Background: Central-line associated bloodstream infections (CLABSI) are a subset of hospital-onset bacteremia and fungemia (HOB), a potential indicator of healthcare-associated infections (HAIs) that can be objectively and directly obtained from electronic health records. We undertook a pilot study to elucidate the causes and determine the preventability of HOB.
Methods: HOB was defined as growth of a microorganism from a blood culture obtained ≥3 calendar days after admission in a hospitalized patient. A random sampling of HOB events across 2 academic hospitals and a pediatric intensive care unit in a third academic hospital were identified between October 1, 2014 and September 30, 2015. Medical records were reviewed to identify potential risk factors and sources of bacteremia. Two physicians used underlying patient factors, microorganism, and other clinical data to rate the preventability of each HOB event in an ideal hospital on a 6-point Likert scale.
Results: Medical records for 60 HOB events (20 in each hospital) were reviewed. The most common organisms were coagulase-negative Staphylococcus (28%) and Candida spp. (17%) (Figure 1). The most likely sources of bacteremia and fungemia included CLABSI (28%) and skin contaminants/commensals (17%) (Figure 2). 49% of HOB events not attributed to skin commensals were rated as potentially preventable (Figure 3). 50% of HOB events randomly sampled across 2 hospitals occurred in an intensive care unit. Central venous catheters, urinary catheters, and mechanical ventilation were present in the previous 2 days among 73%, 20%, and 25% of all HOB events, respectively. Only 10% of all HOB events occurred in a patient without an indwelling device. Only 20% of HOB events resulted in a National Healthcare Safety Network (NHSN) reported CLABSI.
Conclusion: Half of HOB events are potentially preventable in this pilot study. HOB may be an indicator for a large number of preventable HAIs not currently measured by NHSN. Larger studies across a variety of hospital settings are needed assess the generalizability of these results the implications of HOB surveillance for infection prevention practices and patient outcomes.
A. M. Milstone, None
J. T. Jacob, None
S. Chernetsky-Tejedor, None
A. D. Harris, None
S. Leekha, None