Chlamydophila pneumoniae infection and cardiovascular disease.

Atherosclerosis is a multifactorial vascular inflammatory process; however, the inciting cause for inflammation remains unclear. Two decades ago, Chlamydophila pneumoniae (formerly Chlamydia pneumoniae) infection was proposed as a putative etiologic agent. We performed a PubMed search using the keywords Chlamydia and atherosclerosis in a Boolean query to identify published studies on C. pneumoniae and its role in atherogenesis, and to understand research interest in this topic. We found 1,652 published articles on this topic between 1991 and 2011. We analyzed relevant published studies and found various serological, molecular, and animal modeling studies in the early period. Encouraged by positive results from these studies, more than a dozen antibiotic clinical-trials were subsequently conducted, which did not find clinical benefits of anti-Chlamydophila drug therapy. While many researchers believe that the organism is still important, negative clinical trials had a similar impact on overall research interest. With many novel mechanisms identified for atherogenesis, there is a need for newer paradigms in Chlamydophila-atherosclerosis research.

Introduction

Atherosclerosis is a leading cause of global mortality and morbidity, among noncommunicable disorders.[1] Atherosclerosis leads to coronary artery disease (CAD), cerebrovascular disease (CVD), and peripheral vascular disease (PVD), which have a wide clinical spectrum including stable or unstable angina (UA), acute myocardial infarction (AMI) or sudden cardiac death (SCD), transient ischemic attacks (TIA), ischemic strokes, vascular dementias, intermittent claudication, and gangrene. Further, atherosclerosis is responsible for a large burden of chronic kidney disease (CKD) and hypertensive heart diseases. At the population level, nine risk factors [smoking, history of hypertension, diabetes, waist/hip ratio, dietary patterns, physical activity, consumption of alcohol, blood apolipoproteins (Apo), and psychosocial factors] account for more than 90% of the population-attributable risk.[2] Despite these strong associations, interactions between these risk factors and an underlying inciting cause remain inadequately explained.[3]

A number of studies have found that inflammation of the vessel wall is an important mechanism responsible for initiation, progression, sclerosis, erosion, and rupture of atherosclerotic plaques.[4] A low-grade infection either by a virus (cytomegalovirus or human herpes viruses) or a bacterium, [Helicobacter pylori or Chlamydophila pneumoniae (formerly Chlamydia pneumoniae)] has been suggested as a possible etiology for this inflammatory activity,[4] For the last two decades, C. pneumoniae has been the strongest candidate organism. C. pneumoniae is an ubiquitous, obligate intracellular Gram-negative bacterium, and is a common respiratory pathogen.[5] It has been shown that C. pneumoniae infects human mononuclear cells, and this has also been demonstrated after respiratory challenge in animal studies.[6] These infected mononuclear cells transmigrate into circulation [peripheral blood mononuclear cells (PBMC)] and secondarily infect endothelial cells by cell-to-cell transmission of C. pneumoniae. This event could trigger a series of immunological phenomenon leading to atherosclerosis[6] [Figure 1]. Further, it has been demonstrated in animal studies that C. pneumonia infection reduces high-density lipoprotein (HDL) levels and intra-plaque hemorrhages.[7] Another mechanism for development of atherosclerosis is mediated through expression of heat shock proteins (HSP) on the endothelium. HSPs are normally located in the endothelium, and have a protective function. Anti-HSP60 antibodies induce endothelial damage, and smooth muscle proliferation. C. pneumonia infection leads to production of antibodies as bacterial HSP has a sequence homology with human HSP. Thus, molecular mimicry between human HSP60 and bacterial 60 k-Da HSP contributes to atherosclerosis.[8]

Figure 1

Postulated inflammatory pathways and role of Chlamydophila pneumoniae in causation of atherosclerosis

Studies in animal models have isolated C. pneumoniae from coronary,[9-11] carotid,[1213] and peripheral arteries.[14] These isolation studies triggered a debate if association of C. pneumoniae infection and atherosclerosis is causal, or if the infectious agent is merely an innocent bystander.[15-17] In a landmark study, Hu and coworkers[18] demonstrated that Chlamydophila infection could induce atherogenesis in low-density lipoprotein (LDL)-knockout mice, only in the presence of a high-cholesterol diet. This experiment further lent credence to the hypothesis that infection and hypercholesterolemia are essential causal components leading to atherogenesis. The current article reviews Chlamydophila atherosclerosis literature with reference to causal significance of this association and traces investigator interest in this hypothesis from the last two decades.

