doi: 10.1161/CIRCIMAGING.108.778902
Original Articles |
Sequential Changes of Myocardial Microstructure in Patients Postmyocardial Infarction by Diffusion-Tensor Cardiac MR
Correlation With Left Ventricular Structure and Function
From the Faculty of Medicine (M.-T.W., Y.-L.H., K.-R.C.), School of Medicine, National Yang Ming University, Taipei; Department of Radiology (M.-T.W., Y.-L.H., H.-B.P.), Kaohsiung Veterans General Hospital, Kaohsiung; Department of Medical Imaging (M.-Y.M.S., W.-Y.I.T.), National Taiwan University Hospital, Taipei; Institute of Biomedical Engineering (M.-Y.M.S.), National Yang Ming University, Taipei; Section of Cardiology (K.-R.C.), Department of Medicine, Kaohsiung Veterans General Hospital, Kaohsiung; Department of Psychiatry (P.Y.), College of Medicine, Kaohsiung Medical University and Kaohsiung Medical University Hospital, Kaohsiung; Department of Radiation Technology (H.-B.P.), College of Medical Sciences, I-Shou University, Kaohsiung, Taiwan, Republic of China; Department of Radiology (T.G.R., V.J.W.), Martinos Center for Biomedical Imaging, Massachusetts General Hospital, Harvard Medical School, Boston, Mass; and Center for Optoelectronic Biomedicine (W.-Y.I.T.), National Taiwan University College of Medicine, Taiwan, Republic of China.
Correspondence to Ming-Ting Wu, MD, Section of Thoracic and Circulation Imaging, Department of Radiology, Kaohsiung Veterans General Hospital, No. 386, Ta-Chung 1st Rd, Kaohsiung, Taiwan 813, Republic of China. E-mail wu.mingting{at}gmail.com, mingting.wu@isca.vghks.gov.tw, or co-correspondent Wen-Yih I Tseng, MD, PhD.
Received March 11, 2008; accepted November 6, 2008.
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionConclusionsReferences |
|---|
Background— We used diffusion-tensor cardiac MR to investigate myocardial microstructure changes, including tissue integrity (mean diffusivity [MD], fractional anisotropy) and fiber architecture (helix angles) in patients with recent myocardial infarction (MI). This study aimed to investigate the sequential changes of myocardial microstructure and its relationships with changes of macrostructure and function of the left ventricle post-MI.
Methods and Results— Seventeen patients (age, 55.1±11.5 years; all men) participated in the follow-up study. Diffusion-tensor cardiac MR, cine gradient echo for left ventricle function, and late gadolinium enhancement for viability were measured from recent to chronic MI (median interval, 191 days). When compared with the remote zone, the infarct-adjacent zone showed overall increase of MD (2-way MANOVA, F1,16=36.3; P<0.001), decrease of fractional anisotropy (F1,16=5.8; P=0.029), and decrease of mean helix angles (F1,16=62.0; P<0.001). From recent to chronic MI, there was overall sequential decrease of MD (F1,16=22.6; P<0.001) and increase of fractional anisotropy (F1,16=7.8; P=0.013). Multiple linear regression showed that the improvement of wall thickening in the infarct-adjacent zone correlated with sequential decrease of MD in the infarct-adjacent zone (r=–0.70; P=0.002) and increase of mean helix angles (ie, more right-handed helical myofiber reorientation, predominantly subendocardial location) in the remote zone (r=0.60; P=0.011). Likewise, wall thickening in the remote zone correlated with MD in the remote zone (r=–0.72; P=0.001) and mean helix angles in the infarct-adjacent zone (r=0.72; P=0.001).
Conclusion— Diffusion-tensor cardiac MR suggests that sequential zonal improvement of tissue integrity and fiber architecture remodeling both associate with sequential recovery of zonal wall thickening of the left ventricle from recent to chronic MI.
