PRONTO — Procalcitonin + NEWS2 for Sepsis Identification in the ED

1. Publication

• Title: Procalcitonin testing combined with NEWS2 evaluation compared with usual care based on NEWS2 for identification of sepsis and antibiotic initiation in the emergency department in England and Wales (PRONTO): a multicentre, randomised, controlled, open-label, phase 3 trial
• Acronym: PRONTO
• Year & Journal: Lancet Respiratory Medicine, epublished March 22, 2026 (print: 2026;14(5):417-431)
• Citation: Todd S, Euden J, Condie J, et al; PRONTO Trial Group. Lancet Respir Med. 2026;14(5):417-431. doi:10.1016/S2213-2600(25)00433-3

2. Context & Rationale

Background: NEWS2 detects physiological deterioration but is not specific for bacterial sepsis. Prior procalcitonin (PCT) literature suggesting shortened antibiotic exposure and improved survival came mostly from ICU or lower-respiratory-infection cohorts, not undifferentiated ED suspected-sepsis populations. A 2025 ED-focused diagnostic meta-analysis found only moderate PCT performance (pooled sensitivity 0.62, specificity 0.80). Antibiotic overtreatment is a documented problem: in a 2025 seven-hospital ED cohort, 31.5% of broadly treated suspected-sepsis cases were unlikely to have bacterial infection, and 17.3% of definite/probable bacterial infection cases developed possible antibiotic-associated complications.

Research Question/Hypothesis: Does adding rapid point-of-care PCT to NEWS2-based triage reduce IV antibiotic initiation by 3 hours without worsening (non-inferior for) 28-day mortality, compared with NEWS2-based usual care alone, in adults managed as suspected sepsis in UK EDs?

Why This Matters: Sepsis causes ~48,000 deaths annually in England; the decision of whether to give immediate broad-spectrum antibiotics, seek brief diagnostic clarification, or pursue an alternative diagnosis is one of the most consequential and time-critical calls in acute care.

3. Design & Methods

• Study Type: Parallel, two-arm, open-label, individually randomized, pragmatic, multicenter phase 3 trial with co-primary effectiveness/safety endpoints and adaptive group-sequential stopping rules
• Setting & Centers: 20 EDs across 17 NHS Trusts/Health Boards, England and Wales
• Population:
• Inclusion: Adults ≥16y managed by local teams as suspected sepsis per site sepsis pathways (NEWS2 mandatory, local triggers permitted), regardless of baseline NEWS2
• Exclusions: IV antibiotics already started at screening; current immunosuppressive/myeloablative chemotherapy; solid-organ or allogeneic transplant within 3 months; urgent surgical intervention; end-of-life directives excluding antibiotics
• Intervention: Point-of-care PCT (turnaround <20 min) integrated into a NEWS2-based risk algorithm (thresholds <0.5, 0.5–1.9, ≥2.0 μg/L) directing urgency of senior review and antibiotic timing; clinicians retained full discretion to override
• Comparator: Usual care based on NEWS2 plus local standard sepsis assessment, without mandated PCT
• Blinding: Open-label (participants, staff, outcome assessors, and statisticians all unblinded)
• Statistical Power & Follow-Up: Powered on 28-day mortality: 7002 participants needed to exclude a >2.5% absolute mortality increase (assumed control mortality 15.0%, non-inferiority margin 17.5%, 90% power, one-sided α=0.05); co-primary stewardship endpoint needed only 532 to detect a 10% absolute reduction. Recruitment target after interim-analysis inflation: 7676. Modified ITT complete-case analysis; follow-up to day 28 (mortality/safety to day 90).

4. Key Results

Trial was not stopped early despite two interim analyses crossing the mortality superiority boundary — IDMC recommended continuation to full sample size.

Subgroup analyses: No prespecified subgroup significantly modified either co-primary effect (NEWS2, infection site, suspected COVID-19, PCT platform, recruitment epoch, ED crowding). Mortality benefit spanned both infectious and non-infectious discharge diagnoses; post-hoc deprivation analysis suggested greater benefit in more deprived groups (interaction P=0.0064).

5. Internal Validity Assessment

• Randomization & Allocation: Secure web-based 1:1 randomization, site + NEWS2 minimization with 80% random element — strong concealment.
• Protocol Adherence & Separation: Incomplete: PCT result available in 2930/3092 (94.8%), seen by clinicians in 63.5%, algorithm followed in only 46.2%. IV antibiotics within 3h remained virtually identical (48.4% vs 48.2%) despite meaningful risk reclassification (24.6% of NEWS2-high-risk patients reclassified low-risk by PCT).
• Missing Data & Sensitivity Analyses: Deferred consent caused substantial attrition: 1548/7667 (20.2%) never consented; primary analysis population 5453/7667 (71.1%). Mortality among withdrawn participants was balanced between arms (9.2% vs 9.1%).
• Blinding & Detection Bias: Open-label; objective co-primary outcomes limit detection bias, but performance bias is the central concern.
• Overall Internal Validity Conclusion: Moderate — strong randomization and objective endpoints, but substantial non-consent, limited protocol adherence, and minimal antibiotic-exposure separation leave the mortality mechanism unexplained.

