Randomized, Prospective Comparison of Large vs. Small Diameter Esophageal Stents for Palliation of Malignant Dysphagia

This prospective, randomized study was conducted at Tenwek Hospital in Bomet, Kenya, and was approved by the hospital's institutional review board (IRB) as well as the Kenya Medical Research Institute's Ethical Review Committee. Participants with dysphagia were recruited and provided signed informed consent prior to endoscopy. During endoscopy and after dilation of an obstructing esophageal tumor sufficient to permit passage of an endoscope, eligible participants were enrolled and randomized to receive either an 18 mm shaft/23 mm flange or 23 mm shaft/28 mm flange partially covered Ultraflex esophageal stent (Boston Scientific, Natick, Massachusets, USA). Stents were either 10 or 12 cm in length, and the endoscopist chose a stent length based on the length of the malignant stricture. One stent was placed in each subject. Follow-up was then obtained at scheduled intervals after stent placement (7 and 28 days, 3 months, then every 3 months), until the participant's death. Earlier follow-up visits were arranged if participants reported any new health concerns.

Block randomization was performed using the sealed envelope technique, with 10 participants in each block; none of the study personnel at Tenwek Hospital had knowledge of the randomization sequence. Allocation was concealed from participants, caregivers, and study personnel until randomization occurred during an endoscopic procedure. After randomization, stent diameters were known to the endoscopy staff and listed in the medical record. All randomized participants correctly received a stent of the allocated diameter, and remained blinded to the stent diameter they received.

At baseline and each follow-up visit body weight was recorded, as well as Karnofsky performance status score, dysphagia score (0=normal, no dysphagia; 1=can swallow most foods; 2=can swallow a soft diet; 3=can swallow fluids only; 4=unable to swallow saliva), current medications, and the presence or absence of 15 symptoms and 9 diagnoses (weakness, fever, vomiting, vomiting blood, melena, weak voice, difficulty breathing, cough, sputum, palpitations, heartburn, chest pain, hiccoughs, recurrent dysphagia, abdominal pain; anemia, gastrointestinal bleeding, esophago-respiratory fistula, arrhythmia, pneumonia, metastases, stent occlusion, stent migration, gastroesophageal reflux disease). In addition, all interval clinic visits, test results, endoscopy reports, and hospitalizations were reviewed.