ASpirin as a Treatment for Acute Respiratory Distress Syndrome – a multi-centre, randomised, double-blind, placebo-controlled trial (STAR): study protocol

ASpirin as a Treatment for Acute Respiratory Distress Syndrome – a multi-centre, randomised, double-blind, placebo-controlled trial (STAR): study protocol

Philip Toner1, Cecilia O’Kane1, James McNamee2, Rejina Verghis1,3, and Daniel F McAuley1,2

1Wellcome-Wolfson Institute for Experimental Medicine, Queen’s University Belfast, 97 Lisburn Road, Belfast, BT9 7BL
2Regional Intensive Care Unit, Royal Victoria Hospital, Belfast Health and Social Care Trust, Grosvenor Road, Belfast, BT12 6BA
3Northern Ireland Clinical Trials Unit, 1st floor Elliott Dynes Building, Royal Hospitals, Grosvenor Road, Belfast, BT12 6BA

Email: ptoner09@qub.ac.uk

Download PDF

AbstractFull-TextReference ListSupplementary MaterialCitationQR code

Background: Acute respiratory distress syndrome (ARDS) remains a common cause of significant morbidity and mortality in the critically ill, for which there is currently no pharmacological treatment. There is in vivo, in vitro, observational and phase I evidence suggesting aspirin may be of benefit in this condition. The aim of the STAR trial (aSpirin as a Treatment for Acute Respiratory Distress Syndrome) is to test the hypothesis that aspirin 75 mg is both safe and effective in improving important surrogate outcomes in patients with ARDS.

Methods/Design: STAR is a randomised, double-blind, allocation-concealed, placebo-controlled, multi-centred phase II trial. Patients diagnosed with ARDS, as per the Berlin Definition, will be randomised in a 1:1 ratio to receive enteral aspirin 75 mg or placebo for a maximum of 14 days. Randomisation is stratified by vasopressor requirement. The primary endpoint is to evaluate the efficacy of aspirin to improve oxygenation index at day 7. A total of 60 patients will be recruited from intensive care units (ICUs) across Northern Ireland. Plasma, bronchoalveolar lavage (BAL) and urine samples will be obtained to further investigate mechanisms by which aspirin might improve clinical outcomes in these patients.

Keywords: Aspirin; acute respiratory distress syndrome, oxygenation index

Trial Registration: NCT02326350

Funding:  The study is funded by the Health & Social Care Research & Development Division of the Public Health Agency, Northern Ireland.

Background information

Acute respiratory distress syndrome (ARDS) is defined by the Berlin Definition as an acute onset of hypoxia (PaO2/FiO2 < 40 kPa), in the presence of a positive end-expiratory pressure (PEEP) > 5 cmH2O and with bilateral radiological opacities not solely explained by cardiac dysfunction1. Despite advances in treatment2,3,4, mortality remains relatively high at 35-40%5,6 with significant post morbid functional impairment and reduction in quality of life7,8. The high mortality and long-term consequences, coupled with the high financial burden, make the treatment of ARDS a healthcare priority9. Multiple drugs have been investigated as potential treatments, most recently simvastatin10, but as yet there is no effective pharmacological treatment11

The rationale for aspirin as a treatment for ARDS

The use of aspirin as a treatment for ARDS is a novel approach. There are various pathways in which aspirin could attenuate the pathophysiology in ARDS. Firstly, through the inhibition of cyclooxygenase enzymes (COX), aspirin can prevent platelet activation12. Platelets play a significant role in the development of sepsis and ARDS13, especially as the lungs are a major site of platelet maturation and subsequent reservoirs14. Once activated, platelets degranulate and release a pro-inflammatory cocktail promoting further platelet degranulation and aggregation, leucocyte recruitment and oedema formation15. Furthermore, platelet and neutrophil aggregation is essential for the formation of neutrophil extracellular traps (NETs), which, in excess, can directly damage the lung architecture, thus escalating the inflammatory and thrombotic processes16,17. In an ARDS murine model, platelet depletion resulted in reduced neutrophil migration, less oedema formation and improved outcomes15.

