Ventilator Weaning and Discontinuation Practices for Critically Ill Patients | Critical Care Medicine | JAMA | JAMA Network
[Skip to Navigation]
Sign In
Figure.  Patient Flow in a Study of Ventilator Weaning and Discontinuation Practices for Critically Ill Patients
Patient Flow in a Study of Ventilator Weaning and Discontinuation Practices for Critically Ill Patients

aFive participants met 2 exclusion criteria (3 had no intubation time and had a tracheotomy at ICU admission, 1 had a tracheotomy at ICU admission and was readmitted to the ICU, and 1 had no intubation time and was already receiving a spontaneous breathing trial setting).

bData for 1 participant who died before undergoing a discontinuation event were missing.

Table 1.  Patient Characteristics in a Study of Ventilator Weaning and Discontinuation Practices in Critically Ill Patients
Patient Characteristics in a Study of Ventilator Weaning and Discontinuation Practices in Critically Ill Patients
Table 2.  Characteristics of Participating ICUs in a Study of Ventilator Weaning and Discontinuation Practices for Critically Ill Patients (N = 142)
Characteristics of Participating ICUs in a Study of Ventilator Weaning and Discontinuation Practices for Critically Ill Patients (N = 142)
Table 3.  Initial Mechanical Ventilation Discontinuation Strategy and Clinical Outcomes Among Critically Ill Patients
Initial Mechanical Ventilation Discontinuation Strategy and Clinical Outcomes Among Critically Ill Patients
Table 4.  Outcome Based on Success or Failure and Timing of Initial Spontaneous Breathing Trials Among Critically Ill Patients
Outcome Based on Success or Failure and Timing of Initial Spontaneous Breathing Trials Among Critically Ill Patients
1.
Wunsch  H, Linde-Zwirble  WT, Angus  DC, Hartman  ME, Milbrandt  EB, Kahn  JM.  The epidemiology of mechanical ventilation use in the United States.   Crit Care Med. 2010;38(10):1947-1953. doi:10.1097/CCM.0b013e3181ef4460PubMedGoogle ScholarCrossref
2.
Carson  SS, Cox  CE, Holmes  GM, Howard  A, Carey  TS.  The changing epidemiology of mechanical ventilation: a population-based study.   J Intensive Care Med. 2006;21(3):173-182. doi:10.1177/0885066605282784PubMedGoogle ScholarCrossref
3.
Needham  DM, Bronskill  SE, Calinawan  JR, Sibbald  WJ, Pronovost  PJ, Laupacis  A.  Projected incidence of mechanical ventilation in Ontario to 2026: preparing for the aging baby boomers.   Crit Care Med. 2005;33(3):574-579. doi:10.1097/01.CCM.0000155992.21174.31PubMedGoogle ScholarCrossref
4.
Wunsch  H, Wagner  J, Herlim  M, Chong  DH, Kramer  AA, Halpern  SD.  ICU occupancy and mechanical ventilator use in the United States.   Crit Care Med. 2013;41(12):2712-2719. doi:10.1097/CCM.0b013e318298a139PubMedGoogle ScholarCrossref
5.
Melsen  WG, Rovers  MM, Groenwold  RHH,  et al.  Attributable mortality of ventilator-associated pneumonia: a meta-analysis of individual patient data from randomised prevention studies.   Lancet Infect Dis. 2013;13(8):665-671. doi:10.1016/S1473-3099(13)70081-1PubMedGoogle ScholarCrossref
6.
Rengel  KF, Hayhurst  CJ, Pandharipande  PP, Hughes  CG.  Long-term cognitive and functional impairments after critical illness.   Anesth Analg. 2019;128(4):772-780. doi:10.1213/ANE.0000000000004066PubMedGoogle ScholarCrossref
7.
MacIntyre  NR, Cook  DJ, Ely  EW  Jr,  et al; American College of Chest Physicians; American Association for Respiratory Care; American College of Critical Care Medicine.  Evidence-based guidelines for weaning and discontinuing ventilatory support: a collective task force facilitated by the American College of Chest Physicians; the American Association for Respiratory Care; and the American College of Critical Care Medicine.   Chest. 2001;120(6)(suppl):375S-395S. doi:10.1378/chest.120.6_suppl.375SPubMedGoogle ScholarCrossref
8.
Blackwood  B, Burns  K, Cardwell  C, O’Halloran  P.  Protocolized vs. non-protocolized weaning for reducing the duration of mechanical ventilation in critically ill adult patients Cochrane systematic review and meta-analysis.   Cochrane Database Syst Rev. 2014;2014(11):CD006904. doi:10.1002/14651858.CD006904.pub3PubMedGoogle Scholar
9.
Burns  KEA, Soliman  I, Adhikari  NKJ,  et al.  Trials directly comparing alternative spontaneous breathing trial techniques: a systematic review and meta-analysis.   Crit Care. 2017;21(1):127. doi:10.1186/s13054-017-1698-xPubMedGoogle ScholarCrossref
10.
Subirà  C, Hernández  G, Vázquez  A,  et al.  Effect of pressure support vs t-piece ventilation strategies during spontaneous breathing trials on successful extubation among patients receiving mechanical ventilation: a randomized clinical trial.   JAMA. 2019;321(22):2175-2182. doi:10.1001/jama.2019.7234PubMedGoogle ScholarCrossref
11.
Esteban  A, Frutos  F, Tobin  MJ,  et al; Spanish Lung Failure Collaborative Group.  A comparison of four methods of weaning patients from mechanical ventilation.   N Engl J Med. 1995;332(6):345-350. doi:10.1056/NEJM199502093320601PubMedGoogle ScholarCrossref
12.
Burns  KEA, Raptis  S, Nisenbaum  R,  et al; Canadian Critical Care Trials Group.  International practice variation in weaning critically ill adults from invasive mechanical ventilation.   Ann Am Thorac Soc. 2018;15(4):494-502. doi:10.1513/AnnalsATS.201705-410OCPubMedGoogle ScholarCrossref
13.
Esteban  A, Anzueto  A, Frutos  F,  et al; Mechanical Ventilation International Study Group.  Characteristics and outcomes in adult patients receiving mechanical ventilation: a 28-day international study.   JAMA. 2002;287(3):345-355. doi:10.1001/jama.287.3.345PubMedGoogle ScholarCrossref
14.
Esteban  A, Ferguson  ND, Meade  MO,  et al; VENTILA Group.  Evolution of mechanical ventilation in response to clinical research.   Am J Respir Crit Care Med. 2008;177(2):170-177. doi:10.1164/rccm.200706-893OCPubMedGoogle ScholarCrossref
15.
Esteban  A, Frutos-Vivar  F, Muriel  A,  et al.  Evolution of mortality over time in patients receiving mechanical ventilation.   Am J Respir Crit Care Med. 2013;188(2):220-230. doi:10.1164/rccm.201212-2169OCPubMedGoogle ScholarCrossref
16.
Burns  KEA, Rizvi  L, Cook  DJ,  et al; Canadian Critical Care Trials Group.  Variation in the practice of discontinuing mechanical ventilation in critically ill adults: study protocol for an international prospective observational study.   BMJ Open. 2019;9(9):e031775. doi:10.1136/bmjopen-2019-031775PubMedGoogle Scholar
17.
Holm  S,.  A simple sequentially rejective multiple test procedure.   Scand J Stat. 1979:6(2):65-70.Google Scholar
18.
Riker  RR, Picard  JT, Fraser  GL.  Prospective evaluation of the Sedation-Agitation Scale for adult critically ill patients.   Crit Care Med. 1999;27(7):1325-1329. doi:10.1097/00003246-199907000-00022PubMedGoogle ScholarCrossref
19.
Vincent  JL, de Mendonca  A, Cantraine  F,  et al.  Use of the SOFA score to assess the incidence of organ dysfunction failure in intensive care units: results of a multicenter, prospective study.   Crit Care Med. 1998;26(11):1793-1800. doi:10.1097/00003246-199811000-00016PubMedGoogle ScholarCrossref
20.
Tallgren  M, Bäcklund  M, Hynninen  M.  Accuracy of Sequential Organ Failure Assessment (SOFA) scoring in clinical practice.   Acta Anaesthesiol Scand. 2009;53(1):39-45. doi:10.1111/j.1399-6576.2008.01825.xPubMedGoogle ScholarCrossref
21.
Harrell  FE  Jr.  Regression Modeling Strategies: With Applications to Linear Models, Logistic and Ordinal Regression, and Survival Analysis. Springer; 2015.
22.
Van Buuren  S.  Flexible Imputation of Missing Data. Chapman & Hall; 2018.
23.
Van Calster  B, McLernon  DJ, van Smeden  M, Wynants  L, Steyerberg  EW.  Calibration: the Achilles heel of predictive analytics.   BMC Med. 2019;17(1):230. doi:10.1186/s12916-019-1466-7 PubMedGoogle ScholarCrossref
24.
Beduneau  G, Pham  T, Schortgen  F,  et al; WIND (Weaning according to a New Definition) Study Group and the REVA Network.  Epidemiology of weaning outcome according to a new definition: the WIND study.   Am J of Respir Crit Care. 2017;195(6):772-83. doi:10.1164/rccm.201602-0320OCPubMedGoogle ScholarCrossref
25.
Ely  EW, Baker  AM, Evans  GW, Haponik  EF.  The prognostic significance of passing a daily screen of weaning parameters.   Intensive Care Med. 1999;25(6):581-587. doi:10.1007/s001340050906PubMedGoogle ScholarCrossref
26.
Kalassian  KG, Dremsizov  T, Angus  DC.  Translating research evidence into clinical practice: new challenges for critical care.   Crit Care. 2002;6(1):11-14. doi:10.1186/cc1446PubMedGoogle ScholarCrossref
27.
Schultz  MJ, Wolthuis  EK, Moeniralam  HS, Levi  M.  Struggle for implementation of new strategies in intensive care medicine: anticoagulation, insulin, and lower tidal volumes.   J Crit Care. 2005;20(3):199-204. doi:10.1016/j.jcrc.2005.05.007PubMedGoogle ScholarCrossref
28.
Brochard  L, Rauss  A, Benito  S,  et al.  Comparison of three methods of gradual withdrawal from ventilatory support during weaning from mechanical ventilation.   Am J Respir Crit Care Med. 1994;150(4):896-903. doi:10.1164/ajrccm.150.4.7921460PubMedGoogle ScholarCrossref
29.
Tripepi  G, Jager  KJ, Dekker  FW, Zoccali  C.  Selection bias and information bias in clinical research.   Nephron Clin Pract. 2010;115(2):c94-c99. doi:10.1159/000312871PubMedGoogle ScholarCrossref
30.
Ouellette  DR, Patel  S, Girard  TD,  et al.  Liberation from mechanical ventilation in critically ill adults: an official American College of Chest Physicians/American Thoracic Society clinical practice guideline.   Chest. 2017;151(1):166-180. doi:10.1016/j.chest.2016.10.036PubMedGoogle ScholarCrossref
Original Investigation
Caring for the Critically Ill Patient
March 23/30, 2021

