Donate Help Contact The AHA Sign In Home
American Heart Association
Circulation
Search: search_blue_button Advanced Search
« Previous Article | Table of Contents | Next Article »
Circulation. 2006;114:1036-1045
Published online before print August 28, 2006, doi: 10.1161/CIRCULATIONAHA.105.545863
CLINICAL PERSPECTIVE
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Data Supplement
Right arrow All Versions of this Article:
114/10/1036    most recent
CIRCULATIONAHA.105.545863v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Wu, M.-T.
Right arrow Articles by Yang, C.-F.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Wu, M.-T.
Right arrow Articles by Yang, C.-F.
Right arrowPubmed/NCBI databases
Medline Plus Health Information
*Heart Attack
*MRI Scans
Related Collections
Right arrow Structure
Right arrow Contractile function
Right arrow CT and MRI
Right arrow Imaging

(Circulation. 2006;114:1036-1045.)
© 2006 American Heart Association, Inc.

Imaging

Diffusion Tensor Magnetic Resonance Imaging Mapping the Fiber Architecture Remodeling in Human Myocardium After Infarction

Correlation With Viability and Wall Motion

Ming-Ting Wu, MD*; Wen-Yih I. Tseng, MD, PhD*; Mao-Yuan M. Su, MS; Chun-Peng Liu, MD; Kuan-Rau Chiou, MD; Van J. Wedeen, MD; Timothy G. Reese, PhD; Chien-Fang Yang, MD

From the Faculty of Medicine, School of Medicine (M.-T.W., C.-P.L., K.-R.C., C.-F.Y.), Institute of Biomedical Engineering (M.-Y.M.S.), National Yang Ming University, Taipei, Taiwan; Section of Thoracic and Circulation Imaging, Department of Radiology (M.-T.W., C.-F.Y.) and Department of Medicine (C.-P.L., K.-R.C.), Kaohsiung Veterans General Hospital, Kaohsiung, Taiwan; Center for Optoelectronic Biomedicine, National Taiwan University College of Medicine, and Department of Medical Imaging, National Taiwan University Hospital, Taipei, Taiwan (W.-Y.I.T.); Department of Radiology, Martinos Center for Biomedical Imaging, Massachusetts General Hospital, Harvard Medical School, Boston (V.J.W., T.G.R.); and Department of Nutrition and Health Science, Fooyin Universtiy, Kaohsiung, Taiwan, ROC (C.-F.Y.).

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, ROC. E-mail wu.mingting{at}gmail.com

Received March 2, 2005; revision received June 28, 2006; accepted June 30, 2006.


*    Abstract
up arrowTop
 *Abstract
down arrowIntroduction
 down arrowMethods
 down arrowResults
 down arrowDiscussion
 down arrowReferences
 
Background— Diffusion tensor magnetic resonance imaging (DT-MRI) provides a means for nondestructive characterization of myocardial architecture. We used DT-MRI to investigate changes in direction-dependent water diffusivity to reflect alterations in tissue integrity (trace apparent diffusion coefficients [ADCs] and fractional anisotropy [FA]), as well as indicators of remodeling of fiber helix angles, in patients after myocardial infarction.

Methods and Results— Thirty-seven patients (35 men, 2 women; median age, 59) after acute myocardial infarction (median interval from onset, 26 days) were enrolled. DT-MRI was performed at the midventricular level to measure trace ADC, FA, and helix angles of myofibers. Helix angles were grouped into left-handed helical fibers, circumferential fibers, and right-handed helical fibers. Measurements were correlated with viability and regional wall motion assessed by contrast-delay-enhancement and cine MRI, respectively. The infarct zone showed significantly increased trace ADC and decreased FA than the remote zone. The percentage of left-handed helical fibers increased from the remote zone (mean±SD, 13.3±5.8%) to the adjacent zone (19.2±9.7%) and infarct zone (25.8±18.4%) (MANOVA, P=0.004). The percentage of right-handed helical fibers decreased from the remote zone (35.0±9.0%) to the adjacent zone (25.5±11.5%) and infarct zone (15.9±9.2%) (P<0.001). Multiple linear regression showed that the percentage of left-handed helical fibers of the infarct zone was the strongest correlate of infarct size and predictor of ejection fraction.

