Estimating My Heart Disease Risks
Details of my attempts to understand and deal with high cholesterol
Summary
My LDL (“bad”) cholesterol has been high (well above 150 mg/dL) since I started measuring it in my mid-20s. While my HDL-C (“good”) levels, and even the ratio between them has always been in the healthy range, my recent unexplained drop in Apple Watch VO2 Max levels has me wondering if I should revisit the affect this has on my heart health.
My History of High Cholesterol
I have no family history of cardiovascular disease. Both my parents are in their 80s and healthy. My grandparents who lived into their 90s died of other ailments. There are some strokes in the family, but always in people well over 80 years old.
My blood cholesterol levels have been high since I began measuring in my 20s. During my 40s, after marriage and children reminded me of my obligations to my family, I became convinced that I should control my high levels through statins, and for about ten years I took 20mg of simvastatin, with great results.
A major factor in my decision to take statins was the landmark “4S” study that showed a decrease in all-cause mortality statin users. The study convinced me because unlike many so-called “proofs” of statin effectiveness, this was a large randomized trial that looked at overall death rates, including from seemingly unrelated causes. In other words, even if statins come with unknown unexpected side effects, the consequences still result in longer life among those who take them compared to those who don’t. Case closed.
Or so I thought. Eventually as I became more skeptical of drug interventions in general, I noticed an important caveat in the study: the trial participants all had a history of heart disease – many of them quite serious. But that’s not me, and it’s not my family either, so out of a general sense of caution about the long-term consequences of taking powerful medications, I stopped taking the statin. The history of medicine is littered with examples of “expert” advice that turned out later to be incorrect, and it seemed to me that, despite the good evidence that statins are safe and effective, it was dangerous to spend a lifetime taking a powerful therapeutic without good understanding of the root cause for my high cholesterol. Given my healthy family history, is it possible that such a powerful intervention might actually cause other problems, especially if taken daily over a lifetime?
I test myself regularly, so here’s a closer look at my cholesterol changes over the past few years:
Incidentally, my high cholesterol levels are unusual. Roughly 10% of men over age 60 have LDL-C > 200 (according to a 2022 NHANES study). There isn’t an apparently difference by ethnicity, but women are slightly higher than men. Interestingly, the rate is highest (15%) among ages 40-59.
Is My Heart Okay?
It’s important to remember to care about the heart, not cholesterol. A blood test, at best, is an indirect measure of health. If high blood cholesterol is a factor in heart disease, it is only because it has an effect on the heart itself, or on the rest of the circulatory system through hardening or blockage of the arteries. Rather than an indirect measure like LDL-C, wouldn’t it be better to measure those effects directly?
Fortunately, there are many low-cost ways to find evidence of arterial blockage or thickening. For under $100, I found a local imaging center that could test me.
A CT calcium scan is a simple procedure where you lie down, fully clothed, as an expensive machine scans your body for signs of calcium that can form in blood vessels, presumably from long exposure to calcifying cholesterol. The resulting coronary artery calcium (CAC) score is a simple estimate of how hardened the arteries are.
I’ve tested a few times, before and after ending statins and found that my CAC is quite low. Measured on the Agatston scale, my numbers were always well under 50, probably ten times lower than you’d expect if there were problems.
Another test, Carotid Intima-Media Thickness (CIMT), showed similarly very clear arteries (under 0.50 mm).
These results, combined with my lack of other risk factors, made me feel vindicated for having given up statin therapy. I discussed my situation with several different doctors. While nobody was happy with my high LDL-C levels, they agreed that long-term therapies have other consequences; given my otherwise healthy lifestyle, nobody could make a case for why one risk significantly outweighed the other. I continued to get regular exercise, maintaining a healthy diet and weight, and of course kept my regular schedule of annual physicals and frequent contact with my doctor.
Based on evidence that the standard measurements of LCL-C don’t show the whole picture, I tested two other components with contradictory results.
