[Skip to Content]
[Skip to Content Landing]

Extracorporeal Membrane Oxygenation for Severe Acute Respiratory Distress Syndrome and Posterior Probability of Mortality Benefit in a Post Hoc Bayesian Analysis of a Randomized Clinical Trial

Educational Objective
To understand differences in the frequentist and Bayesian approaches to data analysis.
1 Credit CME
Key Points

Question  Can Bayesian analysis clarify the interpretation of clinical trial results?

Findings  In a post hoc Bayesian analysis of the recent EOLIA (Extracorporeal Membrane Oxygenation [ECMO] to Rescue Lung Injury in Severe ARDS) trial, the posterior probability of mortality benefit (relative risk <1) ranged between 88% and 99% given a range of prior assumptions reflecting varying degrees of skepticism and enthusiasm regarding previous evidence for the benefit of ECMO. Probabilities varied according to the definition of minimum clinically important mortality benefit; for example, the posterior probability of relative risk less than 0.67 ranged between 0% and 48% given the same range of prior assumptions.

Meaning  Information about the posterior probability of treatment effect provided by Bayesian analysis may help clarify the interpretation of clinical trial findings.

Abstract

Importance  Bayesian analysis of clinical trial data may provide useful information to aid in study interpretation, especially when trial evidence suggests that the benefits of an intervention are uncertain, such as in a trial that evaluated early extracorporeal membrane oxygenation (ECMO) for severe acute respiratory distress syndrome (ARDS).

Objective  To demonstrate the potential utility of Bayesian analyses by estimating the posterior probability, under various assumptions, that early ECMO was associated with reduced mortality in patients with very severe ARDS in a randomized clinical trial (RCT).

Design and Evidence  A post hoc Bayesian analysis of data from an RCT (ECMO to Rescue Lung Injury in Severe ARDS [EOLIA]) that included 249 patients with very severe ARDS who had been randomized to receive early ECMO (n = 124; mortality at 60 days, 35%) vs initial conventional lung-protective ventilation with the option for rescue ECMO (n = 125, mortality at 60 days, 46%). The trial was designed to detect an absolute risk reduction (ARR) of 20%, relative risk (RR) of 0.67. Statistical prior distributions were specified to represent varying levels of preexisting enthusiasm or skepticism for ECMO and by Bayesian meta-analysis of previously published studies (with downweighting to account for differences and quality between studies). The RR, credible interval (CrI), ARR, and probability of clinically important mortality benefit (varying from RR less than 1 to RR less than 0.67 and ARR from 2% or more to 20% or more) were estimated with Bayesian modeling.

Findings  Combining a minimally informative prior distribution with the findings of the EOLIA trial, the posterior probability of RR less than 1 for mortality at 60 days after randomization was 96% (RR, 0.78 [95% CrI, 0.56-1.04]); the posterior probability of RR less than 0.67 was 18%, the probability of ARR of 2% or more was 92%, and the probability of ARR of 20% or more was 2%. With a moderately enthusiastic prior, equivalent to information from a trial of 264 patients with an RR of 0.78, the estimated RR was 0.78 (95% CrI, 0.63-0.96), the probability of RR less than 1 was 99%, the probability of RR less than 0.67 was 8%, the probability of ARR of 2% or more was 97%, and the probability of ARR of 20% or more was 0%. With a strongly skeptical prior, equivalent to information from a trial of 264 patients with an RR of 1.0, the estimated RR was 0.88 (95% CrI, 0.71-1.09), the probability of RR less than 1 was 88%, the probability of RR less than 0.67 was 0%, the probability of ARR of 2% or more was 78%, and the probability of ARR of 20% or more was 0%. If the prior was informed by previous studies, the estimated RR was 0.71 (95% CrI, 0.55-0.94), the probability of RR less than 1 was 99%, the probability of RR less than 0.67 was 48%, the probability of ARR of 2% or more was 98%, and the probability of ARR of 20% or more was 4%.

Conclusions and Relevance  Post hoc Bayesian analysis of data from a randomized clinical trial of early extracorporeal membrane oxygenation compared with conventional lung-protective ventilation with the option for rescue extracorporeal membrane oxygenation among patients with very severe acute respiratory distress syndrome provides information about the posterior probability of mortality benefit under a broad set of assumptions that may help inform interpretation of the study findings.

