Gunnar K. Lund, MDCharles B. Higgins, MDMichael F. Wendland, PhDNorbert Watzinger, MDHanns-Joachim Weinmann,

PhDMaythem Saeed, DVM, PhD

Index terms:Heart, experimental studies,

511.12143Heart, MR, 511.121413,

511.121416, 511.12143Myocardium, infarction, 511.771Myocardium, ischemia, 511.1939Myocardium, MR, 511.121413,

511.121416, 511.12143

Published online before print10.1148/radiol.2213010533

Radiology 2001; 221:676–682

Abbreviations:LV 5 left ventricleTTC 5 triphenyltetrazolium chloride

1 From the Department of Radiology,University of California, San Francisco,505 Parnassus Ave, Rm L-308, SanFrancisco, CA 94143-0628 (G.K.L.,C.B.H., M.F.W., N.W., M.S.); and Scher-ing AG, Berlin, Germany (H.J.W.). Re-ceived February 28, 2001; revision re-quested April 2; revision received June7; accepted July 5. Supported in partby a gift from Schering AG, Berlin,Germany. G.K.L. supported in part bya scholarship from the University Hos-pital Eppendorf, Hamburg, Germany.N.W. supported in part by a scholar-ship from the Max-Kade Foundation,New York, NY. Address correspon-dence to M.S. (e-mail: maythem.saeed@radiology.ucsf.edu).© RSNA, 2001

Author contributions:Guarantors of integrity of entire study,M.S., G.K.L., C.B.H.; study conceptsand design, M.S., G.K.L., C.B.H.; liter-ature research, M.S., G.K.L.; experi-mental studies, M.S., G.K.L., M.F.W.,N.W., H.J.W.; data acquisition, M.S.,G.K.L.; data analysis/interpretation,G.K.L., M.S.; statistical analysis, G.K.L.,M.S.; manuscript preparation, G.K.L.,M.S.; manuscript definition of intellec-tual content and editing, M.S., G.K.L.,C.B.H.; manuscript revision/review andfinal version approval, all authors.

Assessment of NicorandilTherapy in IschemicMyocardial Injury by UsingContrast-enhanced andFunctional MR Imaging1

PURPOSE: To determine the potential of mesoporphyrin– and gadopentetatedimeglumine–enhanced and functional magnetic resonance (MR) imaging in theassessment of the acute effect of nicorandil on ischemic injury of the myocardium.

MATERIALS AND METHODS: Spin-echo MR imaging was used to monitorchanges in myocardial contrast and function in reperfused myocardial injury. Inver-sion-recovery echo-planar MR imaging was used to depict the injured region.Myocardial injury in rats was produced by using 30 minutes of coronary occlusionfollowed by 24 hours reperfusion. Nicorandil (n 5 9) was infused during occlusionand early reperfusion. Control animals (n 5 11) received no therapy. At 24 hours,after administration of mesoporphyrin and gadopentetate dimeglumine and histo-chemical staining, the function and size of the injured region of the left ventricle (LV)were determined. A t test was used to compare data between groups of animals,whereas regression and Bland-Altman analyses were used to determine correlationand agreement between MR imaging and histomorphometry, respectively.

RESULTS: Treated animals showed reduced infarction size as compared with thecontrol group from 25.6% 6 7.9 (SD) to 7.9% 6 6.8 of LV myocardial area (P ,.001), as defined with mesoporphyrin-enhanced MR imaging; while the size of therim increased from 10.8% 6 10.0 to 16.1% 6 14.4 (P , .05). The diastolic-midventricular cavity area was smaller in treated animals (15.2 mm2 6 4.3) com-pared with the control group (28.5 mm2 6 7.9; P , .001). At functional MRimaging, nicorandil improved systolic reduction in LV cavity area (57.5% 6 17.3)compared with the control group (38.0% 6 16.0; P , .05) and preserved regionalLV wall thickening at the site of injury (12.2% 6 11.1 in treated group vs 0.3% 68.6 in the control group; P , .05).

CONCLUSION: Contrast material–enhanced MR imaging has the potential to dem-onstrate reduction in size of ischemically injured myocardium, whereas functionalMR imaging demonstrated the recovery of LV function 24 hours after nicorandiltherapy.

Nicorandil is an antianginal drug, clinically approved in Europe and Japan, that hascardioprotective properties by causing vasodilation like a nitrate (1) and by openingmyocardial adenosine triphosphate–sensitive potassium channels (2). Recent findings inpatients, by using echocardiography, have indicated that intravenously administerednicorandil preserves microvascular integrity and function of the left ventricle (LV) afteracute myocardial infarction (3) and improves ischemic tolerance during coronary angio-plasty (4). Physiologic studies (4–8) in animal models have indicated that nicorandildecreases LV end-diastolic and arterial pressures. Sato et al (9) found that nicorandil exertsa direct cardioprotective effect on heart muscle cells, an effect mediated by selectiveactivation of mitochondrial adenosine triphosphate–sensitive potassium channels. Otherstudy findings (5–8) have demonstrated that nicorandil reduces the size of infarction.

