Pneumopericardium

Article written and reviewed by Dr Luca Moretti

A pneumopericardium is the presence of gas in the pericardial cavity. It may follow trauma, procedures, mechanical ventilation or pathological communications with air-containing structures; more rarely, it accompanies a gas-producing infection. It must be distinguished from pneumomediastinum, in which air is distributed through mediastinal tissues without necessarily being contained within the sac surrounding the heart.

Clinical significance depends mainly on the pressure generated. A small stable collection may be reabsorbed, whereas continuous entry or a one-way valve mechanism may produce tension pneumopericardium, with impaired filling and obstructive shock. Gas may coexist with fluid, blood or pus, producing situations in which compression, hemorrhage and infection require coordinated interventions.

Assessment must establish simultaneously whether there is a circulatory threat and where the air is coming from. Decompression relieves cardiac tamponade, but a persistent digestive or respiratory fistula requires source control. The rarity of the condition makes it essential to distinguish etiologic settings and interpret case series cautiously, as they often concern severe trauma or individual procedural complications.

Traumatic and iatrogenic causes and pathological communications

In thoracic trauma, air may reach the pericardium through direct injury, a pleuropericardial communication or propagation along bronchovascular structures. Penetrating wounds may be associated with cardiac injury and blood within the cavity; therefore, the presence of gas does not reduce the need to exclude hemorrhage. In blunt trauma, transmitted energy may rupture the pericardium and airways or cause an alveolar leak. Air distribution must be interpreted together with the traumatic mechanism and other injuries.

Alveolar rupture may allow air to dissect connective tissue along bronchovascular bundles, reaching the mediastinum and, in particular circumstances, the pericardial sac. This pathway is related to the so-called Macklin effect, but not every pneumomediastinum results in pneumopericardium. A pleural communication may instead allow passage from a pneumothorax; behavior depends on the size of the defect and pressure gradients. Pleural drainage does not always guarantee relief of pericardial pressure.

Positive-pressure ventilation may contribute to air leakage or transform an initially non-tension collection into a compressive condition. The risk is particularly relevant in neonates with respiratory disease and in patients with pulmonary or tracheobronchial injury. It is not, however, an inevitable complication of ventilation: the anatomy of the communication, applied pressures and tissue compliance determine the result. In an adult trauma patient, intubation may reveal compromise that was previously partly compensated.

Iatrogenic causes include cardiothoracic surgery, airway maneuvers, endoscopic procedures and entry of air through drainage systems. After pericardiocentesis, non-airtight connections or an improperly managed circuit may introduce gas into the cavity. A small post-procedural amount should be compared with the clinical course and previous imaging; it should not automatically be attributed to a harmless residual if it increases, fever appears or the patient develops circulatory compromise.

A digestive fistula may result from esophageal perforation, malignancy, ulcer or complications of surgery and ablation. An esophagopericardial communication introduces not only air but contaminated material, with risk of purulent pericarditis and sepsis. Chest pain, dysphagia, fever or deterioration after a relevant procedure should rapidly direct the investigation. Airway fistulas pose similar questions of localization and closure, although the nature of contamination and repair strategy differ.

Gas-producing infections are rare and should be considered especially when air and purulent collection occur without a clear mechanical explanation. The absence of a visible tract does not prove a microbial origin, however: a small or intermittent fistula may escape the first study. Forms associated with asthma attacks, intense coughing or other pressure increases likewise require exclusion of relevant lesions according to context. A spontaneous form is defined only after a coherent etiologic assessment.

Pathophysiology and recognition of compromise

Gas exerts pressure on the cardiac chambers according to amount, rate of entry and ability to escape. In a one-way valve mechanism, air enters during one respiratory phase or pressure change and is not effectively eliminated during the next. Pericardial pressure rises until the filling gradient is reduced. Radiologic size of the collection is not a direct measure of danger because the volume-pressure relationship varies with sac distensibility.

The hemodynamic result is obstructive shock: cardiac output falls because the heart does not receive adequate venous return, even without primary myocardial injury. Tachycardia and vasoconstriction may temporarily support arterial pressure. Once reserve is exceeded, hypotension, altered mental status and oliguria develop. In extreme cases, pulseless electrical activity may occur; treatment must remove the compression as well as provide resuscitation.

Positive intrathoracic pressure further reduces venous return and may feed the air leak. Sedation and anesthetic induction also remove sympathetic compensatory mechanisms. For this reason, deterioration during intubation should not be interpreted only as a pharmacologic effect, especially when pneumopericardium was already known. Oxygenation and airway protection remain priorities when needed, but should be managed by a team prepared for decompression and circulatory support.

