Sunday, March 28, 2010

Do chylomicrons clog your arteries?

The oral fat tolerance test (OFTT) challenges with a fixed dose of a fixed type of fat. It is very important that the type of fat used and the dose given remains consistent throughout a given study. It tells us completely different things to comparing the chylomicron response to oral loads of differing fat compositions. With a fixed dose of constant fat composition we are then specifically looking at the ability of an individual to use or store fat in general.

The rise in triglycerides after an OFTT is actually chylomicrons (I'll use this term from here on) and gives us an idea of how good we are are putting fat in to use or storage. If we take any standard american chap and make him eat a standardised fat load there will be a surge of chylomicrons in his bloodstream, starting at a couple of hours after eating and going on for a few more hours.

So what determines the size of the rise in chylomicrons? As Dr Volek has pointed out, a ketogenic diet for six weeks markedly reduces your chylomicron concentration after an OFTT, ie you clear the chylomicrons from your blood stream more rapidly. This should hardly be surprising. If you have been in ketosis for 6 weeks you are hardly going to be running your metabolism on sugar. Fat comes in, fat gets used. But it still has to be transported.

How do we transport bulk fat from our gut to our butt? Dietary fat (medium chain triglycerides excepted) is ALWAYS transported in bulk as chylomicrons. "Regulated" fat, for metabolic needs, is ALWAYS transported as free fatty acids. These FFAs can be attached to albumin as their transport molecule in the plasma or can be accessed directly from chylomicrons via lipoprotein lipase, at the sites on the vascular wall where FFAs are needed by the tissues. But we can never transport all of a bulk fat meal as free fatty acids. Even when we are in ketosis. In ketosis we clear our chylomicrons faster because we are using a lot of fat. But we can never get rid of them altogether.

So if apoB48 containing chylomicrons kill, then fat kills. Post prandial chylomicrons kill. Even ketogenic diets kill. Eat anything to generate chylomicronaemia above 100mg/dl and you can kiss your coronary arteries goodbye. Gulp.

But how do we actually know that chylomicrons kill? From Denmark of course. Land of the Danish Pastry.

Denmark hosts The Copenhagen Heart Study. Just take 13,981 people, measure their non-fasting chylomicrons (described as "remnants"), record the results and watch who dies of heart disease and all cause mortality (bit risky that last one, but cardiologists were so naive back in the 70s) over the next 30 odd years. If chylomicrons kill, the higher the chylomicron count after a random meal, the more people should die, especially of heart disease.

Here are the "event" and death Hazard Ratios for men. Women are similar if not worse:

Random chylomicrons after a routine Danish breakfast or lunch are going to kill you. Heart disease. ApoB48. Simple. QED. Convincing enough to a cardiologist!

Woooooaaaaah, just a minute. This is observational. What are we observing? We are observing a group of people and they looked like this when they entered the study:

Look at hypertension incidence, diabetes incidence, physical inactivity prevalence and BMI splurge. All increase across rising chylomicron quartiles! Chylomicrons even make you smoke! They don't seem to make you alcoholic, a little disappointing that last one.

And you just thought apoB48 just caused heart disease!

Does anyone recognise the metabolic syndrome in these patient characteristics? Well I think that the size of your chylomicron surge after an average Danish pastry is determined by how far in to metabolic syndrome you are. However, the authors corrected for all of these factors and STILL chylomicrons kill.

The baseline characteristics recorded were crude in the extreme. Heavy drinking is defined as more than one drink a week (drinking two or more times a week is heavy)! But the fascinating one is diabetes. The definition of diabetes is anyone self reporting themselves as such, anyone who mainlines insulin (perhaps some body-builders got accidentally included here?), sulpha drug usage or having a random post prandial glucose above 11mmol/l.

Quick repeat: Anyone randomly detected with a blood glucose over


(actually, over 198.2mg/dl) was classed as diabetic. Perhaps they missed a few diabetics in their multifactorial adjustments! But perhaps they don't think hyperglycaemia has anything to do with heart disease.

We know from Volek that if you eat a ketogenic diet your OFTT "improves". I'll just say that again. Elevated chylomicrons levels after a fat challenge reduce if you have been eating a VERY high fat diet. I think it is a reasonable extrapolation to say that a high carbohydrate diet might worsen OFTT results.

