Colonoscopic perforation: useful parameters for early diagnosis and conservative treatment

Abstract

The purpose of this study is to determine useful parameters for the early diagnosis of colonoscopic perforation and to select those who would require surgical treatment. We retrospectively reviewed the demographics, clinical and colonoscopic data, diagnostic-surgical interval, operative findings, complications, and hospital stay of patients who developed postcolonoscopy iatrogenic colonic perforation between January 2002 and December 2008. A retrospective multicentric study of patients diagnosed of colonoscopic perforation was performed. Fifty-four patients were found for final analysis (mean age, 71 years (26-91 years). Thirty-four were diagnostic and 20 were therapeutic colonoscopies. Most patients in whom the perforation was noticed during colonoscopy were treated surgically (p = 0.032) within 24 h (p = 0.004) and had a lesser degree of surgical peritonitis (p = 0.033). Those with deficient bowel preparation had more interventions (p < 0.05), ostomies (p = 0.015), and complications (p = 0.023) as well as major clinical (p < 0.001) and surgical peritonitis (p = 0.031). Patients with nonoperative management had fewer complications (p = 0.011) and lower hospital stay (p < 0.048). Surgical treatment within 24 h resulted in a lesser degree of surgical peritonitis (p < 0.001), fewer intestinal resections (p < 0.001), ostomies (p = 0.002) and complications (p < 0.047), and shorter hospital stay (p < 0.05). We recommend a conservative treatment for patients with the following conditions: good general health, unnoticed perforation during endoscopy, early diagnosis, no signs of diffuse peritonitis, proper colonic preparation, and a different injury mechanism to traction. Patients treated surgically after the first 24 h are likely to have a greater degree of peritonitis and more intestinal resections, ostomies, and complications.