Symptoms include chest pain, dyspnea, anxiety and a sense of illness, all nonspecific. Some descriptions report a metallic or splashing auscultatory sound related to the combination of air, fluid and cardiac motion; its absence has no exclusion value. The triad of hypotension, jugular venous distension and muffled heart sounds may be absent. In polytrauma, hypovolemia may attenuate venous congestion while respiratory injuries initially dominate the picture.

The differential diagnosis of shock includes tension pneumothorax, hemorrhage, cardiac contusion or rupture, embolism and other causes. Pneumothorax and pneumopericardium may coexist; improvement after pleural drainage does not prove that every compressive component has resolved. If perfusion remains inadequate, the pericardium and other injuries must be reassessed. Even in non-trauma patients, sepsis from a fistula and tamponade may produce mixed shock requiring more than one therapeutic intervention.

Serial assessment is essential in initially stable cases. Heart rate, blood pressure, mental status, urine output, oxygenation and lactate trends should be interpreted together, recognizing that laboratory abnormalities may appear late. The distinction between simple and tension forms is clinical and hemodynamic, supported by imaging; it does not depend on the isolated presence of a gas halo. Rapid deterioration, a new need for ventilatory support or evidence of continuous entry requires immediate reassessment of surveillance intensity and need for intervention.

Imaging and differential diagnosis of gas collections

Chest radiography may show a radiolucent band surrounding the cardiac silhouette, bounded by the pericardium. Air-fluid levels may be seen when liquid is also present. Distribution may change with position, but an unstable patient does not need to be moved to demonstrate this phenomenon. The finding must be distinguished from mediastinal, pleural or subdiaphragmatic air and interpreted with regard to projection, rotation and technical conditions of bedside imaging.

In pneumomediastinum, air tends to track along mediastinal planes and structures and may extend into the neck; intrapericardial gas is instead confined by pericardial reflections. This distinction is more reliable on CT than radiography, especially when both conditions coexist. The continuous diaphragm sign may occur in more than one setting and should not be considered pathognomonic. Anatomic distribution, not a single isolated sign, allows correct localization.

Echocardiography may be impaired by the air itself, which reflects ultrasound and interrupts visualization. Cyclic appearance and disappearance of the cardiac image, described as the air-gap sign, may suggest interposed gas but is not specific only to pneumopericardium. Failure to visualize the heart does not equal absence of tamponade. Alternative windows should be sought and any fluid collections assessed; a technically inadequate examination must be reported as such, avoiding unsupported reassuring conclusions.

CT precisely defines the site and extent of air, its relationship to the pericardium and the presence of fluid, blood, pneumothorax or mediastinal lesions. In a stable patient it permits targeted investigation for airway disruptions and digestive fistulas. Compression or deformation of the chambers supports suspicion of tension physiology but does not replace hemodynamic assessment. Transfer to CT must not delay decompression in a deteriorating patient.

When esophageal perforation is suspected, contrast-enhanced CT and, when indicated, a CT esophagography protocol may document leakage and contamination. The route of contrast administration and additional studies depend on stability and local expertise. Endoscopy is not an automatic step: insufflation and manipulation may worsen certain communications. After atrial ablation, neurologic signs, fever and suspicion of an atrioesophageal fistula require a dedicated urgent pathway, avoiding maneuvers that may promote gas embolization.

Complementary investigations include blood gas analysis, complete blood count, organ function tests and cultures when sepsis is suspected. Bronchoscopy is selected when the presentation suggests tracheobronchial injury and should be coordinated with ventilatory management. Any drained fluid is sent for microbiology and other relevant tests; gas alone does not provide a complete etiologic diagnosis. Cardiac MRI generally has little role in the emergency setting, whereas review of previous procedures and devices may rapidly identify a correctable cause.

Management of the stable patient and prevention of deterioration

A non-tension pneumopericardium may be managed conservatively when the patient is stable, the cause is understood and no lesion requires repair. This choice implies adequate surveillance, not simple discharge based on the initial absence of hypotension. Trauma context, need for ventilation, increasing gas and rapid access to decompression influence the appropriate care setting. Available series do not define a universally applicable duration of observation.

Source correction may consist of restoring an airtight drainage system, treating an air leak or draining an associated pneumothorax. The latter should be reassessed for position, patency and actual effectiveness, especially if emphysema or compromise persists. A pleuropericardial communication may facilitate venting but does not guarantee it. Stable residual air and a progressive collection require different interpretations even if the initial image appears similar.

