NPC Archive Item: Cholinesterase inhibitors don’t help in mild cognitive impairment

NOTE – This is an archive post from the NPC and has not been updated since first publication. Therefore, some hyperlinks may no longer be working.
MeReC Rapid Review

NPC Logo

Raschetti R, et al.  Cholinesterase inhibitors in mild cognitive impairment: a systematic review of randomised trials. PLoS Medicine 2007; 4(11): e338. doi:10.1371/journal.pmed.0040338

What is the background to this?

Dementia is a distressing and disabling progressive and largely irreversible clinical syndrome that is characterized by a widespread impairment of mental function. The term mild cognitive impairment (MCI) describes patterns of cognitive decline which are greater than the mild and variable cognitive decline associated with normal ageing, but which fall short of standard definitions of dementia. However, widely accepted and validated diagnostic criteria for MCI do not exist.

Estimates of conversion rates between MCI and dementia vary widely because of differences in case definition, sampling size, length of follow-up (and the number of people lost to follow-up), etc in studies.  One high quality review states that more than 50% of people with MCI later develop dementia.

Given the nature of dementia, with its burden for sufferers, carers and care systems, an effective, low risk, economic way of reducing the risk of conversion from MCI to dementia would be very attractive. The cholinesterase inhibitors (ChEIs) donepezil, galantamine and rivastigmine are used in dementia (see below) and some people have suggested they might have a preventative role in people with MCI.

What does this systematic review claim?

The authors conclude that the use of ChEIs in MCI was not associated with any delay in the onset of dementia. In only two studies (for rivastigmine and donepezil) was it possible to derive point estimates of the relative risk of conversion: 0.85 (95% confidence interval [CI] 0.64 to 1.12) and 0.84 (95%CI 0.57 to 1.25) respectively.  These did not reach conventional levels of statistical significance. With regard to other, surrogate, outcome measures, after adjusting for multiple comparisons a statistically significant difference emerged only for differences in rate of brain atrophy (with galantamine).  The clinical significance of this is not clear.

Moreover, the safety profile showed that the risks associated with ChEIs are not negligible.  In all but one of the eight studies examined, fewer subjects in the treatment groups than in the control groups completed the studies and the rate of discontinuation due to side effects was consistently greater among subjects given active drug.

How was the systematic review conducted?

The authors searched electronic databases and clinical trial registers to identify possible randomized controlled trials (RCTs) to include.  Only eight RCTs of ChEIs in MCI were located (of which only three were published in peer-reviewed journals).  The quality of the reporting of the trials was low and a major limitation was the heterogenous nature of the subjects included.

How does this relate to national guidance?

NICE issued clinical guideline 42, which advised on health and social care and support for people with dementia and their carers in 2006, jointly with the Social Care Institute for Excellence (SCIE).  This incorporated a NICE technology appraisal issued that year relating to the ChEIs and memantine.  (This advice was challenged by judicial review, but was substantially upheld. NICE re-issued an amended version of the  guidance (TA111) in September 2007).

NICE guidance is that the ChEIs should be considered only for people with moderate Alzheimer’s disease, ie with a Mini-Mental State Examination (MMSE) score between 20 and 10 (the higher the score, the less severe the disease).  They are not recommended in other types of dementia or in MCI outside properly constructed clinical trials.

So what?

This systematic review indicates that, at present, ChEIs are not appropriate for use as a preventive strategy to reduce the risk of MCI conversion to dementia.  They are not likely to be effective, harmful effects on progression to dementia cannot be excluded, and the risk of side effects (sufficiently unpleasant to lead to withdrawal of motivated subjects from clinical trials) outweighs any possible effects on surrogate outcomes. Its conclusions bear out NICE guidance.

A section devoted to dementia and its management is open on NPC


Healthcare professionals should follow the NICE clinical guideline, and resist requests to prescribe ChEIs to people with MCI (unless as part of a clinical trial).  However, they should consider referring people who show signs of MCI for assessment by memory assessment services, as recommended in the NICE guideline.


Please comment on this blog in the NPC discussion rooms, or using our feedback form.