INTRODUCTION — Mechanical ventilation can be performed using positive pressure or negative pressure. Positive pressure ventilation is the primary type of mechanical ventilation used today. During positive pressure ventilation, the ventilator forces air into the central airways and the resulting pressure gradient causes airflow into the small airways and alveoli. (See "Overview of initiating invasive mechanical ventilation in adults in the intensive care unit".)
Physiologic and pathophysiologic consequences of positive pressure ventilation are discussed in this topic review. Two major consequences of positive pressure ventilation, pulmonary barotrauma and ventilator-associated lung injury, are reviewed separately. (See "Diagnosis, management, and prevention of pulmonary barotrauma during invasive mechanical ventilation in adults" and "Ventilator-induced lung injury".)
PULMONARY EFFECTS
Barotrauma — Pulmonary barotrauma is a well-known complication of positive pressure ventilation. Consequences include pneumothorax, subcutaneous emphysema, pneumomediastinum, and pneumoperitoneum. Pulmonary barotrauma during mechanical ventilation is discussed separately. (See "Diagnosis, management, and prevention of pulmonary barotrauma during invasive mechanical ventilation in adults".)
Ventilator-associated lung injury — Ventilator-associated lung injury (VALI) refers to acute lung injury that occurs during mechanical ventilation. It is clinically indistinguishable from acute lung injury or acute respiratory distress syndrome (ALI/ARDS) due to other causes. VALI is discussed separately. (See "Ventilator-induced lung injury".)
Auto-PEEP — Auto-positive end-expiratory pressure (auto-PEEP, also called intrinsic PEEP) exists when there is positive airway pressure at the end of expiration due to incomplete exhalation [1]. In other words, inspiration is initiated before expiratory airflow from the preceding breath has ceased. (See "Positive end-expiratory pressure (PEEP)".)
Causes — There are numerous reasons that patients receiving positive pressure ventilation are susceptible to developing auto-PEEP (table 1) [2]:
●High minute volume – High minute volume ventilation exists when there are large tidal volumes (greater than the patient's functional residual capacity), a high respiratory rate, or both. Large tidal volumes increase the volume that must be exhaled prior to the next breath. High respiratory rates decrease the duration of expiration. In both situations, the next breath is initiated prior to completion of the last exhalation. A high minute volume may be due to patient factors (eg, fever, infection) or ventilator settings.
●Prolonged inspiratory time – Prolonging the inspiratory time can be used to improve oxygenation in patients with refractory hypoxemia. When the inspiratory time is increased, there is an obligatory decrease in the expiratory time. This can lead to incomplete exhalation and auto-PEEP [3].
●Time-constant inequality – Patients whose lung units empty heterogeneously (eg, patients with obstructive airways disease) are particularly susceptible to developing auto-PEEP during positive pressure ventilation, even at relatively low minute ventilation.
●Expiratory flow resistance – Resistance to airflow (eg, narrow endotracheal tube, ventilator tubing) can cause auto-PEEP by impairing exhalation.
●Expiratory flow limitation (eg, obstructive airways disease) and altered respiratory system compliance (eg, expiratory muscle activity) similarly impede exhalation, causing auto-PEEP. Altered respiratory system compliance may also interfere with accurate measurement of auto-PEEP [4].
Detection — Auto-PEEP can be identified by applying an expiratory breath hold (usually 0.5 to 1 second) and then directly measuring the airway pressure at the end of the breath hold (figure 1). It can also be identified when ventilator waveforms, auscultation, and/or palpation demonstrate continued expiratory airflow from the preceding breath when the next breath is triggered. In a study of 73 patients (503 observations), physical examination identified auto-PEEP with positive and negative predictive values of 95 and 58 percent, respectively [5]. This indicates that physical examination is useful for confirming auto-PEEP, but not for excluding auto-PEEP.
Consequences — Auto-PEEP exacerbates the hemodynamic effects of positive pressure ventilation (discussed below), increases the risk of pulmonary barotrauma, and makes it more difficult for the patient to trigger a ventilator-assisted breath (figure 2). In addition, auto-PEEP can lead to incorrect estimation of the mean alveolar pressure and static lung compliance [6].
