Introduction to the Selby Study

by Dr. William Chamberlin

I share the following with my peers knowing that the following straight forward assessment will, or at least should, make all of us feel somewhat uncomfortable. I believe that Humble Pie is good for our health.

I was recently contacted by a physician with over 25 years of experience in designing and interpreting clinical trials who wanted to find out more about the role of mycobacteria in chronic inflammatory bowel disease. This doctor read the pro and con arguments in the review by the American Academy of Microbiology and we shared some recent unpublished evidence supporting the role of M. paratuberculosis (MAP) as one of the causal agents of the Crohn’s syndrome. Most of the arguments against MAP being a causal agent have been disproven over the past few years. The strongest remaining argument are the negative claims made by the authors of the Australian/Selby Antibiotic Trial. The negative claims were strongly disputed at the time and published by those of us who support the theory that MAP is a causal agent of Crohn’s disease. (See the below links.) In fact, we looked at data interpreted through an “intention-to-treat” analysis and argued that the data showed that properly chosen antibiotics were beneficial in treating Crohn’s disease although they were not curative. All of the pertinent articles were sent to this scientist who analyzed the data and the arguments pro and con.

The following comments are quite interesting and should be reviewed by any physician remotely interested in advancing our understanding of the etiology, pathogenesis and treatment of Crohn’s disease. The pitfalls of interpreting clinical trials by those of us only superficially introduced to the process are apparent – and this statement applies to the overwhelming majority of clinicians. It applies to the academic centers training future gastroenterologists. Both sides can learn from this analysis.

The conclusion is that there were major flaws in the design of the trial. Patients were adequately randomized and the data good only for the first 16 weeks. No conclusions one way or the other should be made. Unfortunately, 99% of gastroenterologists have incorrectly heard that this study proves that Mycobacterium paratuberculosis does not cause Crohn’s disease and that antibiotics have no benefit. The advice is to ignore the Australian trial and learn from its mistakes.

This section includes Selby’s original article, responses to the article refuting his conclusions and the opinion now offered that criticizes claims made by both sides of the issue. I believe it is helpful to lay out the articles and interpretations in one place so that readers can judge for themselves.

I remain grateful for the effort put forward by Dr. Selby and those running the trial. Their efforts definitely moved the ball forward and for that we should all be thankful.