Abstract Rationale Sepsis carries high mortality, and early appropriate antibiotics are critical to improve outcomes. Standard microbial cultures can have long turnaround times which may delay targeted therapy and promote antimicrobial resistance from extended broad empiric antibiotic coverage. In contrast, multiplex PCR (mPCR) panels can rapidly detect a wide range of pathogens and resistance genes within hours. Prior studies have suggested that use of blood culture mPCR panels (first FDA approved in 2012) may shorten time to optimal antibiotics and enable earlier de-escalation of broad-spectrum therapy. We sought to characterize practice patterns and quantify trends in uptake of blood mPCR among patients admitted with sepsis. Methods We performed a retrospective cohort study using the Premier Healthcare Database, an enhanced claims-based database that includes diagnosis codes, charge data, and laboratory and vital signs data from electronic health records from U.S. hospitals. We identified adults ≥18 years admitted between 9/1/2022 to 09/30/2024 with sepsis present on admission (per validated Angus ICD-10 algorithm approximating Sepsis-3) who had blood cultures drawn within 2 days of admission. To ensure a consistent denominator over time, we restricted to hospitals that continuously provided culture data throughout the study period. Blood mPCR testing was identified using standardized LOINC codes corresponding to currently available commercial assays. Temporal trends in mPCR use were assessed by quarter (2022Q4 to 2024Q3) using Cochran-Armitage tests, and patient and hospital factors associated with mPCR testing were evaluated with hierarchical regression models. Results Among 138,799 eligible patients at 84 hospitals, 7,184 (5.2%) received blood mPCR testing at 43 (51.2%) hospitals. From 2022Q4-2024Q3, the proportion of patients receiving mPCR increased significantly from 2.7% to 12.3% (relative increase 355%, p 0.001). Similarly, hospital adoption of mPCR increased as well from 10.6% to 51.2% of hospitals (relative increase 383% p 0.001); see Figure 1. Patients receiving blood mPCR test were more likely to be admitted to ICU (aOR 1.23; 95% CI 1.14-1.32), require vasopressors (aOR 1.36; 95% CI 1.25-1.48), and have cardiac (aOR 1.43; 95% CI 1.34-1.53), renal (aOR 1.28; 95% CI 1.21-1.35), or neurologic (aOR 1.25; 95% CI 1.18-1.33) dysfunction on admission. No significant difference in blood mPCR use were observed by hospital size, teaching hospital status, or insurance payor type. Conclusions We found that use of blood mPCR tests has increased significantly over two years, both at the patient- and hospital-level. Further studies should evaluate clinical outcomes associated with adoption of this test including duration and appropriateness of antibiotics. This abstract is funded by: None
Saul et al. (Fri,) studied this question.