pericardial disease: value of ct and mr imaging1

Revi

ews

and

Com

men

taRy

n R

evie

w

340 radiology.rsna.org n Radiology: Volume 267: Number 2—May 2013

Pericardial disease: Value of CT and MR Imaging1

Jan Bogaert, MD, PhDMarco Francone, MD, PhD

The pericardium represents an important focus of morbid-ity and mortality in patients with cardiovascular disease. Fortunately, in recent years knowledge regarding this enigmatic part of the heart and the diagnosis of related diseases has substantially advanced. To a large extent, this can be attributed to the availability of several non-invasive cardiac imaging modalities. Transthoracic echo-cardiography, which combines structural and physiologic assessment, is the first-line technique for examination of patients suspected of having or known to have pericardial disease; however, cardiac computed tomography (CT) and magnetic resonance (MR) imaging are becoming increas-ingly popular for the study of this part of the heart. Mod-ern multidetector CT scanners merge acquisition speed and high spatial and contrast resolution, with volumetric scanning to provide excellent anatomic detail of the peri-cardium. Multidetector CT is by far the modality of choice for depiction of pericardial calcifications. MR imaging is probably the best imaging modality for the acquisition of a comprehensive view of the pericardial abnormalities. MR imaging combines cardiac and pericardial anatomic assessment with tissue characterization and appraisal of the effects of pericardial abnormalities on cardiac per-formance. This review aims to elucidate the role of the pericardium and its interaction with the remainder of the heart in normal and pathologic conditions. It focuses on the rapidly evolving insights regarding pericardial disease provided by modern imaging modalities, not infrequently necessitating reconsideration of evidence that has thus far been taken for granted.

q RSNA, 2013

Supplemental material: http://radiology.rsna.org/lookup /suppl/doi:10.1148/radiol.13121059/-/DC1

1 From the Department of Radiology, Medical Imaging Research Center, University Hospitals Leuven, Herestraat 49, B-3000 Leuven, Belgium (J.B.); and Cardiovascular MRI Unit, Department of Radiological, Oncological and Patholog-ical Sciences, University of Rome La Sapienza, Rome, Italy (M.F.). Received June 6, 2012; revision requested July 16; revision received August 12; accepted October 2; final version accepted October 31. Address correspondence to J.B. (e-mail: [email protected]).

q RSNA, 2013

Learning Objectives:

After reading the article and taking the test, the reader will be able to:

n Describe the CT and MR appearances of normal pericardial anatomy, related sinuses and recesses, and potential diagnostic pitfalls.

n Explain the contribution of pericardial anatomy and physiology to ventricular interdependence and diastolic filling.

n Apply CT and MR techniques to the diagnosis of constrictive pericarditis, pericardial effusions, neoplasms, and congenital anomalies.

n Differentiate constrictive pericarditis from restrictive cardiomyopathy by assessing septal motion.

Accreditation and Designation Statement

The RSNA is accredited by the Accreditation Council for Continuing Medical Education (ACCME) to provide continuing medical education for physicians. The RSNA designates this journal-based CME activity for a maximum of 1.0 AMA PRA Category 1 Credit TM. Physicans should claim only the credit commensurate with the extent of their participation in the activity.

Disclosure Statement

The ACCME requires that the RSNA, as an accredited provider of CME, obtain signed disclosure statements from the authors, editors, and reviewers for this activity. For this journal-based CME activity, author disclosures are listed at the end of this article.

Online CMESee www.rsna.org/education/ry_cme.html

Note: This copy is for your personal non-commercial use only. To order presentation-ready copies for distribution to your colleagues or clients, contact us at www.rsna.org/rsnarights.

Radiology: Volume 267: Number 2—May 2013 n radiology.rsna.org 341

REVIEW: CT and MR Imaging of Pericardial Disease Bogaert and Francone

direct attachment to the heart itself, having its inner surface lined by a se-rous membrane.” This definition of the pericardium by R. B. Todd in the Cy-clopaedia of Anatomy and Physiology from 1835 is still up to date more than 175 years after it was written (26). This thin flask-shaped membrane is macro-scopically composed of two layers, the inner serosa (also referred as the vis-ceral pericardium or the epicardium) and the outer fibrosa (also referred as the parietal pericardium) (27,28). The serosa forms a complete sac filled with up to 50 mL of plasmatic ultrafiltrate and is separated from the heart by loose epicardial connective tissue and a single layer of mesothelial cells.

Like visceral abdominal fat, the thickness of the epicardial fat is in-creased in obesity, and there is accumu-lating evidence that epicardial fat can directly influence coronary atherogene-sis, because there is no fibrous layer to impede diffusion to the underlying coro-nary arteries of the free fatty acids and adipokines that modulate atherothrom-bosis (29–31). The fat distribution is typically asymmetric, with three to four times more epicardial fat present along the right ventricle (RV) than along the left part of the heart (29). The outer layer, the fibrous pericardium, is at-tached internally to the epicardium and extends cranially above level of the aor-tic root (Fig 1). Moreover, this layer is continuous with the deep cervical fas-cia and is attached to the sternum and the diaphragm by loose ligaments that impede cardiac displacement in the mediastinum. Another anatomic fea-ture of the pericardium is the presence of two major reflections of the visceral layer, which are arranged around two

recognized for almost half a century, allowing both structural assessment and evaluation of the physiologic con-sequences to the heart (4–8). However, echocardiography may fail to allow ad-equate evaluation of the pericardium at times. In particular, the diagnostic accuracy of echocardiography is limited in patients with a suboptimal acoustic window (eg, obese subjects, patients with severe chronic obstructive pulmo-nary disease or skeletal malformations) (9,10). Moreover, transthoracic echo-cardiography is limited in its ability to aid in recognition of focal effusions, as-sessment of pericardial thickness, and characterization of tissue (10,11).

Since the late 1970s, computed to-mography (CT) and magnetic resonance (MR) imaging have emerged as inter-esting alternatives to echocardiography for morphologic assessment of the heart and pericardium (Table 1) (12–15). In recent years, improvements in MR imag-ing hardware and sequence design have further enhanced the diagnostic value of this modality for the study of pericardial disease. For example, fast MR sequences allow one to study cardiac motion and inflow patterns in real time during free breathing, which allows identification of constrictive physiology (15–19). More-over, delayed (or late) gadolinium-en-hanced MR imaging is useful for the de-tection of pericardial inflammation and for monitoring the effects of medical (anti-inflammatory) treatment (20,21). Cardiac CT with multidetector scan-ners and electrocardiographically (ECG) synchronized data acquisition allows an accurate assessment of the coronary arteries and the remainder of the heart and pericardium, including functional cardiac analysis (22–25).

In this review, we will describe how CT and MR imaging have dramatically shaped our view on imaging of pericar-dial disease and how these techniques may contribute to the optimization of current patient care.

Pericardial Anatomy and Physiology

“The pericardium is a fibrous bag sur-rounding the heart and origin of the large blood vessels, but without any

Pericardial disease is an important cause of morbidity and mortality in patients with cardiovascular

disease. It represents a heterogeneous group of congenital and acquired dis-eases, with a widely variable clinical manifestation that often necessitates a specific treatment. Besides isolated peri-cardial disease, the pericardium may be secondarily involved by a large group of organ and systemic diseases, such as in-fective, autoimmune, and neoplastic pro-cesses. Moreover, iatrogenic causes—for example, after cardiac surgery or radia-tion therapy—represent an important cause of pericardial-related morbidity and mortality. Pericardial disease may be an incidental finding (eg, pericardial cyst) but may also necessitate urgent in-tervention in case of hemodynamic com-promise (eg, acute cardiac tamponade, herniation of cardiac contents through a partial pericardial defect).

The diagnosis of pericardial disease frequently remains clinically challeng-ing, requiring integration of medical history and findings from physical ex-amination, imaging, blood analyses, and, eventually, invasive hemodynamic measurements and/or pericardial bi-opsy (1–3). The role of transthoracic echocardiography in helping diag-nose pericardial disease has been well

Essentials

n Despite its relatively simple mac-roscopic and ultrastructural architecture, the pericardium has a crucial but complex role in the normal performance of the heart.

n CT and MR imaging have dramat-ically changed our view of peri-cardial disease imaging and in-creasingly contribute to the care of patients with pericardial disease.

n Although pericardial thickness has traditionally been used as an important criterion for constric-tive pericarditis, this concept is under greater and greater chal-lenge and should be put in per-spective with regard to recent findings.

Published online10.1148/radiol.13121059 Content codes:

Radiology 2013; 267:340–356

Abbreviations:ECG = electrocardiographyLV = left ventricleRV = right ventricleSSFP = steady-state free precession

Conflicts of interest are listed at the end of this article.



but importantly, the pericardial sac, hav-ing a slightly negative pressure, acts not only as an anatomic barrier but also as an important physiologic intermediary between the heart and the remainder of the chest. As such, it functions as a pressure transducer between the pleural spaces and cardiac chambers. In normal conditions, the intrathoracic respiratory changes are directly transmitted to the cardiac chambers. During inspiration, the intrathoracic pressure decreases, which augments systemic venous return and RV filling, while LV filling is not influenced.

the mediastinum (33,34). When surgi-cally incised, the fibrous pericardium spontaneously retracts, indicating that it is under some physical stress. This mild degree of pericardial stretching impedes pathologic overdistention, mainly of the thin-walled cardiac chambers such as the RV and right atrium, and directly influ-ences intracavitary diastolic pressures.

