Scott IS, Sheaff M, Coumbe A, Feakins RM, Rampton DS. (Department of Morbid Anatomy and Gastrointestinal Re- search Unit, St. Bartholomew's and the Royal ...
272 SELECTED SUMMARIES
GASTROENTEROLOGY Vol. 117, No. 1
in nonulcer dyspepsia as being clinically insignificant but then proceeds to promote the efficacy of the other possible treatments for this common condition. However, critical analysis of the literature indicates that the evidence for a beneficial effect of these agents is extremely weak and that the magnitude of any effect over placebo is certainly no greater than that shown with H. pylori eradication treatment. A recent large study (1248 patients) comparing omeprazole, 20 mg/ day, vs. placebo found that resolution of symptoms occurred in 10% more of the actively treated group (95% CI, 4%–16%) (Aliment Pharmacol Ther 1998;12:1055–1056), a magnitude of benefit identical to that with H. pylori eradication treatment. The major study that Dr. Schubert references for the benefit of fedotozine indicates a benefit of active over placebo again of only about 10%. The one original study he quotes in support of the efficacy of cisapride did not include a placebo control group. The only other reference given by Dr. Schubert to support the efficacy of prokinetic agents is to a meta-analysis from Sanofi Winthrop Ltd., the manufacturers of the prokinetic agent motilium, and its conclusion is not supported by independent reviews or results of large methodologically sound studies (Am J Gastroenterol 1996;32:535–540, Can J Gastroenterol 1997;11:127–134, Am J Gastroenterol 1998;93:368–374). The current scientific literature therefore indicates that the magnitude of the symptomatic benefit from eradicating H. pylori infection in nonulcer dyspepsia patients is equivalent to that achieved by any other available treatment. One important advantage of H. pylori eradication therapy is that it produces a long-term cure with a single 1-week course of medication. In light of the current evidence, it seems more appropriate that patients with nonulcer dyspepsia and H. pylori infection should be offered H. pylori eradication therapy as first-line therapy and then consider these noncurative therapies for those with troublesome persisting symptoms. KENNETH E. L. McCOLL, M.D. LILIAN S. MURRAY, Ph.D. EMAD M. EL–OMAR, M.D.
THE APPENDIX IN ULCERATIVE COLITIS: A NOT SO INNOCENT BYSTANDER Scott IS, Sheaff M, Coumbe A, Feakins RM, Rampton DS (Department of Morbid Anatomy and Gastrointestinal Research Unit, St. Bartholomew’s and the Royal London Hospital School of Medicine and Dentistry, Queen Mary and Westfield College, University of London, London, England). Appendiceal inflammation in ulcerative colitis. Histopathology 1998;33: 168–173. Recently, there has been a resurgence in the interest of the role of the appendix in inflammatory bowel disease (IBD). Embryologically, the appendix arises from the cecum. Therefore, it would be anticipated that a disease state that causes colonic inflammation would also involve the appendix. Scott et al. report the results of a retrospective case-controlled study that compared the prevalence and histological features of appendiceal inflammation in cases with IBD compared with a control group of patients with acute appendicitis or colonic carcinoma. Cases consisted of colectomy specimens obtained from the Royal London Hospital between the years 1980 and
1994. Only cases in which the appendix was available for histological review were included in the analysis. Twenty-four of 50 (48%) cases with ulcerative colitis (UC) had evidence of appendiceal inflammation; this was significantly higher than cases of colonic carcinoma (8%) but no different than cases of Crohn’s disease (CD). The inflammatory changes seen within these appendixes were more characteristic of the underlying IBD than changes seen in acute appendicitis. Appendiceal inflammation with cecal sparing was seen in 9 of 24 UC specimens (37%) and 2 of 9 (22%) CD specimens. There had been a previous appendectomy in 3% of UC cases compared with 8% of carcinoma cases and 21% of CD cases. The difference between the number of appendectomies previously done in UC compared with CD patients was statistically significant (P ⬍ 0.01). Based on these results, the authors conclude that appendiceal inflammation commonly occurs as a skip lesion in UC and CD that resembles the colonic disease. In addition, they state that the low prevalence of appendectomy in the UC cases supports the hypothesis that the appendix may have a key role in the pathogenesis of UC. Comment. The appendix has long been considered an evolutionary misfit; a redundant organ that served little purpose in