Low-carbohydrate High-fat diets: Green flags and Red flags.

Fun with flags. But first, a poem!

Atkins Antidote


Eating low carbohydrate what threat that poses

Do my friends think I’m suffering from halitosis?

I’ve got these sticks for measuring ketoacidosis

I’m taking supplements but I don’t know what the dose is

I’m trying hard to keep in a state of ketosis

I’m not sure what the right amount of weight to lose is

I’m sure I’ve put on a pound just through osmosis

Is eating this way risking osteoporosis

Are my kidneys wrestling with metabolic acidosis

My store of liver glycogen I don’t know how low is

Who knows what the glycemic load of oats is

Does anyone know if I can eat samosas?


Ian Turnbull

I do. The answer's "No!" :-D

From https://forum.nationstates.net/viewtopic.php?f=23&t=13567&start=8925

The Green flags...

1. For a person with Insulin Resistance, an ad-libitum low-carb diet results in more weight loss than an ad-libitum high-carb diet.

See How low-carbohydrate diets result in more weight loss than high-carbohydrate diets for people with Insulin Resistance or Type 2 Diabetes , for an explanation.

2. For a person with Type 1 Diabetes Mellitus (T1DM), a lowish-carb (~100g/day), highish fat diet results in minimal disturbances to blood glucose levels and minimal bolus insulin doses.

See Diabetes: which are the safest carbohydrates? , to see which foods should comprise the ~100g/day. N.B. As ~50% of dietary proteins can be converted into glucose by gluconeogenesis, ~100g/day of slow-digesting proteins such as meats, eggs & cheeses can contribute ~50g/day of glucose towards the ~100g/day total.

3. For a person with LADA or MODY, see 2.

4. For a person with Type 2 Diabetes Mellitus (T2DM), a LCLF 600kcal/day Protein Sparing Modified Fast can normalise BG in 1 week and reverse T2DM in 8 weeks (provided there are sufficient surviving pancreatic beta-cells).

See http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3168743/
"After 1 week of restricted energy intake, fasting plasma glucose normalised in the diabetic group (from 9.2 ± 0.4 to 5.9 ± 0.4 mmol/l; p = 0.003)."

"Maximal insulin response became supranormal at 8 weeks (1.37 ± 0.27 vs controls 1.15 ± 0.18 nmol min−1 m−2)."

After 8 weeks, transition to a diet based on whole, minimally-refined animal & vegetable produce.

See also http://www.fend-lectures.org/index.php?menu=view&id=94

As Insulin Resistance is multi-factorial, ALL of the potential causes need to be addressed. Once this has been done, IR should be reversed, allowing restrictions on dietary carbohydrate intake to be lifted. See also Can supplements & exercise cure Type 2 diabetes?

The Red flags...

The low-carb diet is a temporary patch to ameliorate IR/IGT/Met Syn/T2DM, a bit like replacing a failed circuit-breaker by sticking a nail in its place, to allow the house to function while you fix the problem by buying a new circuit-breaker. Although the house functions fine with a nail in place of a circuit-breaker, you wouldn't want to spend the rest of your life without a working circuit-breaker protecting the house.

So, why do low-carbers seem to want to spend the rest of their lives using a temporary patch to ameliorate their IR/IGT/Met Syn/T2DM?

Long-term use of very-low-carb, very-high-fat diets is not recommended.

1. Cortisol level can gradually increase, resulting in increasing fasting BG level. See How eating sugar & starch can lower your insulin needs.

2. If you do too much high-intensity exercise, you may momentarily black-out, fall and hurt yourself. See "Funny turns": What they aren't and what they might be.

3. Some people seem to gradually go bat-shit crazy. See Can very-low-carb diets impair your mental faculties? Read the comments in https://www.facebook.com/TheFatEmperor/posts/1633434020253792. Do the behaviours of Ivor Cummins & Gearóid Ó Laoi seem normal to you?

