Fire-Eater’s Lung

ArticlePlus Click on the links below to access all the ArticlePlus for this article. Please note that ArticlePlus files may launch a viewer application outside of your web browser. https://links.lww.com/MD/A1 Introduction Accidental aspiration during “fire-eating” demonstrations is responsible for the development of an acute pneumonitis known as “fire-eater’s lung.” Different substances called “pyrofluids” are used by fire-eaters, but the most common is the petroleum derivative kerdan, characterized by its reduced viscosity and, unfortunately, its rapid diffusion throughout the bronchial tree after sudden accidental aspiration. This type of accident occurs during the show, as the fire-eater blows out a mouthful of kerdan against a burning stick. Flames seem to emanate directly from the mouth of the fire-eater. Kerdan is composed of petroleum-distillated products like toluene, xylene, ethylbenzene, propylbenzene, and methylbenzene, which differ from kerosene composition. To date, only short series or isolated observations of fire-eater’s lung have been reported. In the present study we report clinical, functional, and radiologic characteristics of 17 patients with fire-eater’s lung, seen immediately after the acute episode and followed for a period of 6 months. Methods Seventeen patients with fire-eater’s lung were included in this study from October 1996 to January 2001. Patients were evaluated at the time of the initial accident, then at 3 and 6 months. All patients were recruited from the Pneumology Department of University Hospitals of Lille, Rennes, Paris, and Marseille, France. Only patients using kerdan as the component for fire-eating were included. All were evaluated in the first 12 hours following the initial kerdan aspiration. Follow-up was performed at 3 and 6 months for clinical (dyspnea, chest pain, fever) and radiologic (chest X-ray and computed tomography [CT] scan) features, and functional respiratory tests, including forced expiratory volume in 1 second (FEV 1 ), slow vital capacity (SVC), FEV 1 /SVC, and diffusing capacity for carbon monoxide. Patients with underlying respiratory disease and patients using petroleum derivatives other than kerdan were not included in the study. The results of the present work were compared with a review of the literature performed using a MEDLINE PubMed computerized search (National Library of Medicine, Bethesda, MD). References were selected using the terms “fire-eater,” “fire-eaters,” and “kerdan.” We added for comparison the cases of fire-eaters who used pyrofluids other than kerdan or pyrofluids not clearly identified (3,4,7,12,18,21,25). Results A total of 17 patients (14 male, 3 female) were included in the study; mean age was 24.5 years (range, 19–42 yr). All patients were previously considered healthy. Ten were current smokers, 3 were drug addicted. Ten of them had been acting as ”fire-eaters” for several years (Table 1). Two had previously experienced kerdan aspiration accidents. Because 1 of the 17 patients was admitted to the intensive care unit (ICU), we describe separately the group of 16 patients admitted to the pneumology section and the 1 patient admitted to the ICU.TABLE 1: Clinical features of fire-eater’s lung (n = 16 of 17 patients)Clinical description (16/17 patients) During the flame-blowing act, the fire-eaters all accidentally aspirated a small but unquantifiable amount of kerdan. The first symptoms appeared immediately, with a burning sensation in the mouth and throat, followed by gagging, choking, coughing, and grunting respirations. After a mean period of 8 hours (4 ± 4), all patients had dramatic pleuritic pain, dyspnea, and cough. Hemoptysis (<20 mL/day) was observed immediately in 1 case and in the first 8 hours in 3 cases, but never required specific therapy. Fever, from 38.5 °C to 39.5 °C, was reported in all cases (see Table 1). Radiologic abnormalities (16/17 patients) Chest X-rays performed in the first 12 hours following the aspiration showed, in all cases, unilateral or bilateral alveolar infiltrates, located on the middle and lower parts of the lungs (Figure 1). Atelectasis and pleural effusion were reported occasionally (Table 2). Twelve patients underwent a high-resolution CT examination, which showed cavitary lesions (pneumatoceles) in 5 cases (generally small cavities less than 2 cm in diameter) in the bilateral dense lung infiltrates (Figure 2).TABLE 2: Radiologic findings in fire-eater’s lung (n = 16)Fig. 1: Chest radiograph of a 25-year-old male fire-eater who accidently inhaled kerdan shows a predominant alveolar infiltrate in the right lower lobe.Fig. 2: Corresponding computed tomography scan from same patient demonstrates a solid infiltrate containing cavitary nodules of various sizes (pneumatoceles) in the right lower lobe.Laboratory tests (16/17 patients) Blood tests showed elevated leukocytosis (up to 12,000 cells/mm 3 in 12 cases) (range, 12,000–33,000 cells/mm 3 ; normal, 4,000–10,000 cells/mm 3 ) with predominance of neutrophils and high serum levels of C-reactive protein (range, 100–250 mg/L) (Table 3). A moderate fall in arterial oxygen tension was noted occasionally (9 cases). No abnormalities of renal, hydroelectrolitic, and hepatic values were described.TABLE 3: Laboratory findings (n = 16)In 3 cases, fiberoptic bronchoscopy with bronchoalveolar lavage (BAL), was performed. Macroscopic