humans. The results from this study raise several issues regarding the importance of the appendix in IBD. Recent studies in patients with UC have implicated the appendix as (1) being a protective or causative factor in UC, (2) playing a central immunologic role in the pathogenesis of UC, and (3) showing the heterogeneity of the disease we call UC. The importance of the appendix in UC is reviewed. First, is the appendix a protective or causative factor in UC? This study claims that the prevalence of previous appendectomy in their specimens is lower than that of the control group. In 1987, Gilat et al. were the first to publish the finding of an inverse relationship between appendectomy and UC (Scand J Gastroenterol 1987;22:1009–1024). In 1994, Rutgeerts et al. revisited this issue in a case-controlled study, which found a 60-fold lower risk of UC in patients who had a previous appendectomy (Gastroenterology 1994;106:1251–1253). Several investigators have confirmed this observation in different populationbased studies (Lancet 1994;343:766–767, J Clin Gastroenterol 1995; 21:283–288, Gut 1995;36[Suppl 1]:A25, Am J Gastroenterol 1996; 91:723–725, Gastroenterology 1997;113:377–382); however, the relationship was not as striking as that shown by the Leuven study. Others have argued that factors that predispose to UC may diminish the risk of the development of acute appendicitis (Inflamm Bowel Dis 1996;2:217–221). The same authors cite that the rate of appendectomy has steadily declined but the incidence of UC has remained stable. Overall, the studies would support a lower incidence of appendectomy in UC compared with various control populations. Whether this means that appendectomy is protective against the development of UC is another matter. A negative correlation with a disease does not in itself imply causation. A plausible biological correlate would strengthen this argument. The origin and pathogenesis of IBD remain an enigma. The most accepted hypothesis is that the disease represents a complex interaction between genetic and environmental influences. Could the appendix somehow be involved in the pathogenesis of UC, and if so, how? The function of the appendix is unknown, but it is a highly vascular organ that represents an important part of gut-associated lymphoid tissue system, together with Peyer’s patches and tonsils. The mucosal lymphoid tissues of the appendix are predominantly com-
July 1999
posed of B cells and CD4 T-helper cells (Gut 1986;27:667–674, Immunology 1987;60:19–28). The appendix is most likely involved in antigen sampling from the lumen. This antigen sampling may regulate the immunologic response in more distal mucosa ( J Clin Invest 1990;86:1255–1260). In UC, the balance between T-helper cells and T-suppressor cells is shifted toward the T-helper cells. Rutgeerts et al. speculated that removing the appendix (a T-helper organ) may tip the balance in favor of T-suppressor cells and therefore protect against the development of UC (Gastroenterology 1994;106: 1251–1253). Recently, this theory has been supported in a murine model of IBD ( J Exp Med 1996;184:707–715). The authors of this study examined the role of the appendix in T-cell receptor ␣ mutant mice who spontaneously develop an IBD similar to human UC. They found that the lymphoid tissue of the appendix was similar to that of Peyer’s patches in these animals. The cellular proliferation in the appendix was 2-fold greater than that seen in Peyer’s patches. Interestingly, removal of the appendix at a young age in these mutant mice suppressed the development of IBD. This line of evidence suggests a central role of the appendix in antigen sampling, processing, and immunologic signaling. Traditionally, UC is considered a disease of mucosal inflammation that begins usually in the rectum and extends proximally in a continuous fashion. There is usually an upper limit of inflammation that can be recognized endoscopically. This upper limit does not necessarily correspond to the transition zone between normal and inflamed mucosa histologically. The finding of rectal sparing or segmental inflammation (endoscopically or histologically) raises the question of CD. Because the appendix is derived embryologically from the cecum, it stands to reason that the appendix would be involved in ulcerative pancolitis involving the cecum. In early reports the appendix was involved in approximately half of the cases that came to colectomy for pancolitis (Gastroenterology 1958;34:381). It was claimed that inflammation of the appendix was always involved in continuity with cecal inflammation (Dis Colon Rectum 1976;19: 345). The present report showed that 37% of cases with a histologically