Inadequate source control and inappropriate antibiotics are key determinants of mortality in patients with intra-abdominal sepsis and associated bacteremia

Document Type


Publication Title

Surgical Infections


Background: Patients with intra-abdominal sepsis and associated bacteremia have a high mortality rate. However, outcomes studies in this population are limited, in part because of the small numbers of such patients. The objective of this study was to describe characteristics of critically ill patients with secondary blood stream infection (BSI) of intra-abdominal origin and identify predictors of mortality. Methods: This retrospective, single-center study evaluated patients admitted between January 2005 and January 2011 with bacteremia because of an intra-abdominal source. Patients were included if they met criteria for severe sepsis based on International Classification of Diseases, 9th Revision (ICD 9) codes for acute organ dysfunction, had a positive blood culture, had at least one ICD 9 code indicative of an intra-abdominal process, and had a confirmed or clinically suspected intra-abdominal infection (IAI) within 72 h of the blood culture. Chart review was performed to confirm the presence of these criteria and also the absence of any other potential source of bacteremia. Data were collected on patient demographics, BSI source, source control procedure details, microorganisms isolated, and antibiotics administered. Multivariable logistic regression analysis was performed to identify independent predictors of mortality. Results: Three hundred six patients with BSI were identified, of which 108 episodes were deemed to be of intra-abdominal origin. Gram-negative organisms comprised 43% of blood isolates, followed by gram-positives (33%), anaerobes (14%), and yeast (9%). Appropriate antimicrobial therapy was administered in 71% of patients. The overall mortality rate was 27.8%. As compared with survivors, non-survivors were older, more likely to have underlying COPD or asthma, and have renal or metabolic failure (p<0.05 for all). Among non-survivors, adequate source control was obtained less frequently (64% vs. 91%, p=0.002) and median time to appropriate antibiotics was longer (23 h vs. 4 h, p=0.004). Logistic regression analysis revealed inadequate source control (p=0.002) and inappropriate antibiotics (p=0.016) to be independently associated with mortality. Conclusions: Critically ill patients with a BSI of abdominal origin are at high risk for mortality. Failure to achieve adequate source control and administration of inappropriate antibiotics were independent predictors of mortality. Thus, these represent potential opportunities to impact outcomes in patients with complicated IAI.

First Page


Last Page




Publication Date