Ventilator Weaning and Discontinuation Practices for Critically Ill Patients

Author Affiliations
  • 1Interdepartmental Division of Critical Care, University of Toronto, Toronto, Ontario, Canada
  • 2Division of Critical Care Medicine, Department of Medicine, Unity Health Toronto, St Michael's Hospital, Toronto, Ontario, Canada
  • 3Li Ka Shing Knowledge Institute, St Michael’s Hospital, Toronto, Ontario, Canada
  • 4Division of Critical Care Medicine, St Joseph’s Hospital, Hamilton, Ontario, Canada
  • 5Departments of Medicine and Health Research Methods, Evidence and Impact, McMaster University, Hamilton, Ontario, Canada
  • 6Applied Health Research Centre, St Michael’s Hospital, Toronto, Ontario, Canada
  • 7Centre for Health Evaluation and Outcome Sciences, Division of Critical Care Medicine, St Paul’s Hospital, University of British Columbia, Vancouver, British Columbia, Canada
  • 8CIBER de Enfermedades Respiratorias, Instituto de Salud Carlos III, Madrid, Spain
  • 9Multidisciplinary Organ Dysfunction Evaluation Research Network, Research Unit, Hospital Universitario Dr Negrin, Las Palmas de Gran Canaria, Spain
  • 10Department of Critical Care Medicine, Guy’s and St Thomas’ NHS Foundation Trust, London, United Kingdom
  • 11Department of Intensive Care, Hinduja National Hospital, Bombay, India
  • 12Intensive Care Unit, Royal Prince Alfred Hospital,, University of Sydney, Camperdown, New South Wales, Australia
  • 13The George Institute for Global Health, Sydney Australia
  • 14Tufts University School of Medicine, Boston, Massachusetts
  • 15Department of Surgical Sciences and Integrated Diagnostics, University of Genoa, Genoa, Italy
  • 16San Martino Policlinico Hospital, IRCCS for Oncology and Neurosciences, Genoa, Italy
  • 17Anaesthesia, Critical Care and Pain Medicine, Edinburgh Royal Infirmary, Edinburgh, Scotland, United Kingdom
JAMA. 2021;325(12):1173-1184. doi:10.1001/jama.2021.2384
Key Points

Question  In critically ill patients who receive invasive mechanical ventilation, how is invasive mechanical ventilation discontinued and do discontinuation practices differ internationally?

Findings  In this prospective observational study that included 1868 patients from 142 intensive care units in Canada, Europe, the US, India, the UK, and Australia/New Zealand from November 2013 to December 2016, 22.7% of patients underwent direct extubation, 49.8% underwent an initial spontaneous breathing trial (of which 81.8% had successful extubation), 8.0% had a direct tracheostomy, and 19.5% died before a weaning attempt. There was notable variation in several aspects of mechanical ventilation weaning practices.

Meaning  Mechanical ventilation weaning practices varied internationally, with nearly 50% of patients undergoing an initial spontaneous breathing trial.

Abstract

Importance  Although most critically ill patients receive invasive mechanical ventilation (IMV), few studies have characterized how IMV is discontinued in practice.

Objective  To describe practice variation in IMV discontinuation internationally, associations between initial discontinuation events and outcomes, and factors associated with the use of select discontinuation strategies and failed initial spontaneous breathing trials (SBTs).

Design, Setting, and Participants  Prospective, multinational, observational study of critically ill adults who received IMV for at least 24 hours from 142 intensive care units (ICUs) in 19 countries within 6 regions (27 in Canada, 23 in India, 22 in the UK, 26 in Europe, 21 in Australia/New Zealand, and 23 in the US).

Exposures  Receiving IMV.

Main Outcomes and Measures  Primary analyses characterized types of initial IMV discontinuation events (extubation, SBT, or tracheostomy) and associations with clinical outcomes (including duration of ventilation, ICU and hospital mortality, and ICU and hospital length of stay). Secondary analyses examined the associations between SBT outcome and SBT timing and clinical outcomes.

Results  Among 1868 patients (median [interquartile range] age, 61.8 [48.9-73.1] years; 1173 [62.8%] men) 424 (22.7%) underwent direct extubation, 930 (49.8%) had an initial SBT (761 [81.8%] successful), 150 (8.0%) underwent direct tracheostomy, and 364 (19.5%) died before a weaning attempt. Across regions, there was variation in the use of written directives to guide care, daily screening, SBT techniques, ventilator modes, and the roles played by clinicians involved in weaning. Compared with initial direct extubation, patients who had an initial SBT had higher ICU mortality (20 [4.7%] vs 96 [10.3%]; absolute difference, 5.6% [95% CI, 2.6%-8.6%]), longer duration of ventilation (median of 2.9 vs 4.1 days; absolute difference, 1.2 days [95% CI, 0.7-1.6]), and longer ICU stay (median of 6.7 vs 8.1 days; absolute difference, 1.4 days [95% CI, 0.8-2.4]). Patients whose initial SBT failed (vs passed) had higher ICU mortality (29 [17.2%] vs 67 [8.8%]; absolute difference, 8.4% [95% CI, 2.0%-14.7%]), longer duration of ventilation (median of 6.1 vs 3.5 days; absolute difference, 2.6 days [95% CI, 1.6-3.6]), and longer ICU stay (median of 10.6 vs 7.7 days; absolute difference, 2.8 days [95% CI, 1.1-5.2]). Compared with patients who underwent early initial SBTs, patients who underwent late initial SBTs (>2.3 days after intubation) had longer duration of ventilation (median of 2.1 vs 6.1 days; absolute difference, 4.0 days [95% CI, 3.7-4.5]), longer ICU stay (median of 5.9 vs 10.8 days; absolute difference, 4.9 days [95% CI, 4.0-6.3]), and longer hospital stay (median of 14.3 vs 22.8 days; absolute difference, 8.5 days [95% CI, 6.0-11.0]).

