Appendix redux

The endless “genetics or environment” debate can get rather convoluted, whether about inflammatory bowel disease (IBD) or any other complex disease.¹ With regard to IBD, a series of epidemiological and genetic “breakthroughs” have barely inched us closer to clarifying this issue.² Indeed, many decades of research on Crohn’s disease and ulcerative colitis have so far revealed just one principal gene, at the NOD2 locus on chromosome 16,³ and one major environmental factor, smoking,⁴ in influencing susceptibility to either of these conditions.

Currently, the “Crohn’s disease gene” is the hottest topic but environmental issues keep pushing themselves into the picture. The “protective” effects of smoking in ulcerative colitis have been old news for a while,⁵ with its deleterious effects in Crohn’s disease only more recently coming to the fore.⁶ But all of the observations on smoking over the past 20 years have not really helped us very much, despite intriguing hypotheses about pathogenesis⁷ and treatment.⁸

Now a newer factor, the role of the appendix, is attracting increasing interest.⁹ Actually, this factor is not particularly “new” at all, since Gilat et al called attention to it as a possible “childhood factor” in 1987,¹⁰ while innumerable clinical and pathological observations about the appendix in IBD were being published throughout at least the 20 years before.¹¹ None the less, it was not until the mid-1990s that intensive investigations began focusing quite sharply on the associations between prior appendicectomy and the incidence of IBD—or, more specifically, ulcerative colitis.

Both of the studies on this subject in the current issue of Gut^12,13 fall thoroughly into step with all previous studies suggesting a “protective” effect of prior appendicectomy against the incidence of ulcerative colitis [see pages 803 and 808]. Each of these two papers in its own way however sheds new light on the topic. Both are meticulous in the application of multivariate analyses to dissect out confounding effects such as age, sex, and smoking, although the French study includesd many more factors in its proportional hazards regression model. Both are equally careful to correct for the bias of appendicectomy at the time of diagnosis.

The Australian study is broader in scope in that it includes Crohn’s disease patients and population controls while the French investigators concentrate exclusively on ulcerative colitis patients; but both series are very large (over 600 cases) and both comprise ambulatory as well as hospitalised cases. Another welcome feature of both papers is that they use clear and appropriate definitions of disease characteristics and severity, including endoscopic data, therapeutic requirements, and rates of surgery. Only the French study calculates the proportion of years in which disease was active as a measure of severity, and only the Australian study incorporates histology in defining disease extent, as well as specifically excluding surgery for neoplasia as a severity measure; but overall both papers employ very reasonable criteria.

So with all these excellent qualities to their credit, just where is the “new light” that these two papers purportedly shed? I would cite two areas of particular illumination. Firstly, from the Australian group we learn for the first time that a large and well controlled study shows a “protective” effect of appendicectomy in Crohn’s disease as well as in ulcerative colitis. This observation leaves smoking as still virtually the only epidemiological factor that works in opposite directions in these two diseases. Secondly, from both of these papers, a combined sample size of 62 ulcerative colitis patients with prior appendicectomy provides confirmation of an earlier suggestion, based on only 21 such cases, that the course of ulcerative colitis seems milder following a history of appendicectomy.¹⁴

But what issues remain in the dark despite these two important new papers? Very simply, what remains obscure is the meaning of it all. The French and Australian authors are quite candid in admitting that their data can be interpreted in two different ways. Either the appendix itself exerts a truly “protective” effect against IBD by virtue of immunological mechanisms that both papers discuss thoroughly; or else young people who develop appendicitis or mesenteric adenitis are somehow physiologically, genetically, or immunologically distinct from the population that is predisposed to IBD. A recent article from Sweden seems to suggest that it is the appendicitis more than the appendicectomy per se that is protective.¹⁵ Radford-Smith et al however tilt slightly in favour of the “appendicectomy” hypothesis while fully acknowledging the validity of the “appendicitis” theory. Cosnes et al lean even further in the same direction, going almost overboard in their recommendation for prophylactic appendicectomy among “patients (sic) genetically at high risk of developing ulcerative colitis.”

In either event, the statistical observations reported in this issue of Gut are certainly consistent with each other as well as with previous studies, and they undoubtedly offer tantalising clues to the pathogenesis of IBD. But Tantalus never did get to eat or drink the food and water surrounding him, and it seems we too are going to have to wait a lot longer before satisfying our own hunger and thirst for understanding everything about IBD.