E

Prsff(cAcfc

g

1

© 2021 Elsevier

Archives of Cardiovascular Disease 114 (2021) 325—335

Available online at

ScienceDirectwww.sciencedirect.com

XPERT CONSENSUS

osition paper on stress cardiac magneticesonance imaging in chronic coronaryyndrome: Endorsed by the Sociétérancaise de radiologie (SFR), the Sociétérancaise d’imagerie cardiovasculaireSFICV) and the Société francaise deardiologie (SFC)vis d’experts sur l’IRM cardiaque de stress dans le syndrome coronairehronique : approuvé par la Société francaise de radiologie (SFR), la Sociétérancaise d’imagerie cardiovasculaire (SFICV) et la Société francaise deardiologie (SFC)

Florent Le Vena,∗, Jean-Nicolas Dacherb,Francois Pontanac,d,e,f, Gilles Barone-Rochetteg,h,i,Laurent Macronj, Jerome Garotk, Olivier Genéel,Damien Mandrym,n, Luc-Philippe Christiaenso,Martine Gilardp, Louis Boyerq, Alain Furberr,Alexis Jacquier s

a Department of Cardiology, Brest University Hospital, EA3878GETBO, Université de BretagneOccidentale, 29609 Brest, Franceb Normandy University, UNIROUEN, INSERM U1096, Department of Medical Imaging, CardiacImaging Unit, Rouen University Hospital, 76000 Rouen, Francec Université de Lille, U1011-EGID, 59045 Lille, Franced INSERM U1011, 59019 Lille, France

Available online 20 April 2021

Abbreviations: CAD, coronary artery disease; CTA, computed tomography angiography; ESC, European Society of Cardiology; LGE, lateadolinium chelate enhancement; LV, left ventricle/ventricular; MRI, magnetic resonance imaging; PTP, pretest probability.∗ Corresponding author. Department of Cardiology, CHU de la Cavale-Blanche, boulevard Tanguy-Prigent, 29609 Brest, France.

E-mail address: florent.leven@chu-brest.fr (F. Le Ven).

https://doi.org/10.1016/j.acvd.2021.02.004875-2136/© 2021 Elsevier Masson SAS. All rights reserved.

Masson SAS.All rights reserved. - Document downloaded on 11/06/2021 by Rodrigues Sébastien (948138). It is forbidden and illegal to distribute this document.

F. Le Ven, J.-N. Dacher, F. Pontana et al.

e Institut Pasteur de Lille, 59000 Lille, Francef Department of Cardiovascular Radiology, CHU de Lille, 59000 Lille, Franceg Department of Cardiology, CHU Grenoble Alpes, 38700 La Tronche, Franceh Radiopharmaceutiques Biocliniques, INSERM U1039, Grenoble Alpes University, 38700 LaTronche, Francei French Alliance Clinical Trial, French Clinical Research Infrastructure Network, 75018 Paris,Francej Department of Imaging, Centre Cardiologique du Nord de Saint-Denis, 93200 Saint-Denis,Francek Cardiac MRI-Institut Cardiovasculaire Paris Sud, Jacques-Cartier Private Hospital-RamsayHealth, 91300 Massy, Francel Pôle Santé Oréliance, Centre Cardiologique d’Orléans, 45770 Saran, Francem Lorraine University, IADI, INSERM U1254, 54000 Nancy, Francen Department of Radiology, Brabois, CHRU Nancy, 54000 Nancy, Franceo Department of Cardiology, CHU de Poitiers, 86021 Poitiers, Francep Department of Cardiology, Brest University Hospital, EA 4324 ORPHY, Université de BretagneOccidentale, 29609 Brest, Franceq Pôle Imagerie Diagnostique et Radiologie Interventionnelle, CHU Gabriel-Montpied, 63000Clermont-Ferrand, Francer Institut MITOVASC, UMR INSERM U1083, CNRS 6015, Équipe PhysiopathologieCardiovasculaire, Service de Cardiologie, CHU d’Angers, Université d’Angers, 49000 Angers,Frances Aix-Marseille Université, Department of Radiology and Cardiovascular Imaging, HôpitalTimone, AP—HM, CNRS, Centre de Résonance Magnétique Biologique et Médicale (CRMBM),13385 Marseille, France

KEYWORDSAngina pectoris;Myocardialischaemia;Stable coronaryartery disease;Cardiac imagingtechniques

Summary This paper is intended to update the former consensus between the French Soci-eties of Radiology and Cardiology about the use of stress cardiac magnetic resonance imaging inchronic coronary syndrome, published in 2009. The Delphi method was used to build the presentconsensus. This expert panel consensus includes recommendations for indications, the proce-dure (with patient preparation), stress-inducing drugs, the acquisition protocol, interpretationand risk stratification by stress magnetic resonance imaging.© 2021 Elsevier Masson SAS. All rights reserved.

MOTS CLÉSAngor ;Ischémiemyocardique ;Coronaropathiestable ;Techniques

Résumé Ce document a pour objectif de mettre à jour l’ancien consensus des Sociétésfrancaises de radiologie et de cardiologie sur l’utilisation de l’IRM cardiaque de stress dans lesyndrome coronaire chronique publié en 2009. La méthode Delphi a été utilisée. Ce consensusd’experts comprend des recommandations pour les indications, la procédure avec préparationdu patient, les produits induisant le stress, le protocole d’acquisition, l’interprétation et lastratification du risque par IRM de stress.© 2021 Elsevier Masson SAS. Tous droits reserves.

B

Ckvvtf

r(so

© 2021 Elsevier

d’imagerie cardiaque

ackground

ardiac magnetic resonance imaging (MRI) has become aey examination in routine clinical practice for assessing

entricular function, valvular regurgitation, extracellularolume and myocardial enhancement [1—3]. The goal ofhis opinion paper issued by a national expert committeerom the French Society of Radiology (Société francaise de

copc

32

Masson SAS.All rights reserved. - Document downloaded on 11/06/2021 by Rodrigues Sébastien

adiologie [SFR]) and the French Society of CardiologySociété francaise de cardiologie [SFC]) is to define howtress cardiac MRI should be positioned in the managementf patients suspected of having or known to have chronicoronary artery disease (CAD). This consensus includes rec-mmendations for indications, the procedure (with patientreparation), stress-inducing drugs, the acquisition proto-ol, interpretation and risk stratification by stress MRI.

