Elevated TMAO levels linked to mortality risk in patients with HF

Patients with stable HF and elevated levels of fasting plasma trimethylamine-N-oxide had an increased risk for all-cause mortality at 5 years, researchers reported in a new study.

For the single-center, prospective cohort study, researchers evaluated data from 720 patients with stable cardiac disease who were treated at the Cleveland Clinic from 2001 to 2007. All patients were undergoing elective, nonurgent coronary angiography and had a history of HF. Levels of fasting plasma trimethylamine-N-oxide (TMAO) and incidence of all-cause mortality were assessed during 5 years of follow-up and compared with a cohort of 300 healthy participants without known cardiac disease.

"There previously had been a connection [made] between the gut and HF, but that tended to focus on a leaky bowel: that congestion in the abdominal cavity impairs the barrier function of the intestines, and leads to microbial pieces entering into the bloodstream and promoting a generalized pro-inflammatory condition that led to adverse effects in patients with HF," Stanley L. Hazen, MD, PhD, from the department of cellular and molecular medicine at Lerner Research Institute, Cleveland Clinic, told Cardiology Today. "What this study is highlighting is a very specific and targetable pathway that involves potential dietary nutrients, specific gut microbes and an enzyme pathway that is important in the generation of TMAO and adverse prognosis in HF."

Stanley L. Hazen

According to the results, patients with HF had a significantly higher median TMAO level compared with those with no HF (5 mcM vs. 3.5 mcM; P<.001). Researchers observed a modest correlation between levels of TMAO and B-type natriuretic peptide (r=0.23) and TMAO and serum levels of uric acid (r=0.29), and a negative correlation with TMAO and arylesterase activity (r=–0.135) and TMAO and estimated glomerular filtration rate (r=–0.55; P<.001 for all).

During follow-up, 207 patients in the HF cohort died. Elevated TMAO was significantly associated with increased 5-year all-cause mortality risk after adjustment for traditional risk factors and B-type natriuretic peptide (HR=2.2; 95% CI, 1.42-3.43). Further adjustment for estimated glomerular filtration rate reduced but did not eliminate the association between TMAO and mortality (HR=1.75; 95% CI, 1.07-2.86).

Assessment of TMAO as a continuous variable also indicated a significant risk increase after adjustment for traditional risk factors (HR=1.18; 95% CI, 1.06-1.31 per SD-unit increase). Researchers also noted that the addition of TMAO to a model incorporating traditional risk factors improved net reclassification (P<.001), and there was a trend toward improvement in area under the curve analysis (P=.096).

"These [results] suggest that, if we can determine or develop a means of lowering TMAO, we might lower HF risk and the adverse prognosis of HF," Hazen said. He noted that certain dietary nutrients are precursors of TMAO, and that a diet low in these precursors — found in red meat, high-fat dairy products and certain cold-water fish, among other sources — may be beneficial.

"We don't know whether or not that would reduce risks in subjects with HF, but the current studies suggest that it's a strong possibility," he said. "I'd like to see interventional studies, whether they be dietary or for a drug, probiotic or prebiotic, to try to actually demonstrate improvement in outcomes in humans [with TMAO reduction]. But that is a long way away."

In a related editorial, Jane A. Cannon, MB, CHB, and John J.V. McMurray, MD, of the British Heart Foundation Cardiovascular Research Centre at University of Glasgow, noted that these findings present several questions, including whether elevated TMAO is simply a marker of renal impairment; the impact of comorbidities such as diabetes on raising TMAO; and the link between TMAO and patient prognosis.

“It is probably unlikely that a proatherogenic mechanism could account for an increase in mortality in such a short time frame, especially in patients with nonobstructive coronary disease,” Cannon and McMurray wrote. “… Clearly, this is only the beginning of the story of TMAO in HF, but one for which we should look forward to further installments.”

For more information:

Cannon JA. J Am Coll Cardiol. 2014;64:1915-1916.

Tang WHW. J Am Coll Cardiol. 2014;64:1908-1914.

Disclosure: See the full study for a list of relevant financial disclosures. Cannon and McMurray report no relevant financial disclosures.