Endoscopic mucosal resection of flat and depressed types of early colorectal cancer

Endoscopic mucosal resection: treatment of neoplasia

Techniques of endoscopic mucosal resection (EMR) can dramatically improve the ability to diagnose and treat superficial lesions in the gastrointestinal (GI) tract. Early cancers, submucosal tumors, and sessile polyps can be safely and completely removed in a single procedure, with long-term outcome results comparable to surgery. This is accomplished with a minimum cost, morbidity, and mortality and with little or no impact on the quality of life of patients. This article provides an overview of the techniques, indications, and outcomes of EMR in the management of GI malignancy.

Flat and depressed lesions of the colorectum

Although flat and depressed-type lesions are found by regular endoscopic view, magnification and pit pattern observation are vital parts of a precise diagnosis of the lesion. The depressed-type lesions have a prominent tendency to show malignant character, and the recognition and timely treatment of this lesion is inevitable in improving the morbidity and mortality from colorectal cancer. A magnifying colonoscope has the capability of a regular colonoscope with an additional feature: magnification. With chromoscopic techniques, the surface pattern of the mucosal pits can be observed. The pit-pattern classification correlates well with actual histologic findings and provides important additional information before endoscopic treatment of the lesion.

Distribution of lymph node metastasis in T1 sigmoid colon carcinoma: should we ligate the inferior mesenteric artery?

OBJECTIVE

In standard oncological sigmoid colectomy, the inferior mesenteric artery is ligated either at its origin or at the level of the left colic artery. However, in patients with early-stage carcinoma, the distribution of metastatic nodes may be limited. The aim of this study was to clarify the prevalence and distribution of lymph node metastasis in T1 sigmoid colon carcinoma and to determine the adequate range of lymph node dissection.

MATERIALS AND METHODS

The study included 121 consecutive patients treated for T1 sigmoid colon carcinoma. Clinicopathologic factors associated with nodal metastasis and the distribution of metastatic nodes were analyzed.

RESULTS

Of 121 patients, 12 (10%) had nodal involvement. The depth of invasion and the presence of lymphatic and vascular invasion were significantly associated with nodal metastasis. Of these 12 patients, 11 (92%) had lymph node metastasis confined to pericolic nodes. Nodes along the sigmoidal artery were involved in one patient. There was no involved node along the superior rectal artery or at the root of the inferior mesenteric artery.

CONCLUSIONS

Lymph node dissection for T1 sigmoid colon carcinoma should be limited to the root of the sigmoidal artery, and the inferior mesenteric artery should be preserved.

Curative resection of T1 colorectal carcinoma: risk of lymph node metastasis and long-term prognosis

PURPOSE

The features of T1 colorectal adenocarcinoma and the risk determination of lymph node metastasis were reviewed. Prognostic factors were assessed to verify whether the risk of lymph node metastasis would influence the long-term prognosis.

METHODS

Patients undergoing curative resection of T1 colorectal adenocarcinoma at the Taipei Veterans General Hospital from December 1969 to August 2002 were retrospectively studied. Patients with synchronous colorectal cancer, distant metastasis, familiar adenomatous polyposis, or inflammatory bowel disease were excluded. The associations between lymph node metastasis and clinicopathologic variables were evaluated univariately using chi-squared test, Fisher's exact test, or Student's t -test, and multivariately using logistic regression. Univariate analysis by the log-rank test and multivariate analysis by Cox regression hazards model determined the factors influencing the overall survival.

RESULTS

A total of 159 patients were included. Sixteen patients (10.1 percent) had lymph node metastasis. The risk of lymph node metastasis included histologic grade (P = 0.005), lymphatic vessel invasion (P = 0.023), inflammation around cancer (P = 0.049), and budding at the invasive front of tumor (P = 0.022). Age (P = 0.001) and number of total sampling lymph nodes (P < 0.0001) were found to be the factors influencing the overall survival.

CONCLUSIONS

Variables that predict lymph node metastasis in surgically resected T1 colorectal carcinoma may not impact the long-term prognosis.

How can colorectal neoplasms be treated during colonoscopy?

AIM

For many physicians who ordinarily treat patients with colonic diseases, colonoscopy is considered a prime study interest. Developments in colonoscopic equipment and methods have led to a large number of endoscopic diagnoses and treatment for colorectal neoplasms. The purpose of this investigation is to evaluate the efficacy and outcomes of endoscopic treatment for colorectal neoplastic lesions.

METHODS

From September 1999 to May 2003, 11 447 consecutive colonoscopic examinations in 9864 patients were gathered; totaling 5502 endoscopic treatments for colorectal neoplasms. Macroscopic characteristics of the neoplasms were classified into protruded (n = 3 953), sessile (n = 1402), lateral spreading tumor (n = 139) and depressed lesions (n = 8). Snare polypectomy was conducted in 3984 lesions, hot forcep removal in 1368 lesions, and endoscopic mucosal resection in 150 lesions.

RESULTS

Histological diagnoses were 4596 neoplastic lesions (4 376 adenomas and 220 adenocarcinomas) and 906 non-neoplastic lesions (891 hyperplastic and 15 inflammatory polyps). For the adenocarcinoma group, 31 instances involved submucosal invasion or resection margin, who received further operations, while 13 surgical specimens discovered no residual tumors. Three perforations and 96 bleedings were found following endoscopic treatment. No procedure-related mortality was found and no recurrent malignancy was found after 14-56 mo follow-up.

CONCLUSION

To lower the incidence and mortality of colorectal cancer, endoscopic treatment for colorectal neoplasms is a simple and safe procedure.

Evaluation of tumor cell dissociation as a predictive marker of lymph node metastasis in submucosal invasive colorectal carcinoma

PURPOSE

Tumor cell dissociation-the histologic finding of small solid carcinoma cell clusters and groups of dissociated dedifferentiated carcinoma cells at the invasive front-is related to tumor metastasis and patient prognosis. However, few previous reports have examined tumor cell dissociation in submucosal invasive colorectal carcinoma. We investigated the relation between tumor cell dissociation and lymph node metastasis in submucosal invasive colorectal carcinoma. We also examined immunohistochemical expression of E-cadherin and beta-catenin in submucosal invasive colorectal carcinoma.

METHODS

Submucosal invasive colorectal carcinoma tissue samples from 20 patients with lymph node metastasis and 100 patients without lymph node metastasis were evaluated. Sections stained with hematoxylin and eosin were evaluated for tumor cell dissociation. Immunohistochemistry was used to determine the expression and cellular distribution of E-cadherin and beta-catenin.

