Complicated Pericarditis

Listen to this manuscript’s

audio summary by

JACC Editor-in-Chief

Dr. Valentin Fuster.

J O U R N A L O F T H E AM E R I C A N C O L L E G E O F C A R D I O L O G Y V O L . 6 8 , N O . 2 1 , 2 0 1 6

ª 2 0 1 6 B Y T H E AM E R I C A N C O L L E G E O F C A R D I O L O G Y F O UN DA T I O N

P U B L I S H E D B Y E L S E V I E R

I S S N 0 7 3 5 - 1 0 9 7 / $ 3 6 . 0 0

h t t p : / / d x . d o i . o r g / 1 0 . 1 0 1 6 / j . j a c c . 2 0 1 6 . 0 7 . 7 8 5

FOCUS SEMINAR: PERICARDIAL AND MYOCARDIAL DISEASE

STATE-OF-THE-ART REVIEW

Complicated Pericarditis
Understanding Risk Factors and Pathophysiologyto Inform Imaging and Treatment
Paul C. Cremer, MD,a Arnav Kumar, MD,a Apostolos Kontzias, MD,b Carmela D. Tan, MD,c E. Rene Rodriguez, MD,c

Massimo Imazio, MD,d Allan L. Klein, MDa

ABSTRACT

Fro

VacD

De

ha

rel

Ma

Most patients with acute pericarditis have a benign course and a good prognosis. However, a minority of patients develop

complicated pericarditis, and the care of these patients is the focus of this review. Specifically, we address risk factors,

multimodality imaging, pathophysiology, and novel treatments. The authors conclude that: 1) early high-dose cortico-

steroids, a lack of colchicine, and an elevated high-sensitivity C-reactive protein are associated with the development of

complicated pericarditis; 2) in select cases, cardiovascular magnetic resonance imaging may aid in the assessment of

pericardial inflammation and constriction; 3) given phenotypic similarities between recurrent idiopathic pericarditis and

periodic fever syndromes, disorders of the inflammasome may contribute to relapsing attacks; and 4) therapies that

target the inflammasome may lead to more durable remission and resolution. Finally, regarding future investigations, the

authors discuss the potential of cardiovascular magnetic resonance to inform treatment duration and the need to

compare steroid-sparing treatments to pericardiectomy. (J Am Coll Cardiol 2016;68:2311–28)

© 2016 by the American College of Cardiology Foundation.

C ompared with other cardiovascular diseases,pericarditis has garnered relatively little in-terest, in part due to the appropriate focus

on reducing cardiovascular mortality. In recent de-cades, however, as mortality related to cardiac dis-ease has continued to decrease (1), morbidity hasbecome increasingly important. This trend coincideswith an increased emphasis on pericarditis, a dis-ease often characterized by significant morbidity.Although most patients with acute pericarditis willhave resolution, some will develop incessant, recur-rent, chronic, or constrictive pericarditis. Many ofthese patients experience a debilitating chronic dis-ease, and to simplify presentation throughout thisreview, are referred to as having complicatedpericarditis.

m the aDepartment of Cardiovascular Imaging, Center for the Diagnos

scular Institute, Cleveland Clinic, Cleveland, Ohio; bDepartment of R

epartment of Anatomic Pathology, Heart and Vascular Institute, Cleve

partment, Maria Vittoria Hospital and Department of Public Health and Ped

s received institutional research grants from Acarpia and SOBI. All other au

evant to the contents of this paper to disclose.

nuscript received April 28, 2016; revised manuscript received July 5, 2016

The improved understanding of complicated peri-carditis is anchored in recent clinical trials, bolsteredby better characterization of pericarditis with multi-modality imaging, and better delineation of the un-derlying pathophysiology through molecular studies.Therefore, given the need for improved diagnosis andmanagement of patients with complicated pericar-ditis, this review focuses on the following questions:

1. Which patients are at risk for complicated diseaseafter acute pericarditis?

2. Which patients with complicated pericarditisbenefit from multimodality imaging?

3. What is the pathological progression of pericar-ditis, and how is autoinflammatory pericarditisdistinct from autoimmune pericarditis?

is and Treatment of Pericardial Disease, Heart and

heumatology, Cleveland Clinic, Cleveland, Ohio;

land Clinic, Cleveland, Ohio; and the dCardiology

iatrics, University of Torino, Torino, Italy. Dr. Imazio

thors have reported that they have no relationships

, accepted July 12, 2016.

https://s3.amazonaws.com/ADFJACC/JACC6821/JACC6821_fustersummary_06




http://crossmark.crossref.org/dialog/?doi=10.1016/j.jacc.2016.07.785&domain=pdf

http://dx.doi.org/10.1016/j.jacc.2016.07.785

ABBR EV I A T I ON S

AND ACRONYMS

CMR = cardiac magnetic

resonance

CT = computed tomography

ECG = electrocardiogram

IVIG = intravenous

immunoglobulin

MI = myocardial infarction

STIR = short-tau inversion

recovery

Cremer et al. J A C C V O L . 6 8 , N O . 2 1 , 2 0 1 6

Complicated Pericarditis N O V E M B E R 2 9 , 2 0 1 6 : 2 3 1 1 – 2 8

2312

4. What are the established treatments forpericarditis, and what are the emergingtherapies for patients with complicateddisease?

IDENTIFYING THE PATIENT WHO IS AT

RISK FOR PROGRESSION BEYOND

LIMITED ACUTE PERICARDITIS

Acute pericarditis is common, and whentreated with appropriate anti-inflammatorymedications, most symptoms resolve within

days to weeks (2,3). However, a significant minorityof patients will either experience adverse eventsrelated to the initial presentation or will be debili-tated by recurrent attacks. To aid in the identificationof these at-risk patients, an understanding of theepidemiology, etiologies, and classifications of peri-carditis is necessary. With this background, risk fac-tors for specific complications are better appreciated.

DEFINING THE STAGES OF PERICARDITIS. Pericar-ditis remains a clinical diagnosis with hallmark fea-tures, including precordial chest pain that is worsewith inspiration and when supine, characteristicST-segment elevation and PR deviation on electro-cardiogram (ECG), a pericardial friction rub, anda pericardial effusion that is more than trivial (4). Adiagnosis of pericarditis requires 2 of these 4 char-acteristics, usually chest pain in conjunction witheither ECG changes or a pericardial effusion (5–7).Supportive findings, although not included in thediagnostic criteria, include elevated inflammatorymarkers or other imaging evidence of pericardialinflammation (5,8,9).

Once diagnosed, pericarditis is categorized ac-cording to the duration of symptoms. Pericarditis thatpersists for more than 4 to 6 weeks is termed inces-sant, and the presence of symptoms for >3 months isconsidered chronic pericarditis. If a patient is free ofsymptoms for at least 4 to 6 weeks, relapse is referredto as recurrent pericarditis (5). As will be discussed,this distinction may be relevant, as the underlyingpathophysiology of a patient with idiopathic recur-rent pericarditis is likely different from a patient withchronic pericarditis related to autoimmune disease.In general, however, these distinct time frames areadmittedly arbitrary and not necessarily mutuallyexclusive. A patient with recurrent pericarditis mayalso develop incessant or chronic disease if thesymptoms are not controlled. Nonetheless, the num-ber and duration of attacks does reflect the morbidityof the disease. In addition, the distinction betweenacute pericarditis, a single recurrence, and multiple

recurrences of pericarditis has provided a usefulframework for the conduct of clinical trials (10–12).

ETIOLOGY AND EPIDEMIOLOGY OF PERICARDITIS.

In the United States and Western Europe, most epi-sodes of pericarditis (80% to 90%) are idiopathic andare often presumed to be post-viral (3). Conversely,tuberculosis is the most common cause of pericarditisin the developing world (13,14). Increasing causes ofpericarditis are the post-cardiac injury syndromes,which include pericarditis after myocardial infarction(MI), percutaneous coronary intervention, electro-physiology procedures, or post-pericardiotomy (15).Less common causes of pericarditis include autoim-mune diseases, chest irradiation, and active cancer.

To establish the cause of pericarditis, an extensivelaboratory analysis may increase the yield (16), but isnot routinely recommended (5). Currently, identifi-cation of the putative viral agent does not informprognosis or management. Similarly, testing forantinuclear antibodies is not recommended, as lowtiter levels are common and nonspecific (17). In pa-tients with a rheumatic disease, involvement of otherorgans is usually apparent before the onset of peri-carditis (18,19).

For the clinician, a central initial question is theoverall likelihood that a patient will develop compli-cated pericarditis. After an episode of acute pericar-ditis, the probability of developing incessantpericarditis or a first recurrence within 18 months isgenerally 15% to 30% (10,20). In patients who havehad an initial recurrence of pericarditis, additionalrecurrence occurs in 25% to 50% (11,12). Likewise,further exacerbations will develop in 20% to 40% ofpatients after 2 or more previous recurrences(Figure 1) (21).

These data regarding the incidence of recurrenceoriginate from randomized trials including approxi-mately 800 patients, and 80% of patients in thesetrials had an idiopathic cause of pericarditis(11,12,20,21). A separate question relates to the risk ofdeveloping a first episode of pericarditis after cardiacinjury. In the current era of early reperfusion foracute MI, late pericarditis (Dressler’s syndrome) isuncommon (<0.5%) (22,23). Early pericarditis afterST-segment elevation MI has also decreased, with anincidence of approximately 4%, although it is morecommon in patients who present after at least 6 h ofsymptoms (14%) and in patients with percutaneouscoronary intervention failure (23%) (23). After peri-cardiotomy, pericarditis is more common, with alikely incidence between 10% and 25% (24,25).

An additional risk of pericarditis is the develop-ment of constrictive pathophysiology. In constrictive

FIGURE 1 Incidence of Short-Term Adverse Events and Complicated Disease After an Episode of Acute Pericarditis

Resolution

MyocardialInvolvement

COMPLICATED

RecurrentPericarditis

Multiple Recurrences~6% (12)

Cardiac Tamponade

PERICARDITIS

ReversibleConstrictivePericarditis

ChronicConstrictive Pericarditis

~1-2% (27)

Acute Pericarditis~1-2% (20,30)~15% (31)

~15-30% (10,20)

?

