The Ornish Myth

Does being a vegan mean you won’t get heart disease?

Heart artery plaque- low fat diet doesn’t prevent this

In 1998 Dean Ornish published his data showing a 3 percent reduction in the plaques seen by coronary angiograms on a select group of patients who followed his diet and “lifestyle” plan.  To be exact: they found 1.75% improvement after one year and 3.1% improvement after five years.  Where the control group increased by 2.3% in one year and 11.8% at five years. This was a group of 28 patients who followed his diet to the letter.

Today people are touting the cardiologist, Dr. Kim Williams, president-elect of the American College of Cardiology,  who became a vegan and watched some of his lipid levels improve. What we do not know about his lipid levels were the other foods he ate, what other changes he might be taking – but the news is that veganism is good for the heart. But is it? The Ornish diet restricts not only meat, but refined carbohydrates like added sugars and white flour, which have been implicated in cardiovascular disease.

Since the paper was published in 1998 no one has reproduced that data. No one. In medicine we see a lot of data come through, when it is not reproduced, or unable to be reproduced by others we look at it with a very jaundiced eye. Or to be blunt – we don’t believe it. Yet, his data, with all the issues it has- is still touted by a few in the popular press as “proof” that the “low fat” works. We have levels of evidence in medicine, and while Ornish attempted to get to the highest level of evidence, by having a control group – he fell short with several major statistical issues: (a) his study does not contain enough people to be anywhere nearly significant (b) one cannot rely on angiographic photographs which are interpreted in many different manners (c) one cannot control outside factors, exercise, BMI, smoking cessation.

In contrast, we now have an entire group of lipid medications. A recent study in New England Journal of Medicine showed how that Crestor had produced a regression of plaque in 63% of the individuals. Reproduced. Although most would say they would rather eat vegetables than take a medicine- and I would agree, except that the reduction in plaque from Dr. Ornish studies is not even close to what medications can do.

Modern methods (since Ornish paper in 1998) of measuring arterial plaque are far more sensitive. Intra-vascular ultrasound where they thread a tiny ultrasound probe into the artery and  measure the plaque precisely.   In the Ornish data, he used photos of angiograms ( show two cardiologist the same angiograms and you will get two different interpretations of it – angiograms are not precise). The medications show specific reductions in plaque- not everyones – unlike Ornish. It is rare that anything does 100 per cent to everyone.

When looking at angiograms- like Ornish did- the interpretation of them is so variable, that no scientific publication today would accept that data, or its interpretation. The small amount of plaque reduction is too small to be anything but observer bias.

Two views of the same image- angiogram interpretation is flawed

From one old paper- Dean Ornish has made a career, being the anti-Atkins, and riding the anti-cholesterol, low-fat band wagon with the same religious fervor as Keyes did thirty years ago (see my earlier post about that). His data is quoted by other vegans, as gospel about how heart disease can change – and yet what is missing: no one has done it again.

The difficulty is this: science has caught up with us, and we know a lot more about how plaque forms and doesn’t form. We know that dietary cholesterol is far less important that what the liver makes. We know that the dietary component may be far more related to the triglycerides – and they are raised far more by the grains and pastas that Ornish loves.

Ornish is the lead health-blogger for Huffington Post, has influenced Bill Clinton (see the previous post) and is favorably mentioned by Dr. Oz. He still argues against those who advocate any “low carbohydrate” solution, based on his “empiric” data.

Personality, the willingness to believe in  your hypothesis no matter what science says, and the desire by the public to see “natural” leads to a great career in politics, and entertainment. For most scientists, Ornish’s paper isn’t a breakthrough, but borders on confabulation.

