Aspirin as a Treatment for ARDS

BackgroundThere is no pharmacologic treatment for ARDS. Platelets play an important role in the pathophysiology of ARDS. Preclinical, observational, and clinically relevant models of ARDS indicate aspirin as a potential therapeutic option.Research QuestionIs enteral aspirin (75 mg, once daily) safe and effective in improving surrogate outcomes in adult patients with ARDS?Study Design and MethodsThis randomized, double-blind (patient and investigator), allocation-concealed, placebo-controlled phase 2 trial was conducted in five UK ICUs. Patients fulfilling the Berlin definition of ARDS were randomly assigned at a 1:1 ratio to receive enteral aspirin (75 mg) or placebo, for a maximum of 14 days, using a computer-generated randomization schedule, with variable block size, stratified by vasopressor requirement. The primary end point was oxygenation index (OI) on day 7. Secondary outcomes included safety parameters and other respiratory physiological markers. Analyses were by intention to treat.ResultsThe trial was stopped early, due to slow recruitment, after 49 of a planned 60 patients were recruited. Twenty-four patients were allocated to aspirin and 25 to placebo. There was no significant difference in day 7 OI [aspirin group: unadjusted mean, 54.4 (SD 26.8); placebo group: 42.4 (SD 25); mean difference, 12.0; 95% CI, –6.1 to 30.1; P = .19]. Aspirin did not significantly impact the secondary outcomes. There was no difference in the number of adverse events between the groups (13 in each; OR, 1.04; 95% CI, 0.56-1.94; P = .56).InterpretationAspirin was well tolerated but did not improve OI or other physiological outcomes; a larger trial is not feasible in its current design.Trial RegistrationClinicalTrials.gov; No.: NCT02326350; URL: www.clinicaltrials.gov There is no pharmacologic treatment for ARDS. Platelets play an important role in the pathophysiology of ARDS. Preclinical, observational, and clinically relevant models of ARDS indicate aspirin as a potential therapeutic option. Is enteral aspirin (75 mg, once daily) safe and effective in improving surrogate outcomes in adult patients with ARDS? This randomized, double-blind (patient and investigator), allocation-concealed, placebo-controlled phase 2 trial was conducted in five UK ICUs. Patients fulfilling the Berlin definition of ARDS were randomly assigned at a 1:1 ratio to receive enteral aspirin (75 mg) or placebo, for a maximum of 14 days, using a computer-generated randomization schedule, with variable block size, stratified by vasopressor requirement. The primary end point was oxygenation index (OI) on day 7. Secondary outcomes included safety parameters and other respiratory physiological markers. Analyses were by intention to treat. The trial was stopped early, due to slow recruitment, after 49 of a planned 60 patients were recruited. Twenty-four patients were allocated to aspirin and 25 to placebo. There was no significant difference in day 7 OI [aspirin group: unadjusted mean, 54.4 (SD 26.8); placebo group: 42.4 (SD 25); mean difference, 12.0; 95% CI, –6.1 to 30.1; P = .19]. Aspirin did not significantly impact the secondary outcomes. There was no difference in the number of adverse events between the groups (13 in each; OR, 1.04; 95% CI, 0.56-1.94; P = .56). Aspirin was well tolerated but did not improve OI or other physiological outcomes; a larger trial is not feasible in its current design. ClinicalTrials.gov; No.: NCT02326350; URL: www.clinicaltrials.gov Take-home PointsStudy Question: Is enteral aspirin (75 mg once daily) safe and effective in improving surrogate outcomes in adult patients with ARDS?Results: Less than 10% of screened patients were recruited, and the study was terminated early due to slow recruitment. Aspirin was well tolerated but had no benefit on physiological markers of lung injury or systemic organ dysfunction. Aspirin had no effect on systemic inflammatory responses.Interpretation: Aspirin appeared to be well tolerated in patients with ARDS. However, recruitment was significantly limited because of exclusion criteria. The feasibility of a large interventional