RESULTS

Tumor cell dissociation was more frequently identified in submucosal invasive colorectal carcinoma cases with lymph node metastasis than in those without lymph node metastasis (P = 0.0001). Decreased membranous expression of E-cadherin occurred more frequently in submucosal invasive colorectal carcinoma cases with lymph node metastasis than in those without lymph node metastasis (P = 0.025). Nuclear expression of beta-catenin tended to be present in submucosal invasive colorectal carcinoma cases with lymph node metastasis (P = 0.063). Decreased membranous expression of E-cadherin occurred more frequently in submucosal invasive colorectal carcinoma cases with tumor cell dissociation than in those without tumor cell dissociation (P = 0.0023).

CONCLUSIONS

Our results suggest that there is a relation between tumor cell dissociation and lymph node metastasis in submucosal invasive colorectal carcinoma. Tumor cell dissociation formation might be related to abnormal expression patterns of E-cadherin and beta-catenin in submucosal invasive colorectal carcinoma. Tumor cell dissociation and decreased membranous expression of E-cadherin would be important predictive markers for lymph node metastasis in submucosal invasive colorectal carcinoma.

Effectiveness of glycerol as a submucosal injection for EMR

BACKGROUND

EMR traditionally is performed by using normal saline solution (NS) as the submucosal fluid cushion. It is thought, however, that NS does not maintain the proper mucosal elevation for EMR of large, flat lesions. We investigated the efficacy of glycerol as the submucosal injection solution.

METHODS

A total of 110 colorectal laterally spreading tumors (LST) were treated by EMR with glycerol. For comparison, 113 LSTs treated by using NS were studied. The en bloc resection, complete resection, and associated complications rates were evaluated retrospectively.

OBSERVATIONS

The en bloc resection rate in the glycerol group was 63.6% (70/110) compared with 48.9% (55/113) in the NS group (p < 0.05). The complete resection rate in the glycerol group was 45.5% (50/110) compared with 24.6% (28/113) in the NS group (p < 0.01). The associated complications rate was similar in both groups.

CONCLUSIONS

It technically was easier and as safe to perform EMR of colorectal LSTs when using glycerol as the submucosal injection solution.

Efficacy of magnifying chromoendoscopy for the differential diagnosis of colorectal lesions

Magnifying chromoendoscopy is an exciting new tool and offers detailed analysis of the morphological architecture of mucosal crypt orifices. In this review, we principally show the efficacy of magnifying chromoendoscopy for the differential diagnosis of colorectal lesions such as prediction between non‐neoplastic lesions and neoplastic ones, and distinction between endoscopically treatable early invasive cancers and untreatable cancers based on a review of the literature and our experience at two National Cancer Centers in Japan. Overall diagnostic accuracy by conventional view, chromoendoscopy and chromoendoscopy with magnification ranged from 68% to 83%, 82% to 92%, and 80% to 96%, respectively, and diagnostic accuracy of accessing the stage of early colorectal cancer using magnifying colonoscopy was over 85%. Although the reliability depends on the skill in magnifying observation, widespread applications of the magnification technique could influence the indications for biopsy sampling during colonoscopy and the indication for mucosectomy. Moreover, the new detailed images seen with magnifying chromoendoscopy are the beginning of a new period in which new optical developments, such as narrow band imaging system, endocytoscopy system, and laser‐scanning confocal microscopy, will allow a unique look at glandular and cellular structures.

Colon cancer: genomics and apoptotic events

Colon cancer is the third most common cancer globally. The risk of developing colon cancer is influenced by a number of factors that include age and diet, but is primarily a genetic disease, resulting from oncogene over-expression and tumour suppressor gene inactivation. The induction and progression of the disease is briefly outlined, as are the cellular changes that occur in its progression. While colon cancer is uniformly amenable to surgery if detected at the early stages, advanced carcinomas are usually lethal, with metastases to the liver being the most common cause of death. Oncogenes and genetic mutations that occur in colon cancer are featured. The molecules and signals that act to eradicate or initiate the apoptosis cascade in cancer cells, are elucidated, and these include caspases, Fas, Bax, Bid, APC, antisense hTERT, PUMA, 15-LOX-1, ceramide, butyrate, tributyrin and PPARgamma, whereas the molecules which promote colon cancer cell survival are p53 mutants, Bcl-2, Neu3 and COX-2. Cancer therapies aimed at controlling colon cancer are reviewed briefly.

Predictive histopathologic factors for lymph node metastasis in patients with nonpedunculated submucosal invasive colorectal carcinoma

PURPOSE

Risk factors for lymph node metastasis in patients with nonpedunculated submucosal invasive colorectal carcinoma remain to be characterized. This study examines the relationship between lymph node metastasis and clinicopathologic factors in nonpedunculated submucosal invasive colorectal carcinoma.

METHODS

The study cohort comprised 155 patients who had undergone surgical treatment for nonpedunculated submucosal invasive colorectal carcinoma. The clinicopathologic factors investigated included gender, age, tumor location, macroscopic type, tumor size, histologic type and grade, intramucosal growth pattern, lymphatic invasion, venous invasion, degree of focal dedifferentiation at the submucosal invasive front, status of the remaining muscularis mucosa, and the depth and width of submucosal invasion.

RESULTS

Lymph node metastases were found in 19 patients (12.3 percent). Univariate analysis showed that lymphatic invasion, focal dedifferentiation at the submucosal invasive front, status of the remaining muscularis mucosa, and depth of submucosal invasion all had a significant influence on lymph node metastasis. Multivariate analysis showed lymphatic invasion (P = 0.014) and high-grade focal dedifferentiation at the submucosal invasive front (P = 0.049) to be independent factors predicting lymph node metastasis. No lymph node metastasis was found in tumors with a depth of submucosal invasion of <1.3 mm.

CONCLUSIONS

Lymphatic invasion and high-grade focal dedifferentiation at the submucosal invasive front are important predictors of lymph node metastasis in patients with nonpedunculated submucosal invasive colorectal carcinoma. Depth of submucosal invasion can be used as an identifying marker for patients who do not require subsequent surgery after endoscopic resection.

Interobserver variability in the pathological assessment of malignant colorectal polyps

Background

Treatment of patients with malignant large bowel polyps is highly dependent on pathological evaluation. The aim of this study was to evaluate interobserver variability in the pathological assessment of endoscopically removed polyps.