Since the ventricles are surrounded by a relatively inelastic pericardium, the RV and left ventricle (LV) do not act in-dependently, although there is a mutual interaction, a phenomenon defined as “ventricular coupling” or “ventricular in-terdependence.” During ventricular fill-ing, the position of the interventricular septum is determined by the pressure difference between the RV and LV. In normal loading conditions, LV diastolic pressures slightly exceed RV pressures, which gives the septum a typically convex shape with a slight right-sided bulge. In many pathologic conditions, this shape may be altered. Septal flattening or in-version may occur in conditions with RV volume (eg, atrial septal defect or pressure overload [acute or chronic cor pulmonale]) and in pericardial abnor-malities where pericardial pressures are increased (eg, cardiac tamponade) or pericardial compliance is decreased (eg, constrictive pericarditis) (27,35). Finally,

complex connected tubes, described as the transverse and oblique sinuses, and include several smaller recesses (Fig 1). These sinuses have a reserve func-tion and may strategically accumulate fluid in the presence of increased fluid content, thus creating pocketlike struc-tures (the so-called pericardial reserve volume) (28,32). The transverse sinus is located behind the ascending aorta and pulmonary trunk and in front of the atria and the superior vena cava. This linear-shaped virtual cavity should not be misinterpreted when filled with fluid as a focal aortic dissection or as an enlarged mediastinal lymph node on CT or MR images (28,32). The oblique sinus is an inverted U-shaped pericar-dial reflection located behind the left atrium (28).

Despite its relatively simple mac-roscopic and ultrastructural architec-ture, the pericardium has a crucial but complex role in ensuring the normal performance of the heart. This role is three fold: mechanical, membranous, and ligamentous. As such, it allows the regulation and maintenance of normal ventricular compliance to minimize ventricular dilatation and to physically protect the heart by means of the pro-duction of fluid and surfactants. It also limits the displacement of the heart in

Table 1

Comparison of Noninvasive Imaging Modalities to Study the Pericardium

Target TTE* CT MR Imaging

PericardiumPericardial thickness Moderate Excellent ExcellentPericardial calcifications Good Excellent Poor or not possiblePericardial inflammation Moderate Good ExcellentMotion layers (adhesions) Good Moderate ExcellentEffusion detection Moderate to good Excellent ExcellentEffusion characterization Moderate Good GoodPericardial masses Moderate Moderate to good Good to excellentCardiac morphology (including tissue

characterization)Good Good Excellent

Cardiac functionSystolic function Good Good† ExcellentDiastolic function Excellent Poor or not possible GoodSeptal motion (coupling) Excellent Poor to moderate Excellent

* TTE = transthoracic echocardiography.† Use of ECG synchronized data acquisition.

Figure 1

Figure 1: Anatomic drawing of interior of pericar-dial sac after removal of the heart. At the reflections of the serous pericardium between the great vessels at base of the heart, the pericardial cavity forms the pericardial sinuses, which are not separate compart-ments but represent extensions of the pericardial cavity. The transverse sinus behind the ascending aorta (Ao) and pulmonary trunk (PT) forms a connection between the recesses adjacent to the right pulmonary veins (PV) and superior vena cava (SVC) with recesses near the left pulmonary vein. The oblique sinus is located behind the left atrium. Double layer of serous pericardium separates the transverse sinus from the oblique sinus. IVC = inferior vena cava. (Illustration by Bettina Conti, MD, Sapienza University, Rome, Italy.)



acquisitions, which improve visuali-zation of the pericardial sinuses and recesses and thus reduce the risk of misinterpreting these pericardial struc-tures as enlarged lymph nodes or focal aortic disease (25,28). Although dy-namic evaluation of ventricular septal motion to evaluate constrictive physiol-ogy is theoretically possible with ECG-synchronized CT, real-time functional imaging is much easier and more accu-rately assessed by using echocardiogra-phy and MR imaging (39).

Comprehensive MR imaging of the pericardium includes (a) morphologic assessment of the heart, pericardium, and surrounding mediastinum; (b) as-sessment of global and regional LV and RV function; (c) assessment of ventric-ular coupling by using real-time imaging during free breathing; (d) assessment of ventricular filling during breath hold-ing or free breathing by using real-time imaging; (e) tissue characterization; and (f) evaluation of pericardial mobility and fusion of pericardial layers (Table 2) (40).

Black-blood T1-weighted spin-echo MR imaging performed with a fast seg-mented sequence is the best approach for morphologic study of the heart,

CT and MR Imaging Techniques

For pericardial imaging by means of CT equipped with multidetector technol-ogy, use of a high-resolution volumetric acquisition with a section thickness of greater than 3 mm generally yields an excellent anatomic depiction of the peri-cardium (Fig 2) (15,37,38). Image arti-facts related to cardiac motion can be minimized by using ECG-synchronized

In patients with constrictive pericarditis, however, the pathologic pericardium im-pedes transmission of intrathoracic pres-sures, causing typical respiratory-related abnormalities in ventricular filling (8). Assessment of the presence and severity of ventricular coupling, as well as eval-uation of the effect of respiration, are crucial in the diagnosis of constrictive pericarditis and its differentiation from restrictive cardiomyopathy (36).

Figure 2

Figure 2: Effect of section thickness on visualization of normal pericardium. Multidetector CT images of the chest obtained from a dual-source scanner (Somatom Definition Flash; Siemens, Erlangen, Germany) with 1.0-mm collimation. Nonionic contrast material (120 mL) was injected at a rate of 5 mL/sec. On images with (a) section thickness of 1 mm and (b, c) reconstructed section thicknesses of (b) 3 mm and (c) 10 mm show pericardium as a thin curvilinear structure (arrows). Because of surrounding fat, pericardium is typically best visible along the right side of the heart. With increasing section thickness, maximal pericardial thickness (measured along the RV) increases from 0.9 mm to 1.0 and 1.6 mm by using section thicknesses of 3 and 10 mm, respectively.

Table 2

Assessment of the Pericardium with MR Imaging

Target MR Sequences

Pericardial morphology T1 weighted, cinePericardial layer and fluid characterization T1 and T2 weighted, cine, T1-weighted contrast agent

enhanced or late gadolinium enhancedPericardial mobility and fusion of pericardial

layersCine, MR tagging

Cardiac morphology T1 weighted, cineCardiac systolic function cine imagingCardiac filling (diastolic function) Phase contrastVentricular coupling Real-time cine, phase contrastOther findings* T1 weighted, T2 weighted, cine, late gadolinium enhanced

* Other findings include myocarditis, myocardial infarction, and myocardial infiltrative or storage disease; caval vein size; mediastinal and pulmonary pathologic conditions; and pleural fluid and ascites.



contours and is surrounded by a variable amount of epicardial fat tissue (Figs 2, 3). Typically, the pericardium is best visualized along the RV, while, because of a sparseness of epicardial fat and the vicinity of pulmonary parenchyma, the pericardium is often difficult to discern along the lateral and posterior LV wall (15,41). Normal pericardial thickness ranges from 1.2 to 1.7 mm on MR im-ages and from 0.7 to 2.0 mm on CT images (25,32,38,51–54). These values slightly overestimate those obtained in anatomic cadaveric studies (ie, 0.4–1.0 mm) (55,56). Differences can be ex-plained by the intrinsically limited spa-tial and temporal resolutions of CT and MR imaging, which do not allow one to fully discriminate between pericardial layers and fluid component. Pericar-dial sinuses and their recesses are fre-quently depicted on CT and MR images (14,28,37,57–59). Their visualization on CT images can be improved by using an ECG-synchronized acquisition and thin-section scanning (28,37). Large recesses are depicted in up to 45% of patients, and smaller recesses are shown in nearly 20% (37). Most recesses are lin-ear when not filled with fluid and band-shaped as the fluid content increases. However, they may also present as crescent, triangle, spindle, ovoid, hemi-sphere, or irregular shapes (37).

Congenital Pericardial Disease

Pericardial Cyst and DiverticulumPericardial cyst and pericardial diver-ticulum are rare congenital abnormal-ities that occur in approximately one in 100 000 individuals and account for 13%–17% of all mediastinal cysts (60). A pericardial cyst is a benign unilocular mass of celomic origin. These cysts are usually incidental findings at chest radi-ography or transthoracic echocardiogra-phy, although symptoms may occur in patients with cardiac compression (Fig 4; Movie 1 [online]) (33,61). A pericardial cyst appears on CT and MR images as an encapsulated fluid-filled structure without internal septa or nodules that generally is directly attached to the peri-cardium but rarely can be attached by

gradient-echo sequences is the currently preferred technique for quantifying global and regional cardiac systolic func-tion and is well suited to rule out associ-ated RV or LV dysfunction (40,41). The high spatial and temporal resolutions of this sequence can be applied in new ap-plications, such as dynamic evaluation of the rigidity of the pericardial layers in patients with constrictive pericarditis. Fast cine sequences enable real-time evaluation of dynamic fast-changing physiologic events such as ventricular coupling (17). MR tagging techniques are valuable in the detection of both fibrotic adhesion of pericardial layers and myocardial involvement in constric-tive pericarditis (48). Assessment of diastolic heart function, usually accom-plished by means of Doppler echocardi-ography, can also be achieved by using velocity-encoded or phase-contrast MR sequences (49,50). Analysis of pulmo-nary and/or systemic venous patterns in combination with cardiac inflow pat-terns through the atrioventricular valves can yield findings that are classic for re-strictive cardiac filling. Real-time acqui-sition during free breathing allows one to identify constrictive physiology (19).

