Over the last 40 years, since the initial description of acute respiratory distress syndrome (ARDS) , there have been significant advances in our management of ARDS, including low tidal volume ventilation, restrictive fluid resuscitation, prone positioning, and higher positive end-expiratory pressure (PEEP) in patients with moderate to severe ARDS [2–5]. In spite of these advances, mortality in ARDS remains high, up to 45% for severe ARDS . Prior clinical trials in ARDS focused on treatment of ARDS to improve mortality after ARDS has developed. However, a multidisciplinary panel of experts convened by the National Heart Lung Blood Institute (NHLBI) in 2010 to discuss the future of ARDS research made a key recommendation to shift the focus from treatment to prevention of ARDS in future clinical trials .
PRIOR APPROACH IN ACUTE RESPIRATORY DISTRESS SYNDROME CLINICAL TRIALS
Most clinical trials in ARDS have enrolled patients in the ICU after the patients have developed ARDS. The timing of enrollment depended partly on the biological rationale. However, even trials aimed at early treatment in ARDS have allowed enrollment for up to 48 h after meeting the last criteria for ARDS. This may mean starting intervention well into the critical illness that predisposed the patients to lung injury originally. Most ARDS patients die of multiple organ failure rather than refractory hypoxemia. But multiple organ failure is not necessarily a consequence of ARDS. Indeed, nonpulmonary organ failure often precedes development of ARDS. The median time from ICU admission to development of ARDS was 1 day [interquartile range (IQR) 0–3 days] and 70% of ARDS patients with acute kidney injury (AKI) presented with acute kidney injury on ICU admission (median 0 days after ICU admission IQR 0–1) often before patients fulfill criteria for ARDS . Furthermore, worsening of organ failure over the first 48 h of ICU hospitalization is a better predictor of mortality than severity of illness on ICU admission. Among those with improvement of SOFA within the first 48 h of ICU, mortality is less than 5%  while an increase in the Acute Physiology and Chronic Health Evaluation physiology score by ICU day 3 has a positive predictive value (PPV) of 97% for ICU mortality .
Thus, by the time of study enrollment, many ARDS patients already have established end organ damage and some have settled into a clinical trajectory of improvement or clinical deterioration. Interventions for such patients may be less likely to change outcomes resulting in a smaller effect size and a ‘negative’ trial where any potential benefit may be too small to be detected with statistical significance . Indeed, inappropriate timing of experimental interventions was one reason given for the large number of negative clinical trials in acute critical illness . The potential for early intervention to improve patient outcomes is being tested now in the new NHLBI Prevention and Early Treatment of Acute Lung Injury (PETAL) Network (U01 HL122998). The PETAL Network is the new clinical trials network succeeding ARDSnet and is designed to conduct clinical trials aimed at the prevention and early treatment of ARDS.
A search of PubMed and Clinicaltrials.gov was done for clinical trials and the keywords ‘prevention’, ‘acute lung injury’, or ‘Acute Respiratory Distress Syndrome’. After a review of the abstracts and manuscripts to confirm the study was a clinical trial with a primary or secondary outcome of acute lung injury or ARDS, we identified 19 published and completed trials in PubMed and 16 unpublished trials registered under ClinicalTrials.gov over the last 20 years (Tables 1 and 2) [13–26,27▪▪,28–30]. These trials enrolled different at-risk patient populations, had interventions that are delivered at different stages of critical illness, and examined different clinical outcomes. Such variability raises several issues about the optimal design of prevention trials.
