J A C C : C A R D I O V A S C U L A R I M A G I N G V O L . 1 1 , N O . 8 , 2 0 1 8

ª 2 0 1 8 B Y T H E A M 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 U N D A T I O N

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

Radiation-Associated Cardiac Disease

A Practical Approach to Diagnosis and Management

Milind Y. Desai, MD,a,* Christine L. Jellis, MD, PHD,a,* Rupesh Kotecha, MD,b Douglas R. Johnston, MD,a

Brian P. Griffin, MDa

ABSTRACT

ISS

Fro

Ins

Me

au

co

Ma

Radiation-associated cardiac disease (RACD) results in complex clinical presentations, unique management issues, and

increased morbidity and mortality. Patients typically present years or even decades after radiation exposure, with

delayed-onset cardiac damage sustained from high cumulative doses. Multimodality imaging is crucial to determine

the manifestations and severity of disease because symptoms are often nonspecific. Comprehensive screening using a

coordinated approach may enable early detection. However, timing of intervention should be carefully considered in

these patients because surgery is often complex and high-risk second surgeries should be minimized in the long-term.

This review aims to provide treating physicians with a comprehensive and clinically focused overview of RACD,

including clinical/imaging manifestations, multi-modality screening recommendations, and management options.

(J Am Coll Cardiol Img 2018;11:1132–49) © 2018 by the American College of Cardiology Foundation.

S everal radiation-sensitive thoracic malig-nancies have obvious spatial orientation to car-diovascular structures including the following:

breast cancer (particularly left-sided), Hodgkin lym-phoma, lung cancer, esophageal cancer, and variousother mediastinal tumors. Although survival andrecurrence data support radiation therapy, it canresult in sustained dose to cardiovascular structures,leading to radiation-associated cardiac disease(RACD) (1–3). Acute radiation damage to the hearthas been recognized since the 1920s, when high-dose, wide-field radiation therapy thoracic portalswere the norm and minimization of cardiovascularirradiation was not necessarily prioritized (4). Thedelayed cardiovascular effects of such therapieshave been recognized more recently, largely due tothe latency of presentation and particularly in thosetreated for Hodgkin’s disease 20 to 40 years ago (5).Contemporary radiation regimens incorporate provi-sions to optimize radiation delivery to the tumor,while minimizing repeated irradiation of surroundingnormal structures, including the heart (6). Theseinclude the following: shielding measures, advanced

N 1936-878X/$36.00

m the aHeart and Vascular Institute, Cleveland Clinic, Cleveland, Ohio; a

titute, Cleveland Clinic, Cleveland, Ohio. Dr. Desai acknowledges the

dicine and the Khouri Family philanthropic gift for Research in Earl

thors have reported that they have no relationships relevant to the cont

ntributed equally to this work and are joint first authors.

nuscript received December 26, 2017; revised manuscript received April 1

respiratory gating techniques with deep inspiratorybreath-holds and activated breathing control, usingsmaller repeated fractions, and using advancedtreatment algorithms with narrow tangential beamsat different angles (Figures 1 and 2, Table 1). Radia-tion alternatives, such as proton therapy, arealso proving to be efficacious and less cardiotoxicoptions (7). Although these measures will likelyreduce the risk of RACD, current outcomes in RACDstill remain considerably influenced by historicalpractices (8,9).

This review aims to provide a clinically focusedoverview of RACD, including clinical/imagingmanifestations, screening recommendations, andmanagement options. Data for this review wereidentified using MEDLINE, Current Contents, andPubMed and using the screening terms “radiation”and “heart disease.” Historical articles published inEnglish from the 1950s onward provide informationabout remote treatment practices, whereas currentmanagement recommendations are based upon arti-cles published during the last 20 years and our owncumulative institutional experience.

https://doi.org/10.1016/j.jcmg.2018.04.028

nd the bDepartment of Radiation Oncology, Cancer

Haslam Family Endowed Chair in Cardiovascular

y Detection of Coronary Artery Disease. All other

ents of this paper to disclose. *Drs. Desai and Jellis

2, 2018, accepted April 13, 2018.

AB BR E V I A T I O N S

AND ACRONYM S

AVR = aortic valve

replacement

CAD = coronary artery disease

CMR = cardiac magnetic

resonance

CT = computed tomography

ECG = electrocardiogram

LVEF = left ventricular

ejection fraction

PCI = percutaneous coronary

intervention

RACD = radiation-associated

cardiac disease

RAPD = radiation-associated

pulmonary disease

TAVR = transcatheter aortic

valve replacement

J A C C : C A R D I O V A S C U L A R I M A G I N G , V O L . 1 1 , N O . 8 , 2 0 1 8 Desai et al.A U G U S T 2 0 1 8 : 1 1 3 2 – 4 9 Radiation-Associated Cardiac Disease

1133

INCIDENCE AND MANIFESTATIONS OF RACD

RACD is a spectrum of deleterious effects, rangingfrom preclinical findings to symptomatic clinicaldisease (Table 2). Acute cardiac inflammation canoccur at the time of treatment, resulting inmyocarditis or pericarditis. If apparent, this maypotentially suggest that an individual has sustainedincreased cardiac dose, or is more susceptible tolonger-term RACD. Late cardiovascular effects man-ifest decades after treatment and result from diffuseinterstitial fibrosis and collagen deposition, alongwith luminal narrowing of both arteries and arteri-oles due to accumulation of myofibroblasts andresultant intimal proliferation. This can result in avariety of cardiovascular complications grouped un-der the umbrella of RACD including the following:myocardial fibrosis, valvular heart disease (regurgi-tation and/or stenosis), vasculopathy including cor-onary artery disease (CAD), pericardial disease, andconduction system dysfunction (Central Illustration).Clinically, there is often overlap of pathologiesmanifesting within individuals. This can contributeto significant management challenges, especiallybecause many patients have additional radiationdamage to their lung parenchyma and present withnonspecific symptoms, such as dyspnea or fatigue,

FIGURE 1 Radiotherapy Arrangements for Chest Radiation in Lymph

A

(A) Traditional anterior-posterior approach, (B) intensity-modulated rad

which can be difficult to differentiate fromcardiac disease and may affect cardiac sur-gical risk. Multimodality imaging is, there-fore, crucial to diagnose and differentiatethe concurrent pathologies, as well as toguide subsequent treatment.

MYOCARDIAL DISEASE. Myocardial damageappears related to total radiation dose,fraction size, and volume of heart in theradiotherapy field. Radiation-related celldamage can result in activation of acute in-flammatory cascades and development of apro-fibrotic milieu, translating into myocar-dial fibrosis, with reduced micro-vascularproliferation and density. This radiation-induced myocardial fibrosis can result inmyocardial dysfunction spanning progres-sive stages of diastolic dysfunction to overt

systolic heart failure. The prevalence of radiation-associated cardiomyopathy is w10%. The anteriorposition of the right ventricle makes it susceptible todamage, although this is often under-recognizeddue to relative wall thinness and suboptimalvisualization.

Biventricular radiation-associated fibrosis isdiffuse and typically follows a nonischemic pattern.

oma

B

iation therapy, where overlapping beams minimize cardiac dose.

FIGURE 2 Adjuvant Radiotherapy After Mastectomy for Breast Cancer

AAbsolute5500.0 cGy5000.0 cGy4500.0 cGy4000.0 cGy

Absolute5500.0 cGy5000.0 cGy4500.0 cGy4000.0 cGy

Isocenter

B

(A) Natural heart position within the radiotherapy field, (B) cardiac-sparing effect of the active-breathing control (ABC) device, where the heart moves outside the

radiation field.

TABLE 1 Adjuvant R

Using Opposed Tange

Dose Reduction

Heart maximum point

Heart mean dose (Gy)

Relative risk change (%

Risk of adverse cardiacBaseline risk ¼ 7.9%

Risk of ischemic heartBaseline risk ¼ 3.3%

ABC ¼ active-breathing co

Desai et al. J A C C : C A R D I O V A S C U L A R I M A G I N G , V O L . 1 1 , N O . 8 , 2 0 1 8

Radiation-Associated Cardiac Disease A U G U S T 2 0 1 8 : 1 1 3 2 – 4 9

1134

However, coexistent radiation-induced micro- andmacro-vascular disease can result in ischemia/infarction and regional replacement-type fibrosis.The detrimental effects of radiation to myocardialfunction can be compounded by the cardio-toxiceffects of chemotherapeutic agents, particularlyanthracyclines, frequently used alongside radiationtherapy (10). These effects appear independent, butcombine to increase the cumulative risk for cardio-toxicity (11). Newer monoclonal antibodies, such asthe HER-2/neu receptor antagonist trastuzumab,have also demonstrated cardiotoxic effects. Recentdata suggest that trastuzumab may have a radio-sensitizing effect on breast cancer cells, although itremains unclear if similar effects occur on normalhealthy cells and early studies monitoring for cardio-toxicity in subjects on combined trastuzumab andradiation therapy have been reassuring (12,13).

adiotherapy Dose After Left Radical Mastectomy for Breast Cancer

nt Fields (No ABC) Versus ABC Device (Marinko et al. [12])

Using ABC Device No ABC ABC

(Gy) 50.51 28.02

4.20 0.80

) per Darby et al. (8) 31.8 5.92

event by 80 yrs 10.41 8.37 (�2.04)

disease death by 80 yrs 4.35 3.50 (�0.85)

ntrol.