Oxygen therapy is titrated according to oxygenation and the respiratory picture. The rationale of nitrogen washout does not automatically justify prolonged high concentrations in a normoxemic patient in the absence of solid condition-specific evidence of clinical benefit. When ventilation is required, pressures compatible with adequate oxygenation are sought while limiting further barotrauma; settings depend on the pulmonary disease. No ventilator adjustment replaces treatment of pericardial compression if it develops.

Antibiotics are not indicated by the mere finding of sterile post-procedural air. They become necessary when digestive perforation, contamination, infection or specific traumatic and surgical indications are present. Empiric therapy should cover plausible organisms and is adapted to cultures and source control. Temporary improvement in fever does not prove fistula closure; absence of initial leukocytosis does not exclude recent contamination.

In penetrating trauma, strategy cannot be inferred automatically from conservative experience with blunt trauma. The trajectory and suspicion of cardiac injury may require exploration or a pericardial window even with preserved blood pressure. Likewise, an apparently modest collection after an esophageal procedure should be interpreted in relation to the possibility of an occult lesion. Indication depends on the underlying disease, not only on the amount of gas one might prefer to observe until it is reabsorbed.

Surveillance includes close clinical assessment and selected imaging to document stability or regression. CT need not be repeated at fixed intervals when the clinical picture and less burdensome tests answer the question; new instability instead requires immediate reassessment. Before transfer to a lower level of care, risk of recurrent leak, oxygen needs and drainage-system function should be clarified. Simple pain reduction is not a sufficient criterion for resolution.

Urgent decompression and definitive treatment of the cause

Tension pneumopericardium requires urgent decompression. The aim is to restore cardiac filling while treating hypoxemia, hemorrhage and other causes of shock. Fluids and vasopressors may temporarily support circulation in selected cases but do not remove the obstruction and must not delay its relief. The team should consider concomitant tension pneumothorax: treating only one cavity may leave a second compressive component unresolved.

Percutaneous aspiration may be a life-saving maneuver or a bridge to definitive treatment, depending on available expertise and anatomy. Gas may interfere with ultrasound guidance, and a mixed or loculated collection may make evacuation incomplete. A blind approach should not be described as a standard procedure applicable to everyone. Position confirmation and avoidance of cardiac, coronary or pleural injury are essential; any catheter must be connected to a system that prevents new air entry.

Surgical decompression allows evacuation, drainage and direct assessment when traumatic lesions, bloody or purulent contents, failure of percutaneous access or a communication requiring repair are present. Window, thoracoscopic access or thoracotomy are selected according to stability and associated injuries. In rapidly deteriorating patients, immediate availability and the ability to control the source take precedence over pursuit of the least invasive approach. Definitive procedures are not interchangeable independently of etiology.

An esophagopericardial fistula requires cessation of oral intake, treatment of contamination and a plan for closure or repair. Pericardial drainage, antibiotics and nutritional support must be integrated with appropriate endoscopic or surgical options. Stents, endoluminal negative-pressure therapy or suturing are not universal solutions: size, location, tissue viability, elapsed time and sepsis influence the choice. Simply evacuating air leaves an ongoing risk if passage of digestive contents persists.

In tracheobronchial injuries, airway control and management of the leak are coordinated with anesthesia and thoracic surgery; repair depends on extent and stability of the lesion. Ventilation should avoid, as far as possible, feeding the communication during preparation for intervention. In infectious cases without a demonstrated fistula, appropriate sampling and drainage accompany antimicrobial therapy. A loculated purulent collection may require wider access than is sufficient for aspirating free gas.

After decompression, hemodynamic recovery must be confirmed through blood pressure, perfusion and available instrumental assessment. Failure to improve requires investigation for residual compression, hemorrhage, myocardial dysfunction or sepsis. Rapid reappearance of gas indicates an uncontrolled source or a circuit problem. Drainage remains effective only if patent and correctly managed; removal is decided after documenting stability and reasonable cessation of the leak, not solely on the initially aspirated amount.

Complications, prognosis and subsequent follow-up

Prognosis varies widely between limited post-procedural air, major trauma and septic fistula. The high mortality reported in some historical series includes ventilated neonates and patients with very severe injuries and cannot be applied to every adult with a small pneumopericardium. Immediate risk depends on development of tension physiology; overall risk includes associated injury, contamination and timeliness of control. Radiologic reabsorption is a useful indicator but does not alone summarize patient recovery.

The most urgent complication is gas tamponade, which may develop after a stable phase. Evolution during ventilation or transfer makes preventive planning necessary in patients at risk. Recurrence after decompression requires verification of the source and drainage patency, without assuming that it is only residual air. Associated cardiac lesions may also produce hemopericardium, changing treatment and the likelihood that simple aspiration will be sufficient.

