Sunday, November 25, 2012

Should I Take a Statin?

This blog post is for informational purposes only and is not meant to diagnose or treat any medical condition.
If you haven't already see Troubleshooting High Cholesterol Part 1 and Part 2
What Are Statins? 

HMG-CoA reductase is part of the mevalonate pathway, which ultimately synthesises cholesterol, but also some other molecules like coenzyme Q10, squalene and Rho.  Statins inhibit HMG-CoA reductase and so are also referred to as HMG-CoA reductase inhibitors.  By inhibiting HMG-CoA reductase statins reduce cholesterol synthesis, but they also reduce the synthesis of other products of the mevalonate pathway.  To compensate the liver increases activity of the LDL receptor, which lowers LDL-C and LDL-P levels. 

Statins are thought to have other cardio-protective independent of lowering cholesterol such as increasing nitric oxide and (which supports endothelial function), having an anti-inflammatory effect [1].  However, statins have adverse side effects which may be related to it decreasing cholesterol and consequently steroid hormone levels like testosterone [2] and decreasing coenzyme Q10 [3] [4], squalene [5] and isoprenes [6

The Benefits vs. the Costs 

Drugs (and other interventions) should improve quality of life (lower morbidity/symptoms) and/or improve longevity (lower total mortality).  However, many drugs have adverse side effects so a drug should either:

  • Improve quality of life and improve longevity (ideal)
  • Improve quality of life greater than it decreases longevity 
  • Improve longevity greater than it decreases quality of life 

The nature of statins and CVD means you’re likely trading an increase in longevity for a decrease in quality of life (although non-fatal heart attacks and strokes decrease quality of life), but to what degree depends on the context. 

The Number Needed to Treat (meta-analysis of statin trials)
Benefits in Percentage
Harms in Percentage
Statin Drugs Given for 5 Years for Heart Disease Prevention (Without Known Heart Disease) [7]
60 for non-fatal heart attack
·         98% saw no benefit
·         0% were helped by being saved from death
·         1.6% were helped by preventing a heart attack
·         0.4% were helped by preventing a stroke
·         2% were harmed by developing diabetes
·         10% were harmed by muscle damage
Statins Given for 5 Years for Heart Disease Prevention (With Known Heart Disease) [8]
83 for mortality
·         96% saw no benefit
·         1.2% were helped by being saved from death
·         2.6% were helped by preventing a repeat heart attack
·         0.8% were helped by preventing a stroke
·         2% were harmed by developing diabetes
·         10% were harmed by muscle damage

However, statins are ineffective at reducing mortality in all women and men aged 80 or over, regardless of whether it’s for primary or secondary preventions.  Also, most of the RCTs for statins were funded by drug companies.  RCTs funded by drug companies are more likely to report better outcomes and fewer side effects, so the benefits are likely exaggerated while the harms are underreported [9].  Something you can’t tell from the NNT data is the minor quality of life issues (muscle pain, memory loss, loss of libido, etc) that can occur from statins and are more common but less severe than muscle damage and type 2 diabetes 

That being said statins are probably more therapeutic for people with FH or ApoE4 genotypes. 

Also, consider your individual risk.  How likely are you to have a heart attack or stroke?  How much atherosclerosis do you currently have?  How much of my list on Part 1 applies to you?  The lower your risk, the less benefit statins are to you 

Remember that diet and lifestyle are far more powerful than most drugs.  Statins come nowhere near close to the 72% reduction in cardiac events, 65% reduction in cardiac deaths and 56% reduction in total mortality observed in the experimental group of the Lyon Diet Heart Study [10

Dealing with the Adverse Side Effects 

Should you decide to take statins it’s in your interests to prevent and deal with the side effects: 

CoQ10.  I mentioned before that statins lower CoQ10, which is a suspected cause of several of the side effects.  However, there have been conflicting results as to whether CoQ10 supplementation is effective at reducing the adverse side effects of statins [3], though probably on balance the evidence is in favour of benefit [11].  That being said, I would err on the side of supplementing the CoQ10 because statins lower CoQ10 levels, the proposed mechanisms are pretty sound, several studies do find a benefit, CoQ10 supplementation is safe and well tolerated and the side effects of CoQ10 are generally only going to be positive.  The only real negative would be the cost as they are more expensive than your average supplement 

