Primary adenocarcinoma of a Brooke ileostomy is a rare finding. The authors report one case of a primary adenocarcinoma and one case of high-grade dysplasia arising in a Brooke ileostomy, both occurring decades after total proctocolectomy for presumed ulcerative colitis and familial adenomatous polyposis, respectively. Both patients were treated successfully. Current recommendations for surveillance and treatment include yearly inspection of any ostomy older than 15 years and wide local excision for invasive disease.

The Brooke ileostomy was introduced in 1951 and quickly became the gold standard for treating patients who required a total proctocolectomy for familial adenomatous polyposis or ulcerative colitis. In 1969, Sigler and Jedd reported the first case of primary adenocarcinoma arising in a Brooke ileostomy.¹ A review article published in 2002 identified 35 cases of such tumors, and a report in 2005 by Quah and associates identified 44 cases.^2,3 In the review of 35 patients, 25 had ulcerative colitis, 8 had familial adenomatous polyposis, and 2 had Crohn's disease.³ The average time from initial operation to presentation with invasive cancer was 26 years, with a range from 3 to 48 years. These neoplasms usually present as an exophytic growth at the mucocutaneous junction of the ileostomy and are associated with bleeding, excoriation, irritation, and pain.^3-8 Often a low index of suspicion results in considerable delays in diagnosis, ranging from several months to as many as 6 years.⁷ This report adds two cases of ileostomy tumors to the literature and provides a current review of the clinical features, pathology, and treatment of this unusual cancer.

Case report

Case 1—A 62-year-old white woman was referred to a colorectal surgeon for pain and irritation at her ileostomy site. She had undergone a total proctocolectomy with a Brooke ileostomy in 1955. Although the diagnosis at the time of her procedure 48 years earlier was unclear, she had reported diarrhea, abdominal cramps, and weight loss. Her family history was significant for a sister who underwent the same procedure for ulcerative colitis.

Upon presenting to our institution, the patient described a rash around her ileostomy that had persisted for several months. She reported no diffuse abdominal pain, cramps, weight loss, or other symptoms. She was otherwise in good health and reported no other significant medical problems.

On physical examination, a well-formed ileostomy in the right lower quadrant with a slightly raised area around the inferior 2/3 of the ostomy was observed. Palpation of the area around the ostomy revealed a small hernia and an area of firmness associated with the visibly raised skin that extended into the subcutaneous tissues. Biopsies obtained from this firm area revealed invasive adenocarcinoma of colonic origin extending from the mucocutaneous border of the ileostomy into the surrounding dermis and subcutaneous tissue.

A complete metastatic workup was performed before surgical intervention. Her complete blood count, liver function tests, carcinoembryonic antigen, and other routine laboratory tests were all normal. A chest radiograph and computed tomography (CT) scans of the abdomen and pelvis revealed no evidence of metastatic disease. An esophagogastroduodenoscopy and terminal ileoscopy also failed to reveal any additional disease or polyposis.

After the workup was completed the patient was taken to the operating room. A wide local excision was performed with a 2-cm margin around the ileostomy. The excised specimen included skin, subcutaneous tissue, muscle, and peritoneum. Along with the ileostomy, a 10-cm segment of distal small bowel was removed. Frozen-section analysis demonstrated no malignancy at the margin of resection. A new ileostomy was placed in the left lower quadrant, and the fascial defect from the original ostomy site was closed in two layers.

The patient's only postoperative complication was a superficial wound infection, which resolved after the wound was opened and packed. Final pathology revealed a well-differentiated adenocarcinoma at the ileostomy site arising from the mucocutaneous junction and invading the surrounding skin and subcutaneous tissues. The tumor measured 2.5 x 2 cm, and the immunohistochemical profile was consistent with colonic adenocarcinoma.