APPROACHES TO PREVENTION IN ACUTE RESPIRATORY DISTRESS SYNDROME
Prevention in epidemiologic and public health terms has several levels: primary, secondary, tertiary, and more recently, quaternary . Each level of prevention has different goals, targets different populations, and is initiated at different times in the course of the disease (see Fig. 1). In primary prevention, the goal is to avoid development of a disease or condition. In general, primary prevention would target patients before or soon after they develop a risk factor for ARDS prior to any signs of respiratory insufficiency (e.g., simvastatin in patients before elective esophagectomies ). In secondary prevention, the goal is to detect and treat a preclinical condition early to reduce the severity and minimize or eliminate the sequelae of the condition. Inherent in secondary prevention is the need for screening to identify patients early. An example of secondary prevention in ARDS may be an intervention targeting patients with hypoxia or severe sepsis but who do not fulfill all criteria for ARDS yet. Tertiary prevention aims to reduce the impact of an existing, recognized condition with the goal of improving survival, function, and quality of life. Tertiary prevention requires accurate diagnosis and follow-up of patients with the condition of interest. Prior trials focused on improving mortality in ARDS represent one aspect of tertiary prevention in ARDS. However, tertiary prevention could also include clinical trials to improve cognitive or physical function in ARDS survivors. Quaternary prevention aims to identify patients at risk for excessive or unnecessary interventions with the goal of preventing additional interventions that may not be beneficial and to suggest ethically acceptable, alternative therapies instead . Quaternary prevention is of high importance in acute critical illness, including ARDS as reflected by the Choosing Wisely campaign in critical care .
Within this framework, both universal, selective, and indicated prevention strategies can be applied. For example, a universal preventive strategy would reach a broad population without regard to individual risk for the condition. Enrolling all patients on mechanical ventilation without ARDS is an example of a universal prevention strategy for ARDS. A selective preventive strategy would target subgroups of patients at increased risk for a condition (e.g., patients admitted to the hospital with pneumonia or sepsis at risk for ARDS). An indicated preventive strategy would target individuals at particularly high risk for the condition with possible early signs of the condition and aims to prevent the individual from progressing to the disease of interest (e.g., pneumonia patients with a history of alcohol abuse requiring more than 4l nasal cannula of oxygen at presentation). A clinical trial that uses a clinical prediction score to identify patients at high risk for developing ARDS for inclusion into a prevention trial would be utilizing an indicated prevention strategy. Given the rapidity with which ARDS develops, there may be overlap between selective and indicative prevention strategy but typically, the incidence of ARDS would be higher among patients targeted with a selected or indicated prevention strategy compared with patients targeted with a universal prevention strategy.
Thus the spectrum of prevention studies in ARDS can span the course of acute critical illness. The various challenges and considerations with prevention trials in ARDS are not unique to prevention trials but the approaches to these challenges will depend on the intervention and the different prevention levels and study population targeted in a prevention trial in ARDS.
CHALLENGES AND CONSIDERATIONS IN PREVENTION TRIALS IN ACUTE RESPIRATORY DISTRESS SYNDROME
Target population in acute respiratory distress syndrome prevention
As the focus shifts from treatment to prevention, the study population will be at-risk patients rather than patients with established ARDS. The at-risk patient population in an ARDS prevention trial depends on the biologic rationale, the risk of the intervention, the feasibility of recruitment, and the incidence of ARDS needed to be able to detect a statistically significant improvement in outcomes.
Several of the prevention studies are primary prevention of ARDS in a selective population. Some examples are trials that focused on patients undergoing elective high-risk surgeries associated with increased development of postoperative ARDS such as cardiac surgery or esophagostomies [22,28,34–36]. In these studies, the preventive intervention was delivered pre or perioperatively before any signs of lung injury or respiratory insufficiency. But the incidence of ARDS after elective surgery is low, occurring in only 2.6% of patients undergoing high-risk surgery . A large sample size would be required to detect an effect, with significant implications for the feasibility and costs of the trial.
Sepsis, pneumonia, aspiration, or trauma are far more common predisposing factors for ARDS but the incidence of ARDS can vary considerably with the inciting injury and with whether screening was done on hospital admission versus ICU presentation (Table 3). Screening on ICU referral or admission will involve sicker patients further along in their course of critical illness with a higher rate of ARDS development and mortality. Although this can help with power of a trial, it may also mean intervening after ARDS has developed. The median time to development of ARDS was 1 day after ICU admission with 38% of the ARDS patients fulfilling criteria for ARDS on ICU admission .
