Established risk factors for coronary artery disease (CAD) such as elevated serum lipids, smoking, hypertension, diabetes, age, gender, and family history are estimated to account for less than half the prevalence and severity of CAD. In searching for additional risk factors, the potential role of infection in the pathogenesis of atherosclerosis has received increasing attention. Based on the evidence from seroepidemological studies, Chlamydia pneumoniae has emerged as the organism with the strongest apparent relationship with CAD. The first serological evidence emerged in 1988. Since then, several more studies have established the prevalence of this organism in patients with CAD. Histopathological studies have provided more direct evidence of C. pneumoniae in atherosclerotic lesions. However, despite evidence of its presence in patients with CAD, the significance of this association and whether or not the organism contributes to the progression and complications of the disease are unknown. Although several mechanisms by which C. pneumoniae may contribute to atherosclerosis have been postulated, the hypothesis that this organism contributes to inflammation in atherosclerosis has resulted in clinical trials of antichlamydial antibiotics in CAD patients. The results of the three trials completed to date have been inconclusive. Further evidence is required to establish the role of antibiotics in CAD. Currently, large secondary prevention trials are under way (WIZARD and ACES) which should help to elucidate whether C. pneumoniae plays a role in the etiology of atherosclerosis and its complications.
Atherosclerosis, the pathological basis for coronary heart disease, is highly prevalent in Western society and a major cause of mortality. Established risk factors for coronary artery disease (CAD) such as elevated serum lipids, smoking, hypertension, diabetes, age, gender, and family history are estimated to account for less than half the prevalence and severity of CAD. Many of the attributable risks for this disease remain unknown. In the search for additional risk factors, the potential role of infection in the pathogenesis of atherosclerosis has received recent attention.
Several organisms have been studied to determine a potential relationship with CAD. The organisms most often implicated include herpes viruses (particularly cytomegalovirus), Helicobacter pylori and C. pneumoniae. Based on seroepidemologic and histopathologic evidence, C. pneumoniae, has emerged as the pathogen that appears to have the strongest apparent relationship. C. pneumoniae is a gram-negative obligatory intracellular bacteria responsible for a wide range of chronic, progressive, inflammatory conditions and is widely recognized as an important cause of respiratory tract infections. This article will provide a brief overview of the evidence for C. pneumoniae in CAD and the preliminary evidence for the role of antibiotics in CAD.
Many seroepidemiological studies have been completed over the past decade that provide evidence for an association between C. pneumoniae and CAD. These studies involve the examination of sera from CAD patients to determine the presence of elevated antibody levels (primarily immunoglobulins (Ig) G and IgA antibodies) to the organism. Following an acute infection, IgG antibody titers rise and then usually decrease slowly, whereas IgA antibodies tend to disappear more rapidly. The characteristics of antibody titers in patients with chronic infection are less well known; however, the persistence of elevated titers is generally considered to be a marker of a chronic infectious process. Thus, these seroepidemiologic studies have examined elevated antibody titers as diagnostic criteria for chronic C. pneumoniae infection.
Saikku et al. provided the first serologic evidence of a relationship between bacteria and CAD in a small observational study in 1988. These investigators examined serum samples from 40 male patients with acute myocardial infarction (MI), 30 male patients with chronic CAD and 41 control subjects. A total of sixty-eight percent of patients with acute MI and 50 % of patients with chronic CAD had elevated IgG and/or IgA antibody titers against C. pneumoniae as compared to 17% of control subjects (p = 0.003 for both acute MI and CAD groups versus control). Following this study, the same group of investigators examined 103 serum samples collected prospectively from patients involved in the Helsinki Heart Study. These serum samples were collected three to six months before a cardiac endpoint (myocardial infarction or sudden cardiac death) and were matched with those from controls. Among case subjects with elevated IgA titers, the odds ratio (OR) for the development of CAD was 2.7 (95% confidence interval [CI] 1.1-6.5, p = 0.02). This relationship was found to be independent of other confounding factors such as age and smoking.
