PURPOSE: We attempted to identify the factors associated with the early mortality of patients with nontraumatic colorectal perforation. METHODS: Eighty patients who underwent surgery for nontraumatic colorectal perforation between May 1986 and December 1999 were retrospectively reviewed. Age, sex, cause of perforation, duration of symptoms, associated preoperative septic shock, concomitant disorders (including cardiac disease, chronic obstructive pulmonary disease, hemodialysis, and steroid treatment), operative findings (such as the site of perforation and the degree of peritonitis), and results of preoperative laboratory blood tests (such as the white blood cell count and platelet count) were analyzed for their association with early outcome using univariate and multivariate analyses. RESULTS: Fourteen of the 80 patients died during hospitalization. According to the univariate analysis, advanced age, preoperative septic shock, concomitant disabling cardiac disease, hemodialysis, diffuse peritonitis, and a low preoperative platelet count were more frequent in the patients who died during hospitalization. According to the logistic regression analysis, preoperative septic shock (odds ratio 8.443, 95% confidence interval (CI) 1.625-43.873), concomitant end-stage renal failure (odds ratio 13.641, 95% CI 1.643-113.244), and diffuse peritonitis (odds ratio 13.212, 95% CI 1.441-121.102) were the most significant factors related to in-hospital mortality. CONCLUSION: Early diagnosis before the patient's general condition deteriorates is a key to improving the early mortality associated with nontraumatic colorectal perforation, especially in patients with concomitant end-stage renal failure.
PURPOSE: We attempted to identify the factors associated with the early mortality of patients with nontraumatic colorectal perforation. METHODS: Eighty patients who underwent surgery for nontraumatic colorectal perforation between May 1986 and December 1999 were retrospectively reviewed. Age, sex, cause of perforation, duration of symptoms, associated preoperative septic shock, concomitant disorders (including cardiac disease, chronic obstructive pulmonary disease, hemodialysis, and steroid treatment), operative findings (such as the site of perforation and the degree of peritonitis), and results of preoperative laboratory blood tests (such as the white blood cell count and platelet count) were analyzed for their association with early outcome using univariate and multivariate analyses. RESULTS: Fourteen of the 80 patients died during hospitalization. According to the univariate analysis, advanced age, preoperative septic shock, concomitant disabling cardiac disease, hemodialysis, diffuse peritonitis, and a low preoperative platelet count were more frequent in the patients who died during hospitalization. According to the logistic regression analysis, preoperative septic shock (odds ratio 8.443, 95% confidence interval (CI) 1.625-43.873), concomitant end-stage renal failure (odds ratio 13.641, 95% CI 1.643-113.244), and diffuse peritonitis (odds ratio 13.212, 95% CI 1.441-121.102) were the most significant factors related to in-hospital mortality. CONCLUSION: Early diagnosis before the patient's general condition deteriorates is a key to improving the early mortality associated with nontraumatic colorectal perforation, especially in patients with concomitant end-stage renal failure.