lifestyle interventions and cognition

there have been several recent articles on interventions to prevent cognitive decline, including 

1. an op-ed piece in the NY Times (see http://www.nytimes.com/2015/10/25/opinion/sunday/can-you-get-smarter.html?emc=edit_th_20151025&nl=todaysheadlines&nlid=67866768&_r=0​ ) . a few of their points, expanded a bit with the references:

-- a 6-week on-line study with 11,430 participants trained several times a week on cognitive tasks: there were 3 groups--one with 6 training tasks emphasizing reasoning, planning and problem-solving; another group was exposed to a broader range of cognitive tasks including tests of short-term memory, attention, visuospatial processing and mathematics, as is found in commercially available brain training devices; and a control group who did not do specific cognitive tasks but answered obscure questions using online resources (ie, like Google). They found that each of these tasks improved through this training, but "no evidence was found for transfer effects to untrained tasks, even when those tasks were cognitively closely related" -- ie, there is no evidence that regular use of computerized specific cognitive tasks improves general cognitive function (see Nature. 2010 June 10; 465: 775)

--exercise does seem to help:

--mice who have regular exercise have more neurons in their hippocampus (that is the site where new memories are formed and converted to long-term memories, and one of the first sites affected by Alzheimer Disease (AD)

--a small study of 86 women with subjective memory complaints were assigned to 3 groups: resistance training, aerobic training, or balance/tone (the control group). those in the aerobic training group remembered more items; both aerobic and resistance training led to better spatial memory performance; and there was a significant correlation between spatial memory performance and overall physical capacity (see doi: 10.1155/2013/861893)

--a study of 155 community-dwelling women 65-75 yo were randomized to resistance training vs balance/tone (control group) and found that resistance training reduced the progression of brain MRI white matter lesions (which are associated with cognitive impairment, and are markers of cerebral small-vessel disease -- see J Am Geriatr Soc 2015; 63(10): 2052)

--they note that BNDF levels (brain-derived neurotrophic factor, which is released in response to neuronal activity) are associated with brain size and function, and is increased with exercise (more below).

2. a JAMA perspective on mitigating cognitive decline (see cognitive decline social issuesJAMA2015 or doi:10.1001/jama.2015.15390​)

--in assessing >1200 brains in 2 different aging studies, there was not a great correlation between anatomic findings and cognitive health:

    --the Rush Memory and Aging Project  [Curr Alzheimer Res. 2012; 9(6): 646] of 1556 elders without dementia enrolled from retirement communities beginning in 1997, tracked cognitive function, relying on clinical diagnoses and an array of 21 cognition tests, found that:

--physical activity was associated with cognition

--incident Alzheimer dementia (AD) was associated with a decline in motor function

--social engagement (social activity and support) was associated with global cognition and a slower rate of cognitive and motor decline and disability

--but at autopsy, the correlation between anatomic changes and cognition were not great: 90% of those meeting clinical criteria of AD did meet pathologic criteria for AD, but 1/2 of those with mild cognitive impairment and 1/3 of those without cognitive impairment also met pathologic criteria for AD.

    --the Religious Orders Study [Curr Alzheimer Res. 2012; 9(6): 628], a study from 1994-2011 of 1162 Catholic nuns, priests, and brothers from 40 groups across the US initially without dementia, similarly used a battery of 21 cognitive tests, also found the same relatively poor correlation as the Rush study between anatomic pathology and AD.

    ​--and, overall overt brain pathology accounted for only 1/2 of the cognitive decline documented

--Animal studies support the role of BDNF in improving neuronal survival and function (especially in the hippocampus and cortex), and improve synaptic plasticity and long-term memory. Human epidemiologic studies find a correlation between low BDNF levels and AD (though unclear which came first). In this light the community-based Framingham Study was reviewed to see if higher BDNF levels in cognitively healthy adults seemed to protect against future development of AD, found that in 10 years of followup, 140 people developed dementia (117 with AD), and each SD increase in BDNF levels was associated with a 33% decreased risk of dementia overall and AD in particular. Comparing the top to bottom quintile of BDNF levels, there was a 51% decrease in risk of dementia (HR 0.49, p=0.01) and AD (HR .46, p=0.02),  specifically in subgroups of women, people >80 yo and those with college degrees (see cognitive function framingham study jamaneurol2014 in dropbox, or JAMA Neurol. 2014; 71(1): 55)

--depression is associated with lower BNDF levels and more hippocampal shrinkage, and antidepressants block the depression-induced drop in BDNF.

