A Prospective, Randomized Study Comparing Two Self-Expanding Stents In The Palliation Of Patients With Dysphagia Due To Inoperable Esophageal Cancer

Introduction: Esophageal cancer (EC) is rarely a curable disease. Self-expanding stents (SES) are used to restore swallowing. The major drawback of metal stents is the high cost. A new expandable plastic stent has been developed. Aim: To compare two different SES in terms of relief of dysphagia (D), complications, and survival. Methods: Seventy-eight consecutive patients (65 males, 13 females; mean age 71.9 yrs) with unresectable EC were randomized to insertion of a covered SES (Ultraflex, 23×28 mm) (U), or a new type of covered SES made of polyester netting, embedded in silicone (Polyflex, 18×23 mm) (P). The primary end point was relief of D (0=no D; 1=some solid food; 2=semisolids; 3=liquids; 4=aphagia). Secondary endpoints were: complications and survival. Results: Stents were inserted in all patients: 42 had U, 36 P. At 7 days, D was relieved in 60/78 (76.9%) patients, and in 50/78 (64.1%) at one month. Following SES placement, the median D score improved from 3 to 1 at 7 days, and was 2 at 30 days, in the U group; and improved from 3 to 1 at 7 days, and was 1 at 30 days in the P group (p=n.s.). The complication rate in U and P groups was 41.5% and 55.6%, respectively (p=n.s.). In the U group complications were: migration (1), and pain (13). In the P group complications were: migration (1), pain (15), and perforation (1) (p=n.s.). Stent related mortality was 2 patients (4.8%) in the U group and 1 (2.8%) in the P group. Recurrent D due to tumor overgrowth occurred in 4 (9.5%) U, and 8 (22.2%) P, after a median follow-up of 3 months (p=n.s.). Dysphagia due to hyperplastic overgrowth was diagnosed in 4 patients treated with U, and in 4 in the P group (p=n.s.). Endoscopic reintervention was required in 10 (23.8%) U patients, and 14 (38.9%) P patients (p=n.s.). In the U group relief of D was obtained with a further SES in 4 (9.5%) patients. Thermal therapy was used in 6 (14.3%) patients. In the P group relief of D was obtained with a second stent in 5 (13.9%). Thermal therapy was used in 6 (16.7%) patients. Median survival was 8.4 months in U group and 13.5 months in P (p=n.s.). Median length of hospital stay was 4 days in the U patients, as compared to 3 in the P group (p=n.s.). Conclusions: The metal and the plastic SES were similarly effective in relieving malignant D in patients with EC. The complication rate was also similar in both groups. A non-significant trend to longer survival was noted in the P group. An advantage of the P stent is its lower cost.