Effusive-constrictive pericarditis

Article written and reviewed by Dr Luca Moretti

The effusive-constrictive pericarditis is a syndrome in which a pericardial effusion coexists with impaired filling due to the pericardium, particularly the visceral component. The fluid and the tissue constraint contribute differently to the hemodynamic disturbance. The classic feature is persistence of elevated filling pressures and constrictive physiology after drainage has normalized pressure in the pericardial cavity. Removal of the fluid therefore eliminates one component of the obstruction, but may not resolve the overall picture.

It is not sufficient for echocardiography to show an effusion and a thickened pericardium at the same time. Diagnosis requires functional correlation and distinction from residual effusion, right ventricular dysfunction, tricuspid regurgitation, and myocardial disease. Modern imaging techniques can identify the phenotype even without catheterization and, in some cases, before drainage. The syndrome lies between cardiac tamponade and constrictive pericarditis, but it is not simply an obligatory transitional phase from one to the other.

Frequency depends on the population and diagnostic method. In a Mayo Clinic cohort of 205 patients undergoing pericardiocentesis, Kim and colleagues identified effusive-constrictive findings in 33 cases, or 16%. This figure concerns people selected for drainage and not all patients with pericarditis or effusions. The course may be reversible when inflammation predominates, or may progress to persistent constriction. Prognosis also depends on the cause, which may be idiopathic, infectious, neoplastic, postoperative, or related to other clinical settings.

Etiology and pathogenesis of the effusive and visceral components

The etiologic process must explain both the fluid collection and the reduction in compliance. In inflammatory forms, increased microvascular permeability and impaired reabsorption produce exudate, while edema, infiltrate, and fibrin alter the behavior of the surfaces. Distribution may be uneven, with loculations and adhesions. The parietal layer does not need to appear as a calcified shell: substantial restriction may depend on the visceral lining adherent to the heart, even when fluid remains free in other portions of the cavity.

The idiopathic or presumably viral forms may occur during acute pericarditis or as it resolves. The mechanism may be predominantly inflammatory, with recovery of compliance after inflammation subsides. An idiopathic designation does not prove a specific viral infection and should not prevent reassessment of the cause if the course changes. A preceding respiratory history is a contextual clue, whereas persistent fever, marked systemic impairment, or an atypical collection requires more targeted investigation.

Tuberculosis is particularly relevant in endemic settings and in immunosuppression. Granulomatous inflammation, fibrin, and organization may simultaneously produce effusion and visceral constraint. Purulent bacterial infections may generate similar physiology through a cell-rich exudate, serosal injury, and adhesions. In these etiologies, control of the infection and its source is an essential part of treatment; immunomodulation cannot replace it. The likelihood of a specific infection must be assessed before interpreting persistence as simple resistance to anti-inflammatory therapy.

In the postoperative or postprocedural setting, hemorrhage, inflammation, and organization may all contribute to the picture. A collection after cardiac surgery may be loculated and regionally compress the chambers, while adherent pericardium limits filling. Proximity to a procedure does not, however, prove that every constrictive sign was caused by the drainage itself: the constraint may already be present and become recognizable only after evacuation. Studies describing the syndrome after pericardiocentesis primarily identify the time of diagnosis, not necessarily the time of its origin.

Neoplasms may produce a collection through infiltration, microcirculatory changes, and lymphatic obstruction, with variable pericardial involvement. Radiotherapy and oncologic treatments may add inflammatory or fibrotic injury; the myocardium and other structures may be involved at the same time. Autoimmune diseases, uremia, and other causes complete the spectrum. In each case, the truly pericardial component must be distinguished from conditions that keep right-sided pressures elevated, because simple anatomic coexistence is not enough to define the mechanism.

Visceral remodeling may remain reversible or may become organized. Fibrin forms a scaffold on which inflammatory and reparative cells interact; fibroblast activation and matrix deposition progressively reduce gliding and deformability. The amount of calcium does not measure this process. Edematous tissue may be temporarily noncompliant, whereas a thin adherent fibrous layer may exert persistent constraint. The ability to distinguish these components with imaging and observation over time underlies the choice between medical and surgical treatment.

