Publication

• Title: Hospital and long-term outcomes for subglottic suction and polyurethane cuff versus standard endotracheal tubes in emergency intubation (PreVent 2): a randomised controlled phase 2 trial
• Acronym: PreVent 2
• Year: 2025
• Journal published in: The Lancet Respiratory Medicine
• Citation: Treggiari MM, Sharp ES, Ohnuma T, et al. Hospital and long-term outcomes for subglottic suction and polyurethane cuff versus standard endotracheal tubes in emergency intubation (PreVent 2): a randomised controlled phase 2 trial. Lancet Respir Med. 2025; published online Nov 27.

Context & Rationale

• Background

  • Ventilator-associated pneumonia (VAP) and related ventilator-associated events (VAE) remain frequent complications of invasive mechanical ventilation, associated with prolonged ventilation, antimicrobial exposure, and resource use.
  • Microaspiration of contaminated oropharyngeal and subglottic secretions around the endotracheal tube (ETT) cuff is a plausible mechanistic target for prevention strategies.
  • ETTs incorporating subglottic secretion drainage (SSD) and higher-performance cuffs (e.g., polyurethane) aim to reduce aspiration, but evidence has focused largely on VAP incidence rather than patient-centred outcomes.
  • Prior trials and meta-analyses suggested SSD can reduce VAP incidence in selected ICU cohorts, but heterogeneity in devices, suction strategies, and diagnostic criteria limits inference, and effects on mortality or long-term function remain uncertain.
  • Emergency intubation populations may differ materially (non-elective intubation, high severity of illness, variable pre-intubation contamination and aspiration risk), and SSD-related airway injury and long-term sequelae (voice, swallowing, cognition) have not been robustly evaluated in this context.
• Research Question/Hypothesis

  • In adults undergoing emergency endotracheal intubation with subsequent ICU admission and mechanical ventilation, does a polyurethane-cuff ETT with continuous SSD (PU-EVAC) improve hospital and 6-month outcomes, compared with a standard polyvinylchloride-cuff ETT (PVC)?
• Why This Matters

  • SSD tubes add device cost, nursing/respiratory therapist workload, and may introduce harms (mucosal injury, secretion plug/occlusion, suction-related trauma); adoption should be anchored to demonstrable net patient benefit.
  • Contemporary practice includes prevention bundles and a shift towards VAE/IVAC surveillance; older VAP-centric evidence may not generalise to current systems or emergency intubation pathways.
  • Long-term outcomes after critical illness (laryngeal injury, swallowing/voice changes, quality of life, cognition) are increasingly prioritised and are essential to assess when evaluating airway devices.

Design & Methods

• Research Question:
  • Does a polyurethane-cuff ETT with continuous SSD (PU-EVAC) versus a standard PVC-cuff ETT reduce ventilator-associated infection outcomes (VAC/IVAC/possible VAP) and improve 6-month laryngeal injury, quality of life, and cognition after emergency intubation requiring ICU mechanical ventilation?
• Study Type:
  • Randomised, controlled, phase 2 trial.
  • Two US academic hospitals (Yale New Haven Hospital; Oregon Health & Science University).
  • Emergency department and in-hospital emergency intubations with subsequent ICU admission.
  • Investigator-initiated; pragmatic enrolment under exception from informed consent (EFIC).
  • Stratified randomisation by hospital and ETT size.
• Population:
  • Adults (≥18 years) requiring emergency endotracheal intubation for acute respiratory distress/failure in the emergency department or hospital, using study intubation kits, with ICU admission and mechanical ventilation.
  • ETT sizes 7.0–8.0 (strata by 7.0 vs 7.5/8.0).
  • Key exclusions: elective operating theatre intubation; intubation not using study kit; pre-existing/permanent tracheostomy; protected populations (children, pregnancy, prisoners); inability/unwillingness to participate at deferred consent stage.
  • Trial registration: NCT03705286.
• Intervention:
  • PU-EVAC ETT (polyurethane cuff with subglottic secretion drainage port).
  • Continuous subglottic suction applied at –20 mm Hg, with manual air bolus (5–10 mL) every 4 hours during intubation period.
  • Standard care co-interventions (e.g., usual VAP prevention bundle elements) were used in both groups.
• Comparison:
  • Standard PVC-cuff ETT without SSD port.
  • Standard care co-interventions (e.g., usual VAP prevention bundle elements) were used in both groups.
• Blinding:
  • Allocation concealment at enrolment via sealed, opaque envelopes and pre-prepared intubation kits.
  • Clinical teams could not be feasibly blinded after intubation because tube features differ; long-term outcome assessor blinding was not reported.
• Statistics:
  • Power calculation: 1074 participants were required to detect a 10% absolute difference in laryngeal injury incidence (assumed 40% in the PVC group) with 80% power at a two-sided 5% significance level.
  • Planned group sequential design with one interim analysis for the primary endpoint.
  • Primary analyses were intention-to-treat; complete-case analyses were used for outcomes requiring follow-up participation.
• Follow-Up Period:
  • Index admission outcomes through ICU/hospital discharge (including ventilator-associated events and length of stay metrics).
  • 6-month follow-up for laryngeal injury and patient-centred outcomes (SF-36 physical/mental component scores; neuropsychological testing battery).