676

Imagawa et al (8) reported that 5-hy-droxydecanoate, the selective mitochon-drial adenosine triphosphate–sensitivepotassium channel inhibitor, abolishesthe infarct size-limiting effect of nicoran-dil.

Contrast material–enhanced magneticresonance (MR) imaging techniques havebeen developed to measure the size ofinfarction (10–12). To our knowledge,Marchal et al (10) first reported that bis-gadolinium-mesoporphyrin (mesoporphy-rin) is a necrosis-specific agent. Authors ofrecent studies (11,12) demonstrated thatmesoporphyrin enhances only necroticmyocardium and provides an accurateestimation of acute (24–28 hours after)infarction. In this time frame, use of extra-cellular nonspecific agents, such as gado-pentetate dimeglumine, results in overesti-mation of infarction size by inclusion of aportion of the area at risk (12,13). The over-estimated hyperenhanced portion of myo-cardium has been recently called “periin-farction zone” (12,14). Accordingly, it wasproposed that subtraction of the gadolin-ium-enhanced region from the mesopor-phyrin-enhanced region provided an esti-mate of the viable periinfarction zone inthe early stage of infarction (14). The pur-pose of this study was to determine thepotential of mesoporphyrin-enhanced, ga-dolinium-enhanced, and functional MRimaging in the assessment of the acute ef-fect of the cardioprotective drug nicorandilon ischemically injured myocardium.

MATERIALS AND METHODS

Myocardial Injury andExperimental Protocols

All experimental protocols receivedprevious approval from the Committeeon Animal Research at our institutionand were performed with the NationalInstitutes of Health guidelines for careand the use of laboratory animals. Rats(Simonson Lab, Modesto, Calif; n 5 20;weight range, 260–300 g) were anesthe-tized by using intraperitoneal adminis-tration of 50 mg of ketamine (Ketaset;Fort Dodge Labs, Fort Dodge, Iowa) perkilogram of body weight and 1.4 mg/kgxylazine (Anased; Lloyd Labs, Shenan-doah, Iowa). A catheter was placed in thetail vein for drug and contrast materialadministration. A thoracotomy on theleft side was performed under mechani-cal ventilation, and the left coronary ar-tery was occluded for 30 minutes by plac-ing a snare loop around the artery.

The two groups of rats were randomlyassigned to receive either a saline or nic-

orandil solution. At 15 minutes after oc-clusion, the animals in the treated group(n 5 9) received a bolus of 100 mg/kgfollowed by infusion of 25 mg/kg/minnicorandil for 60 minutes. This dose ofnicorandil was chosen because it wasfound to produce a 15–20-mm Hg de-cline in mean arterial pressure in dogs foralmost 1 hour after administration (5). Inhumans, the apparent elimination half-life (plasma half-life) of nicorandil is 1hour, and all metabolites are excreted in24 hours (4). The control group (n 5 11)received saline infusion. At 2 hours ofreperfusion, each animal received 0.05mmol/kg mesoporphyrin, and imageswere obtained at 24 hours of reperfusion.

Contrast-enhanced MR Imaging

Mesoporphyrin and gadopentetatedimeglumine were synthesized and sup-plied by Schering AG, Berlin, Germany.Mesoporphyrin is a necrosis-specific me-talloporphyrin that may bind to necrotictissue or cellular debris and, thus, is dis-tinguished from gadopentetate dimeglu-mine, which distributes in necrotic cellsand in the edematous periinfarction zone(10–12). The T1 and T2 relaxivities ofmesoporphyrin are two times higher (8.9and 12 sec21 mM21) than those of gado-pentetate dimeglumine (3.7 and 5.6sec21 mM21). The dose (0.05 mmol/kg)of mesoporphyrin used in this study wasone-sixth of the dose of gadopentetatedimeglumine. The 0.05 mmol/kg meso-porphyrin dose resulted in an equivalentincrease in signal intensity comparedwith a 0.3 mmol/kg gadopentetate dime-glumine dose. Detailed physiochemicalproperties of both contrast materialshave been previously described (15,16).Animals were again anesthetized for im-aging by using 60 mg/kg sodium pento-barbital (Nembutal Sodium Solution; Ab-bott Laboratories, North Chicago, Ill).Copper leads were inserted into a fore-limb and the lower chest wall and con-nected to an electrocardiographic monitor(AccuSync 6L; AMR, Milford, Conn) to pro-vide cardiac gating. Each rat was placedsupine in a custom-made birdcage radio-frequency coil (5.6-cm inner diameter and7.6-cm length), and MR images were ac-quired with a 2.0-T magnet (Bruker Instru-ments, Fremont, Calif). The magnet wasequipped with self-shielded gradient coils(Acustar S-150; Bruker Instruments) (620G/cm, 15-cm diameter). MR imaging wasperformed 24 hours after infarction by us-ing the following sequences.