Sign in to take quiz and track your certificates

Buy This Activity

JN Learning™ is the home for CME and MOC from the JAMA Network. Search by specialty or US state and earn AMA PRA Category 1 CME Credit™ from articles, audio, Clinical Challenges and more. Learn more about CME/MOC

Article Information

Corresponding Author: Ewan C. Goligher, MD, PhD, Toronto General Hospital, 585 University Ave, Peter Munk Bldg, 11th Floor, Room 192, Toronto, ON M5G 2N2, Canada (ewan.goligher@utoronto.ca).

Accepted for Publication: September 26, 2018.

Published Online: October 22, 2018. doi:10.1001/jama.2018.14276

Correction: This article was corrected on November 14, 2018, to correct a degree in the author byline and to add a conflict of interest disclosure that was inadvertently omitted. It was corrected on June 11, 2019, to provide the prior probabilities for RR <0.67 in Table 1 and update an absolute risk reduction value in Table 3.

Author Contributions: Dr Goligher had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Drs Tomlinson and Goligher conducted and are responsible for the data analysis.

Concept and design: Goligher, Tomlinson, Wijeysundera, Jüni, Brodie, Slutsky, Combes.

Acquisition, analysis, or interpretation of data: All authors.

Drafting of the manuscript: Goligher.

Critical revision of the manuscript for important intellectual content: All authors.

Statistical analysis: Goligher, Tomlinson, Hajage.

Administrative, technical, or material support: Slutsky, Combes.

Conflict of Interest Disclosures: Dr Goligher reports receiving travel reimbursement and speaking honoraria from Getinge outside the submitted work. Dr Jüni reports being a tier 1 Canada research chair in clinical epidemiology of chronic diseases. Dr Brodie reports serving as the cochair of the trial steering committee for the VENT-AVOID trial sponsored by ALung Technologies; serving on the medical advisory board for Baxter; and previously serving on the medical advisory boards of ALung Technologies and Kadence (Johnson & Johnson), with all compensation for these activities paid to Columbia University. Dr Slutsky reports serving as a paid consultant for Maquet Critical Care, Baxter, and Novalung/Xenios. Dr Combes reports receiving study grant support from Maquet, lecture fees from Maquet and Baxter, and consulting fees from Hemovent, outside the submitted work. No other disclosures were reported.

Funding/Support: This work is supported by a new investigator award from the Canadian Institutes of Health Research (Drs Fan and Wijeysundera) and a merit award from the department of anesthesia at the University of Toronto (Dr Wijeysundera).

Role of the Funder/Sponsor: The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

