Kudo Classification Research Articles

508 Surveillance Colonoscopy of Neoplasm Associated With Ulcerative Colitis by Chromo Endoscopy and Magnifying Endoscopy

A higher incidence of colorectal cancer has been recognized in patients with long-standing ulcerative colitis (UC). Surveillance colonoscopy with step biopsies for detecting UC-associated neoplasm has been recommended. However, this method needs tremendous efforts with a low detection rate. On the other hand, pit pattern diagnosis (Kudo's classification) is useful for assessing of the sporadic neoplasm, but the surface structures of UC-associated neoplasm have not been well elucidated. Aims and methods: The aim of this study is to investigate the validity of pit pattern diagnosis for UC-associated neoplasm.

Expression of IGF-1R in Colorectal Polyps and its Role in Colorectal Carcinogenesis

Insulin-like Growth Factor Receptor 1 (IGF-1R) may play a role in the neoplastic progression of colorectal cancer because it is related to both cellular proliferation and differentiation. The aim of this study was to further elucidate the role of IGF-1R in colorectal carcinogenesis by evaluating IGF-1R expression in different types of precancerous colorectal polyps and comparing its expression to normal mucosa and colorectal carcinoma. A total of 47 colorectal polyps and their respective adjacent normal mucosa were collected from 32 patients. In addition, 20 colorectal adenocarcinoma tissues were obtained from patients undergoing colorectal resection, and 12 normal non-malignant colorectal mucosal tissues collected from outpatients served as the control group. The pit patterns of polyps were classified by the Kudo classification scheme through magnifying chromoendoscopy. Immunohistochemistry and quantitative real-time RT-PCR were utilized for expression analysis of IGF-1R in colorectal mucosa, polyps, and adenocarcinoma tissue. The results of immunohistochemistry showed no significant differences in IGF-1R expression in inflammatory polyps compared with their surrounding normal mucosa by the Mann-Whitney U test (p=0.251); however, tubular adenoma and villous adenoma tissues exhibited significantly higher levels of IGF-1R expression (p=0.000). The results of real-time RT-PCR showed that IGF-1R was transcribed at a high level in colorectal adenomatous polyps and adenocarcinoma compared with their respective paired normal mucosa. Spearman's rank correlation two-variable analysis was used to demonstrate a significant correlation between the expression of IGF-1R and neoplastic progression from normal mucosa to adenomatous polyps and finally to colorectal cancer (r=0.574, p=0.000). This study suggests that the expression of IGF-1R correlates with the degree of carcinogenesis. In addition, these results demonstrated that there is a significant correlation between the level of IGF-1R expression and pit patterns of polyps (r=0.432, p=0.002). Thus, IGF-1R might be a factor in the morphological change of colorectal mucosal crypts, and it may play an important role in the growth and malignant transformation of precancerous polyps. These results suggest that IGF-1R can be considered a biomarker for the stage and risk of carcinogenesis during neoplastic initiation and progression along the colorectal normal mucosa-polyp-cancer sequence. Inhibitors of IGF-1R are not only a promising targeted anticancer strategy, but also a possible option for the chemoprevention of colorectal cancer.

NBI versus hochauflösende Endoskopie in der Diagnostik von Neoplasien bei Colitis ulcerosa