6

(948138). It is forbidden and illegal to distribute this document.

Archives of Cardiovascular Dis

Table 1 Indications and patient selection for stress car-diac magnetic resonance imaging.

Common indications for stress MRI in chroniccoronary syndromeInitial test to diagnose CAD in symptomaticpatients with a PTP > 15% (Fig. 1)Initial test to diagnose CAD in symptomaticpatients with a PTP of 5—15% after assessingoverall clinical likelihood based on PTP modifiers(Fig. 2)If coronary CTA has shown CAD of uncertainfunctional significance or is not diagnosticIn high-risk asymptomatic adults (with diabetes, astrong family history of CAD or when previousrisk-assessment tests suggest a high-risk of CAD),functional imaging, such as stress MRI, may beconsidered for cardiovascular risk assessmentShould be considered when an adverse evolution ofthe patient’s obstructive CAD is suspected: changein (the severity of the) symptoms and theelectrocardiogram (onset of Q waves, change inrepolarization, onset of left bundle branch block,etc.) or deterioration of LV function, if the site andextent of ischaemia would influence clinicaldecision makingIn stable patients with known CAD, reassessmentof their prognosis can be discussed when the timeelapsed since the last stress test considered topresent a low risk has exceeded its period ofvalidity (3—5 years)

Specific situations where stress MRI may bepreferable to other imaging modalitiesFemale patientsObese patients/poor echogenicityYounger patientsAtrial fibrillationCardiac function and morphology informationneeded (i.e. hypertrophic cardiomyopathy, preciseLV ejection fraction evaluation before CRT, etc.)Cardiac tissue characterization information needed(i.e. fibrosis in hypertrophic cardiomyopathy,precise evaluation of viability in CAD, etc.)

CAD: coronary artery disease; CTA: computed tomographyangiogram; CRT: cardiac resynchronization therapy; LV: leftventricular; MRI: magnetic resonance imaging; PTP: pretestprobability.

Ic

Togi

I

Tromhcded(ltocge

ltpc•••

ttttaaa

Piak5aip

sas

actfr

In asymptomatic patients

Data are lacking on indications and how to manage pos-itive test results in asymptomatic subjects. According to

© 2021 Elsevier

ndications for stress MRI in chronicoronary syndrome

his section mainly refers to the 2019 European Societyf Cardiology (ESC) guidelines for the diagnosis and mana-ement of chronic coronary syndromes [4]. Most commonndications for stress MRI are summarized in Table 1.

32

Masson SAS.All rights reserved. - Document downloaded on 11/06/2021 by Rodrigues Sébastien

ease 114 (2021) 325—335

n symptomatic patients

he preliminary stage in approaching the diagnosis of tho-acic pain or dyspnoea is to determine the probability ofbstructive CAD. Patient assessment should first and fore-ost be based on clinical questions about the patient’s

istory and risk factors, details of the characteristics ofhest pain and the presence of dyspnoea, electrocar-iography, chest X-ray and echocardiography. The initialxamination should also be used to consider any possibleifferential diagnoses and to research functional anginaanaemia, hyperthyroidism, severe blood hypertension witheft ventricular [LV] hypertrophy, valve disease and, in par-icular, aortic stenosis and hypertrophic cardiomyopathyr the possibility of rhythm disorders). If the patient’sondition is deemed unstable, the clinician must refer touidelines on acute coronary syndromes, which are not cov-red in the present document [5,6].

The pretest probability (PTP) of obstructive CAD is calcu-ated, considering age, sex, presence of dyspnoea and theypical or atypical characteristics of the patient’s anginaain. Typical angina pain is defined on the basis of threeriteria:

the presence of chest pain suggesting angina;triggered by exercise or emotional stress;rapidly alleviated when the patient stops exercising or hastaken nitroderivatives, usually via the sublingual route.

Angina pain is described as atypical when only two cri-eria are present; in most cases when chest pain is notriggered by exercise. If only one of these criteria is present,he pain is not considered to be angina. In agreement withhe ESC, we recommend the use of the method of Diamondnd Forrester in its 2019 updated version, which gives a moreccurate assessment of the probability of obstructive CADnd can also be used for elderly patients (Fig. 1) [4,7—10].

Patients with a CAD PTP > 15%, especially a mid-to-highTP, are those who would benefit most from functional non-nvasive imaging testing, such as stress MRI, particularly if

revascularization procedure is likely or if the patient hasnown obstructive CAD [4]. For patients with a PTP between% and 15%, testing for diagnosis may be considered afterssessing the overall clinical likelihood based on the mod-fiers of PTP presented in Fig. 2. There is no indication toerform a diagnostic test if the PTP is < 5%.

Functional imaging for myocardial ischaemia, such astress MRI, is indicated if coronary computed tomographyngiography (CTA) has shown CAD of uncertain functionalignificance or is not diagnostic.

Invasive coronary angiography is recommended as anlternative test to diagnose CAD in patients with a highlinical likelihood, severe symptoms refractory to medicalherapy or typical angina at a low-level of exercise, LV dys-unction and clinical evaluation that indicates high eventisk.

7

(948138). It is forbidden and illegal to distribute this document.

F. Le Ven, J.-N. Dacher, F. Pontana et al.

Figure 1. Pretest probability of obstructive coronary artery disease in 15,815 symptomatic patients according to age, sex and the typeof symptoms (adapted from [4]). The patients who presented with isolated dyspnoea or had it as a predominant symptom were included,in addition to the usual categories of Diamond and Forrester’s algorithm. The cells in dark green show the groups for whom the noninvasivetests are the most relevant (PTP > 15%). The cells in pale green show the groups with a PTP of 5—15%, for whom diagnostic tests should beconsidered after overall clinical probability has been assessed, based on the modifiers shown in Fig. 2.