Dr. Terry Simpson About Dr. Terry Simpson
Dr. Terry Simpson received his undergraduate and graduate degrees from the University of Chicago where he spent several years in the Kovler Viral Oncology laboratories doing genetic engineering. He found he liked people more than petri dishes, and went to medical school. Dr. Simpson, a weight loss surgeon is an advocate of culinary medicine. The first surgeon to become certified in Culinary Medicine, he believes teaching people to improve their health through their food and in their kitchen. On the other side of the world, he has been a leading advocate of changing health care to make it more "relationship based," and his efforts awarded his team the Malcolm Baldrige award for healthcare in 2011 for the NUKA system of care in Alaska and in 2013 Dr Simpson won the National Indian Health Board Area Impact Award. A frequent contributor to media outlets discussing health related topics and advances in medicine, he is also a proud dad, husband, author, cook, and surgeon “in that order.” For media inquiries, please visit www.terrysimpson.com.

Share this article on social media!

Latest Comments

  1. Guy Hibbins says:

    I think that it is just not true that nobody has reproduced Dr Ornish’s results. Look at the work of Caldwell Esselstyn at the Cleveland Clinic, for example. You can find his papers with the PubMed search “Esselstyn CB and coronary”. He even has a website http://www.heartattackproof.com where he displays the angiogram of a senior surgeon at the Cleveland Clinic who basically cleaned out his left anterior descending artery in 3 years. This artery was so badly diseased that it was not considered operable.
    Esselstyn wrote a book about his study which caused Bill Clinton to go low fat according to interviews which Clinton gave on the subject.
    In relation to the use of lack of intravascular ultrasound in Ornish’s studies, it should be noted that his evidence of reverasal was not simply based on angiography but on nuclear medicine perfusion studies.
    For what it is worth, the Current FDA Guidance for Industry on Drug Eluting Coronary Stents does not require or recommend the use of IVUS. In fact if you read the Summary of Safety and Effectiveness documents for any of the current generation stents on the FDA’s website you will see that the regulatory submissions are based upon quantitative coronary angiography (QCA), often 2 dimensional QCA.

  2. thedoc says:

    You have not seen the paper I did about Esslestyn – his work was likewise under par. The use of angiography was poor in both studies and not up to any standard of modern cardiology. Most of the studies they look at are simply subject to interpretation which varies by as much as 25% among readers. Plus the website you cite shows clot in arteries, that is not plaque. In terms of perfusion studies- those vary as the heart recovers over time- and are far from anything one would use to determine coronary artery reversal.

  3. Guy Hibbins says:

    Well I would have to respectfully disagree with your interpretations here. The left anterior descending artery which is shown by Dr Esselstyn was not simply blocked by thrombus. It was considered non-stentable due to diffuse coronary disease with lesions longer than the longest available stents. If it had been stentable it would certainly have been stented right away. This was in a leading heart surgery center.

    In relation to the perfusion studies of Dr Ornish, the patients all had stable angina. They were not recovering from heart attacks; something which would have caused their perfusion scans to improve over time. The only reasonable interpretation would seem to be that they developed collateral circulation and or underwent plaque reversal.

    Also, other trials using IVUS have demonstrated reduciton in coronary plaque volume. The REVERSAL trial which was published in JAMA in March 2004 was a randomised trial conducted across 34 centers and showed plaque reversal on IVUS due to cholesterol lowering on atorvastatin.

    It should be said that the most important endpoint in coronary heart disease trials is coronary events. That is what the therapy is intended to prevent. Everything else is ultimately a surrogate measure.

    In Caldwell Esselstyn’s series of 17 patients, there were 49 coronary events in the 10 years prior to the study, and none in the 12 years after to the study. The 49 prior events which occurred over ten years then that would give an event rate of 0.29 events per patient year. (This is conservative as not all of the patients had had symptoms for 10 years). Using binomial probabilities, the likelihood of no event in the next 120 patient years would be
    (1-0.28)^120 = (1-0.29) ^120 = 1.4 x 10 ^-18 = 0.0000000000000000014.

  4. thedoc says:

    Sorry- it is just a thrombus, and LAD can be stented and can by bypassed.

    Re: Ornish. Angina means you have low perfusion of the heart muscle – and perfusion increases with three things: (a) plaque that has ruptured gets remodeled (b) collateral circulation- although that takes years (c) endothelization and remodeling of the plaque.