study of aspirin in ARDS is questionable, unless further studies are less restrictive in their exclusion criteria. Study Question: Is enteral aspirin (75 mg once daily) safe and effective in improving surrogate outcomes in adult patients with ARDS? Results: Less than 10% of screened patients were recruited, and the study was terminated early due to slow recruitment. Aspirin was well tolerated but had no benefit on physiological markers of lung injury or systemic organ dysfunction. Aspirin had no effect on systemic inflammatory responses. Interpretation: Aspirin appeared to be well tolerated in patients with ARDS. However, recruitment was significantly limited because of exclusion criteria. The feasibility of a large interventional study of aspirin in ARDS is questionable, unless further studies are less restrictive in their exclusion criteria. ARDS remains a considerable cause of mortality1Bellani G. Laffey J. Pham T. et al.Epidemiology, patterns of care, and mortality for patients with acute respiratory distress syndrome in intensive care units in 50 countries.JAMA. 2016; 315: 788-800Crossref PubMed Scopus (2188) Google Scholar and morbidity,2Cheung A.M. Tansey C.M. Tomlinson G. et al.Two-year outcomes, health care use, and costs of survivors of acute respiratory distress syndrome.Am J Respir Crit Care Med. 2006; 174: 538-544Crossref PubMed Scopus (321) Google Scholar with no widely accepted pharmacologic treatment.3Boyle A.J. Mac Sweeney R. McAuley D.F. Pharmacological treatments in ARDS: a state-of-the-art update.BMC Med. 2013; 11: 166Crossref PubMed Scopus (90) Google Scholar Platelets may play a role in the pathophysiology of ARDS.4Toner P. McAuley D.F. Shyamsundar M. Aspirin as a potential treatment in sepsis or acute respiratory distress syndrome.Crit Care. 2015; 19: 1-9PubMed Google Scholar, 5Wei Y. Wang Z. Su L. et al.Platelet count mediates the contribution of a genetic variant in LRRC16A to ARDS risk.Chest. 2015; 147: 607-617Abstract Full Text Full Text PDF PubMed Scopus (34) Google Scholar, 6Idell S. Maunder R. Fein A.M. et al.Platelet- specific α-granule proteins and thrombospondin in bronchoalveolar lavage in the adult respiratory distress syndrome.Chest. 1989; 96: 1125-1132Abstract Full Text Full Text PDF PubMed Scopus (39) Google Scholar, 7Carvalho A.C. Quinn D.A. DeMarinis S.M. et al.Platelet function in acute respiratory failure.Am J Hematol. 1987; 25: 377-388Crossref PubMed Scopus (16) Google Scholar The lungs act as a site for platelet maturation and reservoir for mature platelets.8Lefrançais E. Ortiz-Muñoz G. Caudrillier A. et al.The lung is a site of platelet biogenesis and a reservoir for haematopoietic progenitors.Nature. 2017; 544: 105-109Crossref PubMed Scopus (526) Google Scholar Platelet activation and aggregation recruits additional platelets and leukocytes,9Asaduzzaman M. Lavasani S. Rahman M. et al.Platelets support pulmonary recruitment of neutrophils in abdominal sepsis.Crit Care Med. 2009; 37: 1389-1396Crossref PubMed Scopus (116) Google Scholar and promotes the production of aspirin-triggered lipoxins.10El Kebir D. József L. Pan W. et al.15-Epi-lipoxin A4 inhibits myeloperoxidase signaling and enhances resolution of acute lung injury.Am J Respir Crit Care Med. 2009; 180: 311-319Crossref PubMed Scopus (161) Google Scholar Multiple murine models of ARDS demonstrated that platelet depletion,9Asaduzzaman M. Lavasani S. Rahman M. et al.Platelets support pulmonary recruitment of neutrophils in abdominal sepsis.Crit Care Med. 2009; 37: 1389-1396Crossref PubMed Scopus (116) Google Scholar aspirin administration,11Zarbock A. Singbartl K. Ley K. Complete reversal of acid-induced acute lung injury by blocking of platelet-neutrophil aggregation.J Clin Invest. 2006; 116: 3211-3219Crossref PubMed Scopus (465) Google Scholar and aspirin-triggered lipoxin administration12Ortiz-Muñoz G. Mallavia B. Bins A. et al.Aspirin-triggered 15-epi-lipoxin A4 regulates neutrophil-platelet aggregation and attenuates acute lung injury in mice.Blood. 