jokomp3.blogspot.com - When we consider the health impacts of eating meat, cardiovascular disease is the first thing that comes to mind. Popular diet advocates often hold diametrically opposed views on the role of meat in cardiovascular disease. Even among researchers and public health officials, opinions vary. In this post, I'll do my best to sort through the literature and determine what the weight of the evidence suggests.
Ancel Keys and the Seven Countries Study
Ancel Keys was one of the first researchers to contribute substantially to the study of the link between diet and cardiovascular disease. Sadly, there is a lot of low-quality information circulating about Ancel Keys and his research (1). The truth is that Keys was a pioneering researcher who conducted some of the most impressive nutritional science of his time. The military "K ration" was designed by Keys, much of what we know about the physiology of starvation comes from his detailed studies during World War II, and he was the original Mediterranean Diet researcher. Science marches on, and not all discoveries are buttressed by additional research, but Keys' work was among the best of his day and must be taken seriously.
One of Keys' earliest contributions to the study of diet and cardiovascular disease appeared in an obscure 1953 paper titled "Atherosclerosis: A Problem in Newer Public Health" (2). This paper is worth reading if you get a chance (freely available online if you poke around a bit). He presents a number of different arguments and supporting data, most of which are widely accepted today, but one graph in particular has remained controversial. This graph shows the association between total fat intake and heart disease mortality in six countries. Keys collected the data from publicly available databases on global health and diet:
The graph shows that there was a very strong positive correlation between total fat intake and reported heart disease mortality among the six countries Keys selected for his analysis. Critics Yerushalmy and Hilleboe later pointed out that data for 22 countries were available at the time of Keys' analysis. When all 22 countries were considered together, the relationship between total fat intake and heart disease mortality weakened but remained statistically significant (3).
These data, among others, stimulated Keys to design the most ambitious diet-heart study ever conducted at the time: the Seven Countries Study. This study investigated the link between diet, lifestyle, cardiovascular risk factors, and disease risk among 16 populations in 7 countries.
Keys' diet assessment methods were extremely rigorous. To determine the food and nutrient intake of each population, for one week his team invited selected families to prepare duplicate meals: one for themselves, and one for the study. The extra meals were then weighed and chemically analyzed to determine their nutritional characteristics. This method is far more rigorous that the questionnaires used in most observational studies today.
However, the study design suffered from a critical weakness: Keys' team used data from 20-50 people to extrapolate the average dietary intake of the population as a whole, and these average population data are what they used for their analyses. In other words, rather than comparing the dietary intake and cardiovascular risk of individuals, the Seven Countries Study compared the dietary intake and cardiovascular risk of populations. This is called an ecological study and it's considered to be a weaker study design than the numerous individual-level observational studies that followed it.
What did Keys' team find? Here are some of their main findings:
This is a very strong correlation showing that populations that eat more saturated fat have more heart attacks, and it was confirmed in a 25-year follow-up study of the same populations (5).
The lowest cardiovascular risk occurred on the rural Greek island of Crete, and a rural Japanese farming community at Tanushimaru. At the beginning of the study, both cultures ate starchy omnivorous diets relatively low in animal foods and high in grains and potatoes (6). The diet at Tanushimaru was very low in fat, whereas the Cretan diet was moderate in fat, mostly from extra-virgin olive oil but also from dairy. The diet of Crete was low in meat and fish but averaged about one cup of milk per day and 3 medium eggs per week. The diet at Tanushimaru averaged 1/5 lb of fish per day and 2-3 medium eggs per week. Both cultures consumed polyphenol-rich plant foods such as extra-virgin olive oil and green tea, and exhibited the various lifestyle factors typical of non-industrial cultures (e.g., regular physical activity, sun exposure, absence of processed foods, slower pace of life).
Blue Zones
The diets of Crete and Tanushimaru are consistent with the diets of populations that live in so-called "Blue Zones"-- areas of notably low cardiovascular disease risk and long natural lifespan (7). Blue Zone populations typically eat meat, and often dairy and eggs, in small to moderate quantities (significantly less meat than is typical of affluent Western cultures). There is one Blue Zone in Loma Linda, California, that contains a high proportion of vegetarians, but the other populations all eat meat. None of the Blue Zones are vegan. The diets in these areas are centered around carbohydrate, and more often than not, grains and legumes of some sort. This doesn't prove that their food choices are optimal, but it does prove that such diets are, at a minimum, compatible with health and long life in the context of a more traditional lifestyle.
