What The Literature Should Tell Us About Keto (But Doesn’t)

How mechanisms and frontline observations make it clear who is set to benefit, who needs some support to achieve these and who may want to rethink (for now)

TLDR:

• The literature cannot tell us what results any given individual will get from the ketogenic diet but is clear that it helps via impacts on blood flow, inflammatory control and especially rehabilitation of energy metabolism

• The ketogenic diet offers disproportionate benefits for those with the metabolic blockage that ketogenic diets side-step, and consistently beats other approaches when paired head-to-head

• Not everyone gets great results; considering the impact that keto has on digestive, cellular and hormonal activity allows us to easily identify those who may be disappointed

• Many such obstacles can be overcome through providing additional support to the processes in question, while some obstacles make the premise of keto unsuitable

(this is Part 2 of a 2-part article).

In Part I, we established how Evidence-Based Medicine is not in a position to tell us if the ketogenic diet is ‘good’ or ‘bad’. This is partially because of the hilariously low quality of most of the research (two thirds of the research did even provide a ketogenic diet) and partially because, even when the research methodology was sound, the EBM paradigm means that the studies report the average figure from the group (rather than observe the range of responses and consider who benefitted, who didn’t and the reasons for this).

As usual, we need to consider the evidence as a whole. This means not just the evidence provided by the literature but what we can already learn from real-life. What follows here is a lowdown that the literature should provide… but doesn’t. To that end, this narrative is backed by the literature but is not limited by it.

First things first, what is a ketogenic diet?

While this should be an obvious question, it is important to touch on because of the absence of any sort of definition in the literature. I will go first: I define a ketogenic diet as one that produces ketones.

Quite bizarrely, this simple yet irrefutable definition should be forefront in the mind of researchers when designing dietary interventions for trials. But it is not, which goes to show why the body of literature is so unfit for purpose when it comes to studies on the ketogenic diet. There is no consensus on what constitutes a low-carbohydrate diet and, as a result, we are plagued with so many ‘keto’ studies that did not provide a ketogenic diet. Some get the balance right but most do not, instead just doing what the other researchers did (which happens to be to just deploy the Atkins diet, which is rarely ketogenic).

What dietary balance do we require to induce ketosis? For individuals who are largely sedentary, this is going to be a maximum of 45g carbohydrates, 80-90g protein and the rest of the energy needs coming from fat (which will end up being 160-220g, depending on  lean mass). This equates to 19-9-73% to 15-7-79% for those who are used to thinking in percentages. Those who are physically active can be more flexible with carbohydrate and protein intake. Those with insulin resistance may need to cap the carbohydrates at a lower limit (eg. 25-30g).

One thing that I have seen repeatedly on the frontline is that individuals embarking on a ketogenic diet will normally have the carbohydrate intake well taken care off, but will over-consume protein and under-consume fats (often by a huge margin). This is why I would advise anyone utilising this approach to track their dietary intake, at least for one whole day (and ideally several). The totals you record may shock you.

Having achieved the desired intake, the next (and vital step) is to measure the ketones to determine that they are being generated. You can do this with a breathalyzer device or a pinprick blood monitor. The breathalyzers measure acetoacetate (with the range for nutritional ketosis 40-70 parts per million) and the blood monitors measuring beta-hydroxybutyrate (where the aim is 0.5-3 nmol/L). It’s fair to say that the levels recorded on tests is not the be-all-and-end-all in all individuals and all environments, but it’s an extremely reliable metric to determine success rates.

Urinary ketone measurements are cheap but are close to useless. They only tell us about ‘leftovers’; the net sum of what the body has produced, minus what it has metabolized. It is common to see them rise in the first few days of ketosis but, as the body adapts to making more efficient use of the newly-available fuel, they no longer make it into the urine. They are hugely prone to producing false positives in the early days and false negatives as time goes on.

This is why there is such a poor correlation between blood and urinary levels of ketones and why around half of individuals in ketosis will not show elevated ketones in their urine. The discrepancies between the two testing methods have been known since the 1940s, yet the majority of measurements conducting across the scientific research have used urinary ketones (of course, most papers don’t measure at all… yes, this is as shoddy as it sounds).

What does keto do?