Methods

The records of 88 patients with colorectal cancer who underwent endoscopic removal of malignant polyps were reviewed. Study investigators reviewed the initial pathology report; three experienced gastrointestinal pathologists reviewed all slides in a blinded fashion. Interobserver variability of pathological assessment of malignant polyps was analysed by κ statistics.

Results

Seventy-six (86 per cent) of the 88 patients had malignant polyps and 12 (14 per cent) had carcinoma in situ. Agreement between experienced pathologists was substantial with regard to T stage (κ = 0·725), resection margin status (κ = 0·668) and Haggitt's classification (κ = 0·682), but comparison of initial and experienced pathologists' assessment demonstrated only moderate agreement in these areas (κ = 0·516, κ = 0·555 and κ = 0·578 respectively). Agreement between even experienced pathologists was poor with respect to histological grade of differentiated adenocarcinomas (κ = 0·163) and angiolymphatic vessel invasion (κ = − 0·017).

Conclusion

Pathological assessment of malignant polyps varies between observers. Specialist pathologists appear to have a higher degree of consensus among themselves than with generalist pathologists with respect to T stage. The high interobserver variability with regard to histological grade of differentiated tumours is clinically irrelevant. However, variability in the assessment of angiolymphatic vessel invasion limits the value of this measurement for clinical decision making.

EMR of large sessile colorectal polyps

BACKGROUND

EMR optimizes histopathologic assessment of resected lesions. This study evaluated the outcome of EMR of large sessile colorectal polyps in terms of complications and recurrence.

METHODS

An uncontrolled prospective study was conducted of a cohort of 136 patients with sessile colorectal polyps referred for EMR. After submucosal injection, EMR was performed piecemeal by either snare polypectomy alone or with cap aspiration.

RESULTS

In 136 patients, a total of 139 sessile polyps were resected, 86 of which were in the right colon. Median polyps diameter was 20 mm in the right colon and 30 mm in the other colonic segments. Intraprocedure bleeding occurred after 15 polypectomies (10.8%) and was controlled endoscopically in all cases; there was no delayed bleeding. Post-polypectomy syndrome occurred in 5 patients (3.7%). There was no perforation. Invasive carcinoma was found in 17 sessile colorectal polyps, and surgery was performed in 10 of 17 cases. Follow-up colonoscopy in 93 patients without invasive carcinoma (96 polyps), over a median of 12.3 months, disclosed local recurrence of 21 adenomatous polyps (21.9%). Colonoscopic follow-up in 5 of the 7 patients, who had sessile colorectal polyps with invasive carcinoma and did not undergo surgery, disclosed no local recurrence.

CONCLUSIONS

EMR, including EMR with cap aspiration, is effective and safe for removal of sessile colorectal polyps throughout the colon.

Low Frequency of bcl-2 Expression in Large Non-Polypoid Colonic Neoplasms

Large non-polypoid colorectal adenomas that spread over the mucosa and morphologically flat lesions are included in a group called granule-aggregating tumours. These are uncommon in the West. We aimed to clarify the biological differences between granule-aggregating tumours and colorectal polypoid adenomas. We evaluated the extent of apoptotic cell loss and expression of bcl-2 and p53 oncoproteins in 26 granule-aggregating tumours and 19 polypoid adenomas. The mean apoptosis index value of granule-aggregating tumours was significantly higher than that of polypoid adenomas. Only two (7.7%) granule-aggregating tumours and 13 (68.4%) polypoid colorectal adenomas expressed bcl-2, while 12 (46.2%) granule-aggregating tumours and six (31.6%) polypoid colorectal adenomas were p53-positive. Our results show that the higher apoptosis index and frequent expression of bcl-2 oncoprotein differentiates granule-aggregating tumours from polypoid colorectal adenomas. We propose that large non-polypoid granule-aggregating tumours of the colorectum are a biologically distinct entity.

Pathological features and genetic alterations in colorectal carcinomas with characteristics of nonpolypoid growth

We sought to clarify pathological features and genetic alterations in colorectal carcinomas with characteristics of nonpolypoid growth. Colorectal carcinomas resected at Showa University Hospital in Tokyo included 86 with characteristics of polypoid growth (PG) and 21 with those of nonpolypoid growth (NPG). Mutations of APC, Ki-ras, and p53 genes, as well as microsatellite instability (MSI), were analysed using fluorescence-based polymerase chain reaction-single-strand conformation polymorphism (PCR-SSCP). Carcinomas with an NPG pattern were smaller than PG tumours (P<0.0001). Carcinomas with a PG pattern were more likely to harbour Ki-ras mutations (36%) than NPG tumours (0%; P<0.0001). Mutation types in the APC gene differed significantly between PG and NPG carcinomas (P=0.0189), including frameshift mutations in 66% of PG carcinomas but no NPG carcinomas. Presence of a p53 mutation at a 'hot spot' also was more likely in PG carcinomas (37%) than in NPG carcinomas (0%; P=0.0124). No significant difference in presence of MSI was evident between carcinomas with PG and NPG patterns. In conclusion, significant genetic differences were evident between carcinomas with PG and NPG patterns. Genetic changes in NPG carcinomas differed from those of the conventional adenoma-carcinoma sequence. Assuming that some nonpolypoid growth lesions transform rapidly into advanced carcinomas, 20% of all colorectal carcinomas may progress in this manner.

A prospective evaluation of high-magnification chromoscopic colonoscopy in predicting completeness of EMR

BACKGROUND

EMR is used to treat flat and sessile lesions in the colon. The aim of this study was to prospectively assess the efficacy of high-magnification chromoscopic colonoscopy in predicting complete resection margins after EMR.

METHODS

A total of 1250 patients underwent colonoscopy by using a magnifying colonoscope. Chromoscopy with indigo carmine and crystal violet dye solutions was used to assess mucosal pit patterns. EMR was performed by using the saline solution inject-and-cut technique. After EMR, resection margins were inspected by using high magnification, and completeness of excision was predicted from the surface pit pattern. This was compared with completeness of excision as determined histopathologically.

