Browse by author
Lookup NU author(s): Professor John O'Brien, Dr Michael FirbankORCiD, Dr Timo Erkinjuntti
Full text for this publication is not currently held within this repository. Alternative links are provided below where available.
Background: Both types of cerebral white matter hyperintensities, periventricular (PVL) and deep white matter lesions (DWML) have been previously associated with the development of depression in older subjects. However, it remains controversial as to whether PVL, DWML, or both are most strongly associated with depression and this was the aim of the current study. Methods: In a pan-European multicentre study of 626 older subjects, we examined the relationship between PVL and DWML, depressive symptoms (GDS quintile), cognitive status (MMSE), hypertension and history of stroke. Results: In univariate analysis we found that depressive symptoms as assessed by GDS were associated with both types of white matter lesions (Spearman rho = 0.12 p = 0.002 for DWML and rho = 0.09 p = 0.01 for PVL). Using ordinal logistic regression analysis the total DWML score (p = 0.041), rather than PVL (p = 0.9) was found to predict GDS scores. Conclusions: DWML, but not PVL, were most strongly associated with depressive symptoms in this sample. As DWML (unlike PVL) are associated with vascular ischaemic damage, our findings are consistent with the 'vascular depression' hypothesis. Longitudinal studies are needed to clarify the time course of these relationships, in particular, whether modifying DWML alters the natural history of depression. Copyright © 2006 John Wiley & Sons, Ltd.
Author(s): Krishnan MS, O'Brien JT, Firbank MJ, Pantoni L, Carlucci G, Erkinjuntti T, Wallin A, Wahlund L-O, Scheltens P, van Straaten ECW, Inzitari D
Publication type: Article
Publication status: Published
Journal: International Journal of Geriatric Psychiatry
Year: 2006
Volume: 21
Issue: 10
Pages: 983-989
ISSN (print): 0885-6230
ISSN (electronic): 1099-1166
Publisher: John Wiley & Sons Ltd.
URL: http://dx.doi.org/10.1002/gps.1596
DOI: 10.1002/gps.1596
PubMed id: 16955428
Altmetrics provided by Altmetric