Conclusions and Relevance  In this observational study of invasive mechanical ventilation discontinuation in 142 ICUs in Canada, India, the UK, Europe, Australia/New Zealand, and the US from 2013 to 2016, weaning practices varied internationally.

Trial Registration  ClinicalTrials.gov Identifier: NCT03955874

Introduction

Invasive mechanical ventilation (IMV) is a defining feature of critical illness. In epidemiologic studies conducted in the US between 2002 and 2010, the incidence of IMV ranged from 270 to 314 cases per 100 000 population,1,2 and this incidence is expected to increase in the future.3 IMV increases the complexity and costs of health care.4Quiz Ref ID Cumulative exposure to IMV has been associated with potentially harmful cointerventions (eg, sedation), increased morbidity (eg, ventilator-related complications [such as pneumonia]),5 and long-term functional sequelae and cognitive impairment.6 Thus, limiting the duration of IMV has been identified as a priority area for research.7

Quiz Ref IDIMV can be discontinued abruptly by direct extubation or gradually through 1 or more spontaneous breathing trials (SBTs) or tracheostomy collar trials. Randomized trials have evaluated the use of screening protocols (vs usual care) to identify patients who are ready to undergo an SBT,8 different techniques to conduct SBTs,9 duration of an SBT,10 and strategies to eventually extubate patients whose initial SBT failed.11 Discontinuation involves the skills of diverse clinicians whose roles differ across settings and may be influenced by factors related to patients, institutions, and care processes. An international survey of intensivists’ weaning practices identified significant variation in screening frequency, ventilator modes, SBT techniques, written directives to guide weaning, and the roles of available personnel.12 Although several large-scale observational studies of IMV use have been conducted,13-15 none have focused exclusively on IMV discontinuation.

The objectives of this prospective, multinational, observational study of critically ill adults who were intubated and receiving mechanical ventilation were to describe ventilator weaning and discontinuation practices, investigate associations between discontinuation strategies and outcomes, and identify factors associated with the use of select discontinuation strategies and initial SBT failures.

Methods
Objectives

The primary objectives of this study were to characterize variation in discontinuation practices among patients who received IMV for at least 24 hours, with regard to written directives, daily screening, preferred methods of ventilator support, SBT techniques, and sedation and mobilization practices, and to describe the association between the initial strategy used to discontinue IMV (direct extubation, initial SBT, or direct tracheostomy) and clinical outcomes (duration of ventilation; ICU and hospital mortality; ICU and hospital length of stay; and the percentage of patients who were reintubated, readmitted to the ICU, receiving mechanical ventilation at day 28, and in the ICU at day 28). Secondary objectives included describing the associations between initial SBT outcome (failure vs success), SBT timing (early vs late initial SBT), SBT techniques, and humidification strategies and clinical outcomes. We also described factors associated with selection of alternative discontinuation strategies. In a tertiary objective, we described factors associated with initial SBT failure. The full protocol for this study was published previously.16 Ethics approval was sought by site investigators at all participating sites. Seventeen ICUs required prospective written or verbal consent for participation. The institutional regulatory policy at 1 site enabled collection of data that were deidentified and protected by prior consent or privacy standards. The remaining ICUs were permitted to screen patients for eligibility and collect deidentified data for patients who met study eligibility criteria without written or oral consent.

Patient Population and Participating Centers

Research personnel prospectively followed up all newly admitted patients who received IMV for at least 24 hours.16 Patients were excluded if they were transferred to participating ICUs without a documented intubation time, underwent a tracheotomy at ICU admission, were already using ventilator settings compatible with SBT (T-piece or continuous positive airway pressure [CPAP] ≤5 cm H2O or pressure support [PS] ≤8 cm H2O with or without positive end-expiratory pressure [PEEP] or automatic tube compensation or equivalent) at ICU admission, were at participating ICUs for 24 hours or more, were readmitted during the study period, or participated in other studies with explicit weaning protocols.

A region was defined as a country or collection of countries represented by a critical care society.16 We aimed to include 150 ICUs with similar representation (approximately 25 ICUs) from each region.16 We used a multimodal strategy (information cards enclosed in a previously administered, postal, international weaning survey16 and correspondence with regional collaborators, critical care societies, and site investigators) to identify participating ICUs and permitted flexible start dates to allow for variable approval processes and clinical coverage models.

Data Collection

Data were collected from November 2013 to December 2016. In each ICU, we collected data from at least 10 discontinuation events and all deaths before a weaning attempt. Patients were followed up until successful extubation or disconnection (tracheostomy), death, or ICU discharge or transfer or until day 60 for patients with ventilator dependence.

Outcomes

Data regarding written directives, daily screening, preferred methods of ventilator support, SBT techniques, and sedation and mobilization practices pertaining to IMV and discontinuation were collected. Types of initial IMV discontinuation events were categorized as direct extubation, SBT, or tracheostomy. We defined an SBT as a focused assessment of the patient's capacity to breathe spontaneously with any technique (eg, T-piece, CPAP, PS with minimal assistance) with the goal of discontinuing IMV. We defined successful extubation (disconnection) as the time when unsupported spontaneous breathing began and noninvasive ventilation or IMV was not used for more than 48 hours after extubation or disconnection. Clinical outcomes were duration of ventilation; ICU and hospital mortality; ICU and hospital length of stay; and the percentage of patients who were reintubated, readmitted to the ICU, receiving mechanical ventilation at day 28, and in the ICU at day 28. Duration of mechanical ventilation summarizes the time to successful extubation and is reported for both patients who died during hospitalization and for those who did not.

Research personnel identified eligible patients using a screening log; collated data to characterize patients, discontinuation events, care processes, and clinical outcomes; and entered data into Medidata RAVE (Medidata Solutions). Two investigators (K.B. and L.R.) reviewed all data forms, transmitted queries to clarify discontinuation events and illogical data, and adjudicated outcomes, when required, in collaboration with a third reviewer (M.M. or D.C.).

Statistical Analysis

Data were planned to be collected on at least 10 discontinuation events other than death in each of 150 participating ICUs. We sought to achieve 225 initial SBT failures, anticipating that at least 50% of discontinuation events would involve an initial SBT (expected range, 750-1050 SBTs) and that 70% (n = 525) would be successful.16

We used descriptive statistics to summarize binary (number and percentage) and continuous (mean and SD and median and interquartile range [IQR], when appropriate) variables. We expressed differences in clinical outcomes (SBT outcome and timing) using the absolute difference (AD) with 95% CIs with Yates continuity correction for binary outcomes and median (IQR) with bootstrapped 95% CIs for continuous outcomes. Analyses of mechanical ventilation duration and length of stay were conducted overall as well as limited to patients who died during hospitalization and those who did not.

To examine associations between discontinuation strategies and SBT outcome and timing, we accounted for clustering of ICUs within hospitals. We used the median time to initial SBT to characterize early (≤2.3 days) vs late (>2.3 days) SBTs. P values were adjusted for clustering of ICUs within hospitals by using a mixed-model framework and were corrected for multiple testing using the Holmes-Bonferroni method.17 Continuous outcomes with skewed distributions were log-transformed. Residuals were assessed to ensure the normality assumption was met.

A generalized linear mixed model with a logit link function was used to identify baseline characteristics and factors that developed between ICU admission and discontinuation events related to the use of specific discontinuation strategies (eMethods in the Supplement). Clustering was accounted for by including a random effect for hospital. To identify factors associated with initial SBT failure (vs success), we constructed 3 generalized linear mixed models with a logit link including patient variables (age, SBT duration, MV duration, Sedation-Agitation Scale [SAS] scores,18 modified Sequential Organ Failure Assessment [SOFA] respiratory score,19,20 SBT method, and primary diagnosis; model 1); patient and site variables (screening, hospital type, hospital profit status, written document for SBT conduct and adjusting IMV support, respiratory therapist presence, and region; model 2); and patient, site, and practitioner variables (clinician type and years in ICU; model 3). We used a likelihood ratio test to compare models. Model 2 had the best balance of variables based on discrimination (C statistic) and calibration (calibration slope and calibration-in-the-large) after performing internal bootstrap validation.21,22 We assumed that missing variables in regression analyses (eg, SOFA score at ICU admission, SOFA score before discontinuation event) were missing at random. These variables were imputed using multiple imputation by chained equations.23 In a post hoc analysis, we examined the association between different SAS scores18 (classified as low [SAS score, 1-2; indicates more sedated], middle [SAS score, 3-5], and high [SAS score, 6-7; indicates less sedated]) on initial SBT failure rates. We considered 2-sided P values <.05 to indicate statistical significance. Analyses were performed using R software, versions 3.5.0 and 4.0.2 (R Foundation for Statistical Computing).

