FGF21 - a liver hormone linking sugar cravings and cardiovascular disease.

On Christmas Eve the media carried reports that scientists had identified a hormone, produced by the liver, that switches off sugar cravings, and which might be the answer to sugar addiction.

"Research on mice and monkeys has shown that the hormone, FGF21, signals the brain to avoid seeking sweet foods.

Harnessing the effect, possibly by copying the hormone's action with a drug, could help patients who are obese or suffering from Type 2 diabetes, scientists believe."



We all know that a LCHF diet suppresses the appetite for sugar - one of my dietary epiphanies involved standing in a supermarket, among shelves of garishly packaged chocolates and sweets, and realising that though being exposed to that stuff had always obliged me to buy some piece of it to take away and eat before, now I hadn't a single impulse worth fighting to buy or eat any of it ever again. 

So it was an obvious question (one I may not have asked had I read the full paper first, because in the actual experiment FGF21 is clearly being produced in response to eating carbohydrates) whether a ketogenic diet raises FGF21.
It does - FGF21 is part of the regulatory response to fasting or a ketogenic diet (in mice).
So, eat keto, and your liver produces FGF21, which stops you wanting sweet food. Simple.

But, like that book by Ben Goldacre, I think you'll find it's a bit more complicated than that.

Luckily though, the complexity of FGF21 seems to fit a pattern we've seen before with other hormones.

While I was looking for the link to ketosis, I noticed in my search results a paper linking elevated FGF21 to cardiovascular disease. How can something that stops you from eating sugar be linked to CVD? That was before I knew that FGF21 was produced in response to eating carbohydrates. By the time I read the CVD paper, I knew what I was looking for. Here it is;
The metabolic syndrome has also been associated with
increased serum FGF21 levels, whereas an increase in FGF21
serum levels has been suggested as a new biomarker for
nonalcoholic fatty liver disease or steatohepatitis (17, 54, 60,
62, 116, 118). A study on obese children confirmed that
increased serum FGF21 is correlated to BMI and free fatty
acids (90). When serum FGF21 levels were tested after an oral
load of fructose, it was interestingly shown that FGF21 values
acutely spike, presenting a similar curve as serum glucose and
insulin after a glucose load. This finding shows that FGF21
presents a typical hormonal response possibly mediated by
carbohydrate-responsive element-binding protein that is activated
by fructose (18).
                                                         ...it was shown that FGF21
levels are predictive of combined cardiovascular morbidity and
mortality (Fig. 3) (59). Increased baseline serum levels of this
molecule were found to be associated with a higher risk for
cardiovascular events in patients with type 2 diabetes in the
Fenofibrate Intervention and Event Lowering in Diabetes
(FIELD) study, and interestingly this association tended to be
stronger in the patient group that presented higher total cholesterol
levels (84). The authors speculate that the increased
basal levels of FGF21 in this group of patients may be an
indication of the potential role of FGF21 as a biomarker for the
early detection of cardiometabolic risk and furthermore that it
may reflect a compensatory response or the need of supraphysiological doses of FGF21 as a result of FGF21 resistance, a hypothesis proven in obese mouse models (21).

FGF21 resistance and hyperFGF21aemia. It's a familiar pattern. If FGF21 is produced in response to carbohydrate - maybe when hepatocytes reach glycogen saturation or ATP depletion or some other threshold - but your lifestyle or culture involves eating past the signal, maybe because of some dopamine effect of sugar you're sensitive to, or because you bought the 1.5 litre bottle of Coke because it was cheaper than the 300ml and it shouldn't go to waste, or because your mum or dietitian is telling you to finish your cake or keep eating the low fat food regularly, then maybe your liver eventually, because you stopped listening to it, makes so much FGF21 that the cells that should notice it become insensitive to it, so you eat more sweet carbs, make more FGF21, and get the same vicious cycle that we see with insulin.
(Or maybe there's some other cause for FGF21 resistance, a virus or environmental toxin or food colouring or genetic bad luck. It doesn't really matter unless you finish the cake.)

So what happens when a modern human goes on a keto diet or fasts? FGF21 may not rise at all - instead, sensitivity can be restored by its dropping, much like insulin. 

Raymund Edwards astutely tweeted this study, which shows what happens to FGF21 when humans fast for 10 days. It's not the same as it was with the mice. In the mice on the ketogenic diet, it sometimes looked as if FGF21 played a role in ketogenesis - in the fasting humans, ketones rose first and FGF21 followed days later. This actually makes more sense, because ketones suppress appetite independently of FGF21, and are produced through basic biochemical economics - this shouldn't require some fancy new hormone, just let glucagon dominate over insulin and the Krebs cycle will do the rest.
In some of the humans, FGF21 was elevated at baseline and dropped fairly quickly, as can be seen in the spaghetti plot:


Whereas ketones rose more rapidly:



Anyway, this line of investigation, which I have only skimmed superficially here, gives us two possibilities; we have pathways which we can use to explain the loss of a sweet tooth when carbs are restricted (either FGF21 elevation, or the restoration of FGF21 sensitivity), and, we have an additional connection between sugar and cardiovascular disease.

What does it mean that FGF21 rises so much when fasting (and probably similarly on a keto diet), if elevated FGF21 is associated with CVD and other metabolic diseases?

If you read the CVD paper, FGF21  has a number of beneficial and antiatherogenic properties. It doesn't seem like bad stuff to have elevated, unless it got that way from a high intake of fructose.

My rant about David Katz's double identity and the meaning of consensus.

(I posted the first part of this rant on this RetractionWatch post but the moderator seems to have decided that it doesn't meet their policy. Fair enough as I find it hard to be restrained about such nonsense.)

I don’t care that David Katz wrote the fake review. Fiction about fiction, all very meta.
What I do care about is the Big Lie he repeats in his defenses – that those criticizing the dietary guidelines and the DGAC process “are the very group employing every means at their disposal to scuttle dietary guidance dedicated to public (and planetary) health to serve their own pecuniary interests”.
This is the paranoid underside to the grandiose self-image that Katz displayed in the reviews. Those critiquing the guidelines have few pecuniary interests and I would be surprised if any of them are as wealthy as Dr Katz. That Dr Katz sees fit to call them for “want of qualifications” begs a question – why is it that PHD students, engineers, psychologists, cell biologists, hard-working journalists, and auto-didacts can see and point-out glaring omissions and bias in the way the DGAC selects and interprets evidence, yet someone like Dr Katz, with enough letters after his name to write another novel, refuses to see them? (Indeed, why, with all these qualifications, did Dr Katz get involved with pseudocience in his practice?).
Katz, the CSPI, and the DGAC committee themselves have brought great guns to bear to find a few minor inaccuracies in Nina Teicholz’ long analysis, none of which seem to affect the conclusions. Yet nowhere do we see them addressing the countless accuracies, which surely need to be addressed if the DGAC is to recover its credibility.
Whatever the DGAC may end up recommending in the near future, it will be different from what they currently recommend, meaning that the current recommendations are not supported by current science. If the science is this labile, why were such far-reaching recommendations being made at all? Why not stick to the basics of nutrition – eat foods rich in vitamins and minerals, not refined foods – get enough protein and essential fats, preferably from natural protein-and-fat foods rather than refined foods or grains – and be aware that diabetes and obesity usually indicate an intolerance to carbohydrates, especially refined ones. These are choices that will improve or maintain health in the present. Theories about what will reduce this or that disease at the end of our lives, based on interventions that do not reduce mortality, should never have been allowed to distort nutritional advice.
Nor should unproven theories about what is best for the planet. Someone who wants what is best for the planet won’t tell us to remove the fat from meat and not cook with animal fat – this advice wastes most of the energy produced by raising animals, which then needs to be replaced with energy derived from growing additional plants. In the case of the unsaturated fat energy these experts are so keen on, these nutritionally unnecessary plants need to be processed in an environmentally damaging and industrially hazardous way.This inane suggestion comes from people who claim to be dedicated to planetary health (another grandiose Sci Fi claim when you think about it). Surely it would be better to leave the effect of diet on planetary health aside until it can be addressed by someone with more information, better sense, and no other axe to grind.