Key Results

This trial was not stopped early. Enrolment reached the prespecified target (1074 enrolled; 1068 included in analyses).

• In-hospital ventilator-associated infection endpoints were similar: possible VAP 33/535 (6%) with PU-EVAC vs 29/533 (5%) with PVC (RD 0.01; 95% CI –0.02 to 0.04; P=0.56).
• Hospital length of stay was longer with PU-EVAC: 26.0 (SD 28.1) vs 22.0 (SD 22.0) days (MD 4.0; 95% CI 0.1 to 7.9; P=0.046).
• Long-term endpoints were constrained by follow-up: only 157 participants completed the 6-month assessment; laryngeal injury was numerically higher with PU-EVAC (83% vs 70%; RD 0.11; 95% CI –0.03 to 0.20; P=0.098).

Internal Validity

• Randomisation and Allocation:
  • Computer-generated randomisation with stratification by hospital and ETT size; allocation concealment via sealed, opaque envelopes and prepared intubation kits.
  • EFIC enrolment enabled randomisation at the point of emergency intubation, reducing delays and potential clinician-selection bias at enrolment.
• Drop out or exclusions:
  • Of 1074 enrolled, 1068 were included in analyses; 6 were excluded post-enrolment (protected populations n=4; refusal n=2).
  • For 6-month outcomes, attrition was substantial: 558 deaths, 167 non-consents, and 186 lost to follow-up; only 157 completed follow-up (87 PU-EVAC; 70 PVC), limiting precision and introducing potential selection bias for long-term endpoints.
• Performance/Detection Bias:
  • Blinding of clinical teams after intubation was not feasible; care processes could theoretically differ by perceived group assignment.
  • Key infection outcomes used standardised CDC/NHSN VAE algorithms (VAC/IVAC/possible VAP), reducing subjective adjudication but not eliminating misclassification.
• Protocol Adherence and Separation of the Variable of Interest:
  • PU-EVAC suction protocol: continuous –20 mm Hg with 5–10 mL manual air bolus every 4 hours (dose and timing specified).
  • Crossovers occurred: 515 patients remained in the originally assigned PU-EVAC group and 510 remained in the originally assigned PVC group; 43 crossovers were excluded in a per-protocol analysis with similar findings (per-protocol details reported in appendix).
  • Reintubation was common and managed by re-randomisation; re-randomisation introduces complexity but was accounted for via intention-to-treat analyses by initial assignment for primary reporting.
• Baseline Characteristics and Heterogeneity:
  • Groups were broadly comparable, but there was imbalance in intubation location: emergency department intubation 64% (PU-EVAC) vs 54% (PVC); outside emergency department 36% vs 46%.
  • Emergency intubation cohorts are clinically heterogeneous (indication mix, aspiration risk, comorbidity burden), which can dilute intervention effects and increase variance.
• Timing and Dose:
  • Intervention delivered at the earliest plausible timepoint (index intubation), which is methodologically appropriate for an aspiration-prevention strategy.
  • The chosen suction pressure (–20 mm Hg) represents a specific “dose”; whether this is optimal for secretion clearance versus mucosal safety is uncertain and not directly tested.
• Outcome Assessment and Statistical Rigor:
  • Binary outcomes were reported as risk differences with 95% confidence intervals; continuous outcomes as mean differences; primary analyses were intention-to-treat.
  • Long-term endpoints were effectively underpowered because only 157 participants completed 6-month assessments, despite adequate enrolment for the planned sample size.

Conclusion on Internal Validity: Internal validity is moderate: allocation concealment and pragmatic enrolment support credibility for in-hospital endpoints, but the very low completion of 6-month assessments materially limits inference for long-term outcomes and airway safety signals.

External Validity

• Population Representativeness:
  • Emergency department and in-hospital emergency intubation patients admitted to ICU and mechanically ventilated represent a common real-world pathway.
  • Exclusions (elective operating theatre intubations, protected populations, pre-existing tracheostomy, tube sizes outside 7.0–8.0) may limit applicability to elective peri-operative ventilation, paediatrics, obstetrics, and patients needing smaller/ larger tubes.
• Applicability:
  • Findings are most applicable to resource-adequate systems able to deploy continuous subglottic suction protocols and prevention bundles similar to those in US academic ICUs.
  • Generalisation to settings with different staffing models, baseline VAE/VAP rates, suction equipment, or differing EFIC/consent frameworks may be limited.

Conclusion on External Validity: External validity is moderate: the emergency intubation ICU population is clinically relevant, but the two-centre US academic setting and device/size constraints restrict broader generalisability.