So in Denmark a high post prandial chylomicron count, which can be viewed as a surrogate marker for the metabolic syndrome, correlates positively with your risk of heart disease. The hallmark of metabolic syndrome is hyperinsulinaemia. If that hyperinsulinaemia is inadequate to maintain normoglycaemia in the face of carbohydrate consumption then HbA1c rises and other nasty hyperglycaemic stuff happens. And the bulk fat transport gets the blame. Certainly in Copenhagen and perhaps other places too!

Let's stop bashing those poor apoB48 molecules.


As on a number of other occasions, thanks to Dr Davis for pointing to both of these studies even if I completely disagree with his interpretation of what is happening in Copenhagen. Lipophobes have such strange yet fascinating ideas. But then I love Goth stuff too.

Friday, March 19, 2010

Butter, insulin and Dr Davis

Sigh. Okay, here we go. It's the weekend and I'll correct the typos when I get chance!

Better read Dr Davis' post here here to get the lie of the ground before reading this post. Now, before we get to the Spanish study, let's look at the insulogenic effect of cream (the closest I can find to butter in a study which, unlike the Spanish study, controlled its variables). Please bear in mind that cream contains small amounts of both casein and lactose. So does butter. BTW look what casein does to insulin. But it doesn't budge glucose levels (they should drop!) so there has to be a counter regulatory system here, glucagon was not measured. It's not relevant to the role of palmitic acid in the Spanish study, but it's interesting never the less.

Taken from Dandona's paper here. This is the effect of 300kcal of cream (equivalent to about 30g of butter) or the equivalent in casein calories:

Okay, on 300kcal of cream alone insulin "spikes" from 39.6pmol/l to 49.2pmol/l at 1h (remember the casein and lactose?) and then insulin drops below baseline at 2h and 3h.

During this period there will be palmitic acid in to the blood stream and muscles. Palmitic acid is the primary metabolic signal to switch from glucose burning to fat burning. Because essentially zero carbohydrate is supplied with cream there is neither a rise in blood glucose or in insulin.

The Spanish study uses about 40g of carbohydrate (22% of about 800kcal) with their fat load. It gets eaten along side just under 60 grams of fat or oil.

The rise in glucose is trivial for all groups. It is neither statistically nor biologically significant. We can ignore it.

Now, let's look at insulin. The full figure and caption is here

Butter is the black squares.

Obviously, the best meal for minimising insulin response is the control meal. That's the one with round dots. That's the one we should eat to maximise weight loss, if it is a simple matter of minimising insulin! Ah, but the control meal is 40g of carbohydrate and no fat at all! Eating just 40g of carbs before an eight hour fast drops your insulin levels all right, this is starvation! But does it allow weight loss? Calories in, calories out, 40g of carbohydrate is only roughly 170kcal.

To answer this we have to look at the free fatty Acids (FFAs):

The study started with FFAs around 500micromol/l in all groups. The carb-only 40g snack DROPPED FFAs from fasting levels of 500micomol/l down to 150micomol/l at 2hours and it took until 5 hours for FFAs to get back up even close to the fasting levels seen at the start of the experiment. After a 40g fat-free carbohydrate "snack" the only source of FFAs is lipolysis and we can say that the small 40g carb snack blunts lipolysis, and so weight loss (rather I should say fat loss), for 5 hours! And you're hungry too.

Now the butter group produced the least fall in FFAs while the insulin was elevated from the carbohydrate and also allowed the most sustained rise in levels of FFAs once the carbs were dealt with. The FFAs were still significantly elevated at the end of the study. The area under the curve for chylomicrons (no VLDLS involved in this study) is also bigger and peak chylomicron-triglyceride level is higher in the butter group than after any other fat meal.

Butter provides palmitic acid which is the physiological signal to switch from using glucose to using fat. It also provides medium chain triglycerides which will produce ketone bodies for a few hours, which also produce physiological insulin resistance and a rise in insulin in their own right.

So in the presence of 40 grams of carbohydrate extra insulin is need to maintain normoglycaemia. The insulin should inhibit lipolysis. It certainly does in the 40g carbohydrate snack group! What about the butter group? The butter provides plenty of FFAs to run metabolism on and storing some calories is no big deal. But does this insulin effectively store calories? What if the adipocytes become physiologically insulin resistant with palmitc acid in exactly the same way as muscle cells do?

You know, my mantra: The function of insulin is the inhibition of lipolysis.

This study makes it look as if it is not quite that simple.

