myocardium at risk after acute infarction in humans on cardiac magnetic resonance: quantitative...

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 . 2 , N O . 5 , 2 0 0 9

© 2 0 0 9 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 I S S N 1 9 3 6 - 8 7 8 X / 0 9 / $ 3 6 . 0 0

P U B L I S H E D B Y E L S E V I E R I N C . D O I : 1 0 . 1 0 1 6 / j . j c m g . 2 0 0 8 . 1 1 . 0 1 8

Myocardium at Risk After Acute Infarction inHumans on Cardiac Magnetic ResonanceQuantitative Assessment During Follow-Up and Validation WithSingle-Photon Emission Computed Tomography

Marcus Carlsson, MD, PHD,* Joey F. A. Ubachs, MD,* Erik Hedström, MD, PHD,*Einar Heiberg, PHD,* Stefan Jovinge, MD, PHD,† Håkan Arheden, MD, PHD*

Lund, Sweden

O B J E C T I V E S Our goal was to validate myocardium at risk on T2-weighted short tau inversion

recovery (T2-STIR) cardiac magnetic resonance (CMR) over time, compared with that seen with perfusion

single-photon emission computed tomography (SPECT) in patients with ST-segment elevation myocar-

dial infarction, and to assess the amount of salvaged myocardium after 1 week.

B A C K G R O U N D To assess reperfusion therapy, it is necessary to determine howmuch myocardium

is salvaged by measuring the final infarct size in relation to the initial myocardium at risk of the left

ventricle (LV).

M E T H O D S Sixteen patients with first-time ST-segment elevation myocardial infarction received99mTc tetrofosmin before primary percutaneous coronary intervention. SPECT was performed within 4 h

and T2-STIR CMR within 1 day, 1 week, 6 weeks, and 6 months. At 1 week, patients were injected with

a gadolinium-based contrast agent for quantification of infarct size.

R E S U L T S Myocardium at risk at occlusion on SPECT was 33 � 10% of the LV. Myocardium at risk on

T2-STIR did not differ from SPECT, at day 1 (29 � 7%, p � 0.49) or week 1 (31 � 6%, p � 0.16) but

declined at week 6 (10 � 12%, p � 0.0096 vs. 1 week) and month 6 (4 � 11%, p � 0.0013 vs. 1 week).

There was a correlation between myocardium at risk demonstrated by T2-STIR at week 1 and

myocardium at risk by SPECT (r2 � 0.70, p � 0.001), and the difference between the methods on

Bland-Altman analysis was not significant (�2.3 � 5.7%, p � 0.16). Both modalities identified

myocardium at risk in the same perfusion territory and in concordance with angiography. Final infarct

size was 8 � 7%, and salvage was 75 � 19% of myocardium at risk.

C O N C L U S I O N S This study demonstrates that T2-STIR performed up to 1 week after reperfusion

can accurately determine myocardium at risk as it was before opening of the occluded artery. CMR can

also quantify salvaged myocardium as myocardium at risk minus final infarct size. (J Am Coll Cardiol

Img 2009;2:569–76) © 2009 by the American College of Cardiology Foundation

From the *Cardiac MR Group, Department of Clinical Physiology, and the †Department of Cardiology, Lund UniversityHospital, Lund, Sweden. This study has been funded by the Swedish Research Council, the Swedish Heart and LungFoundation, the Medical Faculty at Lund University, Sweden, and the region of Scania, Sweden.

Manuscript received July 23, 2008; revised manuscript received October 17, 2008, accepted November 21, 2008.

Msiphoa(smfid

dcbtnpwsoTc

Satdof

fsaffhqStrSa

M

SptphaswMa

mSfwcateaafsiwcMwsdeauiCVU

A

A

C

r

D

L

M

P

i

S

c

S

m

T

tau inversion recovery

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 . 2 , N O . 5 , 2 0 0 9

M A Y 2 0 0 9 : 5 6 9 – 7 6

Carlsson et al.

Myocardium at Risk on CMR

570

yocardial infarction (MI) size dependsheavily on duration of occlusion, collat-eral flow, and size of the initial myo-cardium at risk (1–3). Therefore, ST-

egment elevation myocardial infarction (STEMI)s treated with reperfusion therapy as soon asossible. The myocardium at risk, defined as theypoperfused myocardium during acute coronarycclusion, is an important measure, since a vari-ble amount of this area will become infarcted4). Therefore, to assess the efficacy of reperfu-ion therapy, it is necessary to determine howuch myocardium is salvaged by measuring the

nal infarct size in relation to the initial myocar-ium at risk.