Materials and Methods

We performed a PubMed search to identify studies about the role of C. pneumoniae in atherosclerosis. We used a Boolean query [(C. pneumoniae) and (atherosclerosis OR coronary OR stroke OR peripheral vascular disease OR cerebral OR hypertension OR diabetes)] and identified a total of 1,668 studies. We classified published literature by year of publication and described three distinct study designs: (A) Seroepidemiological studies; (B) Molecular studies; and (C) Clinical trials. We excluded animal and cell culture studies and have elaborated human studies alone. We included serology studies, which had used either a case-control study design at a defined cross section in time, or studies nested in well-defined community cohorts. Molecular studies included those that demonstrated C. pneumoniae DNA in circulating mononuclear cells or in vascular tissue using electron microscopy, molecular diagnostics, immunochemistry, or cell line culture techniques. Clinical trials included in the description were randomized control trials, which aimed at evaluating reduction in mortality or cardiovascular events with antibiotic therapy in patients with early atherosclerotic disease.

Results

The chronologic distribution of studies evaluating Chlamydophila-atherosclerosis association shows a peak in the year 2000, and a gradual decline in number of publications since the year 2003 [Figure 2]. Published literature has three overlapping periods, with most serology studies being published in the first decade (1991-2001) followed by accumulation of molecular evidence. Between 2003 and 2006, negative results in large clinical trials heralded a gradual decline in research publications pursuing this hypothesis.[19] Subsequently, molecular studies have continued to demonstrate presence of Chlamydophila antigens in atheromatous tissue, and researchers have argued that failed antibiotic trials do not mean that the hypothesis is refuted. The seroepidemiological, molecular, and clinical trial evidence for and against this hypothesis is detailed below.

Figure 2

Number of PubMed indexed articles (n=1668) by year of publication on Chlamydophila pneumoniae infection and atherosclerosis

Seroepidemiological studies

Various seroepidemiological studies tested the hypothesis of association between exposure to C. pneumoniae and cardiovascular outcomes. Two broad study designs have tested this hypothesis. First, case-control studies with a cross-sectional design are hospital- or community-based studies, where exposure is determined in cases, after the outcome had already taken place. Second, case-control studies nested in well-defined community cohorts or prospective case-control studies, where exposure is determined from serum samples, which had been stored prior to occurrence of the outcome. These later studies characterize temporal relationship between exposure and outcome.

Case-control studies: Cross-sectional design

Of a total of 47 studies identified, 44 were included in two systematic reviews[2021] [Table 1], and a majority of them (23 out of 44) reported an unequivocal positive association. Another 16 studies reported a non-significant positive outcome [odds ratio (OR) for C. pneumoniae seropositivity was greater than 1.0, but the 95% confidence interval (CI) crossed 1, while remaining five studies reported negative point estimates (with OR of less than 1.0)]. The range of point estimates was wide (0.7-17.0), representing a considerable heterogeneity in the results. The results of 29 studies included in the review by Bloemankamp et al.[20] were pooled using the random effects model, and the weighted pooled OR was 2.0 (95% CI 1.5-2.6).

Table 1

Case-control studies that evaluated the relationship between exposure to Chlamydophila pneumoniae seropositivity and cardiovascular risk in a cross-sectional manner

Most of the earlier studies had evaluated immunoglobulin (Ig) G antibodies, which signify past infection. Subsequently, IgA antibodies were shown to be better associated with the presence of C. pneumoniae in the atheromatous tissue, signifying a persistent infection. Different studies showed a higher prevalence of IgA antibodies among patients with established athermanous disease as compared to healthy controls.[22-24] The point estimates risk ratios in these studies were higher and confidence intervals significant even after adjusting for confounders.[25-29]

Case-control studies: Prospective design

These cohorts from which cases and controls were sampled had been followed up for 3-15 years [Table 2]. Eighteen studies with this design were summarized in the two systematic reviews,[2021] and only two of them reported a positive association of C. pneumoniae seropositivity with cardiovascular events [Osserwade et al. (OR 2.8 95% CI 1.3-5.8)[30] and Roivainen et al. (1.7 95% CI 1.2-2.5)[31]]. The point estimates were equivocal in nine studies, and negative in seven others. The weighted pooled odds of 15 of these studies reviewed by Bloemankamp et al.[20] was 1.1 (95% CI 0.8-1.4). In a meta-regression of these studies, the regression coefficient was estimated to be -0.04 (-0.08, -0.01) for every additional year of follow-up of the cohort, implying that for every additional one year of follow-up, the point estimate in these studies was lowered by 0.04. The accuracy of serology for detection of Chlamydophila has also been variable across studies, and at different cut-offs [Table 3].