2014; 124: 2625-2634Crossref PubMed Scopus (120) Google Scholar improved various outcomes including reduced neutrophil migration and pulmonary edema formation and improved survival. Whereas many of these preclinical models involved supratherapeutic systemic doses, several observational studies conclude that low-dose aspirin is associated with better outcomes in patients with ARDS.13Yu H. Ni Y.N. Liang Z.A. et al.The effect of aspirin in preventing the acute respiratory distress syndrome/acute lung injury: a meta-analysis.Am J Emerg Med. 2018; 36: 1486-1491Abstract Full Text Full Text PDF PubMed Scopus (11) Google Scholar, 14Mohananey D. Sethi J. Villablanca P.A. et al.Effect of antiplatelet therapy on mortality and acute lung injury in critically ill patients: a systematic review and meta-analysis.Ann Card Anaesth. 2016; 19: 626-637Crossref PubMed Scopus (11) Google Scholar, 15Boyle A.J. Gangi S. Hamid U. 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; 19: 109Crossref PubMed Scopus (60) Google Scholar, 16O'Neal H.R. Koyama T. Koehler E.S. et al.Prehospital statin and aspirin use and the prevalence of severe sepsis and acute lung injury/acute respiratory distress syndrome.Crit Care Med. 2011; 39: 1343-1350Crossref PubMed Scopus (146) Google Scholar, 17Lösche W. Boettel J. Kabisch B. et al.Do aspirin and other antiplatelet drugs reduce the mortality in critically ill patients?.Thrombosis. 2012; 2012: 720254Crossref PubMed Google Scholar, 18Trauer J. Muhi S. Es M. 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; 45: 1871-1879Crossref PubMed Scopus (28) Google Scholar Aspirin administration reduced alveolar inflammation and injury in a human ex vivo lung perfusion model of ARDS and in a healthy volunteer model of endotoxin-induced lung injury.19Hamid U. Krasnodembskaya A. Fitzgerald M. et al.Aspirin reduces lipopolysaccharide-induced pulmonary inflammation in human models of ARDS.Thorax. 2017; 72: 971-980Crossref PubMed Scopus (38) Google Scholar Low-dose aspirin (75 mg daily) was as effective as high-dose aspirin (1,200 mg daily) in reducing pulmonary neutrophil infiltration and cytokine production in the healthy volunteer model.19Hamid U. Krasnodembskaya A. Fitzgerald M. et al.Aspirin reduces lipopolysaccharide-induced pulmonary inflammation in human models of ARDS.Thorax. 2017; 72: 971-980Crossref PubMed Scopus (38) Google Scholar Recently, the Lung Injury Prevention Study with Aspirin (LIPS-A) trial reported that low-dose aspirin did not reduce the incidence of ARDS within 7 days in patients at high risk (Lung Injury Prediction Score ≥ 4) of developing ARDS.20Kor D.J. Carter R.E. Park P.K. 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; 315: 2406-2414Crossref PubMed Scopus (138) Google Scholar One limitation was the unexpectedly low incidence of ARDS (9.5% compared with the predicted 18%), which limited the ability of the study to detect any statistically significant effect. A subsequent substudy, investigating lipid mediator and leukocyte responses, concluded that of the 367 patients included, 24 developed ARDS following initial administration of the study drug. In this population the incidence of ARDS was numerically (but not statistically significantly) lower in the aspirin-treated group (10 vs 14 patients); however, as before, this substudy was underpowered.21Calfee C.S. Delucchi K.L. Sinha P. et al.Acute respiratory distress syndrome subphenotypes and differential response to simvastatin: secondary analysis of a randomised controlled trial.Lancet Respir Med. 2018; 6: 691-698Abstract Full Text Full Text PDF PubMed Scopus (249) Google Scholar22Abdulnour R.E. Gunderson T. Barkas I. et al.Early intravascular events are associated with development of acute respiratory distress syndrome: a sub-study of the LIPS-A clinical study.Am J Respir Crit Care Med. 2018; 197: 1575-1585Crossref PubMed Scopus (23) Google Scholar Although the LIPS-A trial did not support the use of aspirin in the prevention of ARDS, the STAR (Aspirin as a Treatment for ARDS) study tested the hypothesis that aspirin (75 mg) was an effective treatment option. STAR was a randomized, double-blind (patient and investigator), allocation-concealed, multicenter, placebo-controlled phase 2 trial to determine if