Inflammatory responses and coronary heart diseaseBMJ 1998; 316 doi: https://doi.org/10.1136/bmj.316.7136.953 (Published 28 March 1998) Cite this as: BMJ 1998;316:953
The “dirty chicken” hypothesis of cardiovascular risk factors
- Michael A Mendall, Consultant gastroenterologist and senior lecturer
The “dirty chicken” hypothesis was proposed by Solomons to explain why children reared in poverty, though appearing healthy and receiving adequate nutrition, end up as short adults.1 Based on the observation that antibiotic supplementation reverses poor growth in chickens reared in overcrowded unhygienic conditions, he suggested that chronic subclinical infection induces a low grade systemic inflammation and that this produces a qualitatively similar effect to full blown acute inflammation—that is, chronic anorexia and increased basal metabolic rate, with cytokines being the mediators. What does this have to do with humans reared in relatively overcrowded unhygienic conditions and cardiovascular disease?
There is an increasing interest in the relation between chronic low grade systemic inflammation, as indicated by serum levels of C reactive protein, and mortality from coronary heart disease. 2 3 There has, however, been little knowledge of the determinants of this response and its importance in the pathogenesis of atherosclerosis. Chronic subclinical infection with Chlamydia pneumoniae, Helicobacter pylori, chronic bronchitis, and chronic dental sepsis have been associated with raised values of C reactive protein within the normal range3 and have been implicated as risk factors for coronary heart disease. Non-infective conventional environmental risk factors also associated with low grade acute phase responses include age, low adult social class, smoking, obesity, and childhood social class (a possible mechanism for the association of short stature and coronary heart disease).3
If the dirty chicken hypothesis is true—that is, that qualitatively similar effects are observed during chronic low grade systemic inflammation (occurring in all of us) as in severe acute inflammation—many biological risk factors should be associated with raised serum C reactive protein values in normal subjects. This is indeed the case: raised serum C reactive protein values are associated with raised serum fibrinogen, plasminogen, factor VIII, white blood cell count, fasting insulin, and serum triglyceride values; depressed high density lipoprotein-cholesterol; and raised fasting blood sugar concentrations. 3 4 (The latter cast light on the pathogenesis of non-insulin dependent diabetes.) These associations are not diminished by controlling for body mass index. A common underlying mechanism such as inflammation may explain why different types of cardiovascular risk factors cluster in the same subject—for example, in syndrome X. It might also explain why many environmental cardiovascular risk factors produce changes in several different biological risk factors—for example, smoking or obesity. Nevertheless, atherosclerosis is clearly a multifactorial condition, since not all contributory factors show a clear relation to inflammation—for example, low density lipoprotein cholesterol and hypertension.
We have recently extended these observations on inflammation. Interleukin 6 and tumour necrosis factor α play a key part in regulating the acute phase response by the liver. They also affect lipid metabolism in vivo. Raised serum concentrations of both have similar associations to those observed with serum C reactive protein and were linked to chronic coronary heart disease.5
Inflammatory type reactions and, particularly, cytokines may not deal only with the body's response to tissue damage or environmental stress. Body mass index is correlated with serum concentrations of tumour necrosis factor α, which is consistent with increased synthesis of tumour necrosis factor mRNA by adipocytes from obese subjects.6 Oestrogen has inhibitory effects on interleukin 6 synthesis and on levels of cardiovascular risk factors, perhaps through this mechanism. Alcohol consumption is associated with diminished serum concentrations of tumour necrosis factor α,5 and polyunsaturated fatty acids inhibit cytokine synthesis. Hence levels of inflammation may respond to metabolic change and be influenced by various dietary factors.
But what relation does systemic inflammation generated in response to environmental or metabolic change bear to the risk of coronary heart disease? Cytokines and activated white blood cells originating in the lungs or gut in response to environmental stress could influence the process through effects on conventional risk factors such as fibrinogen. In addition, tumour necrosis factor α and interleukin 6 generated at these sites could have direct effects which promote atherosclerosis and thrombosis at distant sites.7 Alternatively, inflammation may be principally located at the site of the atherosclerotic lesion, being directly influenced by environmental factors that can reach that location, such as smoking, alcohol, diet, and C pneumoniae, with the systemic inflammatory response being an epiphenomenon of this process. Obesity, H pylori infection, and chronic bronchitis cannot act directly at the site of atherosclerotic lesions, supporting the notion that distant inflammation may be important. Whatever the balance of effects between locally and distantly generated cytokines, agents which can influence inflammatory processes are likely to have important therapeutic effects in atherosclerosis, as has recently been suggested for aspirin.2
These observations provide new insights into how environment can influence the risk of atherosclerosis and reduce growth in children. Inflammation and inflammatory cytokines play a fundamental role in the whole body response to environmental stress (infective and non-infective) and metabolic change. These mechanisms are likely to be continuously active, but more so in some who die sooner from coronary heart disease. The dirty chicken hypothesis has the pleasing property of unifying many previously disparate observations about the clustering of cardiovascular risk factors and also of identifying new risk factors such as chronic bronchitis and dental disease. It suggests simple ways in which a whole set of environmental stressors—infective agents—can be treated to reduce risk of coronary heart disease, as well as providing a mechanism for the association of poverty with coronary heart disease. The growth of dirty chickens is augmented by antibiotics, and preliminary studies suggest that inflammatory responses8 and coronary events 9 10 after myocardial infarction are reduced by antibiotic administration.
MM is supported by the British Heart Foundation.