By: Dr. Geoffrey Modest
So, another microbiome blog (I can’t help myself). In a 5-year observational study of 720 patients with heart failure followed at the Cleveland Clinic, elevated levels of TMAO (trimethylamine-N-oxide) were associated with long-term mortality, independent of traditional risk factors and BNP or renal function (see doi.org/10.1016/j.jacc.2014.02.61). Per prior conceptions, the role of the gut in heart failure was basically that splanchnic circulatory congestion led to changes in gut permeability and in intestinal barrier integrity leading to entry into the circulation of lipopolysaccharides from gram negative bacteria, activating cytokines and creating systemic inflammation. The current study basically found that the plasma TMAO levels were increased in those with CHF. TMAO is derived from foods containing carnitine (eg red meats) or lecithin (eg from eggs). (See the blog from 2013 at the end, which found that consumption of red meat actually led to changes in intestinal bacteria that, in the presence of red meat or carnitine, led to TMAO production). Other studies have found that increased TMAO levels is associated with coronary artery disease and poor prognosis, perhaps related to TMAO effects on cholesterol transport, macrophage activity, ?other mechanisms. (And, by the way, it is possible that the association between renal dysfunction and CAD is at least partly mediated by TMAO levels, since TMAO is renally-excreted). And other studies have found relations between the gut microbiome and obesity, type 2 diabetes, and chronic inflammation.
In the current study (sponsored by NIH):
–702 patients (mean age 67, 59% male, 41% with diabetes, 78% hypertension, 64% ischemic cardiomyopathy, BMI 28.4, with baseline meds: ACE-I/ARB 70%, b-blockers 70%, loop diuretics 59%, statins 61%, aspirin 64%; and baseline BNP 300, hsCRP 3.5) undergoing elective non-urgent caths, and followed 5 years
–Median TMAO level was 5.0 mM in this group, but 3.5 mM in a healthy cohort (their healthy cohort was 300 people without known heart disease from a health-screening program)
–Those with higher TMAO levels tended to be sicker: history of diabetes, renal insufficiency, lower HDL levels; and also to have higher BNP levels and diuretic use and lower use of b-blockers.
–BUT, there was a strong, graded association between TMAO levels and mortality, even after adjustment for traditional cardiac risk factors, with mathematical modeling showing that adding TMAO levels significantly improved the cardiac risk classification over the standard cardiac risk calculator (eg Framingham).
–TMAO levels were also significantly associated with mortality, controlling for BNP and eGFR
–Subgroup analysis showed no difference in outcome for those with ischemic or nonischemic heart failure
–Overall, 1.8-fold increase in mortality in patients with fasting TMAO levels >8.5mM, controlling for all of these cardiac and renal risk factors
One interesting feature of this study is that there was no difference in the mortality associated with TMAO in those with ischemic vs. non-ischemic cardiomyopathies, suggesting that the pro-atherosclerotic mechanisms of TMAO really may only explain part of its toxicity in heart failure. Or that there are multifactorial changes with a disease such as heart failure, leading to diverse microbiome changes, with the production of TMAO plus other changes in other bacteria/toxins, etc.
So, not that I am going to run out and measure the TMAO levels on all of my patients with heart failure, but this study adds to other ones highlighting the significant role of the gut microbiome in health and disease. And the microbiome is very susceptible to changes in diet (as in blog below), use of antibiotics, even other meds. For example, there was a study of metformin (see dm metformin microbiome GUT 2014 in dropbox, or doi.org/10.1136/gutjnl-2013-305370), the 7th most prescribed medicine and the first choice for diabetics (in part because of its cardio protection!!). This study was in mice, and found that those on metformin improved their glycemic profile, but significantly changed the gut microbiome to increase the bacterium Akkermansia. In another experiment, just increasing this bacterium (in the absence of metformin) also enhanced glucose tolerance and decreased adipose tissue inflammation, suggesting that an additional mechanism of action for metformin may be through its effect on the microbiome.
Bottom line: a healthy diet (esp. one high in fruits and vegetables — eg Mediterranean diet), exercise, and a generally healthy lifestyle are really important to preserving health, and deviations clearly affect the microbiome, which might be part of the cause of many of the common diseases in our society. What this really means is that our often simplistic/reductionist models (eg that the Mediterranean diet improves lipids, decreases inflammation, decreases oxidative stress, etc. etc.) sometimes leads us to miss the big picture: the better the quality of diet for example has profound effects on the overall, coordinated functioning of our complex bodies in ways we do not fully understand, and that medications which individually improve lipids, decrease inflammation, decrease oxidative stress etc. really do not fully compensate for lack of a good diet. medications are often necessary (and I certainly prescribe my fair share) but they also often have unanticipated changes in the body (eg effects of antibiotics on the gut microbiome). The key is for society to coherently and unambiguously promote these healthy lifestyle components and not rely on medications to compensate for the adverse effects of smoking, sedentary lifestyle, poor diet, job/social stress, etc.
Blog from 2013:
Article in NY times (see here) seems that presumptive coronary heart disease culprit of carnitine (in red meat, same Latin route at “carnivore”) may be only partially correct. Published in Nature Medicine. Turns out to (maybe) be more complex:
–In regular meat eaters, eating a sirloin leads to an increase in TMAO (trimethylamine-N-oxide), which in lab studies leads to accumulation of cholesterol in macrophages and also seems to decrease the ability to excrete excess cholesterol (for excruciating detail, see the paper itself), with some previous data that blood TMAO is assoc with CAD, both in humans and mice
–In vegans who had a similar meal of sirloin (presumably with a significant bribe), no increase in TMAO.
–Larger group of vegetarians given carnitine pill, also no increase in TMAO.
–So, speculation is that it was intestinal bacteria responsible (precursor TMA known to be produced by intestinal bacteria), and that the intestinal flora are different in vegetarians than meat-eaters — mouse studies show that increasing L-carnitine leads to changes in intestinal bacteria. beef has highest level of carnitine. Lesser amounts in fish, chicken, and dairy. Also in high-energy drinks for body-builders
–Meat-eaters pretreated with antibiotics to sterilize the colon did no have TMAO in their blood
–Further analysis of 2600 pts., controlling for traditional risk factors, found that TMAO and not carnitine was assoc with CAD
So, bottom line, might be good idea to pre-treat cows with even more antibiotics….