Long-Term Results of Endoscopic Submucosal Dissection for Early-Stage Gastrointestinal Tumors

Abstract

Introduction: Endoscopic mucosal resection is minimally invasive technique for early stage gastrointestinal tumors with negligible risk of lymph node metastasis, but it is unsuitable for larger lesions. Such lesions may require piecemeal resection because of limitation of snare size, therefore the risk of local recurrence increases by the incorrect diagnosis of the tumor margin and the risk of distant metastasis also increases by the limited diagnosis of the submucosal invasion and lymphovascular invasion, because of the histopathological assessment by piecemeal resection. A new endoscopic procedure, endoscopic submucosal dissection (ESD), using newly developed endo-knives (ex. Flush knife), is expected to enable en-block resection of lesions regardless of tumor size and shape. Aims and Methods: This study aims to examine the effectiveness and feasibility of ESD. A total of 1635 lesions (Esophagus; 138, Stomach; 1136, Colorectum; 361) was removed by ESD between June 2002 and July 2007. The judgment of negligible risk of lymph node metastasis was based on the indication as follows, Stomach; 1) mucosal (m)-cancer (ca), differentiated (diff.) type, ly(-) v(-), and ulceration (Ul) (-), 2)m-ca, diff. type, ly(-) v(-), and ≦3cm in size, Ul(+), 3) submucosal (sm)-ca (<500 μm), diff. type, ly(-) v(-), and ≦3cm in size. Esophagus; m-ca (m1-m2). Colorectum; 1) m-ca 2) sm-ca (<1000μm), well and mod diff. type, ly(-) v(-). After the histopathological assessment, the cases within the indication were followed up, and the others were recommended the surgical operation. We retrospectively evaluated the data including en-block resection rate, tumor size, time required for resection, rate of complication and prognosis. Results: En-block resection rate was Esophagus; 98.6% (136/138), Stomach; 99.3% (1128 /1136), Colorectum; 98.6% (356 / 361). The median tumor size / resected specimen size was Esophagus; 23 mm(1-72) / 45 mm(22 -90), Stomach; 13 mm(1-105) / 42 mm(14-153), Colorectum; 30 mm(6 -158) / 40 mm(16 -165). The time required for resection was Esophagus; 61 min.(25-206), Stomach; 47 min.(11-352), Colorectum; 58 min.(15 -335). The perforation / postoperative bleeding occurred in Esophagus; 0% / 0%, Stomach; 1.9% / 3.3%, Colorectum; 1.9% / 1.7%. We judged 1419 lesions (Esophagus; 123, Stomach; 972, Colorectum; 324) as the cases within the indication. Among the cases within the indication, no local recurrence or distant metastasis was observed during the follow-up (median observation period; 2674days). Conclusion: The new procedure ESD is very effective and feasible, and it is expected to replace the surgical treatment in certain situations.