# 04 - 313 Mechanical Ventilatory Support

### 313 Mechanical Ventilatory Support

patients who survive regain nearly normal lung function. Patients usu­
ally recover maximal lung function within 6 months. One year after 
endotracheal extubation, more than one-third of ARDS survivors have 
normal spirometry values and diffusion capacity. Most of the remain­
ing patients have only mild abnormalities in pulmonary function. 
Unlike mortality risk, recovery of lung function is strongly associated 
with the extent of lung injury in early ARDS. Low static respiratory 
compliance, high levels of required PEEP, longer durations of mechani­
cal ventilation, and high lung injury scores are all associated with less 
recovery of pulmonary function. Of note, when physical function is 
assessed 5 years after ARDS, exercise limitation and decreased physical 
quality of life are often documented despite normal or nearly normal 
pulmonary function. When caring for ARDS survivors, it is important 
to be aware of the potential for a substantial burden of psychological 
problems in patients and family caregivers, including significant rates 
of depression and posttraumatic stress disorder. Investigations into 
sequelae of COVID ARDS have provided additional insight into longterm ICU outcomes. Please see Chap. 205 for information regarding 
COVID prognosis and recovery.

PART 8
Critical Care Medicine
■
■FURTHER READING
ARDS Definition Task Force: Acute respiratory distress syndrome: 
The Berlin definition. JAMA 307:2526, 2012.
ARDS Network: Ventilation with lower tidal volumes as compared 
with traditional tidal volumes for acute lung injury and the acute 
respiratory distress syndrome. N Engl J Med 342:1301, 2000.
Bellani G et al: Epidemiology, patterns of care, and mortality for 
patients with acute respiratory distress syndrome in intensive care 
units in 50 countries. JAMA 315:788, 2016.
Dequin P-F et al: Hydrocortisone in severe community-acquired 
pneumonia. N Engl J Med 388:1931, 2023.
Gorman EA et al: Acute respiratory distress syndrome in adults: 
Diagnosis, outcomes, long-term sequelae, and management. Lancet 
400:1157, 2022.
Matthay MA et al: A new global definition of acute respiratory dis­
tress syndrome. Am J Respir Crit Care Med 209:37, 2024.
Munshi L et al: Noninvasive respiratory support for adults with acute 
respiratory failure. N Engl J Med 387:1688, 2022.
The National Heart, Lung, and Blood Institute Petal Clinical 
Trials Network: Early neuromuscular blockade in the acute respi­
ratory distress syndrome. N Engl J Med 380:1996, 2019.
■
■WEBSITES
ARDS Foundation: www.ardsusa.org
ARDS Support Center for patient-oriented education: www.ards.org
National Health, Lung, and Blood Institute ARDS Clinical Trials infor­
mation: www.ardsnet.org and www.petalnet.org
Scott Schissel

Mechanical Ventilatory 

Support
Mechanical ventilation refers to devices that deliver positive pres­
sure gas, of varying oxygen content, to patients with acute or chronic 
respiratory failure. Hypoxemic respiratory failure refractory to supple­
mental oxygen and requiring mechanical ventilation is most often 
due to ventilation-perfusion mismatch or shunt caused by processes 
such as pneumonia, pulmonary edema, alveolar hemorrhage, acute 
respiratory distress syndrome (ARDS), and sequelae of trauma or 
surgery. Hypercapnic respiratory failure is most frequently caused by 
severe exacerbations of obstructive lung disease, including asthma and 