RESULTS

A total of 684 lesions were treated by EMR (62 piecemeal) in 602 patients. The sensitivity of high-magnification chromoscopic colonoscopy for predicting remnant tissue in the lateral margins and the deep margins were, respectively, 79% and 80%. Specificity for both margins was 97%. The overall accuracy of high-magnification chromoscopic colonoscopy in predicting incomplete resection after EMR in the lateral axis and the deep axis was, respectively, 93% and 95%, where the true respective fractions of incomplete resections were 17% and 10%. Of the single en bloc EMRs performed, 77 (12%) had histopathologic evidence of incomplete resection in either axis, compared with 60 (97%) of the 62 piecemeal resections. Piecemeal resection was more likely to result in incomplete resection compared with en bloc EMR (p < 0.001). Complete resection by EMR was more likely if the lesion was sessile instead of flat (p < 0.001).

CONCLUSIONS

High-magnification chromoscopic colonoscopy as an in vivo modality for prediction of remnant tissue after EMR has a high overall accuracy, but further studies assessing long-term outcome and cost-effectiveness compared with conventional colonoscopic techniques are required.

New indication for endoscopic treatment of colorectal carcinoma with submucosal invasion

BACKGROUND AND AIMS

Although an increasing number of early colorectal cancers (CRC) have been curatively treated by endoscopy, there have been no definitive criteria to decide the effectiveness of such therapy. We retrospectively analyzed clinicopathological factors to establish criteria for curative endoscopic treatment of early CRC.

METHODS

First, risk factors of lymph node metastasis were analyzed in 171 patients who received surgery with postoperative histology of CRC submucosal invasion. The resultant new criteria were evaluated in another 60 patients who experienced endoscopic resection of CRC and surgery according to the current criteria most often used in Japan.

RESULTS

In the first substudy, lymph node metastasis was present in 18 of 171 patients (10.5%). Lymphatic permeation, sprouting and infiltrative growth of cancer cells were identified as histological factors significantly related to lymph node metastasis, and observed in much higher rates when the depth of submucosal invasion was beyond 1,500 micron. The minimum depth with positive lymph nodes was 1,075 micron. In the second group of 60 patients, lymph node metastasis was recorded in none of nine patients who met our new criteria of complete endoscopic treatment: submucosal invasion below 1,500 micron in depth, and no lymphatic permeation, sprouting or infiltrative growth pattern on tumor histology. Lymph node metastasis was positive in three of the other cases who did not meet our new criteria.

CONCLUSIONS

The present study showed that endoscopic treatment of early CRC may be considered complete when submucosal invasion beyond 1,500 micron, lymphatic permeation, sprouting, and infiltrating growth are all denied.

The prevalence of small, flat colorectal cancers in a western population

OBJECTIVE

Small, flat colorectal cancers have been widely reported in the Japanese literature but are thought to occur rarely outside Japan. The aim of this retrospective cohort study was to clarify the prevalence of flat colorectal cancer in a Western population.

METHODS

One thousand and twenty-six consecutive colonoscopies performed by a single experienced endoscopist were retrospectively analysed over a two-year period. The morphology, site and histological appearance of all documented colorectal cancers (CRC) were recorded.

RESULTS

Forty-seven cases of CRC were detected, five of which (10%) demonstrated flat configuration. Flat cancers varied between 8 and 15 mm in diameter (mean 11 mm). Histologically, all flat lesions were moderately differentiated Dukes A adenocarcinomas. Two of these cancers contained no adenomatous component.

CONCLUSION

This study confirms that small, flat colorectal cancers are not an uncommon finding at colonoscopy in Western patients. Compared to polypoid neoplastic lesions, flat cancers appear to undergo malignant change at a smaller size.

A prospective clinicopathological and endoscopic evaluation of flat and depressed colorectal lesions in the United Kingdom

OBJECTIVES

Flat and depressed colorectal lesions are now reported in Western populations. The malignant potential, anatomical distribution, and other clinicopathological features have not been established in this group. This study aimed to assess prospectively the prevalence, clinicopathological, and endoscopic features of flat and depressed colorectal lesions in the United Kingdom.

METHODS

A single endoscopist performed colonoscopy on 850 consecutive patients presenting for routine colonoscopy. All endoscopies were performed using a high magnification colonoscope with chromoscopy to facilitate detection of flat and depressed colorectal lesions.

RESULTS

A total of 458 flat lesions were identified. Of these, 173 (38%) were hyperplastic and 285 (62%) adenomatous or beyond. Of the 173 hyperplastic flat lesions, 162 (94%) were located in the recto-sigmoid region. Of the 267 adenomas, 66 (25%) had areas of high grade dysplasia (HGD), with 54/66 (82%) being present in the right colon. Flat lesions <8 mm in diameter was more likely to contain HGD than those <8 mm (p<0.001). Nine of the 10 (90%) flat invasive adenoacarcinomas were in the right colon and all had a depressed morphological component. In contrast, HGD was observed in 58/466 (12%) of protuberant (sessile/pedunculated) adenomas of which 95% (55/58) were located in the left colon. In addition, HGD was present in 17% of all sessile adenomas versus 44.6% of flat lesions >8 mm in diameter (p=0.001). Of the 14 protuberant carcinomas, 13/14 (93%) were in the left colon. Synchronous lesions were found in 96/816 (12%) of cases. Of the 816 patients with two or more left-sided protuberant adenomas <8 mm (with or without HGD), 89 (11%) had one or more flat lesions in the right colon with HGD.

CONCLUSIONS

Flat adenomas and carcinomas have a high malignant potential compared to protuberant lesions and have a propensity for developing in the right hemi-colon. Total colonoscopy is required to detect such lesions, as only 18% of flat lesions would be in reach of the flexible sigmoidoscope.

The role of flat and depressed colorectal lesions in colorectal carcinogenesis: new insights from clinicopathological findings in high-magnification chromoscopic colonoscopy

High-magnification chromoscopic endoscopy is a new technique which has been recently introduced to the UK. This technology, initially pioneered by the Japanese in the 1980s, has changed our understanding of the pathogenesis of colorectal cancer and our subsequent therapeutic strategies aimed at the secondary prevention of cancer. Magnification colonoscopic techniques when combined with colonic chromoscopy (dye spraying of the colon) permit in vivo assessments of lesions at a magnification and resolution similar to a stereomicroscope. Furthermore, flat/depressed adenomas and cancers can be diagnosed with increasing frequency and new resection practices performed. This technique is known as endoscopic mucosal resection. As gastrointestinal endoscopists adopt these new techniques, close liaison with histopathologists is essential to provide the highest standards of diagnostic accuracy. The histopathologist also needs to be aware of the endoscopic findings when interpreting specimens and hence must understand new endoscopic terminologies and classification systems that accompany the introduction of new technologies and therapeutic techniques. This article describes the controversies relating to the flat and depressed colorectal lesion, where these new endoscopic technologies are ideally suited. It then provides a working description of high-magnification chromoscopic colonoscopy including the Japanese 'pit pattern' and morphological classification system-information which will be provided to histopathologists with specimens obtained by these new techniques. Finally, we describe the procedure of endoscopic mucosal resection, as the type and quality of specimens received for histopathological analysis will be highly influenced by these techniques.

