ABSTRACT
Background
Major haemorrhage remains one of the most preventable causes of early death in prehospital care. While prehospital administration of plasma has shown potential to improve survival, logistical challenges limit its use outside hospitals. Dried plasma offers a practical solution, particularly in rural or remote environments. This statistical analysis protocol (SAP) describes the predefined methods for evaluating the effect of dried plasma versus standard care in bleeding patients treated in the prehospital setting.
Methods
This prospective, randomised parallel trial is registered at ClinicalTrials.gov (NCT07012863) and includes adult patients (≥ 18 years) with suspected bleeding requiring fluid resuscitation. Ambulance vehicles are randomised to either carry dried plasma or standard care (crystalloids), stratified by urban and rural regions. Each ambulance will be reassigned to either standard care or dried plasma after each mission. The primary outcome is 24‐h survival. Secondary outcomes include markers of coagulopathy, transfusion volume, and impact of transport time. All analyses follow the intention‐to‐treat principle and utilize Cox regression, linear or quantile regression, with Bonferroni–Holm adjustment for multiplicity. The trial will enrol a total of 650 patients.
Conclusions
This SAP ensures transparency and prevents analytical bias in a trial that addresses an urgent need for field‐appropriate blood products. The results may inform practice and policy for both civilian and military prehospital care.
Trial Registration
Keywords: coagulopathy, dried plasma, haemorrhage, prehospital resuscitation, randomised controlled study
1. Introduction
Major haemorrhage remains one of the leading causes of preventable early death after injury, and the window for effective intervention often closes in the prehospital phase. Contemporary European guidelines therefore advocate damage‐control resuscitation: permissive hypotension, minimal crystalloid use and early infusion of blood components to avoid dilutional coagulopathy, acidosis and hypothermia [1]. Several reports support that early resuscitation with blood products may save life [2, 3]. Today, Helicopter Emergency Medical Services and physician‐manned rapid response vehicles often carry whole blood or blood components, but ground ambulance services, which account for most assignments, normally have only crystalloid fluids for resuscitation due to logistical challenges in providing fresh blood products. Freeze‐ or spray‐dried plasma containing coagulation factors can be stored for up to 2 years and the possibility of being stored at ambient temperature may be an alternative to crystalloids. The results regarding the beneficial effects of treatment with plasma are ambiguous. In a previous randomised study using fresh plasma versus crystalloids, a 9.8% reduced mortality was shown with resuscitation with plasma. A lower international normalized ratio (INR) and lactate level were also observed among patients treated with plasma [4, 5]. In some studies, no effect on mortality has been shown, e.g., in Rephill, a RCT [6, 7], but in other studies, it has been suggested that plasma may be beneficial in situations with long transport times [8].
Acute traumatic coagulopathy can be observed in at least 25% of severely injured patients who are admitted to a trauma centre [9, 10]. Dilution is often considered likely to worsen coagulopathy, although the exact mechanism and level of aggravation are unknown [11].
Most studies of prehospital bleeding refer to trauma, but other causes of bleeding can also be severe and even fatal, especially in rural areas with long distances, for example, obstetric bleeding, gastrointestinal bleeding, and vascular catastrophes. In a previous observational study of standard care in the ambulance service, we showed that a high proportion of bleeding patients treated with crystalloid fluid resuscitation in the regular ambulance service had a coagulopathy and a high shock index upon arrival at the hospital. We also showed that a high shock index predicted the need for blood transfusions, and prehospital bleeding of more than 500 mL predicted mortality, both adjusted for age [12].
Therefore, the aim of this trial is to report the outcomes of patients receiving dried plasma as a treatment for major prehospital bleeding compared with patients receiving standard care in the ambulance service with crystalloid fluids and to report clinical data for both groups.
2. Methods
2.1. Trial Design and Setting
This trial is an investigator‐initiated, open, prospective randomised parallel trial.
Patents are recruited from ground‐based ambulance services in the following regions in Sweden:
Västra Götalandsregionen, Region Halland, Region Värmland, Region Jönköping, Region Västerbotten, Region Norrbotten, Region Blekinge and Region Skåne.
2.2. Trial Registration
This trial was registered in the clinicaltrials.gov registry (NCT07012863) on 30 June 2025.
2.3. Trial Conduct
The protocol of this trial has been prepared in advance and in accordance with the Standard Protocol Items: Recommendations for Interventional Trials (SPIRIT) guidelines [13]. The SPIRIT protocol is found as a supplement. This trial will be conducted according to the Declaration of Helsinki and its later amendments and according to the Good Clinical Practice guidelines.