Vitamin D.  One of the potential symptoms of vitamin D deficiency (< 20 ng/dl) is myalgia (muscle pain).  There seems to be a relationship between low vitamin D levels (< 30 ng/dl) and statin-induced myalgia, as vitamin D activates enzymes that metabolise some, but not all classes of statins.  Vitamin D supplementation appears to be very therapeutic in these contexts.  (Interestingly some statins appear to increase vitamin D) [12

There seems to be less research on sides effects resulting from a lack of steroid hormones, squalene and isoprenes and on supplementing any of them to prevent some of the adverse side effects.  (I suppose funding research to investigate and minimise the adverse side effects legitimises them, whereas it’s probably in the drug companies interests to sweep them under the rug).  Squalene seems to have anti-cancer effects and protects the skin from UV and singlet oxygen [13].  Squalene is mainly found in olive oil, a fairly health promoting food anyway (squalene may inhibit the cholesterol lowering effect of statins, not sure).

Sunday, November 18, 2012

Troubleshooting High Cholesterol: Part 2

This blog post is for informational purposes only and is not meant to diagnose or treat any medical condition.

If you haven't already see Part 1

Possible Causes of High Cholesterol or LDL-P

If you have high LDL-C or LDL-P, and whether you said yes to any of the things on the list above or not, it’s important to find to why you have high LDL-C or LDL-P as it could suggest an underlying problem that should be fixed 

Losing Abdominal and/or Liver Fat 

If you got your cholesterol tested and it came back high shortly after you changed your diet/lifestyle it could be that active weight loss and/or clearance of liver fat is temporarily increasing your cholesterol or LDL-P levels [1]. 

  • Both active weight loss and clearance of liver fat will increase triglycerides as more of them enter the circulation
  • LDL-P may increase slightly as more LDL particles may be needed to transport the extra triglycerides
  • These LDL particles stay in the bloodstream for longer, which increases their exposure to cholesterol ester transfer protein (CETP).  CETP may then exchange the triglycerides from the triglyceride rich VLDL particles with cholesterol from HDL particles, resulting in higher LDL-C and lower HDL-C 

Simply wait until your weight has stabilised, and then wait another a month or so until your next test.  You could be clearing liver fat even if you haven’t lost weight (especially if you were insulin resistant, drinking a lot of alcohol or had IBS and are now eating more eggs/liver), once again wait a few months.


While thyroid hormone increases both cholesterol synthesis and cholesterol metabolism (increases LDL receptor activity), people with hypothyroidism tend to have higher LDL-C and LDL-P with minimal differences in triglycerides and HDL-C [2].  People with subclinical hypothyroidism (high TSH but in the reference range) also have higher TC, non-HDL-C and triglycerides, but and lower HDL-C [3] 

Key nutrient deficiencies (iodine, iron, selenium) can impair thyroid function, although the most common cause of hypothyroidism in developed countries is autoimmune attacks against the thyroid (like Hashimoto’s thyroiditis).  Likely common causes of subclinical hypothyroidism include calorie restriction, overtraining [4] and very low carb diets [5] [6] [7] 

Insulin Resistance 

Insulin resistance results in higher triglycerides and LDL-P, which in turn can result in smaller increases in LDL-C and decreases in HDL-C [8] probably via an increased activity of CETP.  See Stephan Guyenet’s seven part series on: What Causes Insulin Resistance? 

Elevated LPS 

One function of lipoproteins, particularly LDL, is to neutralise LPS in the bloodstream.  LPS-binding protein binds to LPS and transfers it to either lipoproteins or CD14 expressing monocytes/macrophages [9].  LPS increases LDL-C levels: 

  • Administration of LPS increases cholesterol synthesis, LDL-C and triglycerides, but lowers HDL-C [10]
  • People with periodontitis have higher LDL-C (1.0 mmol/l), lower HDL-C (0.3 mmol/l) and higher triglycerides (0.6 mmol/l) [11] 


One of the functions of the cholesterol transport of LDL is to repair cells and tissues, specifically the vasculature* [12].  So a high cholesterol level could indicate vascular injury.  Besides LPS, other source of inflammation are associated with high cholesterol: 

