Subscribe for unlimited access to DynaMed content, CME/CE & MOC credit, and email alerts on content you follow.

Already subscribed? Sign in now


Acute Respiratory Distress Syndrome (ARDS)

General Information


  • clinical syndrome of lung injury marked by acute onset with hypoxemia and bilateral radiographic infiltrates and without heart failure2,3

Also called

  • respiratory distress syndrome
  • adult respiratory distress syndrome
  • acute lung injury


  • Berlin definition of ARDS (preferred over 1994 American-European Consensus Conference definition; 1994 acute lung injury definition replaced by 3 severities of ARDS)
    • onset within 1 week of known clinical insult or new or worsening respiratory symptoms
    • bilateral opacities not fully explained by effusions, lobar/lung collapse, or nodules on chest x-ray or computed tomography
    • respiratory failure not fully explained by cardiac failure or fluid overload (in the absence of risk factors for ARDS, an objective assessment such as echocardiography is required to exclude these causes of hydrostatic edema)
    • oxygenation status
      • mild ARDS defined as partial pressure of oxygen in arterial blood (PaO2) to fraction of inspired oxygen (FiO2) > 200 mm Hg but ≤ 300 mm Hg with positive end-expiratory pressure (PEEP) or continuous positive airway pressure (CPAP) ≥ 5 cm H2O
      • moderate ARDS defined as PaO2/FiO2 > 100 mm Hg but ≤ 200 mm Hg with PEEP ≥ 5 cm H2O
      • severe ARDS defined as PaO2/FiO2 ≤ 100 mm Hg with PEEP ≥ 5 cm H2O
      • if altitude > 1,000 meters, correction factor is PaO2/FiO2 x (barometric pressure/760)
    • Reference - mdc22797452pJAMA 2012 Jun 20;307(23):2526, editorial can be found in mdc22797455pJAMA 2012 Jun 20;307(23):2542, commentary can be found in mdc23032539pJAMA 2012 Oct 3;308(13):1321
  • ARDS defined as moderate to severe form of acute lung injury (ALI) in older criteria from 1994 American-European Consensus Conference3
    • ALI defined as PaO2/FiO2 < 300 mm Hg
    • ARDS defined as PaO2/FiO2 < 200 mm Hg (measurement of PaO2 on PEEP ≥ 5 cmH2O not required)


General references used

  1. Prabhakaran P. Acute respiratory distress syndrome. Indian Pediatr. 2010 Oct;47(10):861PDF
  2. Saguil A, Fargo M. Acute respiratory distress syndrome: diagnosis and management. Am Fam Physician. 2012 Feb 15;85(4):352-8full-text
  3. Sweeney RM, McAuley DF. Acute respiratory distress syndrome. Lancet. 2016 Nov 12;388(10058):2416-2430

