MKSAP primer: Mechanical ventilation
Adapted from ACP's latest Medical Knowledge Self-Assessment Program
From the June ACP Hospitalist, copyright © 2007 by the American College of Physicians.
|
Sidebar: |
Ventilatory failure refers to the inability of the respiratory system to sustain its ventilatory function. Patients in acute ventilatory failure can require noninvasive or invasive mechanical ventilation.
Noninvasive ventilation
Noninvasive positive-pressure ventilation (NPPV) consists of the provision of positive-pressure ventilation without the need for an invasive airway. Consisting of a ventilator that delivers pressurized gas to the upper airway via tubing attached to a mask strapped to the face, NPPV has assumed an important role in the management of acute and chronic respiratory failure. Although not a form of assisted ventilation because it does not actively assist inhalation, continuous positive airway pressure (CPAP) alone, administered noninvasively, reduces the work of breathing in chronic obstructive pulmonary disease (COPD), thereby counterbalancing intrinsic (auto-) positive end-expiratory pressure (PEEP) and effectively reducing or even eliminating the inspiratory threshold load (Figure). By increasing airway pressure during inspiration, pressure support alone also reduces work of breathing in COPD. However, the combination of pressure support and PEEP reduces the work of breathing in patients with COPD more than either modality alone.

Larger image. Chest wall configuration at functional residual capacity of a normal person (left hemithorax) and a patient with severe COPD (right hemithorax). The patient with COPD has a flattened diaphragm, which increases the radius of curvature and increases the tension for a given pressure. This patient's ribs are horizontal and the zone of apposition between the pleural surfaces is reduced, greatly reducing the diaphragm's efficiency in expanding the chest wall. Also, intrinsic PEEP poses an inspiratory load, adding further to inspiratory work. Exhalation is slowed by airway collapse and the loss of elastic recoil. Reproduced with permission from: Hill NS. Current concepts in mechanical ventilation for chronic obstructive pulmonary disease. Semin Respir Crit Care Med. 1999;20(4):375-395.
Indications for NPPV in critical care
Multiple randomized, controlled trials support the recommendation that NPPV be used as the ventilatory modality of first choice in four conditions: COPD exacerbations, cardiogenic pulmonary edema, acute respiratory failure in immunosuppressed patients and early extubation of patients with COPD who required intubation initially. In COPD exacerbations, several meta-analyses have demonstrated more rapid improvements in respiration rate and heart rate, gas exchange and dyspnea, and reductions in the intubation rate (relative risk reduction, 0.42), hospital length of stay and mortality compared to oxygen-treated controls. NPPV reduces the need for intubation and ICU length of stay and mortality in patients with pneumocystis pneumonia and solid organ and bone marrow transplant. This benefit is attributable to the reduction in the occurrence of health care–acquired pneumonia with NPPV compared to endotracheal intubation. NPPV also permits early removal of the endotracheal tube, mainly in patients with COPD, reducing complications related to the tube. For patients with acute pulmonary edema, both CPAP alone and NPPV more rapidly improve gas exchange, vital signs and dyspnea than conventional therapy. Intubations, but not mortality, are also significantly reduced. Neither modality has demonstrated clear superiority over the other.
NPPV can be used for some other causes of respiratory failure, but evidence for efficacy is not as strong. In acute asthma, one controlled trial found more rapid improvements in FEV1 and reduced hospitalization rates compared to sham-treated controls, but a recent meta-analysis concluded that data are insufficient to make firm recommendations. In hypoxemic respiratory failure, some studies have shown significant reductions in intubation and mortality rates, but the category is very broad (including cardiogenic pulmonary edema) and evidence from these should be applied to individual patients with caution. NPPV is not used routinely to treat the acute respiratory distress syndrome (ARDS) unless the patient is an excellent candidate, can be adequately oxygenated noninvasively and has only single organ system failure. NPPV can be used in COPD and congestive heart failure for patients with do-not-intubate status with the expectation that most such patients will survive the hospitalization. NPPV to treat extubation failure is controversial because of a study that examined use to prevent extubation failure and found increased ICU mortality in the NPPV group, thought to be related to delays in needed intubation. In acute respiratory failure of other causes, NPPV can be tried with caution and for some entities, NPPV should not be used at all.
