| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
(Circulation. 2002;106:957.)
© 2002 American Heart Association, Inc.
Clinical Investigation and Reports |
From the Departments of Adult Cardiology and Radiology, Texas Heart Institute/St Lukes Episcopal Hospital and Baylor College of Medicine, Houston, Tex, and Philips Medical Systems, North America, Bothell, Wash.
Correspondence to Emerson C. Perin, MD, 6624 Fannin, Suite 2220, Houston, TX 77030. E-mail eperin{at}crescentb.net
| Abstract |
|---|
|
|
|---|
Methods and Results We prospectively studied 15 ambulatory patients with stable coronary disease who were candidates for cardiac catheterization. Within 48 hours of EMM, DE-MRI was performed. Using EMM software, we created a bulls eye precisely matched to that generated by DE-MRI. Segment by segment, we compared the MRI results to the corresponding unipolar voltage value for that same segment in the EMM bulls eye. Of 300 total segments, 275 were compared. The segments were divided into normal (n=211), subendocardial scar (n=49), and transmural scar (n=15). We found that subendocardial (6.8±2.9 mV) and transmural (4.6±1.9 mV) scar segments had significantly less unipolar voltage than normal (11.6±4.5 mV) segments (P<0.05 for each comparison). When normal myocardium was compared with myocardium with subendocardial scar, the threshold for differentiating between the two areas was 7.9 mV (sensitivity, 80%; specificity, 80%). Comparison of normal tissue to transmural scar yielded a threshold of 6.9 mV (sensitivity, 93%; specificity, 88%).
Conclusions Our results demonstrate that normal myocardium can be accurately distinguished from myocardium with subendocardial or transmural infarcts on the basis of unipolar voltage values obtained through EMM. This is the first study to validate these results by using cardiac DE-MRI in humans.
Key Words: coronary disease magnetic resonance imaging myocardium
| Introduction |
|---|
|
|
|---|
MRI is rapidly evolving, and its cardiovascular applications range from the assessment of hemodynamic and functional variables (left ventricular mass and volume) to the study of myocardial perfusion and metabolism.12,13 The inherently superior soft-tissue contrast offered by MRI allows effective discrimination between the normal and diseased states of the cardiovascular system. In addition, MRI is noninvasive, imposes no radiation burden to the patient, and often offers better spatial resolution than competing diagnostic modalities, such as nuclear medicine. The combination of these characteristics makes MRI an attractive tool for the assessment of cardiovascular anatomy and physiology.1418 Recently, the introduction of delayed-enhancement MRI technique (DE-MRI) has made it possible to evaluate the presence and extent of irreversibly injured myocardium. Given the submillimeter spatial resolution capabilities of MRI, it is possible to discriminate the extent of myocardial injury as subendocardial or transmural, a capability unmatched by other competing modalities.1921
Several studies have suggested that left ventricular electromechanical mapping (EMM) can also be used to assess myocardial viability on the basis of myocardial voltage.2227 The ability of EMM to distinguish between infarcted and healthy myocardium has been corroborated by SPECT28 and PET imaging.29 Also, the ability of EMM to provide 3D, real-time images makes it possible to guide and deliver direct myocardial therapies using an integrated injection system, such as delivery of vascular endothelial growth factors.3033 However, precise definition of myocardial viability using EMM relies on establishing accurate EMM thresholds. So far, no such thresholds have been established.
In this study, we sought to determine EMM unipolar voltage thresholds for defining myocardial viability and to establish practical cut-off values for differentiating normal tissue from subendocardial and transmural scar tissue. Cardiac DE-MRI was used as the gold standard because of its high spatial and contrast resolution images, which can distinctly differentiate the degree of infarct transmurality.
| Methods |
|---|
|
|
|---|
|
Inclusion Criteria
EMM procedures and cardiac DE-MRI were performed only on patients who were clinically stable. Patients with severe peripheral vascular disease, atrial fibrillation, aortic stenosis, a suspected left ventricular thrombus, or acute coronary syndromes were excluded. Patients were also excluded from the protocol if they had any of the standard contraindications to MRI, such as the presence of a pacemaker, implantable defibrillator, and intracerebral aneurysm clips. St Lukes Episcopal Hospital Institutional Review Board approved the study protocol, and informed consent was obtained from all patients on enrollment.