Clinical outcome of endoscopically removed early colorectal cancer

BACKGROUND AND AIM

Endoscopic resection is widely adopted for early colorectal cancer. Most studies examining the effectiveness of endoscopic treatment of early colorectal cancer are from Japan, and little is known about the success of this treatment in other areas of the world. The authors performed this study to investigate the effectiveness of endoscopic treatment of early colorectal cancer.

METHODS

The medical records of patients with histologically proven early colorectal cancer and who were treated by endoscopic resection between January 1985 and December 2000 were reviewed retrospectively. Information regarding the demographic data of patients, clinicopathological characteristics of the tumors, and follow-up events were recorded and analyzed.

RESULTS

Altogether 91 patients, including 73 with mucosal cancers and 18 with submucosal invasive cancers, were enrolled. Nineteen patients (eight with mucosal cancers and 11 with submucosal invasive cancers) underwent subsequent surgical colectomy. Larger tumors, and those with sessile or flat depressed morphology, and deep submucosal invasion were more difficult to remove completely through endoscopic resection alone. Five patients had local recurrent tumors, occurring between 4 and 12 months after endoscopic resection, and were treated surgically. Furthermore, two patients died of unrelated diseases, one at 1 month and one at 3 years after treatment. All the other patients were well after an average follow-up interval of 40.9 months (range 18-189 months).

CONCLUSIONS

Endoscopic resection safely and effectively removed most early colorectal cancers. Careful histological examinations of resected specimens to determine the indications of subsequent surgical colectomy, and close post-treatment surveillance would improve the outcome and life quality for these patients.

Staging of colonic neoplasms by colonoscopic miniprobe ultrasonography

BACKGROUND AND AIMS

In contrast to the situation in the upper gastrointestinal tract staging of colonic neoplasm by endoscopic ultrasonography (EUS) has not gained importance because until yet preoperative staging is without any clinical consequences. This may change with the introduction of minimally invasive surgical procedures and endoscopic resection techniques as an alternative to conventional (open) surgery.

PATIENTS AND METHODS

We performed EUS with a miniprobe in 54 consecutive patients with colonic tumors who had been referred to our hospital for endoscopic resection or for laparoscopic resection of their lesions. Therefore patients with locally advanced tumors or systemic tumor spread were not included. After detection of the lesion during colonoscopy miniprobe EUS was performed with water-filling of the colonic lumen. The depth of invasion (T classification) and the local lymph node status (positive or negative) was ascertained. Lymph node-negative lesions staged as T1 underwent endoscopic resection whenever this was technically possible. In lymph node-negative T2-3 tumors laparoscopic resection was planned if they were localized at least 10 cm apart from the flexuras. All other lesions were resected by open surgery. The EUS findings were later compared with the final pathological results (pTN classification) of the resected specimen.

RESULTS

In 50 patients (93%) a sufficient EUS evaluation of the colonic tumor was possible. In one patient with a tumor at the left flexura the lesion could not be completely visualized, and in three patients a sufficient water filling of the colon was impossible. The infiltration depth was correctly classified in 17 adenomas, 16 T1, 8 T2, 5 T3, and one T4-carcinoma (EUS accuracy for T staging: 94%). Two T2 and one T3 carcinoma were overstaged by EUS while no understaging was recorded. The lymph node status was correctly classified in 42/50 patients (84%), and a false-negative lymph node status was found in only 4/50 cases (8%). The overall accuracy of EUS was 80%.

CONCLUSION

Miniprobe EUS is suitable and has a sufficient but not optimal accuracy for staging of colonic neoplasm. Its employment makes sense if minimally invasive resection techniques in patients with high-risk for open surgery are planned.

[Risk factors for lymphatic metastasis from pT1 colorectal adenocarcinoma]

During the last 20 years, endoscopic removal of colorectal adenoma has become widely accepted as a replacement for removal by open surgery. Even colorectal adenocarcinomas are not excluded. The key question is when surgical treatment should still be preferred over endoscopic removal as the primary treatment. One good indicator is the frequency of lymph node metastasis, which should be compared with the overall risk involved in the surgical procedure itself. Histological examination allows subdivision of early colorectal adenocarcinomas into low-risk and high-risk groups. Classical parameters for a high-risk situation are lymphatic invasion, poor differentiation, and incomplete removal (R1). Additional risk factors that have recently been discussed are infiltration into the lower third of the submucosal layer (sm3) and dissociation (budding) of the tumour cells at the invasion front. Drawing on the literature and an analysis of our own patients, we demonstrate a positive correlation between these new markers and an elevated risk of the presence of lymph node metastasis.

Pelvic radiological imaging: a surgeon's perspective

Radiological imaging of the pelvis adds an important dimension to our understanding of rectal and perianal disease. By integrating relevant information obtained from these investigations into planning and conduct of surgical procedures, outcomes for patients may be optimised. This review focuses on three areas from a clinical viewpoint. (1) With the increased use of neoadjuvant treatments pretherapeutic staging strategies become central to the management of rectal cancer patients. At present, transrectal ultrasound (TRUS), computerised tomography and magnetic resonance imaging (MRI) serve in combination to provide the essential informations. (2) The advent of endoanal ultrasound and MRI in the diagnostic workup of patients with faecal incontinence has caused a paradigm shift both conceptionally and in the way treatments are tailored to individual patients. (3) Concerning primary perianal fistulas there is little place for endoanal ultrasound or MRI. However, when a recurrent or Crohn's fistula is present, a combination of surgical exploration with either endoanal ultrasound or MRI depending on local expertise and availability may be the optimal approach to maximise benefit for these patients.

Gasless, video endoscopic transanal excision for carcinoid and laterally spreading tumors of the rectum

BACKGROUND

The aim of this study is to determine whether gasless, video endoscopic transanal-rectal tumor excision (gasless VTEM) is a valid treatment for rectal carcinoid and laterally spreading tumors (LST).