chronic obstructive pulmonary disease (COPD); loss of central respi­
ratory drive from acute neurologic events, such as stroke, intracranial 
hemorrhage, or drug overdose; and respiratory muscle weakness from 
diseases such as Guillain-Barré syndrome. Mechanical ventilation may 
also be necessary if patients have an artificial airway placed (an endo­
tracheal tube) due to poor airway protection, such as in coma or in the 
context of a large upper gastrointestinal hemorrhage and vomiting, or 
due to processes leading to large airway obstruction, such as laryngeal 
edema. Finally, since mechanical ventilation can lower the work of 
breathing compared to spontaneous ventilation, it is a useful adjunct 
therapy for shock and multiorgan system failure.
PRINCIPLES OF MECHANICAL 
VENTILATION
Although contemporary mechanical ventilators use positive pressure 
to inflate the lungs, a patient’s response to pressure applied across the 
lung (transpulmonary pressure) depends on the elastic properties of 
their lungs and chest wall; the amount of pressure needed to inflate a 
lung is the same, therefore, whether applied positively via mechanical 
ventilation or negatively using the diaphragm and chest wall muscles. 
In ARDS, for example, lungs are “stiff” or poorly compliant and often 
require much more pressure to achieve a physiologic tidal volume 
(Fig. 313-1), which, over time, may lead to respiratory muscle fatigue. 
If a patient with ARDS is on mechanical ventilation and makes no 
spontaneous respiratory effort, using sedation and neuromuscular 
blockade, the amount of positive pressure needed to inflate the lung 
is equal to the negative inflation pressure required if the patient were 
spontaneously breathing; however, the work of breathing is removed 
on a ventilator, allowing for sustainable ventilation.
Mechanical ventilation can be lifesaving by restoring adequate oxy­
genation and correcting hypercapnia. Optimal application of positivepressure ventilation, however, requires avoiding underinflation, which 
can cause cycles of alveolar recruitment then collapse and, at the 
other extreme, alveolar overinflation (Fig. 313-2); collectively, these 
processes can cause ventilator-induced lung injury by barotrauma and 
volume trauma. Optimal tidal volume ventilation occurs along the lung 
pressure–volume curve where respiratory system compliance is great­
est, or where the smallest change in applied pressure leads to the great­
est increase in lung volume (Fig. 313-2, shaded box). To prevent too 
low lung volumes at end-exhalation, where alveolar collapse occurs, the 
ventilator can be set to maintain a specified positive pressure at endexhalation, or positive end-expiratory pressure (PEEP) (Fig. 313-2B). 
Lower tidal volume ventilation (goal 6 mL/kg of ideal body weight) 

5 L
Normal

Volume (liters and % total lung capacity)

2.5 L

ARDS

0.5 L

–30
–20
–10

Pressure (cmH2O)
FIGURE 313-1  Hypothetical pressure-volume curves of patients with normal lung 
function (normal) and acute respiratory distress syndrome (ARDS). A tidal volume 
breath of 0.5 L in the normal lung requires 8 cmH2O of pressure (open box), but in 
ARDS requires 28 cmH2O (shaded box).

5 L

Volume (liters and % total lung capacity)

B.
Optimal PEEP:
20 cmH2O
D.
Alveolar
overdistension
2.5 L

ARDS

C.
Protective
ventilation
A.
Alveolar
collapse

0.5 L

–30
–20
–10

Pressure (cmH2O)
FIGURE 313-2  Hypothetical pressure-volume curve of a patient with acute 
respiratory distress syndrome (ARDS), demonstrating optimal positive endexpiratory pressure (PEEP) and protective ventilation. A tidal volume breath of 0.5 L 
initiated at a PEEP of 20 cmH2O (B), after the area of greatest alveolar collapse (A). 
End inhalation occurs within the most compliant portion of the pressure-volume 
curve (C) and at a pressure <30 cmH2O, before the area where lung overdistention 
occurs (D), minimizing lung injury.
can help prevent the end-inhalation, or “plateau,” pressure (measured 
just after flow stops at end-inhalation) from exceeding 30 cmH2O; this 
approach minimizes barotrauma and volume trauma–induced lung 
injury, especially in ARDS patients (Fig. 313-2C).
MECHANICAL VENTILATION MODES
Mechanical ventilation entails controlling or monitoring the same 
basic variables involved in spontaneous, negative-pressure breathing, 
including respiratory rate, tidal volume (VT), inspiratory flow rate 
and time, and the fraction of inspired oxygen (Fio2). In addition, the 
PEEP is a variable specific to positive-pressure ventilation and set by 
the clinician. The mechanical ventilation mode determines how much 
control the clinician and ventilator have over these variables versus the 
patient; for example, assist control (AC) mode allows for essentially 
complete operator control of all variables, whereas pressure support 
TABLE 313-1  Key Features of Commonly Used Mechanical Ventilation Modes
VARIABLES SET BY CLINICIAN 
(INDEPENDENT)
MONITORED VARIABLES 
(DEPENDENT)
ADVANTAGES
DISADVANTAGES
MODE
Assist control–volume 
control
VT
Respiratory rate
PEEP
Fio2
Inspiratory flow rate
Peak inspiratory airway 
pressure
End-inhalation (plateau) 
pressure
VE
Assist control–pressure 
control
Inspiratory driving pressure
Respiratory rate
PEEP
Fio2
Tidal volume
VE
Pressure-regulated 
volume control
VT
Respiratory rate
PEEP
Fio2
Peak inspiratory airway 
pressure
End-inhalation (plateau) 
pressure
VE
Pressure support
Inspiratory pressure
PEEP
Fio2
VT
Respiratory rate
VE
Abbreviations: Fio2, fraction of inspired oxygen; PEEP, positive end-expiratory pressure; VE, minute ventilation; VT, tidal volume.