4. Insulin Resistance is bad, mmm-kay? See Lifestyle-induced metabolic inflexibility and accelerated ageing syndrome: insulin resistance, friend or foe?

5. Dyseverything elseaemia isn't addressed. See Type 2 diabetes: between a rock and a hard place , Type 2 diabetes: your good signalling's gonna go bad and When the only tool in the box is a hammer.

6. Dietary deficiencies may develop. See Rigid diets & taking loadsa supplements to compensate for them.

7. High-fat diets with no energy deficit result in high postprandial TG's. Postprandial lipaemia is atherogenic. See Ultra-high-fat (~80%) diets: The good, the bad and the ugly.

There may be more but I'm knackered, so I'm Publishing!

Neovascularization/Neovascularisation: It doesn't ONLY cause CHD.

Another serendipitous moment.

From http://www.aao.org/theeyeshaveit/optic-fundus/retinal-neovascularization.cfm

After talking to someone with Age-related Macular Degeneration (AMD), I Googled the condition and spotted the word neovascularisation. This reminded me of Neovascularization of coronary tunica intima (DIT) is the cause of coronary atherosclerosis. Lipoproteins invade coronary intima via neovascularization from adventitial vasa vasorum, but not from the arterial lumen: a hypothesis.

So I did a search on PubMed for Neovascularization/Neovascularisation. This is what I got:-
http://www.ncbi.nlm.nih.gov/pubmed/?term=%28%28%22neovascularization,%20pathologic%22[MeSH%20Terms]%20OR%20%28%22neovascularization%22[All%20Fields]%20AND%20%22pathologic%22[All%20Fields]%29%20OR%20%22pathologic%20neovascularization%22[All%20Fields]%20OR%20%22neovascularization%22[All%20Fields]%29%20OR%20%28%22neovascularization,%20pathologic%22[MeSH%20Terms]%20OR%20%28%22neovascularization%22[All%20Fields]%20AND%20%22pathologic%22[All%20Fields]%29%20OR%20%22pathologic%20neovascularization%22[All%20Fields]%20OR%20%22neovascularisation%22[All%20Fields]%29%29%20AND%20%28%28hasabstract[text]%20AND%20%22loattrfree%20full%20text%22[sb]%29%20AND%20%222009/09/14%22[PDAT]%20:%20%222014/09/12%22[PDAT]%20AND%20%22humans%22[MeSH%20Terms]%29%20AND%20%28hasabstract[text]%20AND%20%22loattrfree%20full%20text%22[sb]%29%20AND%20%222009/09/14%22[PDAT]%20:%20%222014/09/12%22[PDAT]%20AND%20%22humans%22[MeSH%20Terms]%20AND%20%28%28hasabstract[text]%20AND%20%22loattrfree%20full%20text%22[sb]%29%20AND%20%222009/09/14%22[PDAT]%20:%20%222014/09/12%22[PDAT]%20AND%20%22humans%22[MeSH%20Terms]%29%20AND%20%28%28hasabstract[text]%20AND%20%22loattrfree%20full%20text%22[sb]%29%20AND%20%222009/09/14%22[PDAT]%20:%20%222014/09/12%22[PDAT]%20AND%20%22humans%22[MeSH%20Terms]%29%20AND%20%28%28hasabstract[text]%20AND%20%22loattrfree%20full%20text%22[sb]%29%20AND%20%222009/09/14%22[PDat]%20:%20%222014/09/12%22[PDat]%20AND%20%22humans%22[MeSH%20Terms]%29

Even after restricting results, there were 5253 results. Wow!

It would appear that AMD has a lot in common with CHD. Ischaemia/Ischemia (lack of oxygen) to tissues causes the body to produce an adaptive response by growing new blood vessels (neovascularization/neovascularisation). Unfortunately, the new blood vessels are a bit crap, and cause other problems to develop e.g. CHD & AMD.

Therefore, prevention is better than cure. The things that lower the RR for CHD may also lower the RR for AMD. See Cholesterol And Coronary Heart Disease.

