Publication

• Title: Prehospital Plasma during Air Medical Transport in Trauma Patients at Risk for Hemorrhagic Shock
• Acronym: PAMPer
• Year: 2018
• Journal published in: The New England Journal of Medicine
• Citation: Sperry JL, Guyette FX, Brown JB, et al. Prehospital Plasma during Air Medical Transport in Trauma Patients at Risk for Hemorrhagic Shock. N Engl J Med. 2018;379(4):315-326.

Context & Rationale

• Background

  • Early deaths after major trauma are commonly driven by haemorrhage and trauma-induced coagulopathy, with a narrow window for effective haemostatic resuscitation.
  • In most civilian systems, plasma is initiated after hospital arrival; logistical delays can push “first plasma” beyond the period when haemorrhage mortality is highest.
  • Military and civilian observational data suggested potential benefit from earlier plasma, but confounding by indication and survivorship bias limited inference.
  • The PAMPer design paper pre-specified a physiologic “at-risk for haemorrhagic shock” phenotype for prehospital enrolment, aiming to enrich for patients plausibly able to benefit from earlier plasma delivery.¹
• Research Question/Hypothesis

  • In trauma patients at risk for haemorrhagic shock transported by air, does prehospital transfusion of 2 units of plasma reduce 30-day mortality compared with standard prehospital resuscitation without plasma?
• Why This Matters

  • If effective, “blood-bank-to-the-field” plasma could be a scalable systems intervention (air EMS protocols, blood bank workflows, storage/temperature management) to improve early trauma survival.
  • If ineffective (or harmful), the operational complexity, opportunity costs, and transfusion risks would argue against widespread adoption.

Design & Methods

• Research Question: Among trauma patients at risk for haemorrhagic shock undergoing air medical transport, does prehospital plasma (2 units) reduce 30-day mortality compared with standard care without plasma?
• Study Type: Multicentre, pragmatic, cluster-randomised, crossover trial (air medical base-level assignment alternating in 1-month blocks), investigator-initiated; conducted across 27 air medical bases and 10 receiving trauma centres (9 Level I, 1 Level II) in the United States (May 2014 to October 2017).
• Population:
  • Setting: Prehospital air medical transport (scene responses and interfacility transfers).
  • Core inclusion: Adult trauma (18–90 years) meeting physiologic “at-risk” criteria: systolic blood pressure ≤70 mmHg, or systolic blood pressure 71–90 mmHg with heart rate ≥108 beats/min.
  • Key exclusions: Age <18 or >90 years; known objection to blood products; pregnancy; prisoner; traumatic arrest with CPR >5 minutes; burn injury >5% total body surface area; drowning or hanging; isolated penetrating head injury; isolated ground-level fall.
• Intervention:
  • Prehospital transfusion of 2 units of plasma, initiated during air medical transport (plasma stored onboard and administered as early as feasible during flight), in addition to other standard resuscitation as clinically indicated.
• Comparison:
  • Standard prehospital resuscitation without plasma, consisting of crystalloids, red cells, or both according to local protocols and clinician judgement.
• Blinding: Unblinded (plasma delivery is operationally apparent in the prehospital environment); primary endpoint (mortality) is objective, but co-intervention and performance bias remain plausible.
• Statistics: A total sample size of 564 was planned, incorporating a cluster-design multiplier of 1.75 and 12% anticipated missingness, to provide approximately 88% power to detect a 14-percentage-point absolute mortality reduction (from 22% to 8%) at a two-sided alpha of 0.05 (adjusted to 0.037 for two planned interim analyses); primary analysis used a modified intention-to-treat approach with multivariable modelling (generalised estimating equations) accounting for clustering at air medical base.
• Follow-Up Period: 30 days for the primary endpoint; in-hospital outcomes and early physiological/transfusion outcomes assessed during the index admission (including the first 24 hours).

Key Results

This trial was not stopped early. Recruitment ran from May 2014 through October 2017; 564 patients were eligible for enrolment, 523 were enrolled, and 501 comprised the modified intention-to-treat population for the primary analysis.

• The primary outcome showed a ~10 percentage point absolute reduction in 30-day mortality with prehospital plasma, with early separation of survival curves (hazard ratio 0.64; 95% CI 0.45 to 0.91; P=0.01).
• Secondary mortality endpoints (24-hour and in-hospital) were directionally concordant, but lost statistical support after multiple-comparison adjustment.
• Among secondary endpoints, the clearest biological signal was better early coagulation profile on arrival (lower prothrombin-time ratio: 1.2 vs 1.3; P<0.001).