Key Words: imaging • magnetic resonance imaging • myocardial infarction • remodeling
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionsReferences |
|---|
Postinfarction late remodeling involves the left ventricle (LV) globally and is associated with time-dependent dilatation, distortion of ventricular shape, rearrangement of fiber architecture, and mural hypertrophy.1 Clearly, there is a great need for, and utility in, the in vivo acquisition of a wealth of cardiac microstructures such as geometry, fiber and sheet orientation, and cell populations to improve our ability to design strategies for effective restoration of myocardium postinfarction.2,3
Editorial see p 4
Clinical Perspective see p 32
Recently, diffusion-tensor cardiac MR (DT-CMR) has emerged as a unique method for the nondestructive reconstruction of the fiber structure of the LV, which has been validated to have strong correspondence with fiber orientation by histological correlation4–6 and also has been modified to advance its application to the living human heart.7–10 In a previous report, we used DT-CMR to observe the alteration of tissue integrity, indicated by mean diffusivity (MD, also known as trace-apparent diffusion coefficient), fractional anisotropy (FA), and fiber architecture, indicated by helix angles (HA), in patients post–myocardial infarction (MI) at a median interval of 26 days.11 We found alteration of tissue integrity and fiber architecture, which had a significant correlation with the viability zones and regional wall function. When compared with the remote zone, the infarct-adjacent zone showed significant decrease of wall thickness and wall thickening at the macrostructure level of the LV, as well as significant decrease of tissue integrity, reflected by increased MD and reduced FA, increase of left-handed helical fibers (LHF), and decrease of right-handed helical fibers (RHF) at the microstructure level.11
To investigate the sequential change of tissue integrity and fiber architecture during myocardial healing and remodeling, we repeated the identical DT-CMR in patients from the previous study who were able and willing to participate in the continuing study. By observing sequential changes of macrostructure and microstructure and the relationships between them, we hypothesized that there is biomechanical relevance of the alteration of tissue integrity and fiber architecture participated in the remodeling process of the myocardium postinfarction, which could be revealed by DT-CMR.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionsReferences |
|---|
Subjects
Patients
From the 37 patients enrolled in our previous study,11 we performed a follow-up study on those who satisfied the following inclusion criteria: (1) receiving the initial MR scan after onset of acute MI in the previous study,11 and (2) being clinically stable and cooperative for the MR scan. Twenty patients participated, but 3 patients were excluded due to suboptimal image quality, which were mainly caused by irregular heart rhythm during scanning. As a result, images of 17 patients (age, 55.1±11.5 years; New York Heart Association class I, n=13; class II, n=4; all men) were available for the study. The first MR was taken in the recent MI phase (22±14 days after acute MI) and the follow-up MR was taken in the chronic MI phase (213±59 days after acute MI). The interval between the 2 MR scans was 191±59 days. Five patients received percutaneous coronary intervention before and 7 patients after the first MR (median of interval=18.4 and 13.0 days, respectively); the remaining 5 patients received no percutaneous intervention. All patients were followed-up at the cardiological clinics with optimal post-MI medical therapy and had no recurrent MI. The study was approved by the Institutional Review Board of the hospital. All patients provided informed consent before the study.
Healthy Controls
To investigate the normal range and reproducibility of DT-CMR measurement of myocardial microstructure, we enrolled 7 age-matched healthy men without known history of cardiac disease as a control group (range, 45 to 65 years old). Except late gadolinium-enhanced imaging, we used the same MR pulse sequences as those applied to the patients. All control subjects received 2 MR examinations, 6 months apart, with the same protocol for interscan reliability.
MR Acquisitions
The study was performed on a 1.5 T imager (CVi, GE Healthcare, Milwaukee, Wis) with a dedicated quadrature cardiac surface coil (Nova Medical, Wilmington, Mass). The imaging protocol for the initial and follow-up studies was identical and was described in detail in our previous study.11
Fast Gradient Echo Cine for Macrostructure
Localizers were first obtained to determine imaging planes in both 4-chamber and short-axis views. We retrieved and compared patients’ initial MR to ensure that slice locations were as identical as possible. Fast gradient echo cine in short-axis view (repetition time/echo time/flip angle, 9.1 ms/4.9 ms/30°; slice thickness, 8 mm; views per segment, 8; view sharing, 20 frames per R-R interval; matrix, 256x128; field of view, 240x320 mm; 1 average) was obtained for the assessment of regional LV wall function.
DT-CMR for Microstructure
The pulse sequence design of DT-CMR used in this study was described in detail previously.7 In brief, we used a double-gating stimulated-echo single shot echo planar imaging sequence in which 2 diffusion gradient pulses (oriented in 6 nonopposed edge centers of a cube; gradient strength, 40 mT/m; duration, 2 ms; diffusion sensitivity, 300 s/mm2) were applied at an identical cardiac phase in 2 consecutive heart beats, to avoid signal dropout due to cardiac motion. Previously, we showed that myocardial strain could influence diffusion measurement and proposed to apply diffusion gradient pulses at the midsystolic phase, so called "sweet spot," to minimize the strain effect.8 We reviewed the cine gradient echo in 4-chamber and short-axis view to determine the trigger delays of end-systole and end-diastole. The midsystole was determined as the midpoint between end-diastole and end-systole. The sweet spots of infarct-adjacent zone and remote zone were determined separately. If the sweet spots were different, 2 were averaged and used for trigger delay.
DT-CMR was performed with intermittent breath holds. Each breath hold spanned 14 heartbeats to obtain 6 diffusion-weighted images and a reference image with null gradient. We acquired 3 DT-CMR slices in short-axis view at mid-LV level (slice thickness, 8 mm; interslice gap, 4 mm), therefore composing a "midventricular volume" covering 32-mm thickness at voxel resolution of 1.88x1.88x8 mm. The imaging parameters for echo-planar image were field of view of 240 mm, matrix size of 128x128, repetition time/echo time of 2 R-R interval/42 ms, average of 12, and duration of the signal-readout of 640 µsec. To minimize misregistration of DT images acquired at different breath holds, we coached the patients to hold the breath at "quiet end expiration," and applied a plethysmography to monitor the respiratory movement to ensure images acquired at a consistent breath-holding position.