Imaging Protocols
Electromechanical Mapping
To construct an electromechanical map of the left ventricle, we acquired a series of points at multiple locations on the left ventricular endocardial surface by using the NOGA mapping system (Biosense Webster). The procedure has been described elsewhere.34,35 In this study, we used a filling threshold of 15 mm. After the desired number of points had been acquired, we removed poor data points from the original map by using an automatic software filter at the moderate setting. The following points were removed: those located in the interior of the left ventricular cavity, those that did not fit the standard stability criteria (location stability <4 mm; loop stability <6 mm; cycle length <10%), those obtained during ST-segment elevation, and those unrelated to the left ventricular cavity (eg, in an atrial location).
Mapping Procedure Protocol
After biplane left ventricular angiography was completed, each patient was given 70 U/kg of heparin. The mapping catheter curve (B, D, or F) was selected on the basis of the left ventricular size. Under fluoroscopic guidance, the 7F-mapping catheter was advanced to the descending thoracic aorta, where the catheter tip was fully deflected. The catheter was then advanced around the aortic arch and across the aortic valve into the left ventricle. Then the deflection was relaxed, and the catheter tip was oriented to the left ventricular apex. The initial data point was acquired at the apex. One point each was also acquired at the base of the septum and the lateral wall, thereby completing an initial triangle that defined the borders of the left ventricle. Subsequent points were acquired until all of the endocardial segments had been uniformly sampled; ideally, 3 points were acquired in each of 12 segments, as recommended by the NOGA Mapping Excellence Program.34 (The NOGA Mapping Excellence Program was created by Biosense Webster to assure the quality of the maps. It has been developed and implemented by the authors in conjunction with Biosense-Webster.) Using the above-described parameters, each data point was filtered online immediately after acquisition and during the postprocessing analysis. Figures 1A and 1B show a completed unipolar voltage map with an area of inferobasal scar.
|
MRI
Cardiac MRI was performed during the same hospital admission, within 48 hours after EMM. Personnel performing and interpreting the cardiac MRI were blinded to the results of the EMM procedure.
A 1.5 Tesla commercially available MRI scanner (Philips NT-Intera, Best) with a cardiac software package was used for imaging. A peripheral intravenous line (20-gauge) was inserted, vector-cardiogram (VCG) leads were placed according to standard protocol, and a 5-element, phased-array surface coil was used for enhanced signal to noise ratio. After localization of the heart, ECG-gated cine images were obtained in the following orientations: 2-chamber-long axis, 4-chamber-long axis, and a short axis stack covering the entirety of the left ventricle from the base to apex. Gd-DTPA (0.2 mmol/kg) was administered intravenously, and, after a 15-minute delay, images were obtained in the same orientations using a breath-hold 2D inversion-recovery, T1-weighted gradient-echo sequence (DE-MRI). The DE-MRI sequence had the following acquisition parameters: TR/TE/
=7 ms/3 ms/15 degrees; voxel size=1.2x1.2x10 mm; 32 lines of k-space collected for each R-R interval in end diastole; acquisition time, 16 heart beats; NSA=2. The inversion time was modified iteratively to obtain maximal nulling of remote normal left ventricular myocardium, with an average value of 225 ms.
Data Analysis
EMM Analysis
Each map was displayed in a polar distribution format (bulls eye). After postprocessing was completed, only segments with 2 or more points were considered for the final analysis. For each segment, we obtained a mean value for the unipolar voltage from the points within that segment.
MRI Analysis
Contiguous short-axis delayed-enhancement images covering the entire left ventricle were transferred to a commercially available postprocessing software (MASS, MEDIS software) for additional postprocessing. The endocardial and epicardial boundaries were manually drawn for each slice to isolate the myocardium. Each slice was additionally divided into 9 equi-angular segments, with the segments starting clockwise from the anterior insertion of the right ventricle into the septum. The following bulls eye maps, reflecting the mean signal intensity of each segment, were generated: (1) endocardial bulls eye: the mean signal intensity of the 50% of the myocardium bordering between the blood-pool boundary to the midmyocardium; and (2) epicardial bulls eye: the mean signal intensity of the 50% of the myocardium starting from the midmyocardium to the epicardial boundary. The gray scale window and level were maintained at the same value for the bulls eye maps for each patient, with black reflecting normal myocardium and the bright signal (signal intensity >2 SD above remote normal myocardium) reflecting irreversibly injured myocardium (Figure 1C).
