Imaging of the Pericardium

Breen, Jerome F. M.D

Journal of Thoracic Imaging:
Cardiac Imaging, Part II: Original Articles

Computed tomography (CT) and magnetic resonance imaging (MRI) are modalities well suited for imaging of the pericardium and pericardial disease. Both offer excellent resolution with a wide field of view. Both have advantages and disadvantages when compared with each other and with echocardiography. Establishing the diagnosis of constrictive pericarditis is a common indication for CT or MRI of the pericardium. Pericarditis, neoplasms, effusions, and congenital anomalies are additional conditions involving the pericardium that can be diagnosed with CT and MRI.

Author Information

From the Mayo Clinic Foundation, Rochester, Minnesota.

Address correspondence and reprint requests to Dr. Jerome F. Breen, Mayo Clinic Foundation, Department of Diagnostic Radiology, 200 1st Street SW, Rochester, MN 55905.

Article Outline
Back to Top | Article Outline


The pericardial complex encloses the heart and consists of fibrous and serous components. The fibrous pericardium forms a flask-shaped, tough outer sac (Fig. 1). Much of the ascending aorta, pulmonary trunk, short segments of pulmonary veins, and the cavae are intrapericardial. The tough fibrous outer parietal layer has attachments to the diaphragm, sternum, and costal cartilage. The serous layer is a thin mesothelial layer adjacent to the surface of the heart. The potential space between the two levels normally contains a small amount of clear fluid, typically between 15 and 50 ml (1). The amount and position of this physiologic fluid is variable in normal individuals. This fluid can collect in the various sinuses of pericardial reflections and must be recognized as normal structures that can mimic pathology such as a dissection flap or adenopathy (Fig. 2) (2). The pericardial sac lies between variable amounts of epicardial and pericardial adipose tissue. This provides natural tissue contrast to define the pericardium (Figs. 3, 4).

The pericardium appears to have numerous physiologic functions; however, normal cardiac function can occur in post pericardiectomy patients as well as individuals with congenital absence of the pericardium. It does serve a protective role from the spread of infection and inflammation from adjacent mediastinal structures as well as limits pathologic cardiac displacement by virtue of its various tough fibrous attachments. Other proposed functions are the ability to reduce friction of cardiac motions and a role to limit the degree of acute distension of the heart (3). The presence of a normal pericardium also can be detrimental in situations of fluid rapidly filling the small potential space with limited ability to distend quickly. Slowly accumulating effusions cause the pericardium to stretch and can exceed a liter in volume.

Back to Top | Article Outline


Two-dimensional (2D) echocardiography is the initial imaging modality applied in nearly every case of suspected (or unsuspected) pericardial disease. This is not to say that it is the test of choice or the definitive test for various specific conditions. Transthoracic and transesophageal echocardiography are limited in their ability to image the entire pericardium because of their dependence on restricted acoustic windows to an organ surrounded by air-filled lung and a bony thoracic cage. Despite this limitation, most pathology of the pericardium results in, or is a result of, cardiovascular disease, therefore making echocardiography the first employed modality. It is the initial test of choice for detecting pericardial effusions and for diagnosing tamponade. In experienced hands, 2D transesophageal imaging with respiratory correlated Doppler features can be diagnostic of constrictive pericarditis (Fig. 5) (4–6).

Detailed anatomic display of the entire pericardium is best performed by x-ray computed tomography (CT) or magnetic resonance imaging (MRI) (7–9). A significant advantage of these techniques over echocardiography is the capacity to provide a wide field-of-view of the entire chest. In addition, CT and MRI are less operator dependent. Like echocardiography, these modalities can be quite comprehensive, with the ability to display pericardial-related anatomy, function, and physiology.