Spin-echo MR imaging.—MultisectionT1-weighted transverse spin-echo images

were obtained, after administration ofmesoporphyrin and gadopentetate dime-glumine, to measure the size of the en-hanced region. Three adjacent equidis-tant (2-mm) sections were acquired at theapex, center, and base of the LV. Theacquisition parameters were a repetitiontime of 30 msec, an echo time of 12 msec,a section thickness of 2 mm, a field ofview of 5 3 5 cm, an image matrix of256 3 128 (0.20 mm per pixel), and anacquisition time of 2.5 minutes, depend-ing on the heart rate. After acquisitionof mesoporphyrin-enhanced MR images,gadopentetate dimeglumine was intrave-nously injected without removing theanimal from the magnet. A second set ofimages was acquired 30 minutes after in-jection of gadopentetate dimegluminewith the identical MR imaging parame-ters and section locations as with themesoporphyrin-enhanced images. LVfunction was demonstrated on a singlemidventricular short-axis plane with im-ages obtained at end diastole, defined asthe rise of QRS complex and end-systolicimages acquired at approximately 45% ofR-R interval (14).

Inversion-recovery echo-planar imaging.—Regional T1 was measured on a singlemidventricular section in the transverseplane 22 hours after administration ofmesoporphyrin. MR imaging parameterswere a repetition time of at least 6 sec-onds, an echo time of 10 msec, an inver-sion time of 20–1,000 msec, a slice thick-ness of 2 mm, a field of view of 5 3 5 cm,an image matrix of 64 3 64 (0.78 mm perpixel) and an acquisition time for eachimage of 33 msec. T1 values were ob-tained from a set of 20 images in whichthe time to inversion was incrementedfrom 20 to 1,000 msec, to detect the timeto inversion with null signal. T1 valueswere calculated from the time to inver-sion at the null point by using the rela-tion T1 5 time to inversion with nullsignal/ln 2 (17).

Postmortem Evaluation

After MR imaging, the left coronaryartery was again occluded, and 0.2 mL ofphthalocyanine blue dye (Engelhard,Louisville, Ky) was intravenously admin-istered to define the area at risk. The LVwas transversely sectioned into three2-mm-thick sections at the apex, center,and base that corresponded to MR im-ages. Both upper and lower surfaces ofthe stained slices were scanned with aflatbed scanner (Silverscanner IV; LaCie,Hillsboro, Ore). Each slice was then incu-bated in a 2% triphenyltetrazolium chlo-

Volume 221 z Number 3 Assessment of Nicorandil Therapy in Ischemic Myocardial Injury z 677

ride (TTC) solution (Sigma Chemical, StLouis, Mo) to define infarcted myocar-dium. Both faces of each slice were re-scanned and digitally stored for analysis.

Data Analysis

The size of mesoporphyrin- and gado-linium–enhanced regions on MR images,the true infarction size depicted withTTC staining, and the area at risk de-picted with blue dye were measured byusing computer-assisted planimetry andan imaging program (public domain NIHimage program developed at the UnitedStates National Institutes of Health, avail-able at rsb.info.nih.gov/nih-image/). Theperiinfarction zone was calculated bysubtracting the size of the mesoporphy-rin-enhanced region from the size of thegadolinium-enhanced region (12,14).Wall thickness was measured at end dias-tole and end systole in four regions,namely, the mesoporphyrin-enhancedregion, the rim of gadolinium-enhancedregion, and the remote posterior and sep-tal walls. The LV cavity area was mea-sured during end diastole and end systoleto calculate the systolic reduction of cav-ity size. Regional T1 values were com-puted from the inversion-recovery nullpoint (17).

Statistical Analysis

Data were expressed as the mean plusor minus the SD. The two-sample t testwas used to compare control and nic-orandil-treated animals. Measurementsin each group were analyzed by using thepaired t test. Linear regression and Bland-Altman analysis were used to determinecorrelation and agreement, respectively,between contrast-enhanced MR imagesand histomorphometry. The null hy-pothesis was rejected when the P valuewas less than .05.

