Professor Peter Nordstrom, of Umea University, Umea, Sweden, and colleagues reviewed records on 7,073 Alzheimer's patients taking cholinesterase inhibitors (ChEIs), such as donepezil, rivastigmine and galantamine, which are used for treating mild to moderate Alzheimer's disease.
Side-effects of ChEIs include a beneficial effect on the vagus nerve, which controls the rate at which the heart beats, and some experimental studies have suggested that ChEIs could also have anti-inflammatory properties.
They found that those who were on ChEIs had a 36 percent reduced risk of death from any cause, a 38 percent reduced risk of a myocardial infarction (heart attack) and a 26 percent reduced risk of death from cardiovascular causes such as stroke compared to people not taking ChEIs.
These results included adjustments for various confounding factors such as age, sex, whether the diagnosis was for Alzheimer's dementia or Alzheimer's mixed dementia (where more than one type of dementia occur simultaneously), level of care, and medical history including medications for other conditions.
Prof Nordstrom said: "If you translate these reductions in risk into absolute figures, it means that for every 100,000 people with Alzheimer's disease, there would be 180 fewer heart attacks - 295 as opposed to 475 - and 1125 fewer deaths from all causes - 2000 versus 3125 - every year among those taking ChEIs compared to those not using them."
Patients taking the highest recommended doses of ChEIs had the lowest risk of heart attack or death: 65 percent and 46 percent lower respectively compared with those who had never used ChEIs.
The researchers also checked whether the reduction in risk applied only to the use of ChEIs or was seen in other drug treatments for dementia. Memantine is a drug indicated for use in moderate to advanced Alzheimer's disease and works in a different way to ChEIs. The researchers found it made no difference to the risk of heart attack or death from any cause.
The study was recently published in the European Heart Journal.