Recently more work focused on tools for the early identification of patients at increased risk for progressing to ARDS for prevention trials. This is essentially secondary prevention using an indicated prevention strategy. The Lung Injury Prediction Score (LIPS) uses clinical data present within the first 6 h of presentation to the emergency department to identify patients at increased risk of developing ARDS (Fig. 2) . A LIPS of at least 4 has a sensitivity of 69% and specificity of 78% (area under the curve of 0.80) of identifying ARDS cases. The PPV of LIPS at least 4 is only 18% which means that a large proportion of patients identified by LIPS may still not develop ARDS. This makes LIPS most appropriate for low risk preventive interventions as most patients would not develop ARDS. Increasing the cutoff to LIPS at least 5 improves the PPV slightly from 18 to 23% but it comes with much greater likelihood of missing ARDS cases as the sensitivity decreased from 69 to 53%. Thus, increasing the cutoff in LIPS may not necessarily increase the power of the study to detect a change in rate of ARDS development. Recently, the validity of LIPS to identify patients at high risk for progressing to ARDS from the emergency department was confirmed in an external cohort . The LIPS has since been used to identify at-risk patients for enrollment into the Lung Injury Prevention Study-Aspirin (LIPS-A) . This trial has completed enrollment, demonstrating the feasibility of doing a secondary prevention trial using an indicated preventive strategy.
Other clinical tools to identify patients at increased risk for developing ARDS include the Early Acute Lung Injury score (EALI) and the Surgical Lung Injury Score (SLIP). EALI examined nonintubated patients without ARDS who already show signs of possible early lung injury as indicated by bilateral infiltrates on chest radiograph in the emergency department . An EALI score at least 2 had a sensitivity of 89%, specificity of 75%, and PPV of 53%. However, EALI requires manual titration of oxygen by study staff to determine the lowest level of supplemental oxygen required to maintain oxygen saturation at least 90%. Similarly, SLIP was developed to identify surgical patients at increased risk for developing ARDS, given the low rates of ARDS after surgery . However, neither EALI nor SLIP have been validated in an external cohort or used yet in a clinical trial for prevention of ARDS.
Currently missing is a reliable biomarker-driven approach to identify patients at increased risk for ARDS. Agrawal et al.  found that plasma angiopoietin-2 in combination with LIPS better predict for subsequent development of ARDS than either LIPS or angiopoietin-2 alone. Such a biologic and clinical phenotypic approach can also identify patients that may better respond to an intervention. Calfee et al.  found that the hyper-inflammatory subphenotypes in ARDS predicted for worse clinical outcomes and better response to high PEEP in the ARDSnet ALVEOLI trial. However, for this to be operationalized in a prevention trial with a short window for enrollment there must be a rapid, accurate point of care testing. As yet, there are no prevention clinical trials using this strategy for enrollment.
Timing of intervention in acute respiratory distress syndrome prevention trials
The timing of intervention in an ARDS prevention trial depends greatly on whether the intervention is meant to be primary, secondary, or tertiary prevention (Fig. 1). Primary prevention in ARDS, when the intervention begins before any signs of lung injury are apparent, is only feasible in patients in whom the acute injury predisposing to ARDS can be predicted, such as those undergoing elective high-risk surgery or patients who need blood transfusion [22,28,45,46].
More commonly, ARDS occurs after an acute critical illness like pneumonia or trauma, which often occurs before hospitalization. In such cases, a patient's presentation to the hospital for medical attention may be the first practical point of contact for secondary prevention trials. Only 30% of ARDS patients fulfill criteria within the first 6 h of presentation to the hospital. The majority of patients developed ARDS a median of 2 days (IQR 1–4 days) after hospital admission, representing a potential window of opportunity for early prevention and treatment of ARDS .
