Whether cardiovascular disease is an infectious disease is not clear. A number of infectious agents have been implicated, including Helicobacter pylori, cytomegalovirus and periodontal bacteria, but by far the most studied is Chlamydia pneumoniae. During the past decade, the role of this organism in development of atherosclerosis, coronary heart disease (CHD) and stroke has been extensively explored, and associations with other vascular diseases, such as abdominal aortic aneurysm, have been proposed. Linking C. pneumoniae with the development or outcome of cardiovascular disease would have a substantial effect on antibiotic use. Indeed, a 1999 survey found that up to 4% of physicians in the United States had recommended treating cardiovascular disease with antibiotics.1 As inappropriate use of antibiotics will affect the development of antibiotic resistance, it is crucial to assess the evidence. This review focuses on the relationship between C. pneumoniae and vascular disease and the evidence on antibiotic therapy for cardiovascular conditions.

In 1988, Saikku and colleagues reported serological evidence of an association between C. pneumoniae infection and acute myocardial infarction.2 Since then, numerous studies have examined the relationship between raised C. pneumoniae antibody titres and vascular disease. A review in 1997 identified 18 case–control studies with 2700 cases. Most studies reported an odds ratio (OR) of 2 or more, suggesting a real association between serological markers of C. pneumoniae infection and vascular disease.3 More recent meta-analysis of 16 prospective case–control studies reported a weak association between raised C. pneumoniae IgA titres and CHD (OR, 1.25; 95% CI, 1.03–1.53), and a non-significant association between raised IgG titres and CHD (OR, 1.15; 95% CI, 0.97–1.36).4,5

However, a major difficulty in interpreting these results is the lack of a gold standard for diagnosing chronic C. pneumoniae infection of blood vessels. The gold standard for acute C. pneumoniae infection is the microimmunofluorescence test — an IgM titre greater than 1:16 or a fourfold rise in IgG titre is considered diagnostic. While persistently raised IgG and IgA titres have been proposed as criteria for chronic infection, there are no uniform cutoff titres to define seropositivity, and it is unclear whether seropositivity reflects chronic, or merely past, infection.6 In addition, the microimmunofluorescence test is technically challenging, and may be replaced by new enzyme-linked immunosorbent assays.7

Furthermore, smoking is an independent risk factor for C. pneumoniae seropositivity. There may also be a positive correlation between C. pneumoniae seropositivity and a serum lipid profile associated with an increased risk of atherosclerosis, and between seropositivity and essential hypertension.8-10

The role of seroepidemiology thus remains controversial. Although seropositivity established a link between C. pneumoniae and atherosclerosis, it seems unlikely that serological tests alone will identify individuals at high risk for atherosclerosis.

The pathological mechanisms underlying the proposed atherogenic effect of C. pneumoniae can be viewed in the light of atherosclerotic plaque development. Initially, fatty streaks form in arterial walls through the accumulation of low-density lipoprotein (LDL) particles within the subendothelium. This leads to recruitment of lymphocytes and monocytes, which differentiate into macrophages and subsequently develop into foam cells. Later in life, smooth muscle cells derived from the media of the artery wall migrate to the subendothelium and form a fibrous cap around the foam cells. This cap is continuously degraded and replaced under the influence of the inflammatory cells within. Plaque rupture and thrombus formation is a very common underlying cause of CHD and stroke.

Risk factors for atherosclerosis are multiple, and include a high-fat diet, raised LDL levels, raised blood pressure, smoking, lack of exercise and hereditary factors.11

The role of C. pneumoniae must be elucidated within this context. C. pneumoniae causes respiratory disease, and serological studies reveal that more than half the adult population worldwide has been infected. Most seroconverters for C. pneumoniae are found at ages five to nine years, the same age that fatty streaks begin to form. Chlamydiae are notorious for establishing chronic infections that resist treatment.12 In-vitro studies have shown that C. pneumoniae can grow in macrophages, endothelial cells and vascular smooth muscle cells.13 Some investigators have isolated viable chlamydia from atherosclerotic tissue, but others have not been able to replicate this finding.14 Furthermore, some studies have detected C. pneumoniae in atheromatous tissues through techniques including polymerase chain reaction (PCR) and immunocytochemistry.15 The detection rate was about 50% overall, varying from zero to 100%. In contrast, detection rates in normal arterial tissue were about 1%.15 However, PCR detection does not seem completely reliable. For example, a study comparing detection rates between laboratories found that three of 16 control samples negative for C. pneumoniae were rated positive by PCR analysis, while, at low C. pneumoniae concentrations, only three of 16 positive control samples were rated as positive.16 Despite these problems, the aforementioned studies suggest that C. pneumoniae may establish infection in vascular tissue. This infection could, if chronic, provide the antigen for chronic inflammation.

Detection of C. pneumoniae in atherosclerotic lesions prompted research on the antigen specificity of lymphocytes within the lesions. Several studies have found that lymphocytes propagated from atherosclerotic tissue are responsive to C. pneumoniae, but also to other recall antigens such as tetanus toxoid and purified protein derivative. Interaction between C. pneumoniae and T lymphocytes and subsequent production of inflammatory cytokines, such as interferon gamma, is a proposed mechanism of plaque destabilisation.17,18

Development of foam cells from macrophages is a feature of atherogenesis. An in-vitro study found that exposure of human macrophages to a combination of C. pneumoniae and LDL induced their transformation into foam cells.19

Another suggested pathological mechanism is an autoimmune reaction involving bacterial heat-shock protein (HSP) with a high sequence homology to human HSP 60. The latter is an important product of cells in the arterial wall, protecting them against unfavourable conditions. Presence of antibodies directed against bacterial HSP (including chlamydial HSP) has been shown to be independently associated with prevalence of atherosclerosis, as well as with seropositivity to C. pneumoniae.20

Some of the more compelling arguments for an aetiological role for C. pneumoniae in atherosclerosis come from mouse and rabbit models. In hyperlipidaemic animals (genetically or diet induced) predisposed to develop atherosclerosis, experimental infection accelerates inflammatory progression.21 Other studies have found inflammatory vascular changes in animals that do not normally develop atherosclerosis after single, and particularly repeated, inoculations of C. pneumoniae,22 although it has been suggested that this effect depends on high serum cholesterol levels. An actual atherogenic effect in both disease initiation and disease progression has thus been convincingly proposed in animals.23 Whether this reflects human atherosclerotic development is unclear.

Although causality has not been established between C. pneumoniae infection and cardiovascular disease, studies of the effects of antibiotic treatment on the disease are completed or under way. The optimal treatment regimen for C. pneumoniae infection has not yet been established, but the microbe is susceptible to tetracyclines and macrolides, including azithromycin and roxithromycin.24 These newer macrolides are the most common agents used in prospective trials.

It is important when evaluating the results of these trials to consider that macrolides and tetracyclines have considerable anti-inflammatory as well as antimicrobial activity, which is a potential confounding factor.

Our knowledge of the relationship between C. pneumoniae and cardiovascular disease has expanded greatly. Serological evidence of infection confers a moderately increased risk of atherosclerosis. Identification of the organism and the consequent lymphocytic response in diseased vascular tissue is consistent with an infectious aetiology. Animal studies strongly support an atherogenic link. However, the clinical impact of the associations remains to be clarified. Macrolide treatment may have a short-term effect, particularly in patients with no known history of cardiovascular disease, but this effect may be due to their anti-inflammatory rather than antichlamydial activity. Ongoing trials will not be able to differentiate these effects, but will show whether antibiotics are beneficial in patients who have already had atherosclerotic events. At present, there is no justification for treating cardiovascular disease with antibiotics.