After evaluating the patient and reviewing the final pathology report, the oncologist was concerned that the patient was at risk for local recurrence and distant metastatic disease but admitted that the risk of recurrence was difficult to estimate and the impact of chemotherapy was unclear due to the rarity of the condition. Adjuvant therapy with leucovorin-modulated 5-fluorouracil triplet, administered 5 days per month for 6 months, was attempted as an outpatient therapy, but the patient declined further treatment after developing severe nausea. Currently, the patient is being followed-up with physical examinations and serial CT and positron emission tomography scans. She has no evidence of recurrent disease to date, more than 4 years later.

Case 2—A 44-year-old white woman was referred to the same colorectal surgeon because of an excess growth of tissue around her ileostomy. She had undergone total proctocolectomy and Brooke ileostomy in 1984 for familial adenomatous polyposis. In 2000, the patient presented with irritation at her ileostomy site and increasing difficultly using the stoma appliance. Although her symptoms had been worsening over several years, she reported no pain, obstructive symptoms, or significant bleeding.

On physical examination, there was a circumferential growth of hypertrophic granulation tissue with evidence of irritation and bleeding. Stenosis of the ileostomy at the skin level precluded digital examination. She was scheduled for an elective revision of her ileostomy.

A local excision of the parastomal granulation tissue was attempted but the amount of tissue and the presence of persistent stenosis necessitated a more extensive revision. Initially, the mucocutaneous junction was excised circumferentially, but the stenosis extended deeper into the bowel. The distal 3 cm of the ileum was removed, and a new ileostomy was created in a Brooke fashion at the same site. The patient recovered well and was discharged home without any immediate postoperative complications.

A

B

Figure—Low-power photomicrograph with hematoxylin and eosin staining showing the adenomatous polyp on a fibrovascular stalk (A). Note the squamous epithelium in the bottom left corner of the image. High-power view of the adenomatous polyp showing high-grade dysplasia (B). Note the presence of nuclear crowding and loss of cellular polarity (nucleus of cell is not lined up like a picket fence along the basement membrane).

Final pathology revealed an adenomatous polyp with the presence of fairly extensive high-grade dysplasia of intestinal mucosa at the mucocutaneous junction. No definite stromal invasion was seen, and the margins of resection were free of dysplastic changes (Figure).

Several months later, the patient underwent an esophagogastroduodenoscopy and terminal ileoscopy by her gastroenterologist, which revealed a few small hyperplastic gastric polyps. Fourteen months after her revision, she developed recurrent granulation tissue at the ileostomy site, which was treated with local excision. The pathologic analysis of this tissue revealed vascular congestion and prominent lymphoid aggregates without dysplasia or invasive carcinoma. Currently, the patient is doing well and is being followed-up annually by the colorectal surgeon and the gastroenterologist.

Discussion

Primary adenocarcinoma arising from an ileostomy is a rare finding. Only 44 cases have been reported previously, and a histologic review of these cases found features consistent with colonic adenocarcinoma.² Colonic metaplasia and high-grade dysplasia also have been reported to occur in the mucosa adjacent to an invasive adenocarcinoma.⁸

Inflammatory cap polyps, which are almost always benign, develop as a result of trauma caused by repeated stomal prolapse.⁹ These polyps can be single or multiple and may occur anywhere on or in the ileostomy. These lesions result from mechanical trauma and not an underlying progressive disorder and can occur any time after ileostomy construction. Similar polyps have been described in relation to prolapsing hemorrhoids or colostomy, rectal prolapse, and solitary rectal ulcer syndrome. Gross examination cannot differentiate inflammatory cap polyps from malignant polyps, but microscopic evaluation of inflammatory cap polyps reveals granulation tissue with no evidence of malignancy.

