Article Text

This article has a correction. Please see:


Microorganisms in the aetiology of atherosclerosis
  1. S A Morré1,
  2. W Stooker2,
  3. W K Lagrand3,
  4. A J C van den Brule1,
  5. H W M Niessen1
  1. 1Department of Pathology, University Hospital Vrije Universiteit, PO Box 7057, De Boelelaan 1117, 1007 MB Amsterdam, The Netherlands
  2. 2Department of Cardiac Surgery, University Hospital Vrije Universiteit
  3. 3Department of Cardiology, University Hospital Vrije Universiteit
  1. Dr Niessen email: jwm.niessen{at}


Recent publications have suggested that infective pathogens might play an important role in the pathogenesis of atherosclerosis. This review focuses on these microorganisms in the process of atherosclerosis. The results of in vitro studies, animal studies, tissue studies, and serological studies will be summarised, followed by an overall conclusion concerning the strength of the association of the microorganism with the pathogenesis of atherosclerosis. The role of the bacteria Chlamydia pneumoniae and Helicobacter pylori, and the viruses human immunodeficiency virus, coxsackie B virus, cytomegalovirus, Epstein-Barr virus, herpes simplex virus, and measles virus will be discussed.

  • atherosclerosis
  • Chlamydia pneumoniae
  • Helicobacter pylori

Statistics from

The pathogenesis of atherosclerosis is a subject of much debate, but it is now generally thought to be a chronic inflammatory disease.1 The initiation and progression of atherosclerosis seems to be related to the location and extent of the inflammatory reactions.2–4 Recent publications suggest that infective pathogens also play an important role in the process of atherosclerosis. Thus, local infection could either react directly with the vessel wall or indirectly through initiation of immunological responses.5

In this review we will discuss the role of microorganisms (bacteria and viruses) in the process of atherosclerosis.J Clin Pathol 2000;53:647–654


Chronic dental bacterial infections are strongly associated with atherosclerosis.6–8 Although the chronological sequence of infection and the initiation of atherosclerosis is still not clear, it can be argued that atherosclerosis is the cause and not the consequence of dental infections because individuals with coronary atherosclerosis might also have compromised arterial circulation in the peridontium. However, a recent study by Adams et al strongly argues against this explanation.9

Nevertheless, Meier et al have shown retrospectively that previous use of tetracyclines and quinolones was associated with a lower risk of acute myocardial infarction, providing indirect evidence that infection with microorganisms susceptible to tetracycline antibiotics might be involved in the aetiology of ischaemic heart disease.10

We will focus on the role of bacteria with the supposed strongest association with atherosclerosis, namely Chlamydia pneumoniae and Helicobacter pylori.


In vitro studies

In vitro experiments have shown that C pneumoniae induces human macrophage or foam cell formation, a key event in early atheroma development, via chlamydial lipopolysaccharide, suggesting a role for C pneumoniae in atherogenesis.11,12 Furthermore, it has been found that endothelial cells, smooth muscle cells, and macrophages are capable of supporting the growth of C pneumoniae.13 In addition, an association between C pneumoniae infection and a specific immune response has been suggested.14 Studies in men with coronary heart disease suggest that C pneumoniae possibly contributes to the process of coronary atherosclerosis by chlamydia specific, cell mediated responses, predominantly induced by antigenic structures that are similar among different species of chlamydia.14 Indeed, a coronary strain, C pneumoniae A-03, has been isolated and shown to stimulate production of monocyte chemotactic protein 1, interleukin 8 (IL-8), and soluble intercellular adhesion molecule 1 in vitro.15 In surgical specimens of human carotid atherosclerotic plaques, induction of macrophage functions by C pneumoniae was also found.16 Interestingly, percutaneous transluminal coronary angioplasty also induces stimulation of the humoral immune response against C pneumoniae and supports the idea that plaque disruption during angioplasty might expose hidden C pneumoniae antigens to the immune system.17

These humoral immune reactions might be related to bacterial heat shock proteins (HSPs), such as chlamydia HSP60, which may play an important role in the process of vascular endothelial injury, a key event in the pathogenesis of atherosclerosis.18

Animal studies

In vivo experiments in rabbits showed that intranasal C pneumoniae infection accelerates intimal thickening (which could be inhibited by azithromycin) and inflammatory atherosclerosis like changes in the aorta.19,20 In addition, intranasal inoculations of C pneumoniae in ApoE deficient transgenic mice and C57BL/6J mice (these mice develop atherosclerosis on an atherogenic diet) resulted in C pneumoniae localisation in atheroma, suggesting a tropism of C pneumoniae to the lesion.21 In a murine model system, both C pneumoniae strains, AR39 and MoPn, were detected in the aorta of mice infected with the corresponding strain. However, only mice infected with AR39 had enhanced atherosclerotic lesions, suggesting that this C pneumoniae strain might possess a unique atherosclerotic property.22 Thus, these animal studies strengthen the aetiological link between C pneumoniae and atherosclerosis.