This is why I do not accept the notion of consensus peddled by Katz, Hu, Willett and others after their recent Oldways celebration. It fails to address any of the inconsistencies in the advice given by most of the attendees. In the case of Paleo expert Boyd Eaton's presentation, it is plainly compromise rather than consensus that is being offered. No doubt this is the price Paleo needs to pay to enter the elite plant-based bullshit club, but the idea of "fat-free" dairy replacing whole meat in a future-paleo menu suggests the price is too high, and not even consonant with mainstream nutritional research in the present day.

Perhaps what sticks in the craw most is the call for the media to avoid reporting research with fear-mongering headlines that contradict each other. This is priceless when one of the signatories is Walter Willett, who feeds this stuff to media outlets verbatim.
But not always.

It is really the question of context that bedevils compromise consensus. However you interpret the evidence for phytochemicals, there are people who don't tolerate plants all that well but have no problem with meat. You may think that it's the saturated fat in pizza that makes it artery-clogging, but saturated fat seems to make no difference to metabolic profiles when carbs are restricted (if anything, it makes them non-significantly better). And when you're supplying advice to a population with a high rate of carbohydrate intolerance, you need to take the relative metabolic superiority of fat into account, yet there was no-one at Oldways speaking for LCHF. Saturated fat and whole-grains are the shibboleths, and you still can't enter the hallowed precincts if you can't pronounce their "artery-clogging" and "healthy" prefixes. 

I'm sure there is plenty of room for agreement between all of us - absolutely no-one here thinks that commercially processed food, frequent deep-frying, and a high content of added sugars are a good idea. But people need to eat, and used to cook meals based on meat, eggs, animal fats, and dairy which were easy to prepare using knowledge handed down in families. A media campaign that painted those foods as killers for decades is one factor behind a tragic and damaging decline in basic cooking abillity and increased reliance on a well-and-truly depraved food industry. Industry can place products with added fibre or low in saturated fat as "healthy" with the backing of epidemiological nutritionists, and sell the alternatives as "treats" which are fine in moderation as part of a balanced diet according to the dietitians. With what results we see.

And, seriously, you want to fix this by feeding Americans (and by extension, the English-speaking world) a watered-down but still costly Mediterranean diet, when there are other foods their grandparents ate which will do the job equally as well?

Everyone in nutrition is influenced, more-or-less unscientifically, by their own dietary choices or those of their culture. On the one hand we have a clique of mandarins who were "born on second base and think that they've hit a home run" with regard to diet and metabolic health. On the other hand we have people such as Tim Noakes, on trial for his opinions as I write, who have overcome metabolic disadvantages with the help of diets that have included the prohibited elements. By any objective test, the second narrative should be the more convincing, but perhaps not in a society that worships unearned success. It is obvious enough that the selection and appreciation of evidence in the DGAC process is distorted by unthinking acceptance of the first narrative. We owe a real debt to Nina Teicholz for bringing this out to be debated in the public domain.

What is the right thing to do when this happens? To blame it on "the very group employing every means at their disposal to scuttle dietary guidance dedicated to public (and planetary) health to serve their own pecuniary interests”? To call in the sponsors, circle the wagons, and manufacture consensus for the media? 

Or to hold a full and frank investigation into the reasons for the debacle, one which includes the evidence gathered by those who don't think your conduct of operations has met a satisfactory standard?

Medium Chain Fatty Acids and Brain Metabolism

This post relates in some way to each of the three previous posts.

The definition of MCFA is a little unclear. Wikipedia lists lauric acid as a MCFA, making the range C:6-12, whereas commercial MCTs are almost completely made from C:8 and C:10, as C:6 is not available in any significant amount from coconut oil, and about 32% of lauric acid is not deposited into to the hepatic portal vein, whereas the totality of shorter chain MCFAs is. It is likely that both lauric and myristic (C:14) acids exist in a grey zone where they have partial MCFA properties. It is also relevant that triglycerides that contain longer chain fatty acids are hydrolysed more slowly and a rapid rate of hydrolysis in the gut is one of the properties desired of MCTs.

In a previous post I wrote about a case study where a ketogenic diet prevented symptomatic hypoglycaemia in a child with hyperinsulinism. I wrote at the time that this was as close as we would get to a proof that slightly elevated ketone levels due to carbohydrate restriction are protective against symptomatic hypoglycaemia in people with type 1 diabetes treated with insulin.
I was wrong - there has been a human trial of the concept, using MCT oil.[1]

In this study "A total of 11 intensively treated type 1 diabetic subjects participated in stepped hyperinsulinemic- (2 mU · kg−1 · min−1) euglycemic- (glucose ∼5.5 mmol/l) hypoglycemic (glucose ∼2.8 mmol/l) clamp studies. During two separate sessions, they randomly received either medium-chain triglycerides or placebo drinks and performed a battery of cognitive tests."

"During the medium-chain triglycerides session, a total of 40 g of medium-chain triglycerides (derived from coconut oil containing 67% octanoate, 27% decanaote, and 6% other fatty acids; Novartis) was ingested at 25-min intervals with front loading of 20 g then 10 g twice. During the control session, cherry-flavored water sweetened with sucralose was ingested at identical time intervals."
The beta-hydroxybutyrate level attained 40 minutes after the MCT drinks was about 3.4 mmol/l.

"We conclude that ingestion of medium-chain triglycerides improves cognitive function without affecting the adrenergic hormonal or symptomatic responses to acute hypoglycemia in intensively controlled type 1 diabetic patients. These findings suggest that medium-chain triglycerides could be used as prophylactic therapy for such patients with the goal of preserving brain function during hypoglycemic episodes, such as when driving or sleeping, without producing hyperglycemia."