See page 577

MI size has earlier been quantified bysingle-photon emission computed tomog-raphy (SPECT) after injection of atechnetium-labeled perfusion tracer at rest(5), where MI is indirectly detected as aregion with decreased myocardial perfu-sion. More recently, contrast delayed en-hancement (DE) cardiac magnetic reso-nance (CMR) has emerged as the newreference method for infarct localizationand quantification (6–8).

Myocardium at risk can be measured onSPECT by injection of a technetium-based tracer before opening of the oc-cluded vessel and is currently the mostwidely practiced technique to determinemyocardium at risk (9–11). However,

rawbacks in the use of SPECT to estimate myo-ardium at risk are the availability of a technetium-ased tracer, the need for injection of the isotope inhe acute setting of coronary occlusion, and scan-ing with a gamma camera after completion ofrimary percutaneous coronary intervention (PCI),hich could interfere with patient care in the acute

etting. These factors have limited the applicabilityf SPECT in measurement of myocardium at risk.herefore, new clinical methods to quantify myo-

ardium at risk need to be developed.T2-weighted short tau inversion recovery (T2-

TIR) highlights myocardial edema (12) presentfter MI (13) without the need of tracer adminis-ration. Moreover, CMR with T2-STIR allows theetection of the ischemic zone several days after theccluded coronary artery is opened (14,15). There-

r

y

ssion

ion

t

ore, CMR with T2-STIR can potentially be used s

or quantification of myocardium at risk. Previoustudies have shown the use of T2-STIR imaging forcute infarction in both reperfused and nonreper-used infarcts (12,14–17) and validated T2-STIRor myocardium at risk in animals (14). There are,owever, no validation studies in humans for theuantification of myocardium at risk using T2-TIR CMR. Hence, the purpose of this study waso validate the measurement of the myocardium atisk on T2-STIR over time, in comparison withPECT in humans with acute MI, and to assess themount of salvaged myocardium after 1 week.

E T H O D S

tudy population and design. The study was ap-roved by the local ethics committee, and all pa-ients gave their written informed consent. Sixteenatients (age 64 � 12 years, 14 men) with noistory of MI, presenting with acute STEMI due ton occluded coronary artery, were included in thetudy. All patients were treated by primary PCIith coronary stents, resulting in Thrombolysis Inyocardial Infarction flow grade 3 in the opened

rtery and received glycoprotein IIb/IIIa inhibitor.Before primary PCI, 99mTc tetrofosmin was ad-inistered intravenously, and myocardial perfusion

PECT was performed 3 to 4 h after primary PCIor determination of myocardium at risk. CMRith T2-STIR was undertaken 1 week after revas-

ularization. In 1 patient, adequate T2-STIR im-ges at 1 week were not obtained. To determine theime evolution of the increase in T2-STIR signal,arly imaging was performed at 1 day in 8 patientsnd within 2 days in 2 patients (hereafter describeds day 1), and 9 patients underwent CMRollow-up at 6 weeks and 6 months. For compari-on, all patients underwent DE-CMR with admin-stration of gadolinium at 1 week. The culprit vesselas identified on angiography by 2 observers in

onsensus.yocardial perfusion SPECT. Patients were injectedith 500 to 700 MBq 99mTc tetrofosmin (Amer-

ham Health, Buckinghamshire, United Kingdom),epending on body weight. Myocardial perfusionlectrocardiogram-gated SPECT was performedccording to the standard clinical protocol at rest,sing a dual-head camera. Nine patients weremaged using an ADAC Vertex camera (Milpitas,alifornia), 7 patients with a cardiac dedicated GEentri camera (GE Healthcare, Buckinghamshire,nited Kingdom). The patients were placed in a

B B R E V I A T I O N S

N D A C R O N YM S

MR � cardiac magnetic

esonance

E � delayed enhanced

V � left ventricle/ventricula

I � myocardial infarction

CI � percutaneous coronar

ntervention

PECT � single-photon emi

omputed tomography

TEMI � ST-segment elevat

yocardial infarction

2-STIR � T2-weighted shor

upine position and imaged in steps of 5.6° using a

6AGawlwmsmLdnMwCsGIcseihswScp(h1tgtSg

CocbCamameo1

ss

dmueawawbdwctsmSsuSubrfiospam

R

MaFfDaapd


M A Y 2 0 0 9 : 5 6 9 – 7 6

Carlsson et al.


