Cholesterol is Bad. Isn’t it ? – Part 3

Please go to Part 1 of the post if you have not already read it.

So what is the relationship between Cholesterol and Cardio-Vascular Diseases (CVD)?

Part 2 of the Diet-Heart Hypothesis is that increased Cholesterol levels CAUSE heart disease.

The following picture is from a huge medical study and shows the relationship between Cholesterol on one side and a) deaths from all causes and b) deaths from CVD on another side.

MRFIT Total Chol vs All Cause Mortality

Per the above picture, the relationship is not as linear as it is made to be. It is more an “U” shaped relationship with the ideal (if there such a thing) Cholesterol level to be approx 210 mg/dl or 5.5 mmol/l. If you have too much Cholesterol or too little Cholesterol then the risk is higher.

Most of you will agree that what matters is whether you live or die and not bothered about what underlying causes your death.

The above picture though shows Stats for Men in one trial.

Another study (a Norwegian one  published in 2012 and based on 52,000 people) has the following Stats:

hunt study smoking vs cholesterol

I have highlighted some of the data in the above picture.

For people in this study who are not smokers and who do not have High Blood pressure the risk of death is greater if your Cholesterol levels are lower than 5.5 compared to levels greater than 5.5. Smoking and High Blood pressure increase the risk of death generally and we will talk about them later. In men of age less than 59 there is a small increase in risk if your Cholesterol levels are greater than 5.5 (green areas highlighted).

The following are actual quotations from the Norwegian study (emphasis is mine):

Our study provides an updated epidemiological indication of possible errors
in the CVD risk algorithms of many clinical guidelines. If our findings are generalizable, clinical and public health recommendations regarding the ‘dangers’ of cholesterol should be revised. This is especially true for women, for whom moderately elevated cholesterol (by current standards) may prove to be not only harmless but even beneficial.

In this validation study of current guidelines for CVD prevention, which is based on new epidemiological data from a large and representative Norwegian population, we found total cholesterol to be an overestimated risk factor.

Our results contradict the guidelines’ well-established demarcation line (5 mmol L-1) between ‘good’ and ‘too high’ levels of cholesterol. They also contradict the popularized idea of a positive, linear relationship between cholesterol and fatal disease. Guideline-based advice regarding CVD prevention may thus be outdated and misleading, particularly regarding many women who have cholesterol levels in the range of 5–7 mmol L-1 and are currently encouraged to take better care of their health.

At least in some settings, cholesterol may represent a risk marker and/or a weak risk factor rather than an important one.

Many individuals who could otherwise call themselves healthy struggle conscientiously to push their cholesterol under the presumed ‘danger’ limit (i.e. the recommended cut-off point of 5 mmol L-1), coached by health personnel, personal trainers and caring family members. Massive commercial interests are linked to drugs and other remedies marketed for this purpose. It is therefore of immediate and wide interest to find out whether our results are generalizable to other populations

WHO Data

Another piece of Data – this time WHO data showing the relationship between Cholesterol and mortality for males and females

who all male and chol


The correlation coefficient is poor and in general you can say with a reduced confidence that Cholesterol levels are inversely proportional to CVD death.

While interpreting these graphs you need to know the profile of the underlying subjects – age, gender, smoking habits, medication, blood sugar levels, previous cardiac disease occurrence,  inflammation indicators etc, otherwise we can easily draw incorrect conclusions.

UK has relatively the lowest level of TC (on average) in Europe but one of the highest rates of Heart Disease in Europe.

So if Total Cholesterol (TC) levels do not tell us anything, what about LDL and HDL?

A little background here. A number of you would have read that LDL Cholesterol is BAD and HDL Cholesterol is GOOD.

This was based on the following assumptions:

  1. LDL (Bad Cholesterol) carries Cholesterol from Liver etc. to the tissues and if there is more LDL in the blood, it is more likely to get deposited in the arteries (and thereby causing Plaque) and hence more likely to cause Heart Disease. So LDL transports Cholesterol (called CT for Cholesterol Transport) from Liver
  2. HDL (Good Cholesterol ) carries Cholesterol from tissues to Liver – thereby removing Cholesterol from the Blood and hence is beneficial leading to lower Heart Disease risk. HDL transports Cholesterol in the reverse manner (called RCT –  Reverse Cholesterol Transport)

This model is now thought to be out-dated and the latest research indicates that LDL also undertakes significant Reverse Cholesterol Transport.

But let us examine the above two assertions. If higher levels of LDL were to cause Heart Disease then we should expect higher LDL levels in Heart Disease patients on average. The target LDL level being advocated now is an LDL of less than 100 mg/dl (or approx < 2.5 mmol/l) and this is tough target. Some still use approx 130 mg/dl as the target.

The following picture contains analysis of actual hospital admissions and the Cholesterol levels (courtesy Dr. Peter Attia). Please read the accompanying text and this has been shown in a number of other studies.

HALF THE PATIENTS HAD LDL levels less than TARGET and HALF ABOVE (with a target of 100 mg/dl) and if we use a Target of 130 mg/dl, then almost 75% had less than this level. So if a lower level of LDL is protective then half the patients should not be in the hospital.


What about HDL (Good Cholesterol) then?

The same study shown above also documented the HDL levels. A good HDL level is supposed to be 60 mg/dl (approx 1.5 mmol/l).