RESULTS

Comparison between Gadolinium-and Mesoporphyrin-enhancedMR Imaging Regions

The mesoporphyrin-enhanced regionwas substantially smaller in animalstreated with nicorandil (treated) com-pared with animals not treated (controlanimals) (Fig 1). The sizes of mesopor-phyrin-enhanced regions were 25.6% 67.9 of the LV myocardial areas in controlsand 7.9% 6 6.8 in nicorandil-treated an-imals (P , .001), which is consistent witha 69% reduction in infarction size (Fig 2).Gadolinium-enhanced regions (infarcted

myocardium plus periinfarction zone)also were reduced, 36.4% 6 8.4 in controlanimals to 24.7% 6 10.3 in treated ani-mals (232%, P , .001). However, the sizeof the periinfarction zone (gadolinium-enhanced region minus mesoporphyrin-enhanced region) was larger in treated(16.1% 6 11.4) compared with control(10.8% 6 10.0) animals (P , .05) (Fig 3).The decrease in infarction size and in-crease in periinfarction zone size intreated animals cannot be attributed to adifference in the size of area at risk be-tween the groups, because the size of areaat risk was not different (control animals,48.7% 6 7.8; treated animals, 50.7% 67.3; P . .05) (Fig 2).

Effect of Nicorandil on the Sizeof Infarction

TTC staining confirmed the reductionin infarction size demonstrated on meso-porphyrin-enhanced images in nicoran-dil-treated rats. Linear regression analysisrevealed a strong correlation (Y 5 1.65 10.95 3 X, r 5 0.96, P , .001; standarderror of the estimate 5 3.11) betweenTTC-depicted infarction size and meso-porphyrin-enhanced regions. Bland-Alt-man analysis revealed no systematicmeasurement bias over a wide range ofsizes of infarction. The mean differencein measurements was 0.77% 6 3.1 of LVmyocardial area and the limits of agree-

ment ranged from 25.5% to 17.1%.Conversely, only a fair correlation wasfound between the size of gadolinium-enhanced regions and TTC staining (Y 522.3 1 0.52 3 X, r 5 0.53, P , .001;standard error of the estimate 5 9.39).Bland-Altman analysis revealed an over-estimation of infarction size of 114.0% 610.8 of LV myocardial area by using gad-olinium-enhanced MR imaging. Figure 4shows the area at risk, or region of infarc-tion demonstrated with TCC staining,and mesoporphyrin-enhanced region onthe corresponding midventricular sections.

Effect of Nicorandil on Regional T1Relaxation Time

The difference in regional T1 valuesbetween control and treated animals wasevident on mesoporphyrin-enhanced in-version-recovery echo-planar images.The T1 value of ischemically injuredmyocardium was significantly shorter incontrol (660 msec 6 86) than in treated(960 msec 6 53) animals (P , .001). Theremote normal myocardium revealed nodifference in T1 relaxation time betweencontrol (910 msec 6 14) and treated (900msec 6 26) animals.

Effects of Nicorandil on LVDilatation and Function

MR imaging revealed marked diastolicdilatation of the LV cavity area in control

Figure 1. Multisection mesoporphyrin-enhanced short-axis view MR images acquired in control(top row) and treated (bottom row) rats at end diastole. The size of the mesoporphyrin-enhancedregion (arrowheads) and LV dilatation is substantially reduced in treated compared with controlanimal. The high blood signal intensity in control animal is caused by slow blood flow in the LVdue to reduced global function.

678 z Radiology z December 2001 Lund et al

animals (Table). A positive correlationwas found between end-diastolic LV cav-ity area and infarction size measured on

identical mesoporphyrin-enhanced MRimages (Fig 5). MR images acquired at 24hours after acute myocardial infarction

showed better global and regional con-tractile function in nicorandil-treated an-imals compared with control animals.The systolic reduction in LV cavity areawas 57.5% 6 17.3 in nicorandil-treatedanimals compared with 38.0% 6 16.0 incontrol animals (P , .05). Regional wallthickening was depressed in control ani-mals, with no wall thickening in the meso-porphyrin-enhanced region, and severelyreduced in the rim of the gadolinium–en-hanced region (Fig 6). Nicorandil im-proved wall thickening of ischemically in-jured myocardium in treated animals.There was no significant difference in re-gional wall thickening, in the remote pos-terior and septal walls (Fig 7), between thetwo groups.

DISCUSSION

The major findings of the current studywere the following: (a) In the acute stage(24 hours) of myocardial infarction, thenecrosis-specific MR imaging contrastmaterial, mesoporphyrin demonstrated areduction in the size of infarction. Thedecrease in infarction size demonstratedwith mesoporphyrin was subsequentlyconfirmed by using histochemical TTCstaining. (b) Nicorandil reduced the sizeof the entire gadolinium-enhanced re-gion. The salvage of ischemically injuredmyocardium is reflected by the expan-sion of the periinfarction zone defined bythe difference in mesoporphyrin and ga-dolinium-enhanced regions. (c) Func-

Figure 2. Bar graph depicts the size of area at risk (black bars), truesize of infarction (white bars) by using TTC staining, and mesopor-phyrin-enhanced region (bars with oblique lines). The infarction sizewas reduced in nicorandil-treated rats as demonstrated with meso-porphyrin-enhanced MR imaging and confirmed with TTC staining.The area at risk was not significantly different between the groups,which suggests that the reduction in the size of infarction is notattributed to a difference in the size of area at risk. The P value was lessthan .001 (p) for control versus treated animals.

