Using data from the Alzheimer’s Disease Neuroimaging Initiative (ADNI), Robert Bartha and colleagues at the Robarts Research Institute in London, Ontario, Canada, have evaluated the possibility that the rate of ventricular enlargement over six months could be a useful marker for Alzheimer disease progression. By comparing change in ventricle size with changes in two common psychometric tests, they found that absolute changes in ventricle size provided a more sensitive measure of disease progression than cognitive changes. Their results, published online in Brain on July 11, suggest that rates of ventricular enlargement, determined by serial MRI, could serve as a feasible short-term marker in multicenter clinical trials for AD. A second study on brain size, this one from Jeffrey Burns and colleagues at the University of Kansas in Kansas City, offers a possible modifier of brain tissue loss by showing that fitness level correlates with brain size in patients with early AD.

Researchers in the ADNI study, a multicenter, longitudinal imaging and biomarker discovery program, have accumulated thousands of MRI measures from hundreds of people with AD, mild cognitive impairment (MCI), and normal elderly controls. In the current study, first author Sean Nestor and colleagues applied an automated method of determining ventricular volume to ADNI MRI data from 504 subjects who had both baseline and six-month follow-up scans. Nestor and coworkers found that at baseline, absolute ventricular volume was highest in AD patients, and that in the MCI group it was also significantly elevated over controls. All of the groups showed further enlargement over six months, with the rate being greatest in the patients with AD, and also significantly elevated in the MCI group compared to normal controls. Among AD patients, the test was sensitive to ApoE allele status, where people with ApoE4 showed significantly more ventricular enlargement than those without.

In the MCI group, the change in ventricular volume tracked with cognitive changes and disease progression as measured by the Mini-Mental State Exam (MMSE), the ADAS-cog battery, and conversion to AD. For people with MCI, an increase in ventricular volume correlated with decline in MMSE score and increase in ADAS-cog. People with MCI who progressed to AD during the six-month follow-up showed twice the rate of ventricle enlargement than people with stable MCI.

Among anatomical measures, loss of hippocampal volume has been shown to predict the progression of AD (see ARF related news story). Ventricular enlargement is the flip side of brain shrinkage, and the authors speculate that their measure is so sensitive, in part, because it captures the loss of hippocampal volume. However, measuring ventricle volume has some advantages over determining hippocampal size, the authors write. The calculation of ventricular volume is easily automated because ventricle boundaries show up as a sharp line of contrast between tissue and cerebrospinal fluid in MRI. This automation may contribute to the high reproducibility of the measure even in a multicenter design.

The results suggest that changes in ventricular volume measured over time could provide a more sensitive read of the clinical progression of AD than the currently accepted endpoints of cognitive testing. By the authors’ calculations, the ability to measure significant (20 percent) changes in ventricle size in a clinical trial would require hundreds of subjects, versus thousands for the MMSE or ADAS-cog endpoints. “As a potential measure of disease progression for multicenter studies of both AD and MCI subjects, ventricular enlargement measures would significantly reduce the number of subjects required to demonstrate a change from the natural history of Alzheimer’s disease progression,” the authors conclude.

One finding of the Nestor study is that there are large variations in the rate of ventricular enlargement within the AD or MCI groups. That suggests a range of responses to AD pathology among individuals, which could be affected by environmental factors. The Burns study, which appeared in the July 15 issue of Neurology, asks whether fitness level affects brain changes seen early in AD. The researchers measured the fitness of 57 early-stage AD patients, and a similar number of non-demented subjects, during exercise on a treadmill, and determined whole brain volume by MRI. Their results indicate that subjects with AD had a modestly but significantly reduced maximum oxygen consumption (a measure of cardiorespiratory fitness) compared to non-demented subjects. Fitness level correlated with brain volume in the AD group, so that people with AD who were in better shape had more brain tissue. There was no such relationship in people without dementia, and there was no correlation between brain volume and cognitive measures after adjustment for age.

From the data, the authors conclude, “Increased cardiorespiratory fitness is associated with reduced brain atrophy in Alzheimer disease.” That does not mean keeping fit staves off brain loss, they caution. The association could just as likely result from a situation where loss of fitness accompanies AD, or where some underlying process affects both fitness and AD progression. Studies in mice support the appealing idea that fitness can moderate the attack of AD on the brain—exercise reduces amyloid pathology in mouse models of AD (see ARF related news story and Adlard et al., 2005) and stimulates learning and neurogenesis in normal adult mice (see ARF related news story). However, moving from association to causation in humans will require further studies.—Pat McCaffrey


  1. This study by Burns and colleagues adds optimistically to the growing body of evidence supporting that aerobic-based physical activities, which improve cardiorespiratory fitness, are beneficial for cognitive function. There are certainly numerous possible reasons why exercise training improves cognitive function.