Non-industrialized Agriculturalists and Hunter-gatherers
The populations with the lowest documented cardiovascular risk are agricultural (and horticultural) cultures living a traditional lifestyle that resembles how our ancestors might have lived 5,000 years ago. In an impressive heart autopsy study on thousands of subjects, Lee and colleagues determined that rural West Africans (Nigerians and Ugandans) in the 1940s, 50s, and 60s were essentially immune to heart attacks, even in old age (8). Their coronary arteries also exhibited less atherosclerosis than Americans, including African-Americans. In the same study, urban Asians living in Japan and Korea had a lower rate of heart attacks than Americans, but higher than Africans. A number of other studies have reported similar findings in various traditionally-living agricultural/horticultural societies (9, 10, 11, Trowell and Burkitt. Western Diseases. 1981).
These cultures tend to eat a starch-based omnivorous diet low in animal foods. It's possible that the Nigerian and Ugandan samples may have included some pastoralists with a high intake of animal foods, but the vast majority of people would have followed a starchy low-animal-food diet. The dietary pattern in these agricultural/horticultural cultures may not be the only factor in their resistance to heart attacks, but their diets are at least compatible with exceptional cardiovascular health.
We have much less information about traditional cultures that eat diets higher in meat, for example, most hunter-gatherers. The only autopsy studies we have of hunter-gatherers were performed in a few Inuit (Eskimo) individuals, and they suggest that these individuals tended to suffer from advanced atherosclerosis but that no signs of heart attack are present (12, 13). Atherosclerosis did not appear to translate into a high heart attack risk in semi-traditional Inuit populations that have been studied, suggesting that they may have somehow been protected from the consequences of their vascular disease (14). In any case, the Inuit are probably not representative of hunter-gatherers in general because they ate extreme diets, lived in an extreme environment, and inhaled a lot of indoor smoke.
Although we don't have autopsy studies in more typical hunter-gatherer cultures, we do have some information about their cardiovascular health. Hunter-gatherers invariably show evidence of good cardiovascular health, including low body fatness, low cholesterol, low blood pressure that doesn't rise with age, and high physical fitness (15, Trowell and Burkitt. Western Diseases. 1981). It would be difficult to imagine a high cardiovascular risk in these populations that rely heavily on meat (16), but again we have little direct evidence of this.
The China Study
The China Study was a massive ecological study relating diet and lifestyle to chronic disease risk in China. It has been invoked by researcher and vegan diet advocate Colin Campbell to support the idea that animal foods promote cardiovascular disease and cancer, even in the small quantities that were typical of the regions studied. After having reviewed the study data, the publications based on it, and the various commentaries on it, it appears relatively clear that the China Study does not support the conclusion that meat consumption is associated with cardiovascular disease or cancer risk (17, 18, 19, 20, 21). Everyone seems to agree on that, except Campbell and certain other vegan diet advocates. I won't discuss the China Study further.
Modern Observational Studies
What do modern observational studies have to say about the relationship between meat intake and cardiovascular risk? Overall, they paint a substantially different picture than the Seven Countries Study. Here is a summary of the weight of the evidence, as I understand it:
The Harvard Healthy Eating Pyramid is based primarily on a synthesis of this observational evidence (37). It recommends eating fish, poultry, eggs, and 1-2 servings per day of dairy, but advises limiting red meat, processed meat, and butter. Whole grains, plant oils, and vegetables/fruits form the base of the pyramid, and nuts, seeds, beans and tofu are recommended, particularly as an alternative to red meat. It's notable that the Healthy Eating Pyramid includes meat as a staple food.
Do Vegetarians and Vegans Have Fewer Heart Attacks than Omnivores?
Overall, yes. Although not all studies have supported this conclusion, the weight of the evidence suggests that vegetarians and vegans have a lower heart attack risk than omnivores (38). Risk among lacto-ovo vegetarians, vegans, and seafood eaters appears similar however (39). If we assume for a moment that these associations reflect cause and effect, there appears to be little to gain from completely avoiding animal foods or even avoiding all types of meat.
Many Seventh-Day Adventists avoid eating meat for religious reasons, although they typically eat eggs and dairy. There's a large population of SDAs in California that researchers have studied extensively. These studies demonstrate that SDAs have a substantially lower heart attack risk than the general population, and SDAs that are vegetarian have a lower risk than SDAs in the same community that eat meat (40). Vegetarian SDAs are also noted for their longevity.
Studies of vegetarians and vegans aren't easy to interpret, however. Vegetarians and vegans differ from the general population in many ways besides meat avoidance, and this is particularly true of SDAs. These populations tend to be more health-conscious and have overall healthier lifestyle behaviors. Can we determine how much of their cardiovascular advantage is due to meat avoidance, and how much is due to factors unrelated to meat?
The Health Food Shoppers Study attempted to tease this out. This UK study recruited vegetarians and omnivores from health food stores and vegetarian societies and magazines-- ensuring that both groups were composed of health-conscious people. After 17 years of follow-up, the results showed no statistically significant difference in total mortality, heart attacks, or stroke between vegetarians and non-vegetarians, although there was a trend toward lower heart attack risk for vegetarians (41).