Patient selection
Selection of appropriate patients is key to successful NPPV. The selection factors include the patient's clinical characteristics and risk of failure on NPPV. It becomes a clinical judgment depending largely on physician experience. Predictors of success of NPPV should be considered when selecting patients. The strongest indicator of success is improvement in pH, PaCO2, level of consciousness and respiration rate within the first 2 hours of NPPV initiation. There is a "window of opportunity" for initiating NPPV that opens when the patient needs ventilatory assistance and closes when severe carbon dioxide retention, acidemia, and agitation occur. Therefore, early initiation of NPPV is recommended so that patients have time to adapt and respiratory crises can be averted. However, if begun too early, NPPV might be unhelpful and may waste resources. Selection guidelines suggest first establishing the need for ventilatory assistance according to clinical and blood gas criteria and excluding patients in whom NPPV is contraindicated or likely to fail.
Practical application of NPPV
NPPV requires the selection of a mask, ventilator, and settings and close monitoring in an ICU or step-down unit until the patient stabilizes. Oronasal masks are preferred over nasal masks as the initial choice in the acute setting because they control air leaks through the mouth better and are better tolerated. Either portable pressure-limited "bilevel" or "critical care" ventilators can be used, although the former have limited oxygen-delivering capacity (FIO2 less than 50%) unless they have an oxygen blender. Starting with lower pressures (8 to 12 cm H2O inspiratory/4 to 5 cm H2O expiratory) enhances patient adaptation; the inspiratory pressure can then be adjusted upward as quickly as tolerated to alleviate respiratory distress (up to 20 cm H2O). The difference between inspiratory and expiratory pressure is the pressure support. Monitoring of mask comfort, dyspnea, synchrony with the ventilator, vital signs, continuous oximetry and occasional blood gases is important. Improvements should be seen within 2 hours. If not, intubation should be considered to avoid undue delay and prevent respiratory arrest. Serious complications such as barotrauma are rare because of the relatively low pressures used; the greatest risk is the failure to recognize deterioration and start intubation before respiratory arrest. Common problems include mask discomfort, ulcers of the nasal bridge, gastric insufflation, conjunctivitis and ubiquitous air leaking that may interfere with synchrony and impair effectiveness. Greater institutional experience with NPPV has been associated with lower nosocomial infection and mortality rates.
Invasive mechanical ventilation
Although NPPV should be used for most ventilator starts for patients with COPD, it is appropriate for only about 25% of all patients with acute respiratory failure because most of these patients are not candidates for NPPV on arrival to the hospital and up to a third of good candidates subsequently fail a trial of NPPV. Indications for use of invasive mechanical ventilation for COPD overlap with those for NPPV, which should be used to avoid intubation, not replace it. Ventilator modes traditionally have been targeted to deliver a preset volume or pressure, but some newer hybrid modes can target both to minimize pressure for a given tidal volume. These newer modes have not been demonstrated to be superior to traditional modes, however.
ARDS
The ARDSnet, a consortium supported by the National Institutes of Health to study therapies for ARDS, has heavily influenced the management of invasive mechanical ventilation for ARDS. The ARMA trial, which demonstrated a reduction in ARDS mortality from 40% to 30% with a low (6 mL/kg) rather than a high (12 mL/kg) tidal volume, has established the use of "lung protective" ventilator strategies to avoid ventilator-associated lung injury resulting from excessive stretching of the lung during mechanical ventilation. The ALVEOLI study showed no advantage of a higher PEEP compared to a lower PEEP, both adjusted to maintain adequate oxygenation. Thus, the current recommendation is to use either a volume- or pressure-limited mode with a low tidal volume (6 mL/kg) while monitoring plateau pressure that should be kept less than 30 cm H20. PaCO2 is allowed to rise if necessary to achieve these goals (permissive hypercapnia), and PEEP is adjusted to maintain a FIO2 of 60% or less with a SaO2 greater than 88%. If hypoxemia remains a problem despite these interventions, prone positioning or high frequency oscillation are sometimes used, although no studies have yet demonstrated improved outcomes with these modalities other than improved oxygenation.