METHODS

Eighty-four patients with an adenoma, adenocarcinoma (Tis/T1), or carcinoid tumor of the rectum were divided into three groups: (i) LST ( n = 17 patients), (ii) carcinoid ( n = 11), and (iii) control with other types of tumors ( n = 56).

RESULTS

The LST group had a longer median operating time than in the control group, whereas the carcinoid group had a shorter operating time. Two patients (11.7%) in LST group developed peritoneal entry during the operation, while 2 patients (3.6%) in the control group experienced postoperative complications. During a median follow-up length of 55.2 months, one patient in the LST group developed a recurrence.

CONCLUSIONS

Gasless VTEM is a simple, minimally invasive procedure used to treat LST and carcinoid tumors of the rectum. However, resection for the LST group had a high risk of peritoneal entry during operation.

Detection of colorectal adenomas by routine chromoendoscopy with indigocarmine

OBJECTIVES

Nonpolypoid adenomas, which can be important precursors of colorectal cancers, are difficult to find during routine colonoscopy. The aim of this study was to evaluate the usefulness of routine chromoendoscopy in Korea, where the incidence of colorectal cancer is low compared with western countries.

METHODS

Colonoscopy with chromoendoscopy was performed in 74 consecutive patients (48 men, 26 women; mean age 53.0 yr). After a careful examination of the whole colon, a defined segment of the sigmoid colon and rectum (0-30 cm from the anal verge) was stained with 20 ml of 0.2% indigocarmine solution with a spraying catheter. Nonpolypoid lesions were classified as flat or depressed types. Biopsies were taken from all lesions detected before or after staining with indigocarmine.

RESULTS

Indications for colonoscopy included routine check-up (21 patients), diarrhea or loose stool (14 patients), abdominal pain (12 patients), constipation (7 patients), bleeding (6 patients), and others (14 patients). Before staining, 58 lesions were found in 30 patients (43.2%). Histology showed tubular adenoma in 41 lesions, hyperplastic or inflammatory changes in 14 lesions, adenocarcinoma in 2 lesions, and villous adenoma in 1 lesion. After indigocarmine staining for normal-looking distal 30 cm colorectal mucosa, 176 lesions were found in 46 patients (62.2%). Histologically, 158 lesions were hyperplastic or inflammatory in nature, and 17 lesions (from 11 patients) were tubular adenomas. There was one serrated adenoma. Eighteen adenomas seen only after spraying indigocarmine were 2.6 +/- 0.6 mm in diameter, and all of them were classified as flat adenomas. There was no depressed-type adenoma. No adenoma with high grade dysplasia, villous histology, or cancer was found after staining. Presence of macroscopic adenomatous lesions or carcinoma before staining could not predict the existence of adenoma after staining.

CONCLUSIONS

In a large proportion of patients, flat or depressed adenomas could be found after spraying indigocarmine for normal-looking colorectal mucosa in Korea. The clinical significance of these diminutive adenomas that can be found only after spraying contrast agent needs to be further investigated.

Flat colonic adenomas in Malaysia: fact or fancy?

BACKGROUND

Some two-thirds of colorectal carcinomas arise from adenomatous polyps, and as such, screening by colonoscopy and polyp removal should significantly reduce colorectal cancer. This has not been the case, as evidenced by recent studies, which revealed that endoscopy failed to prevent up to 50% of all subsequent carcinomas. Flat or depressed adenomas, frequently reported from Japan but rarely elsewhere, might explain the 'missed carcinomas.' Detection of flat adenomas has not been previously reported from Malaysia.

METHODS

In the present prospective study, 426 consecutive patients underwent colonoscopic examination between March 1997 and January 2000, for a variety of bowel symptoms. The examinations were performed by an experienced endoscopist using a standard colonoscope and methylene blue dye spraying technique. Macroscopically, flat adenomas were defined using the criteria proposed by Sawada.

RESULTS

Twenty-nine adenomas were identified in 12 patients, of which 15 were polypoid and 14 were flat, with no depressed lesions. Eight polypoidal lesions and all the flat adenomas contained mild or moderate areas of epithelial dysplasia. Seven severely dysplastic polyps were identified. One Duke's A polypoidal cancer and two advanced carcinomas were also found. All the severely dysplastic lesions and Duke's A carcinomas were found in polyps greater than 10 mm in mean size. The flat adenomas were all less than 5 mm in size.

CONCLUSIONS

A significant proportion of colonic adenomas in Malaysian patients appear as small flat lesions, which could easily be missed during endoscopy. Increased recognition and treatment of flat adenomas among colonoscopists is warranted.

An early stage small bowel adenocarcinoma with microsatellite instability phenotype in a case of hereditary nonpolyposis colorectal cancer

BACKGROUND

Hereditary nonpolyposis colorectal cancer (HNPCC) is an autosomal dominantly inherited disease characterized by onset at a relatively early age, excess of synchronous and metachronous tumors, and a variety of extracolorectal malignancies. Small bowel carcinoma is included in the tumor spectrum of HNPCC, but the frequency of occurrence of this tumor in HNPCC patients is comparatively rare.

CASE PRESENTATION

We report the case of a 55-year-old woman who had a history of multiple colon cancers at 33 years of age, sigmoid colon cancer at 47, and endometrial carcinoma at 51. This case fulfills the Amsterdam criteria for HNPCC and followed from the patient's age of 47 at our institute. Surveillance colonoscopy showed a sessile polyp in the ileum that was 9 mm in diameter and located about 10 cm proximal to the ileorectal anastomosis, and that was resected by endoscopic mucosal resection. Histopathological studies showed an adenoma with well-differentiated adenocarcinoma in the mucosa. A molecular analysis of the adenoma component of this polyp was performed, and microsatellite instability was found in four of the nine analyzed loci. The patient was a mutation carrier in hMSH2, one of the mismatch repair genes responsible for HNPCC.

CONCLUSION

Reports of early-stage carcinoma of the small bowel in HNPCC are very rare, and an adenoma-carcinoma sequence was present in the small bowel tumor of this patient. The molecular findings of this tumor are discussed.

Spasmolytic effect of peppermint oil in barium during double-contrast barium enema compared with Buscopan

AIM

To evaluate the efficacy of peppermint oil in barium as a spasmolytic agent during a double-contrast barium enema (DCBE).