Internal Validity

• Randomisation and allocation: Cluster randomisation at air medical base with monthly crossover reduces stable centre-level confounding, but does not conceal allocation from crews and permits month-level behavioural drift.
• Dropout/exclusions: 564 eligible; 523 enrolled; 230 (plasma base) and 271 (standard-care base) included in modified intention-to-treat; exclusions occurred after eligibility screening and intervention assignment, creating potential for selection bias if exclusions were differential.
• Missing primary outcome: Primary outcome data were available for 220/230 (plasma) and 261/271 (standard-care); 10 outcomes were imputed in each group (total imputed n=20).
• Performance/detection bias: Unblinded prehospital teams could influence co-interventions (fluids, red cells, vasopressors); primary endpoint is objective, but secondary endpoints and process measures are susceptible to performance bias.
• Protocol adherence (dose delivery): Among those assigned to plasma bases, 205/230 (89.1%) received 2 units, 21/230 (9.1%) received 1 unit, and 4/230 (1.7%) received none.
• Baseline comparability: Major injury severity was similar (median ISS 22 in both groups), but several prehospital care variables differed (see separation), consistent with pragmatic delivery and unblinded teams.
• Timing: Median air medical transport time was similar (40 minutes plasma vs 42 minutes standard care), supporting comparable exposure windows for prehospital interventions.
• Dose: The tested dose was fixed (2 units plasma), but incomplete delivery occurred in 10.9% (1 unit) and 1.7% (0 units), which would be expected to bias towards the null if benefit is dose-dependent.
• Separation of the variable of interest: Plasma group received prehospital plasma by design; additional measurable separation included prehospital red-cell transfusion (26.1% plasma vs 42.1% standard care; P<0.001) and prehospital crystalloid volume (median 700 mL [IQR 200 to 1100] vs 900 mL [IQR 400 to 1600]; P=0.03).
• Adjunctive therapy use: Not reported (beyond standardised reporting of major transfusion and early physiology); unmeasured differences in other haemostatic adjuncts could contribute to residual confounding.
• Outcome assessment: Mortality endpoints are objective; several secondary endpoints (e.g., multiorgan failure, nosocomial infection) depend on clinical diagnosis and may vary by site.
• Statistical rigor: Primary modelling used generalised estimating equations to account for clustering; prespecified adjustment for baseline imbalances and planned interim analyses (alpha 0.037) were described.

Conclusion on Internal Validity: Overall, internal validity is moderate: the cluster crossover design strengthens causal inference for system-level implementation, but lack of blinding and evident separation in co-interventions (notably prehospital red-cell and crystalloid use) introduce plausible performance-related confounding around a biologically intertwined exposure.

External Validity

• Population representativeness: Adult trauma patients with profound hypotension (or combined hypotension/tachycardia) requiring air medical transport; enriched for patients at risk of haemorrhagic shock (median ISS 22).
• System dependence: The intervention presumes a mature blood bank–air EMS interface (thawed plasma availability, cold-chain logistics, staff training, storage monitoring) and may be less feasible in smaller or resource-limited systems.
• Transport-time dependence: Median transport times were ~40 minutes; generalisability to short urban ground transports is uncertain.
• Exclusions: Important subgroups (pregnancy, burns, drowning/hanging, isolated penetrating head injury, isolated ground-level falls) were excluded, limiting breadth of applicability across “all-comer” trauma.

Conclusion on External Validity: Generalisability is best for air medical systems caring for severely hypotensive trauma patients with non-trivial prehospital times; translation to short-transport ground EMS or markedly different transfusion infrastructures is more limited.

Strengths & Limitations

• Strengths: Pragmatic multicentre design; clinically meaningful primary endpoint (30-day mortality); cluster crossover reducing stable centre confounding; high intervention delivery (89.1% received 2 units); mechanistic signal consistent with haemostatic effect (lower prothrombin-time ratio on arrival).
• Limitations: Unblinded crews (performance and co-intervention bias); cluster design with post-screening exclusions and missing primary outcomes requiring imputation; measurable imbalance in co-interventions (prehospital red cells/crystalloids); mixed-significance secondary outcomes after multiple comparisons; applicability may be transport-time and system dependent.

Interpretation & Why It Matters

• Clinical implication

  In severely hypotensive trauma patients transported by air, initiating plasma during transport was associated with lower 30-day mortality and improved early coagulation metrics, supporting the concept that “time-to-haemostatic resuscitation” matters in select high-risk patients.
• Systems implication

  Operationalising prehospital plasma requires robust storage and transfusion governance; the benefit signal must be weighed against complexity, cost, and potential reactions, and may be best targeted to phenotypes and transport contexts most likely to benefit.
• Mechanistic coherence

  The early reduction in prothrombin-time ratio (1.2 vs 1.3; P<0.001) is directionally consistent with partial mitigation of early coagulopathy, even though many downstream clinical secondary endpoints did not differ materially after adjustment.

