Diagnosis and Treatment of Inhalation Injury

Introduction

Inhalation injury is a nonspecific term that refers to damage to the respiratory tract or pulmonary parenchyma by heat or chemical irritants carried into the airways during respiration.

Along with total body surface area (TBSA) burned and age, inhalation injury is one of the three features most associated with mortality following thermal insult. Issues related to diagnosis and management of inhalation injury have been most recently reviewed by Walker et al.

Inhalation injury may occur in conjunction with cutaneous burns or in isolation. The severity of injury varies depending on the chemical composition of the agent(s) inhaled, the intensity of exposure, and pre-existing comorbidities. There are three basic classes of inhalation injury: direct thermal injury, tissue damage due to inhalation of chemical irritants, and systemic effects of inhaled toxins. The upper airway serves as an efficient heat exchanger that protects lower structures from extremes of heat or cold. Reflex laryngeal closure also protects subglottic areas. As a result, direct thermal injury is generally restricted to the upper airway and rarely involves subglottic structures. Exceptions are the inhalation of steam, due to the much higher specific heat of water vapor, and blast injuries that can force hot gases past the glottic opening. Inhaled irritants are generally present in smoke as a mixture of gases, fumes, and mists, and the chemical composition of smoke produced from various fuels has been described. Fumes consist of particles of various size dispersed in gases. Mists are aerosolized liquids. The intensity of exposure along with the size and chemical composition of these particles and droplets determines how far distally they will migrate in the respiratory tract and, thus, the nature of the tissue injury. Large particles and droplets of lipid-soluble liquids are more likely to adhere to airway surfaces and do not reach as far distally as smaller particles and more water-soluble droplets. Systemic toxicity may occur when toxins such as carbon monoxide (CO) or cyanide are present in the inhaled gases.

The reported incidence of inhalation injuries has varied greatly over time and from region to region. Smith and others reported 19.6% incidence among burn patients in the United States. In Israel Haik and colleagues found as few as 1.9% in association with burns, whereas Luo and others found 8.01% in China. Regional differences are to be expected as a result of differences in local customs, building materials, and other factors.

The presence of inhalation injury is clinically significant for a variety of reasons, as listed in Box 17.1 . Inhalation injury has been found to be an independent risk factor for mortality. It is also associated with hemodynamic instability because volume requirements for resuscitation may be increased by as much as 50% when cutaneous burns are accompanied by inhalation injury. Parenchymal injury from inhaled irritants or hot gases can lead to impaired gas exchange, pneumonia, and acute respiratory distress syndrome (ARDS). When severe, these changes increase the risk of multiorgan failure and mortality. After recovery from inhalation injury pulmonary function disorders may persist due to pulmonary fibrosis or bronchiectasis. Improvements in the survival of patients with inhalation injury have been attributed to better overall burn outcomes, improved ventilator management, and improved management of pneumonias.

Improvements in the care of cutaneous burns have outpaced advancements in the treatment of inhalation injuries. There are several reasons for this disparity. The treatment of pulmonary parenchymal injury is inherently more complex than treatment for cutaneous burns. Necrotic skin can be excised and replaced with substitute materials or autografted skin, and healing can be observed directly. In contrast, treatment of injured lung involves measures to prevent further injury to allow host mechanisms to repair injured tissues. Healing of pulmonary injury is followed more indirectly by observations of blood gas exchange and radiographs. Inhalation injury results both from direct effects of heat and chemical irritants as well as from indirect effects from an inflammatory response to the initial insult. Despite extensive studies, these processes are incompletely understood and no specific therapies have been identified.

Because inhalation injury has such broad and critical clinical implications, it is important that it be diagnosed as early as possible. Early diagnosis can be accomplished by recognition of risk factors revealed by the history and physical examination and confirmed by diagnostic procedures.

Diagnosis

There is no consensus on the diagnostic criteria for inhalation injury. In the clinical setting the diagnosis is a relatively subjective judgment based on history and physical examination, often confirmed by additional diagnostic procedures such as bronchoscopy. One of the reasons for the lack of consensus for early diagnosis is that much of the impaired pulmonary function following inhalation injury results from obstruction of small airways and an inflammatory response to the initial direct injury. These changes develop over a period of days after injury. In addition, it is our clinical impression that progressive respiratory failure is not always proportional to the intensity of smoke exposure. It is also possible for thermally injured patients to experience acute lung injury from the systemic effects of the inflammatory response to severe cutaneous burns. Thus it is not uncommon to see acute lung injury in children with large scald burns when inhalation of hot or caustic gases did not occur. This makes it difficult to determine what component of respiratory failure is due to inhalation injury and what component is an effect of systemic inflammation associated with large cutaneous burns.

On initial presentation, patients with inhalation injury may have relatively normal gas exchange as evaluated by arterial blood gas analysis, and the chest radiograph is often normal. In the absence of evidence of respiratory distress it is important to recognize features from the history and physical examination that reveal risk factors for inhalation injury. Normal gas exchange and chest radiograph on admission do not rule out significant inhalation injury. Early diagnosis is important to recognize the potential for airway compromise, manage fluid resuscitation, and to recognize systemic toxicity that may lead to permanent neurological deficits if not promptly treated.