Late Gadolinium-Enhancement MR for Viability Zone
Late gadolinium-enhancement (LGE) MR was used to identify the infarct area.12 After a delay interval of 8 to 15 minutes after a bolus of 0.2 mmol/kg of gadopentetate dimeglumine, a series of 180° inversion-recovery segmented gradient-echo T1-weighted images was acquired in the same imaging planes as for DT-CMR and cine fast gradient echo.
Figure 1 illustrates a case of MI over the territory of let anterior descending coronary artery, showing LGE-MR for infarct-adjacent zones and DT-CMR for MD, FA, and HA maps at recent MI and chronic MI.
|
Data Analysis
Data Set: Midventricular Volume
Because DT-CMR was performed only on the 3 adjacent slices at mid-LV level, the "midventricular volume," the data analysis of LGE-MR viability map and regional LV function were performed on this volume only.11 We then averaged the measurements from the 3 slices to represent the volume data, which was categorized into 2 viability zones; therefore, a volume data with 2 zones represented the data set of a single subject. Combining the initial and follow-up MR as 1 pair, a total of 17 pairs of data sets entered the data analysis for the 17 patients. For healthy control, similar procedure for midventricular volume was performed; however, there was no difference in terms of viability zones and all measurements were averaged over the whole volume.
Evaluation of Myocardial Macrostructure
Viability Zones and Wall Thickness
High-resolution LGE-MR was used for zonal segmentation of the viability. Using commercial software (MASS, GE Healthcare) on an off-line workstation (Advantage Window 4.2, GE Healthcare), one author (C.K.R.), who was blinded to the infarct-related arteries of the patients, traced the endocardial and epicardial contours of the myocardium. With the modified centerline method, the program automatically generated 100 cords perpendicular to the centerline.13 The infarct segment was defined by the distinct hyperenhanced area visually12 by one author (H.Y.L.) without knowledge of infarct-related arteries. For each subject, an infarct-adjacent zone was defined to include segments with transmural or nontransmural hyperenhancement,14 and a remote zone to include all segments showing no hyperenhancement. The cords encompassed in the LGE segments were then counted and divided by 100 to represent the length fraction of the infarct-adjacent zone. The length fraction of the remote zones was derived from the number of cords uncounted.
The wall thickness measurement on this LGE-MR was used to represent the wall thickness of viability zone. The identical zonal demarcation was applied to gradient echo cine and DT-CMR both for zonal analysis of regional wall thickening and DTI measurements.
Regional LV Wall Thickening
Evaluation of LV wall function was performed on the fast gradient echo cine images with the modified centerline method by a semiautomated program (MASS) as described in the literature.13 One author (C.K.R.) traced the epicardial and endocardial contours without knowledge of the infarct-related arteries and viability maps by LGE-MR. Regional wall thickness and wall thickening were evaluated. The wall thickening was expressed by (wall thickness of end-systole–wall thickness of end-diastole)/wall thickness of end-diastolex100%.13
Evaluation of Myocardial Microstructure
Tissue Integrity and Fiber Architecture
The details of the calculation of MD, FA, and HA were described previously.11 Each DTI data set was analyzed by one author (S.M.Y.) blinded to the results of LV wall function and LGE-MR viability maps. The program was institute-developed using Mathematica (version 4.0, Wolfram Research Inc). We did quality assurance of the images by discarding the images that are morphologically different from a reference image. We coregistered the images of the same diffusion-sensitivity gradient to the reference image by performing in-plane translation to achieve maximum correlation of image intensity. The 2 MR data of a patient were analyzed in pair with the same setting of parameters, but in a random order blinded to observer (S.M.Y.).
To evaluate the myocardial tissue integrity, we measured MD and FA of the diffusion tensor. MD was quantified as the mean of the 3 eigenvalues of the diffusion tensor. FA represents the degree of deviation of a diffusion ellipsoid from a sphere and was quantified as the SD of the eigenvalues of the diffusion tensor normalized by the "magnitude" of the 3 eigenvalues of the diffusion tensor. Myocardial fiber orientation at each pixel was defined by the first eigenvector of the diffusion tensor in the DT-CMR. Fiber HA was determined according to Streeter’s original convention,15 simplified to cylindrical coordinates for image analysis.5,16 To compare the difference in fiber orientations (ie, HA) across the 2 viability zones, we calculated the transmural average of HA as mean HA of each viability zone; in addition, we divided the fiber orientation into 3 groups with continuous-scale pseudocolor encoding on 128x128 matrix: (1) –90° to –30°, as LHF, mainly distributed in the subepicardium and pseudocolored as green hue; (2) –30° to 30°, circumferential fibers (CF), pseudocolored as blue hue; and (3) 30° to 90°, RHF, mainly distributed in the subendocardium, pseudocolored as red hue. Finally, the percentages of LHF, CF, and RHF in each viability zone were expressed as LHF%, CF%, and RHF%, respectively.11
Sequential Changes of the Measurements
The sequential change of wall thickness and wall thickening were defined as (measurement at chronic MI–measurement at recent MI)/measurement at recent MIx100%. The sequential changes of DTI measurement were defined as (measurement at chronic MI–measurement at recent MI).