Overall, the evidence suggests that factors other than meat avoidance account for at least some of the cardiovascular benefit associated with vegetarian and vegan diets. However, there may still be some cardiovascular benefit from avoiding meat, and this would be consistent with the LDL-lowering effect of meat avoidance (discussed below). The observational evidence does not appear to support the idea that avoiding seafood, dairy, or eggs reduces cardiovascular risk.
Vegan Diet Interventions for Cardiovascular Disease
Dean Ornish, Caldwell Esselstyn, John McDougall, and others have used vegan or near-vegan diets as part of a diet, lifestyle, and/or drug strategy to reduce cardiovascular disease risk in high-risk patients. Some of them have published studies (42, 43, 44). These studies vary in quality, but overall they do suggest that these diet/lifestyle/drug strategies may indeed be effective for reducing cardiovascular risk in high-risk patients (typically, people who have had a heart attack or have been diagnosed with severe cardiovascular disease). They are probably more effective than conventional medical therapy.
However, it's difficult to know what aspect of the intervention is responsible for the cardiovascular benefits. Ornish's intervention, for example, involves a comprehensive diet overhaul including reducing fat intake, avoiding processed food, focusing on minimally refined foods, and increasing vegetable intake. The intervention also includes smoking cessation, regular exercise, and stress reduction. Esselstyn's intervention is not quite vegan but it excludes nearly all dietary fat and includes lipid-lowering statins for patients who aren't able to meet stringent blood lipid goals through diet alone.
What role does meat avoidance play in these results? Are patients benefiting from reducing their intake of processed and high-heat cooked meat? Are patients benefiting from reducing their intake of red meat? Are patients benefiting from reducing their intake of all types of meat? There are a lot of remaining questions here, and these studies are not able to answer them.
It is plausible that animal food avoidance could play some role in the therapeutic effect of these diet and lifestyle strategies, but the studies don't allow us to come to that conclusion. There are too many confounding factors. It would be informative to study the effectiveness of these interventions with or without added meat.
Animal Studies and Possible Mechanisms
Beginning with the pioneering studies of Nikolay Anichkov in the early 1900s, researchers have long known that high levels of dietary cholesterol can rapidly promote cardiovascular disease in certain animals. Dietary cholesterol in the human diet comes exclusively from animal foods, with egg yolks being the most concentrated source, so dietary cholesterol is an obvious suspect in the possible link between animal foods and cardiovascular disease.
These artery-clogging high-cholesterol experimental diets generally cause massive 3-10-fold increases in circulating cholesterol in animal models, which is difficult to compare to the human situation. Humans only see small increases in circulating cholesterol when we eat dietary cholesterol, and the increase typically occurs in both "bad" LDL and "good" HDL* (45). Most humans seem to be able to handle normal amounts of dietary cholesterol efficiently, which substantially weakens the likelihood of a meaningful link between dietary cholesterol and cardiovascular disease. However, in a minority of people, dietary cholesterol has a larger effect on circulating cholesterol, and could play a larger role in cardiovascular risk.
Saturated fat is another possible link between animal food consumption and cardiovascular risk. In typical Western diets, most saturated fat comes from animal foods. Saturated fat feeding can exacerbate the impact of high-cholesterol diets on circulating cholesterol and cardiovascular damage in animal models, at least when compared with polyunsaturated fat (the latter of which has a cholesterol-lowering effect). However, when animals are not overfed cholesterol, saturated fat feeding per se doesn't raise circulating cholesterol when compared with monounsaturated fat (as in olive oil), so I'm not sure to what extent saturated fat itself is responsible (46, 47).
In humans, many trials have shown that short-term saturated fat feeding increases circulating cholesterol. This increase occurs in both the LDL and the HDL fraction, although the increase in LDL is often somewhat greater than the increase in HDL (48). However, the fact that observational studies typically find little or no correlation between habitual long-term saturated fat intake and circulating cholesterol (or cardiovascular risk) makes me wonder how durable these effects are (49). Longer-term randomized controlled trials also often show little or no impact of 2-3-fold differences in saturated fat intake on circulating cholesterol, as long as other relevant dietary components including linoleic acid (n6 PUFA) and fiber are kept relatively constant (50). Research is ongoing, but I'm not currently convinced that this mechanism plays a major role in cardiovascular disease.
High levels of animal protein (from meat or dairy) can increase LDL cholesterol and aggravate cardiovascular disease in certain animal models, while plant protein is often protective (51). This appears to be due to the amino acid composition of animal vs. plant protein (lysine-to-arginine ratio), and the phytochemicals and fiber associated with plant protein. Certain plant proteins, such as soy, also tend to lower LDL cholesterol in humans, particularly when they replace animal protein (52, 53). This probably goes a long way toward explaining the LDL-lowering effect of vegetarian and particularly vegan diets, and it is consistent with a protective cardiovascular effect.
