Exercising for half an hour, five times a week could help slow down the effects of Alzheimer's disease.
People with signs of Alzheimer's disease who exercised regularly had a slower progression of the condition compared to those who did not engage in physical activity.
A team of researchers looked at individuals who had an accumulation of a protein called amyloid-beta in the brain - a hallmark feature of the illness.
Although exercise did not slow down the accumulation of the protein, the findings suggest aerobic workouts can at least slow down the effects of the disease, if intervention happens in the early stages.
The study which was conducted by researchers at UT (University Of Texas) Southwestern Medical Centre, compared cognitive function and brain volume between two groups of sedentary older adults with memory issues. One group did aerobic exercise (at least a half-hour workout four to five times weekly), and another group did only flexibility training.
Both groups maintained similar cognitive abilities during the trial in areas such as memory and problem-solving. However, brain imaging showed that people from the aerobic exercise group experienced slightly less volume reduction in their hippocampus - a memory-related brain region that progressively deteriorates as dementia takes hold.
Around 850,000 people in the UK suffer from dementia, and most have Alzheimer's disease. Despite many trials, scientists are yet to find a drug that can prevent, cure or delay the progression of the condition. But the new research suggests regular exercise could help.
"What are you supposed to do if you have amyloid clumping together in the brain? Right now doctors can't prescribe anything," said Dr Rong Zhang, who led the study. "If these findings can be replicated in a larger trial, then maybe one day doctors will be telling high-risk patients to start an exercise plan. In fact, there's no harm in doing so now."
However, Zhang notes that more research is needed.
"I’m excited about the results, but only to a certain degree," Zhang said. "This is a proof-of-concept study, and we can't yet draw definitive conclusions."
The study is published in the Journal of Alzheimer's Disease.