Prevalence of Adenomas and Carcinomas in the Ileal PouchAfter Proctocolectomy in Patients with FamilialAdenomatous Polyposis
Masahiro Tajika & Tuneya Nakamura & Osamu Nakahara & Hiroki Kawai &Kouji Komori & Takashi Hirai & Tomoyuki Kato & Vikram Bhatia & Hideo Baba &Kenji Yamao
Received: 7 October 2008 /Accepted: 6 March 2009 /Published online: 31 March 2009# 2009 The Society for Surgery of the Alimentary Tract
AbstractPurpose Restorative proctocolectomy has become the most common surgical option for patients with familial adenomatouspolyposis (FAP). However, adenomas may develop in the ileal pouch mucosa over time, and even carcinoma in the pouch hasbeen reported. Our aim was to evaluate the prevalence, nature, and etiology of ileal pouch and nonpouch adenomas andcarcinoma in patients with FAP.Patients and methods This was a retrospective study of 31 FAP patients with Kocks continent ileostomy (Kock; n=8),ileorectal anastomosis (IRA; n=7), and ileal pouchanal anastomosis (IPAA) (n=16). All patients were followed with astandardized protocol including chromoendoscopy and biopsies of visible polyps in the ileal pouch and nonpouch mucosa.Results Sixteen of 24 pouch patients (Kock and IPAA) developed adenomas in the ileal pouch mucosa, and all patients withIRA developed adenomas in the rectal mucosa. The prevalence of ileal adenomas was significantly higher in pouch patientsthan in IRA patients (P=0.002). Only one patient with Kock showed adenoma in the prepouch area. Two cases ofadenocarcinomas and one case of advanced adenoma were found in the ileal pouch mucosa.Conclusion Our results show a high frequency of adenomas in the ileal pouch mucosa, with evolution into carcinoma insome patients. Regular endoscopic surveillance of the pouch is recommended at a frequency similar to that for the rectalmucosa after IRA in pouch patients with FAP.
Keywords Familial adenomatous polyposis . Ileal pouch .
Restorative proctocolectomy . Carcinoma . Adenoma
Familial adenomatous polyposis (FAP) is an inheriteddisease characterized by the development of hundreds ofcolorectal adenomas, leading to a 100% lifetime risk ofcolorectal cancer.1 For this reason, a prophylactic colec-tomy is recommended for patients with FAP for theprevention of colorectal cancer. Four surgical strategiesare available for patients with FAP: proctocolectomy withpermanent ileostomy, proctocolectomy with Kocks pouchcontinent ileostomy (Kock), colectomy with ileorectalanastomosis (IRA), and restorative proctocolectomy withileal pouchanal anastomosis (IPAA).2 The option of apermanent ileostomy is usually reserved for cases wherethere is a contraindication to the other procedures. IRA
J Gastrointest Surg (2009) 13:12661273DOI 10.1007/s11605-009-0871-1
M. Tajika (*) : T. Nakamura :H. KawaiDepartment of Endoscopy, Aichi Cancer Center Hospital,1-1 Kanokoden, Chikusa-ku,Nagoya 464-8681, Japane-mail: email@example.com
O. Nakahara :K. YamaoDepartment of Gastroenterology, Aichi Cancer Center Hospital,Nagoya, Japan
K. Komori : T. Hirai : T. KatoDepartment of Gastroenterological Surgery,Aichi Cancer Center Hospital,Nagoya, Japan
V. BhatiaDepartment of Medical Hepatology,All India Institute of Medical Sciences,New Delhi, India
O. Nakahara :H. BabaDepartment of Gastroenterological Surgery,Graduate School of Medical Sciences, Kumamoto University,Kumamoto, Japan
produces good functional results, and this surgery isassociated with less morbidity than the other procedures.3
However, continuing endoscopic surveillance for adenomasin the rectum is necessary, and there is a 13% to 25%cumulative risk of rectal cancer after 1525 years despitesurveillance.46 On the other hand, both Kock and IPAA(pouch patients) theoretically eliminate the risk of colorec-tal cancer and adenomas and perhaps the need for furtherlower gastrointestinal surveillance. However, a recent reportshowed that adenomas or carcinomas appeared not only inthe residual rectal mucosa or anastomosis after IRA but alsoin the ileal pouch mucosa after Kock or IPAA.713 Inaddition, there were five reports of cancers arising from theileal pouch mucosa, as opposed to from the anastomosis, inpatients with FAP.1418
In our center, patients with FAP underwent Kock pouchconstruction or an IRA until 1987. However, since theintroduction of IPAA in our center in 1988, we havefavored IPAA as the operation of first choice for thetreatment of patients with FAP. The aim of this study was todescribe the prevalence, nature, and etiology of adenomasand carcinoma developing in the ileal pouch mucosa andprepouch ileal mucosa in patients with FAP after proctoco-lectomy or colectomy.