Results
Patient and Site Characteristics

Of 1868 participants (median [IQR] age, 61.8 [48.9-73.1] years; 1173 [62.8%] men), 852 (45.6%) had acute respiratory failure and 1093 (58.5%) were receiving IMV at the time of ICU admission (Figure; Table 1). Of the 142 participating sites in 19 countries within 6 regions (27 in Canada [n = 340 {18.2%}], 23 in India [n = 327 {17.5%}], 22 in the UK [n = 311 {16.6%}], 26 in Europe [n = 351 {18.8%}], 21 in Australia/New Zealand [n = 263 {14.1%}], and 23 in the US [n = 276 {14.8%}]), most were medical/surgical (57 [40.1%]) or multidisciplinary (33 [23.2%]) ICUs, and 114 (80%) had a university affiliation (Table 2). There was a median of 13 patients per ICU.11,15

Primary Outcomes
Regional Variation in Discontinuing IMV
Use of Written Directives During Weaning

The percentage of ICUs with written directives for screening for (range, 4.5%-82.6%; P < .001) and conducting (range, 9.1%-78.3%; P < .001) SBTs varied significantly across regions. Written directives to screen for SBT readiness were present in more than half of the participating ICUs in Canada (15 [55.6%]), India (16 [69.6%]), and the US (19 [82.6%]). The percentage of ICUs with written directives for adjusting ventilator support did not vary significantly across regions (range, 9.5%-60.9%; P = .16). However, written directives for SBT conduct and for adjusting ventilator support were present in less than half of the participating ICUs in 5 regions (13 [48.1%] and 11 [40.7%] in Canada, 11 [47.8%] and 9 [39.1%] in India, 2 [9.1%] and 4 [18.2%] in the UK, 12 [46.2%] and 8 [30.8%] in Europe, and 2 [9.5%] and 2 [9.5%] in Australia/New Zealand). Although the percentages of ICUs with written directives for administering sedation (range, 40.9%-78.3%) and interrupting sedation (range, 23.8%-91.3%) were not statistically significantly different, the percentage of ICUs with directives for early mobilization (range, 14.3%-65.2%; P < .001) varied significantly across regions (eTable 1 in the Supplement).

Practices in Screening for SBTs

Patients were not screened for SBTs daily in participating ICUs in 4 regions (5 [18.5%] in Canada, 12 [54.5%] in the UK, 3 [11.5%] in Europe, and 13 [61.9%] in Australia/New Zealand). Patients were screened once daily to undergo an SBT in more than half of the ICUs in Canada (18 [66.7%]), India (17 [73.9%]), and the US (19 [82.6%]) and in less than half of the participating ICUs in Europe (11 [42.3%]), the UK (6 [27.3%]), and Australia/New Zealand (3 [14.3%]) (P < .001). Patients were less frequently screened twice daily (range, 3.7%-19.2%) or more than twice daily (range, 0%-26.9%).

Practices in Conducting SBTs

Initial SBTs most commonly used PS with PEEP (457 of 930 [49.1%]) or T-piece (236 of 930 [25.4%]) and less frequently applied CPAP (100 of 930 [10.8%]) or PS without PEEP (88 of 930 [9.5%]). SBTs were commonly conducted using PS with PEEP in Canada (85 of 177 [48.0%]), the UK (78 of 132 [59.1%]), Australia/New Zealand (57 of 76 [75.0%]), and the US (166 of 214 [77.6%]), and T-piece SBTs were commonly performed in India (99 of 181 [54.7%]) and Europe (77 of 150 [51.3%]) (P < .001).

Ventilator Support Before Initial Discontinuation Attempts

Among 424 patients who underwent direct extubation, most patients in ICUs outside of the US were receiving PS mode before extubation (70 of 92 [76.1%] in Canada, 14 of 27 [51.9%] in India, 52 of 69 [75.4%] in the UK, 78 of 103 [75.7%] in Europe, and 101 of 124 [81.5%] in Australia/New Zealand vs 1 of 9 [11.1%] in the US) (P < .001). Similarly, before an initial SBT, most patients in ICUs outside of the US were receiving PS (133 of 177 [75.1%] in Canada, 104 of 181 [57.5%] in India, 61 of 132 in the UK [46.2%], 99 of 149 in Europe [66.4%], and 38 of 76 in Australia [50.0%] vs 15 of 214 [7.0%] in the US) (P < .001). Before direct extubation, patients in ICUs in the US were most commonly receiving assist-control pressure-regulated volume control/volume control plus (3 of 9 [33.3%]), synchronized intermittent mandatory ventilation volume control with or without PS (2 of 9 [22.2%]), or assist-control volume control (2 of 9 [22.2%]). Before initial SBTs, patients in ICUs in the US were most commonly receiving assist-control pressure-regulated volume control/volume control plus (58 of 214 [27.1%]), synchronized intermittent mandatory ventilation volume control with or without PS (20 of 214 [9.3%]), or synchronized intermittent mandatory ventilation pressure-regulated volume control/volume control plus with or without PS (16 of 214 [7.5%]). Among 150 patients who underwent direct tracheostomy, patients across regions were most commonly supported with PS (59 of 150 [39.3%]) and assist-control volume control (25 of 150 [16.7%]) before tracheostomy.

Personnel Involved in Screening for and Conducting SBTs and Adjusting Ventilator Support

Clinicians who were responsible for daily screening (nurses [range, 9.5%-75.0%; P < .001] and respiratory therapists [range, 0%-95.7%; P < .001]) differed significantly across regions, and the clinician who was indicated to be the most responsible for conducting SBTs (P < .001) and adjusting ventilator support (P < .001) also differed significantly across regions (eTable 2 in the Supplement).

Sedation and Mobilization Practices

There was a significant difference in sedation levels immediately before an initial SBT (most patients had SAS scores of 3 [162 of 929 {17.4%}] or 4 [558 of 929 {60.1%}] vs direct extubation (most patients had SAS scores of 4 [266 of 424 {62.7%}] or 5 [54 of 424 {12.7%}]) (P < .001) (eTable 3 and eFigure 1 in the Supplement). Few patients were actively mobilized immediately before discontinuation events (51 of 424 [12.0%] who underwent direct extubation, 99 of 930 [10.6%] who underwent initial SBT, and 12 of 150 [8.0%] who underwent direct tracheostomy) (eTable 4 in the Supplement). There was also a significant difference in mobilization levels (none vs nonmobility vs mobility physiotherapy) immediately before an initial SBT vs direct extubation (P < .001).

Associations Between Initial Discontinuation Events and Clinical Outcomes

Of the 1868 total patients, 424 (22.7%) underwent direct extubation (including 36 [8.5%] unplanned), 930 (49.8%) underwent an initial SBT (761 [81.8%] successful), 150 (8.0%) underwent direct tracheostomy, and 364 (19.5%) died before a weaning attempt (Figure). Within ICUs, there were a median (IQR) of 3 (1-6) direct extubations,1-12 8 (5-9) initial SBTs,1-17 1 (1-2) direct tracheostomy,1-7 and 3 (2-4) deaths before weaning attempts.1-11 There were no statistically significant differences between men vs women in discontinuation strategies (direct extubation: 272 [64.2%] vs 152 [35.8%]; direct tracheostomy: 102 [68%] vs 48 [32%]; initial SBT: 57 [62.3%] vs 351 [37.7%]; P = .40).

