Flat colon lesions, which include superficial colon tumor and depressed colon tumor, have close relationship with colon cancers. The main type of flat colon tumor is laterally spreading tumor (LST) while the main type of depressed tumor is depressed early colon cancer (IIc type). Little has been reported about the flat colon tumor because it is difficult to be detected and has not yet been broadly recognized by clinicians. Recently, clinicians with the Nanfang Hospital have innovated some novel colonoscopic approaches to improve the diagnosis and treatment of flat colon tumor and accumalate experiences in this regard. Their conclusions are: 1. Improvement in recognizing the physical characteristics of the lesion and the application of chromoscopic colonoscopy are the key points to reveal and identify the flat colon tumors. 2. The best theraputic methods for flat colon lesions are endoscopic mucosal resection and piecemeal endoscopic mucosal resection.

Objective To investigate the effect of p21-activated kinase 1 on chemotherapy sensitivity of 5-fluorouracil. Methods Cell proliferation was measured by CCK8 and apoptosis rate by flow cytometry or Hoechst staining; the expression of Bcl-xl, Bcl-2, XIAP were determined by Western Blot. Results 5-FU combined shRNA-Pak1 group (combination group) could be significantly inhibited in terms of proliferation (P <0.05). The percentage of apoptosis rate in combined group was the highest and the difference among groups indicated statistical significance (P < 0.05). The expression of Bcl-xl, Bcl-2, XIAP in combination group was significantly inhibited compared with 5-FU group or shRNA-Pak1 group. Conclusion PAK1 inhibited by RNA interference can enhance chemotherapy sensitivity of 5-Fu on growth inhibition and apoptosis induction in colon cancer significantly.

Objective To establish a proliferating model of hepatic oval cells (HOCs) with adult Wistar rats, isolate and culture in vitro the HOCs, and to approach the possibility of inducing the HOCs differentiated into hepatocytes. Methods Rats were fed with 0.07% wt/wt ethionine. On day 8, 2/3 partial hepatectomy (2/3 PH) was performed. Isolate, harvest and purify HOCs by type Ⅳ collagenase perfusion with semi in situ 2-step method and Percoll density gradient method. Epidermal growth factor (EGF) and hepatocyte growth factor (HGF) were added to complete Williams' medium E (WE) to culture the HOCs. HOCs was induced differentiation with HGF, oncostatin m (OSM) and fibroblast growth factor-4 (FGF4) into hepatocytes. Results The concentration of cell after purification was about 1.34?105/ml. Most cells were small, about 1/6~1/3 the size of normal hepatocyte. Ovoid, elliptical or polygonal in shap, the nucleus-cytoplasm ratio was relatively large, and clone-like proliferation appeared 2 weeks later. LSCM revealed positive expression of Thy-1 and C-kit in cytoplasm and membrane of HOCs. ICC showed AFP in cytoplasm of HOCs. Stimulated by inducing system, the shape of HOCs changed gradually. The volume enlarged and cells lost their adherence ability. ICC indicated apparent positive stain of cytoplasm Alb 14 days after differentiated induction, and the positive ratio increased along with the extension of induction duration. Cytochemical tests indicated brown or black sediment with G-6-P staining and red particles with PAS staining, respectively. Conclusion The proliferation model of rats' HOCs was established after ethionine feeding and 2/3 PH. HOCs can be obtained with type IV collagenase perfusion and Percoll density gradient isolation and purification. Clone proliferation can be achieved through culturing HOCs in vitro. Under certain conditions, HOCs can be induced and differentiated into certain typical hepatocyte phenotype.

Objectives In order to further investigate the biological characteristics and the developmental mechanisms of laterally spreading tumor, a cell line from the primary culture of human colon LST tissue was isolated. After 70 passages, we identified the cells by morphologic, genetic and immunohistochemistry analysis. Results (1) The origin of LST cells is epithelial, and the majority of which were multiform epithelial cells. It displayed an approximately similar growth characteristics of tumor cells. The nest corpus doubling time was 36 hours. (2) The number of chromosome is 42-66. Eighty-five percent of chromosomes were triploid, showing abnormal karyotypes. (3) Immunohistochemistry showed that ESA and CK20 were expressed. (4) The ultrastructure of LST was almost the same as that of a tumor cell. Conclusion A novel LST cell line has been established and named as LST-R1. It facilitates further study of the biological characteristics of LST cells and the development of colon cancer.

Objective To eraluate the methods and results of endoscopic mucosal resections for colonic laterally spreading tumors with early malignant degeneration. Methods The pit pattern was studied with magnitying colonoscopy and mucosal staining technique for colonic laterally spreading tumors (LST) undergoing early earcinatous degeneration. They were removed with endoscopic mucosal resection techniques. A follow-up study was made. Results Eight patients suffering from early carcinoma on top of colonic laterally spreading tumors were followed-up. ① 75% of the lesions were situated in the rectum, sigmoid colon and decending colon. ② 75% of the lesions were larger than 30mm in diameter. ③ 75% of the lesions belonged to Ⅳpit pattern. ④ 62.5% of the lesions were Villous adenomas and 75% of the onalignant change involved the mucosa only. ⑤ All the lesions were completely removed by EMR, ⑥ No local residual lesion, recurrence or metastasis was discovered in all the patients after a mean follow-up period of 20.7 months. Conclusions Most of the early carcinomas originated from colonic LST involved the mucosa only. Endoscopic mucosal resection may be a curative method for the early cancer arising from colonic LST.

