This is a draft I’m writing in public, so please excuse the rough notes below . If you’d like updates when I publish something new, please subscribe to my mailing list. I’ve challenged myself to write 50 blogs in 2020.
https://twitter.com/KetoAurelius/status/1260204394035007490
https://twitter.com/KetoAurelius/status/1275114152017776645
https://twitter.com/Mangan150/status/1280274442908647424
https://www.tandfonline.com/doi/full/10.1080/17512433.2018.1519391
Here are some links to P. D. Mangan’s collection of writing about cholesterol. You can search his site or his Twitter account, @Mangan150, for related terms such as coronary heart disease, insulin, and cholesterol. I have heard the claim that eating meat and cholesterol leads to heart disease. I wanted to look more closely at the evidence behind it. (For example, a hamburger meal includes more than meat: bread, sauces, and other ingredients can complicate the comparison.) This collection of research shaped my questions about that assumption.
http://roguehealthandfitness.com/higher-cholesterol-associated-with-longer-life/
http://roguehealthandfitness.com/cholesterol-for-brain-health/
http://roguehealthandfitness.com/cholesterol-vs-blood-sugar-in-heart-disease/
Cholesterol is an organic molecule produced by cells throughout the body. The liver produces much of the cholesterol the body uses each day. It is a component of cell membranes and helps build and maintain them. The body also uses cholesterol to make vitamin D and steroid hormones, including cortisol, aldosterone, progesterone, estrogen, and testosterone. It also plays a role in transport and signaling between cells.
The brain produces its own cholesterol, which accounts for a substantial share of the body’s cholesterol. It helps maintain brain cells and the signals involved in thought, movement, and sensation.
Some familiar claims about cholesterol trace back to older research, and newer findings deserve attention too. One common account says that eating meat raises blood cholesterol and thereby contributes to atherosclerosis, the narrowing of arteries. This account often distinguishes “bad” LDL (low-density lipoprotein) from “good” HDL (high-density lipoprotein).
The Seven Countries Study
Ancel Keys helped popularize the view that cholesterol was connected to heart disease. He was also known for developing K-rations for U.S. soldiers during World War II. Keys observed cholesterol in arterial plaque and drew conclusions about its role in disease. This account argues that plaque may also be part of the body’s response to damage and inflammation. Critics of Keys’s Seven Countries Study argue that the countries selected supported his hypothesis about diet and heart health. They point to countries such as Sweden, France, and West Germany, where meat consumption and heart-disease rates did not seem to fit that pattern. The Cretan participants drew particular attention because they had low reported rates of heart disease and appeared to eat relatively little fat. The critique argues that wartime and postwar food shortages may have affected their diet and body weight. Part of the research also took place during Lent, when some Greek Orthodox participants avoided meat and cheese. That timing may have influenced what the study recorded about meat consumption.
The notes I collected also describe Keys’s association with Procter & Gamble, which sold Crisco as an alternative to butter. The product was marketed as a vegetable-based fat. Procter & Gamble sponsored radio programs and advertised products such as soap during the breaks. That sponsorship gave rise to the term “soap opera.” The account says Keys helped promote Crisco, whose price and vegetable-based branding appealed to consumers. This was also a period when heart disease was drawing growing public attention. Ancel Keys also became the head of the American Heart Association, which was then a small organization. The organization became an influential source for journalists covering heart disease. The research funding of those involved increased. The notes argue that dissenting researchers had less influence and less access to funding. I would want to verify how the AHA’s guidance has changed over time before drawing a firm conclusion.
These are rough notes drawn from the links above. The linked articles include graphs and references to studies.
Cholesterol and later research
The cited Japanese research reports an association between higher cholesterol and longer survival in the groups it studied. The lowest-cholesterol group had the highest mortality rate in that analysis.
A Dutch study of 724 people aged 85 or older reported a similar association. The group with the lowest cholesterol, less than 193 mg/dl, had the highest mortality rate, year after year. The group with more than 252 mg had the lowest mortality rate.
A Finnish study followed 490 older adults for six years. Mortality was highest for the group with the least cholesterol, less than 194 mg/dl. Mortality was far lower for the middle group, and lowest for the group with more than 232 mg.