examination of the airways showed a diffuse inflammation of the bronchial mucosa. Total and differential cell counts of BAL were normal in 2 cases. The third case showed a total cell count of 300 × 10 3 cells/mL with 16% neutrophils, 80% macrophages, and 4% lymphocytes. Bacteriologic analysis of BAL was sterile for the 3 cases, although secretions were suppurative in 2 cases. Bacteriologic tests (16/17 patients) Cytobacteriologic analysis of sputum was performed in the first 24 hours and was found sterile in the 16 patients. Because fever persisted, these analyses were repeated in 4 patients after 72 hours, but remained negative. Treatment (16/17 patients) Among the 16 patients, 12 received antibiotics at admission, and 6 were treated with additional systemic corticosteroids (30–50 mg eq. prednisolone per day for 5 days). Patients were hospitalized for a mean (± SD) period of 8 (± 4) days and treated with decreasing corticosteroid doses on 5 days. All of them received supplemental O 2 for a mean period of 3 days. Follow-up and outcome During the hospitalization period, respiratory symptoms usually disappeared within 72 hours, although fever persisted more than 5 days in 4 patients. Chest X-ray examination showed a complete resolution of pulmonary abnormalities within 14 days in 15 of 16 cases. A follow-up of clinical, radiologic, and pulmonary function tests was performed at the end of the third and sixth month, for 13 of 16 patients (3 patients were lost to follow-up). Three months after hospitalization, the CT scan demonstrated discrete scarring of the previous pneumatoceles in 2 of 13 patients. Moreover, restrictive impairment of ventilatory function (2/12) and decreased diffusing capacity (2/12) were observed. One of the 2 patients with a restrictive impairment of ventilatory function had experienced a previous kerdan inhalation. Six months later, radiologic and pulmonary function tests were all normal, and clinical symptoms disappeared completely. No difference in the clinical, functional, and radiologic evolution was observed in patients who underwent corticosteroid treatment compared with those who did not. Patient admitted to the ICU (1/17 patients) One of the 17 patients, who was hemodynamically unstable as a consequence of a severe acute lung injury, was admitted to an ICU. Immediately after kerdan aspiration, hemodynamic shock with fever up to 40 °C was observed. Chest auscultation showed bilateral basal coarse inspiration crackles and decreased breath sounds. Dyspnea was major (level 4 of the NYHA classification), and respiratory status required assisted ventilation with continuous positive airway pressure for 24 hours. Laboratory findings included a peripheral leukocyte count of 3,000 cells/mm 3 and a fall in arterial oxygen tension (50 mmHg). Chest radiographs revealed bilateral, diffuse, and progressive patchy infiltrates. Antibiotics (penicillin), corticosteroids, and oxygen therapy were prescribed. Nevertheless, after 2 weeks, bacterial colonization (Staphylococcus epidermidis in bronchial aspiration and blood), pleural suppurative effusion, and bilateral pneumothorax appeared. Improvement was obtained with adjusted therapy (antibiotics and chest tube). The patient left the ICU with pleural sequelae and a persistent alveolar opacity in the right lower lobe. Unfortunately, he was lost for further follow-up. Discussion Since the first description in 1971 (15), isolated case reports and short series of kerdan aspiration accidents in fire-eaters have been reported, representing a total of 17 published cases. In all cases described in the literature, the follow-up was quite occasional with poor information concerning the radiologic and functional sequelae. In the present study we report 17 cases of kerdan aspiration with favorable evolution, except in 1 case, despite a severe initial presentation. In Table 4, these 17 patients are compared with the 17 patients described in the literature. In all, 17 patients had a follow-up (13 from the present series, 4 from the literature) (2,6,8,10,12,15,19) (Table 5).TABLE 4: Characteristics of 34 patients with fire-eater’s lung, present and previous reportsTABLE 5: Follow-up at 3 months for 13 patients (present study) and 4 patients (literature) (n = 17) †In the present series, kerdan aspiration was observed mainly during professional or semiprofessional demonstrations, rather than during an amateur exercise. Fire-eaters used a volatile petroleum derivative called kerdan. Kerdan is produced from the distillation process of petroleum at 150°C-240°C, in the same way as kerosene and gasoline (16). For flame blowing, the performer blew out a mouthful of pyrofluid against a burning stick, which gave spectators the impression of flames emanating from the performer’s mouth (2). Performers occasionally inhaled a small quantity of kerdan, which acted as a noxious substance on the lung, inducing a pneumonitis essentially related to the kerdan and not to the thermal injury. The kerdan remained in the mouth at ambient temperature. After flame blowing, the fire-eater had a profound inspiration and aspirated the hydrocarbon remaining in the mouth. No case of burning kerdan aspiration was described. Symptomatic episodes occurred only once in most patients, but twice in 2 patients. Preexisting aspiration accidents did not influence or alter the clinical