Wheat, Constipation, Ischaemic Heart Disease, Type 1 Diabetes, Schizophrenia and Autism.

Did you see this coming?

Gliadorphin 7, from http://en.wikipedia.org/wiki/Gliadorphin

The above 7-peptide chain contains 3 molecules of proline (the pentagon with a "N" at one corner), just like:-

Bovine β-casomorphin 7, from http://en.wikipedia.org/wiki/Casomorphin

From Further research for consideration in 'the A2 milk case'.
"Prior to discussion it must be clarified that the hypothetical link between A1 consumption with autistic spectral disorder (ASD) and schizophrenia relates not to the cause of the condition but to the aggravation of symptoms associated with these neurological conditions. More specifically, the hypothesis states that the absorption of food-derived exomorphins such as beta casomorphin 7 (BCM 7) may aggravate symptoms associated with ASD or schizophrenia.

This hypothesis is the basis of 'dietary intervention' that excludes gluten and casein (Knivsberg et al., 2002) from the diet of ASD patients. The former, gluten, has been shown to release gliadamorphin, an exomorphin comparable in opioid activity to BCM-7. A number of laboratories in the United States and Europe offer urine tests, which determine the level of peptides including BCM 7 and other beta casomorphins to serve as an indication of the potential usefulness of dietary intervention in the treatment of ASD patients. One published study reports that a casein- and gluten-free diet was accompanied by improvement in 81% of autistic children within 3 months (Cade et al., 2000)."


According to What is gliadorphin?
"What is gliadorphin? Gliadorphin (also called alpha-gliadin or gluteomorphin) is a substance that resembles morphine. Ordinarily, this is a short-lived by-product from the digestion of gluten molecules (found in wheat, barley, rye, oats, and several other grains). Gliadorphin is very similar to casomorphin. Gliadorphin has been verified by mass spectrometry techniques to be present in unusual quantities in urine samples of children with autism, and are believed by many to be a central part of the system of causes and effects that cause autistic development. The most probable reasons for the presence of these molecules are:
* One or more errors in the breakdown (digestion) process caused by enzyme deficiency and/or
* Abnormal permeability of the gut wall (that would allow these relatively large molecules to enter the bloodstream from the intestine in abnormal quantities)."

Continued on Rheumatoid Arthritis: It's the food!

Cow's milk, Constipation, Ischaemic Heart Disease & Type 1 Diabetes.

Hat-tip to Jamie Scott and https://twitter.com/_Jamie_Scott/status/503383804686262272 , which led to A1 threat to NZ dairy.

From http://en.wikipedia.org/wiki/Milk

There are a few problems with feeding cow's milk to baby humans.

1. It contains bovine beta casein A1. During digestion, this is broken down into a 7-amino acid peptide chain beta-casomorphin7 (BCM7), which appears to cause issues (e.g. increased gut permeability) increasing the RR of diseases like Type 1 Diabetes and Ischaemic Heart Disease. See http://www.ncbi.nlm.nih.gov/pubmed/?term=A1[All%20Fields]%20AND%20%22beta-casein%22[All%20Fields]%20AND%20%22humans%22[MeSH%20Terms]%20AND%20%28hasabstract[text]%20AND%20%22humans%22[MeSH%20Terms]%29 A solution is to use A2 milk, or goat's milk which is apparently naturally A2. See also Further research for consideration in 'the A2 milk case'.


2. It's much higher in protein (4g/100mL) than human breast milk (1.1g/100mL), as baby cows are supposed to grow very rapidly, unlike baby humans. As 80% of the protein in milk is casein, and casein is joined to calcium as calcium caseinate, this increases the calcium intake, and too much calcium relative to magnesium is constipating. A solution is to increase magnesium intake, or dilute 1 part cow's milk with ~3 parts water & add some coconut oil, to get the fat content back up to 4.4g/100mL.

Continued on Cow's milk, Schizophrenia and Autism.

Chowdhury et al, More forests & more trees and more "Eureka!" moments with cheese.