MATERIALS AND METHODS

A total of 383 DCBEs with positive results from occult blood tests were assessed. Patients were assigned to one of four groups: peppermint in barium (n=91), peppermint in tube (n=90), Buscopan (n=105), or no treatment (n=97). After a screening sigmoidoscopy, the DCBEs were performed using air as a distending gas. In the Buscopan group, the DCBE was performed with an intramuscular injection of 20mg Buscopan at the start of the examination. Patients in the no-treatment group underwent DCBE without any spasmolytic agent. A peppermint oil preparation (30ml) was mixed in the barium solution for patients in the peppermint-in-barium group, and the same dose of peppermint oil was included in the enema tube in the peppermint-in-tube group. The presence of spasm on a series of spot films was evaluated without information about the type of spasmolytic agent used.

RESULTS

The percentage of patients in the four groups (no treatment, Buscopan, peppermint in tube, and peppermint in barium) with absence of spasm in the entire colon on the series of spot films was 13.4, 38.1, 41.8, and 37.8%, respectively. In the group using peppermint oil or Buscopan, the rate of patients with non-spasm examination was higher than that in no-treatment group (p<0.0005). Peppermint oil had the same spasmolytic effect as the systemic administration of Buscopan in the transverse and descending colon. Peppermint oil had a stronger effect in the caecum and the ascending colon than a Buscopan injection (p<0.005). There was no advantage to placing peppermint oil in the enema tube over mixing it in the barium solution. A total of 157 polyps were found during the DCBE procedures, and no differences were observed in the number of lesions among the four groups. Peppermint oil did not impair image quality.

CONCLUSION

Barium solution mixed with peppermint oil was safe and effective for the elimination of colonic spasm during the DCBE procedure, and it could be used instead of Buscopan.

Gasless video endoscopic transanal excision of rectal tumors incompletely removed by colonoscopic snare polypectomy

BACKGROUND

This report describes an experience with gasless video transanal endoscopic microsurgery (VTEM) to excise rectal tumors previously incompletely removed with colonoscopic snare polypectomy.

METHODS

Gasless VTEM involves a modification of transanal endoscopic microsurgery (TEM) that incorporates a standard laparoscopic video camera and requires no CO(2) insufflation system. Nineteen patients who had had a rectal tumor removed incompletely by colonoscopic polypectomy with a diathermy snare were enrolled in this prospective study. The patients included 14 men and 5 women whose median age was 63.5 (range, 49-83) years. The rectal tumors included 4 adenomas, 11 adenocarcinomas (Tis, 7; T1, 4), and 4 carcinoid tumors. The median distance from the tumor margin to the dentate line was 5.8 (range, 2.0-13.0) cm.

RESULTS

All rectal lesions were successfully removed by gasless VTEM. No intraoperative complication occurred. The median operating time and blood loss were 40 (range, 15-145) minutes and 5 (range, 0-100) mL, respectively. The median maximal tumor diameter in 9 patients with residual tumors was 1.3 (range, 0.5-2.5) cm. There was no operative mortality. A postoperative complication (bleeding from a suture wound and transient incontinence) developed in 1 (5.3%) of the 19 patients. The median postoperative hospital stay was 5 (range, 2-10) days. Postoperative histology revealed a residual tumor in 10 (52.9%) of the 19 specimens. Complete excision of all tumors was confirmed histologically. During a median follow-up period of 59.5 (range, 12.3-94.9) months, no tumor recurred.

CONCLUSIONS

Gasless VTEM is useful and minimally invasive for the local removal of rectal tumors incompletely resected by colonoscopic snare polypectomy.

Flat and depressed colorectal tumours in a southern Swedish population: a prospective chromoendoscopic and histopathological study

BACKGROUND

Flat and depressed colorectal tumours are common in Japan but are very rare or non-existent in Western countries.

AIMS

To study the occurrence of flat colorectal tumours in a southern Swedish population.

METHODS

In this prospective study, 371 consecutive European patients were examined by high resolution video colonoscopy combined with chromoendoscopy. The nature of the lesions was determined by histopathological examination.

RESULTS

A total of 973 tumours were found; 907 (93.2%) were protruding and 66 (6.8%) were flat or depressed. Of the flat/depressed tumours, five (7.7%) were early adenocarcinomas infiltrating the submucosa. Eleven carcinomas (1.2%) were found among protruding tumours. High grade dysplasia was observed in 18% (n=11) of flat/depressed adenomas in contrast with 7.3% (n=65) of protruding adenomas, and occurred in smaller flat/depressed tumours compared with protruding ones (mean diameter 8 mm v 23 mm, respectively). Furthermore, high grade dysplasia was significantly more common in flat elevated tumours with central depression or in depressed adenomas (35.7%; 5/14) than in flat elevated adenomas (12.8%; 6/47).

CONCLUSION

Flat and depressed tumours exist in a Western population. Future studies should address whether or not chromoendoscopy with video colonoscopy is necessary in the search for flat colorectal neoplasms.

Surgical management of malignant colorectal polyps

The anatomic landmarks of the depth of invasion for pedunculated lesions (Haggitt level) and the Sm system for the sessile lesions give excellent objective information in the management of malignant colorectal polyps. Malignant polyps with low risk of lymph node metastasis include pedunculated lesions with invasion into Haggitt levels 1, 2, and 3. Level 4 pedunculated lesions and sessile lesions in which the invasion is into Sm1 or Sm2 level also have low risk if there are no adverse factors. These lesions can be treated by a complete local excision. Lesions that have high risk of lymph node metastasis are those with invasion into the lower third of the submucosa (Sm3), lesions that contain lymphovascular invasion, and lesions sited in the lower third of the rectum. These lesions require an oncologic colorectal resection. For lesions in the distal third of the rectum, a per anal full-thickness excision followed by an adjuvant chemoradiation may be an alternative. The box below summarizes malignant colorectal polyps requiring oncologic bowel resections:

Endoscopic ultrasound and endoscopic mucosal resection for rectal cancers and villous adenomas

EUS is the most accurate tool for local staging of rectal carcinoma. In addition to providing accurate T- and N-stages, EUS allows assessment of the internal and external anal sphincters. Accurate endosonographic staging directs the optimal method of management of rectal carcinoma, type of resection, and candidacy for neoadjuvant therapy. EMR may be applied to large rectal adenomas as an alternative to surgical resection in selected patients. EUS is important in discriminating lesions suitable for EMR.