Greater awareness of this “multi-hit” phenomenonin oncology and the introduction of more standard-ized screening programs will hopefully allow earlyidentification of dysfunction and modification oftherapy (14).

VALVULAR DISEASE. Mediastinal radiation therapyis associated with significant valvular abnormalitiesranging from 7% to 39% at 10 years and 12% to 60%at 20 years, with mitral and aortic valves the mostaffected (15). This usually manifests as progressivevalve thickening and calcification, resulting in valverestriction presenting as stenosis or regurgitation.Radiation-associated valvular disease usually be-comes symptomatic later than coronary disease, w1to 2 decades after radiation exposure (16). Aware-ness of this latency is important, given thatasymptomatic survivors treated more than 20 yearsago remain at significantly increased risk of aorticregurgitation (60% vs. 4%), tricuspid regurgitation(4% vs. 0%), and aortic stenosis (16% vs. 0%)compared with patients treated within 10 years (17).Radiation-associated valvular thickening and calci-fication are more extensive and may affect multiplevalves. Surrounding structures, such as the valveannulus, subvalvular apparatus, and aorto-mitralcurtain (intervalvular fibrosa), are also frequentlyinvolved. Increasingly, aorto-mitral curtain thick-ening/calcification is being recognized as a hallmarkof previous heart irradiation and its extent isstrongly associated with mortality in subjects un-dergoing cardiac surgery (18) (Figure 3).

TABLE 2 RACD is a Spectrum of Deleterious Effects, Ranging From Preclinical Findings To Symptomatic Clinical Disease

Component of theHeart Effects of Radiation Therapy Examples of Clinical Sequelae

Myocardium Myocardial edema, myocardial inflammation, myocardial fibrosis Cardiomyopathy, systolic dysfunction, diastolic dysfunction, congestiveheart failure

Heart valves Valvular, subvalvular, annular, and aorti-mitral curtain thickening and/orcalcification, restricted leaflet mobility, reduced size of valve orifice

Valve regurgitation, valve stenosis, volume, or pressure overload

Pericardium Pericardial thickening, pericardial calcification, pericardial constriction Pericarditis (acute or chronic), pericardial effusion, chronic pericardialdisease, cardiac tamponade

Coronaryarteries

Myofibroblast replacement and enhanced platelet deposition, plaque rupture Coronary artery disease, accelerated atherosclerosis, coronary stenosis,angina pectoris, ischemic heart disease, myocardial infarction, restingand inducible regional wall abnormalities

Vascular Narrowing of capillary lumens, thrombosis, rupture of vessel walls, damage toendothelial cell membranes, reduced capillary density

Reduced collateral flow, reduced vascular reserve, microvascular disease,myocardial infarction, systemic embolization

Conductionsystem

Fibrosis of conduction pathways, compression from adjacent calcification Conduction defects including arrhythmias, heart block, and sick sinussyndrome; autonomic dysfunction

J A C C : C A R D I O V A S C U L A R I M A G I N G , V O L . 1 1 , N O . 8 , 2 0 1 8 Desai et al.A U G U S T 2 0 1 8 : 1 1 3 2 – 4 9 Radiation-Associated Cardiac Disease

1135

PERICARDIAL DISEASE. Although acute pericarditisis less common with modern radiation protocols,chronic pericarditis may manifest many years aftertreatment, with an estimated prevalence of 8% to30% (19). Patients who receive high doses of radia-tion are at risk for pericardial effusions (incidenceof 36%; 8% symptomatic) (20). Chronic pericardialinflammation can result in a thickened, rigid, andoften calcified pericardial sac. Loss of pericardialdistensibility can then result in ventricular inter-dependence and constrictive physiology. In prac-tice, given that most patients with RACD have atleast some degree of restrictive physiology due tothe myocardial fibrosis, it can be difficult todistinguish between restriction and constriction.Often, invasive evaluation with simultaneous rightand left heart catheterization is necessary todistinguish between the two by assessing hemody-namics. However, multimodality imaging isincreasingly providing a noninvasive alternative todetermine which pathology dominates. Advantagesof different modalities are addressed below and aretypically useful in combination, with particularlyrelevant techniques to differentiate betweenconstriction and restriction being the following:myocardial tissue velocities and strain imaging byecho, pericardial thickening and calcification bycomputed tomography (CT), and pericardialenhancement and septal shift on free breathingcardiac magnetic resonance (CMR) cine imaging(Figures 4 and 5, Online Video 1).

VASCULOPATHY. Radiation therapy is an indepen-dent risk factor for accelerated atherosclerosis (21).This manifests as both micro- and macrovasculardisease and can be progressive, despite not beingclinically apparent until years after treatment. Anyarteries or arterioles within the radiation fieldare potentially at risk. Doses of $0.50 Gy can

initiate atherosclerosis (22), via radiation-inducedinflammation, which results in accumulation ofmyofibroblasts, resulting in intimal proliferation withaggregation of lipid-rich macrophages. These inflam-matory, pro-thrombotic plaques are more likely torupture than more stable, less lipid-rich plaques. Thelikelihood of vasculopathy is magnified and acceler-ated by traditional cardiovascular risk factors,particularly hypercholesterolemia (23).

Radiation-induced coronary vasculopathy, with aprevalence of w85%, typically affects the ostia orproximal aspects of the epicardial coronary arteries,however, the proximal right coronary artery, mid leftanterior descending artery, and mid diagonalbranches are particularly involved among patientswith breast cancer and left-sided radiotherapy (24)(Figure 6). The predilection for ostial left main andright coronary artery likely relates to coronary arteryposition within the anterior radiation fieldand perhaps a greater propensity for intimal prolif-eration more proximally, which may relate toradiation-induced aortitis and fibrocalcific de-rangements. Myocardial ischemia from epicardialcoronary artery disease can compound concurrentmyocardial dysfunction due to radiation-inducedmyocardial fibrosis, thereby potentiating the risk forflash pulmonary edema, especially in those with leftmain disease.

The risk of myocardial infarction proportionallyincreases with duration from radiation exposure andis highest in those who received treatment whenyounger than 20 years of age (25). Microvasculardisease is less well studied, although it appears tocontribute to myocardial dysfunction via ischemiaand resultant fibrosis. Even at low radiation doses,microvascular injury results in reduced capillarydensity resulting in reduced vascular reserve furtherpotentiating macro-vascular ischemia (26).

CENTRAL ILLUSTRATION Multimodality Imaging Manifestations of Radiation-Associated Cardiac Disease

Desai, M.Y. et al. J Am Coll Cardiol Img. 2018;11(8):1132–49.

Listed clockwise 1 to 9: 1) coronary angiography demonstrating severe circumflex stenosis; 2) ‘porcelain’ ascending aortic calcification on cardiac computed tomog-

raphy (CT); 3) severe mitral annular, aorto-mitral curtain, and aortic valve calcification on cardiac CT and 4) transthoracic echocardiography; 5) near-transmural inferior

wall ischemic scar on cardiac magnetic resonance (CMR) in short-axis and 6) vertical long-axis planes; 7) complete heart block on electrocardiography; 8) severe

pericardial calcification on noncontrast CT, and 9) strain ‘bullseye’ plot demonstrating reduced left ventricular anterolateral wall deformation due to tethering.

Desai et al. J A C C : C A R D I O V A S C U L A R I M A G I N G , V O L . 1 1 , N O . 8 , 2 0 1 8

Radiation-Associated Cardiac Disease A U G U S T 2 0 1 8 : 1 1 3 2 – 4 9

1136

FIGURE 3 Latent Valvular Manifestations of Chest Radiation

A B

C D

A 52-year-old man treated with mantle radiation for Hodgkin lymphoma 25 years ago demonstrates: severe calcification of the aortic valve,

aorto-mitral curtain, and mitral valve on 2-dimensional echocardiography (A, arrows) and computed tomography (B, arrow), resulting in

severe mitral (C) and aortic (D) stenosis using Doppler echocardiography.