Like Siri-Tarino et al, Forests & Trees and "Eureka!" moments, Chowdhury et al is a meta-analysis of many studies. See Association of dietary, circulating, and supplement fatty acids with coronary risk: a systematic review and meta-analysis. I don't have access to the full study, but Google Image Search found Figure 2.

From http://annals.org/data/Journals/AIM/929862/6ff2_Figure_2_RRs_for_coronary_outcomes_in_prospective_cohort_studies_of_circulating_fatty_acid.jpeg

All saturated fatty acids have a RR for CHD of 1.06 (95% CI 0.86 - 1.30).
∴ There's no association between saturated fat intake and the RR for CHD.

Before VLC'ers do a dance of joy, consider the Forest plot for individual saturated fatty acids.
Palmitic acid has a RR for CHD of 1.15 (95% CI 0.96 - 1.37).
Stearic acid has a RR for CHD of 1.23 (95% CI 0.93 - 1.61).

Red meat & saturated fats synthesised by DNL aren't looking too good. However...
Pentadecanoic acid has a RR for CHD of 0.94 (95% CI 0.67 - 1.32).
Margaric acid has a RR for CHD of 0.77 (95% CI 0.63 - 0.93).
Pentadecanoic acid and Margaric acid combined have a RR for CHD of 0.81 (95% CI 0.62 - 1.06).

What are Pentadecanoic acid & Margaric acid found in? The clue's in the title:-
Say cheese: saturated fat in dairy may protect against diabetes.

The article in the Telegraph is actually referring to Differences in the prospective association between individual plasma phospholipid saturated fatty acids and incident type 2 diabetes: the EPIC-InterAct case-cohort study.

Of the omega-3 fatty acids...
Eicosapentaenoic acid has a RR for CHD of 0.78 (95% CI 0.65 - 0.94).
Docosahexaenoic acid has a RR for CHD of 0.79 (95% CI 0.67 - 0.93).
Eicosapentaenoic acid and Docosahexaenoic acid combined have a RR for CHD of 0.75 (95% CI 0.62 - 0.89).

Of the omega-6 fatty acids...
Arachidonic acid has a RR for CHD of 0.83 (95% CI 0.74 - 0.92).

Of the trans-fatty acids...
Trans-oleic acid has a RR for CHD of 1.20 (95% CI 0.39 - 3.73).
Trans-linoleic acid has a RR for CHD of 1.36 (95% CI 0.83 - 2.22).

Historical perspectives on the impact of n-3 and n-6 nutrients on health, by Bill Lands.

Here's Fig. 1. from http://www.sciencedirect.com/science/article/pii/S0163782714000253

Relating tissue HUFA balance with blood cholesterol and heart attacks. Results from the 25-year follow-up in the Seven Countries Study [35] were discussed in an earlier review [10] which noted that “Food energy imbalances which elevate blood cholesterol may be fatal only to the degree that omega-6 (n-6) exceeds omega-3 (n-3) in tissue HUFA. Such evidence raises questions about the hypothesis that blood cholesterol levels cause CHD.” Northern Europe and Southern Europe have abbreviations “No.” and “So.”, respectively. The Figure is reprinted with permission of the publisher.

Hat-tip to Dr. Thomas Dayspring for Tweeting this review.

Fig. 1 is interesting, as it shows a significant association between 25-year CHD mortality and Serum Total Cholesterol for every region except Japan. What's different about Japan, compared to Northern Europe, USA, Serbia, Southern Europe & Crete?

According to Measuring Blood Fatty Acids as a Surrogate Indicator for Coronary Heart Disease Risk in Population Studies , Philippines & Iceland have lower % linoleic acid than Japan. Where's the CHD vs TC data?

Could another difference be that the Japanese eat rice, a relatively intact grain, instead of foods made from wheat grain dust (i.e. flour) as their main source of dietary carbohydrates?

See also Using 3–6 differences in essential fatty acids rather than 3/6 ratios gives useful food balance scores , and Omega 3-6 Balance Score.