Strengths & Limitations

• Strengths:
  • Large pragmatic randomised trial (n=1068 analysed) in an emergency intubation pathway where randomisation is logistically challenging.
  • Use of standardised CDC/NHSN VAE framework (VAC/IVAC/possible VAP), aligning outcomes with contemporary surveillance constructs.
  • Prespecified 6-month patient-centred outcomes (laryngeal injury, quality of life, cognition) beyond traditional VAP endpoints.
  • Allocation concealment at point of intubation with stratification by site and ETT size.
• Limitations:
  • Open-label post-intubation design with potential for performance bias, although major infection outcomes were algorithmic.
  • Very low completion of 6-month follow-up (157/1068), limiting power and increasing risk of selection bias for long-term outcomes.
  • Reintubation with re-randomisation introduces analytic complexity and potential contamination across exposure groups, despite intention-to-treat reporting.
  • Device “dose” (–20 mm Hg suction with scheduled air bolus) may not be generalisable to other SSD systems or suction settings.

Interpretation & Why It Matters

• Clinical practice signal

  In emergency intubation ICU patients, PU-EVAC did not reduce VAC/IVAC/possible VAP and was associated with longer hospital length of stay (26.0 vs 22.0 days), arguing against routine “device-first” adoption in this context without clearer benefit.
• Safety and patient-centred outcomes

  Laryngeal injury was common and numerically higher with PU-EVAC (83% vs 70%) among the small subset completing 6-month evaluation; the trial highlights the necessity—and the difficulty—of measuring long-term airway and neurocognitive outcomes in device trials.
• Methodological implications

  PreVent 2 prioritised contemporary surveillance endpoints (VAE/IVAC) and long-term functional outcomes; this approach may better reflect patient value than traditional VAP endpoints, but follow-up feasibility must be built into trial design to preserve interpretability.

Controversies & Subsequent Evidence

• Reassessment of SSD as a default prevention strategy:
  • The accompanying editorial argued that the null findings on VAE/VAP outcomes and lack of demonstrable long-term benefit should prompt reconsideration of SSD tubes as routine practice in emergency intubation pathways.¹
• Guideline uncertainty and evolving recommendations:
  • The 2022 SHEA/IDSA-aligned guidance framed SSD as an “unresolved issue” (downgraded from earlier recommendations), reflecting heterogeneity and uncertainty in older VAP-focused evidence; PreVent 2 adds a large pragmatic emergency intubation trial to this evidentiary tension.²
• Meta-analytic signals versus trial-level outcomes:
  • Updated syntheses have reported reductions in VAP incidence with SSD, but device heterogeneity, varying suction protocols, and outcome ascertainment differences complicate translation to patient-centred endpoints; PreVent 2 suggests that any VAP signal may not map cleanly to VAC/IVAC or long-term outcomes in contemporary care.³
  • Evidence specific to polyurethane cuffs has also been synthesised, but the extent to which cuff material alone affects clinically meaningful outcomes remains uncertain in pragmatic emergency cohorts.⁴
• Next steps:
  • A formal updated meta-analysis incorporating PreVent 2 (and stratifying by setting, suction “dose”, and outcome framework: VAP vs VAE/IVAC) will be essential to reconcile older VAP-focused evidence with contemporary emergency-intubation data.

Summary

• In 1068 emergency intubation ICU patients, PU-EVAC did not reduce VAC, IVAC, or possible VAP compared with standard PVC ETT.
• Hospital length of stay was longer with PU-EVAC: 26.0 vs 22.0 days (MD 4.0; 95% CI 0.1 to 7.9; P=0.046).
• Six-month mortality was similar (51% vs 53%), but effect estimates for survival were not reported.
• Long-term patient-centred outcomes (laryngeal injury, SF-36, cognition) showed no statistically significant between-group differences, but interpretation is limited by very low 6-month follow-up completion (n=157).
• The trial shifts attention from VAP incidence alone towards VAE/IVAC and long-term functional outcomes, highlighting both their importance and the operational challenge of achieving robust follow-up in emergency cohorts.

Overall Takeaway

PreVent 2 is a pragmatic, EFIC-enabled randomised evaluation of an SSD/polyurethane-cuff ETT in an emergency intubation ICU pathway, showing no reduction in VAE/VAP outcomes and a longer hospital stay with the intervention. Its most practice-shaping contribution is methodological and conceptual: it shifts the device-prevention debate from VAP incidence alone towards VAE/IVAC frameworks and patient-centred long-term harms and recovery.

Bibliography

• 1.Klompas M, Branson R. Time to pull the plug on subglottic secretion drainage? Lancet Respir Med. 2025; published online Dec 3.
• 2.Klompas M, Branson R, Cawcutt K, Crist MB, Filice GA, Griesbach D, et al. Strategies to prevent ventilator-associated pneumonia and other ventilator-associated events in acute care hospitals: 2022 update. Infect Control Hosp Epidemiol. 2022;43:687-713.
• 3.Pozuelo-Carrascosa DP, García-Hermoso A, Álvarez-Bueno C, et al. Subglottic secretion drainage for preventing ventilator-associated pneumonia: an updated meta-analysis. Eur Respir Rev. 2020;29:190107.
• 4.Saito H, Doi A, Nakamura K, et al. Efficacy of polyurethane cuffed endotracheal tube for preventing ventilator-associated pneumonia: systematic review and meta-analysis. Medicine. 2021;100:e24906.