In fact, you have to ask some interesting questions about exactly where all of these FFAs come from and where the chylomicrons go to. ALL of the fat meals provided the SAME number of fat calories, but the FFA levels in the butter group where ALWAYS higher than other fat meal groups, even before chylomicrons levels became different between groups. Now, FFAs do not pop in to existence merely to prove that butter is going to kills us through obesity. They come from SOMEWHERE. And chylomicrons. These are lower in the oil meal groups than in the butter meal group. Where are the chylomicrons going to? Because all fat based meals provide the same number of fat calories then either:

a) the butter group has to be allowing more lipolysis from adipocytes or from chylomicrons to get those extra FFAs. Lipolysis means fat loss.


b) the non-butter groups are allowing more fatty acid storage and less FFA release. Insulin sensitive fat cells store fat! Low FFAs means less fat release. You CANNOT burn fat without lipolysis!

As I see it butter produces sustained chylomicronaemia. The chylomicrons are used to provide FFAs to run metabolism on rather than going in to storage. There is no hyperglycaemia to glycate apoB48s, so who cares if they hang around. Either they (or possibly adipocytes) are supplying energy.

Oil based meals can only produce lower levels of chylomicrons if they are storing the chylomicron fat ON YOUR BUTT and lowered FFAs means the INHIBITION of lipolysis from chylomicrons or from YOUR BUTT.

There is no other explanation that I can see, whatever the insulin levels are. Take you pick.

More fat gain and/or less fat loss: The gift of olive oil!

To summarise:

Insulin controls bodyweight. Physiological insulin resistance modifies this.

Of course if you think apoB48 was evolved to kill you, run from the butter and knock back the vegetable oil/fish oil combination. Maybe cut a few calories too!

Me, I'll stick to FFAs and butter as my energy source.


BTW 40g is close to the total daily carb intake for a reasonable LC diet. After that it's fat on its own and, as we all well know, fat on it's own causes ZERO insulin spike and allows FFA run metabolism where fat can be both stored and accessed freely. Okay, add a little protein somewhere along the line.

Thursday, March 18, 2010

Statin plus fibrate sucks

Haven't read the full paper yet but the abstract says it all: Extra lipid lowering does nothing to alter the trivial benefits seen with statin therapy.

"Conclusions The combination of fenofibrate and simvastatin did not reduce the rate of fatal cardiovascular events, nonfatal myocardial infarction, or nonfatal stroke, as compared with simvastatin alone. These results do not support the routine use of combination therapy with fenofibrate and simvastatin to reduce cardiovascular risk in the majority of high-risk patients with type 2 diabetes."


EDIT from the full text, it's not the triglycerides either. Why not just eat less sugar as per Nigel's comment? Simple but scary! To me this is pretty convincing that trigs are harmless, they are a surrogate for carbohydrate/fructose...


Wednesday, March 17, 2010

Paradox: Obesity and heart failure

Just taking a break from the busy stuff going on away from the blog. This paper deserves a brief comment. Here is a mainstream discussion, thanks Elizabeth. [And here is the text for when the blog link goes down].

To begin:

"Being skinny confers no advantage when it comes to the risk of dying suddenly from cardiac causes"

That lead statement about being skinny is somewhat misleading. A 99% increase in risk is not quite neutral....

So how does a lipophile see this paradox?

Well, the first thing is that the population is pre-selected. They have (a) had a heart attack and (b) have cardiac muscle wastage.

It's a little pointless going through the detail of the studies linking hyperglycaemia to heart disease when Jenny Ruhl has very neatly collated the studies that matter. I have mentioned the role of hyperglycaemia in cardiac muscle apoptosis before. Ischaemia and apoptosis is a great recipe for heart failure.

So we can reasonably describe this population as a set of people who have been routinely achieving post prandial blood glucose levels in excess of 8.0mmol/l, ie about 140mg/dl, almost certainly for years.

How they have achieved this seems unimportant, what matters to me is that by definition this is a group of recurrently hyperglycaemic people. Their LDL cholesterol level is of no interest and, thankfully, does not get mentioned here.

Why do the skinny people do so badly?