The time period between when a patient presents to the hospital and when they develop ARDS often occurs in the emergency department or other non-ICU areas of the hospital. Acute critical illness is now increasingly being identified and managed outside of the ICU. Between 2001 and 2009, the number of critical care admissions from the emergency department increased 79% and the critical care time spent on emergency department patients increased from a median of 185 to 245 min per patient . In another study, 14% of adult patients hospitalized outside of the ICU had severe sepsis with a hospital mortality of 13% with 3/4 of these patients having at least two organ failures . This means that any meaningful screening for ARDS prevention trials must occur outside of the ICU. Indeed, the NHLBI PETAL Network recognized this key point and required two investigators from each site: one intensivist and another investigator from another specialty like emergency medicine or surgery who will have access to high-risk patients before they may present to the ICU.
Study design considerations in acute respiratory distress syndrome prevention trials
Prior treatment trials in ARDS have been designed to demonstrate effectiveness with patient randomized controlled trials using strict inclusion and numerous exclusion criteria to minimize risk and to reduce heterogeneity among the study population. As the focus has shifted to prevention trials, newer study designs may be applicable.
For secondary or tertiary prevention of high-risk study population with an intervention that is more than minimal risk, the traditional effectiveness study design may still be appropriate to ensure safety and to maximize the likelihood of detecting a benefit. However, as the risk benefit ratio of preventive interventions can be difficult to predict, new clinical trial designs such as adaptive designs can be used to minimize risk or to find the optimal dose for benefit .
However, in instances where the incidence of ARDS may be low and the intervention is minimal risk, pragmatic trial designs may be possible. Pragmatic trials have broad inclusion criteria, few exclusion criteria, and the intervention is delivered in the usual care setting without the strict adherence typically seen in the traditional effectiveness trials [50–52]. Such trials are usually more conducive to large sample size, important for power given the lower incidence of the outcome and the more heterogeneous population included in the study.
Cluster randomization is another alternative trial design utilized for critical care trials . Instead of randomizing patients individually to an intervention, all patients within a cluster of ICUs or hospitals are randomized to an intervention or control. This is particularly useful for processes of care interventions in the usual care setting that require education of clinical staff and high likelihood of cross contamination of practices between patients randomized to different groups treated in the same clinical area . This is relevant for ARDS prevention as there have been recent interests in how processes of care in the early period of critical illness could influence the development of ARDS. Li et al.  found the incidence of hospital-acquired ARDS decreased significantly after the successful implementation of best care practices in the ICU, including restrictive transfusion strategy, low tidal volume ventilation, protocolized early resuscitation, and antibiotics in severe sepsis. Indeed, the findings from this study led to the development of the Checklist for Lung Injury Prevention (CLIP). CLIP was used to standardize the care that may affect the development of ARDS in the LIPS-A . However, effectiveness of CLIP has not been demonstrated in a clinical trial, but CLIP would be amenable to hospital level cluster-randomized controlled trial done under a quality improvement initiative.
Outcome measures in acute respiratory distress syndrome prevention trials
In ARDS prevention trials, the goal is to reduce the development of ARDS. However, the use of ARDS as a primary outcome may be problematic. ARDS is not a patient-centered outcome. Prevention of ARDS is meaningless to patients if they were to develop prolonged respiratory failure and die later of nonpulmonary organ failure. Another issue with ARDS as an outcome is that clinical determinants of the syndrome can be manipulated without modifying outcome. High PEEP can improve PaO2/FiO2 without resolution of ARDS. Lastly, a patient must survive long enough to develop ARDS so a preventive intervention that ends up with higher early mortality may actually result in a lower rate of ARDS.
One goal of prevention is to improve mortality by preventing a highly fatal condition like ARDS. But powering for mortality based upon deaths averted from preventing ARDS can be problematic because not all deaths are directly related to ARDS. In general, a much larger sample size is needed to demonstrate a mortality benefit than would be required to demonstrate a change in rate of ARDS. For example, in Determan's small trial  low tidal volume ventilation was shown to decrease development of ARDS without any change in mortality. However, if the study intervention improves the survival in both the risk condition leading to ARDS and in patients with ARDS, then a larger effect on mortality would be found.
