The debate by Professor Quirke (Gut 2001;49:757–60) was an interesting review of the hypothesis of a microbiological aetiology of Crohn’s disease. He indicates that “the hypothesis remains controversial and unproved.”
The point is that proof is never absolute, and indeed the objective of research is to disprove the hypothesis rather than to prove it, the latter being an impossible objective and scientifically flawed. He goes on to mention that “for the infectious disease hypothesis to be proved for any organism, Koch’s postulates need to be fulfilled.” This is not correct. Proof is pragmatic not absolute, and in practice it is the fulfilment of a set of predetermined objectives. Koch’s postulates are but an example of this, the Euclidian principle of quod erat demonstrandum, and an extremely important development of scientific philosophy of the 19th century. Koch himself however recognised the weakness of his postulates in that although he felt that cholera was microbiological in causation, he was unable to apply his postulates to it.
It is important to review Koch’s postulates and they are as follows:
(1) “The specific organism should be shown to be present in all cases of animals suffering from a specific disease but should not be found in healthy animals”
This postulate demands a high level of sensitivity of laboratory methods and the clinicopathological identification of a specific disease—can Crohn’s disease be classified as such? At the time of Koch the important concept of a commensal microbe was not developed.
(2) “The specific microorganism should be isolated from the diseased animal and grown in pure culture on artificial laboratory media”
This demands laboratory methods which have not always been achieved at the present time.
(3) “This freshly isolated microorganism, when inoculated into a healthy laboratory animal, should cause the same disease seen in the original animal”
Animal models are not always available for postulated microbial disease and it is recognised that transgenic transmission might cause a different disease.
(4) “The microorganism should be re-isolated in pure culture from the experimental infection”
Once again, laboratory cultures are not always possible at present.
Koch was a great scientist and he recognised so well that there was more to microbiological explanation for disease than his postulates, which have a very high level of specificity but a very low level of sensitivity. I am afraid that microbiological science must look beyond Koch’s postulates for its “proofs” and we must rethink the concepts of proof for the newly recognised microbiological diseases in the present century.
In that there is no such thing as absolute proof, science works in paradigms. These are models which come to be accepted as the best explanation of the phenomena that we see around us. However, there is no real paradigm for Crohn’s disease. I remember farfetched “psychosomatic” concepts in the 1960s, to be replaced by food allergy during the latter part of the 20th century. But these paradigms have fallen into disfavour, being replaced by a lack of any clear idea of the type of disease that we are dealing with in respect of Crohn’s disease. Genetic factors are probably an influence only on susceptibility but not causation.
But in the treatment of patients clinical doctors require a paradigm on which to base explanations and understanding, treatment, and research. The nearest we have to a paradigm of causation of Crohn’s disease is that it is “inflammatory”, and so conforming to the allopathic principle of contraria contrariis curantur we give anti-inflammatory medications. The clinical manifestations of the disease are a direct result of the inflammatory process but is this a protective mechanism in itself? The assumption that it is only damaging led to the paradigm of “autoimmunity”, a concept which itself lacks “proof”, and indeed the criteria of proof in putative autoimmune disease have never been defined.
A paradigm of a microbiological causation of Crohn’s disease must be based on two factors. The first of these is the statistical association between the disease and a putative microbe but the difficulty of this is the lack of robust detection methods for identification. The second is plausibility. It is interesting to reflect that for many years acute hepatitis was accepted as being a viral disease and indeed became known as “viral hepatitis”, well before the viruses had been identified. The plausibility was clear even though microbiological science had not progressed so far to identify the viruses themselves. In respect of Crohn’s disease, we need to continue to think as to whether it is plausible that the disease might be microbiological, even in the absence of a definite microbe. The development of the paradigm and identification of a specific microbe are different scientific processes.
Plausibility is founded on existing knowledge and models, based mainly on epidemiology and pathology, the main foundations of Western clinical medicine. What therefore do we think of the pathology of Crohn’s disease? Firstly, it is clear that Crohn’s disease is not a homogenous pattern of disease but a variety of different patterns of inflammatory disease of the intestinal tract. The hallmark of Crohn’s disease is firstly a patchy inflammation of the gastrointestinal tract, including perioral and perianal areas of skin. Granulomas are another hallmark and fissuring a third. We can go on in this way but the more criteria that we add, the more it would appear that the disease is a heterogynous group. In other words we cannot define Crohn’s disease, we do not really know what it is, and so a concept of causation is going to be based on a very fragile foundation.
However, we can make progress, especially if we look at the “classical” type of Crohn’s disease involving the right side of the colon, the caecum, and the terminal ileum, with fissuring and granulomatous disease. This type of disease looks very much like tuberculosis, so much so that if it presents in an Asian patient the disease is usually called tuberculosis whereas if it presents in a non-Asian patient it is usually called Crohn’s disease. If the similarities to tuberculosis are so powerful, then clearly causation is likely to be very similar. A further important feature is the epidemiological observation of family clustering across genetic boundaries, the husband/wife associations which point very much towards a transmissible agent. Finally, there are the parallels with Johne’s disease in animals which continue to be suggestive of Crohn’s disease being an equivalent in the human. In terms of response to antimicrobial compounds, do we feel that some studies suggesting benefit are more, less, or equally important to those that suggest no benefit? It depends on the attractiveness of the microbiological paradigm to the individual—some people are anxious to find a cause for Crohn’s disease whereas others see no practical advantage of this and are happy to remain without a paradigm other than “inflammatory bowel disease”. Response to one or more given antibiotics cannot be laid down as a criterion of proof of microbiological causation but could help strengthen a paradigm.
What we need to do in respect of Crohn’s disease is consider which is the most plausible hypothesis and then continue to test it, in this case with microbiological scientific efforts, the importance of which Professor Quirke emphasises. As with every other paradigm in science, it must be under continual review and we must always be prepared to reconsider our perceptions of causation. Although we can always be wrong, and indeed we often are, to be totally sceptical denies the opportunities for scientific progress. Research must be based on hypothesis and paradigm.