Quantification of Myocardial Area at Risk With T2-Weighted CMR

Transcription

1 JACC: CARDIOVASCULAR IMAGING VOL. 2, NO. 7, 9 9 BY THE AMERICAN COLLEGE OF CARDIOLOGY FOUNDATION ISSN X/9/$36. PUBLISHED BY ELSEVIER INC. DOI:1.116/j.jcmg Quantification of Myocardial Area at Risk With T2-Weighted CMR Comparison With Contrast-Enhanced CMR and Coronary Angiography Jeremy Wright, MBBS,* Tom Adriaenssens, MD, Steven Dymarkowski, MD, PHD,* Walter Desmet, MD, PHD, Jan Bogaert, MD, PHD* Leuven, Belgium; and Brisbane, Australia OBJECTIVES We sought to quantify the myocardium at risk in reperfused acute myocardial infarction (AMI) in man with T2-weighted (T2W) cardiac magnetic resonance (CMR). BACKGROUND The myocardial area at risk (AAR) is defined as the myocardial tissue within the perfusion bed distally to the culprit lesion of the infarct-related coronary artery. T2W CMR is appealing to retrospectively determine the myocardial AAR after reperfused AMI. Data on the utility of this technique in humans are limited. METHODS One hundred eight patients with successfully reperfused ST-segment elevation AMI were studied between 1 and days after percutaneous coronary intervention (PCI). We compared the volume of hyperintense myocardium on T2W CMR with the myocardial AAR determined by contrastenhanced CMR with infarct endocardial surface length (ESL) and AAR estimated by conventional coronary angiography with the BARI (Bypass Angioplasty Revascularization Investigation) risk score. RESULTS The volume of hyperintense myocardium on T2W CMR (mean 32 16%, range 3% to 67%) was consistently larger than the volume of myocardial infarction measured with contrast-enhanced images (mean 17 12%, range % to 55%) (p.1). Myocardial salvage ranged from 4% to 45% of the left ventricular myocardium (mean 14 1%). The AAR determined by T2W CMR compared favorably with the infarct ESL (r.77) with contrast-enhanced CMR, and there was moderate correlation between the BARI angiographic risk score and infarct ESL (r.42). The time between PCI and CMR did not cause a significant difference in the volume of T2W hyperintense myocardium (r.11, p.27) or the calculated volume of salvaged myocardium (r.12, p.23). CONCLUSIONS T2W CMR performed early after successfully reperfused AMI in humans enables retrospective quantification of the myocardial AAR and salvaged myocardium. (J Am Coll Cardiol Img 9;2:825 31) 9 by the American College of Cardiology Foundation From the Departments of *Radiology and Cardiology, Gasthuisberg University Hospital, Leuven, Belgium; and the Greenslopes Private Hospital, Brisbane, Australia. Manuscript received December 3, 8; revised manuscript received February 1, 9, accepted February 17, 9.

2 826 JACC: CARDIOVASCULAR IMAGING, VOL. 2, NO. 7, 9 ABBREVIATIONS AND ACRONYMS AAR area at risk AMI acute myocardial infarction CMR cardiac magnetic resonance ESL endocardial surface length FOV field of view LV left ventricle/ventricular PCI percutaneous coronary intervention SPECT single-photon emission computed tomography T2W T2-weighted TE echo time TIMI Thrombolysis In Myocardial Infarction TR repetition time The myocardial area at risk (AAR) is defined as the myocardial tissue within the perfusion bed that is distal to the culprit lesion of the infarct-related coronary artery. In humans, the portion of the AAR that is irreversibly injured (i.e., infarcted) ranges from % (aborted infarction) to as much as 88% (1). The proportion of the AAR that survives the salvaged myocardium is dependent on multiple factors including time to reperfusion, See page 832 ischemic pre-conditioning, collateral flow, distal embolization, reperfusion injury, and microvascular dysfunction. Because the extent of myocardial salvage determines final infarct size, quantification of myocardial salvage is an appealing method for the assessment of the efficacy of different therapies for acute myocardial infarction (AMI). However, quantification of AAR remains challenging. Fluorescent microspheres are the reference standard for measuring the AAR in animal studies. In humans, the most widely used technique is single-photon emission computed tomography (SPECT), but there are many drawbacks to its widespread use. These include 24-h isotope availability, additional radiation exposure for the patient and the operator, and suboptimal spatial resolution. Cardiac magnetic resonance (CMR) is an attractive technique to noninvasively quantify the AAR. First, infarct endocardial surface length (ESL) measured with contrast-enhanced CMR has recently been described as a