Superficial rectal tumors could be removed by ESD with two major advantages showed by the literature: high rate of monobloc resection and R0 resection. Hence, recurrence rate was lower than ERM therapy. This technique is actually performed routinely in Japan and experience is limited in Europe, with few data in the literature.

This study is suggested to patients with superficial medium or distal rectal tumors. The lesion was identified and demarcated using white-light endoscopy, magnifying endoscopy, and chromoendoscopy. Then, marking around the lesions was performed. Local injection was made using injection needle, and then mucosal incision was performed around the lesion using endo-knives. Submucosal dissection was performed using endo-knives. Hemostasis and vessel coagulation were practiced using primary hemostatic forceps during the procedure.

Follow-up is one year. For lesion with invasive carcinoma (vessel and lymphatic involvement, undifferentiated, free margin less than 1 mm), a surgical resection is performed.

All the patients have an endoscopic control 3 months after ESD (rectosigmoidoscopy with biopsies):

• For complete resection, an endoscopic control was performed at 1 year.
• For incomplete resection, another endoscopic therapy was attempted (EMR) an endoscopic control was performed at 1 year.

At one year, patient with incomplete resection have a surgical resection.

Colorectal neoplasia prevalence was 35 000 for a total number of 278 000 cancers in France during 2000. Treatment of these cancers is based on local extension especially in the digestive tract (stade) and on co-morbidities of the patient. There are many advantages to treat these cancers at an early stade (involving mucosa or superficial submucosa), since they are associated with a better prognosis with a lower cost-effectivness ratio. Indeed, in 2003 mean cost of treatment of colorectal cancer during the first year following diagnosis was 24 000 €, whereas it was only 18 000 € for early stade and up to 36 000 € for the most advanced stade. Transversal (1 week) and national survey of the French society of endoscopy (SFED) during 8 years demonstrated that among 1 millions of endoscopies (3/4 being performed with anesthesia), 8000 were performed for endosocpic treatment of early digestive cancers by endoscopic mucosal resection (EMR) or more recently by endoscopic submucosal dissection (ESD). One of the problems to diffuse these endoscopic treatments is the multiple tools required to endotherapy, and the fact that this treatment is time-consuming, whereas a simple ablation of polyps (polypectomy) smaller than 1.5 cm needs only 5 to 10 minutes. For exemple, endotherapeutic tools for polypectomy (one needle and one polypectomy loop) costs 180 € and occasionally required a couple of clips when rare (1%) hemorrhage or exceptional (1p1000) perforation occurred. Conversely, endotherapy tools for ESD of early mucosal or submucosal digestive cancers larger than 1.5 cm up to 4-5 cm, costs 1500 €, including needles, submucosal injected products, hook-knife, insulinated-knife and coagulation rasper. So, the recent T2A codification gathers these two kinds of endoscopic treatment under the same code, whatever the size of the lesion, whereas costs of tools required for their treatment are notably different especially when there are performed for inpatients hospitalized less that 2 days which is only charged 480 € added of 192 € and 53€ for salary in private exercise or 914 € in public hospital, by the national and public healthcare system. Beside the expertise required for EMR and ESD of larger early cancers, this cost does not invite to increase the frequency of EMR or ESD, taking into account that this treatment is better practised by two simultaneous endoscopists, one of them manipulating the endoscope en endotherapy tools.

Inclusion Criteria:

• Patients with superficial medium or distal rectal tumors more than 1 cm in size
• Criteria for ESD were determined by the endoscopic characteristics and histological findings of biopsy specimens
• Endoscopic ultrasonography (EUS) also was performed when the lesion was strongly suspected of submucosal invasion.
• Age older than 18, younger than 85 years.

Exclusion Criteria:

• Lesion less than 10 mm in size
• Pedundulated lesion
• Suspicion of submucosal invasion (MRI or EUS)
• Distant metastasis on CTscan
• Coagulopathy
• Pregnancy
• Refusal to participate to the study or inability to consent