aspirin improved surrogate clinical outcomes and was safe in adult patients with ARDS. The study was approved by a national research ethics committee (14/NI/1093) and the Medicines and Health Products Regulation Agency (CTA No. 32485/0025/001-0001, EudraCT No. 2014-002564-32), and was sponsored by the Belfast Health and Social Care Trust. STAR was coordinated through the Northern Ireland Clinical Trials Unit. The trial protocol and statistical analysis plan have been published previously.23Toner P. O'Kane C. Mcnamee J.J. et al.Aspirin as a treatment for acute respiratory distress syndrome: a multi-centre, randomised, study protocol.Crit Care Horizons. 2018; 1: 1-7Google Scholar This report has been prepared in line with the Consolidated Standards of Reporting Trials (CONSORT) guidelines.24Moher D. Hopewell S. Schulz K.F. et al.CONSORT 2010 explanation and elaboration: updated guidelines for reporting parallel group randomised trials.BMJ. 2010; 340: c869Crossref PubMed Scopus (3240) Google Scholar Patients were recruited from five ICUs across Northern Ireland. Invasive mechanically ventilated patients were eligible within 72 h of the onset of ARDS, as defined by acute onset of hypoxemia with a Pao2/Fio2 ratio ≤ 40 kPa with a positive end-expiratory pressure of 5 cm H2O or more, bilateral infiltrates on radiography, not fully explained by pulmonary pathology (Berlin definition of ARDS).25Ranieri V.M. Rubenfeld G.D. Thompson B.T. et al.Acute respiratory distress syndrome: the Berlin definition.JAMA. 2012; 307: 2526-2533Crossref PubMed Scopus (5794) Google Scholar Exclusion criteria included contraindications to aspirin or antiplatelet therapy, or imminent treatment withdrawal. Full inclusion and exclusion criteria are listed in e-Appendix 1. Several protocol amendments were implemented during the course of the study. The first increased the number of recruiting sites. The initial recruiting site was the tertiary center for trauma and neurosurgery, which resulted in a high proportion of exclusions due to head injury or major hemorrhage. The additional sites have a more generalized medical and surgical patient population with the expectation that this would aid recruitment. The next protocol amendment adjusted the platelet cutoff to < 50 × 109/L from < 100 × 109/L, and the exclusion of patients receiving methotrexate was limited to those receiving 15 mg or more per week. Last, the exclusion of patients with active or recurrent peptic ulcer disease was limited to those with active disease within the previous 5 years. Written consent was obtained from each patient's legal representative and followed up by retrospective consent to continue from the patient if that was possible. Where consent to continue was not obtained, consent from the legal representative remained valid, as in previous critical care trials.26McAuley D.F. Cross L.M. Hamid U. et al.Keratinocyte growth factor for the treatment of the acute respiratory distress syndrome (KARE): a randomised, double-blind, placebo controlled trial.Lancet Respir Med. 2017; 5: 484-491Abstract Full Text Full Text PDF PubMed Scopus (51) Google Scholar, 27Craig T.R. Duffy M.J. Shyamsundar M. 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; 183: 620-626Crossref PubMed Scopus (158) Google Scholar, 28McAuley D.F. Laffey J.G. O'Kane C.M. et al.Simvastatin in the acute respiratory distress syndrome.N Engl J Med. 2014; 371: 1695-1703Crossref PubMed Scopus (276) Google Scholar This study was performed in line with the principles of the Declaration of Helsinki, and the study was approved by a national research ethics committee on October 24, 2014 (14/NI/1093). Subjects were randomly assigned to receive enteral aspirin (75 mg) or placebo. Randomization was stratified by vasopressor use: patients were randomized at a 1:1 ratio, using blocks of variable size. Aspirin (75 mg) and placebo were identical in appearance, were packaged identically, and were identified only by a unique pack number. An independent clinical trials pharmacist allocated the next sequential number as per the computer-generated randomization schedule, and was the only person