Serological studies

Serological studies and the detection of C pneumoniae in atheromatous lesions were the first indications of an association between C pneumoniae and atherosclerosis.23 Seropositivity for C pneumoniae was associated with an increased risk for future cardiovascular disease, namely stroke,24 carotid wall thickening,25 and coronary heart disease.26. Chronic C pneumoniae infection seemed to be associated with a serum lipid profile considered to increase the risk of atherosclerosis, supporting the hypothesis that infections do play an (indirect) role in the pathogenesis of atherosclerosis.27 In addition, C pneumoniae DNA can be detected in circulating white blood cells by the polymerase chain reaction (PCR).28,29 Using this technique, an association between coronary heart disease and circulating C pneumoniae DNA in men, but not in women, was found.30

However, in a large scale study, Ridker et al found no evidence of an association between C pneumoniae IgG seropositivity and risk for future myocardial infarction.31 Furthermore, Altman et al found that the presence of IgG antibodies to C pneumoniae in the serum is not predictive of acute arterial complications.32 Therefore, seroepidemiological studies are contradictionary with respect to the role of C pneumoniae in atherosclerosis.

Tissue studies

By means of immunohistochemistry, PCR, and nested PCR, C pneumoniae was detected in carotid endarterectomy specimens, within atherosclerotic plaques, and in endothelial cells, macrophages, and smooth muscle cells, suggesting a direct role of C pneumoniae in carotid artery atherosclerosis.33–36 Using PCR and immunohistochemistry, C pneumoniae was detected in arterial biopsies from femoral, popliteal, and coronary arteries, as well as in the aorta, indicating that the organism is widespread in atherosclerosis of the vascular system.37–42. Bartels et al even found that occluded aorta-coronary venous grafts harbour C pneumoniae (but not cytomegalovirus).43

Others have demonstrated viable C pneumoniae by means of cell culture in coronary atheromas, carotid endarterectomy specimens, and abdominal aneurysm.15,34,44–46 Chlamydia pneumoniae has also been demonstrated in many tissues by electron microscopy, supporting the true evidence of C pneumoniae particles.34,45–48

In contrast, in another study of carotid endarterectomy specimens, bacterial and viral cultures in plaques were negative.49 An Autralian study using PCR in postmortem material did not detect C pneumoniae in coronary arteries or carotid endarterectomy specimens.50 In addition, Lindholt et al could not detect C pneumoniae in symptomatic aneurysms of the abdominal aorta. Therefore, they suggested that these aneurysms and atherosclerosis might be two different disease entities.51

Finally, Andreasen et al could not detect C pneumoniae in calcific or degenerative atherosclerotic aortic heart valve disease.52 However, Nystromrosander et al did detect C pneumoniae in aortic valves using electron microscopy.48


Pilot studies have shown that antibiotic treatment (azithromycin and roxithromycin) improves the clinical outcome in patients with myocardial infarction and acute non-Q wave coronary syndromes.53,54 Recently, the final ROXIS study showed that roxithromycin appeared to extend the clinical benefit of preventing death and re-infarction for at least six months after initial treatment.55

In contrast, Anderson et al have shown that in patients with coronary heart disease, positive for C pneumoniae antibodies, global tests for C reactive protein (CRP), IL-1, IL-6, and tumour necrosis factor α (TNF-α) improved at six months with azithromycin. However, no differences in antibody titres and clinical events were seen.56 Thus, controlled trials are needed to establish the therapeutic role of antibiotics in peripheral arterial disease.57


The Koch-Henle criteria for proof of the aetiology of C pneumoniae infection in atherosclerosis seem to be largely fulfilled: correlation of atherosclerosis with antibodies against C pneumoniae; detection of C pneumoniae in atheromas with different techniques; international studies with macrolides in coronary heart disease were successful; target cells of atherosclerosis can be infected with C pneumoniae in vitro; and animal experiments have been positive.58 However, several studies reported that no C pneumoniae could be detected in atherosclerotic lesions. It is not known whether this is related to technical problems, the specimens that were studied, or geographical differences. Furthermore, it is unclear whether C pneumoniae initiates the process of atherosclerosis, facilitates progression of existing plaques, or merely colonises the lesions.59


