Show Summary Details
Page of

Inhalation injury in the ICU 

Inhalation injury in the ICU
Inhalation injury in the ICU

Silvia Coppola

and Franco Valenza

Page of

PRINTED FROM OXFORD MEDICINE ONLINE ( © Oxford University Press, 2020. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in Oxford Medicine Online for personal use (for details see Privacy Policy and Legal Notice).

date: 25 October 2020

Key points

  • Smoke inhalation causes three types of injuries—thermal injury to the upper airways, chemical injury to the tracheobronchial tree, and systemic poisoning.

  • When a patient presents with known or suspected smoke inhalation, the patient’s airway, breathing, and circulation should be immediately and repeatedly assessed.

  • Carbon monoxide and hydrogen cyanide interfere with the delivery and utilization of oxygen.

  • Prevention or early diagnosis, and treatment of associated life-threatening complication are necessary to decrease morbidity and mortality after inhalation injury.

  • Bronchoscopy is currently a standard diagnostic regimen for inhalation injury.


Inhalation injury is an acute respiratory tract insult caused by steam or toxic inhalant, such as fumes, gases, and mists. Inhalation injury may occur without cutaneous burn injury, although the two injuries usually occur together [1]‌. Inhalation injury continues to be one of the most serious associated injuries, complicating the care of thermally-injured patient. Prevention or early diagnosis, and treatment of this associated life-threatening complication are necessary to decrease its associated morbidity and mortality. Airway injury is present in up to one-third of patients with major burns, and the risk of concurrent pulmonary damage is directly related to the extent of the body surface area burned [2]. Inhalation injury greatly increases the incidence of respiratory failure and acute respiratory distress syndrome (ARDS). It is also the cause of most early deaths in burn victims. [1] The mortality rate following smoke inhalation ranges from 45 to 78% [3,4,5].


Three mechanisms are responsible alone or in combination with inhalation injury—thermal injury to the respiratory tract, chemical injury of airways and lung parenchyma due to exposure to irritating gases and particulate matter, and systemic oxygen supply impairment.

Thermal injury

The inhalation of extremely hot smoke may produce thermal injury to the upper airway. Direct thermal injury below the vocal cords is very unusual because of the efficient heat-exchanging system of the mucous membranes of the upper airway and the limited capacity of dry air to conduct heat. Thermal injury due to smoke inhalation may very rapidly cause erythema, ulceration, and life-threatening intra-oral, pharyngeal, or laryngeal oedema. The oedema may also be progressive during the first 18–24 hours [1]‌. Therefore, early recognition of thermal injury is very important to perform tracheal intubation before oedema progresses to airway obstruction. Lower airway mucosal oedema is usually not clinically evident until 24 hours after exposure. This oedema can cause damage to the tracheal and bronchial mucosa resulting in bronchorrhoea (that may not appear until 36 hours after exposure), ulceration, and damage to the ciliary system. Tracheal and bronchial epithelial sloughing may also occur [5,6,7].

Chemical injury

Chemical injury of the airway can be caused by irritants or cytotoxic compounds. The type and volume of the irritants generated by combustion can vary depending on the material burned, the temperature of the fire, and the amount of oxygen present in the fire environment.

Gases with high water solubility such as ammonia, hydrogen chloride, and sulphur dioxide, react with water in the mucous membranes of the upper airway, and produce strong acids and alkalis leading to irritation, ulceration, and oedema of the mucosal surface. This irritation is well perceived by the victims provoking escape responses. Less water-soluble gases (nitrogen oxides, phosgene, chlorine) are transported to the lower airway. Since these lipid-soluble constituents are not as irritating, protracted exposures are more likely before irritation and inflammatory reaction of the upper and lower airways occur [8]‌. Moreover, smoke contains particulate matter, usually less than 0.5 mm in size, which is formed by the incomplete combustion of organic material. These small particles easily reach all parts of the respiratory tract including the alveoli.

A short duration of exposure to highly reactive irritants may result in loss of cilia and epithelial erosions of the tracheobronchial tree.

When irritant gasses and particulate matter reach the terminal bronchioles, they can initiate an inflammatory reaction [5,7].