with access to the randomization schedule. Patients and investigators were blinded to treatment allocation. A member of the ICU nursing staff not involved in the trial administered the study drug. Patients were given aspirin (75-mg capsule) or placebo. The dose of aspirin was based on similar dosages of aspirin showing benefit in observational studies13Yu H. Ni Y.N. Liang Z.A. et al.The effect of aspirin in preventing the acute respiratory distress syndrome/acute lung injury: a meta-analysis.Am J Emerg Med. 2018; 36: 1486-1491Abstract Full Text Full Text PDF PubMed Scopus (11) Google Scholar,15Boyle A.J. Gangi S. Hamid U. 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; 19: 109Crossref PubMed Scopus (60) Google Scholar as well as on the fact that low-dose aspirin reduced neutrophil-mediated inflammation in a human model of ARDS.19Hamid U. Krasnodembskaya A. Fitzgerald M. et al.Aspirin reduces lipopolysaccharide-induced pulmonary inflammation in human models of ARDS.Thorax. 2017; 72: 971-980Crossref PubMed Scopus (38) Google Scholar As the median [interquartile range (IQR)] duration for mechanical ventilation in ARDS is 6 (2-12) days,29Huber W. Findeisen M. Lahmer T. et al.Prediction of outcome in patients with ARDS: a prospective cohort study comparing ARDS-definitions and other ARDS-associated parameters, ratios and scores at intubation and over time.PLoS One. 2020; 15e0232720Crossref Scopus (9) Google Scholar treatment for a maximum of 14 days was selected to ensure adequate time for the study drug to have effect. The study drug was continued until 14 days after randomization, or was stopped earlier for any of the following reasons: discharge from critical care, death, drug-related adverse event (AE), or as requested by the clinical team or legal representative, on development of a condition requiring aspirin as treatment or if the hemoglobin dropped below 70 g/L. All patients had baseline demographic data recorded including etiology and severity of ARDS, Murray Lung Injury score, Sequential Organ Failure Assessment (SOFA) score, and Acute Physiology and Chronic Health Evaluation II (APACHE II) score. Daily data were collected at or as close to 10:00 am as possible: these included arterial blood gas, ventilator requirements, SOFA score, and clinical laboratory assessments. Plasma, urine, and BAL samples were collected, when possible, for additional exploratory mechanistic analysis. Clinical care was delivered in accordance with local guidelines. The primary outcome was oxygenation index (OI) on day 7. OI is a physiologic index of the severity of ARDS and measures both impaired oxygenation and the amount of mechanical ventilation delivered. OI (cm H2O/kPa) is calculated as [mean airway pressure (cm H2O) × Fio2 × 100] ÷ Pao2 (kPa), and therefore a decreasing OI correlates with improving oxygenation. OI has been shown to be predictive of mortality in patients with ARDS.30Rubenfeld G.D. Caldwell E. Peabody E. et al.Incidence and outcomes of acute lung injury.N Engl J Med. 2005; 353: 1685-1693Crossref PubMed Scopus (2731) Google Scholar,31Seeley E. McAuley D.F. Eisner M. et al.Predictors of mortality in acute lung injury during the era of lung protective ventilation.Thorax. 2008; 63: 994-998Crossref PubMed Scopus (93) Google Scholar Secondary outcomes included OI on days 4 and 14; respiratory compliance and Pao2/Fio2 ratio on days 4, 7, and 14; and change in SOFA score from baseline to days 4, 7, and 14. Patients were assessed for AEs up to 28 days after completion of the drug. The duration of ventilation, length of ICU and hospital stay, and 28- and 90-day mortality were also recorded. However, these are not considered formal outcome measures because the study did not have sufficient power to assess these outcomes. As the population recruited to the STAR study is already critically ill, it was expected that many of the patients would experience AEs. Events that were expected in this population were not reported as AEs. The summary of product characteristics for aspirin was used as the reference safety information. All serious adverse events (SAEs) were reported with the outcome and the association with the underlying clinical condition. Sample size calculations were informed by previous clinical trials evaluating therapies for patients with ARDS.26McAuley D.F. Cross L.M. Hamid U. et al.Keratinocyte growth factor for the treatment of the acute respiratory distress syndrome (KARE): a randomised, double-blind, placebo controlled trial.Lancet Respir Med. 2017; 5: 484-491Abstract Full Text Full Text PDF PubMed Scopus (51) Google Scholar,27Craig T.R. Duffy M.J. Shyamsundar M. 