- Apolipoprotein B (ApoB) 200 mg/dL (“dangerously high”).
- Lipoprotein(a): 41 nmol/L (“optimal”)
Meanwhile, my high HDL-C (“good”), low triglycerides, and the various ratios that appear to matter all seemed to point toward “no problem”. I wasn’t going to worry about it.
Here are my July 2024 detailed results:
But Why am I so Easily Out of Breath?
Then, in the summer of 2020, despite my regular exercise, I began to notice a disturbing trend in both my self-perceived levels of fitness and in the VO2 Max estimates made by my Apple Watch. Climbing stairs quickly never bothered me, but even my wife began to notice how winded I appeared to get after such a simple activity. And I noticed a decisive plunge in my VO2 Max:
Because my Apple Watch tracks my heart rate along with my movement, it’s possible to roughly estimate my overall fitness as measured by the maximum volume (V) of oxygen (O2) my lungs can handle, the so-called “VO2 Max”. Considering the exercise I do, my numbers were mysteriously never all that great for my age, but the recent steep drop was enough for me to consult with my doctors again.
Not only that, but I noticed a long-term increase in my resting heart rate. Apple Watch shows a clear year-over-year pattern:
Cardiologist Advice: Statins!
I take health risks seriously, so in mid-2020 I visited a cardiologist. Although he couldn’t detect anything unusual in his careful examination of my heart at rest, he emphatically insisted that I return to statin therapy immediately. While there are always outliers, my very high LDL-C combined with my body’s tolerance for statins, made this a no-brainer, he said. My low CAC wasn’t necessarily a good indication of my risks, particularly for stroke, since a CT scan won’t see the “soft” cholesterol deposits that haven’t yet formed into calcium. A blood vessel could be 80% blocked, he said, without showing signs of calcification.
So in mid-2020 I taking statins again. The cardiologist recommended atorvastatin (aka Lipitor), at a more aggressive 40mg dose. After a month of daily use, my cholesterol declined, and after two months it was back to the levels I’d seen when I’d used statins a decade before.
Incidentally, as expected note how my LDL-C rises dramatically when eating a low-carb, high-fat (“Ketogenic”) diet.
One more incidental observation: the significantly lower results from the April 1 test followed three days of heavy eating on my version of the Feldman protocol: more than 3K calories/day, at least half of which were fat.
The Case for Statins
The most up-to-date (updated 2019) guidelines from the American College of Cardiology are clear for somebody like me with LDL-C > 200. I am at high risk, period, and should be on aggressive statin therapy regardless of other risk factors.
In patients with severe primary hypercholesterolemia (LDL-C level ≥190 mg/dL aka ≥4.9 mmol/L), without calculating 10-year ASCVD risk, begin high-intensity statin therapy. If the LDL-C level remains ≥100 mg/dL (≥2.6 mmol/L), adding ezetimibe is reasonable. If the LDL-C level on statin plus ezetimibe remains ≥100 mg/dL (≥2.6 mmol/L) and the patient has multiple factors that increase subsequent risk of ASCVD events, a PCSK9 inhibitor may be considered, although the long-term safety (>3 years) is uncertain and economic value is uncertain at mid-2018 list prices.
Years of taking 20mg of simvastatin lowered my LDL-C below 100, so probably I don’t need the extra ezetimibe boost.
American College of Cardiology Lifetime ASCVD Calculator says my lifetime risk is 50% based on my Sept 2020 blood test results.
The dangerous association between high LDL cholesterol and heart disease is abundantly clear. For example, see this lengthy review in Goldstein and Brown (2015):
summarizes the scientific evidence that converged from multiple disciplines to implicate cholesterol-carrying low-density lipoprotein (LDL) as the instigator of atherosclerotic plaques and high dietary fat as the major cause of pathologic LDL levels
Cochrane Reviews (2013) concludes that statins are a good idea for preventing heart attacks:
Of 1000 people treated with a statin for five years, 18 would avoid a major CVD event which compares well with other treatments used for preventing cardiovascular disease. Taking statins did not increase the risk of serious adverse effects such as cancer. Statins are likely to be cost-effective in primary prevention.