Across discontinuation strategies (initial direct extubation, tracheostomy, and SBT), there were differences in ICU mortality [20 (4.7%) vs 23 (15.3%) vs 96 (10.3%)], reintubation [26 (6.3%) vs 42 (28.4%) vs 67 (7.7%)], and the percentage of patients who were receiving mechanical ventilation [11 (2.6%) vs 38 (26.8%) vs 50 (5.5%)] and in the ICU at day 28 [26 (6.2%) vs 51 (35.7%) vs 86 (9.4%)]. There were differences across discontinuation strategies between initial direct extubation, tracheostomy, and SBT in the median total duration of mechanical ventilation overall (2.9 vs 13.2 vs 4.1 days), in patients who did not die during hospitalization (2.9 vs 12.9 vs 3.7 days), and in patients who died during hospitalization (7.3 vs 16.9 vs 7.6 days); in median duration of ICU stay overall (6.7 vs 19.6 vs 8.1 days), in patients who did not die during hospitalization (6.2 vs 18.9 vs 7.8 days), and in patients who died during hospitalization (17.5 vs 22.2 vs 11.2 days); and in median duration of hospital stay overall (16.9 vs 35.4 vs 18.0 days), in patients who did not die during hospitalization (16.3 vs 35.9 vs 17.9 days), and in patients who died during hospitalization (23.2 vs 33.1 vs 18.1 days) (Table 3; eTables 5 and 6 in the Supplement)

Compared with patients who underwent direct extubation, patients who underwent SBT had higher ICU mortality (20 [4.7%] vs 96 [10.3%]; AD, 5.6% [95% CI, 2.6%-8.6%]), longer median duration of mechanical ventilation overall (2.9 vs 4.1 days; AD, 1.2 days [95% CI, 0.7-1.6]) and in those who did not die during hospitalization (2.9 vs 3.7 days; AD, 0.8 days [95% CI, 0.4-1.3]), and longer median ICU stay overall (6.7 vs 8.1 days; AD, 1.4 days [95% CI, 0.8-2.4]) and in patients who did not die during hospitalization (6.2 vs 7.8 days; AD, 1.6 days [95% CI, 0.8-2.6]) (Table 3)

Secondary Outcomes
Association Between Outcome of Initial SBT and Clinical Outcomes

Patients whose initial SBT failed (vs passed) had higher ICU mortality (29 [17.2%] vs 67 [8.8%]; AD, 8.4% [95% CI, 2.0%-14.7%]), were more likely to be still receiving mechanical ventilation at day 28 (19 [11.7%] vs 31 [4.1%]; AD, 7.5% [95% CI, 2.0%-13.0%]) and be in the ICU at day 28 (26 [15.7%] vs 60 [8.0%]; AD, 7.7% [95% CI, 1.4%-13.9%]), and had longer median duration of ventilation (6.1 vs 3.5 days; AD, 2.6 days [95% CI, 1.6-3.6]) and ICU stay (10.6 vs 7.7 days; AD, 2.8 days [95% CI, 1.1-5.2]) (Table 4). Associations of alternative SBT techniques and humidification strategies with clinical outcomes are shown in eTables 7 and 8 in the Supplement.

Compared with patients who underwent an initial SBT earlier (≤2.3 days), those who had an initial SBT after this time were more likely to still be receiving mechanical ventilation at day 28 (15 [3.3%] vs 35 [7.7%]; AD, 4.4% [95% CI, 1.3%-7.6%]) and be in the ICU at day 28 (29 [6.3%] vs 57 [12.6%]; AD, 6.3% [95% CI, 2.3%-10.3%]) and have longer median duration of mechanical ventilation (2.1 vs 6.1 days; AD, 4.0 days [95% CI, 3.7%-4.5%]), ICU stay (5.9 vs 10.8 days; AD, 4.9 days [95% CI, 4.0-6.3]), and hospital stay (14.3 vs 22.8 days; AD, 8.5 days [95% CI, 6.0-11.0]) (Table 4).

Factors Associated With Use of Specific Discontinuation Strategies

In regression analyses, patients who underwent an initial SBT (vs direct extubation) were more likely to be older (odds ratio [OR] per 10-year increase, 1.1 [95% CI, 1.0-1.3]; P = .02) and have solid organ malignancy (OR, 2.6 [95% CI, 1.2-5.2]; P = .01) and hypertension (OR, 1.6 [95% CI, 1.0-2.3]; P = .03). Compared with the US, patients were significantly less likely to undergo an initial SBT vs direct extubation in Australia/New Zealand (OR, 0.01 [95% CI, 0.00-0.05]; P < .001), Canada (OR, 0.03 [95% CI, 0.01-0.17]; P < .001), Europe (OR, 0.03 [95% CI, 0.01-0.18]; P < .001), and the UK (OR, 0.04 [95% CI, 0.01-0.23]; P < .001).

Tertiary Outcomes
Factors Associated With Initial SBT Failure

In regression modeling, 3 factors were identified to be associated with initial SBT failure, including 2 patient-level variables (higher SOFA respiratory score [P = .005] and lower SAS score [P = .02] before an initial SBT attempt) and a site-level variable (not having a protocol to adjust ventilator settings [P = .03]). Missing data in this analysis were less than 1% for most variables, except for SOFA score at ICU admission (279 [30%]) and before discontinuation events (164 [18%]). A post hoc analysis showed that patients with low (more sedated) vs midrange SAS scores were significantly more likely to have a failed initial SBT (OR, 2.4 [95% CI, 1.3-4.5]; P = .006) (eFigure 2 in the Supplement).

Discussion

Quiz Ref IDThis international study identified variability in IMV discontinuation practices across geographical regions with regard to the use of protocols, screening for and conducting SBTs, adjustment of ventilator support, and the responsibility of clinicians involved in weaning. The main findings were that nearly 50% of patients underwent an initial SBT, of which more than 80% were successful; an initial SBT (vs direct extubation) was associated with higher ICU mortality and longer duration of ventilation and ICU stay; failing the initial SBT (vs passing the initial SBT) was associated with higher ICU mortality, longer duration of ventilation, longer ICU stay, and greater likelihood of still receiving ventilation and being in the ICU at day 28; and undergoing a later (vs earlier) initial SBT was associated with longer duration of ventilation, longer ICU and hospital stays, and greater likelihood of still receiving ventilation and being in the ICU at day 28.

Similar to a large international weaning survey, this study found that most patients were screened once daily to identify SBT candidates, and were less often screened twice daily or more frequently.12 There was significant regional variation in the presence of written directives to guide care during weaning and the roles played by available clinicians.12 T-piece SBTs were more common in India and Europe, while PS with PEEP SBTs were more common in Canada, the UK, the US, and Australia/New Zealand. Compared with earlier studies reported by Esteban and colleagues in 2002,13 2008,14 and 2013,15 the current study included data from fewer medical/surgical ICUs (68%-77%13-15 vs 40.1%) and observed similar ICU (7-813-15 vs 8 days) and hospital (16-1713-15 vs 19 days) length of stays, but lower rates of ICU mortality (28%-31%13-15 vs 26.9% [including initial deaths]), reintubation (11%-14%13-15 vs 9%), and tracheostomy (11%-15%13-15 vs 8%). In contrast to their findings, fewer patients in the current study underwent an initial SBT (58%-62%13,14 vs 49.8%). Among those who underwent SBTs, there was more use of PS with PEEP SBTs than T-piece SBTs. Discordant findings may reflect differences in the study populations, inclusion of geographically disparate ICUs in this study, or temporal changes in care. Similar to the Weaning according to a New Definition (WIND) classification, this study also identified higher mortality and duration of mechanical ventilation with progression from WIND group 1 (first separation attempt to weaning termination ≤24 hours: mortality, 3.6%; mechanical ventilation duration, 2.6 days) to WIND group 2 (first separation attempt to weaning termination within 2-6 days: mortality, 13.1%; duration of mechanical ventilation, 5.2 days) and to WIND group 3 (first separation attempt to weaning termination ≥7 days: mortality, 22.9%; duration of mechanical ventilation, 16.6 days) in patients who received IMV for at least 24 hours.24 Notwithstanding, this study differs from the WIND study in important ways. Approximately half of the patients (1543 of 2729 [56.5%]) in the WIND study received IMV for less than 24 hours, while the current study included 1868 patients who received IMV for at least 24 hours, corresponding to WIND groups 2 and 3 (n = 1183). Although discontinuation attempts (extubation, SBT, tracheostomy) in the current study were similar to separation attempts in the WIND study, the current study collected detailed data at the time of discontinuation attempts and further categorized initial SBT attempts prospectively as successful or failed.

The importance of a daily screen failure was highlighted in a previous retrospective, single-center study (n = 300; 2 ICUs) of a trial that compared daily screening with either T-piece or CPAP SBTs to daily screening with SBTs conducted at the clinician’s discretion.25 In that study, Ely et al25 reported that patients who had a failed daily screen, with or without SBT, were more likely to require IMV for more than 21 days and were less likely to be successfully extubated and alive at hospital discharge. The current study focused on failure of an initial SBT, as opposed to a daily screen with or without an SBT, in a large cohort of predominantly unselected patients (n = 930) in 129 ICUs. Quiz Ref IDFailure of an initial SBT was associated with a broader range of less favorable outcomes, more likely as a marker of illness severity than a cause of poor outcome in this observational study.

