Introduction The NHS BCSP uses colonoscopy as the primary colonic investigation following a positive faecal occult blood screening test. The standards for completion of colonoscopy are high within the programme and well above the target of 90%. Whilst there is clear guidance for use of a ‘second test’ to complete colonic imaging in that minority with a ‘failed’ index colonoscopy,1little is known about the outcome of these patients. The aim of this study was to determine the outcome of patients following a failed BCSP colonoscopy.
Method All patients that underwent a BCSP colonoscopy at Royal Liverpool University Hospital between 2008–2013 were identified. All cases of failed colonoscopy were fully audited with a minimum of 12 months follow up and comparisons were made with those within the BCSP with completed examinations (Chi-Squared statistical analysis).
Results In the study period 2885 BCSP colonoscopies were performed, caecal intubation rate was 96%. From the 111 failed procedures, 26 (23%) were due to obstructing colorectal cancer (CRC). 108 patients underwent a second test: 67 (62%) had CT colonography, 23 (21%) had a CT abdomen, 17 (16%) had a repeat colonoscopy, 2 (2%) had a barium enema. 3 (3%) patients did not have a second test: 2 declined, and 1 died of cardiac causes post orthopaedic surgery whilst awaiting their test. Once the patients with significant positive findings on the index colonoscopy were excluded (e.g. cancer diagnosis), the risk of finding additional significant pathology on second test was 49%. Sub-categorised, the risk of CRC, extra-colonic malignancy and polyps was 2.3%, 2.3% and 44% respectively. Whilst the risk of finding CRC on the second test was similar to a completed procedure (2.3% vs. 3.2% respectively; p = 0.66), the risks of finding polyps was unsurprising lower (second test 44% vs. complete colon 61%: p < 0.001). The 12-month outcome of patients with failed colonoscopy found that 53% returned to the BCSP, 19% were enrolled to BCSP surveillance and 25% were offered a definitive treatment after MDT (all CRC patients). The 12-month mortality was just 3%: 2 having died from non-GI malignancy and 1 from cardiac causes.
Conclusion This study emphasises the importance of complete colonic examination within the BCSP. For those with an incomplete colonoscopy not caused by an obstructing tumour, there was a 49% chance of finding significant pathology on the second test. The lower incidence of polyps found on the second test confirms colonoscopy as the gold standard for polyp detection. Overall the outcomes are good with just 3% mortality at 12 months (unrelated to CRC) even with a diagnosis of CRC in almost a quarter of cases.
Disclosure of interest None Declared.
Taylor S et al. Guidelines for the use of imaging in the NHS Bowel Cancer Screening Programme. 2nd edition. 2012.