COPD
Volume-limited or pressure-limited modes can be used, but volume-limited assist/control is the most frequent initial choice. The challenge in ventilating COPD patients is to avoid excessive minute volume that contributes to dynamic hyperinflation (auto-PEEP) and alkalemia that results from the compensatory metabolic alkalosis for chronic hypercarbia. Tidal volume should be kept small (for example, 5 to 7 mL/kg ideal body weight), and backup respiratory rate should be set between 10 and 14 breaths/min. A lower rate increases the cycle time, which permits more time for exhalation and emptying of the lung. Shortening inspiratory time is another way to increase expiratory time, accomplished by increasing the inspiratory flow rate. This strategy is not as fruitful as lowering respiratory rate because substantial increases in flow rate result in only minor increases in expiratory time, and flow rates above 60 L/min may add to patient discomfort.
Asthma
Invasive mechanical ventilation in patients with acute ventilatory failure due to asthma is to be avoided because of the frequency and severity of the potential complications. These include pneumothorax and pneumomediastinum, associated with mortality rates as high as 10%. If invasive mechanical ventilation is necessary, the approach is similar to that used for COPD patients. Excessive respiratory rates and tidal volumes should be avoided, plateau pressures should be kept less than 30 to 35 cm H2O and "permissive hypercapnia" is used, which was first described in invasively ventilated asthma patients. When ventilation remains difficult, heliox or even general anesthesia with bronchodilator anesthetics may be tried.
Weaning from invasive mechanical ventilation
Minimizing the duration of mechanical ventilation is desirable to reduce potential complications. When patients no longer require high levels of oxygen (SaO2 greater than 89% on FIO2 of 40% or less), are hemodynamically stable and are not excessively sedated, spontaneous breathing trials using a T-piece or low levels of CPAP or pressure support should be initiated. If the patient can comfortably tolerate a trial of 30 to 120 minutes without excessive tachypnea, hemodynamic instability or oxygen desaturation, extubation should be undertaken as long as the patient has an adequate cough without excessive airway secretions. For patients who fail to respond to T-piece trials, daily spontaneous breathing trials or gradual reductions in pressure support may be used to facilitate weaning, but use of synchronized intermittent mandatory ventilation alone without pressure support is discouraged. Routine use of weaning protocols has been shown to expedite weaning, as has daily interruption of sedation. If the patient fails to meet extubation criteria but is otherwise a good candidate for noninvasive ventilation, extubation to noninvasive ventilation should be considered.
Failure to wean
About 5% to 10% of intubated patients fail to wean after the first 7 days of mechanical ventilation, and these difficult-to-wean patients consume disproportionate amounts of ICU resources. These patients have multiple comorbidities and an anticipated mortality rate of up to 50%. Tracheostomies are often done in these patients, particularly if the anticipated duration of intubation exceeds 3 weeks. Tracheostomies can be done surgically or percutaneously, the latter leading to more rapid placement, fewer peri-operative bleeding complications and lower costs if the procedure is performed at the bedside. There is interest in performing earlier tracheostomies with prompt transfer out of the ICU to a long-term acute care facility for continued weaning. This approach may shorten the duration of intubation, but the overall superiority of this practice has not yet been established.
Key points
- Noninvasive ventilation has become an important part of ventilatory assistance in ICUs for COPD, acute pulmonary edema and immunosuppressed patients
- In patients with acute lung injury and ARDS treated with invasive mechanical ventilations, lung protective strategies, consisting of low tidal volume (6 mL/kg) and limited plateau pressure (less than 30 cm H2O), should be used.
- In intubated patients with COPD, ventilator settings should minimize adverse effects of auto-PEEP.
More from ACP Press
IM Essentials for Clerkship Students: Augment your learning with ACP's authoritative text and online supplemental resources.
Introducing ACP Summer Session
| The most popular learning formats from the Internal Medicine meeting offered at only $35 for ACP members! Combine CME and leisure time in San Francisco, CA or Orlando, FL | ![]() |


