Historical collection · First published Dec 9, 2024; added to EM Weekly Oct 28, 2026. This retrospective summary may be superseded by later evidence or guidance.
← All articlesBiomarker-Guided Antibiotic Duration for Hospitalized Patients With Suspected Sepsis: The ADAPT-Sepsis Randomized Clinical Trial
Summary
ADAPT-Sepsis randomized 2760 critically ill adults with suspected sepsis in 41 UK ICUs to daily procalcitonin-guided antibiotic discontinuation, daily C-reactive-protein-guided discontinuation, or standard care. PCT guidance reduced mean total antibiotic duration from 10.7 to 9.8 days (mean difference 0.88 days, 95% CI 0.19–1.58; P=.01); 28-day mortality was 20.9% versus 19.4% and met the prespecified 5.4% noninferiority margin (absolute difference 1.57%, 95% CI −2.18 to 5.32; P=.02 at the one-sided .025 threshold). CRP guidance did not reduce total antibiotic duration and its mortality noninferiority result was inconclusive.
Why it matters
For critically ill septic patients after initial antibiotic treatment, a protocolized daily PCT strategy can modestly shorten antibiotic exposure. This supports PCT as a discontinuation aid, not as a reason to delay or withhold empiric antibiotics in the ED.
Study at a glance
- Design
- Randomized trial
- Population
- Adults requiring critical care within 24 hours of starting IV antibiotics for suspected sepsis and expected to receive antibiotics for at least 72 hours at 41 UK NHS ICUs.
- Sample size
- 2760
- Intervention
- Daily procalcitonin-guided antibiotic discontinuation protocol; a separate daily CRP-guided arm was also tested.
- Comparator
- Standard care without protocolized daily biomarker-guided discontinuation.
- Primary endpoint
- Total antibiotic treatment duration to 28 days (effectiveness) and 28-day all-cause mortality (safety/noninferiority).
Main findings
- PCT-guided care reduced mean total antibiotic duration by 0.88 days versus standard care (9.8 vs 10.7 days; 95% CI for mean difference 0.19–1.58; P=.01).
- Twenty-eight-day mortality was 20.9% with PCT guidance versus 19.4% with standard care; absolute difference 1.57% (95% CI −2.18 to 5.32) met the prespecified 5.4% noninferiority margin at P<.025.
- CRP guidance did not reduce total antibiotic duration (10.6 vs 10.7 days; mean difference 0.09, 95% CI −0.60 to 0.79; P=.79), and mortality noninferiority was inconclusive.
- Nine serious adverse events were equally distributed across groups; 90-day mortality did not differ.
Limitations
This was a UK ICU trial, not an ED antibiotic-initiation trial, and the average reduction was less than one day. The upper confidence bound for the PCT mortality difference (5.32%) was very close to the prespecified 5.4% noninferiority margin; later correspondence debated the clinical interpretation of that safety margin.
Read the primary source
Full text: Open-access full text available.
Original sourcePMID: 39652885
Material reviewed: full text. Latest correction: Dec 17, 2025.
Source retrieved Oct 28, 2026. Historical backfill. The current corrected JAMA full text, PubMed/PMC metadata, and the publisher correction were reviewed. JAMA verifies Published Online 2024-12-09; PMID 39652885 and DOI 10.1001/jama.2024.26458 were cross-checked. A Dec 17, 2025 correction clarified the Methods description of the noninferiority approach and replaced the first plot of eFigure 3; this record uses the corrected current text. Most participants originated in the ED, but the randomized intervention occurred in critical care and should not be extrapolated to biomarker-guided antibiotic initiation.
Corrections & review history
Dec 17, 2025 — JAMA corrected the Methods description of the noninferiority approach and replaced the first panel of eFigure 3; the current corrected full text was used for this historical summary.