571

4 � 64 matrix with a pixel size of 5.02 mm for theDAC camera and a pixel size of 6.4 mm for theE Ventri camera. Image acquisition time was

pproximately 15 min, but was extended to 25 minhen imaging was performed after 3 h. Short- and

ong-axis images covering the left ventricle (LV)ere reconstructed. Myocardium at risk was deter-ined using an in-house developed segmentation

oftware (segment version 1.702) (18). The auto-atic segmentation finds the centerline through theV wall and identifies the endocardium and epicar-ium based on an individually estimated wall thick-ess and signal intensity values within the image.anual adjustment of the automatic delineation

as sometimes required in the LV outflow region.MR. CMR was performed on either of 2 1.5-Tystems: Magnetom Vision (Siemens, Erlangen,ermany) with a CP body array coil, or Philips

ntera CV (Philips, Best, the Netherlands) with aardiac synergy coil. All subjects were placed in aupine position, and images were acquired at end-xpiratory breath hold with electrocardiogram gat-ng. Initial scout images were acquired to locate theeart, and a T2-weighted double inversion blooduppressed turbo spin echo sequence (T2-STIR)as employed to depict myocardium at risk. T2-TIR images were acquired in the short-axis view,overing the LV from the base to apex. Imagearameters for T2-STIR were: echo time, 43 msSiemens), or 100 ms (Philips); repetition time, 2eart beats; number of averages, 2; inversion time,80 ms; image resolution, 1.5 � 1.5 mm; slicehickness, 10 mm (Siemens), or 8 mm with a sliceap of 2 mm (Philips). When acquiring images withhe cardiac synergy coil, parallel imaging withENSE � 1 was used to minimize signal inhomo-eneities due to differences in coil sensitivity.

Infarct quantification was performed on DE-MR 30 � 9 min after intravenous administrationf 0.2 mmol/kg extracellular gadolinium-basedontrast agent (gadoteric acid, Gd-DOTA; Guer-et, Gothia Medical AB, Billdal, Sweden). DE-MR with an inversion-recovery turbo fast low-

ngle shot sequence (Siemens) (slice thickness 10m, field of view 380 mm, matrix 126 � 256, flip

ngle 25°, repetition time 100 ms, echo time 4.8s) or an inversion-recovery balanced turbo field

cho sequence (Philips) (slice thickness 8 mm, fieldf view 340 mm, repetition time 3.14 ms, echo time.58 ms) was performed, covering the LV.The CMR images were analyzed using the same

oftware as for the SPECT images (segment ver-
ion 1.702) (19). Observers were blinded to patient c
ata and time of acquisition after infarction. Theyocardium in each LV short-axis slice was man-

ally segmented by tracing the endocardial andpicardial borders. Regions of myocardium at risknd MI were identified as hyperenhanced regions,ithin the T2-STIR images and DE-CMR im-

ges, respectively. The myocardium at risk regionas delineated manually by independent andlinded observers, and the infarcted region waselineated automatically as previously describedith manual adjustment when needed (19). Myo-

ardium at risk size and MI size were defined as theotal amount of myocardium at risk/MI in allhort-axis slices and expressed as percentage of LVass.tatistical methods. Continuous variables are pre-ented as mean � SD. Pearson’s correlation wassed to determine the relationship between T2-TIR and SPECT. Two-tailed paired t test wassed to detect differences in myocardium at risketween techniques, differences in myocardium atisk on T2-STIR at different time points, and thenal infarct size compared with myocardium at riskn T2-STIR. A p value �0.05 was consideredignificant. Agreement between methods was ex-ressed as mean difference � SD, and the limits ofgreement were shown in a Bland-Altman graph asean difference � 2 SD.

E S U L T S

yocardium at risk. Table 1 shows the patient char-cteristics for all 16 patients included in this study.igure 1 shows multislice images from base to apex

rom T2-STIR at week 1, SPECT at day 1, andE-CMR at week 1 from 1 patient. Myocardium

t risk presented as a perfusion defect on SPECTnd hyperenhanced regions on T2-STIR. In allatients, T2-STIR and SPECT identified myocar-ium at risk in the same perfusion territory and in

Table 1. Patient Characteristics

Male gender 14 (88%)

Age (yrs) 64 � 12 (36–83)

Body mass index (kg/m2) 27 � 3 (24–33)

Occluded artery by angiography

LAD 7 (44%)

RCA 8 (50%)

LCX 1 (6%)

Values are presented as n (%) or as mean � SD (range).LAD � left anterior descending artery; LCX � left circumflex artery; RCA �right coronary artery.

oncordance with angiography. The resulting in-

fr

t(r

wT1�wm

at w

performed.


M A Y 2 0 0 9 : 5 6 9 – 7 6

Carlsson et al.


572

arction on DE-CMR was present in the sameegion in all patients.