First off, we have sufficient knowledge from the literature to form a clear idea of the benefits and downsides of a ketogenic diet. Studies show us that, at physiological levels of ketone production (2.0 nmol/L), we can expect a 39% increase in cerebral blood flow. They also demonstrate that, at physiological levels, ketones induce a metabolism-wide anti-inflammatory effect, which has proven clinically useful in a number of chronic inflammatory conditions (such as rheumatoid arthritis).

However, it appears that the most relevant metabolic effect is in promoting an alternative energy production pathway, in doing so ‘side-stepping’ mitochondrial blockages. To this end, it’s worth noting that:

• a) this is especially pertinent if individuals have 1. slowed activity in the PDH/Pyruvate Dehydrogenase enzyme (which occurs due to low B1 status or raised oxidative stress) or 2. impaired activity at Complex I of the electron transport chain (driven by excessive nitric oxide production, itself response to endotoxemia)

• b) these two metabolic blockages are the most common mitochondrial blockages to find when undertaking Organic Acid Testing with clients

We should also recognise that excess fermentation of carbohydrate in the small intestines (aka SIBO/Small Intestinal Bacterial Overgrowth) is well-documented to drive both intestinal symptoms and central metabolic challenges, and that such problems are also side-stepped with any diet that provides less opportunity for fermentation (this is easily achieved with keto).

In a nutshell, there are many reasons to expect that any individual undertaking an actual ketogenic diet can expect to receive the benefits shown in the trials, but with the added advantages of improved blood flow to the brain and enhanced anti-inflammatory control, and without the downside of leaving the central nervous system with low fuel availability.

Of course, those with any mitochondrial blockages or SIBO issues will be those who may show up as ‘outliers’ due to the disproportionate benefits they see.

Keto is better… for those who respond to it

The literature above outlines how the ketogenic diet exerts its effects and the published results allow us to form an idea of how potent this effect is. But we must remember that these results are the average response within the group studied.

In other words, all studies – even those with particularly impressive outcomes – will feature variable responses within the groups they studied; some participants will have hyper-responded while others may have seen no benefits. In other words, plenty of real-life responses are often even better than the literature suggests (but, for each one of these, expect one that disappoints).

The literature provides little answers as to who should apply caution before embarking on a ketogenic diet (again, a further example of the ‘black box’ approach that does not account for how an intervention actually drives a particular outcome and pays no mind to the pathways/systems/factors that such mechanisms are dependent on). So, once again, a combination of mechanistic understanding and real-life observations allow for clarity here.

Warning: it gets more nuanced before it gets more clear.

We’ve established that some individuals do exceptionally well with ketogenic diets. Some don’t. How can we tell who is who?

This is again where we focus on the mechanisms through which the diet works and the pathways that it depends on to elicit these effects.

An extension of this is recognizing that it can be a godsend for those with inflammatory burdens, common impaired mitochondrial impairments, restrictions in blood blow to their brains and disturbed conditions in their small intestines (and especially those with all four), but that the mixed responses tell us that we should not consider it a panacea and highlight the need to identify who does not stand to benefit.

What stops people from benefitting?

People who see disappointing responses fall into the following brackets

• Physical challenges – intestinal spasms and sulphur

• Metabolic blockages – AMPK, stress

• Messaging mismatch – catabolic signalling/insulin

Let’s touch on these issues and what drives them.

Physical challenges – intestinal spasms and sulphur

Intestinal spasms are connected to bile. Bile is released whenever we eat fat, as it is key in breaking the dietary fats into micelles (little droplets) within the digestive tract. It also has antimicrobial effects and triggers the contraction of the muscles in the gut lining. This is normally a positive thing, as this supports healthy intestinal movement. However, should there already be a tendency towards spasm, then bile will amplify this. And the more fat we eat, the more bile we release. This results in bloating (characterized by its round-the-clock nature) and general discomfort. Such problems tends to be resolved through somatic (“body-mind”) approaches that help discharge stored tension and reset brain-muscle connections (such approaches have proved very successful, if often slow going, while other options have largely disappointed).