Secondly, aspirin can downregulate the production of proinflammatory cytokines through the inhibition of NFκB18, subsequently reducing leucocyte recruitment19. Thirdly, aspirin promotes nitric oxide (NO) production, resulting in decreased leucocyte migration, oedema formation and microthrombi, all of which are important features in ARDS20. Finally, aspirin promotes resolution via the production of lipoxins (aspirin triggered lipoxin [ATL])21, a feature absent in ARDS.

Both aspirin and ATL administration reduce pulmonary inflammation and improve outcomes in murine models of ARDS22,23. Observational studies have shown a benefit to prior aspirin use in those admitted to ICU with ARDS or sepsis13,24. In a local single centre retrospective study, aspirin was associated with reduced mortality in ARDS, with an odds ratio 0.42 (95% confidence interval, 0.18 – 0.96)25. Aspirin also significantly reduced bronchoalveolar lavage (BAL) neutrophil count, not only in an ex vivo lung perfusion (EVLP) lipopolysaccharide (LPS)-induced inflammation model of ARDS, but also in a healthy volunteer model of ARDS-induced by inhaled LPS26. These animal models, observational studies, and human models support the hypothesis that aspirin may have a role in the treatment of ARDS.

Aspirin has recently been investigated as a preventive agent for ARDS in high-risk patients presenting to the emergency department. However, it did not significantly reduce the incidence of ARDS at day 727. On closer review, the study was underpowered, thus limiting the impact of this result.

The aim of the STAR trial is to test the hypothesis that aspirin 75 mg, when administered enterally, is both safe and effective in improving important surrogate outcomes in patients diagnosed with ARDS.

Methods/design

STAR is a randomised, double-blind, allocation-concealed, multi-centre, placebo-controlled phase II trial to determine if aspirin is safe and whether it improves important surrogate clinical outcomes in adult patients with ARDS. The trial is sponsored by the Belfast Health and Social Care Trust (BHSCT). The protocol was reviewed and approved by the Office for Research Ethics Committees Northern Ireland (ORECNI) (14/NI/1093) and by the Medicines and Health Care Products Regulatory Authority (MHRA). STAR is registered with ISRCTN (NCT02326350) and with the European Union Drug Regulation Authorities Clinical Trials Database (2014-002564-32). The trial will be conducted in accordance with the ethical principles that have their origin in the Declaration of Helsinki and will be carried out in accordance with the principles of the International Conference on Harmonisation Good Clinical Practice (ICH-GCP) guidelines (https://ichgcp.net/). STAR is coordinated through the Northern Ireland Clinical Trials Unit (NICTU) and has been adopted on to the Clinical Research Portfolio supported by the Northern Ireland Clinical Research Network (NICRN) for Critical Care.

In PICO terms:
Population: adult patients with ARDS
Intervention: aspirin 75 mg
Comparator: placebo
Outcome: safety and physiological indices of efficacy

Figure 1. Trial Schematic

Outcome Measures

As this is a phase II clinical study, several surrogate outcomes will be evaluated.

The primary endpoint is oxygenation index (OI) at day 7. OI is a physiological index of the severity of ARDS, which measures both impaired oxygenation and the amount of mechanical ventilation delivered. and independently predicts mortality in patients with ARDS28. Day 7 was chosen as this time interval will minimise the competing effects of death and extubation, while allowing a sufficient time interval for a biological effect to occur. OI is calculated as (mean airway pressure (cmH2O) x FiO2 x 100) ÷ PaO2 (kPa).

The secondary outcomes are:

  1. OI at days 4 and 14
  2. Physiological indices of ARDS, as measured by respiratory compliance (Crs) and P/F ratio on days 4, 7 and 14
  3. Change in sequential organ failure assessment (SOFA) score from baseline to day 4, 7 and 14
  4. Safety and tolerability, as assessed by the occurrence of suspected unexpected serious adverse reactions (SUSARs)

Duration of ventilation, ventilation-free-days at day 28, mortality at both day 28 and 90, as well as length of ICU stay will also be recorded. These important clinical outcomes are not included as outcome measures as the study is not adequately powered to assess them.