?

?

?

?

A B

C

D

E

After an episode of acute pericarditis, a minority of patients will develop short-term adverse events, including cardiac tamponade or

concomitant myocarditis. Delayed enhancement imaging may delineate associated subepicardial myocardial involvement (A), and transthoracic

echocardiography quickly defines the extent of a pericardial effusion (B). With appropriate treatment of acute pericarditis, the long-term

prognosis is generally good. However, a subset of patients will develop complicated pericarditis, and a few of these patients will have multiple

recurrences. In select cases, delayed enhancement imaging is useful and may demonstrate increased pericardial signal (C). Rarely, patients will

develop chronic constrictive pericarditis requiring pericardiectomy, although the incidence of reversible constrictive pericarditis is unknown.

Early inspiratory septal shift may aid in the diagnosis of constrictive pericarditis (D) (Online Videos 1, 2, 3, and 4), and the extent of calcification

can be defined with computed tomography (E).

J A C C V O L . 6 8 , N O . 2 1 , 2 0 1 6 Cremer et al.N O V E M B E R 2 9 , 2 0 1 6 : 2 3 1 1 – 2 8 Complicated Pericarditis

2313

pericarditis, diastolic heart failure results from peri-cardial restraint that decreases the compliance of thecardiac chambers (26). The eventual development ofconstrictive pericarditis that is severe enough torequire surgery after an episode of acute idiopathicpericarditis is rare, and likely occurs in a minority ofpatients (w1%) (27). However, constrictive pericar-ditis that responds to anti-inflammatory therapy islikely more common (28), although the incidence ofthis reversible constrictive pericarditis is unknown.

ADVERSE EVENTS FROM ACUTE PERICARDITIS AND

RISKS FOR THE DEVELOPMENT OF COMPLICATED

PERICARDITIS. In a patient with acute pericarditis,risk stratification is dichotomized into the likelihoodof adverse events from the initial attack and thelikelihood of developing complicated pericarditis inthe ensuing months to years. Even though morbidityrelated to the disabling pain of acute pericarditis maybe substantial, death in patients hospitalized for

acute pericarditis is uncommon (1%), and mortality ismost often not related to pericarditis (29).

Conversely, a direct complication of pericarditisincludes extension of the inflammation through thepericardial layers and epicardial fat to involve themyocardium. When the inflammation predominantlyinvolves the pericardium, it is termed myoper-icarditis, in distinction to perimyocarditis, in whichmyocardial inflammation supersedes the pericarditis.Clinically, both syndromes require an abnormaltroponin level. In perimyocarditis, patients will alsohave focal or globally reduced left ventricular systolicfunction (30).

Myocardial involvement occurs in approximately15% of patients with acute pericarditis, and risk fac-tors include younger age, male sex, fever,arrhythmia, and ST-segment elevation on ECG. Ofnote, myopericarditis generally has a benign courseand is not associated with an increased risk ofrecurrent pericarditis or pericardial tamponade.

http://jaccjacc.acc.org/video/2016/1698_VID1.avi




TABLE 1 Risks for Developing Complicated Disease After an Episode of

Acute Pericarditis

Treatment-related variables with increased risk in multivariable models

Corticosteroid use

Lack of colchicine

Patient-related variables with increased risk in multivariable models

Incomplete response to NSAIDs

Elevated high-sensitivity C-reactive protein

Patient-related variables without increased risk in multivariable models

Younger age

Sex

Pericardial effusion

Adapted from data in Imazio et al. (20) and Imazio et al. (33).

NSAID ¼ nonsteroidal anti-inflammatory drug.



2314

Moreover, most patients with left ventriculardysfunction will have recovery (30,31) (Figure 1).

Although the long-term outcome is favorable forthese patients, they are generally observed in thehospital. Likewise, patients with large effusions arealso admitted, and hospitalization can be consideredin certain patients with fever, a subacute course,trauma, anticoagulation, or a failure of outpatientanti-inflammatory agents (5,6). After the patient hasbeen risk stratified according to the likelihood ofdeveloping a short-term adverse event, the nextquestion relates to the risk of later developingcomplicated pericarditis. These risks can be catego-rized as either patient- or treatment-related (Table 1).

In regard to treatment, early use of corticosteroidshas been consistently associated with recurrence(32,33), and relapse is especially hastened with high-dose short courses. Colchicine has reduced re-currences of pericarditis and has become a mainstayof treatment. In regard to patient-related factors, at1 week after acute pericarditis, an incompleteresponse to nonsteroidal anti-inflammatory agentsand a persistently elevated high-sensitivity C-reac-tive protein (CRP) are both associated with risk ofrecurrence (33). Of note, other patient-related fac-tors, including younger age, female sex, and peri-cardial effusion, have not been associated withrecurrent pericarditis (33).

The risks for developing constrictive pericarditisare less well defined, primarily because the diagnosisis less common and the epidemiology is not aswell understood. Moreover, most data relate to pa-tients who have had a pericardiectomy, and thesepatients may represent only a subset of patientswho have constrictive pathophysiology. In some pa-tients, the constrictive pathophysiology may not besevere enough to require surgery, whereas otherpatients may have constriction that resolves with

anti-inflammatory treatment (Figure 1) (28). Overall, ahistory of idiopathic pericarditis remains the mostcommon diagnosis in patients who have a peri-cardiectomy for constriction, given the predominantprevalence of this condition, although pericar-diectomy after prior cardiac surgery is increasing (34).Among patients who have a pericardiectomy, thosewith idiopathic pericarditis have the best prognosis.Patients with post-pericardiotomy syndrome havean intermediate outcome, and patients with ahistory of radiation heart disease have the highestmortality (34).

THE POTENTIAL VALUE OF MULTIMODALITY

IMAGING IN COMPLICATED PERICARDITIS

Most patients with acute pericarditis will have alimited and uncomplicated illness, and echocardiog-raphy is the first and only imaging test necessary(Central Illustration) (35). In this setting, becausemany patients have been diagnosed according tocharacteristic chest pain and ECG changes, echocar-diography is performed primarily for risk stratifica-tion. Specifically, echocardiography may identify apericardial effusion, evidence of pericardial tampo-nade, wall motion abnormalities, or features of peri-cardial constriction (Figure 1B) (8).

After this initial risk stratification, only a minorityof patients will develop complicated pericarditis.Some of these patients may benefit from further im-aging, primarily directed at 2 questions that mayremain unresolved after clinical evaluation andechocardiography. First, does the patient still havesignificant pericardial inflammation? Second, doesthe patient have constrictive pathophysiology? Ingeneral, compared with echocardiography, cardiacmagnetic resonance (CMR) has a predominant role inthe evaluation of pericardial inflammation. Forconstriction, CMR is typically an adjuvant test whenechocardiographic data are ambiguous.

To appropriately select patients with complicatedpericarditis for CMR to address either of these ques-tions, clinicians should understand the sequencesused to evaluate the pericardium and also how theseimages correlate with the progressive pathology ofpericarditis. Even though a detailed discussion ofCMR is beyond the scope of this review (36), thelimitations of current techniques should be empha-sized (Table 2). Nonetheless, despite these limita-tions, emerging evidence indicates that the stage andseverity of pericardial inflammation is better assessedwith CMR. This improved assessment may subse-quently lead to better care for the patient withcomplicated pericarditis.

CENTRAL ILLUSTRATION Complicated Pericarditis: Clinical Stages of Pericarditis With Imaging andTreatment Considerations

Clinical Stages of Pericarditis With Imaging and Treatment Considerations

Acute Firstrecurrence

Multiple recurrences

Colchicine-resistant orsteroid dependent

Constrictive

• Echocardiogram for pericardial

myocardial involvement, constriction

• NSAIDS(weeks)

• Colchicine(3 mos.)

• NSAIDS

• Colchicine

• NSAIDS+ Colchicine+ Prednisone(>6 mos., taper steroidas tolerated)

• Consider steroid- sparing agent (warrantsfurther study)

• NSAIDS+ Colchicine+ Prednisone+ Steroid- sparing agent

taper steroidas tolerated)

• Considerpericardiectomy (warrantsfurther study)

• Intensifymedical therapy

• Pericardiectomy if “burnt out”

• Echocardiogram for constriction

• CMR in select cases for pericardial

or constriction

Same as for Same as for Same as for

Plus possibleCT for extent of

and preoperative planning

Stage ofpericarditis

Imaging

Treatment

Cremer, P.C. et al. J Am Coll Cardiol. 2016;68(21):2311–28.

All patients with acute pericarditis should have an echocardiogram for short-term risk stratification, and subsequent echocardiograms can be performed if there is

concern for constrictive pericarditis. In recurrent pericarditis, CMR imaging has an emerging role to assess for pericardial inflammation if the clinical evaluation is

equivocal and to assess for constrictive pathophysiology if the echocardiogram is indeterminate. CT is primarily employed to assess pericardial calcification and for pre-

operative planning. The mainstay of treatment is NSAIDs and colchicine with the addition of low-dose corticosteroids in patients with multiple recurrences. Steroid-

sparing agents can be added in refractory cases. Early use of steroid-sparing agents and pericardiectomy for recurrent pericarditis may be beneficial and warrants further

study. CMR ¼ cardiac magnetic resonance; CT ¼ computed tomography; LV ¼ left ventricle; NSAID ¼ nonsteroidal anti-inflammatory drug; RV ¼ right ventricle.