Conclusions— In vivo DT-MRI of postinfarct myocardium revealed a significant increase in trace ADC and a decrease in FA, indicating altered tissue integrity. The redistribution of fiber architecture correlated with infarct size and left ventricular function. This technique may help us understand structural correlates of functional remodeling after infarction.


Key Words: myocardial infarction • remodeling • magnetic resonance imaging


*    Introduction
up arrowTop
 up arrowAbstract
 *Introduction
down arrowMethods
 down arrowResults
 down arrowDiscussion
 down arrowReferences
 
The unique 3-demensional organization of myocardial fibers of the mammalian ventricle is composed not only of circumferential fibers (CFs) and longitudinal fibers but also of obliquely running fibers that form a helical spiral from base to apex. When viewed from the apex of the left ventricle (LV), the orientation of fibers changes smoothly from left-handed helical fiber (LHF) in the subepicardium to CF in the mid wall to right-handed helical fiber (RHF) in the subendocardium.1,2 This structure is a key determinant of ventricular mechanical properties, including torsion, strain, and stress,3,4 and structure adaptation.5,6 From a clinical point of view, understanding the altered tissue integrity and fiber architecture in diseased myocardium such as myocardial infarction (MI) is critically important because it might shed light on the mechanism of structure-function remodeling after MI. However, conventional histological study of myocardial fiber architecture is time-consuming and destructive and can be done in the excised heart only.

Clinical Perspective 1045

Recently, diffusion tensor magnetic resonance imaging (DT-MRI) has emerged as a candidate method for nondestructive reconstruction of the fiber structure of the LV.7–11 DT-MRI probes orientations of organized microstructures on the basis of the principle that the main orientation of the microstructures parallels the direction of the greatest diffusivity of water molecules, which causes signal attenuation in the presence of a magnetic field gradient. By comparing DT-MRI of animal cardiac specimens and conventional histological results, recent reports have demonstrated a direct correlation between MRI and histological fiber angle measurements, thereby validating the DT-MRI approach for complete characterization of myofiber architecture.7–11 DT-MRI was modified by Edelman et al12 and Pearlman and Brookeman13 to reduce the motion-induced signal dropout to measure the water diffusion of the beating heart. Reese et al14,15 and Tseng et al16–18 have used the same modified DT-MRI to successfully observe the myocardial fiber architecture in the normal living human heart.

Until now, DT-MRI studies have been used mostly to delineate normal myofiber architecture in animal fixed heart in vitro7–11 and, less frequently, in human beating hearts in vivo.14–18 There is only 1 report of its utility in postinfarct remodeling of fixed rat myocardium in vitro.19 In the present study, we performed DT-MRI in 37 patients with recent MI. It is known that diffusion indexes derived from DT-MRI, namely trace apparent diffusion coefficient (trace ADC) and fractional anisotropy (FA), reflect the structural integrity of underlying tissues.20 Trace ADC indicates the mean diffusivity of water molecules; it will change if there is pathological change causing redistribution of intracellular and extracellular space volumes. FA measures the variability of water mobility in different directions and will decrease if the organization of tissue structure is destroyed. We hypothesized that (1) the tissue integrity, as reflected by trace ADC and FA, and fiber architecture, as indicated by helix angle, will be changed after infarction; (2) the alteration would differ between infarct zone and noninfarct zones (the latter can be further divided into adjacent zone and remote zone21); and (3) the alteration would correlate with functional changes. Using contrast-delay-enhancement MRI (CDE-MRI) as a reference of viability maps22,23 and cine MRI for LV functional assessment,24 we aimed to investigate the structure-function interaction in postinfarct myocardial remodeling in living human heart using DT-MRI.