Table 2

Prospective case-control studies that evaluated the relationship between Chlamydophila pneumoniae seropositivity and cardiovascular risk

Table 3

Accuracy of serology with presence of Chlamydophila pneumoniae infection

Molecular studies

Circulating C. pneumoniae DNA

It has been hypothesized that PBMC positive for C. pneumoniae DNA detected by polymerase chain reaction (PCR) techniques[32] could be considered as a surrogate marker for chronic C. pneumoniae infection. Case only studies (which have reported the prevalence of C. pneumoniae–positive DNA in different population subgroups) and case-control studies (where cases include patients with atherosclerosis-associated cardiovascular conditions and healthy controls) have been used to determine risk estimates. Smeija et al.[32] published a systematic review of these studies. A few[1333-37] additional studies have appeared since the publication of this review. The prevalence of C. pneumoniae DNA in PBMC ranges from 2.5% to 46.2% in healthy blood donors and other healthy individuals who were studied in prevalence studies (in one study, all medical student controls were negative). The pooled odds of presence of C. pneumoniae DNA in PBMC as a risk factor for cardiovascular events was 2.03 (95% CI 1.34-3.08).[32]

Identification of C. pneumoniae in vascular tissue

More than 40 studies have used various techniques (immunocytochemistry techniques were used to detect the presence of specific antibodies or antigens in tissue aspirates, PCR, and in situ hybridization to demonstrate the presence of C. pneumoniae DNA; electron microscopy was used to demonstrate the intact bacterium and cell culture to isolate viable Chlamydophila organisms[4]) to determine the presence of C. pneumoniae in vascular tissues. The results of these studies have been compiled by Boman et al.,[4] and in the 2679 specimens analyzed by all authors by different techniques, immunocytochemistry (676 specimens), electron microscopy (97 specimens), and PCR (2294 specimens) showed positive results in 49.7%, 39.1%, and 24.3% cases, respectively. There was a wide heterogeneity in the results across studies [Table 4], and only 7.3% of the 451 specimens were culture positive.

Table 4

Detection of Chlamydophila pneumonia in atheromatous tissue specimens

Mechanism of C. pneumoniae–induced atherosclerosis

It has been hypothesized that respiratory infections by C. pneumoniae may result in hematogenous spread through PBMC. The organism has effects on endothelium and vascular smooth muscle cells, mediated through cytokines [interleukin (IL)-1, IL6, IL8, tumor necrosis factor (TNF)-α, interferon (IFN)-γ, platelet-derived growth factor (PDGF), HSP60, etc) resulting in upregulation of inflammation,[38] endothelial apoptosis,[39] and vascular smooth muscle proliferation. These events are likely precursors of atheroma formation. In the last five years, additional signaling mechanisms [mediated through Interferon regulatory factors 3 and 7, toll-like receptor (TLR)-2/4, IL-8, Intercellular Adhesion Molecule (ICAM)-1, vascular cell adhesion molecule (VCAM)-1, extracellular signal-regulated kinase (ERK)-1/2, nuclear factor-kappa B (NF-kB), IL-23, IL-6, IL-1beta, transforming growth factor (TGF)-beta, and chemokine (C-C motif) ligand (CCL-20)][40-42] have been postulated to have a role in the initiation and progression of atheromatous lesions. C. pneumoniae also has a role in lipid accumulation in the vessel wall by upregulating lecithin-like oxidized LDL receptors (LOX-1) in both endothelial and vascular smooth muscle cells.[4344] In the same period, four studies[45-48] did not find convincing evidence of presence of Chlamydophila genome in the atheromatous tissue, and argued in favor of alternative mechanisms. However, it is likely that C. pneumoniae infection may only be providing an initial trigger and is transient rather than persistent. The mechanism mediated through molecular mimicry between human and bacterial HSP60 and production of anti-HSP antibodies is attractive, as it strengthens ‘hit-and-run’ hypothesis for the organism.

Clinical trials

Various Clinical trials evaluated whether eradication of C. pneumoniae is beneficial in the secondary prevention of cardiovascular events. The evidence that such a therapy could be useful had come from animal studies, where administration of high-dose azithromycin for 10 weeks was associated with a reduction in intimal thickening.

