR imaging in the aging brain: data from the Cardiovascular Health Study. Radiology 1997; 202(1):33–39.[Abstract/Free Full Text]
  8. Breteler MM, van Swieten JC, Bots ML, et al. Cerebral white matter lesions, vascular risk factors, and cognitive function in a population-based study: the Rotterdam Study. Neurology 1994; 44(7):1246–1252.[Abstract/Free Full Text]
  9. Leaper SA, Murray AD, Lemmon HA, et al. Neuropsychologic correlates of brain white matter lesions depicted on MR images: 1921 Aberdeen Birth Cohort. Radiology 2001; 221(1):51–55.[Abstract/Free Full Text]
  10. Starr JM, Leaper SA, Murray AD, et al. Brain white matter lesions detected by magnetic resonance imaging are associated with balance and gait speed. J Neurol Neurosurg Psychiatry 2003; 74(1):94–98.[Abstract/Free Full Text]
  11. Liao D, Cooper L, Cai J, et al. The prevalence and severity of white matter lesions, their relationship with age, ethnicity, gender, and cardiovascular disease risk factors: the ARIC Study. Neuroepidemiology 1997; 16(3):149–162.[Medline]
  12. Fazekas F, Niederkorn K, Schmidt R, et al. White matter signal abnormalities in normal individuals: correlation with carotid ultrasonography, cerebral blood flow measurements, and cerebrovascular risk factors. Stroke 1988; 19(10):1285–1288.[Abstract/Free Full Text]
  13. Liao D, Higgins M, Bryan NR, et al. Lower pulmonary function and cerebral subclinical abnormalities detected by MRI: the Atherosclerosis Risk in Communities study. Chest 1999; 116(1):150–156.[Abstract/Free Full Text]
  14. Deary IJ, Leaper SA, Murray AD, Staff RT, Whalley LJ. Cerebral white matter abnormalities and lifetime cognitive change: a 67-year follow-up of the Scottish Mental Survey of 1932. Psychol Aging 2003; 18(1):140–148.[CrossRef][Medline]
  15. Staff RT, Murray AD, Deary IJ, Whalley LJ. What provides cerebral reserve? Brain 2004; 127(5):1191–1199.[Abstract/Free Full Text]
  16. Hofer SM, Berg S, Era P. Evaluating the interdependence of aging-related changes in visual and auditory acuity, balance, and cognitive functioning. Psychol Aging 2003; 18(2):285–305.[CrossRef][Medline]
  17. Scottish Council for Research in Education. The intelligence of Scottish school children: a national survey of an age group. London, England: University of London Press, 1933.
  18. Chiu KC, Lee NP, Cohan P, Chuang LM. Beta cell function declines with age in glucose tolerant Caucasians. Clin Endocrinol (Oxf) 2000; 53(5):569–575.[CrossRef][Medline]
  19. Khaw KT, Wareham N, Luben R, et al. Glycated haemoglobin, diabetes, and mortality in men in Norfolk cohort of European prospective investigation of cancer and nutrition (EPIC-Norfolk). BMJ 2001; 322(7277):15–18.[Abstract/Free Full Text]
  20. Gupta SR, Naheedy MH, Young JC, Ghobrial M, Rubino FA, Hindo W. Periventricular white matter changes and dementia: clinical, neuropsychological, radiological, and pathological correlation. Arch Neurol 1988; 45(6):637–641.[Abstract]
  21. Lee CK, Weindruch R, Prolla TA. Gene-expression profile of the ageing brain in mice. Nat Genet 2000; 25(3):294–297.[CrossRef][Medline]
  22. Lechner H, Schmidt R, Bertha G, Justich E, Offenbacher H, Schneider G. Nuclear magnetic resonance image white matter lesions and risk factors for stroke in normal individuals. Stroke 1988; 19(2):263–265.[Abstract/Free Full Text]
  23. Brayne C, Gill C, Huppert FA, et al. Vascular risks and incident dementia: results from a cohort study of the very old. Dement Geriatr Cogn Disord 1998; 9(3):175–180.[CrossRef][Medline]
  24. Groeneveld Y, Petri H, Hermans J, Springer MP. Relationship between blood glucose level and mortality in type 2 diabetes mellitus: a systematic review. Diabet Med 1999; 16(1):2–13.[CrossRef][Medline]
  25. Wirta O, Pasternack A, Mustonen J, Laippala P. Renal and cardiovascular predictors of 9-year total and sudden cardiac mortality in non-insulin-dependent diabetic subjects. Nephrol Dial Transplant 1997; 12(12):2612–2617.[Abstract/Free Full Text]
  26. Lee SC, Park SJ, Ki HK, et al. Prevalence and risk factors of silent cerebral infarction in apparently normal adults. Hypertension 2000; 36(1):73–77.[Abstract/Free Full Text]
  27. Beamer N, Giraud G, Clark W, Wynn M, Coull B. Diabetes, hypertension and erythrocyte aggregation in acute stroke. Cerebrovasc Dis 1997;7:144–149.
  28. Wardlaw JM, Sandercock PA, Dennis MS, Starr J. Is breakdown of the blood-brain barrier responsible for lacunar stroke, leukoaraiosis, and dementia? Stroke 2003; 34(3):806–812.[Abstract/Free Full Text]
  29. Starr JM, Wardlaw J, Ferguson K, MacLullich A, Deary IJ, Marshall I. Increased blood-brain barrier permeability in type II diabetes demonstrated by gadolinium magnetic resonance imaging. J Neurol Neurosurg Psychiatry 2003; 74(1):70–76.[Abstract/Free Full Text]
  30. Finch CE, Cohen DM. Aging, metabolism and Alzheimer disease: review and hypotheses. Exp Neurol 1997; 143(1):82–102.[CrossRef][Medline]
  31. Iozzo P, Beck-Nielsen H, Laakso M, Smith U, Yki-Jarvinen H, Ferrannini E. Independent influence of age on basal insulin secretion in nondiabetic humans. European Group for the Study of Insulin Resistance. J Clin Endocrinol Metab 1999; 84(3):863–868.[Abstract/Free Full Text]
  32. Tessari P. Changes in protein, carbohydrate, and fat metabolism with aging: possible role of insulin. Nutr Rev 2000; 58(1):11–19.[Medline]
  33. Hole DJ, Watt GC, Davey-Smith G, Hart CL, Gillis CR, Hawthorne VM. Impaired lung function and mortality risk in men and women: findings from the Renfrew and Paisley prospective population study. BMJ 1996; 313(7059):711–715.[Abstract/Free Full Text]
  34. Sobol BJ, Herbert WH, Emirgil C. The high incidence of pulmonary functional abnormalities in patients with coronary artery disease. Chest 1974; 65(2):148–151.[Abstract/Free Full Text]
  35. Beaty TH, Cohen BH, Newill CA, Menkes HA, Diamond EL, Chen CJ. Impaired pulmonary function as a risk factor for mortality. Am J Epidemiol 1982; 116(1):102–113.[Abstract/Free Full Text]
  36. Schmidt R, Hayn M, Fazekas F, Kapeller P, Esterbauer H. Magnetic resonance imaging white matter hyperintensities in clinically normal elderly individuals: correlations with plasma concentrations of naturally occurring antioxidants. Stroke 1996; 27(11):2043–2047.[Abstract/Free Full Text]
  37. Veldink JH, Scheltens P, Jonker C, Launer LJ. Progression of cerebral white matter hyperintensities on MRI is related to diastolic blood pressure. Neurology 1998; 51(1):319–320.[Free Full Text]
  38. de Leeuw FE, de Groot JC, Oudkerk M, et al. A follow-up study of blood pressure and cerebral white matter lesions. Ann Neurol 1999; 46(6):827–833.[CrossRef][Medline]
  39. Shintani S, Shiigai T, Arinami T. Silent lacunar infarction on magnetic resonance imaging (MRI): risk factors. J Neurol Sci 1998; 160(1):82–86.[CrossRef][Medline]
  40. Dunder K, Lind L, Zethelius B, Berglund L, Lithell H. Increase in blood glucose concentration during antihypertensive treatment as a predictor of myocardial infarction: population based cohort study. BMJ 2003; 326(7391):681.[Abstract/Free Full Text]
  41. Leppala JM, Virtamo J, Fogelholm R, Albanes D, Heinonen OP. Different risk factors for different stroke subtypes: association of blood pressure, cholesterol, and antioxidants. Stroke 1999; 30(12):2535–2540.[Abstract/Free Full Text]
  42. Schatz IJ, Masaki K, Yano K, Chen R, Rodriguez BL, Curb JD. Cholesterol and all-cause mortality in elderly people from the Honolulu Heart Program: a cohort study. Lancet 2001; 358(9279):351–355.[CrossRef][Medline]
  43. Maier H, Smith J. Psychological predictors of mortality in old age. J Gerontol B Psychol Sci Soc Sci 1999; 54(1):P44–P54.[Abstract]