Several other findings warrant comment. First, regional variation existed in the use of written directives to guide weaning and in the roles played by clinicians involved in weaning. Second, data from this study can be used to inform decision-makers regarding outcomes associated with specific discontinuation practices (eg, tracheostomy). Third, these data highlight the challenges of generalizing weaning evidence from randomized trials (screening, SBT conduct, use of protocols) into diverse practice settings with variable personnel.26,27Quiz Ref ID Fourth, an exploratory post hoc analysis showed that patients with low (more sedated) vs midrange SAS scores were more likely to have a failed initial SBT. These findings highlight an important potential interaction between sedation level and SBT outcome. Compared with prior studies, findings from the current study suggest that clinical outcomes have improved over time, but are less favorable than those reported in clinical trials.11,28 These findings may reflect the greater potential for selection bias in randomized trials compared with observational studies.29

To our knowledge, this is the first multinational, prospective, observational study focused specifically on describing IMV discontinuation in critically ill patients who received IMV for at least 24 hours. This study has several strengths. By design, similar representation from each region was achieved and allowed us to draw inferences about regional practice variation. Site-level screening logs helped to identify consecutive patients and minimize the potential for selection bias. Near-complete data with minimal missing data were achieved by using a detailed review and query generation process. These data reflect actual practices and complement the findings of a previous cross-sectional survey reflecting intensivists’ stated practices in IMV weaning from ICUs in the same 6 geographic regions.12

Limitations

This study had several limitations. First, data were collected on 169 SBT failures, representing 75% of the target sample size.16 Second, these data are descriptive and analyses are not adjusted for patient characteristics. As such, they highlight associations between discontinuation events and outcomes and cannot infer causality. For example, severely ill patients may have been more likely to undergo an initial SBT (vs direct extubation) or a late initial SBT. Additional data were not available to describe why patients did not undergo an earlier SBT (eg, failure to be screened, did not pass screening criteria earlier, too ill to undergo an earlier SBT). Third, although the eligibility criteria enabled inclusion of a diverse cohort of critically ill patients in whom IMV was discontinued using several different strategies, the use of a different time threshold for inclusion (12 or 48 hours of IMV) and limiting data collection to 10 events at each site may have affected the patient cohort included, the number and type of discontinuation events, and outcomes reported. Fourth, the study findings may not be generalizable to all ICUs and settings because 80% of included ICUs were university-affiliated ICUs with physicians in training. Fifth, these data were collected between 2013 and 2016; whether they reflect current practice is unclear, but the authors are unaware of any data that has suggested that variation in IMV discontinuation has changed. Few large trials10 or observational studies24 focused on IMV discontinuation have been published since completion of the current study, and the 2017 international weaning guideline was published after data collection for this study was completed.30

Conclusions

In this observational study of invasive mechanical ventilation discontinuation in 142 ICUs in Canada, India, the UK, Europe, Australia/New Zealand, and the US from 2013 to 2016, weaning practices varied internationally.

Section Editor: Christopher Seymour, MD, Associate Editor, JAMA (christopher.seymour@jamanetwork.org).
Back to top
Article Information

Corresponding Author: Karen E. A. Burns, MD, MSc (Epid), Interdepartmental Division of Critical Care Medicine, Unity Health Toronto, St Michael's Hospital, 30 Bond St, Office 4-045 Donnelly Wing, Toronto, ON M5B 1W8, Canada (karen.burns@unityhealth.to).

Accepted for Publication: February 9, 2021.

Author Contributions: Drs Burns and Lebovic had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Burns, Cook, Dodek, Villar, Epstein, Pelosi, Meade.

Acquisition, analysis, or interpretation of data: Burns, Rizvi, Cook, Lebovic, Villar, Slutsky, Jones, Kapadia, Gattas, Epstein, Kefala, Meade.

Drafting of the manuscript: Burns, Rizvi, Villar.

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

Statistical analysis: Lebovic, Villar.

Obtained funding: Burns, Meade.

Administrative, technical, or material support: Burns, Rizvi, Cook, Jones, Epstein.

Supervision: Burns, Villar, Pelosi.

Other - coordinated India contribution: Kapadia.

Other - detailed data review: Cook.

Other - project management - including the following activities: training sites on protocol, data collection form development, data management, developing data analytical plan, and manuscript development: Rizvi.

Conflict of Interest Disclosures: Dr Burns reported receiving grants from Fisher & Paykel, Covidien, and GE Healthcare Ltd during the conduct of the study. Dr Lebovic reported receiving grants from the Canadian Institutes for Health Research during the conduct of the study. Dr Villar reported receiving grants from Maquet/Getinge outside the submitted work. Dr Slutsky reported personal fees for consultancy in relation to extracorporeal membrane oxygenation from Baxter and Novalung/Xenios outside the submitted work. No other disclosures were reported.

Funding/Support: This study was funded through a peer-reviewed industry-partnered grant from the Canadian Institutes of Health Research with contributions from Fisher & Paykel, Covidien, and GE Healthcare Ltd.

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. One industry partner (Fisher & Paykel) reviewed the information being collected on humidification devices during data collection form development.

Meeting Presentation: Data from this study were presented in abstract form at the 2020 American Thoracic Society Conference; May 15-20, 2020.

Additional Contributions: We thank the members of the Canadian Critical Care Trials Group (CCCTG) for their contributions to the study design and guidance during study implementation. In particular, we wish to thank Niall Ferguson, MD, MSc (Unity Health Network, Toronto, Canada), and Philippe Jouvet, MD, PhD (St Justine Hospital, Montreal, Canada), for peer reviewing our manuscript on behalf of the CCCTG Grants and Manuscript Committee. We also thank Shujun (Diana) Ya, HBSc (Applied Health Research Centre, Toronto), for her assistance with the statistical analyses. None of these individuals received compensation for their role in the study. We thank the site investigators and research personnel who participated in the Ventilator Weaning and Discontinuation Practices for Critically Ill Patients study. The individuals listed below were research personnel, some of whom are also members of the CCTG.

