Understanding Smoke Inhalation Injury and the Need for Objective Scoring

Smoke inhalation injury is one of the most dangerous and underappreciated complications of burn injury and fire exposure. While cutaneous burns are immediately visible and straightforward to quantify using tools such as the Rule of Nines or Lund-Browder chart, the pulmonary component of smoke inhalation injury is largely invisible on physical examination in the early hours after exposure. Yet it is the pulmonary injury, not the cutaneous burn, that accounts for the majority of early mortality in fire victims. Inhalation injury is present in approximately 10 to 30 percent of burn admissions and increases mortality by 20 to 40 percent over comparable burns without inhalation injury.

The diagnosis of inhalation injury has traditionally relied on a constellation of clinical findings: history of exposure in an enclosed space, carbonaceous sputum, singed nasal hairs, facial burns, hoarseness, stridor, and bronchoscopic findings of erythema, edema, mucosal sloughing, and carbonaceous deposits in the airway. While bronchoscopy remains the gold standard for diagnosing supraglottic and tracheobronchial injury, it provides limited information about parenchymal lung injury. It also carries procedural risk and cannot be repeated frequently for serial assessment.

Computed tomography (CT) of the chest has emerged as a powerful non-invasive tool for assessing the parenchymal component of smoke inhalation injury. CT can detect subtle early abnormalities such as ground-glass opacity, bronchial wall thickening, air trapping, and developing consolidation that are invisible on plain chest radiograph. The RADS (Radiologist's Score) for Smoke Inhalation Injury is a structured, quantitative CT-based scoring system designed to standardize radiological assessment, improve inter-rater reliability among radiologists, and provide a clinically actionable numerical score that correlates with injury severity and predicts patient outcomes.

Pathophysiology of Smoke Inhalation Injury

To understand what RADS measures and why its individual components matter, it is essential to understand the pathophysiological mechanisms by which smoke injures the respiratory tract.

Thermal Injury to the Upper Airway

Heat from inhaled gases and smoke causes direct thermal injury primarily to the supraglottic airway. The nasopharynx, oropharynx, and larynx are efficient heat exchangers, and by the time inhaled air reaches the vocal cords, it has been cooled substantially. Subglottic thermal injury from inhaled gases is therefore uncommon except with steam inhalation, which carries far more thermal energy than dry air at the same temperature. Upper airway thermal injury causes mucosal edema that can rapidly progress to complete airway obstruction, explaining why early intubation is a priority in patients with signs of upper airway injury.

Chemical Injury to the Lower Airway and Parenchyma

The lower airway and lung parenchyma sustain injury primarily through chemical mechanisms rather than thermal injury. Combustion of structural materials, furniture, plastics, and synthetic fabrics generates hundreds of toxic compounds, including:

• Carbon monoxide (CO): A colorless, odorless gas that binds hemoglobin with 250 times the affinity of oxygen, causing cellular hypoxia and direct cytotoxic effects on the myocardium and nervous system
• Hydrogen cyanide (HCN): Released from combustion of nitrogen-containing materials (wool, silk, polyurethane, nylon); inhibits cytochrome c oxidase and causes histotoxic hypoxia by preventing cellular oxygen utilization
• Acrolein: A highly reactive aldehyde produced from combustion of wood, cotton, and polyolefins; causes direct mucosal injury, bronchoconstriction, and delayed pulmonary edema
• Nitrogen oxides (NOx): Dissolve in airway secretions to form nitric acid and nitrous acid, causing chemical bronchitis and pulmonary edema
• Sulfur dioxide (SO2) and hydrogen chloride (HCl): Dissolve in mucus to form sulfurous and hydrochloric acids, causing bronchospasm and mucosal ulceration
• Particulate matter: Fine particles carry adsorbed chemicals deep into the alveolar space; particles smaller than 2.5 micrometers reach the alveoli and trigger macrophage activation and inflammatory cascades