Initial studies reported between 1997 and 2001 (Gupta et al., ACADEMIC, ROXIS, CLARIFY, and Leowattana et al.) had fewer subjects, shorter duration of antibiotic therapy, and shorter durations of follow-up. Subsequently, larger trials were launched (WIZARD, ACES, ANTIBIO, AZACS, and PROVE-IT) where antibiotics were administered for 3 months to 2 years, and duration of follow-up was 1-4 years [Table 5]. These later trials had greater statistical power to detect smaller differences between the intervention and non-intervention arms. In a meta-analysis summarizing the evidence from 19,217 patients in 11 randomized controlled trials, the point estimates for mortality reduction and reduction in secondary cardiovascular event were 1.02 (95% CI 0.89-1.16) and 0.92 (95% CI 0.81-1.04), respectively.[49]

Table 5

Randomized control trials that studied the impact of antibiotic therapy on subsequent cardiovascular events

Discussion

There is an ongoing debate whether the association between C. pneumoniae infection and cardiovascular outcomes is causal or the organism is merely an innocent bystander. The main arguments supporting a causal role are: biological plausibility and a consistent finding that atherosclerosis associated with vascular inflammation, is inducible by C. pneumoniae in laboratory experiments. The arguments against causality are poor association between seropositivity and cardiovascular outcomes, lack of consistency in demonstration of C. pneumoniae in vascular tissue, and failed attempts to show benefits of eradication of the organism. It, however, may be argued that we may be constrained by traditional Koch's postulates in an attempt to prove causality. In case of C. pneumonia, it is likely that the organism is not a singular causal factor in atherogenesis or its progression, its presence transient after an initial trigger, and eradication a failed aim because the persistence of the organism in atheromatous tissue and penetration of drugs are both questionable.

Seroepidemiological studies are useful in framing initial hypothesis, but these also have a major limitation. An important assumption in seroepidemiological studies is that the presence of anti-C. pneumoniae antibodies is a surrogate measure of chronic C. pneumoniae infection. This assumption may not hold true as there is a poor correlation between serology and detection of C. pneumoniae in vascular tissues [Table 3]. At low titers, serology had a poor specificity, and the number of false-positive results was high. At high titers, serology had a poor sensitivity but a high specificity.[4] Anti-Chlamydophila antibodies are measured using microimmunofluoroscence (MIF) technique, which needs an experienced microscopist to interpret the results.[4] Different studies have used in-house tests, using a variable cut-off to define positivity, and this approach is prone for misclassification error.[50] Owing to these limitations, there is a need for a better serological test to define chronic Chlamydophila infection.[1551]

Infection due to Chlamydophila species is common, has seasonal variations, and different species exhibit antigenic cross-reactivity.[51] This leads to a high background prevalence of seropositivity as well as presence of C. pneumoniae DNA in PBMC. This diminishes the strength of association between evidence of past infection and atherosclerotic diseases.[52] Further, age, smoking status, and socioeconomic status are potential confounders in the relationship between exposure to C. pneumoniae and atherosclerosis. Systematic reviews[2021] have revealed that studies that had adequately adjusted for confounders had a lower point estimate as compared to studies where adjustment was inadequate [1.1 (0.8-1.7) vs. 1.9 (1.2-3.0)]. In a meta-regression,[20] the regression coefficient for the degree of adjustment was -0.10 (-0.23, 0.02), implying that for every additional degree of adjustment, the point estimate is lowered by 0.1.

Despite reasonable molecular and serologic evidence, antibiotic trials failed to improve clinical outcomes. While this is a strong evidence against the Chlamydophila-atherosclerosis hypothesis, and may indicate an absence of the organism from either circulation or atheromatous plaques. The lack of protective effect in these antibiotic trials has been a major setback for the C. pneumoniae-atherosclerosis hypothesis, and calls for a reappraisal of its pathological mechanisms. Four main arguments are proposed to counter these negative results. First, the organism is difficult to eradicate and refractory to current anti-Chlamydophila antibiotics. Second, most patients in antibiotic trials had advanced atheromatous lesions that had already reached an irreversible stage. Third, the bacterium might be acting by a hit-and-run mechanism, in which case secondary prevention strategies are unlikely to be beneficial. Finally, the presence of other causal factors, such as hypercholesterolemia may be essential for the organism to induce atherosclerosis, and hence this mechanism may operate in specific subpopulations. Researchers argue that negative antibiotic trials should not put a premature end to C. pneumoniae–atherosclerosis hypothesis,[719] rather these should stimulate research into newer treatment strategies targeting Chlamydophila-specific proteins and machinery directly involved in their survival, replication, and maintenance.[53]

Infectious disease etiology for atherosclerosis is an attractive hypothesis, as it can cause a paradigm shift in preventive strategies. Current etiologies have led to primary and secondary prevention strategies targeting multiple risk factors, and reinforcing positive and negative behaviors is an intensive life-long process.[107108] On the contrary, a single infectious etiology can stimulate research into vaccine development, and has the potential for better prevention and cure. In recent years, same paradigm shift has been adopted for cervical cancers. However, with multiple-candidate organisms, the most promising of these failing trials, there has never been a setback. Newer approaches to understand its pathogenesis and identifying a successful clinical application must continue.