EXPERT REACTION: Ketogenic diet helps mice live longer - but can also make them obese

Dietary experiments in mice differ from human studies e.g. high fat feeding has different effects in male and female B6 mice so studies usually report only male mice (as often there is no effect in females). 

The high fat diet is purely  experimental as, unlike humans, mice don't eat much fat naturally.

 So these studies of extreme diet change e.g.  ketosis in a male mouse eating excess fat without carbohydrate may invoke "survival" responses which may not occur in a human (especially a female). While the responses raise interesting questions regarding a beneficial effect   of stimulating fat burning while avoiding excess energy intake  which  usually occurs in high fat diet in a mouse by alternating the diet.

In summary, there is much more experimentation to do to test if the beneficial effects of such diet manipulation are applicable in  humans.

Two independent mouse studies published in the same issue of Cell Metabolism conclude that elevated ketones improve longevity and memory in rodents. In both studies the elevated ketones are produced by replacing the carbohydrate content of the diet with fat.

When managing human obesity, it would be inadvisable to prescribe a diet that is high in fat as it could worsen lipids. In any case it is possible to induce ketosis by simply avoiding carbohydrates without an increase in fat intake.

This is because the patients have adequate reserves of fat to burn. The patients simply need to be deprived of carbohydrate for 2-3 days. Studies inducing ketosis without increasing fat intake have not been done and it would be interesting if the effects are even more marked without the high fat intake.

Ketones are a kind of fuel produced from fatty acids (either from our dietary intake or body fat stores), by our liver, when glucose (a sugar) is severely restricted. Dietary regimens that stimulate the production of ketones by the liver are known as 'ketogenic diets'. 

In these new studies, mice in the intervention (ketogenic diet) group were fed diets providing around 90 per cent of energy (kilojoules) from fat and in the control groups were fed more than 65 per cent of energy from carbohydrates (starches and sugars). Those mice fed the ketogenic diet lived longer, and their memory was preserved better with ageing, compared to those fed the high carbohydrate diet. 

Needless to say, mice are not men, and much more research is needed before we know whether ketogenic diets have similar effects in free-living human populations. Short-term studies in humans suggest a small, non-clinically significant improvement in weight loss. There are currently no long-term (i.e. greater than 2 years) studies of ketogenic diets in humans.

Adult Australians consumed an average of 43.5 per cent of energy from carbohydrates in 2011/2 – far less than the >65 per cent of energy fed to the mice in the control diet. Total fat intake averaged 30.9 per cent of energy – far less than the 90 per cent fed to the mice in the intervention diet. It is unlikely that the average Australian would be able to (or would want to) adhere to a ketogenic diet for long periods of time.

Ketogenic diets typically require people to limit total carbohydrate intake to less than 10 per cent of energy (less than 50 g a day for a typical adult), and recommend fat provides around 80% of energy. What this means in real terms is severe restriction of:
*             most fruits, 
*             starchy vegetables (carrots, corn, peas, pumpkin, potatoes, etc...), 
*             cereal-based foods (bread, breakfast cereals, pasta, rice, etc...), 
*             legumes (beans, chickpeas, lentils, etc...),
*             milk and yoghurt.

For a typical adult, 10 per cent of energy, or 50 g of carbohydrate is equal to 1-slice of bread, 1-piece of fruit and 1/3rd cup of pasta or rice for the day. 

Instead of having carbohydrate-containing food, you must therefore eat mostly:
*             Meat, seafood, poultry
*             eggs, 
*             cheese,
*             butter and cream,
*             fats and oils,
*             low-carbohydrate vegetables (greens, onions, peppers, etc...),
*             low-carbohydrate fruits (berries).

Perhaps unsurprisingly, it’s difficult to obtain all of the essential vitamins, minerals and dietary fibres from a ketogenic diet, and supplements are therefore required.

Caloric restriction extends lifespan and healthspan in experimental animals, but for humans is arduous, with few people willing to undertake the ordeal of cutting food intake by 30 per cent.

For decades, researchers have sought a means of producing the same outcome with a ‘pill’. Since caloric restriction shifts metabolism from burning carbohydrates to burning fats for energy production, so causing the formation of ketone bodies, it was reasonable to test whether a ketogenic diet might mimic the benefits of caloric restriction.

In a new study published in the prestigious journal Cell, researchers in California found that indeed it did, at least in mice. It is most probable that this approach could improve health outcomes, slow diseases of ageing and increase human lifespan.

These well conducted and controlled studies in mice suggest that mice provided with very tiny amounts of carbohydrate benefit in terms of better memory to complete tasks such as mazes, more muscle function, and longevity compared to control mice.

There has long been interest in humans to restrict dietary carbohydrate to extremely low levels in order to induce weight loss. In most studies, these diets have been tested in the very short term due to the difficulties that humans have in adhering to extremely low carb diets necessary to maintain ketosis.

This would mean in a human adult only 20-30g of carbohydrates per day - a dietary pattern that would lead to deficiency in vitamins and minerals unless the diet is carefully constructed, and would usually require micro-nutrients and minerals to be provided by supplements.

Feelings of hunger and desire to eat are reduced whilst the body is in the state of ketosis, which may help explain why people eat less. However, in humans who have lost weight using ketosis-inducing diets will re-gain that weight unless the are able to maintain a lower energy intake as appetite returns.* 

The results of the memory improvements are intriguing in these mice experiments and provide a clue as to how the brain may operate using ketones rather than glucose as a fuel. The precise mechanisms need to be understood further as they may be helpful to people with severe diseases of the brain, such as epilepsy. 

So, should people start ketogenic diets? Answer: not unless there is a medical indication to do so and under the guidance of an accredited practising dietitian who can ensure nutritional adequacy.

In terms of ketosis being the cure for memory loss - I doubt anyone would be able to stick to the sort of strict low carb regimen necessary for long term use and therefore its applicability to humans is currently limited by the unpalatability of the dietary pattern that is required.

These studies are interesting in that they provided the dietary changes midlife rather than post-weaning, which is often the case. As such, this would make these studies somewhat more relevant to changing diets during adulthood.

However, as useful as the studies were to understand biological pathways, any extrapolation to humans needs caution. We note that firstly, the mice were lean (unlike most Australians these days), and secondly, that the “control diets” were 65-77% high glycaemic carbohydrates and included a significant amount of sucrose.

So is it that the low carbohydrate/ketogenic diet is relatively effective or is it that the “control diet” is relatively adverse? I might suggest the latter but at least there is an increasing research effort into understanding lower carbohydrate dietary patterns, which have been on the fringes in the scientific literature until now.

Our own work supports the metabolic advantages of less refined carbohydrate diets in humans with a more liberal approach to fat.