(PS) permits the patient to control important variables, such as respira­
tory rate, VT, and flow rates (Table 313-1).

■
■ASSIST CONTROL VENTILATION
AC allows the clinician to control nearly all ventilator variables and 
is widely used when patients cannot safely participate in their own 
ventilatory efforts, such as when deeply sedated or unstable from acute 
respiratory failure or other critical illness. Most AC ventilation is in 
volume control mode where the operator sets a specific VT and respira­
tory rate, thereby assuring a minimum minute ventilation (VE). In addi­
tion to the set rate, patients can get additional, fully supported breaths 
at the set VT by making an inspiratory effort, which is sensed by the 
ventilator and triggers the breath. The inspiratory flow rate is set by the 
operator; thus, a dyspneic patient may meet resistance on inhalation if 
their desired flow rate is higher than the set rate, possibly leading to 
patient distress and increased work of breathing. In AC volume mode, 
the operator also sets the PEEP and Fio2. Importantly, since VT is an 
independent variable in volume control (i.e., set by the clinician), the 
end-inhalation (or plateau) pressure is a dependent variable not con­
trolled by the clinician but rather determined by the compliance of the 
lung. Inspiratory pressures must be monitored, therefore, to minimize 
barotrauma.
CHAPTER 313
Mechanical Ventilatory Support 
Although AC is often volume controlled, it can be used in a pressure 
control mode, also referred to as pressure control ventilation (PCV). 
The key difference between volume control and PCV is that an inspira­
tory (or “driving”) pressure is set instead of a tidal volume in PCV; thus, 
every time the ventilator delivers a breath, it raises the airway pressure 
to the set amount above PEEP until inspiratory flow decreases below a 
set threshold, therefore ending inhalation. The resulting tidal volume 
will, therefore, vary depending on the compliance of the lung. In a 
sedated and paralyzed patient (making no respiratory effort), the pres­
sure required to generate a specific tidal volume (x) using PCV should, 
in the same patient, be equal to the plateau pressure in volume control 
mode where tidal volume is set at x. Importantly, since lung compliance 
can change dynamically, tidal volume may also change with PCV; tidal 
volume and minute ventilation, therefore, must be monitored since 
there is no assurance of delivered ventilation volumes as with volume 
control. PCV is often used to limit peak airway and lung distending 
(plateau) pressures in situations where high pressure can cause harm, 
such as in ARDS or after thoracic surgery with fresh suture lines in the 
airways or lung parenchyma. Importantly, however, inspiratory flow 
rate and volume are dependent variables in PCV, unlike in volume 
control ventilation, and not set by the clinician. Spontaneously breath­
ing patients on PCV can generate a relative negative pressure in the 
Guarantee minimum VT and VE
Control VT, limiting volume trauma
Barotrauma from high plateau pressure
Patient-ventilator dyssynchrony, 
increased work of breathing
Limit barotrauma (if patient 
respiratory efforts minimal)
Inspiratory flow can vary with 
patient effort (improved comfort/
synchrony)
Vt and VE not mandated; must monitor 
closely
Patient’s respiratory effort can lead to 
large VT and volume trauma
Patient effort can vary inspiratory 
flow, increasing comfort, and 
ventilator synchrony
Guarantee minimum VT and VE
Variable patient effort can lead to VT 
larger than set VT; monitor to prevent 
volume trauma
Patient effort preserved and 
controls VT, inspiratory flow, and 
respiratory rate, allowing for 
ventilator synchrony
Apnea and hypoventilation possible; 
must monitor respiratory rate, VT, and 
VE closely

5 L

Volume (liters and % total lung capacity)