Sulphur issues occur if someone has overgrowth of either desulfovibrio or biliphilia species in their gut. Desulfovibrio process dietary sulphur into hydrogen sulphide; while moderate amounts of hydrogen sulphide are not problematic, higher levels that diffuse into the circulation can compete with oxygen at the mitochondria. This gas formation can not only result in bloating after higher-sulphur meals but also, due to the oxygen competition, drives fatigue and brain fog. The ‘red flag’ for this issue is intestinal gas with a rotten egg odour.

Dietary sulphur is a natural trigger for such problems. Although it’s not possible, it can be difficult to moderate sulphur intake on a ketogenic diet (as the primary sources of dietary sulphur are meat, fish, eggs, cruciferous vegetables and onions/garlic). However, the main issue here also comes down to bile; bile is a highly-sulphurous substance. Recall what drives increased bile release: dietary fat.

Solutions here centre on temporary limiting dietary sulphur and the use of anti-microbial agents to kill of the offending microbes; we cannot expect success with a ketogenic diet while this issue is unresolved. 10-12 weeks of a plan that features basic metabolic support (as per individual needs), alongside low-sulphur diet (with sulphur instead provided transdermally, through Epsom Salt baths) and the use of antimicrobials that target the relevant species is normally enough to resolve such problems. (Current/crude evidence pointing to lemongrass as suitable options here, and frontline responses show it to be great option).

Metabolic blockages – AMPK and stress

An obvious detail, but one that deserves pointing out: if we are to remove the carbohydrate supply that the system has been receiving, then we need to ensure that it can still meet its energy demands. The organs/muscles need an ongoing supply of fatty acids, and the brain needs an ongoing supply of ketones (formed in the liver, but only when its energy pool is made up of lots of fatty acids and little glucose).

Supply of fatty acids is therefore an important consideration; specifically, will the individual’s fat cells release sufficient fats when the diet is not supplying them (eg. when we fast overnight). This release is co-ordinated by a messenger called AMPK. It will naturally be impaired if there is any downregulation of AMPK activity in such cells. And it turns out that downregulation of AMPK is a common feature of chronic inflammatory states.

Reduced AMPK activity means reduced release of fatty acids, and this means reduced fatty acid availability at the liver, and this means reduced ketone production (as we noted above, ketone production generally occurs only when there is a high fatty acid-to-glucose ratio); this is where individuals who are particularly stressed (aka have lots of adrenaline dumping higher levels of sugars into their bloodstream) may struggle to produce the ketones needed.

If someone does not release fats from their fat cells at an appropriate rate (eg. low AMPK activity), then herbs such as Gynostemma (which reliably activates AMPK) can make the difference. If individuals are struggling to digest fat, then we can normally overcome this obstacle through providing additional support, in the form of digestive enzymes that both include ox bile and contain a high lipase content.

If they are subject to particularly high stress (specifically, high sympathetic activity and the high adrenaline load that goes with this) then they may need to undertake some life changes. Often, when we see individuals who show pronounced sympathetic activation but ‘don’t feel stressed’, nervous system work will be required first. Calming herbs can sometimes be enough, but are not a reliable solution in themselves.

It’s worth pointing out that there are exceptions here; a particularly potent source of stress is not having sufficient cellular energy, which means that there are times when a ketogenic is the most powerful way to reduce stress in this individual. This therefore needs to be considered on an individual basis; we can normally determine what we need to know within 1-2 weeks of tracking and make a call (pursue or conduct a U-turn) accordingly.

Messaging mismatch – mimicking fasting when our systems have been low on energy for too long

I hope it is obvious from the above that, when people ‘do badly on keto’, it is sometimes due to maladaptive responses to high fat intake (intestinal spasm) or increased bile release (hydrogen sulphide production) but also because they are not actually producing ketones. Not producing enough ketones is extremely common.

It is therefore easy to conclude that, if we have overcome the obstacles in handling the dietary fat/bile release and resolved the impediments in producing ketones, we are now all set to exploit the benefits of the ketogenic diet. And, as we discussed in part I, these can be sufficiently powerful to be life-changing.  

And this is indeed how it tends to play out, at least for a time. This is where it is important to consider the potential downsides of long-term ketosis, which ties into the usual questions: are ketones a better fuel for the body? Is ketosis stressful? These two questions are inter-connected. And this is where it gets most nuanced.