Recommendation grading systems used

  • American Thoracic Society/European Society of Intensive Care Medicine/Society of Critical Care Medicine (ATS/ESICM/SCCM) grading system
    • strength of recommendations
      • Strong recommendation
        • most patients should receive the recommended course of action
        • most patients would want the intervention, only a small proportion would not
      • Conditional/Weak recommendation
        • clinicians should be more prepared to help patients make decisions consistent with patient's values
        • most patients would want the intervention, but many would not
    • quality of evidence determination and implications/confidence in estimate of effect
      • High
        • randomized trial without limitation in study quality, indirectness, important inconsistency, sparse or imprecise data, or high probability of publication bias
        • further research is very unlikely to change confidence in estimate of effect
      • Moderate
        • downgraded randomized trial or upgraded observational study with strong association (no plausible confounders), evidence of a dose-response gradient, or plausible confounders which would have reduced the effect
        • further research is likely to have important impact on confidence in estimate of effect and may change estimate
      • Low
        • well-done observational study with control groups
        • further research is very likely to have important impact on confidence in estimate of effect and is likely to change estimate
      • Very low
        • any other evidence (for example, case reports, case series)
        • very uncertain about estimate
  • Society of Critical Care Medicine/American Society for Parenteral and Enteral Nutrition (SCCM/ASPEN) uses Grading of Recommendations, Assessment, Development, and Evaluation (GRADE) system
    • levels of evidence
      • High-quality evidence - randomized controlled trials (RCTs) without factors that reduce quality of evidence, or well-done observational studies with very large magnitude of effect
      • Moderate-quality evidence - RCT with important inconsistency (heterogeneity across studies, as I2 > 0.5 or some say yes but others say no)
      • Low-quality evidence
        • RCT with some or major uncertainty about directness (outcome variable not direct measure of process)
        • observational study with significant relative risk of > 2 (< 0.5) based on consistent evidence from ≥ 2 observational studies with no plausible confounders
      • Very low-quality evidence
        • RCT with imprecise of sparse data (combined effect size not significant, small number of patients)
        • RCT with high probability of reporting bias
        • observational study with significant relative risk of > 5 (< 0.2) based on direct evidence with no major threats to validity
        • observational study with evidence of dose-response gradient
      • Good Practice Statement - ungraded
    • Reference - SCCM/ASPEN guideline on provision and assessment of nutrition support therapy in the adult critically ill patient (26773077JPEN J Parenter Enteral Nutr 2016 Feb;40(2):159), correction can be found in JPEN J Parenter Enteral Nutr 2016 Nov;40(8):1200, commentary can be found in JPEN J Parenter Enteral Nutr 2016 Nov;40(8):1197
  • Society of Critical Care Medicine/European Society of Intensive Care Medicine (SCCM/ESICM) uses Grading of Recommendations, Assessment, Development, and Evaluation (GRADE)
    • strength of recommendation
      • Strong - benefits clearly outweigh risks and burdens (or vice versa) for most, if not all, patients
      • Conditional - benefits and risks closely balanced and/or uncertain
      • Best practice statement - recommendation not conducive to GRADE process
    • quality of evidence
      • High - randomized trials without factors that reduce quality of evidence, or well done observational studies with very large magnitude of effect
      • Moderate - downgraded randomized trials or upgraded observational studies
      • Low - well done observational studies, or randomized trials with many limitations
      • Very low - downgraded observational studies (results from individual studies at serious or very serious risk of bias, serious inconsistencies in results across studies, indirect evidence, sparse or imprecise data, or likely publication bias)
    • Reference - SCCM/ESICM guideline on diagnosis and management of critical illness-related corticosteroid insufficiency (CIRCI) in critically ill patients (Part I) (28938253Crit Care Med 2017 Dec;45(12):2078), also published in 28940011Intensive Care Med 2017 Dec;43(12):1751, correction can be found in 29476199Intensive Care Med 2018 Mar;44(3):401
  • American College of Critical Care Medicine/Society of Critical Care Medicine (ACCCM/SCCM) uses Grading of Recommendations Assessment, Development, and Evaluation (GRADE) system
    • strengths of recommendation
      • Strong - we recommend
      • Weak - we suggest
      • Good practice statement - no clear alternative exists and consensus among Task Force members
    • levels of evidence
      • High-quality evidence - further research unlikely to change confidence in estimate of effect
      • Moderate-quality evidence - further research likely to have impact on confidence in estimate of effect
      • Low-quality evidence - further research very likely to have important impact on confidence in estimate of effect and is likely to change estimate
      • Very low-quality evidence - estimate of effect very uncertain
    • Reference - ACCCM/SCCM clinical practice guideline on sustained neuromuscular blockage in adult critically ill patient (27755068Crit Care Med 2016 Nov;44(11):2079)
  • American College of Chest Physicians (ACCP) grades
    • Grade 1 - strong recommendation based on clear risk/benefit balance
    • Grade 2 - weak recommendation based on unclear or close risk/benefit balance
    • Grade A - high-quality evidence based on consistent evidence from randomized trials without important limitations or exceptionally strong evidence from observational studies
    • Grade B - moderate-quality evidence based on randomized trials with important limitations (inconsistent results, methodologic flaws, indirect or imprecise results) or very strong evidence from observational studies
    • Grade C - low- or very low-quality evidence based on evidence for ≥ 1 critical outcome from observational studies, case series, or randomized trials with serious flaws or indirect evidence
    • Reference - ACCP evidence-based clinical practice guideline on methodology for development of antithrombotic therapy and prevention of thrombosis (22315256Chest 2012 Feb;141(2 Suppl):53Sfull-text), commentary can be found in 23546508Chest 2013 Apr;143(4):1190