Predisposing factors include immunosuppression, tuberculosis exposure, surgery or trauma, neoplasia, and systemic diseases. A fibrinous or loculated collection and some Doppler signs may be associated with the phenotype, but they are neither independent causes nor sufficient evidence. Individual risk cannot be inferred from a single feature of the fluid. Causal interpretation must integrate history, microbiology or cytology when indicated, anatomy, and physiology, while keeping the origin of the process distinct from its hemodynamic expression.

Pathophysiology before and after drainage

Before drainage, intrapericardial pressure may increase because of the collection and reduce the transmural gradient available for filling. Pressure measured within a chamber is not the same as the pressure that distends it: what matters is the difference from the external pressure. When fluid accumulates rapidly or adaptive capacity is exhausted, the chambers operate with a reduced margin for expansion and cardiac output may fall. The simultaneous presence of a poorly compliant visceral layer adds a second limitation that does not depend only on fluid pressure.

Tamponade may dominate the initial picture and make constriction less recognizable. Elevated venous pressures, respiratory variation, and impaired filling are shared features; some signs specific to tissue constraint may be masked. However, not all patients identified by modern echocardiographic criteria have overt clinical tamponade. The syndrome encompasses a spectrum in which the relative contribution of the effusive and constrictive components changes over time and according to loading conditions.

After effective evacuation, fluid pressure falls toward values close to zero, but the visceral lining may continue to restrict expansion. Right atrial pressure then remains higher than would be expected after resolution of fluid compression alone. Early diastolic filling becomes rapid because of the high initial gradient, followed by an abrupt halt when the volume limit is reached. The y descent may become more prominent and ventricular tracings may acquire a dip-and-plateau morphology.

Respiratory dissociation and ventricular interdependence often become more evident after fluid removal. Changes in thoracic pressure are not transmitted uniformly to the constrained chambers; left-sided filling decreases during inspiration and right-sided filling tends to increase at its expense. The septum shifts and atrioventricular flows vary reciprocally. These findings explain why a complete Doppler reassessment after the procedure may provide information different from the initial urgent examination, which is often focused on a safe access point and signs of tamponade.

Persistent right atrial pressure is not specific, however. Right ventricular dysfunction, pulmonary hypertension, significant tricuspid regurgitation, volume overload, and residual collections can keep it elevated. It must be shown that the fluid has been adequately evacuated and that coherent signs of constriction are present. Otherwise, a limitation due to another mechanism may be incorrectly attributed to the visceral pericardium. The numerical value must be interpreted together with the entire physiology before and after drainage.

Residual congestion may involve the liver, kidneys, and intestine, while cardiac output may improve only partially. This behavior does not necessarily imply an irreversible outcome: if the constraint is mainly due to inflammation and edema, it may regress over the following weeks. When fibrosis predominates, however, normalization of the fluid is not sufficient. The period after drainage is therefore a time for reassessment, during which hemodynamic adaptation, reversible activity, structural damage, and procedural complications must be distinguished.

Clinical manifestations and observation after pericardiocentesis

Presentation may be dominated by dyspnea and fatigue, with reduced exercise capacity, chest pressure, and sometimes orthopnea. Pleuritic pain, fever, and a friction rub indicate an inflammatory component when present, but may be absent in neoplastic, radiation-induced, or fibrotic phenotypes. The history reconstructs the duration of the collection, the rate of worsening, previous pericarditis, and procedures. A course lasting weeks with progressive congestion has a different meaning from a hemorrhagic emergency after a procedure, although both may produce a clinically significant effusion.

In overt tamponade, tachycardia, hypotension, pulsus paradoxus, jugular venous distension, and signs of hypoperfusion may occur. The classic triad is not sufficiently sensitive to wait until it is complete. Loculated collections after surgery may produce regional compression with less typical signs. Urgent assessment must identify the hemodynamic threat and permit drainage when indicated; complete characterization of the constrictive component must not delay a procedure needed to restore perfusion.

The etiologic history includes tuberculosis exposure, prolonged fever, immunosuppression, neoplasia, radiotherapy, rheumatologic diseases, and renal failure. Recent procedures, anticoagulation, and trauma also suggest blood in the cavity, without excluding other mechanisms. Previously used medications are reviewed because corticosteroids and anti-inflammatory drugs may modify symptoms and biomarkers. Weight loss, sweats, lymphadenopathy, and other serositis require a systemic interpretation, particularly when the picture does not follow the expected course of uncomplicated pericarditis.

