METHODS

The central microbiology laboratory identified SAB patients within our health system from 2013 through the first quarter of 2015. Using the electronic medical record system (EMR), SAB patients ≥19 years of age with blood culture(s) positive for S. aureus were retrospectively enrolled in this study. Exclusion criteria included age <19 years, patient refusal of treatment, or any of the following within 4 days of the initial positive blood culture sample: death, initiation of hospice care, or transfer to another facility.Reference Shorr, Tabak, Killian, Gupta, Liu and Kollef ⁶ Primary sites of infection associated with SAB were determined from the EMR by reviewing physical exam information.

Creighton University Medical Center is a 232-bed academic hospital. The 3 community-based hospitals included in this study (631 total beds) became a part of our health system in 2013. Creighton University Medical Center has a multidisciplinary ASP. In 2014, the health system added rapid molecular identification (BioFire FilmArray blood culture identification, bioMèrieux, Marcy-l'Étoile, France) for positive blood cultures. Positive blood culture identification results were electronically placed in the EMR, which generated alerts to pharmacists at the community hospitals. In contrast, technologists contacted the on-call ASP consultant at the academic hospital via pager. Antimicrobial recommendations were made to the primary care physician or resident after speaking to an ID physician. This project was approved by the Institutional Review Board (IRB) at Creighton University.

Deidentified data were collected for all SAB patients, entered into an Excel spreadsheet (Microsoft, Pullman, WA), and uploaded to the statistical software (SPSS version 24, IBM, Armonk, NY). All spreadsheet data were reviewed by C.J.D. prior to SPSS upload.

The 4 key standards of care for SAB include (1) removal of an intravascular focus (if present) within 4 days, (2) obtaining follow-up cultures every 2–4 days after initial positive blood culture, (3) use of parenteral β-lactam therapy for methicillin-sensitive infection, and (4) administration of ≥28 total days of therapy for complicated SAB.Reference Liu, Bayer and Cosgrove ⁸ The appropriate duration of therapy for uncomplicated infections was considered to be 14 days. Vancomycin use was considered inappropriate when used for ≥7 days for a methicillin-sensitive infection in the absence of a β-lactam allergy or intolerance. Oral therapy was considered inappropriate.

Treatment failure included any of the following: (1) microbiological failure, defined as a blood culture growing S. aureus obtained ≥10 days from the initial positive culture specimen and before the completion of antimicrobial therapy; (2) recurrent bacteremia, defined as a blood culture positive for S. aureus obtained after completion of antimicrobial therapy; (3) late metastatic infection, defined as hematogenous seeding ≥10 days after the initial positive blood culture; or (4) death.Reference Holland, Arnold and Fowler ¹ Complications of treatment for SAB included grade 3 or 4 abnormalities in laboratory parameters.

Statistical comparisons of discrete variables regarding SAB management between the community hospital setting and the academic hospital setting were conducted using the χ² test, Fisher’s exact test, or Wilcoxon rank-sum test, where appropriate. The Student t test was used to compare continuous variables. Mean±SD or percentages are reported. P≤0.05 was considered statistically significant.