2.5 L

B

A

0.5 L

PART 8
Critical Care Medicine

–30
–20
–10

Pressure (cmH2O)
FIGURE 313-3  Hypothetical pressure-volume curve of a patient on pressure control 
ventilation, paralyzed (A) and breathing spontaneously (B). (A) Paralyzed patient 
(light shaded box): positive end-expiratory pressure (PEEP), 10 cmH2O; inspiratory 
(driving) pressure, 15 cmH2O; end-inhalation (plateau) pressure, 25 cmH2O; tidal 
volume (VT), 300 mL. (B) Breathing patient (dark shaded box): PEEP, 10 cmH2O; 
inspiratory (driving) pressure, 15 cmH2O; patient effort (negative “pulling” pressure), 
10 cmH2O; end-inhalation (plateau) pressure displayed on ventilator, 25 cmH2O; net 
end-inhalation (transalveolar) pressure, 35 cmH2O; VT, 700 mL.
ventilator circuit, transiently decreasing the positive pressure below the 
set point; the ventilator responds by increasing gas flow until it restores 
the set pressure, resulting in higher inspiratory flow rates, a higher 
tidal volume for that breath, and importantly, increased pressure across 
the alveoli, equal to the absolute (negative) pressure generated by the 
patient plus the positive pressure set by the clinician (Fig. 313-3). 
Since mechanical ventilators do not routinely measure or graphically 
display the negative pressure generated by the patient, clinicians can be 
unaware of this additional transalveolar pressure and potential harm 
by volume and barotrauma; importantly, therefore, clinicians should 
monitor for increases in tidal volume on PCV.
■
■PRESSURE-REGULATED VOLUME 

CONTROL VENTILATION
Advances in ventilator technology, such as flow and pressure sen­
sors and microprocessors, allow for additional modes of mechanical 
ventilation that meld the benefits of volume and pressure control 
ventilation. Pressure-regulated volume control (PRVC) ventilation is 
a fully supported mode of ventilation where the clinician sets a target 
tidal volume, as in volume control ventilation, but it allows a patient 
to make spontaneous respiratory efforts and vary inspiratory flow 
rates, as in PCV, enhancing patient comfort and ventilator synchrony. 
PRVC senses patient inspiratory efforts and delivers the least amount 
of positive pressure to achieve the targeted tidal volume; since patient 
efforts can vary and ventilator adaptation is not instantaneous, tidal 
volumes can vary from breath to breath on PRVC. In disease states 
where tidal volume needs tight control to prevent volume trauma, such 
as in ARDS, PRVC must be used cautiously if the patient can make 
significant respiratory effort.
■
■PRESSURE SUPPORT VENTILATION
Pressure support ventilation (PSV) and PCV are very similar except 
there is no mandated ventilation or set mechanical respiratory rate on 
PSV, and ventilator support is entirely patient triggered and controlled. 
The clinician sets the Fio2, PEEP, and maximum inspiratory pressure. 
When patients make a negative-pressure inspiratory effort on PSV, the 
ventilator senses this pressure change and increases positive pressure 
to the set inspiratory pressure level, maintaining it until flow decreases 
below a set threshold (often ~20% of peak inspiratory flow); at this 
point, inhalation ends and pressure drops back to the set PEEP. The 