BOHB levels for MCT vs Placebo in insulin-induced hypoglycaemia after overnight fast, down arrows = 20g, 10g, 10g MCT or placebo drinks. 100 umol/l = 0.96 mmol/l.

There was another interesting aspect of this experiment.

"In vitro rat hippocampal slice preparations were used to assess the ability of β-hydroxybutyrate and octanoate to support neuronal activity when glucose levels are reduced."

The reason for this is, that the authors wanted to be sure whether the protective effects of MCT oil were due to the brain using ketone bodies or due to the brain's use of MCFAs. It turns out that the MCFAs used in MCTs can cross the blood-brain barrier and be used in brain metabolism. In another rat paper, "We found that oxidation of 13C-octanoate [C:8] in brain is avid and contributes approximately 20% to total brain oxidative energy production."[2]

The C:8 is mainly being oxidised by astrocytes. If this happens in a hypoglycaemic brain, it's possible that due to lack of oxaloacetate ketone bodies will be produced, which can be used by the neurons.

What I really want to know is how coconut oil compares to MCT oil as a means to elevate serum ketone bodies. I suspect that ketone elevation from coconut oil has a slower onset and is more protracted due to the slower rate of hydrolysis of MCFAs from triglycerides with some longer-chain fatty acids, and if so this "time release" effect could be beneficial during sleep.

There is only one study I can find online which shows ketone levels after feeding coconut oil, and this is Mary Newport's n=1 experiment.[powerpoint here]







I don't know what to make of this, beyond the expected drop in glucose (due to insulin response to lauric acid - this wouldn't apply in type 1 diabetes); the levels, though elevated by both interventions, are still within the reference range (and very different from those in the diabetes paper), unless I'm reading the measurements wrong, and the time scale with coconut oil stops short. I'd like to see many more comparisons like this, with higher doses, in healthy volunteers. The coconut oil industry and coconut oil aficionados have been accused of extrapolating from MCT studies in the absence of evidence about coconut oil, for example by the Heart Foundation of New Zealand here. While I don't think it's justifiable to ignore animal studies of coconut oil, which tell us that coconut oil protects the liver and pancreas from chemical injury, totally consistent with the MCT research, I don't see why the coconut oil industry can't fund proper comparative studies of ketogenesis in humans, which would not be at all expensive.

A 1982 review of medium chain triglycerides stated that "MCTs are ketogenic in the normal subject
and even more in the patient with hyperosmolar diabetic syndrome (117). Hence, MCTs should not be given to patients with diabetes. They should also not be given to patients with ketosis or acidosis."[3]
Whilst no-one would treat diabetic ketoacidosis with MCTs, the statement "MCTs should not be given to patients with diabetes" is unfounded. People with type 2 diabetes, due to hyperinsulinaemia, are not at an increased risk of diabetic ketoacidosis*, and the experiment I posted above shows that those with intensively controlled type 1 diabetes may benefit from their use. The reference (117) which is the only reference in this section is a rat experiment; the hyperosmolar diabetic syndrome described is high glucose with normal ketones, not DKA.
A 2010 review cites several reports that "suggest that MCFAs/MCTs offer the therapeutic advantage of preserving insulin sensitivity in animal models and patients with type 2 diabetes".[4]


This is consistent with the Malmö Diet and Cancer study epidemiology I posted here. Which implies that even the small amounts of MCFAs in foods such as coconut and dairy are beneficial for maintaining metabolic homeostasis at a population level.

*Edit: thanks to Carol Loffelman for reminding me of this - type 2 diabetes is a risk factor for ketoacidosis if it's being treated with a SLGT2 inhibitor. See this link, but there are many cases of ketoacidosis on SLGT2 inhibitors where a low carb diet is not involved. I have looked for case studies of ketoacidosis in diabetic patients that were triggered by carbohydrate restriction or MCTs without SLGT2 administration and have not yet found one.
This is a case study of DKA in a woman with decompensated T2D [link] where there is not enough insulin to prevent it. There's no low carb diet or SGLT2i involvement, and my expectation is that a normal calorie very low carbohydrate diet would most likely have prevented the syndrome in this patient as it did in the patients of Newburgh and Marsh back in the day.


[1] Page KA, Williamson A, Yu N et al. Medium-Chain Fatty Acids Improve Cognitive Function in Intensively Treated Type 1 Diabetic Patients and Support In Vitro Synaptic Transmission During Acute Hypoglycemia. Diabetes. 2009 May; 58(5): 1237–1244

[2] Ebert D, Haller RG, Walton ME. Energy contribution of octanoate to intact rat brain metabolism measured by 13C nuclear magnetic resonance spectroscopy. J Neurosci. 2003 Jul 2;23(13):5928-35.

[3] Bach AC, Babayan VK. Medium-chain triglycerides: an update. Am J Clin Nutr. 1982 Nov;36(5):950-62.

[4] Nagao K, Yanagita T. Medium-chain fatty acids: functional lipids for the prevention and treatment of the metabolic syndrome.  Pharmacol Res. 2010 Mar;61(3):208-12. doi: 10.1016/j.phrs.2009.11.007. Epub 2009 Nov 30.

Breastfeeding on a low carb diet - is there an increased risk of ketoacidosis?

[Disclaimer: I have never breast fed a baby and thus have no practical experience of this subject. This blog post details the insights into this problem that can be found in the medical literature and is not intended as medical advice.]

Any disaster that may overtake him, even to the extent of
ground moles getting in his lawn, will be blamed on his "red
meat" diet.
                - Blake Donaldson, Strong Medicine 1961 (download here)



Lactation ketoacidosis is a rare event in humans, and the authors of a 2012 case study could only find four previous cases in the literature.[1]
Non-diabetic ketoacidosis outside of breastfeeding related to diet is also rare, with 3 cases (including one unexplained death) attributed to the Atkins diet. It will be seen from the example given here the role that fluid and electrolyte loss from vomiting can play in the condition.[2]
What all these cases seem to have in common is rapid weight loss; they also (where information is available) tend to involve illness with loss of appetite, fasting, or deliberate undereating. There are also cases involving recent or historical gastric bypass surgery. I am unable to find a case of diet-related ketoacidosis in a male, or of either lactation or LCHF diet-related ketoacidosis in anyone with type 2 diabetes, type 1 diabetes, or gestational diabetes (however diabetic ketoacidosis is an infrequent complication of pregnancies in gestational diabetes). On the other hand, anorexia and bulimia are associated with diabetic ketoacidosis in insulin-dependent diabetics.
The Association of British Clinical Diabetologists states that diagnosis of ketoacidosis should only be confirmed with a concomitant blood glucose over 11.1 mmol/l or known diabetes, and significant acidosis (arterial pH below 7.3 or venous bicarbonate below 15 mmol/l). Thus the non-lactation case in ref [2] qualifies, despite normal HbA1c, but the lactation ketoacidosis cases do not because blood glucose in these cases is at normal or lower than normal levels.
Recently two cases of lactation ketoacidosis in Sweden have been given wide publicity. The first of these is discussed by Andreas Eenfeldt here on his diet doctor blog, and can clearly be attributed to inability to eat for a prolonged period, unrelated to the diet - the woman concerned now says that she continues to breastfeed at 10 months on an LCHF diet; if the diet composition was the cause of the ketoacidosis the risk would have increased as the child's milk requirement grew.
The second case might give us more prima facie reason to suspect a mechanism related to diet composition.

