Neuropsychological Importance of Subcortical White Matter Hyperintensity Larry A. Tupler, PhD; C. Edward Coffey, MD; Patrick
E.
Subcortical hyperintensity on magnetic resonance imaging is a common incidental finding in healthy elderly subjects. The relationship of such changes to cognitive functioning remains unclear, however, because only a small number of studies have examined this issue with conflicting results. We therefore assessed 66 healthy adult volunteers (mean [\m=+-\SD]age, 61.8\m=+-\15.8years) with magnetic resonance imaging scans rated for subcortical hyperintensity, and with two neuropsychological instruments selected a priori on the basis of previous reports in the literature. Findings were highly significant for both the Benton Facial Recognition Test and the Wechsler Adult Intelligence Scale-Revised Digit Symbol. However, in both cases, the majority of variance was accounted for by age and educational level. Effects of subcortical hyperintensity were not significant. We conclude that subcortical hyperintensity in healthy adults does not relate to cognitive functioning, at
Logue, PhD; William
(Arch Neurol. 1992;49:1248-1252)
in subcortical white matter are common inci¬ dental findings on computed tomography and mag¬ netic resonance imaging (MRI) in healthy elderly subjects.14 The relationship of such changes to cognitive functioning remains unclear, however, because only a small number of studies have examined this issue with conflicting results.512 These reports are also difficult to compare, given differences in subject population (sample size, age range, and exclusion criteria), scanning technique, lesion definition and grading, and neuropsychological as¬ sessment (Table 1). We assessed performance on two care¬ fully chosen neuropsychological instruments to examine the important issue of subcortical white matter hyperin¬ tensity on MRI and its relationship to cognitive perfor¬ mance in healthy middle-aged and elderly subjects.
Changes
Djang, MD; Susan
M.
Fagan, EdM
SUBJECTS AND METHODS
\s=b\
least with these two instruments.
T.
Subjects
Sixty-six healthy adult volunteers recruited from the commu¬ nity gave written informed consent to participate in this study following a full explanation of the procedures. A total of 42 women and 24 men, ranging in age from 45 to 84 years (mean [±SD] age, 61.8±15.8 years), participated in the study. All subjects were strongly right handed.13 Each participant was living inde¬ pendently in the community and was either actively employed or retired. In general, the subjects represented a wide educational and socioeconomic cross-section of the population.14 Years of ed¬ ucation for the sample averaged 15.8 years (SD, 2.7 years). Each subject underwent a structured medical and neuropsy¬ chiatrie history interview and examination, including the MiniMental State examination.15 No participant had a history of any cerebral or psychiatric illness (including substance abuse), and all denied use of psychotropic medications. Most subjects (n=36) were free of clinically measurable systemic illness. The remaining 30 subjects were taking medications for one or more of the following conditions: hypertension or organic heart disease (n=15), hypercholesterolemia (n=l), diabetes mellitus (insulin, n=2; oral agent, n=l), postmenopausal symptoms with hormone replacement (n=8), hypothyroidism (n=l), osteoarthritis (n=9), and glaucoma (n=2). Any subject who smoked or who took med¬ ication for hypertension, diabetes mellitus, hypercholesterolemia, or organic heart disease was considered to have a "risk factor" for
vascular disease (see below). In addition to the findings of subcortical white matter hyper¬ intensity described below, many of the subjects also exhibited other structural abnormalities on their brain MRI scans. Neuroradiologic evaluation revealed cortical atrophy in 34 subjects (52%) (rated as mild in 30 and moderate in four subjects) and en¬ largement of the lateral ventricles in 29 subjects (44%) (rated as mild in 17 and moderate in 12 subjects). Both cortical atrophy and lateral ventricular enlargement were present in 22 (33%) of these subjects. No other abnormalities were present on the brain MRI scans.
Brain MRI Procedure All MRI scans of the brain were obtained with one of three 1,5-T (Signa Systems, General Electric Corp, Milwaukee, Wis) at Duke University Medical Center, Durham, NC Head position was oriented in the scanner with use of laser sources fixed to the canthomeatal plane and was stabilized during the scanning pro¬ cedure with Velcro straps and molded foam head supports. Typ¬ ical imaging variables included two excitations, a 256x128 matrix, and a field of view of 20 to 24 cm. Spin-echo pulse sequences were used to produce intermediate (repetition time, 2500 milliseconds; echo time, 40 milliseconds) and T2-weighted (repetition time, 2500 milliseconds; echo time, 80 milliseconds) scanners
Accepted
publication April 27, 1992. From the Departments of Psychiatry (Medical Psychology) (Drs Tupler and Logue and Ms Fagan) and Radiology (Dr Djang), Duke University Medical Center, Durham, NC, and the Departments of Psychiatry (Neuropsychiatry) (Dr Coffey) and Medicine (Neurology) (Dr Coffey), Medical College of Pennsylvania, Allegheny Campus, Pittsburgh, Pa. Reprint requests to the Allegheny Neuropsychiatric Institute, 7777 Steubenville Pike, Oakdale, PA 15071 (Dr Coffey). for
Downloaded From: http://archneur.jamanetwork.com/ by a UCSF LIBRARY User on 06/15/2015
.
Table 1.—Published Studies of
investigators Brant-Zawadzki et
al,5
1985
Neuropsychological
Subjects 14 healthy elderly volunteers (6 M/8 F), aged 59-81 y (mean, 71.7
y); no medical conditions "associated with cognitive loss" (psychiatric histories not
examined)
Steingart
et
1987
al,
105
healthy elderly
volunteers (56
M/49 F), aged 59-91 y (mean [±SD], 71.2 + 5.8 y); no evidence of dementia or stroke;
psychiatric history
not
examined Rao et
al,7
1989
50
healthy middle-aged
volunteers (11 M/39 F),
aged
(mean, 43.7 y); no evidence of major medical, 25-60 y
neurologic,
or
illness
Hunt et 1989
al,
46
psychiatric
healthy elderly volunteers (17 M/29 F) (mean [±SDJ age, 78.2±4.6 y); no evidence of major medical illness
(psychiatric
Test Performance in
Subcortical White Matter*
histories
not
examined)
Healthy Subjects With Imaging Changes in Findings
Imaging/Measurement Technique (0.35-T), standardized severity ratings (5-point scale) of white
MRI
hyperintensity made by 2 (¡nterrater agreement not
matter
raters
given) from intermediate and
T2-weighted
scans
CT, determination of presence of leukoaraiosis
by 1 "blinded"
rater
analysis of neuropsychological test data (10 tests) was conducted; one of the 10 subjects
No statistical
with a rating of 1 or less scored in the "demented range" on the Wechsler Memory Scale (Russell revision) and WAIS-R Block Design; one of the 4 subjects with a rating of 2 or greater had impaired performance on WAIS-R Picture Arrangement
Subjects with leukoaraiosis (n=9) had lower
scores on
the Extended Scale for
Dementia than
subjects without the finding (n=96), even after controlling
for age, sex, education, and presence of ¡nfarct (ANCOVA) MRI (1.5-T), presence of leukoaraiosis Relative to subjects without leukoaraiosis on intermediate and T2-weighted (n=40), subjects with the finding scans (no description of rating (n=10) performed significantly worse methods provided) on 3 of 45 neuropsychological tests (f tests): Benton Facial Recognition Test (P