J A C C : C A R D I O V A S C U L A R I M A G I N G , V O L . 1 1 , N O . 8 , 2 0 1 8 Desai et al.A U G U S T 2 0 1 8 : 1 1 3 2 – 4 9 Radiation-Associated Cardiac Disease

1137

Large-vessel vasculopathy can involve the thoracicaorta and arch branch vessels, manifesting asatherosclerotic disease, although pro-thrombogenicfactors can also result in regional thrombosis leadingto vessel occlusion or embolic stroke (27). Aortic vas-culopathy may preclude management of other RACDby limiting percutaneous intervention or surgical ac-cess (porcelain aorta). In one RACD study, 59% ofpatients had ascending aortic calcification and 13%had severe circumferential calcification (28). In theinoperable PARTNER (Placement of Aortic TraNs-cathetER valve) trial cohort, 15.1% had a porcelainaorta, similar to a Canadian registry of high–/prohib-itive–surgical risk aortic stenosis patients (18.0%),with a high proportion likely due to prior mediastinalradiation (29,30). Severe ascending aortic calcification

may preclude surgical clamping or cannulation,whereas intraluminal atheroma may embolize duringcatheterization or surgical manipulation resulting instroke or peripheral embolic phenomena (31).CONDUCTION SYSTEM DYSFUNCTION. Mediastinalradiation can result in fibrosis of conduction path-ways and subsequent arrhythmias. Up to 75% of long-term survivors who received mediastinal radiationhave conduction defects on electrocardiogram (ECG)(32). Conduction system injury can be related todirect irradiation or secondary to myocardial inflam-mation, ischemia, or fibrosis. At the time of radio-therapy, transient, nonspecific repolarizationabnormalities are common, but typically asymptom-atic (33). Severe conduction abnormalities and ar-rhythmias are usually not evident until years/decades

FIGURE 4 Features of Pericardial Constriction

A

Septal e’= 12 cm/sec

Lateral e’= 7 cm/sec

Strain = –15.4%

R

APeak Systolic strain

H

F

P

L

R L

C E

B D F

(A) Annulus reversus with preservation of septal early diastolic tissue velocity (e’), (B) compared with reduced lateral wall e’; (C) reduced left ventricular free wall strain

due to tethering with reduced longitudinal motion (pink); (D) simultaneous right and left heart catheterization shows equalization of diastolic pressures between the

right and left ventricles (black arrow); cardiac computed tomography demonstrating pericardial calcification anteriorly (white arrow) and laterally extending into the

mitral annulus (yellow arrow) on (E) axial and (F) sagittal reconstructions. See Online Video 1.

Desai et al. J A C C : C A R D I O V A S C U L A R I M A G I N G , V O L . 1 1 , N O . 8 , 2 0 1 8

Radiation-Associated Cardiac Disease A U G U S T 2 0 1 8 : 1 1 3 2 – 4 9

1138

later. Infra-nodal and right bundle branch blocks aremost common, with the anteriorly located rightbundle being particularly susceptible. Ectopy andsustained arrhythmias (supraventricular and ven-tricular) are both more common after radiotherapycompared with age-matched control patients (33).Non-specific T-wave and ST-segment ECG changesare often noted incidentally years after irradiation(34). Autonomic dysfunction has been poorly studied,although inappropriate sinus tachycardia has beenrecognized as a sign of extensive RACD. The use ofECG or telemetry screening remains uncertain.

MULTIMODALITY IMAGING FOR

RISK STRATIFICATION, SCREENING,

AND DIAGNOSIS

Patients should be educated about risk of long-termRACD before treatment commencement. For thosewho will potentially receive >30 Gy, consideration

could be given to a cardiology consultation beforetreatment. This may be especially relevant to thosewho have additional risk factors including thefollowing: younger age at exposure, concomitantcardio-toxic chemotherapy, underlying structuralheart disease, and traditional cardiac risk factors ofdiabetes, hypertension, obesity, and smoking (35).Additional risk factors include type of prior radiationsource (e.g., cobalt) and time since exposure.

Given the protracted course and poor outcomes ofRACD, even despite surgery, serial evaluation ofcancer survivors with appropriate screeningprograms is recommended in the expert consensusdocument (35). We have incorporated these recom-mendations into practice and advocate that ongoingscreening and management should be performed byexpert physicians at centers experienced in RACD(Figure 7). Transition from pediatric to adultservices is also always important to ensure adequatefollow-up.

FIGURE 5 Distinguishing Features of a Restrictive Cardiomyopathy Due to Underlying Myocardial Fibrosis

A

Septal e’= 4 cm/sec

Lateral e’= 5 cm/sec

Strain = –9.1%

Peak Systolic Strain

C

BD

Reduced (A) septal and (B) lateral early diastolic tissue velocities (e’); (C) grade 3 restrictive mitral inflow, with a short E-wave (white arrow)

deceleration time (<150 ms) and small A-wave (yellow arrow); (D) reduced global two-dimensional longitudinal strain (pink region).

J A C C : C A R D I O V A S C U L A R I M A G I N G , V O L . 1 1 , N O . 8 , 2 0 1 8 Desai et al.A U G U S T 2 0 1 8 : 1 1 3 2 – 4 9 Radiation-Associated Cardiac Disease

1139

Reports suggest 42% of chest-irradiated patientshave significant asymptomatic valvular disease and14% have stress-induced myocardial ischemia,although the true incidence is likely higher due tounder-recognition (25,36). Typically, screening forCAD should commence within 5 years of radiationexposure (8). Due to its later presentation, screeningfor valvular disease is delayed until 10 years afterradiotherapy, with subsequent imaging then per-formed at 5-year intervals (35). Modifiable risk factorssuch as smoking, obesity, hypercholesterolemia, hy-perglycemia, and hypertension should be aggres-sively managed because they act synergistically withradiation exposure, to increase the risk for majorcoronary events from 2% to 7% (8). The potential roleof various imaging modalities in diagnosis and man-agement of RACD is discussed later in this article.Also, Table 3 lists the published data behind the roleof multimodality evaluation in RACD.

ECHOCARDIOGRAPHY. Echocardiography is themost common screening tool used for detection andmonitoring of RACD. Recommended frequency ofechocardiographic screening varies, but is typicallyperformed every 2 years in asymptomatic individualsand more often once symptomatic. Particular featuresof RACD on echo include the following: biventricularsystolic and diastolic dysfunction, multi-valvularinvolvement with mixed valvular dysfunction,prominent calcification (pericardial, valvular,annular, aorto-mitral curtain, and aortic), wall motionabnormalities, and pericardial disease. Specific fea-tures of pericardial constriction include thefollowing: bi-atrial enlargement, pericardial thick-ening (>3 mm) and calcification, myocardial teth-ering, an early diastolic septal bounce and plethora ofthe inferior vena cava and hepatic veins, withrestricted respiratory variation. A pericardial effusionmay be also noted. Significant respiratory variation of

FIGURE 7 Screening Algorithm for RACD

YES

NO

CARDIOLOGY CONSULT

Could consider dose minimization tospecific cardiac structures of concern,including pacemakers

• • • Optimization of modifiable factors

Treat pre-existing cardiac disease

Annual clinical history andexamination by physician

experienced in RACD

RADIATION THERAPY

Pre-radiation screening for cardiacdisease and cardiac risk factors. Include:

examination, electrocardiogram, andechocardiography

Patient referred for chestradiation therapy

YES

YES

Pre-existing cardiovascular condition

Specific assessment for signs andsymptoms of RACD including: Myocardial,

valvular, pericardial, coronary artery,vascular, or conduction system disease.

Screening for and optimizationof new or pre-existing cardiac

risk factors

Targeted further investigationwith appropriate resting and/or

stress imaging andelectrophysiological testing.

Management of specific cardiaccondition according to

guidelinerecommendations

5 years post-exposure –commence screening for CADwith consideration of stress

testing every 2 years.

10 years post-exposure – commencescreening for valvular heart disease with

echocardiography every 2 years.

Recommended algorithm for monitoring of cardiac disease before and after chest radiation therapy. CAD ¼ coronary artery disease;

RACD ¼ radiation-associated cardiac disease.

FIGURE 6 Radiation-Associated Coronary Artery Disease

LAO 27 CRAN 13 LAO 2 CRAN 26

Severe right coronary ostial stenosis (arrows) after radiotherapy for Hodgkin lymphoma (surgical clips from splenectomy).