The reproducibility of repeated DTI measurements in healthy control was evaluated by variability, defined as absolute difference divided by average of 2 measurementsx100%; and coefficient of variance, defined as SD of absolute difference between 2 measurements divided by the average of the 2 measurements.
Statistical Analysis
Values of measurements of macrostructure (wall thickness and thickening) and microstructure (MD, FA, mean HA, LHF%, CF%, and RHF%) of each zone are expressed as mean±SEM. Otherwise, values such as age, days, ms are expressed as mean±SD. The viability zonal effect (2 zones, infarct-adjacent versus remote zone) and sequential effect (2 time points, recent versus chronic) within subject on the macrostructure and microstructure measurements were evaluated by 2-way (ie, zone and time point) MANOVA and paired t test. Relationships between sequential changes of microstructure versus that of macrostructure were evaluated by Pearson’s correlation coefficients. Multiple linear regression analysis was used to search MD, FA, and mean HA as potential associations of changes of wall thickness and thickening. P<0.05, 2-sided, was considered as significant. Statistical analysis was performed with SPSS 11.0 (SPSS Inc, Chicago, Ill).
The authors had full access to and take full responsibility for the integrity of the data. All authors have read and agree to the manuscript as written.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionsReferences |
|---|
General Patients Characteristics
The infarct-related arteries involved left-anterior-descending artery in 9, circumflex artery in 2, and right coronary artery in 6 cases, and involved infracted walls anterior in 10, septal in 3, inferior in 7, and lateral in 4 cases. Of the midventricular volume of interest where the microstructure also measured, the size of infarct-adjacent area was 31.5±10.5% in recent MI and 28.2±9.5% in chronic MI. The ejection fraction was 44±7% in recent MI and 43±9% in chronic MI.
Changes of Myocardial Macrostructure
Table 1 lists all the measurements of macrostructure and microstructure of the 2 viability zones, and the sequential pairwise comparison between recent and chronic MI. The differences between the remote zone and infarct-adjacent zone were also listed and compared for sequential changes.
|
There was overall decrease of wall thickness in the infarct-adjacent zone relative to the remote zone (2-way MANOVA, F1,16=29.9; P<0.001). Interval changes from recent MI to chronic MI showed sequential decrease of wall thickness in the infarct-adjacent zone (9.6±0.5 to 8.8±0.5 mm; P=0.003), whereas sequential increase in the remote zone (10.4±0.6 to 11.2±0.5 mm; P=0.005). The difference of wall thickness between the remote zone and infarct-adjacent zone was 0.7±0.3 mm in recent MI, and the difference progressed significantly to 2.4±0.3 mm in chronic MI (P<0.001).
There was overall decrease of wall thickening in the infarct-adjacent zone relative to the remote zone (F1,16=21.3; P<0.001). There was no significant sequential changes of wall thickening in the infarct-adjacent zone (39.0±3.5 to 32.4±4.6%; P=0.17) and in the remote zone (55.9±5.9 to 62.9±5.5%; P=0.15); whereas the difference of wall thickening between the remote zone and infarct-adjacent zone showed progression from 16.9%±4.2 in recent MI to 30.5%±6.2 in chronic MI (P=0.020).
The above findings indicated that the LV postinfarction underwent progressive eccentric remodeling of wall thickness and wall thickening from recent MI to chronic MI.
Changes of Myocardial Microstructure
There was overall increase of MD in the infarct-adjacent zone relative to the remote zone (F1,16=36.3; P<0.001). There was a sequential decrease of MD from recent MI to chronic MI in the infarct-adjacent zone (9.2±0.5 to 7.4±0.3x10–6 cm2/s; P<0.001) and in the remote zone (8.1±0.5 to 6.7±0.3x10–6 cm2/s; P=0.001) (Figure 2A). The difference of MD between infarct-adjacent zone and remote zone showed a trend of reduction from recent MI to chronic MI (–1.1±0.2 to –0.7±0.1x10–6 cm2/s; P=0.06; Table 1; Figures 2A through 2C).
|
There was overall decrease of FA in the infarct-adjacent zone relative to the remote zone (F1,16=5.8; P=0.029). There was a trend of increase of FA from recent MI to chronic MI in the infarct-adjacent zone (0.25±0.01 to 0.27±0.01; P=0.07) and in the remote zone (0.27±0.01 to 0.29±0.01; P=0.018; Figure 2B). The difference of FA between infarct-adjacent zone and remote zone had no significant interval change (0.02±0.01 to 0.02±0.01; P=0.59).