measure of AAR (2). Assuming the widely accepted wave front phenomenon of irreversible injury during coronary artery occlusion (3), AAR is estimated by measuring the circumferential extent of endocardial necrosis (as a percentage of total left ventricular [LV] endocardial circumference) on contrast-enhanced magnetic resonance images. The results compared favorably with 2 angiographic risk score estimates of AAR (2). Second, it has been demonstrated that T2-weighted (T2W) CMR, or edema-weighted imaging in animal models and patients, performed after myocardial infarction can retrospectively identify the AAR (4,5). The purpose of this study was to quantify AAR in patients with reperfused AMI with T2W CMR and to compare results with both angiographic risk scores and infarct ESL. Moreover, we evaluated the influence of the time delay between percutaneous coronary intervention (PCI) and T2W CMR-derived AAR measures. METHODS Patient population. One hundred thirty-four patients with ST-segment elevation AMI and successful PCI were prospectively enrolled between May 3 and October 7. Patients were included if they had cumulative ST-segment elevation of 6 mm or more, PCI between 2 and 24 h after symptom onset, and significant LV dysfunction (hypokinesia of at least 3 contiguous myocardial segments). Exclusion criteria included pulmonary edema, cardiogenic shock, prior coronary artery bypass grafting, significant renal dysfunction (glomerular filtration rate 6 ml/ min/1.73 m 2 ), or other major comorbidities. One hundred nineteen consecutive patients underwent CMR during the in-hospital phase. This study comprises 18 of the 119 patients in whom T2W CMR-derived AAR quantification was able to be performed. We obtained informed written consent from all patients, and the study was approved by the ethics review board of the Gasthuisberg University Hospital, Leuven, Belgium. CMR imaging. The CMR studies were performed on a 1.5-T system (Intera, Philips Medical Systems, Best, the Netherlands) with commercially available CMR software, electrocardiographic triggering, dedicated cardiac surface coils, and inspiratory breath holds. After determining the cardiac axes with localizers, vertical and horizontal long-axis and short-axis cine CMR was performed with a balanced steady-state free procession sequence. Imaging parameters were as follows: repetition time (TR): 3.6 ms; echo time (TE): 1.8 ms; turbo field echo: 14; flip angle: 6 ; slice thickness: 8 mm; matrix: ; field of view (FOV): 3 mm; number of phases: 3. Next, T2W CMR was performed by encompassing the LV in cardiac short-axis direction with a dark-blood T2W short tau inversion-recovery fast spin echo sequence. Imaging parameters were TR: 2 heart beats; TE: 1 ms; turbo factor: 33; FOV: 35 mm; slice thickness: 8 mm. Finally, a breath-hold, T1- weighted 3-dimensional contrast-enhanced inversionrecovery gradient-echo sequence after intravenous administration of.2 mmol/kg gadolinium-diethylene triamine pentaacetic acid in cardiac short-axis was used to depict the presence, location, and extent of myocardial infarction. Imaging parameters were as

3 JACC: CARDIOVASCULAR IMAGING, VOL. 2, NO. 7, follows: TR: 4.5 ms; TE: 1.3 ms; flip angle: 15 ; contiguous slices, slice thickness: 5 mm; matrix: ; and FOV: 35 mm. Images were obtained 1 to min after contrast administration, and the inversion time was tailored to null signal from normal myocardium. Image analysis. All CMR images were analyzed on an offline workstation (ViewForum, Philips Electronics, Best, the Netherlands), and the results are the consensus of 2 experienced observers blinded to the angiographic data (S.D. and J.B.). To avoid interpretation bias, T2W images and contrastenhanced CMR images were presented separately to the readers in a random way, respecting an interval of 3 days between readings. Moreover, CMR readers were blinded to the clinical and angiographical data. Endocardial and epicardial borders were delineated on the end-diastolic and -systolic short-axis slices for quantification of LV volumes, function, and mass. The AAR was quantified on the T2W images by delineation of myocardium with signal intensity 2 SDs above the mean signal obtained in the remote noninfarcted myocardium. If present, a central core of hypointense signal within the area of increased signal intensity (hemorrhagic infarction) was included in the AAR (6,7). Increased signal intensity from the blood pool adjacent to the endocardium due to slow flow was excluded. The volume of T2W increased signal intensity was normalized for