Myocardium at risk on SPECT was 33 � 10% ofhe LV and on T2-STIR at day 1 it was 29 � 7%p � 0.49 compared with SPECT). Myocardium atisk on week 1 was 31 � 6% (p � 0.16 compared

Figure 1. Myocardium at Risk by SPECT and T2-STIR CMR and F

Short-axis slices at the same ventricular level of single-photon emisinversion recovery (T2-STIR) week 1, and delayed enhanced-cardiacright coronary occlusion resulting in an inferior infarct. The epicardaffected region is traced in yellow. Note the similarity in size of thesion by SPECT and T2-STIR CMR 1 week later, showing that T2-STIR

yocardium at Risk and Percentage Infarcted and Salvaged Myoc

T2-STIR

SPECT(%)

Day 1(%)

Week 1(%)

Week 6(%)

28 25 — 16

38 — 36 —

37 40 36 0

24 27 27 27

26 NA 24 13

38 30 36 0

28 — 23 —

39 25 26 30

34 34 34 0

39 39 31 —

20 18 25 5

20 27 25 —

46 — 42 —

31 — 29 0

25 — 30 —

56 — 43 —

3 � 10 29 � 7 31 � 6 10 � 12

p � 0.49 p � 0.16 p � 0.00

rcent of myocardium salvaged in comparison with T2-STIR week 1; †salvaged m

ith SPECT) (Table 2). Myocardium at risk on2-STIR at day 1 and week 1 did not differ (p �.00). Over time, myocardium at risk declined to 10

12% at week 6 (p � 0.0096 vs. T2-STIR at 1eek and p � 0.0026 vs. SPECT) and 4 � 11% atonth 6 (p � 0.0013 vs. T2-STIR at 1 week and

Infarct Size by DE-CMR in 1 Typical Patient

computed tomography (SPECT) day 1, T2-weighted short taunetic resonance (DE-CMR) week 1 in a patient with reperfusedis traced in green, the endocardium is traced in red, and thected region between perfusion defect size during coronary occlu-eek 1 can be used to quantify myocardium at risk.

ium

DE

Month 6(%)

Week 1(%)

SalvagedMyocardium*

(%)

0 8 68.0†

— 16 55.6

0 2 94.4

0 16 40.7

5 7 70.8

0 19 47.2

— 1 95.7

32 10 61.5

— 2 94.1

0 0 100.0

— 2 92.0

— 1 96.0

— 19 54.8

0 6 79.3

0 10 66.7

— 7 83.7

4 � 11 8 � 7 75 � 19

p � 0.001

rdium in comparison with T2-STIR day 1.

inal

sionmagiumaffe

Table 2. Percentage M ard

Patient #

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

Mean � SD 3

vs. SPECT 26

*Salvaged myocardium: pe yocaDE � delayed enhanced; NA � unable to appoint occluded vessel; SPECT � single-photon emission computed tomography; T2-STIR � T2-weighted short tau inversion recovery; – � no imaging

pwtt2rSsdSp

wss1aba��Mws(dwOomp1wbaapt

abbreviations as in Figure 1.


M A Y 2 0 0 9 : 5 6 9 – 7 6

Carlsson et al.


573

� 0.001 vs. SPECT). In patient 5, the observersere unable to designate the culprit vessel within

he T2-STIR images at day 1; however, at week 1he myocardium at risk could be identified. FigureA shows the relationship between myocardium atisk demonstrated on T2-STIR at week 1 andPECT (r2 � 0.70, p � 0.001). There was notatistical significant difference between myocar-ium at risk on T2-STIR compared with that ofPECT on Bland-Altman analysis (�2.3 � 5.7,� 0.16) (Fig. 2B).The evolution of myocardial edema in a patient

ith left anterior descending artery occlusion ishown in Figure 3; SPECT is shown for compari-on. The size of the affected region is similar at dayand week 1 on T2-STIR, but cannot be detected

t week 6 in this patient. Figure 4 displays the ratioetween T2-STIR and SPECT at the differentcquisition times, day 1 (0.97 � 0.20), week 1 (0.97

0.18), week 6 (0.35 � 0.40), and month 6 (0.110.27).yocardial salvage. The mean infarct size at week 1as 8 � 7% LV (range 0% to 19%), significantly

maller than myocardium at risk (p � 0.001)Table 2). Myocardial salvage, calculated as theifference of T2-STIR and DE-CMR at week 1as, on average, 75 � 19% (range 41% to 100%).ne patient did not have any infarcted myocardium

n DE-CMR after the acute revascularization,eaning that all myocardium was salvaged by

rimary PCI; salvage in this patient was therefore00%. This patient underwent revascularizationithin 1 h after pain onset. In patient 1, infarct sizey DE-CMR was compared with that of T2-STIRt day 1 in absence of T2-STIR data at week 1. Themount of salvaged myocardium is exemplified in 1atient in Figure 5. Note that the salvage occurs inhe subepicardial layers of the myocardium.