Inflammatory Response and Secondary Injury

The initial chemical injury triggers a robust inflammatory response characterized by neutrophil infiltration, cytokine release, reactive oxygen species generation, and disruption of the alveolar-capillary barrier. This leads to:

• Increased alveolar capillary permeability and non-cardiogenic pulmonary edema
• Loss of surfactant function, causing alveolar collapse and atelectasis
• Bronchial wall edema and secretion accumulation, predisposing to mucus plugging and lobar collapse
• Impaired mucociliary clearance, increasing the risk of secondary bacterial pneumonia
• Progression to acute lung injury (ALI) and acute respiratory distress syndrome (ARDS) in severe cases

The timeline of these events is important. Immediate clinical findings may be subtle while pathophysiological injury is rapidly progressing. CT performed in the first 24 to 48 hours after exposure captures the initial injury pattern and can detect abnormalities before they are clinically apparent, making early imaging a valuable prognostic and management tool.

The Role of Chest CT in Smoke Inhalation Assessment

Plain chest radiography has limited sensitivity for early smoke inhalation injury. In the first 24 hours, the chest radiograph is often normal or shows only subtle peribronchial haziness, even in patients with significant parenchymal injury confirmed by CT. The relatively low sensitivity of chest radiography is attributable to its inability to resolve subtle density differences and its superimposition of anterior and posterior structures, which obscures focal opacities, early air trapping, and bronchial wall changes.

High-resolution CT (HRCT) of the chest provides exquisite detail of bronchial wall anatomy, airspace opacification patterns, air trapping on expiratory imaging, and subtle ground-glass changes that precede consolidation. CT allows the specific pattern, distribution, and extent of injury to be characterized in a way that has direct prognostic implications.

The key CT findings in smoke inhalation injury and their pathophysiological correlates are:

Development of the RADS Scoring System

Prior to the development of RADS, radiological assessment of smoke inhalation injury was largely qualitative and unstandardized. Different radiologists used different terminology, described findings in different ways, and provided assessments that varied widely in their reproducibility and clinical utility. There was no validated tool that a clinician could use to translate a radiological report into a meaningful severity score or a prognostic estimate.

The RADS was developed to address these limitations by creating a structured, reproducible, and quantitative scoring framework applicable to any chest CT performed in a smoke inhalation patient. The system was designed with several principles in mind: it should be based on CT findings with established pathophysiological significance, it should cover the entire lung in a systematic zone-based approach, the individual scoring elements should be simple enough to evaluate reliably without extensive training, and the total score should correlate with clinically meaningful outcomes such as the need for mechanical ventilation, ICU length of stay, and mortality.

RADS Scoring Methodology

Lung Zone Division

The RADS system divides each lung into three zones based on anatomical landmarks visible on CT:

• Upper zone: Lung tissue above the carina
• Middle zone: Lung tissue between the carina and the inferior pulmonary veins
• Lower zone: Lung tissue below the inferior pulmonary veins

This creates six zones in total (three per lung: right upper, right middle, right lower, left upper, left middle, left lower). Each zone is evaluated independently for each of the scoring parameters.

Scored CT Parameters

For each of the six lung zones, the following CT findings are assessed and scored:

GGO and consolidation are each scored from 0 to 3 based on the percentage of the zone involved. Bronchial wall thickening and air trapping are binary (0 or 1) per zone. The maximum score per zone is therefore 3 + 3 + 1 + 1 = 8. With six zones, the maximum possible total RADS score is 48.

Score Calculation

The total RADS score is the sum of all individual zone scores across all six zones:

RADS Total = Σ (GGO score + Consolidation score + Bronchial wall thickening score + Air trapping score) for each of the 6 zones

A higher total RADS score indicates more extensive and more severe pulmonary involvement. The score is designed to be calculated from a single CT acquisition, typically the first CT obtained after presentation or stabilization.