Imaging of Normal Pericardium

The normal pericardium appears on CT and MR images as a thin curvilin-ear structure that follows the myocardial

pericardium, and mediastinum (41,42). The use of a small field of view and a sat-uration pulse positioned on the frontal chest wall may improve pericardial visu-alization. To guarantee complete depic-tion of the heart, it is advisable to image the heart and pericardium in two perpen-dicular planes. T2-weighted spin-echo imaging, preferably performed by using a short-tau inversion-recovery sequence (also called triple-inversion spin echo), enables depiction of pericardial fluid and/or edema of the pericardial layers in patients with inflammatory pericardi-tis. Moreover, concomitant myocardial edema may be identified in patients with an acute coronary syndrome or acute myocarditis (41,43,44). MR imaging after administration of a paramagnetic contrast agent is useful in patients sus-pected of having pericardial masses or inflammatory pericarditis and to depict concomitant myocardial pathologic con-ditions (eg, myocarditis) and is of great value in depicting persistent chronic in-flammation in patients with constrictive pericarditis (3,20,21,41,44).

Although T1-weighted spin-echo sequences can be used for contrast-en-hanced imaging, we recommend use of late (or delayed) gadolinium-enhanced inversion-recovery MR imaging, as has been proposed for imaging myocardial infarct and viability (20,45–47). Cine MR imaging performed with balanced steady-state free precession (SSFP)

Figure 3

Figure 3: Normal pericardium. Black-blood T1-weighted fast spin-echo MR images (repetition time/echo time msec = one heartbeat/30, 90° flip angle, 1.4 3 2.0-mm in-plane resolution) in horizontal (a) long-axis and (b) short-axis planes show pericardium as a thin curvilinear hypointense structure (arrows), best visible when surrounded by fatty tissue.



failure, renal and liver insufficiency, in-flammation, infection, neoplastic disease, trauma, and myocardial infarction (39). Imaging is often necessary to confirm the presence, severity, and extent of fluid; to characterize the nature of fluid (transu-dates vs exudates); to rule out pericardial inflammation; to determine the hemody-namic effect on the heart; and ultimately to guide pericardiocentesis.

Although transthoracic echocardiog-raphy is the preferred first-line modality for this evaluation, CT and MR imaging often allow for a more comprehensive

location of the defect (Fig 5) (67,68). Since herniation is often intermittent in time, positional changes to the left lat-eral decubitus position, for example, can be helpful in diagnosing pericardial de-fects. A functional examination may also be helpful in establishing the diagnosis of congenital pericardial defect (69).

Acquired Pericardial Disease

Pericardial EffusionAbnormal accumulation of pericardial fluid may occur in patients with heart

a peduncle. Most commonly, it is found in the right cardiophrenic angle (70% of cases), and less frequently in the left cardiophrenic angle (20% of cases) (62). Pericardial diverticulum is a focal out-pouching of the pericardial sac, which can be differentiated from a congenital cyst by the presence of direct communi-cation with the pericardial cavity, usually identified by changes in size related to body position (60).

Pericardial DefectA congenital defect of the pericardium is an extremely rare anomaly that oc-curs as a consequence of premature atrophy of the cardinal veins supplying the pleuropericardial folds during em-bryogenesis, with subsequent failure of the membrane to develop. The extent of the defect, likely determined by the timing of vascular degeneration, ranges from partial to complete absence of the pericardial sac (63,64). Partial defects (large ones are more common than small ones) are far more frequent than total defects, and the left side is more commonly affected than the right side (Fig 5) (25,33,63). Although patients are often asymptomatic, symptoms may arise when cardiac structures are tran-siently entrapped or incarcerated in the defect (65). Herniation of the left atrial appendage through a small defect may lead to infarction of the appendage, or the left coronary artery might be com-pressed, leading to ischemia especially during exercise (66).

In about one-third of cases, pericar-dial defects are associated with other congenital abnormalities such as bron-chogenic cysts, ventricular septal defect, patent ductus arteriosus, or diaphrag-matic hernia. On chest radiographs, a left-sided pericardial defect causes typ-ical levodisplacement of the heart and aortic knob, with the trachea remaining at the midline. The problem in CT and MR imaging of left-sided pericardial de-fects is the inability of these modalities to visualize the pericardium in this part of the heart. Therefore, the diagnosis of left-sided defects usually relies on other signs, such as an abnormal position of cardiac structures with excessive levo-rotation or cardiac indentation at the

Figure 4

Figure 4: Typical appearance of a pericardial cyst. (a) Frontal chest radiograph shows sharply defined opacity in the left cardiophrenic angle. (b) Axial contrast-enhanced CT image shows presence of well-de-fined, oval-like, fluid-filled structure (attenuation: 3 HU) in broad contact with pericardium. (c) Cyst has low signal intensity on T1-weighted spin-echo MR image (one heartbeat/30, 90° flip angle, 1.4 3 2.0-mm in-plane resolution) and (d) high signal intensity on T2-weighted short tau inversion-recovery spin-echo MR image (two heart beats/100, 90° flip angle, 1.4 3 2.0-mm in-plane resolution). Both c and d were obtained in horizontal long-axis plane. No cystic wall is discernible on b–c. Note presence of breast implants. Relation-ship of pericardial cyst to LV can be well appreciated on Movie 1 (online).



may lead to cardiac chamber compres-sion, compromising diastolic filling and causing a reduction of cardiac output with subsequent hypotension, tachy-cardia, and potentially progression to cardiogenic shock (71). This condition can be triggered by multiple causes, including trauma, inflammation, aortic dissection, neoplastic involvement of the pericardial space, acute myocardial infarction, and cardiac surgery (71,72). Symptoms may arise acutely or more chronically and are determined by the absolute volume of fluid, the speed of fluid accumulation, and the physical characteristics (ie, elasticity) of the pericardium and epicardium.

Since pericardial layers have a limited capacity for stretching, the fast accumulation of a relatively limited amount of fluid can cause a substan-tial elevation in pericardial pressures, which can potentially be lethal minutes after onset (Fig 7). In contrast, a slow accumulation of a considerable amount of fluid (up to 2 L or more) may have relatively unremarkable effects on ven-tricular filling.

The diagnosis of pericardial tam-ponade is a clinical one that is usually confirmed with echocardiography. Typ-ical features are diastolic collapse of the RV free wall, indicating that the pericardial pressure exceeds ventricu-lar pressures; right atrial compression during early systole; exaggerated respi-ratory variation in mitral and tricuspid inflow as a consequence of increased ventricular coupling; and distention of the inferior vena cava and hepatic veins with diastolic flow reversal (10,41). However, the radiologist should be fa-miliar with the appearance on CT and MR images, although their role in the diagnosis of acute cardiac tamponade is usually limited. Patients with tampon-ade usually have a large pericardial ef-fusion (or, as already mentioned, other contents such as air or blood) at the time of presentation. Typical findings of tamponade include flattening or in-version of the right atrial or RV wall, with compression of these chambers; inversion of the interventricular sep-tum; distention of the superior vena cava and inferior vena cava; and reflux

shorten the rate of T2 relaxation (lower signal intensity). Hemopericardium can be suspected in patients who have pre-viously undergone aortic or cardiac sur-gery or who have a history of trauma or neoplastic disease and is characterized by high signal intensity on T1-weighted images and inhomogeneous low signal intensity on cine SSFP images. Because of motion artifacts, however, pericar-dial fluid characterization is not always feasible. Bright-blood dynamic cine MR imaging often allows a better apprecia-tion the intrapericardial contents, such as the visualization of fibrinous strands or of the presence of coagulated blood (40). Third, both CT and MR imaging en-able accurate depiction of the pericardial layers, with assessment of thickness and composition. As such, CT and MR im-aging allow the differentiation of simple pericardial effusions from inflammatory effusive pericarditis or malignant pericar-dial diseases.

Cardiac TamponadeCardiac tamponade may be a life-threat-ening condition that requires urgent therapy. It is caused by a pericardial ac-cumulation of fluid, blood, pus, or gas that increases pericardial pressure and

assessment (Fig 6). First, both techniques are superior to echocardiography for de-tection of the amount and distribution of fluid accumulation. In general, a peri-cardial thickness greater than 4 mm is regarded as abnormal. However, because fluid can accumulate in the pericardial recesses, a relationship between the measured width of the pericardial space and the total volume of fluid is not clear cut. For instance, a pericardial space an-terior to the RV that is greater than 5 mm corresponds to a moderate effusion of 100–500 mL of fluid (40,70). Second, characterization of pericardial fluid can, to some extent, be achieved by measur-ing attenuation values on CT images and signal intensity on MR images. If the CT attenuation value is greater than that of water, then an effusion is more likely to be due to hemopericardium, malignancy, purulent exudates, or hypothyroid-relat-ed effusion (25). Low-attenuation peri-cardial effusions have been reported in cases of chylopericardium (15). Transu-dates typically manifest with low signal intensity on T1-weighted MR images and with high signal intensity on T2-weighted images. Exudates, having high protein and cell content, increase the rate of T1 relaxation (higher signal intensity) and

Figure 5

Figure 5: Congenital absence of pericardium. (a) Axial and (b) coronal T1-weighted spin-echo MR images (two heart beats/43, 90° flip angle, 1.4 3 2.0-mm in-plane resolution) show pronounced left-sided rotation and displacement of the heart, filling anterior portion of left hemithorax. The pericardial defect as such is usually difficult to visualize, but large defects such as this can be suspected on the basis of the abnormal location of the heart in the absence of other predisposing factors. (Images courtesy of Luigi Lovato, MD, S. Orsola Malpighi Hospital, Bologna, Italy.)



include iatrogenic posttraumatic peri-carditis after radiation therapy for breast cancer and mediastinal tumors and cardiac interventions such as car-diac surgery, percutaneous coronary in-terventions, pacemaker insertion, and catheter ablation (78). Pericarditis may occur early after infarction in approxi-mately 10% of patients and is typically

specific causes accounts for about 5% of unselected cases of pericarditis. A high prevalence of tuberculous-related peri-carditis is reported in developing coun-tries, and this disease accounts for up to 90% of cases when the disease is as-sociated with human immunodeficiency virus infection (74–77). In developed countries, leading causes of pericarditis

of contrast material into the azygos vein and inferior vena cava on contrast-en-hanced CT images (Movie 2 [online]) (72). Finally, cardiac tamponade should be differentiated from effusive-con-strictive pericarditis. In these patients with a pericardial effusion at the time of presentation, complaints are caused by a pathologic noncompliant pericar-dium rather than by the effusion itself. Thus, care should be taken to look for features of constriction, which can oc-cur transiently in the resolution phase, after pericardiocentesis or with orga-nized effusions (35).