The MRI myocardial segments were categorized into 3 groups: normal, subendocardial scar tissue, and transmural scar tissue. Normal tissue was defined by the absence of bright signal in both endocardial and epicardial bulls eye. Subendocardial scar was defined to be present when bright signal was present in the endocardial bulls eye and absent in the epicardial bulls eye. Bright signal present in both endocardial and epicardial bulls eyes defined the presence of transmural scar.
Comparative Analyses of EMM and Cardiac MRI Data
We analyzed the EMM and cardiac DE-MRI data using the bulls eye representations. Using EMM software, we created a bulls eye precisely matched to that generated by DE-MRI for comparative analysis. Segment by segment, the MRI results as delineated above were compared in a blinded fashion to the corresponding unipolar voltage (uniV) value for that same segment in the EMM bulls eye. Voltage values were tabulated into 3 groups according to the DE-MRI analysis (normal, subendocardial scar, and transmural scar), and the uniV values (mean±SD) for each group were obtained.
Statistical Analyses
Comparison among segments that were normal, subendocardial scar, or transmural scar were performed using ANOVA. Determination of thresholds comparing normal tissue to subendocardial and transmural scar was done using area under the curve of receiver operating curves (ROCs). P<0.05 was considered significant.
| Results |
|---|
|
|
|---|
|
When ROCs were constructed to compare normal myocardium and myocardium with subendocardial scar, the threshold was 7.9 mV (sensitivity, 80%; specificity, 80%) (Figure 2). Similarly, comparison of normal tissue versus transmural scar tissue yielded a threshold of 6.9 mV (sensitivity, 93%; specificity, 88%) (Figure 3). The comparison between voltage values in subendocardial scar segments and transmural scar segments did not reach statistical significance.
|
|
| Discussion |
|---|
|
|
|---|
This study is the first to compare left ventricular EMM with cardiac DE-MRI in humans. Cardiac DE-MRI precisely delimits the presence and extent of myocardial scar tissue.36 The results demonstrated that normal myocardium is accurately distinguished from subendocardial infarcts or transmural infarcts on the basis of unipolar voltage values. This additionally corroborates the finding of Wolf et al37 that EMM can accurately predict the transmural extent of a myocardial infarction. Interestingly, in the canine model used by these researchers, the parameter associated with a higher sensitivity to infarction transmurality was bipolar voltage; however, unipolar voltage, unipolar voltage slew rate, and impedance could also determine the extent of infarcted myocardium. The design of the study by Wolf et al, using thorough mapping technique and a higher number of points per segment (obtainable with an animal model), allowed a more detailed analysis of transmurality of infarcts (<30%, 31% to 60%, and 61% to 100%).
Previous clinical studies22,28 using EMM have characterized normal myocardium as that having uniV values >12 mV. Clinical studies2629,38 of electrical activity and myocardial viability found values for UniV ranging from 6.4 mV to 7.5 mV to define viable myocardium. In the present study, the finding of a threshold of 6.9 mV to detect transmural infarction is associated with higher sensitivity and specificity than has previously been obtained. Nonetheless, these findings are in concordance with previous viability studies and additionally validate UniV as an indicator of myocardial health.
To be classified as a subendocardial infarct, scar tissue had to involve <50% of the myocardial thickness by DE-MRI. Even with this more practical approach (fewer points per segment) and a less accurate assessment of local electrical activity (unipolar voltage) in comparison with the animal study mentioned above, endocardial EMM distinguished between normal and subendocardial scar tissue and between normal and transmural scar tissue in patients. However, EMM could not distinguish between subendocardial and transmural scar. More importantly, when ROC analysis was applied to our data, high specificity and sensitivity were obtained for thresholds differentiating normal from scarred myocardium. Thus, the present data confirm the utility of EMM as a valuable tool for diagnostic evaluation of myocardial viability in the invasive cardiology setting where clinical decisions involving high-risk revascularization procedures are frequently entertained on an immediate basis.