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; 183: 620-626Crossref PubMed Scopus (158) Google Scholar The mean (SD) OI on day 7 in patients with ARDS is 62 (51) cm H2O/kPa.27Craig T.R. Duffy M.J. Shyamsundar M. 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; 183: 620-626Crossref PubMed Scopus (158) Google Scholar A sample size of 56 participants has 80% power at a two-tailed significance level of .05 to detect a difference in OI of 39 cm H2O/kPa. Assuming a dropout rate of 3% (in keeping with previous ARDS trials32Reade M.C. Angus D.C. Elbourne D. et al.PAC-Man: game over for the pulmonary artery catheter?.Crit Care. 2006; 10: 303Crossref PubMed Scopus (18) Google Scholar), it was planned to recruit a total of 60 patients. In a previous phase 2 study of similar size we found that an intervention can demonstrate a change in OI of similar magnitude, confirming that a treatment effect of this size can be achieved.27Craig T.R. Duffy M.J. Shyamsundar M. 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; 183: 620-626Crossref PubMed Scopus (158) Google Scholar Analyses were on an intention-to-treat basis. The primary outcome measure was OI on day 7. Day 7 was chosen because this time interval will minimize the competing effects of death and extubation, while allowing a sufficient time interval for a biological effect to occur. The primary analysis looked at OI on day 7, using only data available at that specific time point. To account for attrition, that is, patients who did not have an OI measured on day 7 because of death or extubation, a further analysis of OI on day 7 used the last available data before the patient discontinued from the study to form an imputed day 7 OI. Secondary outcomes were analyzed using data available at the specified time point without imputation. Unadjusted analyses used an independent t test. Adjusted analyses were undertaken using analysis of covariance to adjust for the baseline values. The ventilator-free days score, duration of mechanical ventilation, and length of stay (LOS) outcomes were summarized as median and interquartile range (25th percentile to 75th percentile) and tested by Wilcoxon rank-sum test. The categorical variables were tested by Fisher exact test. A linear mixed model was used to compare change, over time, in a range of clinical laboratory assessments including CBC count, renal function, and coagulation screen. All available data points were considered in the model; treatment was the fixed effect and the patient was the random effect in the mixed model. This model provides an estimate of fixed effects while adjusting for the nonindependence due to repeated measurements on each subject. Patients were recruited from February 10, 2015 to November 25, 2018. During this 45-month period, 593 potential participants were screened, with 49 (8.3%) successfully recruited (reasons for exclusion listed in e-Table 1). The Data Monitoring Committee (DMC) reviewed the unblinded data throughout the study, and in November 2018 the DMC recommended closure of the study after 49 of the planned 60 patients had been recruited, due to slow recruitment. No safety concerns were highlighted by the DMC, but it was thought that to continue the study would not significantly alter the outcome. Twenty-four patients were allocated to the aspirin group and 25 to the placebo group. All patients received their allocated medications and no patient withdrew or was lost to follow-up (Fig 1). The baseline demographic and clinical characteristics for each randomized group are shown in Table 1. Although the two groups were largely comparable, the aspirin