Oxygen delivery impairment

The tissue hypoxia is multifactorial, including the inspiration of air with a FiO2 of less than 0.15 during the fire and the impaired delivery and utilization of oxygen (O2) by the tissues. Fires occurring in enclosed spaces rapidly remove the available O2 from the environment. Asphyxiant gases lower the ambient oxygen tension. Both mechanisms lead to hypoxaemia in the victims.

Carbon monoxide (CO) and hydrogen cyanide are the two major tissue asphyxiants, which interfere with the delivery and utilization of O2.

The decrease of the O2 delivery to the brain and other organs interferes with mental and physical capabilities. Gases that produce significant hypoxaemia are indirectly toxic to the heart. If hypoxaemia progresses, the aerobic metabolism shifts to the anaerobic pathway leading to metabolic acidosis.

Injury activates the inflammatory cascade resulting in histologically evident inflammatory changes of the respiratory mucosa within 2 hours of injury. In addition, inflammatory mediators such as thromboxane have been shown to decrease mucociliary activity. Both direct toxic injury, as well as the release of inflammatory mediators, allow particles and toxins to exert their effects on other local defence mechanisms and initiate a cascade of parenchymal damage and bacterial infection [6]‌.

It is thought that inhalation of burn products, as well as respiratory tract thermal injury can impair surfactant function and production. Lung injury models have shown that increased capillary permeability leads to reduced surfactant production causing a loss of force that maintain alveolar patency thus resulting in alveolar collapse [7]‌.

Of particular importance is the fact that inhalation injuries have been proven to increase the risk of pneumonia in burned ICU patients. The risk for pneumonia depends on the impaired function of alveolar macrophages, polymorphonuclear leukocytes, and mucociliary clearance mechanisms. Age >60 years and total burn surface area >20% have been shown to be associated with the development of pneumonia [9]‌. Both factors might prolong the length of stay and, thus, the risk of pneumonia. The mortality for the combination of inhalation injury and nosocomial pneumonia can reach 50–86% [4]. The root cause of this synergistic effect has to do with both direct lung injury from inhalation, as well as systemic inflammation and immune dysfunctions. Endogenous organisms that often cause infections are those present in the oral and respiratory tract or in the gut at the time of admission. These include Staphylococcus aureus, Streptococcus pneumoniae, Haemophilus influenzae, Proteus mirabilis, and Escherichia coli. Exogenous organisms include methicillin-resistant S. aureus, Acynetobacter, Pseudomonas aeruginosa, and other opportunistic organisms. Early infections tend to originate from endogenous organisms, whereas infections at a later time tend to be from exogenous organisms. Recognition and understanding of the pathogens involved in inhalation injury and the time for the onset of infections are important to tailor effective antimicrobial therapy and avert serious complications. ARDS is a complication that may develop several days after the exposure. During thermal injury, inflammatory mediators increase vascular permeability, recruit immune cells, and reduce surfactant function [10]. Thus, burn and inhalation injury carry a significant risk in the development of ARDS that results in a fairly high mortality rate (50–60%) [9].


An accurate diagnosis of the inhalation burn injury in the early stage is essential for achieving a good prognosis. However, this is not entirely true because the inhalation burn injury has a latent period for 3–4 days prior to the occurrence of respiratory complications. For this reason, inhalation injury is difficult to diagnose and need to be suspected on the basis of a history of smoke exposure in an enclosed space [11].

Patients can be symptomatic within minutes or be asymptomatic for several hours. Symptomatic patients may show some of the following symptoms—mental confusion, unconsciousness, burns to face, lips, mouth and neck, burned nasal or facial hair, soot in the mouth, around the nares or in sputum, hoarseness, and laryngeal stridor. Symptoms of inhalation injury below the glottis include—dyspnoea, bronchorrhoea, wheezing sounds, productive cough, and increased labour in breathing. Bronchospasm and upper airway oedema can occur rapidly. Lower airway oedema can be asymptomatic for up to 24 hours [1]‌.