Many ARDS treatment trials have utilized the composite outcome of ventilator-free days (VFD), a continuous outcome that combines mortality with time to successful liberation which may allow for smaller sample sizes . However, depending on the target at-risk population in an ARDS prevention trial, not all patients will be ventilated and intubation may occur on different days into the study. The potential bias from this on VFD needs to be determined. VFD works best if one can assume that the treatment can reduce both duration of mechanical ventilation and mortality, but a treatment that results in better survival at the cost of longer duration of mechanical ventilation may result in no difference in VFD even if there is a survival difference. Lastly, VFD is not a patient-centered outcome. Although one might argue that patients would prefer to minimize the duration of mechanical ventilation, it is unclear whether patients would consider death before 28 days to be comparable with survival with prolonged mechanical ventilation.
Similar to VFD is hospital-free days which would give priority to interventions likely to return patients to home. However, hospital length of stay even for similar patients may differ by hospital, insurance status, and whether long-term acute care hospitals are easily available .
Another potential outcome measure is the persistent organ dysfunction plus death . This composite outcome combines death with persistent organ dysfunction defined as ongoing vasopressor need, mechanical ventilation, or dialysis and was shown to correlate with 6-month mortality, longer hospital and ICU stays, and worse quality of life at 3 and 6 months. Notably, it allowed for smaller sample size than that required to demonstrate a statistically significant reduction in mortality or VFD.
Tertiary prevention studies in ARDS can also focus on reducing the complications and loss of function associated with ARDS. At 1 year from ARDS, up to 70% and 31–36% of ARDS patients suffer from cognitive and functional impairment [58–60]. Given this high burden of complications experienced by ARDS survivors, clinical trials aimed at improving the long-term functional outcomes of ARDS survivors will be important. However, preventing ARDS without significantly decreasing the duration of mechanical ventilation or the extent of nonpulmonary organ failure may not improve functional and cognitive outcomes in patients.
Traditionally clinical trials in ARDS have focused on improving mortality after development of ARDS. There is now an exciting shift in emphasis toward earlier identification of patients at risk for ARDS and prevention trials aimed at preventing the development of ARDS and its sequelae. Prevention in ARDS can vary according to the at-risk population and timing of intervention. While this presents an opportunity, it also presents challenges in selecting the appropriate at-risk patient population, the timing of implementation of prevention, study design, and outcome measures. With the recent increase in the number of trials aimed at ARDS prevention and the NHLBI PETAL network, hopefully many of these challenges will be overcome and cost-effective treatments identified. This shift from management of established organ failure to prevention of organ failure has enormous potential to improve public health and improve the efficiency of the healthcare system.
We would like to thank Brittany Gary MD for her assistance in the literature search.
B.T.T. is currently receiving a grant from NHLBI (UO1HL123009) and M.N.G. is currently receiving grants from NHLBI (U01 HL122998 and UH3 HL125119).
Financial support and sponsorship
The work is supported by the National Heart, Lung, Blood Institute (NHLBI) at the National Institute of Health.
Conflicts of interest
There are no conflicts of interest.
REFERENCES AND RECOMMENDED READING
Papers of particular interest, published within the annual period of review, have been highlighted as:
- ▪ of special interest
- ▪▪ of outstanding interest
1. Ashbaugh DG, Bigelow DB, Petty TL, Levine BE. Acute respiratory distress in adults. Lancet 1967; 2:319–323.
2. Brower RG, Matthay MA, Morris A, et al. Ventilation with lower tidal volumes as compared with traditional tidal volumes for acute lung injury and the acute respiratory distress syndrome. New Engl J Med 2000; 342:1301–1308.
3. Taccone P, Pesenti A, Latini R, et al. Prone positioning in patients with moderate and severe acute respiratory distress syndrome: a randomized controlled trial. JAMA 2009; 302:1977–1984.