"A 32-year-old white woman presented to our county hospital with a history of nausea, vomiting, heart palpitations, trembling and extremity spasms. She had started a strict LCHF diet, with an estimated carbohydrate intake of less than 20g per day, 10 days before admittance, lost 4 kilograms and had felt growing malaise. She was breastfeeding her son of 10 months of age.
An arterial blood gas was taken. It revealed pH 7.20, base excess (BE) −19, partial pressure of carbon dioxide (pCO 2 ) 2.8 kPa, glucose 3.8mmol/l and lactate 1.0mmol/l. Her blood ketones were 7.1mmol/l (reference 0 to 0.5mmol/l). The primary diagnosis was thought to be ketoacidosis due to starvation induced by the LCHF diet."[3]

The authors concluded "A lactating woman has a high demand of substrate to produce milk. A LCHF diet limits the amount of substrate and results in a negative energy balance. This kind of diet should thus be avoided during lactation."

This is vague. What is the substrate(s), and are they limited by the LCHF diet itself, or by negative energy balance, which is something that can occur on any weightloss diet or as a result of undereating for any reason? There is a case study of lactation ketoacidosis occurring in someone eating a normal diet with a healthy appetite (the possible triggers were feeding twins, albeit with some formula feeding, and gastric bypass surgery some 5 years previously).[4] What aspects of LCHF if any would counsel avoidance during lactation, and what if anything can be done to modify these, given that ketogenic LCHF will be some mothers' choice of a natural treatment for potentially serious medical conditions, and not only a way to regain one's original silhouette?

The substrates for milk production are amino acids for protein synthesis, triglycerides (fatty acids and glycerol) and glucose for fat synthesis, and glucose (or other sugars) and glycerol for lactose synthesis. In the carbohydrate-fed state all lactose (a disaccharide formed by joining one molecule of glucose and one molecule of galactose) can be made from glucose, in the fasting state glycerol is the substrate for the majority of the galactose portion, but not the glucose.[5] If galactose is present in the diet it will be incorporated into lactose in the breast (rather than being converted to glucose in the liver).

Lactation ketoacidosis is sometimes called "bovine" ketoacidosis because it is similar to the ketoacidosis seem in milking cows. However there is an important difference as acetate is the main substrate in ruminants. Hexoneogenesis involves the interconversion of glucose or glycerol to glucose or galactose via the normal glycolytic or pentose-phosphate pathways, so does not require the removal of oxaloacetate from the TCA cycle, whereas generation of glucose and galactose from acetate in ruminants involves removing oxaloacetate from the TCA cycle, increasing the rate of ketogenesis.
In starvation and carbohydrate restriction, ketogenesis from fatty acid oxidation in the liver is in large part a byproduct of gluconeogenesis - when there is little glucose in the diet, the liver needs to make glucose from amino acids and glycerol. Removal of oxaloacetate from the TCA cycle to form glucose means that all the acetyl-CoA yielded from fatty acid beta-oxidation cannot enter the TCA cycle by condensing to citrate, and some is converted to ketone bodies which are exported from the liver instead.
It is the removal of blood glucose for breast milk production and compensatory production of extra glucose by the liver that may have the potential to raise ketones above the level usually seen on ketogenic diets. The removal of glycerol for fasting breast milk production also means that this substrate for hexoneogenesis, which also supplies oxaloacetate to the TCA cycle and is the main substrate for fasting hepatic glucose production, needs to be plentiful in the LCHF diet (i.e. by keeping fat intake high).
Because blood glucose may be unusually low, insulin sensitivity is high and insulin is low. Whereas ketoacidosis is reversed with glucose and insulin. This may explain why lactation ketoacidosis and diet-related ketoacidosis isn't seen (as far as I can tell from the literature) in people with type 2 diabetes, in whom high insulin and high glucose will continue to suppress ketosis.

Ref [3] may also indicate that the period of ketoadaptation presents a increased risk of lactation ketoacidosis (or it may not, given how rare such cases are).

Ketoacidosis is unlikely in a fat-adapted person, because ketone and non-hepatic FFA clearance has increased, electrolytes are back in balance, and stress hormones have normalised. Perhaps trying to ketoadapt while lactating is like trying to ketoadapt while running a marathon.
In the case of lactation, the doubled demand for glucose GNG could cause a sudden rise in ketones during ketoadaptation - making it too easy to get into deep ketosis before the body has fully adapted to utilize ketones or regulate ketogenesis.  Meanwhile, the loss of electrolytes during ketoadaptation (due to the diuretic effects of glycogen loss), especially in someone who doesn't eat enough salt, could result in a decreased ability to buffer serum ketoacids. Electrolytes are also being incorporated into the milk and have been contributing to fetal growth before that.
It seems to me that in a woman who has already adapted to the LCHF diet, especially if she has a hyperinsulinaemic condition as her reason for LCHF eating, the conditions that predispose to ketoacidosis are reduced, especially if sudden weight loss is avoided.
On the other hand lactation, by diverting sugars into milk, will also increase glucose tolerance, and breastfeeding decreases a woman's future risk of type 2 diabetes.[6]

A comment (by greensleeves21) on the Diet Doctor blog includes the following observation:
"The old Atkins community group heard Dr. Atkins once caution against this theoretically possible scenario & ever since their official recommendation was that breastfeeding moms should never fast, get 3 square meals a day & eat 60-70 carbs a day to avoid it. So I guess there is some validity to that old recommendation to avoid such rare cases."

If one was to take this advice, what food source of carbohydrate would be best for the extra 30-50g? It seems to me from the studies I've read that cow's milk would be ideal, at least for part of it, IYTI. In the first place, the galactose in milk, 50% of its carbohydrate content, will be directly incorporated into breast milk with little effect on blood sugar (this is likely true of glucose as well, as when a fat-adapted athlete sips a glucose gel in the middle of an endurance event, and the sugar is skimmed off the top into the muscles and does not interfere with fat adaptation).[7] Milk also supplies glycerol and short- and medium-chain fats, and as short-chain and medium-chain fats are not present in large amounts in body stores and are synthesised by the breast, with glycerol and glucose as potential substrates, including them in the diet by eating full fat dairy foods seems prudent.[8, 9] Galactose is also present in appreciable amounts in non-starchy leaf and root vegetables, especially beetroot and celery.