group had more male participants (aspirin, 58.3% vs placebo, 40%), had a higher baseline APACHE II score (aspirin, 24.4 vs placebo, 22.0), and had a lower vasopressor requirement (aspirin, 54.2% vs placebo, 64%). The main cause of ARDS was pneumonia, but the placebo group had more sepsis-induced ARDS than did the aspirin group (aspirin, 20.8% vs placebo, 40%). These differences probably likely represent a chance finding, due to the lower than planned sample size.Table 1Baseline Demographics and Clinical CharacteristicsDemographic/CharacteristicAspirinPlaceboPatients, No.2425Age, mean (SD), y55.4 (12)56.4 (17.7)Male patients, No. (%)14 (58.3)10 (40)APACHE II score, mean (SD)24.4 (7.3)22.0 (7.7)Mean arterial pressure, mean (SD), mm Hg64.3 (7.2)65.0 (5.3)Tidal volume, mean (SD), mL/kg PBW7.2 (2.3)7.6 (2.4)Vasopressor required (yes), No. (%)13 (54.2)16 (64.0)Plateau pressure, mean (SD), cm H2O23.5 (6.0)24.1 (4.9)Oxygenation index, mean (SD), kPa59.3 (30.3)66.3 (21.0)Worst Pao2/Fio2 ratio, mean (SD), kPa13.6 (6.4)13.1 (6.1)SOFA score, mean (SD)9.4 (3.3)10.3 (4.0)Etiology of ARDS, No. (%)aSome patients had more than one recorded cause of ARDS; in the aspirin group there were 30 causes of ARDS for 24 patients and in the placebo group there were 37 causes of ARDS for 25 patients. Smoke/toxin inhalation1 (4.2)2 (8.0) Gastric content aspiration6 (25.0)6 (24.0) Near drowning0 (0.0)0 (0.0) Thoracic trauma2 (8.3)2 (8.0) Pneumonia14 (58.3)13 (52.0) Sepsis5 (20.8)10 (40.0) Pancreatitis1 (4.2)0 (0.0) Nonthoracic trauma0 (0.0)1 (4.0) OtherbOther causes include out-of-hospital cardiac arrest, blood transfusion-related cause, and intraabdominal surgery.1 (4.2)3 (12.0) No etiology identified00Data represent No. (%) or mean (SD). APACHE II = Acute Physiology and Chronic Health Evaluation II; OI = oxygenation index; PBW = predicted body weight; SOFA = Sequential Organ Failure Assessment.a Some patients had more than one recorded cause of ARDS; in the aspirin group there were 30 causes of ARDS for 24 patients and in the placebo group there were 37 causes of ARDS for 25 patients.b Other causes include out-of-hospital cardiac arrest, blood transfusion-related cause, and intraabdominal surgery. Open table in a new tab Data represent No. (%) or mean (SD). APACHE II = Acute Physiology and Chronic Health Evaluation II; OI = oxygenation index; PBW = predicted body weight; SOFA = Sequential Organ Failure Assessment. The median duration (IQR) of treatment was 9 (6-13) days in the aspirin group and 8 (6-14) days in the placebo group (e-Table 2). The most common reasons for discontinuation of the study drug were completion of the treatment course on day 14 (aspirin, 21% vs placebo, 36%), occurrence of an AE (aspirin, 29% vs placebo, 24%), or discharge from critical care (aspirin, 29% vs placebo, 20%). Eight protocol deviations were recorded. Four patients (one in the aspirin group and three in the placebo group) received an extra dose of the study drug in error, either after it had been discontinued or after the total 14-day course was complete. Two patients (one in the aspirin group and one in the placebo group) had a dose of their study drug omitted in error. Two other protocol deviations were recorded, both of which involved accidental discarding of the extra study drug instead of the study drug being returned to the pharmacy after completion of the study (e-Table 2). There was no significant difference between the mean observed OI in the aspirin group [mean, 54.4 (SD 26.8)] and the placebo group on day 7 [mean, 42.4 (SD 25); mean difference, 12.0; 95% CI, –6.1 to 30.1; P = .19]. OI on day 7 was available for 17 patients in each group (see e-Table 3 for reasons OI was not available). Similarly, there was no significant difference between the mean OI when the last observation was carried forward, in the aspirin group [mean, 45.7 (27.0)] and the placebo group on day 7 [mean, 46.8 (SD 33.8); mean difference, –1.1; 95% CI, –18.7 to 16.6; P = .90]. Furthermore, there was no significant difference in mean observed OI difference between the aspirin and the placebo groups on days 4 and 14. There were no significan