BACKGROUND AND STUDY AIMS: Controversy exists about which colonoscopic technique is most sensitive for the diagnosis of neoplasia in patients with ulcerative colitis. We compared new-generation narrow-band imaging (NBI) to high-definition endoscopy (HDE) for the detection of neoplasia and evaluated NBI for the differentiation of neoplastic from non-neoplastic mucosa. PATIENTS AND METHODS: Randomized crossover trial in which patients with ulcerative colitis under-went both NBI and HDE colonoscopy in random order with at least 3 weeks between the two procedures, which were performed by different endoscopists. Lesions detected during the first examination were left in situ in order to enable detection during the second examination as well. Main outcome measures were (1) neoplasia detection, and (2) diagnostic accuracy of NBI for -differentiating neoplastic from non-neoplastic mucosa by using the Kudo classification and vascular pattern intensity (VPI). RESULTS: Twenty-five patients were randomized to undergo HDE first and 23 to undergo NBI first. Of 16 neoplastic lesions, 11 (69 %) were detected by HDE and 13 (81 %) by NBI (P = 0.727). Of 11 patients with neoplasia, 9 (82 %) were diagnosed by HDE and 8 (73 %) by NBI (P = 1.0). The sensitivity, specificity, and accuracy of the Kudo classification were 76 %, 66 % and 67 %. Corresponding figures for VPI were 80 %, 72 % and 73 %. CONCLUSION: NBI does not improve the detection of neoplasia in patients with ulcerative colitis compared to HDE. In addition, NBI proves unsatisfactory for differentiating neoplastic from non-neoplastic mucosa.

Mo1522 The Diagnosis of Colorectal Cancer With NBI Magnifying Endoscopy, Magnifying Chromoendoscopy and Endocytoscopy

Mo1522 The Diagnosis of Colorectal Cancer With NBI Magnifying Endoscopy, Magnifying Chromoendoscopy and Endocytoscopy

Tu1550 Accuracy of Predicting Polyps Histology Using Narrow Band Imaging (NBI) With Magnification in Routine Clinical Practice

Tu1550 Accuracy of Predicting Polyps Histology Using Narrow Band Imaging (NBI) With Magnification in Routine Clinical Practice

Magnifying Chromoendoscopic Findings of Early Gastric Cancer and Gastric Adenoma

In the field of colorectal cancer and adenoma, Kudo's classification of pit pattern with magnifying chromocolonoscopy using crystal violet (CV) staining is now accepted. Magnifying endoscopy using narrow band imaging has been used for the diagnosis of gastric carcinoma; the characteristic findings of microvascular patterns have been demonstrated. However, there was limited information on magnified endoscopic findings with CV staining for gastric neoplasms in terms of their pit patterns. Magnifying chromoendoscopy with CV was performed in 175 patients with early gastric cancer and 18 with gastric adenoma, prior to treatment. Surface patterns of gastric tumors were classified into five types: (1) long tubular pit pattern, (2) irregular size pit pattern, (3) small round pit pattern, (4) destroyed pit pattern, and (5) non-structural pattern. Long tubular pit pattern was most common in gastric adenoma. Well differentiated adenocarcinoma and papillary adenocarcinoma tended to show different size of pit pattern or destroyed pit pattern. Small round pit pattern was most commonly seen in moderately differentiated adenocarcinoma. Non-structural pattern was most frequently observed in poorly differentiated adenocarcinoma and signet ring cell carcinoma (P<0.0001). For gastric neoplasms, magnifying endoscopy may help predict histopathological type.

Narrow-band imaging versus high-definition endoscopy for the diagnosis of neoplasia in ulcerative colitis

Controversy exists about which colonoscopic technique is most sensitive for the diagnosis of neoplasia in patients with ulcerative colitis. We compared new-generation narrow-band imaging (NBI) to high-definition endoscopy (HDE) for the detection of neoplasia and evaluated NBI for the differentiation of neoplastic from non-neoplastic mucosa. Randomized crossover trial in which patients with ulcerative colitis underwent both NBI and HDE colonoscopy in random order with at least 3 weeks between the two procedures, which were performed by different endoscopists. Lesions detected during the first examination were left in situ in order to enable detection during the second examination as well. Main outcome measures were (1) neoplasia detection, and (2) diagnostic accuracy of NBI for differentiating neoplastic from non-neoplastic mucosa by using the Kudo classification and vascular pattern intensity (VPI). Twenty-five patients were randomized to undergo HDE first and 23 to undergo NBI first. Of 16 neoplastic lesions, 11 (69 %) were detected by HDE and 13 (81 %) by NBI ( P = 0.727). Of 11 patients with neoplasia, 9 (82 %) were diagnosed by HDE and 8 (73 %) by NBI ( P = 1.0). The sensitivity, specificity, and accuracy of the Kudo classification were 76 %, 66 % and 67 %. Corresponding figures for VPI were 80 %, 72 %, and 73 %. NBI does not improve the detection of neoplasia in patients with ulcerative colitis compared to HDE. In addition, NBI proves unsatisfactory for differentiating neoplastic from non-neoplastic mucosa.