Mahommad, Sunemag and Haymond have provided most of the experimental research into the metabolic pathways involved in lactation and its adaptation to fasting and low carb diets that I have drawn on. They have tested the effect of a hypocaloric low carb high fat diet (1800 kcal, 31%E or 137 g/d CHO) on weight loss and milk composition during lactation in a short crossover study (n=7).[10]
The high fat diet included eggs, butter, cheese and cream. The average infant in this study consumed 486 kcal of milk, including 44g lactose, 11g protein, and 29g fat per day, on the low-carb diet (this was not significantly different from the milk in the high carb arm).

There are other considerations when using low carb purely for weightloss that are discussed on this La Leche League post.[link] In particular the warning to avoid sudden weight loss because of potential mobilization of persistent environmental toxins stored in body fat makes me speculate whether this effect played a role in the sickness that stopped some of the subjects in the case studies from eating. We'll never know of course.

Some of the foods commonly eaten on LCHF diets, such as fermented meats, shellfish, pre-packed salads and some cheeses, present a listeria risk during pregnancy and need to be avoided.  There is a list of these foods here.

To summarise - 

- Ketoacidosis can occur on rare occasions due to sudden weight loss or inability to eat while breastfeeding on a variety of diets - undereating should thus be avoided. 

- The increased demand for sugars for milk lactose synthesis may play a role in the strict LCHF cases but this need is small (the Atkins recommendation of 60-70g/day would cover it). 

- Sugars from milk and some non-starchy vegetables and medium-chain fats from full-fat dairy can be incorporated into human milk with minimal effects on glucose tolerance.

- No cases of LCHF ketoacidosis or lactation ketoacidosis in type 2 diabetics (or any diabetics) could be found, possibly because type 2 diabetics have higher blood glucose and glycerol and higher fasting insulin than non-diabetics.

- Adapting to a very low carbohydrate ketogenic diet is best done when physiological and emotional stresses in one's life are minimal, if this is at all possible. Adaptation to other degrees of carbohydrate restriction that are effective for weight control and metabolic health presents minimal challenges.

From a Mexican science paper - look what your dietary guidelines have done America.

[1] Learning from errors. A severe case of iatrogenic lactation ketoacidosis. Szulewski A, Howes D, Morton AR. BMJ Case Reports 2012; doi:10.1136/bcr.12.2011.5409

[2] Ketoacidosis during a Low-Carbohydrate Diet. Shah, P, Isley, WL. N Engl J Med 2006; 354:97-98 January 5, 2006; doi:10.1056/NEJMc052709 

[3] Ketoacidosis associated with low-carbohydrate diet in a non-diabetic lactating woman: a case report. von Geijer L, Ekelund M. Journal of Medical Case Reports (2015) 9:224 doi: 10.1186/s13256-015-0709-2

[4] A Case of Lactation "Bovine" Ketoacidosis. Heffner AC, Johnson DP. The Journal of Emergency Medicine, Vol. 35, No. 4, pp. 385–387, 2008. doi:10.1016/j.jemermed.2007.04.013
[5] Precursors of hexoneogenesis within the human mammary gland. Mohammad MA, Maningat P, Sunehag AL, Haymond MW. Am J Physiol Endocrinol Metab 308: E680–E687, 2015. doi:10.1152/ajpendo.00356.2014

[6] Lactation Intensity and Postpartum Maternal Glucose Tolerance and Insulin Resistance in Women With Recent GDM. The SWIFT cohort. Gunderson AP et al.  Diabetes Care January 2012 vol. 35 no. 1 50-56. doi: 10.2337/dc11-1409
[7] Galactose promotes fat mobilization in obese lactating and nonlactating women.
Mohammad MA, Sunehag AL, Rodriguez LA, Haymond MW. Am J Clin Nutr. 2011 Feb;93(2):374-81. doi: 10.3945/ajcn.110.005785. Epub 2010 Dec 1.
[8] De novo synthesis of milk triglycerides in humans. Mohammad MA, Sunehag AL, Haymond MW. Am J Physiol Endocrinol Metab. 2014 Apr 1; 306(7): E838–E847.  doi:  10.1152/ajpendo.00605.2013

[9] Acute effects of dietary fatty acids on the fatty acids of human milk.
Francois CA, Connor SL, Wander RC, Connor WE. Am J Clin Nutr. 1998 Feb;67(2):301-8.

[10] Effect of dietary macronutrient composition under moderate hypocaloric intake on maternal adaptation during lactation. Mohammad MA, Sunehag AL, Haymond MW. Am J Clin Nutr June 2009 vol. 89 no. 6 1821-1827. [link]

Do moderate ketone levels from low carb protect against symptomatic hypoglycemia in type 1 diabetes? A relevant case study.

Before many can know something, one must know it. 

- Dr Stockmann, in Ibsen's An Enemy of the People.

One of the benefits of a very low carbohydrate diet for type 1 diabetes is a much lower rate of hypoglycemic episodes, because of the need for less insulin, lower insulin doses and longer acting insulin overall (a greater proportion of the insulin used is basal dose). This is predicted by the laws of small numbers.
It is also reported anecdotally that hypoglycemia, when it does occur by the meter, is often non-symptomatic. The milder symptoms aren't easily ignored, and severe hypoglycaemia can lead to seizures, unconsciousness, and death. This is because the brain requires glucose at a steady rate. The brain can also use ketone bodies, which are generated by the liver from the incomplete combustion of fatty acids and some amino acids on a low carb diet. On the Bernstein diet carbohydrate intake is around 30g/day and this would provide deep ketosis if the protein intake were not high. As it is, the ketone level in someone on the Bernstein diet has been reported as 2 mmol/L.

Is this enough to protect against symptoms in insulin-induced hypoglycaemia? Does an oversupply of insulin suppress ketogenesis in the absence of carbohydrate?

These questions are impossible to test in someone with Type 1 Diabetes, no ethics committee would approve the experiment.

Luckily a case study has just been published that seems to answer them. Thanks to Bill Lagakos @caloriesproper for tweeting this.

Ketogenic diet in a patient with congenital hyperinsulinism: a novel approach to prevent brain damage. 
Maiorana, A, Manganozzi, L, Barbetti, F, Bernabei, S, Gallo, G, Cusmai, R, Caviglia, S, Dionisi-Vici, C. Orphanet Journal of Rare Diseases 2015, 10:120  doi:10.1186/s13023-015-0342-6

The subject is a child with congenital hyperinsulinism, that is, she has chronic high insulin levels (without apparent insulin resistance) and eating carbohydrate to relieve hypoglycaemia causes a further increase in insulin.

In addition to increased peripheral glucose utilization, dysregulated insulin secretion induces profound hypoglycemia and neuroglycopenia by inhibiting glycogenolysis, gluconeogenesis and lipolysis. This results in the shortage of all cerebral energy substrates (glucose, lactate and ketones), and can lead to severe neurological sequelae.
A child with drug-resistant, long-standing CHI caused by a spontaneous GCK activating mutation (p.Val455Met) suffered from epilepsy and showed neurodevelopmental abnormalities. After attempting various therapeutic regimes without success, near-total pancreatectomy was suggested to parents, who asked for other options. Therefore, we proposed KD in combination with insulin-suppressing drugs.We administered KD for 2 years. Soon after the first six months, the patient was free of epileptic crises, presented normalization of EEG, and showed a marked recover in psychological development and quality of life.