In vitro study

In an in vitro study it has been shown that polyunsaturated fatty acids inhibit the growth of H pylori and prevent/arrest atherosclerosis.60

Serum studies

Laurila et al found that serum triglyceride and total cholesterol concentrations were significantly higher in men with IgG and IgA against H pylori than in men with no signs of infection, supporting the hypothesis that chronic infection with H pylori might modify the serum lipid profile in a way that increases the risk of atherosclerosis.61

In duplex ultrasound studies that established carotid atheroma load in combination with serum H pylori measurements, it was concluded that chronic H pylori infection is an independent risk factor for ischaemic cerebrovascular disease and may act, at least in part, by increasing atherosclerosis.62 Another study suggested that it is possible that exposure to H pylori and other microorganisms leads to an increased risk of clinically manifest coronary artery disease by an autoimmune process (anti-HSP 60/65 antibodies), in patients submitted for routine angiography.63,64

In contrast, in other prospective case cohort designs among middle aged men and women, H pylori seropositivity was not associated with coronary heart disease, or with increased mean intima-media thickness of the carotid artery, suggesting that H pylori is probably not an important contributor to clinical coronary heart disease events.65,66

Tissue study

In patients with aortic aneurysms, seropositivity for H pylori was found. However, PCR for H pylori was negative, making the possibility of a direct involvement of H pylori in atherosclerotic aortic aneurysms less tenable.38


To date, there is no convincing evidence supporting the presence of H pylori within atherosclerotic plaques, and seroepidemiological evidence is contradictionary.


Pesonen et al investigated 175 children (0–15 years of age) who had undergone necropsy and found that infections in general and viral infections in particular seem to be associated with intimal thickening, which might predispose coronary arteries to atherosclerosis.67 We will focus on the role of human immunodeficiency virus, coxsackie B virus, cytomegalovirus, Epstein-Barr virus, herpes simplex virus, and measles virus in atherosclerosis.


Serological studies

Constans et al have shown that, although no clinically relevant atherosclerotic lesions were found, plaques occurred more often in patients with HIV than in controls.68 During postmortem examination of eight HIV seropositive male patients, Paton et al found major atherosclerosis in coronary arteries in the absence of an associated cardiovascular risk factor.

They concluded that viral infection, either HIV or coexisting herpesviruses, played a role in the development of the coronary lesions.69


It is difficult to establish whether HIV itself, or an opportunistic pathogen, or both, are causally related to the process of atherosclerosis.


Animal study

Recent murine model work support the idea that CBV, a member of the enterovirus genus, and immune cells cooperate and play a role in arterial lipid accumulation, possibly acting as initiating factors for atherosclerosis.70

Serological studies

One of the most recent studies on CBV showed that high concentrations of enterovirus specific antibodies were associated with a risk of myocardial infarction in men.71 However, an association between antibodies to CBV and myocardial infarction has not been found in all studies.72,73


Further studies are needed to evaluate whether, and by which mechanism, enterovirus infections are involved in the pathogenesis of atherosclerosis and the development of myocardial infarction.


In vitro studies

In vitro studies have shown that CMV infection can potentially lead to infection of blood vessel endothelium, thereby inducing damage to endothelium, and infection of smooth muscle cells, suggesting a role in atherosclerosis.74–78 Indeed, in cultures of smooth muscle cells of carotid artery plaques, CMV (but not herpes simplex virus 1 or 2) was detected by means of immunohistochemistry in 25% of cells.79 However, no replicating CMV was found by electron microscopy, suggesting that the artery walls might be a site of CMV latency79.

Animal studies

Inoculation of BALB/c mice with CMV resulted in the development of atherosclerosis; that is, immune injury and high low density lipoprotein cholesterol.80 In rats, it was found that an active CMV infection of arterial smooth muscle cells can be established in vivo. Interestingly, neointimal smooth muscle cells were far more susceptible to CMV infection than medial smooth muscle cells.81

Serological studies

In a case control study and the ARIC (atherosclerosis risk in communities) study, an association was found between the serum CMV antibody titre and carotid intimal-medial thickness, which is in line with a causal role for CMV in atherosclerosis.82,83 CMV infection has also been identified as an independent risk factor in re-stenosis after coronary angioplasty.84,85 In men undergoing vascular surgery for atherosclerosis, the prevalence of CMV antibodies was higher in the surgical group than in the control group (not significant for herpes simplex virus 1 or 2), suggesting that periodically activated virus might have a role in the pathogenesis of atherosclerosis.86 In addition, serum CMV antibody titres were higher in patients with atherosclerosis and diabetes than in patients without diabetes, suggesting that CMV might play a role in the development of clinical atherosclerosis in patients with diabetes mellitus.87