Treatment — Immediate intervention is necessary if significant auto-PEEP is detected (table 2):
●Change ventilator settings – The ventilator settings should be changed in an effort to reduce or eliminate auto-PEEP. The most helpful maneuvers are those that increase the duration of expiration: increasing the inspiratory flow rate, decreasing the respiratory rate, or both. Decreasing the tidal volume or using applied PEEP to overcome auto-PEEP may also be helpful. The use of applied PEEP in this setting is discussed separately. (See "Positive end-expiratory pressure (PEEP)", section on 'Treatment'.)
●Reduce ventilatory demand – Ventilatory demand can be decreased by reducing carbohydrate intake, anxiety, pain, or fever. This may decrease the minute volume, thereby reducing auto-PEEP.
●Reduce expiratory flow resistance – Reduction of expiratory flow resistance by suctioning, administration of bronchodilators, and use of a wide endotracheal tube can reduce auto-PEEP.
Heterogeneous ventilation — The distribution of positive pressure ventilation is never uniform because the amount of ventilation is a function of three factors that vary from region to region within the lungs: alveolar compliance, airway resistance, and dependency (upper versus lower lung zones). Compliant, non-dependent regions with minimal airway resistance will be best ventilated. In contrast, stiff, dependent regions with increased airway resistance will be least ventilated. The heterogeneity of ventilation is accentuated in patients who have both airways disease and parenchymal lung disease.
Ventilation/perfusion mismatch — Mechanical ventilation can alter two opposing forms of ventilation/perfusion mismatch (V/Q mismatch), dead space (areas that are overventilated relative to perfusion; V>Q) and shunt (areas that are underventilated relative to perfusion; V<Q). By increasing ventilation (V), the institution of positive pressure ventilation will worsen dead space but improve shunt.
Increased dead space — Dead space reflects the surface area within the lung that is not involved in gas exchange. It is the sum of the anatomic plus alveolar dead space. Alveolar dead space (also known as physiologic dead space) consists of alveoli that are not involved in gas exchange due to insufficient perfusion (ie, overventilated relative to perfusion). Positive pressure ventilation tends to increase alveolar dead space by increasing ventilation in alveoli that do not have a corresponding increase in perfusion, thereby worsening V/Q mismatch and hypercapnia.
Reduced shunt — An intraparenchymal shunt exists where there is blood flow through pulmonary parenchyma that is not involved in gas exchange because of insufficient alveolar ventilation. Patients with respiratory failure frequently have increased intraparenchymal shunting due to areas of focal atelectasis that continue to be perfused (ie, regions that are underventilated relative to perfusion). Treating atelectasis with positive pressure ventilation can reduce intraparenchymal shunting by improving alveolar ventilation, thereby improving V/Q matching and oxygenation. This is particularly true if PEEP is added. (See "Positive end-expiratory pressure (PEEP)" and "Measures of oxygenation and mechanisms of hypoxemia", section on 'Ventilation-perfusion mismatch'.)
Diaphragm — Mechanical ventilation itself causes diaphragmatic muscle atrophy, a phenomenon called ventilator induced diaphragmatic dysfunction (VIDD). Controlled mechanical ventilation may lead to a very rapid type of disuse atrophy involving the diaphragmatic muscle fibers, which can develop within the first day of mechanical ventilation. An observational study of 22 patients compared the size of diaphragmatic muscle fibers from patients who received positive pressure ventilation for more than 18 hours to those from patients who received positive pressure ventilation for fewer than three hours [7]. The mean cross sectional area of both fast twitch diaphragmatic muscle fibers (1871 versus 3949 micron2) and slow twitch diaphragmatic muscle fibers (2025 versus 4725 micron2) was significantly smaller among those patients who received positive pressure ventilation for a longer duration. These findings were supported by a subsequent study that found that diaphragmatic strength decreased progressively during mechanical ventilation and that long-term mechanical ventilation (defined as >24 hours) was associated with diaphragmatic muscle injury, atrophy, and proteolysis compared to short-term mechanical ventilation (defined as two to three hours) [8]. Patients receiving a greater percentage of controlled ventilation (>25 percent) during the first 48 hours of assist-control mechanical ventilation exhibit a greater degree of narrowing in diaphragmatic thickness [9]. VIDD appears to be mediated by oxidative stress. Diaphragmatic lipid accumulation is seen in conjunction with mitochondrial biogenesis and content down regulation [10]. Speculation exists as to whether these events are triggered by metabolic oversupply. The optimal approach to potentially obviate this phenomenon clinically is unknown [10].
