Introduction The introduction of the Bowel Cancer Screening Programme (BCSP) in 2006 has resulted in the detection and removal of increasing numbers of malignant polyps (MPs). Following endoscopic polypectomy, further management is variable among institutions, and it has been suggested that further surgery after polypectomy may be unnecessary. The aim of this study was to document the management and outcome of patients following endoscopic removal of MPs within the six London Bowel Cancer Screening Centres.

Methods All individuals who had MPs reported to be removed completely at colonoscopy from the start of screening to May 2010 were identified from the six London centres. Patient notes were used to obtain data on colonoscopies, histopathology of polyps and surgical specimens, surgical complications, and follow-up.

Results 117 MPs were identified in 115 patients. There was no significant variation in the percentages of patients recommended surgery between centres. However, there were observed differences between the polyp characteristics used to guide the decision for surgery. The presence or absence of tumour budding was not recorded for any patient at three centres, nor for 24 (96%) of patients at another centre, but for all 19 (100%) of patients at a separate centre. Surgery was carried out in 54 patients. Of the available data from 51 patients, 9 (17.6%) had lymph node metastases or residual cancer in the resection specimen; 5 (9.8%) had lymph node involvement, 3 (5.9%) had residual cancer and 1 (2.0%) had both lymph node involvement and residual cancer. No significant associations were found between polyp characteristics and the presence of residual disease. There was, however, a trend for residual disease in the resection specimens of patients with MPs removed from the sigmoid or rectum. Of the data available for 48 patients, 6 (12.5%) had surgical complications; 3 (6.3%) anastomotic leaks, 1 (2.1%) infarcted colon, 1 (2.1%) hospital acquired pneumonia and 1 (2.1%) death. Follow-up data were available for 21 patients who underwent surgery and 47 patients managed conservatively. To date, none has had a recurrence of cancer (median follow up of 12 months, range 1-28 months).

Conclusion The proportion of patients recommended further surgery was similar across London, although factors influencing the decision varied. Despite perceived complete endoscopic resection, residual malignancy was seen in a significant number of cases (17.6%). In addition there was a considerable surgical complication rate (12.5%). Further research is needed to clarify the factors that more accurately predict residual cancer following polypectomy, thus enabling improved selection of patients for surgery.