Inhalation injury should be suspected in the presence of a high carboxyhaemoglobin level after exposure to smoke in an enclosed space. Nevertheless, carboxyhaemoglobin levels may not be reliable if O2 has been administered or significant time has passed from initial exposure to carbon monoxide (CO). Arterial blood gases may be initially normal. The presence of hypoxaemia and hypercarbia is suggestive of pulmonary injury. All victims of smoke inhalation should be evaluated for cyanide and CO poisoning. Carbon monoxide and hydrogen cyanide poisoning may produce metabolic acidosis and a decreased difference in arteriovenous O2 content due to tissue hypoxia.

Until 1–2 days after the onset of inhalation burn injury, abnormal findings cannot be identified on a chest radiography or on arterial blood gas analysis. Chest X-ray may show some abnormalities 24–48 hours after the injury, such as atelectasis, and interstitial and alveolar infiltrates that could suggest the presence of pulmonary infection or oedema. Computed tomography may be useful for the diagnosis of delayed respiratory sequelae, such as bronchiectasis, pulmonary fibrosis [12]. Laryngoscopy and bronchoscopy can be performed and have a very high accuracy in the diagnosis of the full extent of the airway injury. The initial exploration may not reveal injured areas for the first 24 hours. This technique may show mucosa erythema, oedema, blisters, ulcerations, presence of particulate matter. Inhalation injury should be assessed by examination of both the upper and lower airway, since any of those may be affected independently. It has been reported that the findings seen on bronchoscopy are useful in predicting the progression of acute lung injury. Fibre optic bronchoscopy is considered the most direct diagnostic method for the definitive diagnosis of inhalation injury and is considered to be more accurate than the diagnosis based on clinical manifestations and signs [13,14].

The literature suggests that the shock phase of burning can lead to colour changes in the airway epithelium that are caused by poor perfusion affecting the correct estimation of the depth of the injury. Fibrescopic bronchoscopy should be performed 3 days after injury, after the shock phase, in order to obtain more accurate results. In conclusion, fibre optic bronchoscopy can evaluate airway epithelial congestion, oedema, erosion haemorrhage, epithelial necrosis, and slough-off. For patients with malignant arrhythmias, refractory hypoxaemia, or severe, uncorrectable bleeding diathesis, bronchoscopy is not recommended [14].

Ventilation–perfusion radionuclide scanning with Xenon-133 is a reliable method for identification of small airway obstruction. This technique is indicated in case of normal findings on chest X-ray and bronchoscopy to determine the extent of the inhalation injury [4]‌.

Pulmonary function tests may give information about respiratory rate, lung compliance, and decreased vital capacity [10]. However, they are usually difficult to perform due to the critical condition of the patients.


When a patient presents with smoke inhalation, immediate assessment of the patient’s airway, breathing, and circulation is indicated. This should take only a few seconds to perform.

The cornerstone of management includes management of airways patency, adequate fluid resuscitation and mechanical ventilation when required, and vigilant surveillance for infectious complications [6]‌.

Intubation is justified if any of the following signs are present—stridor, use of accessory respiratory muscles, respiratory distress, hypoventilation, deep burns to the face or neck, or blistering or oedema of the oropharynx. If these findings are absent, the oropharynx should be examined, followed by laryngoscopy if there is erythema. Some centres routinely perform bronchoscopy, rather than laryngoscopy. Other centres consider that laryngoscopy is preferable to bronchoscopy, because thermal injuries tend to be limited to the supraglottic airways and the appearance of the subglottic airways does not definitively affect management or predict the need for ventilator support [15].

Upper airway oedema or blistering seen during laryngoscopic examination should prompt intubation. Intubation with a large lumen endotracheal tube is preferable to facilitate optimal management of secretions and bronchoscopy. Humidified oxygen may help avoid inspissation [11].

In contrast, in the absence of upper airway oedema or blistering, close observation for 24 hours is reasonable, particularly if serial laryngoscopies are performed. If upper airway oedema is going to occur, it will usually manifest within 24 hours of the exposure.

For patients who are intubated, the endotracheal tube should be left in place until resolution of the upper airway oedema has been documented. Changing the endotracheal tube and failed extubation requiring re-intubation are dangerous in the presence of upper airway oedema and should be avoided [3]‌.

