section name header

Info


A. Indications for Intubation [1]

  1. Noninvasive ventilatory methods may be appropriate and safer in many patients (see below)
  2. Clinical Use of Mechanical Ventilation
    1. Acute respiratory failure (66%)
    2. Coma (15%)
    3. Acute exacerbation of chronic obstructive pulmonary disease (COPD; 13%)
    4. Neuromuscular Disorders (5%)
  3. Acute Respiratory Failure (~10% for each)
    1. Non-cardiogenic pulmonary edema (ARDS)
    2. Cardiogenic Pulmonary Edema (congestive heart failure, CHF)
    3. Pneumonia
    4. Sepsis
    5. Surgery
    6. Trauma
  4. Hypercarbia
    1. COPD
    2. Neuro-muscular disease - respiratory muscle fatigue and/or dysfunction
    3. Drug depressant effects
    4. Endocrinopathies - leading to alveolar hypoventilation
  5. Hypoxemia
    1. Shunt physiology predominant: refractory hypoxemia (including pulmonary embolism)
    2. Pulmonary Edema
    3. Pulmonary Hemorrhage
    4. Pulmonary Infection
    5. Carbon Monoxide Poisoning (may use hyperbaric oxygen)
    6. Asthma
  6. Airway protection
    1. Decreased level of consciousness - Cardiac Arrest, Stroke, Status Epilepticus, Drugs
    2. Increased airway secretions - Toxic Inhalation (airway edema), Infections
  7. Acute Medical and Post-Surgical Cardiovascular Syndromes
    1. Mechanical ventilation is used as an adjunct to decrease work of breathing
    2. Breathing can require a substantial amount of energy expenditure
  8. If there is doubt or moderate concern patients should be intubated or non-invasive ventilation provided
    1. Failure to support ventilation in a timely fashion can lead quickly to respiratory arrest
    2. Elective intubation far more desirable than emergent
    3. Once condition has stabilized, patient may be extubated
  9. Critical and Chronic Illness
    1. Illness and stress associated with increased protein turnover, negative nitrogen balance
    2. Skeletal muscle function comprised by protein breakdown
    3. Respiratory muscle weakness leads to increased problems and complications
    4. High levels of APR especially TNFa are likely etiologies
    5. Treatment of critically ill adults with recombinant human growth hormone to reverse these catablic effects has lead to 1.9-2.4 fold increased mortality [2]

B. Ventilatory Modes [1]

  1. Classification of Modes
    1. Volume Controlled - Assist/Control (A/C), Intermittanent Manditory Ventilation (IMV)
    2. Pressure Controlled - delivers preset airway pressure support (PSV; no specific volume)
  2. Assist / Control (A/C)
    1. Usually selected first
    2. Ventilator delivers breath when triggered by patients inspiratory effort ("assist") OR
    3. independently if such an effort does not occur within a specified time
    4. Volume to be delivered is set with a backup ("control") rate
    5. Patient may initiate a breath with negative inspiratory force (sensed by ventilator)
    6. Ventilator delivers a fixed volume when patient initiates breath
    7. Often more comfortable for the patient, since they initiate the breath
    8. Allows better synchrony between patient and ventilator
    9. Therefore, it is preferred in patients "fighting" the ventilator
    10. If patient does not initiate sufficient number of breaths, then backup rate will kick in
    11. If patient's "spontaneous" rate exceeds control rate, then no "control" breaths are given
    12. This is not a good weaning mode, because patient's contribution to minute ventilation is not known
    13. Higher rates (>8-10 bpm) may lead to Autopeep, especially in COPD, ARDS (see below)
  3. Intermittent Mandatory Ventilation (IMV)
    1. Synchronized IMV (SIMV) is usually used
    2. Set rate and volume of each breath (which is constant)
    3. This is usually the starting mode of ventilation after A/C
    4. Should include baseline pressure support of ~6cm to overcome tube resistance
    5. If the patient takes a breath between ventilator breaths, machine will resynchronize
    6. Allows patient to breath spontaneously without fighting ventilator initiated breaths
    7. Excellent mode for most situations, patient's progress is easily followed
    8. Very good weaning mode
  4. Pressure Support Ventilation (PSV)
    1. Patient triggered and patient cycled modes
    2. Set breath rate and pressure support (usually 10-30cm H20) for each breath
    3. Ventilator delivers a variable volume to the pressure level set
    4. The volume is determined by the mechanics (impedance) of the ventilatory system
    5. Good for patients with COPD to ease the work of breathing
    6. Weaning is more "physiology" than SIMV: that is, low pressure, high volume breathing
    7. Study results conflict about benefit of PSV during weaning phases
  5. Pressure Control Ventilation
    1. Similar to IMV (volume ventilation) except that pressure is set
    2. Thus, ventilator delivers a preset pressure limit with each breath
    3. Breath is time cycled and volume delivered is variable, depending on sytem impedance
    4. This leads to lower peak airway pressure and avoidance of barotrauma
    5. Especially useful in patients with low compliance / high pressures (such as ARDS)
    6. Set airway pressure level, FiO2, PEEP and rate (inspiratory time and flow rates)
  6. Other Methods of Ventilation
    1. High frequency jet ventilation
    2. Extra-corporeal membrane oxygenation (ECMO)
    3. Extra-corporeal carbon dioxide removal (ECCO2R)
    4. Partial liquid ventilation (Perflubron) - premature infants that fail surfactant