2315

FORM AND FUNCTION IN MULTIMODALITY IMAGING

OF THE PERICARDIUM. In a CMR evaluation forpericardial disease, morphological characteristics arecomplemented by insight into the hemodynamicconsequences of pericardial constraint. In contrast,cardiac computed tomography (CT) generally pro-vides only anatomic information. For example, CTmay occasionally aid in the percutaneous approach topericardiocentesis and may also help with surgicalplanning prior to pericardiectomy (37). Moreover, CTis particularly well-suited to define the extent ofpericardial calcification (Figure 1E), but generallyprovides limited and indirect hemodynamic infor-mation (38). Alternatively, complete anatomicdelineation of pericardial abnormalities with

echocardiography is often inferior when comparedwith CT and CMR, although hemodynamic assess-ment with echocardiography is excellent (8).

In pericardial disease, the noninvasive hemody-namic evaluation relates to the diagnoses of pericar-dial tamponade and constriction. Echocardiographyis essential in the rapid assessment of a patient withsuspected pericardial tamponade, and CMR has norole in these patients. In the stable patient with sus-pected constrictive pathophysiology, echocardiogra-phy and CMR data may be complementary, althoughredundant data collection should be avoided. Echo-cardiographic features of constriction include respir-ophasic septal shift, annulus reversus or paradoxuson tissue Doppler imaging, significant respiratory

TABLE 2 Practical Considerations in the Current Use of CMR to Evaluate Pericarditis

Assessment Method Strengths Limitations

Pericardial thickness T1- or T2-weighted fast-spinecho images (36)

� Extent of pericardial thickness appreciated� Often abnormal in patients with constrictive

pericarditis

� In-plane spatial resolution limits assess-ment of normal or mildly thickenedpericardium (46)

� Some patients with constrictive pericarditismay have normal pericardial thickness

Pericardial edema T2 short-tau inversion-recoveryfast-spin echo images (36)

� Likely a specific finding for active pericarditis � Difficult to distinguish from pericardialeffusion (48)

Pericardial inflammation Late gadolinium enhancement(36,51)

� Increased pericardial enhancement reflectsincreased vascularity (53)

� Potential use of “ruling in” or “ruling out”pericarditis in equivocal cases (49,56)

� Correlation with different stages ofpericarditis still being defined

Ventricular interdependence Cine imaging with short-axisimages at the basal level withpatients instructed to breathedeeply (36)

� Relative septal excursion may be a specificfinding for constrictive pericarditis (45,46)

� Patient respiratory effort can influencediagnostic yield

CMR ¼ cardiac magnetic resonance.



2316

variation in tricuspid and mitral inflow, preservedglobal longitudinal strain, a dilated inferior venacava, and prominent diastolic expiratory flow reversalin hepatic veins (39–43).

If a patient has all of these findings, CMR may behelpful to assess for ongoing pericardial inflamma-tion, but the diagnosis of constrictive pericarditis issecure. Yet, few patients demonstrate all of thesefeatures, and the best multiparametric assessment todiagnose constrictive pericarditis with echocardiog-raphy is unclear (44). Therefore, CMR is likely bestreserved for patients with suggestive, but inconclu-sive evidence of constrictive pericarditis on echocar-diography (Central Illustration).

The imaging features of constrictive pericarditis,whether assessed by echocardiography or CMR,reflect the consequences of pericardial constraint.Specifically, this constraint renders the cardiacchambers noncompliant, results in prominent respi-ratory variation in flow, and accentuates early dia-stolic filling. With CMR, this physiology is reflected inventricular coupling during real-time cine imagingwith free breathing (Figure 1D, Online Video 1)(45,46). During early inspiration in patients withconstrictive pericarditis, the septum shifts toward theleft ventricle (Online Video 2). An increased relativeseptal excursion has been shown as a specific findingfor constrictive pericarditis (Table 2) (45,46). Inpractice, though, this technique may be limited if thepatient’s breathing does not generate negative intra-thoracic pressure that falls within an acceptablerange. For example, if a patient can take only shallowbreaths, the change in intrathoracic pressure is less-ened, and the sensitivity of septal excursion isdecreased (Online Video 3). Conversely, witha particularly vigorous inspiration, mild septal

excursion may be evident in a patient withoutconstrictive pericarditis, thus decreasing specificity(Online Video 4).

To assess anatomy, the CMR evaluation usuallybegins by assessing pericardial thickness, often withblack blood preparation turbo spin-echo sequencesthat include a breath-hold (Figures 2A, 2D, and 2G)(46). On CMR, normal pericardial thickness is <4 mm(36). However, anatomic studies suggest that thenormal parietal pericardium is thinner, approxi-mately #1 mm or less (Figure 3) (47). CMR may thusoverestimate pericardial thickness, possibly due tomotion or chemical shift artifacts at the fat-fluidinterface (48). Importantly, given the normal dimen-sion of parietal pericardium, the in-plane spatialresolution of CMR is limited (49). Therefore, normalpericardium can be difficult to delineate on CMR andoften appears “pencil-thin.” In addition, given theselimitations, reporting a pericardial thickness of 3 mmas normal and a thickness of 4 mm as abnormal isclearly fraught with error. Nonetheless, markedlythickened pericardium is usually evident on CMR,and has been described in patients with constrictivepericarditis (Figure 2G) (46,50). Although the extentof pericardial thickness is better appreciated on CMRcompared with echocardiography, the imprecision ofthis measurement should be emphasized, especiallyin patients with seemingly normal or slightlyincreased pericardial thickness (Table 2).

Another potential advantage of CMR is the abilityto assess pericardial edema (Figures 2B, 2E, and 2H).On T2-weighted short-tau inverted recovery (STIR)fast spin-echo sequences, pericardial edema willappear bright (Figure 2E) (48). Occasionally, theseedema-weighted images are helpful, but their clinicaleffect is often limited for 2 reasons (Table 2). First,





FIGURE 2 Cardiovascular Magnetic Resonance of the Pericardium

In a patient with a normal pericardium (A to C), the pericardium is often difficult to delineate, even on a breath-hold double inversion-recovery,

dark-blood, short-axis image (A). With an edema-weighted, short-tau inversion recovery, fast spin-echo image, a hyperintense signal is seen

only in the area of a trivial pericardial effusion at the inferior margin over the anterior right ventricle (B, white arrowhead). With a phase-

sensitive inversion-recovery technique and no fat suppression, there is no delayed pericardial enhancement (C). In this patient with active

pericarditis (D to F), abnormal pericardial thickness is not seen (D), but there is diffuse hyperintense signal on edema-weighted imaging

(E, white arrow), and near circumferential delayed enhancement of the pericardium (F, yellow arrow). Finally, in this patient with chronic

constrictive pericarditis (G to I), the pericardium is abnormally thickened (G, yellow diamond). There is no hyperintense signal on edema-

weighted imaging (H), and in this image with a fat suppression preparation, there is no delayed pericardial enhancement (I).


2317

patients with dramatically increased pericardial sig-nals on T2-STIR images typically have severe peri-carditis on the basis of history, examination, andinflammatory markers. Therefore, this finding doesnot usually inform the diagnosis, and, at present, theeffect on patient management and prognosis is un-clear. Second, in a patient who also has a pericardial

effusion, the interpretation of superimposed peri-cardial edema is difficult because both will appearbright on T2-STIR images. Cine steady-state free pre-cision images typically define the extent of a peri-cardial effusion, and short-axis comparisons of thesecine images with short-axis T2-STIR images may helpto distinguish pericardial effusion from pericardial

FIGURE 3 Normal Pericardial Histology

(A) The parietal pericardium is normally <1 mm in thickness. The parietal pericardium is lined with a layer of mesothelial cells, forming the

serosal pericardium depicted on the top portion of the section. The fibrous pericardium forms the tough pericardial sac and consists of dense

collagen bundles (yellow) with interspersed elastic fibers (black). A variable amount of adipose tissue is present outside of the fibrous peri-

cardium facing the mediastinum. The layer of mesothelial cells at the bottom portion of the section represents the serosal surface of the

mediastinum facing the pleural cavity (original magnification �100, Movat pentachrome stain). (B) The visceral pericardium, also called the

epicardium, consists of a layer of mesothelial cells, which rests on a basement membrane supported by thin fibrous tissue (yellow) with elastic

fibers (black). In some areas of the heart, the myocardium is seen immediately beneath the epicardium. The mesothelial cells have a fuzzy

border due to well-developed surface microvilli (original magnification �600, Movat pentachrome stain). (C) In most other areas of the heart,

there is a subepicardial layer of adipose tissue separating the epicardium from the myocardium. This intervening layer of adipose tissue contains

blood vessels, lymphatic vessels, and nerves (original magnification �400, Movat pentachrome stain).



2318

edema. In practice, however, a comment on pericar-dial edema in a patient with a pericardial effusion isdifficult to report with confidence.

In addition, late gadolinium enhancement caninform the presence and severity of pericardialinflammation (Figures 1C, 2C, 2F, and 2I) (51). Typi-cally, this protocol is similar to imaging for myocar-dial scar. Consequently, when there is a concern ofmyocardial involvement and echocardiography isinconclusive, CMR may define the extent of associ-ated myocarditis (Figure 1A). With severe pericarditis,the inflammation can also extend into the surround-ing epicardial fat. This epicarditis may be seen on

CMR with increased delayed enhancement of thepericardium and epicardial fat. However, the kineticsof gadolinium egress from the pericardium are un-clear. Therefore, the optimal timing for imaging,especially in different stages of pericarditis, is notdefined (Table 2). These uncertainties highlight thenascent state of pericardial CMR and the need forfurther research. Despite these limitations, early workhas correlated pericardial histopathology with CMR.

DELAYED ENHANCEMENT OF THE PERICARDIUM AND

THE PATHOLOGICAL STAGES OF PERICARDITIS. Thepericardium responds to injury with increased

FIGURE 4 Gross Pathology of Pericarditis

(A) The classic appearance of acute fibrinous pericarditis is shown, with fibrin deposits projecting into the pericardial space as they line both the

parietal and visceral pericardium. In the early stage, there is no adhesion between the 2 layers of the pericardium. (B) Organization of the

fibrinous exudate can lead to diffuse obliteration of the pericardial space or focal adhesions by a fibrotic repair process. The fibrinous exudate

shown here has a more variegated appearance with pale and dark areas as the repair process ensues. (C) Healed pericarditis results in thin

adhesions, appearing as depressed gray translucent soft tissue between the parietal pericardium and epicardial fat.