tidal volume on PSV is monitored but not assured, is determined by 
lung compliance, and depends on the patient’s sustaining an inspira­
tory effort. Tidal volume, minute ventilation, and respiratory rate, 
therefore, must be closely monitored on PSV to detect hypopnea/apnea 
and hypoventilation. PSV is often used when patients are less sedated 
and able to participate in respiratory work, such as when transitioning 
off mechanical ventilation or on a ventilator only for airway support.
■
■NONINVASIVE POSITIVE PRESSURE 
VENTILATION
Noninvasive ventilation (NIV) is historically referred to positivepressure ventilation and is delivered via a nasal or full-face mask at 
a continuous pressure (continuous positive airway pressure [CPAP]) 
or at different inspiratory and expiratory pressures (bi-level positive 
airway pressure [BiPAP]). Most current noninvasive ventilators, how­
ever, can function in full support modes, including volume control 
ventilation. NIV is particularly beneficial for acute respiratory failure 
where the underlying cause responds quickly to treatment, minimizing 
the need for prolonged mechanical ventilatory support. For moderate 
acute hypercarbia, blood pH between 7.25 and 7.35, due to exacerba­
tions of chronic obstructive pulmonary disease (COPD), NIV, for 
example, reduces the need for endotracheal intubation and shortens 
hospital length of stay; more severe acute respiratory acidosis from 
COPD exacerbations (blood pH <7.2) generally requires mechanical 
ventilation with an endotracheal tube. NIV can also be an important 
adjunct treatment for respiratory failure from acute cardiogenic pul­
monary edema, where interventions, such as diuresis and vasodilator 
therapy, can rapidly improve gas exchange and respiratory mechanics. 
NIV, particularly with volume support modes, is effective in managing 
chronic respiratory failure from restrictive lung diseases, such as severe 
scoliosis and respiratory muscle weakness, and in COPD complicated 
by chronic hypercapnia, where nocturnal NIV reduces COPD-related 
hospital admissions. Despite the technical innovations in NIV and 
expanding clinical applications, several important contraindications to 
using mechanical ventilation without a secure airway, such as an endo­
tracheal tube or tracheostomy tube, include delirium, difficulty manag­
ing respiratory secretions, and hemodynamic instability (Table 313-2).
STRATEGIES TO OPTIMIZE GAS EXCHANGE 
ON MECHANICAL VENTILATION
■
■ARTERIAL OXYGENATION
The optimal partial pressure of arterial oxygen (Pao2) and oxygen 
saturation measured by pulse oximetry (Spo2) during mechanical 
ventilation remain uncertain. Although tissue hyperoxia can cause 
oxidative injury with some clinical studies of mechanically ventilated 
patients suggesting worse clinical outcomes with higher Fio2 and when 
Pao2 frequently reaches supraphysiologic levels, randomized studies 
comparing conservative oxygen delivery to a more liberal oxygen strat­
egy have not demonstrated a clear advantage to conservative oxygen 
delivery. In ARDS, targeting a lower Pao2 of 55–70 mmHg (or Spo2 of 
88–92%) versus a higher, but more physiologic, Pao2 of 90–105 mmHg 
(or Spo2 >96%) did not lower mortality, with adverse events being 
more frequent in the lower Pao2 group, including mesenteric ischemia. 
Pao2 and Spo2 targets, therefore, should be individualized to patients 
considering circumstances where even mild hyperoxia may be harm­
ful, such as in recovery from ischemic brain injury, and, conversely, 
where lower Pao2 levels (<55–70 mmHg) may be less optimal, such as 
in patients with ARDS and evidence of bowel dysfunction. Regardless 
TABLE 313-2  Common Contraindications to Noninvasive Ventilation
Inability to protect the airway, such as severe encephalopathy
High risk for aspiration, such as vomiting or severe upper gastrointestinal 
bleeding
Difficulty clearing respiratory secretions
Facial trauma or surgery
Upper airway obstruction or compromise
Significant hemodynamic instability

