Point-of-care presepsin and procalcitonin to rule out sepsis in patients with infectious symptoms
Authors' conclusion
Use of presepsin yielded high negative predictive value for sepsis, especially when used in combination with PCT.
Study context
- Population
- Adult Emergency Department (ED) patients with acute infectious symptoms (antibiotics ordered, infectious source on imaging or lab testing, or SIRS)41 with sepsis
- Participants
- 149 patients
- Comparison
- Point-of-care presepsin, procalcitonin (PCT) and their combination for ruling out sepsis
- Outcome
- Diagnosis of sepsis
Methodological detailssingle site, prospective, pilot observational cohort, clinical team blinded to biomarker results
- Single site
- Prospective
- Pilot observational cohort
- Clinical team blinded to biomarker results
Performance
As reported in the abstract
Sepsis by biomarker risk category; NPV for sepsis
| Test or model | Value | Sepsis in moderate- or high-risk | NPV | Sepsis without low-risk for both |
|---|---|---|---|---|
| Presepsin | 5.1% | 35.5% (39/110) | 94.9% (95% CI of 88.0%, 100%) | |
| PCT | 9.2% | 60.4% (29/48) | 90.8% (95% CI of 85.1%, 96.5%) | |
| Combination | 0% | 100% (95% CI of 90.3%, 100%) | 34.5% (38/110) |
Value
Presepsin5.1%
PCT9.2%
Combination0%
Sepsis in moderate- or high-risk
Presepsin35.5% (39/110)
PCT60.4% (29/48)
Combination— not stated in the abstract
NPV
Presepsin94.9% (95% CI of 88.0%, 100%)
PCT90.8% (95% CI of 85.1%, 96.5%)
Combination100% (95% CI of 90.3%, 100%)
Sepsis without low-risk for both
Presepsin— not stated in the abstract
PCT— not stated in the abstract
Combination34.5% (38/110)
- Presepsin low-risk vs Presepsin moderate- or high-risk: P < 0.0001 — chi-square
- PCT low-risk vs PCT moderate- or high-risk: P < 0.0001 — chi-square
- Both low-risk vs Not low-risk for both: P < 0.0001 — chi-square
Original publication
Point-of-care presepsin and procalcitonin to rule out sepsis in patients with infectious symptoms. J Crit Care. Published 22 May 2026. doi:10. 1016/ j. jcrc. 2026. 155616
Composed by AI from the published abstract and checked against it. How Briefs are made →