Desai et al. J A C C : C A R D I O V A S C U L A R I M A G I N G , V O L . 1 1 , N O . 8 , 2 0 1 8

Radiation-Associated Cardiac Disease A U G U S T 2 0 1 8 : 1 1 3 2 – 4 9

1140

TABLE 3 Published Data Demonstrating the Role of Multimodality Assessment in Radiation-Associated Cardiac Disease

Modality (Ref. #) First Author Summary of Findings

Electrocardiography (33) Larsen et al. 12-lead and 24-h ambulatory electrocardiograms recorded in 73 patients who received anthracyclines, radiotherapy, orboth. Increased frequency of QTc prolongation, supraventricular premature complexes, supraventricular tachycardia,ventricular premature complexes, couplets, and ventricular tachycardia when compared with an age-matched healthypopulation.

(34) Strender et al. In 197 patients with prior chest radiotherapy, at 10-yr follow-up after radiotherapy for breast cancer, 45% demonstratedT-wave abnormalities, ST-segment depression, and ectopic beats.

Echocardiography (18) Desai et al. In 173 patients with documented RACD who underwent cardiothoracic surgery, greater AMC thickness independentlypredicted mortality, and the subgroup with AMC thickness of at least 0.6 cm was associated with highest mortality.

(39) Erven et al. In 75 patients, LV-GLS detected a subclinical decrease in cardiac function up to 14 months after left breast radiotherapy,despite no change on conventional echo parameters.

(40) Armstrong et al. Evaluation of 1,820 adult survivors of childhood cancer (chemotherapy ¼ 1,050, chest-directed radiotherapy ¼ 306 orboth ¼ 464) revealed that one-third of survivors with normal LVEF had cardiac dysfunction by strain imaging, diastolicgrading, or both. Incidence was doubled in those with metabolic syndrome.

(41) Chirakarnjanakornet al.

In 163 patients with documented RACD who underwent cardiac surgery, abnormal LV-GLS was associated with highermortality, despite preserved LVEF.

(58) Donnellan et al. 172 patients with prior mediastinal radiation presenting with severe aortic stenosis undergoing surgical aortic valvereplacement had significantly worse longer-term survival vs. a matched cohort (at 6 � 3 yrs of follow-up, there were48% deaths in radiotherapy vs. 7% in comparison group).

Stress imaging (48) Heidenreich et al. In 294 patients undergoing screening with stress echocardiography and radionuclide perfusion imaging for coronaryartery disease after mediastinal irradiation for Hodgkin disease, 21.4% demonstrated abnormal resting ventricularfunction and 14% inducible ischemia. Screening led to bypass graft surgery in 7 patients, whereas there were23 coronary events during 6.5 yrs of follow-up.

Chest CT (28) Desai et al. In 117 RACD patients undergoing cardiothoracic surgery, worsening pulmonary fibrosis, quantified on chest CT, wasindependently associated with increased mortality. Pulmonary complications were noted in 39% on follow-up.

(29) Leon et al. In the inoperable Placement of Aortic TraNscathetER valve trial cohort, 15.1% had a porcelain aorta, with a highproportion likely due to prior mediastinal radiotherapy.

(44) Kamdar et al. In this study of 167 patients, routine use of pre-operative chest CT to detect high-risk findings had a strong associationwith adoption of preventive surgical strategies in high-risk patients undergoing redo cardiac surgery.

CMR (49) Cremer et al. Article describing the role of CMR in pericardial disease.

Left heartCatheterization

(24) Nilsson et al. In 199 patients irradiated for left breast cancer, there was an increased risk for significant LAD (mid/distal), diagonalbranch (distal), and RCA (proximal) stenosis, with location of stenosis dependent on field of radiation.

Percutaneous coronaryintervention

(65) Reed et al. In a matched cohort study of 314 patients, compared with patients with typical atherosclerotic coronary artery disease,patients with radiation-associated coronary artery disease are at higher risk for mortality after PCI.

Arterial ultrasound andangiography

(51) van Son et al. In 10 patients, despite previous chest irradiation, the internal mammary artery remained a viable long-term conduit formyocardial revascularization, when preoperative assessment with ultrasound or angiography showed patency.

AMC ¼ aorto-mitral curtain; CMR ¼ cardiac magnetic resonance; CT ¼ computed tomography; LAD ¼ left anterior descending; LVEF ¼ left ventricular ejection fraction; PCI ¼ percutaneous coronaryintervention; RACD ¼ radiation-associated cardiac disease; RCA ¼ right coronary artery.

J A C C : C A R D I O V A S C U L A R I M A G I N G , V O L . 1 1 , N O . 8 , 2 0 1 8 Desai et al.A U G U S T 2 0 1 8 : 1 1 3 2 – 4 9 Radiation-Associated Cardiac Disease

1141

mitral/tricuspid valve inflow velocities and pulmo-nary or systemic venous flow patterns also supportsconstrictive physiology. Reduced lateral mitralannular e’ velocity, with preserved/increased septalmotion, suggests free wall tethering and results in aninverse relationship between the E/e’ ratio and fillingpressures (37,38).

In addition to left ventricular ejection fraction(LVEF), deformation imaging (strain and strain rate)enables further assessment of myocardial functionand can be useful to distinguish constrictive fromrestrictive cardiomyopathies (Figures 3 and 4). Inasymptomatic or subclinical disease, left ventricularstrain may be reduced, despite normal LVEF (39). Alink between previous chemo-radiotherapy exposure,abnormal strain, and diastolic dysfunction has alsobeen noted (40). Mortality is higher in those withabnormal strain, even when LVEF is normal, sodeformation imaging may be useful in identifying

those who may benefit from an earlier intervention(41).

Stress echocardiography enables evaluation formyocardial ischemia and dynamic assessment ofradiation-induced valvular heart disease. Exercisestress is preferred over pharmacological agents, tomimic normal physiological conditions. Radiation-induced CAD may manifest as resting or inducibleregional wall motion abnormalities in typical coronarydistributions. However, balanced ischemia frommulti-vessel disease may present with globaldysfunction and cavity enlargement at peak stress.Stress valvular assessment is typically reserved forsymptomatic subjects withmild ormoderate disease atrest, whose symptoms appear proportionally worsethan expected. Stress may demonstrate increasedvalvular regurgitation, trans-valvular gradients, orpulmonary pressures, along with impaired ventricularcontractile reserve (42). Simultaneous evaluation of

Desai et al. J A C C : C A R D I O V A S C U L A R I M A G I N G , V O L . 1 1 , N O . 8 , 2 0 1 8

Radiation-Associated Cardiac Disease A U G U S T 2 0 1 8 : 1 1 3 2 – 4 9

1142

functional capacity on cardiopulmonary testing pro-vides further stratification and can help differentiateunderlying radiation-induced pulmonary disease.Limitations of echocardiography in RACD include thefollowing: chest wall deformities, poor acousticwindows, and difficulty with breath holding due toconcurrent pulmonary disease and prominent tissuecalcification resulting in acoustic shadowing/artifact.

MULTIDETECTOR CARDIAC CT. Dedicated cardio-thoracic CT is also used for evaluation of aortic,valvular, myocardial, and pericardial calcification oneither contrast or noncontrast imaging. Preoperativeassessment for aortic calcification (e.g., porcelainaorta) is important to determine suitability for aorticcross-clamping and cannulation in patients withRACD undergoing cardiac surgery. Significantvalvular and/or annular calcification may precludevalvular repair. With recent advent of transcatheteraortic valve replacement (TAVR), 4-dimensional cinechest CT has become crucial for preprocedural plan-ning, including assessment of shape and size of theannuli (aortic, mitral, and tricuspid) and iliofemoralvasculature (43). Chest CT is also very useful toevaluate extra-cardiac structures for surgical plan-ning, especially for redo surgeries (44,45). Extensivemediastinal fibrosis or lack of a safety margin be-tween the sternum and adjacent structures maynecessitate abandoning a median sternotomyapproach for an alternative surgical/transcatheterstrategy. The presence and degree of pulmonaryfibrosis on CT has been demonstrated to adverselyaffect surgical risk and mortality in RACD (28). Peri-cardial calcification, thickening, inferior vena cavaenlargement, and ventricular conical deformity aresuggestive of pericardial constriction (46). Althoughretrospective CT acquisition enables multi-phase CTimaging to evaluate for constrictive physiology, ven-tricular volumes, and regional wall motion abnor-malities, the higher radiation dose means that it isusually reserved for those with contraindications toCMR. Finally, coronary CT angiography can be usefulin RACD for its negative predictive value, with noatherosclerosis, suggesting a very low risk (47).However, assessment for luminal stenosis becomesmore difficult in the presence of coronary calcifica-tion, due to calcium ‘blooming’ artifact.