    Yes- drugs that decrease the numbers of lipo proteins do cause plaque regression.

    Coronary events is open to interpretation. Why do you think that Esselstyn’s study was in a family practice journal and not a coronary journal (where it was rejected). Why do you think he has a highly select group of patients? Why do you think he got rid of other patients? No- this is not something that can be determined by probabilities. It is a high degree of selection in a group of patients who were clearly well selected. It has not been reproduced- and one would think “everyone would do it.”
    Nice you like statistics- but this was bias and selection – and your statistics about probability simply do not work with this

  5. Guy Hibbins says:

    Actually I would respectfully,disagree again on the point about the LAD in the angiogram being treatable. I do most of the new coronary stent regulaotry assessments for the Australian Government and there are definite contraindications to the length of artery which can be stented. This is included of the instructions for use which are approved no matter where the stents are marketed and you can read them on the FDA website in the approved instrucitons. You cannot simply give people stainless steel arteries. The reason that Joe Crowe, the surgeon whose artery is shown in the angiogram was not stented is that the lesion was too long for stenting.

    The only seleciton bias in Esselsytn’s case series was that the patients described complied with the diet. If we had a case series of any treatment group, it would include subanalyses of those who were compliant and those who were not. This is a normal way of reporting results in case series. In this regard the statistics are a perfectly valid way to compare events in the patients who followed the diet with the number of events which otherwise might otherwise have been expected in those patients. The fact that some patients did not follow the diet is noted but it does not invalidate the outcomes in those who did.

    This was, after all, not a randomized trial as, apart from anything else, you cannot successfully randomize people to follow diets any more than you can successfully randomize people to believe in a particular religion, political viewpoint or philosophy. They need to be persuaded.

    Large numbers of randomized dietary trials with very poor compliance attest to this. The much touted A to Z trial is an excellent example.

    Esselstyn got rid of the other patients because they were not interested in following his diet and they were thus not testing the hypothesis in question. He states however that he is in the process of preparing all the patients for publication. This is a much larger case series involving both compliant and non-compliant patients.

    In relation to where Esselstyn’s work is published, most his eight cardiology papers are in the American Journal of Cardiology.

    The idea that “everyone would do it” misses the point that many people believe that the changes required to get their total cholesterol below 150mg/dl (3.9mmol/L), the level below which noone in the Framingham study suffered a coronary event, would be too hard. If the investigators themselves would not eat a low fat diet then they could hardly expect their patients to do so.

    In reality compliance with a low fat diet is highly individual and is dependent upon how convinced the person is of the benefits of such a diet. Personally, I reduced my total cholesterol by 40% almost immediately and I did not find it very difficult although it did require significant dietary changes like avoiding fried foods.

    It should be said that in populaiton studies of people on traditional Asian diets such as the Cornell-China Heart Study the average cholesterol is around 127mg/dl (3.3mmol/L). In that particular study the age standardized rate of coronary disease was one sixteenth the US rate in males.

  6. thedoc says:

    You should always have a control- and he got rid of the control. He had selection bias, and that selection bias makes it an “interesting” paper but not worthy of science or any change. The numbers are small, the techniques are poor, and there are plenty of vegans who have heart disease, get cancer, and live to a younger age than people who eat fish. And, from biostatistics- you have too many other variables that are not accounted for in this – so it does not rise to a high level of evidence based medicine (a reason the paper isn’t cited as a major accomplishment in the fight against heart disease).
    His work was poorly done, and is just a story. You can read our evaluation of him here.
    Dietary Cholesterol is not related to plaque, and for most dietary cholesterol has little or no relationship to blood levels. That is rather old, dated thinking which is why the views that they had was antiquated – and cute.
    That lesion did not need stenting- it was a clot- not a plaque- it resolved probably within an hour after that angiogram was taken – as can be seen by the follow up angiogram.