Safety

The frequency with which the following events occur will be reported to the Data, Monitoring and Ethical Committee (DMEC):

  1. fall in haemoglobin below 70 g/l
  2. incidence of acute kidney injury
  3. all adverse events (AEs), including serious adverse events (SAEs) and occurrences of suspected unexpected serious adverse events (SUSARs)

Safety monitoring blood tests including full blood count (FBC), urea and electrolytes (U&E), coagulation profile, liver function tests (LFTs) and clinical assessment is undertaken on a daily basis for 14 days. All AEs are reported up to 28 days from completion of study drug.

Biological mechanisms

An exploratory study will provide insight into the mechanism by which aspirin may be an effective treatment in ARDS. Blood, BAL and urine will be taken on days 0 and 4, with further blood and urine samples taken on days 7 and 14 while patients continue to receive the study drug. Samples will be analysed for biological markers of pulmonary and systemic inflammation, as well as pulmonary and systemic epithelial and endothelial function. Furthermore, we will measure several lipid inflammatory mediators in addition to assessing the pharmacokinetics and pharmacodynamics of aspirin in the critically ill.

It may not be possible to collect all samples from all patients at all time points. If samples are not collected, this will not be recorded as a protocol violation.

Eligibility criteria

Inclusion

Patients are screened on a daily basis to determine if they fulfil the following inclusion criteria:

  1. receiving invasive mechanical ventilation
  2. ARDS as defined by the Berlin Definition1
    • onset within 1 week of an identified insult
    • within the same 24 hour time period
      • hypoxic respiratory failure (PaO2/FiO2 ratio < 40 kPa on PEEP > 5 cmH20)
      • bilateral infiltrates on chest radiograph consistent with pulmonary oedema not explained by another pulmonary pathology
      • no evidence of heart failure or volume overload

Exclusion

Patients fulfilling any of the criteria below will be excluded from the trial:

  1. greater than 72 hours from the onset of ARDS
  2. age < 16 years
  3. patient is known to be pregnant
  4. participation in a clinical trial of an investigational medicinal product within 30 days
  5. current treatment with aspirin or within the past 4 weeks
  6. platelet count < 50 x 109/l
  7. haemophilia or other haemorrhagic disorder or concurrent therapeutic anticoagulant therapy
  8. history of aspirin-sensitive asthma or nasal polyps associated with asthma
  9. active or history of recurrent peptic ulcer and/or gastric or intestinal haemorrhage, or other kinds of bleeding, such as cerebrovascular haemorrhage
  10. traumatic brain injury
  11. active gout
  12. currently receiving methotrexate
  13. severe chronic liver disease with Child-Pugh score > 12
  14. known hypersensitivity or previous adverse reaction to salicylic acid compounds or prostaglandin synthetase inhibitors
  15. physician decision that aspirin is required for proven indication
  16. contraindication to enteral drug administration, e.g. patients with mechanical bowel obstruction
  17. treatment withdrawal imminent within 24 hours
  18. consent declined.

Power and sample size

The primary outcome measure will be the difference in OI between the aspirin and placebo treated groups at day 7. Based on data from a recently completed clinical trial in ARDS, the mean (standard deviation; SD) OI at day 7 in patients with ARDS is 62 (51) cmH2O/kPa29. A sample size of 56 subjects (28 in each group) will have 80% power, at a two-tailed significance level of 0.05, to detect a clinically significant difference of 39 cmH2O/kPa in OI between groups. In a previous phase II study of similar size, we found that an intervention can demonstrate a change in OI of a similar magnitude confirming a treatment effect of this size can be achieved29. Although we anticipate few withdrawals or loss to follow-up, we have allowed for this in the sample size calculation. In our previous single centre study of simvastatin in ARDS, there were no withdrawals. In a multi-centre UK study of pulmonary artery catheters in ICU patients (PACMan), no patients were lost to follow up, and only 3% withdrew consent after recovering competency30. Therefore, a drop-out rate of 5% has been estimated and the study will require a total of 60 patients (30 in each group).