Inflammatory PericarditisInflammation of the pericardium (peri-carditis or inflammatory pericarditis) manifests in many clinical settings and has a wide range of causes (35). Although the true incidence and prev-alence of pericarditis are difficult to measure, a prevalence of 1% in au-topsy studies suggests that pericarditis may frequently be subclinical (35). In North America and Western Europe, most cases of acute pericarditis are id-iopathic (80%–85%) (73). These cases are generally presumed to be viral. Ma-jor nonidiopathic causes include tuber-culosis, neoplasia, and systemic (gener-ally autoimmune) disease. Each of these

Figure 7

Figure 7: Acute cardiac tamponade due to ruptured aortic dissection. (a) Axial non–ECG-gated arterial phase CT image shows intimal flap in the ascending aorta extending into the aortic arch, corresponding to Stanford type A aortic dissection. No pericardial effusion is present. (b) Axial CT image acquired 60 seconds after contrast agent administration shows large high-attenuation pericardial effusion corresponding to mas-sive hemopericardium. Weakening of aortic wall with subsequent rupture has led to massive extravasation of iodinated contrast agent to the pericardial sac. The patient died immediately after CT angiography. (Image courtesy of G. F. Gualdi, MD, and C. Valentini, MD, DEA Policlinico Umberto I, Rome, Italy.)

Figure 6

Figure 6: Malignant pericardial effusion in patient with primary pulmonary adenocarcinoma. (a) Frontal chest radiograph shows massive cardiomegaly with retrocardiac opacity in right hemithorax (arrows). (b) Axial contrast-enhanced CT image confirms presence of large pericardial effusion. A centrally excavated mass in right lower lobe (arrow) and bilateral (paraneoplastic) pulmonary emboli (arrowheads) are also present. (c) Coronal contrast enhanced CT image shows extent of the pericardial effusion. Histologic examination of a pericardial biopsy showed presence of malignant cells, likely metastases from pulmonary adenocarcinoma.



administration. The attenuation of peri-cardial fluid may be similar to that of water or may be increased in case of exudative or purulent forms. Signs of cardiac tamponade may be present. In chronic forms of pericarditis, pericar-dial layers tend to be irregularly thick-ened and effusions may be loculated owing the presence of adhesions (20).

Similar morphologic findings can be shown on MR images when a combina-tion of T1-weighted and bright-blood se-quences is used (Fig 9, Movie 3 [online]). T2-weighted short tau inversion-recovery spin-echo MR imaging is helpful for vi-sualizing edema of the inflamed pericar-dial layers, while contrast-enhanced MR studies—in particular those obtained with the inversion-recovery gradient-echo sequences—are useful in depicting pericardial inflammation, as has been re-cently shown in several publications (Figs 8, 9) (3,20,21). Pericardial enhancement reflects inflammation at histologic exami-nation and is correlated with elevated in-flammatory markers. Contrast-enhanced MR imaging, moreover, allows evaluation of the extent of inflammation into the surrounding fat and adjacent myocardial tissue.

Because acute pericarditis is a main differential diagnosis of acute chest pain,

a variable amount of pericardial fluid is usually present, “dry pericarditis” (also known as pericarditis sicca) may found as well and is thought to produce the typical pericardial friction rub on auscul-tation. The natural history of acute peri-carditis is commonly benign, and most patients respond favorably to nonsteroi-dal anti-inflammatory drugs, although recurrent bouts of pericardial pain are found in up to 30% of patients. Pericar-ditis may progress to chronic sclerosing pericarditis, which is characterized by fibroblasts and collagen deposition and ultimately leads to end-stage chronic fibrosing pericarditis, which consists of a stiff pericardium that constricts the heart (constrictive pericarditis).

Although echocardiography is con-sidered the modality of choice if addi-tional imaging is needed in the diagno-sis of pericarditis or to guide diagnostic or therapeutic procedures (pericar-diocentesis or pericardial biopsy), the value of CT and MR imaging is probably underestimated owing to the ability of improved lesion visualization and char-acterization with these modalities. Im-aging findings depend on the pathologic substrate. On CT images, pericardial layers are thickened and show diffuse enhancement after contrast material

found in patients with transmural in-farction (“epistenocardiac pericarditis”) (Fig 8) (79). This condition should be differentiated from late postinfarc-tion pericarditis (Dressler syndrome). Whereas acute postinfarction pericar-ditis has a close temporal relation with the acute event due to the pericardial spread of infarct-related inflammation, Dressler syndrome has an autoimmune origin without a close temporal relation with myocardial infarction.

The manifestation of pericarditis can be acute, but it may manifest in subacute, recurrent, and chronic forms as well. The diagnosis of acute pericar-ditis is usually clinically suspected on the basis of a combination of typical and often severe chest pain, pericar-dial rub at physical examination, the presence of widespread ST elevation on the ECG trace. Diagnosis of acute pericarditis is supported by elevation of serologic markers of inflammation with or without evidence of infection (35). Symptoms are mainly related to the se-verity of pericardial inflammation. Histo-logically, the inflamed pericardial layers are composed of a highly vascularized granulation tissue with fibrin deposi-tion that may cause fibrinous adhesion of the pericardial layers (20). Although

Figure 8

Figure 8: Early postinfarction pericarditis in 73-year-old patient with recent anteroapical myocardial infarction. (a) Short-axis, (b) horizontal long-axis, and (c) ver-tical long-axis late gadolinium-enhanced MR images (repetition time msec/echo time msec/inversion time msec, 4.3/1.3/270) show strong myocardial enhancement in apical part of ventricular septum, LV anterior wall, and LV apex (arrows). Note presence of large centrally located zone of microvascular obstruction. In addition, prominent enhancement of entire pericardium is shown, reflecting pericardial inflammation (arrowheads). Although the patient had no specific complaints, C-reactive protein level was markedly increased (123 mg/L [1171.45 nmol/L]).



even acute forms of pericardial constric-tion have been described, this disease usually clinically manifests years after an initial pathologic trigger. The risk of constrictive pericarditis after acute peri-carditis is relatively low in viral or idio-pathic acute pericarditis (, 0.5%) but is relatively frequent in purulent and tu-berculous pericarditis, particularly in pa-tients with an incessant course and large pericardial effusions (85). The spectrum of causes of pericardial constriction has shifted over time from infectious causes (particularly tuberculous) to postirradi-ation and postoperative forms, which have now become the most frequent causes of the disease (86,87).

The diagnosis of constrictive pericar-ditis, even in the modern era, remains challenging. In clinical practice today, a combined approach is used to evaluate the morphologic pericardial (and car-diac) abnormalities in conjunction with assessment of the functional and hemo-dynamic consequences. Although this is commonly achieved by using a multimo-dality imaging approach, comprehensive MR imaging offers an almost complete appreciation of constrictive pericardi-tis, with exception of demonstration of

involvement can be suspected in the presence of atypical ECG changes asso-ciated with transient regional and global wall motion abnormalities and increase in the levels of cardiac enzymes.

MR imaging has today become the reference for diagnosis and follow-up of myocarditis (82,83). This technique allows appreciation of the spread of inflammation throughout the heart, as-sessment of disease activity, and eval-uation of the functional implications of the myocardial involvement. In particu-lar functional myocardial abnormalities may not be prominent and, therefore, often not recognized at echocardiogra-phy (44). Although patterns of myocar-dial enhancement similar to those seen on MR images have been described for CT images in patients with myocarditis, the role of CT in this diagnosis is cur-rently undefined (84).

Constrictive PericarditisConstrictive pericarditis reflects a condi-tion in which the compliance of the peri-cardium is decreased, which may result in impaired ventricular filling, severe diastolic dysfunction, and right heart failure (33,35). Although subacute and

as is observed in about 5% of patients admitted to emergency department, the radiologist should be familiar with this entity since an increasing number of cardiac CT scans are performed in the emergency setting to rule out significant coronary artery disease, or to triple rule out, in addition, aortic dissection and pulmonary embolism (73,80); alterna-tively, these patients undergo MR imag-ing primarily to exclude ischemia-related myocardial damage (77).