CT scanning has provided clinically useful detail of the pericardium for nearly two decades. A significant advantage of CT over other modalities is the ability to detect pericardial calcifications that are often helpful to make the diagnosis of constriction (Figs. 6–8). There has been a continuous evolution and development of faster and higher-resolution scanners. The temporal resolution of both electron beam CT and the more recently developed multi-row detector spiral CT scanners has made possible motion-free images of the pericardium and underlying cardiac structures (Fig. 9). High-quality sagittal and coronal reconstructions are now readily available on workstations provided with modern CT scanners. Cine CT acquisitions can provide very valuable dynamic functional and physiologic information (10). Until recently, cine CT was only available on electron beam CT scanners, which have imaging times as short as 50 milliseconds. Retrospectively gated “cine-like” protocols are in development on the more widely available multirow detector spiral CT scanners. The major disadvantage of CT scanning of the heart is the frequent need for iodinated intravenous contrast administration to best display the associated findings of pericardial pathology. A minor disadvantage of CT is the ionizing radiation dose to the patient. This dose is almost trivial in comparison with that of a prolonged fluoroscopic catheterization study that might be required if noninvasive studies are nondiagnostic. A limitation of CT imaging of the pericardium is the occasional difficulty in differentiating fluid from thickened pericardium (Fig. 10). This limitation is typically minimized with forehand knowledge of an echocardiography report of the presence or absence of an effusion. The normal pericardium is seen as a very thin linear density surrounding the heart but is often not visualized over much of the left ventricle, where there can be a paucity of pericardial and epicardial fat. The thickness of the normal pericardium measured by CT is less than 2 mm (11).

MRI scanning can provide a detailed and comprehensive examination of the pericardium and heart without the need for iodinated contrast material or the use of ionizing radiation. Functional cine acquisitions are widely available. An advantage over CT is superior ability to characterize any effusion present using a combination of T1, T2, and cine sequences. Image acquisition can be obtained in any plane with superior detail to what is ordinarily available by CT reconstructions. A major drawback to MRI is the necessity to cardiac gate the acquisition for optimal images. Whereas CT is a “snapshot,” a high-quality MRI of the heart requires anywhere from 12 to 256 or more regular heartbeats. Breath-hold fast spin echo sequences in individuals with regular rhythms can produce very high-quality images with less motion-related blurring than a conventional CT scan, but irregular rhythms are not uncommon, because there is a high association with atrial fibrillation and pericardial disease. MRI may be diagnostic in patients with arrhythmias, but often the examination is less than ideal. The “normal” pericardium as seen by MRI has been reported to be up to 4 mm in thickness (12,13). This measurement most likely includes the entire pericardial complex, with physiologic fluid representing a significant component of the thickness measured. Like with CT, the pericardium over the left ventricle may not be visible.

Back to Top | Article Outline


The request to “rule-out constriction” is nearly a daily occurrence at our large multispecialty practice. The presentation of patients with constrictive pericarditis can mimic numerous clinical disease states, including restrictive cardiomyopathy, heart failure, and liver disease. The causes for constriction are also numerous, including postsurgical, postradiation, posttraumatic, and postinfectious presentations (14,15). Often the cause is labeled idiopathic. The demonstration of thickened pericardium with or without calcifications in the proper clinical setting is basically diagnostic of constriction (Figs. 11, 12). Additional findings seen with constriction include distorted contours of the ventricles, tubular-shaped ventricles, hepatic venous congestion, ascities, pleural effusions, and occasionally some pericardial effusion. Often there is dilatation of the atria, coronary sinus, inferior vena cava, and hepatic veins. Cine acquisitions, by MRI or electron-beam CT, show an abnormal motion of the interventricular septum in early diastole. The pericardium can be globally thickened, but frequently it is only focally thickened (Fig. 13). Compression of the right ventricle caused by focal thickening is more common than left-sided compression. Descriptive reports of the location of the focal thickening will aid the surgical approach to release the constricted heart chambers (16). The presence of thickened pericardium alone without clinical evidence of impaired diastolic filling indicates current or remote pericarditis and not necessarily constriction. The absence of thickened pericardium argues against the diagnosis of constriction but does not rule it out. Pericarditis can resolve, leaving a minimal residual thickening but with impairment of normal compliance of the pericardium and filling of the ventricles (Fig. 14). The presence of any calcification in an individual suspected of constriction should be considered significant. CT will detect minute amounts of pericardial calcium, and MRI can miss significant deposits. This fact is the primary reason why at our institution MRI for the diagnosis of constriction is reserved for those patients who have contraindications to iodinated contrast material.