In addition, the present development of new techniques of direct myocardial therapy aiming toward angiogenesis or restoration of contractility either through cell transplantation with stem cells39 and myoblasts40,41 or growth factor injection42 holds the potential for significant advancement in the treatment of ischemic heart disease. Although cell transplantation for cardiac repair and angiogenesis shows great preclinical promise,43,44 its future will profoundly depend on conducting carefully controlled clinical trials with appropriate methods and end points. A critical issue in that regard is the mode of delivery of these therapies. In the cell-transplantation era, the endocardial mapping system might play an important role, because it is a less invasive approach with promising results.45 In light of the above, in the present study, the definition of practical unipolar voltage thresholds for assessment of viability may increase the accuracy of therapy delivery. In theory, the demarcation of appropriate target zones for treatment is one of the keys to the success of the procedure. Moreover, along the boundaries of the predefined target zone, each potential individual injection site could be interrogated before therapy delivery to ascertain delivery into viable myocardium. Conversely, delivery into a presumed region of nonviable tissue might be targeted if EMM interrogation reveals attractive voltage thresholds.
Study Limitations
This study has two primary limitations: a relatively small study population and lack of confirmation of functional recovery in segments classified as viable. Despite the small study population, EMM unipolar voltages could reliably differentiate both subendocardial and transmurally infarcted regions of myocardium based on DE-MRI findings. This finding adds to the body of knowledge validating the use of EMM. As noted previously, DE-MRI is an accurate predictor of functional recovery in myocardium without transmural scar formation and can be performed in a resting study without the need for radiation or potentially nephrotoxic contrast agent. As such, it provides a practical and reliable technique for predicting viability.
Conclusions
When compared with cardiac DE-MRI, electromechanical voltage maps can distinguish normal myocardial tissue from subendocardial and transmural scar tissue with high sensitivity and specificity. Electromechanical voltage maps may be useful in assessing the viability of myocardial tissue when considering the application of direct myocardial therapies and in other scenarios where immediate assessment of underlying viability in the catheterization laboratory is paramount to clinical success.
| Footnotes |
|---|
Received March 29, 2002; revision received June 5, 2002; accepted June 5, 2002.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
M. Penicka, P. Tousek, B. De Bruyne, W. Wijns, O. Lang, J. Madaric, M. Vanderheyden, J. Tintera, M. Maly, P. Widimsky, et al. Myocardial positive pre-ejection velocity accurately detects presence of viable myocardium, predicts recovery of left ventricular function and bears a prognostic value after surgical revascularization Eur. Heart J., June 1, 2007; 28(11): 1366 - 1373. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. P. Finn, R. Saleh, S. Thesen, S. G. Ruehm, M. H. Lee, J. Grinstead, J. S. Child, and G. Laub MR Imaging with Remote Control: Feasibility Study in Cardiovascular Disease Radiology, November 1, 2006; 241(2): 528 - 537. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. J.M. Westenberg, H. J. Lamb, R. J. van der Geest, G. B. Bleeker, E. R. Holman, M. J. Schalij, A. de Roos, E. E. van der Wall, J. H.C. Reiber, and J. J. Bax Assessment of Left Ventricular Dyssynchrony in Patients With Conduction Delay and Idiopathic Dilated Cardiomyopathy: Head-to-Head Comparison Between Tissue Doppler Imaging and Velocity-Encoded Magnetic Resonance Imaging J. Am. Coll. Cardiol., May 16, 2006; 47(10): 2042 - 2048. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Mansour, M. Vanderheyden, B. De Bruyne, B. Vandekerckhove, L. Delrue, I. Van Haute, G. Heyndrickx, S. Carlier, G. Rodriguez-Granillo, W. Wijns, et al. Intracoronary Delivery of Hematopoietic Bone Marrow Stem Cells and Luminal Loss of the Infarct-Related Artery in Patients With Recent Myocardial Infarction J. Am. Coll. Cardiol., April 18, 2006; 47(8): 1727 - 1730. [Full Text] [PDF] |