Diagnostic accuracy of narrow-band imaging and pit pattern analysis significantly improved for less-experienced endoscopists after an expanded training program

Previous reports assessing diagnostic skill using narrow-band imaging (NBI) and pit pattern analysis for colorectal polyps involved only highly experienced endoscopists. To evaluate diagnostic skills of less-experienced endoscopists (LEE group) for differentiation of diminutive colorectal polyps by using NBI and pit pattern analysis with and without magnification after an expanded training program. Prospective study. This study involved 32 patients with 44 colorectal polyps (27 adenomas and 17 hyperplastic polyps) of < or =5 mm that were identified and analyzed by using conventional colonoscopy as well as non-magnification and magnification NBI and chromoendoscopy followed by endoscopic removal for histopathological analysis. Before a training course, 220 endoscopic images were distributed in randomized order to residents with no prior endoscopy experience (NEE group) and to the LEE group, who had performed colonoscopies for more than 5 years but had never used NBI. The 220 images were also distributed to highly experienced endoscopists (HEE group) who had routinely used NBI for more than 5 years. The images were distributed to the NEE and LEE groups again after a training class. Magnification NBI and chromoendoscopy images were assessed by using the Sano and Kudo classification systems, respectively. Diagnostic accuracy and interobserver agreement for each endoscopic modality in each group. Diagnostic accuracy was significantly higher, and kappa (kappa) values improved in the LEE group for NBI with high magnification after expanded training. Diagnostic accuracy and kappa values when using high-magnification NBI were highest among endoscopic techniques for the LEE group after such training and the HEE group (accuracy 90% vs 93%; kappa = 0.79 vs 0.85, respectively). Study involved only polyps of < or =5 mm. Using high-magnification NBI increased the differential diagnostic skill of the LEE group after expanded training so that it was equivalent to that of the HEE group.

S1587: Surveillance Colonoscopy of Neoplasm Associated With Ulcerative Colitis by Chromoendoscopy and Magnifying Endoscopy

A higher incidence of colorectal cancer has been recognized in patients with long-standing ulcerative colitis (UC). Surveillance colonoscopy with step biopsies for detecting UC associated dysplasia has been recommended. However, this method needs tremendous efforts with a low detection rate. The chromoendoscopy and magnifying endoscopy are helpful to detect colonic neoplasm. Pit pattern diagnosis classified (Kudo's classification) is also useful for assessing of the colonic neoplasm. On the other hand, the surface structures of UC neoplasm have not yet been well elucidated.

S1567: Validation of a Novel Polyp Classification System and Comparison of Outcomes Between a Japanese and Western Expert: A Prospective Study

Small hyperplastic Polyps have a very low malignant potential. However, they are currently being removed due to the lack of in-vivo histology techniques. Japanese endoscopists have published excellent results based on Kudo's pit pattern. However, most of the published literature has used zoom endoscopy and vital stains, such as cresyl violet, and Kudo's classification is difficult and inpractical to apply in a UK setting.