2.4. Randomisation
Randomisation will occur at the level of individual ambulances. Each ambulance will be randomly assigned in a 1:1 ratio to one of two groups: intervention group (given dried plasma) and control group (given standard care). A computer‐generated randomisation list will be created prior to study initiation and used to assign all eligible ambulances across the participating ambulance stations. Randomisation will be stratified by geographic region and operational volume to ensure balance between treatment groups across different service environments. After completing a mission, the ambulance will be reassigned again to either standard care or dried plasma.
2.5. Blinding
Blinding of prehospital ambulance personnel is not possible due to the inherent characteristics of the intervention. However, allocation concealment is maintained for the statistician performing the primary data analysis. To minimise interpretive bias, a preliminary abstract summarising the study findings will be drafted prior to unblinding, and this document will be included as an appendix in the final publication.
2.6. Inclusion and Exclusion Criteria
All patients with suspected bleeding will be eligible for inclusion in the trial. The patient will be allocated to either intervention or standard care.
2.6.1. Inclusion Criteria
Patients with clinical signs of bleeding triggering intervention with fluid resuscitation by the ambulance service at the site.
Patients > 18 years of age.
2.6.2. Exclusion Criteria
Patients who lack signs of severe clinical bleeding. Patients who have received ≥ 1 unit blood component before randomised ambulance arrival will not be eligible to take part in the study.
Patients < 18 years of age.
2.7. Trial Interventions
The timeline describing the eligibility, randomisation, and the study intervention is presented in Figure 1. The observation time is 24 h from the administration of either dried plasma or standard care.
FIGURE 1.
Timeline that describes the different times and events that are recorded. The counting of survival time starts from the emergency site to 24 h after arrival at the hospital (time 2 + time 3) In addition to this, 30‐day mortality is also recorded.
2.7.1. Intervention Group
Patients with bleeding in a prehospital setting with suspected or verified bleeding where fluid resuscitation is indicated will be given dried plasma. The patient will be given two units of dried plasma (200 mL×2).
2.7.2. Control Group
Patients with bleeding in a prehospital setting where fluid resuscitation with crystalloids is indicated will be treated with standard care according to the guidelines for the actual ambulance service.
2.8. Outcome Measures
2.8.1. Primary Outcome Measures
The primary goal is to compare the survival rate at 24 h following prehospital administration of dry plasma with that following standard treatment with clear fluids in bleeding patients at the emergency site.
2.8.2. Secondary Outcome Measures
To compare the occurrence of coagulopathy defined as PK/INR > 1.2 and/or a platelet count < 150 × 109/L and/or APTT > 34 s or each value alone. INR = international normalized ratio; PK = prothrombin complex. APTT = activated partial thromboplastin time between the two groups.
To compare the difference in the amount of blood transfusion within 24 h after arrival at the hospital between the two groups.
The survival rate at 24 h, explained by the interaction with time from emergency call to ED arrival (minutes) and treatment, is compared.
These three secondary outcomes (the occurrence of coagulopathy, the need for blood transfusion and the impact of time), together with the primary outcome, will be adjusted for multiplicity by the Bonferroni–Holm method [14, 15].
2.9. Exploratory Outcomes
The following data from each group will be compared: Patients with estimated bleeding > 500 mL.
Survival and all‐cause mortality within 30 days following prehospital administration of dried plasma or crystalloid fluids.
Coagulation parameters on arrival at the emergency department (INR, platelet count, and viscoelastic test parameters).
The total volumes of crystalloids, prehospital, and blood components transfused within the first 24 h.
Transfusion‐related adverse events and acid–base status (lactate, pH) on arrival.
Subgroup analyses for (a) transport time > 20 min, (b) nontraumatic haemorrhage, and (c) baseline evidence of coagulopathy.
The level of lactate upon arrival at the hospital. The number of intensive care days. The number of patients in each group given tranexamic acid. The survival rate at 30 days.
2.10. Data Collection
Data are obtained from ambulance records, hospital medical records, hospital laboratories, and official death registries. Subjects are included by the local ambulance personnel in each ambulance district. The ambulance personnel will use an electronic platform connected to a Redcap database.