*    Methods
up arrowTop
 up arrowAbstract
 up arrowIntroduction
 *Methods
down arrowResults
 down arrowDiscussion
 down arrowReferences
 
Patients
We prospectively enrolled patients suffering from acute MI who fulfilled all of the inclusion criteria: (1) acute MI diagnosed by both ECG and biomarkers; (2) documented culprit lesion by coronary catheter angiography; (3) stable vital signs, regular sinus rhythm, and ability to hold breath for 15 seconds; and (4) no contraindication for MRI study. Thirty-seven patients (35 men, 2 women) were enrolled. Details of the clinical characteristics are listed in Table 1.


View this table:
[in this window]
[in a new window]

  		TABLE 1. Clinical Characteristics

MRI Acquisitions
All MRI studies were performed on a 1.5-T imager (CVi, Signa, GE Healthcare, Milwaukee, Wis) with a dedicated quadrature cardiac surface coil (Nova Medical, Wilmington, Mass). Patients were positioned off-centered right so that the heart aligned to the isocenter of the magnet. To minimize image misregistration between multiple breathholds, we coached the patients to hold their breath at "quiet end expiration" and applied a plethysmography to monitor the respiratory movement to ensure that the images were acquired at desirable position.

Definition of LV Long Axis and Short Axis
In the present study, the LV long axis was determined as the line bisecting the LV in both 4- and 2-chamber views under the assumption that the LV is axially symmetric. The short axis was defined perpendicular to the long axis.

Cine Fast Gradient Echo
Localizers were first obtained to determine imaging planes in the 4-chamber and short-axis views. Breathhold cine MRI in these 2 views was performed with segmented k-space spoiled gradient echo sequence (repetition time, 9.1 ms; echo time, 4.9 ms; flip angle, 30°; slice thickness, 8 mm; views per segment, 8; view sharing, 20 frames per R-R interval; matrix, 256x128; field of view, 240x320 mm; and 1 average). The short-axis view was obtained from 1 cm below the level of the mitral valve insertion and then every 8 mm throughout the whole LV. One breathhold was obtained for 1 slice, and a series of 10 to 14 slices was obtained for evaluation of LV function.

Cardiac DT-MRI
Details of the pulse sequence design were given previously.14,15 In brief, cardiac DT-MRI was performed using double-gating stimulated-echo single-shot echo planar imaging (Figure 1). To avoid signal dropout resulting from cardiac motion and to minimize strain effects on diffusion measurement, 2 diffusion gradient pulses were ECG triggered at an identical phase of mid systole in 2 consecutive cardiac cycles. The diffusion-sensitive gradient pulses were oriented in 6 nonopposed edge centers of a cube, ie, {1, ±1, 0}, {0, ±1, 1}, {±1, 0, 1}, with the intensity |gD|=40 mT/m and duration {delta}D=2 ms adjusted to yield diffusion sensitivity: b={int} |kD|2;dt=300 s · mm–2, where kD={gamma}gD{delta} and {gamma} is the proton gyromagnetic ratio.


Figure 1177604
View larger version (11K):
[in this window]
[in a new window]

  		Figure 1. Pulse sequence of cardiac diffusion MRI. The sequence is a stimulated-echo single-shot echo planar imaging with 2 ECG triggers to locate the diffusion-encoding gradient pulses (Gd) at identical phase delays, {Psi}, in consecutive cardiac cycles. The time between the onsets of the 2 diffusion-encoding gradients is the diffusion time, {Delta}, precisely equal to the R-R interval.

