Research Day
38IS BLOODSTREAM INFECTION AN EPIPHENOMENON IN SURGICAL PATIENTS?
Document Type
Abstract
Date
2021
Abstract
Background: Bloodstream infection (BSI) is not uncommon in the ICU setting. It seems that rather than being another part of the milieu of critical illness in surgical ICU patients, BSI may be an important predictor of mortality.
Methods: Data on all infections treated between 1997 and 2017 in a single Surgical Intensive Care Unit (ICU) were prospectively collected. Patients were categorized by primary site of infection and presence or absence of associated BSI (same organisms grown from primary site and blood). Demographic and outcome variables were compared using Student's t-test or Chi-square analysis. Logistic regression analysis was used to determine independent predictors of death, including the following variables: Age, sex, trauma vs. non-trauma, APACHE II score, days from admission to infection treatment, and presence or absence of BSI.
Results: Over 6,000 patients were studied during this 20 year period, and BSI was associated with increased mortality in patients with concurrent (detected within 72 hours) infections of the abdomen, lung, urine, and other primary site infections, as demonstrated in the attached table. Regression analysis demonstrated that even after correcting for the above variables, BSI was associated with increased mortality (OR = 1.51, 95% CI = 1.21-1.89).
Conclusion: This study demonstrates a statistically significant association that cannot be overlooked. BSI was an independent predictor of mortality, despite patients receiving a longer duration of therapy. Thus, surgeons may consider obtaining blood cultures promptly if a significant primary site infection is detected; this may lead to earlier initiation of treatment and improved outcomes.