C. Considerations when Setting Ventilator

  1. Two crucial parameters are tidal volume (TV) and rate
  2. Tidal Volume (TV) [6]
    1. Minute Ventilation Setting = TV x Rate (alveolar oxygen delivery)
    2. TV originally set at 8-15cc/kg body mass (~1.5-3X normal tidal volume)
    3. Accumulating data support the use of 5-8cc/kg and allow some hypercapnea
    4. Overdistension of lungs leads to significant inflammatory / cytokine responses [24]
    5. These inflammatory responses likely exacerbate pulmonary dysfunction
    6. Recommend lower tidal volumes and inspiratory plateau pressures <28 cm [6,24,26]
    7. Peak lung pressures <30cmH20 in ARDS and emphysema generally recommended [3]
    8. In ARDS, TV 6mL/kg recommended, with plateau pressures <30cc
    9. Weight used is predicted body weight (PDW) based on height, not actual body weight
    10. PDW(men)=50.0+0.91*(height in cm-152.4); PDW(women)=45.5+0.91*(height cm-152.4)
    11. Permissive hypercapnea is increasingly recognized as beneficial in critical patients [3,7]
    12. Maintenance of higher end expiratory pressures may be beneficial
    13. Hypocarbia must be avoided unless brain herniation is being treated
  3. Rate - usually 18-22 breaths per minute; permissive hypercapnea may occur
  4. Oxygen Level
    1. Fraction of inspired oxygen (FiO2)
    2. Usually begin at 100% FiO2 and reduce as tolerated
    3. For patients without vascular disease, low normal pO2 levels are probably beneficial
    4. Elevated arterial oxygen levels over extended periods are detrimental [7]
  5. Inspiratory to Expiratory Ratio (I:E ratio) - usually set ~1:2
  6. Positive End Expiratory Pressure (PEEP)
    1. Application of a fixed amount of positive pressure to mechanically ventilated cycle
    2. Adjunct to above settings (may be added on)
    3. Maintains a fixed amount of pressure in lungs at the end of expiration
    4. Therefore, helps to prevent alveolar collapse, increase lung compliance
    5. Excellent for very "wet" lungs, which have a high tendency toward alveolar collapse
    6. Use in ARDS, cardiogenic shock with pulmonary edema, to increase FRC
    7. PEEP reduces venous return to the heart which may lead to decreased cardiac output
    8. About 30% of patients do not benefit from PEEP, and actually may worsen
  7. PEEP in ARDS [17,24]
    1. Higher PEEP added to low tidal volumes may provide optimal lung protection in ARDS
    2. PEEP levels of 8cm versus 13cm are not associated with outcome differences in ARDS [4]
    3. Patients with >9% of potentially recruitable lung had poorer oxygenation, respiratory system compliance, higher dead space, and higher death rates
    4. Response to PEEP of 15cm or 5cm dependent upon recruitable lung space
    5. Percentage of recruitable lung space should guide selection of PEEP levels, with more PEEP for higher percentage of recruitable lung space in ARDS
    6. In general, tidal volumes in ARDS should be 6cc/kg to minimize overdistension, allow PEEP
    7. Maintain plateau pressures >30cm and probably no higher than 40cm with PEEP [42,43]
    8. With tidal volume 6cc/kg, PEEP set to reach plateau pressure 28-30cm superior to PEEP set at 5-9cm on oxygenation, duration of mechanical ventilation and organ failure [43]