F ronari

Ebbrv

F

Ttarwitlo

mHwt

pl

Pc

Ttrwstdwhfct

© 2021 Elsevier

igure 2. Determinants of the clinical likelihood of obstructive con style of image provided.]. a If available.

SC guidelines [4], functional imaging or coronary CTA maye considered in ‘‘high-risk asymptomatic adults (with dia-etes, a strong family history of CAD or when previousisk-assessment tests suggest a high-risk of CAD) for cardio-ascular risk assessment’’.

ollow-up of patients with known CAD

he long-term prognosis for patients with CAD depends onhe demographics and clinical features, on LV function andngiographically defined coronary lesions and, finally, on theesults of stress imaging. The latter should be consideredhen an adverse evolution of the patient’s obstructive CAD

s suspected: change in (the severity of the) symptoms andhe electrocardiogram (onset of Q waves, change in repo-arization, onset of left bundle branch block) or worseningf LV function.

There is no randomized study focused on the value of

onitoring stable patients by periodical stress imaging.owever, a reassessment of their prognosis can be discussedhen the time elapsed since the last stress test considered

o present a low risk has exceeded its period of validity. A

cpsp

32

Masson SAS.All rights reserved. - Document downloaded on 11/06/2021 by Rodrigues Sébastien

y artery disease (adapted from [4]). [To be drawn as a flow diagram,

eriod of 3 to 5 years was proposed in the 2019 ESC guide-ines [4].

atient selection for stress MRI in chronicoronary syndrome

he most exhaustive meta-analyses reveal that MRI stressesting and positron emission tomography-computed tomog-aphy are the most sensitive and specific imaging modalitieshen invasive coronary angiography shows that the patient’s

tenosis is > 50% or the fractional flow reserve is posi-ive [11]. A strategy guided by MRI has the advantage ofecreasing the number of invasive coronary angiographiesithout compromising patient prognosis [12]. The Britishealthcare system considers that an MRI-guided strategyor exploring patients referred for angina yields the bestost-effectiveness ratio compared with the other alterna-ives. A recent study demonstrated that the use of stress MRI

ompared with invasive fractional flow reserve in patientsresenting with stable angina decreased the number of inva-ive coronary angiographies without compromising patientrognosis [13].

8

(948138). It is forbidden and illegal to distribute this document.

r Dis

taaamaospscwnpMpmcrbdbtNtcta[tttithciMawc

T

S

Ai(w[wtcmbup

ms

•

•

•

•

B

ApctpicactpNbttbbsftastdnnm

D

Tatfcspinician is allowed via microphones. Headphones are placed

© 2021 Elsevier

Archives of Cardiovascula

Despite the higher diagnostic performance of stress MRI,he type of imaging to use will essentially depend on avail-bility, local expertise, the presence of contraindicationsnd patient choice. However, MRI should be considered as

more appropriate choice in some clinical situations, sum-arized in Table 1. MRI does not produce ionizing radiation,

nd has no sensitivity to breast or diaphragm attenuationr sex-based differences in accuracy [14]. Because of itsafety, stress MRI is especially useful in younger and femaleatients. In obese patients, stress echocardiography mayuffer from poor echogenicity, and single-photon emissionomputed tomography can be limited by attenuation,ith a higher incidence of false-positive results. Coro-ary computed tomographic angiography is hampered byatient morphology and higher radiation. Stress perfusionRI is feasible and safe, and has accurate discriminativerognostic value, even in morbidly obese patients (bodyass index ≥ 40 kg/m2) [15]. Both single-photon emission

omputed tomography and stress echocardiography haveeduced diagnostic accuracy in patients with left bundleranch block. Dobutamine stress MRI may have greateriagnostic accuracy than dobutamine echocardiographyecause of the comprehensive examination with the addi-ion of perfusion and late gadolinium enhancement [16].evertheless, data on accuracy of stress MRI using vasodila-ors in patients with left bundle branch block is limited, andannot be recommended over another imaging modality forhose patients. Stress MRI is feasible in patients with atrialrrythmia, and has good discriminative prognostic value17]. Finally, the main advantage of stress MRI is the abilityo accurately evaluate cardiac morphology, function andissue characteristics during the same test. For instance,his modality should be considered if stress imaging isndicated for a patient with hypertrophic cardiomyopathy,o identify focal or diffuse fibrosis, or for a patient witheart failure, to measure LV ejection fraction (beforeardiac resynchronization therapy and/or defibrillatormplantation, etc.) and right ventricular function. StressRI also offers precise evaluation of myocardial viability as

result of high spatial resolution and excellent correlationith histology, which provides valuable information whenonsidering revascularization.

he procedure

afety of MRI stress tests

bundant literature shows that myocardial ischaemia-nducing tests (dobutamine) or coronary reserve testsadenosine, regadenoson, dipyridamole) can be performedith MRI in perfectly acceptable conditions for the patient

10]. The vasodilators should be preferred to dobutamine,hich carries a higher complication rate in clinical prac-

ice. Dobutamine can be used at a low dose to researchontractile reserve and to explore the viability of the

yocardium. Our group regrets that vasodilators have noteen granted a visa for MRI stress tests. In France, theirse is justified by many articles in the literature, the Euro-ean recommendations, several randomized studies and

oce

32

Masson SAS.All rights reserved. - Document downloaded on 11/06/2021 by Rodrigues Sébastien

ease 114 (2021) 325—335

eta-analyses and the recommendations from internationalocieties [12,13,18—22].

