'Good cholesterol' theory challenged

“Good cholesterol” doesn’t lower heart attack risk, the Daily Mail has reported.

A great deal of research has previously suggested that higher levels of “good” HDL cholesterol reduce your risk coronary heart disease, while higher levels of “bad” LDL cholesterol increase your risk of a heart attack. However, it has been hard to tell whether HDL cholesterol directly reduces coronary heart disease risk as other medical, biological or lifestyle factors could be involved. To get round this, researchers conducted a complex study to identify genes that raise levels of HDL cholesterol, then looked at whether carrying these genes influenced heart disease risk.

Researchers first identified genetic variants associated with high HDL levels and tested for them in several thousand people, including some who had had a heart attack. They found that carrying these "HDL cholesterol genes" had no effect on the risk of a heart attack. From this, the researchers concluded that there is no direct relationship between HDL cholesterol and coronary heart disease and, therefore, that other factors must be involved.

This complex study challenges the commonly held belief that having higher HDL cholesterol will reduce heart attack risk. However, as it only looked at a particular set of genetic variations, it cannot provide the whole answer and tell us whether HDL cholesterol does or does not affect coronary heart disease, and how this effect might come about. An important question is whether things that increase HDL cholesterol levels during our lifetime (i.e. after our genetics are determined), such as exercise and certain medications, can then improve our heart disease risk.

Where did the story come from?

The study was carried out by researchers from Harvard Medical School and was funded by the US National Institutes of Health, Wellcome Trust, European Union, British Heart Foundation and German Federal Ministry of Education and Research. The study was published in the peer-reviewed medical journal The Lancet.

The media generally oversimplified what is a complex analysis. Also, captions referring to cholesterol intake through diet have no direct relevance to this research, which examined the genetic factors that determine HDL cholesterol levels and not the influence of dietary sources.

What kind of research was this?

There are two broad types of cholesterol in the body that are each associated with altered risk of cardiovascular problems: high-density lipoprotein (HDL) and low-density lipoprotein (LDL). LDL cholesterol is often referred to as “bad” cholesterol, as research has found that raised levels of LDL are associated with an increased risk of heart attacks. Conversely, previous observational studies have tended to show that people with higher levels of HDL (“good”) cholesterol have a lower risk of coronary heart disease (CHD).

However, it is difficult to prove that HDL cholesterol directly lowers people’s risk of CHD. For example, other factors in a person’s health and lifestyle might influence both HDL levels and CHD risk, so could be responsible for the apparent relationship between the two.

This study used a complex genetic analysis concept, called “mendelian randomisation analysis”, to investigate the relationship between genes, HDL cholesterol and CHD. Broadly speaking, mendelian randomisation analysis looks at whether genetics that determine one factor (such as HDL cholesterol levels) are directly associated with the risk of an outcome (such as heart disease).

In this case, the researchers considered the theory that if increased HDL directly reduces CHD risk, then carriers of genetic variants that confer a high concentration of HDL cholesterol should have a reduced risk of CHD. If genetic determinants of HDL cholesterol had no relationship to CHD risk, then there isn’t a causal relationship between the two, and other factors are likely to be involved.

This mendelian analysis has the important limitation that in looking at purely genetic factors, it does not consider how environmental, health and lifestyle factors influence both HDL levels and CHD risk (in other words, everything that occurs after our genetics are determined at conception).

What did the research involve?

The researchers first identified a certain rare form of a gene called the endothelial lipase gene (LIPG Asn396Ser). This particular form of the gene, carried by about 2.6% of the population, was associated with levels of HDL cholesterol. Carriers of this gene variant had consistently higher levels of HDL (good) cholesterol compared to non-carriers, but no difference in their levels of LDL (bad) cholesterol or other blood fat levels. Based on the influence that carrying this LIPG variant had on HDL cholesterol levels, the researchers calculated that if the relationship between HDL cholesterol and CHD was causal, then they would expect carriers of this variant to have a 13% reduced risk of CHD.