Acute pericarditis is often accompa-nied by some degree of myocarditis. In clinical practice, both pericarditis and myocarditis may coexist because they share common etiologic agents, mainly cardiotropic viruses (73). However, they are rarely present with equivalent in-tensity, giving rise to clinical syndromes that are mainly pericarditic or myocar-ditic. The term myopericarditis indicates a primarily pericarditic syndrome, while perimyocarditis indicates a primarily myocarditic syndrome. Recognition of associated myocarditis may be clinically relevant, and is a negative prognostic predictor in patients with pericarditis, often requiring hospitalization and a full etiologic search (75,81). Myocardial

Figure 9

Figure 9: Horizontal long-axis MR images of bacterial pericarditis in patient with a history of myocardial infarction and stent implantation in left anterior descend-ing coronary artery. (a) T1-weighted spin-echo MR image (one heartbeat/30, 90° flip angle, 1.4 3 2.0-mm in-plane resolution) shows moderate-sized pericardial effusion mainly along the right heart and thickening of the pericardial layers (arrows). (b) SSFP cine MR image (repetition time msec/echo time msec = 2.7/1.4, 55° flip angle, 1.4 3 2.0-mm in-plane resolution) allows differentiation of pericardial fluid and pericardial layers (arrows). Note presence of several fibrinous strands in the pericardium. (c) Late gadolinium-enhanced MR image (4.3/1.3, 260-msec inversion time) shows transmural myocardial enhancement in apical half of the inter-ventricular septum and LV apex (black arrows), reflecting healed transmural myocardial infarction. Prominent enhancement of thickened pericardial layers is also seen (white arrows), and a small left-sided pleural effusion is present. The effect of pericardial abnormalities on the heart can be well appreciated on Movie 3 (online).



pericarditis is characterized by a thin-ning of the chronically inflamed peri-cardium. In the study by Feng et al (3), pericardial thickness was significantly less in patients with persistent constric-tive pericarditis than in those with re-versible constrictive pericarditis (2 mm 6 1 vs 4 mm 6 1; P , .001). Similar results were reported by Zurick et al (21), who found a thinner pericardium in patients with end-stage constrictive pericarditis than in those with evidence of persistent chronic inflammation (3.1 mm 6 1.7 vs 4.6 mm 6 1.4; P = .021). As mentioned previously, both studies showed the value of contrast-enhanced MR imaging in depicting residual in-flammation in patients with constric-tive pericarditis (20). These patients have “reversible” or “transient” forms of constrictive pericarditis that respond to optimized anti-inflammatory treatment. This underscores the fact that not all patients with constrictive pericarditis need to undergo pericardiectomy (89). It should also be noted that the degree of pericardial thickening is only weakly related to the degree of cardiac constric-tion (35).

Pericardial calcifications are consid-ered an important sign of constrictive

cine imaging. In the end stage of chronic fibrosing forms of constrictive pericardi-tis there is no enhancement after con-trast material administration, whereas pericardial enhancement is suggestive of residual inflammation (3,20,21). Differ-entiation between pericardial thickening and effusion is usually straightforward on MR images.

Although pericardial thickness is traditionally used as an important cri-terion for constrictive pericarditis (ie, pericardial thickness 2 mm is normal, . 4 mm is suggestive of pericardial con-striction in patients with the appropriate clinical presentation, and . 5–6 mm is highly specific for constriction [52–54]), this concept has been increasingly chal-lenged and should be put in perspective with regard to recent findings. Although fibrosis with or without calcification is nearly always present at pathologic ex-amination (96%) in cases of pericardial constriction, the maximal pericardial thickness shows a wide range (1–17 mm; mean, 4 mm) with up to 20% of patients showing a normal (2-mm) thickness (Fig 11, Movie 4 [online]) (54,74).

The findings from two recent studies (3,21) indicate that evolution toward an end-stage irreversible chronic fibrosing

pericardial calcifications. First, all other causes of (right) heart failure (eg, pulmo-nary hypertension, severe tricuspid in-sufficiency, myocardial infarction) should be excluded. Second, it should be deter-mined whether the pericardium is caus-ing constriction, thereby impeding car-diac filling. Third, the optimal treatment needs to be determined (eg, pericardial stripping vs medical treatment) (40). In this workup, which is tailored to each pa-tient’s specific needs, the differentiation of constrictive pericarditis from restric-tive cardiomyopathy is crucial.

Although, to our knowledge, there are no prospective studies in which medical and surgical management are compared, pericardiectomy is the ac-cepted treatment to improve patient he-modynamics. Long-term survival after pericardiectomy is related to the under-lying cause. In particular, postoperative survival in postirradiation constrictive pericarditis is poor (86). It is also im-portant to mention that a number of patients with constrictive pericarditis may not benefit from pericardiectomy because of associated myocardial com-pliance abnormalities (ie, restrictive cardiomyopathy), myocardial atrophy after prolonged constriction, or other myocardial processes, thus emphasizing the need for a complete evaluation of the heart and pericardium (88).

Morphologic abnormalities.—The typical morphologic manifestation of con-strictive pericarditis is that of generalized thickening of the pericardium with or without calcifications. Abnormalities are usually most pronounced over the right heart (RV and anterior atrioventricular groove), and pericardial delineation is often irregular (Fig 10). The underly-ing cardiac cavities may be constricted by the abnormal pericardium, having a flattened or tubular-shaped appearance (Fig 10). Indirectly, as a result of the in-creased cardiac filling pressures, unilat-eral or bilateral atrial enlargement, dila-tation of caval and hepatic veins, pleural effusion, and ascites are encountered. These abnormalities may be easily rec-ognized on CT images. The thickened fibrotic and/or calcified pericardium has low signal intensity on T1-weighted and T2-weighted spin-echo MR images and at

Figure 10

Figure 10: Calcified constrictive pericarditis with constriction and tubular flattening of the RV. (a) Axial T1-weighted spin-echo MR image (one heartbeat/30, 90° flip angle, 1.4 3 2.0-mm in-plane resolution) shows irregularly thickened appearance of pericardium, primarily along laterobasal part of RV and LV (arrows) with subsequent flattening of RV cavity (∗). (b) Short-axis T1-weighted spin-echo MR image (one heartbeat/30, 90° flip angle, 1.4 3 2.0 mm in-plane resolution) shows extent of pericardial abnormalities along inferior part of the heart (arrows).



forms include localized and occult forms of constrictive pericarditis (93).

Functional and hemodynamic conse-quences.—Encasement of the heart by a noncompliant, rigid pericardium leads to (a) dissociation between intracardiac and intrathoracic pressure, which iso-lates the heart from normal respiratory changes in intrathoracic pressure; (b) increased ventricular coupling; and (c) increased cardiac filling pressures with pressure equalization in all four cardiac chambers. Although the hemodynamic consequences are assessed in clinical practice by means of echocardiography and cardiac catheterization, MR imaging has great potential, because information regarding pericardial-cardiac morphol-ogy and tissue characteristics can be merged with functional hemodynamic information (8,35,75). Phase-contrast MR imaging of the tricuspid valve in-flow shows a restrictive filling pattern of enhanced early filling and decreased or absent late filling, depending on the degree of pericardial constriction and increased filling pressures. Also, flow in the inferior vena cava shows restrictive physiology with diminished or absent forward—or even reversed—systolic flow, increased early diastolic forward flow, and late reversed flow. Constrictive

complications such as the intramyocar-dial extent of the fibrocalcific process, which may hamper the success of a peri-cardiectomy (Fig 12; Movies 5, 6 [on-line]). Preoperative CT may be useful to provide a detailed depiction of both the severity of thickening and the presence and location of calcifications, thus allow-ing better surgical planning and stratifi-cation of procedural risk (2). Pericardial calcifications, however, are less common these days than in the past, which is probably related to the decrease in rates of tuberculosis and the increase in iatro-genic causes of constriction. The authors of two recent studies (74,90) reported pericardial calcifications in 27% and 28% of patients with histologically con-firmed constrictive pericarditis. Since tuberculosis was ruled out in nearly all cases in these patient populations, these results suggest that the occurrence of calcification should be considered a non-specific response to chronic inflamma-tion (90).

Finally, patients clinically suspected of having constrictive pericarditis may show atypical manifestation of pericardial ab-normalities such as effusive-constrictive pericarditis manifesting with a combina-tion of symptoms related to tamponade and constriction (91,92). Other atypical

pericarditis on chest radiographs. To-day, CT is the most appropriate tech-nique to depict even minute amounts of pericardial calcium and to visualize

Figure 11

Figure 11: Minimally thickened constrictive pericarditis. (a) Horizontal long-axis T1-weighted spin-echo MR image (one heartbeat/30, 90° flip angle, 1.4 3 2.0-mm in-plane resolution) shows minimally thickened pericardium (2.5 mm) along right side of the heart and laterobasally on the left (arrows). There is mild constriction of RV, and both atria appear enlarged. (b, c) Short-axis real-time SSFP cine MR images (2.3/1.16, 50° flip angle, 2.8 3 2.8-mm in-plane resolution) obtained during (b) inspiration and (c) expiration show inspiratory septal inversion during early ventricular filling (arrow in b), and an enhanced right-sided septal shift during expira-tion (arrow in c). Total septal shift measured 28% (see Movie 4 [online]).

Figure 12

Figure 12: Extensive pericardial calcifications focally invading myocardium in a patient with history of radiation therapy for Hodgkin disease. Short-axis contrast-enhanced CT image shows thickened and heavily calcified pericardium (arrows) with focal fanlike invasion of myocardium (arrowheads) along lateral LV wall and RV outflow tract. Pericardial and myocardial calcifications can be well appreciated on Movie 5 (online), while Movie 6 (online) shows three-dimensional spread of pericardial calcifications with regard to the beating heart.



and lymphoma. Benign tumors in-clude fibroma, teratoma, hemangioma, and lipoma and are extremely rare. Malignant mesothelioma is the most common primary lesion and is usu-ally characterized by the presence of hemorrhagic effusion with pericardial nodular lesions or plaques (Fig 17)

time, is of great help in clinical routine practice. Two extreme examples illus-trate its usefulness. On the one hand, a pericardium may have a normal or near-normal thickness but cause severe constriction and thus lead to marked in-crease of ventricular coupling, while on the other hand a thick pericardium may be not constrictive and can show normal ventricular coupling at real-time cine im-aging. Without this essential information one risks wrongly categorizing patients. Finally, MR tagging enables one to better appreciate the fibrotic fusion of the peri-cardial layers and the extent of the fibro-calcific process into the underlying myo-cardium (Fig 14; Movies 7, 8 [online]) (48). Whereas in normal pericardium, tag lines rapidly become discontiguous during the cardiac cycle due to shear motion of the inner and outer pericar-dial layer, persistence of these tag lines is indicative of fibrotic fusion.