There are exceptions. For example, a 2016 Review (Ravnskof et al) says that, among people over age 60, LDL appears not to make a difference in mortality. My mother, well into her 80s has my blood type (A+) and remains healthy and active despite cholesterol levels similar to mine.
The Case Against Statins
But maybe LDL-C isn’t the whole story. I’ve tested a few other parameters, so what do they say about my overall risk?
My apoB is between 140 mg/dL (Sep 2020) and 200+ (Jul 2024) which “constitutes a risk-enhancing factor”. Although I should be careful relying on this number because some labs may not measure it reliably, different labs show consistent results for me, so I think it’s likely I’m at risk. According to Grundy et al (2019):
Because apoB is the major apolipoprotein embedded in LDL and VLDL, several investigators identify strength of association between apoB and ASCVD
On the other hand, in Summer 2024 my Lp(a) is around 41 nmol/L, though up from 22 nmol/L a few years ago, is still well out of the risk range. (Optimal is < 75 nmol/L)
Still, the guidelines suggest you only bother testing Lp(a) if you’re unsure about the other risk factors.
InsideTracker (2024) thinks LDL particle size is not yet worth measuring:
Based on current research, Apolipoprotein B has emerged as the superior marker for assessing cardiovascular disease risk. LDL particle size also has quality evidence behind it, but as it’s a metric that’s not yet as standardized nor as strongly predictive as ApoB, ApoB remains a better assessment of heart health than LDL particle size alone.
Similarly, coronary artery calcium (CAC) is useful for borderline cases, but any Agatston score above 0 “favors statin, especially after age 55”. My scores hover below 50, but never 0, so once again this is a case for statin therapy.
It’s not hard on the internet to find credentialed cardiologists who think LDL-C by itself isn’t especially predictive. For example Dr. Mark Houston is a cardiologist who says
If you look at total LDL you cannot predict anything about risk, what the treatment is going to do and how far you need to drive the LDL down. So total levels, no matter whether you’re looking at, HDL or LDL, to my reading, are non useful information.
Instead, he thinks insulin response is more important. (My insulin levels are fine).
What About Others?
For white males, incidence is 7.26 per year after age 65. This is per 1000 people.
Also note that approximately 34% of adults aged 65 to 75 are on statins for primary prevention. Without knowing for sure which of the statin-users are having heart attacks, it’s difficult to say whether the statin is really doing any good.
That said, if that many Americans are taking statins, wouldn’t you expect a dramatic decline in heart disease? Look at charts comparing the steady decline in heart disease to smoking, and see if you notice anything different since statins became widely prescribed. Spoiler: at a population level, you won’t see a difference.
Statins in older adults
This 2019 from American Family Physician: Overuse of Statins in Older Adults and adds
The U.S. Preventive Services Task Force guideline states that current evidence is insufficient to assess the balance of benefits and harms of statins in people older than 75 years who have no history of stroke or heart attack.2 For people 65 to 75 years of age with one or more risk factors, the U.S. Preventive Services Task Force recommends that clinicians selectively prescribe statins for those with at least one risk factor and a 7.5% to 10% risk of a cardiovascular event in 10 years.2
Summary of Absolute Numbers
Assuming a hypothetical population of 10,000 men over 60, ChatGPT thinks I have a roughly 25-30% chance of a heart attack. That’s awful of course, but it’s not that much more awful than the 10-15% chance I’d have anyway. And if I lower my LDL “artificially” through statins, does that really improve my chances?
