Background: Although endoscopic submucosal dissection (ESD) has become accepted for the treatment of superficial esophageal cancer, the incidence of stricture formation caused by ESD for widespread lesions is high and leads to a low quality of life. A few studies reported that locoregional steroid injections are useful for the prevention of such stricture formation compared with historical controls. We evaluated the efficacy of prophylactic locoregional steroid injections for stricture formation caused by ESD using quasi-randomized analysis.
Methods: This matched case-control study included 461 superficial esophageal cancers from 305 patients who underwent ESD between 2006 and 2013. We used two methods of locoregional steroid injection to prevent stricture formation after ESD. A propensity score matching analysis was performed to reduce the effects of a selection bias for steroid injections and other potential confounding factors. In addition, generalized estimating equations were used to analyze repeated measures data. We compared the incidence of stricture formation with or without steroid injections.
Results: Forty-two lesions were treated with locoregional steroid injection (dexamethasone/triamcinolone, 23/19) after ESD and esophageal stricture formation occurred in 36 lesions. Fifty-six lesions treated with or without steroid injections were matched after propensity score matching. Locoregional steroid injection reduced the incidence of stricture formation to 10.7% (3/28) of patients compared with 35.7% (10/28) in the control group (odds ratio 4.63, 95% confidence interval 1.11-19.25, p = 0.035).
Conclusions: Locoregional steroid injections could be efficient for the prevention of stricture formation after ESD for superficial esophageal cancer.
Keywords: Endoscopic submucosal dissection; Esophageal cancer; Esophageal stricture; Propensity score; Steroid injection.