NUCLEAR SCINTIGRAPHY. Various radionuclideimaging strategies, including single photon emissionCT and positron emission tomography, have beenused to assess myocardial ischemia in RACD. Studieshave been limited by small numbers and broad rangesof radiation dose and interval from therapy, butshow that 12% of asymptomatic patients have

stress-induced perfusion defects 15 � 7 years after amean radiation dose of 43.5 � 3.4 Gy (48). As with CT,the benefits of additional radiation exposure for im-aging purposes always require careful consideration.

CMR. There is a paucity of data regarding the use ofCMR in RACD. However, CMR can provide usefulsimultaneous functional and structural data, enablingdetection of radiation-induced coronary, valvular,and pericardial disease. Cine imaging using steady-state free precession gradient echo sequences allowsassessment of ventricular mass, volumes, function,and regional wall motion abnormalities, which canthen be correlated with late gadolinium enhancementto establish regions of viability, scar, and non-ischemic fibrosis. Other novel CMR techniques toassess regional function include deformation imagingwith tissue tagging or strain imaging. Valvular func-tion can be assessed using calculation of trans-valvular gradients and regurgitant volumes usingvelocity-encoded sequences. Multiple CMRsequences are useful for evaluation of radiation-induced pericardial thickening, effusions, and fea-tures of constrictive physiology including thefollowing: ventricular conical deformity, diastolicseptal bounce, diastolic chamber restraint, and infe-rior vena cava enlargement. A free breathingsequence can also assess for constriction-associatedrespiro-phasic septal shift, whereas increased peri-cardial signal intensity on edema weighted T2 imag-ing and late gadolinium enhancement suggestspericardial inflammation (49). First-pass perfusionimaging (using pharmacological stressors) can iden-tify underlying myocardial ischemia related toradiation-induced CAD. T1 mapping techniquesenable quantitation of diffuse fibrosis using native orpost-contrast myocardial T1 relaxation times. Utilityhas already been demonstrated in infiltrative cardio-myopathies and valvular heart disease and may proveuseful for radiation-induced myocardial disease, butrequires further investigation (50).

CMR can be precluded by claustrophobia and somemetal implants including noncompatible pace-makers/implantable defibrillators. Performance ofCMR may also be limited in the setting of cardiacfailure if the patient cannot lie supine, in radiation-induced pulmonary disease where recurrent breathholding may be problematic and in subjects withsignificant renal dysfunction where gadolinium iscontraindicated due to the risk for nephrogenic sys-temic sclerosis. Cardiac CMR is not optimized forassessment of surrounding structures including lungsand poorly shows calcification as a region of signalvoid. Although CMR is not routinely recommended in

J A C C : C A R D I O V A S C U L A R I M A G I N G , V O L . 1 1 , N O . 8 , 2 0 1 8 Desai et al.A U G U S T 2 0 1 8 : 1 1 3 2 – 4 9 Radiation-Associated Cardiac Disease

1143

all patients with suspected RACD and accessibilitymay be limited in some centers, it demonstrates mostuse for: 1) assessment of ischemic/nonischemicmyocardial fibrosis; 2) pericardial constriction; and 3)as an adjunct to echocardiography for quantificationof ventricular and valvular function, particularly insubjects with technically difficult acoustic windowsafter prior chest radiation and surgery.

LEFT AND RIGHT HEART CATHETERIZATION.

Invasive catheterization provides complementaryand confirmatory information to noninvasive imag-ing, especially in cases where invasive therapies arecontemplated. Left heart catheterization allowsassessment of coronary stenosis severity, as well asenabling intervention on amenable, discrete, prox-imal- to mid-vessel lesions. Right heart studies areuseful for calculation of intracardiac and pulmonarypressures in valvular RACD and radiation-associatedpulmonary disease (RAPD). Simultaneous left andright heart measurements allow for evaluation ofconstrictive physiology and cardiac index, with a lowindex prompting evaluation for restrictive cardio-myopathy. Proximal CAD may be underappreciated,especially if ostial in location. Hence, there should bea low threshold for using intravascular ultrasound,particularly in the setting of pressure damping orcontrast reflux. Catheterization should be consideredto exclude occult CAD or underlying pulmonary hy-pertension when symptoms are disproportionate toknown valvular or myocardial disease.

EVALUATION OF EXTRACARDIAC VASCULAR

STRUCTURES. Many instances of extensive radio-therapy involve the carotid and subclavian arteries.As a result, the clinical threshold to perform ultra-sound of these vessels should be low. Additionally,pre-operative arterial (of the internal mammary ar-teries [51]) and vein mapping allows for assessment ofthe quality and availability of coronary bypassconduits.

EVALUATION OF RAPD. Subjects with a history ofthoracic radiation should be screened for RAPD,especially if symptomatic. Typically, this manifests aspulmonary fibrosis with traction bronchiectasis insevere cases. Concurrent RAPD is independentlyassociated with reduced survival in RACD (28).Screening for RAPD should be performed in consul-tation with physicians experienced in RAPD. It isimportant to exclude a contributory pulmonic causefor symptoms without simply attributing them toRACD. In all likelihood, subjects with RACD will havealso sustained a degree of RAPD and symptoms areoften multi-factorial. Clinical examination, chestx-ray, pulmonary function tests, including lung

volume measurement and diffusion lung capacity,and dedicated high-resolution CT chest imaging aregenerally recommended. RAPD should be particularlyconsidered when determining suitability for cardiacsurgery because pulmonary complications are a majorsource of perioperative morbidity and mortality (28).This is especially problematic after repeat cardiacsurgery, where recurrent pleural effusions, severelyreduced lung volumes, and ventilation impairmentare commonly observed. Those with RACD and sig-nificant RAPD may be better managed with nonsur-gical or percutaneous approaches, even if cardiacissues cannot be completely resolved.

MANAGEMENT

All patients with a prior cancer history should bequestioned about radiation or cardiotoxic chemo-therapy. Often, radiation exposure is only realizedwhen cardiac testing suggests a more extensivecalcific or fibrotic process than typical for age. Phy-sicians treating complex patients with RACD have theimportant, but difficult, task of resetting patients’expectations and educating regarding the pooreroutcomes in RACD. An experienced team of cardiol-ogists, imaging specialists, interventionalists, andcardiothoracic surgeons is recommended to guidetherapeutic strategies. Medical therapy of RACD istypically undertaken according to standard treatmentguidelines. Pericardial constriction may warrant atrial of anti-inflammatory therapy, in case of revers-ibility. The benefit of heart failure pharmacotherapyin subclinical myocardial dysfunction remainsunknown.

CARDIAC SURGERY

Cardiac surgery in RACD is often complex and,therefore, best undertaken by surgeons experiencedin this arena. Because radiation exposure is hetero-geneous, patients’ conditions cannot be uniformlymanaged and often require individualized surgicalapproaches. Studies of long-term outcomes (with orwithout cardiac surgery) in patients with RACDdemonstrate increased morbidity and mortality(5,15,25,36,52–55). Previous surgical reports havedemonstrated various predictors of short-term(constrictive pericarditis, reduced preoperative ejec-tion fraction, longer cardiopulmonary bypass times)and long-term outcomes (radiation dose, duration ofradiation, tangential vs. mediastinal) (5,25,36,56). Wehave recently demonstrated that the longer-termmortality of patients with RACD undergoing cardiacsurgery was significantly worse compared with age-and sex-matched patients undergoing similar cardiac

Desai et al. J A C C : C A R D I O V A S C U L A R I M A G I N G , V O L . 1 1 , N O . 8 , 2 0 1 8

Radiation-Associated Cardiac Disease A U G U S T 2 0 1 8 : 1 1 3 2 – 4 9

1144

surgeries (55% vs. 28%), despite a low perioperativemortality (57). Additionally, the outcomes of patientswith RACD undergoing redo cardiac surgery wereparticularly worse compared with a matched group(71% vs. 43%). In patients with severe symptomaticaortic stenosis undergoing surgical aortic valvereplacement (AVR), we have demonstrated similarworse survival in patients with RACD compared withage- and sex-matched patients undergoing AVR (48%vs. 7%) (58). However, this data does not reflect thecrucial technical considerations paramount to suc-cessful surgery. In our longitudinal experience, thefollowing practical ‘pearls’ are importantconsiderations.

TIMING OF INTERVENTION. Concurrent manifesta-tions of radiation injury to the heart can have animpact on surgical planning and outcome. In ourexperience, surgery should be delayed in patientswith RACD, especially in multi-valve disease, whereone valve may have severe dysfunction and anothermild or moderate. Reoperative surgery in RACD por-tends significantly increased operative risk andmorbidity compared with non-RACD surgery, soevery attempt should be made to address all issueswith a complete operation the first time.