Figure 3. Bar graph depicts the gadolinium-enhanced region (blackbars) (infarcted myocardium and periinfarction zone) and the calcu-lated periinfarction zone (white bars) (gadolinium-enhanced regionminus mesoporphyrin-enhanced region). The gadolinium-enhancedregion was smaller in nicorandil-treated animals. However, nicoran-dil treatment increased the size of the viable periinfarction zone. TheP value was less than .001 (p) and less than .05 (†) for control versustreated animals.

Figure 4. Photographs of tissue specimens and short-axis view MR images obtained at midven-tricular level in control and treated animals. In both sets of animals, the location and size of themesoporphyrin-enhanced region (right, bright regions defined with arrowheads) correspondedwell to the true infarction size demonstrated with TTC staining (center, gray regions defined witharrowheads). The size of infarction was substantially smaller in the nicorandil-treated comparedwith control animal. The area at risk (left, unstained light regions defined with arrowheads) wascomparable in size in both sets of animals, which indicates the reduction in infarction size wasnot attributed to a difference in area at risk.

Volume 221 z Number 3 Assessment of Nicorandil Therapy in Ischemic Myocardial Injury z 679

tional MR imaging revealed reduction inLV cavity area and preservation of re-gional wall thickening in nicorandil-treated animals. (d) Shortening in re-gional T1 relaxation time indicatedaccumulation of mesoporphyrin in isch-emically injured regions of control, butnot in nicorandil-treated animals.

The use of necrosis-specific radiotrac-ers, such as technetium 99m pyrophos-phate and labeled monoclonal antibodies,for the identification and determination ofthe size of acute myocardial infarction iswell established in nuclear medicine. How-ever, the diagnostic accuracy of these nu-clear tracers is limited by difficulties in thedetection of small or nonreperfused infarc-tions, interference with overlying rib up-take, residual blood activity, and limitedspatial resolution of scintigraphic tech-nique (18). Measurement of the size ofinfarction by using mesoporphyrin-en-hanced MR imaging has overcome some ofthe limitations of scintigraphic techniquesand nonspecific extracellular MR imagingcontrast material.

Mesoporphyrin used with new or con-ventional MR imaging sequences pro-vides for accurate sizing of reperfusedand occlusive infarctions (10–12,14). In-tracellular binding, or precipitation ofthe contrast material, demonstrates thespecificity of mesoporphyrin for depict-ing necrotic tissues and cellular debris,such as denatured protein, nucleotide, orcalcium precipitates of necrotic cells (19).Conversely, gadopentetate dimegluminehas no binding site and distributes pas-sively into edematous and infarcted myo-cardium and enhances both regions(12,14,20). Its influence on signal inten-sity is attributed to a greater accessibilityto free water in edematous and infarctedmyocardium. Thus, tandem administra-tion of mesoporphyrin and a high doseof gadopentetate dimeglumine was usedto map the edematous periinfarctionzone, as described in a previous report(12,14). Overestimation of the size ofacute myocardial infarction (24 hours)was observed after administration of ex-tracellular MR imaging contrast materialas demonstrated in the current study andin previous reports (12–14). This overes-timation cannot be attributed to the vol-ume averaging effect of new or conven-tional MR imaging techniques becausemesoporphyrin- and gadolinium-en-hanced MR imaging was performed inone session without removing the ani-mal from the magnet, by using a singlepulse sequence and identical section po-sition and thickness. Furthermore, thetheory that a gadolinium-enhanced re-

gion encompasses viable and nonviableportions, especially, in reperfused infarc-tions older than a couple of days, is stillcontroversial.

Contrast-enhanced MR Imaging inDemonstrating the Reduction inthe Size of Infarction

The strategy used in the current studywas to administer nicorandil in the mid-dle of the period of coronary artery oc-clusion and in the 1st hour of reperfu-sion. With the use of this approach,nicorandil statistically reduced the sizeof infarction, as shown at contrast-enhanced MR imaging and histomor-phometry. Furthermore, inversion-recov-

ery echo-planar MR imaging revealed amuch greater reduction in T1 in ischemi-cally injured myocardium in control an-imals compared with that in nicorandil-treated rats. These findings indicate thatnicorandil preserved large portions of thearea at risk and subsequently reduced theaccumulation of mesoporphyrin in ne-crotic myocardium.