D. Suggested Ventilator SettingsTable: Initial settings for rate and tidal volume for mechanical ventilation

Patient TypeRate breaths/minTidal Vol cc/kgPeak Pressures cmH20
Normal Lungs8-12<8<60
COPD (increased expiratory)<8-10<12<30
Chronic Pulmonary Restriction>12-20<10<30
Severe Acute Lung Injury>12-206<40

E. Adequacy of Ventilation
  1. Arterial Blood Gas (ABG)
    1. Used to monitor patient's status - arterial catheters may be very helpful
    2. Should be correlated with pulse oximetry early on to decrease need for ABG's
    3. Note that pulse oximetry is unreliable in situations with poor peripheral perfusion
    4. pH and pCO2 may be followed in venous blood gases
    5. Venous pH ~ 0.05<arterial pH; Venous CO2 ~ 5-10mm > arterial CO2
    6. Hypocarbia should be avoided in critically ill patients except in brain herniation cases
  2. Permissive Hypercapnia [7]
    1. Deliberate limitation of ventilatory support to avoid lung overdistension
    2. Allowing pCO2 to rise to higher than normal levels (target to 50-100 mmHg)
    3. Permissive hypercapnia helps prevent lung injury from high airway pressures
    4. Permissive hypercapnia may also be safe in ARDS, where lung pressures are high
    5. Permissive hypercapnia may be problematic in increased intracranial pressure
    6. Has led to improved outcomes in patients with ARDS and other disorders [7]
    7. Hypercapnia also has anti-inflammatory effects
    8. Hypercapnia stimulates beneficial oxygenation, cardiovascular and pulmonary effects
  3. Chest Radiograph
    1. Should be obtained initially to check position of endotracheal tube
    2. Should be obtained with any acute change in ventilatory status
    3. May be obtained at any time to recheck tube status and assess for barotrauma
    4. Frequent (eg. daily) radiographs are rarely necessary
  4. Pulmonary Pressures
    [Figure] "Pulmonary Pressure and Volume Changes"
    1. Airway Pressure = PA = VI·Raw + VT/Crs + PEEP
    2. VI is inspiratory flow rate (L/Min)
    3. Raw unitscm / liter / sec = cm·sec/liter, which is airway Resistance
    4. Crs unitsliters / cm., which is lung Compliance (distensibility)
    5. VT is tidal volume Raw·Crs = time constant (seconds)
    6. On inspiration, VT =0, so PA ~ VI·Raw + PEEP: pressure depends on resistance in airway
    7. At end inspiration, VI =0, so PA = VT/Crs , or pressure depends on lung compliance
    8. Therefore, one can estimate airway R and alveolar (lung) compliance separately
    9. In most cases, maximal airway pressure occurs on early inspiration
    10. Thus, PMAX ~ VI·Raw; Plateau Pressure ~ VT/Crs
  5. Pressure Changes
    1. In acute pressure changes, assess whether plateau pressure is increased
    2. Plateau pressure increased - parenchymal disease (loss of volume low compliance)
    3. Loss of volume: pneumonia, pneumothorax, atelectasis
    4. Compliance decreased: pulmonary edema (cardiogenic versus ARDS), pneumonia
    5. If plateau pressure shows little increase, then airway resistance has increased
    6. This is most common in bronchitis, hyperactive airway disease
    7. Failure to completely exhale (eg. in COPD) before next breath leads to pressure buildup
    8. This is "Auto-PEEP" and can be reduced/eliminated by prolonging expiratory time
  6. Pressure Support Level - for pressure support ventilatory mode
    1. Alveolar overdistention can produce cell and membrane damage
    2. In general, plateau pressures should not exceed 35cm H2O
    3. Newer data suggest that even at <35cm pressure, alveolar hyperdistension occurs
    4. Overdistension may be more toxic than hypercapnea, acidosis
  7. Peak pressures should not exceed ~60cm or risk of barotrauma is highly elevated
  8. Sedation
    1. Often required to allow adequate ventilation
    2. Sedatives are often used to reduce fear and anxiety and allow invasive procedures
    3. Sedative anxiolytics reduce stress responses which can be deleterious
    4. Lorazepam (Ativan®) currently recommended for sustained sedation of mechanically ventilated patients
    5. Sedatives should be discontinued once daily until patients are awake [32]
    6. This discontinuation led to reduced time on ventilator and in intensive care unit [32]
    7. Dexmedetomidine, highly selective alpha2-adrenergic receptor agonist, produces sedation and analgesia; showed superior clinical results in mechanically ventilated brain injury [41]
    8. Paralysis is uncommonly required

