Colonic perforation due to colonoscopy: a retrospective study of 48 cases

Abstract

Background and study aims: The aim of this retrospective study was to analyze data on the treatment of 48 cases of colonic perforation, with a view to defining the criteria for choosing between medical and surgical treatment.

Patients and methods: A questionnaire requesting information about complications of colonoscopy and their treatment was sent out to four hospital gastroenterological and surgical units.

Results: From January 1979 to December 1993, we reviewed the records of 48 cases of colonic perforation following colonoscopy (24 perforations occurred after diagnostic colonoscopy and 24 after therapeutic colonoscopy). Diagnosis of perforation was delayed in 42% of the patients, with a mean delay of two days (range 0.5-7 days). The treatment was surgical in 35 cases, including eight in which previous medical treatment had been unsuccessful. The perforation was in the sigmoid colon in 74% of the surgical population. Operations were carried out using two procedures, including colostomy, in the case of 20 patients (57%). Colostomy closure was performed in 12 patients (60%) with no mortalities. Surgical mortality occurred in five patients (14%), in four cases due to preexisting medical diseases. Medical treatment was attempted in 21 cases, and was successful in 13, mainly in cases in which perforation had occurred after therapeutic colonoscopy (12 patients).

Conclusion: The choice of the right type of treatment for colonoscopic perforation seems to depend on the size of the lesion. Surgical treatment is appropriate when the perforation has occurred during diagnostic colonoscopy, since the lesion in this case is usually a large colonic laceration, whereas nonsurgical treatment seems to be justified after polypectomy, as long as there is rapid clinical improvement.