Suspicion increases when, after drainage, there is persistent congestion disproportionate to the residual fluid. Blood pressure may improve and pulsus paradoxus diminish, while jugular venous distension, dyspnea, edema, or ascites persist. Venous pressure may show a more prominent y descent and a previously subtle Kussmaul sign. These changes are suggestive but require confirmation, because right ventricular dysfunction and valvular disease may also cause incomplete recovery.

The physical examination is therefore repeated with attention to perfusion, volume status, and respiration. Jugular veins, blood pressure, heart rate, urine output, and symptoms are compared with the previous picture. Persistence of a dilated vena cava or edema during the first hours may have different causes and is not sufficient by itself for diagnosis. The time elapsed, the amount actually drained, and the distribution of remaining fluid must be known. Sudden deterioration first requires exclusion of acute complications before it is attributed to the constrictive component.

A pericardial decompression syndrome, rare but important, may present with ventricular dysfunction and deterioration after removal of a collection and is not the same as persistent constriction. Rebleeding, cardiac injury, pneumothorax, arrhythmias, and pulmonary edema must also be considered. Prompt echocardiography and hemodynamic assessment help distinguish these conditions. Awareness of a possible effusive-constrictive form should not become an automatic explanation for every problem occurring after the procedure.

During more stable phases, functional capacity and nutrition are assessed. Hepatic and intestinal congestion may reduce appetite and cause muscle loss, while edema masks weight loss. Recovery after drainage may be gradual, especially if the underlying disease remains active. The frequency of visits should depend on symptoms, physiology, and ongoing treatment, not solely on the initial presence of a large effusion. Longitudinal assessment makes it possible to recognize both spontaneous resolution and constriction that persists despite control of the effusion.

Diagnostic criteria, echocardiography, and hemodynamics

The classic definition is hemodynamic and is based on comparison before and after pericardiocentesis. Normalization of pressure in the cavity should be accompanied by an appropriate reduction in filling pressures; when this does not occur and constrictive signs are present, the dual mechanism is recognized. ESC guidelines describe this criterion without requiring every patient to undergo catheterization. Multimodality imaging can in fact support a noninvasive clinical diagnosis when the data are coherent.

Classic hemodynamic definition of effusive-constrictive pericarditis, as reported in ESC recommendations.

• Reduction of intrapericardial pressure to values close to 0 mmHg through effective evacuation of the fluid.
• Persistence of elevated right atrial pressure without the expected response, namely a decrease of at least 50% or a value below 10 mmHg.

The criterion must be attributed to a residual pericardial constraint, after alternative explanations for elevated right-sided pressure have been excluded. Measuring a high value with a central catheter is not sufficient without knowing the effectiveness of drainage and the physiology. Persistently compressive collections, severe tricuspid regurgitation, pulmonary hypertension, or right ventricular dysfunction may make interpretation equivocal. Demonstration of ventricular interdependence and respiratory dissociation increases diagnostic coherence in invasively studied cases.

Before the procedure, echocardiography defines the location, size, and distribution of the fluid, any clots or loculations, chamber collapse, and respiratory consequences. Findings such as respiration-dependent septal shift, preserved medial e', expiratory diastolic flow reversal in the hepatic veins, and fibrinous features may suggest a constrictive component. Some findings are also shared with tamponade and should not be overinterpreted. In an emergency, priority remains assessment of hemodynamic compromise and selection of safe access.

The examination after drainage should be more complete and comparable. It is first verified that the collection has been removed sufficiently, then respiration-dependent septal motion, variation in atrioventricular flows, and hepatic venous flow are assessed. Persistence of these signs after compression from fluid has resolved points toward constriction. A relatively elevated medial e' and annulus reversus, when present, support the picture without being mandatory. Isolated dilation of the inferior vena cava is poorly specific and may regress more slowly than other findings.

Doppler parameters must be interpreted according to rhythm, respiration, and loading conditions. Inspiratory variation in mitral flow may be attenuated by elevated left-sided pressures or altered by atrial fibrillation. Pulmonary disease with marked intrathoracic pressure swings may produce exaggerated variations without true constraint. Tricuspid regurgitation mainly produces systolic flow reversal in the hepatic veins, which must be distinguished from the expiratory diastolic component. Positive-pressure ventilation alters respiratory relationships. Diagnosis derives from concordance of findings, not from an isolated threshold applied out of context.