Material and Methods
Endoscopic and medical records of all patients with FAP (n=70) treated in Aichi Cancer Center Hospital, Nagoya,between January 1965 and December 2002 were reviewed.FAP was defined by the presence of more than 100colorectal adenomas (all patients) and a family history ofFAP. Thirty-one patients were enrolled in endoscopicsurveillance and were included in this study. Fourteenpatients had undergone Kock and IRA until May 1987.After March 1988, 16 patients had undergone IPAA and onepatient had Kock as he had advanced cancer in the lowerrectum. These patients were subjected to regular endoscopicexamination of the ileal pouch or the rectal stump. Patientdemographic data, surgical data, details of pathologicalspecimens, and details of upper gastrointestinal endoscopywere obtained from the medical records. All patientssubmitted informed consent for collection and subsequentuse of data for research purpose, and the study was carriedout in accordance with the Helsinki Declaration.
The interval between surgery and adenoma appear-ance was defined as the time from surgery to the firstreport showing histologically confirmed adenomas in theileal mucosa. The number, size, and histology ofadenomas occurring in the ileal mucosa were deter-mined based on the last report, or the last report beforetreatment. For each patient, the most advanced histo-
logic diagnosis was taken as valid. The examinationwas performed with a flexible colonoscope. Themonitoring procedure included systematic chromoendo-scopy using 0.5% indigo carmine and biopsies of thevisible polyps. A thorough examination of the pouch,the distal 15 to 20 cm of the afferent limb, and the analcanal was made. Polyps were classified into three sizegroups: 14, 59, and 10 mm in diameter. Advancedadenomas were defined as adenomas 10 mm ingreatest diameter and/or with high-grade dysplasia.
During the follow-up of the ileal pouch or the rectalstump, endoscopic treatment of any adenoma that wasfound was decided according to its size and shape, aswell as the number of synchronous adenomas. Alladenomas
the mean duration of ileal endoscopic follow-up was 2.04.4 years (range 0.5 to 22 years).
Table 1 shows the characteristics of the pouch patients(Kock and IPAA) and IRA patients. Although the medianage and median follow-up duration of IRA patients waslonger than that of the pouch patients, there was nostatistically significant difference. Furthermore, there wereno significant differences in the median polyp count atinitial treatment not only in colon but also in the rectumbetween the patients who underwent pouch reconstructionand the IRA patients. Only the median bowel frequencywas significantly lower in IRA patients compared to thepouch patients.
The number, size, shape, and histology of polyps foundin each patient and the age of the patient and pouch areshown in Table 2. In patients with ileal pouch, adenomasdeveloped in 16 of 24 patients (67%), ranging in numberfrom 1 to 300. The size of the adenomas ranged (ranging insize) from 2 to 20 mm (Fig. 1). Two cases of adenocarci-noma and one case of advanced adenoma developed in theileal pouch of Kock and IPAA patients, respectively. Thesetumors developed in the ileal pouch mucosa itself, asopposed to the ileoanal anastomosis site. Tiny polyps ofsize 1 to 3 mm were observed in the prepouch ileal mucosain five of 24 patients, one of these were adenomas with lowgrade atypia. In patients with IRA, from one to tenadenomas were observed in all cases in the rectum; sizesvaried from 2 to 10 mm. No patient had adenomas in theileal mucosa above the IRA site. Only one patient had alymphoid polyp in the ileal mucosa.
There were no significant differences in the median ageor the median time to adenoma development since pouchsurgery in pouch patients (Kock and IPAA) and IRApatients. However, the prevalence of ileal adenomas wassignificantly higher in pouch patients, especially in thepouch mucosa as compared to the IRA patients (P=0.002),and there was a significant relationship between the number
of ileal polyps and the duration since pouch surgery inpouch patients (P=0.016).
The risk of adenoma development in the ileal pouch was13%, 43%, and 72% at 5, 10, and 20 years of follow-up,respectively, after proctocolectomy with Kock and IPAA(Fig. 2). The risk of rectal adenoma after colectomy withIRA was 14%, 57%, and 85%, at 5, 10, and 20 years offollow-up, respectively. There was no significant differencein the cumulative prevalence of ileal pouch adenomas andrectal adenomas.
Characteristics of patients who developed pouch adeno-mas were compared with those who did not develop pouchadenomas in pouch patients (Table 3). There were nosignificant differences between the ages of patients,duration of follow-up, severity of colon disease, presenceof gastric polyps and duodenal adenomas, type of pouchconstruction, median bowel frequency, and presence ofpouchitis.