These people are skinny. They don't eat huge amounts of calories, but what they do eat spikes their blood sugar. They may well not be hyperinsulinaemic. After all, they are not squirreling away fat and they are allowing their blood sugar to rise... Those suggest, if anything, a blunted insulin response. Is someone with a BMI of 21, who has just had a heart attack, going to set out to lose a serious amount of weight? No, I doubt it. Cutting the fat would probably be the standard advice, obviously cutting saturated fat specifically. But backing off on fat (that scary stuff which Ancel Keys told us causes heart attacks, and cardiologists still believe him!) will invariably lead to increased carbohydrate intake. We know these people already develop post prandial hyperglycaemia. Low fat means more hyperglycaemia. Hyperglycaemia = death.

What about the healthier fatties?

Obviously these people are recurrently hyperglycaemic too, again because they are in the same cardiovascular situation as the skinnies. But they must have been eating a bit more of everything in the past than the thin people. They have to have eaten more to (a) eat enough calories to move themselves around, (b) eat enough calories to provide enough for their basal metabolic rate and (c) eat enough fat to go in to storage in their adipocytes.

However, they do have considerable scope for weight loss and ANY reduction in calorie intake is likely to reduce carbohydrate calories somewhat, as well as fat calories. The two tend to go together. Any decrease in carbohydrate calories will reduce hyperglycaemia in a person who is proven to develop post prandial hyperglycaemia.

Even cutting fat can be good. Less omega 6 PUFA and less trans fats from the diet, both of which, during weight loss, get replaced by an excellent mix of saturated fat with some mono unsaturated fat from butt fat. Butt fat, once again, is an excellent and healthy source of calories.

I have no idea whether surviving a heart attack and having a pacemaker implanted in your chest might make you think about losing a little weight if you are officially obese. Perhaps it concentrates the mind a little. Or maybe overhearing what the doctor's skinny receptionist called you when she thought you were out of earshot might help!

Obesity should become protective when someone starts to use their stored fat, because human adipose tissue is a health resource which can usefully replace anything with a heart-healthy logo on its plastic wrapping. It doesn't spike glucose! The more weight you have to start with, the longer it takes before the corrosive effect of a low fat diet kicks in. This kicks in when you stop losing weight!

If someone is losing weight on a low fat diet there will obviously come a time when weight loss stops and, at that time, they then simply join the ranks of the initially skinny low-fat eaters but, of course, a fair few of those will be dead by then....

Leaves some space in the queue I guess.


Wednesday, March 10, 2010

Getting fat is good: Official

This paper (thanks Elizabeth):

"Gluttony, sloth and the metabolic syndrome: a roadmap to lipotoxicity"
Roger H. Unger and Philipp E. Scherer

doesn't seem to be published yet, so I'm not sure if this will actually be the abstract:

"Once considered divine retribution for sins, comorbidities of obesity (metabolic syndrome) are today attributed to obesity-induced metabolic defects. Here, we propose that obesity and hyperleptinemia protect lipid-intolerant nonadipose organs against lipotoxic lipid spillover during sustained caloric surplus. Metabolic syndrome is ascribed to lipotoxicity caused by age-related resistance to antilipotoxic protection by leptin.

"The wrath of God came upon them, and slew the
fattest of them. . ." 78th Psalm, Verse 31."

Great start, even I have to admit. It gets better.

I got as far as the this diagram before having to stop. Head banging is fine for rock concerts, not so good on a hard desk.

Why are Texans so fat? According to Messrs Unger and Scherer:

Woo hoo! It's gluttony and sloth. With the number of guns that there are supposed to be in Texas (I wouldn't really know) I'd be careful about throwing such generalised insults around! Though I guess insulting people in a scientific journal is a lot safer than doing the same thing in a bar in down town Dallas.

They are also, being diabetologist and/or obesity experts, utter lipophobes (see the bottom line and text of their illustration) and struggle manfully with trying to show lipids (and even, OMG, cholesterol) are directly toxic. As I say, these folks are not too bright. I doubt they would understand glucotoxicity if it kicked them in the pancreas. But then, they are probably both on a statin.

I briefly re drew their illustration for them:

I had to split the slide in two to fit in the basic cause at the top of slide 1!

The ability to confuse symptoms with causes is hysterical. They really should read Good Calories Bad Calories, but it might be a little technical (and dispiriting!) for them.

However the reason for this post is that we have, on page 3 under "Protective role of obesity", this AMAZING quote:

"Thus, we propose that adipogenesis delays, rather than causes, the metabolic syndrome induced by chronic caloric surplus."

