Most investigators will concede that Crohn's disease is unlikely a single disease entity and that, in all probability, Crohn's disease represents a syndrome with multiple etiologies. By definition, Crohn's disease is an etiopathic disease, i.e., it has no known etiology. Hence, if the etiology of some cases of Crohn's disease is discovered and defined (e.g., M. paratuberculosis-associated), by definition that disease is no longer Crohn's disease. Thus, if the association of M. paratuberculosis and Crohn's disease proves to be a causal relationship, those cases cannot, by definition, be Crohn's disease. The proper perspective is that some cases of M. paratuberculosis intestinal infection are being misdiagnosed as Crohn's disease.

This perspective is not trivial as it serves to circumvent and "rise above" the entire controversy. To state that Crohn's disease is caused by M. paratuberculosis is therefore likely erroneous and is, in large part, the driving force behind the controversy.

Cases of idiopathic bowel disease (including Crohn's Disease) are diagnosed subjectively by individual clinicians and pathologists; there is no specific test for the disease. Because it is a subjective diagnosis, there exists inherent error. In fact, there is a known 20% diagnostic error in Crohn's disease. Although the study of Crohn's Disease in general has no option other than to base their research on this subjectively defined patient population, this is not the case at hand where a specific etiologic agent is being investigated.

If you were to conduct a study on tomatoes, you would not include all gardens in that study, but just those gardens that were growing tomatoes. The comparison is certainly relevant and the inclusion of all cases subjectively defined as Crohn's disease in a study of M. paratuberculosis-associated infection is no different than including all gardens in the study of tomatoes. The generation of objective data from a subjective starting point cannot yeild objective end-point data.

It is an undisputed fact that M. paratuberculosis can be detected in low numbers in some cases of Crohn's disease and this finding is a marker and objective "diagnostic" tool for investigative study. However, no one has ever determined the reproducibility of those findings. If patient A is positive for M. paratuberculosis today, will patient A be positive tomorrow? This is a critical question that not only addresses causality but defines the ability to conduct research on a defined and objective study population.

While readily easy to determine scientifically, there are patient and cost concerns to achieving this goal. A patient would need to undergo consecutive intestinal biopsies, which is not a pleasant procedure, and obtaining patient consent would be problematic. In addition, the endoscopic procedures and biopsies would be performed solely for research purposes, not part of routine diagnostics, and the costs ensued by such would need to be paid by the research; not an inexpensive endeavor. Nevertheless, it is a critical question that must be addressed.

This is an open question and a MAJOR problem in the entire hypothesis. Most bacterial infections are associated with large numbers of organisms and the whole premise of M. paratuberculosis-associated intestinal disease is that low numbers of this agent are the cause. It is a major and valid criticism since there are no good examples on which to draw a comparison (excluding toxin-producing bacteria). It is a threshold that must be overcome in establishing causality.

In 1998, the National Institute of Allergy and Infectious Disease (NIAID)expressed concern related to the findings of M. paratuberculosis in Crohn's disease in that, in the history of science/medicine, no organism has ever been searched for with the same intensity as M. paratuberculosis has been searched for in Crohn's disease. As such, there exist no comparative basis to interpret the results. Yet, 10-years later, nobody has addressed this issue.

What would one find if the same search intensity was applied to other organisms? Without this basis of comparison, the significance of detecting M. paratuberculosis in Crohn's disease will remain unknown. After all, low numbers of infectious agents are found in normal and diseased tissues and the morphologic features with antigen trapping can account for the increased prevalence.

Without the ability to experimentally infect humans or other comparable animal model (Johne's disease in animals is not an appropriate or comparative model), establishing causality is problematic. A host of experiments could be conducted to suggest a role of M. paratuberculosis in Crohn's disease, but such findings would not distinguish between a causal or secondary role. We need to draw on the past and learn from other diseases; for example, Helicobacter pylori and Tropheryma whippeli (the causative agent of Whipple's disease).

Both of these diseases, among others, have had their etiology defined and confirmed by demonstrating a correlation between clinical/pathologic disease improvement with the progressive clearance of the causative agent. This approach can be applied to M. paratuberculosis and Crohn's disease.

By the above methods, an objective patient population can be defined and the role of M. paratuberculosis in Crohn's disease elucidated. However, the entire premise of this solution rests on the assumption that the detection of M. paratuberculosis is reproducible. If it is not, then the role of M. paratuberculosis maybe an incidental finding.