Research Collaborators: Canada: Hamilton General Hospital (Dr M. Meade, L. Hand); St. Michael’s Hospital (Dr K. Burns, Dr J. Marshall, O. Smith, M. Wang, & J. Lee); The Ottawa Hospital (Dr L. McIntyre, I. Watpool, L. Leclair, R. Porteous, T. McArdle); St. Paul’s Hospital (Dr P. Dodek, B. Ashley); St. Joseph’s Hospital (Dr D. Cook, F. Clarke); Thunder Bay Regional Health Sciences Centre (Dr M. Scott, C. Pelletier, S. Stoger, C. Cryderman, M. Roussos, A. Holts); Kingston General Hospital (Dr J. Muscedere, J. Podilchak, M. Hunt, S. Fleury); Institut Universitaire de Cardiologie et de Pneumologie de Quebec (IUCPQ) (Dr F. Lellouche, P. Lizotte, PA. Bouchard, MC. Ferland, K. Collard); CSSS Champlain—Charles-Le Moyne (Charles LeMoyne Hospital) (Dr G. Poirier, S. Spearson, I. Neas); Montreal Heart Institute (Dr A. Denault, MM. Pires, S. Grenier); Capital Health - Queen Elizabeth II Health Sciences Centre (Dr R. Hall, L. Julien); McGill University Health Centre (Dr K. Khwaja, J. Campisi, C. Keropian, Dr LC. Banici, L. Garcia); Mount Sinai Hospital (Dr S. Mehta, B. Giacomino, S. Shah, M. Jakab); Hopital de l'enfant Jesus (Dr A. Turgeon, SJ. Breton, N. Langis, MC. Temblay); Universite de Sherbrook (Dr M. Mayette, QM. Hector); Toronto General Hospital (Dr M. Herridge, A. Matte); Oakville Trafalgar Memorial Hospital (Dr S. Kohli, L. Sterling, K. Khwaja); Centre hospitalier de l'Université de Montréal (Dr P. Aslanina, R. Rigal, F. Benettaib, C. Dube, S. Furdui); Centre de santé et de services sociaux de Trois-Rivières (Dr E. Charbonney, M. Bergeron). India: Saifee Hospital (Dr D. Aggarwal, M Patel); Medanta-The Medicity (Dr D. Govil, Dr S Gupta); Sri Ramachandra Medical Centre (Dr R. Prabakar); Apollo Hospitals (Dr S. Haranath, Dr Padmaja); Hinduja Hospital (Dr F. Kapadia, R. Kumar); Convenient Hospital Ltd (Dr N. Jain, M. Pahuja); Fortis Hospital (Dr M. Sircar, J. Shukla); Apollo Hospitals Bhubaneswar (Dr S. Sahu, M. Singh); Rajasthan Hospitals (Dr M. Harunbhai);Health care Global Multispecialty Hospital (Dr B. Shah, N. Ponkia, P. Khatri); Sanjeevan Hospital (Dr S. Dixit, Dr R. Rhayakar); Sanjay Gandhi Postgraduate Institute of Medical Sciences (SGPGIMS) (Dr M. Gurjar); Saket City Hospital (Dr P. Saxena, B. Sharma); Jamdar Hospital PVT. LTD. (Dr C. B. Oak); Terna Sahyadri Speciality Hospital (Dr V. Kumar). UK: Oxford University Hospitals: John Radcliffe (JR) and Churchill (CH) Hospital (Dr JL. Millo, A. Chadwick); Royal Surrey County Hospital (Dr B. Creagh-Brown, L. Montague, D. Hull); James Cook University Hospital (Dr IM. Gonzalez, C. Emanuel, A. Inma, S. Bonner); Hull and East Yorkshire Hospital NHS Trust (Dr AP. Gratrix, ND Smith, C. Abernathy); Royal Liverpool University Hospital (Dr PA. Hampshire, K. Williams, A. Walker); Medway NHS Foundation (Dr P. Hayden, C. Plowright, J. Cullinane, R. Box, R. Box, T. Hatton); Morriston Hospital (Dr H. Jones, T. Ghuman, C. Fagan); Royal Sussex Country Hospital (Dr O. Boyd, L. Ortiz-Ruiz de Gordoa); Royal Infirmary of Edinburgh (Dr K. Kefala, DP. Hope); University Hospital Lewisham (Dr BO rose, RA Rosie, D. Grannell, C. Harris); Royal Victoria Hospital (Dr J. Silversides, M. Maguire); Belfast City Hospital (Dr J. Silversides, Dr A. Callaghan, Dr C. McClure); Great Western Hospital (Dr RE. Prout, H. Brown, S. Hughes, L. Matter, A. Leslie, RE. Jones); Wirral University Teaching Hospital (Dr J. Gannon, R. Jacob, S. Christopher, P. Brassey); St George’s Hospital NHS Trust (Dr S. Leaver, C. Ryan, J. Mellinghoff); Sunderland Royal Hospital (Dr P. Hersey, J. Furneval, G. Whitehead); Watford General Hospital (Dr VJ. Page, C. Annalisa); Antrim Area Hospital (Dr CL. Nutt, O. O’Neill, E. McKay); Barnet Hospital, Royal Free London Hospitals NHS Foundation Trust (Dr RR. Jha, Dr N.Hooker); Craigavon Area Hospital (Dr J. Ferguson, L. Espie); Aberdeen Royal Infirmary (Dr C. Kaye, T. Scott). Europe: Bashkir Republic Hospital G.G. Kuvatov (Dr K. Zolotukhin); Hospital General Castellon (Dr AB. Muncharaz); Centre Hospitalier Universitaire -Mont-Godinne (Dr P. Bulpa, S. Bouhon, A. Romnee); Mater Dei Hospital (Dr S. Sciberras, L. Chircop); Centre Hospitalier Universitaire Amiens (Dr D. Herve, Dr M. Stephanie); Medisch Spectrum Twente (Dr A. Beishuizen, Dr JW. Vermeijden); Inselspital, Bern University Hospital (Dr J. Takala, Dr S. Jakob, P. Venetz, T. Merz); Groupe Hospitalier Paris Saint Joseph (Dr B. Misset, J. Fournier); Charité – University Medicine Berlin (Dr St. Weber-Carstens, Dr A. Goldmann); Hospital Beatriz Angelo (Drs. A. Messias, M. Aguiar); San Raffaele Scientific Institute (Dr G. Landoni, Dr P. Bergonzi); Centre Hospitalier Universitaire - Estaing (Dr S. Perbet, Dr JM Constantin, C. Bigot, Dr R. Chabanne, Dr S. Colomb); Xeral-Cies Universitary Hospital (Dr A. Castro); Hospital Univeristari Mutua de Terrassa (Dr M. Fernandez, A. Pous); Hospital Universitario Araba-sede Txagorritxu (Dr JML. Sanchez); St Vincent’s University Hospital (Dr A. Nichol, E. Meaney, K. Brickell); Istituto di Ricovero e Cura a Carattere Scientifico Azienda Ospedaliera Universitaria San Martino – IST (Dr Pelosi, C. Micalizzi, M. Mandelli); Città della Salute e della Scienza (Dr L. Brazzi, A. Trompeo, Dr L. Amendolia, Dr S. Dalmasso, Dr E. Di Cristofaro, Dr D. Ferrero); Ospedali Riuniti di Foggia (Dr L. Mirabella, Z. Difiore); Azienda Ospedaliera Universitaria Arcispedale Sant’ Anna Ferrara (Dr S. Spadaro, Dr E. Ferri). Australia/New Zealand: Sir Charles Gairdner Hospital (Dr S. Baker, B. Roberts); Royal Hobart Hospital (Dr PD. Cooper, RE. McAllister); Canberra Hospital (Dr F. van Haren, H. Rodgers, M. Nourse); Wellington Hospital (Dr P. Young, L. Andrew, A. Hunt, S. Hurford); Hawke's Bay Hospital (Dr R. Freebairn, L. Chadwick, L. Thomas); Royal Adelaide Hospital (Dr M. Chapman, S. O’Connor, K. Glasby); Royal Prince Alfred Hospital (Dr DJ. Gattas, H. Buhr, D. Hutch); Westmead Hospital (Dr V. Nayyar, C. Skelly, J. Kong); Auckland City Hospital (Dr C. McArther, L. Newby, Y. Chen); Nepean Hospital (Dr N. Nguyen, L. Weisbrodt, R. Gresham); Royal North Shore Hospital (Dr L. Quinn, F. Bass); Barwon Health, The Geelong Hospital (Dr N. Orford, A. Bone); Christchurch Hospital (Dr S. Henderson, JE. Mehrtens); Concord Hospital (Dr M. Kol, Dr D. Milliss, H. Wong); Blacktown Hospital (Dr G. Reece, T. Sara, K. Nand); The Queen Elizabeth Hospital (Dr S. Peake, J. McIntyre, T. Williams); Royal Darwin Hospital (Dr LT. Campbell, J. Thomas); Middlemore Hospital (Dr A. Williams, C. Hogan, R. Song, A. Tisley); Flinders Medical Centre (Dr S. Bihari, E. Matheson, K. Schwartz). US: Creighton University Medical Center (Dr A. Modrykamien, M. Pote); Mayo Clinic - St. Mary’s Hospital (Dr R. Hubmayr, S. Holets); University of Michigan Health System (Dr R. Hyzy, C. Haas, K. Nelson, R. Eakin); Los Angeles County-University of Southern California Medical Center (Dr J. Leibler, J. Zhu, S. Milstein); United Hospital Systems (Dr A. Answar, J. Baughman, J. Rosen); Saint Barnabas Medical Center (Dr P. Yodice, ME. Dimi); Loma Linda University Medical Center (Dr T. O’Callaghan, X. Luo-Owen); Grady Memorial Hospital (Dr E. Honig, S. Lawoyin, LB. Abderson); Freeman Health System (J. Keener); St. John Medical Center (Dr MS. Charles); Spectrum Health Hospital (Dr S. Fitch, A. Dutkiewicz, F. Carrier, A. Boersen, N. Graff); The University of Kansas Hospital (Dr S. Berry, J. Penn, A. Dalton); University of California at Los Angeles Medical Center, San Monica (Dr S. Chang); Charleston Area Medical Center (Dr R. Crisalli, K. Sutphin A. Barnes); Thomas Jefferson University Hospital (Dr F. Rincon, Dr U. Mukhtar, J. Jaeger, J. Reyes); Nebraska Medical Center (Dr K. Buesing, K. Peterson); Swedish Health Services (Dr K. Koo, J. Wallick).