This is ABSOLUTELY crucial. These people have finally gotten the message! And they are idiots! When a concept is so clear cut that even morons can see it, there really is hope for the world.


Then (thanks, Hege) there is George Bray. Bray is one of the architects of the current obesity epidemic. Probably believes in gluttony and sloth as Unger and Scherer do. Anyway, have a read at Michael Eades' post to find a little more about what Bray is like. If you haven't already got him on the list (come the revolution).

Bray is now blaming fructose for the obesity epidemic.

When dinosaurs move, well, there will be progress!

It's a good year so far!


Sunday, March 07, 2010

Lipoprotein(a) and the Fairies at the bottom of my garden

Just a brief giggle. You recall that a reduced fat diet, packed with plant antioxidants raises Lp(a). Without the plant toxins it works even better to raise Lp(a). This paradox is explained by the immense power of the low fat diet to physically tear Lp(a) out of atheromatous plaque and place it in to the circulation in antigenically recognisable form. It's healing. Snigger.

This is powerful medicine. I don't know how many covalent bonds the lysines and glutamines in apo(a) have made to the lysines and glutamines in fibronectin. Let's say quite a few. But the low fat diet appears to be able to tear several hundred of these covalent bonds apart and then it reassembles the apo(a) molecule in to its original shape in the plasma. Gasp in awe. Well that's incredible.

So incredible that anyone who believes it does not realise what incredible means.

Incredible means unbelievable. Better not to believe unbelievable things. These are the believers who published (just a little name and shame here):

Silaste ML, Rantala M, Alfthan G, Aro A, Witztum JL, Kesäniemi YA, Hörkkö S.

and the believers who editorialised:

Mohamad Navab; Srinivasa T. Reddy; Brian J. Van Lenten; Alan M. Fogelman

Even the Fairies at the bottom of my garden tell me not to believe stuff this stupid!


Saturday, March 06, 2010

Lipoprotein(a) and tissue transglutaminase

"Lp(a) levels are low at birth and rise significantly between 0 and 7 days post partum; in this newborn population, a continuous rise of the mean Lp(a) levels was observed until 180 days"

Well, that's it. The smoking gun of arteriosclerosis. As we know, arteriosclerosis is essentially non existent for humans in-utero and it starts soon after birth (in the 1950s anyway, we may do better at damaging unborn children nowadays). Lp(a) starts low at birth and rises soon after. You only need to be a cardiologist to see the obvious cause and effect here.

OMG it must be so embarrassing to realise that LDL-C has nothing to do with the process, it's been that nasty Lp(a) all the time!

So the obvious explanation must be that Lp(a) worms its way through the arterial endothelium and rips and tears the elastin fibres of the intima to shreds. Simple and utterly plausible. Probably carries a flick-knife.

Except for the tissue transglutaminase (TTG) issue.

We have all heard of this enzyme as an antibody-ligand when testing for coeliac disease, but what does TTG actually do?

Well there is an awful lot of information in this paper. This is what they looked at:

"Because of its intimate interaction with fibrin and/or cell surface oriented structures, we asked whether Lp(a) could be a substrate for transglutaminases (e.g. Factor XIII, tissue transglutaminase). These enzymes catalyze cross-linking between endo-gamma-glutaminyl and endo-epsilon-lysyl residues of proteins (16) resulting in irreversible homo or heteropolymerization of susceptible proteins."

Note the heteropolymerization. Hetero means mixed. If TTG really acts on Lp(a), it doesn't just bind it to more Lp(a), it will bind it to other TTG substrate proteins too. Many of the other TTG substrates are physically part of the cement between cells. Molecules like fibronectin.

"First, substrate specificity was compared with known transglutaminase substrates including fibronectin (17) and alpha 2-plasmin inhibitor (30) and substrates of poor or unknown transglutaminase susceptibility including IgG, alpha 1-proteinase inhibitor, and albumin. This showed that Lp(a) had amine acceptor qualities comparable to that of fibronectin and alpha 2-plasmin inhibitor where as IgG, alpha 1-proteinase inhibitor and albumin failed to incorporate significant amounts of DSC."

So Lp(a) will heteropolymerise with fibronectin:

"FXIII or tissue transglutaminase may catalyze cross-linking and deposition of Lp(a) to surface oriented structures (e.g. connective tissue matrix proteins) and/or cell surfaces."

