Therapeutics
Dispositionof Spontaneous Pneumomediastinum BRENT A. SMITH, MD, DOUGLAS B. FERGUSON, MD In the evaluation of spontaneous pneumomediastinum, it is important to exclude pathological causes of pneumomediastinum, including Soerhaave’s syndrome, which carries a high mortality rate. The literature varies greatly as to the care of patients with presumed spontaneous pneumomediastlnum. The authors present an Illustrative case of spontaneous pneumomedlastlnum treated with intravenous antlblotics and intenslve care unit (ICU) admission despite a normal esophagram. Spontaneous pneumomediastinum is a benign entity that may not require observation In an ICU, and patients may do as well with close outpatient follow-up. For those patients in which the etiology of the pneumomediastlnum Is unclear, a contrast esophagram is recommended to assess for esophageal dlsruption. The signs, symptoms, pathophplology, and dlsposition of spontaneous pneumomediastinum are discussed. (Am J Emerg Med 1&&1;9:256-259. Copyright 0 1991 by W.B. Saunders Com&asy)
The discovery of subcutaneous emphysema and pneumomediastinum in a patient demands prompt diagnosis. The principle dilemma is the differentiation between spontaneous pneumomediastinum, which has a benign course, and conditions that result in pathological disruption of the tracheobronchial or alimentary tracts. On initial presentation, it is also important to exclude associated conditions such as asthma, pulmonary infection, pneumothorax, and illicit drug use. Whereas the cause of pneumomediastinum and subcutaneous emphysema may be self-evident in many cases, the diagnosis of Boerhaave’s syndrome may sometimes be difficult to make. The literature varies greatly in the initial and subsequent
management
neous pneumomediastinum.
of patients
with presumed
sponta-
DISCUSSION
l-3
CASE REPORT A 27-year-old active-duty soldier presented to the emergency department with a 3-day history of pleuritic chest pain, neck pain,
From the Department of Emergency Medicine, Emergency Medicine Residency Program, Darnall Army Community Hospital, Fort Hood, TX. Manuscript received April 9, 1990; revision accepted September 19, 1990. The opinions and assertions contained herein are those of the authors and should not be construed as official or as representing the opinions of the Department of the Army or the Department of Defense. Address reprint requests to Dr Smith, 6535 Pendragon, San Antonio, TX 78250. Key Words: Pneumomediastinum, spontaneous, Boerhaave’s syndrome. Copyright 0 1991 by W.B. Saunders Company 0735-6757/9110903-0014$5.0010
256
dysphagia, and voice hoarseness. He had had an intermittent cough productive of blood-tinged mucus. There was no history of fever, vomiting, trauma, asthma, instrumentation, or illicit drug use. Vital signs were blood pressure, 126184 mm Hg; pulse rate, 84 beats/min; respiration rate, 20 breaths/min; temperature, 98.5”F. The neck was remarkably tender on both sides and range of motion was reduced secondary to pain. There was no crepitus or masses. The oropharynx was hyperemic without exudates or other lesions. An intermittent friction rub sound coinciding with the heart beat was heard in the anterior left chest. No rhonchi or wheezes were heard. A chest roentgenogram showed mediastinal air without pulmonary infiltrates. effusion, and pneumothorax. Air was seen in the soft tissues of the neck on cervical plain films. A Gastrogratin (Squibb, Princeton, NJ) esophagram showed no extravasation of contrast material or esophageal displacement. Although mandatory screening for recreational drug use is done regularly for all military personnel, a current drug screen was not obtained. Initial and subsequent complete blood counts were within normal limits. The patient was admitted to the intensive care unit (KU) and placed on intravenous gentamicin and clindamycin. Two days after admission, a computed tomography (CT) scan of the neck demonstrated generalized emphysema of the muscular, prevertebral. and subcutaneous compartments of the neck. No masses or abscesses were present. His cough subsided in several days and there was no further blood-tinged mucus production. A follow-up chest roentgenogram on the fifth hospital day showed near resolution of the pneumomediastinum. The patient’s chest and neck pain gradually diminished, and he was discharged home on the seventh hospital day. No specific etiology was found for the transient blood-tinged cough.