--and what increases BDNF levels??? increased physical activity, reduced caloric intake, social support (and lack thereof is associated with lower BDNF levels). A review found that aerobic exercise increases BDNF as well as improves hippocampal atrophy, improves memory function, and reduces depression [see Neuroscientist 2012: 18(1): 82].  by the way, there are also data showing exercise is associated with decreases in b-amyloid in a transgenic Alzheimer mouse model.

--Increased BDNF also reduces risk of stroke (i could not find this reference to review the primary data).

--also there are data finding that blood VEGF levels (vascular endothelial growth factor) may be protective of brain function. for example, a study found that VEGF levels in the CSF were higher in those with higher cognitive abilities, in comparing those with normalcognition to mild cognitive impairment to those with AD. and, exercise boosts VEGFlevels... ​

3. a more recent article undercut the exercise/dementia link: the LIFE trial (Lifestyle Interventions and Independence for Elders) looked at the effects of a 24-month physical activity program for 1635 community-living people in the US, aged 70-89, who were at risk for mobility disability but able to walk 400 meters (see cognitive decline physical activity JAMA2015 in dropbox, or JAMA. 2015;314(8):781-790). they assessed cognitive function as a secondary outcome (though this study was not specifically powered for this outcome) using several instruments, especially the Digit Symbol Coding subtest of Wechsler Adult Intelligance Scale and the revised Hopkiins Verbal Learning Test (12-item word list recall), comparing those in the structured moderate-intensity physical activity program of walking, resistance training, and flexibility exercises, to those in a health education program. They found no overall benefit to exercise, except in those >80 yoand in those with poorer baseline physical performance, noting an improvement in executive function in these groups (p=0.01). However, they did not measure BDNFlevels, which may be particularly important in a group of patients who were physically challenged at baseline and perhaps unable to exercise very much. also, only 57 patients overall developed dementia though 132 developed mild cognitive impairment (this LIFE was a well-educated cohort, and therefore more likely to show less cognitive decline by the cognitive tests). 

so, things are usually more complicated than they seem. we are all (myself certainly included) prone to look for simple solutions that make sense physiologically. it makes sense that AD and dementia in general are neurologic degenerative disorders associated with specific anatomic pathology, but the actual correlation is not so clear. it also makes sense that keeping the brain active, through suduko or other cognitive puzzles would help preserve brain function, but it seems that overall brain function does not improve (though performance of the specific task, suduko in this case, does get better....).  however, the above studies suggest that there do seem to be a large array of potential benefits from exercise, including decreased depression, improved memory, decreased progression of cognitive decline. and the data also support the benefit of a Mediterranean diet. bottom line: i think it is not so useful to try to break down the specific components of diet or exercise which is beneficial (a tad reductionist), that cognitive decline is undoubtedly multifactorial (again arguing against a reductionist approach), but that nutrition and exercise are very likely necessary for brain health and likely more so than the too-often-advertized easy fixes of on-line memory games, etc. And this realization is a bit of a game-changer, in that perhaps our primary therapeutic focus in elderly people concerned about cognitive decline really should be on a healthy diet and exercise, which not only are good for the body but also for the mind.

for some other blogs on cognitive function, see:

http://gmodestmedblogs.blogspot.com/2014/07/chocolate-and-cognitive-function-etc.html/ showing that chocolae improves cognitive function

http://gmodestmedblogs.blogspot.com/2015/11/mediterranean-diet-and-brain-volume.html  reviewed the Mediterranean diet and cogntive decline, showing decreased AD and also larger brain volumes, which tracked with the degree of adherence to a Mediterranean diet.