Smoke Inhalation Injury Workup

Updated: May 24, 2017
  • Author: Keith A Lafferty, MD; Chief Editor: Joe Alcock, MD, MS  more...
  • Print

Approach Considerations

In the workup of inhalation injuries caused by toxic smoke, the primary investigation focuses on the pulmonary system. Other tests may be clinically indicated based on history, physical examination, and underlying health problems. Initial blood tests should include lactate and CO-oximetry in addition to electrolytes and arterial blood gases. Carbon dioxide levels also may be monitored, since patients with prior lung disease such as asthma and chronic obstructive pulmonary disease may be affected more severely and are at greater risk to retain carbon dioxide.

Studies may include the following:

  • Pulse oximetry and CO-oximetry
  • Arterial blood gases (ABGs)
  • Carboxyhemoglobin level
  • Lactate
  • Complete blood cell count (CBC)
  • Chest radiography
  • Electrocardiogram
  • Serial cardiac enzymes (in patients with chest pain)
  • Pulmonary function testing
  • Direct Laryngoscopy and fiberoptic bronchoscopy

Pulse Oximetry and CO-oximetry

Pulse oximetry readings can be misleading in the setting of carbon monoxide (CO) exposure or methemoglobinemia because these devices use only 2 wavelengths of light (the red and the infrared spectrum), which detect oxygenated and deoxygenated hemoglobin only and not any other form of hemoglobin. Readings are falsely elevated by CO-bound hemoglobin (carboxyhemoglobin).

In methemoglobinemia, light reflection is similar to that in reduced hemoglobin. Pulse oximetry may show a depressed oxygen saturation, but the decrease does not accurately reflect the level of methemoglobinemia. In fact, as methemoglobin levels reach 30% or higher, the pulse oximetry reading converges on approximately 85%.

CO-oximeters use 4 wavelengths of light and are capable of detecting carboxyhemoglobin and methemoglobin as well as hemoglobin and oxyhemoglobin. Some newer co-oximeters use 5 wavelengths and are also able to measure sulfhemoglobin. The percent of oxyhemoglobin measured by CO-oximetry is an accurate measure of the arterial oxygen saturation. The difference between saturations obtained by CO-oximetry and calculated figures is known as the saturation gap and is an indicator of dyshemoglobinemia.


Arterial Blood Gases

Arterial oxygen tension (partial pressure of arterial oxygen [PaO2]) does not accurately reflect the degree of CO poisoning or cellular hypoxia. The PaO2 level reflects the oxygen dissolved in blood that is not altered by the hemoglobin-bound CO. Because dissolved oxygen makes up only a small fraction of arterial oxygen content, a PaO2 level within the reference range may lead to serious underestimation of the decrement in tissue oxygen delivery and the degree of hypoxia at the cellular level that occurs when CO blocks the delivery of oxygen to the tissues.

With most blood gas machines, the oxygen saturation is calculated on the basis of the PaO2 level. Thus, such a reading does not give an accurate determination of oxygen saturation, which must come from CO-oximetry.

ABG measurements are nonetheless useful to assess the adequacy of pulmonary gas exchange. Although the presence of a PaO2 level that is within the reference range may not exclude significant tissue hypoxia due to the effects of CO, the presence of a low PaO2 (< 60 mm Hg in room air) or hypercarbia (alveolar [arterial] carbon dioxide pressure [PaCO2] level of 55 mm Hg) indicate significant respiratory insufficiency. Metabolic acidosis suggests inadequate oxygen delivery to the tissues.

The difference between the partial pressure of oxygen in the alveolus and that measured on an ABG is the alveolar-arterial (A-a) gradient. This value, usually less than 5-10 mm Hg, may be several hundred mm Hg in the setting of significant pulmonary injury and can be used to assess improvement or deterioration in lung function when measured at a stable fraction of inspired oxygen (FiO2).

The alveolar gas equation can be used to estimate the efficiency of pulmonary oxygen delivery to the arterial circulation in the presence of supplemental oxygen administration. This formula determines the alveolar oxygen pressure.

The formula is as follows: PaO2 = (FiO2)(PB - PH2 O) - (PaCO2/RQ). PB represents barometric pressure, PH2 0 represents the partial pressure of water vapor (47 mm Hg at body temperature, ambient pressure), and RQ represents the respiratory quotient (estimated at 0.8).


