A prospective, randomized, controlled trial of covered expandable metal stents in the palliation of malignant esophageal obstruction at the gastroesophageal junction

OBJECTIVE: Palliation of malignant esophageal obstruction is an important c linical problem. Expandable metal stents are a major advance in therapy, bu t many stents become obstructed because of tumor ingrowth. The aim of this study was to compare a new, membrane-covered expandable metal stent to conv entional prostheses in a randomized controlled trial. METHODS: Sixty-two patients with malignant inoperable esophageal obstructio n at the gastroesophageal junction participated in the study. Patients were randomly assigned to covered or uncovered stents. The principal outcome me asure was the need for reintervention because of recurrent dysphagia or mig ration. Secondary endpoints were relief of dysphagia measured by a dysphagi a score (grade 0 = no dysphagia, grade 1 = able to eat solid food, grade 2 = semisolids only, grade 3 = liquids only, grade 4 = complete dysphagia) an d the rate of complications and functional status. All patients were observ ed at monthly intervals until death or for 6 months. RESULTS: One week after stenting the dysphagia score improved significantly in both the uncovered (n = 32, 3 +/- 0.1 to 1 +/- 0.1 [means +/- SEMs], p < 0.001) and covered (n = 30, 3 +/- 0.1 to 1 +/- 0.2 [means +/- SEMs], p < 0.001) stents. Obstructing tumor ingrowth was significantly more likely in the uncovered stent group (9/30) than in the covered group (1/32) (p = 0.00 5). Significant stent migration occurred in 2/30 patients with uncovered st ents, as compared with 4/32 patients in the covered group (p = 0.44). Reint erventions for tumor ingrowth were significantly greater in the uncovered s tent group (27%), as compared with 0% in the covered group (p = 0.002). Lif e table analysis showed similar survival in both groups. CONCLUSION: Membrane-covered stents have significantly better palliation th an conventional bare metal stents because of decreased rates of tumor ingro wth that necessitate endoscopic reintervention for dysphagia. (C) 2001 by A m. Cell. of Gastroenterology.