If you are wondering what fibronectin really does, from wiki:

"fibronectin also binds extracellular matrix components such as collagen, fibrin and heparan sulfate proteoglycans" (as well as intergins). Oh, look, proteoglycans...

And integrins, from wiki:

"Integrins are receptors that mediate attachment between a cell and the tissues surrounding it, which may be other cells or the extracellular matrix (ECM). They also play a role in cell signaling and thereby define cellular shape, mobility, and regulate the cell cycle."

So TTG binds Lp(a), almost certainly to fibronectin, one the main proteins which glue our tissues together. Lp(a) is not randomly sticky. It is enzymatically and specifically integrated in to the matrix of exposed extracellular material in the wall of a damaged artery.

If we reject the hypothesis that Lp(a) circulates with an elastin-targeted flick-knife in its pocket, we can look at the specific and deliberate incorporation of Lp(a) in to a cross-linked network of tissue proteins. This looks like an emergency repair kit to me. Elastin is broken by mechanical stress. Birth, growth, pulsation of arteries, hypertension, relative movement of the supply artery against fixed arteries like the intercostals, branch points in arteries etc are all mechanical stressors. Broken elastin implies stretching of the artery beyond what the elastin will tolerate. Damage to the elastin comes with damage to the endothelium. Disrupted endothelium exposes extracellular matrix and needs a clot to patch it and the clot needs strengthening. An area of damage intrinsically means that the location was too weak. Adding some fibrous tissue and a stronger muscle around damage is an adaptive stratagem in an injury-disrupted provenly-weak arterial area.

Looking at Lp(a) as a repair kit you can make certain predictions, especially if the repair kit is rather helpful.

First is that not having any Lp(a) is bad. Well, we know that's the case.

Next is that having some Lp(a) is good, we know that too from the same graph.

If you are genetically well endowed (with repair kits) they will not show up as repair patches on your arterial wall unless you are actively damaging that arterial wall. ie High Lp(a), (genetic low kringle IV repeat numbers) means nothing if you are not damaging your arteries. Thanks to Kurt for that anecdote!

If you do lots of damage, you will need lots of repair kits. Lp(a) goes up with carb intake and down with saturated fat intake. Check DELTA.

EDIT: Also from DELTA, your body anticipates damage, or detects actual damage, if you replace the saturated fat of the SAD with monounsaturated fat. So it makes more Lp(a). That's it. You liver is worried by MUFA. Extra virgin olive oil is heart healthy? Dump it for lard or, better still, beef dripping! Less plant antioxidants for your liver to eliminate too.

If you do some thing stupid, like pushing your carb intake beyond what is acceptable (even as a non industrialised form of carbs), your liver will make extra repair kits and they will be both needed and used on your arteries. Remember the vegetarian Bantu farmers getting nearly 90% of their calories from complex carbohydrate? Blood pressure rises with age. Not so at 70% from carbs in the Bantu fishermen.

The more damage, the more repair kits get made, the more repair kits get used. The more damage, the more blood pressure rises with age.

Lp(a) rises soon after birth because birth is when pressure induced damage starts. Life, especially being born, is a damaging process. Some lives are more damaging than others.

Are you going to blame the rising blood pressure on the repair kits? Have you forgotten to take your statin today? You may also have forgotten where you left your blood pressure tablets.


BTW, obviously these stable fibrous arteriosclerotic lesions have nothing to do with heart attacks. Heart attacks happen in the elderly, not in infants!

Friday, March 05, 2010

Intellectual honesty vs obfuscation

This papragraph is taken from Eric Westman's paper on LC for managing diabetes. Thanks to Valtsu for the heads up. Concise, accurate, comprehensive, numerical:

"Prior to the study intervention, the mean ± SD dietary intake for both groups was 2128 ± 993 kcal, 245 ± 136 g of carbohydrate (46% of daily energy intake), 86 ± 33 g of protein (18% of daily energy intake), 88 ± 57 g of fat (36% of daily energy intake). Over the 24-week duration of the intervention, the LCKD group consumed 1550 ± 440 kcal per day, 49 ± 33 g of carbohydrate (13% of daily energy intake), 108 ± 33 g of protein (28% of daily energy intake), 101 ± 35 g of fat (59% of daily energy intake). In comparison, the LGID group consumed 1335 ± 372 kcal per day, 149 ± 46 g of carbohydrate (44% of daily energy intake), 67 ± 20 g of protein (20% of daily energy intake), 55 ± 23 g of fat (36% of daily energy intake). There was no difference in self-reported exercise between the groups: the mean number of exercise sessions per week increased from 2.0 ± 2.0 to 3.0 ± 2.0 for the LCKD group and from 2.2 ± 2.2 to 3.8 ± 2.9 for the LGID group (p = 0.39 for comparison)."