4. Wiedemann HP, Wheeler AP, Bernard GR, et al. Comparison of two fluid-management strategies in acute lung injury. New Engl J Med 2006; 354:2564–2575.
5. Briel M, Meade M, Mercat A, et al. Higher vs lower positive end-expiratory pressure in patients with acute lung injury and acute respiratory distress syndrome: systematic review and meta-analysis. JAMA 2000; 303:865–873.
6. Force ADT, Ranieri VM, Rubenfeld GD, et al. Acute respiratory distress syndrome: the Berlin Definition. JAMA 2012; 307:2526–2533.
7. Spragg RG, Bernard GR, Checkley W, et al. Beyond mortality: future clinical research in acute lung injury. Am J Respir Crit Care Med 2000; 181:1121–1127.
8. Soto GJ, Frank AJ, Christiani DC, Gong MN. Body mass index and acute kidney injury in the acute respiratory distress syndrome. Crit Care Med 2012; 40:2601–2608.
9. Ferreira FL, Bota DP, Bross A, et al. Serial evaluation of the SOFA score to predict outcome in critically ill patients. JAMA 2001; 286:1754–1758.
10. Afessa B, Keegan MT, Mohammad Z, et al. Identifying potentially ineffective care in the sickest critically ill patients on the third ICU day. Chest 2004; 126:1905–1909.
11. Rubenfeld G. Confronting the frustrations of negative clinical trials in acute respiratory distress syndrome. Ann Am Thorac Soc 2015; 12 (Supplement 1):s58–s63.
12. Vincent JL. We should abandon randomized controlled trials in the intensive care unit. Crit Care Med 2010; 38 (10 Suppl):S534–538.
13. Weigelt JA, Norcross JF, Borman KR, Snyder WH 3rd. Early steroid therapy for respiratory failure. Arch Surg 1985; 120:536–540.
14. Luce JM, Montgomery AB, Marks JD, et al. Ineffectiveness of high-dose methylprednisolone in preventing parenchymal lung injury and improving mortality in patients with septic shock. Am Rev Respir Dis 1988; 138:62–68.
15. Slotman GJ, Burchard KW, D’Arezzo A, Gann DS. Ketoconazole prevents acute respiratory failure in critically ill surgical patients. J Trauma 1988; 28:648–654.
16. Yu M, Tomasa G. A double-blind, prospective, randomized trial of ketoconazole, a thromboxane synthetase inhibitor, in the prophylaxis of the adult respiratory distress syndrome. Crit Care Med 1993; 21:1635–1642.
17. Nathens AB, Neff MJ, Jurkovich GJ, et al. Randomized, prospective trial of antioxidant supplementation in critically ill surgical patients. Ann Surg 2002; 236:814–822.
18. Schuster DP, Metzler M, Opal S, et al. Recombinant platelet-activating factor acetylhydrolase to prevent acute respiratory distress syndrome and mortality in severe sepsis: Phase IIb, multicenter, randomized, placebo-controlled, clinical trial. Crit Care Med 2003; 31:1612–1619.
19. Manzano F, Fernandez-Mondejar E, Colmenero M, et al. Positive-end expiratory pressure reduces incidence of ventilator-associated pneumonia in nonhypoxemic patients. Crit Care Med 2008; 36:2225–2231.
20. Esteban A, Frutos-Vivar F, Ferguson ND, et al. Sepsis incidence and outcome: contrasting the intensive care unit with the hospital ward. Crit Care Med 2007; 35:1284–1289.
21. Bulger EM, Jurkovich GJ, Nathens AB, et al. Hypertonic resuscitation of hypovolemic shock after blunt trauma: a randomized controlled trial. Arch Surg 2008; 143:139–148.discussion 149.
22. Shyamsundar M, McAuley DF, Shields MO, et al. Effect of simvastatin on physiological and biological outcomes in patients undergoing esophagectomy: a randomized placebo-controlled trial. Ann Surg 2014; 259:26–31.
