—  SPECIALTY CONFERENCE  —

Gastrointestinal Pathology

Case 5 - Either Crohn's Colitis or a Crohn's-like Variant of Diversion Colitis

Henry Appelman
University of Michigan
Ann Arbor, Michigan


Click on each slide thumbnail image for an enlarged view
Clinical History:
4 years ago, this teenager had a terminal ileal and right colonic resection with an ileostomy for a traumatic perforation. Except for the acute perforation, the resected colon and ileum were normal. The rest of the colon from the hepatic flexure through the rectum was left in place as a diverted segment. About 3-˝ years later, occasional stool drainage from the rectum began, and this lasted for about 6 months up to the present. Recently, there has been some blood spotting mixed superimposed on the drainage. A colonoscopic exam was performed, but the endoscopist did not see any abnormalities. The whole colon and rectum were resected about 6 weeks after the colonoscopy. The reason for this resection is also not important for this presentation. In the resected specimen, there were numerous small intramural nodules scattered throughout. In addition, there were several small partial strictures due to mural thickening and some of these had small ulcers, which tended to be situated over the points where the vessels penetrate the wall. In the provided slide, there are 2 sections of the colon. One of these sections has several of the small gross nodules. The other section contains one of the strictures.


Case 5 - Figure 1 - Most of the colon looked like this. There was prominent lymphoid hyperplasia at the mucosal-submucosal junction. Some follicles contained small granulomas, and some had small foci of mucosal inflammation including tiny ulcers, in other words, aphthous lesions.
Case 5 - Figure 2 - Superimposed on the features in the first slide, there were segments of transmural inflammation with ulcers, some of which were fissure ulcers. The inflammation obliterated the submucosa, extended through the muscularis propria, involved the pericolic adipose tissue and produced strictures. Loose granulomas were present from the submucosa through the pericolic adipose tissue.


Case 5 - Figure 3 - In the submucosa of these strictures, there were numerous lymphocytic aggregates.
Case 5 - Figure 4 - In some places, numerous large loose granulomas almost filled the submucosa.

Microscopic Findings:
The gross nodules are hyperplastic lymphoid nodules at the junction of mucosa and submucosa. A few of these have small foci of inflammation in the overlying mucosa, including tiny ulcers, the aphthous distribution pattern. A few nodules contain small granulomas. The intervening, non-distorted mucosa has too many plasma cells in the lamina propria. The second section is very different. It has three ulcers, one of which extends perpendicularly into the bowel wall as a fissure. In these ulcerated areas, there is transmural inflammation that is a mixture of lymphoid aggregates, granulomas, edema and irregular proliferation of the muscularis mucosae. In the outer muscularis propria and mesentery, the granulomas tend to be loose, edematous and contain lymphocytes. Loose fibrous septa extend into the mesentery.

Diagnosis:
Diversion colitis (the lymphoid hyperplasia plus the aphthous lesions and the plasmacytosis) with superimposed transmural, chronic, granuloma-rich, ulcerating colitis, either Crohn's colitis or a Crohn's-like variant of diversion colitis.

Discussion:
Diverted segments of normal colon, usually Hartmann's pouches composed of rectum closed at the proximal end, often develop inflammation of several types that is collectively referred to as "diversion colitis." (Glotzer, Ma, Edwards) The diagnosis can only be made with confidence if the colon was non-inflamed at the time of the surgical diversion. Diversion colitis is thought to be due to deficiencies of short-chain fatty acids normally produced by colonic bacteria from ingested proteins and starches. These fatty acids are the usual sustenance of colonic epithelium. Usually, when continuity is restored, the inflammation disappears, and this response is probably the best test for this disease. One of these inflammatory variants is nodular lymphoid hyperplasia with aphthous lesions as in part of this case, a set of alterations that seems to be more common in children than adults. Another variant looks identical to ulcerative colitis with diffuse plasmacytosis in the lamina propria and architectural distortion. (Geraghty, Haque, Komorowski, Ma) Focal flat mucosa lesions, including crypt abscesses, distorted crypts, were actually the dominant feature in the earliest published series. (Glotzer) Mucosal granulomas have also been described. (Ma) Such focal inflammations, including granulomas, are common in Crohn's disease mucosal biopsies. There is a report of a transmural inflammation vaguely resembling Crohn's disease in several diverted segments where the original disease was typical ulcerative colitis, but the inflammation was purely lymphoid aggregates and lacked typical Crohn's features of lymphoid nodules hugging the outer muscularis and muscular and neural hyperplasia in the submucosa. Furthermore, this was not so easily diagnosed as diversion colitis, since the excluded segment had ulcerative colitis in the first place. (Warren-Shepherd) As best as I can tell, there do not seem to be any reports of transmural, granuloma-rich, Crohn's-like inflammation occurring in a diverted colon that was previously normal.

Although the diagnosis of diversion colitis can only be made with confidence when the diverted segment is known to have been normal at the time of diversion, it is likely that it occurs in segments that have had ulcerative colitis and Crohn's disease. However, the criteria for making such a diagnosis are not clear, since diversion changes can mimic either disease in biopsies. Several situations exist in which there is presumptive evidence occurring in patients with either UC or Crohn's include:

  1. A patient with Crohn's disease elsewhere in the gut develops inflammation that is not typical for Crohn's disease in an excluded rectum that was normal prior to diversion
  2. A patient with ulcerative colitis who has known disease without lymphoid hyperplasia in the rectum prior to diversion, develops prominent lymphoid hyperplasia in additional to the ulcerative colitis changes in the diverted rectum

How will we ever know whether this case is Crohn's colitis or Crohn's-like diversion colitis?

We have no choice except to wait and see what happens to the patient over the next 20 years or so, particularly to see if small bowel Crohn's disease develops. If it does not, then we still will not know, but we will make diversion colitis the preferred choice. Maybe a fool-proof set of antibodies or genetic markers will be developed prior to that, so we will not have to wait so long.

References

  1. Glotzer DJ, Glick ME, Goldman H: Proctitis and colitis following diversion of the fecal stream. Gastroenterology 80:438-441;1981.
  2. Ma CK, Gottlieb C, Haas PA: Diversion colitis: A clinicopathologic study of 21 cases. Hum Pathol 21:429-436;1990.
  3. Edwards CM, George B, Warren B: Diversion colitis—new light through old windows. Histopathology 34:1-5;1999.
  4. Geraghty JM, Talbot IC: Diversion colitis: histological features in the colon and rectum after defuntioning colostomy. Gut 32:1020-1023;1991.
  5. Haque S, Eisen RN, West AB: The morphologic features of diversion colitis: studies of a pediatric population with no other disease of the intestinal mucosa. Hum Pathol 24:211-219;1993.
  6. Komorowski RA: Histologic spectrum of diversion colitis. Am J Surg Pathol 14:548-554;1990.
  7. Warren BF, Shepherd NA, Bartolo DCC, et al: Pathology of the defunctioned rectum in ulcerative colitis. Gut 34:514-516;1993.