The relation of cerebral magnetic resonance signal hyperintensities to Alzheimer's disease

https://doi.org/10.1016/0022-510X(96)00169-4Get rights and content

Abstract

To further elucidate the relation of cerebral magnetic resonance signal hyperintensities to Alzheimer's disease (AD) we performed a case-control comparison between 30 consecutive patients with probable AD (age range 49–76, mean 65 years) and 60 asymptomatic volunteers matched for age, sex, and major cerebrovascular risk factors. We used a 1.5T magnet and determined the extent of morphologic abnormalities both by visual grading and measurement. AD patients showed comparable grades of deep/subcortical white matter hyperintensities (WMH) and a similar extent of the total WMH area as controls (3.3 cm2 ± 8.8 vs. 2.0 cm2 ± 4.6). They had significantly more often a ‘halo’ of periventricular hyperintensity (PVH) (p < 0.0005) and an increased mean PVH thickness (3.0 mm ± 1.9 vs. 1.3 mm ± 1.2; p < 0.001). This PVH thickness correlated significantly with measures of ventricular enlargement. While univariate logistic regression also suggested a significant association of PVH thickness with a diagnosis of AD this association was lost against atrophy measures in a multivariate analysis. Our results confirm a significantly greater extent of PVH in AD patients than controls even when matched for cerebrovascular risk factors. However, this abnormality was not independently related to the disease but rather appears to be an epiphenomenon of brain atrophy.

Introduction

Magnetic resonance imaging (MRI) of patients with Alzheimer's disease (AD) may show focal hyperintensities in the deep and subcortical white matter (WMH) and bright caps or rims around the lateral ventricles, so-called periventricular hyperintensity (PVH) (Brant-Zawadzki et al., 1985; Zimmerman et al., 1986; Fazekas et al., 1987). In an early report of ourselves, we found no difference in the extent of deep and subcortical WMH between AD and control subjects. However, AD patients showed a significantly increased frequency of a smooth ‘halo’ of PVH (Fazekas et al., 1987). Subsequent studies yielded conflicting results. Some confirmed the presence of more pronounced PVH in AD patients without a difference in WMH (Harrell et al., 1991; Mirsen et al., 1991; Waldemar et al., 1994). Some reported a higher rate of both PVH and WMH (Bowen et al., 1990; Schmidt, 1992), yet others found no difference in signal abnormalities at all (Kozachuk et al., 1990; Leys et al., 1990; Erkinjuntti et al., 1994). Finally, we ourselves could not fully replicate our initial observation (Fazekas et al., 1991).

Discrepancies in the rate of WMH may be largely attributable to differences in the distribution of cerebrovascular risk factors among the investigational groups. Studies reporting a similar frequency of WMH in AD patients and controls were in general those which had excluded individuals with major cerebrovascular risk factors from analysis (Kozachuk et al., 1990; Leys et al., 1990; Fazekas et al., 1991; Waldemar et al., 1994). The relation of PVH to AD seems less clear (Harrell et al., 1991; Erkinjuntti et al., 1994). Histopathologic data suggest a non-vascular etiology for those types of PVH which have been described in AD (Fazekas et al., 1993). We previously speculated on an association between a periventricular ‘halo’ and pronounced ventricular enlargement (Fazekas et al., 1991), but the relation of signal abnormalities to atrophic changes of AD brains has not yet been assessed in detail.

We therefore graded and measured the MRI signal hyperintensities of AD patients and related their extent to other morphologic brain changes and the clinical diagnosis. Controls matched for age, sex and major cerebrovascular risk factors served for comparison in order to minimize potentially confounding effects of these variables.

Section snippets

Patients

This study was performed on 30 patients with probable AD ranging from 49–76 years and 60 matching participants of the Austrian Stroke Prevention Study. The study population's demographic variables including the frequency of cerebrovascular risk factors are shown in Table 1.

AD patients consisted of consecutive admissions to the Memory Clinic of the Department of Neurology, Karl-Franzens University, who had undergone full clinical and laboratory evaluation following the recommendations of the MRC

Results

Visual rating of MRI signal hyperintensities showed a similar proportion of WMH grades in AD patients and controls. Periventricular caps/lining and a smooth ‘halo’ of PVH were significantly more often seen in AD patients, however (Table 2). The morphometric measures confirmed this finding. PVH thickness was significantly greater in AD patients than controls (p < 0.001) while there was no difference in the WMHA between both investigational groups (Table 3). AD patients also showed a significant

Discussion

In extension of previous studies this investigation compared the graded and measured extent of MRI hyperintensities in the brains of AD patients with results obtained from a large cohort of controls carefully matched in respect to age, sex, and risk factors. In accordance with studies excluding patients with cerebrovascular risk factors from the analysis (Kozachuk et al., 1990, Leys et al., 1990, Fazekas et al., 1991; Waldemar et al., 1994) we found no significant difference in the frequency of

References (28)

  • M.F Folstein et al.