Score Interpretation and Clinical Thresholds

These thresholds serve as clinical guidance and should always be interpreted in the context of the patient's overall clinical presentation, oxygenation status, ventilatory mechanics, and associated injuries. A patient with a low RADS score but significant upper airway burns may still require intubation for airway protection. Conversely, a patient with a moderate RADS score but excellent reserve and no co-morbidities may tolerate a trial of non-invasive respiratory support.

Clinical Correlations and Prognostic Value

Prediction of Mechanical Ventilation

One of the primary applications of RADS is identifying patients at high risk of requiring mechanical ventilation. Studies evaluating CT-based pulmonary scoring in smoke inhalation patients have demonstrated that quantitative CT scores at presentation are significantly associated with subsequent need for intubation and mechanical ventilatory support. Patients with higher CT scores are more likely to deteriorate over the first 24 to 72 hours as inflammatory edema, surfactant dysfunction, and secretion accumulation compound the initial injury.

The ability to predict ventilatory requirements early is clinically important because preemptive intubation in a controlled setting (before respiratory failure necessitates emergent intubation) allows for optimal patient positioning, pre-oxygenation, and choice of anesthetic agents. In burn patients with large surface area burns, early intubation also facilitates wound care and surgical procedures.

Prediction of ICU Length of Stay

Higher RADS scores correlate with prolonged ICU length of stay, reflecting the greater nursing intensity, respiratory therapy requirements, and time needed for resolution of pulmonary injury in patients with more severe CT findings. This correlation has resource planning implications: burn centers with high-RADS patients can anticipate extended ventilator days, increased nursing ratios, and the potential need for advanced respiratory support modalities such as high-frequency oscillatory ventilation (HFOV) or extracorporeal membrane oxygenation (ECMO).

Association with Mortality

Smoke inhalation injury is independently associated with mortality in burn patients, and the degree of pulmonary injury quantified by CT correlates with this mortality risk. Patients with severe CT-defined pulmonary injury (high RADS) face a substantially elevated risk of death compared to those with minimal or no CT abnormalities, even after controlling for burn size and patient age. This prognostic information can support conversations with families about prognosis and the goals of care.

RADS in the Context of the Baux Score

The Baux Score (age + total body surface area burned) is the traditional and most widely used predictor of mortality in burn patients. Revised Baux Score adds an additional 17 points for the presence of inhalation injury, acknowledging the independent mortality contribution of pulmonary injury. RADS complements the Baux framework by quantifying the severity of the inhalation component rather than simply recording its presence or absence. A patient with a Revised Baux Score elevated by 17 points for inhalation injury will have a very different prognosis if their RADS score is 2 versus 22.

Individual Scoring Parameters: Clinical Significance

Ground-Glass Opacity

Ground-glass opacity (GGO) on CT represents partial filling of the airspace with fluid, exudate, or cellular material, or thickening of the alveolar walls without complete opacification. In smoke inhalation, early GGO reflects the initial exudative response of the alveolar-capillary unit to inhaled toxins. It precedes consolidation and may represent a potentially reversible phase of injury if the offending stimulus is removed and supportive care is optimized.

The extent of GGO is a sensitive indicator of early parenchymal involvement. A patient presenting within hours of exposure with bilateral diffuse GGO has sustained significant alveolar-capillary injury and is at high risk of rapid progression to ARDS. Serial CT can track whether GGO is improving (decreasing extent and density) or worsening (progression to consolidation).

Consolidation

Consolidation indicates complete opacification of the airspace, reflecting dense filling with exudate, inflammatory cells, hemorrhage, or a combination. In smoke inhalation, early consolidation may reflect direct alveolar flooding from chemical injury to the alveolar-capillary barrier. Later consolidation (24–72 hours after exposure) more commonly represents secondary pneumonia, aspiration, or the proliferative phase of ARDS.