of the approach, there is no evidence that a supraphysiologic Pao2 
(>100 mmHg) has clinical benefit; thus, sustained hyperoxia should 
be avoided.
Arterial hypoxemia refractory to standard mechanical ventilation 
techniques is common in severe acute lung disease, especially ARDS. 
In general, if the Fio2 requirement is >0.6 or the Pao2:Fio2 ratio is 
<150 mmHg, additional interventions should be considered to improve 
arterial oxygenation. The application of adequate PEEP to prevent 
alveolar collapse during exhalation improves oxygenation by decreasing 
ventilation/perfusion (V. /Q. ) mismatch and shunt in areas of atelectatic 
lung. PEEP should ideally be set at the lower inflection point of the 
most compliant region of the lung pressure-volume curve (Fig. 313-2B). 
Although optimal PEEP may improve arterial oxygenation, achieving 
best PEEP has not been shown to improve clinical outcomes defini­
tively and may have deleterious effects, including barotrauma with 
pneumothorax and hypotension from decreasing venous return to the 
right ventricle. Patients with refractory hypoxemia are often dyspneic 
on mechanical ventilation and make significant respiratory efforts 
dyssynchronous with the ventilator despite deep sedation, leading to 
poor ventilation and preventing optimal V./Q. matching. In this context, 
neuromuscular blockade can be very effective at restoring effective 
mechanical ventilation and optimizing gas exchange. Although a nec­
essary intervention at times, neuromuscular blockade does not improve 
overall outcomes in ARDS, can contribute to critical illness myopathy, 
and requires adequately deep sedation to prevent conscious paralysis; 
thus, it should be used only when necessary to treat refractory hypox­
emia. In ARDS, diseased lung is predominantly dependent, and placing 
the patient in a prone position for extended periods can significantly 
improve arterial oxygenation. The role of prone positioning in other 
disease states is unknown and can be associated with adverse events 
unless performed by a trained team, such as dislodging endotracheal 
tubes and central venous catheters. Delivery of pulmonary vasodilator 
medications through the airway can improve perfusion to ventilated 
alveolar units, therefore improving V./Q. matching and arterial oxygen­
ation. Inhaled prostacyclins, such as epoprostenol, and nitric oxide are 
commonly used to treat refractory hypoxemia and can increase, on 
average, the Pao2:Fio2 ratio by 20–30 mmHg. Hypoxemia refractory to 
these multiple interventions may require consideration of transitioning 
to extracorporeal membrane oxygenation (ECMO; see below).
■
■HYPERCAPNIA
Except for rare circumstances of excess CO2 production (VCO2), which 
can occur in the setting of fever, sepsis, overfeeding, and thyrotoxicosis, 
most hypercapnia is due to inadequate alveolar ventilation (VA) from 
an increase in the fraction of dead space (VD) [the volume of each 
breath not participating in CO2 exchange] relative to the total minute 
ventilation (VE), expressed as VA = VE (1 – VD/VT). Normal physiologic 
dead space is approximately 150 mL (~2 mL/kg), making the VD/VT 
for a 500-mL tidal volume breath 0.3. In acute respiratory failure due 
to ARDS, for example, VD may increase due to poorly perfused but 
ventilated portions of lung while ventilation strategies lead to low VT; 
thus, a modest increase in VD to 200 mL and a low VT of 300 mL will 
result in a VD/VT of 0.66, a situation where hypercapnia may easily 
develop. Hypercapnia in the context of low tidal volume (6 mL/kg) 
ventilation for ARDS often causes acute respiratory acidosis that can be 
managed with higher respiratory rates, up to 30 breaths/min. Respira­
tory acidosis is often tolerated down to a pH of 7.2, so-called “permis­
sive hypercapnia,” but progressive acidosis may require intravenous 
alkalinizing therapy (e.g., sodium bicarbonate or tromethamine) or 
accepting an increase in VT. In severe exacerbations of obstructive lung 
disease, COPD, and status asthmaticus, hypercapnia and acute respira­
tory acidosis are common despite mechanical ventilation, with average 
Paco2 values of 65 mmHg and blood pH of 7.20 after initial endotra­
cheal intubation. Poor alveolar ventilation is primarily due to dead 
space created by alveolar capillary compression in areas of alveolar 
overdistension and lung hyperinflation. Increasing minute ventilation 
by increasing the respiratory rate or tidal volume will, therefore, often 
paradoxically worsen hypercapnia by increasing gas trapping and VD/
VT. The optimal ventilator strategy for severe obstructive lung disease 