Pericardial (Pseudo)-MassesPericardial masses represent a hetero-geneous group of cystic lesions, hema-tomas, complex organized effusions, masslike structures, and primary and secondary malignancies that affect or involve the pericardium. CT and MR imaging are often necessary in the di-agnostic workup, because they provide an accurate description of the pericar-dial abnormalities and the relationship to the surrounding structures and facil-itate understanding of the underlying cause, establishment of the diagnosis and differential diagnosis, and assess-ment of complications such as cardiac tamponade (Fig 15, Movie 8 [online]). Because of the superior anatomic de-tail, these techniques allow one to eas-ily differentiate “true” pericardial mass-es from “pseudo” masses—for example, the presence of abundant epicardial fat, which was originally misinterpreted at transthoracic echocardiography (Fig 14).

While primary pericardial tu-mors are extremely rare, secondary involvement has been described in up to 10%–12% of patients (Fig 16) (98–100). Primary tumors are more frequently malignant and include meso-thelioma, angiosarcoma, liposarcoma,

pericarditis, in contrast to restrictive cardiomyopathy, is typically character-ized by a strong respiratory-related var-iation in cardiac filling (ie, enhanced RV filling on inspiration, enhanced LV filling on expiration). Real-time phase-contrast MR imaging is an attractive alternative to Doppler echocardiography to assess the effects of respiration on cardiac fill-ing (19).

The lack of pericardial stretch in constrictive pericarditis leads to in-creased ventricular coupling, which is characterized by septal flattening or inversion (“septal bounce”) at early diastolic ventricular filling (Movie 3 [online]) (16,36,94,95). Because of the dissociation between intrathoracic and intracardiac pressure, this pattern is strongly influenced by respiration (35,95). Septal abnormalities are most pronounced at the onset of inspiration and rapidly fade away, while at onset of expiration an opposite (right-sided) septal shift occurs. Abnormalities are most pronounced in the basal septum, leading to an S-like septal motion on a horizontal long-axis view (Movie 3 [online]).

We exploited the clinical potential of novel real-time cine sequences to study the effects of free breathing on ventricular coupling (17). Patients with constrictive pericarditis showed the typical respiratory pattern of septal abnormalities, while those with restric-tive cardiomyopathy showed a pattern similar to that of healthy volunteers (Figs 11, 13; Movie 4 [online]). Quan-tification of the total septal excursion between inspiration and expiration was very helpful in differentiating between constrictive pericarditis and restrictive cardiomyopathy (Fig 13) (18). More-over, patients with inflammatory peri-carditis also often show increased sep-tal excursion, most likely related to the decreased compliance of the inflamed pericardial layers. The added value of real-time imaging during free breathing to evaluate the hemodynamic effect of the pericardium on cardiac filling has become accepted as an essential part of an MR examination (96,97).

This sequence, which takes approxi-mately 10 seconds of extra measurement

Figure 13

Figure 13: Box-and-whiskers plots of total excur-sion of interventricular septum between inspiration and expiration at early ventricular filling. Data were derived by using real-time SSFP cine MR imaging. In patients with constrictive pericarditis (CP), the ex-cursion is significantly increased, as compared with normal excursion and that in patients with restrictive cardiomyopathy (RCM). Patients with inflammatory pericarditis (IP) also show increased excursion, most likely owing to decreased compliance of inflamed pericardial layers. (Adapted, with permission, from reference 18.)

Figure 14

Figure 14: Diffuse pericardial thickening in pa-tient with clinical presentation of cardiac decompen-sation. Horizontal long-axis T1-weighted spin-echo MR image (one heartbeat/30, 90° flip angle, 1.4 3 2.0-mm in-plane resolution) shows thickened pericardium (maximal, 6 mm) mainly along right side of the heart (arrows). Movies 7 and 8 (online) show dynamic evaluation of the pericardium.



is needed to establish the definitive diagnosis.

Other entities that have a masslike appearance are pericardial hematoma and pericardial gossypiboma. The CT and MR imaging characteristics of a pericardial hematoma depend on the age of the collection (41,100). The pos-sibility of a pericardial gossypiboma or foreign body granuloma (eg, surgical sponge) should always be considered in patients with a history of cardiac surgery. A final entity to be mentioned is pericardial fat necrosis. This benign entity of unknown cause typically man-ifests with sudden onset of chest pain, mimicking other diseases such as myo-cardial infarction and pulmonary em-bolism. The typical CT appearance is a lesion of fat attenuation surrounded by the increased attenuation of the an-terior mediastinal paracardiac fat adja-cent to the pericardium (102).

Conclusion

Although pericardial disease often rep-resents a diagnostic challenge, CT and MR imaging have notably contributed in the demonstration of pericardial path-ologic conditions and greatly improved our understanding of this enigmatic part of the heart. Since pericardial dis-eases have substantial morbidity and

(eg, lymphoma), or heart (eg, angiosar-coma); (b) hematologic spread (eg, ma-lignant melanoma, lymphoma, breast cancer); or, rarely, (c) venous exten-sion (usually renal cell or hepatocellular carcinomas). Malignant invasion can be seen as focal interruption of pericardial layers invading the pericardium from outside (eg, lung cancer) or inside (eg, angiosarcoma). The latter generally arise from the right atrium and rapidly and massively invade the pericardium. In most pericardial tumors, biopsy

(101). As a general rule, pericardial masses are characterized by a variable amount of effusion that is dispropor-tionally greater than the size of the solid lesion, is very often hemorrhagic, and is, therefore, characterized by in-homogeneous high signal intensity on both T1-weighted and T2-weighted MR images and high attenuation on unen-hanced CT images. Secondary pericar-dial involvement may occur by means of (a) direct invasion of neoplasms originating in the lung, mediastinum

Figure 16

Figure 16: Pericardial metastasis with myocardial invasion in a patient with Ewing sarcoma. Short-axis T1-weighted spin-echo MR images (one heartbeat/30, 90° flip angle, 1.4 3 2.0-mm in-plane resolution) obtained (a) before and (b) after administration of gadolinium chelate shows large (53 3 46 3 33-mm) pericardial mass (arrows) with diffuse invasion and thickening of the basal LV inferolateral wall (arrowheads). After contrast agent administration, marked inhomogeneous enhancement of the metastatic mass is noted. A moderate sized pericardial effusion is also seen (maximal width, 19 mm).

Figure 15

Figure 15: Hematoma in epicardial fat in a patient with acute lateral transmural myocardial infarction. (a) Short-axis and (b) horizontal long-axis SSFP cine MR images (2.7/1.4, 55° flip angle, 1.4 3 2.0-mm in-plane resolution) and (c) horizontal long-axis late gadolinium-enhanced MR image (4.3/1.3, 270-msec inversion time) show thinning of entire LV lateral wall, with inhomogeneous transmural myocardial enhancement after contrast agent adminis-tration (arrows). Diffuse edema of epicardial fat along lateral and inferior heart border is present, with slightly hyperintense appearance, as compared with adjacent myocardium (arrowheads). Moderate to severe pericardial effusion is also noted. Location and nature of abnormalities (hematoma, effusion) can be well defined on MR images (see also Movie 9 [online]).



cardium: normal and pathologic findings. Radiology 1984;150(2):469–474.

15. Wang ZJ, Reddy GP, Gotway MB, Yeh BM, Hetts SW, Higgins CB. CT and MR imag-ing of pericardial disease. RadioGraphics 2003;23(Spec No):S167–S180.

16. Giorgi B, Mollet NR, Dymarkowski S, Rademakers FE, Bogaert J. Clinically suspected constrictive pericarditis: MR imaging assessment of ventricular sep-tal motion and configuration in patients and healthy subjects. Radiology 2003; 228(2):417–424.

17. Francone M, Dymarkowski S, Kalantzi M, Bogaert J. Real-time cine MRI of ventric-ular septal motion: a novel approach to assess ventricular coupling. J Magn Reson Imaging 2005;21(3):305–309.

18. Francone M, Dymarkowski S, Kalantzi M, Rademakers FE, Bogaert J. Assessment of ventricular coupling with real-time cine MRI and its value to differentiate constric-tive pericarditis from restrictive cardiomy-opathy. Eur Radiol 2006;16(4):944–951.

19. Thavendiranathan P, Verhaert D, Walls MC, et al. Simultaneous right and left heart real-time, free-breathing CMR flow quantification identifies constrictive physiology. JACC Car-diovasc Imaging 2012;5(1):15–24.

20. Taylor AM, Dymarkowski S, Verbeken EK, Bogaert J. Detection of pericardial in-flammation with late-enhancement cardiac magnetic resonance imaging: initial results. Eur Radiol 2006;16(3):569–574.

21. Zurick AO, Bolen MA, Kwon DH, et al. Pericardial delayed hyperenhancement with CMR imaging in patients with con-strictive pericarditis undergoing surgical

6. Ellis K, King DL. Pericarditis and pericar-dial effusion. Radiologic and echocardio-graphic diagnosis. Radiol Clin North Am 1973;11(2):393–413.

7. Candell-Riera J, García del Castillo H, Per-manyer-Miralda G, Soler-Soler J. Echocar-diographic features of the interventricular septum in chronic constrictive pericarditis. Circulation 1978;57(6):1154–1158.