Back to Top | Article Outline


Primary pericardial tumors are rare, with mesotheliomas and sarcomas being among the more “common” (17). Metastatic involvement of the pericardium is relatively frequent in autopsy series, but the demonstration of metastatic deposits by any imaging method is unusual. Most often the source is carcinoma of the lung or breast, but melanoma and renal metastases are among the more frequently visible by CT or MRI. Pericardial involvement caused by direct extension of pulmonary and mediastinal malignancies is more commonly shown with imaging. Focal obliteration of the pericardial line with or without an effusion indicates extension. If the effusion is hemorrhagic, extension is certain. This is easily seen by MRI with high signal intensity on spin-echo imaging. Common paracardiac masses include pericardial cysts, prominent pericardial fat pads, diaphragmatic eventration, and hernia, pulmonary malignancies, and lymphoma. CT and MRI are both well suited for imaging of these conditions (Figs. 15, 16).

Back to Top | Article Outline


Pericarditis is an inflammatory response by the pericardium associated with a large variety of clinical conditions. There may be little or no increase in pericardial fluid. Chest pain is often the presenting symptom. MRI and CT are both useful in establishing the diagnosis and in monitoring for resolution or possible progression to constriction (Figs. 17, 18). MRI is superior in differentiating fluid from thickened pericardium (Fig. 19). MRI can provide some additional tissue characterization of pericardial effusions with simple transudative effusions having no or very little signal intensity of T1-weighted images (18). Hemorrhagic or exudative effusions will often have medium or high signal on T1 sequences.

Back to Top | Article Outline


Pericardial cysts and diverticula are benign developmental structures that are well marginated containing water density fluid. Cysts are more commonly located in the right anterior cardiophrenic angle but can be located throughout the mediastinum (Fig. 20). Their identification by CT or MRI is typically not challenging. As with other fluid collections, MRI has the better capacity for characterization. Cysts are thought to form when a portion of the embryonic pericardium is redundant and pinched off. Absence of the pericardium can be complete but is more often partial, with the defects more often left-sided. Individuals with partial or complete absence of the pericardium are usually asymptomatic (19). When most of the pericardium is absent, the heart axis will usually shift to the left and posteriorly (Fig. 21). Partial absence of the left side can result in a prominent-appearing left atrial appendage or pulmonary artery segment (20). Herniation with strangulation of the appendage through the defect is rare. A congenital left atrial appendage aneurysm is even more rare, but appears similar by chest radiograph to a partial pericardial defect (Fig. 22) (21,22). The aneurysm has been associated with a risk of embolic stroke. CT and MRI may be helpful in determining the presence or absence of a segment of pericardium, but lack of visualization of the pericardium over the left ventricle and atrial appendage is not sufficient to make the diagnosis. Herniation of most of the heart through a surgical pericardial defect can be life-threatening and is often diagnosed by simple chest radiograph and clinical history.

