Evolving management of colonoscopic perforations

Abstract

Background: Perforations of the large bowel during diagnostic or therapeutic colonoscopy are a rare but significant complication. Their treatment has evolved over the last decade, but there are still no specific guidelines for their optimal management.

Materials and methods: Retrospective review of 105,786 consecutive colonoscopies performed in a 21-year period allowed assessment of the medical records in all patients treated at our institution for colonoscopic perforation.

Results: Thirty-five patients suffered perforation (perforation rate 0.033%) during colonoscopy from January 1986 to October 2007 (14 men, 21 women; mean age 69.4 years). Twenty-four of the perforations occurred during diagnostic colonoscopy, whereas 11 during therapeutic colonoscopy. Twenty-three (66%) of the patients underwent operative treatment and 12 (34%) were managed nonoperatively. The average length of stay was 15.2 days, and there was one death (2.9% 30-day mortality rate) among the patients.

Conclusions: Perforations from diagnostic colonoscopy usually are large enough to warrant surgical management, whereas perforations from therapeutic colonoscopy usually are small, leading to successful nonoperative treatment. Over the last decade, the surgical treatment of colonoscopic perforations has evolved, as there has been a trend that favors primary repair versus bowel resection with successful outcome. Careful observation and clinical care adherent to strict guidelines for patients treated nonoperatively is appropriate in order to minimize morbidity and mortality and identify early those who may benefit from operation. Each treatment, however, has to be individualized according to the patients' comorbidities and clinical status, as well as the specific conditions during the colonoscopy that lead to the perforation.