Spontaneous pneumomediastinum is an uncommon problem representing 1 in 7,000 to 12,000 hospital admissions.’ Although sometimes reported in young healthy adults without a definite precipitating cause, it is usually seen in association with the acute production of high intrathoracic pressures. Thus, asthma exacerbations, coughing, child-birth, the valsalva maneuver, vomiting, seizures and inhalational drug abuse are common associated factors. l-6 The pathophysiology of this disorder is related to excessive intraalveolar pressures leading to rupture of perivascular alveoli.4 Air escapes into the perivascular connective tissue with subsequent dissection into the mediastinum (Figure I). Located within the mediastinal visceral space, air may then dissect superiorly into the visceral, retropharyngeal, and subcutaneous spaces of the neck. The prevertebral space is less accessible for air dissection. From the neck, the subcutaneous compartment is continuous over the body so air can diffuse very widely. Mediastinal air can also pass
SMITH AND FERGUSON n SPONTANEOUS PNEUMOMEDIASTINUM
1. Excessive intraalveolar pressure leads to rupture of perivascular alveoli with subsequent dissection of air into the me-
FIGURE
diastinum .
inferiorly into the retroperitoneum and other extraperitoneal compartments. A perforated bowel may be simulated radiographically by the subdiaphramatic dissection of air.’ If the mediastinal pressure rises abruptly or decompression is not sufficient, the mediastinal parietal pleura may rupture and cause pneumothorax.’ The usual presenting symptoms of spontaneous pneumo-
257
mediastinum are acute pleuritic chest pain, dyspnea, and neck pain. 1,9Unusual presentations include isolated dysphagia, dysphonia, or abdominal ~ain.~~” Physical findings include subcutaneous emphysema, initially present in the suprasternal notch, and mediastinal crepitation (Hamman’s sign). Patients can be relatively asymptomatic and physical findings may be absent.” The roentgenogram is essential for making the diagnosis of spontaneous pneumomediastinum. Thin, lucent, vertically oriented streaks of air within the mediastinum are seen on anterior-posterior (A-P) chest films. In addition, the aortic knob and left cardiac border are highlighted by mediastinal air. The air is more easily detected in the lateral projection, particularly in the retrostemal location (Figure 2). An A-P chest film may miss pneumomediastinum 50% of the time.” A minimal pneumothorax or pneumopericardium can be difficult to differentiate from pneumomediastinum, but a lateral decubitus film will resolve the issue. In both pneumothorax and pneumopericardium, the air is free to rise to the highest position, but the air remains fixed in pneumomediastinum.’ Like pneumothorax, pneumomediastinum is better seen on expiratory films. Pleural effusions and pulmonary infiltrates are not generally seen in spontaneous pneumomediastinum unless another process coexists. Other laboratory findings in spontaneous pneumomediastinum are nonspecific and frequently are normal. Fever,
FIGURE 2. Thin, lucent, vertically oriented streaks of air are seen within the mediastinum on A-P and lateral chest roentgenograms. On the lateral projection, a retrostemal air collection is commonly seen. The left cardiac border is highlighted by mediastinal air on the A-P view.
258
AMERICAN
JOURNAL OF EMERGENCY
leukocytosis and electrocardiographic (ECG) S-T changes have been reported but are not common or diagnostic.” The diagnosis of spontaneous pneumomediastinum implies the absence of other causes of pneumomediastinum and subcutaneous emphysema. Other causes include soft tissue infection of the head and neck by gas-producing organisms as well as traumatic or surgical disruption of the oropharyngeal, esophageal, and respiratory mucosa.5’8 Whereas the majority of these are self-evident, such is not always the case with Boerhaave’s syndrome.’ Boerhaave’s syndrome classically presents with a history of overindulgence, vomiting, chest pain, subcutaneous air, and rapid toxicity of the patient. ‘l,‘* The chest film usually shows pneumomediastinum, basilar pulmonary infiltrates and left-sided pleural effusion. Pneumothorax is common.1’,‘3 More “atypical” presentations reported include orbital emphysema,14 isolated abdominal pain, ’ ’ and the absence of vomiting.” The lack of classic symptoms in esophageal rupture, coupled with the delays in diagnosis, contribute to an overall mortality rate of 64% to 70% even with surgical intervention.15 The natural history of spontaneous pneumomediastinum is generally benign. The chest pain and mediastinal air usually resolve over 2 to 7 days without treatment.‘.’ Although the usual course is self-limited, there is some concern about the occasional patient who may manifest serious sequelae. The pneumomediastinum” by entity termed “malignant Macklin16 in 1944 is referenced in much of the literature on this subject. Malignant pneumomediastinum supposedly manifests when unvented mediastinal or pulmonary adventitial air exerts such great pressure that it leads to circulatory or ventilatory failure. It is noteworthy that all of these patients with fatal outcomes had an associated serious illness such as Boerhaave’s syndrome, pneumothorax, tuberculosis, influenza, or an airway foreign body. With idiopathic spontaneous pneumomediastinum, patients uniformly recover. To our knowledge, there are no reports in the recent literature of fatal outcomes in patients with spontaneous pneumomediastinum in the absence of underlying illness. However, associated pneumothorax should be sought as it may occur in 10% to 18% of patients.lX5 The disposition of the patient who presents to the emergency department with pneumomediastinum is predicated on the underlying illness. Boerhaave’s syndrome should always be considered in these patients because a missed diagnosis carries such a high mortality. A contrast esophagram is a useful, easily performed test that can be done emergently to assess for esophageal disruption.’ Whereas a diagnosis of esophageal perforation can be made by a contrast esophagram in the vast majority of patients,” such testing may not be necessary in patients who present with a clear-cut cause for their pneumomediastinum. Barium is a more sensitive agent for delineating mucosal tears or small perforations but most authors believe that a water-soluble agent is better tolerated ifextraluminal leakage of contrast occurs.” For those patients with a normal esophagram and no underlying illness, the prognosis is uniformly good. The literature varies greatly in the care of patients with presumed spontaneous pneumomediastinum, ranging from outpatient follow-up after chest radiographic examination, to KU admission despite a normal contrast esophagram. In our
MEDICINE
n Volume 9, Number 3 W May 1991
an emergent esophagram was effective in excluding esophageal disruption, and he had no associated illness affecting his prognosis. In retrospect, our case may have been safely managed by simple observation-perhaps even as an outpatient. The discrepancy in treatment of these patients may result from misinterpretation of the older literature that suggests isolated spontaneous pneumomediastinum per se may carry a poor prognosis. This has not been supported in subsequent literature. If this is the case, do these patients require hospital admission, or can they be treated with close outpatient follow-up? Further evaluation of this issue would be helpful for optimum utilization of hospital beds. case,
Pneumomediastinum in individuals without preexisting or associated disease has been termed spontaneous pneumomediastinum. The care of these patients requires consideration of any associated illness and the exclusion of pathological causes of pneumomediastinum such as disruption of the tracheobronchial or alimentary tracts. A contrast esophagram can effectively screen patients for esophageal perforation. The literature does not support ICU admission for patients with isolated spontaneous pneumomediastinum as there is no evidence of a poor outcome in this patient population. Although it is current practice to admit such patients for observation, they may do equally well with close outpatient follow-up. The authors thank Dr Donald M. Yealy for his advice in preparation of this manuscript.
REFERENCES 1. Halperin AK, Deichmann RE: Spontaneous pneumomediastinum: A report of 10 cases and review of the literature. N Carolina Med J 1985;46(1):21-23 2. Jabourian Z, Mckenna EL, Feldman M: Spontaneous pneumomediastinum and subcutaneous emphysema. J Otolaryngol 1988;17:50-53 3. Werne C, Ulreich S: An unusual presentation of spontaneous pneumomediastinum. Ann Emerg Med 1985;14:10101013 4. Pellinen TJ, Karjalainen JE: Spontaneous pneumomediastinum. Acta Med Stand 1982;211:139-140 5. Munsell WP: Pneumomediastinum. JAMA 1967;202:129133 6. Seaman ME: Barotrauma related to inhalational drug abuse. J Emera Med 1990;8:141-149 7. Schulm& A, Fataar.S, Van Der Spuy JW, et al: Air in unusual places: Some causes and ramifications of pneumomediastinum. Clin Radio1 1982;32:301-306 8. Maunder RJ, Pierson DJ, Hudson LD: Subcutaneous and mediastinal emphysema: Pathophysiology, diagnosis, and management. Arch Intern Med 1984;144:1447-1453 9. Rogers LF, Puig WA, Dooley BN, et al: Diagnostic considerations in mediastinal emphysema: A pathophysiologicroentgenologic approach to Boerhaave’s syndrome and spontaneous pneumomediastinum. AJR 1972;115:495-511 10. Shuster MJ: Pneumomediastinum as a cause of dysphagia and pseudodysphagia. Ann Emerg Med 1981;10:648-651 11. Walker WS, Cameron EWJ, Walbaum PR: Diagnosis and management of spontaneous transmural rupture of the oesophagus (Boerhaave’s syndrome). Br J Surg 1985;72:204-207 12. Jaworski A, Fischer R, Lippmann M: Boerhaave’s syndrome. Arch Intern Med 1988;148:223-224
SMITH AND FERGUSON W SPONTANEOUS PNEUMOMEDIASTINUM
13. delCastillo J, Boyar C, Hess F, et al: Atraumatic panmural rupture of the esophagus: Boerhaave’s syndrome. Ann Emerg Med 1983;12:385-390 14. Schneider SM, Goodman D: Spontaneous rupture of the esophagus presenting with unilateral proptosis. Ann Emerg Med 1984;13:374-377 15. Singh GS, Slovis CM: “Occult” Boerhaave’s syndrome. J Emerg Med 1988;6:13-16
259
16. Macklin MT, Macklin CC: Malignant interstitial emphysema of the lungs and mediastinum as an important occult complication in many respiratory diseases and other conditions: An interpretation of the clinical literature in the light of laboratory experiment. Medicine 1944;23:281-358 17. Foley MJ, Ghahremani GG, Rogers LF: Reappraisal of contrast media used to detect upper gastrointestinal perforations. Radiology 1982;144:231-237