In contrast, in serological studies of immune responses to CMV, combined with the correlation of angiographically assessed atherosclerosis, it was concluded that multiple reactivation of latent viruses might be a consequence rather than a cause of atherosclerosis.88 Furthermore, another study showed that prior infection with CMV is not a major risk factor for angiographically demonstrated primary coronary artery atherosclerosis.89 In addition, Tiran et al found that anti-CMV positivity is not a major risk factor at the time of disease manifestation, implying that CMV does not play an important role in the pathogenesis of atherosclerosis.90

Tissue studies

Using PCR, 90% of the samples obtained from patients with grade III atherosclerosis were shown to contain CMV viral nucleic acids, compared with 53% of patients with grade I atherosclerosis, indicating a role for this virus in the pathogenesis of atherosclerosis.91 However, CMV was also found (by means of in situ hybridisation and immunohistochemistry) in abdominal aortas, femoral, and coronary arteries in atherosclerotic as well as control material, suggesting that the human arterial wall might be a site of latency of this virus.92 Other findings suggest that the persistent expression of CMV immediate early genes in the vessel wall might play a role in the vascular cellular responses, including progression of atherosclerosis or vasculitis in vivo.93


Heart transplant recipients who are immunosuppressed and who are also actively infected with CMV are prone to develop accelerated atherosclerosis in the transplanted organ,94 with more frequent rejection.95 In serological tests after cardiac transplantation CMV infection was found in 77% of patients, suggesting a relation between CMV infection and rapidly progressive coronary atherosclerosis after cardiac transplantation.96 Another study of patients with cardiac transplants found serological evidence of past C pneumoniae and CMV infection. However, C pneumoniae does not appear to have an independent role or synergistic relation to CMV in the development of transplant associated atherosclerosis.97


In vitro and immunohistochemical studies have provided mainly circumstantial evidence for the involvement of CMV in human atherosclerosis, whereas seroepidemiologic studies suggest that a periodically activated latent infection of CMV is present in patients with atherosclerosis.98 In general, the conclusions of seroepidemiological studies are contradictionary.


Serological studies

A study of antibodies against EBV in sera, combined with angiography, suggested that reactivation of latent EBV is a consequence rather than a causal factor of atherosclerosis.99

Tissue study

Using immunohistochemistry and PCR, EBV could not be found in early and advanced atheromatous changes in coronary arteries, or in lesion free areas in the thoracic aorta of young trauma victims.100


An association between atherosclerosis and a previous EBV infection was demonstrated in heart transplants.101


Although seroepidemiological studies implicate a role of EBV in atherosclerosis, to date, EBV has not been found in atherosclerotic lesions.


In vitro studies

An in vitro study in HUVEC cells (cultured primary endothelial cells from the human umbilical vein) indicated that latent infection of vascular cells with HSV might play a pathogenic role in the development of human atherosclerosis.102 Another study suggested that the efficiency of HSV-1 infection of smooth muscle cells derived from adult rats is greater than of those from pup rats. This could be important because age is a risk factor for the development of atherosclerosis.103 It has also been shown that HSV infection leads to lipid accumulation in vascular cells. Furthermore, HSV infection of endothelial cells attracts leucocytes, with subsequent inflammatory damage. HSV also induces procoagulant changes on endothelium, with increased thrombin generation and platelet adhesion. These changes are more or less characteristic for atherosclerosis.104–108

Animal study

It has been shown that HSV causes atherosclerosis in experimental animals.94,109 These authors suggest that lipid depositions in atherosclerosis may result in part from virus induced changes in the cholesterol metabolism of host cells, as was found in cultured smooth muscle cells.

Tissue studies

Histological studies and in situ DNA hybridisation in young trauma victims have shown HSV associated with areas showing early and advanced atheromatous changes in coronary arteries and with lesion free as well as lesional areas in the thoracic aorta. This supports the concept that HSV might play a direct or indirect role in the pathogenesis of human atherosclerosis.100 In coronary biopsies obtained during coronary artery bypass graft surgery, immunohistochemistry and situ hybridisation demonstrated HSV-2 in 45% of biopsies and also one case of HSV-1.110


In humans, clinical, epidemiological, and molecular biology studies implicate a relation between herpesviruses and atherosclerosis.94,111 However, it also has been hypothesised that mechanical abrasion might reactivate latent HSV (and CMV) infection in endothelial cells, particularly those exposed to high shearing forces—for example, at vessel bifurcations. This mechanism might be responsible for the endothelial damage, clotting, and atheroma formation often found at these sites.112


In vitro study

An in vitro study in isolated endothelium/smooth muscle cells indicated that measles virus infection might be a risk factor for atherosclerosis, by means of damaging endothelial cells and initiating proliferation of smooth muscle cells.113


The role of measles virus has not been studied extensively and, to date, no in vivo data are available.


