The Patient Was…

Haven’t we heard enough about cholesterol?

The answer, like it or not, is no.

Few topics in health or medical care have received the attention that has been given to cholesterol. Although no precise number is available, various sources state that hundreds of thousands of scientific papers have been written about it over the past century. And an estimated twenty thousand have been published in just the last twenty years.

Enough already! Enough already? Uh – uh.

One important but largely neglected area regarding cholesterol, which has important implications for practical medical care, is the very reliability of cholesterol testing. Lifelong treatment decisions may be based on tests in which reproducibility and reliability may be lacking.

Consider, for a moment, scientific measurements that are generally regarded as absolute and immutable. The boiling point of water is 212 degrees Fahrenheit (or 100 degrees Centigrade). Right? Yes, but only under standard atmospheric pressure at sea level. If you measure the boiling point of water at higher altitudes, the boiling point temperature is lower. And if you make the measurement below sea level (imagine you’re at the Dead Sea), the boiling point is higher.

The point is, if you change the conditions of the test the result changes.

Look at it another way. We all know that weather forecasting is notoriously imprecise. The exact high temperature prediction is often a bit off. Actual rain amounts may vary considerably from what is forecast. Broader predictions, like rain versus no rain are more likely to be correct, but the small variations are less predictable. If the forecaster says the wind will be 10 miles per hour, I’ll wear my broad- brimmed Peruvian hat. But if there is an unexpected 20 mile per hour gust, I may be chasing my fedora down Broadway. The practical implications are real, but don’t usually have lasting and significant effects on us. They may be momentarily annoying  —  darn it, I should have brought my umbrella!  —  but the big picture of our lives isn’t altered.

So how does this relate to cholesterol? And why might it be important to our health and medical care?

There are three things that I would like you to consider: First, how precise is the measurement of cholesterol; second, how reproducible is the result; and third, how might the precision and reproducibility affect your treatment?

The first thing to understand is that cholesterol measurements are reported to you and your doctor as specific and precise numbers.  The report doesn’t say the value is probably between x and y. The report gives a very specific number, for example 100 mg/dl (meaning milligrams per deciliter). To confuse things more, in Europe and many countries around the world, values are reported in mmol/L, or millimoles per liter, numbers that are extremely different from mg/dl. A cholesterol level of 200 mg/dl, for example, is the same as 5.17 mmol/L.

The very specific value reported implies a very precise measurement. But the fact is that cholesterol measurements are not that precise at all. There is an accepted level of variation or error that can be close to 10 percent. So, a value of 100 mg/dl might actually be as high as 110 mg/dl or as low as 90 mg/dl. To compound the problem, different laboratories may measure cholesterol somewhat differently, introducing another variable into the mix. And some laboratories don’t actually directly measure all the lipid fractions (fatty substances in the blood) including LDL, the so-called bad cholesterol; HDL, the so-called good cholesterol; and triglycerides, another part of the blood lipids. Rather, they use mathematical equations to estimate some of the values they report. And the equations used may be different in different laboratories.

Well, if the measurements aren’t really precise, are they at least reproducible? In other words, is the imprecision consistent? If the value reported is 10 percent off the actual level of cholesterol in the blood, will repeated testing at least maintain the same level of inaccuracy, so that fluctuations in reported values may be considered to represent real changes in the patient? Here, again, the answer is no.

Repeating the test within a short time of the initial test, with no recognizable change in the patient or the conditions of the test, will not necessarily give the same result.

In a reputable study, in 2024, of over 400 patients who underwent repeated cholesterol testing within a short time period, over half of the subjects had a difference greater than the acceptable total error of 8.9% in LDL-C calculation. In fact, there was a mean 25.8% difference between the two tests.

More recently, an expert consensus panel established “error goals” for lipid tests: Total Cholesterol tests were permitted an error level of up to 9%, LDL cholesterol tests were allowed a range of variability of up to 12%, and HDL cholesterol tests up to 15%.

If cholesterol tests are neither precise nor reproducible, how does that affect the treatment that you and your doctor agree upon?

To answer that, first consider how treatment choices regarding cholesterol are usually made. There are two major components that affect treatment decisions in regard to cholesterol: The first is your personal level of risk for heart disease; the second is the level of cholesterol in your blood that is considered therapeutically appropriate for your level of heart disease risk.

Here, again, lack of precision rules. Calculating an individual’s risk for developing coronary heart disease, for which elevated cholesterol is the leading and most potent risk factor, is a very imprecise exercise. There have been several formulas developed in recent years to calculate one’s risk of developing coronary disease, but even the most established and widely used ones still omit some key factors that can play a major role in determining risk. Inflammation, for example, which markedly increases one’s risk of developing coronary heart disease, is only recently being recognized in formulas and algorithms for risk assessment. [Note: The issue of inflammation and heart disease is addressed in an upcoming posting in this blog, ThePatientWas.com].

Based on recent scientific studies, experts have recommended specific levels of blood cholesterol for varying degrees of individual patient risk of developing coronary heart disease. Without going into too much detail, suffice to say that there are currently three major recommended blood levels of LDL cholesterol (bad cholesterol), depending on whether a patient is considered low, intermediate or high risk. These levels are 100 mg/dl, 70 mg/dl, and 55 mg/dl.

Note that these recommended levels of blood cholesterol, however, are like those reported in blood tests: very specific and precise.  But the reality, as we’ve been discussing, is that there is no precision or reliable reproducibility in the measurements of a patient’s blood cholesterol. Now, think about the implications for treatment.

Suppose, based on your doctor’s assessment of your individual risk for developing coronary heart disease  —  often an imprecise assessment, at best —   he or she selects your “desirable goal” of LDL cholesterol as 100 mg/dl. Now, suppose your blood test reports a value of 105 mg/dl. Does that mean you need to embark on a treatment regimen to lower your level to 100 mg/dl?

We just learned that due to imprecise measurement, your level might actually be below 100 mg/dl, say 95. Would you then not need any treatment? The same dilemma could apply at any level of reported LDL cholesterol and the desirable or recommended goal level. If your goal is 55 mg/dl due to an assessment of high risk, and through various medications you attain a reported level of, say, 58 mg/dl, do you still need more intensive treatment? Remember, the imprecision of the blood test means your actual level could already be at, or even below, the goal.

Before concluding that there is too much uncertainty around cholesterol, and too much reliability on imprecise numbers, so you should just forget about it, please consider the following.  I believe that elevated blood cholesterol is the major risk factor for coronary heart disease which, in turn, is the leading cause of mortality in the world. Cholesterol testing is a valuable tool in determining a patient’s risk for developing coronary disease; in most instances, the imprecision and non-reproducibility in testing cholesterol levels will not affect risk assessment and treatment decisions.  But in instances where minor differences in numbers may have important implications for treatment decisions that may impact the rest of your life, consider detailed discussions with your doctor, repeat testing, and recognition that clinical medicine is always a humanistic experience requiring knowledge, empathy, understanding, and patience.