Endoscopic resection of superficial esophageal neoplasms is already a reality and presents important advantages when compared to esophagectomy as fewer complications and better quality of life. However, extensive resections can lead to difficult-to-manage stenoses. There are several therapies available in order to prevent this stenosis but, to date, there is no definition of the gold standard.

The objective of this study was to compare the use of intralesional steroid injection versus oral prednisolone after endoscopic submucosal dissection and to evaluate the stenosis rate, number of dilations to resolve the stenosis and complications.

Endoscopic resection of superficial esophageal neoplasms is widely used as an alternative to esophagectomy, since it is less invasive, besides presenting good clinical results. Compared with esophagectomy, patients submitted to endoscopic resection present shorter hospitalization time, lower incidence of complications and better quality of life in the long term.

However, repair of esophageal ulcer, caused by endoscopic resection, which occupies three quarters or more of the circumference of the organ, can result in the formation of stenosis.

In the past, there was no consensus on the use of preventive therapies for esophageal stenosis after extensive ESD. However, Oliveira et al recently demonstrated through systematic review and meta-analysis that the use of these therapies reduces the rate of stenosis (40% on average), decreased the number of dilations to resolve the stenosis (8 sessions less ), Without altering the number of complications.

Theoretically, corticosteroids are the most appropriate choice due to their mechanism of action, modulating wound healing by preventing inflammation, by reducing prolyl hydroxylase, which helps reduce collagen production.

However, treatment with corticosteroids, especially at high oral doses, can cause several adverse effects, such as immunosuppression, diabetes, psychiatric disorders, osteoporosis, optic lesion and peptic ulcer. Thus, the use of local corticosteroid injection could minimize these side effects. However, local injection implies risks of bleeding and perforation, and is of limited use in patients receiving anticoagulant or antiplatelet therapy.

The objective of this study was to compare the local corticosteroid injection and the use of oral corticosteroids to prevent stenosis after extensive submucosal endoscopic resection of superficial esophageal carcinoma, in relation to the stenosis rate, number of dilations necessary to resolve the stenosis and frequency of complications.

Inclusion Criteria:

• Patients with a diagnosis of superficial esophageal neoplasm submitted to submucosal endoscopic resection greater than 3/4 of the organ circumference;
• Absence of lymph node or distant metastases, evaluated through echoendoscopy, CT and PET-CT;
• Signed informed consent form

Exclusion Criteria:

• Presence of invasive esophageal neoplasia
• Hepatical cirrhosis
• Diabetes mellitus with fasting glycemia above 200mg%
• Use of corticosteroids in the 30 days prior to ESD
• INR> 1.5
• Platelet count less than 50,000
• Active gastrointestinal ulcer
• Severe psychiatric illness
• Glaucoma
• History of allergy or hypersensitivity to corticosteroids or proton pump inhibitor