Complications Associated With Colonoscopy in an Open-Access Referral Center

Abstract

Introduction: Although the benefit of colonoscopy has been well documented, rates of complications related to colonoscopy have been heterogeneous. The San Francisco VA medical center’s (SFVAMC) open-access colonoscopy referral program performs a large number of therapeutic colonoscopies. The aim of this study was to quantify the incidence and to characterize the type and severity of colonoscopyrelated complications in an open-access referral system with a high adenoma detection rate (>40%) and snare polypectomy rate (>40%). Methods: All patients undergoing colonoscopy from January 2004 to December 2013 identified as having a colonoscopy-related complication within the SFVAMC GI quality improvement database were selected for in-depth chart review. Colonoscopy-related complications (hemorrhage, post-polypectomy syndrome, perforation, cardiopulmonary event, and other medical complaints attributed to colonoscopy for which patients sought medical advice) occurring within 30 days of procedure were included. Complications were captured by a variety of mechanisms: 48 hour post-procedure follow-up phone calls, emergency department visits and hospitalization within the VA, and billing by out-of-network hospitals to the VA for emergency department and inpatient care directly related to procedure complication. Results: Eighty-five colonoscopies resulted in 87 complications among a total of 17,393 colonoscopies (5.00 per 1,000 colonoscopies [95% confidence interval (CI) 4.06-6.17]). The median time to all complications was one day (IQR 0-2 days). Complications requiring hospitalization occurred in 3.56 per 1,000 colonoscopies (95% CI 2.78-4.57). Hemorrhage occurred in 2.18 per 1,000 colonoscopies (95% CI 1.59-3.00) with 12% (4/33) of all hemorrhage due to forceps biopsy (Table 1). Post-polypectomy syndrome occurred in 0.52 per 1,000 colonoscopies (95% CI 0.27-0.99), all of which required hospitalization. There were 2 deaths within 30 days of the procedure (0.11 per 1,000 [95% CI 0.03-0.46]), one due to myocardial infarction and one of unknown etiology.Table 1Conclusion: In an open access referral colonoscopy program with a high proportion of therapeutic colonoscopies, the rate of all reported complication was 5.0 per 1,000 colonoscopies, and rate of serious complications 3.4 per 1,000 colonoscopies. These complication rates are similar to rates reported in the literature in other settings.