    Regarding the China Study- see my comments about that- but that study is just silly. Part one I review here. But the data, when examined showed inverse correlation between cholesterol and cardiac disease – second, the study was done in rural China, where even to this day they state they do not have adequate numbers of cardiac disease. Third, the study looked at deaths during the end of the cultural revolution- most people were still dying of starvation related illness.

  7. mhikl says:

    Great repartee, Dr Simpson. “Touché”, an honourable man would have replied.

    Dean Ornish? My he does like to find agreement, doesn’t he? Someday some bright person is going to be able to break the code in his sophistry: “Let’s agree that . . .” the man speaketh by myth and obfuscation, and as suggested, his data do well fudge the meaning of Truth with wonky pictures and reprehensible factoids tainted further with his own curious interpretations. Is it that his clever jibber-jabber got him out of childhood jams that should have served excuse for severe bonkings most blokes would have endured for lies dared told? His reality is so out of wack that in earlier times, and out of kindness, he would be locked up for his own safety and the protection of lesser fools within his sphere of influence. But today his spiel is freely let loose on the public and his supporters buy into his distorted explanations with blind admiration. But, lies and deceptions are eventually figured out and my only hope is that it is done earlier than later, preferably before he “haue shuffel’d off this mortall coile”.

    Dean-O must have a brain or he wouldn’t have survived the rigours of study it takes to get a medical diploma—so the big, really big question is: “Does he actually believe the science that he so unscrupulously wrenches to such high degrees of distortion that a side-show carny would stand in awe and profound admiration; or is he caught up in a mad labyrinth from which he is incapable of extricating himself, his name and the monumental legacy of scorn that he surely knows is to serve his final epitaph. As his lies unravel, Dean Ornish will be the epic example of the ultimate sham-man in all history of medical research. I am reminded of King lear and his fool- nix the insights; Ornish being the crazy of both bound into one. There is a great farce to be written on this ambitious hoaxer.

  8. RosiesDad says:

    And yet, you don’t read-see-hear much from individuals with heart disease who have followed the regimen of Dr. Ornish (or Dr. Esselstyn, for that matter) who complain that they didn’t feel better nearly immediately after going on an ultra low fat diet.

  9. Dr. Terry Simpson says:

    You don’t hear much from anyone following those diets because there are so few of them. One universal truth about people who follow some diet plan: they tell you how much better they feel – except the high protein diet folks, who feel horrible but don’t admit it

  10. Willis James says:

    The author, Dr. Simpson, makes most of his money doing lap-band surgery.

    Hmmm… I think I’ll go with Ornish as a more natural way to health. Both for heart health and weight loss.

  11. Ben Brown says:

    I am a physician who has worked with many of the original research participants, most of whom lived another 20-30 years after their physicians predicted they would die. For your flow data that you would like, look at the coronary blood flow or PET scan paper, published in JAMA, Sept 20, 1995-Vol 274, No 11. Very small changes in diameter (as you mentioned above) caused amazing changes in blood flow as flow is a 4th power function of radius. 300-400% increases in blood flow.

  12. Dr. Terry Simpson says:

    You need to look at Pouseuille’s law would disagree at the point they are talking. And since 2.5% is in the area of inability to determine if it is true or not, and the PET scan blood flow is not that good- this is not a good argument.

  13. Ashwani Garg MD says:

    Dr. Dean Ornish’s programs are definitely revolutionary, and he emphasized not just a plant based diet, but yoga / meditation, exercise, and cultivation of loving relationships. These factors were seen as essential to his program. This approach is echoed by Dan Buettner of Blue Zones, who saw these factors as key to the vitality and longevity of the Blue Zones. Dr. Esselstyn has a summary of the peer-reviewed nutritional interventions to date: http://dresselstyn.com/JFP_06307_Article1.pdf

