Impact of white matter hyperintensities scoring method on correlations with clinical data: The LADIS study

Elisabeth C.W. Van Straaten*, Franz Fazekas, Egill Rostrup, Philip Scheltens, Reinhold Schmidt, Leonardo Pantoni, Domenico Inzitari, Gunhild Waldemar, Timo Erkinjuntti, Riita Mäntylä, Lars Olof Wahlund, Frederik Barkhof

*Corresponding author for this work

Research output: Contribution to journalArticleAcademicpeer-review

Abstract

Background and Purpose - White matter hyperintensities (WMH) are associated with decline in cognition, gait, mood, and urinary continence. Associations may depend on the method used for measuring WMH. We investigated the ability of different WMH scoring methods to detect differences in WMH load between groups with and without symptoms. Methods - We used data of 618 independently living elderly with WMH collected in the Leukoaraiosis And Disability (LADIS) study. Subjects with and without symptoms of depression, gait disturbances, urinary incontinence, and memory decline were compared with respect to WMH load measured qualitatively using 3 widely used visual rating scales (Fazekas, Scheltens, and Age-Related White Matter Changes scales) and quantitatively with a semiautomated volumetric technique and an automatic lesion count. Statistical significance between groups was assessed with the χ2 and Mann-Whitney tests. In addition, the punctate and confluent lesion type with comparable WMH volume were compared with respect to the clinical data using Student t test and χ2 test. Direct comparison of visual ratings with volumetry was done using curve fitting. Results - Visual and volumetric assessment detected differences in WMH between groups with respect to gait disturbances and age. WMH volume measurement was more sensitive than visual scores with respect to memory symptoms. Number of lesions nor lesion type correlated with any of the clinical data. For all rating scales, a clear but nonlinear relationship was established with WMH volume. Conclusions - Visual rating scales display ceiling effects and poor discrimination of absolute lesion volumes. Consequently, they may be less sensitive in differentiating clinical groups.

Original languageEnglish
Pages (from-to)836-840
Number of pages5
JournalStroke
Volume37
Issue number3
DOIs
Publication statusPublished - 1 Mar 2006

Cite this