I'm gearing up for a talk later this month for the Massachusetts Psychiatric Society Geriatric Interest Group, which means I'm trying to keep abreast of all the latest diet and dementia studies. A couple new big ones came out, both from groups with which we are familiar. The first is from the Nurse's Health Study which purports that eating berries can protect us from cognitive decline via the magical power of flavinoids. The second paper is from the same group who linked consumption of organ meats with dementia in northern Manhattan in one of those dreaded dietary pattern studies.
They have similar designs. Follow a large group of people without dementia at baseline for a number of years, tracking demographic and dietary information along the way. Track new diagnoses of dementia or test people with basic cognitive screens on a regular basis. In addition, the members of the Washington Heights/Hamilton Heights Columbia Aging Project (WHICAP) had plasma beta amyloid peptide measured. It is thought that blood levels of beta amyloid correlate with brain levels, and other studies have shown that higher beta amyloid in the plasma correlates with the onset of dementia.
In the WHICAP crew, those who had higher intake of omega 3 fatty acids had lower levels of plasma beta amyloid, even after adjusting for confounders. This finding would make biologic sense, as omega 3 fatty acids in lipid rafts help cleave amyloid precursor protein (APP) into harmless bits, whereas arachidonic acid in the same place allows for APP to be made into beta amyloid, the component of those plaques that build up in the brain. However, the dietary sources of omega 3 PUFAs are listed as "salad dressing, fish, poultry, margarine, and nuts," which, excepting the fish, are generally terrific dietary sources of omega 6 PUFA, not omega 3. To get an idea of the confounders, "Participants with higher plasma levels of [beta amyloid] peptide were older and less educated and had lower intakes of omega 3 PUFA, omega 6 PUFA, and MUFA…" and "Participants with higher omega 3 PUFA, higher omega 6 PUFA, or higher MUFA intakes had a higher education, were more likely to be white or black and less likely to be Hispanic."
The researchers tried to adjust for subtype of omega 3 PUFA (bless their hearts, considering they were working from food frequency questionnaires) and "none of the subtypes of omega 3 PUFA was significantly associated with the level of beta amyloid, suggesting that overall intake of omega 3 PUFAs might play a more important role." (Or maybe it all just suggests that people who eat "healthy" in general eat fairly healthy who are also better educated and wealthier have a lower risk of dementia and less inflammation on board, as plasma beta amyloid is linked with inflammation and oxidative stress.)
In rodents models, a bit of omega 3 in the chow can reduce plasma beta amyloid by a whopping 70% in a matter of weeks compared to "low DHA control chow" and plaque burden in the brain is "reduced by 40.3%." In human experiments (only one of which measured plasma beta amyloid and have been small), there hasn't been much benefit to adding omega 3s in folks with dementia, though those with mild cognitve impairment might be helped. I'm guessing that it is difficult for a human to be quite as depleted of omega 3 and topped off with omega 6 as much as those lab rats with their shudder-worthy processed rat chow.
In the Nurses Health Study, blueberries and strawberries take center stage. Nurses who admit to eating buckets of them (chock full of the antioxidant anthocyanidin) had better cognitive scores as the study went forward over the decades. And while according to Walter Willett the confounders are easily accounted for, you may not be surprised to know that the berry eating nurses were wealthier, more likely to exercise, more likely to eat fish, and more likely to eat more calories period.
But, in the interest of some discussion, yes, bioflavinoids are very sexy. In the Nurses Health Study, the riches sources were strawberries and blueberries, but tea, oranges, and apples are also common sources. The anthocyanidins in the berries, in particular, are known to be able to cross the blood brain barrier and work in the hippocampus, a central brain area of learning and memory. Not only are they antioxidants, but might also directly deactivate cellular inflammatory mechanisms.
In short, I concur that fish and berries are good for you. However, these studies were not designed to prove that by any means, but rather to point out some interesting population trends as related to dementia and cognitive decline. Consider them pointed out.
I'm a bit sick of these sorts of studies, to be honest.