2319

vascularity, a predominant infiltration of neutrophils,formation of granulation tissue, and depositionof fibrin. Grossly, in early pericarditis, the fibrindeposits line the parietal and visceral pericardium,but there is no adhesion between the 2 layers. Orga-nization of the fibrinous exudate can lead to focaladhesions that become more confluent as the peri-carditis heals (Figure 4). On histopathology, acutepericarditis is characterized by disintegration andexfoliation of the mesothelial cells lining the peri-cardium (Figure 5). As the inflammatory processcontinues, the fibrinous exudate organizes, andeventually, deposition of extracellular matrix leads tofibrosis of the pericardium (Figure 5).

Typically, given the lack of vascularity in normalpericardium, gadolinium uptake in the pericardiumis absent or minimal. As pericarditis leads toincreasing neovascularization, delayed enhance-ment of the pericardium also increases (Figures 1Cand 2F). This relationship is distinct from delayedenhancement of the myocardium. In myocardialdiseases, delayed enhancement reflects the degreeof myocardial fibrosis (52). Conversely, end-stagepericarditis may have prominent pericardialfibrosis and resultant pericardial thickening, butlimited vascularity and active inflammation. In this

scenario, pericardial delayed enhancement will beminimal (Figure 2I).

This putative correlation between pericardialvascularity and delayed enhancement is on the basisof small studies of patients with CMR and pericardialhistology (51,53). In patients who had peri-cardiectomy, patients with more active pericarditis,termed “organizing pericarditis with fibroplasia,”were more likely to have delayed enhancement. Inparticular, pericardial delayed enhancement wasassociated with fibroblastic proliferation, neo-vascularization, and granulation tissue. Conversely,patients without delayed enhancement were morelikely to have organized fibrous pericarditis, charac-terized by pericardial fibrosis and calcification (53).

In addition to the small size, a notable limitation ofthese studies is that histological correlation is typi-cally performed from samples obtained during peri-cardiectomy. Because pericardiectomy is most oftenperformed for constrictive pericarditis, the validity ofthese data in other inflammatory pericardial diseasesis limited. Moreover, even when pericardiectomy isperformed for medically refractory pain related torecurrent pericarditis, patients have typically beentreated aggressively with anti-inflammatory agents,not only to reduce their pain, but also to decrease

FIGURE 5 Histopathology of Pericarditis

(A) An example of fibrinous pericarditis with abundant deposition of fibrinous material (*) that contains scattered inflammatory cells. The

fibrous pericardium with mesothelial lining (arrowhead) is present at the lower portion of the microphotograph (original magnification �100,

hematoxylin-eosin stain). (B) In acute pericarditis, there is disintegration and exfoliation of the mesothelial cells (arrowhead) lining the

pericardial space. The initial inflammatory cell response to injury is predominantly composed of neutrophils (arrows). The fibrin exudate (*) is

seen as eosinophilic (bright pink) meshwork separating the mesothelial cells, and is also present in the background of inflammatory cells

(original magnification �400, hematoxylin-eosin stain). (C) Organization of the fibrinous exudate (*) is indicated by neovascularization and

ingrowth of fibroblasts, with deposition of extracellular matrix that leads to fibrosis of the pericardium (original magnification �200,

hematoxylin-eosin stain). Healing with dense fibrosis may lead to pericardial constriction. (D) Depending on the extent and duration of the

insult leading to injury, the fibrin exudate can be completely resorbed. The resolution of inflammation and the tissue repair process can lead to

obliteration of the pericardial space. In some instances, as in this case of post-cardiac surgery pericarditis, the resulting fibrosis is loose in nature

and does not cause significant fibrous thickening of the pericardium. The extracellular matrix contains sparse collagen fibers and mild neo-

vascularization (original magnification �100, hematoxylin-eosin stain). Grossly, this healed pericarditis results in loose fibrous adhesions, as

seen in Figure 4C.



2320

complications related to operating on an intenselyinflamed pericardium. Consequently, in patients withthe most severe pericardial inflammation, the rela-tionship between histology and pericardial delayedenhancement is not well defined.

PATIENTS WHO BENEFIT FROM AN ASSESSMENT OF

DELAYED PERICARDIAL ENHANCEMENT. With theselimitations as background, the next question centerson which patients with complicated pericarditisbenefit from an assessment of delayed pericardialenhancement. Specifically, how will patient man-agement change on the basis of the presence orabsence of significant delayed pericardial enhance-ment? Currently, CMR is most appropriate for thisindication in 2 scenarios. The first applies to a patient

in whom the presence of active pericardial inflam-mation is uncertain. The second applies to a patientwith constrictive pericarditis in whom the severity ofactive pericarditis is unclear.

In a patient with a history of pericarditis, chestpain reminiscent of a previous attack may recur, butother diagnostic characteristics may be lacking. In acohort of 275 patients with previous idiopathic acutepericarditis, approximately 10% developed chest painwithout other clinical evidence of pericarditis (54).This presentation was more common in women, pa-tients treated with glucocorticoids, and patients witha history of multiple attacks. Eventually, a third ofthese patients fulfilled typical diagnostic criteria forpericarditis. Therefore, in a patient who has chestpain and a history of pericarditis, delayed pericardial


2321

enhancement may favor continued or intensifiedanti-inflammatory treatment. If CMR does not showdelayed pericardial enhancement, then tapering ofmedications may continue, and other diagnoses maybe considered (Central Illustration).

For patients with recurrent pericarditis, pre-liminary observational data support this CMR-guidedapproach to treatment. In a single-center cohort of507 patients with recurrent pericarditis, approxi-mately one-half had CMR (55). Overall, patients withCMR subsequently received less treatment with glu-cocorticoids. These data suggest that clinicians aremore comfortable tapering glucocorticoids when CMRdoes not show significant delayed pericardialenhancement. However, these results should beconsidered suggestive, given their observational andretrospective nature without adjustment for potentialconfounders.

CMR also has the potential to affect managementin patients with constrictive pericarditis when thedegree of active pericardial inflammation is uncertain(Central Illustration). In 2 small observational studies,increased pericardial delayed enhancement wasassociated with reversible constrictive pericarditis(49,56), and a quantitative assessment of pericardialdelayed enhancement may be superior to visualanalysis (49). In patients with constrictive patho-physiology and increased pericardial delayedenhancement, active inflammation may be contrib-uting to the constriction. As this inflammationresolves with anti-inflammatory therapy, theconstriction may also resolve. In patients withconstrictive pathophysiology and without increasedpericardial delayed enhancement, advanced fibrosismay cause constriction. This fibrotic state is unlikelyto respond to anti-inflammatory therapy.

However, for patients with established constrictivepericarditis, CMR to assess for pericardial inflamma-tion is not necessary in every patient. In a patientwith heart failure, no chest pain, and prominentcircumferential calcification on chest x-ray or CT,CMR is unlikely to add diagnostic value. The calcifi-cation represents the end stage of the inflammatoryprocess, and pericardiectomy is likely necessary forsymptom relief.

The patient with constrictive pericarditis at theother end of the inflammatory spectrum also deriveslittle benefit from CMR. Medical therapy is indicatedin a patient with typical pericarditic pain andelevated inflammatory markers. Currently, there isincreasing interest in modulating the intensity ofanti-inflammatory therapy according to the severityof pericardial inflammation on CMR, but publisheddata are lacking. However, many patients with

constrictive pericarditis are neither severely inflamednor obviously “burnt-out.” These patients appear tobenefit most from CMR to further characterize theirpericardial inflammation. In fact, patients may nothave elevated inflammatory markers, but will stillhave significant delayed pericardial enhancement(49). Moreover, these patients may improve withanti-inflammatory therapy, and referring the patientfor pericardiectomy may be unnecessary or prema-ture (Central Illustration). Even if the patient does notimprove enough clinically to avoid pericardiectomy,pre-treatment with anti-inflammatory agents mayfacilitate a more successful surgery for organized, asopposed to fibrinous, pericarditis, although this hy-pothesis is on the basis of anecdotal experience.

SIMILARITIES AND DIFFERENCES

IN AUTOINFLAMMATORY AND

AUTOIMMUNE PERICARDITIS

Despite an improved understanding of the histo-pathological progression of recurrent pericarditis,and early studies on the correlation of pathology withclinical and imaging features, the underlying patho-physiology of recurrent pericarditis remains poorlyunderstood. In general, recurrent pericarditis resultsfrom an interplay between environmental triggersand the innate and adaptive immune systems in agenetically susceptible host (57). Recurrent attacksmay result from an inability to clear the presumedviral infection with increased viral replication, a hy-pothesis supported by the increased risk of relapse inpatients treated with glucocorticoids (58). Alterna-tively, infections may trigger an autoimmuneresponse via molecular mimicry, where a foreign an-tigen shares sequences or structural similarities withself-antigens. This similarity may result in cross-reactivity and recurrent inflammatory attacks (59).In addition, superantigens may be generated by mi-crobes or virus-infected cells, which can result inactivation of T cells, irrespective of antigen speci-ficity. Moreover, following tissue injury, cell death,and oxidative stress, self-proteins may be recognizedas foreign due to mutation, altered expression, mis-folding, or post-translational modification. Finally,epitope spreading and a tolerance break may occurthrough enhanced processing and presentation ofself-antigens, independent of specific T-cell receptorstimulation (60).

Although these mechanisms emphasize an inap-propriate adaptive immune response, more recentinvestigations suggest that innate immunity and itseffector mechanisms may be at the epicenter ofthe pathogenesis of recurrent idiopathic pericarditis.



