Prognosis in intra-abdominal infections. Multivariate analysis on 604 patients

Objectives: To identify factors that influence mortality in patients who are affected by intra-abdominal infections (IAIs) and to make a comparison among three different scoring systems: the Acute Physiology and Chronic Health Evaluation (APACHE) II score, the sepsis score of Elebute and Stoner, and the Mannheim peritonitis index. Design: Case series. Setting: Both primary and referral hospital care. Patients: The hospital records of 604 patients who consecutively underwent emergency operations for unequivocal IAIs, both spontaneous and postoperative, from 1981 to 1993, were retrospectively reviewed. Patients with IAIs that were related to peritoneal dialysis or infected ascites, those patients who were affected by primary peritonitis from a distant site, and those patients who underwent operations for acute appendicitis or acute cholecystitis without peritoneal contamination were excluded from the study. Univariate and multivariate analyses were used to calculate the prognostic significance of the following variables: age (< or = 70 vs > 70 years); sex; type (spontaneous vs postoperative) and extent (localized vs diffuse) of infection; preoperative serum levels of albumin, cholesterol, and hemoglobin; preoperative total lymphocyte count; amount of intraoperative blood loss; presence of preoperative organ impairment; the APACHE II score; the sepsis score of Elebute and Stoner; and the Mannheim peritonitis index. Main outcome measure: Death was the outcome variable that was studied. Results: Multivariate logistic regression analysis showed that the APACHE II score, the Mannheim peritonitis index, hypoalbuminemia, hypocholesterolemia, and preoperative organ impairment were independent predictors of death. Conclusions: Results showed a significant dominance of host-related factors over the type and source of infection on the prognosis of patients with IAIs. Both the APACHE II score and the Mannheim peritonitis index correctly graded IAI severity and were strongly and independently associated with the outcome; however, the latter score has the advantage of being easier to calculate.