TABLE 313-3  Adverse Effects of Hypercapniaa
Pulmonary arterial vasoconstriction (possible worsening of right heart failure)
Rightward shift of the oxyhemoglobin curve
Cerebral vasodilation
Increased intracranial pressure
Sympathetic-adrenal stimulation
Reduced cardiac contractility (especially in the presence of β-adrenergic 
blocking therapy)
aSome effects decrease if cellular pH is corrected.
physiology entails using lower respiratory rates, usually 9–12 breaths/
min, and moderate tidal volumes (7–9 mL/kg) to maintain a minute 
ventilation of ~10 L/min; higher minute ventilation usually worsens 
hyperinflation and can cause barotrauma. To prevent dyspneic patients 
from driving hyperventilation, deeper sedation and occasionally neu­
romuscular blockade are necessary until severe bronchial obstruction 
responds to medical therapy. Although permissive hypercapnia can 
minimize barotrauma and volume trauma during mechanical ventila­
tion, hypercapnia has adverse effects including increased intracranial 
pressure, pulmonary artery vasoconstriction, and even depressed 
cardiac contractility (Table 313-3). The benefits and risks of a hyper­
capnia ventilatory strategy must, therefore, account for the individual 
patient’s comorbid medical conditions, for example, acute neurologic 
injury and risk of critical increases in intracranial pressure.
CHAPTER 313
Mechanical Ventilatory Support 
COMPLICATIONS OF MECHANICAL 
VENTILATION
■
■AIRWAY
Endotracheal intubation and mechanical ventilation can lead to several 
pulmonary and extrathoracic complications, especially when patients 
remain on mechanical ventilation for >7 days. Upper airway compli­
cations from endotracheal tube placement include vocal cord trauma 
(edema, avulsion, paralysis), tracheal stricture due to granulation tis­
sue, and tracheomalacia. Vocal cord injury can lead to postextubation 
stridor (PES) and need for replacement of an endotracheal tube. PES 
risk factors include prolonged (>7 days) or traumatic intubation, large 
endotracheal tube size, previous episode of PES, and head/neck surgery 
or trauma. Patients with PES risk factors should have the balloon cuff 
deflated on their endotracheal tube and assessed for air passing across 
the balloon (so-called “cuff leak test”). Patients with no cuff leak have 
an approximate 30% risk of PES and may need further assessment for 
causes of PES, with endotracheal tube removal delayed until the under­
lying process is treated.
■
■ADVERSE CARDIOPULMONARY EFFECTS OF 
POSITIVE-PRESSURE VENTILATION
High positive intrathoracic pressure, such as sustained inspiratory 
plateau pressures >30 cmH2O or high PEEP, can cause several manifes­
tations of lung barotrauma, including worsening of acute lung injury, 
pneumomediastinum, pneumothorax, and even pneumoperitoneum. 
Although positive-pressure ventilation can improve left-sided heart 
failure by decreasing left ventricular preload and afterload, right 
ventricular failure and pulmonary arterial hypertension can worsen 
due to inadequate right ventricular preload and an increase in right 

ventricular afterload and pulmonary vascular resistance; these effects 
on the right ventricular and pulmonary circulation should be con­
sidered when choosing a ventilatory strategy in patients with severe 
right-sided heart disease. In addition, blunted central venous return 
can cause upper and lower extremity edema, especially in the setting 
of aggressive IV fluid resuscitation and vascular leak related to the 
underlying critical illness.
■
■VENTILATOR-ASSOCIATED PNEUMONIA
Several factors during mechanical ventilation, such as violation of natural 
airway defenses, sedation with depressed cough, and micro-aspiration, 
all increase the risk of bacterial entry into the lower respiratory tract and 
development of pneumonia. Ventilator-associated pneumonia (VAP)

occurs in up to 15% of mechanically ventilated patients and causes death 
in nearly 50% of patients. VAP is a lower respiratory tract infection that 
occurs ≥48 h after initiating mechanical ventilation and requires the fol­
lowing: (1) new pulmonary opacities on chest x-ray, (2) a clinical change 
consistent with pneumonia (fever, increased sputum, leukocytosis, or 
increase in ventilator support, such as increased Fio2 or PEEP), and (3) 
positive microbial culture obtained from the lower respiratory tract via 
deep endotracheal suctioning or bronchoscopy specimen (bronchoal­
veolar lavage or protected endobronchial brushing). Most VAP pathogens 
are typical hospital-acquired bacteria including Staphylococcus aureus, 
Pseudomonas aeruginosa, and several other enteric gram-negative rods. In 
cases of suspected VAP, early empiric antibiotic therapy generally requires 
an intravenous β-lactam with broad gram-negative rod activity, such as 
piperacillin-tazobactam, cefepime, or ceftazidime. Empiric therapy for 
methicillin-resistant S. aureus (MRSA) with vancomycin or linezolid or 
for multidrug-resistant enteric gram-negative rods with a carbapenem 
should depend on local intensive care unit (ICU) infection control data 
or individual patient risk for these resistant bacteria. If possible, based on 
respiratory cultures, empiric antibiotic regimens should be narrowed and 
total treatment duration should be 7 days. Given the significant morbidity 
and mortality for VAP, prevention strategies are paramount and should 
be part of standardized care or “bundles.” VAP prevention interventions 
supported by clinical trial evidence include head-of-bed elevation to at 
least 30–45° (70% VAP reduction compared to supine position), special­
ized endotracheal tube use with a suction port above the cuff to minimize 
aspirated secretions (50% VAP reduction), minimization of ventilator 
circuit tubing changes (prevents bacterial entry), and hand hygiene before 
handling the ventilatory circuit. Practices with uncertain value in reduc­
ing VAP but still reasonable include limiting deep tracheal suctioning, 
daily sedation interruption, and routine mouth and dental care.