8. Hatle LK, Appleton CP, Popp RL. Dif-ferentiation of constrictive pericarditis and restrictive cardiomyopathy by Dop-pler echocardiography. Circulation 1989; 79(2):357–370.

9. D’Cruz IA, Constantine A. Problems and pitfalls in the echocardiographic assess-ment of pericardial effusion. Echocardiog-raphy 1993;10(2):151–166.

10. Verhaert D, Gabriel RS, Johnston D, Lytle BW, Desai MY, Klein AL. The role of mul-timodality imaging in the management of pericardial disease. Circ Cardiovasc Imag-ing 2010;3(3):333–343.

11. Ling LH, Oh JK, Tei C, et al. Pericardial thickness measured with transesophageal echocardiography: feasibility and poten-tial clinical usefulness. J Am Coll Cardiol 1997;29(6):1317–1323.

12. Houang MT, Arozena X, Shaw DG. Dem-onstration of the pericardium and pericar-dial effusion by computed tomography. J Comput Assist Tomogr 1979;3(5):601–603.

13. Pugatch RD, Braver JH, Robbins AH, Faling LJ. CT diagnosis of pericardial cysts. AJR Am J Roentgenol 1978;131(3):515–516.

14. Stark DD, Higgins CB, Lanzer P, et al. Magnetic resonance imaging of the peri-

mortality, both techniques have an in-creasingly important role in decision making, particularly in determination of the optimal treatment for patients with constrictive pericarditis.

Disclosures of Conflicts of Interest: J.B. Finan-cial activities related to the present article: none to disclose. Financial activities not related to the present article: received funds for travel from GE Healthcare. Other relationships: none to disclose. M.F. No relevant conflicts of interest to disclose.

References 1. Maisch B, Seferović PM, Ristić AD, et al.

Guidelines on the diagnosis and manage-ment of pericardial diseases executive sum-mary; The Task force on the diagnosis and management of pericardial diseases of the European society of cardiology. Eur Heart J 2004;25(7):587–610.

2. Yared K, Baggish AL, Picard MH, Hoff-mann U, Hung J. Multimodality imaging of pericardial disease. JACC Cardiovasc Imag-ing 2010;3(6):650–660.

3. Feng DL, Glockner J, Kim K, et al. Cardiac magnetic resonance imaging pericardial late gadolinium enhancement and elevated in-flammatory markers can predict the revers-ibility of constrictive pericarditis after anti-inflammatory medical therapy: a pilot study. Circulation 2011;124(17):1830–1837.

4. Feigenbaum H, Waldhausen JA, Hyde LP. Ultrasound diagnosis of pericardial effu-sion. JAMA 1965;191:711–714.

5. Stein L, Shubin H, Weil MH. Recognition and management of pericardial tampon-ade. JAMA 1973;225(5):503–506.

Figure 17

Figure 17: Pericardial mesothelioma in a 42-year-old man with history of pericardial effusion (presumably postviral) complicated by constriction. CT scans were obtained without ECG gating. (a) Before pericardiectomy, diffuse pericardial thickening can be seen (arrows). (b, c) Four years later, the patient presented with symptoms of right heart failure similar to those before surgery. Follow-up CT scans show several nodular masses diffusely spread around in the pericardial sac (arrows). Biopsy showed epitheloid pericardial mesothelioma. (Image courtesy of Scott H. Yang, MD, PhD, Santa Rosa, Calif.)



function by cardiovascular magnetic reso-nance. Am Heart J 2002;144(2):198–205.

51. Sechtem U, Tscholakoff D, Higgins CB. MRI of the normal pericardium. AJR Am J Roentgenol 1986;147(2):239–244.

52. Bogaert J, Duerinckx AJ. Appearance of the normal pericardium on coronary MR angiograms. J Magn Reson Imaging 1995;5(5):579–587.

53. Delille JP, Hernigou A, Sene V, et al. Maximal thickness of the normal human pericardium assessed by electron-beam computed tomography. Eur Radiol 1999; 9(6):1183–1189.

54. Talreja DR, Edwards WD, Danielson GK, et al. Constrictive pericarditis in 26 patients with histologically normal pericardial thick-ness. Circulation 2003;108(15):1852–1857.

55. Ferrans VJ, Ishihara T, Roberts WC. Anatomy of the pericardium. In: Reddy PS, Leon DF, Shaver JA, eds. Pericardial disease. New York, NY: Raven, 1982; 77–92.

56. Hort W, Braeun H. Studies on the di-mensions, wall thickness and microscopic structure of the pericardium in normal and pathological conditions [in German]. Arch Kreislaufforsch 1962;38:1–22.

57. Im JG, Rosen A, Webb WR, Gamsu G. MR imaging of the transverse sinus of the pericardium. AJR Am J Roentgenol 1988;150(1):79–84.

58. McMurdo KK, Webb WR, von Schulthess GK, Gamsu G. Magnetic resonance imaging of the superior pericardial recesses. AJR Am J Roentgenol 1985;145(5):985–988.

59. Solomon SL, Brown JJ, Glazer HS, Mirow-itz SA, Lee JKT. Thoracic aortic dissection: pitfalls and artifacts in MR imaging. Radi-ology 1990;177(1):223–228.

60. Akiba T, Marushima H, Masubuchi M, Kobayashi S, Morikawa T. Small symp-tomatic pericardial diverticula treated by video-assisted thoracic surgical resec-tion. Ann Thorac Cardiovasc Surg 2009; 15(2):123–125.

61. Feigin DS, Fenoglio JJ, McAllister HA, Madewell JE. Pericardial cysts. A radiolog-ic-pathologic correlation and review. Radi-ology 1977;125(1):15–20.

62. Jeung MY, Gasser B, Gangi A, et al. Imaging of cystic masses of the mediastinum. Radio-Graphics 2002;22(Spec No):S79–S93.

63. Drury NE, De Silva RJ, Hall RM, Large SR. Congenital defects of the pericardium. Ann Thorac Surg 2007;83(4):1552–1553.

64. Yamano T, Sawada T, Sakamoto K, Naka-mura T, Azuma A, Nakagawa M. Magnetic resonance imaging differentiated partial

36. Nishimura RA. Constrictive pericarditis in the modern era: a diagnostic dilemma. Heart 2001;86(6):619–623.

37. Kodama F, Fultz PJ, Wandtke JC. Com-paring thin-section and thick-section CT of pericardial sinuses and recesses. AJR Am J Roentgenol 2003;181(4):1101–1108.

38. Bull RK, Edwards PD, Dixon AK. CT di-mensions of the normal pericardium. Br J Radiol 1998;71(849):923–925.

39. Ghersin E, Lessick J, Litmanovich D, et al. Septal bounce in constrictive pericarditis. Diagnosis and dynamic evaluation with multidetector CT. J Comput Assist Tomogr 2004;28(5):676–678.

40. Bogaert J, Francone M. Cardiovascular magnetic resonance in pericardial diseases. J Cardiovasc Magn Reson 2009;11:14.

41. Bogaert J, Dymarkowski S, Taylor AM, Muthurangu V. Clinical cardiac MRI. 2nd ed. Berlin, Germany: Springer, 2012.

42. Misselt AJ, Harris SR, Glockner J, Feng D, Syed IS, Araoz PA. MR imaging of the peri-cardium. Magn Reson Imaging Clin N Am 2008;16(2):185–199, vii.

43. Francone M, Carbone I, Agati L, et al. Utility of T2-weighted short-tau inversion recovery (STIR) sequences in cardiac MRI: an overview of clinical applications in isch-aemic and non-ischaemic heart disease. Radiol Med (Torino) 2011;116(1):32–46.

44. Yelgec NS, Dymarkowski S, Ganame J, Bo-gaert J. Value of MRI in patients with a clinical suspicion of acute myocarditis. Eur Radiol 2007;17(9):2211–2217.

45. Simonetti OP, Kim RJ, Fieno DS, et al. An improved MR imaging technique for the visualization of myocardial infarction. Ra-diology 2001;218(1):215–223.

46. Klein C, Graf K, Fleck E, Nagel E. Im-ages in cardiovascular medicine. Acute fibrinous pericarditis assessed with mag-netic resonance imaging. Circulation 2003;107(13):e82.

47. Bogaert J, Taylor AM, Van Kerkhove F, Dymarkowski S. Use of inversion recovery contrast-enhanced MRI for cardiac imag-ing: spectrum of applications. AJR Am J Roentgenol 2004;182(3):609–615.

48. Kojima S, Yamada N, Goto Y. Diagnosis of constrictive pericarditis by tagged cine magnetic resonance imaging. N Engl J Med 1999;341(5):373–374.

49. Rademakers FE, Bogaert J. Cardiac dys-function in heart failure with normal ejec-tion fraction: MRI measurements. Prog Cardiovasc Dis 2006;49(3):215–227.

50. Paelinck BP, Lamb HJ, Bax JJ, Van der Wall EE, de Roos A. Assessment of diastolic

pericardiectomy: a case series with his-topathological correlation. JACC Cardio-vasc Imaging 2011;4(11):1180–1191.

22. Achenbach S, Raggi P. Imaging of coronary atherosclerosis by computed tomography. Eur Heart J 2010;31(12):1442–1448.

23. American College of Cardiology Foundation Task Force on Expert Consensus Documents, Mark DB, Berman DS, et al. ACCF/ACR/ AHA/NASCI/SAIP/SCAI/SCCT 2010 expert consensus document on coronary computed tomographic angiography: a report of the American College of Cardiology Foundation Task Force on Expert Consensus Docu-ments. Circulation 2010;121(22):2509–2543.