To this end, ketones are not a better fuel than glucose. They are a different fuel, and have benefits and drawbacks versus glucose. The benefits are that they side-step common metabolic blockages, mimic the beneficial effects of fasting (in autophagy, aka helping breakdown cellular debris) and show anti-inflammatory effects, while providing excellent opportunities to reverse insulin resistance. The drawbacks are that they mimic some of the undesirable effects of fasting.

The eagle-eyed reader will no doubt have noticed that the fasting-mimicking effects of ketosis can be both an advantage or a disadvantage. As per usual, it depends heavily on context. Let’s look at two stereotypical examples, the so-called Average Westerner and the individual who has been seeking better health for some time, thinks carefully about diet and supplementation but hasn’t quite seen the results they want.

Average Westerner (overfed, inflamed, stressed, insulin resistant, subject to symptoms but still functional) tends to benefit from fasting and interventions that mimic it. This is because their storage sites have become overly full, aka lipotoxic. This lipotoxicity drives oxidative stress in the cell and forces the cell into action that protects against further oxidative stress; namely, downregulation of the insulin receptor (aka insulin resistance) to limit the entry of further/excess energy.

The fasting-mimicking effects of the ketogenic diet can be a wonderful tonic in these circumstances. It removes the ongoing burden of energy-burdened cells having to process incoming glucose, as well as encourages the clearance of the excess energy in storage; this means less oxidative stress going forward, and no further need to protect the cell through inducing insulin resistance. You really couldn’t conceive of a more well-suited strategy when it comes to reversing insulin resistance (and it is therefore no wonder that we see papers documenting how the dietary approach reversed type II diabetes in just 90 days). Add in the anti-inflammatory effects of ketone bodies and you have a multi-faceted solution to the metabolic syndromes that are progressively becoming the norm.

On the other hand, those on a journey back to health have often already read up on these benefits and employed intermittent fasting for some time. What’s more, it’s commonly seen how such individuals have raised levels of adrenaline and prominent intestinal disturbances (see here and here), obstacles we touched on above. This is highly relevant on the frontline; in my experience, more than half of individuals with complex/chronic issues will have substantially undereaten for some time and, due to a lack of hunger, will not even be aware of it until I get them to track their intake.

Sustained low caloric intake triggers our evolved survival mechanisms, centralized in the ‘energy management centre’ of the body, the hypothalamus. This is where we see increased activation of the hypothalamic-pituitary-adrenal axis, reduced thyroid activity (specifically, reduced T3 formation and reduced T3 uptake into cells) and reduced investment in steroid hormone production. We also see increased activation of the sympathetic nervous system (‘fight-and-flight’-type stress response) although, as the lack of calories is sustained over longer periods of time, we can expect a shift in the type of stress response towards increased activation of the parasympathetic stress response.

While often misinterpreted as a desirable thing - ‘higher HRV scores are good, right?’ - this is parasympathetic stress. This is often referred to as the ‘freeze’-type stress response and can be compared to a ‘waking hibernation’ as the system attempts to save energy. This can obviously affect our sense of energy, but also mood and concentration – think sluggishness and brain fog – but also limits energy investment into non-emergency tasks. In other words, it suspends a whole host of healing tasks. Too much of this response is not what we want if our goals are healing and optimal health.

It’s clear that, in any underfed environment, the benefits of the ketogenic diet are likely to be less evident. Or absent. This is especially the case for anyone who has spent long periods of time on calorie-controlled diets; their metabolism needs help moving out of energy-saving mode, not further environmental inputs that energy is scarce.

It is also relevant for anyone who goes though phases of high energy needs. eg. athletic competition (or anything that activates the sympathetic nervous system to this extent, which can include emotional challenge/life stress). In these circumstances, they need to regenerate the ATP quickly, and this means that their body will need glucose. Put another way, ketones/fatty acids will not be sufficient on their own to meet demands, aka they will leave someone under stress (defined as any circumstances where energy demands are greater than energy availability). The more stress the individual faces, the more we will see the elegant-but-often-unwanted stress response that conserves energy (at the cost of healing, hormonal balance, performance and mood).