Risk Factors for Lymph Node Metastasis in Patients with Submucosal Invasive Colorectal Carcinoma

Purpose: This study was designed to determine the risk factors of lymph node (LN) metastasis in patients with submucosal invasive colorectal cancer (SICC). Methods: Between January 1998 and January 2009, we reviewed patients who had undergone radical colon resection with LN dissection for SICC. Results: There were 36 males and 40 females (mean age, 61.1 years; range, 35∼86 years). In the univariate analysis, the risk of LN metastasis was related to the depth (absolute and relative), lymphovascular invasion, tumor budding, and tumor differentiation (P＜0.05). The relative depth by Kudo classification and lymphovascular invasion were significant predictors of LN metastasis both in univariate and multivariate analysis. In SICC with an absolute depth ＜1,800㎛, no LN metastasis was detected. Regardless of the size of the SICC, tumors that invaded within the sm2 layer and had no lymphovascular invasion had no LN metastasis. Conclusion: In the SICC, lymphovascular invasion and depth of submucosal invasion are strong predictors of LN metastasis. If deep invasion exceeds sm2 and positive lymphovascular invasion exists in the resected specimen, additional colectomy with LN dissection appears to be necessary.

Endoscopic features and carcinoembryonic antigen expressions of colorectal serrated adenomas

Objective To explore the endoscopic features and carcinoembryonic antigen （CEA） expressions of colorectal serrated adenomas （SA）. Methods From June 2005 to July 2008, 27 patients with colorectal polyps and 26 cases of advanced colorectal cancer （ACC） were enrolled in the study. The pit pattern of the suspected lesions were observed with 0. 4% indigo carmine sodium stain and classified according to Kudo classification. The polyps were removed by biopsy, high-frequency electrical excision or endoscopic mucosa resection and the samples of ACC were collected with biopsy forceps. All specimens underwent rou- tine pathological examination and CEA expression was detected by immunohistochemistry. Results There were 47 SAs and 27 other types of polyps in polyp group. Most SAs located in left-side colon and were characterized by the bulge semipedunculation. The diameters of 45 serrated adenomas ranged from 3 to 8mm, and the maximal diameter of other 2 were more than lOmm. The most common type of pit pattern in SA was mixed type Ⅱ and ⅢL, with type II predominant in a certain percentage. Expression of CEA in SA glands was significantly higher than that in proliferative polyps （P 〈 0. 01 ）. Conclusion SA is an unique type of colorectal adenoma, which is capable of expressing CEA, and displays malignant potential and deserves great attention. Key words: Adenoma; Endoscopy, digestive system; Carcinoembryonic antigen

Ultrastructure of colon pit pattern classification

Objective To explore morphological changes in evolution of colorectal pit patterns and their relation with polyp development by observing the ultrastructure of colorectal polyps. Methods Analyses were based on data from 50 subjects with colorectal polyps diagnosed with magnifying colonoscopy. Colonoscbpy combined with endoscopic lesions staining was employed to determine the pit pattern of colorectal lesions, according to Kudo classification. Colorectal lesions were removed through endoscopy or surgery. All samples were analyzed with histopathology, scanning and transmission electron microscopy. Results The pit phttern or 50 samples included 10 cases of type Ⅰ , 8 cases of type Ⅱ, 15 cases of type Ⅲ, 9 cases of type Ⅳ, 5 cases of type Ⅴ and 3 cases of mixed-type. The positive predictive rate and the accuracy rate of differential diagnosis of neoplastic polyps, verified with histological diagnosis, were 86% and 94% , respectively. Under scanning electron microscopy, crypt morphology in type Ⅰ and type Ⅱ exhibited regular round hole shape, with epithelial cells in uniform size, abundant absorptive cells with rich microvilli and goblet cells. Crypt in type Ⅲ, type Ⅳ and type Ⅴ appeared gradual deformation accompanied by increasingly atypical hyperplasia in the epithelial cells and heterotypic tissues. Transmission electron microscopy revealed epithelium of type Ⅰ and type Ⅱ arranged regularly and connected closely, with neat microvilli at the top of cells, affluent mitochondria and endoplasmic reticulum in the cytoplasm, oval nuclear at the base, and smooth and continuous basement membrane. While the epithelium of type Ⅲ, type Ⅳ and type Ⅴ was characterized by sparse and irregular villi, swelling mitochondria and expanded endoplasmic reticulum, lysosomes of increased number, heterotypic nuclear and mitotic nuclear, and broken and disappearing substrate. Conclusion Morphological changes in pit pattern of colorectal polyps is affected by surface structure of glandular epithelial cells, the proportion and composition of the crypt surface cellsn. Ultrastructural changes in epithelial cells are the material basis of morphological changes in pit pattern and polyp development. Key words: Scanning electron microscopy; Transmission electron microscopy; Pit pattern

