Additional surgery for non-curative resection after endoscopic submucosal dissection for gastric cancer: a retrospective analysis of 200 cases

• Published in: Surgery today (Tokyo, Japan) Vol. 47; no. 2; pp. 202 - 209
• Main Authors: Sunagawa, Hideki, Kinoshita, Takahiro, Kaito, Akio, Shibasaki, Hidehito, Kaneko, Kazuhiro, Ochiai, Atsushi, Ohtsu, Atsushi, Nishida, Toshirou
• Format: Journal Article
• Language: English
• Published: Tokyo Springer Japan 01.02.2017
• Subjects: Adult; Aged; Aged, 80 and over; Cohort Studies; Endoscopy, Gastrointestinal; Female; Gastrectomy; Humans; Lymph Node Excision; Lymphatic Metastasis; Male; Medicine; Medicine & Public Health; Middle Aged; Neoplasm, Residual; Original Article; Palliative Care; Predictive Value of Tests; Retrospective Studies; Stomach Neoplasms - pathology; Stomach Neoplasms - surgery; Surgery; Surgical Oncology; Lymph node metastasis; Endoscopic submucosal dissection; Additional gastrectomy; Gastric cancer
• ISSN: 0941-1291; 1436-2813
• DOI: 10.1007/s00595-016-1353-1

Purpose Endoscopic submucosal dissection is recommended for early gastric cancer with a low risk of lymph node metastasis. When the pathological findings do not meet the curative criteria; then, an additional gastrectomy with lymph node dissection is recommended. However, most cases have neither lymph node metastasis nor a local residual tumor during an additional surgery. Methods This was a single-institutional retrospective cohort study, analyzing 200 patients who underwent an additional gastrectomy after non-curative endoscopic submucosal dissection from January 2005 to October 2015. We reviewed the patients’ clinicopathological data and evaluated the predictors for the presence of a residual tumor. Results Histopathology revealed lymph node metastasis in 15 patients (7.5 %) and a local residual tumor in 23 (11.5 %). A multivariable analysis revealed macroscopic findings (flat/elevated type) ( p  = 0.011, odds ratio = 4.63), lymphatic invasion ( p  < 0.0001, odds ratio = 14.2), and vascular invasion ( p  = 0.04, odds ratio = 4.00) to be predictors for lymph node metastasis. A positive vertical margin ( p  = 0.0027, odds ratio = 3.26) and horizontal margin ( p  = 0.0008, odds ratio = 5.74) were predictors for a local residual tumor. All cases with lymph node metastasis had lymphovascular invasion with at least one other non-curative factor. Conclusions The risk of a residual tumor can, therefore, be estimated based on the histopathology of endoscopic submucosal dissection samples. Lymphovascular invasion appears to be a pivotal predictor of lymph node metastasis.