To test whether carrying the gene variant had this great an effect, they used a case-control study that included 20,913 people who had had a heart attack (the cases) and 95,407 control participants. They examined whether, as they expected, carriers of the variant had around a 13% reduced risk of being among the cases and to have had a heart attack.

In another part of the study, they examined further gene variants in what they called a “genetic score”. They identified the 14 gene variants that were most commonly associated with HDL cholesterol levels, and the 13 gene variants that were most commonly associated with LDL cholesterol. They tested these variants in a further 12,482 cases who had had a heart attack and 41,331 controls.

What were the basic results?

Carriers of the LIPG genetic variant (Asn396Ser) had HDL cholesterol levels that were slightly higher than people who did not carry this gene (about 0.14mmol/L higher). However, while this led researchers to  expect that people carrying this variant would have around a 13% reduced odds of having had a heart attack, they found that carrying this variant had no significant effect on risk of a heart attack (odds ratio [OR] for heart attack 0.99, 95% confidence interval [CI] 0.88 to 1.11).

Following this phase, the researchers looked at a person’s carriage of up to 14 variants that were associated with higher HDL cholesterol levels. They once again found that an increased "HDL genetic score" was not significantly associated with odds of having a heart attack. However, when they examined the LDL genetic score (based on a person’s carriage of up to 13 variants associated with higher LDL cholesterol levels), they found that this was associated with increased odds of having a heart attack (OR 2.13, 95% CI 1.69 to 2.69). In short, genetic variants that increased a person’s LDL cholesterol level were associated with higher CHD risk, as expected.

How did the researchers interpret the results?

The researchers concluded that certain genetic variants that raise blood HDL cholesterol do not seem to be related to the risk of heart attacks. They said that this data “challenges the concept” that raising HDL cholesterol levels will directly translate into reduced risk of a heart attack.

Conclusion

Previous research has tended to show that HDL cholesterol is “good” for you and higher levels reduce your risk coronary heart disease, while LDL cholesterol is “bad” for you and higher levels increase your risk of a heart attack. This complex research aimed to avoid the problem of the influence of other medical, biological or lifestyle influences by concentrating on genetics linked to HDL cholesterol and how closely they related to the risk of heart disease. If HDL cholesterol is directly related to CHD risk, then genes associated with high HDL levels should be directly associated with lower heart attack risk. Researchers carried out their study based on the theory that because our genetics are randomly assigned, participants can be considered to be randomly allocated to their circumstances and, therefore, equal.

However, the research did not find that HDL genetics determines the risk of heart disease. Instead, the gene variants that were associated with higher HDL cholesterol levels had no association with heart attack risk. This suggests that there may be no direct relationship between HDL cholesterol and coronary heart disease and, therefore, that other factors must be involved.

When the researchers examined gene variants that caused a person to have higher LDL (“bad”) cholesterol levels, they found that carriers of these variants were more likely to have had a heart attack than people without the variants. This would suggest that there is a direct causal relationship between LDL cholesterol and coronary heart disease, but not HDL cholesterol.

This complex study challenges the commonly held belief that having higher HDL cholesterol will reduce heart attack risk. However, this study alone cannot provide the whole answer and tell us whether HDL cholesterol has any effect on coronary heart disease, and how this effect might be medicated. Also, only a few gene variants were examined and there may be many other genetic influences on HDL cholesterol and other blood fats.

Importantly, while our genetics are determined at conception, the environment that we live in for the rest of our lives is likely to have an influence. Therefore, it is not possible to say how much our genetics influence our cholesterol compared with the many other risk factors for heart disease (such as diabetes and lifestyle factors including smoking, alcohol and exercise). Exercise in particular is thought to raise HDL levels during our lifetime, regardless of our genetic assignment at conception. This study cannot tell us how raising HDL cholesterol levels in adult life could influence coronary heart disease risk.