Search

Background Procalcitonin (PCT) can be elevated with certain bacterial infections. Debate continues as to how to best use this biomarker to guide antibiotic use. The primary objective of this study was to evaluate the correlation of PCT levels and the presence of bacterial infection on admission in the total population and in different disease states. Methods This was a multicenter retrospective cross-sectional study of patients admitted with specified infectious diagnoses to two VA Medical Centers from 4/1/2019 to 7/1/2021. Patients were stratified into 4 cohorts for analysis; those with COVID-19, sepsis from respiratory source_(S-R), sepsis from non-respiratory source (S-NR), and respiratory source without sepsis (R). Electronic medical records were reviewed to collect the following: initial procalcitonin, cultures, SIRS criteria, comorbidities (CKD, ESRD, HF, immunosuppressed, surgery within the 7 days), and c-reactive protein. PCT elevation was defined as ≥0.25 ng/mL. The frequency of positive cultures within 72 hours was evaluated for patients with elevated and normal PCT levels to determine the diagnostic performance of PCT overall and for each cohort. Results 632 of 664 patients were evaluated in this study. PCT is elevated twice as often in the septic groups as compared to the non-septic groups (figure 1). Positive predictive value (PPV) varies from 27% to 63% as compared to negative predictive value (NPV) 53%-79% among the disease state groups (figure 2). Although small numbers, the NPV of PCT improves to 83% in patients with elevated temperature and white blood cells (WBC) (figure 3). Figure 1Figure 2Figure 3 Conclusion The findings that NPV of PCT appears to be better than PPV, support current recommendations against using this as a diagnostic tool, but rather as a tool to assist with antibiotic de-escalation. Further studies are necessary to confirm whether there are specific markers such as temperature or WBC which may improve the NPV. Our data suggests PCT is less helpful in identifying the presence or absence of bacterial infection in septic versus non-septic patients. Disclosures All Authors: No reported disclosures.