MRI stress tests require strict safety measures:the hospital must be equipped with a cardiology intensivecare unit;the patient must be duly informed of how the examinationwill be carried out, the fact that the drug is being usedoff-label and the risks entailed, and an informed consentform must be signed;the medical staff should respect the contraindications,dosage and routes of administration of the pharmacolog-ical products and contrast agents involved, and be awareof their side effects and the withdrawal criteria;contraindications for provoked myocardial ischaemiatesting must be respected (Table 2).

efore the stress test

physician specialized in cardiovascular imaging shouldractice stress cardiac MRI in cooperation with a physi-ian trained in cardiorespiratory resuscitation who is able toake care of the patient immediately if necessary [23]. Thearamedical team should include two operators, one trainedn the use of the emergency trolley and a radiology techni-ian. An area located beside the MRI suite must be availablend equipped with all items required for emergency pro-edures: oxygen and suction drainage, and an emergencyrolley equipped with all drugs and antidotes (amino-hylline, beta-blockers, salbutamol, adrenalin, nitrates).asotracheal intubation equipment and a defibrillator muste available. The different emergency procedures, usefulelephone numbers and names of the different opera-ors should be displayed in the room. The patient shoulde given a 12-channel reference electrocardiogram beforeeing installed in the room. Twelve to 24 hours before atress MRI, the patient must not take any medication orood that is likely to inhibit drugs or change the interpre-ation. For dobutamine, beta-blockers and nitroderivativesre prohibited. For dipyridamole, adenosine and regadeno-on, caffeine (coffee, tea, chocolate or drinks containinghese, food or drugs containing caffeine) aminophylline,ipyridamole and nicotine are contraindicated. Fasting isot mandatory; a light meal is advised before the exami-ation. The vasodilators frequently induce side effects thatust be explained to the patient.

uring the stress test

wo venous lines (flexible 16 or 20 gauge catheters) are madevailable, one for injecting the stress-inducing product andhe other for the gadolinium-based contrast medium (exceptor regadenoson, for which a single venous line is suffi-ient). Patient monitoring (amagnetic device mandatory)hould include continuous three-channel electrocardiogra-hy, arterial pressure and a digital oximeter. An alarm buttons placed in the patient’s hand. Oral contact with the tech-

n patient’s ears to reduce noise. If urgent care or resus-itation is required, the patient must be removed from thexamination room immediately.

9

(948138). It is forbidden and illegal to distribute this document.

F. Le Ven, J.-N. Dacher

Table 2 Stress magnetic resonance imagingcontraindications.

Provoked ischaemia testing contraindicationsMyocardial infarction or recent acute coronarysyndrome (< 5 days)Uncontrolled rhythm disorderSerious conduction disorders (atrioventricularblock ≥ 2)Known significant untreated stenosis of the left mainarteryNon-controlled heart failure, severe aortic stenosis,obstructive cardiomyopathyPatient refusal

Vasodilator (dipyridamole, adenosine, regadenoson)contraindicationsSecond- or third-degree atrioventricular block or sinusdysfunctionSystolic pressure < 90 mmHgSevere systemic arterial hypertension(≥ 220/120 mmHg)Sinusal bradycardia (heart rate < 40 beats/min)Hypersensitivity to the active principle or one of theexcipientsKnown hypersensitivity to stress agentActive bronchoconstrictive or bronchospastic diseasewith regular use of inhalersa

Dobutamine contraindicationsSevere systemic arterial hypertension(≥ 220/120 mmHg)Unstable angina pectorisSevere aortic valve stenosisComplex cardiac arrhythmias, including uncontrolledatrial fibrillationHypertrophic obstructive cardiomyopathyMyocarditis, endocarditis or pericarditisUncontrolled heart failure

Atropine contraindicationsNarrow-angle glaucomaMyasthenia gravisObstructive uropathyObstructive gastrointestinal disorders

MRI absolute contraindicationsIncompatible cardiac implantable electronic device(PM, ICD and CRT)Metallic intraocular foreign bodiesImplantable neurostimulation systemsCochlear implants/ear implantDrug infusion pumpsCatheters with metallic componentsMetallic fragments such as bullets, shotgun pelletsand metal shrapnelCerebral artery aneurysm clips

Magnetic dental implant, tissue expander, artificiallimb, hearing aid, piercing

CRT: cardiac resynchronization therapy; ICD: implantable car-dioverter defibrillator; MRI: magnetic resonance imaging; PM:pacemaker.a Regadenoson has been demonstrated to be safe to use inpatients with mild-to-moderate chronic obstructive pulmonarydisease and asthma (see text for details).

A

CtdTtrgtouaabli

Sc

Asipsategeev

D

DTitdmwds

AFbpi1

RFc

33

© 2021 Elsevier Masson SAS.All rights reserved. - Document downloaded on 11/06/2021 by Rodrigues Sébastien

, F. Pontana et al.

t the end of the stress test

are should be taken when the patient stands up because ofhe risk of orthostatic hypotension. A 12-channel electrocar-iogram should be performed to check return to baseline.he patient should remain under surveillance until the elec-rocardiogram returns to baseline, with no chest pain andecovery from side effects. An aminophylline injection is notiven systematically when the stress agent is dipyridamole,he half-life of aminophylline being much longer than thatf dipyridamole; it can be helpful if the patient experiencesnpleasant side effects. Aminophylline is not indicated afterdministration of adenosine. Aminophylline is not advisedfter injection of regadenoson in patients with epilepsy,oth drugs having proconvulsive effects. The patients caneave the department after a period of surveillance if nontercurrent clinical event has occurred.

tress-inducing drugs: precautions andontraindications

study suggested that dipyridamole was less sensitive andpecific than adenosine or regadenoson for MRI stress test-ng [24]. Adenosine or regadenoson should therefore bereferred, and dipyridamole should only be used if adeno-ine or regadenoson are not available in the hospital. Inddition, regadenoson is a drug that is easier to use, ashere is only one dosage for adults; the likelihood of dosagerrors is therefore decreased. Regadenoson specifically tar-ets the cardiac receptors, thus mitigating the benign sideffects connected to this vasodilator. This group of expertsstimates that regadenoson is currently the most suitableasodilator to be used in stress cardiac MRI.

ose

obutaminehe expert group does not advise researching myocardial

schaemia with dobutamine, and suggests that vasodila-ors should be preferred. However, it is possible to useobutamine to research contractile reserve and assessyocardial viability. The maximum dose is 15 �g/kg/min,hich should be reached by steps of 2.5—5 �g/kg/min. Theuration of each step should be between 2 and 3 minutes,tarting from an initial dose of 5 �g/kg/min.

denosineor adenosine, the dose is 140 �g/kg/min. The dose cane increased to 210 �g/kg/min if, after 2—3 minutes, theatient’s heart rate does not increase by 10 beats/min orf systolic blood pressure does not decrease by at least0 mmHg.

egadenosonor regadenoson, a single dose is used in adults, withoutonsidering size or weight: 0.4 mg by intravenous injection.