LV mass and expressed as a percentage. Myocardial infarction was quantified by delineation of the areas of hyperenhancement on the contrast-enhanced CMR images and was also normalized for LV mass. If present, the dark subendocardial zone in the center of the hyperenhancement (microvascular obstruction) was included in the infarct volume. The infarct ESL was calculated as the percentage of summed lengths of endocardial hyperenhancement/total LV endocardial surface length (again, the hypoenhanced surface of microvascular obstruction was included in the infarct). Coronary angiography. Cardiac catheterization with selective injection of right and left coronary arteries was performed before PCI. The angiograms were retrospectively assessed independently by 2 angiographers (not present at the PCI) blinded to the CMR data (J.W. and T.A.). The anatomic area at risk distally to the culprit lesion in the infarct-related artery was estimated with the adapted angiographic BARI (Bypass Angioplasty Revascularization Investigation) scoring system (8). Right ventricular marginal branches, atrial branches, and septal branches of the posterior descending coronary artery were excluded from the analysis. All remaining terminating coronary arteries were qualitatively graded from to 3 on the basis of relative length. The BARI risk score was defined as the percentage of the sum scores of the vessels distal to the culprit lesion divided by the total LV score. Noninfarctrelated artery stenoses did not contribute to the AAR (Fig. 1). Consensus between angiographers was used for subsequent analysis. When the Thrombolysis In Myocardial Infarction (TIMI) flow grade was before PCI, the culprit vessel was verified by the 12-lead electrocardiogram and left ventriculography in 2 orthogonal planes. Statistical analysis. Continuous data are reported as mean SD, and categorical data are reported as frequencies and percentages. Correlations between continuous variables were assessed with Pearson s correlation coefficient r. The technique proposed by Bland and Altman was used to compare T2W AAR and BARI risk score with infarct ESL. Discrepancies between T2W AAR and infarct ESL correlated to clinical factors with Pearson s r (continuous variables) and 1-way analysis of variance for categorical variables. All tests were 2-tailed, and statistical significance was accepted at p.5. Data were analyzed with SPSS version 15. (SPSS Inc., Chicago, Illinois). RESULTS The demographic data of the 18 patients are presented in Table 1. The culprit vessel was the left anterior descending coronary artery in 48 (44%) patients, the left circumflex coronary artery in 12 (11%), and the right coronary artery in 48 (44%) patients. All patients had a region of increased signal intensity detected on T2W CMR (Fig. 2). Three patients had no increased signal on contrastenhanced CMR (aborted infarcts), and 4 patients had evidence of prior infarction (in different vascular territories) on contrast-enhanced CMR. Microvascular obstruction was identified in 69 patients, range to 8 g (mean g). Time from symptom onset to PCI ranged from 1 to 7 min (mean min). The CMR was usually performed within 1 week of PCI, range 22 to 496 h (mean h). Comparison of infarct size and (T2W AAR). Infarct size ranged from % to 55% of the LV myocardium (mean 17 12%). The T2W AAR ranged from 3%

4 828 JACC: CARDIOVASCULAR IMAGING, VOL. 2, NO. 7, 9 Table 1. Patient Demographic Data Sex, male:female 91:17 Age, yrs Hypertensive subjects (%) 35 (32) Current smokers (%) 62 (57) Hyperlipidemia (%) 73 (68) Diabetes mellitus (%) 11 (1) Family history of coronary artery disease (%) 53 (49) Previous myocardial infarction 4 Pre-hospital thrombolysis (%) 6 (6) TIMI flow grade before PCI (%) 61 (57) I 9 (8) II (18) III 18 (17) Multivessel disease (%) 58 (53) Peak CK (U/l) 2,57 1,843 Peak CKM ( g/l) LVEDV (ml) LVESV (ml) LV ejection fraction (%) 49 8 Data are n (%) SD or mean unless otherwise stated. CK creatine kinase; CKM creatine kinase mass; LV left ventricular; LVEDV left ventricular end-diastolic volume; LVESV left ventricular end-systolic volume; PCI percutaneous coronary intervention; TIMI Thrombolysis In Myocardial Infarction. Figure 1. Hemorrhagic Myocardial Infarction in a 69-Year-Old Man Coronary angiography before (A) and after (B) percutaneous coronary intervention, T2-weighted (T2W) cardiac magnetic resonance (CMR) (C). Left anterior oblique cranial view shows a proximal left anterior descending coronary artery (LAD) occlusion with thrombotic appearance (A). Presence of 6% stenosis on