Trial conduct

Consent

Informed consent will be obtained before conducting any trial specific procedures. Due to the incapacitating nature of the condition, the patients typically lack the capacity to give consent. In this situation, informed consent will be sought from a Personal Legal Representative (Per LR) or Professional Legal Representative (Pro LR), in keeping with requirements from the EU clinical trails directive. Once they regain capacity, patients will be informed of their participation in the trial, have the trial explained to them, and consent to continue in the trial will be sought. Where consent to continue is not obtained, consent from the legal representative will remain valid. Similar consent mechanisms have been used successfully in other critical care trials10,29,30.

Randomisation and study drug design

Victoria Pharmaceuticals will prepare the drug packs. Aspirin 75 mg and the placebo will have an identical appearance. All trial drugs will be packaged identically and identified by a unique pack number.

After informed consent, the researcher will contact the clinical trials pharmacist who will allocate the next sequential number as per the randomisation schedule, maintaining blinding. Randomisation will be stratified by vasopressor use and subjects will be randomised in a 1:1 ratio using blocks of variable size. The researcher will then register the recruited patient with the CTU.

Study drug administration

Subjects will be randomised to receive aspirin 75 mg capsule or a placebo capsule enterally for up to 14 days. The first dose of the study drug will ideally be administered within 4 hours of  randomisation and subsequent doses will be as close to 10 am as possible starting on the following calendar day. The clinical staff administering the drug will not be involved in any of the study specific assessments.

 

Post randomisation withdrawals and exclusions

Patients may withdraw, or be withdrawn by their representative, from the trial at any time without prejudice. Consent will be requested to use the data collected to that point. If a patient, or their representative, requests termination of the trial drug during the treatment period, the drug will be stopped but the patient will continue to be followed-up as part of the trial, unless they also explicitly request withdrawal from follow-up.

Study drug termination criteria

The study drug will be continued until one of the following is met:

  1. 14 days after randomisation (maximum treatment period)
  2. study drug related adverse event
  3. critical care discharge
  4. death or discontinuation of active treatment
  5. request from Per LR or Pro LR to withdraw the patient from the study
  6. decision by the attending clinician on safety grounds
  7. clinical indication for treatment with aspirin e.g. myocardial infarction.

Study drug compliance

Any omission of the study drug will be recorded in the case report form (CRF) to monitor compliance.

Clinical management of patients in the trial

All patients will receive standard management with regards to nutrition, antibiotic policy, fluid management and weaning. It is recommended patients will be managed using a standardised mechanical ventilation protocol aiming for tidal volumes of 6 ml/kg predicted body weight2. Rescue therapies, such as extra-corporeal membrane oxygenation, can be used according to local policy.

Study procedures for unblinding

As STAR is a randomised, placebo-controlled, double-blind trial, the research staff, treating clinical staff, and patients, will be blinded as to which arm of the study the patient is allocated. All trial drugs have identical appearance and are only identified by a unique pack number. The investigator or treating physician may unblind a participant’s treatment assignment in the case of an emergency, when knowledge of the study treatment is essential for the appropriate clinical management or welfare of the subject. Emergency unblinding will be performed by telephone contact with the pharmacy in the BHSCT. The date and reason for unblinding will be recorded in the CRF.

Data collection

All data for an individual patient will be collected by the research team and recorded in the CRF. The majority of the data will be obtained from the patient’s hospital record. Data will be collected from the time the patient is considered for entry into the trial through to their discharge from hospital and recorded on a secure, backed up custom database. Data censorship will occur after 90 days post randomisation.