  14. Terry Simpson says:

    His data does not rise to the level of evidence based medicine – and is anecdotal at best. When it comes to longevity, it isn’t measured in telomeres – it is measured in lifespan. The EPIC study showed meat eaters have no less lifespan, no more risk of cancer or heart disease than vegetarians – and the issue is processed meats, not regular meat. In terms of “vitality” – well, just no data for that. I prefer to go where science takes me- not to cherry pick data. Most of what Dr. Ornish says is incorrect – when obesity increased the number of calories from protein and fat actually decreased.In a 2007 clinical trial led by Gardner researchers randomly assigned 311 individuals to four groups: One group was assigned the high-fat, high-protein and low-carbohydrate Atkins diet; the second was assigned Ornish’s very low-fat vegetarian diet, which requires consuming fewer than 10 percent of calories from fat; the third was assigned the Zone diet, which aims for a 40/30/30 percent distribution of carbohydrate, protein and fat; and the fourth was assigned the high-carbohydrate, low–saturated fat LEARN (for: lifestyle, exercise, attitudes, relationships, nutrition) diet. The participants all had trouble adhering to their regimens, but all lost about the same statistically significant amounts of weight, and when compared head to head, the Atkins dieters saw greater improvements in blood pressure and HDL cholesterol than the Ornish dieters did.
    Does Ornish claim hold up? Nope – in his study of 48 people who were randomized not only did his group change diet- but they quit smoking and reduced stress. The control group did not. Given statistics, the overwhelming Relative Risk of smoking and the small numbers mean that his work does not meet significance. In addition, he claims reduction of 2.5% of the occlusion of vessels, which is well BELOW what people can actually read or agree with in an angiogram, especially one done back then – not even with todays modern digital angiography can we find such significance.
    Most people cannot stay on that diet, and there is no good evidence for them to.
    The number of people who claim this are small, but vocal.
    The real issue isn’t that it is plant vs. non-plant – although some make it. We need more diversity in diet. Ornish didn’t show a breakthrough – but be honest – stopping smoking was the single, most overwhelming thing his patients did. In terms of yoga – absolutely no evidence based medicine that it reduces the risk of any disease.

  15. Ashwani Garg MD says:

    Dr. Simpson, thank you for your comments and very respectful tone and nice discussion. My personal experience and experience with treatment of patients in my family practice is that the more plant foods, the better. One does not have to be 100% to be healthy, but I would probably propose at least 80-90% plant, and 10-20% animal origin as far as calories. While I personally choose to make it 100%, I know that this is not easy for many patients, and I am very willing to work with everyone to improve their diet and lifestyle. I like to borrow aspects from Dr. Ornish because when they exercise, do yoga, and have healthy relationships they are more likely to be healthy. I don’t need hard RCT data to recommend lifestyle changes, because you can’t really do a randomized double blind study on this. The predominance of evidence is good for me. One resource I like to point patients to is “Full Plate Diet” on http://www.fullplateliving.org – this plan advocates for a much higher amount of fiber in the diet than any other, and goes away from portion control, dieting, etc. in favor of a larger volume diet composed of mostly plant foods. As far as the number of people advocating for a plant-based diet, it is growing rapidly, with many conferences around the country. One is at the end of May in Rosemont about plant-based approaches to CVD, I am looking forward to it. Having done a plant-based diet for the last 2 years, I am still learning and 1 issue I have learned is not to go crazy about reducing healthy “fats” like nuts, seeds, avocado and olives, but I still eliminate processed oils. Thanks again for the discussion.

  16. Dr. Terry Simpson says:

    My pleasure. There is nothing wrong with meats – and they are a great biosource for ingredients. Plant based diets are getting more noise on the internet, but much of that is rehashing the same things that were on the internet. You cannot do randomized controls, but good data does not show an issue with meat. Most people who advocate for plant based diets are vegans, and veganism is simply a political movement. But that is another discussion – if you look for information about plant based diets you will find more on the internet, as everyone who is on them seems to write a blog about their lovely experience. You can do randomized controls for diets, you just cannot blind them. When those are done, there is no better results with plant based diets. Nutrition is more complex than just “eat your vegetables,” and so are diseases. I object to the nonsense that if you eat a plant based diet you have less cancer or heart disease – or can reverse it, as some of their advocates claim – without data. If I were to recommend a lifestyle I like the DASH diet the best.

Leave a Reply