Figure 3. T2-STIR Over Time in 1 Typical Patient

Midventricular short-axis slices of automatically delineated perfusiondelineation of T2-STIR over time, in the same patient with a left antendocardium is traced in red. The myocardium at risk is delineated
imaging is similar at day 1 and week 1, but disappears at week 6. Abbr
SPECT (%LVM)10 20 30 40 50 60

A

B

SPECT (%LVM)

T2-

ST

IR -

SP

EC

T (

%LV

M)

T2-

ST

IR (

%LV

M)

10

20

30

40

50

60

0

20

20 50 60

40

-40

-20

r =0.70p<0.001

2

30 40

Figure 2. Agreement Between T2-STIR and SPECT

(A) T2-STIR at week 1 versus perfusion defect during coronary occlusion bySPECT (r2 � 0.70, p � 0.001) with the line of identity. (B) Bland-Altmangraph showing the difference between myocardium at risk quantified onT2-STIR (week 1) and SPECT versus the reference method SPECT. The differ-ence between T2-STIR and SPECT was �2.3 � 5.7%. Solid lines � mean ofT2-STIR � SPECT; dotted lines � �2 SD. LVM � left ventricular mass; other

defect during coronary occlusion by SPECT and blinded manualerior descending occlusion. The epicardium is traced in green; thein yellow. Note that the signal of the affected region on T2-STIR

eviations as in Figure 1.

D

TmpSrtaacitMhfrT

rtIaePr(scTtOhs

ca(ddaripseiS(tsr

iair

abbreviatio


M A Y 2 0 0 9 : 5 6 9 – 7 6

Carlsson et al.


574

I S C U S S I O N

his study is the first to validate the quantification ofyocardium at risk on T2-STIR over time in com-

arison with that of SPECT, in patients withTEMI. The main finding is that T2-STIR accu-ately identifies and quantifies myocardium at risk upo 1 week after opening of the occluded coronaryrtery. Thus, the success of reperfusion therapy can bessessed on CMR within 1 image session utilizing theombination of DE and T2-STIR to measure finalnfarct size and myocardium at risk, respectively, andhereby measuring myocardial salvage.yocardium at risk. Earlier studies (12,14–17,20)ave demonstrated the use of T2-weighted imagingor acute infarction, in both reperfused and non-eperfused infarcts. The current study showed that2-STIR accurately assessed myocardium at risk in

CMR day 1 CMR week 1 CMR week 6 CMR month 6

2-STIR Over Time in All Patients in Relation to Perfusioncclusion by SPECT Imaging

een T2-STIR and SPECT at day 1, week 1, week 6, and month 60.20; 0.97 � 0.18; 0.35 � 0.40; and 0.11 � 0.27, respectively,

at the T2-STIR signal at day 1 and week 1 agree with perfusioncclusion by SPECT imaging. The presence of edema at 6 monthsin 2 of 9 patients. CMR � cardiac magnetic resonance; otherns as in Figure 1.

Figure 5. Myocardial Salvage by CMR

Midventricular short-axis slices in a patient with a left circumflexdium is traced in red. (Left) T2-STIR image showing the myocardcardial infarction (delineated in yellow). (Right) T2-STIR image w
myocardium (blue area). Abbreviations as in Figure 1.
eperfused infarcts in patients using SPECT at theime of coronary occlusion as the reference method.n agreement with studies on humans (21,22) andnimals (12,14,23), the area with high T2 signalxceeded that of irreversible injury in acute MI.athology studies in humans have shown complete

esorption of edema after acute MI within 5 weeks13), and Aletras et al. (14) showed that edema wastill present at the 2-month follow-up CMR in aanine model. No evidence of edema was found on2-STIR in 7 of the 9 patients at 6 months after

he acute coronary occlusion in the present study.ne may hypothesize that these 2 patients may

ave had residual or recurring ischemia within theame myocardial region.

Myocardial ischemia increases cellular and extra-ellular osmolarity, alters plasma membrane perme-bility, causes cell swelling and interstitial edema24). Quantification of interstitial edema has beenemonstrated in experimental studies of myocar-ium at risk after reperfusion by light microscopy,utoradiography, and contrast-enhanced inversionecovery echo-planar CMR (25). Once reperfusions established, an inflammatory reaction within theerfusion bed of the culprit vessel takes place, whichignificantly increases tissue edema. Myocardialdema is therefore a consistent feature of acuteschemia (26), which can be demonstrated on T2-TIR (12) although the myocardium is reperfused14,15). The results of the present study showedhat the amount of edema present after 1 week wasimilar to the amount of edema at 1 day aftereperfusion.

