Published papers

Endoscopic submucosal dissection (ESD) for early stage gastric cancer (EGC) has improved the success rate of en bloc resection but results in perforation more often than does endoscopic mucosal resection. We report a novel technique of ESD using an external grasping forceps. A total of 265 lesions with EGC or gastric adenoma were enrolled in this study. Sixteen lesions were located in the upper third portion of the stomach, 114 in the middle third portion, and 135 in the lower third portion. After submucosal injection followed by circumcision of the lesions with a flex knife, the external grasping forceps was introduced with the help of a second grasping forceps and anchored at the margin of the lesion. Oral traction applied with this forceps could elevate the lesion and make the submucosal layer wider and more visible, thereby facilitating dissection of the submucosal layer under direct vision. The mean lesion size was 15.0 mm (range: 5-50 mm). All but 11 lesions (95.8%) could be resected en bloc with free margins. Mean procedure time was 45 min (range: 20-180 min). It was difficult to carry out this procedure when the lesions were located in the cardia, lesser curvature, or posterior wall of the upper third of the gastric body. Bleeding after ESD occurred in 10 patients (3.8%) and perforation occurred in one patient (0.4%). The endoscopic submucosal dissection using an external grasping forceps for superficial gastric neoplasia is efficacious and safe.

AIM: Endoscopic hemostasis using hemoclips is useful, but there are technical difficulties because the angle of the approach is tangential. A transparent hood facilitates the observation and treatment of these lesions, and a shorter hood provides a wider visible field. Endoscopic hemoclipping of hard lesions with hemoclips of the conventional size does not reliably result in sustained hemostasis because the clips slip. Short clips, however, can be easily clamped on protruded visible vessels without slip. The aim of the present study was to evaluate the efficacy of endoscopic hemostasis with a short transparent hood and short clips. METHODS: Subjects were 198 patients with 214 lesions of non-variceal upper gastrointestinal bleeding at Keio University Hospital. We used a video endoscope with a short transparent hood attached to its distal tip and carried out hemostasis using short hemoclips. RESULTS: The short transparent hood provided a good visual field. If the lesions were in the tangential, the short hood made it possible to observe them in the frontal view and made clip hemostasis much easier. The short clip could be securely clamped against protruded visible vessels. Of 214 lesion, 211 (98.6%) had temporal hemostasis. Rebleeding occurred in 13 of 211 lesions (6.2%), and 205 of 214 lesions (95.8%) had permanent hemostasis. Nine cases were endoscopically difficult. CONCLUSION: Endoscopic hemostasis with a short transparent hood and short clips is useful for non-variceal upper gastrointestinal bleeding.

Although enhanced lymphocyte trafficking is associated with colitis formation, little information about its regulation is available. The aim of this study was to examine how the murine liver and activation-regulated chemokine (mLARC/CCL20) contributes to lymphocyte recruitment in concert with vascular adhesion molecules in murine chronic experimental colitis. T and B lymphocytes isolated from the spleen were fluorescence-labelled and administered to recipient mice. Lymphocyte adhesion to microvessels of the colonic mucosa and submucosa was observed with an intravital microscope. To induce colitis, the mice received two cycles of treatment with 2% dextran sodium sulphate (DSS). In some of the experiments antibodies against the adhesion molecules or anti-mLARC/CCL20 were administered, or CC chemokine receptor 6 (CCR6) of the lymphocytes was desensitized with excess amounts of mLARC/CCL20. Significant increases in T and B cell adhesion to the microvessels of the DSS-treated mucosa and submucosa were observed. In chronic colitis, the accumulation of lymphocytes was significantly inhibited by anti-mucosal addressin cell adhesion molecule (MAdCAM)-1 mAb, but not by anti-vascular cell adhesion molecule-1. In DSS-treated colonic tissue, the expression of mLARC/CCL20 was significantly increased, the blocking of mLARC/CCL20 by monoclonal antibody or the desensitization of CCR6 with mLARC/CCL20 significantly attenuated the DSS-induced T and B cell accumulation. However, the combination of blocking CCR6 with MAdCAM-1 did not further inhibit these accumulations. These results suggest that in chronic DSS-induced colitis, both MAdCAM-1 and mLARC/CCL20 may play important roles in T and B lymphocyte adhesion in the inflamed colon under flow conditions.

A second-line chemotherapy for advanced gastric cancer has not been established. We report a case of good response in a 39-year-old woman who had recurrent pleural effusion and positive cytology of type 4 gastric cancer and was treated with TS-1, a DPD inhibitory fluoropyrimidine, in combination with weekly taxane. After a partial response for type 4 gastric cancer from the treatment with 2 courses of TS-1 plus low-dose cisplatinum (CDDP), followed by outpatient chemotherapy with TS-1 alone or TS-1 plus weekly CDDP, left pleural effusion appeared and CA19-9 increased during the 7th course of the chemotherapy. Cytology of the effusion was class IV. The patient was treated with a course of TS-1 (120 mg/day, day 1-21) plus paclitaxel (50 mg/m2, day 1, 8) followed by 2 week washout. In the following courses, paclitaxel was replaced with docetaxel (30 mg/m2, day 1 and 8) and the course was continued in the outpatient setting. After 2 courses, the left pleural effusion disappeared and remained absent after 6 courses. Gastric biopsy showed no cancer cells and abdominal CT showed no recurrence. Serum CA19-9 doubled 1 week after taxane treatment and decreased gradually week by week during the course. This case suggests that a combination of TS-1 and taxane is effective against recurrent pleural effusion of advanced gastric cancer and useful as a second-line chemotherapy.

It has recently been suggested that C-C chemokines may play a role in the organ-specific homing of lymphocytes, but there is not enough in vivo evidence in intestinal mucosa. The aim of this study was to examine whether thymus-expressed chemokine (TECK)/CCL25 and its ligand CCR9 are involved in T-lymphocyte interaction with microvessels of murine intestinal mucosa. T lymphocytes from the small intestine were fluorescence labeled, and their adhesion to mucosal microvessels was observed by intravital microscopy. Lamina proprial lymphocytes (LPL) and intraepithelial lymphocytes (IEL) adhered to both the small intestine and colon, and desensitization of CCR9 with TECK/CCL25 or anti-TECK/CCL25 antibody significantly inhibited these adhesions only in small intestine. At both sites, TNF-alpha significantly increased LPL adhesion but not IEL adhesion. Desensitization of CCR9 or anti-TECK/CCL25 antibody also attenuated the TNF-alpha-induced LPL adhesion in the small intestine. Increased expression of TECK/CCL25 by TNF-alpha was observed in the lamina propria of small intestine. TECK/CCL25 may thus play an important role in the adherence of mucosal lymphocytes to the microvessels of the small intestine but not the colon under uninflamed as well as inflamed conditions.