Another potential mechanism for harmful effects of meat is the compounds that are produced during high-heat cooking. Some of these may be carcinogenic and contribute to cardiovascular disease (54). We might be better off focusing on low-heat-cooked meat rather than grilled, browned, roasted, or fried meats-- but this also applies to plant foods.
Red meat is very high in iron, which some researchers have proposed could compromise cardiovascular health. Free (unbound) iron is particularly good at catalyzing powerful free radical reactions that damage surrounding molecules. While iron is an essential nutrient and high-iron foods can be particularly beneficial for pre-menopausal women, many men and post-menopausal women may get too much dietary iron. Personally I think excess dietary iron is probably detrimental to overall health, but the evidence is not currently very convincing that body iron status plays a major role in cardiovascular health in particular (55, 56). The case isn't closed yet.
Sialic acids are unique sugars that are used as building blocks for certain classes of molecules in the body. Over the course of evolution, humans have lost the ability to produce a particular sialic acid called Neu5gc. However, we absorb it from animal foods, particularly red meat, and it is incorporated into our (glyco)proteins and other molecules. The human body recognizes it as foreign and produces antibodies against it. Some researchers have proposed that the immune reaction that the body mounts toward Neu5gc from meat causes inflammation and contributes to disease risk (57). Although the paper introducing this mechanism made a big splash when it was first published, to my knowledge the finding hasn't been followed up sufficiently to make a compelling case that Neu5gc actually contributes to human disease. Perhaps future research will clarify that.
There are many scary-sounding mechanisms by which meat intake could potentially contribute to disease. However, I'd like to point out that you can find something wrong with any food if you look hard enough. Melissa McEwen illustrated this well in her satire piece "Just Kale Me". Mechanisms are relevant, but they only serve to explain or bolster demonstrated effects. For example, if poultry consumption doesn't actually increase cardiovascular risk, there's no point digging around for mechanisms. When they have been identified, mechanisms can increase our confidence in an empirically demonstrated effect on health.
Synthesis and Conclusion
There is a lot of evidence we can bring to bear on this question, and not all of it is consistent. This inconsistency is why we see different groups interpreting the research in opposite ways.
Here is my best attempt to synthesize the overall evidence, as I see it. Vegetarian and vegan diets probably do reduce cardiovascular disease risk, consistent with the LDL-lowering effect of replacing meat with plant protein. However, the risk reduction afforded by avoiding meat per se appears modest (as opposed to living the whole lifestyle of a vegetarian or vegan), and we have no examples of healthy vegan cultures. There is little evidence that poultry or seafood promote cardiovascular disease, and some evidence suggesting that seafood is actually protective. Although the evidence on red meat isn't very consistent, and we need more and better research, it's probably prudent to keep intake modest for now. It's probably a good idea to limit processed meat intake as well (attn LCers and Paleos: bacon).
For people at high cardiovascular risk, it may be useful to replace some animal protein with plant protein such as beans and nuts, and focus meat intake on seafood and poultry. Randomized controlled trials such as the Oslo Diet-Heart study have supported the ability of similar dietary patterns to reduce cardiovascular risk (58).
The ideal strategy would also incorporate other factors that are relevant to cardiovascular risk, such as controlling body fatness, eating potassium- and polyphenol-rich fruit and vegetables, focusing on unrefined food, exercising regularly, avoiding prolonged sitting, managing stress, avoiding cigarette smoking, and working with a doctor to monitor and possibly control blood pressure and lipids.
So, does eating meat increase cardiovascular risk? Yes! And no!
In the next post, we'll examine the impact of meat consumption on obesity risk.
* Nuance: we know that people with high HDL cholesterol have fewer heart attacks, but we still don't know whether a diet-induced increase in HDL cholesterol is actually protective. Recent drug trials have shown that using drugs to simply pack more cholesterol into HDL particles does not reduce cardiovascular risk in humans, and it may even increase risk. This demonstrates that the protective function of the HDL particle is not always associated with its cholesterol content. We still have a lot to learn about why high HDL cholesterol is associated with cardiovascular protection in humans, and what diet/lifestyle factors influence its protective actions.
Ancel Keys and the Seven Countries Study
Ancel Keys was one of the first researchers to contribute substantially to the study of the link between diet and cardiovascular disease. Sadly, there is a lot of low-quality information circulating about Ancel Keys and his research (1). The truth is that Keys was a pioneering researcher who conducted some of the most impressive nutritional science of his time. The military "K ration" was designed by Keys, much of what we know about the physiology of starvation comes from his detailed studies during World War II, and he was the original Mediterranean Diet researcher. Science marches on, and not all discoveries are buttressed by additional research, but Keys' work was among the best of his day and must be taken seriously.
