The role of low-density lipoprotein in heart disease

Updated: Mar 16

There is now indisputable evidence that cholesterol-rich low-density lipoprotein (LDL) is causal in the pathology of cardiovascular disease (CVD). This is backed by decades of research from converging lines of evidence that consistency show LDL as an accurate and valid variable that is indicative of CVD risk.


The relationship between LDL, ApoB-100 and CVD

The relationship between plasma cholesterol levels and heart disease was first recognised in1953 by Oliver and Boyd (1), since then, and with the development in research methodologies in human epidemiology, clinical trials and genetic studies, the association has since been firmly solidified with this relationship ubiquitous in all lines of evidence.


ApoB-100 is a structural component of lipoproteins (which contain cholesterol) and each lipoprotein particle contains one ApoB-100 particle. Because LDL remains in circulation for a longer duration compared to its precursors (very-low-density lipoproteins (VLDL) and intermediate-density lipoproteins (IDL)), 90% or more of all ApoB-100 participles are attached to LDL. Therefore, ApoB-100 particles are an accurate measurement of LDL and studies will usually refer to LDL as either LDL-C or ApoB-bound LDL-C (2).


The role of LDL in atherosclerosis

Atherosclerosis refers to the hardening of an artery provoked by atheromatous plaque. The clinical manifestations of atherosclerosis include myocardial infarction and ischaemic stroke, collectively referred to as CVD. First, circulating LDL particles in the bloodstream influx within the subendothelium space in the artery wall. LDL is then retained and oxidized, where it actively participates in the excessive inflammatory and fibroproliferative response, causing atheromatous plaque formation (see Fig.1.) (3).


  • Figure 1. A visual representation of how LDL causes atheromatous plaque formation

With LDL-C being one driver of atherosclerosis, it has since been a target for drug development driving the production of LDL-C lowering agents in the late 1980's such as statins and cholesterol absorption inhibitors.


The relationship between statins & LDL

The liver is the primary target site for statins where they competitively block and bind to the active site of HMG-CoA reductase, a key rate-limiting enzyme in the mevalonate pathway responsible for cholesterol biosynthesis. Once bound, statins prevent HMG-CoA reductase from attaining a functional structure. The inhibition of HMG-CoA reductase not only reduces intracellular cholesterol but also increases the gene expression for LDL receptors responsible for LDL uptake from the bloodstream. Therefore, the reduction of cholesterol accumulation in the liver (hepatocytes) plus increased LDL receptor activity facilitates increased LDL-C clearance (and its precursors) from the bloodstream and increased cholesterol uptake by tissue (4). Statins can reduce circulating LDL-C levels by up to 55%, and there's a clear dose-response relationship between statin dose and LDL-C lowering (see Fig.2.)


  • Figure 2. The dose-response relationship between statin dose and degree of LDL-C lowering

The degree of LDL-C lowering & CVD risk

As the LDL-C lowering properties of statins are well established, how does this translate to heart disease risk? Well, there is a clear linear, dose-response relationship between the degree of LDL-C lowering and reduction in CVD risk (see Fig.3.). This means the reduction in LDL-C is proportional to the reduction in CVD risk. The relationship between LDL-C and CVD has led to a consensus amongst experts, with the latest position of the American Heart Association and The American College of Cardiology being “lower is better” for LDL [5].

  • Figure 3. The dose-response relationship between LDL-C lowering and CVD risk reduction (6)


To further emphasise the importance of LDL-C lowering therapies as primary prevention for CVD, even subtle changes in LDL-C translate to a clinically important risk reduction in CVD. Findings from Mendelian Randomisation trials help confirm this relationship.

  • Mendelian Randomization is a technique for examining the causal effect of a risk factor (LDL-C) on a disease (CVD) using a variation in a gene that codes for a known function (LDL biosynthesis).

Participants with specific genetic variants that result in lower circulating LDL-C have a significantly lower risk of experiencing a CVD event. This is an extremely intricate relationship to a point where even long-term exposure to 1 mmol/l (38.7 mg/dl) lower LDL-C is associated with a 55% reduction in CVD risk (see Fig.4.). Mendelian Randomisation studies by design reduce the inherent bias present in experimental clinical trials, and thus, can be a helpful addition to experimental data.


  • Figure 4. Proportional risk reduction of CHD for each LDL exposure allele (7)


There are in fact many genetic polymorphisms in cell surface receptors, enzymes, nucleur transcription factors, apoproteins and signalling intermediates (to name a few) that all influence lipid metabolism and have since been identified, characterized and catalogued to further our understanding of the relationship between lipids and CVD (See Table below 8). What is clear is all genetic variants that are associated with a reduction in LDL-C are also strongly associated with a reduction in CVD risk (8).



Is the relationship between LDL-C and CVD causal?

Determining causality in science can be ambiguous as no single study can provide a basis to isolate a causal variable. Instead, a convergence of all lines of evidence is needed with a logical interpretation of the consistency in which a marker influences an outcome. In short, if a particular exposure is consistently indicative of an outcome, then we can logically infer causality. While I won't discuss causality in-depth, Shaun Ward provides a great synopsis of causality in research which you can find here if interested in further reading.


As touched on earlier, there are plausible mechanisms as to how LDL-C initiates atherosclerosis and CVD. More importantly, the human outcome data is clear, the consistency is overwhelming with every LDL-C lowering therapy reducing CVD risk (see Fig. 5.)


  • Fig.5. The relation between achieved low-density lipoprotein cholesterol of control (square) and treatment (tri- angle) groups with per annum major cardiovascular event rate in select clinical trials where the annual event rate could be calculated (9).


An important point and part of the conversation that is often left out when discussing causality in disease is the multifactorial pathophysiology of disease. CVD is unquestionably multifactorial with various risk factors that tend to cluster and interact in individuals which is ultimately key in determining the level of CVD risk an individual risk factor poses.