Diaphragmatic atrophy during mechanical ventilation may be associated with prolonged mechanical ventilation, difficulty weaning, prolonged ICU stay, and a higher risk of complications [11]. (See "Management of the difficult-to-wean adult patient in the intensive care unit", section on 'Investigational strategies (inspiratory respiratory muscle training)'.)
Respiratory muscles — Respiratory muscle atrophy can develop in patients undergoing positive pressure ventilation. The mechanism of respiratory muscle weakness is probably similar to that of general neuromuscular weakness in critically ill patients, which is discussed elsewhere. Expiratory muscle weakness occurred in 22 percent of patients during the first week of mechanical ventilation in one series; it was not associated with any specific clinical parameter and did not correlate with diaphragmatic thickness [12]. (See "Neuromuscular weakness related to critical illness".)
Mucociliary motility — Positive pressure ventilation appears to impair mucociliary motility in the airways. In a series of 32 patients, bronchial mucus transport velocity was measured using technetium 99m-labeled albumin microspheres during the first three days of mechanical ventilation [13]. Bronchial mucus transport was frequently impaired and associated with retention of secretions and pneumonia.
The administration of mucolytics is poorly studied and in general, is not routine. One randomized trial of 922 patients receiving mechanical ventilation who were not expected to be extubated within 24 hours reported that routine, compared with as needed, nebulization of the mucolytic, acetylcysteine, together with salbutamol did not reduce the number of ventilator-free days, length of stay, or mortality but did result in more tachyarrhythmias and agitation [14]. Whether the tachyarrhythmias were beta-2 agonist-related is unclear.
SYSTEMIC EFFECTS
Hemodynamics — Positive pressure ventilation frequently decreases cardiac output, which may cause hypotension. There are several mechanisms that contribute to the fall in cardiac output:
●Decreased venous return – The amount of venous return is determined by the pressure gradient from the extrathoracic systemic veins to the right atrium. Intrathoracic and right atrial pressure increase during positive pressure ventilation, thereby reducing the gradient for venous return. This effect is accentuated by auto-PEEP, applied PEEP, or intravascular hypovolemia [15].
●Reduced right ventricular output – Alveolar inflation during positive pressure ventilation compresses the pulmonary vascular bed. This increases pulmonary vascular resistance, thereby reducing right ventricular output. In a study of 21 patients with acute respiratory distress syndrome (ARDS), titrating the PEEP from 5 cm H2O to achieve a plateau pressure of 30 cm H2O was associated with a fall in cardiac output and an increase in right ventricular afterload [16]. This effect was mitigated by increasing the central venous blood volume via a passive leg raise maneuver.
●Reduced left ventricular output – Increased pulmonary vascular resistance can shift the interventricular septum to the left, impair diastolic filling of the left ventricle, and reduce left ventricular output.
In contrast to these adverse effects, positive pressure ventilation may be beneficial in patients with left ventricular failure. Specifically, increased intrathoracic pressure can improve left ventricular performance by decreasing both venous return and left ventricular afterload [17].
These hemodynamic effects are the result of positive airway pressure being transmitted to the surrounding structures of the thorax. The extent to which this occurs varies according to chest wall and lung compliance. Transmission of airway pressure is greatest when there is low chest wall compliance (eg, fibrothorax) or high lung compliance (eg, emphysema); it is least when there is high chest wall compliance (eg, sternotomy) or low lung compliance (eg, ARDS, heart failure).
Monitoring — Another consequence of positive airway pressure being transmitted to surrounding intrathoracic structures is that hemodynamic measurements may be artificially elevated. PEEP plays a particularly prominent role because most hemodynamic measurements are performed at the end of expiration when PEEP is the primary source of positive airway pressure.
The effect of positive pressure ventilation on hemodynamic measures has been best studied using the pulmonary capillary wedge pressure (PCWP), although it appears to have a similar impact on the central venous pressure (CVP). The PCWP is measured by a pulmonary artery catheter (Swan-Ganz catheter). When a patient is receiving positive pressure ventilation, the PCWP is artificially elevated and not reflective of the true transmural filling pressure.
The true transmural filling pressure can be estimated by subtracting one-half of the PEEP level from the PCWP if the lung compliance is normal, or one-quarter of the PEEP level if lung compliance is reduced [18]. As an example, for a patient with normal lung compliance who is receiving a PEEP of 12 cm H2O and whose PCWP is measured as 18 mmHg, the true PCWP is estimated to be 12 mmHg.
