Infections May Be Causal in the Pathogenesis of Atherosclerosis
by Uffe Ravnskov1 and Kilmer S. McCully2,3
1Independent Investigator, Magle Stora Kyrkogata 9, 22350 Lund, Sweden; 2Pathology and Laboratory Medicine Service, Boston Veterans Affairs Healthcare System, West Roxbury, MA; 3Department of Pathology, Harvard Medical School, Boston MA, USA
Published by athero.org, the home of the INTERNATIONAL ATHEROSCLEROSIS SOCIETY. It is a short version of our previous paper published in Am J Med Sci; see reference 2.
According to the current view, atherosclerosis is an inflammatory disease initiated by endothelial dysfunction caused by hypercholesterolemia, hyperhomocysteinemia, or other toxic factors. Endothelial dysfunction is said to allow LDL-cholesterol and monocytes to enter the arterial wall, where LDL-cholesterol becomes oxidized and taken up by macrophages. However, there is no association between the concentration of cholesterol in the blood and the degree of endothelial dysfunction, and neither autopsy nor angiographic studies nor electron beam tomography have found an association between degree of atherosclerosis and total or LDL cholesterol . Furthermore, high cholesterol is not a risk factor for women and old people; in fact, more than twenty studies have shown that old people with high cholesterol live the longest . Several authors have pointed out that the inflammation does not start in the intima, but in the adventitia, and they have suggested that the crucial factor is a narrowing or total obstruction of vasa vasorum causing hypoxia of the arterial wall [1-8]. If so, the question is, what causes obstruction of the vasa vasorum?
Arguments for an Infectious Etiology
More than one hundred reviews have pointed to an association between infections and cardiovascular disease (CVD), and remnants of more than 50 bacterial species and several viruses have been identified in athero-sclerotic arteries, but none in normal arteries. Most authors consider it as a secondary phenomenon, but many observations suggest that the infections may indeed be causal [1,2].
Cardiovascular mortality increases during influenza epidemics and a third of patients with acute myocardial infarction or stroke have had an infect-ious disease immediately before onset . Periodontal infections are associated with an increased risk of CVD, and their treatment improves endothelial function and reduces the intima-media thickening of the arte-ries . Serological markers of infection are often elevated in patients with CVD and are also risk factors for such diseases. The coronary arteries of children who die from an infectious disease are narrowed, and their walls are thickened in those who survive . A strong argument is also that early signs of atherosclerosis have been produced by experimental infections in chickens, mice, and mini-pigs .
The classical study of early atherosclerosis in young American soldiers killed in Korea is frequently cited as proof that atherosclerosis starts in early adulthood. In that study, 77.3% had gross evidence of coronary disease and 15% had more than 50% luminal narrowing. However, such severe changes have never been observed in autopsy studies of young people who have died from other causes. The explanation may be that many of these soldiers had severe, infected wounds before they died. As the author stated, “thrombosis occurred especially in cases in which extensive trauma and shock exerted their influence.”
In a post-mortem study of several thousand victims in the concentration camp in Dachau, extensive atherosclerosis was seen in individuals younger than 35 years. Many had severe infections, and the degree of arterio-sclerosis was related to the duration of internment in the camp. Other than severe stress, there was no dietary cholesterol or saturated fat, no smoking, no lack of exercise, no obesity or other risk factors for arteriosclerosis 
Lipoproteins Are Anti-Infectious
To understand why the microorganisms are localized to the arterial wall and why their presence leads to atherosclerosis, it is necessary to consider the importance of the innate lipoprotein immune system. Although documented for decades by more than a dozen research groups, it is little known that the lipoproteins partake in the immune system by binding and inactivating all kinds of microorganisms and their toxic products. In ani-mals HDL is the predominant actor; in man it is LDL. Human LDL inac-tivates up to 90% of Staphylococcus aureus alpha-toxin, and it inac-tivates an even larger fraction of bacterial lipopolysaccharide (LPS). The import-ance of this system has been documented in laboratory studies, animal experiments, and observations in patients with CVD [1,2,9].
The presence of foam cells in the arterial wall is said to be due to uptake of oxidized LDL (OxLDL) by macrophages, but test tube experiments have shown that lipopolysaccharides from various microorganisms are able to convert macrophages to foam cells in the presence of human LDL, indi-cating that oxLDL may be created as a side effect during the oxidation of the microorganisms inside the macrophages.
The protective role of the lipoproteins has been documented in many ways. For instance, hypocholesterolemic rats injected with LPS have a markedly increased mortality compared with normal rats, but they survive if injected by purified human LDL; and hypercholesterolemic mice challenged with LPS or live bacteria have an eight-fold increase of LD50, compared with normal mice. In agreement with these experiments, low cholesterol in man is a risk factor for mortality due to infectious diseases .
Why Vasa Vasorum Becomes Obstructed
Microorganisms form complexes with lipoproteins, producing aggregates that contain microbial remnants and lipoproteins. The size of such aggregates may impede their passage through capillary networks, in par-ticular the vasa vasorum of the artery walls because of the high extra-capillary pressure. The size of the complexes may increase in the presence of hyperhomocysteinemia because homocysteine reacts with LDL to form homocysteinylated LDL aggregates. Autoantibodies against homocystein-ylated or oxLDL may also enhance the aggregation. Furthermore, the lumens of vasa vasorum are narrowed by hyperhomocysteinemia, which causes endothelial dysfunction [1,2,9].