Synthesized Recommendation Grading System for DynaMed Content

  • The DynaMed Team systematically monitors clinical evidence to continuously provide a synthesis of the most valid relevant evidence to support clinical decision-making (see 7-Step Evidence-Based Methodology).
  • Guideline recommendations summarized in the body of a DynaMed topic are provided with the recommendation grading system used in the original guideline(s), and allow users to quickly see where guidelines agree and where guidelines differ from each other and from the current evidence.
  • In DynaMed content, we synthesize the current evidence, current guidelines from leading authorities, and clinical expertise to provide recommendations to support clinical decision-making in the Overview & Recommendations section.
  • We use the Grading of Recommendations Assessment, Development and Evaluation (GRADE) to classify synthesized recommendations as Strong or Weak.
    • Strong recommendations are used when, based on the available evidence, clinicians (without conflicts of interest) consistently have a high degree of confidence that the desirable consequences (health benefits, decreased costs and burdens) outweigh the undesirable consequences (harms, costs, burdens).
    • Weak recommendations are used when, based on the available evidence, clinicians believe that desirable and undesirable consequences are finely balanced, or appreciable uncertainty exists about the magnitude of expected consequences (benefits and harms). Weak recommendations are used when clinicians disagree in judgments of relative benefit and harm, or have limited confidence in their judgments. Weak recommendations are also used when the range of patient values and preferences suggests that informed patients are likely to make different choices.
  • DynaMed synthesized recommendations (in the Overview & Recommendations section) are determined with a systematic methodology:
    • Recommendations are initially drafted by clinical editors (including ≥ 1 with methodological expertise and ≥ 1 with content domain expertise) aware of the best current evidence for benefits and harms, and the recommendations from guidelines.
    • Recommendations are phrased to match the strength of recommendation. Strong recommendations use "should do" phrasing, or phrasing implying an expectation to perform the recommended action for most patients. Weak recommendations use "consider" or "suggested" phrasing.
    • Recommendations are explicitly labeled as Strong recommendations or Weak recommendations when a qualified group has explicitly deliberated on making such a recommendation. Group deliberation may occur during guideline development. When group deliberation occurs through DynaMed Team-initiated groups:
      • Clinical questions will be formulated using the PICO (Population, Intervention, Comparison, Outcome) framework for all outcomes of interest specific to the recommendation to be developed.
      • Systematic searches will be conducted for any clinical questions where systematic searches were not already completed through DynaMed content development.
      • Evidence will be summarized for recommendation panel review including for each outcome, the relative importance of the outcome, the estimated effects comparing intervention and comparison, the sample size, and the overall quality rating for the body of evidence.
      • Recommendation panel members will be selected to include at least 3 members that together have sufficient clinical expertise for the subject(s) pertinent to the recommendation, methodological expertise for the evidence being considered, and experience with guideline development.
      • All recommendation panel members must disclose any potential conflicts of interest (professional, intellectual, and financial), and will not be included for the specific panel if a significant conflict exists for the recommendation in question.
      • Panel members will make Strong recommendations if and only if there is consistent agreement in a high confidence in the likelihood that desirable consequences outweigh undesirable consequences across the majority of expected patient values and preferences. Panel members will make Weak recommendations if there is limited confidence (or inconsistent assessment or dissenting opinions) that desirable consequences outweigh undesirable consequences across the majority of expected patient values and preferences. No recommendation will be made if there is insufficient confidence to make a recommendation.
      • All steps in this process (including evidence summaries which were shared with the panel, and identification of panel members) will be transparent and accessible in support of the recommendation.
    • Recommendations are verified by ≥ 1 editor with methodological expertise, not involved in recommendation drafting or development, with explicit confirmation that Strong recommendations are adequately supported.
    • Recommendations are published only after consensus is established with agreement in phrasing and strength of recommendation by all editors.
    • If consensus cannot be reached then the recommendation can be published with a notation of "dissenting commentary" and the dissenting commentary is included in the topic details.
    • If recommendations are questioned during peer review or post publication by a qualified individual, or reevaluation is warranted based on new information detected through systematic literature surveillance, the recommendation is subject to additional internal review.

DynaMed Editorial Process

  • DynaMed topics are created and maintained by the DynaMed Editorial Team and Process.
  • All editorial team members and reviewers have declared that they have no financial or other competing interests related to this topic, unless otherwise indicated.
  • DynaMed content includes Practice-Changing Updates, with support from our partners, McMaster University and F1000.

Special acknowledgements

On behalf of the American College of Physicians
  • Barbara Turner, MD, MSEd, MACP, ACP Deputy Editor, Clinical Decision Resource, as part of the ACP-EBSCO Health collaboration, managed the ACP peer review of the Overview and Recommendations section and related clinical content in this topic.
  • DynaMed topics are written and edited through the collaborative efforts of the above individuals. Deputy Editors, Section Editors, and Topic Editors are active in clinical or academic medical practice. Recommendations Editors are actively involved in development and/or evaluation of guidelines.
  • Editorial Team role definitions
    Topic Editors define the scope and focus of each topic by formulating a set of clinical questions and suggesting important guidelines, clinical trials, and other data to be addressed within each topic. Topic Editors also serve as consultants for the internal DynaMed Editorial Team during the writing and editing process, and review the final topic drafts prior to publication.
    Section Editors have similar responsibilities to Topic Editors but have a broader role that includes the review of multiple topics, oversight of Topic Editors, and systematic surveillance of the medical literature.
    Recommendations Editors provide explicit review of Overview and Recommendations sections to ensure that all recommendations are sound, supported, and evidence-based. This process is described in "Synthesized Recommendation Grading."
    Deputy Editors oversee DynaMed internal publishing groups. Each is responsible for all content published within that group, including supervising topic development at all stages of the writing and editing process, final review of all topics prior to publication, and direction of an internal team.

How to cite

National Library of Medicine, or "Vancouver style" (International Committee of Medical Journal Editors):

  • DynaMed [Internet]. Ipswich (MA): EBSCO Information Services. 1995 - . Record No. T113803, Acute Respiratory Distress Syndrome (ARDS); [updated 2018 Dec 04, cited place cited date here]. Available from Registration and login required.

Published by EBSCO Information Services. Copyright © 2020, EBSCO Information Services. All rights reserved. No part of this may be reproduced or utilized in any form or by any means, electronic or mechanical, including photocopying, recording, or by any information storage and retrieval system, without permission.

EBSCO Information Services accepts no liability for advice or information given herein or errors/omissions in the text. It is merely intended as a general informational overview of the subject for the healthcare professional.


Subscribe for unlimited access to DynaMed content.
Already subscribed? Sign in