2.10.1. Data Management
All collected data are handled confidentially and in accordance with national laws and regulations. All patient data are anonymized, using an individual study code. The code lists are kept locked up and separated from the registered study data. Original records will be retained at the trial sites for the time specified by the national regulations. A common high‐security encrypted electronic database (eCRF) with all recorded data is kept in Gothenburg, Sweden. The overall timeline for the study is described in Figure 2.
FIGURE 2.
This figure describes the timeline for the trial. Patient inclusion–Data analysis–Manuscript and reporting.
2.11. Statistical Analysis
2.11.1. General Analytical Principles
We will conduct all analyses according to the modified intention‐to‐treat (mITT) principle; that is, all randomised participants, given a first dose per ambulance and site, will be included in all analyses. We do not include covariates that are multicollinear. For example, there is a risk that time and urban/rural areas are strongly correlated, and the NACA score and SI score are strongly correlated. In that case, we use the model with the lowest AIC. If the assumptions of the models are not fulfilled, we adjust them accordingly.
We will set statistical significance to 0.05 and report two‐sided p‐values and 95% confidence intervals. Continuous variables will be checked for normality and linearity. If the normality assumption is not fulfilled, we will use quantile regression. We will check for the proportional assumption when modelling binary variables, with Cox regression. If the assumption is not fulfilled, we will adjust the model (e.g., by a time‐dependent variable).
2.11.2. Primary Outcome
The primary endpoint is the time to death from treatment (intervention or standard care) at the emergency site and up to 24 h at the hospital. This endpoint will be analysed via Cox regression adjusted for treatment, age, mechanism of injury, NACA score, urban/rural area and the SI score. As a sensitivity analysis, the crude rate (a Cox regression adjusted by treatment and urban/rural area) is analysed.
2.11.3. Secondary Outcomes
The occurrence of coagulopathy is to be analysed via Cox regressions adjusted for treatment, age, mechanism of injury, NACA score, urban/rural area, and SI score. As a sensitivity analysis, the crude rate (a Cox regression adjusted by treatment and urban/rural area) is analysed.
The difference in the amount of blood transfusion within 24 h after arrival at the hospital between the two groups was analysed via regression analysis (either linear or quantile regression), depending on whether the normal assumption was fulfilled. The regressions are adjusted for treatment, age, mechanism of injury, NACA score, urban/rural area, and SI score.
The influence of the time between the emergency call and ED arrival on the survival rate after 24 h at the hospital was measured via Cox regression, adjusting for urban/rural area, treatment, the time between the emergency call and hospital arrival, and the interaction between treatment and the time between the emergency call and hospital arrival. 95% confidence intervals and p values will be provided.
The three tests, together with the primary test, will be analysed with a Bonferroni–Holm adjustment.
2.11.4. Exploratory Outcomes
Binary endpoints will be analysed via Cox regressions adjusted for treatment, age, mechanism of injury, NACA score, urban/rural area and the SI score. As a sensitivity analysis, the crude rate (a Cox regression adjusted by treatment and urban/rural area) is analysed. A 95% confidence interval per outcome will be provided.
Continuous endpoints are analysed via regression analysis (either linear or quantile regression), depending on whether the normal assumption is fulfilled. The regressions are adjusted for treatment, age, mechanism of injury, NACA score, urban/rural area, and SI score. As a sensitivity analysis, the crude rate (a Cox regression adjusted by treatment only) is analysed. A 95% confidence interval per outcome will be provided.
2.11.5. Clinical Data
All clinical and background data collected will be presented in tables, per treatment group (standard care or dried plasma), as well as total data.
Qualitative data are presented as numbers (%), and quantitative data are presented as the means (SDs) and medians (IQRs).
2.11.6. Missing Data
If a patient dies during transport to the hospital, a quarter of the time between arrival at the hospital (time 3) and the emergency call (time 1) is estimated as the survival time.
If the time at the emergency site is unknown, time 1 and time 2 are estimated as half the time between arrival at the hospital and the emergency call, respectively. We do not expect to see other missing data.
2.12. Sample Size
A sample size calculation was performed to compare two independent groups with respect to a binary outcome. The expected event rate in the group with standard treatment is 0.33, whereas the intervention group (dried plasma) is anticipated to have an event rate of 0.23. To detect this difference with a statistical power of 80% and a two‐sided significance level of 0.05, a total of 316 participants per group was needed. A robust proportional test was used, and the sample size calculation was performed in R (version 4.4.3.).
2.13. Trial Profile
The flow of the patients participating in this trial will be reported according to the CONSORT statement [16].