Objective The laterally spreading tumor (LST) of the large intestine have received a great deal of study in Japan, but no report of such cases was found in China. We here of report 46 cases of LST (47 lesions, because one of the patients was found to have two lesions) diagnosed in our hospital in last two years. The morphological features under endoscopy, pit pattern, pathology and treatment under endoscopy were studied. Methods The 47 lesions of LST were found in a total of 5 735 patients who underwent colonoscopic examination in the last two years (from 01-12-2000 to 30-11-2002). The site, size and morphological features of LST were recorded, and the pit pattern was classified with magnification. 42 patients (totally 42 lesions) received colonoscopic surgery, and the rest refused to accept endoscopic therapy, therefore only a biopsy was done for pathological study. In those patients who accepted colonoscopic surgery, 14 patients were treated by mucosal resection (EMR) after injection of saline solution, 24 were treated by piecemeal mucosal resection (EPMR) after injection of saline solution, in 3 patients the lesion was removed by high frequency loop resection, and Cap-EMR was performed using distal hyaline hat attachment fitted to the end of endoscope. The resected or biopsy specimens were studied pathologically. Results 47 lesions of LST were found in a total of 5863 patients who underwent colonoscopic examination. The distrihution of LST lesions were as follows. 22 lesions were situated in the rectum, 10 lesions in the sigmoid colon, 7 lesions in the descending colon, 4 lesions in the transverse colon, 2 lesions in the ascending colon, and 2 lesions were located in the cecum. Macroscopic type: 25 lesions were homogeneous granular type, 12 lesions were mixed nodular granular type, 8 lesions were flat elevated type, and 2 lesions were pseudo-depressed type. The pit pattern as revealed under magnifying endoscopy indicated that 2 lesions were of type II pit, 8 lesion of type II+III L pit, 9 lesions of type III L pit, and 28 lesions of type IV. The pathological diagnosis revealed that 28 lesions were villous adenomas in which 7 lesions showed early malignant degeneration (6 intramucosal cancer and 1 submucosal cancer). 11 lesions were tubular adenomas, in which 1 was intramucosal cancer, 2 lesions were hyperplastic polyp, and 6 lesions were mixed hyperplastic polyps (serrated adenomas). Conclusion The LST manifest some special features, and their incidence may be higher in China than in Japan. EMR or EPMR may be considered as a good method for treatment for LST.

Objective To evaluate the efficacy and safety of high-frequency electric dissection for colorectal cysts.Methods The personal information,clinical data,operation methods and postoperative complications of patients who were diagnosed as having colorectal cysts and underwent high-frequency electric dissection in Nanfang hospital and Zhujiang Hospital of Southern Medical University between January 1st,2005 and July 1st,2015 were analysed.All patients enrolled in the study were followed up to obtain their resuits of colonoscopy.Results A total of 63 patients were enrolled into our study,9 lesions located in the ileocecus,17 in the ascending colon,19 in the transverse colon,10 in the descending colon,7 in the sigmoid colon and 1 in the rectum.The maximum diameter of the cysts was 20.2+7.5 mm (5-40 mm).All patients underwent high-frequency electric dissection to remove the cysts completely or part of the cyst wall for drainage.Hemorrhage occurred in only one patient and bleeding stopped after being clipped by Titanium clip.Forty-five patients were followed up and there were no delayed complications or recurrence during a postoperative follow-up of 24.1 + 14.3 months (6-87 months).Conclusion High-frequency electric dissection is a safe and effective procedure for the treatment of colorectal cysts.