Helical reorientation of fiber architecture was negative mean HA (more left-handed helical) in the infarct-adjacent zone and positive mean HA (more right-handed helical) in the remote zone (F1,16=62.0; P<0.001). There was no significant interval change of mean HA between recent MI and chronic MI in the infarct-adjacent zone (–10.4±3.6° to –10.8±2.1°; P=0.91) and in the remote zone (13.1±1.5° to 15.4±2.1°; P=0.31; Figure 2C). The difference of mean HA between infarct-adjacent zone and remote zone showed no significant interval change from recent MI to chronic MI (23.5±3.9° to 26.1±2.9°; P=0.31).
Correlation of Sequential Changes of Myocardial Macrostructure Versus Microstructure
In associating with the sequential changes of wall thickness, there was only one sequential change of microstructure showing significant correlation, ie, the sequential change of RHF% (r=0.66; P=0.005; Table 2; Figure 3).
|
|
There were several significant correlations between the sequential changes of wall thickening and microstructure. The sequential change of wall thickening in the infarct-adjacent zone correlated inversely with the sequential change of MD in the infarct-adjacent zone (r=–0.70; P=0.002) and the remote zone (r=–0.53; P=0.029), and correlated with the sequential change of mean HA in the remote zone (r=0.60; P=0.011) and the sequential change of RHF% in remote zone (r=0.61; P=0.010). Multiple linear regression analysis showed that the change of wall thickening in the infarct-adjacent zone was associated with the change of MD in the infarct-adjacent zone (the local wall) (β=–0.180; 95% CI=–0.292, –0.067) and the change of mean HA in the remote zone (the distant wall) (β=0.022; 95% CI=0.004, 0.041; Table 2; Figure 3A and 3B).
The sequential change of wall thickening in the remote zone was correlated inversely with the sequential change of MD in the infarct-adjacent zone (r=–0.60; P=0.011) and the remote zone (r=–0.72; P=0.001), and correlated with the sequential change of mean HA in the infarct-adjacent zone (r=0.72; P=0.001). Multiple linear regression analysis showed that the change of wall thickening in the remote zone was associated with the change of MD in the remote zone (the local wall) (β=–0.153; 95% CI=–0.296, –0.009) and the change of mean HA in the infarct-adjacent zone (the distant wall) (β=0.017; 95% CI=0.002, 0.031; Table 2; Figure 3C and 3D).
Normal Values and Reproducibility of DT-CMR in Healthy Control
The mean of MD of the global myocardium in the healthy controls (N=7) were 6.3±0.2x10–6 cm2/s in the first MR and 6.5±0.3x10–6 cm2/s in the repeated MR (P=0.32). The coefficient of variation was 4.5%. In the first and repeated MR, the mean of FA were 0.34±0.01 and 0.33±0.02 (P=0.54), respectively. The coefficient of variation was 7.0%. The mean of HA were 0.2±2.1° and 0.3±2.2° (P=0.58), respectively. The coefficient of variation was 11.2%. The RHF% were 26.6±3.2% and 28.2±2.8% (P=0.48), CF% were 52.8±3.6% and 49.8±4.4% (P=0.26), and LHF% were 20.6±1.1% and 22.0±1.8% (P=0.25), respectively.
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionsReferences |
|---|
This follow-up study on the sequential changes of myocardial macrostructure and microstructure from recent MI to chronic MI supported and extended our previous observations. From recent MI to chronic MI, in macrostructure, the wall thickness and wall thickening showed progressive decrease in the infarct-adjacent zone and increase in the remote zone. In microstructure, there was global recovery of tissue integrity as reflected by the sequential decrease of MD. In correlation with functional changes, improvement of tissue integrity and increased right-handed helical reorientation correlated with recovery of the wall thickening.
MD indicates mean diffusivity of water molecules and reflects the redistribution of intracellular and extracellular space volumes.17–19 We postulate that the increase of MD in the infarct-adjacent zone in recent MI results from initial increase in extracellular space due to cell lysis, followed by a decrease in the chronic MI as the healing process ensues18,20 We believe our finding on sequential change of MD is reliable because (1) repeated measurements of MD of normal myocardium in our healthy control (6.3±0.2 to 6.5±0.3x10–6 cm2/s, coefficient of variation=4.5%) were well reproducible; (2) our normal value of myocardial MD is close to a previous study (approximately 6.0±1.0x10–6 cm2/s) in healthy volunteers using a more advanced cardiac DTI technique using a pair of bipolar diffusion gradient pulses, which is inherently insensitive to cardiac motion and strain21; and (3) our normal values of myocardial MD is close to that of perfused myocardium of rabbits (7.2±0.7x10–6 cm2/s).22 Therefore, in our chronic MI, MD in the remote zone (6.7±0.3x10–6 cm2/s) was found to come closer to that of the normal myocardium, suggesting that in chronic MI, the tissue integrity of the remote zone was undergoing recovery toward the normal condition.