References
1.
Wunsch  H, Linde-Zwirble  WT, Angus  DC, Hartman  ME, Milbrandt  EB, Kahn  JM.  The epidemiology of mechanical ventilation use in the United States.   Crit Care Med. 2010;38(10):1947-1953. doi:10.1097/CCM.0b013e3181ef4460PubMedGoogle ScholarCrossref
2.
Carson  SS, Cox  CE, Holmes  GM, Howard  A, Carey  TS.  The changing epidemiology of mechanical ventilation: a population-based study.   J Intensive Care Med. 2006;21(3):173-182. doi:10.1177/0885066605282784PubMedGoogle ScholarCrossref
3.
Needham  DM, Bronskill  SE, Calinawan  JR, Sibbald  WJ, Pronovost  PJ, Laupacis  A.  Projected incidence of mechanical ventilation in Ontario to 2026: preparing for the aging baby boomers.   Crit Care Med. 2005;33(3):574-579. doi:10.1097/01.CCM.0000155992.21174.31PubMedGoogle ScholarCrossref
4.
Wunsch  H, Wagner  J, Herlim  M, Chong  DH, Kramer  AA, Halpern  SD.  ICU occupancy and mechanical ventilator use in the United States.   Crit Care Med. 2013;41(12):2712-2719. doi:10.1097/CCM.0b013e318298a139PubMedGoogle ScholarCrossref
5.
Melsen  WG, Rovers  MM, Groenwold  RHH,  et al.  Attributable mortality of ventilator-associated pneumonia: a meta-analysis of individual patient data from randomised prevention studies.   Lancet Infect Dis. 2013;13(8):665-671. doi:10.1016/S1473-3099(13)70081-1PubMedGoogle ScholarCrossref
6.
Rengel  KF, Hayhurst  CJ, Pandharipande  PP, Hughes  CG.  Long-term cognitive and functional impairments after critical illness.   Anesth Analg. 2019;128(4):772-780. doi:10.1213/ANE.0000000000004066PubMedGoogle ScholarCrossref
7.
MacIntyre  NR, Cook  DJ, Ely  EW  Jr,  et al; American College of Chest Physicians; American Association for Respiratory Care; American College of Critical Care Medicine.  Evidence-based guidelines for weaning and discontinuing ventilatory support: a collective task force facilitated by the American College of Chest Physicians; the American Association for Respiratory Care; and the American College of Critical Care Medicine.   Chest. 2001;120(6)(suppl):375S-395S. doi:10.1378/chest.120.6_suppl.375SPubMedGoogle ScholarCrossref
8.
Blackwood  B, Burns  K, Cardwell  C, O’Halloran  P.  Protocolized vs. non-protocolized weaning for reducing the duration of mechanical ventilation in critically ill adult patients Cochrane systematic review and meta-analysis.   Cochrane Database Syst Rev. 2014;2014(11):CD006904. doi:10.1002/14651858.CD006904.pub3PubMedGoogle Scholar
9.
Burns  KEA, Soliman  I, Adhikari  NKJ,  et al.  Trials directly comparing alternative spontaneous breathing trial techniques: a systematic review and meta-analysis.   Crit Care. 2017;21(1):127. doi:10.1186/s13054-017-1698-xPubMedGoogle ScholarCrossref
10.
Subirà  C, Hernández  G, Vázquez  A,  et al.  Effect of pressure support vs t-piece ventilation strategies during spontaneous breathing trials on successful extubation among patients receiving mechanical ventilation: a randomized clinical trial.   JAMA. 2019;321(22):2175-2182. doi:10.1001/jama.2019.7234PubMedGoogle ScholarCrossref
11.
Esteban  A, Frutos  F, Tobin  MJ,  et al; Spanish Lung Failure Collaborative Group.  A comparison of four methods of weaning patients from mechanical ventilation.   N Engl J Med. 1995;332(6):345-350. doi:10.1056/NEJM199502093320601PubMedGoogle ScholarCrossref
12.
Burns  KEA, Raptis  S, Nisenbaum  R,  et al; Canadian Critical Care Trials Group.  International practice variation in weaning critically ill adults from invasive mechanical ventilation.   Ann Am Thorac Soc. 2018;15(4):494-502. doi:10.1513/AnnalsATS.201705-410OCPubMedGoogle ScholarCrossref
13.
Esteban  A, Anzueto  A, Frutos  F,  et al; Mechanical Ventilation International Study Group.  Characteristics and outcomes in adult patients receiving mechanical ventilation: a 28-day international study.   JAMA. 2002;287(3):345-355. doi:10.1001/jama.287.3.345PubMedGoogle ScholarCrossref
14.
Esteban  A, Ferguson  ND, Meade  MO,  et al; VENTILA Group.  Evolution of mechanical ventilation in response to clinical research.   Am J Respir Crit Care Med. 2008;177(2):170-177. doi:10.1164/rccm.200706-893OCPubMedGoogle ScholarCrossref
15.
Esteban  A, Frutos-Vivar  F, Muriel  A,  et al.  Evolution of mortality over time in patients receiving mechanical ventilation.   Am J Respir Crit Care Med. 2013;188(2):220-230. doi:10.1164/rccm.201212-2169OCPubMedGoogle ScholarCrossref
16.
Burns  KEA, Rizvi  L, Cook  DJ,  et al; Canadian Critical Care Trials Group.  Variation in the practice of discontinuing mechanical ventilation in critically ill adults: study protocol for an international prospective observational study.   BMJ Open. 2019;9(9):e031775. doi:10.1136/bmjopen-2019-031775PubMedGoogle Scholar
17.
Holm  S,.  A simple sequentially rejective multiple test procedure.   Scand J Stat. 1979:6(2):65-70.Google Scholar
18.
Riker  RR, Picard  JT, Fraser  GL.  Prospective evaluation of the Sedation-Agitation Scale for adult critically ill patients.   Crit Care Med. 1999;27(7):1325-1329. doi:10.1097/00003246-199907000-00022PubMedGoogle ScholarCrossref
19.
Vincent  JL, de Mendonca  A, Cantraine  F,  et al.  Use of the SOFA score to assess the incidence of organ dysfunction failure in intensive care units: results of a multicenter, prospective study.   Crit Care Med. 1998;26(11):1793-1800. doi:10.1097/00003246-199811000-00016PubMedGoogle ScholarCrossref
20.
Tallgren  M, Bäcklund  M, Hynninen  M.  Accuracy of Sequential Organ Failure Assessment (SOFA) scoring in clinical practice.   Acta Anaesthesiol Scand. 2009;53(1):39-45. doi:10.1111/j.1399-6576.2008.01825.xPubMedGoogle ScholarCrossref
21.
Harrell  FE  Jr.  Regression Modeling Strategies: With Applications to Linear Models, Logistic and Ordinal Regression, and Survival Analysis. Springer; 2015.
22.
Van Buuren  S.  Flexible Imputation of Missing Data. Chapman & Hall; 2018.
23.
Van Calster  B, McLernon  DJ, van Smeden  M, Wynants  L, Steyerberg  EW.  Calibration: the Achilles heel of predictive analytics.   BMC Med. 2019;17(1):230. doi:10.1186/s12916-019-1466-7 PubMedGoogle ScholarCrossref
24.
Beduneau  G, Pham  T, Schortgen  F,  et al; WIND (Weaning according to a New Definition) Study Group and the REVA Network.  Epidemiology of weaning outcome according to a new definition: the WIND study.   Am J of Respir Crit Care. 2017;195(6):772-83. doi:10.1164/rccm.201602-0320OCPubMedGoogle ScholarCrossref
25.
Ely  EW, Baker  AM, Evans  GW, Haponik  EF.  The prognostic significance of passing a daily screen of weaning parameters.   Intensive Care Med. 1999;25(6):581-587. doi:10.1007/s001340050906PubMedGoogle ScholarCrossref
26.
Kalassian  KG, Dremsizov  T, Angus  DC.  Translating research evidence into clinical practice: new challenges for critical care.   Crit Care. 2002;6(1):11-14. doi:10.1186/cc1446PubMedGoogle ScholarCrossref
27.
Schultz  MJ, Wolthuis  EK, Moeniralam  HS, Levi  M.  Struggle for implementation of new strategies in intensive care medicine: anticoagulation, insulin, and lower tidal volumes.   J Crit Care. 2005;20(3):199-204. doi:10.1016/j.jcrc.2005.05.007PubMedGoogle ScholarCrossref
28.
Brochard  L, Rauss  A, Benito  S,  et al.  Comparison of three methods of gradual withdrawal from ventilatory support during weaning from mechanical ventilation.   Am J Respir Crit Care Med. 1994;150(4):896-903. doi:10.1164/ajrccm.150.4.7921460PubMedGoogle ScholarCrossref
29.
Tripepi  G, Jager  KJ, Dekker  FW, Zoccali  C.  Selection bias and information bias in clinical research.   Nephron Clin Pract. 2010;115(2):c94-c99. doi:10.1159/000312871PubMedGoogle ScholarCrossref
30.
Ouellette  DR, Patel  S, Girard  TD,  et al.  Liberation from mechanical ventilation in critically ill adults: an official American College of Chest Physicians/American Thoracic Society clinical practice guideline.   Chest. 2017;151(1):166-180. doi:10.1016/j.chest.2016.10.036PubMedGoogle ScholarCrossref
×