For anyone who has slogged through the Ben-Gurion study, including the full text, looking for exactly this information, it simply is not there. You can stop hunting now. In a weight loss study there is NO REPORT of the absolute calories consumed! Either the DIRECT group does not include anyone who can present data or they are too scared of their own data to actually present it! But for people with a LC bias and honest data, there is no need for fear. Just generate the data, let the truth speak and and expect to be ignored! But not for ever....

A few IMT changes in a LCKD group over 6m would be nice too! I guess they are on their way if Dr Westman has anything to do with it.


Wednesday, March 03, 2010

Intimal wall volume reductions with weight loss

Thanks to Chris for the heads up on this one.

Just briefly: Low fat, Mediterranean and low carbohydrate: Statistically significant reductions in carotid vessel wall volume for all of them, with no significant differences between groups, so long as you lose weight:

"with no differences in the low-fat, Mediterranean, or low-carbohydrate groups (-60.69 mm(3), -37.69 mm(3), -84.33 mm(3), respectively; P=0.28"

But the LC group lost more intimal vessel wall volume than the other two groups, even if this wasn't statistically significant. I've also not got access the information about what they mean by low carbohydrate. As we know this could be anything up to 150g/d in some people's book!

2010 could be a good year for LC and honesty!


A tale of two abstracts

I've kindly re written the abstract for the high fat, low carbohydrate vs semi starvation paper by Eckel's group of clowns in Colorado. I've not attempted to re-title it as the current title needs to be discarded rather than corrected. Especially describing a non significant rise in LDL cholesterol as "hypercholesterolaemia" is not simply incorrect, it is completely dishonest. These people are, like Black's group in Belfast, not stupid. They're just bent.

The original:

Background: Little is known about the comparative effect of weight-loss diets on metabolic profiles during dieting. Objective: The purpose of this study was to compare the effect of a low-carbohydrate diet (<20 g/d) with a high-carbohydrate diet (55% of total energy intake) on fasting and hourly metabolic variables during active weight loss. Design: Healthy, obese adults (n = 32; 22 women, 10 men) were randomly assigned to receive either a carbohydrate-restricted diet [High Fat; mean +/- SD body mass index (BMI; in kg/m2): 35.8 +/- 2.9] or a calorie-restricted, low-fat diet (High Carb; BMI: 36.7 +/- 4.6) for 6 wk. A 24-h in-patient feeding study was performed at baseline and after 6 wk. Glucose, insulin, free fatty acids (FFAs), and triglycerides were measured hourly during meals, at regimented times. Remnant lipoprotein cholesterol was measured every 4 h. Results: Patients lost a similar amount of weight in both groups (P = 0.57). There was an absence of any diet treatment effect between groups on fasting triglycerides or on remnant lipoprotein cholesterol, which was the main outcome. Fasting insulin decreased (P = 0.03), and both fasting (P = 0.040) and 24-h FFAs (P < 0.0001) increased within the High Fat group. Twenty-four-hour insulin decreased (P < 0.05 for both groups). Fasting LDL cholesterol decreased in the High Carb group only (P = 0.003). In both groups, the differences in fasting and 24-h FFAs at 6 wk were significantly correlated with the change in LDL cholesterol (fasting FFA: r = 0.41, P = 0.02; 24-h FFA: r = 0.52, P = 0.002). Conclusions: Weight loss was similar between diets, but only the high-fat diet increased LDL-cholesterol concentrations. This effect was related to the lack of suppression of both fasting and 24-h FFAs.

