British Journal of Surgery zyxwvutsrqp 1995, zyxwvutsrq 82, 1678-1681 Self-expanding metal stents for the palliation of dysphagia due to inoperable oesophageal carcinoma J. P. M. ELLUL, A. WATKINSON*, R. J. K. KHAN, A. ADAM* and R. C. MASON Departments of Surgery and *Radiology, zyxwvutsr Guy’s Hospital, London SEl 9RT UK Correspondence to: zyxwvutsrqpo Mr R. C. Mason Adequate palliation of dysphagia due to inoperable oesophageal carcinoma is difficult to achieve with low morbidity. Thirty-three patients (21 men and 12 women of mean(s.e.m.) age 69(2) years) with inoperable carcinoma of the oesophagus underwent insertion of self-expanding metal stents. In 22 patients the tumours were in the lower third of the oesophagus, in eight in the middle third and in three in the upper third. A stent was inserted as primary palliative therapy in 14 patients, after failed laser therapy in 13 and after oesophageal perforation following other treatments in six. Patients presented with dysphagia of grade 3 or 4. Three types of stent were used: Wallstent, Strecker and Gianturco; stents were inserted under fluoroscopic guidance after balloon dilatation of the stricture. All attempted insertions of metal stents were successful. Dysphagia reduced from grade 3 or 4 to 0 or 1. There were no perforations related to insertion. Patients who had stents inserted to seal previous perforations left hospital a median 7 days later. Dysphagia recurred in six patients, due to migration of the stent (three), blockage by food bolus (one) and tumour overgrowth (two). These problems were easily treated. Self-expanding metal stents seem to offer excellent palliation with minimal morbidity for patients with inoperable carcinoma of the oesophagus. Carcinoma of the oesophagus is relatively uncommon in the UK, with a mean incidence of 7.5 per 100000 population1. However, management of the condition is difficult and the prognosis poor, mainly because patients present at a late stage in the disease. Some 50-60 per cent of patients have incurable cancer at presentation* and are suitable only for palliative therapy. Palliation of oesophageal carcinoma consists mainly of symptomatic treatment of dysphagia. Radiotherapy, insertion of conventional plastic endoprostheses and endoscopic laser therapy have been used, but all have severe limitations. Radiotherapy is useful only for squamous cell carcinoma of the oesophagus; a reasonable degree of palliation after oesophageal dilatation is achieved in only 40 per cent of patients and this may take 2 months3. Conventional plastic prostheses may be difficult to insert and their use is associated with considerable morbidity4. They provide at best limited palliation, allowing patients to swallow only liquidized or semiliquid food. They may also become blocked and migrate. Endoscopic laser therapy has recently been the mainstay of treatment for malignant dysphagia in the authors’ unit. Laser treatment enables near-normal swallowing and is associated with a lower incidence of complications than is intubation5. However, laser therapy has to be repeated at regular intervals, usually every 6 weeks, and in some patients the dysphagia-free period becomes progressively shorter as the disease progresses. Furthermore, laser therapy may be associated with a 6-8 per cent perforation rate. Self-expanding metal stents &table for insertion into the oesophagus have recently become available6-I4. Results from these studies, with mainly Gianturco stents, have demonstrated improvement in dysphagia at least as good as that achieved with plastic stents, and reduced morbidity. The present paper describes experience with self-expanding metal stents placed under radiological Paper accepted 12 March 1995 1678 control to treat patients with inoperable oesophageal cancer. Patients and methods Patients Thirty-four patients with inoperable oesophageal cancer or who were unfit for surgery were considered for treatment with an expandable metal stent. Twenty-one of them were men; the mean(s.e.m.) age was 69(2) (range 40-89) years. Fourteen of the thirty-four patients had stents inserted as primary palliative treatment, in a further seven adequate dilatation could not be achieved by laser therapy, six had extensive disease requiring increasingly frequent laser sessions and a further six had an oesophageal perforation following other treatments. One woman with extensive carcinoma invading the cricopharyngeus was considered but thought to be unsuitable for an expandable stent. Twelve patients had adenocarcinoma of the oesophagus, 18 had squamous cell carcinoma and three undifferentiated or anaplastic carcinoma. Twenty-two patients had tumours situated in the lower third of the oesophagus or cardia of the stomach, eight had tumours in the middle third of the oesophagus and four tumours in the upper third. Two patients had recurrence of adenocarcinoma after oesophagectomy 12 and 15 months earlier. The rest had primary tumours of the oesophagus. Four patients received adjuvant chemotherapy, a further two had radiotherapy and one had palliative radiotherapy to a lung metastasis. Furthermore, one patient had severe Parkinson’s disease and another had had a previous carcinoma of the colon (Dukes C), which had been treated by surgery and radiotherapy. Stents Three different types of oesophageal stent were used. The polythene-covered Gianturco stent (William Cook Europe, Bjaeverkov, Denmark) has a 16-mm lumen and is available in lo-, 12- and 14-cm lengths mounted on a 24-Fr introducer sheath with a tapered dilator. The polyurethane-covered Wallstent (Telestep Device; Schneider, Bulach, Switzerlandj comes in two sizes: 20 mm wide and 11 cm long, and 25 mm diameter and 10.5 cm long, mounted on 18-Fr and 22-Fr delivery systems respectively. The third type used was the Strecker oesophageal stent (Boston Scientific, St Albans, UK). This is a knitted mesh of a self-expanding metallic alloy of nickel and titanium. It has a z 0 1995 Blackwell Science Ltd 11/14 Single copy made for private research or study by RACS Library on 14/11/17