PLANNING. Planning for cannulation, aortic crossclamping, and managing valvular calcium and calci-fication of the cardiac fibrous skeleton are crucial inRACD (Figure 8). Patients with porcelain aorta are athigh risk for embolic stroke due to manipulation ofaortic atheroma during surgery (59). For patientsundergoing isolated coronary bypass, a “no touch”approach is typically used, often with off-pumptechniques, arterial grafting, and radial graft as aside Y or T graft, if needed (60). With surgical AVR,however, cardiopulmonary bypass and aortic manip-ulation are necessary. Mild, focal calcification of theascending aorta suggests medial calcium and allowsfor safe aortic clamping, whereas dense, circumfer-ential calcification merits planning for circulatoryarrest and replacement of the ascending aorta. Thesurgeon needs to be prepared to remove or workaround areas of calcification. A flexible perfusion andmyocardial protection strategy allows for dealingwith unexpected reconstruction problems in oftenlong, multicomponent operations and might includethe following: cannulation of the right axillary arterywith a side graft, bicaval cannulation, and directcannulation of the coronary sinus for retrogradecardioplegia.

SURGICAL TECHNIQUES. Coronary artery bypass. Des-pite the internal thoracic arteries often lying within

the radiation field, the majority can still be used withgood results unless they appear small and fibrotic(51). Pre-operative vein mapping allows for assess-ment of the quality and availability of alternativevenous conduits. Radial artery conduits can be usedsimilarly to non-RACD bypass surgery. Althoughcoronary targets are likely to be diffusely calcified,finding an adequate graft touchdown site is usuallynot a problem.Valve surgery. Given the susceptibility to calcificationof the aortic valve, aorto-mitral curtain, and mitralvalve annulus, consideration should be given toreplacing both valves, even if disease of one is onlymild to moderate. This is advocated as intraoperativeoptions may be limited due to calcium spanning be-tween the two valves, and because rapidly progres-sive disease may result in a well-functioningprosthesis and other severe valve disease within afew years. Replacement is favored over repairbecause irradiated valve tissue is abnormal and tendsto progressively fibrose and calcify, therebyincreasing the risk of transforming a repaired regur-gitant valve into a stenotic one. Given the increasedrisks of reoperation, mechanical prostheses areappealing, especially for younger patients. However,if other comorbidities preclude lifelong anti-coagulation, consideration may be given tobioprosthetic valve replacement with subsequentvalve-in-valve transcatheter therapy.Aorto-mitral curtain reconstruction. Confluent fibrousskeleton calcification extending from the aorticannulus, across the aorto-mitral curtain and to theanterior mitral valve leaflet can complicate sutureplacement in the mitral annulus and preclude safereplacement. These patients also often have a smallaortic root and small annular sizes, possibly related toradiation exposure during childhood, progressivefibrosis, and/or scar shrinkage. These combined is-sues make double valve replacement an attractiveapproach. Division of the aorto-mitral curtain andanterior mitral leaflet also allows for better exposureof the posterior mitral annulus for debridement ofcalcification, suture placement, and reconstruction.In the “Commando” operation, a patch of autologousor bovine pericardium is fashioned to repair andexpand the dome of the left atrium, the mitralannulus, aorto-mitral curtain, aortic annulus, andaortic valve (Figure 9). This allows for repair afteraggressive calcium debridement, adequate sealing ofthe 2 prosthetic valves, and size increase of bothaortic and mitral valves for more physiological he-modynamics. Adequate bioprosthetic valve sizing iscrucial and, especially in AVR, consideration should

FIGURE 8 Radiation-Associated Aortic Disease

C

A BA

P

R L

H

F

R L

H

F

A P

*

D

Severe calcification of the ascending aorta on cardiac CT (axial [A], coronal [B], and sagittal [C] views; yellow arrows) and at the time of

surgery (D) (white arrows). *Calcification of the aorto-mitral curtain extending into the anterior mitral valve leaflet. Abbreviation as in the

Central Illustration.

J A C C : C A R D I O V A S C U L A R I M A G I N G , V O L . 1 1 , N O . 8 , 2 0 1 8 Desai et al.A U G U S T 2 0 1 8 : 1 1 3 2 – 4 9 Radiation-Associated Cardiac Disease

1145

be given to enlarging the aortic root with the inten-tion of implanting the largest possible prosthesisconsidering future valve-in-valve transcatheterinterventions.

Pericardiectomy. Pericardiectomy is reserved forconstriction with fibro-calcification or severe recur-rent pericarditis despite medical therapy. Surgery istechnically challenging, due to comorbidities and theextensive, piecemeal debridement often required forremoval of pericardial calcification. Pericardiectomycan become even more technically challenging whenthere is ongoing inflammation. For this reason,subjects with evidence of acute or subacute peri-carditis by CMR may warrant treatment with acourse of anti-inflammatory therapy pre-operatively.Outcomes from pericardiectomy are worse in pa-tients with RACD, with 5-year survival ratespost-pericardiectomy of 79.8% for idiopathic, 55.9%

for post-operative, and only 11.0% for post-radiationpericardial disease (61).Transplantation. Cardiac transplantation in the settingof RACD is associated with poor outcomes due toconcomitant advanced lung pathology (often necessi-tating combined more complicated heart-lung trans-plantation) and increased frequency of recurrentmalignancies in the setting of prolonged immuno-suppression. As a result, we do not routinely advocatecardiac transplantation in patients with RACD.

POST-OPERATIVE CONSIDERATIONS. Chronic pleuraland pericardial effusions are common and treatedwith repeated drainage, although soft drainage cath-eters may be placed intra-operatively and left in situfor several weeks post-discharge. Post-operativeintrathoracic fluid retention likely relates to radiationinjury to the lungs and pleura, with resultantlymphatic dysfunction. Achieving post-operative

FIGURE 9 Commando Procedure

A B

C D E

(A) Aortic valve, mitral valve, and aorto-mitral curtain exposed and excised, (B) mitral prosthesis implanted, (C) aorto-mitral curtain reconstructed using pericardium or

synthetic patch, (D) aortic valve prosthesis implantation, with patch reconstruction of the aortic annulus, (E) ascending aortic patch closure.

Desai et al. J A C C : C A R D I O V A S C U L A R I M A G I N G , V O L . 1 1 , N O . 8 , 2 0 1 8

Radiation-Associated Cardiac Disease A U G U S T 2 0 1 8 : 1 1 3 2 – 4 9

1146

diuresis is often difficult and delayed due to restric-tive myocardial dysfunction from fibrosis. Conduc-tion system disturbances are common, especiallywith aggressive reconstruction. Prolonged temporarypacing may be required and permanent left ventric-ular epicardial pacing leads should be considered.Nodal-blocking agents should be used with caution aspatients with RACD are often rate-dependent forcardiac output because variation in stroke volume islimited by ventricular fibrosis. Higher pacemakerheart rates should, therefore, also be considered.

TRANSCATHETER TECHNIQUES

Transcatheter valve replacement techniques providealternative management strategies in RACD and canbe deployed via femoral artery, transaortic, or

transapical approaches (62). Pre-procedural multi-modality imaging stratification is of paramountimportance to determine technical feasibility. Inmost patients with severe AS and porcelain aorta,TAVR has become the preferred option. Among theinoperable Placement of Aortic TraNscathetER valvetrial cohort, porcelain aorta was the most commonreason for technical inoperability, and proceduraloutcomes were similar in these patients (63). Inaddition, even trans-aortic TAVR may be possible insome patients with porcelain aorta, if there is nosignificant calcium at the anterior and lateral aspectof the distal ascending aorta (64). Anecdotally,excellent medium-term outcomes have been ach-ieved with TAVR in individual RACD cases, butlonger-term outcomes in this specific population arenot yet available. Application of the technique to

J A C C : C A R D I O V A S C U L A R I M A G I N G , V O L . 1 1 , N O . 8 , 2 0 1 8 Desai et al.A U G U S T 2 0 1 8 : 1 1 3 2 – 4 9 Radiation-Associated Cardiac Disease

1147

mitral valve replacement in RACD remains limitedwith no longer-term outcomes data.