References
1.
Spiegelhalter  DJ, Myles  JP, Jones  DR, Abrams  KR.  Methods in health service research: an introduction to Bayesian methods in health technology assessment.  BMJ. 1999;319(7208):508-512. doi:10.1136/bmj.319.7208.508PubMedGoogle ScholarCrossref
2.
Bland  JM, Altman  DG.  Bayesians and frequentists.  BMJ. 1998;317(7166):1151-1160. doi:10.1136/bmj.317.7166.1151PubMedGoogle ScholarCrossref
3.
Wijeysundera  DN, Austin  PC, Hux  JE, Beattie  WS, Laupacis  A.  Bayesian statistical inference enhances the interpretation of contemporary randomized controlled trials.  J Clin Epidemiol. 2009;62(1):13-21. doi:10.1016/j.jclinepi.2008.07.006PubMedGoogle ScholarCrossref
4.
Spiegelhalter  DJ, Freedman  LS, Parmar  MK.  Applying Bayesian ideas in drug development and clinical trials.  Stat Med. 1993;12(15-16):1501-1511. doi:10.1002/sim.4780121516PubMedGoogle ScholarCrossref
5.
Brophy  JM, Joseph  L.  Placing trials in context using Bayesian analysis: GUSTO revisited by Reverend Bayes.  JAMA. 1995;273(11):871-875. doi:10.1001/jama.1995.03520350053028PubMedGoogle ScholarCrossref
6.
Quintana  M, Viele  K, Lewis  RJ.  Bayesian analysis: using prior information to interpret the results of clinical trials.  JAMA. 2017;318(16):1605-1606. doi:10.1001/jama.2017.15574PubMedGoogle ScholarCrossref
7.
Combes  A, Hajage  D, Capellier  G,  et al; EOLIA Trial Group, REVA, and ECMONet.  Extracorporeal membrane oxygenation for severe acute respiratory distress syndrome.  N Engl J Med. 2018;378(21):1965-1975. doi:10.1056/NEJMoa1800385PubMedGoogle ScholarCrossref
8.
Vincent  J-L, Brochard  LJ.  Do we need randomized clinical trials in extracorporeal respiratory support? we are not sure.  Intensive Care Med. 2017;43(12):1869-1871. doi:10.1007/s00134-017-4930-xPubMedGoogle ScholarCrossref
9.
Dalton  HJ, MacLaren  G.  Extracorporeal membrane oxygenation in pandemic flu: insufficient evidence or worth the effort?  Crit Care Med. 2010;38(6):1484-1485. doi:10.1097/CCM.0b013e3181e08fffPubMedGoogle ScholarCrossref
10.
Gattinoni  L, Vasques  F, Quintel  M.  Use of ECMO in ARDS: does the EOLIA trial really help?  Crit Care. 2018;22(1):171. doi:10.1186/s13054-018-2098-6PubMedGoogle ScholarCrossref
11.
Hardin  CC, Hibbert  K.  ECMO for severe ARDS.  N Engl J Med. 2018;378(21):2032-2034. doi:10.1056/NEJMe1802676PubMedGoogle ScholarCrossref
12.
Mi  MY, Matthay  MA, Morris  AH.  Extracorporeal membrane oxygenation for severe acute respiratory distress syndrome.  N Engl J Med. 2018;379(9):884-887. doi:10.1056/NEJMclde1804601PubMedGoogle ScholarCrossref
13.
United States Census Bureau. US and world population clock. https://www.census.gov/popclock/. Accessed September 10, 2018.
14.
Rubenfeld  GD, Caldwell  E, Peabody  E,  et al.  Incidence and outcomes of acute lung injury.  N Engl J Med. 2005;353(16):1685-1693. doi:10.1056/NEJMoa050333PubMedGoogle ScholarCrossref
15.
Bellani  G, Laffey  JG, Pham  T,  et al; LUNG SAFE Investigators; ESICM Trials Group.  Epidemiology, patterns of care, and mortality for patients with acute respiratory distress syndrome in intensive care units in 50 countries.  JAMA. 2016;315(8):788-800. doi:10.1001/jama.2016.0291PubMedGoogle ScholarCrossref
16.
Li  X, Scales  DC, Kavanagh  BP.  Unproven and expensive before proven and cheap: extracorporeal membrane oxygenation vs prone position in acute respiratory distress syndrome.  Am J Respir Crit Care Med. 2018;197(8):991-993. doi:10.1164/rccm.201711-2216CPPubMedGoogle ScholarCrossref
17.
Brower  RG, Matthay  MA, Morris  A, Schoenfeld  D, Thompson  BT, Wheeler  A; Acute Respiratory Distress Syndrome Network.  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;342(18):1301-1308. doi:10.1056/NEJM200005043421801PubMedGoogle ScholarCrossref
18.
Peek  GJ, Mugford  M, Tiruvoipati  R,  et al; CESAR trial collaboration.  Efficacy and economic assessment of conventional ventilatory support vs extracorporeal membrane oxygenation for severe adult respiratory failure (CESAR): a multicentre randomised controlled trial.  Lancet. 2009;374(9698):1351-1363. doi:10.1016/S0140-6736(09)61069-2PubMedGoogle ScholarCrossref