2322

This insight stems from similarities between recur-rent pericarditis and prototypical autoinflammatorydiseases, including familial Mediterranean fever(FMF) and tumor necrosis factor–associated periodicsyndrome (TRAPS) (61). These syndromes are char-acterized by seemingly unprovoked attacks of multi-system inflammation and are triggered by innateimmunity perturbations in the absence of antigen-specific T cells or high titers of autoantibodies.To better understand the pathophysiology of auto-inflammatory disease and its distinction from pre-dominant autoimmune disease, research has focusedon the pivotal role of the inflammasome (62).

PUTATIVE ROLE OF THE INFLAMMASOME AS A

DANGER AND PATHOGEN SIGNAL SENSOR. Inflam-masomes are integral players in innate immunity andrespond to a wide array of damage- and pathogen-associated molecular patterns. In addition,pathogen-associated molecular patterns are recog-nized by endosomal and extracellular Toll-like re-ceptors that can also activate inflammasomes. Inbasic structure, the inflammasome is a cytosolicmacromolecule composed of the adaptor protein ASC,procaspase 1, and, importantly, a sensor molecule(Figure 6). This sensor molecule contains anucleotide-binding oligomerization domain–like re-ceptor (NLR), and is triggered by diverse stimuli.

The best-characterized inflammasome has an NLRpyrin domain-containing 3 (NLRP3) sensor molecule.Gain-of-function mutations in NLRP3 result in thecryopyrin-associated periodic syndromes, a group ofautosomal-dominant autoinflammatory diseases.Seemingly disparate diseases can also activateNLRP3, including monosodium urate crystals fromgout, cholesterol related to atherosclerosis, and manyviruses (63–65). In fact, some of these viruses,including adenovirus, influenza A, herpesvirus, andcytomegalovirus, are associated with pericarditis.Once activated, the NLRP3 inflammasome leads to therelease of interleukin (IL)-1 (Figure 6). IL-1 then re-cruits effector cells of a myeloid lineage, namelyneutrophils, monocytes, and macrophages, to the siteof injury. In autoinflammatory syndromes, this skewtoward IL-1 expression predominates. Conversely,autoimmune diseases, such as systemic lupus ery-thematosus (SLE), primarily have a type I interferonsignature (62).

RECURRENT PERICARDITIS SECONDARY TO

AUTOINFLAMMATORY DISEASE. TRAPS is an auto-somal dominant periodic fever syndrome mostcommonly caused by missense mutations in the re-ceptor for tumor necrosis factor (TNF)-a (66).These mutations lead to downstream activation of

TNFa-dependent signaling pathways that up-regulatethe inflammasome (61). Typically, patients developsymptoms every 5 to 6 weeks, and symptoms last 1 to3 weeks. Common symptoms include fever, migratorymyalgia, rash, and serositis. In an international reg-istry of 138 patients with TRAPS, 7% had pericarditisand 25% had chest pain (67). Of note, a family historyof recurrent pericarditis should prompt screening, asincomplete TRAPS phenotypes and low-penetrancemutations have been described (68,69). In patientswith TRAPS, attacks may be unprovoked or may beprecipitated by injury, minor infection, stress, exer-cise, or hormonal changes (67).

Another prototypical autoinflammatory disease,FMF is caused by missense mutations in the MEFVgene, which encodes pyrin, a component of theNLRP3 inflammasome (70). Pyrin mutations renderthe NLRP3 inflammasome constitutively active, andtherefore allow caspase 1 to cleave pro–IL-1b to IL-1b(Figure 6) (71). Despite this constitutive activation ofthe NLRP3 inflammasome, FMF is characterized byintermittent inflammatory attacks lasting 1 to 3 dayswith fever, serositis, and oligoarthritis. In a registry of346 pediatric patients with FMF, pericarditis occurredin approximately 18%, although chest pain was morecommon (56%) (72).

In addition to a similar pattern of relapse andremission, the clinical link between FMF and idio-pathic recurrent pericarditis includes the effective-ness of colchicine in both diseases. Colchicine inhibitsthe activation of P2X2 and P2X7 pores, 1 of the firstsignals of NLRP3 activation. Furthermore, colchicineinhibits caspase 1, as well as the release of TNFa andreactive oxygen species (73). Finally, evidence tosupport a predominant role of the inflammasome inrecurrent idiopathic pericarditis relates to the role ofIL-1. As a final common pathway of inflammasomeproduction, IL-1 has been a pharmacological target forautoinflammatory diseases. In fact, as will be dis-cussed, antagonism of the IL-1 receptor may havepromise in the treatment of patients with refractoryrecurrent idiopathic pericarditis.

RECURRENT PERICARDITIS SECONDARY TO AN

AUTOIMMUNE DISEASE. Pericarditis also occurs inthe context of systemic autoimmune disease, oftenduring flares. In SLE, pericardial inflammation oreffusion can occur in as many as 20% to 50% of pa-tients, with a higher prevalence of pericardialinvolvement when autopsy is performed (18,74).Usually, other features of SLE, such as rash, arthritis,pleuritis, or leukopenia, accompany pericarditis.Glucocorticoids are generally effective for pericardialsymptoms, but more aggressive immunosuppressive

FIGURE 6 Activation of the Inflammasome, a Proposed Pathogenic Mechanism in Recurrent Idiopathic Pericarditis

The initial stimulus can be either microbial (PAMPs) or sterile (DAMPs). This stimulus is then recognized by innate immunity receptors found either at the cell surface

(TLRs) or inside the cell (NLRs). NLRs are then integrated into the structure of the inflammasome. Variants in the genes encoding inflammasome proteins can render

them constitutively active or lower the threshold upon which the macromolecular structure is assembled in response to DAMPs and PAMPs. TLR signaling also leads to

NF-kB activation and IL-1b production via ROS. Once primed by signals such as ATP, the threshold may be lower for inflammasome activation, resulting in recurrent

attacks. ATP ¼ adenosine triphosphate; CARD ¼ caspase activation and recruitment domain; DAMPs ¼ damage-associated molecular patterns; IL ¼ interleukin;

JNK ¼ Jun amino-terminal kinase; Kþ ¼ potassium ions; NF-kB ¼ nuclear factor–kappa B; NLRP3 ¼ nucleotide-binding oligomerization domain–like receptor pyrin

domain-containing 3; PAMPs ¼ pathogen-associated molecular patterns; ROS ¼ reactive oxygen species; TLR ¼ Toll-like receptors; TNF ¼ tumor necrosis factor;

TNFR ¼ tumor necrosis factor receptor.


2323

therapy may be necessary, depending on overall dis-ease activity (75).

Rheumatoid arthritis may also frequently involvethe pericardium, but is symptomatic in <10% of pa-tients (76). Pericarditis is more common in severe,deforming, nodular rheumatoid arthritis with hightiters of rheumatoid factor and cyclic citrullinatedpeptide antibodies. Targeted biologic therapy maybe necessary to control other symptoms, but such

therapy specifically directed at pericarditis is un-usual, and constrictive pericarditis is rare in the cur-rent era.

In conclusion, this dichotomization betweenautoimmune and autoinflammatory disease can be auseful construct, but it is an oversimplification ofcomplex biological processes. In fact, both innate andadaptive effector mechanisms act in concert, andseveral genetic polymorphisms of inflammasome



2324

components are associated with susceptibility, ac-tivity, and treatment responses of autoimmune dis-eases (77,78). Nevertheless, in the approach tounderstanding and treating the patient with pericar-ditis, the diathesis toward autoinflammatory orautoimmune disease is important. Autoinflammationis likely a more important mechanism in many pa-tients with recurrent idiopathic pericarditis, whereasautoimmunity may be paramount in patients withchronic pericarditis and systemic autoimmunedisease.

ESTABLISHED AND NOVEL TREATMENTS FOR

PERICARDIAL INFLAMMATION

Given the predominant autoinflammatory mecha-nisms and relapsing episodes that characterizerecurrent idiopathic pericarditis, the mainstay ofmedical therapy has understandably involved anti-inflammatory therapy, with the goal of not onlycontrolling symptoms during a flare, but alsodecreasing recurrence (11,20). This latter objective ofsecondary prevention should be emphasized. Manypatients with recurrent idiopathic pericarditis have achronic disease. They may take anti-inflammatoriesfor months or years, only to relapse randomly orwhen their medications are tapered. For these pa-tients, an understanding of therapies with estab-lished efficacy is essential, and the potential fordisease-modifying therapies should be investigated.

ESTABLISHED TREATMENTS FOR PERICARDITIS.

Over 25 years ago, nonsteroidal anti-inflammatorydrugs (NSAIDs) were shown to be effective in a ran-domized trial of patients with post-pericardiotomysyndrome (79). In a single-center, double-blind trial,149 patients were randomized to ibuprofen or indo-methacin versus placebo. The diagnosis required 2 of3 characteristics, including fever, anterior chest pain,and a friction rub. The primary outcome was definedas resolution of 2 of these characteristics at 48 h.Ibuprofen and indomethacin had similar efficacy,90% and 89%, respectively, and both were signifi-cantly more effective than placebo (63%).

Even before this pivotal randomized trial, NSAIDshad been first-line therapy for many patients withpericarditis (80), although a randomized trial ofNSAIDs in post-viral pericarditis has never been per-formed. In routine clinical practice, aspirin or anNSAID, in conjunction with gastroprotection, is givenfor rapid control of symptoms (Central Illustration) (5).Initially, an attack dose is recommended every 8 h toalleviate symptoms, and usually continues until CRPhas normalized (7,33). This approach allows for indi-vidualized therapy, and tapering typically begins

after the patient is quiescent in terms of symptomsand inflammatory markers (3).

More recently, when added to an NSAID, colchicinehas also become an established first-line therapy forpericarditis (81). An ancient drug, colchicine wasapproved by the U.S. Food and Drug Administrationfor the treatment of FMF and gout in 2009, althoughcolchicine had been used to treat the latter conditionfor centuries (82). Thirty years ago, after noting suc-cess in treating and preventing the polyserositis at-tacks in FMF, colchicine was used to effectively treat3 patients with recurrent pericarditis (83).