23. Determann RM, Royakkers A, Wolthuis EK, et al. Ventilation with lower tidal volumes as compared with conventional tidal volumes for patients without acute lung injury: a preventive randomized controlled trial. Critical care 2010; 14:R1.
24. Onorati F, Santini F, Mariscalco G, et al. Leukocyte filtration ameliorates the inflammatory response in patients with mild to moderate lung dysfunction. Ann Thoracic Surg 2011; 92:111–121.discussion 121.
25. Sopena N. [Diagnostic methods and therapeutic approach in nonventilated nosocomial pneumonia patients]. Enferm Infecc Microbiol Clin 2005; 23:517–518.
26. Sopena N, Sabria M, Neunos Study G. Multicenter study of hospital-acquired pneumonia in non-ICU patients. Chest 2005; 127:213–219.
27▪▪. Futier E, Constantin JM, Paugam-Burtz C, et al. A trial of intraoperative low-tidal-volume ventilation in abdominal surgery. N Engl J Med 2013; 369:428–437.
A multicenter randomized controlled trial of lung protective ventilation in abdominal surgery. The development of ARDS was not statistically significant (0.5% in lung protective ventilation group versus 3% in control group), but the incidence of ARDS is too low for them to have sufficient power to detect an effect. The trial did find that low tidal volume ventilation results in lower rates of pulmonary and nonpulmonary complications.
28. Perkins GD, Gates S, Park D, et al. The beta agonist lung injury trial prevention
. A randomized controlled trial. Am J Respir Crit Care Med 2014; 189:674–683.
29. Hall WB, Willis LE, Medvedev S, Carson SS. The implications of long-term acute care hospital transfer practices for measures of in-hospital mortality and length of stay. Am J Respir Crit Care Med 2012; 185:53–57.
30. Seguin P, Laviolle B, Dahyot-Fizelier C, et al. Effect of oropharyngeal povidone-iodine preventive oral care on ventilator-associated pneumonia in severely brain-injured or cerebral hemorrhage patients: a multicenter, randomized controlled trial. Crit Care Med 2014; 42:1–8.
32. Pandve HT. Quaternary prevention
: need of the hour. J Family Med Prim Care 2014; 3:309–310.
33. Halpern SD, Becker D, Curtis JR, et al. An official American Thoracic Society/American Association of Critical-Care Nurses/American College of Chest Physicians/Society of Critical Care Medicine policy statement: the Choosing Wisely top 5 list in critical care medicine. Am J Respir Crit Care Med 2014; 190:818–826.
34. Kim JC, Hong SW, Shim JK, et al. Effect of N-acetylcysteine on pulmonary function in patients undergoing off-pump coronary artery bypass surgery. Acta Anaesthesiol Scand 2011; 55:452–459.
35. Kim JC, Shim JK, Lee S, et al. Effect of combined remote ischemic preconditioning and postconditioning on pulmonary function in valvular heart surgery. Chest 2012; 142:467–475.
36. Li C, Xu M, Wu Y, et al. Limb remote ischemic preconditioning attenuates lung injury after pulmonary resection under propofol-remifentanil anesthesia: a randomized controlled study. Anesthesiology 2014; 121:249–259.
37. Kor DJ, Warner DO, Alsara A, et al. Derivation and diagnostic accuracy of the surgical lung injury prediction model. Anesthesiology 2011; 115:117–128.
38. Gong MN, Thompson BT, Williams P, et al. Clinical predictors of and mortality in acute respiratory distress syndrome: potential role of red cell transfusion. Crit Care Med 2005; 33:1191–1198.
39. Gajic O, Dabbagh O, Park PK, et al. Early identification of patients at risk of acute lung injury: evaluation of lung injury prediction score in a multicenter cohort study. Am J Respir Crit Care Med 2011; 183:462–470.
40. Agrawal A, Matthay MA, Kangelaris KN, et al. Plasma angiopoietin-2 predicts the onset of acute lung injury in critically ill patients. Am J Respir Crit Care Med 2013; 187:736–742.