Mendall et al. demonstrated that 22 % of men with angiographically proven CAD (n = 100) had elevated titers of IgG antibody to C. pneumoniae as compared with 4.7% of age-matched controls. The IgG titre associated with disease was independent of all other assessed risk factors at titers above 1:64. In another larger cross-sectional study of 388 men, a significant association between raised IgG titers to C. pneumoniae and CAD was demonstrated (OR 3.06; 95% CI 1.33-7.01), after adjustment for potential confounding factors. Thom et al. evaluated 171 men, 55 years of age or younger, and women, 65 years of age or younger, who had at least one coronary artery lesion and compared them to 120 control patients. Their results indicated that the OR for CAD associated with the presence of anti-C. pneumoniae antibody was 2.6 (95% CI 1.4-4.8).
Melnick et al. evaluated the sera from a subgroup of patients with asymptomatic carotid wall thickening involved in the Atherosclerosis Risk in Communities (ARIC) study and found that among the 326 case-control pairs, 73% of atherosclerotic cases had serologic evidence of C. pneumoniae infection versus 63% of controls (OR 1.76, 95% CI 1.21-2.57). After adjusting for confounding risk factors, the OR was 2.00 (95% CI 1.19-3.35). Further studies have subsequently identified similar associations between this organism and disease, adding to the growing body of seroepidemiological evidence.
Examination of plaques from coronary and other arteries has provided more direct evidence of the possible involvement of C. pneumoniae in atherosclerotic disease. In 1992 Shor et al. first detected C. pneumoniae in fatty streaks and atheromatous lesions in seven autopsy cases. There are now over seventeen reports describing the detection of C. pneumoniae in cardiovascular lesions. The presence of C. pneumoniae has been demonstrated in the atherosclerotic plaque using a variety of techniques, such as polymerase chain reaction (PCR), electron microscopy, immunostaining, and culture of the organism. A wide range of detection rates exists between different investigators and with various techniques. Gibbs et al. conducted a summation analysis of data from fifteen papers that examined the prevalence of C. pneumoniae in atherosclerotic arterial lesions (n=597). Eleven of these papers involved a comparison with non-atherosclerotic controls (n=131). The analysis demonstrated that 303 (50.8%) of the atherosclerotic arterial specimens were positive for C. pneumoniae against five (3.8%) of controls. Other authors have quoted similar figures in summarizing the histopathologic data. The results indicate a greater prevalence of C. pneumoniae in atherosclerotic specimens; however, possible publication bias (i.e. non-publication of negative studies) may affect the rate of detection in this summation analysis. Finally, the presence of C. pneumoniae in atherosclerotic lesions is not proof that it has a role in the atherosclerotic process.
Despite evidence of its presence in patients with CAD and in atherosclerotic lesions, the actual role of C. pneumoniae in atherosclerosis remains unknown. There is no direct evidence that these organisms can cause the lesions of atherosclerosis. However, several mechanisms by which C. pneumoniae may contribute to atherosclerosis have been proposed. Generally, infectious agents are thought to promote atherogenesis through both direct and indirect effects on vascular wall cells. It is postulated that the presence of C. pneumoniae or its antigens within cells of the arterial wall acts as a stimulus to chronic inflammation. C. pneumoniae commonly infects monocytes and chlamydial species can infect epithelial cells persistently under certain conditions. The cytokine response from endothelial and smooth muscle cells, as well as from leukocytes, is regarded as an important mechanism in atherosclerosis-associated inflammation. C. pneumoniae in human monocytes has been shown to induce production of cytokines, including tissue necrosis factor (TNF), interleukin (IL) 1-beta and IL‑6, as part of the cell-mediated immune response. In circulation, these cytokines promote a hypercoagulable state by activating platelets, increasing hepatic synthesis of acute-phase proteins (including fibrinogen) and inducing tissue factor release from endothelium with subsequent activation of the coagulation cascade. TNF also inhibits the action of lipoprotein lipase leading to altered lipid metabolism, accumulation of serum triglycerides and a decrease in serum high density lipoprotein cholesterol (HDL-C), all of which produce a favourable environment for atherogenesis. Seropositivity to C. pneumoniae has also been associated with elevated C-reactive protein (CRP), a marker of inflammation that is produced in response to IL-1 and IL-6.
