Oesophageal resection for high-grade dysplasia in Barrett's oesophagus

Background: The aims of this study were to evaluate the prevalence of invas ive cancer in patients with high-grade dysplasia in Barrett's oesophagus an d to verify whether a second endoscopy with multiple biopsies could improve the accuracy of preoperative diagnosis. In addition, the mortality, morbid ity and survival rates in patients with high-grade dysplasia having oesopha geal resection were recorded. Methods: Fifteen patients were observed from 1982 to 1998; the first seven patients were offered primary oesophageal resection after diagnosis. The ot her eight patients underwent a second endoscopy with a median of 12 biopsie s examined. All later underwent oesophageal resection. Results: Invasive adenocarcinoma was found in five patients, with a minimal difference between the first and second periods (two of seven versus three of eight). There were no perioperative deaths. Early morbidity was observe d in eight patients and late morbidity in four. The actuarial survival rate was 79 per cent at 5 years. The Karnofsky status was unchanged from preope rative values in 13 of 15 patients after a median follow-up of 46 months. Conclusion: These patients with high-grade dysplasia had a 33 per cent prob ability of harbouring invasive oesophageal carcinoma but even a second endo scopy failed to identify patients with invasive tumour. Oesophagectomy was performed with no deaths and remains a rational treatment in patients fit f or surgery.