Microorganisms as aetiological agents might provide new insights into some unexplained aspects of atherosclerosis. A higher incidence of coronary heart disease in young men coincides with the remarkable androtropism of bacterial diseases, whereas the low incidence of coronary artery disease in France could be explained by a much higher use of antichlamydial antibiotics. However, the low incidence of atherosclerosis in the tropics, despite a high frequency of chlamydial infection, is difficult to explain.58

It has also been suggested that low grade infections might be one of the causes of the inflammatory reaction observed in atherosclerotic lesions and acute ischaemic syndromes, reflected in raised concentrations of CRP,109,114 suggesting that the rise in CRP is not a risk factor, but a sign of an active chronic infection. However, immunohistochemical studies of atherosclerotic lesions suggested that CRP might promote atherosclerosis locally by activating the complement system and inducing foam cell formation.115 This contradicts the idea that the rise in CRP merely reflects baseline inflammation.

Indeed, seroepidemiological data suggest an association between pathogens and clinical events related to atherosclerosis (table 1). However, they cannot distinguish between a causal relation and secondary infection. Furthermore, both serological and epidemiological data are contradictionary in different studies (table 2).

Table 1

Studies supporting the aetiological link between microorganisms and atherosclerosis

Table 2

Studies contradicting the aetiological link between microorganisms and atherosclerosis

Some tissue studies have found antigens, genetic material, or cultivatable infectious agents in association with the inflammatory lesions. However, it is unclear whether these agents initiate arterial lesions or exacerbate those lesions already present. It is also unknown whether infectious agents in atherosclerotic lesions are directly pathogenic or act through immune responses to microbial antigens, which crossreact with normal human antigens. Recently, an interesting study has shown that a peptide from the murine heart muscle specific α myosin heavy chain has sequence homology to the 60 kDa, cysteine rich, outer membrane of C pneumoniae, C psittaci, and C trachomatis, leading to the production of autoantibodies to heart muscle specific epitopes in mice.116 This suggests that chlamydia mediated heart disease is induced by antigenic mimicry of a heart muscle specific protein.

Overall, C pneumoniae is the agent with the most evidence for a causal association with atherosclerosis (provided by seroepidemiological, pathological, and animal models, and in vitro studies). Although Koch's postulates seem to be fullfilled for C pneumoniae infection and atherosclerosis, definitive proof that a particular microorganism causes atherosclerosis may not come from this direction, but from the prevention of primary infection by vaccination or the eradication of the agent by antimicrobials. However, if successful, there is a danger that an increase in prescription of these antibiotics could result in an increase in resistence to antibiotics.

In animals, polyclonal immunoglobulin preparations inhibited atherosclerosis via modulation of T cell activity and/or antibody production. Therefore, immunomodulation might be another effective way to prevent and/or treat atherosclerosis.117 In addition, studies of the prophylactic use of antiviral agents, such as ganciclovir or CMV vaccine, especially in patients at high risk of developing atherosclerosis (such as heart transplant patients), will allow an alternative prevention strategy for coronary heart disease.

If antibiotic, antiviral, and/or immunomodulatory agents do appear to attenuate the atherosclerotic process, the public health implications will be enormous.

In conclusion, arguments for and against the role of infection in atherosclerosis have appeared with equal regularity. However, for some of these infectious agents (CMV, but especially C pneumoniae), evidence for their role in atherosclerosis seems relatively strong. Nevertheless, the mechanism by which they induce their pathological effect is still unclear. A direct effect would have important implications for this widespread disease because adequate antibiotic treatment is possible. If the effect is indirect—for example, by the process of molecular mimicry, as has been postulated recently, innovative ways of exploring the pathogenesis of human disease in general will be needed.


This work was supported by the Netherlands Heart Foundation, grant no 93–119. Dr Niessen is a recipient of the Dr E Dekker programme of the Netherlands Heart Foundation.


View Abstract

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Linked Articles

  • BMJ Publishing Group Ltd and Association of Clinical Pathologists