Consolidation has a greater impact on gas exchange than GGO because consolidated lung represents complete anatomical shunt: blood passing through consolidated lung segments receives no oxygen. Extensive consolidation (particularly bilateral and lower lobe predominant) predicts severe hypoxemia, high ventilatory requirements, and prolonged ventilator dependence.

Bronchial Wall Thickening

Bronchial wall thickening on CT reflects mucosal edema, submucosal inflammatory cell infiltration, and peribronchial inflammation in the larger airways. It is a direct CT correlate of the chemical bronchitis and airway inflammation seen on bronchoscopy. While bronchial wall thickening itself does not impair gas exchange as directly as alveolar opacification, it has important consequences: impaired mucociliary clearance, increased secretion production, and predisposition to mucus plugging and secondary bacterial pneumonia.

In patients with bronchial wall thickening, aggressive airway humidification, chest physiotherapy, and early bronchoscopic suctioning are important management priorities. Routine prophylactic antibiotics are not indicated for bronchial wall thickening alone, but a high index of suspicion for early-onset pneumonia is appropriate.

Air Trapping and Mosaic Attenuation

Air trapping and mosaic attenuation on CT reflect small airway obstruction from edema, bronchospasm, secretions, or early mucus plugging. These findings are best appreciated on expiratory CT sequences, where normal lung decreases in volume and increases in attenuation while air-trapped regions remain hyperlucent. On inspiratory CT, mosaic attenuation (alternating regions of higher and lower lung density) may be seen when small airway obstruction is present.

Functionally, air trapping and small airway obstruction produce obstructive physiology: increased residual volume, dynamic hyperinflation, intrinsic positive end-expiratory pressure (auto-PEEP), and ventilation-perfusion mismatch. In intubated patients, air trapping contributes to dynamic hyperinflation, which can impair venous return, reduce cardiac output, and increase the risk of barotrauma. Recognition of air trapping on CT should prompt adjustment of ventilator settings to allow adequate expiratory time.

RADS in the Context of Combined Burn and Inhalation Injury

Patients who sustain both significant cutaneous burns and smoke inhalation injury represent the highest-acuity subgroup in burn care. The combination produces a synergistic pathophysiological effect that is not simply additive. Large cutaneous burns cause massive systemic inflammatory response, cytokine release, and fluid shifts that amplify the pulmonary inflammatory response initiated by direct inhalation injury. Aggressive fluid resuscitation required for burn management (following Parkland or modified Brooke formulas) increases the risk of pulmonary edema, which can exacerbate CT findings and the RADS score over the first 24 to 48 hours.

In combined injury patients, the RADS score provides an objective baseline assessment of the pulmonary contribution to overall severity. Serial RADS calculation on repeat CT (where clinically indicated) can differentiate worsening pulmonary injury from burn-related fluid overload, pneumonia superimposed on inhalation injury, or other complications such as pulmonary embolism. Each of these complications has a different CT signature and a different management approach.

Differential CT Patterns in Smoke Inhalation Subtypes

Chemical Bronchitis Pattern

Patients with predominantly large airway chemical injury (from highly water-soluble gases such as ammonia, chlorine, or sulfur dioxide that dissolve in the proximal airway secretions) exhibit bronchial wall thickening as the dominant CT finding. Parenchymal opacification may be minimal or absent. The RADS score in these patients may be modest in the GGO and consolidation components but elevated in the bronchial wall thickening component. Clinically, these patients present with productive cough, bronchospasm, and stridor rather than hypoxemia.

Parenchymal Injury Pattern

Patients exposed to poorly water-soluble gases (such as nitrogen dioxide, phosgene, or ozone) that penetrate to the alveolar level without significant proximal airway absorption present with predominantly parenchymal injury. GGO and consolidation dominate the CT picture, and bronchial wall thickening may be less prominent. These patients may have a deceptive initial clinical presentation (minimal immediate symptoms) followed by abrupt deterioration 6 to 24 hours after exposure as non-cardiogenic pulmonary edema develops. The RADS score in these patients may be initially low and then dramatically increase on repeat CT, highlighting the value of serial imaging.

