Novel removable internally fully covered self-expanding metal esophageal stent: feasibility, technique of removal, and tissue response in humans.

OBJECTIVES: Internally fully covered self-expandable metal stents (FCSEMSs) have recently become available offering the possibility of removal. However, to our knowledge, the feasibility, techniques for endoscopic removal, and tissue reaction to FCSEMS have not been reported in humans. Our aims are (1) to report reasons for and success of FCSEMS removal in consecutive patients, (2) to describe methods of esophageal stent retrieval, and (3) to describe stent reactions induced by FCSEMS in humans.
METHODS: Removable FCSEMSs (ALIMAXX-E, Alveolus Inc., Charlotte, NC) were placed in consecutive patients requiring palliation of dysphagia symptoms with locally advanced esophageal or gastro-esophageal junction cancers, refractory benign esophageal strictures, postoperative mediastinal leaks, or tracheoesophageal fistulas. Dysphagia scores were assessed at 1, 3, and 6 months. Esophagogram and endoscopy follow-up were performed routinely at 1 month or to assess new symptoms.
RESULTS: A total of 36 stents were placed in 31 patients over 16 months (mean age 62.8 years, 61% white, 58% male). Indications included the following: malignant strictures (n=17), refractory benign strictures (n=7), tracheoesophageal fistulas (n=8), postoperative mediastinal leaks (n=2), and iatrogenic perforations (n=2). Dysphagia scores improved (decreased) significantly at 1 (mean difference 2.09; 95% confidence interval (CI) 1.65-2.52, P=0.0001), 3 (mean difference 2.1; 95% CI 1.57-2.62, P=0.0001), and 6 months (mean difference of 2; 95% CI 1.28-2.73, P<0.0001). Clinical success rates for various indications were: malignant strictures 88%, refractory benign strictures 29%, tracheoesophageal fistula 88%, postoperative mediastinal leaks 100%, and iatrogenic perforations 50%. Sixty one percent (n=22) of the stents were removed (removal was successful in 100% of cases attempted). Twenty three percent (n=5 stents in four patients) were removed due to chest pain or abdominal pain. The stents satisfied their purposes in all of these four patients. The remainder were removed on schedule or had satisfied their purpose and/or migrated. Fifty percent of removed stents had ulcers (median size=2 cm) at the distal edge of the stent, whereas 23% had ulcers at the proximal edge of the stent. Pseudo-polyps formed at either the proximal or distal edge of the stent in four patients. Tissue reaction was severe in only two patients. Stents were embedded in four cases but were easily lifted from tissue. Our preferred method of retrieval was the double-channel therapeutic endoscope with two rat-tooth forceps or rat-tooth forceps with a snare.
CONCLUSIONS: FCSEMSs palliate dysphagia from both benign and malignant conditions and seal postoperative tracheoesophageal fistulas and mediastinal leaks. Tissue reaction to the stents was common but remained clinically insignificant in the majority of patients. Close follow-up of patients with FCSEMS is required and repeat endoscopy with stent retrieval might be necessary with the development of new symptoms or complications from the stent.