One of Keys' earliest contributions to the study of diet and cardiovascular disease appeared in an obscure 1953 paper titled "Atherosclerosis: A Problem in Newer Public Health" (2). This paper is worth reading if you get a chance (freely available online if you poke around a bit). He presents a number of different arguments and supporting data, most of which are widely accepted today, but one graph in particular has remained controversial. This graph shows the association between total fat intake and heart disease mortality in six countries. Keys collected the data from publicly available databases on global health and diet:
The graph shows that there was a very strong positive correlation between total fat intake and reported heart disease mortality among the six countries Keys selected for his analysis. Critics Yerushalmy and Hilleboe later pointed out that data for 22 countries were available at the time of Keys' analysis. When all 22 countries were considered together, the relationship between total fat intake and heart disease mortality weakened but remained statistically significant (3).
These data, among others, stimulated Keys to design the most ambitious diet-heart study ever conducted at the time: the Seven Countries Study. This study investigated the link between diet, lifestyle, cardiovascular risk factors, and disease risk among 16 populations in 7 countries.
Keys' diet assessment methods were extremely rigorous. To determine the food and nutrient intake of each population, for one week his team invited selected families to prepare duplicate meals: one for themselves, and one for the study. The extra meals were then weighed and chemically analyzed to determine their nutritional characteristics. This method is far more rigorous that the questionnaires used in most observational studies today.
However, the study design suffered from a critical weakness: Keys' team used data from 20-50 people to extrapolate the average dietary intake of the population as a whole, and these average population data are what they used for their analyses. In other words, rather than comparing the dietary intake and cardiovascular risk of individuals, the Seven Countries Study compared the dietary intake and cardiovascular risk of populations. This is called an ecological study and it's considered to be a weaker study design than the numerous individual-level observational studies that followed it.
What did Keys' team find? Here are some of their main findings:
- Circulating cholesterol was strongly correlated with cardiovascular risk.
- Dietary saturated fat intake was strongly correlated with circulating cholesterol.
- Dietary saturated fat intake was strongly correlated with cardiovascular risk.
- Animal foods were the primary source of dietary saturated fat.
- Meat intake, except fish, was correlated with cardiovascular risk, although that was mostly explained by its saturated fat content.
This is a very strong correlation showing that populations that eat more saturated fat have more heart attacks, and it was confirmed in a 25-year follow-up study of the same populations (5).
The lowest cardiovascular risk occurred on the rural Greek island of Crete, and a rural Japanese farming community at Tanushimaru. At the beginning of the study, both cultures ate starchy omnivorous diets relatively low in animal foods and high in grains and potatoes (6). The diet at Tanushimaru was very low in fat, whereas the Cretan diet was moderate in fat, mostly from extra-virgin olive oil but also from dairy. The diet of Crete was low in meat and fish but averaged about one cup of milk per day and 3 medium eggs per week. The diet at Tanushimaru averaged 1/5 lb of fish per day and 2-3 medium eggs per week. Both cultures consumed polyphenol-rich plant foods such as extra-virgin olive oil and green tea, and exhibited the various lifestyle factors typical of non-industrial cultures (e.g., regular physical activity, sun exposure, absence of processed foods, slower pace of life).
Blue Zones
The diets of Crete and Tanushimaru are consistent with the diets of populations that live in so-called "Blue Zones"-- areas of notably low cardiovascular disease risk and long natural lifespan (7). Blue Zone populations typically eat meat, and often dairy and eggs, in small to moderate quantities (significantly less meat than is typical of affluent Western cultures). There is one Blue Zone in Loma Linda, California, that contains a high proportion of vegetarians, but the other populations all eat meat. None of the Blue Zones are vegan. The diets in these areas are centered around carbohydrate, and more often than not, grains and legumes of some sort. This doesn't prove that their food choices are optimal, but it does prove that such diets are, at a minimum, compatible with health and long life in the context of a more traditional lifestyle.
Non-industrialized Agriculturalists and Hunter-gatherers
The populations with the lowest documented cardiovascular risk are agricultural (and horticultural) cultures living a traditional lifestyle that resembles how our ancestors might have lived 5,000 years ago. In an impressive heart autopsy study on thousands of subjects, Lee and colleagues determined that rural West Africans (Nigerians and Ugandans) in the 1940s, 50s, and 60s were essentially immune to heart attacks, even in old age (8). Their coronary arteries also exhibited less atherosclerosis than Americans, including African-Americans. In the same study, urban Asians living in Japan and Korea had a lower rate of heart attacks than Americans, but higher than Africans. A number of other studies have reported similar findings in various traditionally-living agricultural/horticultural societies (9, 10, 11, Trowell and Burkitt. Western Diseases. 1981).
