Each morning at the retirement community, the healthy 88-year-old man received a delivery of 25 soft-boiled eggs, which he would consume during his day. This had been his way for many years. He'd had one experience of chest pain that might have been angina, but aside from that, he had a healthy cardiovascular system. He recognized that his only problem was psychological: "Eating these eggs ruins my life, but I can't help it."
I think of the Eggman, a brief case report from 1991 in the New England Journal of Medicine, whenever "news" of cholesterol's unsuitability as a one-size-fits-all biomarker resurfaces, as it does every few years and did again just last month.
Fred Kern Jr., MD, a gastroenterologist at the University of Colorado School of Medicine, heard about the man and saw an opportunity to study individual differences in how diet affects serum cholesterol level. And so the 88-year-old egg eater joined such famous patients as French-Canadian explorer Alexis St. Martin and Henrietta Lacks.
St. Martin accidentally shot himself in the abdomen in 1822, and when a hole remained after surgery, provided a window through which US Army surgeonWilliam Beaumont could observe and document how the stomach lining changes during digestion. And of course the celebrated cellular descendants of Henrietta Lacks' cervical cancer live on in labs throughout the world. Cholesterol — not the enemy?
First, Dr. Kern tested the man's lipid levels, which were normal: 200 milligrams per deciliter total cholesterol and 142 for LDL. Then he compared the extent to which the man's body compensated for the cholesterol overload in the 25 daily eggs to that of 11 volunteers who had their cholesterol tested under their usual eating habits and 16 to 18 days after adding five eggs a day. The Eggman was too obsessed to try to go without to help the study. Dr. Kern's analysis, albeit limited, revealed a lot about how our bodies differ in what we do with lipids.
The 11 volunteers on average absorbed 54.6% of the dietary cholesterol while on their normal diets and 46.4% while on the egg diet. But the 88-year-old absorbed only 18% of his dietary cholesterol. Plus, his liver made less cholesterol than is common and shunted more of it into bile acids (which aid digestion) than did the livers of the volunteers, a metabolic triple whammy that Dr. Kern called "extremely efficient compensatory mechanisms."
Basically the man's metabolism adjusted to the overload in a way that kept his blood lipid levels both healthy and constant, the very definition of homeostasis. And he likely isn't alone.
Genes lie behind our biochemistry, including lipid metabolism. The Eggman from 1991 is a perfect example of how personalized/precision medicine can not only identify individuals at elevated risk for specific diseases, but also those who are genetically protected. Despite recent presidential recognition and the breathless news coverage it spawned, precision medicine is hardly a new idea.It's genetics. And geneticists have known for decades that in some families, people can have very high serum cholesterol levels yet healthy hearts.
Back in 1991, the case report of the Eggman intrigued me because I, too, ate eggs every day when everyone was pushing cereal and low-fat diets, and my cholesterol levels were just fine. Nor did any of my relatives have any sort of heart disease, not even hypertension. But I still felt guilty eating omelets.
I continued to eat eggs, and like so many in the US in the 1990s who tried to follow low-fat diets, believing the anti-cholesterol mantra, I gained weight. We were drowning in high-fructose corn syrup, added to everything. But when I dug back into the biochemistry I'd learned in grad school, it quickly became apparent that cutting fats was not at all the best way to lower serum cholesterol or to lose weight. So I began researching the Atkins diet and its low-carb brethren.
In 2004, I began the South Beach Diet. Six months later, my weight and triglycerides had plunged.
Why carbs are bad for blood vessel linings is a matter of logic as well as biochemistry. But it's a bit circuitous, which may have contributed to the persistent perception that cholesterol is the enemy. It isn't as simple as what you eat showing up unchanged in your blood or on your frame.
Yes, cholesterol is a major part of the plaque that clogs arteries. It comes from the outside (diet; exogenous) or is made in the body (the liver; endogenous). Here is a good review. I'll summarize.
Triglycerides, sometimes called just "fat," consist of one glycerol group bonded to 3 fatty acid tails. They are dismantled during digestion, solubilized by bile acids in the intestines, absorbed into the bloodstream at the intestinal villi, and then the freed fatty acids are used to power muscles or are stored in adipose tissue.
Dietary cholesterol is ultimately ferried to the liver. Meanwhile, the liver is making its own cholesterol — from digested triglycerides. So a low-fat diet can reduce dietary cholesterol, and a statin drug can reduce the liver's production of cholesterol by blocking the rate-controlling enzyme HMG-CoA reductase.But consider these facts and relationships:
- Most of the cholesterol in the circulation comes from endogenous production (the liver), not from eating cheeseburgers.
- The liver makes cholesterol from triglycerides.
- 95% of lipids in the diet are triglycerides.
- Carbs (the low-fiber white kind: rice, pasta, potatoes) increase triglyceride levels.
Shouldn't we be limiting carbs, and not dietary cholesterol? It's the triglyceridesthat matter. So how did the vilification of dietary cholesterol arise?
Might the anti-cholesterol push have something to do with the $29 billion global market for statins? (Despite its suffering from "severe generic erosion" as patents expire.) Rampant fear of cholesterol is making statin manufacturers a lot of money.