Mixed Pattern

House fires and industrial fires typically generate a complex mixture of combustion products across a wide range of solubility and reactivity. Most clinical smoke inhalation patients exhibit a mixed pattern on CT with contributions from all scoring parameters. The RADS score in these patients reflects the combined burden of bronchial injury (wall thickening, air trapping) and parenchymal injury (GGO, consolidation) and is most predictive of overall clinical severity.

Management Implications of RADS Stratification

Low RADS Score (0–4): Conservative Management

Patients with minimal CT findings can generally be managed conservatively with supplemental oxygen, airway humidification, chest physiotherapy, and close monitoring. Bronchodilators are appropriate if bronchospasm is present clinically or suggested by air trapping on CT. These patients should be monitored for 24 to 48 hours for clinical deterioration, as pulmonary injury from smoke inhalation can progress during this window even from a low initial baseline.

Moderate RADS Score (5–12): Intermediate Intervention

Patients with moderate CT findings require a higher level of monitoring and proactive respiratory support. Non-invasive positive pressure ventilation (CPAP or BiPAP) may provide a bridge for those with significant GGO and early consolidation who are not yet in respiratory failure. Bronchoscopy should be considered both diagnostically (to assess the degree of airway mucosal injury) and therapeutically (to remove carbonaceous deposits and secretions). The threshold for early intubation should be lower in these patients, as they are at high risk for rapid deterioration, particularly with superimposed pneumonia or fluid mobilization from burn resuscitation.

High RADS Score (≥13): Aggressive Intervention

Patients with high RADS scores are at high risk of respiratory failure and should be managed in a burn ICU with early intubation and lung-protective ventilation. Lung-protective ventilatory strategies (tidal volumes 6 mL/kg ideal body weight, plateau pressure below 30 cmH2O, PEEP titrated to oxygenation) should be initiated from the outset. Prone positioning should be considered early in patients with severe bilateral infiltrates and refractory hypoxemia meeting ARDS criteria. For patients failing conventional mechanical ventilation, HFOV, airway pressure release ventilation (APRV), or ECMO may be considered at centers with appropriate expertise.

Nebulized heparin alternating with N-acetylcysteine (NAC) has been studied in smoke inhalation injury and has shown benefit in reducing cast formation and improving pulmonary function in some trials; patients with high RADS scores are the most likely to benefit from this adjunctive approach. Systemic corticosteroids are generally avoided in acute smoke inhalation injury outside of specific indications (such as reactive airways dysfunction syndrome or bronchospasm refractory to bronchodilators) due to the risk of secondary infection.

Timing and Repeat Imaging Considerations

The RADS score is ideally calculated from a CT performed within the first 24 hours of exposure, before fluid resuscitation-related pulmonary edema confounds the picture and before secondary pneumonia develops. However, clinical priorities (airway management, hemodynamic resuscitation) may delay CT acquisition, and initial CT may be performed after intubation and initiation of mechanical ventilation, which itself changes the CT appearance.

Repeat CT imaging is appropriate in patients who fail to improve clinically, who develop new fever suggesting secondary pneumonia, or in whom a change in management is being considered (such as tracheostomy planning or escalation to ECMO). Repeat RADS calculation allows quantitative comparison of injury extent over time and can document both improvement and deterioration with greater precision than clinical assessment or plain radiography alone.

It should be noted that CT transport carries risks in critically ill burn patients, including hemodynamic instability, ventilator disconnection, and hypothermia. The decision to perform CT must weigh the clinical benefit of the information against these procedural risks. For hemodynamically unstable patients, portable chest radiography may be the only feasible imaging option.