0

(948138). It is forbidden and illegal to distribute this document.

r Dis

DFi

C

SAsttmoscrvrmia

P

Aaaid

flnpMhr

qil

A

ClEl

C

Tso1oists

mt

Sp

Peooor

ico0aRgciaiasbub1ls

L

Tiacbgwpa6t

I

Invdt

© 2021 Elsevier

Archives of Cardiovascula

ipyridamoleor dipyridamole, the dose is 0.56 to 0.82 mg/kg via slowntravenous injection over a period of 3 minutes.

ontraindications

tress agent contraindications are displayed in Table 2.ll three vasodilators (dipyridamole, adenosine, regadeno-on) share common contraindications. However, accordingo available data from observational studies as well as con-rolled clinical trials, the use of regadenoson in patients withild-to-moderate asthma and mild-to-moderate chronic

bstructive pulmonary disease is safe [25—29]. Regadeno-on should be used very cautiously in patients with severehronic obstructive pulmonary disease, in patients whoequire 24-hour/day home oxygen administration, have pre-iously been intubated for respiratory failure, have hadecent exacerbations or have required uptitration of theiredication regimen within a 1-month period, because data

n these populations are limited. Regadenoson should bevoided in patients with severe bronchial asthma.

otential side effects

t low doses, dobutamine rarely causes complications, butt higher doses (20—40 �g/kg/min) can induce chest painnd palpitations. More serious complications (myocardialnfarction, ventricular fibrillation and ventricular tachycar-ia) are rare.

Adenosine, regadenoson and dipyridamole can induce hotushes, headaches, precordial pain, palpitations and dysp-oea. These side effects occur frequently (in around 30% ofatients), but they are usually benign and rapidly reversible.ore serious side effects (transient conduction disorders,ypotension, sinusal tachycardia and bronchospasm) arearer.

The side effects described for adenosine occur less fre-uently with regadenoson, but the half-life of regadenosons longer (so the patient should be monitored for slightlyonger with regadenoson than with adenosine).

cquisition protocol

ine MRI sequences associated with first-pass perfusion andate enhancement form the basis of any stress cardiac MRI.xamples of acquisition protocols with approximate time-ines are proposed in Fig. 3.

ine MRI

his part is based on fast acquisition cine sequences usingteady-state free precession (slice thickness 6—8 mm, withr without 2—4 mm interslice gaps [to make a total of0 mm]; temporal resolution ≤ 45 ms between phases toptimize evaluation of wall motion) [30]. Parallel imag-

ng shortens the acquisition process. These pulse sequenceshould include at least: slices covering the whole left ven-ricle (LV) in its short axis from base to apex (the most basallice must be immediately proximal to the position of the

oTMp

33

Masson SAS.All rights reserved. - Document downloaded on 11/06/2021 by Rodrigues Sébastien

ease 114 (2021) 325—335

itral valve); and three long-axis slices of the LV, includingwo-chamber, four-chamber and LV outflow tract views.

tress test using vasodilators: myocardialerfusion imaging

erfusion MRI is based on a qualitative (visual) analysis of thenhancement of the myocardial signal during the first-passf a bolus injection of gadolinium chelate. The temporal res-lution of perfusion sequences should allow the acquisitionf three to five slices within an R-R space. Saturation-ecovery is used as prepulse.

The following should be performed: at least three slicesn the short axis of the LV, and most often a slice in the verti-al plane of the long axis and another in the horizontal planef the long axis, after intravenous injection of a bolus of.05—0.1 mmol.kg−1 of gadolinium chelate (4—5 mL/s usingn automatic injector); these slices are acquired on each R-

interval during the minute after injection of the bolus ofadolinium. The acquisitions are performed under pharma-ological stimulation during the last minute of the adenosinenjection, 3—5 minutes after the end of the slow intravenousdministration of dipyridamole or 1—4 minutes after thentravenous bolus of regadenoson. Pulse sequences must bedapted to the drug-induced tachycardia. A perfusion acqui-ition at rest after injection of the vasodilators is optional,ut may be superfluous if the result of the stress perfusion isnequivocal (normal, ischaemia). Aminophylline could alsoe injected after the first perfusion acquisition (minimum

minute after gadolinium injection) to reverse the vasodi-ation effect if dipyridamole or regadenoson is used as thetress agent.

ate gadolinium chelate enhancement (LGE)

wo-dimensional or three-dimensional phase-sensitivenversion recovery sequences can be used, 10—15 minutesfter the injection of 0.1—0.15 mmol.kg−1 gadoliniumhelate (dose depending on medical staff choice). It mighte reasonable to reinject gadolinium chelate to performood quality LGE after stress perfusion (if rest perfusionas not performed). Inversion time is optimized for eachatient, so that the signal for healthy myocardium is zerot the time of acquisition. The slice thickness must be—8 mm, and the in-plane spatial resolution must be lowerhan 1.4—1.8 mm.

nterpretation

nterpreting stress cardiac MRI is a synthesis that isot restricted to analysing first-pass perfusion underasodilators. The clinical, electrical and angiographicalata (invasive coronary angiography or coronary computedomography) must be considered, together with all featuresf the MRI examination (hypokinesia, late enhancement).herefore, in addition to experience in cardiac MRI, stress

RI should only be performed by teams with clinical com-etence in CAD.

1

(948138). It is forbidden and illegal to distribute this document.