Magnetic resonance imaging characterizes the pericardium and myocardium, looking for edema, late gadolinium enhancement, and interdependence during free breathing. It is useful when estimating a treatable inflammatory component or clarifying a mixed picture. LGE may persist and does not by itself demonstrate clinically active inflammation or irreversibility. CT better defines calcifications, the distribution of collections, masses, and preoperative anatomy. Normal thickness does not exclude visceral constraint, while thickening alone does not prove that residual right-sided pressure is pericardial in origin.

Simultaneous catheterization may be reserved for unresolved cases, documenting intrapericardial and intracardiac pressures when appropriate. After evacuation, a dip-and-plateau pattern and respiratory discordance of right- and left-ventricular systolic pressures may emerge. Diastolic equalization alone is less discriminating. Measurements should be obtained with correct references and under known conditions because volume status, ventilation, and rhythm may alter them. It is not justified to drain an effusion without another indication solely to perform a diagnostic test when imaging can adequately answer the clinical question.

Etiologic investigations and assessment of reversibility

Fluid obtained for a clinical indication represents an opportunity for etiologic diagnosis. Cytology, cultures, and targeted investigations are selected according to suspected neoplastic, bacterial, or tuberculous disease. Hemorrhagic appearance, high protein content, or cellularity do not identify the cause by themselves: blood may occur in neoplasia, trauma, procedures, and other conditions. The absence of malignant cells in a single sample does not exclude all infiltration, while microbiologic interpretation also depends on previous therapy and specimen quality. Tests should be planned before the specimen is handled in a manner unsuitable for the clinical question.

When tuberculosis is suspected, epidemiology, disease at other sites, immune status, and fluid or tissue findings must be integrated. Immunologic tests of exposure are not equivalent to demonstration of pericardial localization. In purulent infections, blood cultures and a search for the source complete the investigation. A negative test does not erase a strong clinical suspicion, but neither does it justify labeling every persistent collection as infectious. Treatment choice requires an explicitly recognized level of probability and documentation.

The search for neoplasia or systemic disease is guided by the history and findings. Thoracic and abdominal imaging, oncologic assessment, or rheumatologic investigations may be necessary. Previous radiotherapy makes myocardial and valvular characterization important, because the benefit of correcting the pericardial component may be partial. An effusion in a person with cancer is not necessarily malignant: infections, treatments, and unrelated conditions must remain in the differential. Likewise, an isolated autoantibody is not sufficient to diagnose an autoimmune etiology.

Reversibility is estimated from clinical activity, CRP, edema on magnetic resonance imaging, duration, and the serial course of physiology. A substantial inflammatory component makes medical treatment reasonable in stable patients, whereas persistent congestion with a fibrotic substrate and lack of recovery points in a different direction. Reduction of the fluid is not the only outcome to measure: venous pressure, exercise capacity, and Doppler signs should also be followed. CRP may be poorly informative during drug therapy or in forms with localized expression and is not a sole arbiter.

Assessment of organ damage includes renal function, liver status, albumin, electrolytes, and nutritional status. An increase in creatinine may result from congestion, low cardiac output, contrast, NSAIDs, or other causes, with different implications. Ascites and hypoalbuminemia require hemodynamic and systemic interpretation. Myocardial dysfunction, arrhythmias, and valvular disease should be defined because they may explain part of an incomplete recovery. Before surgery, it is particularly important to distinguish still-reversible injury from advanced impairment that increases risk and limits benefit.

Pericardial biopsy is considered in selected cases when the diagnosis remains uncertain and a tissue result could change treatment, or during an already indicated procedure. Visceral involvement may make sampling and dissection more complex. Histologic fibrosis alone does not establish the entire etiology, and a limited biopsy may not represent regional lesions. The aim of further investigation is to reach an operational description: probable or demonstrated cause, amount of treatable inflammation, residual constraint, and clinical consequences to be corrected.

Treatment of the effusion and residual constriction

In the presence of clinical tamponade, urgent drainage takes priority. A suspected constrictive component must not delay removal of fluid pressure that threatens perfusion. The procedure is imaging-guided and adapted to the location, contents, and condition of the patient. Loculated collections, clots, pus, or some traumatic complications may require a surgical approach. The amount needed to restore stability and subsequent drain management are decided according to the clinical response, avoiding treatment of rapid complete evacuation as a goal independent of safety.