F. Weaning Parameters

  1. Patients may be weaned off of ventilator in volume (SIMV) or Pressure (PS) Modes
  2. Parameters to predict successful weaning have been developed
  3. Best overall test is the Frequency:Tidal Vol Ratio <105 [1]
    1. Frequency is breaths per minute (spontaneous)
    2. Tidal Volume is given as average in liters (unassisted)
    3. Positive Predictive Value ~75-83%, Neg Pred Value ~95%
    4. Sensitivity >95%, Specificity ~60% (30 minutes into spontaneous breathing)
    5. Ratio is best ascertained 30 minutes after spontaneous breathing begun
  4. Other parameters suggesting weaning (in decreased order of positive and negative predictive values)
    1. Tidal Volume >325cc
    2. Maximum inspiratory pressure < -15cm
    3. Respiratory frequency <39 breaths per minute
  5. Comparison of Weaning Methods [7]
    1. Patients were deamed ready for weaning and randomized to various protocols
    2. Once daily trial of spontaneous breathing led to extubation 2-3X faster than others
    3. Other modes tested were pressure support and IMV (volume support)
    4. No benefit to >1 trial daily of spontaneous breathing
    5. Findings suggest that once or twice daily spontaneous breathing attempted for 3 days
    6. Spontaneous breathing (2 hours) trial in selected patients led to earlier extubation
    7. Noninvasive positive pressure ventilation did not prevent need for reintubation in patients with respiratory failure after extubation [5]
    8. Protocol with coordinated sedation and weaning, "wake up and breathe protocol," with interruption of sedatives results in better outcomes than standard approaches [22]
  6. Factors Which May Impair Weaning
    1. Hypoalbuminemia - malnutrition, severe weight Loss, renal or liver disease
    2. Electrolyte and Acid-Base Disorders
    3. Increased airflow resistance
    4. Medications - sedatives (see above), narcotics, paralytics
    5. Anemia - transfusion or erythropoietin may be beneficial [15]
    6. Limited cardiac reserve
    7. Systemic infection - fever, hypothermia
    8. Hemodynamic Instability
    9. Parasis - muscle weakness and/or neuropathy [10]
    10. CNS Disorders (including iatrogenic)
    11. Gastrointestinal Disorders - ileus, ascites, constipation, diarrhea
    12. Thyroid Disorders - difficult to accurately quantitate
  7. Methylprednisolone 20mg IV beginning 12 hour before extubation, then q4 hours, reduced post-extubation laryngeal edema (3% versus 22%) and retintubation (4% versus 8%) [40]

G. Specific Circumstances

  1. Highly compliant lungs due to tissue destruction (such as emphysema)
    1. Lower initial lung volumes (8-10cc/kg)
    2. Consider pressure support (decrease barotrauma and improve work of breathing)
  2. "Wet" Lungs
    1. Consider diuresis to improve oxygenation
    2. Choose higher tidal volumes ± PEEP to prevent alveolar collapse
    3. PEEP should be tested at low (~5cm) levels initially (may cardiac output)
  3. Acute Increase in Ventilator Pressures
    1. Attempt to determine whether peak, plateau, or both pressures are elevated
    2. Check ABG, Chest Radiograph, ECG (if cardiac ischemia suggested)
    3. Bronchodilator often helpful with increased resistance (consider steroids also)
    4. Consider diuresis, nitrates, and/or PEEP with decreased compliance
  4. Severe Restrictive Lung Disease
    1. Includes respiratory muscle disease and chest wall anomalies
    2. May be treated chronically and effectively wiht non-invasive positive pressure systems