Cardiac DT-MRI was performed with intermittent breathholds. Each breathhold spanned 14 heartbeats to obtain 6 diffusion-weighted images and a reference image with null gradient. We acquired 3 DT-MRI slices in the short-axis view at the mid LV level (each slice thickness, 8 mm; interslice gap, 4 mm) that composed a "midventricular volume" covering 32-mm thickness (see online-only Data Supplement, section 1) (field of view, 240 mm; matrix size, 128x128; repetition time, 2 R-R intervals; echo time, 42 ms; and readout duration, 0.64 ms). With a number of excitations equal to 12, a total of 252 images, ie, 3 levels times 7 directions times 12 averages, at spatial resolution of 1.88x1.88x8 mm were obtained in &30 minutes.

CDE-MRI for Viability Map
CDE-MRI was used to estimate infarct size.22,23 A bolus of 0.2 mmol gadopentetate dimeglumine (Magnevist, Schering, Berlin, Germany) per 1 kg body weight was administrated into an antecubital vein by hand injection. After a delay interval of 8 to 15 minutes, a series of segmented 180° inversion-recovery gradient-echo T1-weighted images was acquired in the same views as for cine images in the short-axis view. The inversion time, typically &250 to 300 ms, was optimized to null the signal of normal myocardium to maximize the detection of contrast-enhanced infarcted area. Trigger delay was 400 ms, and voxel size was 1.41x1.41x8 mm. The high-resolution viability map was used for zonal segmentation for the infarct, adjacent, and remote zones.

Data Analysis
Midventricular Volume
Because DT-MRI was performed only on the 3 adjacent slices at the mid LV level, the midventricular volume (see the online-only Data Supplement, section 1), data analysis of CDE viability map, and regional LV function were performed on this volume only. We then averaged the measurements from the 3 slices to represent the volume data, which were categorized into 3 viability zones on the basis of the CDE viability map; therefore, a volume data with 3 zones represented the data set of a single subject. Thus, 37 data sets entered analysis.

Tissue Integrity and Myocardial Fiber Architecture
Each DT-MRI data set was analyzed by one author (M.-Y.M.S.) blinded to the results of LV function and CDE viability map. The program was developed with Mathematica (version 4.0, Wolfram Research, Inc, Champaign, Ill). We ensured the quality of the images by discarding the images that were morphologically different from the reference images. We then coregistered all the selected images of the same diffusion-sensitivity gradient to the reference images by performing in-plane translation to achieve maximum correlation of image intensity. The symmetrical 3-dimensional diffusion tensor, D, was reconstructed from 6 diffusion-attenuated images, Ii (i =1,...,6) obtained from 6 diffusion-sensitizing gradients, gi, and a reference image with null gradient, I0, by the following equation25: equation


Formula 1

Cross interaction between diffusion and imaging gradients, given immediate refocusing of the gradients, was negligibly small.

To evaluate the myocardial tissue integrity, we measured trace ADC and FA of the diffusion tensor. Trace ADC, denoted by <d>, was quantified as the mean of the 3 eigenvalues of the diffusion tensor d1, d2, and d3: equation


Formula 2

FA represents 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: equation


Formula 3

Myocardial fiber orientation at each pixel was defined by the first eigenvector of the diffusion tensor in the cardiac DT-MRI. Fiber helix angle was determined according to the convention described by Streeter and Hanna26,27 and Scollan et al.9,11 Briefly, fiber orientation was assigned to have LHF from the subepicardium to the mid wall and as RHF from the mid wall to the subendocardium. Fiber helix angle at each pixel was the angle between the local fiber vector and local circumferential vector. The circumferential vector was perpendicular to both the long axis of the LV and the local radial (or transmural) vector. Globally, fiber orientations of the LV wall show a circularly symmetric pattern, from about –90° at the subepicardium through 0° at the mid wall to 90° at the subendocardium. To compare differences in fiber orientations among the 3 viability zones, we divided the fiber orientation into 3 groups with continuous-scale pseudocolor encoding on a 128x128 matrix: (1) –90° to –30° as LHF, which was distributed mainly over the subepicardial layer and was pseudocolored green; (2) –30° to 30° as CF, pseudocolored blue; and (3) 30° to 90° as RHF, which was distributed mainly over the subendocardial layer and was pseudocolored red.