Neoplastic changes at the ileostomy site almost always arise at the mucocutaneous border and range from simple polyps to invasive masses. Several decades usually pass between creation of the ileostomy and the diagnosis of cancer. Adenocarcinoma of a Brooke ileostomy is thought to proceed in the following sequence: normal mucosa to metaplasia to dysplasia to invasive adenocarcinoma. Histology almost always reveals well-differentiated mucinous colonic adenocarcinoma. Invasive adenocarcinoma typically develops in a background of well-described colonic metaplasia. In the surrounding mucosa, changes such as the absence of villi, the development of tubular glands lined by goblet cells, and the change in the mucin profile from sialomucins to sulfomucins, exemplify the development of metaplastic epithelium that occurs in the small bowel mucosa exposed to the external environment.^8,10 Precancerous dysplastic changes have been noted in the setting of colonic metaplasia, supporting this described sequence of progression to invasive adenocarcinoma.^8,9

In contrast, Starke and associates discuss the possibility that ileal carcinoma would have developed in these cases regardless of the creation of an ileostomy. They theorize that ileostomy carcinoma represents the random development of carcinoma of the terminal ileum. This theory does not seem plausible when examining statistics. The incidence of small bowel malignancies is 0.7 per 100,000 patients, and the terminal ileum is the least common site of occurrence, whereas the incidence of ileostomy carcinoma is much higher, with approximately 2 to 4 cases occurring per 1,000 ileostomies.⁷

The theory regarding sequential progression from normal mucosa to cancer is not clearly understood. Gilson and colleagues, as well as Stark and associates, propose that exposure of small bowel mucosa to an abnormal external environment results in chronic inflammation and subsequent metaplastic changes.^5,7 In almost all ileostomies, chronic inflammatory changes, such as inflammatory polyps and leukocytic infiltrates, have been attributed to both the external environment and mechanical trauma associated with stomal appliances.^4,7,9 Additionally, ileostomies are known to harbor bacterial flora more consistent with the colon than the terminal ileum.¹¹ These findings support the contention that environmental changes contribute to the observed mucosal changes.

The contribution of the underlying disease process to the interval development of adenocarcinoma must be considered. Ileostomy adenocarcinoma has been diagnosed in patients with both familial adenomatous polyposis and ulcerative colitis. Patients with familial adenomatous polyposis are known to exhibit polyps throughout the gastrointestinal tract, and the malignant degeneration of a preexisting adenoma could account for the interval development of cancer at the ostomy. Others contend that the synergistic combination of colonic metaplasia and underlying familial adenomatous polyposis predisposes a patient to the interval formation of neoplastic polyps and masses.⁷ Cases of adenocarcinoma of the terminal ileum in patients with ulcerative colitis have been reported but most of these have been attributed to severe backwash ileitis. Ulcerative colitis predisposes patients to dysplasia and adenocarcinoma of the colon, but not of the small bowel. Apparently, none of the reported patients who developed adenocarcinoma of the ostomy exhibited backwash ileitis before proctocolectomy⁸ Most likely, a combination of underlying disease and altered environment results in the formation of colonic metaplasia and eventual invasive adenocarcinoma in the mucosa of the terminal ileum over many years.

The development of adenocarcinoma of an ileostomy likely represents the end point of a prolonged pathophysiologic process that begins with the creation of the ileostomy. The incidence appears to be increasing, with the majority of reported cases having arisen over the past two decades. Gadacz and colleagues note that the increase likely reflects the completion of latent biologic processes that begin with construction of the ileostomy, proceed through stages of metaplasia, and end with invasive adenocarcinoma.⁸ They surmise that the incidence will continue to rise as patients with Brooke ileostomies age.

The Brooke ileostomy, introduced in 1951, was largely replaced by the Koch continent ileostomy in the 1970s. More recently, the ileal pouch anal anastomosis has become the procedure of choice in treating patients with familial adenomatous polyposis and ulcerative colitis. If the same pathophysiologic process that leads to adenocarcinoma in a Brooke ileostomy applies to patients who undergo either Koch ileostomy or ileal pouch anal anastomosis, there is a risk of finding cancer in a Koch pouch or an ileal pouch. Similar chronic inflammatory changes, microbial alteration, and colonic metaplasia have been identified in ileal pouchs.^7,12 Furthermore, adenocarcinoma of a Koch continent ileostomy has been reported.¹² Two cases of cancer of ileal pouches also have been reported but these are thought to have resulted from retained rectal mucosa.¹²