mid left circumflex coronary artery and % stenosis on the fourth inferolateral branch. Successful reperfusion of LAD (B). The T2W CMR shows area of increased signal intensity in anteroseptal left ventricular wall with hypointense core consistent with post-reperfusion intramyocardial hemorrhage (C). The BARI (Bypass Angioplasty Revascularization Investigation) risk score is 43.5%, T2W area at risk 48.7%. to 67% (mean 32 16%). The T2W AAR was significantly larger than infarct size (p.1) (Fig. 3). The single exception was a patient with a modest-sized infarct (11% of LV myocardium) who had CMR 166 h after PCI. There was no correlation between the T2W AAR and the delay between PCI and CMR (r.11, p.27). Myocardial salvage ranged from 4% to 45% of the LV myocardium (mean 14 1%). There was no correlation between myocardial salvage and the time from onset of symptoms to PCI (r.11, p.29) or between myocardial salvage and the delay between PCI and CMR (r.12, p.23). Comparison of different measures of area at risk. The BARI risk score ranged from 14% to 48% of the LV myocardium (mean %). There was moderate correlation between the BARI risk score and infarct ESL (r.42, p.1). There was strong correlation between T2W AAR and infarct ESL (r.77, p.1) (Fig. 4). The correlation between T2W and infarct ESL was significantly stronger than the correlation between BARI and infarct ESL (r.77 vs..42, Z 3.99, p.3). Bland-Altman analysis demonstrated that the agreement between T2W AAR and infarct ESL (mean difference %) was similar to the

5 JACC: CARDIOVASCULAR IMAGING, VOL. 2, NO. 7, Figure 2. T2-Weighted and Contrast-Enhanced CMR Images The T2W CMR (A) and contrast-enhanced CMR (B) in a 77-year-old male patient with an extensive anteroseptal acute myocardial infarction. On short-axis T2W CMR the area at risk is visible as a hyperintense (i.e., bright gray) area and is slightly larger than the infarct area as visible on contrast-enhanced CMR (i.e., 29.7% vs. 26.2%, respectively). The BARI risk score is 21.7%, and the infarct endocardial surface length is 24%. Abbreviations as in Figure 1. agreement between BARI and infarct ESL (mean difference %) (Fig. 5). Discrepancies between T2W AAR and infarct ESL. There was no significant correlation of the differences between T2W AAR and infarct ESL with time to reperfusion, time between PCI and CMR, LV ejection fraction, or infarct size. The mean difference between T2W AAR and infarct ESL did not seem to be predicted by TIMI flow grade before PCI, the culprit vessel, or single- versus multivessel disease (Table 2). problems. We found that it was possible to achieve diagnostic image quality in 18 of 119 patients. Friedrich et al. (12) recently reported the feasibility of quantifying salvaged myocardium at risk in reperfused AMI. Our study confirms their findings and strengthens the validity of the technique by comparison with 2 other measures of myocardial area at risk (i.e., infarct ESL and BARI angiographic risk score). Infarct ESL is a recently described novel technique for estimating the area at risk (2) and is measured on contrast-enhanced DISCUSSION This study demonstrates that it is possible to quantify the volume of myocardium that has increased signal intensity on T2W CMR after AMI in humans. Animal studies have demonstrated that this portion of the myocardium correlates closely with the AAR as measured by fluorescent microspheres (4) and is thought to encompass reversibly and irreversibly injured myocardium. Contrastenhanced CMR with late (or delayed) imaging for quantification of irreversible myocardial injury has been extensively validated in acute and chronic settings (9 11). Thus, comparing the volume of delayed enhancement with the volume of T2W AAR it is possible to determine the proportion of myocardium that has been salvaged. The T2W CMR has long been considered technically challenging, due to the long echo times required and reduced spatial resolution and signalto-noise ratios. Current pulse sequences with dedicated cardiac coils have largely overcome these T2W AAR (% of LV Myocardium) 6 6 Infarct Size (% of LV Myocardium) Figure 3. Myocardium at Risk and Infarct Size The percentage of myocardium at risk during acute myocardial infarction was consistently greater than the irreversibly injured myocardium. The volume of myocardium at risk was measured with T2W CMR, and irreversibly injured myocardium was measured with contrast-enhanced CMR. The dashed line indicates the line of identity. AAR area at risk; LV left ventricular; other abbreviations as in Figure 1.