Estudo comparativo entre MBI (FICE®) e a magnificação com índigo-carmin no diagnóstico diferencial de lesões neoplásicas e não-neoplásicas de cólon e reto

Multiband imaging (MBI)/Fuji Intelligent Color Enhancement (FICE) is a spectral image processing technology that helps in vivo diagnosis of colorectal neoplasias. To compare the diagnostic accuracy of the magnification with either the electronic chromoendoscopy or indigo carmine dye in the differential diagnosis of neoplastic and non-neoplastic colorectal lesions. Seventy five patients with 157 colorectal lesions were prospectively evaluated. The capillary pattern, as well as the pit pattern according to the Kudo classification, of colorectal lesions were evaluated by means of the FICE system. Absence and presence of meshed capillary networks were labeled as non-neoplastic and neoplastic lesions, respectively. Afterwards, indigo carmine 0.8% was instilled and a new evaluation of the pit pattern was carried out. One hundred and sixteen of the 157 lesions were classified as positive meshed capillary network, 115 of them were confirmed histologically as neoplasia. Other 32 lesions out of 41 with negative meshed capillary network were non-neoplastic. Sensitivity, specificity and accuracy were, respectively, 92.7%, 97% and 93.6%. Pit patterns I and II were confirmed as non-neoplastic lesions, and patterns III to V were confirmed as neoplasias. Sensitivity, specificity and accuracy for the electronic chromoendoscopy were, respectively, 94.4%, 97% and 94.9%. Meanwhile, the figures for the magnification with indigo carmine were, respectively, 97.6%, 93.9% and 96.8%. Both methods, either the MBI/FICE system or the use of indigo carmine dye with magnification, achieved a high accuracy for the differential diagnosis between neoplastic and non-neoplastic colorectal lesions.

Combining Autofluorescence Imaging and Narrow-Band Imaging for the Differentiation of Adenomas from Non-Neoplastic Colonic Polyps Among Experienced and Non-Experienced Endoscopists

Endoscopic tri-modal imaging incorporates high-resolution white-light endoscopy (HR-WLE), narrow-band imaging (NBI), and autofluorescence imaging (AFI). Combining these advanced techniques may improve endoscopic differentiation between adenomas and non-neoplastic polyps. In this study, we aimed to assess the interobserver variability and accuracy of HR-WLE, NBI, and AFI for polyp differentiation and to evaluate the combined use of AFI and NBI. First, still images of 50 polyps (22 adenomas; median 3 mm) were randomly displayed to three experienced and four non-experienced endoscopists. All HR-WLE and NBI images were scored for Kudo classification and AFI images for color. Second, the combined AFI and NBI images were assessed using a newly developed algorithm by six additional non-experienced endoscopists. The outcomes measured were interobserver agreement and diagnostic accuracy using histopathology as reference standard. Experienced endoscopists had better interobserver agreement for NBI (kappa=0.77) than for AFI (kappa=0.33), whereas non-experienced endoscopists had better agreement for AFI (kappa=0.58) than for NBI (kappa=0.33). The accuracies of HR-WLE, NBI, and AFI among experienced endoscopists were 65, 70, and 74, respectively. Figures among non-experienced endoscopists were 57, 63, and 77. The algorithm was associated with a significantly higher accuracy of 85% among all observers (P<0.023). These figures were confirmed in the second evaluation study. Non-experienced endoscopists have better interobserver agreement and accuracy for AFI than for HR-WLE or NBI, indicating that AFI is easier to use for polyp differentiation in non-experienced setting. The newly developed algorithm, combining information of AFI and NBI together, had the highest accuracy and obtained equal results between experienced and non-experienced endoscopists.

