Regular exercise increases cardiorespiratory endurance, tolerance to physical exertion, high-density lipoprotein cholesterol levels, and insulin sensitivity while decreasing adiposity, blood pressure levels, triglyceride levels, and inflammatory markers.1 All of this is probably common knowledge, but it also seems that consistent exercise may reduce the risk of Alzheimer disease (AD) and other types of dementia. This conclusion rests on evidence accumulating from 3 perspectives: observational studies, randomized clinical trials, and animal studies.
Observational Studies
Typical of observational studies is a recent report about members of the Group Health Cooperative (GHC) in Seattle.2 The 1740 study participants were older than 65 years, had no cognitive impairment, and were monitored for a mean of 6.2 years. During follow-up, dementia developed in 158 participants (AD developed in 107 of these participants). The incidence of dementia was 13.0 per 1000 person-years for participants who self-reported exercising 3 or more times per week compared with an incidence of 19.7 per 1000 person-years for those who self-reported exercising fewer than 3 times per week. The hazard ratio for participants who exercised at least 3 times a week was 0.68, or a 32% reduction in risk of dementia. The interaction of regular exercise and AD was similar, with a hazard ratio of 0.69. This study excluded GHC members older than 65 years without cognitive impairment who scored below the 25th percentile on the initial Cognitive Ability Screening Instrument to minimize the possibility that the prodromal phase of dementia might be associated with a decrease in exercise.
Randomized Clinical Trials
A number of clinical trials investigating the effect of regular exercise on the development of dementia can be found in the literature. In most of these trials, relatively sedentary persons older than 60 years are randomly assigned to an exercise group (usually involving aerobic exercises, such as walking, swimming, or bicycling) and are compared with a control group (offering far less exercise—eg, toning or stretching). Participants typically exercise an hour a day several days a week and may continue to do so for months or years. Cognition is examined before and after the exercise and control interventions.
A meta-analysis of 18 such intervention studies published between 1966 and 2001 showed that fitness training had robust benefits for cognition.3 Participants in combined strength- and aerobic-training regimens improved to a reliably greater degree than those following an aerobic-training regimen alone. Exercise resulted in beneficial effects on cognition only when it was continued for more than 30 minutes per session. Programs that continued long-term (more than 6 months) were more effective than those lasting 1 to 3 months or 4 to 6 months. The clinical literature does not explain why exercise improves cognition. For example, a recent meta-analysis confirmed the beneficial effect of exercise, but concluded that meta-regression techniques did not show a reliable relationship between aerobic fitness and cognitive performance.4
So what is the mechanism by which exercise helps the brain? If it is not just better cardiopulmonary function (and presumably better oxygenation of the brain), what is so helpful about exercise? The recent results of brain imaging studies are beginning to suggest some answers. In a 2-part study, MRI measurements of cerebral blood volume in humans showed that exercise differentially expands the dentate gyrus, a hippocampal subregion important for memory known to be adversely affected by aging.5 Using the same MRI techniques on mice and conducting a postmortem analysis, the researchers found that exercise correlates with neurogenesis in the dentate gyrus of adult mice.
Animal Studies
Animal models are increasingly important in the study of dementia. Rodents, in particular, are used because they can be cognitively assessed (ie, in the Morris Water Maze, in which maze cues are used to determine the location of a submerged platform), exercised or not, genetically altered (eg, special transgenic mice that produce β-amyloid plaques similar to those seen in persons with AD), and eventually examined postmortem for a better understanding of brain mechanisms.6,7