HDL Levels Hospitalisation CAD

Almost 90% had less than the Target level of 60 mg/dl. So perhaps HDL can better predict the CVD risk.

Now that we have discussed LDL and HDL let us also consider the following example (and the fallacy of using Total Cholesterol levels and Heart Disease risk)

Day 1: A gentleman has the following Lipid Profile

TC = 190 LDL = 110 HDL = 30 TG = 250 ( TC = LDL + HDL + TG/5)

The above are considered “Very Good” according to official standards.

Let us say that the gentleman works very hard and increases his Good Cholesterol (because his Doctor said so) to 80 for example and everything else remains the same.

His TC is now  110 + 80 + 50 (250/5) = 240.

He will be considered at High Risk given the TC level but the only thing he has done is increase Good Cholesterol !! He may even be prescribed statins if TC is used as the measure.

What is the latest on Cholesterol Research?

There is some recent research that shows that the AMOUNT of LDL cholesterol does not matter but the LDL PARTICLE Number and SIZE matter much more.

LDL Particle Number is abbreviated as LDL-P and LDL Cholesterol amount is known as LDL-C.This is based on the thinking that small, dense LDL particles are much more harmful (atherogenic) compared to big, fluffy LDL particles of lesser density because the smaller size and the larger number causes more of them to be deposited in the artery wall leading to Plaque, Rupture,  Clot and Heart Attack.

    1. It is very much possible in some instances that even if your LDL Cholesterol amount (LDL-C) decreases you are at a greater risk because your LDL particle Number (LDL-P) has gone up.
    2. Conversely, it is very much possible that your LDL Cholesterol amount increases but you are at a lesser  risk because your LDL particle Number (LDL-P) has gone down.

The following diagram (courtesy Dr. Peter Attia) shows relative risks associated with LDL-P given a particular amount of LDL-C. The X Axis shows the particle size.

This shows 2 graphs in one – 1) for people NOT suffering from Metabolic syndrome and 2) for people suffering from Metabolic syndrome.

It is very evident from the picture below that for a given LDL amount, the risks are very much higher for people with metabolic syndrome because the LDL particle size is much greater on average.

Microsoft PowerPoint - ADA Otvos LDL size talk_modified.ppt [Com

The next picture shows the cumulative incidence or occurrence of CVD events in 3 different groups.


The above picture shows that the highest incidence is in people with the largest LDL-P (particle number and also the smallest size) and the LOWEST LDL amount and the lowest incidence is in people with the smallest or lowest LDL-P but the highest LDL-C (amount).

This is also perhaps a reason why the LDL-C goes up a little when you eat a lot of Saturated Fat but LDL-P goes down. To this extent, if the LDL-C increase is around 10%-15% but with LDL-P decreasing (and TGs decreasing and HDL increasing) then there may not be a lot to worry about – this could happen due to dietary changes.

We will talk more about the metabolic syndrome in the near future but it is generally characterised by high glucose (and insulin levels), high TGs, large waist size, high Blood Pressure, Insulin Resistance etc.

So what are better predictors of CVD risk?

I personally think (based on the current available information) the following will predict the risk better than LDL or TC:

  1. Tri-glycerides (TG)
  2. HDL
  3. LDL Particle Number and Size

Of the data I have seen there seems to be an inverse relationship, in general, between the above two i.e. people who have very high TGs seem to have a low HDL.

The following are good (and tough) targets:

HDL ( > 60 mg/dl or > 1.5 mmol/l)

TGs ( < 90 mg/dl or < 1 mmol/l)

Low LDL Particle number and size

Unfortunately, estimating LDL-P reliably is an expensive process and not available to all. The most reliable method currently available involves nuclear magnetic resonance spectroscopy, or NMR for short.

I have also seen that the level of TGs increase (and HDLs decrease) a lot in people who have

  1. High Blood glucose levels
  2. High Insulin levels
  3. Insulin Resistance

It is also quite likely that these people have a lot of fat around their waist/hip/thighs and also eat a Carb rich (particularly sugary, processed and refined food based) Diet.

If you can achieve the above targets then the TC and LDL do not matter much, I think.

So if I have a High Cholesterol level it does not matter, is it?

Not entirely.

  1. If you are woman with no history of Heart Disease then it is quite likely that your TCs does not matter – certainly if you are above 50-55 years of age. Also if your blood sugars, insulin levels and TGs are low and HDL is high, then you perhaps have little to worry about TC.
  2. If you are a man and above 50  and with no history of Heart Disease again it appears that TCs do not matter as much. Also if your blood sugars, insulin levels and TGs are low and HDL is high, then you perhaps have little to worry about TC.
  3. But if you are a man and less than 50 years of age and your TCs are high, it does indicate that there is some underlying problem and you better be cautious. Whatever is causing the Cx levels to go high might also be causing some other damage elsewhere. Aim to get your HDL high (> 60 mg/dl) and TGs low (< 90 mg/dl). You may also consider medication while sorting out your Diet and Lifestyle.
  4. If you have already had a Heart Attack, you need to watch your Cholesterol levels not so much because they cause a problem but they are sort of saying something wrong is going in the body


continued in part 4


Leave a Reply

Your email address will not be published. Required fields are marked *