  • Copper deficiency promotes CVD [13] and increases LDL-C [14]
  • Homocysteine is a risk factor for CVD.  In children with high homocysteine, folate supplementation decreased homocysteine and cholesterol levels (0.6 mmol/l) [15] 
  • Severe vitamin C deficiency increases LDL-C [16] 
  • Ferritin is associated with LDL-C [17] [18] [19] 

* This function was actually one of the main ideas behind the response to injury hypothesis, where vascular injury (desquamation of endothelial cells) increased cholesterol transport to that area and if there was prolonged injury then the cholesterol would build up and form plaques. 


Stress seems to raise cholesterol: 

  • Mental stress (10 minutes of mental arithmetic with harassment) raised TC, HDL-C, LDL-C and triglycerides (by ~20%), which was partially due to a reduction in plasma volume [20] 
  • Cholesterol levels in male medical students were on average 11% higher during examination week [21] [22] and 16.5% higher during the winter quarter exams [22] 

* On the effect of plasma volume on cholesterol levels: there is a minor seasonal variation, where cholesterol levels are slightly lower during summer, probably due to an increase in plasma volume during summer [23].  Also, salt restriction causes a minor increase in TC, LDL-C and triglycerides [24], which may be related to a slightly lower plasma volume 

See Chris Kresser’s blog post on 9 Steps to Perfect Health – #6: Manage Your Stress 


Like many things, genes explain a fair bit of the individual variation in cholesterol levels.  Some genetic determinants of high cholesterol include: 

  • Familial hypercholesterolemia (FH) is a rare inherited disorder, often resulting from a defect in the LDL receptor.  Only 1 in 500 people have the milder form (heterozygous), which results in very total cholesterol levels (~8-10 mmol/l) and LDL-C.  People with FH will have had very high cholesterol for their whole lives 
  • ApoE is a protein on LDL particles.  The ApoE4 polymorphism has a frequency of 14.7% and results in higher LDL-C, particularly in the context of a high fat or high cholesterol diet [25], and a greater risk of other diseases such as Alzheimer’s disease

See Should I Take a Statin?

Sunday, November 11, 2012

Troubleshooting High Cholesterol

This blog post is for informational purposes only and is not meant to diagnose or treat any medical condition. 

Cholesterol      1 mmol/l = 38 mg/dl
Triglycerides   1 mmol/l = 88 mg/dl 

Before reading this post you should probably read the links on Cardiovascular Disease if you haven't already

Firstly, Is Your Cholesterol Actually High? 

Make sure you took the test in a fasted state as this can affect blood lipids, particularly triglycerides 

Total Cholesterol 

The threshold for ‘high cholesterol’ is being pushed lower and lower, likely to boost drug sales more than anything else. 

If you take a look at the graphs on The Lipid Hypothesis: Total Cholesterol you’ll see a meta-analysis that found all-cause mortality (what we’re interested in) is lowest between 160-239 mg/dl (4.2-6.3 mmol/l), and yet ‘high cholesterol’ is often defined as >200 mg/dl or >5 mmol/l, which is within the bracket for lowest all-cause mortality [1]. 

You would think that based on total cholesterol vs. all-cause mortality data that ‘high cholesterol’ should probably be defined as 240 mg/dl at the minimum, and probably more like 260 or 280 mg/dl 

The Total:HDL-C Ratio 

Unfortunately many people only look at total cholesterol.  But the ratio of total cholesterol to HDL cholesterol is a better risk factor than total cholesterol, HDL-C, LDL-C or triglycerides [2] [3].  The reason why, is that it accounts for LDL-C, IDL-C, VLDL-C (non-HDL-C) and the lower risk of having higher HDL-C.  It also indicates how long the LDL particles stay in the bloodstream 

“So what would the total-to-HDL cholesterol mean? The longer LDL stays in the blood, the more two things happen: it is exposed to oxidants, and as its limited supply of antioxidants run out, the polyunsaturated fatty acids in its membrane oxidize, leading to the further oxidation of its proteins and cholesterol; it is exposed to cholesterol ester transfer protein (CETP), which transfers cholesterol from HDL to LDL, thus boosting the total-to-HDL cholesterol ratio.” - Chris Masterjohn [4] 

