The Incidence of Bowel Perforations Is Positively Related to the Intensity of Pain Experienced During Simulated Colonoscopy

Abstract

Background: Painful colonoscopy is due to looping (90%) in the sigmoid colon (56%). i Shah SG, Brooker JC, Thapar C, Williams CB, Saunders BP. Patient pain during colonoscopy: an analysis using real-time magnetic endoscope imaging. Endoscopy. 2002 Jun;34(6):435-40 iShah SG, Brooker JC, Thapar C, Williams CB, Saunders BP. Patient pain during colonoscopy: an analysis using real-time magnetic endoscope imaging. Endoscopy. 2002 Jun;34(6):435-40 The incidence of bowel perforation is estimated at 0.1% in a recent survey of 5000 colonoscopies. This study aims to look at the relationship of intensity of pain with the incidence of bowel perforation, in simulated colonoscopy. Such a test on real patients would be difficult for ethical reasons. Methods: Operators of variable experience; novice (< 10 real procedures done), intermediate (11-100), and experienced (>100) performed 264 episodes of colonoscopy on the HT Immersion simulator at random. All the procedure recorded the percentage of time the simulator sensed no pain, mild, moderate, severe, or extreme pain and if bowel perforated. Each operator performed three to five colonoscopy episodes. Severity of pain was weighed according to the frequency of occurrence for each procedure, and an increasing severity scale was determined from 1-3. Chi square was calculated to evaluate statistical significance. Data was also evaluated according to the level of experience of the operator. Results: 38 cases developed bowel perforation. It was infrequent to get severe or extreme pain and most procedures were done with mild or no pain. However the probability of bowel perforation progressively increased with experience of increasing intensity of pain (6.3%, 29.3% and 41.2% on scale 1-3 respectively). The probability of safe endoscopy without bowel perforation progressively diminished with increasing severity of pain (93.7%, 70.7% and 58.9%). This difference was statistically highly significant (p < 0.000). There were no perforations in the experienced group but both in the novice and intermediate the results were not affected by the experience of the operator. Discussion: The exact nature of what is measured as pain during simulated colonoscopy is unclear but this study validates it as possibly useful in predicting possibility of bowel perforation during simulation. This study shows that increasing severity of pain is statistically significant predictors of bowel perforation during simulated colonoscopy regardless of the experience of the operator. The results have a potential role in assessment at the time of the completion of training in colonoscopy or for re-accreditation.