The maps of trace ADC, FA, and fiber helix angle of an example slice of normal myocardium at the midventricular level are shown in Figure 2.


Figure 2177604
View larger version (36K):
[in this window]
[in a new window]

  		Figure 2. DT-MRI of normal myocardium. A, Trace ADC map; B, FA map; C, pseudocolored continuous-scale helix angle map. Arrow indicates papillary muscle. Note the homogeneity of trace ADC and FA and the symmetry of the helix angles. The red fiber represents mainly LHF; blue, mainly CF, and green, mainly RHF.

Definition, Segmentation, and Quantification of Viability Zones by CDE-MRI
High-resolution CDE-MRI was used for zonal segmentation of the viability maps (Figure 3A). Using commercial software (MASS, GE Healthcare, Milwaukee, Wis) on an offline workstation (Advantage Window 4.1, GE Healthcare), one author (K.R.C.) who was blinded to the infarct-related artery of the patients traced the endocardial and epicardial contours of the myocardium. Using the modified centerline method, the program automatically generated 100 cords perpendicular to the centerline.24 The infarct zone was defined by the distinct hyperenhanced area visually22,23 by the consensus of 2 authors (C.-P.L., M.-T.W.) without knowledge of the infarct-related artery. The number of cords encompassed in the infarct zone was then counted and divided by 100 to represent the length fraction of the infarct zone (Figure 3B). We further divided the noninfarct area into the adjacent zone and remote zone.21 In the present study, we defined the adjacent zones to be one-fourth the length of the infarct zone bilaterally. The remaining area was the remote zone. The identical zonal segmentation was applied to assess the regional LV function and DT-MRI measurements. Area percentages of the infarct, adjacent, and remote zones were calculated on the basis of the length fraction and mean wall thickness of the respective zones (see online Data Supplement, section 2).


Figure 3177604
View larger version (121K):
[in this window]
[in a new window]

  		Figure 3. CDE-MRI at mid ventricle level. A, Hyperenhanced area indicating infarct area. B, Segmentation of viability map. Epicardial and endocardial contours were manually traced, and the program automatically generated 100 cords perpendicular to the central line. The infarct zone (labeled I) was the hyperenhanced area segmented by a pair of arrows. The adjacent zones (labeled A) were defined as one-fourth the of length of the infarct zone on bilateral sides (between arrow and arrowhead). The remaining area was the remote zone (labeled R). The asterisk indicates cord 0 in C and D. C, Profile of wall thickness measured from B. D, Profile of wall motion expressed by the displacement of centerlines of each cord between end systole and end diastole on cine fast gradient echo images (not shown) at a slice corresponding to B. The infarct zone revealed thinner wall and less wall motion.

Evaluation of Regional LV Function
Evaluation of LV function was performed on the cine gradient-echo images with the modified centerline method by a semiautomated program (MASS) as described previously. One author (K.-R.C.) traced the epicardial and endocardial contours without knowledge of the infarct-related artery and viability map by CDE-MRI. Regional wall thickness and wall motion were evaluated. Wall thickness was measured by high-resolution CDE-MRI (Figure 3C). Wall motion was expressed by displacement of center points of each pair of the 100 cords between end systole and end diastole (Figure 3D).24

Ejection fraction of each midventricular volume was calculated by the relative difference in LV cavity volume between end diastole and end systole with respect to end-diastolic volume.

Zonal Segmentation of DT-MRI Measurements and Regional LV Function
After the above data analysis, the viability map by CDE-MRI was used as a reference to segment the data of DT-MRI measurement and LV function by one author (W.-Y.I.T.) without knowledge of the results of the DTI measurement and LV function.