Contrast-enhanced MR Imaging inDetermining the Reduction in theSize of the Injured Region

Our study was performed to assess theperiinfarction zone in the early stage ofreperfused infarction at 24 hours aftercoronary occlusion by using contrast-en-

End-diastolic LV Cavity Area Measured at MR Imaging in Controland Nicorandil-treated Animals

Location Control (mm2) Nicorandil (mm2) P Value

Apex 22.1 6 11.7 11.9 6 3.7 ,.05Center 28.5 6 7.9 15.2 6 4.3 ,.001Base 29.4 6 7.5 17.1 6 4.7 ,.001

Note.—Data are the mean 6 SD.

Figure 5. Graph shows correlation between infarction size (meso-porphyrin-enhanced region) and end-diastolic LV cavity area de-picted on identical MR images (60 images from 20 rats). A directcorrelation was made between infarction size and extent of LV dila-tation from each single image (Y 5 13.3 10.44 3 X, r 5 0.5, P , .001,standard error of the estimate 5 8.48). Nicorandil-treated animals(white diamonds) showed smaller infarction size and diminished LVdilatation in almost all images compared with control animals (blackdiamonds), underlining the beneficial effect of infarction size reduc-tion with early LV remodeling.

680 z Radiology z December 2001 Lund et al

hanced and functional MR imaging incontrol and treated animals. To obtain asizable periinfarction zone, a relativelyshort period of coronary artery occlusionwas used. A relatively short period of cor-onary artery occlusion was also used tomimic the clinical setting of nontrans-

mural infarction. This occlusion timeprovided a situation in which a substan-tial amount of edematous but salvageablemyocardium existed in the area at risk(14,20). The existence of the periinfarc-tion zone has been very recently docu-mented in rats (by using MR imaging)

(12,14), cats (by using MR imaging andelectron microscopy) (21), pigs (by usingMR imaging and electron microscopy)(22), and humans (by using thallium-201single photon emission computed to-mography and combined positron emis-sion tomography with tagged MR imag-ing) (23,24).

The effect of therapy on the periinfarc-tion zone by using noninvasive imagingmethods, as found in this study, to ourknowledge, has not been previously de-scribed. There are several possible reasonsfor the increase in the periinfarction zonedepicted in nicorandil-treated animals.First, nicorandil increased the size of theperiinfarction zone at the cost of the sizeof infarction, which was substantially de-creased in treated animals. Second, theaugmented periinfarction zone may bedue to an increased tolerance to ischemiaduring coronary occlusion. This explana-tion is supported by the observation ofSchultz et al (25), who found that nic-orandil has a pharmacologic precondi-tioning effect by opening adenosinetriphosphate–sensitive potassium chan-nels. Similarly, Matsubara et al (4) foundan increased ischemic tolerance in pa-tients treated with nicorandil subjectedto coronary angioplasty. Finally, nicoran-dil may have caused a diminution ofedema in the periinfarction zone duringocclusion and reperfusion in treated ani-mals, which resulted in a smaller gado-linium-enhanced region compared withthat in control animals. Results of previ-ous reports have indicated that edema isinitiated during occlusion (26) and reper-fusion (27) after extravasation of plasmaproteins into the extracellular space. Nic-orandil may impede the accumulation ofplasma proteins in the ischemically in-jured region by preserving microvascularfunction. Authors of a recent clinicalstudy (3), using echocardiography, sug-gested a microvascular protection by nic-orandil in reperfused acute myocardialinfarction.

Functional MR Imaging inDetermining the Improvementof LV Function

The adaptive cardiac responses to acutemyocardial infarction are well described inthe literature (28–31). The main changesinclude infarct expansion, regional dilata-tion, and thinning of the infarct zone,which occurs as early as 1 day after theischemic event (31). Chronic infarctionmodels have shown that the size of infarc-tion is the main predictor of late LV remod-eling (29). Results of the current study

Figure 6. Functional MR images in short-axis view show the mesoporphyrin- and gadolinium-enhanced regions during diastole and systole in control and nicorandil-treated rats. Note thereduced LV dilatation and improved LV wall thickening in nicorandil-treated rat compared withthe control rat. Furthermore, the gadolinium-enhanced region (arrowheads) is substantiallylarger than the mesoporphyrin-enhanced region (arrowheads) in the nicorandil-treated animal.

Figure 7. Bar graph shows regional LV wall thickening in controland nicorandil-treated animals. There was no wall thickening in themesoporphyrin-enhanced anterolateral wall (white bars) in controlanimals. Nicorandil significantly (P , .05) improved the function inthe anterolateral wall in the treated animals. Additionally, better wallthickening was also observed in the gadolinium-enhanced rim (barswith oblique lines) of the nicorandil-treated animals (P , .05). Nodifference in wall thickening was observed in the remote nonin-farcted posterior (gray bars) and septal (black bars) regions of the LV.