This article has been cited by other articles:


Home page
 RadiologyHome page
N. Altaf, P. S. Morgan, A. Moody, S. T. MacSweeney, J. R. Gladman, and D. P. Auer
Brain White Matter Hyperintensities Are Associated with Carotid Intraplaque Hemorrhage
Radiology, July 1, 2008; 248(1): 202 - 209.
[Abstract] [Full Text] [PDF]

Home page
 Arch Intern MedHome page
M.-C. Corti, G. Baggio, L. Sartori, G. Barbato, E. Manzato, E. Musacchio, L. Ferrucci, G. Cardinali, D. Donato, L. J. Launer, et al.
White Matter Lesions and the Risk of Incident Hip Fracture in Older Persons: Results From the Progetto Veneto Anziani Study
Arch Intern Med, September 10, 2007; 167(16): 1745 - 1751.
[Abstract] [Full Text] [PDF]

Home page
 StrokeHome page
X. Guo, L. Pantoni, M. Simoni, D. Gustafson, C. Bengtsson, B. Palmertz, and I. Skoog
Midlife Respiratory Function Related to White Matter Lesions and Lacunar Infarcts in Late Life: The Prospective Population Study of Women in Gothenburg, Sweden
Stroke, July 1, 2006; 37(7): 1658 - 1662.
[Abstract] [Full Text] [PDF]

Home page
 StrokeHome page
D. G. Munoz
Leukoaraiosis and Ischemia: Beyond the Myth
Stroke, June 1, 2006; 37(6): 1348 - 1349.
[Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
2371041496v1
237/1/251    most recent
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Murray, A. D.
Right arrow Articles by Whalley, L. J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Murray, A. D.
Right arrow Articles by Whalley, L. J.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
RADIOLOGY RADIOGRAPHICS RSNA JOURNALS ONLINE