In the intervening decades, several multicentertrials have shown that colchicine can hasten theresponse to medical therapy, increase the duration ofremission, and decrease the recurrence risk(11,12,20,21). Overall, in a meta-analysis including 8randomized trials and 1,635 patients, colchicinedecreased the relative risk of recurrent pericarditis byabout 50% (84). In clinical trials, gastrointestinalintolerance develops in 5% to 10% of patients, andtolerance may be improved if colchicine is used at alower dose without loading. This intolerance maylimit its use, but in general, colchicine is recom-mended for 3 months in patients with acute pericar-ditis and for at least 6 months in patients withrecurrent pericarditis (Central Illustration) (5).

Given the rapid resolution of symptoms and peri-cardial effusions, corticosteroids were oncecommonly used to treat pericarditis. Unfortunately,unopposed corticosteroids increase the risk of recur-rence and prolong the course of disease (58,85). Inparticular, high doses at an equivalence of 1.0 to1.5 mg/kg/day of prednisone are associated with se-vere side effects in 25% of patients. Furthermore,brisk tapering can lead to relapse (86). Conversely,when the combination of colchicine and an NSAID isineffective, low-dose corticosteroids may preventfurther recurrences (87).

Consequently, corticosteroids should be consid-ered only after failure of NSAIDs and colchicine,although there are a few notable exceptions. Forexample, patients with a systemic autoimmune dis-ease may benefit from corticosteroids, especially ifalready on a maintenance dose for another indication.In addition, pregnant patients and patients with renalfailure may need to avoid NSAIDs and colchicine.Similarly, concomitant NSAIDs are a relative contra-indication for patients who take anticoagulants.

In patients on corticosteroids, after remission withresolution of symptoms and normalization of CRP isattained, the dose must be slowly tapered. Often,pericarditis recurs at doses of prednisone <15 mg/day.If tolerable, the dose of NSAID or colchicine should be

TABLE 3 Emerging Therapies for Steroid-Dependent and

Colchicine-Refractory Pericarditis

Therapy Initial Dosing Duration

Azathioprine Started at 1 mg/kg/day, then graduallyincreased to 2 to 3 mg/kg/day

At least 6 months

Human immunoglobulins 400 to 500 mg/kg/day iv dailyfor 5 days

5 days, possiblyrepeated course

Anakinra 1 to 2 mg/kg/day up to 100 mg/dayin adults

At least 6 months

Pericardiectomy Not applicable Not applicable

iv ¼ intravenous.


2325

increased, rather than increasing the corticosteroid(88). In difficult situations, this approach may not bepossible, and steroid-sparing regimens should beconsidered (Central Illustration).

NOVEL TREATMENTS FOR REFRACTORY PERICARDITIS.

In this subset of patients with pericarditis that is re-fractory to standard medical therapy, novel treat-ments are needed not only to control symptoms, butalso to avoid the long-term side effects of glucocor-ticoids. Unfortunately, the evidence to support otheranti-inflammatory therapies in recalcitrant pericar-ditis consists primarily of case series and expertopinion (88). Among the emerging treatments, 3promising options include azathioprine, humanintravenous immunoglobulin (IVIG), and anakinra(Table 3).

Although commonly used to treat organ transplantrecipients and patients with autoimmune disease, thelargest study of azathioprine in recurrent pericarditisis a single-center retrospective report of 46 patients(89). At a dose of 1.5 to 2.5 mg/kg/day and a meanduration of treatment of over 1 year, azathioprine wasassociated with stable remission following steroiddiscontinuation in more than 50% of patients.Azathioprine was reasonably well-tolerated, withtransient hepatocellular dysfunction in 10%, leuko-penia in 7%, and transient gastrointestinal symptomsin 5% of patients. Overall, azathioprine is a poten-tially inexpensive and efficacious therapy in patientswho do not respond to conventional treatments. Inparticular, azathioprine may facilitate a gradualtapering of corticosteroids.

In select cases, IVIG may be an effective treatmentfor refractory pericarditis. In patients with a history ofmultiple recurrences, especially in the setting of anunderlying autoimmune disease, IVIG can act rapidlyto improve symptoms during an acute attack (90).Typically, patients receive infusions of 400 to 500mg/kg/day for 5 days, and if necessary, this regimencan be repeated after 1 month (91). However, despitebroader use with a good safety profile in rheumato-logical diseases, IVIG therapy has been reported inonly 30 patients with recurrent pericarditis (90).Given this limited experience, coupled with a highcost, IVIG is currently best reserved for pericarditispatients who have an indication on the basis of anunderlying autoimmune disease.

Among biological agents, anakinra is increasinglyused to treat patients with pericarditis that is re-fractory to NSAIDs, colchicine, and corticosteroids.An IL-1 receptor antagonist, anakinra is currentlyapproved by the U.S. Food and Drug Administrationfor the treatment of rheumatoid arthritis. Given the

central role of IL-1 in perpetuating an auto-inflammatory state, anakinra may have a role in thetreatment of recurrent idiopathic pericarditis.Currently, anakinra is administered as a daily sub-cutaneous injection at 1 to 2 mg/kg/day, up to 100 mg,for at least several months, although the optimalduration is unknown. With this regimen, efficacy hasbeen demonstrated in children and adults, althoughthe number of treated patients is small (92–94). Withanakinra, symptoms typically improve rapidly, butunfortunately, recurrence is common after the drug isdiscontinued. Given the concern that patients mayneed to be on a subcutaneous immunomodulatorydrug for the long-term, coupled with the high costand limited published experience, anakinra iscurrently reserved for patients with the most debili-tating and refractory pericarditis, especially if theyare corticosteroid-dependent and colchicine-resistant(Central Illustration). Even though these patientsrepresent a small percentage of the overall cohort ofpatients with pericarditis, they experience severemorbidity related to their pericarditis, and clinicaltrials are needed to define the optimal disease-modifying treatment.

Finally, after medical therapy has failed, peri-cardiectomy is rarely considered for refractory painrelated to recurrent pericarditis (Central Illustration).To attenuate the pain, as much pericardium as can beaccessed should be removed. Therefore, radical per-icardiectomy with resection extending beyond thephrenic nerves is often performed. In a single-centerretrospective study, 58 patients had pericardiectomyafter failed medical therapy (95). There were noperioperative deaths, and after a mean follow-up of5 years, the surgical group had a decreased relapserate compared with medically treated patients. Again,like the other observational studies mentioned,uncontrolled confounding could explain this associ-ation. Nonetheless, these preliminary data argue thatpericardiectomy could be considered as a potential



2326

treatment in prospective trials of patients with re-fractory recurrent pericarditis.

CONCLUSIONS AND FUTURE DIRECTIONS

If treated with appropriate anti-inflammatory ther-apy, the majority of patients with acute pericarditiswill have a benign course. About 15% of patients willhave myocardial involvement and 1% to 2% will havepericardial tamponade, but following acute manage-ment, these patients have a course that is indistinctfrom patients without these complications (20,30,31).After acute pericarditis, complicated pericarditis willdevelop in 15% to 30% of patients, and risks forrecurrence include early use of corticosteroids, a lackof response to NSAIDs, and a high CRP (33). Most ofthese patients will also achieve sustained remission,but unfortunately, some patients will develop multi-ple debilitating recurrences. These patients have achronic disease with substantial morbidity related notonly to intractable symptoms, but also to side effectsfrom medications, specifically long-term corticoste-roid use.

In patients with complicated pericarditis, a betterunderstanding of the causes and pathological pro-gression of pericardial inflammation is needed. Aftersuccess was observed in FMF, colchicine was given topatients with recurrent idiopathic pericarditis (83),and clinical trials have subsequently shown aconsistent relative risk reduction of w50% (84). Thisclinical success has led to a focus on the inflamma-some, a site of action for colchicine. In addition, givenits central role in periodic febrile syndromes thathave phenotypic similarities to recurrent idiopathic

pericarditis, the inflammasome may be an importanttarget for future therapies. This emerging focus onthe inflammasome has also highlighted an importantdistinction in patients with complicated pericardialdisease: those with a tendency toward inappropriateautoinflammation versus those with predominantautoimmune disease.

Finally, although still in its infancy given thelimitations noted, CMR has the potential to provideinsight into the progression of pericarditis by im-aging pericardial thickness, edema, and inflamma-tion. Currently, the assessment of pericardialinflammation with delayed enhancement imaging isessentially binary. A patient either does or does nothave significant pericardial delayed enhancement.The patient with an inflamed pericardium warrantsanti-inflammatory therapy or, in the setting ofconstrictive pericarditis, deferral of pericardiectomyto observe the response to intensified medicaltherapy. In the future, CMR may help to furtherstage pericarditis and guide both the intensity andduration of treatment. Thus, in conjunction withclinical data, CMR could identify the patients whoare at risk for recalcitrant and protracted pericar-ditis. These patients are most in need of noveltherapies and could be the focus of multicentertrials and registries.

REPRINT REQUESTS AND CORRESPONDENCE: Dr.Allan L. Klein, Center for the Diagnosis and Treat-ment of Pericardial Diseases, Heart and VascularInstitute, Cleveland Clinic, 9500 Euclid Avenue, DeskJ1, Cleveland, Ohio 44195. E-mail: [email protected].

RE F E RENCE S

1. Mozaffarian D, Benjamin EJ, Go AS, et al., forthe American Heart Association Statistics Com-mittee; Stroke Statistics Subcommittee. Heartdisease and stroke statistics—2016 update: areport from the American Heart Association. Cir-culation 2016;133:e38–360.

2. Imazio M, Spodick DH, Brucato A, et al.Controversial issues in the management of peri-cardial diseases. Circulation 2010;121:916–28.

3. Imazio M, Gaita F, LeWinter M. Evaluation andtreatment of pericarditis: a systematic review[erratum in JAMA 2015;314:1978]. JAMA 2015;314:1498–506.

4. LeWinter MM. Clincal practice. Acute pericar-ditis. N Engl J Med 2014;371:2410–6.

5. Adler Y, Charron P, Imazio M, et al. 2015 ESCguidelines for the diagnosis and management ofpericardial diseases. Eur Heart J 2015;36:2921–64.

