Donepezil for mild cognitive impairment

Problems with memory which do not meet the diagnostic criteria for dementia, usually called mild cognitive impairment (MCI), can be the first sign of an impending dementia, particularly Alzheimer's disease (AD). There is no consensus on a definition or diagnostic criteria for MCI, and MCI remains a vague term and those so described are a heterogeneous population, consisting of people who may rapidly progress to dementia but also of people with stable cognitive deficits and some who may actually improve. Treatment in the very earliest stages of AD may delay progression to AD. Donepezil (Aricept, E2020), a cholinesterase inhibitor, has been shown to benefit all severities of AD including mild and it would be reasonable to investigate its efficacy for those with MCI.

To assess the effects of donepezil in people with mild cognitive impairment but no diagnosis of dementia.

The trials were identified from a search of the Specialized Register of the Cochrane Dementia and Cognitive Improvement Group on 6 January 2006. This register contains records from major health care databases like CENTRAL, MEDLINE, EMBASE, CINAHL and PsycINFO and many ongoing trial databases and is updated regularly.

All double blind, randomized trials in which treatment with donepezil was compared with placebo for patients with mild cognitive impairment.

Data were extracted from the published reports of the included studies, pooled where appropriate and the treatment effects or the risks and benefits estimated.

The two included studies, with a total of 782 patients, all with a MMSE greater than 23 points, identified similar patients for inclusion, but were quite different with respect to design and objective. Pooling results in a meta-analysis was not possible. In the first study the 13-item ADAS-Cog showed benefit associated with 10 mg/day donepezil compared with placebo at 24 weeks (MD 1.90, 95% CI 0.51 to 3.29, p=0.007), but four other measures of cognitive function did not. The analysis of withdrawals before the end of treatment at 24 weeks, withdrawals due to an adverse event, and numbers experiencing an adverse event, showed a significant difference between the donepezil group and the placebo group in favour of placebo, (43/133 donepezil 23/137 placebo, OR 2.37, 95% CI 1.33 to 4.22, p=0.003), (29/133 donepezil 10/137 placebo, OR 3.54, 95% CI 1.65 to 7.60, p=0.001), (116/133 donepezil, 100/137 placebo, OR 2.52 95% CI 1.34 to 4.76, p=0.004). Various adverse effects were recorded, and several types of event, diarrhoea, nausea, vomiting, leg cramps and abnormal dreams, were reported more frequently in the donepezil group compared with the placebo. In the second study there was a significant difference between the number of patients diagnosed with AD or another dementia between the donepezil group and the placebo group in favour of donepezil after one year of treatment (16/253 donepezil 38/259 placebo) (OR 0.39, 95% CI 0.21 to 0.72, p=0.003), but no difference after 3 years of treatment (63/253 donepezil 73/259 placebo) (OR 0.84, 95% CI 0.57 to 1.25, p=0.4).

There are two included studies. One study demonstrated a modest treatment effect in cognitive function as assessed by ADAS-Cog13 but not for other outcomes assessing different domains of cognitive function. Donepezil was associated with significantly more adverse effects compared with placebo, mostly gastrointestinal. From the second study, there is no evidence that donepezil delays the onset of AD. There is no evidence to support the use of donepezil for patients with MCI. The putative benefits are minor, short lived and associated with significant side effects.