evolution. Usually symptoms of lung involvement occurred in the first 12 hours after aspiration, progressed during the first 24 hours, and persisted during the following week. Intense chest pain, dyspnea, and cough were quite usual and severe. Hemoptysis has been reported occasionally (26.6%) but was not considered a symptom of increased severity, as no patient required specific therapy. In the course of a few hours, a fever appeared (>38.5 °C), which persisted for several days and did not respond to corticosteroids in all cases. Chest auscultation revealed crackles and decreased breath sounds. In the present study, no extrarespiratory symptoms were detected. Extrathoracic clinical symptoms have been described occasionally in the literature, represented by vomiting, stomach pain, dysphagia (6), or vertigo and fainting (8). In the case of 1 patient who had worked as a fire-eater during his youth (19), pyrofluid inhalation was not symptomatic, and the diagnosis of fire-eater’s lung was proposed several years after inhalation when pleural effusions were detected. Only 1 of our patients developed an acute lung injury. None of the patients who recurrently inhaled kerdan developed clinical complications. Radiologic abnormalities developed with a mean delay of 12 hours. Chest X-ray showed mottled or patchy, local or diffuse opacities. They were often localized in lower lungs and predominated on the right side. After 24 hours, opacities became confluent and dense as pneumonitis. Later on, the major radiologic findings were represented in 5 cases by cavitary lesions (pneumatoceles). CT scans obtained in 60% of the 17 followed cases have been useful and efficient for diagnosing cavitary lesions and for the follow-up of initial lesions. Indeed, all pneumatoceles disappeared in the course of a few weeks or months. Histologically, pneumatocele formations are thought to result from coalescing areas of bronchiolar necrosis (2) or from bronchiolar inflammation with subsequent partial obstruction of the bronchial lumen (5). In the literature and the present study, pneumatoceles (11 cases), pleural effusion (9 cases), and atelectasis (4 cases), have been found as transitory radiologic complications. Pleural effusion was limited, except for the patient who developed suppurative pleural effusion and bilateral pneumothorax consecutive to an acute lung injury in our series and for a recently reported patient (14) who developed a bronchopleural fistula with spontaneous pneumothorax as a consequence of kerosene inhalation. The presence of a durable fever might raise the suspicion of an infectious complication. However, repeated bacteriologic analysis of sputum remained negative, except in 1 case with severe sepsis. No bacterial pneumonitis was recognized, except in 1 case in which Staphylococcus aureus was detected in bronchial aspirations and the patient received antibiotics and corticosteroids. This is the only case of fire-eater’s lung involving kerdan in which the patient presented with an acute respiratory distress syndrome, from our series and the literature. Indeed, in accidental hydrocarbon ingestion cases, acute lung mortality was less than 1%(1). Fiberoptic bronchoscopy was performed in 5 cases in the literature; 3 of them had suppurative secretions without any pathogenic bacteria. Bronchoalveolar lavage in 3/5 cases showed a slightly increased total cell count and neutrophilic alveolitis (26%) in 1 case (8). Elevated macrophage values were present in 2 cases with cytoplasmic vacuolation (8,10). In the present study, only 1 of the 3 patients with BAL had an increased total cell count, with a neutrophilic influx. Bacteriologic BAL analysis performed in our 3 patients was negative, despite the suppurative aspect in 2 cases. In the 3 examinations, macroscopic aspect failed to detect abnormalities linked to burns of the tracheobronchial tree. Systemic corticosteroid and broad-spectrum antibiotic therapy has been used commonly in the treatment of acute hydrocarbon pneumonitis. Inhaled corticosteroid treatment has been proposed in the case of acute lung injury caused by inhalation of toxic gases (8,10). Corticosteroid therapy in hydrocarbon pneumonitis did not appear to be effective (11,24). Experimental models of kerosene pneumonia have been described (9,23). The results provided no evidence that currently recommended corticosteroids and antibiotics were effective in treating fulminant cases of hydrocarbon aspiration (24). Among the 17 patients in the present study, no difference in the evolutive profile was observed when patients received antibiotics and corticosteroids compared with isolated antibiotics (data not shown). In 1 patient with severe sepsis, administration of both drugs did not prevent the acute lung injury and was associated with a bacterial infection. Antibiotics such as penicillin have been recommended frequently for prophylactic use in hydrocarbon pneumonitis, presumably on the assumption that superinfection due to Haemophilus influenzae, Staphylococcus aureus, Diplococcus pneumoniae, or Streptococcus pyogenes may occur, but the routine use of antibiotics is not justified (11). Some authors recommend clinical and bacteriologic surveillance and differential white blood cell counts to guide antibiotic therapy (9). Specific methods of mechanical ventilation are documented, due to altered surface tension after