6. Imazio M, Demichelis B, Parrini I, et al.Day-hospital treatment of acute pericarditis: a

management program for outpatient therapy. J AmColl Cardiol 2004;43:1042–6.

7. Imazio M, Gaita F. Diagnosis and treatment ofpericarditis. Heart 2015;101:1159–68.

8. Klein AL, Abbara S, Agler DA, et al. AmericanSociety of Echocardiography clinical recommen-dations for multimodality cardiovascular imagingof patients with pericardial disease: endorsed bythe Society for Cardiovascular Magnetic Reso-nance and Society of Cardiovascular ComputedTomography. J Am Soc Echocardiogr 2013;26:965–1012.e15.

9. Cosyns B, Plein S, Nihoyanopoulos P, et al.European Association of Cardiovascular Imaging(EACVI) position paper: multimodality imaging inpericardial disease. Eur Heart J Cardiovasc Imaging2015;16:12–31.

10. Imazio M, Bobbio M, Cecchi E, et al. Colchicinein addition to conventional therapy for acutepericarditis: results of the COlchicine for acute

PEricarditis (COPE) Trial. Circulation 2005;112:2012–6.

11. ImazioM, BobbioM, Cecchi E, et al. Colchicine asfirst-choice therapy for recurrentpericarditis: resultsof the CORE (COlchicine for REcurrent PericarditisTrial). Arch Intern Med 2005;165:1987–91.

12. Imazio M, Brucato A, Cemin R, et al., for theCORP (COlchicine for Recurrent Pericarditis) In-vestigators. Colchicine for Recurrent Pericarditis(CORP): a randomized trial. Ann Intern Med 2011;155:409–14.

13. Mayosi BM, Burgess LJ, Doubell AF. Tubercu-lous pericarditis. Circulation 2005;112:3608–16.

14. Mayosi BM, Ntsekhe M, Bosch J, et al., for theIMPI Trial Investigators. Prednisolone and Myco-bacterium indicus pranii in tuberculous pericar-ditis. N Engl J Med 2014;371:1121–30.

15. Imazio M, Hoit BD. Post-cardiac injury syn-dromes. An emerging cause of pericardial diseases.Int J Cardiol 2013;168:648–52.

mailto:[email protected]

http://refhub.elsevier.com/S0735-1097(16)35320-7/sref1





























































2327

16. Gouriet F, Levy PY, Casalta JP, et al. Etiologyof pericarditis in a prospective cohort of 1162cases. Am J Med 2015;128:784.e1–8.

17. Imazio M, Brucato A, Doria A, et al. Antinuclearantibodies in recurrent idiopathic pericarditis:prevalence and clinical significance. Int J Cardiol2009;136:289–93.

18. Prasad M, Hermann J, Gabriel SE, et al. Car-diorheumatology: cardiac involvement in systemicrheumatic disease. Nat Rev Cardiol 2015;12:168–76.

19. Hoffman M, Fried M, Jabareen F, et al. Anti-heart antibodies in postpericardiotomy syndrome:cause or epiphenomenon? A prospective, longitu-dinal pilot study. Autoimmunity 2009;35:241–5.

20. Imazio M, Brucato A, Cemin R, et al., for theICAP Investigators. A randomized trial of colchi-cine for acute pericarditis. N Engl J Med 2013;369:1522–8.

21. Imazio M, Belli R, Brucato A, et al. Efficacy andsafety of colchicine for treatment of multiple re-currences of pericarditis (CORP-2): a multicentre,double-blind, placebo-controlled, randomisedtrial. Lancet 2014;383:2232–7.

22. Shahar A, Hod H, Barabash GM, et al. Disap-pearance of a syndrome: Dressler’s syndrome inthe era of thrombolysis. Cardiology 1994;85:255–8.

23. Imazio M, Negro A, Belli R, et al. Frequencyand prognostic significance of pericarditisfollowing acute myocardial infarction treated byprimary percutaneous coronary intervention. Am JCardiol 2009;103:1525–9.

24. Imazio M, Trinchero R, Brucato A, et al., for theCOPPS Investigators. COlchicine for the Preven-tion of the Post-pericardiotomy Syndrome(COPPS): a multicentre, randomized, double-blind,placebo-controlled trial. Eur Heart J 2010;31:2749–54.

25. Imazio M, Brucato A, Ferrazzi P, et al., for theCOPPS-2 Investigators. Colchicine for preventionof postpericardiotomy syndrome and post-operative atrial fibrillation; the COPPS-2 random-ized clinical trial. JAMA 2014;312:1016–23.

26. Syed FF, Schaff HV, Oh JK. Constrictivepericarditis—a curable diastolic heart failure. NatRev Cardiol 2014;11:530–44.

27. Imazio M, Brucato A, Maestroni S, et al. Risk ofconstrictive pericarditis after acute pericarditis.Circulation 2011;124:1270–5.

28. Haley JH, Tajik AJ, Danielson GK, et al. Tran-sient constrictive pericarditis: causes and naturalhistory. J Am Coll Cardiol 2004;43:271–5.

29. Kytö V, Sipilä J, Rautava P. Clinical profile andinfluences on outcomes in patients hospitalizedfor acute pericarditis. Circulation 2014;130:1601–6.

30. Imazio M, Brucato A, Barbieri A, et al. Goodprognosis for pericarditis with and withoutmyocardial involvement: results from a multi-center, prospective cohort study. Circulation 2013;128:42–9.

31. Imazio M, Cecchi E, Demichelis B, et al. Myo-pericarditis versus viral or idiopathic acute peri-carditis. Heart 2008;94:498–501.

32. Imazio M, Cecchi E, Demichelis B, et al. In-dicators of poor prognosis of acute pericarditis.Circulation 2007;115:2739–44.

33. Imazio M, Brucato A, Maestroni S, et al. Prev-alence of C-reactive protein elevation and timecourse of normalization in acute pericarditis: im-plications for the diagnosis, therapy, and prog-nosis of pericarditis. Circulation 2011;123:1092–7.

34. Bertog SC, Thambidorai SK, Parakh K, et al.Constrictive pericarditis: etiology and cause-specific survival after pericardiectomy. J Am CollCardiol 2004;43:1445–52.

35. Douglas PS, Garcia MJ, Haines DE, et al.ACCF/ASE/AHA/ASNC/HFSA/HRS/SCAI/SCCM/SCCT/SCMR 2011 appropriate use criteria for echocardi-ography: a report of the American College ofCardiology Foundation Appropriate Use Criteria TaskForce, American Society of Echocardiography,American Heart Association, American Society ofNuclear Cardiology, Heart Failure Society of Amer-ica, Heart Rhythm Society, Society for Cardiovascu-lar Angiography and Interventions, Society ofCritical Care Medicine, Society of CardiovascularComputed Tomography, and Society for Cardiovas-cular Magnetic Resonance. J Am Coll Cardiol 2011;57:1126–66.

36. Kramer CM, Barkhausen J, Flamm SD, et al.,for the Society for Cardiovascular Magnetic Reso-nance Board of Trustees. Task Force on Stan-dardized Cardiovascular Magnetic Resonance.Standardized cardiovascular magnetic resonance(CMR) protocols 2013 update. J Cardiovasc MagnReson 2013;15:91.

37. Cremer PC, Kwon DH. Multimodality imagingof pericardial disease. Curr Cardiol Rep 2015;17:24.

38. Bogaert J, Francone M. Pericardial disease:value of CT and MR imaging. Radiology 2013;267:340–56.

39. Hatle LK, Appleton CP, Popp RL. Differentia-tion of constrictive pericarditis and restrictivecardiomyopathy by Doppler echocardiography.Circulation 1989;79:357–70.

40. Garcia MJ, Rodriguez L, Ares M, et al. Differ-entiation of constrictive pericarditis from restric-tive cardiomyopathy: assessment of leftventricular diastolic velocities in longitudinal axisby Doppler tissue imaging. J Am Coll Cardiol 1996;27:108–14.

41. Reuss CS, Wilansky SM, Lester SJ, et al. Usingmitral ‘annulus reversus’ to diagnose constrictivepericarditis. Eur J Echocardiogr 2009;10:372–5.

42. Kusunose K, Dahiya A, Popovi�c ZB, et al.Biventricular mechanics in constrictive pericarditiscomparison with restrictive cardiomyopathy andimpact of pericardiectomy. Circ Cardiovasc Imag-ing 2013;6:399–406.

43. Sengupta PP, Krishnamoorthy VK,Abhayaratna WP, et al. Disparate patterns of leftventricular mechanics differentiate constrictivepericarditis from restrictive cardiomyopathy. J AmColl Cardiol Img 2008;1:29–38.

44. Welch TD, Ling LH, Espinosa RE, et al. Echo-cardiographic diagnosis of constrictive pericarditis:Mayo Clinic criteria. Circ Cardiovasc Imaging 2014;7:526–34.

45. Francone M, Dymarkowski S, Kalantzi M, et al.Assessment of ventricular coupling with real-timecine MRI and its value to differentiate constrictivepericarditis from restrictive cardiomyopathy. EurRadiol 2006;16:944–51.

46. Bolen MA, Rajiah P, Kusunose K, et al. CardiacMR imaging in constrictive pericarditis: multi-parametric assessment in patients with surgicallyproven constriction. Int J Cardiovasc Imaging2015;31:859–66.

47. Ferrans VJ, Ishihara T, Roberts WC. Anatomyof the pericardium. In: Spodick DH, editor. Peri-cardial Disease. New York: Raven Press, 1982:15–39.

48. Bogaert J, Francone M. Cardiovascular mag-netic resonance in pericardial diseases.J Cardiovasc Magn Reson 2009;11:14.

49. Cremer PC, Tariq MU, Karwa A, et al. Quanti-tative assessment of pericardial delayed hyper-enhancement predicts clinical improvement inpatients with constrictive pericarditis treated withanti-inflammatory therapy. Circ Cardiovasc Imag-ing 2015;8:e003125.