PART 8
Critical Care Medicine
■
■OTHER
The systemic physiologic stress associated with mechanical ventilation 
and necessary adjunctive therapies, such as sedation and neuromus­
cular blockade, can cause significant extrathoracic complications. The 
more common disorders include gastrointestinal stress ulcers and 
bleeding, deep venous thrombosis and pulmonary embolism, sleep 
disruption and delirium, and critical illness–associated myopathy that 
sometimes leads to prolonged mechanical ventilation. To minimize the 
risk of these adverse events, ICUs should institute care bundles includ­
ing daily interruption of sedatives and assessment for extubation and 
prophylaxis for deep venous thrombosis.
Daily assessment
Ready to extubate?
Continue
mechanical
ventilation
No
Spontaneous
breathing trial
Passed?
Failure/
reintubation
Extubation
Recurrent respiratory
failure or high risk*?
No
Yes
High-flow O2
or NIV
Stable/improved
respiratory
status?
SUCCESS
(off mechanical
ventilation)
Yes
FIGURE 313-4  Algorithm for discontinuing mechanical ventilation. APACHE-II, Acute Physiology and Chronic Health Enquiry II; BMI, body mass index; COPD, chronic 
obstructive pulmonary disease; PEEP, positive end-expiratory pressure; NIV, noninvasive ventilation.

LIBERATION FROM MECHANICAL 
VENTILATION
Discontinuing mechanical ventilation and transitioning a patient back 
to spontaneous breathing is often referred to as ventilator “weaning,” 
implying dependency on positive-pressure ventilation once started. 
Although patients on prolonged mechanical ventilation can develop 
respiratory muscle weakness, this occurs in a minority of patients. 
Approaching removal of ventilator support as a “wean” extends unnec­
essary mechanical ventilation time up to 40%. Liberating a patient 
from mechanical ventilation, therefore, should be more active by 
frequently assessing a patient’s readiness for spontaneous breathing, 
determined largely by resolution of the underlying process causing 
respiratory failure (Fig. 313-4). Important criteria indicating a patient 
may be ready for extubation include the following: underlying disease 
process has improved, patient is awake and largely off sedative medica­
tions, Fio2 ≤0.5, PEEP <8 cmH2O, and Sao2 >88%, stable hemodynam­
ics, and manageable respiratory secretions with adequate cough. These 
criteria should be assessed daily, and if achieved, patients should have 
a spontaneous breathing trial (SBT), which is a maneuver wherein 
positive pressure is set to a minimum to compensate for endotracheal 
tube resistance (usually 5–7 cmH2O) and the patient breathes sponta­
neously from 30 to 120 min. A patient “passes” the SBT if they appear 
comfortable overall (no marked anxiety or diaphoresis) and have a 
respiratory rate <35, Sao2 >90%, systolic blood pressure between 90 
and 180 mmHg, and heart rate change of <20%. Patients passing an 
SBT have a >70% chance of successful extubation. Incorporating extu­
bation “readiness” screening followed by SBT into a care protocol leads 
to 25% fewer ventilator days and a 10% decrease in ICU length of stay 
compared to traditional ventilator weaning. Although many physi­
ologic variables correlate with successful liberation from mechanical 
ventilation, such as minute ventilation, negative inspiratory force gen­
eration, and the respiratory rate–to–tidal volume ratio (Tobin index), 
overrelying on these measures versus the outcome of an SBT leads to 
unnecessary delays in extubation. Risk factors for failing extubation 
even after a successful SBT include age >65, congestive heart failure, 
COPD, Acute Physiology and Chronic Health Enquiry (APACHE-II) 
score >12, body mass index (BMI) >30, significant secretions, more 
than two medical comorbidities, and >7 days on mechanical ventila­
tion. Patients with these risk factors transitioned immediately after 
extubation to noninvasive respiratory support using either high-flow 
oxygen or positive-pressure NIV have significantly lower rates of rein­
tubation and need to resume mechanical ventilation. Although NIV 
• Underlying process improved
• Awake, minimal sedation
• FIO2, <0.5, PEEP <8 cmH2O
• SaO2 >88%
• Stable hemodynamics
• Minimal secretions/good cough
Yes
*High-risk for respiratory failure
• Age >65
• Congestive heart failure
• COPD
• APACHE-II score >12
• BMI >30
• Significant secretions
• >2 medical comorbidities
• >7 days on mechanical
  ventilation
Yes
No