The Clinical and Pathological Features of Depressed Type Colorectal Neoplasms and the Significance of Pit Pattern Diagnosis

Introduction: At one time, most colorectal carcinomas were thought to develop from protruded polyps. Recent advances in endoscopic diagnosis and colonoscope technology have revealed the existence of many flat and depressed types of neoplasms. GIE, published in Oct. 2008, widely introduced those lesions as nonpolypoid mucosal colorectal neoplasms (NMCN). We have detected more than 600 depressed type neoplasms at Akita Red Cross Hospital and our Unit during 1985-2008. Depressed type neoplasms are assumed to emerge directly from normal epithelium without going through a stage of adenoma, which is mentioned as the “de novo” carcinogenesis. “De novo” sequence possibly may be main route of development of colorectal cancer. Aim: To evaluate the clinical and pathological natures and validity of pit pattern diagnosis (Kudo's classification) on depressed type colorectal neoplasms. Methods: From Apr. 2001 to Jun. 2008, we retrospectively evaluated 10,212 colorectal neoplasms, excluding advanced cancers, which removed endoscopically or surgically at our Unit. We evaluated a correlation between submucosal invasive rates and size, morphological features and pit patterns. Result: The submucosal invasive rate of depressed type neoplasms reached 64.1% (98/153). Meanwhile, the overall invasive rates of flat and protruded type neoplasms were 2.6% (98/3772) and 3.2% (204/6288), respectively. The submucosal invasive rates were 6.5% in lesions not exceeding 5 mm, 60.8% in those from 6-10 mm, and 89.2% in those from 11-15 mm in diameter. In contrast, the invasive rates in flat type and protruded type neoplasms were 0% and 0.04% for those less than 5mm in diameter. Most (90.9% and 91.7%) of the protruded and flat type neoplasms showed type IIIL or IV pit patterns, whereas 94.3% of the depressed type neoplasms were characterized by type IIIS, VI, or VN pit patterns. Only 0.63% of the lesions with type IIIL, and IV pit patterns were invasive cancers. On the other hand, 2.4% of the lesions with type IIIS pit pattern, 34.5% of those of type VI and 91.1% of those of type VN were invasive carcinomas regardless of its gross appearance. Type VI reflected histological structural atypism outgrowth of cancerous glands, and type VN, which lacks of superficial microstructure, reflected exposing desmoplastic reaction of deeply invasive submucosal components to the surface. Conclusion: We have revealed that depressed type neoplasms grow rapidly and invade the submucosa even at an early stage. With magnifying colonoscopies, the pit pattern diagnosis, which correlated well with the histological diagnosis, is available for assessing the depth of the lesions and selecting appropriate therapies.

A Comparison of Magnifying Chromoendoscopy Versus Histopathology of Forceps Biopsy Specimen in the Diagnosis of Minute Flat Adenoma of the Colon

Having noted a discrepancy between endoscopic and histopathological diagnoses in cases of minute adenomas of the colon, a prospective study was designed to clarify which is appropriate, magnifying chromoendoscopy or histopathology of a specimen obtained by biopsy forceps. A total of 208 patients comprised the study population. The endoscopic diagnoses were performed with magnifying colonoscopies. We separated the detected lesions with type III(L) pit pattern following Kudo's classification into two groups at random: in group A (n = 104) resected specimens were fixed with 20% buffered formalin without being flattened, whereas in group B (n = 104) the resected specimens were flattened using forceps before fixation and the specimens were cut under observation of their surface structure with stereomicroscopy. Comparison of the initial diagnoses between groups A and B showed that a total of 84.6% (88/104) of the lesions were diagnosed to be tubular adenomas histopathologically in group A, compared with 100% (104/104) in group B (P < 0.0001). Results for comparison of the secondary diagnoses between group A and group B showed that 14 of the 16 lesions were diagnosed as tubular adenomas histopathologically. Thereafter, 98.1% (102/104) of the lesions were diagnosed to be tubular adenomas histopathologically in group A (P = 0.4976). In conclusion, high-resolution magnifying chromoendoscopy is an appropriate procedure for the diagnosis of minute adenomas in comparison with histopathology of specimens obtained by biopsy forceps in this prospective study.

