An evidence-based treatment algorithm for colorectal polyp cancers: results from the Scottish Screen-detected Polyp Cancer Study (SSPoCS).

OBJECTIVES: Colorectal polyp cancers present clinicians with a treatment dilemma. Decisions regarding whether to offer segmental resection or endoscopic surveillance are often taken without reference to good quality evidence. The aim of this study was to develop a treatment algorithm for patients with screen-detected polyp cancers. DESIGN: This national cohort study included all patients with a polyp cancer identified through the Scottish Bowel Screening Programme between 2000 and 2012. Multivariate regression analysis was used to assess the impact of clinical, endoscopic and pathological variables on the rate of adverse events (residual tumour in patients undergoing segmental resection or cancer-related death or disease recurrence in any patient). These data were used to develop a clinically relevant treatment algorithm. RESULTS: 485 patients with polyp cancers were included. 186/485 (38%) underwent segmental resection and residual tumour was identified in 41/186 (22%). The only factor associated with an increased risk of residual tumour in the bowel wall was incomplete excision of the original polyp (OR 5.61, p=0.001), while only lymphovascular invasion was associated with an increased risk of lymph node metastases (OR 5.95, p=0.002). When patients undergoing segmental resection or endoscopic surveillance were considered together, the risk of adverse events was significantly higher in patients with incomplete excision (OR 10.23, p<0.001) or lymphovascular invasion (OR 2.65, p=0.023). CONCLUSION: A policy of surveillance is adequate for the majority of patients with screen-detected colorectal polyp cancers. Consideration of segmental resection should be reserved for those with incomplete excision or evidence of lymphovascular invasion.

Does the location of colorectal carcinoma differ between screened and unscreened populations?

AIM: Screening for colorectal malignancy using faecal occult blood testing is established across the UK. In NHS Grampian the programme was introduced in 2007. Previous studies have reported no difference in anatomical locations of cancers detected by screening programmes compared with those in unscreened populations. This study aims to review the location of tumours detected in an established screening programme compared with those diagnosed through symptomatic presentation within the same population. METHOD: All patients discussed at the regional multidisciplinary meeting between June 2007 and August 2011 were included. Data were collated prospectively from multidisciplinary team records while site of tumour was documented from radiology, endoscopy, operative and pathology reports. Comparative statistics (χ(2) ) were performed using spss 19. RESULTS: Of 1487 patients included 255 were detected via the screening programme and 1232 from symptomatic presentation. More left sided tumours (splenic flexure to rectosigmoid) were detected via screening (P=0.005). Of non-screened patients (n=1232), 456 (37%) tumours were right sided (caecum to distal transverse colon), 419 (34%) were left sided and 357 (29%) were in the rectum. This compares with the screened group (n=255): right sided 74 (29%), left sided 113 (44%) and rectal 68 (27%). CONCLUSION: More left sided tumours appear to be detected in screened patients compared with symptomatic presentation, contrary to previously published work. These results may be worthy of further consideration given the ongoing debate on the optimal means of screening.

Self-expanding metallic stents for large bowel obstruction.

BACKGROUND: Self-expanding metallic stents (SEMS) may relieve colonic obstruction as definitive therapy or as a bridge to elective surgery. METHODS: This was a retrospective longitudinal cohort study of patients undergoing insertion of SEMS for large bowel obstruction at one institution. Scrutiny of the radiology department's coding system allowed identification of all patients undergoing colonic stent insertion between 2002 and 2008. Data were extracted from patient case notes and investigation reports. RESULTS: Eighty-two patients with a median age of 75 (interquartile range (i.q.r.) 43-94) years underwent stent insertion, 71 for palliation and 11 as a bridge to surgery. Obstruction was due to malignant disease in 67 patients and had a benign cause in 15. Median survival in the palliative setting was 103 (i.q.r. 44-317) days. Complications occurred in 43 patients, of whom 22 underwent unplanned surgery. High-grade obstruction (relative risk (RR) 2·05; P = 0·055) and benign disease (RR 3·45; P < 0·001) were associated with risk of adverse events. CONCLUSION: SEMS should not be used for large bowel obstruction with benign pathology.