Volume 221 z Number 3 Assessment of Nicorandil Therapy in Ischemic Myocardial Injury z 681

demonstrate that nicorandil effectivelyprevents early LV dilatation after myocar-dial infarction and improved regional LVfunction.

In conclusion, contrast-enhanced MRimaging has the potential to demonstratereduction in size of ischemically injuredmyocardium, whereas functional MR im-aging demonstrates the recovery of LVfunction 24 hours after therapy. The ben-eficial effect of newly developed cardio-protective agents can be noninvasivelymonitored by using contrast-enhancedMR imaging.

Furthermore, beside prompt restora-tion of coronary blood flow with throm-bolysis or angioplasty, early therapy withintravenously administered nicorandilmay be an ideal adjunctive therapy toreduce the size of infarction and to im-prove LV function. The early reduction inthe size of infarction and LV dilatationdepicted may attenuate late LV remodel-ing. Further MR imaging studies are war-ranted to investigate the beneficial effectof early nicorandil treatment to limit lateLV remodeling and to improve the prog-nosis of patients.

Practical application: There are anumber of potential applications pertain-ing to these results. If necrosis-specificMR imaging contrast material were ap-proved for clinical use, it would providethe following advantages over nonspe-cific extracellular MR imaging contrastmaterial: (a) MR images could be ob-tained in a wide time window after ad-ministration of this type of agent, allow-ing flexible scheduling for MR imaging.(b) Persistent enhancement providedwith mesoporphyrin enables a correla-tion between the size of infarction andthe extent of wall motion abnormalitymeasured on identical MR image sec-tions. Chronologic measurements of re-gional wall motion abnormality and sizeof infarction will provide further insightinto the time course of functional recov-ery in ischemically injured but viablemyocardium after therapy. (c) Dual con-trast-enhanced imaging enables charac-terization of the periinfarction zone. Theissue of toxicity of porphyrins has to beresolved before this new contrast mate-rial can be applied in humans. Phase Iclinical trials are expected in the nearfuture.

Acknowledgments: Mitsuaki Chujo, PhD(Chugai Pharmaceutical, Tokyo, Japan) and

Hanns-Joachim Weinmann, PhD (Schering AG,Berlin, Germany) kindly provided nicorandiland mesoporphyrin, respectively, as gifts.

References1. Taira N. Nicorandil as a hybrid between ni-

trates and potassium channel activators.Am J Cardiol 1989; 63:18J–24J.

2. Yoneyama F, Satoh K, Taira N. Nicorandilincreases coronary blood flow predominantlyby K-channel opening mechanism. CardiovascDrugs Ther 1990; 4:1119–1126.

3. Ito H, Taniyama Y, Iwakura K, et al. Intrave-nous nicorandil can preserve microvascularintegrity and myocardial viability in patientswith reperfused anterior wall myocardial in-farction. J Am Coll Cardiol 1999; 33:654–660.

4. Matsubara T, Minatoguchi S, Matsuo H, et al.Three-minute, but not one minute, ischemiaand nicorandil have a preconditioning effectin patients with coronary artery disease. J AmColl Cardiol 2000; 35:345–351.

5. Auchampach JA, Gross GJ. Anti-ischaemicactions of potassium channel openers in ex-perimental myocardial ischemia/reperfusioninjury in dogs. Eur Heart J 1993; 14(supplB):10–15.

6. Lamping KA, Christensen CW, Pelc LR, War-ltier DC, Gross GJ. Effects of nicorandil andnifedipine on protection of ischemic myo-cardium. J Cardiovasc Pharmacol 1984;6:536–542.

7. Mizumura T, Nithipatikom K, Gross GJ. Ef-fects of nicorandil and glyceryl trinitrate oninfarct size, adenosine release, and neutro-phil infiltration in the dog. Cardiovasc Res1995; 29:482–489.

8. Imagawa J, Baxer GF, Yellon DM. Myocardialinfarction afforded by nicorandil and isch-emic preconditioning in a rabbit infarctmodel in vivo. J Cardiovasc Pharmacol 1998;31:74–79.

9. Sato T, Sasaki N, O’Rourke B, Marban E. Nic-orandil, a potent cardioprotective agent, actsby opening mitochondrial ATP-dependentpotassium channels. J Am Coll Cardiol 2000;35:514–518.

10. Marchal G, Ni Y, Herijgers P, Flameng W, etal. Paramagnetic metalloporphyrins: infarctavid contrast agents for diagnosis of acutemyocardial infarction by MRI. Eur Radiol1996; 6:2–8.

