Conversion from subtypes of mild cognitive impairment to Alzheimer dementia
Fischer P, Jungwirth S, Zehetmayer S, Weissgram S, Hoenigschnabl S, Gelpi E, Krampla W, et al.;
Commented by , 22 Feb 2007
Aim of the study
To compare the rate of conversion to Alzheimer's disease (AD) and other types of dementia between subtypes of mild cognitive impairment (MCI).
Method
Community-based birth (born between May 1925 and June 1926) cohort living in Vienna, Austria, investigated at age 75 and followed up after 30 months (Vienna Trans-Danube Aging Study). Neuropsychological testing was done using CERAD battery and the Trail Making Test B.
The diagnosis of probable or possible AD as well as of other specific dementias was made using research criteria and in consensus conferences using all available information, including cerebral MRI.
Results
From 581 nondemented subjects who completed the neuropsychological examination at baseline, 476 could be reassessed after 30 months and 35 had died. Out of the 141 at baseline with MCI, 41 progressed to AD. Amnestic MCI carried a risk of conversion to AD of 48.7% (CI: 32.4 to 65.2) over the 30 months (annual conversion rate of 19%), and nonamnestic MCI of 26.8% (CI: 9.4 to 16.3). Out of the 390 cognitively normal at baseline 12.6% (CI: 9.4 to 16.3) developed AD.
The comparaison of frequencies of conversion to AD for the three subgroups (cognitively normal, amnestic MCI and nonamnestic MCI) was significant (Χ2 = 38.742, df = 2). One subject with amnestic MCI developed vascular dementia (VaD) and two developed Dementia with Lewy Bodies (DLB). 21.5% of subjects with MCI reverted back to normal cognition.
Professor Gauthier's comments
It is already established that MCI as a syndrome carries a higher risk of progression to dementia, particularly in carriers of the apoE4 genotype (ref. 1). What is yet unresolved is whether the subtype (amnestic single domain, amnestic plus other domains, non-amnestic) is a prodrome to different types of dementia. This differentiation is based on the neuropsychological profile rather than clinical or biological characteristics.
It has been suggested that non-amnestic MCI may be a relatively specific prodrome of non-AD dementias such as Fronto-temporal Dementia (FTD), DLB or VaD. If true, the amnestic MCI should be the sole target for very early AD (or pre-dementia AD) therapeutic research and eventually treatment with potential disease-modifying drugs such as tramiprosate, flurbiprofen or various amyloid monoclonal antibodies.
This study demonstrates that nonamnestic MCI can also lead to AD, and that amnestic MCI can lead to non-AD dementias. We may thus need to refine our classification of prodromal dementia states to include biological measures such as hippocampal volume using MRI and/or CSF biomarkers such as β-amyloid fragments and tau.
Finally, it should be noted that 21.5% of subjects with MCI reverted back to normal cognition, and there should be caution in explaining what "MCI" means to individual subjects: it is a risk state, not a specific diagnosis. We should also refrain from using cholinesterase inhibitors which have not proven to have adequate safety/efficacy ratio in this population.
References
1. Gauthier S, Reisberg B, Zaudig M, Petersen RC, Ritchie K, Broich K, et al. Mild cognitive impairment. The Lancet 2006; 367; 1262-1270