| Category | LDLC/ApoB Level | Estimated Cardiac Events in 10 Years |
|---|---|---|
| High LDLC | ≥160 mg/dL | 2,000-3,000 |
| High ApoB | ≥120 mg/dL | 2,500-3,500 |
| Normal/Low LDLC | <130 mg/dL | 1,000-1,500 |
| Normal/Low ApoB | <100 mg/dL | 1,000-1,500 |
Cholesterol Code
Dave Feldman is a fellow software engineer and, like me, a curious amateur, who proposes a condition he calls Lean Mass Hyper Responder (LMHR), characterized by people just like me with lipids:
- LDL of 200 mg/dL (5.17 mmol/L) or higher
- HDL of 80 mg/dL (2.07 mmol/L) or higher
- Triglycerides of 70 mg/dL (0.79 mmol/L) or lower
His Cholesterol Code Report shows I’m at low risk on every one of the studies he tracks. I remain at low risk even after after a Keto Diet, where my LDL-C shot up to 300 nmol/dL.
I’m now following their 6K+ member Facebook group for more detailed discussions.
Test for Plaque at Home
Test your arterial plaque levels at home using the Ankle Brachial Index. Using an at-home blood pressure cuff, measure the blood pressure of both arms and both ankles. Look up a chart at Ankle Brachial Pressure Index (ABPI): An update for practitioners
I tested this using a $30 iHealth blood pressure monitor.
My readings were pretty variable. It’s important to align the air bladder with the artery, which can be tricky to find on the legs. After multiple tries, here’s what I found
| R Arm | 115/85 |
| R Ankle | 146/72 |
| L Ankle | 137/71 |
| L Arm | 117/84 |
The important number is the ratio between the ankle and arm. Take the highest of that first, systolic, number for the ankle and divide by the highest of the systolic number for the arm.
Roughly speaking, I’m about 1.2 ratio which is basically normal. While this doesn’t entirely get me off the hook, I’m encouraged that at least I didn’t get an abnormal result. If this test had shown an abnormal result, I would want to get a professional to look at me again.
Echo Stress Test
The ultimate test of heart health is an echo stress test, and so in March 2021 I found myself on a treadmill, plugged into an EGK and blood pressure cuff. Under carefully controlled conditions, I ran as hard as I could so a specialist could study my heart with a live ultrasound.
Results
The stress test said my heart is fine:
- Stress ECG findings: No chest pain or ECG changes of ischemia noted. Exercise time: 12:37 min on Bruce protocol. Duke treadmill score = +12, which predicts a low risk for future adverse cardiac events at 5 years.
- Arrhythmias: No significant arrhythmias.
- Hemodynamic findings: Blunted BP response to stress
- Stress Echo findings: There are no stress induced wall motion abnormalities with an adequate exercise workload. Stress echo is negative for ischemia.
- Risk: This study predicts a low short-term risk for future major adverse cardiac events
- Functional aerobic impairment (FAI): < -20 %
- Age equivalent: 36 years
It completely contradicts the Apple Watch-estimated VO2Max that had been driving my concerns.
Exercise capacity: Overall, the patient’s exercise capacity was excellent. The estimated VO2 max is 45.09 ml/min/kg.
and, based on the performance of others who’ve taken this test, it computes:
- Workload achieved: 12.8 METS
- Baseline heart rate: 85 bpm
- Baseline blood pressure: 110/70 mmHg
- Peak heart rate: 182 bpm
- Peak blood pressure: 130/60 mmHg
- Target heart rate: 139 bpm
- % of max predicted heart rate achieved: 112%
- Max predicted heart rate (220-Pt age): 163 bpm
- Estimated VO2 max: 45.09 ml/kg/min
- One minute heart rate recovery: 160 bpm
In other words, I passed the functional heart test with flying colors.
Conclusion
Other than stubbornly-high total and LDL cholesterol levels, I find it difficult to make a case for why I should be concerned about cardiovascular disease, at least in my current healthy state. I worry about my high LDL, but I worry about the long-term effects of a statin too.