6 83 JACC: CARDIOVASCULAR IMAGING, VOL. 2, NO. 7, 9 T2W AAR (% LV Myocardium) 6 y = x r =.77 6 Infarct ESL (% Total Endocardial Surface Length) Figure 4. T2W AAR and Infarct ESL The AAR measured with T2W CMR correlates strongly with the infarct endocardial surface length (ESL) (r.77). The solid line is the regression line, whereas the dotted line is the line of identity. Abbreviations as in Figures 1 and 3. A T2W AAR Minus Infarct ESL (% LV Myocardium) B BARI Risk Score Minus Infarct ESL (% LV Myocardium) Mean T2W AAR and Infarct ESL (% LV Myocardium) Mean BARI Risk Score and Infarct ESL (% LV Myocardium) Figure 5. Bland-Altman Plots of T2W AAR With Infarct ESL and BARI Risk Score With Infarct ESL Both T2W AAR (7.8 1.%) (A) and BARI risk score ( %) (B) show a positive bias compared with infarct ESL. The solid lines are the mean difference, whereas the dashed lines represent 2 SD from the mean. Abbreviations as in Figures 1 and 3. CMR images. In our study, T2W AAR and infarct ESL showed good correlation across the entire range of infarct sizes (r.77). The Bland-Altman plot demonstrated a bias of 8% toward T2W AAR with SD 1%. This bias is in keeping with the observations of Lee et al. (1) of a small lateral border of at-risk but viable myocardium at the lateral edges of the infarcted myocardium. The SD we observed is consistent with the findings of Aletras et al. (4) in animals their Bland-Altman analysis of T2W AAR versus fluorescent microspheres demonstrated a bias of.4% with SD of 9%. We observed no relationship between myocardial salvage and time to reperfusion (r.11, p.29). Friedrich et al. (12) found myocardial salvage had a weak (although statistically significant) negative correlation with time between onset of symptoms and reperfusion (r.37). Within the limitations of the small study size and mean time to reperfusion of 4 h, these observations are consistent with factors other than time to reperfusion having a significant influence on the volume of myocardial salvage. The angiographic BARI risk score has recently been validated as a measure of myocardium at risk, by comparison with infarct ESL (2). We found only moderate correlation between BARI risk score and infarct ESL, in contrast to the results reported by Ortiz-Perez et al. (2). This might be due in part to different study populations, because Ortiz-Perez et al. (2) only studied patients with TIMI flow grade before PCI. Additionally, the BARI risk score does not take into account stenoses in noninfarctrelated vessels or collateral vessel flow. Moreover, it should be emphasized that inferior infarcts not infrequently extend to the right ventricular inferior wall (13). Because T2W AAR and infarct ESL measurements were limited to the LV, this might at least partially explain the poor agreement with BARI AAR estimates. In the absence of a reference standard (e.g., microspheres), these issues are difficult to resolve. The optimal timing for measurement of T2W AAR and myocardial salvage is unknown. On the one hand, reperfusion injury (14) is thought to increase the volume of irreversibly injured myocardium within the first 48 h after restoration of epicardial coronary artery flow; thereafter, infarct shrinkage will reduce the volume of delayed enhancement over the next 6 to 8 weeks (9). On the other hand, edema will resolve over time. In our study CMR was performed between 1 and days after PCI (mean and median 4 days). Despite this variation we observed the T2W AAR to be consis-

