The Conversion of Planned Colonoscopy to Sigmoidoscopy and the Effect of this Practice on the Measurement of Quality Indicators

Abstract

A cecal intubation rate (CIR) of >90% is a well-accepted quality indicator of colonoscopy and is consequently monitored within endoscopy units. Endoscopists' desire to meet this target may mean that incomplete colonoscopies are recorded as flexible sigmoidoscopies. The aim of this study was to examine whether the conversion of requested colonoscopies is a clinically significant phenomenon and whether this impacts upon the measurement of quality indicators. A retrospective review of all flexible sigmoidoscopies performed between 1 January 2015 and 31 December 2015 at Nottingham University Hospitals, Sheffield Teaching Hospitals, and Cambridge University Hospitals was performed. Where a colonoscopy was requested but a flexible sigmoidoscopy performed, the patient's records and endoscopy reports were reviewed to determine whether this conversion was decided before the start of the procedure and documented. During the 12-month period, 6,839 flexible sigmoidoscopies were performed by 125 endoscopists. The original requests of 149 sigmoidoscopies could not be retrieved and were therefore excluded from this analysis. Of the 6,690 sigmoidoscopy requests reviewed, 2.8% (n=190) procedures were originally requested as a colonoscopy. On review of patient records, 85 conversions were appropriate according to pre-defined criteria. However, 105 conversions were deemed inappropriate, occurring in patients who had a valid documented indication for colonoscopy and had undergone full bowel preparation. The most common reasons cited included poor bowel preparation (n=37), technically challenging procedure (n=24), at the endoscopist's discretion based on clinical factors (n=21), and obstructing patology (n=8). A clear reason for conversion was not apparent in 11 cases. During the study period, 21,271 colonoscopies were performed and so conversions represent 0.45% of the total requests. When inappropriate conversions were included in individuals' performance data, 15 endoscopists fell to ≤90% target cecal intubation target. A small, but significant number of colonoscopies are converted to flexible sigmoidoscopies at the time of the procedure. This study demonstrates the conversion of colonoscopy to sigmoidoscopy as being a potential limitation of relying on CIR alone. Endoscopy units should consider monitoring the rate of inappropriate conversions to ensure quality.