Ratios less than 4 are associated with a lower risk of CVD while ratios greater than 4 are associated with a higher risk of CVD 

Calculating LDL-C 

LDL-C is calculated based by using the Friedewald equation: 

LDL-C = TC – HDL-C – k x Triglycerides
(where k is 0.45 if measured in mmol/l and 0.20 if measured in mg/dl) 

However, the Friedewald equation overestimates LDL-C in people who have triglycerides below 1.1 mmol/l (100 mg/dl), whereas the Iranian equation is more accurate at those triglyceride levels [5].  The Iranian equation is: 

LDL-C = TC/1.19 – HDL-C/1.10 + Triglycerides/0.81 – 0.98 (mmol/l)
LDL-C = TC/1.19 – HDL-C/1.10 + Triglycerides/1.90 – 38 (mg/dl) 

VLDL-C isn’t measured either, but rather is calculated by: VLDL-C = TC – HDL-C – LDL-C.  On a standard blood test it will be equal to triglycerides/5 if mg/dl or triglycerides/2.2 if mmol/l (so there’s no point mentioning both triglycerides and calculated VLDL-C) 

Not that this really matters because non-HDL-C and the total:HDL-C are better risk factors than LDL-C, and in both risk factors it doesn’t matter whether the cholesterol is in LDL, IDL or VLDL, as it’s still in an ApoB lipoprotein, which is the measurement that LDL-C and non-HDL-C tries to approximate [6] 

Intra-individual Variation in Cholesterol Levels 

Even if diet and lifestyle are kept constant, there can be a huge variation in an individual’s blood lipids.  Chris Masterjohn discusses the mean intra-individual variation is as follows 

Blood Lipid
2 Standard Deviations (mg/dl)
2 Standard Deviations (mmol/l)
Total Cholesterol

“If you’ve only measured it two times, you should expect to see an increase or a decrease greater than 35 mg/dl before you can be 95% confident that your cholesterol has increased or decreased” – Chris Masterjohn 

So your average cholesterol level could be within ‘normal range’ but due to intra-individual variation you could take a blood test and it comes back high, then you’re prescribed a statin or get a higher premium for life insurance.  This also makes it difficult to judge whether a diet or lifestyle change has actually changed your cholesterol levels. 

What’s mentioned is the mean intra-individual variation.  Not everyone has the same degree of variation and you can get a rough idea how variable your cholesterol levels are with repeated tests (while controlling diet and lifestyle) 

If I have High Cholesterol, Am I at Risk of a having a Heart Attack? 

LDL-P > non-HDL-C > LDL-C 

The value of both LDL-C and LDL particle size as risk factors are abolished when you look the number of LDL particles (LDL-P) [7].  The reasons are that: 

  • LDL-C and LDL particle size are only approximate measures of LDL-P 
  • Elevated LDL-P has a legitimate mechanism for atherosclerosis (the response to retention hypothesis), whereas elevated LDL-C or small, dense LDL particles do not. 

The response to retention hypothesis is where some LDL particles penetrate the endothelium, generate an inflammatory response, become oxidised, then macrophages come along and engulf the LDL particle, turn into foam cells, which develop into atherosclerotic plaque.  The penetration of LDL particles into the endothelium is driven by a concentration gradient, in other words: more LDL-P more penetration [8] 

So the next thing to do if you’re concerned is to check to see whether you have a high number of LDL particles or not.  If you have high LDL-C you probably do, just as if you have low LDL-C you probably don’t, but you can also have discordant LDL-C and LDL-P.  Two main reasons for discordance are insulin resistance and triglyceride levels: 

Insulin inhibits LDL particle production and increases the activity of the LDL receptor.  So in insulin resistance there is an increased secretion and decreased clearance of LDL particles, which elevates LDL-P [9] 

LDL particles also contain triglycerides so if you have higher triglyceride levels you need more LDL particles to transport them.  So one could have: normal LDL-P despite slightly elevated LDL-C because they have low triglycerides; high LDL-P despite low or normal LDL-C because they have high triglycerides [10].  While carbohydrates (relative to fats) can increase triglycerides, the real culprit of high triglycerides is insulin resistance 

…And If I Have High LDL-P? 