6. External Validity Assessment

• Population Representativeness: Broad, real-world suspected-sepsis ED population (median age 72–73y, Charlson index ~4), closer to real-world sepsis-alert populations than confirmatory-infection cohorts.
• Practice Context: Generalizable to NEWS2-based, nurse-led triage systems with rapid point-of-care PCT availability; less applicable to systems using different screening tools or lacking fast biomarker platforms. Predominantly White population (84.5% vs 85.2%) and entirely NHS-workflow-embedded, which may limit portability to other health systems.
• Overall External Validity Conclusion: Moderate-to-good for UK-like EDs with NEWS2-centred pathways; limited for highly immunocompromised, urgent surgical sepsis, or non-NHS systems lacking equivalent biomarker infrastructure.

7. Strengths & Limitations

Strengths:

• Largest randomized ED PCT-guided suspected-sepsis trial to date
• Highly pragmatic, broad-inclusion, deferred-consent design
• Objective co-primary outcomes capturing both effectiveness and safety
• Mortality signal stable across adjusted, complete-case, imputed, and compliant-average causal-effect analyses

Limitations:

• Open-label; performance bias is the principal interpretive concern
• 20.2% post-randomization non-consent/attrition
• Minimal antibiotic-exposure separation despite meaningful biological risk reclassification
• Single-country, NEWS2-centric, predominantly White cohort limits portability
• Mortality mechanism not directly measured — the most important finding is the least well explained

8. Interpretation & Practice Impact

• Clinical Implications: Baseline PCT did not reduce early antibiotic initiation (the intended stewardship goal was negative), but the mortality benefit argues for serious consideration of rapid diagnostic support at ED presentation.
• Mechanistic Coherence: Benefit despite unchanged antibiotic timing suggests value may come from improved diagnostic orientation, prioritization, or escalation rather than antibiotic withholding — a mechanism not directly measured by this trial.
• Systems-Level Takeaway: Future biomarker trials should measure downstream diagnostic workup, source control, and escalation pathways, not antibiotic timing alone.

9. Controversies & Subsequent Evidence

• Editorial Commentary: An accompanying editorial (Cabral & Rhee) argued that unmeasured improvements in diagnostic clarity or escalation prioritization are more plausible explanations than antibiotic stewardship itself, and flagged that the one-tailed non-inferiority framework with 90% CIs, while prespecified, warrants careful interpretation.
• Guideline Integration: The 2026 Surviving Sepsis Campaign international guideline was published one day after PRONTO (March 23, 2026) and is the current guideline backdrop, but not yet a mature consensus response to PRONTO's specific findings. Related ICU evidence remains mixed: ADAPT-Sepsis showed serial PCT safely shortened antibiotic duration by 0.88 days; MULTI-CAP failed its primary antibiotic-free-days endpoint despite a 3-day reduction in cumulative duration.

10. Summary & Executive Takeaway

Summary: PRONTO randomized suspected-sepsis ED patients across 20 NHS EDs to PCT-guided NEWS2 care versus NEWS2-alone usual care. The stewardship co-primary endpoint was negative (IV antibiotics by 3h: 48.4% vs 48.2%), but 28-day mortality fell from 16.6% to 13.6% (adjusted RD −3.12 percentage points, P=0.0009), with benefit persisting to 90 days.

Overall Takeaway: PRONTO overturns the simplest expectation for admission PCT — antibiotics did not fall, yet mortality did — shifting the field from "can a biomarker suppress prescribing" to "how does rapid diagnostic support change early clinical reasoning and escalation." Practice-informing and research-directing, not yet a mandate for unqualified universal implementation pending mechanistic replication.

11. Bibliography

• Euden J, Thomas-Jones E, Aston S, et al. PRONTO protocol. BMJ Open. 2022;12(6):e063424.
• Dark P, Hossain A, McAuley DF, et al; ADAPT-Sepsis Collaborators. JAMA. 2025;333(8):682-693.
• Prescott HC, Antonelli M, Alhazzani W, et al. Surviving Sepsis Campaign 2026 guideline. Crit Care Med. 2026. doi:10.1097/CCM.0000000000007075.
• Cabral S, Rhee C. Rethinking the role of procalcitonin in suspected sepsis [editorial]. Lancet Respir Med. 2026.