Hospital Medicine

Complications of mechanical ventilation

I. Problem/Challenge.

Most complications of mechanical ventilation are related directly to the disruption of the normal cardiopulmonary physiology. Normal pulmonary physiology involves creating a negative pressure in the airways by the diaphragm and the chest wall, whereas positive pressure ventilation (i.e., the vast majority of modern ventilators) involves forcing air into the airways under pressure.

Others are a combination of the above and being critically ill. It is essential to be aware of these complications so that they can be recognized early and intervened upon.

II. Identify the Goal Behavior

Most complications of mechanical ventilation can be avoided and prevented.

III. Describe a Step-by-Step approach/method to this problem.


Barotrauma refers to alveolar rupture due to elevated transalveolar pressure. This can manifest as pneumothorax, pneumoperitoneum, subcutaneous emphysema, pneumomediastinum and can sometimes progress to bronchopleural fistula or tension pneumothorax.

High plateau pressures (as opposed to peak airway pressures) predispose to barotrauma. Hence it is patients with obstructive airway disease or diseases of the lung parenchyma which lead to low compliance (like ARDS or interstitial lung diseases) who are at greatest risk.

Most ventilators display the peak airway pressure. To measure the plateau pressure an inspiratory breath hold is applied (i.e., by closing the exhalation valve at the end of inspiration and holding for a second). The plateau pressure is usually slightly lower than the peak pressure; however, this difference can be higher in bronchospasm or plugged airways.

Sudden desaturation, tachypnea, tachycardia, hypotension, rise in peak airway pressures, reduced breath sounds on one side, or tracheal deviation in the mechanically ventilated patient are suggestive of barotrauma and should prompt an investigation with a chest X-ray. Alternately ultrasonography by the trained physician can also be used to rule out pneumothorax since it can be usually be done much faster and may help make prompt management decisions in the critically unstable. Chest ultrasonography is more sensitive than X-ray for detecting pneumothorax and it may also be useful for identifying other causes for the changing clinical picture.

To prevent barotrauma, it is generally recommended to maintain the plateau pressure below 35 cm H2O. Plateau pressure can be lowered by treating the underlying cause; bronchodilators (in those with obstructive airway disease), sedation (in those dyssynchronous with the ventilator), lowering tidal volume, reducing flow rate, or a combination of all these. In patients who are on assist control mode of ventilation, it is easier to use pressure control and then to titrate the ventilator to get target volumes as opposed to using volume control.

Occasionally patients who continue to have very high airway pressures despite the above may need to be paralyzed to allow greater ventilator synchrony.

Most of the consequences of barotrauma are self-limited and need no intervention other than close monitoring. Pneumothorax needs closer observation and prompt intervention if hemodynamic compromise is seen. Pneumothorax can progress rapidly to tension pneumothorax in the setting of positive pressure ventilation and then needs decompression with thoracostomy.


Ventilator-associated lung injury (VALI) refers to alveolar injury caused by mechanical ventilation. This is distinct from barotrauma (which is alveolar rupture caused by high pressures). VALI is associated with alveolar edema and increased permeability caused by large tidal volumes irrespective of airway pressures.

Patients particularly prone to develop VALI are those receiving large tidal volumes, those with underlying restrictive lung diseases, those with ALI/ARDS and those who have received blood transfusions.

Even physiologic or low tidal volumes can lead to VALI in some patients. This is because in patients with atelectasis, air tends to preferentially flow towards more compliant alveoli (i.e. the ones that are already open) and hence may overdistend them. Furthermore those portions of the lung which are atelectatic but are being opened with each breath (cyclic atelectasis) are also prone to lung injury due to shear forces associated with this.

VALI can be prevented by applying two strategies. These strategies are based on studies on patients with ARDS, but are unproven in others.

The first is to prevent alveolar overdistension by employing low tidal volume ventilation. The general recommendation is to use tidal volumes of 8mls/kg of ideal body weight (based on height and gender) and to gradually keep lowering this by 1 ml/kg/ibw (ideal body weight) to get to the lowest tidal volume which the patient tolerates whilst providing acceptable oxygenation and ventilation. Mild acidosis and hypercapnia should be tolerated.

The second is to prevent collapse of alveoli during expiration and hence preventing cyclic atelectasis. The amount of positive end expiratory pressure (PEEP) needed to overcome cyclic atelectasis is higher than the lower inflection point on compliance curve (volume/pressure curve). Determining this is difficult and requires paralysis. Practically, however a small amount of PEEP is applied (3 to 5cm H2O is determined to be physiologic, i.e., equivalent to a closed glottis in a normal healthy person). This can then be titrated upwards until greater static compliance is seen (Vt/Plat-PEEP).


AutoPEEP refers to hyperinflation of the lungs due to air trapping. It is caused by initiation of inspiration before expiration is complete. It can be caused by large tidal volumes, high respiratory rate (insufficient time for expiration), obstructive air disease or narrow endotracheal tube.

Unchecked AutoPEEP can lead to barotrauma as well as worsening of the hemodynamic effects of positive pressure ventilation (PPV). Increased intrathoracic pressure leads to decreased venous return which in turn leads to decreased cardiac output and hypotension. This effect is further exacerbated in the hypovolemic patient.

AutoPEEP can also worsen ventilation-perfusion (V/Q) mismatch by compressing capillaries in the healthy part of the lung and diverting blood to the diseased lung. Work of breathing may also be increased because in pressure cycled settings it makes it harder to trigger a breath.

Most modern ventilators are also designed to measure AutoPEEP by subtracting applied PEEP from end expiratory breath hold.

AutoPEEP should be suspected when ventilator/patient dyssynchrony is seen. Breaths may not be triggered despite visible inspiratory effort. AutoPEEP can also present as sudden hypotension or desaturation.

AutoPEEP can be seen by looking at the flow versus time graph. If inspiratory flow begins before expiratory flow has stopped, then AutoPEEP will develop.

Treatment is to address the underlying cause. In patients with high minute ventilations, lowering the tidal volume, respiratory rate or both may help. Increasing the inspiratory flow rate may also help by allowing more time for expiration.

In patients with obstructive airway disease, if bronchodilators or steroids are not helpful and the above strategies have also failed applying PEEP may be useful. Applying PEEP to match intrinsic PEEP by 50% keeps the airways from collapsing during expiration.

Hemodynamic effects

Positive pressure ventilation causes decreased cardiac output by decreasing venous return (worsened with high PEEP). PPV also compresses the pulmonary vasculature leading to reduced right ventricular output. This in turn leads to reduced left cardiac output.

Applied PEEP also artificially elevates central venous pressure (CVP) and pulmonary capillary wedge pressure (PCWP) measurements.

In general, fluid resuscitation seems to correct hypotension caused by PPV. However, be aware of flash pulmonary edema after extubation (because sudden removal of PEEP leads to a large venous return). Hence, T piece trial before extubation may be useful in patient who had been mechanically ventilated on a high applied PEEP.


Ventilator associated pneumonia (VAP) is defined as pneumonia which occurs after 48-hours hours of intubation and mechanical ventilation. The incidence is between 9–27% and it is associated with considerable mortality (up to 50%). Thus early recognition and prompt treatment are important. The risk rises with duration of ventilation.

Oropharyngeal secretions and leakage of secretions around the cuff are the primary routes of infection. Stomach and the sinuses may also act as reservoirs. Hematogenous spread or infections from aerosolized medications is rare.

Efforts should be made to minimize the risk of aspiration. Elevating the head of the bed to 30°, minimizing sedation or paralysis, frequent suctioning of subglottic secretions and maintaining the cuff pressure at least 20cm H2O are measures which may limit aspiration. In addition, there is evidence that decontaminating the oral cavity with chlorhexidine swabs has reduced incidence of VAP. Prophylactic antibiotics too can reduce incidence of VAP but are not routinely recommended because of the risk of developing resistant pathogens.

Initiation of enteral nutrition too poses a risk of aspiration. However, the risks and benefits of enteral nutrition need to be determined individually per patient.

Common pathogens include aerobic gram negative bacilli, such as P. aeruginosa, E. coli, K. pneumoniae, Acinetobacter sp, and gram positive cocci such as S. aureus. Fungal pneumonias are unlikely in the immunocompetent but patients with organ transplants or immunosuppression are at risk.

New onset of fever, purulent sputum, leucocytosis, and desaturation should prompt further investigation. Demonstration of an infiltrate on radiography consistent with a consolidation together with fever, leukocytosis and purulent sputum (two of the latter three) is enough to initiate empiric treatment.

The incidence of colonization of the airways is common in intubated patients and antibiotics for decontamination are strongly discouraged. Routine aspirate for cultures in the asymptomatic patients is also discouraged.

To determine the etiology, the lower airway tract needs to be sampled. This can be done by endotracheal aspiration, bronchoalveolar lavage or by using a protected specimen brush (during bronchoscopy). There is no evidence that bronchoscopic sampling improves mortality, length of stay or duration of mechanical ventilation; thus, bronchoscopy simply for the purpose of sampling the airway is discouraged.

A sterile tracheal aspirate rules out pneumonia and in a patient suspected of having an infection, further investigation should be undertaken to look for source elsewhere.

Empiric antibiotics should be initiated with the consideration for the risk of multidrug resistant organisms in a given patient. Patients at risk for multidrug resistant organisms include those intubated longer than 5 days, having received antibiotics in the preceding 90 days, on chronic dialysis, recent hospitalization or resident of nursing home and high frequency of drug resistant pathogens in the community. Initial antibiotics should be broad and should cover both gram positive cocci and gram negative rods. Inappropriate antibiotics are associated with significant morbidity and mortality. Patients at risk for pseudomonal infections should receive a beta-lactam as well as either a quinolone or an aminoglycoside in order to increase the likelihood of susceptibility to at least one antibiotic. Antibiotics can then be de-escalated or modified later based on culture date or response. Optimal duration of therapy is 8 days based on a multicenter randomized trial.

