Affiliations: [a]
Department of Neurology, Tergooi Hospitals, Blaricum, The Netherlands
| [b]
Department of Geriatrics, Tergooi Hospitals, Blaricum, The Netherlands
| [c]
Department of Medical Neuropsychology, Tergooi Hospitals, Blaricum, The Netherlands
| [d]
Department of Radiology, Tergooi Hospitals, Blaricum, The Netherlands
| [e]
Department of Neurology, Alzheimer Center, VU University Medical Center, Neuroscience Campus Amsterdam, Amsterdam, The Netherlands
Correspondence:
[*]
Correspondence to: Philip Scheltens, Alzheimer Center PK -1 Z 44, VU University Medical Center, de Boelelaan 1118, 1081 HZ Amsterdam, The Netherlands. Tel.: +31 20 4440816; Fax: +31 20 4440715; E-mail: p.scheltens@vumc.nl.
Note: [1] These authors contributed equally to this work.
Abstract: Background:Evidence suggests that cerebral white matter lesions (WML) play a role in cognitive decline. Objective:To assess the impact of cerebral WML on cognitive function relative to absence or presence of medial temporal atrophy (MTA) in a large single-center memory clinic population. Methods:Patients included had subjective cognitive impairment (SCI, n = 333), mild cognitive impairment (MCI, n = 492) and Alzheimer’s disease (AD, n = 832). The relationships between visually rated WML (Fazekas scale, 0–3) on brain Computed Tomography and CAMCOG memory and non-memory function were investigated with regression analysis adjusted for age, gender and education in combined patient groups. We assessed possible interaction versus addition effects of these relationships with visually rated MTA (Scheltens scale). Results:The highly statistical significant relationship between WML and memory function was no longer significant when MTA was taken into account. However, the strong significant relationship between WML and non-memory function remained significant after adjustment for MTA, but the explained variance attributed to WML was only 1.3%. There was no interaction between WML and MTA on CAMCOG test scores. In addition, shown by a 2×2 factorial model by presence versus absence of WML and MTA, WML affected non-memory function only in the presence of MTA. Conclusion:Our data suggest that presence of WML is associated with lower non-memory cognitive function but this effect is conditional on the presence of pre-existing MTA. The very small explained variance suggests little impact of WML to the clinical profile of a memory clinic patient.