Adverse event reporting

As STAR is recruiting in a population that is already in a life-threatening situation, it is expected that many of the patients will experience AEs. Events that are suspected in this population (i.e. events in keeping with the underlying condition) will not be reported as AEs. The adverse effects as listed in the summary of product characteristics (SmPC) for aspirin will be used as the reference safety information. AEs that occur between trial entry and up to 28 days after completion of the study drug will be reported. SAEs and SUSARs will be reported within 24 hours of becoming aware of their occurrence and the sponsor will inform the regulatory authorities as per the regulatory requirements.

End of trial

The trial will end when 60 patients have been recruited and completed 90-day follow-up. The trial will be stopped prematurely if mandated by the Ethics Committee, MHRA, or the sponsor e.g. following recommendations from the DMEC.

Statistical analysis plan

A full statistical analysis plan is available in the STAR Supplementary Data.

Trial oversight

The Chief Investigator will have overall responsibility for the conduct of the study. The Trial Management Group will have responsibility for the daily running of the trial. An independent DMEC will monitor the safety of the participants through regular review of AEs, deaths and any other data as requested by the DMEC. They will report any issues pertaining to safety to the Chief Investigator. It will be the responsibility of the Chief Investigator to inform the sponsor who will take appropriate action to halt the trial if concerns exist about patient safety.

Trial status

The trial has been successfully initiated and as of August 2018 46 patients have been successfully enrolled. There has been one major amendment to the protocol design and eligibility criteria, which has been approved by the MHRA and local ethics committee. That amendment was to extend from a single site to a multi-centre trial to ensure adequate recruitment and to adjust the absolute platelet exclusion count from 100 x109/l to 50 x109/l.

Author’s contributions

DFM and CO’K conceived the study. All authors made a substantial contribution to the protocol development. All authors have read and approved the manuscript.

Acknowledgements

The study is funded by the Health & Social Care Research & Development Division of the Public Health Agency, Northern Ireland. The authors thank the Data Monitoring and Ethics Committee:

  • Dr P Glover (Chair – Consultant, Belfast Health and Social Care Trust)
  • P McKeown (Consultant/Professor, Belfast Health and Social Care Trust/Queen’s University Belfast), and
  • M Clarke (Professor, Queen’s University Belfast).