F. Le Ven, J.-N. Dacher, F. Pontana et al.

Figure 3. Examples of stress magnetic resonance imaging protocols according to vasodilator agent used. A. Adenosine. B. Dipyridamole.C. Regadenoson. Gd: gadolinium; LGE: late gadolinium chelate enhancement. a Optional. b It might be reasonable to reinject gadoliniumchelate to perform good quality LGE after stress perfusion (if rest perfusion was not performed).

Figure 4. Stress magnetic resonance imaging under regadenoson in a young adult with stenosis of the left anterior descending artery.Three sequential images are shown. A. Gadolinium just reached the left ventricular cavity; no myocardial contrast is visible. B. Whereasnormal myocardium is enhanced, the septum and posterior wall perfusion is obviously decreased. C. Hyposignal (ischaemia) persisted duringseveral cardiac cycles.

332

© 2021 Elsevier Masson SAS.All rights reserved. - Document downloaded on 11/06/2021 by Rodrigues Sébastien (948138). It is forbidden and illegal to distribute this document.

r Disease 114 (2021) 325—335

D

TfammttcdaiGtTcodic

ahwftaaicmL

hseaossspprt

V

Aygttaaestocm

Figure 5. Gibbs artefact in a young patient with hypertension;hyposignal is visible in the phase-encode direction when high con-trast gadolinium appears in the left ventricle (thin arrow). Thehc

ednn

Ae

Wofcfttvr

pnahoaiom

R

Ie

© 2021 Elsevier

Archives of Cardiovascula

iagnosis of ischaemia

he semiology of a first-pass perfusion sequence isairly unequivocal. After injection, the gadolinium chelateppears successively in the right heart chambers, the pul-onary parenchyma and then the myocardium. Normalyocardium picks up the contrast clearly and evenly. Some-

imes a perfusion gradient is observed from the epicardiumo the endocardium; this feature must be fleeting to beonsidered as normal. In the case of coronary reserveeficit, persistent hypoperfusion is found in a segmentalrea (Fig. 4). This hypoperfused area is researched by brows-ng (slowly) through the first-pass sequence image by image.oing through the images using fast cine mode could mislead

he observer, and fleeting hypoperfusion could be missed.he perfusion images should be classified using the Ameri-an Heart Association 17-segment model [31]. The numberf diseased segments is used to assess the risk of adverse car-iovascular events. Hypoperfusion in at least two segmentss a sign of significant ischaemia, and carries a high-risk ofardiovascular events.

The presence of ischaemia is univocal when acquisitionsnd injections are correctly performed in a patient whoas been previously vasodilated. Interpretation is facilitatedhen using motion compensation algorithms. When rest per-

usion images are acquired, it is important to make sure thathe vasodilators are no longer effective; this is one of thedvantages of adenosine, which has a very short half-lifend thus does not require the use of an antagonist. Whenschaemia is detected, the first-pass perfusion should beompared with cine sequences (wall motion abnormality andyocardial thinning are searched) and, above all, with the

GE.If the resting perfusion is normal and no LGE is visible, the

ypoperfusion is reversible. Revascularization is a possibleolution if at least two adjacent segments appear as hypop-rfused during stress imaging. If the hypoperfused segmentsre perfectly superimposable with LGE, the diagnosis is thatf ischaemic heart disease with necrotic sequalae. Occa-ionally, especially when infarct size is small, hypoperfusedegments may not be identified during a first-pass perfu-ion sequence, although a typical ischaemic LGE pattern isresent [19]. If the hypoperfused area is wider than the LGE,erilesional stress hypoperfusion is present; it means that aevascularization procedure could be discussed according tohe invasive coronary angiography findings.

iability assessment

stress cardiac MRI test must be completed by an anal-sis of myocardial viability using late enhancement afteradolinium injection. A segmental hypersignal with < 25% ofransmural extension indicates a strong probability of func-ional recovery after revascularization, whereas if > 75% isffected, no improvement can be expected [32]. A meta-nalysis of 331 patients showed that < 50% of transmuralxtension is predictive of functional recovery, with 95% sen-itivity and 51% specificity [33]. The contractile reserve

est can be useful when the necrotic scar involves 25—50%f the thickness of the wall. The contractile reserve testonsists of injecting a low dose of dobutamine (maxi-um 15 �g/kg/min in consecutive steps of 2.5—5 �g/kg/min

tsfp

33

Masson SAS.All rights reserved. - Document downloaded on 11/06/2021 by Rodrigues Sébastien

yposignal, usually septal, is transient and limited to the endo-ardium. No ischaemia was present in this patient.

very 2 to 3 minutes), analysing the kinetics of the segmenturing the injection [34]. The thickness of the myocardium isot sufficient to assess myocardial viability because a thick-ess of < 5 mm is not specific for non-viability [35].

rtefacts and hypoperfusion withoutpicardial coronary stenosis

hen a linear hyposignal is visible on a single slice, typicallyn the septum (Fig. 5), this can suggest a truncation arte-act (or Gibbs artefact). True myocardial hypoperfusion isharacterized by a front wave-shaped hyposignal, persistingor several R-R intervals (at least three). A resting acquisi-ion can be performed, and compared image by image withhe stress acquisition. The truncation artefact is generallyisible and unchanged in both series (stress and rest); it isestricted to the subendocardium.

Perfusion anomalies can be observed on the first-passerfusion images, even if the patient does not have sig-ificant epicardial coronary artery stenosis, and should notlways be considered as ‘‘false-positives’’: if the patient’seart is hypertrophic, a subendocardial hyposignal can bebserved, and could be the substrate of functional angina;lso, endothelial dysfunction may lead to subendocardialschaemia (the hyposignal is often circumferential, and isbserved in the context of hypertension and/or diabetesellitus).

isk stratification by stress MRI

n line with ESC recommendations, the expert group consid-rs that an ischaemia threshold of at least two segments is

he most relevant. A negative (perfusion or dobutamine) MRItress test is associated with an annualized event rate ≤ 1%or a follow-up period of > 2 years [36,37]. This ‘‘guarantee’’eriod is assessed on the basis of a heterogeneous popula-

3

(948138). It is forbidden and illegal to distribute this document.

cher

tro

aaaMdd

asasero

C

Cpcdcslfbocsawas

S

N

D

T

R

[

[

[

[

[

[

[

[

[

© 2021 Elsevier

F. Le Ven, J.-N. Da

ion, and should thus be adapted according to the clinicalisk factors, such as age, female sex, presence of diabetesr the severity of coronary lesions [38].

