Retrospective Observational Cohort of ED/Inpatient “Sepsis”: Is This Really Sepsis?
Authors: Barker H, Goodacre S
Journal: Emergency Medicine Journal (EMJ), November 2025
Conclusions:
-
About 1 in 4 patients labeled “sepsis” (meeting Sepsis-3) likely had organ dysfunction from another cause (comorbidity flare or direct effect of infection), not a dysregulated host response.
-
These “alternative explanation” patients were less acutely ill and had lower in-hospital mortality, suggesting that tighter phenotyping could focus aggressive sepsis care on those most likely to benefit.
-
Practical takeaway: pair Sepsis-3 with clinical adjudication (trajectory, inflammatory phenotype, lactate trend, shock/vasopressor need) to avoid over-treating low-risk phenotypes while still not missing true septic shock.
Results:
-
Cohort: 373 admissions; 303 (81.2%) fulfilled Sepsis-3.
-
Among Sepsis-3 positives, 78/303 (25.7%; 95% CI 21.4–30.0%) had alternative organ-dysfunction explanations:
-
Comorbidity exacerbation: 28 (9.2%)
-
Direct effects of infection: 42 (13.9%)
-
Respiratory dysfunction adjudicated from “normal” SpO₂: 8 (2.6%)
-
Severity/outcomes:
-
NEWS: median 5 (IQR 3–8) vs 7 (IQR 5–10) for presumed dysregulated host response (p < 0.001).
-
In-hospital mortality: 19.2% vs 34.7% (p = 0.011).
Methods:
-
Single-centre, retrospective observational study (Jan–Dec 2022).
-
Included hospitalized patients diagnosed with sepsis; confirmed Sepsis-3 and clinically adjudicated whether organ dysfunction reflected dysregulated immune response vs alternative cause (comorbidity flare, direct infection effects, or misclassification of respiratory dysfunction).
Barker H, Goodacre SRetrospective observational cohort study of patients diagnosed with sepsis: is this really sepsis?Emergency Medicine Journal Published Online First: 28 November 2025. doi: 10.1136/emermed-2025-214894