References

  1. Ranieri VM, Rubenfeld GD, Thompson BT, Ferguson ND, Caldwell E, Fan E, et al. Acute respiratory distress syndrome: the Berlin Definition. JAMA. 2012 Jun;307(23):2526–2533.  https://doi.org/10.1001/jama.2012.5669
  2. Brower RG, Matthay MA, Morris A, Schoenfeld D, Thompson BT, Wheeler A. Ventilation with lower tidal volumes as compared with traditional tidal volumes for acute lung injury and the acute respiratory distress syndrome. N Engl J Med. 2000 05;342(18):1301–1308. https://doi.org/10.1056/NEJM200005043421801
  3. Papazian L, Forel JM, Gacouin A, Penot-Ragon C, Perrin G, Loundou A, et al. Neuromuscular blockers in early acute respiratory distress syndrome. N Engl J Med. 2010 Sep;363(12):1107–1116. https://doi.org/10.1056/NEJMoa1005372
  4. Wiedemann HP, Wheeler AP, Bernard GR, Thompson BT, Hayden D, deBoisblanc B, et al. Comparison of two fluid-management strategies in acute lung injury. N Engl J Med. 2006 Jun;354(24):2564–2575. https://doi.org/10.1056/NEJMoa062200
  5. Rubenfeld GD, Caldwell E, Peabody E, Weaver J, Martin DP, Neff M, et al. Incidence and outcomes of acute lung injury. N Engl J Med. 2005 Oct;353(16):1685–1693. https://doi.org/10.1056/NEJMoa050333
  6. Bellani G, Laffey JG, Pham T, Fan E, Brochard L, Esteban A, et al. Epidemiology, Patterns of Care, and Mortality for Patients With Acute Respiratory Distress Syndrome in Intensive Care Units in 50 Countries. JAMA. 2016 Feb;315(8):788–800. https://doi.org/10.1001/jama.2016.0291
  7. Mikkelsen ME, Christie JD, Lanken PN, Biester RC, Thompson BT, Bellamy SL, et al. The adult respiratory distress syndrome cognitive outcomes study: long-term neuropsychological function in survivors of acute lung injury. Am J Respir Crit Care Med. 2012 Jun;185(12):1307–1315. https://doi.org/10.1164/rccm.201111-2025OC
  8. Cheung AM, Tansey CM, Tomlinson G, Diaz-Granados N, Matte A, Barr A, et al. Two-year outcomes, health care use, and costs of survivors of acute respiratory distress syndrome. Am J Respir Crit Care Med. 2006 Sep;174(5):538–544. https://doi.org/10.1164/rccm.200505-693OC
  9. Gao Smith F, Perkins GD, Gates S, Young D, McAuley DF, Tunnicliffe W, et al. Effect of intravenous β-2 agonist treatment on clinical outcomes in acute respiratory distress syndrome (BALTI-2): a multicentre, randomised controlled trial. Lancet. 2012 Jan;379(9812):229–235. https://doi.org/10.1016/S0140-6736(11)61623-1
  10. McAuley DF, Laffey JG, O’Kane CM, Perkins GD, Mullan B, Trinder TJ, et al. Simvastatin in the acute respiratory distress syndrome. N Engl J Med. 2014 Oct;371(18):1695–1703. https://doi.org/10.1056/NEJMoa1403285
  11. Boyle AJ, Mac Sweeney R, McAuley DF. Pharmacological treatments in ARDS; a state-of-the-art update. BMC Med. 2013 Aug;11:166. https://doi.org/10.1186/1741-7015-11-166
  12. Floyd CN, Ferro A. Mechanisms of aspirin resistance. Pharmacol Ther. 2014 Jan;141(1):69–78. https://doi.org/10.1016/j.pharmthera.2013.08.005
  13. Toner P, McAuley DF, Shyamsundar M. Aspirin as a potential treatment in sepsis or acute respiratory distress syndrome. Crit Care. 2015 Oct;19:374. https://doi.org/10.1186/s13054-015-1091-6
  14. Lefrancais E, Ortiz-Munoz G, Caudrillier A, Mallavia B, Liu F, Sayah DM, et al. The lung is a site of platelet biogenesis and a reservoir for haematopoietic progenitors. Nature. 2017 04;544(7648):105–109. https://doi.org/10.1038/nature21706
  15. Asaduzzaman M, Lavasani S, Rahman M, Zhang S, Braun OO, Jeppsson B, et al. Platelets support pulmonary recruitment of neutrophils in abdominal sepsis. Crit Care Med. 2009  Apr;37(4):1389–1396. https://doi.org/10.1097/CCM.0b013e31819ceb71
  16. McDonald B, Urrutia R, Yipp BG, Jenne CN, Kubes P. Intravascular neutrophil extracellular traps capture bacteria from the bloodstream during sepsis. Cell Host Microbe. 2012 Sep;12(3):324–333. https://doi.org/10.1016/j.chom.2012.06.011
