Magnetic Resonance Imaging of the Brain in Diabetes: The Cardiovascular Determinants of Dementia (CASCADE) Study

Pathophysiology

Magnetic Resonance Imaging of the Brain in Diabetes

¹Department of Neurology, Medical University Graz, Graz, Austria
²Laboratory of Epidemiology, Demography, and Biometry, National Institute on Aging, National Institutes of Health, Bethesda, Maryland
³Department of Epidemiology and Biometry, Erasmus Medical Center, Rotterdam, the Netherlands
⁴Department of Psychology, Stockholm University, Stockholm, Sweden
⁵Department of Epidemiology and Population Studies, Institute of Public Health, Jagiellonian University Medical School, Kraków, Poland
⁶Institute of Health Studies, Department of Health and Social Security, Barcelona, Spain
⁷Institute of Epidemiology and Social Medicine, University of Muenster, Muenster, Germany
⁸Institut National de la Santé et de la Recherche Médicale (INSERM) Unit 360, Epidemiological Research in Neurology and Psychopathology, Hopital La Salpetriere, Paris, France
⁹Department of Epidemiology and Public Health, University College, London, U.K
¹⁰Istituto Superiore di Sanità, Laboratory of Epidemiology and Biostatistics, Rome, Italy

Address correspondence and reprint requests to Dr. Reinhold Schmidt, Department of Neurology, Medical University Graz, Auenbruggerplatz 22, A-8036 Graz, Austria. E-mail: reinhold.schmidt{at}uni-graz.at

Download figure
Open in new tab
Download powerpoint
FIG. 1.
Results from linear regression analysis. The figure shows diabetes and cortical atrophy in CASCADE subcohorts and in the entire CASCADE cohort stratified by the absence and presence of hypertension. The black rectangles represent β-coefficients adjusted for age, sex, years of education, smoking status, coronary heart disease, BMI, and cholesterol level. 95% CIs are denoted by lines.

TABLE 1

Description of the study cohorts

Site	n	Sex (% male)	Age (mean years)	Education (mean years)	Diabetes	Hyper-tension (%)	CHD (%)	Ever smoker (%)	BMI (mean)	Cholesterol (mean mmol/l)
Austria	169	53.8	67.9	11.1	10.7	39.6	34.3	43.8	26.8	6.0
France	186	39.2	68.4	10.7	7.0	38.2	11.8	41.9	25.9	6.3
Italy	167	44.9	69.8	4.5	9.6	46.1	13.8	37.7	28.6	6.1
The Netherlands	220	51.8	70.1	9.3	4.5	45.5	21.4	72.3	26.3	6.7
Poland	102	56.9	70.5	7.4	17.6	74.5	36.3	42.2	28.4	5.4
Spain	110	53.6	70.1	5.9	19.1	38.2	12.7	50.0	28.1	5.9
Sweden	139	47.5	68.6	9.1	7.9	60.4	15.1	48.9	26.7	6.8
U.K.	159	74.8	66.3	12.0	4.4	26.4	8.8	51.6	24.7	6.7
Total	1,252	52.3	68.9	9.0	9.1	44.6	18.8	49.7	26.8	6.3

TABLE 2

Demographics and risk factors in nondiabetic and diabetic subjects

Factor	Nondiabetic subjects	Diabetic subjects	P value
n	1,138	114	—
Sex (% male)	51.3	62.3	0.03*
Age (years)	68.8 ± 3.2	69.3 ± 3.0	0.08†
Education (years)	9.0 ± 4.2	8.5 ± 4.7	0.34†
Hypertension (%)	43.1	60.5	<0.0001*
Coronary heart disease (%)	17.4	33.3	<0.0001*
Smoking status (%)
Never	51.3	44.6
Former	36.6	42.9
Current	12.1	12.5	0.37*
BMI (kg/m²)	26.7 ± 3.9	28.1 ± 3.8	<0.0001†
Total cholesterol (mmol/l)	6.3 ± 1.2	6.3 ± 1.4	0.59†

TABLE 3

Magnetic resonance imaging findings in nondiabetic and diabetic subjects

MRI finding	Nondiabetic subjects	Diabetic subjects	P value
n	1,138	114
Focal lesions
Total periventricular WML-score	2.0 (1.0–3.0)	2.0 (1.0–4.0)	0.10*
Severe periventricular WML (%)	18.7	25.4	0.08†
Total volume of subcortical WML	0.16 (0.02–1.07)	0.31 (0.02–1.79)	0.18*
Severe subcortical WML (%)	25.7	29.8	0.35†
Presence of infarcts (%)	7.2	9.6	0.34†
Measures of brain atrophy
Total cortical atrophy-score	7.2 ± 3.0	8.2 ± 3.2	0.001*
Severe cortical atrophy (%)	22.6	35.1	0.003†
Subcortical atrophy
Ventricle-to-brain ratio	0.32 ± 0.04	0.33 ± 0.04	0.03*
Severe subcortical atrophy (%)	23.8	29.8	0.16†

Data are median (interquartile range) or means±SE, unless otherwise noted.
*
* Mann-Whitney U test,
†
† χ² test. Severe abnormalities were considered if the extent of changes belonged to the upper quintile of the distribution.

March 2004, 53(3)