H. Complications of Mechanical Ventilation

  1. Barotrauma (Ventilator Associated Lung Injury) [3]
    1. Particularly affects patients with ARDS and COPD (bullous emphysema)
    2. Risk greatly increased with peak pressure >60cm
    3. High plateau pressures are more predictive than peak pressures for barotrauma
    4. However, pneumothorax and other air leaks to not correlate mortality in ARDS
    5. Target tidal volumes 6mL/kg, pressures <30 cm H20 in ARDS
    6. Disuse atrophy of diaphragmatic muscle fibers occurs within >70 hours of mechanical ventilation [44]
  2. Oxygen Toxicity (including ARDS)
  3. Endotracheal Tube Complications
    1. Laryngeal Injury
    2. Tracheal Stenosis
    3. Tracheomalacia
    4. Sinusitis (nasal airway)
    5. Injury to teeth
    6. Aspiration pneumonia - smaller nasogastric tube size does not reduce risk [23]
  4. Ventilator Associated Pneumonia (see below)
  5. Reducing Antibiotic Resistance in ICU [27]
    1. Antibiotic resistance in the intensive care unit (ICU) is an increasing problem
    2. Most commonly seen in mechanically ventilated patients
    3. The following should be instituted to reduce development of antibiotic resistance:
    4. Limit unnecessary antibiotic administration
    5. Optimize antimicrobial effectiveness
    6. Reduce length of mechanical ventilation (use noninvasive ventilation whenever possible)
    7. Increase vaccination of adults to pneumococcus, influenza virus, and H. influenzae
    8. Reduce duration of therapy (8 versus 15 days) [8]
  6. Hypotension on intubation (exacerbated by sedative agents)
  7. Reduced cardiac output - usually with PEEP (or Auto-PEEP) high airway pressure

I. Ventilator Associated Pneumonia (VAP) [12,14,29]

  1. Risks
    1. Failure to clear secretions properly
    2. Supine position particularly with enteral nutrition
    3. Mechanical ventilation >7days
    4. Glascow coma scale score <9
  2. Detection [14,39]
    1. Physical exam is only minimally beneficial
    2. Changes in gas exchange parameters should always prompt evaluation
    3. Combination of new chest x-ray (CXR) infiltrate with at least 2 of fever, leukocytosis or purulent sputum increases VAP likelihood 2.8X [14]
    4. Absence of new infiltrate on CXR reduces VAP likelihood to 0.35X [14]
    5. Early bronchoscopy with sampling for suspected VAP rather than expectant "clinical" management" reduces antibiotic use, organ dysfunction, and mortality [31]
    6. Bronchoalveolar lavage (BAL) with quantitative culture of BAL fluid or endotracheal aspiration with nonquantitative aspirate culture have similar clinical outcomes [39]
    7. <50% neutrophils on cell counts of lower pulmonary secretions makes VAP <10% likely [14]
    8. Levels of soluble triggering receptor expressed on monocytes (sTREM-1) in BAL fluid can be used to predict presence of VAP [18]
    9. Thus, either BAL or endotracheal aspiration are reasonable for diagnosis of VAP
  3. Prevention
    1. Orotracheal intubation (rather than nasotracheal)
    2. Incidence of pneumonia is reduced 6.8 fold for semi-recumbant versus supine position
    3. Change ventilator circuits only for new patients or if circuit is soiled (not routinely)
    4. Heat and moisture exchangers unless contraindicated
    5. Selective (topical) digestive tract decontamination (SDD) may reduce incidence [9] but clear evidence is lacking [29]
    6. SDD reduced ICU and overall mortality in medical and surgical patients [9]
    7. Sucralfate is generally preferred over H2-blockers for patients at high risk of gastrointestinal bleeding, but clear evidence is lacking
  4. Treatment
    1. Antibiotics to cover mixed flora, particularly hospital acquired infections
    2. Antibiotics for VAP for 8 days as effective as 15 days with less antibiotic use and less development of resistance [8]