Ancel B. Keys' critique of "Diet and coronary thrombosis. Hypothesis and fact, by John Yudkin. The Lancet, 1957."

Ancel B. Keys has come in for a lot of flak recently over alleged "cherry-picking" of data for his 6/7 Countries studies. Here's Keys' critique:- "SUCROSE IN THE DIET AND CORONARY HEART DISEASE" of Dr. John Yudkin's "15 Countries" article.

Keys accuses Yudkin of bias, cherry-picking countries that fit his own hypothesis.

Here are some plots from Keys' 11 Countries article.

5-Year CHD cases/1,000 men vs Sucrose %E.

5-Year CHD cases/1,000 men vs Sat Fats %E.

Sucrose %E vs Sat Fats %E.

So there you have it.

Diet and coronary thrombosis. Hypothesis and fact, by John Yudkin. The Lancet, 1957.

Twitter did it again. From http://www.abc.net.au/catalyst/heartofthematter/download/Yudkinssugartheory.pdf

This looks like bad news for the fat-lovers.

There's good correlation between Coronary mortality and total fat intake, for countries 15 to 7. For countries 7 to 1, there's no correlation between Coronary mortality and total fat intake, suggesting that other differences (e.g. quality of health-care, social stress, antioxidant status etc) are significant factors.

This looks like bad news for the meat/fowl/fish/cheese/egg-lovers.

This looks like bad news for the sugar-lovers.

Of course, association ≠ causation.

This looks like bad news for rich people.

In conclusion, total fat intake, animal protein intake, sugar intake & annual income are all associated with increased Coronary mortality, over a certain range of values.

Saturated fats Saturated fats Saturated fats.

George Henderson left the following comment. I think that the information in it deserves a bigger audience.

Saturated fats seem to get the blame for everything nowadays. "Saturated fats clogged my arteries". "Saturated fats gave me cancer". "Saturated fats stole my job". O.K, I've done that joke before.

There are saturated fats, there are saturated fats, there are saturated fats, there are saturated fats, there are saturated fats and there are saturated fats. Saturated fats are an ester of Glycerol (a 3-carbon alcohol) and three saturated fatty acids (SFA's). There are roughly six categories of SFA's.

1) Short chain SFA's such as Acetic acid, Propionic acid, Butyric acid (found in butter and also what soluble fibre ferments into in the colon) and Caproic acid.
2) Medium chain SFA's such as Caprylic acid, Capric acid, Lauric acid and Myristic acid.
3) Long chain SFA's such as Stearic acid.
4) SFA's behaving like Palmitic acid.
5) Odd chain SFA's such as Pentadecylic acid and Margaric acid.
6) Very long chain SFA's such as Behenic acid.

See http://en.wikipedia.org/wiki/List_of_saturated_fatty_acids

In foods, the above SFA's are associated with different things.
1) and 2) don't get associated with much polyunsaturated fatty acids (PUFA's), e.g. dairy and tropical nuts.
3) and 4) are more likely to be associated with long-chain PUFA's, e.g. meats, poultry, temperate nuts.
5) is associated with CLA and not much PUFA's, e.g. dairy from grass-fed animals.

See also Siri-Tarino et al, Forests & Trees and "Eureka!" moments and Chowdhury et al, More forests & more trees and more "Eureka!" moments with cheese.

Siri-Tarino et al, Forests & Trees and "Eureka!" moments.

Here's Fig. 2 from Meta-analysis of prospective cohort studies evaluating the association of saturated fat with cardiovascular disease:-

Risk ratios and 95% CIs for fully adjusted random-effects models examining associations between saturated fat intake in relation to coronary heart disease and stroke.

The above "Forest" plot has a subtotal RR of 1.07 (95% CI 0.96 1.19). The overall conclusion is that there's no association between saturated fat intake and the RR for CHD. Hmmm.

