BACKGROUND AND AIMS: Endoscopic resection of large colorectal neoplasms is increasingly being used as an alternative to surgery. However data on failure rates, safety and long-term outcomes remain limited. The aim of the study was to report short- and long-term outcomes from endoscopic resection of large colorectal neoplasms from a single centre and use a model to predict mortality had surgery been performed. METHODS: Consecutive patients referred for endoscopic resection of large (> or = 20 mm) colorectal neoplasms from January 2001 to February 2008 were included. Resection details were recorded in a prospectively maintained database. Data was collected on 30-day complication rates, and follow-up colonoscopy findings. The Colorectal-POSSUM score was used to estimate mortality from open surgery. RESULTS: There were 154 large neoplasms in 140 patients. Mean age was 68 years (range 22-94). Mean neoplasm size was 26 mm (range 20-80 mm, 24 > or = 40 mm). Complete endoscopic removal was achieved in 95% of cases. Twenty patients were referred for surgery (14%). In the endoscopy group, there were no deaths within 30 days. Twelve patients had a complication including two perforations. Endoscopic follow-up data was available in 90% of cases and five patients (4%) were found to have residual adenoma that was treated endoscopically with subsequent clearance. If surgery had been performed, the mean predicted mortality was 2.2% (range 0.5-10%). There were two deaths (10%) in patients who underwent elective surgery within 30 days. CONCLUSION: Endoscopic resection of large colorectal neoplasms is safe and effective even for very large benign neoplasms. When the lesion is endoscopically resectable this should be the preferred treatment.
BACKGROUND AND AIMS: Endoscopic resection of large colorectal neoplasms is increasingly being used as an alternative to surgery. However data on failure rates, safety and long-term outcomes remain limited. The aim of the study was to report short- and long-term outcomes from endoscopic resection of large colorectal neoplasms from a single centre and use a model to predict mortality had surgery been performed. METHODS: Consecutive patients referred for endoscopic resection of large (> or = 20 mm) colorectal neoplasms from January 2001 to February 2008 were included. Resection details were recorded in a prospectively maintained database. Data was collected on 30-day complication rates, and follow-up colonoscopy findings. The Colorectal-POSSUM score was used to estimate mortality from open surgery. RESULTS: There were 154 large neoplasms in 140 patients. Mean age was 68 years (range 22-94). Mean neoplasm size was 26 mm (range 20-80 mm, 24 > or = 40 mm). Complete endoscopic removal was achieved in 95% of cases. Twenty patients were referred for surgery (14%). In the endoscopy group, there were no deaths within 30 days. Twelve patients had a complication including two perforations. Endoscopic follow-up data was available in 90% of cases and five patients (4%) were found to have residual adenoma that was treated endoscopically with subsequent clearance. If surgery had been performed, the mean predicted mortality was 2.2% (range 0.5-10%). There were two deaths (10%) in patients who underwent elective surgery within 30 days. CONCLUSION: Endoscopic resection of large colorectal neoplasms is safe and effective even for very large benign neoplasms. When the lesion is endoscopically resectable this should be the preferred treatment.