A more precise way to estimate the true transmural PCWP in patients requiring positive pressure ventilation utilizes the respiratory related variation of PCWP to estimate the transmission of alveolar pressure to the pulmonary vessels [19]. This measure is called the index of transmission:
Index of transmission =
(end inspiratory PCWP - end expiratory PCWP) / (plateau airway pressure - total PEEP)
Measurement of the plateau airway pressure is described separately. (See "Diagnosis, management, and prevention of pulmonary barotrauma during invasive mechanical ventilation in adults", section on 'Prevention'.)
Once the index of transmission is calculated, the true PCWP can be estimated:
Transmural PCWP =
end-expiratory PCWP - (index of transmission x total PEEP)
This estimate may be unreliable if the respiratory variation of the PCWP is greater than that of the pulmonary arterial pressure tracing (figure 3) [20]. (See "Pulmonary artery catheterization: Interpretation of hemodynamic values and waveforms in adults".)
Gastrointestinal — Positive pressure ventilation for greater than 48 hours is a risk factor for clinically important gastrointestinal bleeding due to stress ulceration. (See "Stress ulcers in the intensive care unit: Diagnosis, management, and prevention".)
Positive airway pressure (especially PEEP) is also associated with decreased splanchnic perfusion [21]. The mechanism underlying this association is unknown, but may be related to decreased cardiac output [22]. Decreased splanchnic perfusion manifests as elevated plasma aminotransferase and lactate dehydrogenase levels.
Other gastrointestinal complications seen in patients receiving positive pressure ventilation include erosive esophagitis, diarrhea, acalculous cholecystitis, and hypomotility [23,24]. It is uncertain whether these complications are due to mechanical ventilation or the critical illness. Hypomotility usually manifests as intolerance to enteral feeding. Correction of electrolytic abnormalities and avoidance of drugs that adversely affect gastric motility (eg, opiates) can improve gastrointestinal motility. While methylnaltrexone has been used intermittently for opioid-induced hypomotility, its routine addition to regular laxatives appears to be of no value [25].
Renal — Mechanical ventilation is associated with the development of acute renal failure. In a prospective cohort study of 29,269 critically ill patients, positive pressure ventilation was an independent risk factor for acute renal failure (odds ratio 2.11, 95% CI 1.58-2.82) [26]. The mechanism for renal injury during positive pressure ventilation is unknown, but it is likely multifactorial. Hypotheses include renal injury through the release of inflammatory mediators (eg, interleukin-6) and impaired renal blood flow due to decreased cardiac output, increased sympathetic tone, or activation of humoral pathways [27].
Central nervous system — Positive pressure ventilation increases intracranial pressure (ICP). This is probably the result of elevated intrathoracic pressure impairing cerebral venous outflow. In an animal model, mechanical ventilation was found to trigger hippocampal neuronal apoptosis. As hippocampal changes are frequently seen in delirious patients undergoing autopsy, this finding raises the possibility of a similar phenomenon in critically ill mechanically ventilated patients [28].
Weakness — Systemic muscular weakness is common among patients who undergo mechanical ventilation. Potential causes include immobilization, prolonged use of sedatives, use of neuromuscular blocking agents, and critical illness. It is unknown whether the mechanical ventilation can cause systemic weakness independently.
Weakness is associated with long-term disability and the need for protracted rehabilitation [29]. However, early mobilization and exercise (ie, physical and occupational therapy) may increase the likelihood that a patient will return to an independent functional status. This was best demonstrated by a trial that randomly assigned 104 patients to either standard care or mobilization plus exercise beginning within 72 hours of the initiation of mechanical ventilation [30]. Patients who received mobilization and exercise were more likely to return to an independent functional status by hospital discharge (59 versus 35 percent, odds ratio 2.7, 95% CI 1.2-6.1).
The causes, diagnosis, and management of systemic weakness in patients in the intensive care unit (ICU) are discussed in detail elsewhere. (See "Neuromuscular weakness related to critical illness".)
Immune system — Positive pressure ventilation appears to induce inflammation. In a randomized trial of 44 patients, patients who received positive pressure ventilation using large tidal volumes and low PEEP had higher concentrations of inflammatory mediators in their blood and bronchoalveolar lavage fluid than patients who received a smaller tidal volumes and high PEEP [31].