Natural history of colorectal nonpolypoid adenomas: a prospective colonoscopic study and relation with cell kinetics and K-ras mutations

OBJECTIVE

Serial colonoscopic observations were prospectively conducted to elucidate the natural history of nonpolypoid tumors. Furthermore, to clarify whether cell kinetic status affects the tumor development, proliferative indices, apoptotic indices, and K-ras codon 12 point mutations on biopsy specimens were investigated.

METHODS

Seventy-five colorectal tumors, 13 polypoid and 62 nonpolypoid type (56 flat elevated and six depressed type) were studied. Proliferating and apoptotic cells were detected with anti-Ki-67 antibody and the terminal deoxynucleotidyl transferase-mediated dUTP nick-end labeling method, respectively. Point mutations at K-ras codon 12 were examined by enriched polymerase chain reaction-based restriction fragment length polymorphism assay.

RESULTS

The average follow-up period was 22 months (range 1-50). The lesions of subsequent exophytic growth, unchanged shape, depressed growth, and disappearance were observed in 0%, 92%, 0%, and 8% of polypoid type, in 39%, 39%, 13%, and 9% of flat elevated type, and in 33%, 67%, 0%, and 0% of depressed type, respectively. There was no significant difference in tumor size between initial and follow-up colonoscopy. Nonpolypoid tumors apparently changed to the exophytic growth during 2 yr or more. The tumors with exophytic growth had significantly higher proliferative indices/apoptotic indices ratios than those with unchanged morphology and disappearance/depressed growth (p < 0.05, respectively). K-ras codon 12 point mutations did not correlate with tumor development.

CONCLUSIONS

Cell kinetic status plays an important role in determining minute colorectal tumor development, but not K-ras codon 12 mutations. Minute nonpolypoid adenomas frequently tend to grow slowly, and nearly 40% of those become the exophytic growth with time. Most of minute nonpolypoid tumors seem to follow the adenoma-carcinoma sequence.

[Efficacy of contrast chromoendoscopy of the colon after oral administration of indigo carmine dye]

INTRODUCTION

Indigo carmine dye is usually spread directly over the colon in many chromoscopic techniques aiming better visualization of a lesion already detected by conventional colonoscopy. Examination of the colon already stained by oral administration of indigo carmine dye may increase detection of small lesions resulting in higher sensibility of the colonoscopy in diagnosing diminutive lesions.

OBJECTIVE

Analyze the results regarding the quality of chromoscopic technique and the indigo carmine dye distribution over the colon after oral administration.

PATIENTS AND METHODS

Fifty patients undergoing colonoscopy were evaluated. A capsule containing 100 mg of indigo carmine dye was offered to these patients 30 min before oral mannitol prep routinely used. The indigo carmine dye contrast effect was graded as bad, regular or good according to preestablished criteria in three segments of the colon: right and left colon and the rectum.

RESULTS

In the right colon, good indigo carmine dye contrast effect was observed in only 9 (18.8%) patients, while it was considered regular and bad in 32 (66.6%) and in 7 (14.6%) patients, respectively. A good indigo carmine dye contrast effect was never observed in this series for the left colon or in the rectum. As a matter of fact, no indigo carmine dye was observed in the left colon in 80.9% and in the rectum in 92% of patients in this series.

CONCLUSION

Although it may be simple and desirable, oral administration of indigo carmine dye seems ineffective for enhancing detection of diminutive lesions by chromoscopy as result of poor colonic distribution of indigo carmine dye mainly at distal colonic sites.

Is carbon dioxide as an insufflation agent for double contrast barium enema still worthwhile after a same-day flexible sigmoidoscopy?

PURPOSE

The aim of the study was to evaluate the efficacy of a double-contrast barium enema (DCBE) with CO(2) as a distending gas after a screening flexible sigmoidoscopy for reducing abdominal distension after an examination.

PATIENTS AND METHODS

A total of 165 DCBEs for patients with positive faecal occult blood tests were randomly allocated to a CO(2) group (n = 83) and a control group (n = 82) using air. Abdominal radiograph were taken before the DCBE, immediately following it, and 30 minutes later to determine residual bowel gas. The degree of abdominal discomfort was recorded using a scoring system.

RESULTS

The patients in the CO(2) group had significantly smaller amounts of residual gas 30 minutes after the DCBE than did the control group, both in the colon (P < 0.02) and in the small intestine (P < 0.001). There was significantly less abdominal distension after the DCBE in the CO(2) group compared to the control group (P < 0.001). The abdominal discomfort score at 30 minutes after CO(2) correlated with the residual gas in the small intestine (r = 0.390,P = 0.003) but not in the colon (r = 0.155, P = 0.15). The quality of the images and the number of polyps detected did not differ between two groups.

CONCLUSIONS

We concluded that using CO(2) when administering a DCBE after a screening sigmoidoscopy was useful for relieving abdominal discomfort following the examination; furthermore, it did not impair the diagnostic ability of the DCBE.

High-resolution chromoendoscopy for classifying colonic polyps: a multicenter study

BACKGROUND

Chromoendoscopy may reliably separate adenomatous from nonadenomatous polyps. The aim of this multicenter trial was to determine the accuracy of high-resolution chromoendoscopy for the determination of colonic polyp histology.

METHODS

This multicenter trial included 4 academic centers and a primary care practice. In 299 patients referred for routine colonoscopy or sigmoidoscopy, 520 polyps 10 mm in size were sprayed with indigo carmine dye. Using a high-resolution endoscope, the endoscopist predicted the histology of each polyp based on its surface characteristics. Hyperplastic polyps had a "pitted" surface pattern of orderly arranged "dots" that resembled surrounding normal mucosa. Adenomatous polyps had at least one surface "groove" or "sulcus." Each polyp was subsequently resected for histopathologic evaluation.

RESULTS

The resected polyps were comprised by 193 adenomas (37%), 225 hyperplastic polyps (43%), and 102 "other" types (20%). Forty polyps (7.7%) could not be classified by high resolution chromoendoscopy with indigo carmine dye. For the remaining polyps, the sensitivity, specificity, and negative predictive value of indigo carmine dye staining for adenomatous polyps were, respectively, 82%, 82%, and 88%. The results were consistent among the academic centers and the primary care practice.

CONCLUSIONS

High-resolution chromoendoscopy with indigo carmine dye demonstrates morphologic detail of diminutive colorectal polyps that can reliably be used to separate adenomatous from nonadenomatous polyps.