Advantages of T2-STIR over SPECT in assess-ng myocardium at risk are: 1) no need of tracerdministration; 2) no complicated imaging interfer-ng with patient care in the acute setting; 3) noadiation dose; and 4) higher spatial resolution.

ery occlusion. The epicardium is traced in green; the endocar-at risk (delineated in yellow). (Middle) DE-CMR showing myo-

inclusion of infarcted region showing the amount of salvaged

Rat

io T

2-S

TIR

at

bas

elin

e an

d d

uri

ng

follo

w-u

p/ S

PE

CT

at

bas

elin

e

0.2

0.4

0.6

0.8

1.0

1.2

1.4

Figure 4. TDefect at O

Ratio betwwere 0.97 �

showing thdefect at owas found

artiumith

ATsTMpacwfirauftwastmsDsaCd(tsascmsTpoHpfMoSofifptpOmH

smeaStuiirptdaSPcs(iiccpm

C

Tqiadwmttcbu

ATALt

R

DH


M A Y 2 0 0 9 : 5 6 9 – 7 6

Carlsson et al.


575

lthough quantification of myocardium at risk by2-STIR at day 1 and week 1 did not show any

ignificant difference, image quality was in favor to2-STIR imaging at week 1.yocardial salvage. The clinical usefulness of theresent study is mainly that the size of myocardiumt risk and MI, and therefore myocardial salvage,an be assessed at the same image session within 1eek after acute coronary occlusion, without inter-

ering with patient care in the acute setting. In acutenfarcts, the necrotic region is surrounded by aegion of reversible injury, characterized by edema,nd consists of myocardial tissue that has notndergone irreversible injury. This region is re-erred to as the salvageable myocardium at risk, andhe region of irreversible injury, if left untreated,ill expand in the first hours after acute MI in

ccordance with the wavefront phenomena de-cribed by Reimer et al. (27). To assess reperfusionherapy, it is necessary to determine how muchyocardium is salvaged by measuring final infarct

ize in relation to the initial myocardium at risk.E-CMR has the ability to quantify final infarct

ize (6,7), and, as shown in this study, T2-STIRccurately quantifies myocardium at risk. Therefore,MR has the ability to calculate salvaged myocar-ium within the same imaging acquisition session21). This can be used to evaluate new drugs andherapeutic procedures aimed at reducing infarctize without interfering with patient care in thecute setting. Indeed, Ibanez et al. (28) recentlyhowed that metoprolol administered before revas-ularization increased myocardial salvage in a pigodel where myocardium at risk and final infarct

ize were assessed with T2-STIR and DE-CMR.he results of the current study in patients withrimary PCI showed that 75% (range 41% to 100%)f the initial myocardium, on average, was salvaged.ence, this line of research can be applied in

atient populations providing a salvageable indexor each patient undergoing primary PCI after acute

I and would potentially increase the knowledgen infarct-related tissue injury.tudy limitations. The present study was performedn a limited number of patients, all presenting withrst-time STEMI and undergoing successful reper-usion. Thus, how this would translate to otheropulations, for example unsuccessful revasculariza-ion, patients treated with thrombolytic therapy, oratients with previous MIs, is not known (15).nly 2 women were included in this study, andore data on female subjects are, therefore, needed.
owever, T2-STIR images from the 2 female a
ubjects showed similar enhancement as the 14ale patients, and there is no a priori reason to

xpect a gender difference in edema formation aftercute MI. Quantification of myocardium at risk byPECT has previously been performed using ses-amibi tracers. In the present study, tetrofosmin wassed; however, sestamibi and tetrofosmin are usednterchangeably in patient studies (29). Image qual-ty can be a limitation in assessing myocardium atisk by T2-STIR, and the image quality of theresent study did not allow for automated segmen-ation. New sequences, however, are continuouslyeveloped to overcome this problem (30,31). Im-ges were acquired using a body array coil with theiemens scanner and a cardiac surface coil with thehilips scanner. It has been suggested that a bodyoil has a more homogenous reception whereas theurface coil has an inherent signal intensity gradient21). However, image reconstruction with parallelmaging techniques uses differences in coil sensitiv-ty and, therefore, compensates inhomogenous re-eption. Thus, the images obtained with the surfaceoil in the present study was acquired using aarallel imaging factor (SENSE factor) of 1 toinimize this effect.