Of course, this is conditional in that this stress can be fairly framed as simply ‘asking ketones to do something that they do not do’ and something that causes little problems if such demands are only placed here and there. This makes a rational case for just not over-taxing the system while on a ketogenic diet… but… there are few of us that can remain in a sufficiently relaxed for the majority of the time. The pattern I see here is the longer someone has been on a ketogenic diet, the more vulnerable they are to rounds of these stress responses; this links into what we see in the literature, where the recorded weight loss often starts off at breakneck speed but then gradually plateaus at around 5 months (indicating an adaptive response within the metabolism to the signal of lower energy supply).

So is the masterplan therefore to remove any metabolic obstacles, enjoy the benefits, then add back carbohydrates at 5 months in? One obvious caveat is that, for some people (those with insulin sensitivity issues and challenges with carbohydrate metabolism, etc), this not-quite-got-enough stress that can occur on a ketogenic diet is infinitely preferable to the nowhere-near-enough stress they personally face when using a carbohydrate-based diet. And it also speaks to the potential utility of a cyclical ketogenic diet.

It is therefore not a fair generalization to say that the ketogenic diet is stressful on the system. More that it can result in stress if our system is asked to replace ATP stores quicker than ketones/fatty acids can supply them, but that the stress of dysfunctional energy production is far greater.

The practical takeaway is that, in these circumstances, what we really want is to provide the system with inputs that demonstrate energy safety. This is where increased insulin signalling – an ‘insulin hit’ - is a great tonic in alleviating the stress of a low energy state. Insulin has been unfairly painted as the bad guy and, while we are indeed asking for trouble if our insulin levels are consistently pushed into the high zone, insulin has a whole host of benefits: at baseline, it supports blood pressure regulation, healthy immune responses, formation of glutathione and formation of T3, among many other functions. When low insulin activity is restored, it normalizes the overactivation of the HPA axis and normalizes steroid hormones.

What if we exploited the abundant benefits of a ketogenic diet, while also periodically nourishing the system with the benefits of healthy insulin signalling?

A more general note here is how the human body, like everything in nature, has evolved to move through cycles. This includes cycling between energy excess and energy deficit; there are advantages to both states, and clear downsides if we live only in one phase of this great cycle. Cycling in and out of ketosis has worked very well for many; the ketogenic state can provide a platform to repair carbohydrate metabolism and therefore tee up a scenario where we can now benefit from both ketone- and carbohydrate-based states.

(SIDE NOTE: if this is surprising to hear that insulin has positive effects, be aware that there is so much written that conflates insulin activity with insulin resistance; one does good things, one causes problems… and yes, they are related, but this is a discussion for another time).

Summing up – Concluding on what the whole body of evidence says

The evidence demonstrates that the ketogenic diet is a powerful tool, which elicits a number of complementary effects that are remarkably well-matched to the problems that run rampant across Western society (a society whose consumption tends to exploit just one part of the energy deficit-energy excess cycle, despite possessing genes that prime them to benefit from phasic intake). These mechanisms centre on inflammation, insulin signalling and mitochondrial energy production, making it exceptionally well-placed to rehabilitate energy metabolism.

In short, if our systems have been limited due to metabolic blockages AND have suffered the consequences of sustained energy excess AND we successfully produce ketones, then we can expect no intervention to come close to the effects of a well-run ketogenic diet.

I hope this successfully explains how we can see the spectacular outcomes observed in so many studies (outlined in Part I) – and why head-to-head dietary trials for weight loss see the ketogenic diet leading 35-0 against other approaches - but why enough individuals attempting a ketogenic diet fail to see such benefits and, when this is the case, why some astute screening can determine which individuals can unlock these benefits with some key adjustments and who should steer clear (for the time being, at least).

Put another way, the argument of ‘good/bad’, ‘effective/dangerous’ is bogus. It’s clear that the ketogenic diet is an astonishingly effective tool when provided to those who need to revive their energy metabolism, but we may not see benefits if we do not account for our individual circumstances and instead assume that it automatically works on a ‘feed keto in, get great results out’ basis.

NEWS - If you enjoyed learning about this topic, then I suggest taking a look at the upcoming Academy event (two-day courses in functional health). Next date: Jan 10-11th 2025, attendance in London and via Zoom. Details at marekdoyle.com/academy/.