Morphogenesis of a colorectal neoplasm with a type IIIS pit pattern inferred from isolated crypts

The goal of this study was to clarify morphogenetic development in colorectal neoplasms with depressions by examining isolated crypts and their pit patterns. Twelve colorectal neoplasms with depressions were examined endoscopically, stereomicroscopically, and histopathologically. The pit pattern was defined according to Kudo's classification. The HCl digestion method was used to isolate the crypts. For scanning electron microscopy observations of the surface structure, specimens were mounted on brass stubs, coated with gold, and then observed. In isolated crypts with bifurcation extending to the middle or all the way to the lesion surface, the crypt orifice was round with a type IIIS pit pattern. In those with bifurcations extending up to the upper portion, the crypt orifice was oval with a type IIIS pit pattern. Some crypts were almost completely separated into two. The orifice diameter of a nonbifurcated crypt was 95.1 +/- 28.1 microm, whereas the diameters of crypts with bifurcations extending to the middle or the upper part were 114.7 +/- 40.8 and 208.1 +/- 71.4 microm, respectively. Thus, the orifice diameter of a crypt with the bifurcation reaching the upper part was about twice that of orifices of nonbifurcated crypts or those with bifurcations extending to the middle. Our results indicate that colorectal neoplasms with depressions may enlarge by a process of crypt bifurcation in which a single crypt divides into two crypts.

Comparison of white light and narrow band high definition images in predicting colon polyp histology, using standard colonoscopes without optical magnification

Narrow band imaging (NBI) with optical magnification is useful in predicting colon polyp histology. As magnifying endoscopes are not routinely available, we investigated the use of NBI and high definition white light imaging in determining polyp histology, using images obtained with colonoscopes without optical magnification. Images (white light and NBI) of colon polyps less than 10 mm in diameter were collected prospectively from patients undergoing screening colonoscopy and digitally stored. Two endoscopists later reviewed all images and predicted polyp histology as neoplastic or non-neoplastic using a modified Kudo classification. Comparison was made with histopathology. Separate white light and NBI images of 80 polyps (49 neoplastic, 31 non-neoplastic) from 63 patients were recorded. Mean polyp size was 5.1 +/- 2.1 mm (5.4 +/- 2.2 neoplastic; 4.4 +/- 1.8 non-neoplastic; P = 0.02). In a pooled analysis, NBI correctly predicted neoplastic histology in 93 of 98 images (sensitivity 95 %, positive predictive value [PPV] 94 %) whereas white light did so in 58 of 98 images (sensitivity 59 %, PPV 79 %). NBI correctly predicted non-neoplastic histology in 56 of 62 images (specificity 90 %, negative predictive value [NPV] 92 %) whereas white light did so in 47 of 62 images (specificity 76 %, NPV 54 %). NBI without optical magnification was more accurate in predicting colon polyp histology compared with white light imaging. Image quality and confidence in histology were significantly higher in the NBI group. NBI without optical magnification may be useful in predicting colon polyp histology.

Is Observation of Vascular Appearance of the Tumor Surface Useful for Diagnosis of the Depth of Early Colorectal Cancer?

Background: Early colorectal cancer with depth of the submucosal invasion is under 1000 μm can usually be treated with endoscopic mucosal resection (EMR) in Japan, because most of such lesion has no metastasis including local lymphnodes. However, submucosal invasive cancer which is deeper than 1000 μm requires surgical resection. The tumor surface pit pattern of Kudo's classification is widely accepted as an useful method for diagnosis of colorectal lesion. Although the type V pit pattern lesion suggested the presence of colorectal cancer, the depth of tumor invasion is difficult to diagnose by the structure pit pattern.