Letters to the Editor

To the Editor—We enjoyed the recent case report by Hyman and colleagues describing an adenocarcinoma of the sigmoid colon seeding into a chronic anal fistula. Our unit has recently treated a 72-year-old Afro-Caribbean male who presented with a welldifferentiated mucinous carcinoma of the sigmoid colon and a synchronous mucinous adenocarcinoma within a chronic anal fistula; the fistula was identified in the United Kingdom 4 years earlier but was untreated. The patient’s self-referral had stemmed from a hard, painless lump recently noted in his longstanding fistula-in-ano. We treated him with sigmoid colectomy and wide local excision of the fistula with sphincter preservation, unlike the report by Hyman et al. in which an extended abdominoperineal excision was performed and in which they found no detectable carcinoma in the anal end of the resection specimen. Our patient also underwent sandwich chemoradiotherapy with irradiation of the pelvic and inguinal portals. He has no sign of recurrent disease at 14 months follow-up, highlighting the prospect that sphincter preservation may be possible in this rare presentation. The presence of an unexpected perianal malignancy of mucinous histology can be multifactorial. The coloproctologist should distinguish synchronous implantation of carcinoma in a preexistent anal fistula from a primary malignant degeneration within a longstanding anal fistula. In the latter circumstance, mucinous histology seems to be the commonest finding. The criteria reported by Rosser for this latter diagnosis still apply: the fistula should be of several years standing (he actually stated 10 years minimum), there is no evidence of tumor within the intervening rectum or anal canal mucosa, and the internal fistula opening is free of cancer. Equally, cutaneous metastases to the anal skin from colonic tumors (as part of wide-spread metastatic disease) have been reported, as has malignant transformation within an unrecognized rectal duplication presenting at the perianum. Furthermore, implantation metastases onto fresh perianal wounds may occur as a direct consequence of rectal cancer surgery, and it is wise to heed this oft forgotten warning of Hyman and colleagues, particularly during stapled low anterior resection. It is, therefore, of some importance to decide the precise origin of the perianal tumor when it presents synchronously with a proximal colonic cancer, because primary mucinous adenocarcinoma of the anal ducts has a generally poor survival, whereas implantation metastases onto wounded perianal skin seem to have a more favorable prognosis if adequately resected. Moreover, we were able to find two other cases similar to ours and that of Hyman et al., in addition to the other report referred to by these authors, reminding us of the merits of routine biopsy of all fistulas-in-ano.