Our study found that in the recent MI (median, 21 days), the MD of the remote region was elevated (8.1±0.5x10–6 cm2/s) when compared with the normal values (6.5±0.3x10–6 cm2/s). Additionally, the decrease of MD over time (from 8.1±0.3x10–6 cm2/s to 6.7±0.3x10–6 cm2/s) was correlated with the improvement of wall thickening (Table 2). These findings suggest that in recent MI, myocardial edema is present even in the remote region. The edema subsides over time and is accompanied with the recovery of the regional wall thickening. Recent advance in T2WI cardiac MR could detect the presence of myocardial edema. It was reported that the area of edema is generally larger than the myocardial necrosis and may extend to the adjacent region.19 In our study, we found MD progressively increased from remote to infarct myocardium, suggesting that MD might be a sensitive marker for myocardial edema and could detect the subtle increase of edema in the remote region. Our finding is not consistent with an ex vivo DT-CMR study on a porcine model by Wu et al23 in which no significant difference in MD was found when compared with the control group in the adjacent and remote regions. The discrepancy may be related to the dehydration procedure performed in formalin fixation, which may reduce the myocardial water content and consequently MD.
FA is considered a sensitive marker for microstructure integrity of neuronal axons and has been validated histologically. However, the validity of FA in representing integrity of myocardial fibers is not as clear as neuronal axons. This may be due to the fact that the volume fraction of the extracellular space in the myocardium varies spatially, leading to heterogeneity of FA over the ventricular walls. As Jiang et al24 demonstrated in the sheep myocardium, FA varies transmurally; it remained relatively constant from the epicardium to the midwall and then decreased steadily toward the endocardium. The relative low FA values in the inner wall may confound the decrease of FA due to myocardial ischemia, which is predominantly located in the subendocardium.
As shown in Figure 3 and Table 2, the recovery of wall thickening in the infarct-adjacent zone was associated with the sequential changes of MD in the infarct-adjacent zone and mean HA in the remote zone. Likewise, the recovery of wall thickening in the remote zone was associated with the sequential changes of MD in the remote zone and mean HA in the infarct-adjacent zone. The above findings suggested that microstructural improvement of local tissue integrity (decrease of MD) and remodeling of distant fiber architecture (ie, more right-handed helical reorientation) might both associate with the functional improvement of wall thickening.
The current study has made 2 important observations regarding the increase of RHF% in the remote zone of myocardium postinfarction. First, the sequential increase of wall thickness in the remote zone had significantly correlated with the sequential increase of RHF% (r=0.66; P=0.005); second, improvement of wall thickening in the infarct-adjacent zone was correlated with the increase of mean HA in the remote zone, which was mainly due to increase of RHF% (r=0.61; P=0.010; Table 2). The increase of RHF% presented a preferential increase of subendocardial fiber, which has also been observed in the LV after heart failure or exercise-training as a stress adaptive response.25,26
Our finding of fiber architecture adaptation after MI is at variance with those observed by Chen et al20 at ex vivo animal study of DT-CMR post-MI. We found it difficult to apply direct comparison between Chen’s versus our results (see online-only Data Supplement). In brief, species-distinct myocardial healing and remodeling do exist27,28; specimen preparation-relevant factors such as permanent ligation versus reperfusion therapy29 and ex vitro versus in vivo measurement may affect; and DT-CMR relevant factors such as spatial resolution and anisotropy of voxel dimensions4 may all contribute to the discrepancy. Likewise, there is variance between results by Geerts et al6 and Chen et al.20
The LV asynchrony post-MI may have potential influence on the measurement of microstructure by DT-CMR. With calculation and simulation of the potential effects in our cases, we found that asynchrony has no significant impact on our conclusion (see online-only Data Supplement).
Study Limitation
In this study, only a half of the patients from the initial study were enrolled because of the inclusion and exclusion criteria used in patient selection, as described in the Patient section. The observation was made in a group of patients with stable hemodynamic conditions post-MI. Therefore, the conclusions drawn from this study might not be applicable to patients with poor outcome of functional recovery. Our scan range was limited to the middle ventricular volume to constrain the scan time. In addition, we did not conduct spatial correlation analysis on the slice-level measurements, therefore might loss of power and possibly bias in comparison with approaches that use each individual slice measurements30 accounting for the within-heart correlation of macrostructures and microstructures. Lastly, several borderline results, such as sequential change of wall thickening, MD and FA, may be hindered by type-II error due to small case number.
|
Conclusions |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionsReferences |
|---|
We used DT-CMR to investigate the sequential changes of myocardial macrostructure and microstructure from recent MI to chronic MI. Microstructure improvement of tissue integrity and fiber architecture remodeling toward right-handed helical reorientation might both associate with improvement of wall thickening in macrostructure. This observation may shed light on the management involving restoration of normal scaffold of fiber architecture in myocardium postinfarction.2
|
Acknowledgments |
|---|
The authors thank Yi-Hsin Connie Yang, PhD, Director of Statistical Analysis Laboratory, Department of Clinical Research, Kaohsiung Medical University for statistical consultation. The authors also thank MR Technologists Chia-Chi Hsiao, Mei-Han Chang, Wen-Chuan Huang, Research Assistant Chin-Ying Tsai, and GE healthcare engineers Steven Sheua and Richard Huang for technical assistance.