O N C L U S I O N S

his is the first study to validate T2-STIR foruantification of myocardium at risk against anndependent method (SPECT) in patients withcute MI after reperfusion therapy. The resultsemonstrate that T2-STIR performed up to 1eek after reperfusion can accurately determineyocardium at risk as it was before opening of

he occluded artery. The result of reperfusionherapy can, therefore, be assessed clinically byalculating myocardial salvage as the differenceetween myocardium at risk and final infarct sizesing CMR.

cknowledgmentshe authors would like to acknowledge Ann-Helenrvidsson and Christel Carlander, both with theund Cardiac MR Group, for their skillful assis-

ance with image acquisition.

eprint requests and correspondence: Dr. Håkan Arheden,epartment of Clinical Physiology, Lund Universityospital, Lund SE-22185, Sweden. E-mail: hakan.

[email protected].

mailto:[email protected]

mailto:[email protected]


M A Y 2 0 0 9 : 5 6 9 – 7 6

Carlsson et al.


576

R

1

1

2

2

2

2

2

2

2

3

3

Ky

E F E R E N C E S

1. Hedstrom E, Frogner F, Åström-Olsson K, Öhlin H, Arheden H.Myocardial infarct size in relation tomyocardium at risk versus duration ofischemia in humans: comparison withdifferent species (abstr). J CardiovascMagn Reson 2007;9:363.

2. Feiring AJ, Johnson MR, KioschosJM, Kirchner PT, Marcus ML, WhiteCW. The importance of the determi-nation of the myocardial area at risk inthe evaluation of the outcome of acutemyocardial infarction in patients. Cir-culation 1987;75:980–7.

3. Lowe JE, Reimer KA, Jennings RB.Experimental infarct size as a functionof the amount of myocardium at risk.Am J Pathol 1978;90:363–79.

4. Kloner RA, Bolli R, Marban E, ReinlibL, Braunwald E. Medical and cellularimplications of stunning, hibernation,and preconditioning: an NHLBI work-shop. Circulation 1998;97:1848–67.

5. Gibbons RJ, Valeti US, Araoz PA, JaffeAS. The quantification of infarct size.J Am Coll Cardiol 2004;44:1533–42.

6. Arheden H, Saeed M, Higgins CB, etal. Reperfused rat myocardium sub-jected to various durations of ischemia:estimation of the distribution volume ofcontrast material with echo-planar MRimaging. Radiology 2000;215:520–8.

7. Kim RJ, Fieno DS, Parrish TB, et al.Relationship of MRI delayed contrastenhancement to irreversible injury, in-farct age, and contractile function.Circulation 1999;100:1992–2002.

8. Carlsson M, Wijns W, Arheden H,Higgins C, Saeed M, Magnetic reso-nance imaging as a potential goldstandard for infarct quantification. JElectrocardiol 2008;41:614–20.

9. De Coster PM, Wijns W, Cauwe F,Robert A, Beckers C, Melin JA. Area-at-risk determination by technetium-99m-hexakis-2-methoxyisobutyl isoni-trile in experimental reperfusedmyocardial infarction. Circulation1990;82:2152–62.

0. Gibbons RJ, Verani MS, BehrenbeckT, et al. Feasibility of tomographic99mTc-hexakis-2-methoxy-2-methyl-propyl-isonitrile imaging for the as-sessment of myocardial area at riskand the effect of treatment in acutemyocardial infarction. Circulation1989;80:1277–86.

1. Sinusas AJ, Trautman KA, Bergin JD,et al. Quantification of area at riskduring coronary occlusion and degreeof myocardial salvage after reperfusionwith technetium-99m methoxyisobu-tyl isonitrile. Circulation 1990;82:
1424–37.
12. Garcia-Dorado D, Oliveras J, Gili J,et al. Analysis of myocardial oedemaby magnetic resonance imaging earlyafter coronary artery occlusion with orwithout reperfusion. Cardiovasc Res1993;27:1462–9.

13. Fishbein MC, Maclean D, MarokoPR. The histopathologic evolution ofmyocardial infarction. Chest 1978;73:843–9.

14. Aletras AH, Tilak GS, Natanzon A,et al. Retrospective determination ofthe area at risk for reperfused acutemyocardial infarction with T2-weighted cardiac magnetic resonanceimaging: histopathological and dis-placement encoding with stimulatedechoes (DENSE) functional valida-tions. Circulation 2006;113:1865–70.

