RESUMEN
BACKGROUND AND AIMS: There is a high incidence of stricture after endoscopic submucosal dissection (ESD) for cervical esophageal cancer. We aimed to elucidate the risk factors for stricture and to evaluate the efficacy of steroid injection for stricture prevention in the cervical esophagus. METHODS: We retrospectively analyzed 100 patients who underwent ESD for cervical esophageal cancer to (1) identify the factors associated with stricture among patients who did not receive steroid injection, and (2) compare the incidence of stricture between patients with and without steroid injection. RESULTS: Among 48 patients who did not receive steroid injection, there were significant differences in tumor size (P = .026), resection time (P = .028), and circumferential extent of the mucosal defect (P = .005) between patients with stricture (n = 5) and without stricture (n = 43). Compared with patients without steroid injection, patients with steroid injection had a significantly lower incidence of stricture when the post-ESD mucosal defect was <3/4 and ≥1/2 (40% versus 8%; P = .039). For the patients with a post-ESD mucosal defect of ≥3/4 (n = 13), local steroid injection was performed for all of them, and 6 (46%) developed stricture. CONCLUSIONS: Patients who underwent ≥1/2 circumferential resection were at high risk of cervical esophageal stricture. Steroid injection had a stricture prevention effect in patients with <3/4 and ≥1/2 circumferential resection, but seemed to be insufficient in preventing stricture in patients with ≥3/4 circumferential resection.