Sources of Funding
This work was supported by the Medical Research and Advancement Foundation in memory of Dr Chi-Shuen Tsou (VTY92-G3-03 and VGHUST97-G3-2); Kaohsiung Veterans General Hospital Research Program in Taiwan (VGHKS96-060 to M.T.W.).
Disclosures
None.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionConclusionsReferences |
|---|
1. Sutton MG, Sharpe N. Left ventricular remodeling after myocardial infarction: pathophysiology and therapy. Circulation. 2000; 101: 2981–2988.
2. Buckberg GD, Weisfeldt ML, Ballester M, Beyar R, Burkhoff D, Coghlan HC, Doyle M, Epstein ND, Gharib M, Ideker RE, Ingels NB, LeWinter MM, McCulloch AD, Pohost GM, Reinlib LJ, Sahn DJ, Sopko G, Spinale FG, Spotnitz HM, Torrent-Guasp F, Shapiro EP. Left ventricular form and function: scientific priorities and strategic planning for development of new views of disease. Circulation. 2004; 110: e333–e336.
3. Burton RA, Plank G, Schneider JE, Grau V, Ahammer H, Keeling SL, Lee J, Smith NP, Gavaghan D, Trayanova N, Kohl P. Three-dimensional models of individual cardiac histoanatomy: tools and challenges. Ann N Y Acad Sci. 2006; 1080: 301–319.[CrossRef][Medline]
4. Garrido L, Wedeen VJ, Kwong KK, Spencer UM, Kantor HL. Anisotropy of water diffusion in the myocardium of the rat. Circ Res. 1994; 74: 789–793.
5. Scollan DF, Holmes A, Winslow R, Forder J. Histological validation of myocardial microstructure obtained from diffusion tensor magnetic resonance imaging. Am J Physiol. 1998; 275: H2308–H2318.[Medline]
6. Geerts L, Bovendeerd P, Nicolay K, Arts T. Characterization of the normal cardiac myofiber field in goat measured with MR-diffusion tensor imaging. Am J Physiol Heart Circ Physiol. 2002; 283: H139–H145.
7. Reese TG, Weisskoff RM, Smith RN, Rosen BR, Dinsmore RE, Wedeen VJ. Imaging myocardial fiber architecture in vivo with magnetic resonance. Magn Reson Med. 1995; 34: 786–791.[Medline]
8. Tseng WY, Reese TG, Weisskoff RM, Wedeen VJ. Cardiac diffusion tensor MRI in vivo without strain correction. Magn Reson Med. 1999; 42: 393–403.[CrossRef][Medline]
9. Tseng WY, Wedeen VJ, Reese TG, Smith RN, Halpern EF. Diffusion tensor MRI of myocardial fibers and sheets: correspondence with visible cut-face texture. J Magn Reson Imaging. 2003; 17: 31–42.[CrossRef][Medline]
10. Tseng WY, Dou J, Reese TG, Wedeen VJ. Imaging myocardial fiber disarray and intramural strain hypokinesis in hypertrophic cardiomyopathy with MRI. J Magn Reson Imaging. 2006; 23: 1–8.[CrossRef][Medline]
11. Wu MT, Tseng WY, Su MY, Liu CP, Chiou KR, Wedeen VJ, Reese TG, Yang CF. Diffusion tensor magnetic resonance imaging mapping the fiber architecture remodeling in human myocardium after infarction: correlation with viability and wall motion. Circulation. 2006; 114: 1036–1045.
12. Kim RJ, Wu E, Rafael A, Chen EL, Parker MA, Simonetti O, Klocke FJ, Bonow RO, Judd RM. The use of contrast-enhanced magnetic resonance imaging to identify reversible myocardial dysfunction. N Engl J Med. 2000; 343: 1445–1453.
13. Holman ER, Buller VG, de Roos A, van der Geest RJ, Baur LH, van der LA, Bruschke AV, Reiber JH, van der Wall EE. Detection and quantification of dysfunctional myocardium by magnetic resonance imaging. A new three-dimensional method for quantitative wall-thickening analysis. Circulation. 1997; 95: 924–931.