The established association between C. pneumoniae infection and CAD determined by seroepidemological and histopathological studies has led to the hypothesis that this organism causes or contributes to inflammation in atherosclerosis. This, in turn, has resulted in the completion of several clinical trials involving the use of antichlamydial antibiotics in patients with coronary artery atherosclerosis.
The first of these studies was a prospective, randomized, double-blind, placebo-controlled trial involving 220 male patients. In this trial, Gupta et al examined the effect of azithromycin, a macrolide antibiotic known to have good activity against C. pneumoniae, in a subgroup of post-MI patients. Antibodies to C. pneumoniae (”anti-Cp antibodies”) were measured at baseline and at three months. Those patients with persisting seropositivity at greater than or equal to a 1:64 dilution on both occasions were randomized to receive either azithromycin 500 mg daily for three days for one course (n = 28), the same azithromycin regimen for two courses at a three month interval (n = 12), or placebo (n = 20). Patients without anti-Cp seropositivity on both occasions were not randomized to a treatment group. Adverse cardiovascular events (nonfatal MI, unstable angina, coronary angioplasty, urgent coronary artery bypass surgery or cardiovascular death) were monitored for 18 months from the original Cp titre determination. Both the nonrandomized and the randomized patients were followed. Those patients who received azithromycin demonstrated a five-fold reduction in cardiovascular events (OR 0.2 95% CI 0.05-0.8, p = 0.03) as compared with the combined group of non-randomized seropositive patients and those seropositive patients who received placebo only. The investigators rationalized the combination of data for the placebo and non-randomized seropositive patients based on the observation that there was no significant difference in cardiovascular risk factors or events between these two groups. Further justification for this data manipulation is not provided nor is there a clear definition of what type of patients comprised the non-randomized seropositive group. It appears that the non-randomized seropositive patients were those patients that had antibody titres greater than or equal to a 1:64 dilution at baseline but not at three months (and therefore not eligible for randomization) but again this is not clear. There was no difference between the treatment groups receiving either one or two courses of azithromycin with respect to cardiovascular event rate or proportion of patients with a decrease in anti-Cp titer.
Gurfinkel et al. conducted a randomized, double-blind trial involving 202 patients with unstable angina or non-Q-wave MI (Randomised Trial of Roxithromycin in non-Q wave Coronary Syndromes [ROXIS trial]). Patients were randomly assigned to receive either roxithromycin 150 mg orally twice daily or placebo twice daily for thirty days. Randomization was independent of C. pneumoniae serostatus and patients were followed for six months. The primary end-point was a composite of three clinical outcomes including cardiac death, acute MI, and severe recurrent ischemia. The incidence of the composite endpoint was 9% in placebo recipients as compared to 2% in roxithromycin recipients (p = 0.032) at 30 days (end of treatment period). The difference between composite endpoint rates were not statistically significant during the follow-up period at day 90 (12.5% versus 4.37%, placebo versus roxithromycin recipients, respectively, p = 0.065) and at day 180 (14.6% versus 8.69%, p = 0.259). (19) Baseline C. pneumoniae serology was performed and IgG antibodies were present at a titer of 1:64 or more in 47% of roxithromycin recipients and 49% of placebo recipients. No subset analysis on the effect of roxithromycin in seropositive versus seronegative subjects was performed. C. pneumoniae IgG titers, measured at baseline and at day 31, were unchanged in both groups. The authors refer to this study as a ‘pilot’ as it was not adequately powered to find significance for the primary endpoint. This coupled with the lack of correlation with IgG titres make it difficult to draw conclusions from this study. Although not statistically significant, the difference in composite endpoint rates at thirty days for the roxithromycin group were postulated by the authors to be due either to an antichlamydial activity or anti-inflammatory activity.