J. Noninvasive Ventilation (NIV) [25]

  1. Ventilatory support without placement of endotracheal airway
  2. NIV can be used on the general medicine / respiratory wards
    1. Averts trauma and hazards of invasive mechanical intubation
    2. However, requires trained personnel to initiate and maintain
    3. When non-emergent, patient training may be instituted
  3. Noninvasive positive pressure ventilation (NIPPV) is increasingly used [20,21]
    1. Support both acute and chronic respiratory failure
    2. Usually given through a face mask or nasal mask
    3. Multiple types of masks available, but nasal masks are most comfortable
    4. Air does not escape through mouth because soft palate flops against tongue
    5. Reduces intubation and ICU length of stay in acute respiratory failure [16]
    6. Overall improved mortality and reduced hospital stay with NIPPV versus MV [21]
  4. Clinical Utility
    1. Chronic respiratory failure, including intermittent daily use
    2. Clear benefits in hospitalized patients with severe (but not mild) COPD exacerbation [13]
    3. Congestive heart failure (CHF) / Pulmonary edema - mainly in acute settings [19,38]
    4. CPAP and Bilevel NIPPV showed reduced mortality and reduced need for mechanical ventilation in patients with acute cardiogenic pulmonary edema (ACPE) [38]
    5. CPAP and NIPPV are similarly effective for ACPE [19,38,45]
    6. CPAP and NIPPV similarly effective and superior to oxygen therapy alone with reduced dyspnea and acidosis, but no effect on length of hospital stay in ACPE [45]
    7. NIPPV can be used to help wean COPD patients with respiratory failure [20]
    8. NIV early in course of patients with COPD exacerbation and mild to moderate acidosis reduced mortality, time to pH correction, and normal respirations [13,33]
    9. Improved outcomes in hypercapnic respiratory failure including reduced hospital stay and reduced mortality [21,35]
    10. Effective in sleep apnea syndromes (particularly with concomitant CHF), and in patients with sleep apnea with daytime sleepiness; may be used at night
    11. CPAP is not effective in sleep apnea without daytime sleepiness [28]
    12. NIPPV is superior to endotracheal intubation in solid organ transplant patients with acute hypoxemic respiratory failure [30]
    13. NIPPV did not improve outcomes in acute hypoxemic nonhypercapnic respiratory failure [34]
    14. Reduced incidence of nosocomial pneumonia and other infections in hypercapnic pulmonary edema and COPD exacerbations [35]
    15. NIPPV with PEEP reduced need for intubation and improved respiratory parameters more rapidly than standard oxygen therapy in moderate cardiogenic pulmonary edema [36,38]
    16. Intermittent NIV reduced need for intubation in immunosuppressed patients with pneumonitis, fever, and early acute respiratory failure [37]
    17. NIPPV did not prevent need for reintubation in patients with respiratory failure after extubation [5]
    18. CPAP may reduce incidence of intubation and other complications in patients with hypoxemia after elective major abdominal surgery [11]
    19. Strongly consider NIPPV/CPAP in ALL patients with acute respiratory failure [16,19,25,38]
  5. NIPPV can be controlled in various ways
    1. Volume
    2. Pressure support
    3. Bilateral positive airway pressure (bilevel PAP or BiPAP)
    4. Continuous positive airway pressure (CPAP)
  6. Volume Ventilation
    1. Ventilator delivers set volume (250-500mL or 4-8mL/kg) per breath
    2. Poor tolerance in general, with variable pressures generated
  7. Pressure Ventilation
    1. Pressure support or control (as above), 8-20 cm H20
    2. End-expiratory pressure of 0-6cm H20 (PEEP)
    3. Variable volume
    4. Better tolerated than volume ventilation
    5. BiPAP and CPAP are even more comfortable to most patients
  8. BiPAP
    1. Continuous high flow variable positive airway pressure (PAP)
    2. The PAP varies between low and high pressures
    3. Spontaneous modes have bilevel PAP responding to patient's own flow rates
    4. Thus, inspiration has high flow (pressure), exhalation has low flow (pressure)
    5. Expiratory pressure is equivalent to PEEP
    6. Inspiratory pressure is equal sum of PEEP and pressure support level
    7. Supplemental oxygen is diluted with high air flow through system
    8. BiPAP is usually the best tolerated and most efficient non-invasive system
  9. CPAP
    1. Usually 5-12cm of water, with constant pressures
    2. Volumes vary
    3. Has been shown to improve outcomes in cardiogenic pulmonary edema
    4. Also improves outcomes in patients with COPD exacerbations
    5. Addition of PEEP to CPAP is really BiPAP and is generally preferred over CPAP
    6. Failure rates are 30-50% requiring invasive mechanical ventilation

