The failure of antibiotics to prevent heart attacks

Two recent trials have shed important light on the theory that the respiratory pathogen Chlamydia pneumoniae might cause atherosclerotic cardiovascular disease.¹,² The first, by Grayston et al, was a trial of azithromycin or placebo taken each week for one year by 4012 patients with stable coronary artery disease who were followed up for four years. The second, by Cannon et al, was a trial of gatifloxacin or placebo taken for 10 days each month for two years by 4162 patients who were in hospital with an acute coronary syndrome. In neither trial did the antibiotic therapy reduce the occurrence of serious cardiovascular events, confirming the conclusion of an earlier meta-analysis of smaller studies.³

Three facts are beyond question in the relation between C pneumoniae and atherosclerosis. Firstly, C pneumoniae DNA and/or antigen have been detected, mainly by polymerase chain reaction technology, in 40% or more of atherosclerotic plaques of patients in various parts of the world, detection being recorded from about the age of 15 onwards.⁴ Secondly, mice and rabbits inoculated with C pneumoniae have developed inflammatory lesions in arteries,⁴ although, admittedly, these models may not mimic faithfully what happens in humans. Thirdly, antibiotics do not reduce cardiovascular events among patients who are already at risk.

Early reports of the benefit of short courses of antibiotics for patients who had experienced an acute coronary event⁵,⁶ have not been borne out by the new large trials of long duration.¹,² Moreover, the early promise that treatment with antibiotics might inhibit the development of abdominal aortic aneurysms and thickening of the carotid artery, both of which are associated with the presence of C pneumoniae, does not seem to have been fulfilled by prospective trials,⁷,⁸ although this aspect of research warrants further exploration.

Thus, although C pneumoniae organisms, whole or in part, often exist in diseased arteries, antibiotics with antichlamydial activity have no protective effect. Why might this be? Human atherosclerosis develops progressively from an early age. Taking antibiotics too late in the inflammatory process is unlikely to have an effect, a point acknowledged by the authors of the two recent trials.¹,² Detection of viable or metabolically active organisms in atherosclerotic lesions has proved to be difficult and achieved rarely.⁴ By inference, it would seem likely that such organisms are sparse in plaques so that antibiotics—however active and for whatever duration taken—may have no chance of having a discernable effect on the occurrence of cardiovascular events.

Furthermore, C pneumoniae or at least its DNA is probably carried from the respiratory tract to atherosclerotic lesions by peripheral blood mononuclear cells. The presence of C pneumoniae DNA in these cells has been associated with cardiovascular disease,⁹ but more studies are required to validate the notion that such bacterium carrying cells can be considered as a specific marker for the presence of C pneumoniae in atherosclerotic plaques. This would add weight and greater meaning to using the polymerase chain reaction to detect C pneumoniae in these cells¹⁰ in order to identify infected patients in trials and to measure the effect of antichlamydial treatment. In one study, however, azithromycin or rifampicin did not inhibit chlamydial carriage in peripheral blood mononuclear cells, an observation that undoubtedly needs confirmation.¹¹

The result of a trial of clarithromycin in patients waiting for coronary artery bypass surgery is also discouraging because the antibiotic did not reduce the prevalence of the main outer protein of C pneumoniae in vascular tissue taken at surgery (although the value of this finding is weakened because C pneumoniae was not detected by a polymerase chain reaction technique).¹² Lastly, not even the most avid proponent of a role for C pneumoniae in atherosclerosis would claim that it is the only factor involved. Factors unaffected by antibiotics are likely to cloud any beneficial antichlamydial effect if, indeed, it exists.

So, is this the end of the road? It is unless a different tack is taken. Certainly, the same kind of antibiotic trials have no future and, as said before, animal modelling is too remote from human reality. Danesh proposes that prospective epidemiological studies based on serology should help to better distinguish cause from consequence.¹³ We are sceptical that this will help to determine whether C pneumoniae has any role in causing coronary heart disease. The odds certainly seem stacked against such a possibility, but it is too early to draw a final conclusion while the current evidence only concerns the late stages of disease. In this context, it is a sobering thought that the role of Treponema pallidum in tertiary syphilis is not questioned despite the fact that antibiotic therapy does not alter the pathological changes at the late stage of disease.

Although large advances in dealing with heart disease may come in the future from gene therapy and stem cell research, it will still be important to determine beyond question whether antibiotic therapy can or cannot reduce C pneumoniae carriage by peripheral blood mononuclear cells. A finding that carriage can be greatly reduced or abolished significantly would provide a logical basis for trials of antibiotic treatment, not in patients with myocardial infarcts, but in young people without well established atherosclerosis. Of course, vaccination of the young against C pneumoniae, although hard to contemplate at the moment, is another avenue that could be taken. Such kinds of research will undoubtedly prove difficult but without implementation, the role, if any, of C pneumoniae in cardiovascular disease will remain shrouded in mystery for many years to come.