Back to 2008 Program and Abstracts
Selective Management of Iatrogenic Colonoscopic Perforations
Dimitrios V. Avgerinos*, Omar H. Llaguna, Andrew Y. Lo, I. Michael Leitman
Surgery, Beth Israel Medical Center, New York, NY
Introduction: Colonoscopy is a safe procedure in the hands of experienced physicians. However, perforations are unavoidable and occur in rare occasions, with a reported incidence rate of 0.03-0.19% in the literature. The management of those perforations remains controversial. The present study evaluates the treatment options for patients with colonoscopic perforations based on their clinical presentation and indications for colonoscopy.
Methods: After Institutional Review Board approval, a retrospective analysis of all the patients that were treated for colonic perforation after diagnostic or therapeutic colonoscopy at a tertiary care teaching hospital over a twenty one-year period (from January 1986 to October 2007). Patient demographics, past medical and surgical history, type and indications for colonoscopy, colonoscopic findings, clinical presentation after suspected perforation, type of operation and intra-operative findings when operative management was undertaken, and method of non-operative approach were analyzed. Injury was verified by ability to visualize the intra-peritoneal space or diagnostic imaging. The decision to perform surgery was based upon the physical findings of peritonitis, signs of infection or hemodynamic changes.
Results: A total of 105,786 colonoscopies during the study period. Thirty-five perforations occurred (0.033% perforation rate). Twenty (57%) of these perforations followed routine diagnostic colonoscopy, whereas the rest 15 (43%) occurred following some type of therapeutic colonoscopy. Nineteen (95%) of the patients that previously underwent diagnostic colonoscopy required operative treatment, whereas only three patients (20%) of the patients that had perforation following therapeutic colonoscopy required operation and were managed conservatively with antibiotics and bowel rest. Both groups had comparable outcomes, with minimal morbidity and zero mortality.
Conclusions: Although many surgeons recommend operation for all colonoscopic perforations, the present study shows that careful selection of patients who suffer bowel perforations during colonoscopy may be managed without surgery. Perforations resulting from therapeutic colonoscopies usually result in small injuries to the bowel wall, which often heal without the need for surgical intervention. On the other hand, perforations from diagnostic colonoscopies are usually large linear lacerations that require surgical repair or resection.