Corrected version:

Background: Much is known about the comparative effect of weight-loss diets on metabolic profiles during dieting, though our research group seem peculiarly ignorant of the literature. Objective: The purpose of this study was to compare the effect of a low-carbohydrate diet (<20 g/d) with a high-carbohydrate diet (55% of total energy intake) on fasting and hourly metabolic variables during active weight loss. Design: Healthy, obese adults (n = 32; 22 women, 10 men) were randomly assigned to receive either a carbohydrate-restricted diet which was unrestricted in calories or fat and was consumed to satiation [High Fat; mean +/- SD body mass index (BMI; in kg/m2): 35.8 +/- 2.9] or a severely calorie-restricted, low-fat diet (High Carb; BMI: 36.7 +/- 4.6) for 6 wk. A 24-h in-patient feeding study was performed at baseline and after 6 wk. Glucose, insulin, free fatty acids (FFAs), and triglycerides were measured hourly during meals, at regimented times. Remnant lipoprotein cholesterol was measured every 4 h. Results: Patients lost a similar amount of weight in both groups (P = 0.57), the excess weight loss in the calorie and fat unrestricted High Fat diet not reaching statistical significance. There was an absence of any diet treatment effect between groups on fasting triglycerides or on remnant lipoprotein cholesterol, which was the main outcome. Fasting insulin decreased (P = 0.03) only in the High Fat diet, and both fasting (P = 0.040) and 24-h FFAs (P < 0.0001) increased within the High Fat group, as is appropriate for a fatty acid based metabolic profile. Twenty-four-hour insulin decreased (P < 0.05 for both groups), the decrease within the High Fat group being twice that seen in the High Carb group. Fasting LDL cholesterol decreased in the High Carb group only (P = 0.003), representing an increase in the atherogenic sdLDL particle subgroup. In both groups, the differences in fasting and 24-h FFAs at 6 wk were significantly correlated with the change in LDL cholesterol (fasting FFA: r = 0.41, P = 0.02; 24-h FFA: r = 0.52, P = 0.002). Conclusions: Weight loss was similar between diets despite unrestricted calories and fat intake in the High Fat group. The high-fat diet did not significantly increase LDL-cholesterol concentrations (P = 0.13). High Fat diets increase LDL lipoprotein size non significantly, potentially decreasing atherogenicity, which is possibly related to the increase in both fasting and 24h FFAs.


Arteriosclerosis images (2): Models

All of these images are taken from Dr John Duguid's monograph The Dynamics of Arterisclerosis. I'm not sure they are available elsewhere, so here come some horrible copyright infringements.

Duguid never uses the term mucopolysaccharide or GAGs, but it is pretty clear from the pictures that this is what the hyaline substance in Duguid's rabbit images represents. The body replaces the fibrin of blood clots with GAG, as it does in the healing process of fractures. Oddly enough the parallel goes on to calcification too but I think I might be stretching a point there...

Please bear in mind that the pulmonary emboli are a model. They are not arteriosclerosis, but just look at the vascular wall changes over the months after embolisation. Obviously they are in the pulmonary artery as this is where all emboli from peripheral veins end up!

Duguid has a whole series of picture of microthrombi in human arterial sections which produce similar changes to the model but in minute amounts. Hence in the previous post the clot over the frayed elastic membrane does not look like the large fibrin particle in the rabbit artery here.

Just for a giggle I've also included two of his images of the cholesterol fed rabbit. Classic.

The titles are self explanatory. The fibrin clot is huge and it's the lump on the lower left side in the lumen of the artery. The paler granular stuff is just blood in the artery. All of the stains are H and E and don't show elastin (not needed as the model does not use intimal trauma to produce the clots).

We now know that circulating endothelial progenitor cells rapidly coat thrombi, or other nasties (like foam cells), so finding a covering by six days is not surprising.

By two months the lesion is looking more like a typical childhood arteriosclerosis lesion:

By six months the lesions do look quite like the mature lesions seen in young children:

I would suggest that there is still a lump of what is probably GAG, but the bulk of the lesion (bottom of the section) is mostly just fibrous thickened intima.

It would have been lovely to have Prussian Blue stained sections and Sudan Red sections to confirm the presence of GAG and the near absence of cholesterol, but you don't get what's not in the book.

Don't forget this is a model!

Now, for the current prefered model in cardiovasciular research, fanfare please for the Cholesterol Fed (poisoned?) Rabbit:

All of the bubbly stuff is cholesterol. There is no GAG, there is no fibrin. EPCs coat these foam cells as well as they do fibrin clots, hence the layers of epithelial cells between the foam cells.

And, if your rabbit survives long enough, what does its aorta look like? Here's one (on the left) I prepared earlier:

Duguid does say quite explicitly that this particular rabbit was an exceptional example. Perhaps it has FH as well as cholesterol poisoning!

Now, did this rabbit die of a heart attack?