Neoplasms developing after ileocystoplasty are indeed rare; only 10 cases of carcinomas developing after ileocystoplasty have been reported in the English language medical literature [7-9]. However, their characteristics are similar to those developing after ureterosigmoidostomy. Of the 10 cases of carcinomas developing after ileocystoplasty, 8 were adenocarcinoma. In 8 of these 10 cases, the tumor developed at the ileovesical junction, and in the native bladder of the remaining 2 cases. The latent period from the original surgery to the development of cancer ranges from 4 to 32 years [7-9]. The peri-anastomotic location of the tumor in the case herein presented is in keeping with the above reported location of most tumors arising after ileocystoplasty. Though the evidence for an increased absolute risk of malignancy after ileocystoplasty is inconclusive from meta-analytic review [6], our finding of tubular adenoma with high-grade dysplasia and previous reports of adenocarcinomas in the ileal neobladder [7-9] suggests that the development of gastrointestinal malignancy after ileocystoplasty is a constant, low, but still distinct, risk, even in the absence of mixture of the urine and fecal stream. Since an estimated 153,270 new cases of large bowel cancers occurred compared to an estimated 6,170 new cases of small bowel cancers in the year 2006 (an approximately 25-fold difference in incidence) [13], small intestinal urinary derivations are considered to be at low risk of malignancy. Therefore, small bowel segments are being used increasingly in lower urinary tract reconstruction.

Two main hypotheses have been proposed for the increased risk of neoplastic changes following ureterosigmoidostomy and ileocystoplasty [1,14,15]. The first hypothesis, carcinogenic N-nitrosamine hypothesis, states that the excessive production of N-nitrosamines of known high carcinogenicity by the action of bacteria on urinary nitrate activated malignant change [1,14]. N-nitrosamines are produced when urinary nitrates are reduced to nitrites by bacteria, especially by Escherichia coli. Nitrites and urinary amines react to form N-nitrosamines. Experimental studies using animal models have reported both supportive [14,16] and contradictory [17,18] evidence for this hypothesis. It is generally believed that malignant change occurring in association with urinary reconstruction using intestinal segments is a result of urinary stagnation and bacteriuria, chronic inflammation and carcinogenic substances, such as N-nitrosamine.

Ureterosigmoidostomy patients have consistently showed substantially higher levels of N-nitrosamine in their faeces or urine compared to control subjects, and was associated with chronic inflammatory changes and altered mucin secretion at the anastomotic site [5]. However, the administration of vitamin C (an inhibitor of N-nitrosamine formation) in a subsequent small preliminary clinical study resulted in only a modest, statistically insignificant, decline in the N-nitrosamine content of rectal slurry [19]. Ileocystoplasty patients showed villous atrophy with chronic inflammation and had high concentration of N-nitrosamine in urine samples compared to control subjects [20]. The levels of N-nitrosamines in the urine of patients after augmentation ileocystoplasty have recently been reported to be greatest among those with continued urinary infection, those with sterile pyuria, those using intermittent catheterization, and those not taking prophylactic antibiotics [21]. The levels of N-nitrosamines within urine from ileal neobladder are intermediate between those found in normal controls and patients with ureterosigmoidostomy [21,22]. Additionally, a significantly higher prevalence of bacteriuria (47%) has been reported in patients after augmentation ileocystoplasty compared to controls [20].

The second hypothesis, inflammatory response hypothesis, states that the inflammatory response stimulated by the anastomosis of differing mucosa lead to nuclear instability, perhaps through the generation of reactive oxygen species [1,14]. However, direct supportive experimental and clinical evidence for this hypothesis are lacking [1,14,15]. One of the most intriguing facts of the previously reported cases of adenocarcinoma or tubular adenomas [7-9] and of the present case of tubular adenoma with high-grade dysplasia following ileocystoplasty is that almost all tumors developed at the anastomotic site and not elsewhere in the intestinal segment. Interestingly, the same situation occurs in patients following ureterosigmoidostomy where most adenocarcinomas developed at the site of the anastomosis [1,3]. Additionally, though spontaneous ileal malignancies are very rare, ileal malignancies have been more frequently reported at the site of surgical anastomosis such as ileo-ileostomy [23]. The inflammatory response hypothesis has been suggested as accounting for the increased susceptibility of development of tumors at the line of anastomosis between the intestinal segment and native bladder, but the supporting evidence is inconclusive and the exact mechanism is not well understood [1,14,15]. We hypothesize that the highest levels of carcinogenic N-nitrosamines are present closest to the anastomosis (on account of stasis or reduced motility at that site postoperatively) coupled with the inflammatory response stimulated by the anastomosis of differing mucosa lead to nuclear instability and the initiation of the metaplasia-adenoma-carcinoma sequence. However, additional long-term clinicohistological and animal studies are required to confirm our hypotheis.