15. Tilak GS, Hsu LY, Hoyt RF Jr., AraiAE, Aletras AH. In vivo T2-weightedmagnetic resonance imaging can accu-rately determine the ischemic area atrisk for 2-day-old nonreperfused myo-cardial infarction. Invest Radiol 2008;43:7–15.

16. Abdel-Aty H, Zagrosek A, Schulz-Menger J, et al. Delayed enhancementand T2-weighted cardiovascular mag-netic resonance imaging differentiateacute from chronic myocardial infarc-tion. Circulation 2004;109:2411–6.

17. Takahashi N, Inoue T, Oka T, et al.Diagnostic use of T2-weighted in-version-recovery magnetic resonanceimaging in acute coronary syndromescompared with 99mTc-pyrophosphate,123I-BMIPP and 201TlCl single photonemission computed tomography. Circ J2004;68:1023–9.

18. Soneson H, Ubachs J, Ugander MAr-heden J, Heiberg E. An improvedmethod for automatic segmentation ofthe left ventricle in myocardial perfusionSPECT. J Nucl Med 2009;50:205–13.

19. Heiberg E, Ugander M, Engblom H,et al. Automated quantification ofmyocardial infarction from MR im-ages by accounting for partial volumeeffects: animal, phantom, and humanstudy. Radiology 2008;246:581–8.

20. Stork A, Muellerleile K, BansmannPM, et al. Value of T2-weighted, first-pass and delayed enhancement, and cineCMR to differentiate between acute andchronic myocardial infarction. Eur Ra-diol 2007;17:610–7.

21. Friedrich MG, Abdel-Aty H, TaylorA, Schulz-Menger J, Messroghli D,Dietz R. The salvaged area at risk inreperfused acute myocardial infarctionas visualized by cardiovascular mag-netic resonance. J Am Coll Cardiol2008;51:1581–7.

22. Stork A, Lund GK, Muellerleile K, et
al. Characterization of the peri-infarction zone using T2-weighted m
MRI and delayed-enhancement MRIin patients with acute myocardial in-farction. Eur Radiol 2006;16:2350–7.

3. Saeed M, Wagner S, Wendland MF,Derugin N, Finkbeiner WE, HigginsCB. Occlusive and reperfused myo-cardial infarcts: differentiation withMn-DPDP–enhanced MR imaging.Radiology 1989;172:59–64.

4. Steenbergen C, Hill ML, JenningsRB. Volume regulation and plasmamembrane injury in aerobic, anaero-bic, and ischemic myocardium invitro. Effects of osmotic cell swellingon plasma membrane integrity. CircRes 1985;57:864–75.

5. Arheden H, Saeed M, Higgins CB, etal. Measurement of the distributionvolume of gadopentetate dimeglumineat echo-planar MR imaging to quan-tify myocardial infarction: comparisonwith 99mTc-DTPA autoradiographyin rats. Radiology 1999;211:698–708.

6. Zhao ZQ, Corvera JS, Halkos ME, etal. Inhibition of myocardial injury byischemic postconditioning during reper-fusion: comparison with ischemic pre-conditioning. Am J Physiol Heart CircPhysiol 2003;285:H579–88.

7. Reimer KA, Lowe JE, RasmussenMM, Jennings RB. The wavefrontphenomenon of ischemic cell death. 1.Myocardial infarct size vs duration ofcoronary occlusion in dogs. Circula-tion 1977;56:786–94.

8. Ibanez B, Prat-Gonzalez S, SpeidlWS, et al. Early metoprolol adminis-tration before coronary reperfusion re-sults in increased myocardial salvage:analysis of ischemic myocardium atrisk using cardiac magnetic resonance.Circulation 2007;115:2909–16.

9. Adams GL, Shaw LK, Tuttle RH,Hanson MW, Pagnanelli R, Borges-Neto S. Prediction of mortality inpatients with coronary artery diseaseundergoing vasodilator stress testing:a comparison between 99mTc-tetrofosmin and 99mTc-sestamibi.Nucl Med Commun 2007;28:457–63.

0. Kellman P, Aletras AH, Mancini C,McVeigh ER, Arai AE. T2-preparedSSFP improves diagnostic confidence inedema imaging in acute myocardial in-farction compared to turbo spin echo.Magn Reson Med 2007;57:891–7.

1. Aletras AH, Kellman P, DerbyshireJA, Arai AE. ACUT2E TSE-SSFP: ahybrid method for T2-weighted im-aging of edema in the heart. MagnReson Med 2008;59:229–35.

ey Words: myocardium at riskT2-STIR y CMR y salvaged
yocardium.

myocardium at risk after acute infarction in humans on cardiac magnetic resonance: quantitative...

Documents