50. Young PM, Glockner JF, Williamson EE, et al.MR imaging findings in 76 consecutive surgicallyproven cases of pericardial disease with CT andpathologic correlation. Int J Cardiovasc Imaging2011;28:1099–109.

51. Taylor AM, Dymarkowski S, Verbeken EK, et al.Detection of pericardial inflammation with late-enhancement cardiac magnetic resonance imag-ing: initial results. Eur Radiol 2006;16:569–74.

52. Mewton N, Liu CY, Croisille P, et al. Assess-ment of myocardial fibrosis with cardiovascularmagnetic resonance. J Am Coll Cardiol 2011;57:891–903.

53. Zurick AO, Bolen MA, Kwon DH, et al. Peri-cardial delayed hyperenhancement with CMR im-aging in patients with constrictive pericarditisundergoing surgical pericardiectomy; a case serieswith histopathologic correlation. J Am Coll CardiolImg 2011;4:1180–91.

54. Imazio M, Demichelis B, Parrini I, et al.Recurrent pain without objective evidence of dis-ease in patients with previous idiopathic or viralacute pericarditis. Am J Cardiol 2004;94:973–5.

55. Alraies MC, AlJaroudi W, Yarmohammadi H,et al. Usefulness of cardiac magnetic resonance-guided management in patients with recurrentpericarditis. Am J Cardiol 2015;115:542–7.

56. Feng D, Glockner J, Kim K, et al. Cardiacmagnetic resonance imaging pericardial late gad-olinium enhancement and elevated inflammatorymarkers can predict the reversibility of constrictivepericarditis after antiinflammatory medical ther-apy: a pilot study. Circulation 2011;124:1830–7.

57. Imazio M. Idiopathic recurrent pericarditis asan immune-mediated disease: current insights intopathogenesis and emerging treatment options.Exp Rev Clin Immunol 2014;10:1487–92.

58. Artom G, Koren-Morag N, Spodick DH, et al.Pretreatment with corticosteroids attenuates theefficacy of colchicine in preventing recurrentpericarditis: a multi-centre all-case analysis. EurHeart J 2005;26:723–7.







































































































































































































2328

59. Cusick MF, Libbey JE, Fujinami RS. Molecularmimicry as a mechanism of autoimmune disease.Clin Rev Allerg Immunol 2012;42:102–11.

60. Sfriso P, Ghirardello A, Botsios C, et al. In-fections and autoimmunity: the multifacetedrelationship. J Leukoc Biol 2010;87:385–95.

61. Park H, Bourla AB, Kastner DL, et al. Lightingthe fires within: the cell biology of auto-inflammatory diseases. Nat Rev Immunol 2012;12:570–80.

62. van Kempen TS, Wenink MH, Leijten EFA, et al.Perception of self: distinguishing autoimmunityfrom autoinflammation. Nat Rev Rheumatol 2015;11:483–92.

63. Martinon F, Pétrilli V, Mayor A, et al. Gout-associated uric acid crystals activate the NALP3inflammasome. Nature 2006;440:237–41.

64. Muruve DA, Pétrilli V, Zaiss AK, et al. Theinflammasome recognizes cytosolic microbial andhost DNA and triggers an innate immune response.Nature 2008;452:103–7.

65. Duewell P, Kono H, Rayner KJ, et al. NLRP3inflammasomes are required for atherogenesis andactivated by cholesterol crystals. Nature 2010;464:1357–61.

66. McDermott MF, Aksentievich I, Galon J, et al.Germline mutations in the extracellular domains ofthe 55 kDa TNF receptor, TNFR1, define a family ofdominantly inherited autoinflammatory syn-dromes. Cell 1999;97:133–44.

67. Lachmann HJ, Papa R, Gerhold K, et al., for thePaediatric Rheumatology International TrialsOrganisation (PRINTO), the EUROTRAPS and theEurofever Project. The phenotype of TNFreceptor-associated autoinflammatory syndrome(TRAPS) at presentation: a series of 158 cases fromthe Eurofever/EUROTRAPS international registry.Ann Rheum Dis 2014;73:2160–7.

68. Cantarini L, Lucherini OM, Baldari CT, et al.Familial clustering of recurrent pericarditis maydisclose tumour necrosis factor receptor-associated periodic syndrome. Clin Exp Rheuma-tol 2010;28:405–7.

69. Cantarini L, Rigante D, Merlini G, et al. Theexpanding spectrum of low-penetrance TNFRSF1Agene variants in adults presenting with recurrent in-flammatory attacks: clinical manifestations and long-term follow-up. SemArthritis Rheum2014;43:818–23.

70. Booty MG, Chae JJ, Masters SL, et al. FamilialMediterranean fever with a single MEFV mutation:where is the second hit? Arthritis Rheum 2009;60:1851–61.

71. Yu JW, Wu J, Zhang Z, et al. Cryopyrin andpyrin activate caspase-1, but not NF-kB, via ASColigomerization. Cell Death Diff 2006;13:236–49.

72. Ozen S, Demirkaya E, Amaryan G, et al., for thePaediatric Rheumatology International TrialsOrganisation (PRINTO), Eurofever Project. Resultsfrom a multicentre international registry of fa-milial Mediterranean fever: impact of environmenton the expression of a monogenic disease in chil-dren. Ann Rheum Disease 2014;73:662–7.

73. Leung YY, Hui LLY, Kraus VB. Colchicine—update on mechanisms of action and therapeuticuses. Semin Arthritis Rheum 2015;45:341–50.

74. Miner JJ, Kim AHJ. Cardiac manifestations ofsystemic lupus erythematosus. Rheum DiseaseClin North Amer 2014;40:51–60.

75. Doria A, Iaccarino L, Sarzi-Puttini P, et al.Cardiac involvement in systemic lupus erythema-tosus. Lupus 2005;14:683–6.

76. Voskuyl AE. The heart and cardiovascularmanifestations in rheumatoid arthritis. Rheuma-tology (Oxford) 2006;45 Suppl 4:iv4–7.

77. Nathan C. Neutrophils and immunity: chal-lenges and opportunities. Nat Rev Immunol 2006;6:173–82.

78. Yang CA, Chiang BL. Inflammasomes and hu-man autoimmunity: a comprehensive review.J Autoimmun 2015;61:1–8.

79. Horneffer PJ, Miller RH, Pearson TA, et al. Theeffective treatment of postpericardiotomy syn-drome after cardiac operations. A randomizedplacebo-controlled trial. J Thorac Cardiovasc Surg1990;100:292–6.

80. Lotrionte M, Biondi-Zoccai G, Imazio M, et al.International collaborative systematic review ofcontrolled clinical trials on pharmacologic treat-ments for acute pericarditis and its recurrences.Am Heart J 2010;160:662–70.

81. Imazio M. Colchicine for pericarditis. TrendsCardiovasc Med 2015;25:129–36.

82. Imazio M, Gaita F. Colchicine for cardiovascu-lar medicine. Future Cardiol 2016;12:9–16.

83. Rodríguez de la Serna A, Guindo Soldevila J,Martí Claramunt V, et al. Colchicine for recurrentpericarditis. Lancet 1987;2:1517.

84. Verma S, Eikelboom JW, Nidorf SM, et al.Colchicine in cardiac disease: a systematic reviewand meta-analysis of randomized controlled trials.BMC Cardiovasc Disord 2015;15:96.

85. ImazioM,AdlerY. Treatmentwithaspirin,NSAID,corticosteroids, and colchicine in acute and recurrentpericarditis. Heart Fail Rev 2013;18:355–60.

86. Imazio M, Brucato A, Cumetti D, et al. Corti-costeroids for recurrent pericarditis: high versuslow doses: a nonrandomized observation. Circu-lation 2008;118:667–71.

87. Imazio M, Brucato A, Trinchero R, et al.Corticosteroid therapy for pericarditis: a double-edged sword. Nat Clin Pract Cardiovasc Med2008;5:118–9.

88. Imazio M, Lazaros G, Brucato A, et al. Recur-rent pericarditis: new and emerging therapeuticoptions. Nat Rev Cardiol 2016;13:99–105.

89. Vianello F, Cinetto F, Cavraro M, et al.Azathioprine in isolated recurrent pericarditis: asingle centre experience. Int J Cardiol 2011;147:477–8.

90. Imazio M, Lazaros G, Picardi E, et al. Intrave-nous human immunoglobulins for refractoryrecurrent pericarditis: a systematic review of allpublished cases. J Cardiovasc Med (Hagerstown)2016;17:263–9.

91. Moretti M, Buiatti A, Merlo M, et al. Usefulnessof high-dose intravenous human immunoglobulinstreatment for refractory recurrent pericarditis. AmJ Cardiol 2013;112:1493–8.

92. Picco P, Brisca G, Traverso F, et al. Successfultreatment of idiopathic recurrent pericarditis inchildren with interleukin-1b receptor antagonist(anakinra): an unrecognized autoinflammatorydisease? Arthritis Rheum 2009;60:264–8.

93. Lazaros G, Vasileiou P, Koutsianas C, et al.Anakinra for the management of resistantidiopathic recurrent pericarditis. Initial experi-ence in 10 adult cases. Ann Rheum Dis 2014;73:2215–7.

94. Lazaros G, Imazio M, Brucato A, et al. Ana-kinra: an emerging option for refractory idiopathicrecurrent pericarditis: a systematic review ofpublished evidence. J Cardiovasc Med (Hagers-town) 2016;17:256–62.

95. Khandaker MH, Schaff HV, Greason KL, et al.Pericardiectomy vs medical management in pa-tients with relapsing pericarditis. Mayo Clin Proc2012;87:1062–70.

KEY WORDS autoimmune diseases,cardiovascular magnetic resonance imaging,colchicine, inflammasome, inflammation,pericardiectomy

APPENDIX For supplemental videos andtheir legends, please see the online version ofthis article.




















































































































































Documents

Complicated Pericarditis