2.14. Data Monitoring and Safety Committee
The trial will be monitored in accordance with Good Clinical Practice (GCP) guidelines. An independent monitor, not involved in data analysis or patient care, will conduct regular audits of trial procedures, data entry, and source documentation to ensure protocol compliance and data integrity.
All adverse reactions (ARs) and serious adverse events (SAEs) will be recorded and reviewed. The need for a Data Monitoring Committee (DMC) was assessed and deemed unnecessary due to the limited risk profile and short duration of the intervention. Any protocol deviations will be documented and reported to the principal investigator and, if required, to the regional ethics board.
Data will be stored in a secure, password‐protected electronic database with restricted access, and backups will be performed regularly.
2.15. Interim Analyses
No interim analyses will be conducted.
3. Discussion
Prehospital bleeding is a common cause of death in many patients. It is still not clear if the availability of plasma would be beneficial for the bleeding prehospital patient. This trial will present clinical data regarding these patients as well as outcome data.
The prospective design of the trial is a strength, as well as the fact that the trial will include not only trauma patients (divided into blunt and penetrating) but also other causes of bleeding, such as gastrointestinal bleeding, obstetric/gynaecological bleeding, aortic aneurysm, and other causes of bleeding. The recruitment of patients from both metropolitan regions and rural areas is also a strength, as it can demonstrate whether there are differences between them related to transport times.
Prehospital patients with severe illness cannot always communicate adequately, which is why there is a risk that we initially miss morbidity other than the acute condition. Whether the patients are on anticoagulant medication will also be unknown in part and will therefore affect the analyses. However, this reflects the prehospital reality, as many factors are unknown at the first contact with the patient.
There are several difficulties in randomizing a patient pre‐hospital while the patient is simultaneously being treated. This has been noted in several similar studies, such as RePHILL (UK), PAMPer (US), and COMBAT (US) [5, 6, 17]. By using the method of assigning each ambulance to either standard care or dried plasma instead of randomizing patients at the point of injury, we believe that we can avoid many of the problems, such as practical and logistical problems and ethical problems.
The design includes relatively many secondary and exploratory outcomes. Statistical adjustment for multiple testing is used to reduce the risk of false positive findings.
Despite its limitations, this study provides valuable data on patients with prehospital bleeding.
Author Contributions
G. S. drafted the first manuscript. G. S. and C. W. revised the statistical details of the statistical analysis plan. G. S., G. S. K. and A. W. were the principal investigators and contributed to the study design. All the authors contributed to the writing of this manuscript and approved the final version.
Disclosure
The results of this trial will be submitted to a peer‐reviewed medical journal regardless of the results.
Ethics Statement
This study will be conducted in compliance with ICH Good Clinical Practice, national laws, and applicable regulatory requirements and in accordance with the ethical principles of the Declaration of Helsinki. The study was approved by the Swedish Ethical Review Authority (2023‐06839‐01). Patients included in the treatment group are likely to receive better treatment than standard care. Patients included in the control group will receive standard care according to the current treatment protocol.
Conflicts of Interest
The authors declare no conflicts of interest.
Acknowledgements
The authors would like to thank the clinical and research personnel at each of the included ambulance services, the hospitals involved in each region, and the participating patients.
Skallsjö G., Wikman A., Wessman C., and Sandström G., “Prehospital Treatment With Dried Plasma in Patients With Major Bleeding—A Prospective Randomised Controlled Multicentre Trial: Statistical Analysis Protocol,” Acta Anaesthesiologica Scandinavica 69, no. 9 (2025): e70120, 10.1111/aas.70120.
Funding: This work was supported by the Centre for Disaster Medicine at the University of Gothenburg, the Carnegie Foundation in Stockholm, the Swedish Military Medical Association, and local ALF (Agreement for Medical Education and Research) funding from Södra Älvsborg Hospital, Västra Götaland Region. The funders had no role in the design of the study, data collection, analysis, interpretation of data, or in writing the manuscript.
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.