Positive pressure ventilation may also promote translocation of tracheal bacteria into the bloodstream, according to one animal study [32]. Translocation was most pronounced during ventilation with large tidal volumes and low PEEP.
Sleep — Disordered sleep is common among patients in the ICU [33,34]. This was illustrated by a prospective cohort study of 20 patients who were being mechanically ventilated for acute lung injury [33]. None of the patients experienced normal sleep according to 24-hour polysomnography. Conventional sleep staging isn't possible in up to a third of awake, unsedated mechanically ventilated patients without delirium. Such patients lack typical sleep stage-2 markers and are felt to have previously uncharacterized states of "atypical sleep" or "pathological wakefulness" with sleep fragmentation and the absence of rapid eye movement (REM) [35]. Patients over the age of 65 years who sleep more during the day than at night may have greater cognitive impairment [36].
Environmental, disease-related, and treatment-related factors have been proposed as possible causes of the disordered sleep seen in critically ill patients. Among these factors, environmental factors may be the least important. This was demonstrated by a study that assessed seven mechanically ventilated patients and six healthy subjects in the intensive care unit using both 24-hour polysomnography and time synchronized environmental monitoring [37]. Patient care activities and noise accounted for less than 30 percent of arousals and awakenings.
The mode of mechanical ventilation may also influence sleep quality, particularly pressure support ventilation [38,39]. Experts suggest avoiding PSV-induced hyperventilation in central apnea especially in those with chronic respiratory and cardiac failure [40]. Similarly, guidelines from the Society of Critical Care Medicine endorse the use of assist-controlled ventilation rather than pressure support ventilation during the night in critically ill patients [41]. Although newer modes such as proportional assist ventilation (PAV) and neurally-adjusted ventilation (NAVA), may improve ventilator asynchrony, their effect on sleep is unknown. For those receiving noninvasive ventilation, administering NIV via the NIV-dedicated ventilator or through the standard ICU- ventilator is appropriate to improve sleep quality.
How to optimize sleep quality in the ICU is unknown since measuring sleep and classifying sleep disordered breathing in ICU patients is difficult and the impact of altering factors such as reducing sedatives, maintaining normal circadian rhythms, and limiting external stimulants are unknown. Thus, such issues need to be addressed before strong recommendations can be made about improving sleep in this population.
Other — Weakness is not the only consequence of prolonged bedrest. Prolonged bedrest has also been associated with insulin resistance [42], venous thromboembolic disease, and joint contractures [43]. During mechanical ventilation, the head of the bed is frequently raised to prevent aspiration and ventilator-acquired pneumonia, which may increase the risk of sacral pressure ulcers by increasing the pressure on the skin in the sacral region [44].
SUMMARY AND RECOMMENDATIONS
●Adverse pulmonary effects of positive pressure ventilation include pulmonary barotrauma, ventilator-associated lung injury, intrinsic positive end expiratory pressure (auto-PEEP), heterogeneous ventilation, altered ventilator/perfusion mismatch (increased dead space, decreased shunt), diaphragmatic muscle atrophy, respiratory muscle weakness, and diminished mucociliary motility. (See 'Pulmonary effects' above.)
●Auto-PEEP exists when there is positive airway pressure at the end of expiration due to incomplete exhalation. It exacerbates the hemodynamic effects of positive pressure ventilation, increases the risk of pulmonary barotrauma, and makes it more difficult for the patient to trigger a ventilator-assisted breath. (See 'Auto-PEEP' above.)
●Detection of auto-PEEP should prompt immediate ventilator setting changes, efforts to reduce ventilatory demand, and efforts to reduce expiratory flow resistance (table 2). The most helpful maneuvers are those that increase the duration of expiration: increasing the inspiratory flow rate, decreasing the respiratory rate, or both. Decreasing the tidal volume or using applied PEEP to overcome auto-PEEP may also be helpful. (See 'Auto-PEEP' above and "Positive end-expiratory pressure (PEEP)", section on 'Applied PEEP'.)
●Positive pressure ventilation may reduce cardiac output and impair hemodynamic monitoring. In addition, it is associated with gastrointestinal stress ulceration, decreased splanchnic perfusion, gastrointestinal hypomotility, fluid retention, acute renal failure, increased intracranial pressure, weakness, inflammation, and disordered sleep. (See 'Systemic effects' above.)