Let's take high blood pressure, a well-established risk factor, as an example. In animal models with normal and hypercholesteremia, increasing blood pressure increases LDL retention in the subendothelium layer, a key early event in atherosclerosis. This highlights how one risk factor (blood pressure) can mediate the propensity to which another risk factor (LDL) influences CVD risk (by increasing LDL retention) (10).


Similarly, the degree of inflammation influences the degree of LDL needed to pose a risk, and vice versa (11). Importantly, all risk factors afore mention are independently causal in CVD, and if someone experiences a CVD event with low LDL-C yet high inflammation then this cannot negate the role of LDL - it simply highlights the multifactorial component of the disease. We do in fact have nice data controlling for all other risk factors except LDL-C which nicely builds on the mountain of evidence to highlight its independent causal role (12).


Expert consensus & coherence in data

While I am no expert in CVD, there are plenty of people who are. The role of LDL-C in CVD amongst experts in the field is widely accepted. In fact, The European Atherosclerosis Panel have nicely outlined several key points as to why LDL-C is causal in CVD (13), with some of the most striking as follows:

  • ''The consistency in the data: Over 200 studies involving more than 2 million participants with over 20 million person-years of follow-up and more than 150 000 cardiovascular events consistently demonstrate a dose-dependent, log-linear association between the absolute magnitude of exposure to LDL and the risk of atherosclerosis cardiovascular disease.

  • Reduction in risk with intervention: More than 30 randomized trials involving over 200 000 participants and 30 000 ASCVD events evaluating therapies specifically designed to lower LDL (including statins, ezetimibe, and PCSK9 inhibitors) consistently demonstrate that reducing LDL cholesterol (LDL-C) reduces the risk of ASCVD events proportional to the absolute reduction in LDL-C.

  • Coherence in the data: Monogenic lipid disorders, prospective cohort studies, Mendelian randomization studies, and randomized intervention trials all show a dose-dependent, log-linear association between the absolute magnitude of exposure to LDL and risk of ASCVD.''


Convergence of data from different study designs

Converging data from different research methodologies provides the most accurate reflection of the true relationship between an exposure and outcome. While unknowns may still remain, science is about making inferences based on what is possible to measure with the tools we have available. Each study design has its own unique value in research and when converging different lines of evidence, the limitations that are concomitant with each design are often attenuated as different designs account for the limitations found in others.


While RCT's provide a controlled experimental environment, they often lack the long-term exposure needed to access chronic disease. Prospective cohort studies provide that long-term measure, and with new advances in research, such as the identification of dietary biomarkers, we are further increasing the validity of our measurements from epidemiological data. Mendelian Randomization studies as mentioned before reduce many of the biasses present in experimental trials. Each design provides one piece of the puzzle, and when coherence exists across all study designs then causal and clinically important relationships can be identified (See Fig.6.).

  • Log-linear association per unit change in low-density lipoprotein cholesterol (LDL-C) and the risk of cardiovascular disease as reported in meta-analyses of Mendelian randomization studies, prospective epidemiologic cohort studies, and randomized trials. (14)



The great cholesterol-con con

Maybe it's somewhat pessimistic to presume all who review the data on LDL-C and CVD will come to the same conclusion, however, there are some who believe and affirm their belief to large audiences that LDL-C is not a marker of concern. These people typically hold tightly pockets of data that are either irrelevant or misconstrued.


You will often see populations presented whose cholesterol is high yet CVD incidence is low as evidence to refute the cholesterol hypothesis. A classic example of this is the 'Japenese paradox' (see below), even though this is not a paradox as it can be easily explained once critically analysing the data.


  • Considerably lower mortality rates from coronary heart disease (CHD) in the Japenese population compared to the U.S.A population, despite only marginal differences in mean total cholesterol

Context is everything here, of which these observations, or conclusions I should say, lack. Indeed, the Japenese Paradox dataset can easily be explained by the fact that the large majority of the sample size with elevated cholesterol is in the youth, not the elderly. As CVD is a chronic condition and a result of life-long high LDL-C exposure, it's no surprise the elevated cholesterol isn't correlated with an increase in CVD incidence if it's predominantly in the younger generation (15). In addition to this, the Japenese data set also observed a dramatic reduction in both both blood pressure and smoking dramatically throughout the study duration.


As demonstrated below (16), although total serum cholesterol is higher in the Japense popuation at age 25, it's clear from age 35 onwards there's a markedly reduced serum cholesterol across the Japenese population compared to the U.S.A - again, a simple observation that disproves the 'Japenese Paradox'.


Even if it were true that the Japanese population have high LDL-C and low incidence of CVD, how would one argue that further LDL-C reduction would not be beneficial when we have data showing how LDL-C reduction in the Japanese population further reduces CVD risk? (16)


In science, being wrong is good, it's only by being wrong do we learn what's right. However, when we are presented with evidence that shows us why we're wrong, with good faith, we must change our stance until proven otherwise.


As G.R. Thompson wrote in 2009, ''The foremost critics of the lipid hypothesis are now deceased but unfortunately for many of the patients with hypercholesterolaemia and coronary heart disease it took the best part of 50 years to disprove the sceptics. '' However, over recent years a resurgence of newfound sceptics are encouraging the public to dismiss the cholesterol hypothesis which will result in serious consequences for those that follow. If someone doesn't believe elevated LDL-C is bad, then why would they take statins? Why would they reduce their saturated fat intake?


Conclusion

CVD is multifactorial and LDL-C is one of many independent risk factors. This relationship is consistent across all lines of evidence and there is no coherent evidence to disprove this hypothesis.




Keep an eye out for part 2, where we'll delve deep into saturated fat and CVD.


Recent Posts

See All