Note the points that a) insulin suppresses ketogenesis by suppressing lipolysis, b) neurologists know that lactate is a cerebral energy substrate.

At the age of 3 years laboratory tests were performed and showed hypoglycemia with hyperinsulinemia (blood glucose 1.95–2.3 mmol/L; plasma insulin 5.5–10.2 μUI/ml).

 A further association with slow-release carbohydrate to drugs did not elicit any clinical improvement, and the patient continued to present hypoglycemic episodes (0.5–1.6 mmol/L), seizures and absence epilepsy regardless of glycemic values, that required frequent hospitalizations.

Furthermore, at the age of 10 years the patient quickly gained 11.5 kg within 12 months, becoming mildly obese (BMI z-score: >97 th centile). Overall, her quality of life was very poor. The lack of response to drug therapy with risk of permanent and severe brain sequelae made us to consider a near-total pancreatectomy, that was discussed with parents with the warning of no guarantee to achieve normoglycemia and of the increased hazard of secondary diabetes. At parents’ request to avoid surgery, we then proposed a trial with KD, explaining that it was aimed to prevent neuroglycopenic epilepsy and to improve neurological status by providing ketone bodies as an alternative energy source for neurons, as seen in GLUT1 deficiency.


Six months after KD was started, maintenance of blood ketones between 2–5 mmol/L (Fig. 1, panel a) fully resolved neuroglycopenic signs with parallel disappearance of both epileptic crisis and absence epilepsy, despite blood glucose levels permanently below 5.5 mmol/L even after meal, and close to 2.2–2.7 mmol/L most of time (Fig. 1, panel b). EEG improved and became normal within the first year on KD, showing no alteration even during episodes of hypoglycemia (Fig. 2). During the first 6 months of KD the patient lost 9 kg and her BMI normalized. Psychological evaluation revealed a strengthening of social, cognitive and verbal capacities (Fig. 3). The child and her family reported an improvement of physical and psychosocial well-being, reduction of fear of hypoglycemic symptoms and awareness of a lower risk of neurological injury, with an overall amelioration of the quality of life related to the management of disease. Diazoxide was discontinued, and currently the patient is given octreotide, reduced to 25 μg/kg/day, without any neuroglycopenic symptoms. KD was well tolerated over a period of 24 months, with no side-effects and no changes in laboratory tests.

Thus we see some relevant findings - excessive levels of insulin doesn't suppress ketogenesis completely on a ketogenic diet (but I would really like to know what the insulin levels were on the KD). Ketones rise steadily proportionate to the ketogenic ratio, there is no cut-off. Ketones can replace glucose in insulin-induced hypoglycaemia, as expected - and were still present in sufficient amounts to do so. Insulin doesn't cause weight gain in the absence of carbohydrate.
An interesting question is whether the insulin resistance that is supposed to be caused by a high-fat diet played a role in this child's recovery by increasing lipolysis, as shown by the weight loss.

This is a neat study that confirms the anecdotal accounts that hypoglycemic episodes in type 1 diabetes are less symptomatic on a very low carb diet (that they are much fewer is already confirmed). Of course we would like more information about this, but this young girl's ordeal, and her convincing recovery, is compelling evidence that we haven't been barking up the wrong tree, given the difficultly of gathering any evidence at all about what happens in this situation.



For support and information about the Bernstein Diet for type 1 diabetes, join the TypeOneGrit group on Facebook. For Dr Bernstein's book, look here. For more information, see Dr Bernstein's YouTube Channel "Dr Bernstein's Diabetes University".

Before many can know something, one must know it. 

How dairy fats and coconut protect against type 2 diabetes

Type 2 diabetes, in the etiology laid out by Professor Roy Taylor, is (in its usual form at any rate) a condition of fat accumulation in the pancreas, liver, and muscle cells, which causes insulin resistance, hyperinsulinaemia, hyperglucagonaemia, and a vicious cycle of glucotoxicity and lipotoxicity.[1]

It is thus one of a constellation of associated lipid accumulation disorders connected with hyperinsulinaemia, the others including atherosclerosis, NAFLD, obesity. That these conditions are linked in some way was recognised as early as the 1880s by the great German physiologist Wilhelm Ebstein, the father of the modern LCHF diet.[2]

In the recent saturated fat and disease meta-analysis by de Souza et al, higher intake of ruminant trans-palmitoleic acid, a marker of dairy fat consumption, was inversely associated with type 2 diabetes (0.58, 0.46 to 0.74).[3] This is consistent with many studies of serum biomarkers of dairy fat consumption, also including odd-chain saturated fatty acids.

The recent results from Malmö, the third largest city in Sweden, give more detail about these correlations. The Malmö Diet and Cancer cohort was studied using a 7-day food diary and a 1 hour interview as well as an FFQ. This makes the results more reliable than other epidemiological diet studies, which normally use only the FFQ. In Malmö, greater consumption of dairy fat (including butter) had a protective association with T2D. The association was strongest for shorter-chain fatty acids (from 4:0, butyrate, to 14:0, myristic acid) and there was also a protective effect of a higher ALA/LA ratio.[4]

In a separate analysis of the Malmö cohort, it was found that adherence to dietary recommendations to limit saturated fat to 14% or less of energy was associated with a 15% increased risk of T2D in men and a slightly smaller increase in women. There was a small association in men between adherence to recommendations to limit added sucrose and T2D.[5] (I see a future paper here titled “food sources of sucrose may clarify the inconsistent role of dietary sucrose intake for incidence of type 2 diabetes”. After all, chocolate consumption has beneficial associations not seen with sugar sweetened beverages.)

Is there some simple, mechanical explanation that begins to explain the relationship? If T2D is the result of excess lipid storage, are some lipids easier to store than others? NAFLD research suggests that short- to medium-chain fatty acids are not easily stored. Wistar rats fed coconut oil under NAFLD-generating conditions ate an incredible 143% extra calories across the board with no increase in hepatic lipid accumulation, while butter-fed rats managed an extra 30%.[6]

A team led by George Bray looked at rates of fatty acid oxidation in humans, and made two findings - 1) the shorter the chain length of a saturated fat, the faster the rate of oxidation, 2) the more double bonds in an unsaturated fat, the faster the rate of oxidation. Thus, lauric acid (12:0) was oxidised at a much higher rate than stearate (18:0), and ALA (18:3) was oxidised at a faster rate than LA (18:2).[7]
The faster a fatty acid is oxidised the harder it is to store; this phenomenon discourages lipid accumulation, with benefits to the risk of lipid accumulation disorders.

A second consideration is that fat displaces carbohydrate in the diet, and carbohydrate is the nutrient that, by inducing insulin secretion, increases lipid synthesis and lipid conservation, something that (without the insulin bit) Wilhelm Ebstein understood in the 1880s. Levels of serum triglycerides are directly associated with %E from carbohydrate, and this triglyceride component is the source of pancreatic fat in the model of Professor Roy Taylor.[1]

A third consideration is that saturated fats are resistant to peroxidation and oxidative stress plays a role in promoting beta-cell failure. Saturated fats are protective against beta-cell failure in the alloxan-treated rat.
http://caloriesproper.com/diet-diabetes-and-death-oh-my/

References

[1] Taylor, R. Type 2 Diabetes. Etiology and reversibility. Diabetes Care April 2013;36(4): 1047-1055

[2] Wilhelm Ebstein. Corpulence and its treatment on physiological principles. 1882. https://archive.org/details/corpulenceitstre00ebst

[3] de Souza, RJ, Mente, A, Maroleanu, A, Cozma, AI, Ha, V, Kishibe,T, et al.  Intake of saturated and trans unsaturated fatty acids and risk of all cause mortality, cardiovascular disease, and type 2 diabetes: systematic review and meta-analysis of observational studies. BMJ 2015;351:h3978

[4] Ericson, U, Hellstrand, S, Brunkwall, L, Schulz, C-A, Sonestedt, E, Wallström, P, et al. Food sources of fat may clarify the inconsistent role of dietary fat intake for incidence of type 2 diabetes. AJCN 2015;114.103010v1

[5] Sonestedt, E et al. A high diet quality based on dietary recommendations does not reduce the incidence of type 2 diabetes in the Malmo Diet and Cancer cohort. EADS2015 ePoster #322 http://www.easdvirtualmeeting.org/resources/a-high-diet-quality-based-on-dietary-recommendations-does-not-reduce-the-incidence-of-type-2-diabetes-in-the-malmo-diet-and-cancer-cohort--3

[6] Romestaing, C, Piquet, MA, Bedu, E, Rouleau, V, Dautresme, M, Hourmand-Ollivier, I et al. Long term highly saturated fat diet does not induce NASH in Wistar rats. Nutr Metab (Lond). 2007; 4: 4

[7] DeLany, JP, Windhauser, MW, Champagne, CM, Bray, GA. Differential oxidation of individual dietary fatty acids in humans. Am J Clin Nutr October 2000;72(4):  905-911

This Mendelian Randomisation - I think it does not mean what you think it means.

"LDL may or may not correlate to cardiovascular outcomes,”

-  Dr. Kim Allan Williams, president of the American College of Cardiologists

“God, grant me the serenity to accept the things I cannot change,

The courage to change the things I can,

And the wisdom to know the difference.”

- Reinhold Niebuhr’s Serenity Prayer

Last year there was a good discussion of the saturated fat issue on Otago University's Public Health Expert blog (here) that continued into the comments.

David Brown pointed out that

"Evaluation of the overall health effects of saturated fat requires consideration of markers in addition to LDL-cholesterol. Isocaloric replacement of carbohydrate with any type of fat results in decreased triglycerides and increased HDL-cholesterol, the effect on HDL-cholesterol being greater for saturated fat compared to unsaturated fat. Reductions in saturated fat also adversely affect HDL subpopulations by decreasing larger HDL2-cholesterol concentrations, whereas increases in saturated fat increase this antiatherogenic fraction. "

In response to this, Tony Blakely, professor of public health, commented

"A note of caution on HDL. An important and massive analysis of many studies published in the Lancet in 2012 found no association of HDL with myocardial infarction (or heart attacks). This study used a genetic technique call Mendelian Randomisation, which strips away (in theory, and I believe in practice in this paper) all the confounding that plagues observational studies. Thus, it appears that it is LDL – not HDL – that has a causal association with coronary heart disease.

Reference: Voight BF, Peloso GM, Orho-Melander M, et al. Plasma HDL cholesterol and risk of myocardial infarction: a mendelian randomisation study. The Lancet 2012;380(9841):572-80 doi: 10.1016/s0140-6736(12)60312-2[published Online First: Epub Date]|."

At the time I thought fine, when the evidence runs against your hypothesis, invent a new statistical method that obscures the fact. Nothing to see here. Then a few weeks ago Bill Barendse tweeted a paper that used Mendelian Randomisation and Bill to put it mildly has expertise in the field of genetics, so I thought it might not hurt to look into this again.

Though I am doing so in a superficial way, using logic rather than any deep understanding of the genome, I want to question whether the idea that HDL is not causal in CVD, if it is true, should make any difference at all to how we interpret risk factors or the effect of dietary or lifestyle changes.

Mendelian Randomisation is a method for identifying drug targets. If people with X polymorphism have the same cardiovascular risk as everyone else, then there's not much point in developing a drug that targets X. Fair enough - drugs that elevate HDL by tweaking the enzymes associated with it have been a big disappointment.

The problem is, the cardioprotective association of HDL remains for all that. And, as for "all the confounding that plagues observational studies", HDL is measured directly, whereas LDL is calculated in diverse ways; and it's hard to see how confounding applies to blood tests in the sense that it applies to diet epidemiology. The examples here involve controlled conditions and short-term hard outcomes in secondary prevention. This is as good as it gets - high HDL is indeed a solid marker for cardioprotection and lots of other good things (albeit the HDL increase in response to alcohol probably weakens this in some populations). Does it matter whether the HDL particle itself is protective in a causal fashion? (and, if the particle that removes cholesterol from foam cells is really a dud, where does that leave the lipid hypothesis?)
Should we rely on LDL alone to assess cardiovascular risk? (TG isn't convincingly associated with risk in Mendelian Randomisation either) Well, not if we have those other CVD risk factors, insulin resistance or type 2 diabetes.

"When compared with IS, the IR and diabetes subgroups exhibited a two- to threefold increase in large VLDL particle concentrations (no change in medium or small VLDL), which produced an increase in serum triglycerides; a decrease in LDL size as a result of an increase in small and a reduction in large LDL subclasses, plus an increase in overall LDL particle concentration, which together led to no difference (IS versus IR) or a minimal difference (IS versus diabetes) in LDL cholesterol; and a decrease in large cardioprotective HDL combined with an increase in the small HDL subclass such that there was no net significant difference in HDL cholesterol. We conclude that 1) insulin resistance had profound effects on lipoprotein size and subclass particle concentrations for VLDL, LDL, and HDL when measured by NMR; 2) in type 2 diabetes, the lipoprotein subclass alterations are moderately exacerbated but can be attributed primarily to the underlying insulin resistance; and 3) these insulin resistance-induced changes in the NMR lipoprotein subclass profile predictably increase risk of cardiovascular disease but were not fully apparent in the conventional lipid panel."

("LDL may or may not correlate to cardiovascular outcomes")

Garvey WT et al. Effects of insulin resistance and type 2 diabetes on lipoprotein subclass particle size and concentration determined by nuclear magnetic resonance. Diabetes. 2003 Feb;52(2):453-62.

Are there other interpretations of Mendelian Randomisation in the literature?

"Observed associations between serum CRP and insulin resistance, glycemia, and diabetes are likely to be noncausal. Inflammation may play a causal role via upstream effectors rather than the downstream marker CRP."

Brunner EJ et al. Inflammation, Insulin Resistance, and Diabetes—Mendelian Randomization Using CRP Haplotypes Points Upstream. Plos Medicine August 12, 2008 DOI: 10.1371/journal.pmed.0050155

In other words, those markers that don’t have genetic links to the incidence of a condition should be considered as downstream effects of the true cause. A drug that blocks CRP synthesis won’t prevent diabetes (or any other inflammatory condition) – why should it? Most likely CRP is just doing its job, and things would not suddenly be brilliant if it was removed.
Let’s walk this idea back to lipids and CVD. The TG/HDL ratio isn’t determined by your lipid genes, it’s a downstream effect of dietary carbohydrate (non-genetic) and insulin resistance (genes linked to IR and hyperglycaemia do correspond to CVD). The association between LDL and cardiovascular risk is modified by carbohydrate which increases TG-rich VLDL, the end product of which is the small, dense LDL particle, which is cleared more slowly than larger LDL particles and is thus exposed to peroxidation. Another effect of having a high output of TG-rich VLDL being that HDL gets loaded with TGs and is cleared from circulation faster (hence high TG, low HDL). Half of your LDL-associated risk can be traced to genes (like ApoE4) which you can try to tweak with drugs if you like, and half belongs to Beta-apolipoprotein, sdLDL, oxLDL etc, which are modified by the carbohydrate factor. Saturated fat effects on VLDL and LDL may differ depending on the foods they are in or the other macronutrients , especially carbohydrate.

From Siri-Tarino et al 2015

Your liver is downstream from your gut and has first pass at the nutrients you absorb there; its uptake of fats, sugars and proteins determines the triglycerides, cholesteryl esters, and apolipoptoteins the liver produces and its types of HDL and LDL species. Genetics has more influence on the LDL species than on the HDL or TG, and if you are insulin-resistant the effect of high-carbohydrate diet on HDL, TG-rich VLDL, and atherogenic LDL subspecies is magnified; this is the pathology that hyperlipidaemia, MetSyn, and diabetic lipid patterns have in common.

This is how Mendelian Randomization of LDL and HDL was presented in a recent butter and cholesterol paper [here]
“The LDL-cholesterol concentration is a true risk factor for CVD. A meta-analysis of 26 trials showed that, for every 1-mmol/L reduction in LDL cholesterol, there was a 20% relative reduction in deaths that were due to coronary heart disease (RR: 0.80; 99% CI: 0.74, 0.87) (30). Thus, our result of an increase in LDL-cholesterol concentration of 0.16 mmol/L was not negligible. In addition, butter resulted in a concomitant increase in HDL cholesterol compared with the habitual diet. An increase in HDL cholesterol of the butter diet rich in long-chain SFAs was expected because these SFAs are known to increase HDL cholesterol…
 According to the literature, the HDL cholesterol concentration is associated with a protective effect on CVD (31, 32). However, studies that used Mendelian randomization showed that genetically decreased HDL cholesterol did not increase risk of myocardial infarction and questioned a causal association between the HDL concentration and CVD (33–35). Thus, it is necessary to be careful with interpreting low HDL-cholesterol concentrations as a CVD risk factor. However, as a marker of cardiovascular health, changes in HDL cholesterol concentrations need to be included when interpreting the effect of SFAs in the diet. It is possible to speculate that an unbeneficial increase in LDL cholesterol may partly be counteracted by the beneficial effect of SFAs on HDL cholesterol, which suggests that dairy and saturated fat may be less harmful in relation to CVD than previously thought, as reported in recent meta-analysis (8, 9).”Engel S and Thorstrup T (2015) Butter increased total and LDL cholesterol compared with olive oil however resulted in higher HDL cholesterol than habitual diet. Am J Clin Nutr. ajcn112227

Another suggestion is that HDL functionality is the important variable. HDL functionality is increased by CLA in butter and ruminant fat, olive oil polyphenols,[1] and the action of vitamin E (found in nuts and vegetable oils and other sources of linoleic acid) on protein kinase-C.[2] As substitution of all fats for carbohydrates tends to raise HDL, there will be a correspondence between intake of natural fats, HDL, and HDL functionality. Polyphenols administered without fat reduce inflammation but do not increase HDL or HDL functionality.[3]

Thus there is evidence for a rather neat correspondence between the quality of dietary fat and the cardioprotection associated with HDL -

[1] Hernáez Á et al (2014) Olive oil polyphenols enhance high-density lipoprotein function in humans: a randomized controlled trial. Arterioscler Thromb Vasc Biol. 2014 Sep;34(9):2115-9. doi: 10.1161/ATVBAHA.114.303374. Epub 2014 Jul 24.[2] Mendez AJ et al (1990) Protein Kinase C as a Mediator of High Density Lipoprotein Receptor dependent Efflux of Intracellular Cholesterol (1990) Journal of Biological Chemistry Vol. 266, No. 16, Issue of June 5, pp. 10104-10111,199


[3] Nicod N et al (2014) Green tea, cocoa, and red wine polyphenols moderately modulate intestinal inflammation and do not increase high-density lipoprotein (HDL) production. J Agric Food Chem. 2014 Mar 12;62(10):2228-32. doi: 10.1021/jf500348u. Epub 2014 Mar 4.

From a 2014 paper that failed to find a causal relationship between HDL and CVD:

The estimates of LDL-C from instrumental variable analysis showed that a long-term genetically increased LDL-C, regardless of the analytical strategy used (unrestricted, restricted, or unrestricted score plus sequential adjustments) resulted in an increased causal OR for CHD, which is similar in magnitude to that reported in randomized trials of statin-lowering therapies in individuals at low risk of vascular disease1 and is further evidence of the validity of our various analytical approaches.For triglycerides, the findings for the unrestricted and restricted allele scores were concordant, with both showing association with CHD. However, the unrestricted score adjusted for HDL-C diminished the association to null.This could mean that a treatment that targets a triglyceride pathway that has no effect on HDL-C may not be beneficial, whereas a treatment that targets a triglyceride pathway that both reduces triglycerides and increases HDL-C could have a role in prevention of CHD events. An alternative explanation is that HDL-C could mark long-term triglyceride concentrations, but this hypothesis requires further investigation.

Holmes, MV, et al. Mendelian randomization of blood lipids for coronary heart disease. 2014. DOI: http://dx.doi.org/10.1093/eurheartj/eht571

Clearly, the complete implications of mendelian randomization for cardiovascular risk related to diet are far from clear.
But I'd putting money on this; these analyses don’t change the meanings of metabolic risk factors that are affected by diet and lifestyle, and if anything they support their usefulness as measures of improvement.