Controversies & Subsequent Evidence

• Discordance with a contemporaneous ground-transport RCT: The COMBAT trial reported no mortality benefit from prehospital plasma during urban ground transport, raising concern that PAMPer’s signal may be transport-time or system dependent.⁴
• Editorial interpretation emphasised “logistics and phenotype”: The accompanying editorial framed PAMPer as an operationally important proof-of-concept for getting the blood bank to the field while stressing that “who” and “when” (including transport time) likely determine benefit.²
• Correspondence highlighted unmeasured adjuncts and mortality context: Letters questioned the absence of detailed reporting/regulation of adjunct haemostatic therapies (e.g., tranexamic acid, cryoprecipitate) and the relatively high mortality in the standard-care arm; the author reply noted these therapies were not regulated/monitored within the pragmatic design and could vary by system.³
• Pooled, post hoc randomised-data signal for transport time: A post hoc analysis pooling PAMPer and COMBAT suggested the association between prehospital plasma and survival was concentrated when transport times exceeded 20 minutes, supporting transport time as an effect modifier hypothesis rather than a settled conclusion.⁶
• Subsequent RCT evidence remains mixed: The RePHILL trial evaluated prehospital packed red cells plus lyophilised plasma; its results contribute to an overall evidence base in which benefit appears context- and implementation-dependent rather than uniform.⁵
• Guideline uptake is cautious and conditional: European trauma bleeding guidelines discuss early balanced blood component resuscitation and recognise the potential role of prehospital blood products in appropriate systems, while reflecting persisting uncertainty and heterogeneity of trial results.⁷⁸

Summary

• PAMPer was a multicentre, pragmatic, cluster-randomised, monthly crossover trial of prehospital plasma delivered by air medical services to severely hypotensive trauma patients.
• The primary endpoint favoured plasma: 30-day mortality 23.0% vs 32.8% (absolute difference −9.8 percentage points; 95% CI −18.6 to −1.0; P=0.03).
• Early mortality and in-hospital mortality were directionally concordant (24-hour: 13.9% vs 22.1%; in-hospital: 22.2% vs 32.5%), but secondary endpoints were generally not robust after multiple-comparison adjustment.
• Arrival coagulation profile improved: prothrombin-time ratio 1.2 vs 1.3 (P<0.001).
• Interpretation requires attention to pragmatism and co-interventions: prehospital red-cell and crystalloid use differed between groups, and external validity is strongest for air systems with longer prehospital times.

Overall Takeaway

PAMPer provided high-impact randomised evidence that, in a US air medical setting with severely hypotensive trauma patients, initiating plasma during transport was associated with improved 30-day survival and better early coagulation indices. Its interpretation is inseparable from pragmatic prehospital delivery, co-intervention differences, and the broader mixed trial landscape, which together suggest that prehospital plasma is most defensible when targeted to high-risk phenotypes and longer transport contexts.

Bibliography

• 1.Brown JB, Guyette FX, Neal MD, et al. Taking the blood bank to the field: the design and rationale of the Prehospital Air Medical Plasma (PAMPer) trial. Prehosp Emerg Care. 2015;19:343-350.
• 2.Cannon JW. Prehospital Plasma — The Blood Bank Goes to the Battlefield. N Engl J Med. 2018;379(4):387-388.
• 3.Yeung J, et al. Prehospital Plasma in Trauma Patients at Risk for Hemorrhagic Shock. N Engl J Med. 2018;379(18):1783.
• 4.Moore EE, Moore HB, Kornblith LZ, et al. Plasma-first resuscitation to treat haemorrhagic shock during emergency ground transportation in an urban area: a randomised trial. Lancet. 2018;392:283-291.
• 5.Crombie N, Doughty HA, Bishop JR, et al. Resuscitation with pre-hospital blood products in patients with trauma-related haemorrhagic shock: the RePHILL randomised controlled trial. Lancet Haematol. 2022;9:e250-e261.
• 6.Pusateri AE, Moore EE, Moore HB, et al. Association of prehospital plasma transfusion with survival in trauma patients with haemorrhagic shock when transport times are longer than 20 minutes: a post hoc analysis of the PAMPer and COMBAT clinical trials. JAMA Surg. 2020;155(2):e195085.
• 7.Rossaint R, Afshari A, Bouillon B, et al. The European guideline on management of major bleeding and coagulopathy following trauma: sixth edition. Crit Care. 2023;27:80.
• 8.Spahn DR, Bouillon B, Cerny V, et al. The European guideline on management of major bleeding and coagulopathy following trauma: fifth edition. Crit Care. 2019;23:98.