Creation of the Vulnerable Plaque
Obstruction of the vasa vasorum by aggregated lipoprotein complexes containing microbial remnants may lead to cell death because of localized ischemia of the vascular wall. Vasa vasorum may rupture, causing hemo-rrhage and a release of the microbes and their toxic products. With a healthy immune system, the microorganisms may be eliminated, new capillaries will enter the lesion, and reparative processes will convert the dead tissue into a stable, fibrous plaque. But in case of an insufficient clearing of the microorganisms and the ensuing inflammatory response, cell death may accelerate creating a vulnerable plaque, the preferential site for rupture and occluding thrombosis [1,2].
Clinical and Pathological Observations
Thus, in our view LDL-cholesterol does not enter the artery through the endothelium, but via the capillary web of the vasa vasorum. Oxidation of LDL does not take place before LDL has entered the macrophage but occurs after phagocytosis, as part of the process of inactivating micro-organisms by oxidation with reactive oxygen species. A reason for con-sidering the vulnerable plaque to be a type of micro-abscess, as originally suggested by William Osler 100 years ago, is that its temperature is higher than that of the surrounding tissue. Whereas neutrophils, the hallmark of pyogenic infections, are rare in stable plaques, they are always found in and around the core of vulnerable plaques, and there are just as many neutro-phils in the intact as in the ruptured plaques, contradicting the assumption that their presence is secondary to rupture .
Our interpretation explains the clinical and laboratory similarities between myocardial infarction and myocarditis. It also explains the frequent occur-rence of bacteremia and sepsis in myocardial infarction complicated with cardiogenic shock. It explains why fever, diaphoresis, leukocytosis, and elevation of inflammatory markers in the blood, including CRP, the classi-cal symptoms of an infectious disease, are common findings in myocardial infarctions. Our interpretation agrees with the almost constant finding of neutrophils in the myocardium 24 hours after an acute myocardial infarc-tion, as well as in infarctions of other organs caused by thrombosis secondary to rupture of vulnerable plaques.
Fatty streaks are not necessarily the precursors of atherosclerotic plaques because they are present in the fetus and are more frequently found in early than late childhood, presumably reflecting a normal and reversible response to infections. Hydrodynamic pressure is usually cited as the reason that atherosclerosis is localized only within systemic arteries. This explanation is probably correct, not because the arterial pressure damages the endothelium, but because the lipoprotein complexes are trapped more easily in vasa vasorum of the systemic arteries where the tissue pressure is much higher than around the veins and the pulmonary arteries. The focal occurrence of atherosclerotic lesions is also in better accordance with a microbial genesis. If elevated LDL cholesterol were the most important cause, atherosclerosis should be a more generalized disease. That inflam-mation should be the cause of atherosclerosis is also unlikely, because all trials with anti-inflammatory drugs have increased CVD mortality. This fact is in accord with an infectious origin of atherosclerosis, however, be-cause inflammation is a necessary step for healing of infections.
Our interpretation is in accord with several of the classical risk factors for CVD. For instance, infections are more prevalent in smokers and diabetics. Hyperhomocysteinemia is found in B vitamin deficiency, smoking, hypertension, hypothyroidism, renal failure, mental stress and aging, all classical risk factors for CVD. Mental stress also stimulates the production of cortisol, and an excess of cortisol promotes infections. The suggestion that excess iron is a risk factor for vascular disease is also in accordance, because bacterial growth is stimulated by the presence of free iron.
1, Ravnskov U, McCully KS. 2009. Vulnerable plaque formation from obstruction of vasa vasorum by homocysteinylated and oxidized lipoprotein aggregates complexed with microbial remnants and LDL autoantibodies. Ann Clin Lab Sci 39: 3-16.
2. Ravnskov U, McCully KS. 2012. Infections may be causal in the pathogenesis of atherosclerosis. Am J Med Sci 344: 391-94.
3. Martin JF, Booth RFG, Moncada S. 1990. Arterial wall hypoxia following hyperfusion through vasa vasorum is an initial lesion in atherosclerosis. Eur J Clin Invest 20: 588-92.
4. Barker SGE, Talbert A, Cottam S, Baskerville PA, Martin JF. 1993. Arterial intimal hyperplasia after occlusion of the adventitial vasa vasorum in the pig. Arterioscl Thromb 13: 70-77.
5. Simanonok JP. 1996. Non-ischaemic hypoxia of the arterial wall is a primary cause of atherosclerosis. Med Hypotheses 46: 155-61.
6. Maiellaro K, Taylor WR. 2007. The role of the adventitia in vascular inflammation. Cardiovasc Res 75: 640-48.
7. Ritman EL, Lerman A. 2007. The dynamic vasa vasorum. Cardiovasc Res 75: 649-58.
8. Mulligan-Kehoe MJ. 2010. The vasa vasorum in diseased and nondiseased arteries. Am J Physiol Heart Circ Physiol 298: H295-H305.
9. Han R. 2010. Plasma lipoproteins are important components of the immune system. Microbiol Immunol 54: 246-53.