7 JACC: CARDIOVASCULAR IMAGING, VOL. 2, NO. 7, tently larger than infarct size measured with contrast-enhanced CMR. The only patient with T2W AAR less than infarct size was an 82-year-old man. Though unproven, the likely cause is prior infarction due to the same culprit vessel. The delay between PCI and CMR did not cause systematic differences in the measured T2W AAR or myocardial salvage, although serial studies of individual patients were not performed. CONCLUSIONS It is possible to retrospectively quantify the myocardium at risk and salvaged myocardium after reperfused ST-segment elevation AMI with T2W CMR. This should enable new insights into reperfusion therapy and contribute to clinical decisionmaking. Reprint requests and correspondence: Dr. Jan Bogaert, Department of Radiology, Gasthuisberg University Hospital, Herestraat 49, B-3 Leuven, Belgium. jan.bogaert@uz.kuleuven.ac.be. Table 2. Pearson Correlates and Univariate Predictors of Discrepancies Between T2W AAR and Infarct ESL T2W AAR Infarct ESL r p Value Time from symptom onset to reperfusion.2.85 Time from PCI to CMR.3.75 LV ejection fraction.1.31 Infarct size (% LV myocardium).29.3 Mean SD Coronary anatomy Single-vessel disease Multivessel disease Culprit vessel LAD LCx RCA TIMI flow grade before PCI I II III AAR area at risk; CMR cardiac magnetic resonance; ESL endocardial surface length; LAD left anterior descending coronary artery; LCx left circumflex artery; RCA right coronary artery; T2W T2-weighted; other abbreviations as in Table 1. REFERENCES 1. Lee JT, Ideker RE, Reimer KA. Myocardial infarct size and location in relation to the coronary vascular bed at risk in man. Circulation 1981;64: Ortiz-Perez JT, Meyers SN, Lee DC, et al. Angiographic estimates of myocardium at risk during acute myocardial infarction: validation study using cardiac magnetic resonance imaging. Eur Heart J 7;28: Reimer KA, Jennings RB. The wavefront phenomenon of myocardial ischaemic cell death. II. Transmural progression of necrosis within the framework of ischemic bed size (myocardium at risk) and collateral flow. Lab Invest 1979;: Aletras AH, Tilak GS, Natanzon A, et al. Retrospective determination of the area at risk for reperfused acute myocardial infarction with T2-weighted cardiac magnetic resonance imaging: histopathological and displacement encoding with stimulated echoes (DENSE) functional validations. Circulation 6;113: Dymarkowski S, Ni Y, Miao Y, et al. Value of T-2 weighted magnetic resonance imaging early after myocardial infarction in dogs: comparison with bis-gadolinium-mesoporphyrin enhanced T1-weighted magnetic resonance imaging and functional data from cine magnetic resonance imaging. Invest Radiol 2;37: Lotan CS, Miller SK, Bouchard A, et al. Detection of intramyocardial hemorrhage using high-field proton (1H) nuclear magnetic resonance imaging. Cathet Cardiovasc Diagn 199;: Lotan CS, Bouchard A, Cranney GB, Bishop SP, Pohost GM. Assessment of postreperfusion myocardial hemorrhage using proton NMR imaging at 1.5T. Circulation 1992;86: Alderman EL, Stadius M. The angiographic definitions of the Bypass Angioplasty Revascularization Investigation. Coron Artery Dis 1992; 3: Kim RJ, Fieno DS, Parrish TB, et al. Relationship of MRI delayed contrast enhancement to irreversible injury, infarct age, and contractile function. Circulation 1999;1: Wu E, Judd RM, Vargas JD, Klocke FJ, Bonow RO, Kim R. Visualization of presence, location, and transmural extent of healed Q-wave and non-q wave myocardial infarction. Lancet 1:357: Wagner A, Mahrholdt H, Holly TA, et al. Contrast-enhanced MRI and routine single photon emission computed tomography (SPECT) perfusion imaging for detection of subendocardial myocardial defects: an imaging study. Lancet 3;361: Friedrich MG, Abdel-Aty H, Taylor A, Schulz-Menger J, Messroghli D, Dietz R. The salvaged area at risk in reperfused acute myocardial infarction as visualised by cardiovascular magnetic resonance. J Am Coll Cardiol 8;51: Larose E, Ganz P, Reynolds HG, et al. Right ventricular dysfunction assessed by cardiovascular magnetic resonance imaging predicts poor prognosis late after myocardial infarction. J Am Coll Cardiol 7;49: Yellon DM, Hausenloy DJ. Myocardial reperfusion injury. N Engl J Med 7;357: Key Words: area at risk y cardiac magnetic resonance y myocardial infarction y edema.