History pertinent to the diagnosis of inhalation injury includes information regarding the mechanism of injury and the intensity of exposure. Mechanisms of injury that carry significant risk of inhalation injury include not only exposure to smoke from a fire, but blast injury that can force hot gases past the larynx, steam burns that can not only burn the upper airway but carry heat to structures below the larynx, and exposure to caustic fumes, as occurs in some industrial accidents. Information regarding the mechanism of injury also includes the source of combustion, which could identify specific chemical irritants. The history can also provide information regarding the intensity of exposure. Duration of exposure is an important determinant of intensity of exposure. When a victim's avoidance behavior is impaired, as when trapped in an enclosed space, intoxicated, unconscious, or in the case of extremes of age, exposure to injurious inhalants is increased.

History of the mechanism of injury is especially important in the case of scalds due to ingestion of hot liquids. Although patients may appear asymptomatic initially, oropharyngeal scalds have led to delayed fatal airway occlusion. Intraoral scalds can present in a manner similar to epiglottitis. The larynx should be examined for evidence of compromise in all patients who present with significant risk (either by history or physical exam) of intraoral scald.

The physical examination can reveal additional risk factors for inhalation injury. We guard our face vigorously, and the presence of burns to the face or singed eyebrows or nasal hair implies a very intense exposure to heat. When gases hot enough to burn tissue are near the airway inlet it suggests that oropharyngeal or nasopharyngeal structures may also suffer thermal injury. Soot deposits on the face and carbonaceous sputum suggests inhalation of smoke. Physical examination may reveal signs and symptoms such as stridor, hoarseness, drooling, and dysphagia that are considered classic evidence of thermal injury to the oropharynx. Presence of these findings, however, does not always indicate that tracheal intubation is necessary. However, as described below, when patients are considered at risk for upper airway thermal injury and occlusion, a priority is to evaluate the upper airway for impending occlusion that may be prevented by early tracheal intubation.

In addition to the history and physical examination, there are diagnostic tools that may be used to confirm a diagnosis of inhalation injury or to follow the progression of injury. Since manifestations of respiratory dysfunction may be delayed after inhalation injury, pulse oximetry and arterial blood gas analysis are insensitive indicators of lung injury during the initial stages. Despite this, it is important to employ these tools as soon as possible. Early impairment of gas exchange is an ominous sign of severe injury and requires early and aggressive intervention. Diagnosis of CO or cyanide toxicity may be facilitated by blood gas analysis. It is also important to have baseline values to judge progress.

Lee and O'Connel assessed the chest radiographs of 45 patients admitted to their hospital following injury in an enclosed space fire. Of those patients, 33 had abnormal findings consistent with inhalation injury. They suggested that the initial radiograph is an important predictor of injury and need for mechanical support of respiration. However, Wittram and Kenny examined admission chest radiographs over a 3-year period for all patients requiring ventilatory support for inhalation injury. . Out of 25 patients, 12 had normal initial chest radiographs despite ultimately requiring mechanical ventilation. The initial chest radiograph is considered an insensitive early indicator of parenchymal injury after smoke inhalation. Although an admission chest radiograph should be obtained in all patients suspected of inhalation injury a normal study does not rule out the possibility of significant pulmonary injury.

Flexible fiberoptic bronchoscopy was early recognized as a powerful tool in the diagnosis of inhalation injury. Fiberoptic bronchoscopy allows direct visualization of tissue damage to the upper airway and bronchi from heat and chemical irritants. This procedure can quickly and reliably identify patients with upper airway compromise who will benefit from intubation and, at the same time, avoid intubation of patients who will not benefit. Bronchoscopic evidence of inhalation injury includes soot deposits, erythema, edema (as indistinct tracheal rings and/or blunting of the carina), mucosal blisters and erosions, hemorrhages, and bronchorrhea ( Fig. 17.1 ). Flexible fiberoptic bronchoscopy has been considered the gold standard for diagnosis of inhalation injury and is often used to confirm the diagnosis of inhalation injury. However, Hunt noted that, in some cases, bronchoscopy performed soon after injury may not show mucosal injury. In addition, because acute lung injury and tracheobronchitis can be a result of systemic inflammation due to cutaneous burns, endoscopic changes after 36–48 hours may be caused by mechanisms other than inhalation of chemical irritants. As an example, a small fraction of young pediatric patients with large scald injuries develop acute lung injury and require mechanical ventilation. Bronchoscopic examination of these patients can reveal inflammatory changes characteristic of smoke inhalation. Moreover, although fiberoptic bronchoscopy can definitively identify tissue damage from inhalation injury, it has been recognized that the observed changes are relatively proximal and may be more severe than more peripheral parenchymal injuries. As a result, a bronchoscopic diagnosis of inhalation injury does not always identify which patients will experience progressive respiratory dysfunction.

