Incidence, risk, management, and outcomes of iatrogenic full-thickness large bowel injury associated with 56,882 colonoscopies in 14 Lithuanian hospitals

Abstract

BACKGROUND: The primary goal of this hospital-based retrospective multicenter case series study was to determine the incidence of large bowel full-thickness injury associated with colonoscopy in Lithuania. We assessed characteristics of patients who were treated as a result of this complication; management and outcomes were the secondary goals of this study. METHODS: The medical records of patients with iatrogenic large bowel perforations resulting from colonoscopy within the period January 1, 2007, to December 31, 2011, were retrospectively reviewed. Representatives of 14 Lithuanian public and private hospitals participated in the survey. RESULTS: A total of 56,882 colonoscopies were performed. Forty patients (23 female and 17 male patients) were reported to have iatrogenic full-thickness large bowel injury. Diagnostic and therapeutic colonoscopies resulted in perforation for 28 of 49,795 patients and 12 of 7,087 patients, respectively. A mean age of 70 years and a female preponderance for this complication was revealed. Sigmoid colon and rectosigmoid junction was perforated in 28 patients. All patients underwent surgical management, either primary repair (70.0 %) or bowel resection (30.0 %). Postoperative complications were diagnosed in 15 patients. Immediate treatment resulted in fewer intestinal resections and shorter hospital stays (p < 0.05). Smoking [odds ratio (OR) 14.4, 95 % confidence interval (CI) 1.16-179.8] and a large size perforation site (15 ± 10 vs. 8 ± 5 mm; OR 1.19, 95 % CI 1.03-1.38) were risk factors for developing a postoperative complication after curative surgery. Six patients died. All deaths were related to diagnostic colonoscopy. CONCLUSIONS: Total incidence of large bowel full-thickness injury in Lithuanian hospitals is 0.07 %. Incidence of this complication after diagnostic and therapeutic colonoscopies is 0.056 and 0.169 %, respectively. [...].