Efficacy of endoscopic ultrasound-guided fine-needle aspiration and core needle biopsy in the diagnosis of upper gastrointestinal submucosal lesions.

INTRODUCTION

Endoscopic ultrasonographically guided fine-needle aspiration (EUS-FNA) has been increasingly utilized to evaluate submucosal lesions of the upper gastrointestinal (UGI) tract. Our study aims to determine the efficacy of UGI EUS-FNA/core needle biopsy (CNB), including the frequency and cytomorphologic features of encountered submucosal lesions, and to investigate contributing factors including the role of rapid on-site evaluation (ROSE).

METHODS

We analyzed all UGI submucosal lesions diagnosed at our institution by EUS-FNA/CNB from September 2008 through August 2015.

RESULTS

During this 8-year study period, 94 patients underwent 110 UGI EUS-FNA/CNB, including 89 (81%) gastric, 11 (10%) duodenal, and 10 (9%) esophageal lesions. Twenty-seven (25%) were gastrointestinal stromal tumors (GISTs), followed by 13 (12%) leiomyomas, 5 (5%) schwannomas, 4 (4%) gastric adenocarcinomas, 3 (3%) neuroendocrine tumors (NETs), and 3 (3%) pancreatic heterotopias. All GISTs, leiomyomas, and NETs were ultimately diagnosed by EUS biopsies, as well as 75% of adenocarcinomas, 60% of schwannomas, and 33% of pancreatic heterotopias. The specificity of EUS-FNA/CNB for these 6 most commonly encountered lesions was 100%, with sensitivity of 82%. Sensitivity was 100% for esophageal and duodenal biopsies, and 80% and 75% for gastric and gastroesophageal procedures, respectively. Factors that contributed to poor yield included the lack of ROSE, small lesional size, lesion location and histology, and needle type. Neither number of needle passes nor operator experience appeared to influence specimen adequacy.

CONCLUSION

EUS-FNA/CNB is an effective modality for diagnosing UGI submucosal lesions. Awareness of potential errors due to sampling of the bowel wall, lesional cystic degeneration, as well as pancreatic heterotopia and Brunner gland hamartoma is essential in order to avoid false diagnoses.