In January 2012, having not had a check-up in years, I saw a new primary care provider, a nurse practitioner. She did a thorough history, and when my cholesterol came back a few days later at the high end of normal and my HDL not as high as it could have been, she insisted that I needed a statin, stat. That plus a low-fat diet, she admonished.
"Why?" I answered. "I have zero family history of cardiovascular disease and have no other risk factors. I exercise an hour a day. Low HDL is no longer considered a biomarker. And you mean a low-carbdiet, correct?"
She dismissed family history, not the best thing to say to a geneticist. Her refusal to consider my personal risk factors, or lack of them, and her prescribing a drug with rare although serious adverse effects was not something that appealed to me. So, "against medical advice," I went about my low-carb, statin-free lifestyle.
Three months later, the NP called to pitch statins again. Had I thought it over and changed my mind?
I referred her to an article in a recent Lancet showing that 14 genetic markers combined into a risk score to indicate high HDL – supposedly a sign of hearth health – did not lower heart attack risk, in many patients. This wasn't news; the report confirmed results of another from 2010. In fact a study of an HDL-boosting drug was halted in 2007 because it actually increased "cardiovascular events." Perhaps low HDL wasn't a valid surrogate for heart disease after all? Perhaps the NP should have been reading the medical literature instead of believing drug sales reps, but I suspect she didn't have the time.
Statins do lower endogenous cholesterol synthesis, and do save lives. But wouldn't they work best if prescribed only to people who actually need them?
My father, for example, was prescribed a statin at age 84, he too with no risk factors, and he developed muscle pain. That was in 2004. But I just checked the Lipitor package insert, and statins are indeed prescribed to people like him, with no personal or family history of heart disease and the only other risk factor his age — and maybe living into one's 80s indicates a healthy heart!
How many people are being over- or inappropriately treated by taking these drugs? How many health care providers actually determine an individual patient's risks? I know health insurers are not big fans of genetic testing, but how many years of one-size-fits-all statin therapy would it take to equal the cost of genetic tests to direct prescribing?
Lots of labs offer tests for single-gene variants that are important in heart health, such as apolipoprotein E (apoE) and angiotensin converting enzyme (ACE). Panels of tests are available too. GENESIS Center for Medical Genetics, Laboratory of Molecular Genetics screens for 8 genes and Vantari Genetics tests expression of several genes to guide drug selection (pharmacogenomics). GeneDx tests for dozens of single-gene conditions that affect cardiovascular health.
In the past, ordering genetic tests for extremely rare diseases involving lipid metabolism when evaluating an average junk food junkie for statin use made little economic sense. Lecithin cholesterol acyltransferase deficiency, for example, affects fewer than one in a million people. It causes very low levels of HDL cholesterol and very high total cholesterol – bad news! — but no associated cardiovascular disease.
But the gene-by-gene approach is headed towards extinction, now that the price of exome sequencing (the 85% of the genome that includes most disease-causing genes) has reached the $1,000 mark. A full genome sequencing can now be done in under a day.
Genome analysis takes longer than sequencing, of course, but cloud storage of DNA data will change that. And since tests for single genes that affect lipid metabolism and cardiovascular disease risk have been around for years, why not develop a probe for the relevant subset of the genome to screen statin candidates? Again, could the money insurance companies save on screening out people whom statins wouldn't help underwrite the cost of sequencing?
Not knowing much about health insurance, I googled to see if any would cover genetic testing to stratify patients for statin use. I didn't find any, but was thrilled when a statement from one major insurer popped up about genetic testing to identify people at high risk for the severe adverse effect of myopathy, which keeps some people from taking the drugs. Those on high statin doses face a 6-fold increased risk and those over 65 an additional 4-fold increased risk. Some insurers do cover the cost, or for detecting broken-down muscle in the bloodstream every few months, which seems a little late to me.
Paragraph 1 introduces the gene that affects myopathy risk, SLCO1B1. Paragraph 2 claims that "use of statins is associated with approximately 30% reduction in cardiovascular events in a wide variety of populations." But with genetic testing, aren't we talking about individuals, not populations?
The document ends with two statements in both italics and boldface. Important! One, that this stuff is complicated so a patient should speak to a physician (patients are dumb, doctors are smart, geneticists don't exist). Statement two, at the very end after the consumer presumably understands that genetic testing can avoid terrifically painful muscle degradation, I must quote, because it shattered my belief that the insurer was actually going to cover the cost of SLCO1B1 testing, because they bothered to explain it:
"Genetic testing for statin-induced myopathy is considered not medically necessary." This insurer "does not provide coverage for not medically necessary services or procedures."
So there you go. With pharmaceutical giants pushing drugs like statins to a supposedly genetically homogeneous population via doctors who might not have the time or expertise to explore the finer points of lipid genetics, and with insurers years behind the state of DNA science, we might need the hype surrounding precision medicine after all.
Caveat: I'm not an MD. None of the above is meant as medical advice. See your doctor after doing your own research to learn your disease and drug reaction risks. But I'm going to continue to eat my eggs every morning, exercise an hour or more each day, eat veggies and avoid white carbs – and not worry about my cholesterol.
This article first appeared on PLoS Blogs and is republished here under Creative Commons license. Image by Daniel G. under Creative Commons license.