Carboxyhemoglobin Level

Carboxyhemoglobin levels in the blood and the corresponding clinical manifestations are as follows [1] :

  • 0-10% - Usually no symptoms
  • 10-20% - Mild headache, atypical dyspnea
  • 20-30% - Throbbing headache, impaired concentration
  • 30-40% - Severe headache, impaired thinking
  • 40-50% - Confusion, lethargy, syncope
  • 50-60% - Respiratory failure, seizures
  • Greater than 70% - Coma, death

Blood carboxyhemoglobin levels may underestimate the degree of CO intoxication because of oxygen administered to the patient before arrival to the hospital. Smokers may have baseline levels up to 5-10% and may experience more significant CO poisoning for the same level of exposure as nonsmokers. Finally, correlation between carboxyhemoglobin levels and eventual neurologic outcome is poor.


Lactate and Other Blood Studies

Elevated lactate levels may result from metabolic acidosis secondary to the following:

  • Hypoxia
  • CO
  • Cyanide (CN)
  • Methemoglobinemia
  • Inadequate resuscitation
  • Unrecognized trauma

Lactate levels associated with CN poisoning have been reported as being above 8 mmol/L. [12] The concentration of lactate increases proportionally with the degree of CN poisoning, and lactate levels higher than 10 mmol/L are a sensitive indicator of CN levels higher than 1 mg/mg. [39] Note that in most institutions, CN levels can take hours to days for results; therefore, one must rely on clinical and indirect laboratory data.


Other Blood Studies

Electrolyte testing can identify an anion gap acidosis. In patients who require large-volume fluid resuscitation, measure electrolytes at regular and frequent intervals to monitor for the electrolyte abnormalities that may occur in these patients. Use results to adjust both fluid and electrolyte replacement.

Blood urea nitrogen (BUN) and creatinine levels should be obtained for baseline renal function determination in patients in shock or with rhabdomyolysis. Patients with large cutaneous burns, crush injuries, or prolonged immobilization should have their serum creatine kinase (CK) checked and, if appropriate, urine myoglobin.

Exposure to zinc oxide warrants baseline liver function tests on initial presentation. Liver function should be followed over the course of hospitalization if exposure is severe enough to warrant admission.

Thermal degradation products of various compounds, including phosphorus-based fire retardants, are capable of impairing cholinesterase activity. A prospective study measured serum erythrocyte cholinesterase activity at the scene of residential fires for 49 victims. A significant lower level of cholinesterase activity was noted in these patients as compared to controls. Obviously, further investigation into the clinical significance of this lower enzymatic activity is needed before it can be used clinically.

Lead-containing paint is common in structures built before 1977, and this element can become aerosolized and absorbed directly into the bloodstream from the lungs. While it is true that severe smoke inhalation has been shown to increase serum lead levels more than 2-fold, no evidence suggests that these elevations are clinically relevant. [40]


CBC Count

A baseline CBC count is warranted, as certain types of smoke are associated with a significant drop in hemoglobin and hematocrit beginning at 1 week postexposure. [41] A baseline white blood cell count can also be used for comparison when concerns arise about infection.

Hemoconcentration resulting from fluid losses is common immediately following injury. Adequate restoration of intravascular volume results in a progressive fall in hematocrit. Severe anemia may require blood transfusion, particularly in the presence of significant hypoxia or hemodynamic instability.


Cyanide Levels

Cyanide levels correlate closely with the level of exposure and toxicity, but these values may not be readily available. Many hospitals send out tests for cyanide levels, and results may not return for several days to a week. In a setting consistent with potential cyanide exposure, institute specific empiric therapy while waiting for laboratory confirmation of the diagnosis.

Findings indicative of cyanide intoxication include the following:

  • Persistent neurologic dysfunction unresponsive to use of supplemental oxygen
  • Cardiac dysfunction
  • Severe lactic acidosis, particularly in the presence of high mixed venous oxygen saturation
  • “Arterialization” of the venous blood gas, with PO 2 values similar to arterial levels due to lack of oxygen utilization by tissues


Obtain chest radiographs in patients with a history of significant exposure or pulmonary symptoms. The chest film is likely to be normal—initial studies have only 8% sensitivity for smoke inhalation—but it provides a baseline for subsequent comparison in cases of significant injury. Radiographic evidence of pulmonary injury typically does not appear until 24-36 hours after the inhalation.

When present, abnormal findings may include atelectasis, pulmonary edema, and acute respiratory distress syndrome (ARDS). Hyperinflation may suggest injury of the smaller airways and air trapping.

Individuals with fume fever often are sent home after 4 hours of observation and with a clear chest radiograph, only to return after the initial recovery and latent phase with more severe dyspnea and florid noncardiogenic pulmonary edema. The chest film in a patient with significant zinc oxide exposure may not show any abnormality until 4-6 hours post exposure. Radiographic abnormalities in these patients may improve slowly with supportive care or advance to a long-standing diffuse interstitial fibrosis.