normal cecum had evidence of appendiceal inflammation. Cohen et al. first described the skip lesion in 1974 as a single case report (Am J Gastroenterol 1974;62:151–155). Since the initial description, many investigators have recognized and described this entity in retrospective case series (Histopathology 1990;16:93–95 and 1990;17: 286, Mod Pathol 1992;5:607–610 and 1994;7:322–325). Patchy cecal inflammation has also been described at the time of endoscopy in patents with left-sided UC (Am J Gastroenterol 1997;92:1275–1279 and 1998;93:2405–2410). Gastroenterologists who perform colonoscopy on UC patients have noted this phenomenon of cecal patch for many years in the region of the appendix (personal communication). This endoscopic appearance may represent cecal spillover of an inflammatory response that is occurring in the appendix analogous to the inflammation seen in the cecum of patients with terminal ileal CD or the inflammation of backwash ileitis. The description by the present authors that the inflammation seen in the appendix was similar to the underlying IBD supports that UC can occur as a skip lesion as well as support the increasing belief that what we call UC and CD represents a heterogeneous group of disorders. In conclusion, recent studies have shown that the appendix is involved in UC either as a continuous or skip lesion. Animal models suggest that it may play a role in the pathogenesis of UC. This combined with the data showing that the incidence of appendectomy is lower in UC patients make it difficult to believe that this organ is an innocent bystander. The controversy regarding the role of the appendix will undoubtedly continue. Further investigation into the
SELECTED SUMMARIES 273
exact role of the appendix in the pathogenesis, immunology, and possible therapy of UC is warranted. REMO PANACCIONE, M.D. WILLIAM J. SANDBORN, M.D.
IN PURSUIT OF THE HIDDEN, THE OCCULT, AND THE OBSCURE Chak A, Koehler MK, Sundaram SN, Cooper GS, Canto MI, Sivak MV (Division of Gastroenterology, University Hospitals of Cleveland, Cleveland, Ohio). Diagnostic and therapeutic impact of push enteroscopy: analysis of factors associated with positive findings. Gastrointest Endosc 1998;47:18–22. In a retrospective study, Chak et al. reviewed the clinical impact of push enteroscopy in a group of 164 patients (70 men and 94 women; mean age, 61 years) during a 2-year period. The indications for enteroscopy included anemia caused by occult gastrointestinal bleeding (n ⫽ 65), overt gastrointestinal bleeding (n ⫽ 64), diarrhea or steatorrhea of suspected small bowel origin (n ⫽ 20), and suspected small bowel mucosal disease (n ⫽ 15). In all patients evaluated for occult or overt gastrointestinal bleeding, enteroscopy was performed after upper endoscopy (esophagogastroduodenoscopy [EGD]) and colonoscopy. An overtube was used in 127 patients with advancement and positioning of the enteroscope performed under fluoroscopic guidance. In 92 of 164 (56%) patients, a potentially responsible disorder was diagnosed by small bowel enteroscopy. Of the abnormalities detected, approximately 60% were located in the jejunum and the remainder in the proximal gastrointestinal tract. Proximal lesions were within reach of a standard gastroscope and presumably overlooked during the prior endoscopy. Missed or overlooked lesions included Cameron erosions associated with large hiatal hernias (n ⫽ 10), peptic ulcer disease (n ⫽ 10), and upper gastrointestinal tract vascular ectasias (n ⫽ 9). Eleven patients had abnormalities detected in both the upper gastrointestinal tract and jejunum. The most common jejunal findings included vascular ectasias and neoplasms. Other less common findings included bacterial overgrowth, sprue, mucosal ulcerations, infection, and lymphoma. In the subgroup of 129 patients with occult or overt gastrointestinal bleeding who were age 65 years or older, there was a significantly greater incidence of vascular ectasias than in patients younger than 65 years (40% vs. 12%, respectively; P ⬍ 0.001). Jejunal vascular ectasias were detected significantly more frequently in patients with a history of vascular ectasias on prior EGD or colonoscopy (34% vs. 15%, respectively; P ⬍ 0.01). Vascular ectasias were also more common in patients with chronic renal failure (54% vs. 27%; P ⬍ 0.05). The authors reported no significant difference in the detection of vascular ectasias when the group with overt bleeding was compared with patients with occult gastrointestinal bleeding. Patients with overt gastrointestinal bleeding using aspirin or nonsteroidal anti-inflammatory drugs (NSAIDs) were more likely to have upper tract lesions than those reporting no