Objective To explore clinical pathological characteristics and treatment changed with time in patients with colorectal laterally spreading tumors (LST) from 2001 to 2015.Methods The clinical data of 549 patients with colorectal LST who received endoscopic resection or surgical operation between 2001 and 2015 were retrospectively collected.According to the time of diagnosis,patients were divided into 2001 to 2005,2006 to 2010 and 2011 to 2015 groups.The gender,age,lesion size and lesion subtypes,clinical pathologic features and their therapeutic methods were analyzed.Chi-square test was used for statistical analysis.Results The detective rates of colorectal LST in 2001 to 2005 period,2006 to 2010 period and 2011 to 2015 period were 0.38％ (50/13 319),(0.60％ (144/23 912) and 0.79％ (355/ 44 715),respectively,and the differences were statistically significant (x2 =29.34,P ＜ 0.01).During these three period,the male to female ratio was about 1:1,mean age about 59 years old,and the mean maximum diameter of the LST lesions remained about 30 mm.The percentages of granular type laterally spreading tumor (LST-G) in 2001 to 2005 period,2006 to 2010 period and 2011 to 2015 period were 82.4％(42/51),67.7％ (105/155) and 78.2％(283/262),respectively;while those of non-granular type laterally spreading tumor (LST-NG) were 17.6 ％ (9/51),32.3 ％ (50/155) and 21.8 ％ (79/362),respectively;and the differences were statistically significant (x2 =7.77,P =0.02).The proportions of LST located at the proximal colon in the three periods were 21.6 ％ (11/51),34.2 ％ (53/155) and 41.4 ％ (150/362),respectively;while the percentages of LST at distal colon were 78.4％ (40/51),65.8％ (102/ 155) and 58.6 ％ (212/362),respectively;and the differnces were statistically significant (x2 =8.61,P=0.01).The percentages of high grade neoplasia (HGN) in the three periods were 13.7 ％ (7/51),21.9 ％(34/155) and 48.6％(176/362),respectively;while the percentages of invasive carcinoma were 2.0％(1/51),5.2％ (8/155) and 8.3％ (30/362),respectively;and the differnces were statistically significantly (x2 =58.89,P＜0.01).The percentages of endoscopic mucosal resection (EMR) in the three periods were 56.9％(29/51),58.7％ (91/155) and 32.0％ (116/362),respectively;the percentages of endoscopic piecemeal mucosal resection (EPMR) were 41.2 ％ (21/51),23.9 ％ (37/155) and 14.1％ (51/362),respectively;the percentages of endoscopic submucosal dissection (ESD) were 0,12.3％ (19/155) and 46.1 ％ (167/362),respectively;the percentages of surgical operation were 0,5.2 ％ (8/155) and 7.7 ％ (28/362),respectively;and the differences were statistically significant (x2 =112.46,P＜ 0.01).Conclusions From 2001 to 2015,the clinical pathological features and therapeutic methods of colorectal LST changed along with time.The proportion of colorectal LST located at proximal colon increased,and the percentage of LST-G decreased.ESD became the primary treatment,and the proportion of pathological diagnosis of HGN and invasive carcinomas increased after operation.

Objective To investigate whether there has any difference of gastric and small bowel transit time and completion rates between two capsule endoscopes with different size and weight. Methods Clinical data of patients who had undergone OMOM or MiroCam (smaller and lighter than OMOM) capsule endoscopy were retrospectively studied. Comparison of gastric and small bowel transit time and completion rates were made between the two kinds of capsule endoscopy. Results 1, 448 patients (628 in OMOM group and 820 in MiroCam group) were finally includ-ed. In patients with Crohn's disease or suspected Crohn's disease, gastric transit time of OMOM was significantly longer than that of MiroCam [(53.4 ± 52.6) minutes vs (41.1 ± 47.9) minutes, = 0.022]. In patients with gastroin-testinal bleeding, gastric transit time in OMOM was significantly shorter than that in MiroCam [(42.1 ± 44.8) minutes vs (62.0 ± 78.6) minutes, = 0.016). No significant difference in small bowel transit time or completion rate was found. Conclusions We conclude that the differences of gastric transit time, small bowel transit time and completion rates between the two kinds of capsule endoscopy with different size and weight are not significantly. Whereas, in patients with Crohn's disease or suspected Crohn's disease, gastric transit time of smaller and lighter capsule en-doscopy is shorter in patients with gastrointestinal bleeding, but longer of gastric transit time in smaller and lighter capsule endoscopy.

Objective To investigate the preventive effect of endoclips and endoloops on postpolypectomy bleeding of large colorectal polyps.Methods Data of patients,who underwent polypectomy during January 2013 to March 2016,were retrospectively collected.The diameters of all polyps were more than 10 mm.Cases were divided into 4 groups.Before large pedunculated (with thick stalks) polyps were resected,endoclips were used to ligate the pedicles of polyps in Group A,and endoloops were used in Group B.After large sessile and pedunculated (without thick stalks) polyps were resected,endoclips were used to close the incision of polypectomy in Group C,but not in Group D.The immediate and delayed postpolypectomy bleeding rate and clinicopathologic features were studied.Articles about endoclip or endoloop on preventing post-polypectomy bleeding in PubMed in last five years were searched and analyzed.Results A total of 2 006 polyps were included.The immediate bleeding rate was 3.4％ (5/147) and 3.8％ (5/132) of Group A and B,respectively.The delayed bleeding rate was 6.1％ (9/147) and 7.6％ (10/132) of Group A and B,respectively.The delayed bleeding rate of Group C and D was 3.2％ (28/888) and 1.9％ (16/839),respectively.None of the bleeding cases needed a surgical operation.And no perforation occurred.Six articles were included for analysis.Most of articles revealed that endoclip and endoloop were effective tools in prevention of post-polypectomy bleeding.Conclusion Endoclips and endoloops are useful to prevent bleeding after resection of large pedunculated (with thick stalks) polyps.For large sessile and pedunculated (without thick stalks) polyps (diameter＞ 10 mm),the effect of endoclips to prevent postpolypectomy bleeding still needs further discussion.