Longitudinal Outcomes of Intra-abdominal Infection Complicated by Critical Illness

Background: Critically ill surgical patients remain at high risk of adverse outcomes as a result of intra-abdominal infections, including prolonged length of stay, organ dysfunction, and death despite advances in critical care and innovations in management of the peritoneal cavity. We evaluated the causes and consequences of intra-abdominal infections among critically ill surgical patients in a single tertiary-care intensive care unit (ICU) over a decade. Methods: Prospective study of 465 critically ill surgical patients with hollow viscus perforation and peritonitis or abscess from 1991-2002. Data collected were age, gender, admission APACHE III score, multiple organ dysfunction score, ICU and hospital length of stay, abscess (yes/no), site and type of perforation (colon vs. other), de novo vs. nosocomial origin, and mortality. Statistical analysis was by univariate ANOVA for coordinate data, Fisher exact test for continuous data, and logistic regression analysis. Results: The incidence of intra-abdominal infection was 5.75%, 73.7% of the patients developed organ dysfunction, and mortality was 22.6%. Females comprised 46.8% of the patients. De novo infection represented 71.8% of cases, whereas nosocomial infection comprised 28.2% of cases. Perforations were of the colon (including the appendix) 49.9% of the time. An abscess formed in 22.3% of patients; the remainder had peritonitis but no abscess. Patients in the cohort with peritonitis were older (p = 0.0157), sicker on admission (p = 0.0411) and developed more organ dysfunction (p = 0.0072), but had the same rate of mortality. Despite steadily increasing acuity since 1991 (r 2 = .71, p < 0.0001), the magnitude of organ dysfunction (r 2 = 0.11) and the mortality rate remained constant (r 2 = .01). By logistic regression, abscess correlated with less severe organ dysfunction (score ≥ 5 [odds ratio 0.54, 95% CI 0.33-0.90] and ≥9 points [odds ratio 0.38, 95% CI 0.20-0.74]), and increasing magnitude of organ dysfunction was associated with mortality (each point [odds ratio 1.46, 95% CI 1.32-1.61]). Conclusions: Although outcomes are improving, generalized peritonitis still causes high organ dysfunction-related mortality among critically ill surgical patients. Further improvements in resuscitation, surgical technique, and pharmacotherapy of severe intra-abdominal infections are needed.