Patients who have undergone total proctocolectomy with ileostomy formation should undergo annual examinations of their ileostomies and biopsies of any suspicious lesions. Quah and colleagues recommend starting surveillance 15 years after the creation of the ileostomy² Any invasive cancer should be treated with wide local excision and creation of a new ostomy. The need for adjuvant therapy is not well established and will likely continue to be administered on a case-by-case basis. The rarity of these cases and the lack of long-term follow-up preclude accurate evaluation of the role of adjuvant therapy or the prediction of long-term survival. The prognosis appears variable. Both patients reported in this article continue to do well, but 5 of the 35 patients reported by Iizuka and associates died. One died of unresectable disease, three died from eventual widespread metastatic disease; the fifth patient succumbed to bleeding complications.³

Conclusion

The development of adenocarcinoma in a Brooke ileostomy after total proctocolectomy is a rare finding. Our two cases highlight the biologic changes that occur in the externally placed small bowel mucosa over a period of many years. Review of the literature supports a prevailing theory that small bowel mucosa of individuals predisposed to development of adenocarcinoma may undergo a gradual sequence of changes beginning with colonic metaplasia and eventually progressing to dysplasia and invasive adenocarcinoma. This sequence has been fairly well established in patients with a Brooke ileostomy. The risk of cancer in patients with either a Koch pouch or an ileal pouch is not as clear, but the identification of similar premalignant mucosal alterations in these patients suggests that the same biological principles apply. It is postulated that the incidence of adenocarcinoma arising in either a Koch pouch or an ileal pouch will increase as these patients continue to age, clearly demonstrating the need for continued surveillance of existing Brooke ileostomies and future surveillance of both Koch pouches and ileal anal pouches.

References

1. Sigler L, Jedd FL. Adenocarcinoma of the ileostomy occurring after colectomy for ulcerative colitis: report of a case. Dis Colon Rectum. 1969;12(1):45-48.
2. Quah HM, Samad A, Maw A. Ileostomy carcinomas a review: the latent risk after colectomy for ulcerative colitis and familial adenomatous polyposis. Colorectal Dis. 2005;7(6):538-544.
3. Iizuka T, Sawada T, Hayakawa K, et al. Successful local excision of ileostomy adenocarcinoma after colectomy for familial adenomatous polyposis: report of a case. Surg Today. 2002;32(7): 638-641.
4. Carey PD, Suvarna SK, Baloch KG, et al. Primary adenocarcinoma in an ileostomy: a late complication of surgery for ulcerative colitis. Surgery. 1993;113(6):712-715.
5. Gilson TP, Sollenberger LL. Adenocarcinoma of an ileostomy in a patient with familial adenomatous polyposis. Report of a case. Dis Colon Rectum. 1992;35(3):261-265.
6. Johnson JA 3rd, Talton DS, Poole GV Adenocarcinoma of a Brooke ileostomy for adenomatous polyposis coli. Am J Gastroenterol. 1993;88(7):1122-1124.
7. Starke J, Rodriguez-Bigas M, Marshall W, et al. Primary adenocarcinoma arising in an ileostomy. Surgery. 1993;114(1):125-128.
8. Gadacz TR, McFadden DW, Gabrielson EW, et al. Adenocarcinoma of the ileostomy: the latent risk of cancer after colectomy for ulcerative colitis and familial polyposis. Surgery. 1990;107(6):698-703.
9. Attanoos R, Billings PJ, Hughes LE, et al. Ileostomy polyps, adenomas, and adenocarcinomas. Gut. 1995;37(6):840-844.
10. Berman JJ, Ullah A. Colonic metaplasia of ileostomies. Biological significance for ulcerative colitis patients following total colectomy. Am J Surg Pathol. 1989;13(11):955-960.
11. Gorbach SL, Nahas L, Weinstein L, et al. Studies of intestinal microflora. IV The microflora of ileostomy effluent: a unique microbial ecology. Gastroenterology. 1967;53(6):874-880.
12. Cox CL, Butts DR, Roberts MP, et al. Development of invasive adenocarcinoma in a long-standing Koch continent ileostomy: report of a case. Dis Colon Rectum. 1997;40(4):500-503.