Clinical studies examining the cellular origin of tumors found at the line of anastomosis between the intestinal segment and native bladder have reported contradictory findings [24,25]. Though, one study suggested the origin of the adenocarcinoma from metaplastic bladder mucosa [24], another study suggested that the neoplastic changes were consistent with an intestinal origin on the basis of histochemical staining [25]. In a follow-up prospective study of 24 patients with orthotopic ileal neobladder who underwent regular endoscopic biopsy for 5 years after surgery, progressive changes to the ileal mucosa were seen, with thinning, increased goblet cells, and villous atrophy, but no dysplasia and no malignancy [26]. Three similar studies of patients with orthotopic ileal neobladder (4-year follow-up study of 30 patients [27], 7-year follow-up study of 90 patients [28,29], and 10-year follow-up study of 20 patients [30]) found similar progressive changes to the ileal mucosa, but with no atrophy, no dysplasia and no malignancy. Additionally, histochemical staining revealed the increased presence of mucin (sulfomucins and sialomucins) immediately postoperatively which gradually reduced with extended follow-up [26-30]. These 5 recent follow-up studies suggested that the ileal mucosal changes were adaptive rather than premalignant [26-30].

Therefore, evidence in support of either of the two hypotheses of pathogenesis for malignancies following ureterosigmoidostomy and ileocystoplasty is incomplete and conflicting [1,14-22,26-30]. Recent animal studies do not suggest a significant role for nitrosamines as the carcinogenic agent [14,16-18], whilst histological studies have not shown clear evidence of premalignant mucosal change subsequent to inflammation of the anastomotic site [26-30]. Although nitrosamine excretion and bacterial load is increased in ileocystoplasty patients [20-22], this does not seem to lead to features of premalignant mucosal change during medium-term follow-up. Transposed intestinal segments exposed to urine undergo a well-characterized series of adaptive changes toward a more transitional phenotype over a period of 4–10 years [26-30]. Early inflammatory changes associated with surgery and seen in more acute animal models soon disappear without any evidence of the continued secretion of abnormal sialomucins [26-30]. These progressive changes result in a relatively thin intestinal epithelium with villous atrophy but without dysplastic or neoplastic features, although the extent of sampling of clinical material is necessarily limited for ethical reasons [1,6,28]. However, our finding of tubular adenoma with high-grade dysplasia adjacent to adaptive colonic metaplasia 34 years after ileocystoplasty and the previous report of adenocarcinoma 20 years after ileocystoplasty [7] suggest that the adaptive changes may progress to premalignant and malignant changes. Therefore, the adaptive colonic changes observed up to 10 years in the 5 studies above [26-30] may progress to premalignant and malignant changes in the ileal mucosa as late complications after ileocystoplasty. As with all malignancies, the pathway for carcinogenesis following ureterosigmoidostomy and ileocystoplasty is likely to be complex and multifactorial, and requires additional long-term histological studies to suggest further experimental approaches.

Immunohistochemical, the high-grade dysplastic ileal epithelial cells in the case herein presented expressed both DNA MMR proteins MSH2 and MLH1. The presence of protein expression of these 2 major human DNA MMR proteins indicates the absence of pathogenic mutations in these MMR genes and the absence of microsatellite instability (MSI) at this stage in this neoplastic process. Therefore, MSI was not contributory to the pathogenesis of the high-grade dysplastic adenomatous changes in the case herein presented, but does not exclude the possibility that MSI may contribute to the subsequent progression to invasive adenocarcinoma. Indeed, a recent report documented increased microsatellite levels in urine from a patient who developed an invasive adenocarcinoma within an augmented bladder after cystoplasty [31]. Loss of heterozygosity was identified in both urine and tumor samples from this patient using PCR-based microsatellite analysis, suggesting that urinary microsatellite analysis may be useful as a monitoring tool to detect neoplastic change in patients after augmentation cystoplasty [31].