Statistical Analysis
The clinical characteristics are expressed as median and 25th and 75th percentiles. Measurements are expressed as mean and SD. Data with nonnormal distribution was proceeded by square-root transformation28 before analysis. To account for repeated measurements from the 3 viability zones, the MANOVA model was used to identify the differences in DT-MRI variables, LV function, and percentage of infarct area, respectively, using the 3 viability zones as the within-subject factor. Post hoc pairwise comparisons of measurements between the 3 zones were performed with Bonferroni correction. Values of P<0.05 were regarded as significant.

Pearson’s correlation coefficients were computed for the fiber architecture and myocardial tissue integrity versus the regional wall motion on a viability zone basis. In addition, multiple linear regressions with the stepwise method were conducted to assess the relationship between wall motions and DT-MRI measurements for each viability zone, respectively.

Statistical analysis was performed with SPSS 11.0 (SPSS Inc, Chicago, Ill). The authors had full access to the data and take responsibility for their integrity. All authors have read and agree to the manuscript as written.


*    Results
up arrowTop
 up arrowAbstract
 up arrowIntroduction
 up arrowMethods
 *Results
down arrowDiscussion
 down arrowReferences
 
Viability Map by CDE-MRI
CDE-MRI detected the infarcted areas in all patients. The distribution of infarcted areas is shown in Table 1. The average percent of the infarct zone was 26.7±15.8% (Table 2).


View this table:
[in this window]
[in a new window]

  		TABLE 2. Zonal Differences of Myocardial Tissue Integrity, Fiber Architecture, and Regional Wall Motion

Regional LV function
There were significant differences in wall thickness and wall motion among the 3 zones, except there was no significant difference (P=0.102) in wall motion between the infarct zone and adjacent zone (Table 2). The average ejection fraction was 44.2±11.7%. A representative case is shown in Figure 3.

Myocardial Tissue Integrity and Fiber Architecture by DT-MRI
Trace ADC
Trace ADC showed significant differences among the 3 viability zones (F(2,72)=49.4; P<0.001), with a transitional increase from the remote zone to the infarct zone. Post hoc pairwise comparison showed that all the zonal differences were significant (all P<0.001) (Table 2 and Figure 4A).


Figure 4177604
View larger version (26K):
[in this window]
[in a new window]

  		Figure 4. Distribution of (A) trace ADC and (B) FA among the 3 viability zones. Each pair of dots-lines indicates 1 case. Probability values were calculated by paired t test. C, Scatterplot of trace ADC vs FA with linear regression line. Significant negative correlation was found in the infarct zone (r=–0.64, P<0.001), adjacent zone (r=–0.55, P<0.001), and remote zone (r=–0.58, P<0.001).

FA Findings
FA showed significant difference among the 3 viability zones (F(2,72)=10.5; P<0.001), with a transitional decrease from the remote zone to the infarct zone. Pairwise comparisons showed that a significant difference in FA between the adjacent and remote zones (P=0.002) and between the infarct and remote zones (P<0.001). No difference between the infarct and adjacent zones was noted (P=0.183) (Table 2 and Figure 4B).

Relation Between Trace ADC and FA
Figure 4C showed the overlaid scatterplots of trace ADC versus FA in the 3 zones. A negative correlation was found between trace ADC and FA in the infarct zone (r=–0.64, P<0.001), adjacent zone (r=–0.55, P<0.001), and remote zone (r=–0.58, P<0.001).

Fiber Architecture: Helix Angle Map
Among 3 groups of fibers, the percentage of RHF (RHF%) had the most significant difference in the 3 zones (F(2, 72)=42.3; P<0.001), followed by the percentage of LHF (LHF%) (F(2, 72)=10.1; P=0.004) and no significant difference for the percentage of CF (CF%) (F(2, 72)=2.3; P=0.114). The ratio of helical fibers to CFs, as expressed by (LHF%+RHF%)/CF%, was maintained and showed no difference among the 3 zones (F(2, 72)=1.2; P=0.304) (Table 2).