In phase III of oxides of nitrogen exposure, a noncardiogenic pulmonary edema pattern may be seen on the chest radiograph. The chest radiograph may also show a pattern similar to military tuberculosis, which corresponds to a pathologic finding of classic bronchiolitis fibrosa obliterans. Fibrotic changes either may clear spontaneously or proceed to severe respiratory failure.

Cervical spine radiography is indicated to investigate neck injury in all unconscious patients and in those with a potential mechanism of injury (eg, a patient who jumped from a window to escape fire or fell down stairs).


Computed Tomography

Chest computed tomography (CT) scans may show ground-glass opacities in a peribronchial distribution and/or patchy peribronchial consolidations. These findings may be present on CT scan as early as a few hours after inhalation injury. [42]

A CT scan of the brain may show signs of cerebral infarction due to hypoxia, ischemia, and hypotension. An interesting and well-reported finding for severe CO toxicity is bilateral globus pallidus low-density lesions. These lesions may not appear until several days after the exposure days. This finding is highly specific for CO insult—unlike focal cortical hypoperfusion, which is nonspecific.


Radionuclide Scintigraphy

Delayed or inhomogeneous clearance of 133Xenon can be used to detect small-airway parenchymal injury. However, this study adds little to the clinical management and is not known to offer any particular therapeutic advantage. [43]

Likewise, increased clearance of aerosolized technetium-99m–labeled diethylenetriaminepentaacetate (99mTcDTPA) is a sensitive indicator of injury to the alveolar capillary membrane. However, its clinical use is not yet established.


Pulmonary Function Testing

Perform baseline pulmonary function tests (PFTs) once the patient is stable. In the ED, serial peak flow readings may be helpful. Later, PFTs allow evaluation and comparison of lung function and reversibility with bronchodilators and potentially steroids. If the patient develops dyspnea on exertion, then perform PFTs with exertion if PFTs at rest cannot explain the symptoms.

Pulmonary function test results become abnormal soon after inhalation injuries. Vital capacity, pulmonary compliance, and functional residual capacity are reduced. In patients with bronchospasm, forced expiratory volume in 1 second (FEV1), peak flow, and midexpiratory flow rates are reduced. Diagnostic accuracy is 91%.

In patients with cutaneous burns, the reduction in vital capacity and FEV1 correlates closely with the extent of surface burns. Full resolution of pulmonary function test result abnormalities may take several months. Some agents, particularly chlorine gas, may result in reactive airways syndrome, with subsequent development of airflow obstruction.


Direct Laryngoscopy and Fiberoptic Bronchoscopy

A significant number of patients may present with a paucity of upper airway signs or symptoms but may still have serious subglottic injury. The threshold for performing diagnostic bronchoscopy should be low. Bronchoscopy can be diagnostic as well as therapeutic, particularly when lobar atelectasis is present.

Bronchoscopy is the criterion standard for diagnosis of smoke inhalation injury. [42] This procedure examines the airways from the oropharynx to the lobar bronchi. Although it may be performed in the ED, the intensive care unit or burn unit may be a more appropriate setting, especially in patients who are intubated.

Erythema, charring, deposition of soot, edema, and/or mucosal ulceration may be present, although severe vasoconstriction from hypovolemia may mask significant injury. Impending airway obstruction may be inferred. Diagnostic accuracy is reported to be 86%. Fiberoptic bronchoscopy can also be used to facilitate endotracheal tube placement, even in the technically difficult airway.

Studies have shown up to a 96% correlation between bronchoscopic findings and the triad of closed-space smoke exposure, carboxyhemoglobin levels of 10% or greater, and carbonaceous sputum. In another study, serial bronchoscopy was twice as sensitive for diagnosing inhalation injury as clinical findings alone. Bronchoscopy is more sensitive and accurate than clinical examination alone in diagnosing inhalation injury and is, therefore, particularly useful in cases in which the decision to perform endotracheal intubation is unclear.

The use of bronchoscopy in patients with inhalation injury complicated by pneumonia is associated with a decreases in the duration of mechanical ventilation, length of intensive care unit stay, and overall hospital cost. [44] Serial bronchoscopy can help remove debris and necrotic cells in cases with aggressive pulmonary toilet or when suctioning and positive pressure ventilation are insufficient.

Bronchoscopy in children requires the use of a bronchoscope with a relatively small diameter, in order to accommodate the narrow pediatric airway. Extremely small diameter fiberoptic bronchoscopes with a suction port (capable of entering an endotracheal tube sized for a small toddler or infant) have only recently become available, and whether these limit the ability to remove heavy particulate matter is unclear.