When intubation is necessary, the use of succinylcholine or other depolarizing agents may be appropriate for patient intubation if neuromuscular blockade is required. These drugs are contraindicated for intubation in burn patients 48 hours after injury, since they may worsen post-burn hyperkalaemia.

Patients who do not require intubation should receive supplemental oxygen at a fraction of inspired oxygen of 100%. The purpose of a high concentration of supplemental oxygen is to quickly reverse tissue hypoxia, and to displace CO and cyanide from protein binding sites [6]‌.

Subsequent management of the patient with smoke inhalation consists of monitoring the patient for the development of upper airway compromise due to thermal injury, as well as the development of lower airway sequelae due to direct toxin damage. The former tends to occur within 24 hours of exposure and is managed by intubation until the upper airway oedema subsides, which generally occurs in 3–5 days [3]‌. The latter tends to occur within 12–36 hours, has a variable duration and is managed by aerosolized bronchodilatators [8].

The hallmark of ventilator management during the treatment of inhalation injury is to minimize further damage and inflammation to lung tissue, and provide adequate ventilation and oxygenation. The increased capillary permeability, coupled with changes in surfactant, results in increased opening alveolar pressures and extensive atelectasis. Studies have shown that increasing positive end expiratory pressure (PEEP) above hydrostatic pressures can prevent collapse of these regions. However, because hydrostatic pressures are not equally distributed and atelectasis tends to occur in dependent lung regions, increasing PEEP to overcome the collapse in these regions could lead to overdistention of other regions, resulting in barotrauma [11]. One way of counteracting the mechanical ventilation-induced damage to lung parenchyma could be the high frequency ventilation (HFV) that uses rapid respiratory rates and small tidal volumes. Several trials of HFV in patients with acute lung injury have shown improvements in oxygenation and ventilation. However, sample sizes for these studies have not been large enough to show a significant survival benefit [6]‌. Extracorporeal membrane oxygenation is used in situations in which mechanical ventilation fails to provide adequate oxygenation or elimination of carbon dioxide.

Because of the increased pulmonary vascular permeability, the fluid management strategy incorporates providing minimal amounts of fluid to maintain adequate haemodynamic parameters and urine output [6]‌.

Inhalation injury predisposes the patient to nosocomial infections by opportunistic organisms. Prophylactic antibiotic coverage did not show any benefits; on the contrary, it may lead to increased antimicrobial resistance by these organisms. Currently, broad spectrum antibiotics are used when infections or sepsis is suspected. Once an infection agent is identified by culture or Gram stain, the antibiotic therapy is directed at that source.

The presence of CO poisoning, a non-inflammatory form of inhalation injury, should always be considered in those patients suspected of having inhalation injury. CO interferes with oxidative metabolisms by decreasing the oxygen-carrying capacity of the blood, shifting the oxygen–haemoglobin dissociation curve to the left and binding to cytochrome oxidase. For these reasons, any patients suspected of inhalation injury should receive 100% oxygen via a tight-fitting non-rebreathing mask until CO poisoning is excluded and the carboxyhaemoglobin level is less than 10%. The half-life of carboxyhaemoglobin is reduced from approximately 240 minutes at an FiO2 of 0.21 to around 75–80 minutes at an FiO2 of 1.0. Hyperbaric oxygen decreases the half-life of carboxyhaemoglobin to approximately 20 minutes [16]. Hyperbaric oxygen treatment has been shown to have an advantage over normobaric oxygen treatment for CO poisoning [6].

A release of hydrogen cyanide also occurs in the smoke of residential fires during combustion of polyurethane, acrylonitrile, and nylon. Hydrogen cyanide can be present in 50% of all fires and its toxicity is synergic with that of carbon monoxide. Cyanide poisoning is nearly impossible to confirm during the initial hours following smoke inhalation because cyanide levels cannot be measured soon enough to be clinically useful. The symptoms and signs are non-specific, and may be due to CO poisoning or an alternative inhaled toxin (including coma, central apnoea, cardiac dysfunction, severe acidosis, high mixed venous oxygen, and low arteriovenous O2 content difference) [17].