  17. Fuchs TA, Brill A, Duerschmied D, Schatzberg D, Monestier M, Myers DD, et al. Extracellular DNA traps promote thrombosis. Proc Natl Acad Sci USA. 2010 Sep;107(36):15880–15885. https://doi.org/10.1073/pnas.1005743107
  18. Yoo CG, Lee S, Lee CT, Kim YW, Han SK, Shim YS. Effect of acetylsalicylic acid on endogenous I kappa B kinase activity in lung epithelial cells. Am J Physiol Lung Cell Mol Physiol. 2001 Jan;280(1):3–9. https://doi.org/10.1152/ajplung.2001.280.1.L3
  19. Weber C, Erl W, Pietsch A, Weber PC. Aspirin inhibits nuclear factor-kappa B mobilization and monocyte adhesion in stimulated human endothelial cells. Circulation. 1995 Apr;91(7):1914–1917. https://www.ncbi.nlm.nih.gov/pubmed/7534663
  20. Taubert D, Berkels R, Grosser N, Schroder H, Grundemann D, Schomig E. Aspirin induces nitric oxide release from vascular endothelium: a novel mechanism of action. Br J Pharmacol. 2004 Sep;143(1):159–165. https://doi.org/10.1038/sj.bjp.0705907
  21. El Kebir D, Jozsef L, Pan W, Wang L, Petasis NA, Serhan CN, et al. 15-epi-lipoxin A4 inhibits myeloperoxidase signaling and enhances resolution of acute lung injury. Am J Respir Crit Care Med. 2009 Aug;180(4):311–319. https://doi.org/10.1164/rccm.200810-1601OC
  22. Ortiz-Munoz G, Mallavia B, Bins A, Headley M, Krummel MF, Looney MR. Aspirin-triggered 15-epi-lipoxin A4 regulates neutrophil-platelet aggregation and attenuates acute lung injury in mice. Blood. 2014 Oct;124(17):2625–2634. https://doi.org/10.1182/blood-2014-03-562876
  23. Zarbock A, Singbartl K, Ley K. Complete reversal of acid-induced acute lung injury by blocking of platelet-neutrophil aggregation. J Clin Invest. 2006 Dec;116(12):3211–3219. https://doi.org/10.1172/JCI29499
  24. Trauer J, Muhi S, McBryde ES, Al Harbi SA, Arabi YM, Boyle AJ, et al. Quantifying the Effects of Prior Acetyl-Salicylic Acid on Sepsis-Related Deaths: An Individual Patient Data Meta-Analysis Using Propensity Matching. Crit Care Med. 2017 Nov;45(11):1871–1879. https://doi.org/10.1097/CCM.0000000000002654
  25. Boyle AJ, Di Gangi S, Hamid UI, Mottram LJ, McNamee L, White G, et al. Aspirin therapy in patients with acute respiratory distress syndrome (ARDS) is associated with reduced intensive care unit mortality: a prospective analysis. Crit Care. 2015 Mar;19:109. https://doi.org/10.1186/s13054-015-0846-4
  26. Hamid U, Krasnodembskaya A, Fitzgerald M, Shyamsundar M, Kissenpfennig A, Scott C, et al. Aspirin reduces lipopolysaccharide-induced pulmonary inflammation in human models of ARDS. Thorax. 2017 11;72(11):971-980. https://doi.org/10.1136/thoraxjnl-2016-208571
  27. Kor DJ, Carter RE, Park PK, Festic E, Banner-Goodspeed VM, Hinds R, et al. Effect of Aspirin on Development of ARDS in At-Risk Patients Presenting to the Emergency Department: The LIPS-A Randomized Clinical Trial. JAMA. 2016 Jun;315(22):2406–2414. https://doi.org/10.1001/jama.2016.6330
  28. Seeley E, McAuley DF, Eisner M, Miletin M, Matthay MA, Kallet RH. Predictors of mortality in acute lung injury during the era of lung protective ventilation. Thorax. 2008 Nov;63(11):994–998. https://doi.org/10.1136/thx.2007.093658
  29. Craig TR, Duffy MJ, Shyamsundar M, McDowell C, O’Kane CM, Elborn JS, et al. A randomized clinical trial of hydroxymethylglutaryl- coenzyme a reductase inhibition for acute lung injury (The HARP Study). Am J Respir Crit Care Med. 2011 Mar;183(5):620–626. https://doi.org/10.1164/rccm.201003-0423OC
  30. Reade MC, Angus DC. PAC-Man: game over for the pulmonary artery catheter? Crit Care. 2006 Feb;10(1):303. https://doi.org/10.1186/cc3977

Additional file 1: Statistical analysis plan: STAR_Supplementary_Data

Cite this article as follows:

Toner P, O’Kane C, McNamee J, Verghis R, McAuley DF. ASpirin as a Treatment for Acute Respiratory Distress Syndrome – a multi-centre, randomised, double-blind, placebo-controlled trial (STAR): study protocol. Critical Care Horizons 2018: 1-7.