Pericardial contamination from the digestive tract may cause purulent pericarditis, mediastinitis and sepsis, requiring multiple drains and prolonged monitoring of repair. Late consequences may include adhesions and constriction. Persistent fever, inflammatory markers or organ failure after apparent decompression requires investigation for undrained collections and still-open fistulas. Normalization of blood pressure is not sufficient to consider a septic condition resolved.

Procedural complications include bleeding, injury to adjacent structures, catheter infection and displacement. In patients with mixed collections, major fluid evacuation also requires attention to rare pericardial decompression syndrome. Monitoring is adapted to drained volume, ventricular function and previous severity, without attributing every deterioration to decompression before excluding more immediate causes. Documentation of the access route and treated lesion facilitates subsequent follow-up.

Radiologic follow-up should confirm regression of gas and absence of new collections, choosing the least burdensome study that answers the question. After a fistula, safe resumption of feeding depends on control of the communication and specialist assessment, not merely disappearance of the pericardial gas halo. After trauma, pulmonary, pleural and cardiac injuries are also followed. Duration varies with these conditions because no specific schedule has been validated for all forms of pneumopericardium.

Prevention mainly concerns airtight management of circuits, early recognition of leaks and attention to injuries after high-risk procedures. Once an isolated mechanical form has resolved, prolonged anti-inflammatory drugs or antibiotics are not indicated without a separate reason. New dyspnea, increasing pain, fever, syncope or worsening exercise tolerance require reassessment. Closing the care pathway presupposes stability of the treated cause and instructions consistent with residual risk, not merely improved chest imaging.

References
1. Schulz-Menger J et al. 2025 ESC Guidelines for the management of myocarditis and pericarditis. European Heart Journal. 2025;46(40):3952-4041.
2. Klein AL et al. Pericardial Diseases: International Position Statement on New Concepts and Advances in Multimodality Cardiac Imaging. JACC: Cardiovascular Imaging. 2024;17(8):937-988.
3. Adler Y et al. 2015 ESC Guidelines for the diagnosis and management of pericardial diseases: The Task Force for the Diagnosis and Management of Pericardial Diseases of the European Society of Cardiology (ESC). Endorsed by: The European Association for Cardio-Thoracic Surgery (EACTS). European Heart Journal. 2015;36(42):2921-2964.
4. Cummings RG et al. Pneumopericardium resulting in cardiac tamponade. Annals of Thoracic Surgery. 1984;37(6):511-518.
5. Sciarretta JD et al. Pneumopericardium following severe thoracic trauma. Injury. 2024;55(5):111303.
6. Rak M et al. Pneumopericardium, a Heart in a Trap. Journal of Clinical Medicine. 2024;13(24):7636.
7. Reid CL et al. Echocardiographic detection of pneumomediastinum and pneumopericardium: the air gap sign. Journal of the American College of Cardiology. 1983;1(3):916-921.
8. Coccolini F et al. Thoracic trauma WSES-AAST guidelines. World Journal of Emergency Surgery. 2025;20:78.
9. Spodick DH. Acute cardiac tamponade. New England Journal of Medicine. 2003;349(7):684-690.
10. Madhivathanan PR, Corredor C, Smith A. Perioperative implications of pericardial effusions and cardiac tamponade. BJA Education. 2020;20(7):226-234.
11. Ristić AD et al. Triage strategy for urgent management of cardiac tamponade: a position statement of the European Society of Cardiology Working Group on Myocardial and Pericardial Diseases. European Heart Journal. 2014;35(34):2279-2284.
12. Chirica M et al. Esophageal emergencies: WSES guidelines. World Journal of Emergency Surgery. 2019;14:26.
13. Worrell S et al. The Society of Thoracic Surgeons (2026) Expert Consensus Document on Management of Esophageal Perforation. Annals of Thoracic Surgery. 2026. Advance online publication, September 17, 2026.
14. Tzeis S et al. 2024 European Heart Rhythm Association/Heart Rhythm Society/Asia Pacific Heart Rhythm Society/Latin American Heart Rhythm Society expert consensus statement on catheter and surgical ablation of atrial fibrillation. Europace. 2024;26(4):euae043.
15. Sagristà-Sauleda J et al. Purulent pericarditis: review of a 20-year experience in a general hospital. Journal of the American College of Cardiology. 1993;22(6):1661-1665.
16. Sarode K et al. Pericardial Decompression Syndrome: A Comprehensive Review of a Controversial Entity. International Journal of Angiology. 2024;33(3):139-147.