These cultures tend to eat a starch-based omnivorous diet low in animal foods. It's possible that the Nigerian and Ugandan samples may have included some pastoralists with a high intake of animal foods, but the vast majority of people would have followed a starchy low-animal-food diet. The dietary pattern in these agricultural/horticultural cultures may not be the only factor in their resistance to heart attacks, but their diets are at least compatible with exceptional cardiovascular health.
We have much less information about traditional cultures that eat diets higher in meat, for example, most hunter-gatherers. The only autopsy studies we have of hunter-gatherers were performed in a few Inuit (Eskimo) individuals, and they suggest that these individuals tended to suffer from advanced atherosclerosis but that no signs of heart attack are present (12, 13). Atherosclerosis did not appear to translate into a high heart attack risk in semi-traditional Inuit populations that have been studied, suggesting that they may have somehow been protected from the consequences of their vascular disease (14). In any case, the Inuit are probably not representative of hunter-gatherers in general because they ate extreme diets, lived in an extreme environment, and inhaled a lot of indoor smoke.
Although we don't have autopsy studies in more typical hunter-gatherer cultures, we do have some information about their cardiovascular health. Hunter-gatherers invariably show evidence of good cardiovascular health, including low body fatness, low cholesterol, low blood pressure that doesn't rise with age, and high physical fitness (15, Trowell and Burkitt. Western Diseases. 1981). It would be difficult to imagine a high cardiovascular risk in these populations that rely heavily on meat (16), but again we have little direct evidence of this.
The China Study
The China Study was a massive ecological study relating diet and lifestyle to chronic disease risk in China. It has been invoked by researcher and vegan diet advocate Colin Campbell to support the idea that animal foods promote cardiovascular disease and cancer, even in the small quantities that were typical of the regions studied. After having reviewed the study data, the publications based on it, and the various commentaries on it, it appears relatively clear that the China Study does not support the conclusion that meat consumption is associated with cardiovascular disease or cancer risk (17, 18, 19, 20, 21). Everyone seems to agree on that, except Campbell and certain other vegan diet advocates. I won't discuss the China Study further.
Modern Observational Studies
What do modern observational studies have to say about the relationship between meat intake and cardiovascular risk? Overall, they paint a substantially different picture than the Seven Countries Study. Here is a summary of the weight of the evidence, as I understand it:
- Total meat, saturated fat, and dietary cholesterol intake typically show little or no relationship with circulating cholesterol, over 2-3 fold differences in intake (22, 23, 24, 25, 26, 27, 28). It remains unclear whether this reflects a lack of a long-term causal relationship, or limitations of the study methods.
- Total saturated fat intake is not associated with cardiovascular risk (29).
- Intake of eggs and dairy (whether full-fat or reduced-fat) are not associated with cardiovascular risk (30, 31).
- Intake of seafood is typically associated with reduced cardiovascular risk (32).
- Intake of poultry is associated with neutral or reduced cardiovascular risk (33).
- Intake of fresh red meat is inconsistently associated with higher cardiovascular risk (34, 35), although meta-analyses suggest the effect size is small.
- Intake of processed meat (e.g., hot dogs, bacon, salami) is typically associated with higher cardiovascular risk (36).
The Harvard Healthy Eating Pyramid is based primarily on a synthesis of this observational evidence (37). It recommends eating fish, poultry, eggs, and 1-2 servings per day of dairy, but advises limiting red meat, processed meat, and butter. Whole grains, plant oils, and vegetables/fruits form the base of the pyramid, and nuts, seeds, beans and tofu are recommended, particularly as an alternative to red meat. It's notable that the Healthy Eating Pyramid includes meat as a staple food.
Do Vegetarians and Vegans Have Fewer Heart Attacks than Omnivores?
Overall, yes. Although not all studies have supported this conclusion, the weight of the evidence suggests that vegetarians and vegans have a lower heart attack risk than omnivores (38). Risk among lacto-ovo vegetarians, vegans, and seafood eaters appears similar however (39). If we assume for a moment that these associations reflect cause and effect, there appears to be little to gain from completely avoiding animal foods or even avoiding all types of meat.
Many Seventh-Day Adventists avoid eating meat for religious reasons, although they typically eat eggs and dairy. There's a large population of SDAs in California that researchers have studied extensively. These studies demonstrate that SDAs have a substantially lower heart attack risk than the general population, and SDAs that are vegetarian have a lower risk than SDAs in the same community that eat meat (40). Vegetarian SDAs are also noted for their longevity.
Studies of vegetarians and vegans aren't easy to interpret, however. Vegetarians and vegans differ from the general population in many ways besides meat avoidance, and this is particularly true of SDAs. These populations tend to be more health-conscious and have overall healthier lifestyle behaviors. Can we determine how much of their cardiovascular advantage is due to meat avoidance, and how much is due to factors unrelated to meat?