||||
![]() |
H. F.R. Dohmann, E. C. Perin, C. M. Takiya, G. V. Silva, S. A. Silva, A. L.S. Sousa, C. T. Mesquita, M.-I. D. Rossi, B. M.O. Pascarelli, I. M. Assis, et al. Transendocardial Autologous Bone Marrow Mononuclear Cell Injection in Ischemic Heart Failure: Postmortem Anatomicopathologic and Immunohistochemical Findings Circulation, July 26, 2005; 112(4): 521 - 526. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. Verma, O. M. Wazni, N. F. Marrouche, D. O. Martin, F. Kilicaslan, S. Minor, R. A. Schweikert, W. Saliba, J. Cummings, J. D. Burkhardt, et al. Pre-existent left atrial scarring in patients undergoing pulmonary vein antrum isolation: An independent predictor of procedural failure J. Am. Coll. Cardiol., January 18, 2005; 45(2): 285 - 292. [Abstract] [Full Text] [PDF] |
||||
![]() |
H. Samady, C. J. Choi, M. Ragosta, E. R. Powers, G. A. Beller, and C. M. Kramer Electromechanical Mapping Identifies Improvement in Function and Retention of Contractile Reserve After Revascularization in Ischemic Cardiomyopathy Circulation, October 19, 2004; 110(16): 2410 - 2416. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Graf, M. Gyongyosi, A. Khorsand, S. G. Nekolla, C. Pirich, K. Kletter, R. Dudczak, D. Glogar, G. Porenta, and H. Sochor Electromechanical Properties of Perfusion/Metabolism Mismatch: Comparison of Nonfluoroscopic Electroanatomic Mapping with 18F-FDG PET J. Nucl. Med., October 1, 2004; 45(10): 1611 - 1618. [Abstract] [Full Text] [PDF] |
||||
![]() |
E. C. Perin, H. F.R. Dohmann, R. Borojevic, S. A. Silva, A. L.S. Sousa, G. V. Silva, C. T. Mesquita, L. Belem, W. K. Vaughn, F. O.D. Rangel, et al. Improved Exercise Capacity and Ischemia 6 and 12 Months After Transendocardial Injection of Autologous Bone Marrow Mononuclear Cells for Ischemic Cardiomyopathy Circulation, September 14, 2004; 110(11_suppl_1): II-213 - II-218. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. Vahlhaus, H. J. Bruns, G. Breithardt, and T. Wichter Direct epicardial mapping predicts the recovery of left ventricular dysfunction in chronic ischaemic myocardium: Reply Eur. Heart J., August 1, 2004; 25(15): 1367 - 1368. [Full Text] [PDF] |
||||
![]() |
M. Saeed, R. Lee, A. Martin, O. Weber, G. A. Krombach, S. Schalla, M. Lee, D. Saloner, and C. B. Higgins Transendocardial Delivery of Extracellular Myocardial Markers by Using Combination X-ray/MR Fluoroscopic Guidance: Feasibility Study in Dogs Radiology, June 1, 2004; 231(3): 689 - 696. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. S. Lee and R. R. Makkar Stem-Cell Transplantation in Myocardial Infarction: A Status Report Ann Intern Med, May 4, 2004; 140(9): 729 - 737. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. Vahlhaus, H.-J. Bruns, J. Stypmann, T. D.T Tjan, F. Janssen, M. Schafers, H. H Scheld, O. Schober, G. Breithardt, and T. Wichter Direct epicardial mapping predicts the recovery of left ventricular dysfunction in chronic ischaemic myocardium Eur. Heart J., January 2, 2004; 25(2): 151 - 157. [Abstract] [Full Text] [PDF] |
||||
![]() |
E. C. Perin, H. F.R. Dohmann, R. Borojevic, S. A. Silva, A. L.S. Sousa, C. T. Mesquita, M. I.D. Rossi, A. C. Carvalho, H. S. Dutra, H. J.F. Dohmann, et al. Transendocardial, Autologous Bone Marrow Cell Transplantation for Severe, Chronic Ischemic Heart Failure Circulation, May 13, 2003; 107(18): 2294 - 2302. [Abstract] [Full Text] [PDF] |
||||
![]() |
E. C. Perin, Y.-J. Geng, and J. T. Willerson Adult Stem Cell Therapy in Perspective Circulation, February 25, 2003; 107(7): 935 - 938. [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Circulation Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search Copyright © 2002 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited. |