19.
Pham  T, Combes  A, Rozé  H,  et al; REVA Research Network.  Extracorporeal membrane oxygenation for pandemic influenza A(H1N1)-induced acute respiratory distress syndrome: a cohort study and propensity-matched analysis.  Am J Respir Crit Care Med. 2013;187(3):276-285. doi:10.1164/rccm.201205-0815OCPubMedGoogle ScholarCrossref
20.
Noah  MA, Peek  GJ, Finney  SJ,  et al.  Referral to an extracorporeal membrane oxygenation center and mortality among patients with severe 2009 influenza A(H1N1).  JAMA. 2011;306(15):1659-1668. doi:10.1001/jama.2011.1471PubMedGoogle ScholarCrossref
21.
Munshi  L, Telesnicki  T, Walkey  A, Fan  E.  Extracorporeal life support for acute respiratory failure: a systematic review and meta-analysis.  Ann Am Thorac Soc. 2014;11(5):802-810. doi:10.1513/AnnalsATS.201401-012OCPubMedGoogle ScholarCrossref
22.
Su  YS, Yajima  M. R2jags: using R to run “JAGS”: R package version 0.5-7. https://cran.r-project.org/web/packages/R2jags/index.html.
23.
Plummer  M. JAGS. 2004. http://mcmc-jags.sourceforge.net.
24.
Allmark  P.  Bayes and health care research.  Med Health Care Philos. 2004;7(3):321-332. doi:10.1007/s11019-004-0804-4PubMedGoogle ScholarCrossref
25.
US Food and Drug Administration. Guidance for the Use of Bayesian Statistics in Medical Device Clinical Trials. https://www.fda.gov/MedicalDevices/ucm071072.htm. Accessed September 10, 2018.
26.
Harrington  D, Drazen  JM.  Learning from a trial stopped by a data and safety monitoring board.  N Engl J Med. 2018;378(21):2031-2032. doi:10.1056/NEJMe1805123PubMedGoogle ScholarCrossref
27.
Kass  RE, Greenhouse  JB.  [Investigating therapies of potentially great benefit: ECMO]: comment: a Bayesian perspective.  Stat Sci. 1989;4(4):310-317. doi:10.1214/ss/1177012386Google ScholarCrossref
28.
Lilford  RJ.  Ethics of clinical trials from a Bayesian and decision analytic perspective: whose equipoise is it anyway?  BMJ. 2003;326(7396):980-981. doi:10.1136/bmj.326.7396.980PubMedGoogle ScholarCrossref
29.
Fan  E, Del Sorbo  L, Goligher  EC,  et al.  An official American Thoracic Society/European Society of Intensive Care Medicine/Society of Critical Care Medicine clinical practice guideline: mechanical ventilation in adult patients with acute respiratory distress syndrome.  Am J Respir Crit Care Med. 2017;195(9):1253-1263. doi:10.1164/rccm.201703-0548STPubMedGoogle ScholarCrossref
30.
Head  ML, Holman  L, Lanfear  R, Kahn  AT, Jennions  MD.  The extent and consequences of P hacking in science.  PLoS Biol. 2015;13(3):e1002106. doi:10.1371/journal.pbio.1002106PubMedGoogle ScholarCrossref
31.
Berry  DA.  Bayesian clinical trials.  Nat Rev Drug Discov. 2006;5(1):27-36. doi:10.1038/nrd1927PubMedGoogle ScholarCrossref
32.
Johnson  SR, Tomlinson  GA, Hawker  GA, Granton  JT, Feldman  BM.  Methods to elicit beliefs for Bayesian priors: a systematic review.  J Clin Epidemiol. 2010;63(4):355-369. doi:10.1016/j.jclinepi.2009.06.003PubMedGoogle ScholarCrossref
If you are not a JN Learning subscriber, you can either:
Subscribe to JN Learning for one year
Buy this activity
jn-learning_Modal_LoginSubscribe_Purchase
If you are not a JN Learning subscriber, you can either:
Subscribe to JN Learning for one year
Buy this activity
jn-learning_Modal_LoginSubscribe_Purchase
With a personal account, you can:
  • Access free activities and track your credits
  • Personalize content alerts
  • Customize your interests
  • Fully personalize your learning experience
Education Center Collection Sign In Modal Right

Name Your Search

Save Search
With a personal account, you can:
  • Track your credits
  • Personalize content alerts
  • Customize your interests
  • Fully personalize your learning experience
jn-learning_Modal_SaveSearch_NoAccess_Purchase

Lookup An Activity

or

My Saved Searches

You currently have no searches saved.

With a personal account, you can:
  • Access free activities and track your credits
  • Personalize content alerts
  • Customize your interests
  • Fully personalize your learning experience
Education Center Collection Sign In Modal Right
Topics
State Requirements