41. Kor DJ, Talmor DS, Banner-Goodspeed VM, et al. Lung Injury Prevention
with Aspirin (LIPS-A): a protocol for a multicentre randomised clinical trial in medical patients at high risk of acute lung injury. BMJ Open 2012; 2:e001606.
42. Levitt JE, Calfee CS, Goldstein BA, et al. Early acute lung injury: criteria for identifying lung injury prior to the need for positive pressure ventilation*. Crit Care Med 2013; 41:1929–1937.
43. Kor DJ, Lingineni RK, Gajic O, et al. Predicting risk of postoperative lung injury in high-risk surgical patients: a multicenter cohort study. Anesthesiology 2014; 120:1168–1181.
44. Calfee CS, Delucchi K, Parsons PE, et al. Subphenotypes in acute respiratory distress syndrome: latent class analysis of data from two randomised controlled trials. Lancet Respir Med 2014; 2:611–620.
45. Hajjar LA, Vincent JL, Galas FR, et al. Transfusion requirements after cardiac surgery: the TRACS randomized controlled trial. JAMA 2010; 304:1559–1567.
46. Watkins TR, Rubenfeld GD, Martin TR, et al. Effects of leukoreduced blood on acute lung injury after trauma: a randomized controlled trial. Crit Care Med 2008; 36:1493–1499.
47. Herring AA, Ginde AA, Fahimi J, et al. Increasing critical care admissions from U.S. emergency departments. Crit Care Med 2013; 41:1197–1204.
48. Rohde JM, Odden AJ, Bonham C, et al. The epidemiology of acute organ system dysfunction from severe sepsis outside of the intensive care unit. J Hosp Med 2013; 8:243–247.
49. Chow SC. Adaptive clinical trial design. Ann Rev Med 2014; 65:405–415.
50. Patsopoulos NA. A pragmatic view on pragmatic trials. Dialogues Clin Neurosci 2011; 13:217–224.
51. Treweek S, Zwarenstein M. Making trials matter: pragmatic and explanatory trials and the problem of applicability. Trials 2009; 10:37.
52. Thorpe KE, Zwarenstein M, Oxman AD, et al. A pragmatic-explanatory continuum indicator summary (PRECIS): a tool to help trial designers. CMAJ 2009; 180:E47–E57.
53. Noto MJ, Domenico HJ, Byrne DW, et al. Chlorhexidine bathing and healthcare-associated infections: a randomized clinical trial. JAMA 2015; 313:369–378.
54. Christie J, O’Halloran P, Stevenson M. Planning a cluster randomized controlled trial: methodological issues. Nurs Res 2009; 58:128–134.
55. Li G, Malinchoc M, Cartin-Ceba R, et al. Eight-year trend of acute respiratory distress syndrome: a population-based study in Olmsted County, Minnesota. Am J Respir Crit Care Med 2011; 183:59–66.
56. Schoenfeld DA, Bernard GR, Network A. Statistical evaluation of ventilator-free days as an efficacy measure in clinical trials of treatments for acute respiratory distress syndrome. Crit Care Med 2002; 30:1772–1777.
57. Heyland DK, Muscedere J, Drover J, et al. Persistent organ dysfunction plus death: a novel, composite outcome measure for critical care trials. Crit Care 2011; 15:R98.
58. Hopkins RO, Weaver LK, Collingridge D, et al. Two-year cognitive, emotional, and quality-of-life outcomes in acute respiratory distress syndrome. Am J Respir Crit Care Med 2005; 171:340–347.
59. Herridge MS, Cheung AM, Tansey CM, et al. One-year outcomes in survivors of the acute respiratory distress syndrome. N Engl J Med 2003; 348:683–693.
60. Herridge MS, Tansey CM, Matte A, et al. Functional disability 5 years after acute respiratory distress syndrome. N Engl J Med 2011; 364:1293–1304.