There is no expectation that one experiment, no matter how innovative, will definitively prove or disprove the hypothesis of an etiologic role for Chlamydia pneumoniae in atherosclerosis. Rather a gradual accumulation of data from many investigators will lead to acceptance or rejection of the hypothesis. Proof will not be absolute but an overwhelming consensus of knowledgeable scientists. Koch's postulates cannot be expected to be fulfilled. Sir Branford Hill's criteria will be helpful. They look at the possible etiologic association in terms of strength, consistency, specificity, temporality, plausibility, coherence, analogy, and experimental evidence. There are three stages of atherosclerosis—any or all of which could be potentially caused by C. pneumoniae (table 1): initiation of the disease process, acceleration of the process, and complications of the disease (e.g., myocardial infarction). If C. pneumoniae is involved in initiation of the disease, it would be a cause of atherosclerosis. If it plays a role in the acceleration of the disease that has already been initiated, then it could be considered a risk factor. Three stages of atherosclerosis by age group. It is clear that the experiments of greatest interest and significance to cardiologists (a most important constituency) are the large-scale, secondary prevention, antibiotic treatment trials. If the clinical trials clearly show a positive effect of antibiotic treatment of coronary artery disease (CAD), and animal model studies also show a favorable effect of specific treatment of C. pneumoniae, it is likely that the hypothesis will be accepted as true. If the clinical trials fail to show an effect, there will be a significant loss of interest in the basic hypothesis. The reaction to the results of the clinical trials, either positive or negative, will probably be in excess of what is justified. If positive, the results will prove little about the cause of atherosclerosis and much about C. pneumoniae being involved in the cascade of events leading to acute complications of atherosclerosis (table 2). The unlikely possibility will exist that the antibiotic effect is due to inhibition of an unidentified organism other than C. pneumoniae. If the results are negative they will say little about C. pneumoniae in the initiation of atherosclerosis and an indeterminate amount about C. pneumoniae in acceleration of atherosclerosis. Potential results of secondary prevention antibiotic treatment trials for coronary artery disease (CAD) events. A definitive demonstration of C. pneumoniae as the cause of atherosclerosis could result from the use of an effective C. pneumoniae vaccine in childhood. If the vaccine was used widely in a population and atherosclerosis was virtually eliminated, the hypothesis would be proven. Some data on the occurrence of atherosclerosis might be obtained in 20 years from pathologic examination of traumatic deaths, but it would probably take 40–50 years for definite proof. Use of vaccine with or without antibiotics in adults to stop progress of atherosclerosis could be considered. An answer to the effectiveness of this approach would have a shorter time frame. Consideration would have to be given to a possible deleterious immune reaction to organisms already in atheroma. The most likely scenario, and a more timely approach to determining the role of C. pneumoniae in atherosclerosis, will come from animal model experiments and studies of basic mechanisms. Strengthening elements of Hill's criteria through more precisely directed observational studies (seroepidemiology and microbiologic pathology) would be useful, but experimental studies will offer stronger evidence of etiology. Animal model studies to date have almost always reported some positive result for C. pneumoniae infection in the atherosclerotic process. Studies on initiation of atherosclerosis are sparse and limited to rabbits. Conversely there are many successful studies in which C. pneumoniae infection is combined with elevated lipids. We need more and more varied animal model experiments. The methods for respiratory infection of animals that promote systemic spread of the organism to arteries have been developed. Repeated inoculations are more successful than single inoculations. Longer observation of infected animals will provide more convincing pathology. New investigators with different backgrounds should be able to devise new approaches based on past experiments. There is also a clear need for experiments with larger animals, such as pigs and primates. Work with rodents may not be generalizable to humans. A continuing restraint on research is the difficulty of laboratory microbiologic work with chlamydia. Studies of basic mechanisms that explore ways in which C. pneumoniae could cause atherosclerosis are accelerating in both numbers and variety. Already such studies have left several “smoking guns.” Mechanism studies should be aided by knowledge of the genome sequence recently completed. If proof is not to be absolute but an accumulation of data that is convincing to most knowledgeable scientists, where do we stand and what can we expect? It was pointed out that chlamydia is an obligate pathogen and in most other Chlaymydia-related diseases, the organism has been shown to be causal. This is true of respiratory tract infections with C. pneumoniae. There are some workers in the field who believe current evidence adequately supports the hypothesis. Most will not be convinced without additional data. How long it will take to accumulate adequate data is a matter of widely different estimates—from a year or 2 to decades. Even with additional positive evidence convincing to knowledgeable scientists, acceptance of the hypothesis by the scientific and medical community at large may not follow. The well-entrenched lipid hypothesis will not be easily replaced or added to. Analogies can be made to the ways in which papilloma virus was accepted as the cause of cervical cancer and Helicobacter pylori as the cause of peptic ulcer. One was almost entirely basic science mechanistic studies and the other mainly clinical treatment trials. In both cases, acceptance was slow. Education and time will be required. The fact that peptic ulcers and cervical cancer are caused by infection will be helpful. Acceptance will be speeded if the clinical treatment trials show a reduction in disease events. There is a clear need for funding agencies to aggressively support the needed experimental studies. Attention should also be given to improvement, simplification, and standardization of the microbiologic methods used in studies of C. pneumoniae.
No takes yet. Share an insight, caveat, or question.
J. Thomas Grayston (2000) studied this question.