A study published in BMJ Open examined LDL cholesterol and mortality among older adults. Some analyses reported an inverse association. The negative connection was seen in most of the people above the age of sixty. In those groups, higher LDL was associated with lower mortality; that observation alone does not establish a cause.
The authors discuss possible explanations, including a role for cholesterol in infection and cancer, but these remain questions rather than conclusions here.
The notes also cite studies reporting an association between lower cholesterol and violent behavior. For people with a cholesterol value of 180 mg/dl or lower, one reported odds ratio was 15.49.
Several studies have found a connection between low cholesterol and a high suicide rate. One study reported that the lowest-cholesterol quartile had more than six times the suicide risk of the highest-cholesterol quartile.
The commercial role of statins, medicines used to lower cholesterol, is another issue these notes raise. Funding sources are worth examining when reading cholesterol research. Industry-funded studies can be more likely to favor their sponsors.
The notes say that changes to research-transparency rules in 2004 affected how trial results were reported. They also claim that newer studies reported smaller effects of statins on heart outcomes.
The notes speculate about how population growth and food policy might influence nutrition advice. I would want stronger evidence before treating claims about grains and inflammation as settled.
The role of meat and fish in early human diets is another part of this argument. The notes contrast those diets with later grain agriculture, though the history is more complex than this brief account can show. They also link early agriculture to changes in average height. Archaeological remains can offer clues about joints and teeth, but the causes of those changes need careful interpretation.
Doctor Ravnskov's Findings
Uffe Ravnskov, a Danish physician who worked at Lund University in Sweden and specialized in internal medicine and nephrology, has contributed to research cited in these notes. In an article titled People With High Cholesterol Live the Longest, he summarizes evidence he believes favors higher cholesterol levels.
Ravnskov cites a report by Harlan Krumholz of Yale University in which older people with lower cholesterol had a higher heart-attack mortality rate than those with higher cholesterol. He points to eleven other studies he says show a similar pattern. He also discusses seven studies available at the time that did not link higher cholesterol to greater mortality, with more research published since.
Ravnskov writes that at least fifteen studies found an inverse association between mortality and total or LDL cholesterol. In those studies, higher cholesterol was associated with longer survival.
A meta-analysis cited in these notes examined nineteen studies and more than 68,000 deaths. It reported higher mortality from some digestive and respiratory conditions among people with lower cholesterol.
Jacobs and colleagues at the University of Minnesota followed more than 100,000 people for fifteen years and reported more infection-related hospitalizations among those with lower cholesterol.
A sixteen-year study cited here reported higher AIDS-related mortality in its lowest-cholesterol group than in its highest. The notes also discuss other possible factors, including prior infections and liver disease.
Children with Smith-Lemli-Opitz syndrome can have low cholesterol and a high rate of infections. The cited account reports fewer and less severe infections after cholesterol was raised through treatment or diet.
Researchers at the University of California examined records from 137,000 patients at 541 hospitals after a heart event. Their reported cholesterol levels were lower than the author expected. The average value was a low 174 mg/dl, and the "bad" LDL cholesterol was much lower than usual. The notes describe it as a large study of cholesterol among heart patients.
Cholesterol can fall after an acute heart event, which complicates the interpretation. The account says the lower values were recorded before the expected decline two to three days later. It estimates the later decline at no more than 15 percent.
A study at the Henry Ford Heart and Vascular Institute in Detroit reported a similar pattern. About half of 500 patients hospitalized with a heart condition had LDL below 105 mg/dL. After three years, 26 patients in the lower-cholesterol group and 12 in the higher-cholesterol group had died. The account also notes differences in statin use between groups, another factor to consider.
Ravnskov and fifteen colleagues reviewed nineteen studies that followed older adults over several years. They reported no consistent association between higher LDL and heart-disease mortality in those studies. Many of the studies reported longer survival in groups with higher LDL.
Cholesterol vs Blood Sugar in Coronary Artery Disease
The cited study reports these average values among U.S. patients with coronary artery disease:
Total cholesterol: 174
LDL cholesterol: 105
HDL cholesterol: 40
Triglycerides: 161
In that account, total and LDL cholesterol are not elevated, while HDL and triglycerides fall outside the ranges the author favors. The point is that many of these patients did not have especially high LDL levels. The figures come from a study of 136,905 patients.