    ‘Mini-mental state’: a practical method for grading the cognitive state of patients for the clinician

    J. Psychiatr. Res.

    (1975)
  • American Psychiatric Association, Committee on Nomenclature and Statistics (1987) Diagnostic and Statistical Manual of...
  • B.C Bowen et al.

    MR signal abnormalities in memory disorder and dementia

    Am. J. Neuroradiol.

    (1990)
  • M Brant-Zawadzki et al.

    MR imaging of the aging brain: patchy white-matter lesions and dementia

    Am. J. Neuroradiol.

    (1985)
  • T Erkinjuntti et al.

    Lack of difference in brain hyperintensities between patients with early Alzheimer's disease and control subjects

    Arch Neurol.

    (1994)
  • F Fazekas et al.

    MR signal abnormalities at 1.5 T in Alzheimer's dementia and normal aging

    Am. J. Neuroradiol.

    (1987)
  • F Fazekas et al.

    Magnetic resonance imaging hyperintensities in Alzheimer's disease

    Letter. Arch. Neurol.

    (1991)
  • F Fazekas et al.

    Pathologic correlates of incidental MRI white matter signal hyperintensities

    Neurology

    (1993)
  • V.C Hachinski et al.

    Cerebral blood flow in dementia

    Arch. Neurol.

    (1975)
  • L.E Harrell et al.

    The relationship of high-intensity signals on magnetic resonance images to cognitive and psychiatric state in Alzheimer's disease

    Arch. Neurol.

    (1991)
  • W.E Kozachuk et al.

    White matter hyperintensities in dementia of the Alzheimeŕs type and healthy subjects without cerebrovasclar risk factors a magnetic resonance imaging study

    Arch. Neurol.

    (1990)
  • J.P Laissy et al.

    Midsaggital MR measurements of the corpus callosum in healthy subjects and diseased patients: a prospective survey

    Am. J. Neuroradiol.

    (1993)
  • D Leifer et al.

    Clinicopathologic correlations of cranial magnetic resonance imaging of periventricular white matter

    Neurology

    (1990)
  • D Leys et al.

    Periventricular and white matter magnetic resonance imaging hyperintensities do not differ between Alzheimer's disease and normal aging

    Arch. Neurol.

    (1990)
  • Cited by (85)

    • Cerebrovascular disease, neurodegeneration, and clinical phenotype in dementia with Lewy bodies

      2021, Neurobiology of Aging
      Citation Excerpt :

      Whereas several studies showed that the WMH burden is higher in patients with probable DLB than in healthy controls (Barber et al., 1999; Joki et al., 2018; Kim et al., 2015; Koikkalainen et al., 2016; Sarro et al., 2017), no differences in WMH burden have also been reported (Burton et al., 2006; De Reuck et al., 2016; Oppedal et al., 2012). Further, it is unclear whether the WMH burden in DLB is due to primary vascular pathology or is secondary to neurodegeneration (Barber et al., 2000; Chimowitz et al., 1992; Fazekas et al., 1996). Two studies showed that WMH burden in DLB is associated with increased global brain atrophy and atrophy in the medial temporal lobes (Barber et al., 2000; Joki et al., 2018).

    • White matter lesions and sudden sensorineural hearing loss

      2019, Journal of Clinical Neuroscience
      Citation Excerpt :

      All images were evaluated, for WMLs, by two experienced neuroradiologists, independently, and the results of each evaluation were then discussed and matched. When present, these were graded according to Fazekas and Wahlund’s scale [17,18]. WMLs score based on Fazekas’s criteria ranges from 0 to 6 and on Wahlund from 0 to 30.

    • Corpus callosum atrophy associated with the degree of cognitive decline in patients with Alzheimer's dementia or mild cognitive impairment: A meta-analysis of the region of interest structural imaging studies

      2015, Journal of Psychiatric Research
      Citation Excerpt :

      To study the potential impact of brain size on the results of the CC midsagittal area between patients and healthy controls, studies with the same methodology for brain size correction were grouped together for subgroup meta-analyses. A total of 1787 studies were initially identified with our aforementioned search strategy, and 25 of them met our inclusion criteria for AD analysis (Anstey et al., 2007; Biegon et al., 1994; Black et al., 2000; Cuenod et al., 1993; Fazekas et al., 1996; Frederiksen et al., 2011; Gootjes et al., 2006; Hanyu et al., 1999; Hensel et al., 2002; Janowsky et al., 1996; Kaufer et al., 1997; Li et al., 2008; Lyoo et al., 1997; Pantel et al., 1999; Pitel et al., 2010; Teipel et al., 2003; Thomann et al., 2006; Thompson et al., 1998; Tomaiuolo et al., 2007; Vermersch et al., 1996; Wang et al., 2006; Wiltshire et al., 2005; Yamauchi et al., 1993, 2000; Zhu et al., 2012). However, the data for effect size calculation of two studies (Anstey et al., 2007; Janowsky et al., 1996) were unavailable after contacting the authors and they were excluded in final analysis.

    View all citing articles on Scopus
    View full text