Numerous attempts have been made to grade severity of inhalation injury based on bronchoscopic findings in order to identify patients who may need increased levels of support regarding airway management, respiratory support, or increased volume of fluid during the initial resuscitation. In their retrospective study, Hassan et al. found that mortality among patients with inhalation injury correlated with severity of bronchoscopic findings. However, Spano and colleagues also performed a retrospective review to evaluate the effectiveness of bronchoscopy to predict outcomes for patients with inhalation injuries. They used an abbreviated injury score (AISD INH) introduced by Endorph and Gameli that grades the severity of bronchoscopic findings associated with smoke inhalation injury. They also reviewed the three previous studies that specifically compared the AIS INH with clinical outcome. All these studies found trends that were suggestive of poorer clinical outcome in patients with more severe bronchoscopic changes, but these trends were not statistically significant. Spano et al. suggested that further studies to allow refinement in descriptions of severity grades will be needed before bronchoscopic evaluation produces reliable prognostic information.

It is generally accepted that patients with smoke inhalation injury require increased fluid volume for resuscitation of cutaneous burns. Increases of up to 40–50% have been observed. While this has been a fairly consistent observation, the value of bronchoscopy in predicting increased fluid needs has been inconsistent. Endorf and Gamelli found that an initial P:F ratio of less than 350 was a more reliable predictor of increased fluid requirements than was diagnosis of inhalation injury by bronchoscopy. In their patients, fluid requirements during initial resuscitation were not correlated with severity of findings at initial bronchoscopy. Cancio and colleagues found that diagnosis of inhalation injury per se was not related to increased fluid needs but that mechanical ventilation was related. They suggested that mechanical ventilation may be a surrogate variable for more severe inhalation injury, and this could explain its closer correlation with fluid needs. These observations reinforce the possibility that findings of bronchoscopy are relatively proximal and may not always reflect the severity of more distal parenchymal injury. Need for mechanical ventilation and decreased P:F are more dependent on parenchymal injury and therefore may be more accurate predictors of inhalation injury.

Mackie and colleagues have offered an alternative mechanism for increased fluid requirements in burn patients who also have inhalation injury. They found that ventilated patients with burns but no inhalation injury required more fluid than did patients with similar burns but who did not receive mechanical ventilation. Fluid balance was not significantly affected in patients who also had burns, inhalation injury, and were mechanically ventilated. Mackie suggests that positive pressure ventilation increases intrathoracic pressure, which impairs venous return to the heart. More intravenous fluid is then required to maintain cardiac preload. These findings are consistent with a greater effect of mechanical ventilation on fluid balance than inhalation injury.

Although bronchoscopy does not reliably predict respiratory failure, endoscopic assessment of the upper airway has been found to be highly useful in identifying patients with glottic or supraglottic changes who would benefit from early intubation. Just as important, this exam also helps avoid unnecessary intubations that expose patients to serious risk without benefit.

Radionuclide studies represent an additional tool that has been used to provide evidence of pulmonary injury distal to the more proximal views permitted by flexible bronchoscopy. Intravenously administered xenon-133 is excreted by the lungs and exhaled. Delayed clearance of xenon-133 is a sensitive indicator of inhalation injury. Lung scintigraphy using technetium-99 aerosol inhalation has also been used to identify areas of pulmonary dysfunction in patients with respiratory dysfunction after smoke inhalation. Delayed clearance and inhomogeneous lung distribution of radioactivity are evidence of injury. These studies are sensitive indicators of inhalation injury, but interpretation can be confounded by pre-existing lung disease, and it may be difficult to perform these studies in critically ill burn patients.

The poor correlation of clinical outcome with severity of bronchoscopic findings in patients with smoke inhalation injury has led some clinicians to evaluate the possibility that chest computed tomography (CT) could provide more accurate prognostic information. CT can reveal regional differences in structural changes that impair pulmonary function, such as atelectasis, consolidation, and fibrosis, more effectively than chest radiographs. Yamamura and others used admission chest CT to measure bronchial wall thickness in patients with suspected inhalation injury. Increased bronchial wall thickness was associated with total number of ventilator days, ICU days, and pneumonia. Oh and colleagues obtained CT scans shortly after admission for patients at risk for smoke inhalation. A grading system was used to provide a score for each CT scan based on interstitial markings, ground-glass opacification, and consolidation. This score was compared with clinical outcomes described as a composite endpoint based on pneumonia, ALI/ARDS, and mortality. Injury detected by bronchoscopy alone was associated with an 8.3-fold increase in the composite score, while the combination of bronchoscopic evidence of inhalation injury and a high CT score was associated with a 12.7-fold increase in the composite endpoint. The authors were cautiously optimistic that with additional clinical experience and refinement of image analytical techniques, CT scans alone or in combination with bronchoscopic evaluation may provide more accurate early prognosis for patients with smoke inhalation injury. The risks of taking an acutely ill patient to the CT scanner were also acknowledged.