Controlled drainage continues with monitoring of blood pressure, rhythm, urine output, and ventricular function, while checking for residual fluid and possible complications. A large, slowly accumulated collection may require particular attention to loading changes after decompression. Temporary circulatory support does not replace correction of compression. Persistent jugular venous distension after a technically effective procedure requires reassessment, but does not mean that further aspiration is needed from a cavity that no longer contains a significant collection.

Treatment of the cause is started or adjusted as soon as the clinical picture permits. Antibiotics and source control are essential in purulent pericarditis; tuberculosis requires an antimycobacterial regimen; neoplasia and autoimmune disease require a coordinated specialist strategy. A pericardial window may control some recurrent effusions, but does not necessarily eliminate a constrictive visceral sheath. It is therefore necessary to clarify whether the proposed intervention addresses the fluid problem, the tissue constraint, or both.

In stable patients with residual inflammation, aspirin or NSAIDs and colchicine may be used according to the indications for the associated pericarditis, with monitoring of renal function, gastrointestinal risk, and interactions. Usual adult regimens include ibuprofen 600-800 mg every eight hours or aspirin 750-1,000 mg every eight hours during the active phase, followed by tapering according to response. Colchicine is generally dosed at 0.5 mg once daily for body weight below 70 kg or twice daily from 70 kg upward, with the necessary adjustments. These regimens are not indiscriminately applicable to infected collections, advanced renal failure, or fibrosis without activity.

Corticosteroids may be appropriate for specific indications or when the conventional regimen cannot be used, after infectious risk has been addressed. Dose and tapering should reflect activity and comorbidities, avoiding rapid courses that leave disease unresolved. Interleukin-1 blockade may be discussed in a selected idiopathic inflammatory phenotype, particularly when incessant disease or difficult recurrences coexist. Randomized evidence in recurrent forms, however, does not constitute specific proof of efficacy in every effusive-constrictive pericarditis and does not support its use as a universal antifibrotic treatment.

A period of therapeutic observation is reasonable when the patient is stable, follow-up is feasible, and there are signs of reversibility. Follow-up should assess physiology, not only reduction in CRP. A progressive decrease in venous pressure and Doppler signs, with better exercise capacity and lower diuretic requirement, supports continued medical treatment. Weeks or months may be needed for recovery, but this does not justify rigid waiting in the presence of deterioration. Refractory congestion and progressive organ damage require prompt reconsideration of the strategy.

Diuretics may reduce residual congestion, but should be used cautiously because excessive reduction in filling may lower cardiac output and impair renal function. Weight, blood pressure, symptoms, electrolytes, and creatinine are monitored. Improvement with diuresis does not prove that the constraint has disappeared. In malnourished or hypoalbuminemic patients, fluid management may be particularly complex and should be accompanied by correction of the causal process. Supportive medications are used to stabilize the patient, not to replace a necessary correction.

Pericardiectomy is indicated in cases of clinically significant persistent constriction that is not reversible or controllable, after weighing benefit and risk. In this syndrome, the key technical problem may be the visceral pericardium: parietal resection alone may leave the constraint adherent to the heart. Visceral release, sometimes described as epicardiectomy, is more delicate because of its relationship with the myocardium and coronary arteries. The operation should be entrusted to an experienced team, with a plan adapted to the extent of adhesions and the hemodynamic conditions.

Preoperative preparation includes anatomic characterization, ventricular function, associated diseases, and renal, hepatic, and nutritional status. In radiation-induced or neoplastic forms, benefit may be limited by nonpericardial injury. Any use of cardiopulmonary bypass depends on complexity and the need for associated procedures. Obtained material should be sent for histology and, when relevant, microbiology. Relief of the constraint does not guarantee immediate normalization of every pressure, particularly after a long history of congestion or in the presence of myocardial disease.

Follow-up after treatment compares symptoms, venous pressure, organ function, and imaging with the initial picture. Physical activity is resumed gradually after inflammation is controlled and hemodynamic stability is achieved, with more cautious criteria when there is myocardial involvement. Medication reduction follows the documented response rather than simply the time elapsed since drainage. Reaccumulation of fluid, persistent congestion, or new deterioration requires identification of the responsible mechanism before automatically repeating the previous procedure.

