Prospective multicenter study of the esophageal triamcinolone acetonide-filling method in patients with subcircumferential esophageal endoscopic submucosal dissection
Abstract
Background and Aim
The esophageal triamcinolone acetonide (TA)-filling method is a novel local approach for stenosis prevention after extensive esophageal endoscopic submucosal dissection (ESD). We evaluated this method after subcircumferential ESD.
Methods
We enrolled 20 patients with esophageal cancer requiring subcircumferential ESD in a prospective multicenter study. Esophageal TA filling was carried out 1 day and 1 week after ESD, with follow-up endoscopy every 2 weeks. We treated severe stenosis preventing endoscope passage with endoscopic balloon dilatation (EBD) and additional TA filling, and mild stenosis allowing endoscope passage with additional TA filling only. Primary endpoint was incidence of severe stenosis; secondary endpoints were total number of EBD, rate of additional TA filling, time to stenosis and complete re-epithelialization, dysphagia score, and adverse events. Horizontal resection grade was divided into grades 1 (≥ 9/12 and <10/12 of the circumference), 2 (≥ 10/12 and <11/12), and 3 (≥ 11/12 but not circumferential) and analyzed statistically for correlation with endpoints.
Results
Incidence of severe stenosis was 5.0% (1/20; 0.1–24.8%) and was treated with three EBD. Six patients showed mild stenosis. Additional TA filling was carried out in these seven patients: 0% (0/9) for grade 1 resection, 40% (2/5) for grade 2, and 83% (5/6) for grade 3 (P < 0.05). Median time to stenosis and re-epithelialization was 3 and 7 weeks, respectively. Dysphagia score deteriorated in one patient. No adverse events occurred.
Conclusions
The esophageal TA-filling method prevented stenosis after subcircumferential ESD. Grade ≥2 resection showed a high risk for stenosis, but additional TA filling for mild stenosis inhibited stenosis progression (UMIN000024384).
Conflicts of Interest
Authors declare no conflicts of interest for this article.