14. Kramer CM, Lima JA, Reichek N, Ferrari VA, Llaneras MR, Palmon LC, Yeh IT, Tallant B, Axel L. Regional differences in function within noninfarcted myocardium during left ventricular remodeling. Circulation. 1993; 88: 1279–1288.
15. Streeter DD Jr, Hanna WT. Engineering mechanics for successive states in canine left ventricular myocardium. I. Cavity and wall geometry. Circ Res. 1973; 33: 639–655.
16. Tseng WY, Reese TG, Weisskoff RM, Brady TJ, Wedeen VJ. Myocardial fiber shortening in humans: initial results of MR imaging. Radiology. 2000; 216: 128–139.
17. Le Bihan D. Looking into the functional architecture of the brain with diffusion MRI. Nat Rev Neurosci. 2003; 4: 469–480.[CrossRef][Medline]
18. Li G, Xiang B, Dai G, Shaw A, Liu H, Yang B, Jackson M, Deslauriers R, Tian G. Tissue edema does not change gadolinium-diethylenetriamine pentaacetic acid (Gd-DTPA)-enhanced T1 relaxation times of viable myocardium. J Magn Reson Imaging. 2005; 21: 744–751.[CrossRef][Medline]
19. Aletras AH, Tilak GS, Natanzon A, Hsu LY, Gonzalez FM, Hoyt RF Jr, Arai AE. 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. 2006; 113: 1865–1870.
20. Chen J, Song SK, Liu W, McLean M, Allen JS, Tan J, Wickline SA, Yu X. Remodeling of cardiac fiber structure after infarction in rats quantified with diffusion tensor MRI. Am J Physiol Heart Circ Physiol. 2003; 285: H946–H954.
21. Dou J, Reese TG, Tseng WY, Wedeen VJ. Cardiac diffusion MRI without motion effects. Magn Reson Med. 2002; 48: 105–114.[CrossRef][Medline]
22. Forder JR, Bui JD, Buckley DL, Blackband SJ. MR imaging measurement of compartmental water diffusion in perfused heart slices. Am J Physiol Heart Circ Physiol. 2001; 281: H1280–H1285.
23. Wu EX, Wu Y, Nicholls JM, Wang J, Liao S, Zhu S, Lau CP, Tse HF. MR diffusion tensor imaging study of postinfarct myocardium structural remodeling in a porcine model. Magn Reson Med. 2007; 58: 687–695.[CrossRef][Medline]
24. Jiang Y, Guccione JM, Ratcliffe MB, Hsu EW. Transmural heterogeneity of diffusion anisotropy in the sheep myocardium characterized by MR diffusion tensor imaging. Am J Physiol Heart Circ Physiol. 2007; 293: H2377–H2384.
25. Gerdes AM, Capasso JM. Structural remodeling and mechanical dysfunction of cardiac myocytes in heart failure. J Mol Cell Cardiol. 1995; 27: 849–856.[CrossRef][Medline]
26. Natali AJ, Wilson LA, Peckham M, Turner DL, Harrison SM, White E. Different regional effects of voluntary exercise on the mechanical and electrical properties of rat ventricular myocytes. J Physiol. 2002; 541: 863–875.
27. Jugdutt BI, Joljart MJ, Khan MI. Rate of collagen deposition during healing and ventricular remodeling after myocardial infarction in rat and dog models. Circulation. 1996; 94: 94–101.
28. Kumar D, Hacker TA, Buck J, Whitesell LF, Kaji EH, Douglas PS, Kamp TJ. Distinct mouse coronary anatomy and myocardial infarction consequent to ligation. Coron Artery Dis. 2005; 16: 41–44.[CrossRef][Medline]
29. Richard V, Murry CE, Reimer KA. Healing of myocardial infarcts in dogs. Effects of late reperfusion. Circulation. 1995; 92: 1891–1901.
30. Bowman FD, Waller LA. Modeling of cardiac imaging data with spatial correlation. Stat Med. 2004; 23: 965–985.[CrossRef][Medline]
|
Footnotes |
|---|
The online-only Data Supplement is available at http://circimaging.ahajournals.org/cgi/content/full/2/1/32/DC1.
Related Articles
Circ Cardiovasc Imaging 2009 2: 32-40.
Circ Cardiovasc Imaging 2009 2: 4-5.
This article has been cited by other articles:
 |
|
 |
|
  D. J. Pennell Cardiovascular Magnetic Resonance Circulation, February 9, 2010; 121(5): 692 - 705. [Full Text] [PDF]
|
|
|
|
 |
|
 R. J. Gibbons, P. A. Araoz, and E. E. Williamson The Year in Cardiac Imaging J. Am. Coll. Cardiol., February 2, 2010; 55(5): 483 - 495. [Full Text] [PDF]
|
|
|
|
 |
|
 C. M. Kramer Insights into Myocardial Microstructure During Infarct Healing and Remodeling: Pathologists Need Not Apply Circ Cardiovasc Imaging, January 1, 2009; 2(1): 4 - 5. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||