In practice, in the case of unknown obstructive CAD, positive MRI stress test indicates that invasive coronaryngiography should be performed to confirm the diagnosisnd possibly to treat the involved vessel. A non-conclusiveRI stress test requires an alternative method, such as car-iac CTA or invasive coronary angiography, according to theegree of clinical probability.

In the case of known obstructive CAD, invasive coronaryngiography should only be performed if the MRI stress testhows ischaemia in at least two segments, which is a sign of

high likelihood of events, or if there are other criteria ofeverity, such as the onset of symptoms at a low threshold ofxercise (also a sign of a high-risk of events) or if the patientemains symptomatic despite optimal antiangina treatmentr if LV function is compromised.

onclusions

ardiac stress testing by MRI has become the high-erformance technology of choice for the diagnosis andlassification of risk in patients with chronic coronary syn-rome. Although vasodilators are the preferred products toarry out these procedures, because they are simple andafe to use, their administration in the MRI suite remains off-abel in France. This group promotes the use of regadenosonor stress cardiac MRI. The latest recommendations issuedy the ESC encourage the use of noninvasive imaging whenbstructive CAD is suspected, to the detriment of theonventional treadmill test. This change in practice willtimulate the development of MRI stress tests that offer thedvantages of relatively low cost and no radiation exposure,hile offering a detailed analysis of the heart’s morphologynd function and the viability of the myocardium during theame examination.

ources of funding

one.

isclosure of interest

he authors declare that they have no competing interest.

eferences

[1] Alis D, Guler A, Yergin M, Asmakutlu O. Assessment ofventricular tachyarrhythmia in patients with hypertrophic car-diomyopathy with machine learning-based texture analysis oflate gadolinium enhancement cardiac MRI. Diagn Interv Imag-ing 2020;101:137—46.

[2] Capron T, Cautela J, Scemama U, et al. Cardiac magneticresonance assessment of left ventricular dilatation in chronic

severe left-sided regurgitations: comparison with standardechocardiography. Diagn Interv Imaging 2020;101:657—65.

[3] Habert P, Capron T, Hubert S, et al. Quantification of rightventricular extracellular volume in pulmonary hypertension

[

33

Masson SAS.All rights reserved. - Document downloaded on 11/06/2021 by Rodrigues Sébastien

, F. Pontana et al.

using cardiac magnetic resonance imaging. Diagn Interv Imag-ing 2020;101:311—20.

[4] Knuuti J, Wijns W, Saraste A, et al. 2019 ESC Guidelines for thediagnosis and management of chronic coronary syndromes. EurHeart J 2020;41:407—77.

[5] Collet JP, Thiele H, Barbato E, et al. 2020 ESC Guidelines forthe management of acute coronary syndromes in patients pre-senting without persistent ST-segment elevation. Eur Heart J2020;42:1289—367.

[6] Ibanez B, James S, Agewall S, et al. 2017 ESC Guidelines forthe management of acute myocardial infarction in patientspresenting with ST-segment elevation: the Task Force for themanagement of acute myocardial infarction in patients pre-senting with ST-segment elevation of the European Society ofCardiology (ESC). Eur Heart J 2018;39:119—77.

[7] Diamond GA, Forrester JS. Analysis of probability as an aid inthe clinical diagnosis of coronary artery disease. N Engl J Med1979;300:1350—8.

[8] Genders TS, Steyerberg EW, Alkadhi H, et al. A clini-cal prediction rule for the diagnosis of coronary arterydisease: validation, updating, and extension. Eur Heart J2011;32:1316—30.

[9] Genders TS, Steyerberg EW, Hunink MG, et al. Prediction modelto estimate presence of coronary artery disease: retrospectivepooled analysis of existing cohorts. BMJ 2012;344:e3485.

10] Montalescot G, Sechtem U, Achenbach S, et al. 2013 ESC guide-lines on the management of stable coronary artery disease:the Task Force on the management of stable coronary arterydisease of the European Society of Cardiology. Eur Heart J2013;34:2949—3003.

11] Knuuti J, Ballo H, Juarez-Orozco LE, et al. The perfor-mance of noninvasive tests to rule-in and rule-out significantcoronary artery stenosis in patients with stable angina: a meta-analysis focused on post-test disease probability. Eur Heart J2018;39:3322—30.

12] Greenwood JP, Ripley DP, Berry C, et al. Effect of care guidedby cardiovascular magnetic resonance, myocardial perfusionscintigraphy, or NICE guidelines on subsequent unnecessaryangiography Rates: the CE-MARC 2 randomized clinical trial.JAMA 2016;316:1051—60.

13] Nagel E, Greenwood JP, McCann GP, et al. Magnetic resonanceperfusion or fractional flow reserve in coronary disease. N EnglJ Med 2019;380:2418—28.

14] Greenwood JP, Motwani M, Maredia N, et al. Comparison ofcardiovascular magnetic resonance and single-photon emis-sion computed tomography in women with suspected coronaryartery disease from the Clinical Evaluation of Magnetic Res-onance Imaging in Coronary Heart Disease (CE-MARC) Trial.Circulation 2014;129:1129—38.

15] Kinnel M, Garot J, Pezel T, et al. Prognostic value ofvasodilator stress perfusion CMR in morbidly obese patients(BMI ≥ 40 kg/m2) without known CAD. JACC Cardiovasc Imaging2020;13:1276—7.

16] Mordi I, Stanton T, Carrick D, et al. Comprehensive dobu-tamine stress CMR versus echocardiography in LBBB andsuspected coronary artery disease. JACC Cardiovasc Imaging2014;7:490—8.

17] Pezel T, Sanguineti F, Kinnel M, et al. Feasibility and prognosticvalue of vasodilator stress perfusion CMR in patients with atrialfibrillation. JACC Cardiovasc Imaging 2021;14:379—89.