Prognosis, complications, and evolution over time

The natural history is not uniform. In the 2004 prospective series by Sagristà-Sauleda and colleagues, seven of fifteen patients required pericardiectomy and three had spontaneous resolution. In the echocardiographic cohort of Kim and colleagues, only two of the thirty-three patients with effusive-constrictive pericarditis required surgery during a median follow-up of 3.8 years. The difference does not demonstrate the superiority of a single therapy: selection, diagnostic definitions, and etiologies differed. It does, however, show that recognition of the phenotype does not mandate immediate pericardiectomy for everyone.

The underlying cause weighs on prognosis at least as much as the hemodynamic behavior. Neoplasms, invasive infections, and radiation injury may lead to unfavorable outcomes through mechanisms that are not completely correctable by drainage or surgery. In idiopathic inflammatory forms, recovery of compliance is possible. This possibility should be verified by follow-up, not assumed on the basis of young age alone or the response of pain. Persistent functional limitation and congestion after inflammation has subsided instead suggest reassessment for a structural component.

Recurrent tamponade may occur if fluid reaccumulates, drainage is incomplete, or the cause is not controlled. Compartmentalized collections may be difficult to identify from a single echocardiographic window. A fall in blood pressure or worsening dyspnea requires reassessment of anatomy and physiology, without attributing every event to the known constraint. Prevention depends on etiologic control, selection of the appropriate procedure, and surveillance, not on repeated drainage at predetermined intervals.

Persistent constriction may lead to hepatic congestion, renal injury, ascites, hypoalbuminemia, and muscle loss. These consequences reduce reserve and may increase surgical risk. An initial period of reasoned observation should therefore not become inertia in the face of progressive deterioration. Assessment of response should include diuretic requirement and the ability to eat and perform activities, in addition to biomarkers and imaging. Control of an effusion does not equal control of the entire syndrome.

Procedural complications include bleeding, cardiac or coronary injury, arrhythmias, pneumothorax, and decompression syndrome after drainage; visceral surgery adds risks related to dissection of adherent tissue and low cardiac output. Immediate vigilance makes it possible to distinguish them from residual constrictive physiology. Medications may cause renal, gastrointestinal, metabolic, or infectious toxicity, especially in an already compromised patient. The balance of treatment therefore requires simultaneous assessment of hemodynamic efficacy and safety.

Long-term recovery includes reduction of congestion, nutritional recovery, and a gradual return to activities. It may be incomplete if advanced myocardial or organ damage exists, even when the pericardial component has been corrected. Documenting disappearance of constrictive signs helps avoid prolonged therapies without a target; their persistence guides further decisions. A clear plan for follow-up and urgent reassessment in case of syncope, hypotension, increasing dyspnea, or fever makes management safer during a phase in which the phenotype may still change.

Bibliography
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. Sagristà-Sauleda J et al. Effusive-constrictive pericarditis. New England Journal of Medicine. 2004;350(5):469-475.
5. Kim KH et al. Effusive-Constrictive Pericarditis After Pericardiocentesis: Incidence, Associated Findings, and Natural History. JACC: Cardiovascular Imaging. 2018;11(4):534-541.
6. Welch TD et al. Echocardiographic diagnosis of constrictive pericarditis: Mayo Clinic criteria. Circulation: Cardiovascular Imaging. 2014;7(3):526-534.
7. Talreja DR et al. Constrictive pericarditis in the modern era: novel criteria for diagnosis in the cardiac catheterization laboratory. Journal of the American College of Cardiology. 2008;51(3):315-319.
8. Spodick DH. Acute cardiac tamponade. New England Journal of Medicine. 2003;349(7):684-690.
9. Mayosi BM et al. Tuberculous pericarditis. Circulation. 2005;112(23):3608-3616.
10. 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.
11. Saab J et al. Diagnostic yield of cytopathology in evaluating pericardial effusions: Clinicopathologic analysis of 419 specimens. Cancer Cytopathology. 2017;125(2):128-137.
12. Brucato A et al. Effect of Anakinra on Recurrent Pericarditis Among Patients With Colchicine Resistance and Corticosteroid Dependence: The AIRTRIP Randomized Clinical Trial. JAMA. 2016;316(18):1906-1912.
13. Klein AL et al. Phase 3 Trial of Interleukin-1 Trap Rilonacept in Recurrent Pericarditis. New England Journal of Medicine. 2021;384(1):31-41.
14. Roberts WC. Pericardial heart disease: its morphologic features and its causes. Proceedings (Baylor University Medical Center). 2005;18(1):38-55.