Kock and IPAA have been used for patients with FAP afterproctocolectomy because they theoretically eliminate therisk of colorectal cancer and adenomas and the need forfurther lower gastrointestinal surveillance. However, devel-opment of ileal adenomas and adenocarcinomas afterproctocolectomy is becoming evident.1013 In previousreports, the prevalence reached 1357% at a medianfollow-up of 4 to 6 years after surgery.6,7,12 Groves et al.estimated that the prevalence of adenomas in the ilealpouch increased by 6.6% per year of age and 20% per yearof follow-up.12 Parc et al. showed that the risk of adenomadevelopment in the ileal pouch was 7%, 35%, and 75% at5, 10, and 15 years follow-up, respectively.11 In our study,the incidence of ileal adenomas was as high as 50% inKock and 75% in IPAA at a median follow-up of 14.7 years
Factor Pouch patients (n=24) IRA patients (n=7) P value
Median age, years (range) 46.0 (3370) 59.4 (4771) NS
Age, years (mean SD) 50.713.9 60.47.3
Median follow-up, years (range) 15.1 (4.630.8) 23.7 (17.328.4) NS
Median polyp count at treatment
Total 2,934 (25020,000) 4,789 (5709,436) NS
Colon 2,630 (21018,300) 4,182 (4209,340) NS
Rectuma 408 (52,520) 165 (11,071) NS
Gastric polyp 18 (75.0%) 5 (71.4%) NS
Papillary adenoma 15 (62.5%) 4 (57.1%) NS
Extrapapillary adenoma 11 (45.8%) 2 (28.6%) NS
Median bowel frequency per day 5 (210) 3 (26) 0.04
Table 1 Characteristics of PouchPatients and IRA Patients
IPAA ileal pouchanal anasto-mosis, Kock Kocks continentileostomya Except for lower rectum inpatients with IRA
1268 J Gastrointest Surg (2009) 13:12661273
after surgery, and the risk of adenoma in the pouch was13%, 43%, and 72% at 5, 10, and 20 years of follow-up,respectively (Fig. 2). In a recent report, Moussata et al.showed the high prevalence of ileal pouch adenoma (17/23,74%) in FAP patients with IPAA at a median interval of8 years after surgery.13 Our study of the high prevalence ofileal adenomas supports these recent results. To helpexplain the high prevalence of ileal adenomas, Moussataet al. emphasized the importance of chromoendoscopyusing indigo carmine; this procedure can help in identifyingflat and, in some rare cases, extensive lesions (Fig. 1).13
In contrast to adenomas in the ileal pouch, developmentof adenomas in the ileal segment immediately above theIPAA (prepouch) has rarely been reported. In previouspublications, development of prepouch adenomas has beenreported in ten of 26 (4%) patients by Wu et al.,7 in two of20 patients (10%) by Groves et al.,12 and in one of 24patients (4%) by Thompson-Fawcett et al.10 In this study,we found only one ileal adenoma in the mucosa above thepouch in 24 pouch patients (4%) at a median follow-up of15.1 years after surgery. It seems that development ofprepouch adenomas is rare compared with that of pouchadenomas, although based on the present study. It isdifficult to recommend reduced surveillance because ofour small patient numbers.
The development of neoterminal ileal adenomas wassignificantly higher (P=0.002) when an ileal pouch wasconstructed (as in Kock and IPAA), compared with thenonpouch patients (IRA). It has been suggested that pouchpatients by nature would be more likely to have ilealadenomas because of their selection for pouch surgeryrather than IRA. In this study, there was no difference inpolyp count at colectomy not only in colon but also inrectum. Moreover, in support of our findings, a previousstudy has reported that in pouch patients, adenomas werelimited to the pouch and were not commonly seen in theprepouch ileum mucosa of the same patients.7,10,12 Thissuggests that the pouch itself is important for enhancedadenoma risk.
The reason why ileal adenomas including prepouchadenomas are uncommon may be because of the rapidtransit of the small bowel contents through this area of thegastrointestinal tract. When fecal stasis occurs such as in areconstructed pouch, the incidence of neoplasia in ilealmucosa may increase. Several authors have implicatedcolonic metaplasia as the reason for the development ofileal adenomas8,19,20 and even carcinomas in the pouch ofpatients with FAP.2123 Colonic metaplasia was frequentlyreported in the earlier descriptions of changes observed inthe ileal pouch mucosa, and some considered it an adaptive
Table 2 Characteristics of Polyps in Pouch Patients (Kock and IPAA) and Nonpouch Patients (IRA) with FPA
Pouch patients (n=24) IRA patients (n=7)
Ileal pouch mucosa (n=16) Prepouch mucosa (n=5) Rectal mucosa (n=7) Ileal mucosa (n=1)
Median age, years (range) 41.0 (3370) 42.1 (3969) 59.4 (4771) 62.4
Age, year 48.314.4 46.87.4 60.47.3 62.4
Greatest polyp size, n
14 mm 5 5 6 1
59 mm 5 0 1 0
10 mm 6 0 0 0No. of polyps
response of the pouch to its role as a neorectum. Furtherinvestigations have shown that colonic transformation isonly partial. Small-bowel brush border disaccharidaseactivity is preserved, as is the ability to absorb vitaminB12, D-xylose, phenylalanine, and bile acids.20,2426 The
mucosal change is now described as colonic metaplasia andis likely a response to chronic inflammation caused bychanges in luminal contents. If colonic phenotypic changesare not the stimuli for the development of adenomas in theileal pouch, adenomas may form as a result of changes inthe luminal contents. Stasis in the pouch causes a change inluminal contents that are in contact with the ileal mucosa.In FAP, t...