Although percutaneous coronary intervention(PCI) may be preferred for reduced morbidity, CAD inRACD is typically diffuse and extensively calcified(rather than discrete single vessel disease), makingstenting less appropriate. When long-term PCI out-comes were compared with nonirradiated controlpatients, those with RACD had significantly highermortality, with balloon angioplasty or bare metalversus drug-eluting stent placement (SYNTAXscore $11; functional class $III being additional fac-tors associated with increased mortality) (65). Incontrast, another analysis found that radiotherapywas not associated with increased target vesselrevascularization whether administered before or af-ter stenting, and there was no difference in subse-quent cardiac mortality or all-cause mortality (66).However, a separate analysis of 76 patients withradiation-associated CAD suggested there may be adose-dependent effect of radiation on survival (67). Ina recent study, there were no differences in mortalitybetween patients with RACD and matched controlpatients undergoing PCI (68). However, this studyonly included patients who received radiotherapyand PCI at their center, hence, potentially having se-lection bias. There are no studies directly comparingoutcomes of PCI versus bypass in RACD patients.

ELECTROPHYSIOLOGY

RACD-associated conduction abnormalities aremanaged according to standard recommendations.

This may include anti-arrhythmic agents, permanentpacemakers, or resynchronization therapy andimplanted defibrillators for prevention of suddencardiac death.

CONCLUSIONS

Due to a legacy effect, the incidence of recognizedRACD is likely to exponentially increase over the nextdecade, thus, comprehensive multimodality imaging-based screening programs will be essential toadequately identify those at risk, plan interventions,and evaluate treatment response. Management ofRACD remains challenging due to increased rates ofmorbidity and mortality. Coordinated managementby an experienced team of providers at a Centre ofExcellence is strongly advocated. Timing of surgicalintervention must be individualized, based upon thecomplexity of the radiation-associated disease pro-cess, comorbidities, and technical difficulty. Percu-taneous options are increasingly available, althoughtheir use and suitability in RACD requires furthervalidation.

ACKNOWLEDGMENT The authors acknowledge Dr.Patrick Collier for aiding in manuscript review.

ADDRESS FOR CORRESPONDENCE: Dr. Milind Y.Desai, Department of Cardiovascular Medicine,Desk J1-5, Heart and Vascular Institute, ClevelandClinic, 9500 Euclid Avenue, Cleveland, Ohio 44195.E-mail: desaim2@ccf.org. Twitter: #JACCCVIMG,#radiationheartdisease.

RE F E RENCE S

1. McGale P, Taylor C, Correa C, et al. Effect ofradiotherapy after mastectomy and axillary sur-gery on 10-year recurrence and 20-year breastcancer mortality: meta-analysis of individual pa-tient data for 8135 women in 22 randomised trials.Lancet 2014;383:2127–35.

2. Lu NN, Li YX, Wu RY, et al. Dosimetric andclinical outcomes of involved-field intensity-modulated radiotherapy after chemotherapy forearly-stage Hodgkin’s lymphoma with mediastinalinvolvement. Int J Radiat Oncol Biol Phys 2012;84:210–6.

3. Gaya AM, Ashford RF. Cardiac complications ofradiation therapy. Clin Oncol (R Coll Radiol) 2005;17:153–9.

4. Kunkler PB, Whitfield AG. Radiation reactions inthe heart. Br Heart J 1957;19:53–8.

5. Handa N, McGregor CG, Danielson GK, et al.Coronary artery bypass grafting in patients withprevious mediastinal radiation therapy. J ThoracCardiovasc Surg 1999;117:1136–42.

6. Emami B, Lyman J, Brown A, et al. Tolerance ofnormal tissue to therapeutic irradiation. Int JRadiat Oncol Biol Phys 1991;21:109–22.

7. MacDonald SM. Proton therapy for breast can-cer: getting to the heart of the matter. Int J RadiatOncol Biol Phys 2016;95:46–8.

8. Darby SC, Ewertz M, McGale P, et al. Risk ofischemic heart disease in women after radio-therapy for breast cancer. N Engl J Med 2013;368:987–98.

9. Patt DA, Goodwin JS, Kuo YF, et al. Cardiacmorbidity of adjuvant radiotherapy for breastcancer. J Clin Oncol 2005;23:7475–82.

10. Velensek V, Mazic U, Krzisnik C, et al. Cardiacdamage after treatment of childhood cancer: along-term follow-up. BMC Cancer 2008;8:141.

11. Guldner L, Haddy N, Pein F, et al. Radiationdose and long term risk of cardiac pathologyfollowing radiotherapy and anthracyclin for achildhood cancer. Radiother Oncol 2006;81:47–56.

12. Marinko T, Dolenc J, Bilban-Jakopin C. Car-diotoxicity of concomitant radiotherapy and tras-tuzumab for early breast cancer. Radiol Oncol2014;48:105–12.

13. Adamo V, Ricciardi GR, Adamo B, et al. The riskof toxicities from trastuzumab, alone or in com-bination, in an elderly breast cancer population.Oncology 2014;86:16–21.

14. Plana JC, Galderisi M, Barac A, et al. Expertconsensus for multimodality imaging evaluation ofadult patients during and after cancer therapy: areport from the American Society of Echocardi-ography and the European Association of Cardio-vascular Imaging. Eur Heart J Cardiovasc Imaging2014;15:1063–93.

15. Veeragandham RS, Goldin MD. Surgical man-agement of radiation-induced heart disease. AnnThorac Surg 1998;65:1014–9.

16. Hull MC, Morris CG, Pepine CJ, et al. Valvulardysfunction and carotid, subclavian, and coronaryartery disease in survivors of hodgkin lymphoma

Desai et al. J A C C : C A R D I O V A S C U L A R I M A G I N G , V O L . 1 1 , N O . 8 , 2 0 1 8

Radiation-Associated Cardiac Disease A U G U S T 2 0 1 8 : 1 1 3 2 – 4 9

1148

treated with radiation therapy. JAMA 2003;290:2831–7.

17. Heidenreich PA, Hancock SL, Lee BK, et al.Asymptomatic cardiac disease following medias-tinal irradiation. J Am Coll Cardiol 2003;42:743–9.

18. Desai MY, Wu W, Masri A, et al. Increasedaorto-mitral curtain thickness independently pre-dicts mortality in patients with radiation-associated cardiac disease undergoing cardiacsurgery. Ann Thorac Surg 2014;97:1348–55.

19. Piovaccari G, Ferretti RM, Prati F, et al. Cardiacdisease after chest irradiation for Hodgkin’s dis-ease: incidence in 108 patients with long follow-up. Int J Cardiol 1995;49:39–43.

20. Fukada J, Shigematsu N, Takeuchi H, et al.Symptomatic pericardial effusion after chemo-radiation therapy in esophageal cancer patients.Int J Radiat Oncol Biol Phys 2013;87:487–93.

21. Stewart FA, Heeneman S, Te Poele J, et al.Ionizing radiation accelerates the development ofatherosclerotic lesions in ApoE-/- mice and pre-disposes to an inflammatory plaque phenotypeprone to hemorrhage. Am J Pathol 2006;168:649–58.

22. Hendry JH. Threshold doses and circulatorydisease risks. Ann ICRP 2015;44:69–75.

23. Stewart FA. Mechanisms and dose-responserelationships for radiation-induced cardiovasculardisease. Ann ICRP 2012;41:72–9.

24. Nilsson G, Holmberg L, Garmo H, et al. Dis-tribution of coronary artery stenosis after radiationfor breast cancer. J Clin Oncol 2012;30:380–6.

25. Hancock SL, Tucker MA, Hoppe RT. Factorsaffecting late mortality from heart disease aftertreatment of Hodgkin’s disease. JAMA 1993;270:1949–55.

26. Darby SC, Cutter DJ, BoermaM, et al. Radiation-related heart disease: current knowledge and futureprospects. Int J Radiat Oncol Biol Phys 2010;76:656–65.

27. Gillette EL, Powers BE, McChesney SL, et al.Response of aorta and branch arteries to experi-mental intraoperative irradiation. Int J RadiatOncol Biol Phys 1989;17:1247–55.

28. Desai MY, Karunakaravel K, Wu W, et al. Pul-monary fibrosis on multidetector computedtomography and mortality in patients withradiation-associated cardiac disease undergoingcardiac surgery. J Thorac Cardiovasc Surg 2014;148:475–81.e3.

29. Leon MB, Smith CR, Mack M, et al. Trans-catheter aortic-valve implantation for aortic ste-nosis in patients who cannot undergo surgery.N Engl J Med 2010;363:1597–607.

30. Rodes-Cabau J, Webb JG, Cheung A, et al.Transcatheter aortic valve implantation for thetreatment of severe symptomatic aortic stenosis inpatients at very high or prohibitive surgical risk:acute and late outcomes of the multicenter Ca-nadian experience. J Am Coll Cardiol 2010;55:1080–90.