There’s more to atherosclerosis and getting heart attacks than LDL-P and since I don’t know you, I can’t say what you’re risk is.  Look at this list and consider how many of these apply to you: 

  • Are you overweight/obese
  • Do you have insulin resistance or type 2 diabetes
  • Do you have hypertension or evidence of endothelial dysfunction
  • Do you have an autoimmune disease, allergies or asthma
  • Do you have GI symptoms (GERD, IBS, etc) 
  • Do you have poor dental health (bleeding gums, gum recession, gingivitis, periodontitis) 
  • Do you smoke
  • Do you eat a poor diet (high in processed foods, refined vegetable oils, refined grains, refined sugar or alcohol)
  • Are you physically inactive
  • Do you have poor sleep or not much of it (< 7 hours)
  • Do you have low HDL-C or HDL-P (ApoA) or high triglycerides
  • Are your liver enzymes elevated on a standard blood test (bilirubin, GGT, AST or ALT)
  • Do you show evidence of poor kidney function (low GFR, high urea/creatinine, etc)
  • Are your markers of inflammation high (such as CRP/hsCRP, Lp-PLA2, etc)
  • Are your iron levels high (such as ferritin > 150)
  • Do you have low vitamin D (< 30 ng/dl or < 75 nmol/l)
  • Do you have hypothyroidism or low thyroid function
  • Do you have high homocysteine 

The more of the above you have (and the worse each one is) I would suspect the greater your risk, whether you have high LDL-P or not.  If you have high LDL-P and none of the other issues then you’re probably not at risk, but if you do have some of them you should try harder to correct both the LDL-P and the other issues.  The good thing is those are all modifiable (unlike age, sex, ethnicity and family history) 

Also, using non-invasive tests to see your level of atherosclerosis and tracking that over time is better than any blood test.  Discuss with your doctor to find out what options are available to you