11. Pislaru SV, Ni Y, Pislaru C, et al. Noninvasivemeasurements of infarct size after thrombol-ysis with a necrosis-avid MRI contrast agent.Circulation 1999; 99:690–696.

12. Saeed M, Bremerich J, Wendland WF, Wein-mann HJ, Higgins CB. Reperfused myocar-dial infarction as seen with use of necrosis-specific versus standard extracellular MRcontrast media in rats. Radiology 1999; 213:247–257.

13. Schaefer S, Malloy CR, Katz J, et al. Gadolin-ium-DTPA-enhanced nuclear magnetic reso-nance imaging of reperfused myocardium:identification of the myocardial bed at risk.J Am Coll Cardiol 1988; 12:1064–1072.

14. Saeed M, Lund G, Wendland MF, BremerichJ, Weinmann HJ, Higgins CB. Magneticresonance characterization of the peri-in-farction zone of reperfused myocardial in-farction with necrosis-specific and extracel-lular nonspecific contrast media. Circulation2001; 103:871–876.

15. Weinmann HJ, Brasch RC, Press WR, WesbeyGE. Characteristics of gadolinium-DTPA

complex: a potential NMR contrast agent.AJR Am J Roentgenol 1984; 142:619–624.

16. Hilger C, Maier F, Ebert W, et al. Germanpatent DE 42 32 925 A1. Berlin, Germany;1992.

17. Wendland MF, Saeed M, Lauerma K, et al.Alterations in T1 of normal and reperfusedinfarcted myocardium after Gd-BOPTA ver-sus GD-DTPA on inversion recovery EPI.Magn Reson Med 1997; 37:448–456.

18. Marcus ML, Braunwald E. Marcus cardiac im-aging: a companion to Braunwald’s heartdisease. 2nd ed. Philadelphia, Pa: Saunders,1996.

19. Ni Y, Petre C, Miao Y, et al. Magnetic reso-nance imaging-histomorphologic correlationstudies on paramagnetic metalloporphyrins inrat models of necrosis. Invest Radiol 1997; 32:770–779.

20. Arheden H, Saeed M, Higgins CB, et al.Reperfused rat myocardium subjected to var-ious durations of ischemia: estimation of thedistribution volume of contrast materialwith echo-planar MR imaging. Radiology2000; 215:520–528.

21. Choi SII, Choi SH, Kim ST, et al. Irreversibledamaged myocardium at MR imaging with anecrosis tissue-specific contrast agent in a catmodel. Radiology 2000; 215:863–868.

22. Choi SII, Jiang CZ, Lim KH, et al. Applicationof breath-hold T2-weighted, first pass perfu-sion and gadolinium-enhanced T1-weightedMR imaging for assessment of myocardialviability in a pig model. J Magn Reson Imag-ing 2000; 11:476–480.

23. Bogaert J, Maes A, Van de Werf F, et al. Func-tional recovery of subepicardial myocardialtissue in transmural myocardial infarctionafter successful reperfusion: an importantcontribution to the improvement of regionaland global left ventricular function. Circula-tion 1999; 99:36–43.

24. Elhendy A, Cornel JH, Roelandt JR, et al.Dobutamine thallium-201 SPECT imagingfor assessment of peri-infarction and remotemyocardial ischemia. J Nucl Med 1996; 37:1951–1956.

25. Schultz JE, Yao Z, Cavero I, Gross GJ. Glib-enclamide-induced blockade of ischemicpreconditioning is time dependent in intactrat heart. Am J Physiol 1997; 272:2607–2615.

26. al-Haboubi HA, Tomlinson DR, Ward BJ. Theinfluence of hypoxia on transvascular leak-age in the isolated rat heart: quantitative andultrastructural studies. J Physiol 1995; 482:157–166.

27. Inauen W, Payne DK, Kvietys PR, GrangerDN. Hypoxia/reoxygenation increases thepermeability of endothelial cell monolayers:role of oxygen radicals. Free Radic Biol Med1990; 9:219–223.

28. Mill JG, Stefanon I, Leite CM, Vassallo DV.Changes in performance of the survivingmyocardium after left ventricular infarctionin rats. Cardiovasc Res 1990; 24:748–753.

29. Pfeffer JM, Pfeffer MA, Fletcher PJ, Braun-wald E. Progressive ventricular remodeling inrat with myocardial infarction. Am J Physiol1991; 260:H1406–H1414.

30. McKay RG, Pfeffer MA, Pasternak RC, et al.Left ventricular remodeling after myocardialinfarction: a corollary to infarct expansion.Circulation 1986; 74:693–702.

31. Weisman HF, Bush DE, Mannisi JA, BulkleyBH. Global cardiac remodeling after acutemyocardial infarction: a study in the ratmodel. J Am Coll Cardiol 1985; 5:1355–1362.

682 z Radiology z December 2001 Lund et al