RADS Versus Other Pulmonary Injury Assessment Tools

Flexible Bronchoscopy Grading Systems

Several bronchoscopic grading systems for airway injury in inhalation patients have been developed, including the Abbreviated Injury Scale (AIS) for inhalation injury and the Grades 0–4 system by Endorf and Gamelli. These systems score mucosal erythema, edema, carbonaceous deposits, bronchorrhea, and mucosal ulceration visible through the bronchoscope. Bronchoscopic grading characterizes the tracheobronchial injury but provides no information about parenchymal disease. RADS and bronchoscopic scoring are therefore complementary rather than competing assessment tools.

PaO2/FiO2 Ratio

The PaO2/FiO2 (P/F) ratio is the most widely used physiological metric for quantifying the degree of gas exchange impairment and is central to ARDS severity classification (Berlin Definition). The P/F ratio reflects the functional consequence of pulmonary injury but is influenced by ventilator settings (PEEP, FiO2), patient position, and hemodynamics in addition to the structural injury itself. RADS provides a structural assessment of injury extent that is independent of ventilator management and therefore complements rather than replaces the P/F ratio.

Murray Lung Injury Score

The Murray Lung Injury Score combines chest radiograph findings, P/F ratio, PEEP level, and respiratory system compliance into a composite score for acute lung injury severity. Unlike RADS, it incorporates physiological parameters and therefore reflects both the structural injury and the functional response. Murray Score is most useful for serial assessment of ALI/ARDS severity in intubated patients; RADS is most useful for initial structural characterization of inhalation-specific injury at presentation.

Important Limitations of the RADS System

• CT timing dependence: The RADS score is sensitive to the timing of CT acquisition relative to exposure. Very early CT (within hours of exposure) may underestimate injury severity because inflammatory edema and exudation have not yet fully developed. Late CT (more than 48 hours after exposure) may overestimate initial injury severity due to superimposed complications such as secondary pneumonia, fluid overload, or ARDS from causes other than direct inhalation injury.
• Inter-rater variability: Although RADS was developed to standardize radiological assessment, some degree of inter-rater variability in scoring GGO and consolidation extent (particularly the 25% and 50% cutoffs) persists. Formal inter-rater reliability studies have shown moderate to good agreement, but disagreement in borderline cases can affect total score and clinical categorization.
• Radiation exposure: CT of the chest involves ionizing radiation exposure. In burn patients who may require multiple imaging studies over a prolonged ICU course, cumulative radiation dose is a consideration. This concern is less acute in the emergency setting but should be factored into decisions about serial imaging.
• No direct assessment of airway injury severity: RADS does not score the degree of tracheobronchial mucosal injury visible on bronchoscopy. A patient with severe endobronchial burns and carbonaceous deposits may have a low RADS score if the parenchymal injury is minimal. In such patients, RADS may underestimate overall respiratory system injury severity.
• Confounding by pre-existing lung disease: Patients with pre-existing chronic obstructive pulmonary disease (COPD), asthma, interstitial lung disease, or prior pneumonia may have CT abnormalities at baseline that contribute to the RADS score independent of acute smoke inhalation injury. In these patients, comparison with any available prior imaging is essential to distinguish acute from chronic findings.
• Not validated for pediatric populations: Most RADS validation data come from adult burn patients. Pediatric airways and lung parenchyma have different anatomy, compliance characteristics, and reserve, and the prognostic thresholds derived from adult data may not apply directly to children.
• Does not capture extrapulmonary effects of smoke exposure: RADS is a pulmonary scoring system only. It does not capture the systemic effects of carbon monoxide toxicity, hydrogen cyanide poisoning, systemic inflammatory response, or neurological injury from hypoxia that are equally important determinants of outcome in smoke inhalation patients.
• Not a substitute for clinical judgment: RADS provides important supplementary information but does not replace clinical assessment. Decisions about intubation, ICU admission, and escalation of care must integrate RADS findings with the full clinical picture, including oxygenation, work of breathing, hemodynamics, associated injuries, and patient comorbidities.