Endoscopic submucosal dissection in a patient with esophageal adenoid cystic carcinoma.

We report the first use of endoscopic submucosal dissection (ESD) for the treatment of a patient with adenoid cystic carcinoma of the esophagus (EACC). An 82-year-old woman visited our hospital for evaluation of an esophageal submucosal tumor. Endoscopic examination showed a submucosal tumor in the middle third of the esophagus. The lesion partially stained with Lugol's solution, and narrow band imaging with magnification showed intrapapillary capillary loops with mild dilatation and a divergence of caliber in the center of the lesion. Endoscopic ultrasound imaging revealed a solid 8 mm × 4.2 mm tumor, primarily involving the second and third layers of the esophagus. A preoperative biopsy was non-diagnostic. ESD was performed to resect the lesion, an 8 mm submucosal tumor. Immunohistologically, tumor cells differentiating into ductal epithelium and myoepithelium were observed, and the tissue type was adenoid cystic carcinoma. There was no evidence of esophageal wall, vertical stump or horizontal margin invasion with pT1b-SM2 staining (1800 µm from the muscularis mucosa). Further studies are needed to assess the use of ESD for the treatment of patients with EACC.

Gastroesophageal flap valve status distinguishes clinical phenotypes of large hiatal hernia.

AIM: To investigate two distinct clinical phenotypes of reflux esophagitis and intra-hernial ulcer (Cameron lesions) in patients with large hiatal hernias. METHODS: A case series study was performed with 16 831 patients who underwent diagnostic esophagogastroduodenoscopy for 2 years at an academic referral center. A hiatus diameter ≥ 4 cm was defined as a large hernia. A sharp fold that surrounded the cardia was designated as an intact gastroesophageal flap valve (GEFV), and a loose fold or disappearance of the fold was classified as an impaired GEFV. We studied the associations between large hiatal hernias and the distinct clinical phenotypes (reflux esophagitis and Cameron lesions), and analyzed factors that distinguished the clinical phenotypes. RESULTS: Large hiatal hernias were found in 49 (0.3%) of 16,831 patients. Cameron lesions and reflux esophagitis were observed in 10% and 47% of these patients, and 0% and 8% of the patients without large hiatal hernias, which indicated significant associations between large hiatal hernias and these diseases. However, there was no coincidence of the two distinct disorders. Univariate analysis demonstrated significant associations between Cameron lesions and the clinico-endoscopic factors such as nonsteroidal anti-inflammatory drug (NSAID) intake (80% in Cameron lesion cases vs 18% in non-Cameron lesion cases, P = 0.015) and intact GEFV (100% in Cameron lesion cases vs 18% in non-Cameron lesion cases, P = 0.0007). In contrast, reflux esophagitis was linked with impaired GEFV (44% in reflux esophagitis cases vs 8% in non-reflux esophagitis cases, P = 0.01). Multivariate regression analysis confirmed these significant associations. CONCLUSION: GEFV status and NSAID intake distinguish clinical phenotypes of large hiatal hernias. Cameron lesions are associated with intact GEFV and NSAID intake.