Cyanide inhalation is a potentially life-threating occurrence that requires immediate intervention.

Cyanide, by virtue of its uncoupling of oxidative phosphorylation, is a known cellular toxin. In the patient with an unexplained severe acidosis, tachycardia, and tachypnoea, yet a normal arterial oxygen content, cyanide poisoning may be present and treatment should be initiated. The first treatment consists of creating a cyanide link in the form of a ferric ion on haemoglobin by the delivery of inhaled amyl nitrite (Amyl Nitrite®) or intravenous sodium nitrite (Nithiodote®). Further treatment includes supplying substrate, such as thiosulphate, which transfers a sulphur group to cyanide and converts it to thiocyanate, which is excreted by the kidneys. A third therapeutic agent, is hydroxycobalamin (Cyanokit®) that combines with cyanide to form the inactive compound cyanocobalamin and is usually given intravenously as a 5-g dose, which should be doubled for patients with blood cyanide levels greater than 40 μ‎mol/L [17].


1. El-Helbawy RH and Ghareeb FM. (2011). Inhalation injury as prognostic factor for mortality in burn patients. Annals of Burns and Fire Disasters, 24(2), 82–8.Find this resource:

2. Soejmima K, Schmalstieg FC, Sakurai H, et al. (2001). Pathophysiological analysis of combined burn and smoke inhalation injuries in sheep. American Journal of Physiology—Lung Cellular and Molecular Physiology, 280, L1233–41.Find this resource:

3. Heimbach DM and Waeckerle JF. (1988). Inhalation injuries. Annals of Emergency Medicine, 17, 1316–20.Find this resource:

4. Shirani KZ. (1987). The influence of inhalation injury and pneumonia on burn mortality. Annals of Surgery, 205, 82–7.Find this resource:

5. Haponik EF, Crapo RO, Herdon DN, et al. (1988). Smoke inhalation. American Reviews of Respiratory Diseases, 138, 1060–3.Find this resource:

6. Monafo WW. (1996). Initial management of burns. New England Journal of Medicine, 335, 1581–6.Find this resource:

7. Weiss SM and Lakshiminarayan S. (1994). Acute inhalation injury. Clinical Chest Medicine, 15, 103–6.Find this resource:

8. Toon MH, Maybauer MO, Greenwood JE, et al. (2010). Management of acute smoke inhalation injury. Critical Care and Resuscitation, 12, 53–61.Find this resource:

9. Ching-Chun Lin AA, Liem Cho-Kai Wu, Yi-Fan Wu, Jui-Yung Yang, and Chung-Ho Feng. (2011). Severity score for predicting pneumonia in inhalation injury patients. Burns, 38, 203–7.Find this resource:

10. Slutzer AD, Kinn R, and Said SI. (1989). Bronchiectasis and progressive respiratory failure following smoke inhalation. Chest, 95, 1349.Find this resource:

11. Clark WR Jr. (1992). Smoke inhalation: diagnosis and treatment. World Journal of Surgery, 16, 24–9.Find this resource:

12. Tasaka S, Kanazawa M, Mori M, et al. (1995). Log-term course of bronchiectasis and bronchiolitis obliterans as late complication of smoke inhalation. Respiration, 62, 40.Find this resource:

13. Bingham Hg, Gallagher TJ, and Powell MD. (1987). Early bronchoscopy as a predictor of ventilatory support for burned patients. Journal of Trauma, 27, 1286–8.Find this resource:

14. American Burn Association (2003). Inhalation injury: diagnosis. Journal of the American College of Surgeons, 196, 307–12.Find this resource:

15. Masanes MJ, Legendre C, Lioret N, et al. (1995). Using bronchoscopy and biopsy to diagnose early inhalation injury. Macroscopic and histologic findings. Chest, 107, 1365–9.Find this resource:

16. Ilano AL and Raffin TA. (1990). Management of carbon monoxide poisoning. Chest, 97, 165–9.Find this resource:

17. Beasley DMG and Glass WI. (1998). Cyanide poisoning: pathophysiology and treatment recommendations. Occupational Medicine, 48, 427–31.Find this resource: