Discussion of low carb diets and cholesterol

** Mod edit to say that this post (and 6 others in this thread) have been moved from a T2 thread on weight loss here https://www.diabetes.co.uk/forum/threads/weight-loss.209977/#post-2774512 so as to keep that thread on track for the OPs question. **

The strong mainstream consensus among medical professionals is that a very low carb diet rich in saturated fat from full fat dairy, eggs and red meat causes elevated LDL cholesterol, which is an established causal factor for heart disease, which we are already at higher risk of as diabetics.




This forum has quite a strong lean towards some alternative ideas. I don't see anything wrong with sharing these theories, and it is possible that some very widely held and evidence supported scientific beliefs somehow turn out to be misguided, but it is irresponsible to present them as something more than fringe thinking.

I have done a keto diet for periods in the past, and it makes keeping good blood glucose control amazingly easy, as well as causing rapid fat loss. However, after learning more, I no longer think these benefits are worth the dangers.

What works for me now is cutting out most refined carbs like bread, pasta and rice, as well as full fat dairy, and instead eating mainly beans and pulses, lots of veg and fruit, fish, nuts and seeds, whey isolate and some chicken. I also eat a lot of oats, though as a T1 I do find they need a fair amount of insulin to counter, so would understand why some T2s avoid them.
I'm feeling great, happy digestion, getting leaner, performance in the gym is good, and my glucose control is the best it has ever been, plus my LDL is going down.

(and of course it should go without saying that the central thing in weight loss is always consuming less energy than you use)

I fully expected some pushback, as I'm familiar with how prevalent these views are within this particular forum. I'll try and respond to some, but as you see, in a short space of time, we have 7 responses in support of these views, with a mix of personal 'anecdata' and links to multiple studies, each of which will take some time to look at. I'm clearly in a minority here (and not by coincidence, diabetes.co.uk sells low carb programs), but I repeat, what I am sharing is the consensus view in mainstream science. This forum is very far from representative.

My own views on this have shifted - I actually used to be quite pro keto, and liked using it myself. My change in thinking is relatively recent, so I welcome the chance to refine it through some debate, but don't present myself as an expert on these matters.

Anyway, to try and address a few of the points-
The paper supporting the low carb program in lovinglife's link (of which one of the authors is incidentally employed by this site), reports the results of the diet, and cites another meta-study mentioning positive changes in HDL and trigs, but omits any mention of LDL (and it is specifically LDL I'm talking about here as the CVD risk factor).
I haven't dug down into the individual studies cited by that metastudy, but one thing to watch out for in some studies which show little change is that they use patients starting with high LDL. There's a big difference between starting in a bad place and not getting much worse and a diet which actually improves the bad ones and doesn't worsen the good ones.

As for anecdotes - results from one person are are not always indicative of overall patterns - individual responses to cholesterol are widely variable. It's not surprising that some people get lucky and don't follow the trends. If your LDL is good (I know <70 is the target for diabetics now), I can understand why you might be less bothered. The statistical data looks pretty clear to me though.

Also, I think with all this anecdata users are sharing, it's worth distinguishing between the variants of low carb diet different people are following re the sources of fats. I'd expect a big difference between those eating high dairy, red meat and eggs, compared to one more based on things like olive oil, avocados and nuts.

What I actually referred to in my post was "a very low carb diet rich in saturated fat from full fat dairy, eggs and red meat"
I see the question of whether some low carb diets could be healthy for some people as more open.

Bird digestion functions very differently to humans'. Cows build muscle while eating grass. I'd be cautious of thinking goose rearing is too relevant to human diet (also - I would say the defining feature of foie gras production is the force feeding of vastly more food than they need, grain just happens to be cheap).

The papers from Kenny will take more time to respond to properly, and at this rate I'm unlikely to keep up with everything, but a few quick thoughts from a skim:

The first paper seems to focus on the whole LDL-particle-size-matters-quantity-doesn't idea, when it comes to the lipids discussion, which I don't believe is well supported now. From what I've read, particle size does matter a bit in terms of how likely they are to cause heart disease, but not enough to ignore elevated levels.

The Nature paper actually gives evidence supporting the consensus view that high LDL causes greater all cause mortality and heart disease. That U-curve they talk about goes up sharply to the right!
I know there is some data linking very low LDL with mortality - though if I recall correctly (I'll have a look at some point for the studies), this correlation can be largely explained by people dying of other things having their levels drop as internal systems fail, rather than the low LDL causing death.

The BMJ paper re-examining an old study gives supporting evidence that saturated fat intake raises LDL. It compares it to a diet where saturated fat is just replaced with large quantities of vegetable oil(!) and finds that patients died at around the same rate. I'm unconvinced that provides any meaningful support for the meat+cheese vs olive oil+avocados question.

The NCBI Norwegian study again shows the U-curve, reinforcing the evidence that too high total cholesterol is harmful. It does not appear to address the effect of LDL levels specifically at all.

I only very briefly looked at the 'blue zones' Sardinia paper, but it looks like it's about a group of old people in a specific location who live to an old age despite some of them having elevated cholesterol.
It seems reasonable that a generally healthy lifestyle in good climate, with good social life, good quality food etc, could lead to longer life despite some raised LDL.

I think Gil also goes into it a bit here:

I'm not keen on providing youtube links as supporting evidence in general, but Gil Carvalho (https://www.youtube.com/@NutritionMadeSimple) is a medical doctor(in contrast with say Dave Feldman, a popular proponent of alternative lipid theories, who is a software engineer) who I think does evaluate the science in a fair and balanced way and with an open mind.

I doubt anyone would have an issue with saying that there is a strong medical consensus that smoking causes lung cancer, yet it's easy to find many cases of individuals who smoked heavily yet lived long healthy lives.

Likewise, high intake of saturated fat from animal sources doesn't necessarily elevate LDL in absolutely every single individual, just most of them. In a pro-low-carb setting like here, I think we maybe there is a skew towards people presenting their labs more when they support the forum view, but there are copious studies with large numbers of participants painting a statistical picture.

When presenting data, it's also important to consider all the factors that accompanied the change. Losing a significant amount of body fat could lead to a long term reduction in LDL, as a different thing from the other cholesterol effects of the diet that lead to that fat loss. In some people one effect might be bigger than the other and in others vice-versa.

For those considering high animal fat keto diets based on recommendations here, I'd urge them to take advice found here with a pinch of salt (or should that be a pinch of fibre?), consider a wider range of information sources outside the bubble, keep a close eye on their labs (and maybe check other things like ApoB, LipoA, calcium score to assess your individual risk level), and run from those that tell you LDL doesn't matter at all.