References

  1. Tobin MJ. 2001. NEJM. 344(26):1986 abstract
  2. Takala J, Ruokonen E, Webster NR, et al. 1999. NEJM. 341(11):785 abstract
  3. Pinhu L, Whitehead T, Evans T, Griffiths M. 2003. Lancet. 361(9354):332 abstract
  4. NHLBI ARDS Clinical Trials Network. 2004. NEJM. 351(4):327 abstract
  5. Esteban A, Frutos-Vivar F, Ferguson ND, et al. 2004. NEJM. 350(24):2452 abstract
  6. Malhotra A. 2007. NEJM. 357(11):1113 abstract
  7. Laffey JG and Kavanagh BP. 1999. Lancet. 354(9186):1283 abstract
  8. Chastre J, Wolff M, Fagon JY, et al. 2003. JAMA. 290(19):2588 abstract
  9. De Jonge E, Schultz MJ, Sapnjaard L, et al. 2003. Lancet. 362(9389):1011 abstract
  10. De Jonghe B, Sharshar T, Lefaucheur JP, et al. 2002. JAMA. 288(22):2859 abstract
  11. Squadrone V, Coha M, Cerutti E, et al. 2005. JAMA. 293(5):589 abstract
  12. Collard HR, Saint S, Matthay MA. 2003. Ann Intern Med. 138(6):494 abstract
  13. Keenan SP, Sinuff T, Cook DJ, Hill NS. 2003. Ann Intern Med. 138(11):861 abstract
  14. Klompas M. 2007. JAMA. 297(14):1583 abstract
  15. Corwin HL, Gettinger A, Pearl RG, et al. 2002. JAMA. 288(22):2827 abstract
  16. Keenan SP, Sinuff T, Cook DJ, Hill NS. 2005. Crit Care Med. 32:2516
  17. Gattinoni L, Caironi P, Cressoni M, et al. 2006. NEJM. 354(17):1775 abstract
  18. Gibot S, Cravoisy A, Levy B, et al. 2004. NEJM. 350(5):451 abstract
  19. Masip J, Roque M, Sanchez B, et al. 2005. JAMA. 294(24):3124 abstract
  20. Nava S, Ambrosino N, Clini E, et al. 1998. Ann Intern Med. 128(9):721 abstract
  21. Antonelli M, Conti G, Rocco M, et al. 1998. NEJM. 339(7):429 abstract
  22. Girard TD, Kress JP, Fuchs BD, et al. 2008. Lancet. 371(9607):126 abstract
  23. Ferrer M, Bauer TT, Torres A, et al. 1999. Ann Intern Med. 130(12):991 abstract
  24. Ranieri VM, Suter PM, Tortorella C, et al. 1999. JAMA. 282(1):54 abstract
  25. Brochard L. 2002. JAMA. 288(8):933
  26. Wyncoll DLA and Evans TW. 1999. Lancet. 354(9177):497 abstract
  27. MH and Fraser VJ. 2001. Ann Intern Med. 134(4):298 abstract
  28. Barbe F, Mayoralas LR, Duran J, et al. 2001. Ann Intern Med. 134(11):1015 abstract
  29. Dodek P, Keenan S, Cook D, et al. 2004. Ann Intern Med. 141(4):305 abstract
  30. Antonelli M, Conti G, Bufi M, et al. 2000. JAMA. 283(2):235 abstract
  31. Fagon JY, Chastre J, Wolff M, et al. 2000. Ann Intern Med. 132(8):621 abstract
  32. Kress JP, Pohlman AS, O'Connor MF, Hall JB. 2000. NEJM. 342(20):1471 abstract
  33. Plant PK, Owen JL, Elliott MW. 2000. Lancet. 355(9219):1931 abstract
  34. Delclaux C, L'Her E, Alberti C, et al. 2000. JAMA. 284(18):2352 abstract
  35. Girou E, Schortgen F, Delclaux C, et al. 2000. JAMA. 284(18):2361 abstract
  36. Masip J, Betbese AJ, Paez J, et al. 2000. Lancet. 356(9248):2126 abstract
  37. Hilbert G, Gruson D, Vargas F, et al. 2001. NEJM. 344(7):481 abstract
  38. Peter JV, Moran JL, Phillips-Hughes J, et al. 2006. Lancet. 367(9517):1155 abstract
  39. Canadian Critical Care Trials Group. 2006. NEJM. 355(25):2619 abstract
  40. Francais B, Bellissant E, Gissot V, et al. 2007. Lancet. 369(9567):1083 abstract
  41. Pandharipande PP, Pun BT, Herr DL, et al. 2007. JAMA. 298(22):2645
  42. Meade MO, Cook DJ, Guyatt GH, et al. 2008. JAMA. 299(6):637 abstract
  43. Mercat A, Richard CJ, Vielle B, et al. 2008. JAMA. 299(6):646 abstract
  44. Levine S, Nguyen T, Taylor N, et al. 2008. NEJM. 358(13):1327 abstract
  45. Gray A, Goodacre S, Newby DE, et al. 2008. NEJM. 359(2):142 abstract