The third published trial to examine he effects of antibiotics in CAD is the Azithromycin in Coronary Artery Disease: Elimination of Myocardial Infection with Chlamydia (ACADEMIC) study. In this study, 302 CAD patients with positive C. pneumoniae IgG titers (³ 1:16) were randomized to receive either azithromycin 500 mg daily orally for three days followed by 500 mg per week for three months (n = 150) or placebo (n = 152). Global inflammatory marker scores (a composite of the circulating inflammatory markers CRP, IL-1, IL-6, and TNF-a), C. pneumoniae antibody titers, and cardiovascular events were assessed at three months and again at six months. At baseline, treatment groups were found to be similar, with an average age of 64 years and similar distribution of atherosclerotic risk factors. No difference in global inflammatory marker scores at the end of the three-month treatment period was observed. However, at six months the global inflammatory marker scores were lower in the azithromycin group, both for change scores (p = 0.03) and for actual value scores (p = 0.01). Antibody titers were found to be unaffected by therapy and the incidence of clinical cardiovascular events at six months did not differ between treatment groups (9 vs 7 events for active drug and placebo groups, respectively). The authors reasoned that although the study was powered to assess the laboratory end points, for the clinical end point the event rate at six months was too low to demonstrate a clear therapeutic benefit. The investigators recognized that the event rate would be too low at six months and, therefore, the primary clinical endpoint was extended to two years. These results are not yet available.
The effect of various antibiotics on MI event rate has also been retrospectively assessed by Meier et al. using a general practice database in the United Kingdom. A sample of 3,315 patients with a diagnosis of first time acute MI were compared with 13,139 controls without MI. Each patients’ medical history for the previous three years was retrospectively examined. These investigators determined that the MI cases were less likely than the control cases to have used a tetracycline or a quinolone antibiotic in the past. In contrast, there was no association between MI occurrence and past use of macrolides, sulfonamides, penicillins or cephalosporins. The apparent lack of reduced risk with macrolides appears surprising; however, the author attributes this to the principal macrolide in use in this study sample, erythromycin (75% of total macrolide utilization), which is stated to be less efficacious against C. pneumoniae. These results appear to support the hypothesis that certain chronic bacterial infections may play a role in the etiology of ischemic heart disease but further conclusions cannot be drawn.
Seroepidemological studies have confirmed an association between C. pneumoniae and atherosclerosis. Furthermore, histopathological studies have provided evidence for the presence of this microorganism in the atherosclerotic lesion. Although the association between C. pneumoniae and atherosclerosis has been established, the significance of this association and whether or not the organism contributes to the progression and complications of the disease are still unknown. Studies evaluating the effect of antibiotic treatment on cardiovascular events can be utilized to shed light on the potential etiologic role of C. pneumoniae. The three prospective antibiotic trials conducted to date provide inconclusive evidence. There are now at least two larger secondary prevention antibiotic trials underway namely the WIZARD (Weekly Intervention with Zithromax against Atherosclerotic-Related Disorders) and ACES (Azithromycin Coronary Events Study) trials that may be adequately powered to determine an effect on cardiac events. If C. pneumoniae plays a role in the etiology of atherosclerosis and its complications, these secondary prevention studies should provide useful data to help us further elucidate this relationship. However, even with this evidence several issues will remain unanswered. The choice of optimal antibiotic regimens, duration of therapy, the relevance of the potential for non-antimicrobial effects of antibiotics and the role of other anti-chlamydial antibiotics will require further investigation. As well, the potential for widespread antibiotic resistance if chronic use of broad-spectrum drugs in this setting becomes common will need consideration.