I looked at the data in Table 3. Of the 16 studies contributing to the CHD results, only 3 of them specify high sat fat intakes over a wide range. The results from these 3 studies are as follows:-

Pietinen et al: RR=0.93 (95% CI 0.6, 1.44).
Mann et al: RR=2.77 (95% CI 1.25, 6.13).
Boniface et al: Pooled RR = 1.37 (95% CI 1.17, 1.65).

The results from Pietinen et al are statistically-insignificant (95% CI values are way above & below 1) with an overall slight protective effect. The results from Mann et al have a RR >> 1 with both 95% CI's >1 and the results from Boniface et al have a RR >1 with both 95% CI's >1.

Other studies either have sat fat intakes varying from very low to low, or specify mean/median sat fat intakes without values for highest & lowest tertiles/quartiles/quintiles etc. Other studies have results that are statistically-insignificant.

However, there are some studies that show a slight protective effect of small amounts of sat fats. How come?

Thanks to George Henderson, I had a "Eureka!" moment. He posted a link to Dietary intake of saturated fat by food source and incident cardiovascular disease: the Multi-Ethnic Study of Atherosclerosis.

Here's Fig. 1 from that study.

HRs and 95% CI's of CVD risk according to quintiles of energy-adjusted SF from different sources (n = 5209).

The Meat SF plot has a net positive slope (bad news, but the range of intake is very small), the Butter & Plant SF plots are random, but the Dairy SF plot has a net negative slope (good news). Dairy saturated fats in amounts of up to 10g/day are protective against CHD. As the Dairy sat fat intake is too small to have a significant effect on lipids, what's the mechanism? I think that it's Vitamin K2. See Chowdhury et al, More forests & more trees and more "Eureka!" moments with cheese.

When you average out the results from all studies, the result is null. This is data dilution statistics.

EDIT: See also Study: Saturated Fat as Bad as Sugar!

I'm NOT a lipophobe, I'm a very naughty boy!

First, postprandial triglycerides again. From Fasting Compared With Nonfasting Triglycerides and Risk of Cardiovascular Events in Women, here's a plot of HR for future CHD vs TG's at various times after eating.

Hazard ratio (HR) and 95% confidence interval (CI) for highest vs lowest tertiles of triglyceride level (see Table 3 for values), adjusted for age, blood pressure, smoking, hormone use, levels of total and high-density lipoprotein cholesterol, diabetes mellitus, body mass index, and high-sensitivity C-reactive protein level.

Notice how the HR falls with increasing time from last meal. As TG's ≥12 hours after eating are a surrogate for Insulin Resistance (IR) and the HR is only 1.04 (95% CI 0.79 - 1.38), this strongly suggests that IR is not a significant factor.

It's been suggested that IR might increase PP TG's in the 2 - 4 hour period due to impaired clearance. According to Fig. 3B in Extended effects of evening meal carbohydrate-to-fat ratio on fasting and postprandial substrate metabolism, TG clearance in healthy men doesn't significantly start until after 4 hours has elapsed. Therefore, an impairment in TG clearance isn't going to make a significant difference to TG level in the 2 - 4 hour period.

Second, the reason why I'm having to repeat myself is due to Cholesterol: Do chylomicrons clog your arteries? (2), where I've been called "my resident lipophobe". As I drink Gold Top milk (5.2g of fat/100mL) and eat pork including belly slices (you know, those strips of pork with a lot of fat on them), I'm being attacked for something that I'm not.

What I'm criticising is dietary extremism. Eating fats in foods is fine by me, but eating sticks of Kerrygold butter and/or dumping loads of butter and/or MCT oil into coffee to achieve "Nutritional Ketosis" is not a good idea. Anyway, here's an amusing spoof on Bulletproof coffee.

False dichotomies: serum cholesterol level vs all-cause mortality. Cause or effect?

Here are some plots from the MRFIT study.

