A randomized controlled trial of endoscopic steroid injection for prophylaxis of esophageal stenoses after extensive endoscopic submucosal dissection

Abstract

Background: Esophageal stenosis following endoscopic submucosal dissection (ESD) is a serious adverse event that makes subsequent management more difficult.

Methods: This parallel, randomized, controlled, open-label study was designed to examine whether local steroid injection is an effective prophylactic treatment for esophageal stenoses following extensive ESD. This single center trial was conducted at the Keiyukai Hospital, a tertiary care center for gastrointestinal disease in Japan [University Hospital Medical Network Clinical Trial Registry (UMIN-CTR) on 15 September 2011 (UMIN000006327)]. Thirty-two patients with mucosal defects involving ≥75% of the esophageal circumference were randomized to receive a single dose of triamcinolone acetonide injections (n = 16) or be treated conventionally (n = 16). The primary outcome was the frequency of stricture requiring endoscopic dilatation; the surrogate primary endpoint was the number of dilatation sessions needed. Secondary outcomes included adverse event rates, the minimum diameter of the stenotic area and the duration of the course of dilatation treatments.

Results: The frequency of stricture was not significantly different between the groups because of insufficient statistical power, but the number of dilatation sessions required was significantly less in the steroid group (6.1 sessions [95% confidence interval, CI 2.8-9.4] versus 12.5 [95% CI 7.1-17.9] sessions in the control group; P = 0.04). The perforation rate was similar in both groups. The minimum diameter of stenotic lumens was significantly greater in the treatment group than controls (11.0 mm versus 7.1 mm, respectively; P = 0.01). The perforation rate was not significantly different between the groups (1.0% versus 0.5% in the treatment and control group, respectively). Steroid injection was effective in cases of mucosal defects encompassing the entire esophageal circumference.

Conclusions: Prophylactic endoscopic steroid injection appears to be a safe means of relieving the severity of esophageal stenoses following extensive ESD.

Figure 1

Typical endoscopic views of the esophagus in a patient in the injection group. a. A superficial esophageal carcinoma in the middle esophagus. The entire circumference of the lesion was marked out by electrocautery using a needle-knife at least 1 mm from the tumor border, confirmed by a Lugol-unstained region. b. This tumor encompasses half the circumference of the esophagus, as seen in the center of the lesion. c. The artificial ulcer encompassed the entire circumference after ESD. d. Injection of triamcinolone into the ulcer (white arrow).

Figure 2

Endoscopic view 6 days after endoscopic submucosal dissection. Some injected triamcinolone is evident in the ulcer (white arrows). The right picture (a) is a magnification of the left picture (b).

Figure 3

Esophageal stenosis assessed by esophagography. The white lines indicate the stricture caused by resection. The stricture had substantially improved 1 month later. The left esophagogram (a) was taken 2 months after endoscopic submucosal dissection (ESD); the right esophagogram (b) was taken 3 months after ESD.

Figure 4

Barium esophagography 1 month after endoscopic submucosal dissection. The yellow lines indicate the narrow lumens owing to resection. a. Patient allocated to the control group, who developed a severe esophageal stricture that required 13 sessions of dilatation therapy. b. Patient allocated to the treatment group, who did not require dilatation.

Figure 5

Typical endoscopic view 1 month after endoscopic submucosal dissection. Severe stenosis of the esophagus did not develop. The right picture (a) is a magnification of the left picture (b).

Figure 6

Study flow chart. Patients with an expected circumferential mucosal defect involving ≥75% of the circumference of the esophagus after ESD were eligible. Patients were excluded if they had received additional adjuvant treatments, such as surgery or chemoradiation therapy, or if they were not adequately followed-up.