Abstract

Title: Impact of Methicillin-resistant Staphylococcus aureus Polymerase Chain Reaction Nasal Screening in Non-critically Ill Patients with Sepsis of Unknown Origin

Purpose: Vancomycin should be included as part of empiric therapy in treating sepsis of unknown origin when risk factors for MRSA infection are present. MRSA PCR screening for colonization has utility for de-escalation of therapy with indications such as pneumonia, but its value for sepsis remains to be elucidated. This study evaluated the impact of MRSA colonization data on the duration of vancomycin in non-critically ill patients with sepsis of unknown origin.

Methods: This was a single-center, observational study at a 900-bed community academic medical center from November 2023 through September 2025. Patients were included if they were ≥ 18 years and received at least one dose of vancomycin with an indication of sepsis of unknown origin. Patients were excluded if they were on mechanical ventilation, had vasopressor usage, admitted to the intensive care unit (ICU), pregnant, incarcerated, or had concomitant aztreonam use. The primary outcome was discontinuation of anti-MRSA therapy within 72 hours. Secondary outcomes included culture-confirmed MRSA infection, MRSA PCR screening rates, vancomycin duration, hospital length of stay (LOS), 30-day readmission for infection, and 30-day all-cause in-hospital mortality. The study was exempted as a quality improvement initiative by the institutional review board.

Results: Of 1,468 patients screened, 1,168 were excluded primarily due to ICU admission. A total of 300 patients were included—150 received MRSA screening and 150 had no screening. More patients admitted to oncology and stepdown received screening compared patients in general medicine and cardiology floors who were more likely to not receive screening (p=0.01). Other baseline demographics were similar. Of the whole cohort, 50% had two or more MRSA risk factors, 28% had one risk factor, and 22% had no risk factors. Anti‑MRSA therapy discontinuation occurred in 106 (71%) of screened patients compared with 99 (66%) without screening (p=0.46). Culture‑confirmed MRSA infections were identified in 7 (5%) screened versus 6 (4%) unscreened patients (p=1.00), and 19% of screened patients had positive results. Median vancomycin duration was 2 days in both groups (IQR=1-3, p=0.70). Median hospital LOS was 13 days (IQR=6-22) in patients who were screened, compared to 8 days (IQR=5-15) in patients who were not screened (p=< 0.01). Thirty‑day hospital readmission for suspected infection (7% vs 5%, p=0.81) and 30-day all-cause in-hospital mortality (4% vs 1%, p=0.37) was similar in both groups.

Conclusion: In a cohort of non-critically ill patients with sepsis of unknown origin who were prescribed empiric vancomycin, no difference was observed for vancomycin duration of treatment in the presence or absence of MRSA colonization data.

Publication Date

5-14-2026

Presented At:

Florida Residency Conference 2026

Content Type

Presentation

Download

Open Access

Available to all.

Share

COinS