The Health Food Shoppers Study attempted to tease this out. This UK study recruited vegetarians and omnivores from health food stores and vegetarian societies and magazines-- ensuring that both groups were composed of health-conscious people. After 17 years of follow-up, the results showed no statistically significant difference in total mortality, heart attacks, or stroke between vegetarians and non-vegetarians, although there was a trend toward lower heart attack risk for vegetarians (41).
Overall, the evidence suggests that factors other than meat avoidance account for at least some of the cardiovascular benefit associated with vegetarian and vegan diets. However, there may still be some cardiovascular benefit from avoiding meat, and this would be consistent with the LDL-lowering effect of meat avoidance (discussed below). The observational evidence does not appear to support the idea that avoiding seafood, dairy, or eggs reduces cardiovascular risk.
Vegan Diet Interventions for Cardiovascular Disease
Dean Ornish, Caldwell Esselstyn, John McDougall, and others have used vegan or near-vegan diets as part of a diet, lifestyle, and/or drug strategy to reduce cardiovascular disease risk in high-risk patients. Some of them have published studies (42, 43, 44). These studies vary in quality, but overall they do suggest that these diet/lifestyle/drug strategies may indeed be effective for reducing cardiovascular risk in high-risk patients (typically, people who have had a heart attack or have been diagnosed with severe cardiovascular disease). They are probably more effective than conventional medical therapy.
However, it's difficult to know what aspect of the intervention is responsible for the cardiovascular benefits. Ornish's intervention, for example, involves a comprehensive diet overhaul including reducing fat intake, avoiding processed food, focusing on minimally refined foods, and increasing vegetable intake. The intervention also includes smoking cessation, regular exercise, and stress reduction. Esselstyn's intervention is not quite vegan but it excludes nearly all dietary fat and includes lipid-lowering statins for patients who aren't able to meet stringent blood lipid goals through diet alone.
What role does meat avoidance play in these results? Are patients benefiting from reducing their intake of processed and high-heat cooked meat? Are patients benefiting from reducing their intake of red meat? Are patients benefiting from reducing their intake of all types of meat? There are a lot of remaining questions here, and these studies are not able to answer them.
It is plausible that animal food avoidance could play some role in the therapeutic effect of these diet and lifestyle strategies, but the studies don't allow us to come to that conclusion. There are too many confounding factors. It would be informative to study the effectiveness of these interventions with or without added meat.
Animal Studies and Possible Mechanisms
Beginning with the pioneering studies of Nikolay Anichkov in the early 1900s, researchers have long known that high levels of dietary cholesterol can rapidly promote cardiovascular disease in certain animals. Dietary cholesterol in the human diet comes exclusively from animal foods, with egg yolks being the most concentrated source, so dietary cholesterol is an obvious suspect in the possible link between animal foods and cardiovascular disease.
These artery-clogging high-cholesterol experimental diets generally cause massive 3-10-fold increases in circulating cholesterol in animal models, which is difficult to compare to the human situation. Humans only see small increases in circulating cholesterol when we eat dietary cholesterol, and the increase typically occurs in both "bad" LDL and "good" HDL* (45). Most humans seem to be able to handle normal amounts of dietary cholesterol efficiently, which substantially weakens the likelihood of a meaningful link between dietary cholesterol and cardiovascular disease. However, in a minority of people, dietary cholesterol has a larger effect on circulating cholesterol, and could play a larger role in cardiovascular risk.
Saturated fat is another possible link between animal food consumption and cardiovascular risk. In typical Western diets, most saturated fat comes from animal foods. Saturated fat feeding can exacerbate the impact of high-cholesterol diets on circulating cholesterol and cardiovascular damage in animal models, at least when compared with polyunsaturated fat (the latter of which has a cholesterol-lowering effect). However, when animals are not overfed cholesterol, saturated fat feeding per se doesn't raise circulating cholesterol when compared with monounsaturated fat (as in olive oil), so I'm not sure to what extent saturated fat itself is responsible (46, 47).
In humans, many trials have shown that short-term saturated fat feeding increases circulating cholesterol. This increase occurs in both the LDL and the HDL fraction, although the increase in LDL is often somewhat greater than the increase in HDL (48). However, the fact that observational studies typically find little or no correlation between habitual long-term saturated fat intake and circulating cholesterol (or cardiovascular risk) makes me wonder how durable these effects are (49). Longer-term randomized controlled trials also often show little or no impact of 2-3-fold differences in saturated fat intake on circulating cholesterol, as long as other relevant dietary components including linoleic acid (n6 PUFA) and fiber are kept relatively constant (50). Research is ongoing, but I'm not currently convinced that this mechanism plays a major role in cardiovascular disease.