    The results showed that participants who were more physically fit had less brain shrinkage than less-fit participants. However, they did not do significantly better on tests of mental performance. Although the authors were surprised by the small effect of exercise on cognitive performance, this should not be taken as unusual. There are several reasons why there was no significant effect between fitness and cognitive performance. For example: 1) the tests used were not sensitive enough to capture the subtle changes of exercise on cognitive function; 2) the study results were based on participants who were evaluated only once rather than repeatedly over time; and 3) the study sample was too small to show a statistically significant effect.

    Even though the study presents several limitations, these results are of clinical and public health relevance since the brain shrinks with normal aging, and that rate is doubled in people with Alzheimer's. A recent review by the Cochrane Library concluded that the same aerobic exercise that is good for your heart may also have a positive effect on cognitive function—specifically, motor function, auditory attention, and memory—in healthy older adults (Angevaren et al., 2008).

    Hosts of mechanisms are thought to be responsible for this activity-induced cognitive change. Animal data suggest that running can lead to an increase in new brain cells in the hippocampus, an area that plays a large role in learning and memory. Also, "cardiorespiratory trained" transgenic mice that model Alzheimer's showed decreased amyloid protein buildup in the rodents' brain (Adlard, 2005). Although researchers cannot count brain cells in studies of live humans, a study by Colcombe and colleagues (2003) showed that aerobically fit adults had reduced white and grey matter tissue loss as compared to a match sedentary group. Many important neurochemicals are influenced by exercise, including neurotransmitters and growth factors, which are being investigated for their role in cognition, and brain function. Even runner's high, that elusive euphoria that some people experience after prolonged or intensive running, is becoming clearer—literally. A study done in Germany (Boecker et al., 2008) used PET scans to look at the brains of 10 athletes following a two-hour run. The scans confirmed that during the run, endorphins were released in certain parts of the brain known to be involved with the processing of information and emotions.

    The study by Burns and colleagues adds to the literature and stresses the need to further these results into the design of rigorous randomized trials to explore whether exercise and physical fitness can slow the progression of Alzheimer’s. In addition, we should start to investigate different exercise training paradigms, specifically the dose-response association between exercise and cognition since the question that needs to be answered is, How much exercise is needed to get these wonderful benefits?

    Though experts do not arrive at a consensus about the right exercise protocol, recommendations, and guidelines for exercise training practices for brain health, considering overall health, any physical activity is better than nothing. When I reviewed several clinical trials dedicated to test the effects of exercise on older adults with cognitive impairments (Heyn et al., 2004), the majority of the protocols included 30 to 60 minutes of exercise sessions performed three to five times per week. The exercise protocols were based on performing movements requiring a large range of motions that use large muscle groups (such as walking, running, and stationary bicycle). Some studies support that some mental health benefits can be observed after 20 minutes of physical activity, though the more exercise and higher intensity, the better the effects. This means that following the Center of Disease Control/American College of Sports Medicine recommendations of 30 minutes of aerobic activity per day will have a positive effect on your brain as well as your heart.


    . Voluntary exercise decreases amyloid load in a transgenic model of Alzheimer's disease. J Neurosci. 2005 Apr 27;25(17):4217-21. PubMed.

    . The runner's high: opioidergic mechanisms in the human brain. Cereb Cortex. 2008 Nov;18(11):2523-31. PubMed.

    . The effects of exercise training on elderly persons with cognitive impairment and dementia: a meta-analysis. Arch Phys Med Rehabil. 2004 Oct;85(10):1694-704. PubMed.

    . Physical activity and enhanced fitness to improve cognitive function in older people without known cognitive impairment. Cochrane Database Syst Rev. 2008;(3):CD005381. PubMed.

    View all comments by Patricia C. Heyn

Make a Comment

To make a comment you must login or register.


News Citations

  1. Biomarker Roundup: Collecting Clues from MRIs to RNAs
  2. Sorrento: More Fun, Less Amyloid for Transgenic Mice
  3. Exercise Helps Mouse Elders Learn, Generate New Neurons

Paper Citations

  1. . Voluntary exercise decreases amyloid load in a transgenic model of Alzheimer's disease. J Neurosci. 2005 Apr 27;25(17):4217-21. PubMed.

Further Reading


  1. . Volumetric MRI and cognitive measures in Alzheimer disease : comparison of markers of progression. J Neurol. 2008 Apr;255(4):567-74. PubMed.

Primary Papers

  1. . Ventricular enlargement as a possible measure of Alzheimer's disease progression validated using the Alzheimer's disease neuroimaging initiative database. Brain. 2008 Sep;131(Pt 9):2443-54. PubMed.
  2. . Cardiorespiratory fitness and brain atrophy in early Alzheimer disease. Neurology. 2008 Jul 15;71(3):210-6. PubMed.