Sinusitis is another infection commonly associated with mechanical ventilation. Particularly susceptible are those with NG tubes. Sinusitis should be suspected in patients who are febrile without apparent cause and have purulent discharge from the nose. This can be confirmed radiographically with an X-ray or computed tomography (CT) scan (preferred) of the sinuses, which would show opacification of the sinuses. Cultures should be obtained by endoscopic-guided meatal aspiration since culturing secretions from the nares is unreliable. Pathogens that cause sinusitis are similar to those that cause pneumonia in intubated patients. Nasal decongestants or steroids for prophylaxis are not recommended.

Complications caused by the endotracheal (ET) tube

Laryngeal injury can be caused by traumatic intubation, large ET tube, movements caused by coughing or transporting the patient, high inflation pressures, or prolonged intubation. They can range from laryngeal edema, mucosal ulcers, granulomas to even vocal cord paralysis caused by compression of the recurrent laryngeal nerve.

Laryngeal injury presents with hoarseness after extubation which resolves in a few days to weeks. Most patients have a benign course and recover completely. Some patients can have post-extubation stridor and may need to be reintubated because of respiratory failure. Risk factors for laryngeal injury are traumatic intubation (due to multiple attempts, inexperienced operator or large ET tube) or intubation prolonged beyond three days. Most laryngeal injuries resolve on their own. Laryngeal edema significant enough to cause respiratory distress requires steroids. Some granulomas can be large enough to require surgical removal.

Two rare but serious complications are tracheal stenosis and tracheoesophageal fistula. Both of these are caused by high ET tube pressures and prolonged intubation. ET tube cuff pressure should be maintained between 18-25 cm H2O and should be checked frequently to minimize this risk.

Tracheal stenosis presents with dyspnea, usually within five weeks of extubation. Pulmonary function tests may show upper airway obstruction however definitive diagnosis requires bronchoscopy or laryngoscopy. Treatment may require surgical excision or stenting.

Tracheoesophageal fistula is a devastating but rare complication of endotracheal intubation. It should be suspected in intubated patients with recurrent pneumonias. Adding methylene blue to the enteral feed and finding it in the endotracheal aspirate supports the diagnosis, however, does not confirm it. This should be followed with a barium swallow, esophagoscopy or bronchoscopy. The fistula needs to be repaired surgically which may not be feasible in the critically ill. Until it is, conservative management involves frequent suctioning, elevation of the head end of the bed and advancing the ET tube distal to the opening of the fistula.

IV. Common Pitfalls.

Early recognition of barotrauma and anticipation of auto-PEEP remain mainstays of treatment in patients without acute lung injury. As a result, the major pitfall is lack of recognition of these problems when they arise.

V. National Standards, Core Indicators and Quality Measures.

Prevention of ventilator associated pneumonia through positioning, oral care, etc. is not yet a national standard but is part of a commonly used "ventilator bundle." The ARDSNet protocol for patients with suspected acute lung injury is increasingly considered standard of care, although it is not yet a standard for accreditation or public reporting efforts.

IV. What's the evidence?

Anueto, A, Frutos-Vivar, F, Esteban, A. "Incidence, risk factors and outcome of barotrauma in mechanically ventilated patients". Intensive Care Med. vol. 30. 2004. pp. 612-9.

Villar, J, Kacmarek, RM, Pérez-Méndez, L, Aguirre-Jaime, A. "A high positive end-expiratory pressure, low tidal volume ventilatory strategy improves outcome in persistent acute respiratory distress syndrome: a randomized, controlled trial". Crit Care Med. vol. 34. 2006. pp. 1311.

Serpa Neto, A, Cardoso, SO, Manetta, JA. "Association between use of lung-protective ventilation with lower tidal volumes and clinical outcomes among patients without acute respiratory distress syndrome: a meta-analysis". JAMA. vol. 308. 2012. pp. 1651.

Nagarsheth, K, Kurek, S. "Ultrasound detection of pneumothorax compared with chest x-ray and computed tomography scan". Am Surg Apr;. vol. 77. 2011. pp. 480-4.

Niederman, MS, Ferranti, RD, Zeigler, A. "Respiratory infection complicating long-term tracheostomy: The implication of persistent gram-negative tracheobronchial colonization". Chest. vol. 85. 1984. pp. 39.

Urgences, R. "International Consensus Conferences in Intensive Care Medicine: Ventilator-associated Lung Injury in ARDS". Am. J. Respir. Crit. Care Med. vol. 160. 1999. pp. 2118-2124.

Amato, MB, Barbas, CS, Medeiros, DM. "Effect of a protective-ventilation strategy on mortality in the acute respiratory distress syndrome". N Engl J Med. vol. 338. 1998. pp. 347.

Gajic, O, Dara, SI, Mendez, JL. "Ventilator-associated lung injury in patients without acute lung injury at the onset of mechanical ventilation". Crit Care Med. vol. 32. 2004. pp. 1817-24.

Colice, GL. "Laryngeal complications of prolonged intubation". CHEST. vol. 96. 1989. pp. 877-884.

Guyton, DC, Barlow, MR, Besselievre, TR. "Influence of airway pressure on minimum occlusive endotracheal tube cuff pressure". Crit Care Med. vol. 25. 1997. pp. 91-4.

You must be a registered member of Psychiatry Advisor to post a comment.

Sign Up for Free e-newsletters