High levels of animal protein (from meat or dairy) can increase LDL cholesterol and aggravate cardiovascular disease in certain animal models, while plant protein is often protective (51). This appears to be due to the amino acid composition of animal vs. plant protein (lysine-to-arginine ratio), and the phytochemicals and fiber associated with plant protein. Certain plant proteins, such as soy, also tend to lower LDL cholesterol in humans, particularly when they replace animal protein (52, 53). This probably goes a long way toward explaining the LDL-lowering effect of vegetarian and particularly vegan diets, and it is consistent with a protective cardiovascular effect.
Another potential mechanism for harmful effects of meat is the compounds that are produced during high-heat cooking. Some of these may be carcinogenic and contribute to cardiovascular disease (54). We might be better off focusing on low-heat-cooked meat rather than grilled, browned, roasted, or fried meats-- but this also applies to plant foods.
Red meat is very high in iron, which some researchers have proposed could compromise cardiovascular health. Free (unbound) iron is particularly good at catalyzing powerful free radical reactions that damage surrounding molecules. While iron is an essential nutrient and high-iron foods can be particularly beneficial for pre-menopausal women, many men and post-menopausal women may get too much dietary iron. Personally I think excess dietary iron is probably detrimental to overall health, but the evidence is not currently very convincing that body iron status plays a major role in cardiovascular health in particular (55, 56). The case isn't closed yet.
Sialic acids are unique sugars that are used as building blocks for certain classes of molecules in the body. Over the course of evolution, humans have lost the ability to produce a particular sialic acid called Neu5gc. However, we absorb it from animal foods, particularly red meat, and it is incorporated into our (glyco)proteins and other molecules. The human body recognizes it as foreign and produces antibodies against it. Some researchers have proposed that the immune reaction that the body mounts toward Neu5gc from meat causes inflammation and contributes to disease risk (57). Although the paper introducing this mechanism made a big splash when it was first published, to my knowledge the finding hasn't been followed up sufficiently to make a compelling case that Neu5gc actually contributes to human disease. Perhaps future research will clarify that.
There are many scary-sounding mechanisms by which meat intake could potentially contribute to disease. However, I'd like to point out that you can find something wrong with any food if you look hard enough. Melissa McEwen illustrated this well in her satire piece "Just Kale Me". Mechanisms are relevant, but they only serve to explain or bolster demonstrated effects. For example, if poultry consumption doesn't actually increase cardiovascular risk, there's no point digging around for mechanisms. When they have been identified, mechanisms can increase our confidence in an empirically demonstrated effect on health.
Synthesis and Conclusion
There is a lot of evidence we can bring to bear on this question, and not all of it is consistent. This inconsistency is why we see different groups interpreting the research in opposite ways.
Here is my best attempt to synthesize the overall evidence, as I see it. Vegetarian and vegan diets probably do reduce cardiovascular disease risk, consistent with the LDL-lowering effect of replacing meat with plant protein. However, the risk reduction afforded by avoiding meat per se appears modest (as opposed to living the whole lifestyle of a vegetarian or vegan), and we have no examples of healthy vegan cultures. There is little evidence that poultry or seafood promote cardiovascular disease, and some evidence suggesting that seafood is actually protective. Although the evidence on red meat isn't very consistent, and we need more and better research, it's probably prudent to keep intake modest for now. It's probably a good idea to limit processed meat intake as well (attn LCers and Paleos: bacon).
For people at high cardiovascular risk, it may be useful to replace some animal protein with plant protein such as beans and nuts, and focus meat intake on seafood and poultry. Randomized controlled trials such as the Oslo Diet-Heart study have supported the ability of similar dietary patterns to reduce cardiovascular risk (58).
The ideal strategy would also incorporate other factors that are relevant to cardiovascular risk, such as controlling body fatness, eating potassium- and polyphenol-rich fruit and vegetables, focusing on unrefined food, exercising regularly, avoiding prolonged sitting, managing stress, avoiding cigarette smoking, and working with a doctor to monitor and possibly control blood pressure and lipids.
So, does eating meat increase cardiovascular risk? Yes! And no!
In the next post, we'll examine the impact of meat consumption on obesity risk.
* Nuance: we know that people with high HDL cholesterol have fewer heart attacks, but we still don't know whether a diet-induced increase in HDL cholesterol is actually protective. Recent drug trials have shown that using drugs to simply pack more cholesterol into HDL particles does not reduce cardiovascular risk in humans, and it may even increase risk. This demonstrates that the protective function of the HDL particle is not always associated with its cholesterol content. We still have a lot to learn about why high HDL cholesterol is associated with cardiovascular protection in humans, and what diet/lifestyle factors influence its protective actions.
other source : http://log.viva.co.id, http://wholehealthsource.blogspot.com, http://stackoverflow.com
0 Response to "Is Meat Unhealthy? Part III"
Post a Comment