Abstract
Background
Predicting the depth of invasion of superficial Barrett adenocarcinoma (s-BA) is important for choosing an appropriate treatment. This study aimed to evaluate the endoscopic and histopathological characteristics related to s-BA submucosal invasion.
Methods
We retrospectively reviewed 67 lesions in 63 cases with pathologically defined s-BA (SSBE, n = 56; LSBE, n = 7) that underwent endoscopic resection at our hospital from January 2004 to December 2017. Initial treatment included endoscopic mucosal resection (EMR) (n = 4), endoscopic submucosal dissection (ESD) (n = 99), and surgery (n = 33). We grouped 133 lesions into two groups based on depth of tumor invasion: group M comprised 87 intramucosal tumors and group SM comprised 49 submucosal tumors. We defined characteristic criteria for submucosal invasion as follows: tumor size ≥ 21 mm, complex macroscopic type; composed of > 2 macroscopic types, biopsy-por; biopsy specimens including poorly differentiated adenocarcinoma. Endoscopic ultrasound (EUS) was performed only in cases in which predicting the depth of tumor invasion was difficult.
Results
In group M, the median tumor diameter was 13 (range, 1–82) mm and included 68 SSBEs and 19 LSBEs. In group SM, the median tumor diameter was 23 (range, 4–55) mm and included 41 SSBEs and 8 LSBEs. Tumors larger than 21 mm were seen in 12 (13.8%) patients in group M and 25 (51.0%) in group SM. Complex macroscopic type tumors were present in 20 patients (23.0%) in group M and 30 (61.2%) in group SM. Biopsy-por was present in 2 (2.3%) in group M and 12 (24.5%) in group SM. Multivariate analysis indicated the above three characteristics as independent predictors of submucosal invasion; in particular, biopsy-por was highly significant (P < 0.001, odds ratio, 10.81). EUS was performed in 55 lesions including 28 tumors invading the submucosa. Sensitivity, specificity, positive predictive value, and negative predictive value of EUS for predicting submucosal invasion were 46.4%, 70.4%, 61.9%, and 57.5%, respectively.
Conclusion
Tumor size ≥ 21 mm, complex macroscopic type, and biopsy specimens including poorly differentiated adenocarcinoma were independent predictors of submucosal invasion. Specificity of EUS was relatively high for cases that were difficult to predict depth of tumor invasion.
Disclosure
All authors have declared no conflicts of interest.
Achalasia Secondary to Submucosal Invasion by Poorly Differentiated Adenocarcinoma of the Cardia, Siewert II: Consideration on Preoperative Workup
