12th May 2018

[A classic black swan]

If your hypothesis is that all swans are white, the discovery of one black swan refutes your hypothesis. That is how science works. Or at least that is how science should work. In the real world, scientists are highly adept at explaining away contradictions to their favoured hypotheses. They will use phrases such as, it’s a paradox. Or, inform you that you didn’t measure the correct things, or there are many other confounding factors – and suchlike.

Anyway, accepting that the finding of someone with a very high LDL level, and no detectable atherosclerosis, will always be dismissed - in one way or another - I am still going to introduce you to a ‘case history’ of a seventy-two-year-old man with familial hypercholesterolaemia, who has been studied for many, many, years. Try as they might, the researchers have been unable to discover any evidence for cardiovascular disease (CVD) – of any sort.

In the past, I have spoken to many people with very high LDL and/or total cholesterol levels who are CVD free, even in very old age. The mother of a friend of mine has a total cholesterol of level of 12.5mmol/l (483mg/dl). She is eighty-five, continues to play golf and has not suffered from any cardiovascular problems.

However, none of these people had been studied in any detail. Which means that they can, and are, dismissed as irrelevant ‘anecdotes’. Yes, the widely used and highly exasperating phrase that I often encounter is that ‘the plural of anecdote is not data’. This, of course, is completely untrue, or at least it is untrue if you start dismissing detailed individual cases as anecdote.

Whilst an anecdote may simply be a story, often second hand, a case history represents a painstaking medical history, including biochemical and physiological data. In reality, the plural of case histories is data. That is how medicine began, and how most medical breakthroughs have been made. We look at what happens to real people, over time, we study them, and from this we can create our hypotheses as to how diseases may be caused and may then be cured.

So, a single case is NOT an anecdote, and cannot be lightly dismissed with a wave of the hand and a supercilious smirk.

In fact, the man who is the subject of this case history has written to me on a few occasions, to tell me his story. I have not written anything about him before, as I knew his case was going to be published, and I did not want to stand on anyone’s toes. With that in mind, here we go.

The paper was called ‘A 72-Year-Old Patient with Longstanding, Untreated Familial Hypercholesterolemia but no Coronary Artery Calcification: A Case Report.’ 1

The subject has a longstanding history of hypercholesterolemia. He was initially diagnosed while in his first or second year as a college student after presenting with corneal arcus and LDL-C levels above 300 mg/dL [7.7mmol/l]

He reports that pharmacologic therapy with statins was largely ineffective at reducing his LDL-C levels, with the majority of lab results reporting results above 300 mg/dL and a single lowest value of 260 mg/dL while on combination atorvastatin and niacin. In addition to FH-directed therapy, our subject reports occasionally using baby aspirin (81 mg) and over-the-counter Vitamin D supplements and multivitamins.

In the early 1990s, our patient underwent electron beam computed tomography (EBCT) imaging for CAC following a series of elevated lipid panels. Presence of CAC (coronary artery calcification) was assessed in the left main, left anterior descending, left circumflex, and right coronary arteries and scored using the Agatston score.

His initial score was 0.0, implying a greater than 95% chance of absence of coronary artery disease. Because of this surprising finding, he subsequently undertook four additional EBCT tests from 2006 to 2014 resulting in Agatston scores of 1.6, 2.1, 0.0, and 0.0, suggesting a nearly complete absence of any coronary artery calcification. In February of 2018, he underwent multi-slice CT which revealed a complete absence of coronary artery calcification.

So, here we have a man who has an LDL consistently three to four times above ‘average’. He had tried various LDL lowering agents over the years. None of which had done anything much to lower LDL. Therefore, his average LDL level over a twenty-year period has been 486mg/dl (12.6mmol/l.

Despite this he has absolutely no signs of atherosclerotic plaque, in any artery, no symptoms of CVD and is – to all intents and purposes - CVD free. What of his relatives? If he has FH, so will many others in his family.

‘He has one sister three years his senior who also has FH and a history of high lipid levels. She also has no history of myocardial infarction, angina, or other symptoms of coronary artery disease. His mother had FH, although she died of pancreatic cancer at age 77. She and her three siblings were never treated for, and had no history of, cardiovascular disease. The patient reports that his father had one high cholesterol score (290s), but was never diagnosed with FH, had no history of cardiovascular disease and died in his 80s during surgery for hernia repair.’

What to make of this? Well, I know that the ‘experts’ in cardiology will simply ignore this finding. They prefer to use the ‘one swallow does not a summer make’ approach to cases like this. For myself I prefer the black swan approach to science. If your hypothesis is that a raised LDL causes CVD, then finding someone with extremely high LDL, and no CVD, refutes your hypothesis.

Unfortunately, but predictably, the authors of the paper have not questioned the LDL approach. Instead, they fully accept that LDL does cause CVD. So, this this man must represent ‘a paradox’. They have phrased it thus:

‘Further efforts are underway to interrogate why our patient has escaped the damaging consequences of familial hypercholesterolemia and could inform future efforts in drug discovery and therapy development.’

To rephrase their statement. We know that high LDL causes CVD. This man has extremely high LDL, with no CVD, so something must be protecting him. I have an alternative, and much simpler explanation: LDL does not cause CVD. My explanation has the advantage that it fits the facts of this case perfectly, with no need to start looking for any alternative explanation.

And just in case you believe this is a single outlier, something never seen before or since. Let me introduce you to the Simon Broome registry, set up in the UK many years ago to study what happens to individuals diagnosed with familial hypercholesterolaemia (FH). It is the longest, if not the largest, study on FH in the world.

It has mainly been used as one of the pillars in support of the cholesterol hypothesis. However, when you start to look closely at it – fascinating things emerge. One of the most interesting is that people with FH have a lower than expected overall mortality rate – in comparison to the ‘normal’ population. Or, to put this another way. If you have FH, you live longer than the average person.

Even if we look at death from heart disease (those with FH have never been found to have an increased rate of stroke) we find that in the older population, the rate of death from Coronary Heart Disease (CHD) was actually lower than the surrounding population in some age groups.

For instance, in the male population aged 60 – 79 (who were CHD free on entry to the study) the rate of death from heart attacks was lower than the surrounding population. Not significantly, but it certainly was not higher.

In fact, in the total male population aged 20 – 79 with FH, the rate of death from CHD was virtually identical to the surrounding population. Over a period of 13,717 years of observation, the expected number of fatal heart attacks was calculated to be 46. The actual observed number was 50.

In women, the expected number of heart attacks in the population aged 20 – 79 was 40, the actual number of observed fatal heart attacks was 40. Which means that FH was not found to be a risk factor for CHD in those enrolled in the study – who had no diagnosed heart disease prior to enrolment2.

Which represents, I suggest, another fully grown black swan. There you go. Two in one day.

1: https://www.cureus.com/articles/11752-a-72-year-old-patient-with-lo...

2: https://academic.oup.com/eurheartj/article/29/21/2625/530400