See Part 2 for some possible causes of high cholesterol or LDL-P

Sunday, November 4, 2012

Cardiovascular Disease

Cholesterol is essential for several functions in the body.  Lipoproteins, such as HDL and LDL, transport cholesterol through the bloodstream (like ships carrying cargo).  HDL and LDL aren’t different types of cholesterol they are different types of lipoproteins.  The cholesterol carried by them is exactly the same molecule
Atherosclerosis is a central part of many cardiovascular diseases, particularly coronary artery disease, which may develop into ischemic heart disease, but also ischemic strokes, peripheral artery disease and hypertension.  Regular LDL particles don’t cause atherosclerosis, whereas modified LDL particles do. Oxidation is one possible modification of LDL.  Particular oxidants such as peroxynitrite and hypochlorous acid seem to be the most likely to oxidise LDL
The lipid hypothesis is based off data showing the greater one’s cholesterol the greater their risk of CVD.  While total cholesterol is associated with CVD, the relationship between total cholesterol and all-cause mortality is a U-shaped curve.  It’s also important to remember that the lipid hypothesis is just a hypothesis.  Things that lower cholesterol needs shouldn’t be assumed to reduce the risk of cardiovascular disease, but to be tested as some things that reduce cholesterol don’t actually reduce cardiovascular disease
The total cholesterol to HDL cholesterol ratio (Total:HDL-C ratio) is probably the strongest blood lipid risk factor for cardiovascular disease (the higher the ratio the greater the risk).  However, once again we shouldn’t assume that things that lower the ratio reduce cardiovascular disease as drugs such as niacin and CETP inhibitors have failed to do so.
In addition to the exceptions with lowering total cholesterol and the Total:HDL-C ratio, almost half of people hospitalised for cardiovascular disease have LDL-C lower than 100 mg/dl.  One idea is small, dense LDL particles penetrate the endothelium, stick there and oxidise but the value of small, dense LDL and LDL-C as a risk factor is abolished when you adjust for LDL-P.  The reason why LDL-C is so variable and sdLDL depends on LDL-P is because the cholesterol in LDL doesn’t initiate CVD, the LDL particle does.  The proposed mechanism for LDL-P is the more LDL particles in the bloodstream the greater the chance of them penetrating the endothelium and oxidising (response-to-retention hypothesis).  The reason why LDL-C has value as a risk factor is because it correlates with LDL-P
The ability of HDL to remove cholesterol from macrophages is more strongly associated with lower risk of CVD than HDL-C and HDL-P and is independent of HDL-C and HDL-P.  HDL has many beneficial functions which can be impaired under conditions of oxidative stress, inflammation, infection and high triglycerides.  Regulatory T cells promote HDL function by increasing cholesterol efflux by HDL particles rather than have macrophages form atherosclerotic plaque
Nitric oxide is secreted by endothelial cells and protects against cardiovascular disease by inhibiting LDL oxidation, endothelial permeability, inflammation and monocyte adhesion.  Superoxide and combine with nitric oxide, which depletes nitric oxide, thereby impairing endothelial function and forms peroxynitrite.  Peroxynitrite can oxidise LDL and has other pro-atherogenic effects.  One of the main sources of superoxide is the mitochondria and in mitochondrial dysfunction more superoxide is generated and antioxidant defenses are overwhelmed.
Bacterial infections and LPS can promote cardiovascular disease through inflammation and oxidative stress.  In addition myeloperoxidase products (hypochlorous acid and chlorinated tyrosine) can oxidise LDL and HDL and are strongly associated with cardiovascular disease
The diet heart hypothesis suggests that: because high cholesterol/Total:HDL-C ratio is associated with CVD; and saturated fat raises cholesterol/Total:HDL-C ratio, therefore saturated fat increases CVD.  However, saturated fat doesn’t seem to raise cholesterol in the long term and SFA is not associated with CVD in meta-analysis of observational studies and clinical trials.  In the clinical trials when saturated fat and TFA are replaced with omega 6 polyunsaturated fat what happens is CVD mortality increases by 16%.  Seeing as TFA is bad, think what would happen if just saturated fat was replaced by omega 6 polyunsaturated fat.
One of the reasons omega 6 polyunsaturated fat increases CVD in clinical trials is because polyunsaturated fats are susceptible to oxidation.  The more polyunsaturated fat there is in LDL the more lipid peroxidation products there are.  Lag time is how long it takes for LDL to become oxidises and is related to the amount of polyunsaturated fat in the membrane and the amount of vitamin E
Some Strategies for Cardiovascular Disease
This is for informational purposes only and is not meant to diagnose or treat any medical condition.
Reduce LDL-P if Elevated
The higher your LDL-P the more LDL particles penetrate the endothelium, thereby potentially starting the events that lead to atherosclerosis.  See Troubleshooting High Cholesterol
Reduce Bacterial and LPS Translocation
Bacteria and LPS are inflammatory, but also increase myeloperoxidase.  See this post and for some ways to reduce LPS
Reduce Chronic Inflammation
Chronic inflammation impairs HDL, endothelial and mitochondrial function and is a major factor in insulin resistance.  See Causes of Inflammation
* Homocysteine (Hcy) is a risk factor for CVD but trials to lower Hcy with folic acid and sometimes B12 have been unsuccessful.  However, the trials may have been too short (only 2 years long) and folate may promote inflammation in an inflammatory environment [1].  Even if Hcy turns out to not be a factor in CVD, I’m only suggesting you get an adequate intake of several B vitamins
Improve Mitochondrial Function
Mitochondrial dysfunction can increase inflammation and insulin resistance, and mitochondrial superoxide can merge with nitric oxide, impairing endothelial function and forming peroxynitrite.  See Mitochondrial Dysfunction (mainly the second half)
Improve Immune Regulation
Having good immune regulation may reduce the likelihood of oxLDL forming atherosclerotic plaque and increase the likelihood of HDL doing reverse cholesterol transport.  Improving immune regulation also reduces the severity and duration of inflammation.  See Immune Dysfunction
Reduce Iron Levels if Elevated
Iron can modify LDL, ferritin (a marker of stored iron levels) is associated with CVD and the sex difference in CVD could be largely attributed to iron stores (see graph) [2].  See Chris Kresser’s AHS 2012 talk for information on the causes, problems, testing and treatment of iron overload