24. Maffei E, Messalli G, Martini C, et al. Left and right ventricle assessment with Car-diac CT: validation study vs. Cardiac MR. Eur Radiol 2012;22(5):1041–1049.

25. O’Leary SM, Williams PL, Williams MP, et al. Imaging the pericardium: appear-ances on ECG-gated 64-detector row car-diac computed tomography. Br J Radiol 2010;83(987):194–205.

26. Todd RB, ed. Heart. In: Cyclopaedia of anatomy and physiology. Vol 2. London, England: Sherwood, Gilbert, and Piper, 1835; 597–598.

27. Spodick DH. Macrophysiology, microphysiol-ogy, and anatomy of the pericardium: a syn-opsis. Am Heart J 1992;124(4):1046–1051.

28. Groell R, Schaffler GJ, Rienmueller R. Peri-cardial sinuses and recesses: findings at electrocardiographically triggered electron-beam CT. Radiology 1999;212(1):69–73.

29. Rabkin SW. Epicardial fat: properties, function and relationship to obesity. Obes Rev 2007;8(3):253–261.

30. Iacobellis G, Corradi D, Sharma AM. Epicardial adipose tissue: anatomic, bio-molecular and clinical relationships with the heart. Nat Clin Pract Cardiovasc Med 2005;2(10):536–543.

31. Sacks HS, Fain JN. Human epicardial adipose tissue: a review. Am Heart J 2007;153(6):907–917.

32. Truong MT, Erasmus JJ, Gladish GW, et al. Anatomy of pericardial recesses on multidetector CT: implications for on-cologic imaging. AJR Am J Roentgenol 2003;181(4):1109–1113.

33. Spodick DH. The pericardium: a comprehen-sive textbook. New York, NY: Dekker, 1997.

34. Ivens EL, Munt BI, Moss RR. Pericardial disease: what the general cardiologist needs to know. Heart 2007;93(8):993–1000.

35. Troughton RW, Asher CR, Klein AL. Peri-carditis. Lancet 2004;363(9410):717–727.



91. Hancock EW. A clearer view of effusive-constrictive pericarditis. N Engl J Med 2004;350(5):435–437.

92. Sagristà-Sauleda J, Angel J, Sánchez A, Permanyer-Miralda G, Soler-Soler J. Effu-sive-constrictive pericarditis. N Engl J Med 2004;350(5):469–475.

93. Hancock EW. Differential diagnosis of re-strictive cardiomyopathy and constrictive pericarditis. Heart 2001;86(3):343–349.

94. Santamore WP, Bartlett R, Van Buren SJ, Dowd MK, Kutcher MA. Ventricular cou-pling in constrictive pericarditis. Circula-tion 1986;74(3):597–602.

95. Hurrell DG, Nishimura RA, Higano ST, et al. Value of dynamic respiratory changes in left and right ventricular pressures for the diagnosis of constrictive pericarditis. Circulation 1996;93(11):2007–2013.

96. Hundley WG, Bluemke D, Bogaert JG, et al. Society for Cardiovascular Magnetic Resonance guidelines for reporting cardio-vascular magnetic resonance examinations. J Cardiovasc Magn Reson 2009;11:5.

97. American College of Cardiology Foun-dation Task Force on Expert Consensus Documents, Hundley WG, Bluemke DA, et al. ACCF/ACR/AHA/NASCI/SCMR 2010 expert consensus document on cardiovas-cular magnetic resonance: a report of the American College of Cardiology Founda-tion Task Force on Expert Consensus Doc-uments. Circulation 2010;121(22):2462–2508.

98. Syed IS, Feng D, Harris SR, et al. MR im-aging of cardiac masses. Magn Reson Imag-ing Clin N Am 2008;16(2):137–164, vii.

99. Fussen S, De Boeck BW, Zellweger MJ, et al. Cardiovascular magnetic resonance imaging for diagnosis and clinical man-agement of suspected cardiac masses and tumours. Eur Heart J 2011;32(12):1551–1560.

100. Maksimović R, Dill T, Seferović PM, et al. Magnetic resonance imaging in pericar-dial diseases. Indications and diagnostic value. Herz 2006;31(7):708–714.

101. Grebenc ML, Rosado de Christenson ML, Burke AP, Green CE, Galvin JR. Primary cardiac and pericardial neoplasms: radio-logic-pathologic correlation. RadioGraphics 2000;20(4):1073–1103; quiz 1110–1111, 1112.

102. Hernandez D, Galimany J, Pernas JC, Llauger J. Case 170: Pericardial fat ne-crosis. Radiology 2011;259(3):919–922.

78. Imazio M, Brucato A, Trinchero R, Adler Y. Diagnosis and management of peri-cardial diseases. Nat Rev Cardiol 2009; 6(12):743–751.

79. Doulaptsis C, Goetschalckx K, Masci PG, Florian A, Janssens S, Bogaert J. Assess-ment of early post-infarction pericardial injury using cardiac magnetic resonance (CMR). JACC Cardiovasc Imaging (in press).

80. Halpern EJ. Triple-rule-out CT angiography for evaluation of acute chest pain and pos-sible acute coronary syndrome. Radiology 2009;252(2):332–345.

81. Spodick DH. Risk prediction in pericar-ditis: who to keep in hospital? Heart 2008;94(4):398–399.

82. American College of Cardiology Founda-tion Task Force on Expert Consensus Doc-uments, Hundley WG, Bluemke DA, et al. ACCF/ACR/AHA/NASCI/SCMR 2010 ex-pert consensus document on cardiovascular magnetic resonance: a report of the Ameri-can College of Cardiology Foundation Task Force on Expert Consensus Documents. J Am Coll Cardiol 2010;55(23):2614–2662.

83. Cocker M, Friedrich MG. Cardiovascular magnetic resonance of myocarditis. Curr Cardiol Rep 2010;12(1):82–89.

84. Brett NJ, Strugnell WE, Slaughter RE. Acute myocarditis demonstrated on CT coronary angiography with MRI correlation. Circ Cardiovasc Imaging 2011;4(3):e5–e6.

85. Imazio M, Brucato A, Maestroni S, et al. Risk of constrictive pericarditis af-ter acute pericarditis. Circulation 2011; 124(11):1270–1275.

86. Bertog SC, Thambidorai SK, Parakh K, et al. Constrictive pericarditis: etiology and cause-specific survival after pericardiectomy. J Am Coll Cardiol 2004;43(8):1445–1452.

87. Ling LH, Oh JK, Schaff HV, et al. Con-strictive pericarditis in the modern era: evolving clinical spectrum and impact on outcome after pericardiectomy. Circulation 1999;100(13):1380–1386.

88. Schwefer M, Aschenbach R, Heidemann J, Mey C, Lapp H. Constrictive pericarditis, still a diagnostic challenge: comprehensive review of clinical management. Eur J Car-diothorac Surg 2009;36(3):502–510.

89. Haley JH, Tajik AJ, Danielson GK, Schaff HV, Mulvagh SL, Oh JK. Transient constric-tive pericarditis: causes and natural history. J Am Coll Cardiol 2004;43(2):271–275.

90. Ling LH, Oh JK, Breen JF, et al. Calcific constrictive pericarditis: is it still with us? Ann Intern Med 2000;132(6):444–450.

from complete absence of the left pericar-dium in a case of leftward displacement of the heart. Circ J 2004;68(4):385–388.

65. Glover LB, Barcia A, Reeves TJ. Congen-ital absence of the pericardium. A review of the literature with demonstration of a previously unreported fluoroscopic finding. Am J Roentgenol Radium Ther Nucl Med 1969;106(3):542–549.

66. Amplatz K, Moller JH. Radiology of con-genital heart disease. St Louis, Mo: Mosby-Year Book, 1993.

67. Abbas AE, Appleton CP, Liu PT, Sweeney JP. Congenital absence of the pericardium: case presentation and review of literature. Int J Cardiol 2005;98(1):21–25.

68. Scheuermann-Freestone M, Orchard E, Francis J, et al. Images in cardiovascu-lar medicine. Partial congenital absence of the pericardium. Circulation 2007; 116(6):e126–e129.

69. Psychidis-Papakyritsis P, de Roos A, Kroft LJM. Functional MRI of congenital absence of the pericardium. AJR Am J Roentgenol 2007;189(6):W312–W314.

70. Frank H, Globits S. Magnetic resonance imaging evaluation of myocardial and peri-cardial disease. J Magn Reson Imaging 1999;10(5):617–626.

71. Spodick DH. Acute cardiac tamponade. N Engl J Med 2003;349(7):684–690.

72. Restrepo CS, Lemos DF, Lemos JA, et al. Imaging findings in cardiac tampon-ade with emphasis on CT. RadioGraphics 2007;27(6):1595–1610.

73. Imazio M, Trinchero R. Myopericarditis: Etiology, management, and prognosis. Int J Cardiol 2008;127(1):17–26.

74. Oh KY, Shimizu M, Edwards WD, Taz-elaar HD, Danielson GK. Surgical pathol-ogy of the parietal pericardium: a study of 344 cases (1993-1999). Cardiovasc Pathol 2001;10(4):157–168.

75. Little WC, Freeman GL. Pericardial disease. Circulation 2006;113(12):1622–1632. [Published correction appears in Circulation 2007;115(15):e406.]

76. Restrepo CS, Diethelm L, Lemos JA, et al. Cardiovascular complications of human immunodeficiency virus infection. Radio-Graphics 2006;26(1):213–231.

77. Dall’Armellina E, Karamitsos TD, Neu-bauer S, Choudhury RP. CMR for char-acterization of the myocardium in acute coronary syndromes. Nat Rev Cardiol 2010;7(11):624–636.