31. Daitoku K, Fukui K, Ichinoseki I, et al.Radiotherapy-induced aortic valve disease associ-ated with porcelain aorta. Jpn J Thorac CardiovascSurg 2004;52:349–52.

32. Adams MJ, Lipsitz SR, Colan SD, et al. Car-diovascular status in long-term survivors ofHodgkin’s disease treated with chest radiotherapy.J Clin Oncol 2004;22:3139–48.

33. Larsen RL, Jakacki RI, Vetter VL, et al. Elec-trocardiographic changes and arrhythmias aftercancer therapy in children and young adults. Am JCardiol 1992;70:73–7.

34. Strender LE, Lindahl J, Larsson LE. Incidenceof heart disease and functional significance ofchanges in the electrocardiogram 10 years afterradiotherapy for breast cancer. Cancer 1986;57:929–34.

35. Lancellotti P, Nkomo VT, Badano LP, et al.Expert consensus for multi-modality imagingevaluation of cardiovascular complications ofradiotherapy in adults: a report from the EuropeanAssociation of Cardiovascular Imaging and theAmerican Society of Echocardiography. Eur Heart JCardiovasc Imaging 2013;14:721–40.

36. Handa N, McGregor CG, Danielson GK, et al.Valvular heart operation in patients with previousmediastinal radiation therapy. Ann Thorac Surg2001;71:1880–4.

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

38. Ha JW, Oh JK, Ling LH, et al. Annulus para-doxus: transmitral flow velocity to mitral annularvelocity ratio is inversely proportional to pulmo-nary capillary wedge pressure in patients withconstrictive pericarditis. Circulation 2001;104:976–8.

39. Erven K, Florian A, Slagmolen P, et al. Sub-clinical cardiotoxicity detected by strain rate im-aging up to 14 months after breast radiationtherapy. Int J Radiat Oncol Biol Phys 2013;85:1172–8.

40. Armstrong GT, Joshi VM, Ness KK, et al.Comprehensive echocardiographic detection oftreatment-related cardiac dysfunction in adultsurvivors of childhood cancer: results from the St.Jude Lifetime Cohort Study. J Am Coll Cardiol2015;65:2511–22.

41. Chirakarnjanakorn S, Popovic ZB, Wu W, et al.Impact of long-axis function on cardiac surgicaloutcomes in patients with radiation-associatedheart disease. J Thorac Cardiovasc Surg 2015;149:1643-51.e1-2.

42. Leung DY, Griffin BP, Stewart WJ, et al. Leftventricular function after valve repair for chronicmitral regurgitation: predictive value of preoper-ative assessment of contractile reserve by exerciseechocardiography. J Am Coll Cardiol 1996;28:1198–205.

43. Otto CM, Kumbhani DJ, Alexander KP, et al.2017 ACC Expert Consensus Decision Pathway forTranscatheter Aortic Valve Replacement in theManagement of Adults With Aortic Stenosis: AReport of the American College of Cardiology TaskForce on Clinical Expert Consensus Documents.J Am Coll Cardiol 2017;69:1313–46.

44. Kamdar AR, Meadows TA, Roselli EE, et al.Multidetector computed tomographic angiog-raphy in planning of reoperative cardiothoracicsurgery. Ann Thorac Surg 2008;85:1239–45.

45. Gada H, Desai MY, Marwick TH. Cost-effectiveness of computed tomographic angiog-raphy before reoperative coronary artery bypassgrafting: a decision-analytic model. Circ Car-diovasc Qual Outcomes 2012;5:705–10.

46. Verhaert D, Gabriel RS, Johnston D, et al. Therole of multimodality imaging in the managementof pericardial disease. Circ Cardiovasc Imaging2010;3:333–43.

47. Sarwar A, Shaw LJ, ShapiroMD, et al. Diagnosticand prognostic value of absence of coronary arterycalcification. J Am Coll Cardiol Img 2009;2:675–88.

48. Heidenreich PA, Schnittger I, Strauss HW,et al. Screening for coronary artery disease aftermediastinal irradiation for Hodgkin’s disease. J ClinOncol 2007;25:43–9.

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. Messroghli DR, Moon JC, Ferreira VM, et al.Clinical recommendations for cardiovascularmagnetic resonance mapping of T1, T2, T2* andextracellular volume: a consensus statement bythe Society for Cardiovascular Magnetic Reso-nance (SCMR) endorsed by the European Associ-ation for Cardiovascular Imaging (EACVI).J Cardiovasc Magn Reson 2017;19:75.

51. van Son JA, Noyez L, van Asten WN. Use ofinternal mammary artery in myocardial revascu-larization after mediastinal irradiation. J ThoracCardiovasc Surg 1992;104:1539–44.

52. Jaworski C, Mariani JA, Wheeler G, et al. Car-diac complications of thoracic irradiation. J AmColl Cardiol 2013;61:2319–28.

53. Mousavi N, Nohria A. Radiation-induced car-diovascular disease. Curr Treat Options CardiovascMed 2013;15:507–17.

54. Bouillon K, Haddy N, Delaloge S, et al. Long-term cardiovascular mortality after radiotherapyfor breast cancer. JAmColl Cardiol 2011;57:445–52.

55. van Nimwegen FA, Schaapveld M, Janus CP,et al. Cardiovascular disease after Hodgkin lym-phoma treatment: 40-year disease risk. JAMAIntern Med 2015;175:1007–17.

56. Chang AS, Smedira NG, Chang CL, et al. Car-diac surgery after mediastinal radiation: extent ofexposure influences outcome. J Thorac CardiovascSurg 2007;133:404–13.

57. Wu W, Masri A, Popovic ZB, et al. Long-termsurvival of patients with radiation heart diseaseundergoing cardiac surgery: a cohort study. Cir-culation 2013;127:1476–85.

58. Donnellan E, Masri A, Johnston DR, et al. Long-term outcomes of patients with mediastinalradiation-associated severe aortic stenosis and sub-sequent surgical aortic valve replacement: amatchedcohort study. J Am Heart Assoc 2017;6:e005396.

59. Davila-RomanVG,MurphySF,NickersonNJ, et al.Atherosclerosis of the ascending aorta is an indepen-dent predictor of long-term neurologic events andmortality. J Am Coll Cardiol 1999;33:1308–16.

60. Sirin G, Sarkislali K, Konakci M, et al. Extra-anatomical coronary artery bypass grafting in

J A C C : C A R D I O V A S C U L A R I M A G I N G , V O L . 1 1 , N O . 8 , 2 0 1 8 Desai et al.A U G U S T 2 0 1 8 : 1 1 3 2 – 4 9 Radiation-Associated Cardiac Disease

1149

patients with severely atherosclerotic (Porcelain)aorta. J Cardiothorac Surg 2013;8:86.

61. George TJ, Arnaoutakis GJ, Beaty CA, et al.Contemporary etiologies, risk factors, and out-comes after pericardiectomy. Ann Thorac Surg2012;94:445–51.

62. Gillinov AM, Lytle BW, Hoang V, et al. Theatherosclerotic aorta at aortic valve replacement:surgical strategies and results. J Thorac CardiovascSurg 2000;120:957–63.

63. Makkar RR, Jilaihawi H, Mack M, et al. Strati-fication of outcomes after transcatheter aorticvalve replacement according to surgical inoper-ability for technical versus clinical reasons. J AmColl Cardiol 2014;63:901–11.

64. Bapat VN, Attia RQ, Thomas M. Distribution ofcalcium in the ascending aorta in patients under-going transcatheter aortic valve implantation andits relevance to the transaortic approach. J AmColl Cardiol Intv 2012;5:470–6.

65. Reed GW, Masri A, Griffin BP, et al. Long-termmortality in patients with radiation-associatedcoronary artery disease treated with percuta-neous coronary intervention. Circ CardiovascInterv 2016;9:e003483.

66. Liang JJ, Sio TT, Slusser JP, et al. Outcomesafter percutaneous coronary intervention withstents in patients treated with thoracic externalbeam radiation for cancer. J Am Coll Cardiol Intv2014;7:1412–20.

67. Sio TT, Liang JJ, Chang K, et al. Dosimetriccorrelate of cardiac-specific survival among pa-tients undergoing coronary artery stenting afterthoracic radiotherapy for cancer. Am J Clin Oncol2017;40(2):133–9.

68. Fender EA, Liang JJ, Sio TT, et al. Percutaneousrevascularization in patients treated with thoracicradiation for cancer. Am Heart J 2017;187:98–103.

KEY WORDS diagnosis, management,radiation heart disease, review

APPENDIX For a supplemental video,please see the online version of this paper.