Back to Top | Article Outline


1. Edwards ED. Applied anatomy of the heart. In: Giulaini ER, Fuster V, eds. Cardiology: fundamentals and practice. 2nd ed. St. Louis, MO: Mosby-Year Book, 1991:47–51.
2. Groell R, Schaffler GJ, Rienmueller R. Pericardial sinuses and recesses: findings at electrocardiographically triggered electron-beam CT. Radiology 1999; 212 (1):69–73.
3. Spodick DH. The normal and diseased pericardium: current concepts of pericardial physiology, diagnosis and treatment. J Am Coll Cardiol 1983; 1:240–51.
4. Ling LH, Oh JK, Tei C, Click RL, Breen JF, Seward JB, Tajik AJ. Pericardial thickness measured with transesophageal echocardiography: feasibility and potential clinical usefulness. J Am Coll Cardiol 1997; 29 (6):1317–23.
5. Seward JB, Khandheria BK, Oh JK, et al. Transesophageal echocardiography: technique, anatomic correlations, implementation, and clinical applications. Mayo Clin Proc 1988; 63:649–80.
6. Oh JK, Hatle LK, Seward JB, Danielson GK, Schaff HV, Reeder Gs, Tajik AJ. Diagnostic role of Doppler echocardiography in constrictive pericarditis. J Am Coll Cardiol 1994;23(1):154–62, 1994.
7. Olson MC, Posniak HV, McDonald V, Wisniewski R, Moncada R. Computed tomography and magnetic resonance imaging of the pericardium. Radiographics 1989; 9:633–49.
8. Olson MC, Posniak HV, McDonald V, Wisniewski R, Moncada R. Computed tomography and magnetic resonance imaging of the pericardium. Radiographics 1989; 9 (4):633–49.
9. White CS. MR evaluation of the pericardium. Top Magn Reson Imaging 1995; 7 (4):258–66.
10. Oren RM, Grover-McKay M, Stanford W, Weiss RM. Accurate preoperative diagnosis of pericardial constriction using cine computed tomography. J Am Coll Cardiol 1993; 22 (3):832–8.
11. Bull RK, Edwards PD, Dixon AK. CT dimensions of the normal pericardium. Br J Radiol 1998; 71 (849):923–5.
12. Sechtem U, Tscholakoff D, Higgins CB. MRI of the normal pericardium. Am J Roentgenol 1986; 147:239–44.
13. Sechtem U, Tscholakoff D, Higgins CB. MRI of the abnormal pericardium. Am J Roentgenol 1986; 147:245–52.
14. Cameron J, Oesterle SN, Baldwin JC, Hancock EW. The etiologic spectrum of constrictive pericarditis. Am Heart J 1987; 113:354–60.
15. Vaitkus PT, Kussmaul WG. Constrictive pericarditis versus restrictive cardiomyopathy: a reappraisal and update of diagnostic criteria. Am Heart J 1991; 122 (5):1431–41.
16. Rienmuller R, Gurgan M, Erdmann E, Kemkes BM, Kreutzer E, Weinhold C. CT and MR evaluation of pericardial constriction: a new diagnostic and therapeutic concept. J Thorac Imag 1993; 8:108–21.
17. Hancock EW. Neoplastic pericardial disease. Cardiol Clin 1990; 8:673–82.
18. Mulvagh SL, Rokey R, Vick GW, Johnson DL. Usefulness of nuclear magnetic resonance imaging for evaluation of pericardial effusions, and comparison with two-dimensional echocardiography. Am J Cardiol 1989; 64:1002–9.
19. Rusk RA, Kenny A. Congenital pericardial defect presenting as chest pain. Heart 1999; 81 (3):327–8.
20. Rozenshtein A, Boxt LM. Plain-film diagnosis of pericardial disease. Semin Roentgenol 1999; 34 (3):195–204.
21. Gassner I, Judmaier W, Fink C, Lener M, Waldenberger F, Scharfetter H, Hammerer I. Diagnosis of congenital pericardial defects, including a pathognomic sign for dangerous apical ventricular herniation, on magnetic resonance imaging. Br Heart J 1995; 74 (7):60–6.
22. Vargas-Barron J, Sanchez-Ugarte T, Keirns C, Vasquez-Sanchez J, Fernandez-Vazquez F. The differential diagnosis of partial absence of left pericardium and congenital left atrial aneurysm. Am Heart J 1989; 118 (6):1348–50.
Back to Top | Article Outline
Section Description

Editor: Jeffrey S. Klein

Associate Editors: Ann Leung, MD

David Lynch, MD, Jung-Gi Im, MD

Michio Kono, MD, Charles White, MD

Guest Editor: William Stanford, M.D.

Cited By:

This article has been cited 4 time(s).

Cardiology in Review
The Utility of Cardiovascular Magnetic Resonance in Constrictive Pericardial Disease
Ariyarajah, V; Jassal, DS; Kirkpatrick, I; Kwong, RY
Cardiology in Review, 17(2): 77-82.
PDF (696) | CrossRef
Journal of Computer Assisted Tomography
Septal Bounce in Constrictive Pericarditis: Diagnosis and Dynamic Evaluation with Multidetector CT
Ghersin, E; Lessick, J; Litmanovich, D; Ofer, A; Elhasid, R; Lorber, A; Dragu, R; Rispler, S; Haimi, M; Engel, A
Journal of Computer Assisted Tomography, 28(5): 676-678.

PDF (119)
Clinical Nuclear Medicine
Pericarditis Associated With Cryptococcal Meningitis on Ga-67 Scintigraphy
Hsiao, E; Mansberg, R; Huynh, T; Roach, P
Clinical Nuclear Medicine, 29(2): 119-120.
PDF (342) | CrossRef
Journal of Thoracic Imaging
Incidental Cardiac and Pericardial Abnormalities on Chest CT
Lee, S; Seo, JB; Kang, J; Chae, EJ; Park, SH; Lim, T
Journal of Thoracic Imaging, 23(3): 216-226.
PDF (2911) | CrossRef
Back to Top | Article Outline

Pericardium; CT; MRI

© 2001 Lippincott Williams & Wilkins, Inc.