From http://sph.bu.edu/otlt/MPH-Modules/PH/PH709_Heart/PH709_Heart5.html

Although the relative risk (RR) for coronary heart disease (CHD) and cardiovascular disease (CVD) mortalities increase with serum total cholesterol (TC) level, all-cause mortality follows a U-curve.

According to Low Serum Cholesterol and Mortality: Which Is the Cause and Which Is the Effect?, certain illnesses that increase mortality lower TC levels. This is the Iribarren hypothesis.

According to Cholesterol and all-cause mortality in elderly people from the Honolulu Heart Program: a cohort study, TC that's low and is still low 20 years later results in a 64% increase in the RR for mortality relative to TC that's intermediate and is still intermediate 20 years later.

Table 4 Relative risk for mortality based on change in cholesterol between examinations three and four

Is low TC level the cause of, or the effect of fatal illnesses? I think that it's both. Cholesterol is an important substance, as a severe lack of it is bad news, as per Smith–Lemli–Opitz syndrome. If certain illnesses result in a depletion of cholesterol and cholesterol synthesis is too low, there's insufficient cholesterol to allow recovery.

Interestingly, TC that's low but is intermediate 20 years later results in a 30% increase in the RR for mortality, whereas TC that's low but is high 20 years later results in a 5% increase in the RR for mortality.

P.S. There's a false dichotomy for vitamin D level vs illness. Ditto for carbohydrates vs calories.

Completing the trine: Which are the safest fats?

First, watch this video by Chris Masterjohn.

Diets very high in pufas (polyunsaturates) are not beneficial to heart health or longevity. Flora? No thanks!

Here's a chart. The beige & grey bars represent pufas (omega-6 & omega-3).

Comparison of dietary fats

The yellow bars represent monounsaturates and the orangey-red bars represent saturates. I consider these to be harmless, as long as you don't go mad and eat them in such large amounts that you gain weight.

Bearing in mind the information in the video, plus the information in Fats: Spawn of Satan or Dogs' Doodads? , I use only fats from the bottom 6 for cooking (olive oil and butter, actually).

Flaxseed oil can be used as an omega-3 supplement for vegetarian/vegan women, as omega-3 pufas are as rare as rocking-horse poo in most foods (apart from oily fish).

Non-vegetarian/vegan people can get their long-chain omega-3 pufas (EPA & DHA) from oily fish. As vegetarian/vegan men barely produce any DHA from the omega-3 in flaxseed oil, they should get it from algal DHA supplements. See Extremely Limited Synthesis of Long Chain Polyunsaturates in Adults: Implications for their Dietary Essentiality and use as Supplements.

\ curves and U curves: Vitamins D3 and K2 again.

Here are some curves relating to Vitamin D. Ref: http://www.ncbi.nlm.nih.gov/pubmed/23601272

Hazard Ratios (HRs) vs serum Vitamin D level

The solid lines are the 95% confidence intervals (CI) & mean for all-cause mortality. 95% CI's are the values within which 95% of the subjects tested fall. 2.5% fall below the lower CI and 2.5% fall above the upper CI. The dashed lines are the 95% CIs & mean for coronary heart disease (CHD) mortality. Most of the curves follow a \ curve, indicating that more Vitamin D is better, up to 66ng/mL (150nmol/L, the level that I'm at). The interesting curve is the upper dashed line, which follows a U curve.

The U curve indicates that a Vitamin D level of greater than 30ng/mL (75nmol/L) increases the Hazard Ratio (HR) for CHD in the top 2.5% of subjects only, relative to 30ng/mL, even though the mean HRs for CHD & all-cause mortality (the more important parameter) are decreasing, up to 66ng/mL. What's occurring?

See Vitamin K. The increase in HR for CHD mortality above 30ng/mL in the top 2.5% of subjects only is almost certainly due to calcification within artery walls, due to under-carboxylation of osteocalcin in bone Matrix Gla Proteins, caused by insufficient Vitamin K2 rather than excessive Vitamin D. This is why I supplement with Vitamin K2. See also Vitamin D toxicity redefined: vitamin K and the molecular mechanism.