18] Fratz S, Chung T, Greil GF, et al. Guidelines and protocols forcardiovascular magnetic resonance in children and adults withcongenital heart disease: SCMR expert consensus group on con-genital heart disease. J Cardiovasc Magn Reson 2013;15:51.

19] Greenwood JP, Herzog BA, Brown JM, Everett CC, Plein S. Car-diovascular magnetic resonance and single-photon emission

computed tomography in suspected coronary heart disease.Ann Intern Med 2016;165:830—1.

4

(948138). It is forbidden and illegal to distribute this document.

r Dis

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[38] Hachamovitch R, Hayes S, Friedman JD, et al. Determinants ofrisk and its temporal variation in patients with normal stress

© 2021 Elsevier

Archives of Cardiovascula

20] Jaarsma C, Leiner T, Bekkers SC, et al. Diagnostic perfor-mance of noninvasive myocardial perfusion imaging usingsingle-photon emission computed tomography, cardiac mag-netic resonance, and positron emission tomography imaging forthe detection of obstructive coronary artery disease: a meta-analysis. J Am Coll Cardiol 2012;59:1719—28.

21] Schwitter J, Wacker CM, van Rossum AC, et al. MR-IMPACT:comparison of perfusion-cardiac magnetic resonance withsingle-photon emission computed tomography for the detec-tion of coronary artery disease in a multicentre, multivendor,randomized trial. Eur Heart J 2008;29:480—9.

22] Schwitter J, Wacker CM, Wilke N, et al. MR-IMPACT II: Mag-netic Resonance Imaging for Myocardial Perfusion Assessmentin Coronary artery disease Trial: perfusion-cardiac magneticresonance vs. single-photon emission computed tomographyfor the detection of coronary artery disease: a comparativemulticentre, multivendor trial. Eur Heart J 2013;34:775—81.

23] Vignaux O, Deux JF, Chabrillat Y, et al. [Cardiac MRI: technicalconsiderations]. J Radiol 2009;90:1133—43.

24] Vasu S, Bandettini WP, Hsu LY, et al. Regadenoson andadenosine are equivalent vasodilators and are superior thandipyridamole — a study of first-pass quantitative perfusioncardiovascular magnetic resonance. J Cardiovasc Magn Reson2013;15:85.

25] Husain Z, Palani G, Cabrera R, et al. Hemodynamic response,arrhythmic risk, and overall safety of regadenoson as a pharma-cologic stress agent for myocardial perfusion imaging in chronicobstructive pulmonary disease and bronchial asthma patients.Int J Cardiovasc Imaging 2012;28:1841—9.

26] Kwon DH, Cerqueira MD, Young R, et al. Lessons from regadeno-son and low-level treadmill/regadenoson myocardial perfusionimaging: initial clinical experience in 1263 patients. J NuclCardiol 2010;17:853—7.

27] Prenner BM, Bukofzer S, Behm S, Feaheny K, McNutt BE. Arandomized, double-blind, placebo-controlled study assessingthe safety and tolerability of regadenoson in subjects withasthma or chronic obstructive pulmonary disease. J Nucl Car-diol 2012;19:681—92.

28] Salgado Garcia C, Jimenez Heffernan A, Sanchez de Mora

E, et al. Comparative study of the safety of regadenosonbetween patients with mild/moderate chronic obstructive pul-monary disease and asthma. Eur J Nucl Med Mol Imaging2014;41:119—25.

33

Masson SAS.All rights reserved. - Document downloaded on 11/06/2021 by Rodrigues Sébastien

ease 114 (2021) 325—335

29] Thomas GS, Tammelin BR, Schiffman GL, et al. Safety ofregadenoson, a selective adenosine A2A agonist, in patientswith chronic obstructive pulmonary disease: a randomized,double-blind, placebo-controlled trial (RegCOPD trial). J NuclCardiol 2008;15:319—28.

30] Kramer CM, Barkhausen J, Bucciarelli-Ducci C, Flamm SD,Kim RJ, Nagel E. Standardized cardiovascular magnetic res-onance imaging (CMR) protocols: 2020 update. J CardiovascMagn Reson 2020;22:17.

31] Cerqueira MD, Weissman NJ, Dilsizian V, et al. Standardizedmyocardial segmentation and nomenclature for tomographicimaging of the heart. A statement for healthcare professionalsfrom the Cardiac Imaging Committee of the Council on Clini-cal Cardiology of the American Heart Association. Circulation2002;105:539—42.

32] Kim RJ, Wu E, Rafael A, et al. The use of contrast-enhancedmagnetic resonance imaging to identify reversible myocardialdysfunction. N Engl J Med 2000;343:1445—53.

33] Romero J, Xue X, Gonzalez W, Garcia MJ. CMR imaging assessingviability in patients with chronic ventricular dysfunction due tocoronary artery disease: a meta-analysis of prospective trials.JACC Cardiovasc Imaging 2012;5:494—508.

34] Nagel E, Schuster A. Myocardial viability: dead or alive is notthe question! JACC Cardiovasc Imaging 2012;5:509—12.

35] Shah DJ, Kim HW, James O, et al. Prevalence of regionalmyocardial thinning and relationship with myocardialscarring in patients with coronary artery disease. JAMA2013;309:909—18.

36] Gargiulo P, Dellegrottaglie S, Bruzzese D, et al. The prog-nostic value of normal stress cardiac magnetic resonance inpatients with known or suspected coronary artery disease: ameta-analysis. Circ Cardiovasc Imaging 2013;6:574—82.

37] Lipinski MJ, McVey CM, Berger JS, Kramer CM, Salerno M. Prog-nostic value of stress cardiac magnetic resonance imaging inpatients with known or suspected coronary artery disease:a systematic review and meta-analysis. J Am Coll Cardiol2013;62:826—38.

myocardial perfusion scans: what is the warranty period of anormal scan? J Am Coll Cardiol 2003;41:1329—40.

5

(948138). It is forbidden and illegal to distribute this document.