hydrocarbon aspiration (5). The use of continuous airway distending pressures in the treatment of severe cases may be warranted. Intermittent positive pressure breathing has been recommended for episodes of pulmonary edema even if the risk of pneumothorax is great (5). Radiologic and functional follow-up of fire-eaters who inhaled hydrocarbons is rarely reported in the literature. Three and 6 months after the accident, functional and radiologic tests were performed in only 4/17 of the previously published cases, and in 13/17 patients in our series (see Table 5). The most frequent radiographic pattern was represented by pneumatoceles which completely disappeared 6 months after the initial accident. Five patients had a moderate restrictive pattern, and 2 of them had a decreased diffusing capacity of CO. In most patients, accidental inhalation of pyrofluids occurred just once. A recurrent inhalation might have induced more fibrotic changes (2). But presently, the risk of chronic lung injury related to inadvertent inhalation of pyrofluids in fire-eaters has not been adequately evaluated. In the present work, only 2 patients experienced a recurrent accidental inhalation of kerdan, and only 1 of them had transitory restrictive trouble with an altered diffusion capacity. Fire-eaters prepared individual mixtures of pyrofluids composed of volatile substances other than kerdan, or used other forms of petroleum distillate. These other ingredients might contribute to lung damage and alter the clinical evolution. In the present work, except for the patient who developed acute respiratory distress syndrome, kerdan was not responsible for lung damage at distance of aspiration. At 6 months, none of the 17 patients followed in the literature (4 patients) and in the present study (13 patients) had persistent radiologic or lung function abnormalities. The present work excluded pneumonitis linked to hydrocarbons other than kerdan. For this reason, we did not include the cases reported in the study by Personne et al (21), as the petroleum distillate used by fire-eaters in that study was not clearly identified as kerdan. (Results of our literature review, including studies excluded from the present study because they did not identify kerdan as the pyrofluid, are summarized in Table 6.) Physical properties of hydrocarbons are different, in terms of surface tension and viscosity, with variable levels of toxicity. Respiratory injuries from hydrocarbons are due to mixtures of aromatic and C5-C6 aliphatic hydrocarbons. Those components (kerosene, turpentine, gasoline, dry cleaning fluids, furniture polish, and lighter fluid) are responsible for accidental ingestion of hydrocarbons, especially in In substance of is more including xylene, toluene, gasoline, and petroleum type of pneumonia is a related to forced inhalation of essentially in patients with or in who inhaled review of fire-eater’s and tension are more toxic than more hydrocarbons, as they the of the and and of hydrocarbons from the airways to the Moreover, they the of in to pulmonary edema (16). The most pneumonitis induced by hydrocarbons is the of kerosene aspiration, frequently reported in and in No on lung of pneumonitis induced by kerdan have been described. analysis of hydrocarbon pneumonitis has revealed an acute within 24 hours, and a chronic within 1 or 2 weeks after the initial of symptoms (8). The acute is characterized by and of with The chronic or is characterized by and by or pneumatocele The of inflammation in the course of a few kerosene was in and with 2 doses The authors that volatile petroleum had induced lung injury and was responsible for the of at both Nevertheless, they that only high doses had induced functional respiratory impairment (9). with kerosene and gasoline, kerdan a volatile petroleum derivative that is in bronchial airways and more than and damage are more severe as petroleum is at the of alveolar of surface to with were not seen in kerdan aspiration in the present We that inhalation of kerdan, a has a favorable evolution, although may severe chest and radiologic and lung function abnormalities. Only 1 case was by severe and acute lung injury mechanical Treatment with antibiotics and corticosteroids not seem to the evolution Accidental aspiration of the petroleum derivative kerdan a common of “fire-eating” and an acute pulmonary In the present study we report clinical, radiologic, and respiratory function features in 17 patients during fire-eating demonstrations, accidentally inhaled variable of kerdan. Clinical symptoms included acute with pleuritic and often associated with in the first hours following kerdan aspiration. chest X-rays showed basal lung with consecutive in 5 cases. Among the 17 patients, 6 were treated with antibiotics and 6 with antibiotics and systemic corticosteroids. was no difference the in terms of clinical and functional evolution, except for 1 patient who developed an acute lung injury mechanical of 17 patients were at 3 and 6 radiologic and pulmonary function abnormalities completely at 3 months in of the 13 patients. At the sixth month, chest X-rays and pulmonary function tests were normal in all 13 patients. Kerdan despite the severe initial clinical symptoms and radiologic rarely and all 13 patients who were followed had a favorable evolution at 6 months.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.