References
- 1. Rossaint R., Afshari A., Bouillon B., et al., “The European Guideline on Management of Major Bleeding and Coagulopathy Following Trauma: Sixth Edition,” Critical Care 27, no. 1 (2023): 80. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Dishong D., Sperry J. L., Spinella P. C., Triulzi D. J., and Yazer M. H., “Administration of Blood Products in the Prehospital Setting Can Decrease Trauma Patient Mortality,” Transfusion 62, no. 4 (2022): 725–727. [DOI] [PubMed] [Google Scholar]
- 3. Holcomb J. B., Donathan D. P., Cotton B. A., et al., “Prehospital Transfusion of Plasma and Red Blood Cells in Trauma Patients,” Prehospital Emergency Care 19, no. 1 (2015): 1–9. [DOI] [PubMed] [Google Scholar]
- 4. Canton S. P., Lutfi W., Daley B. J., et al., “Lactate as a Mediator of Prehospital Plasma Mortality Reduction in Hemorrhagic Shock,” Journal of Trauma and Acute Care Surgery 91, no. 1 (2021): 186–191. [DOI] [PubMed] [Google Scholar]
- 5. Sperry J. L., Guyette F. X., Brown J. B., et al., “Prehospital Plasma During Air Medical Transport in Trauma Patients at Risk for Hemorrhagic Shock,” New England Journal of Medicine 379, no. 4 (2018): 315–326. [DOI] [PubMed] [Google Scholar]
- 6. Crombie N., Doughty H. A., Bishop J. R. B., et al., “Resuscitation With Blood Products in Patients With Trauma‐Related Haemorrhagic Shock Receiving Prehospital Care (RePHILL): A Multicentre, Open‐Label, Randomised, Controlled, Phase 3 Trial,” Lancet Haematology 9, no. 4 (2022): e250–e261. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Jost D., Lemoine S., Lemoine F., et al., “French Lyophilized Plasma Versus Normal Saline for Post‐Traumatic Coagulopathy Prevention and Correction: PREHO‐PLYO Protocol for a Multicenter Randomized Controlled Clinical Trial,” Trials 21, no. 1 (2020): 106. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Pusateri A. E., Moore E. E., Moore H. B., et al., “Association of Prehospital Plasma Transfusion With Survival in Trauma Patients With Hemorrhagic Shock When Transport Times Are Longer Than 20 Minutes: A Post Hoc Analysis of the PAMPer and COMBAT Clinical Trials,” JAMA Surgery 155, no. 2 (2020): e195085. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. MacLeod J. B., Lynn M., McKenney M. G., Cohn S. M., and Murtha M., “Early Coagulopathy Predicts Mortality in Trauma,” Journal of Trauma 55, no. 1 (2003): 39–44. [DOI] [PubMed] [Google Scholar]
- 10. Niles S. E., McLaughlin D. F., Perkins J. G., et al., “Increased Mortality Associated With the Early Coagulopathy of Trauma in Combat Casualties,” Journal of Trauma 64, no. 6 (2008): 1459–1463. [DOI] [PubMed] [Google Scholar]
- 11. Schols S. E., Heemskerk J. W., and van Pampus E. C., “Correction of Coagulation in Dilutional Coagulopathy: Use of Kinetic and Capacitive Coagulation Assays to Improve Hemostasis,” Transfusion Medicine Reviews 24, no. 1 (2010): 44–52. [DOI] [PubMed] [Google Scholar]
- 12. Skallsjo G., Sandstrom G., Sato Folatre J. G., et al., “Bleeding Resuscitation in the Ambulance Service, an Observational Study of Standard Care in Sweden,” Scandinavian Journal of Trauma, Resuscitation and Emergency Medicine 33, no. 1 (2025): 130. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Chan A. W., Tetzlaff J. M., Altman D. G., et al., “SPIRIT 2013 Statement: Defining Standard Protocol Items for Clinical Trials,” Annals of Internal Medicine 158, no. 3 (2013): 200–207. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Dunn O. J., “Multiple Comparisons Among Means,” Journal of the American Statistical Association 56, no. 293 (1961): 52–64. [Google Scholar]
- 15. Holm S., “A Simple Sequentially Rejective Multiple Test Procedure,” Scandinavian Journal of Statistics 6, no. 2 (1979): 65–70. [Google Scholar]
- 16. Moher D., Hopewell S., Schulz K. F., et al., “CONSORT 2010 Explanation and Elaboration: Updated Guidelines for Reporting Parallel Group Randomised Trials,” Journal of Clinical Epidemiology 63, no. 8 (2010): e1–e37. [DOI] [PubMed] [Google Scholar]
- 17. Moore H. B., Moore E. E., Chapman M. P., et al., “Plasma‐First Resuscitation to Treat Haemorrhagic Shock During Emergency Ground Transportation in an Urban Area: A Randomised Trial,” Lancet 392, no. 10144 (2018): 283–291. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.