Luca Mascitelli and colleagues discuss studies of cholesterol-lowering treatments other than statins and their effects on coronary mortality. The treatments included clofibrate and hormone therapy for women. (The notes report that clofibrate was discontinued after studies raised concerns about mortality and suicide risk.)
Blood Sugar
Acute coronary syndrome includes heart attack and unstable angina. The cited figures suggest that 57 percent of patients hospitalized with ACS had abnormal glucose metabolism. About 25–30 percent had diabetes. Others had impaired glucose tolerance.
In a cited study of 1,300 patients with impaired glucose tolerance, those treated with acarbose had fewer reported cardiovascular problems.
Acarbose affects blood glucose and may also influence insulin levels and sensitivity. In this account, improved sensitivity means the body needs less insulin for the same task. The notes report higher rates of fatal coronary disease among people with diabetes. Type 2 diabetes often involves reduced insulin sensitivity, also called insulin resistance. Diet is one possible contributor, though insulin resistance has multiple causes. (The notes describe a process in which the body needs more insulin to regulate blood glucose and, over time, may struggle to keep glucose within a healthy range.)
Gerald Reaven was an influential researcher of insulin resistance and related metabolic conditions. He and his colleagues followed 208 adults for 6.3 years. The group with the lowest insulin resistance, the lowest tertile, had no reported cases of the listed outcomes in this account. The middle tertile had twelve cases, and the highest had twenty-eight.
The study links insulin resistance to later health outcomes. It does not, by itself, establish how much sugar each participant consumed. It made me want to pay attention to blood glucose and insulin resistance as well as cholesterol.
These notes also mention intermittent fasting as a possible way to affect insulin levels. The proposed mechanism involves changes in insulin sensitivity.
Cholesterol for Brain Health
The brain makes up a small share of body weight but contains a substantial share of the body’s cholesterol. Cholesterol helps maintain brain cells and signaling between them.
Some studies of older adults report an association between higher cholesterol and better cognitive performance. Those findings do not by themselves show that low cholesterol causes cognitive decline.
In one cited group aged 85 and over, higher serum cholesterol was associated with lower dementia risk.
Another cited study of middle-aged women reports an association between memory performance and total and LDL cholesterol.
A study of Finnish men reported an association between egg consumption and lower rates of dementia and Alzheimer’s disease. The reported estimate was an 11 percent lower dementia risk per additional half egg a day; it was an association, not proof of cause.
Beatrice Golomb reviewed studies from 1965 to 1995 and discussed reported links between lower cholesterol and violent behavior or exposure to violence. Some trials she reviewed reported more violent deaths in groups assigned to cholesterol-lowering interventions. The review also mentioned animal studies of aggression and cholesterol. (One proposed explanation involves serotonin signaling, but these notes do not establish that mechanism.)
One reported comparison found a higher suicide risk in the lowest-cholesterol quartile, below 162 mg/dL.
Some studies report an association between lower cholesterol and depression among older men.
Other cited observations found low cholesterol among people with severe depression.
A Mexican study cited here reported associations between low cholesterol, high triglycerides, severe depression, and attempted suicide.
The notes also cite a small study in which some people taking statins reported memory problems after two months. Some participants reported improvement after stopping the medicine under study conditions. A few reported a return of symptoms when they resumed it.
Another cited series of 174 patients reported cognitive improvement in many participants after statins were stopped. The notes describe apparent improvements even among some people who had received dementia diagnoses; that would need careful verification.
Some observational research has reported an association between statin use and Parkinson’s disease. Other studies cited here report a lower Parkinson’s risk among people with higher cholesterol.
The final cited comparison reports higher rates of several diagnoses among statin users than among nonusers: hyperlipidemia (86.3%), hypertension (69.6%), diabetes (36.0%), osteoarthritis (31.5%), coronary artery disease (26.1%), hypothyroidism (21.5%), and depression (19.3%). Those figures alone do not show that statins caused the conditions.