Figure 5 shows the DT-MRI analysis of the same case as in Figure 3. The infarct zone showed increased trace ADC and decreased FA compared with the adjacent zone and remote zone. The infarct zone showed substantial loss of RHF% and increase in LHF%.


Figure 5177604
View larger version (35K):
[in this window]
[in a new window]

  		Figure 5. DT-MRI of the same heart shown in Figure 3. A, Trace ADC map; B, FA map; C, continuous-scaled pseudocolored helix angle map. The infarcted area, as identified in Figure 3A, showed increased trace ADC and decreased FA. The infarct zone showed a decrease in RHF (red) and an increase in LHF (green), whereas the remote zone showed hypertrophy with an increase in RHF (red). The arrow indicates papillary muscle.

Correlation of Myocardial Tissue Integrity and Fiber Architecture With Regional Wall Motion, Infarcted Area Size, and Ejection Fraction
Table 3 shows the correlation matrix between regional wall motion versus myocardial tissue integrity and fiber architecture across the 3 viability zones. In the infarct zone, wall motion correlated negatively with trace ADC of all 3 zones (all P<0.05). It also correlated negatively with LHF% and positively with CF% in the infarct zone. With multiple linear regression, LHF% of the infarct zone was the only predictor (r=–0.33, P<0.05) of regional wall motion of the infarct zone.


View this table:
[in this window]
[in a new window]

  		TABLE 3. Correlation Matrix of Myocardial Tissue Integrity and Fiber Architecture Versus Left Ventricular Function

In the adjacent zone, wall motion correlated negatively with trace ADC of all 3 zones (all P<0.05). It also correlated negatively with LHF% and positively with CF% of the infarct zone. With multiple linear regressions, trace ADC of the adjacent zone was the only predictor (r=–0.36, P<0.05) of regional wall motion of the adjacent zone.

In the remote zone, only LHF% of the adjacent zone correlated negatively with regional wall motion of the remote zone. The infarcted area percentage correlated positively with trace ADC of all 3 zones (all P<0.05). The infarcted area percentage also correlated negatively with CF% of the infarct and adjacent zones and positively with LHF% of the infarct zone. With multiple linear regression, the LHF% of the infarct zone was the only predictor of the infarct area percentage (r=0.71, P<0.001). In addition, ejection fraction was correlated negatively with trace ADC of all 3 zones, negatively with LHF% of infarct and adjacent zones, and positively with CF% of the infarct and adjacent zones. With multiple linear regression, LHF% of the infarct zone was the only predictor (r=–0.61, P<0.001) of ejection fraction.


*    Discussion
up arrowTop
 up arrowAbstract
 up arrowIntroduction
 up arrowMethods
 up arrowResults
 *Discussion
down arrowReferences
 
Trace ADC indicates the mean diffusivity of water molecules, which reflects the redistribution of intracellular and extracellular space volumes.20 In the present study, we found that trace ADC was significantly increased in the infarct zone. This finding is consistent with the findings of DT-MRI study of the rat heart.19 In formalin-fixed rat myocardium, the infarcted area was found to have increased trace ADC compared with the noninfarct zone. Histological examination revealed that few viable myocytes remained in the infarct zone, accompanied by increased extracellular space as a result of cell death and subsequent scar formation. Therefore, the increased trace ADC in scar tissues is attributable to the increase in extracellular space as water diffusion becomes less restricted in the extracellular space.19 The association between increased water diffusion and tissue necrosis is well recognized and validated in cerebral infarction.29

FA, because it measures the variability of water mobility in different directions, is a common index for structure integrity and architectural organization of tissues.20 In the present study, we found that FA of the infarct zone was significantly decreased compared with the remote zone. This finding again is in line with the finding from the rat heart showing a negative correlation between degree of anisotropy and microscopic fiber disarray.19 The decrease in FA related to infarction may be multifactorial such as swollen myocytes, myocardial fibers that are disarrayed in orientation, and an increase in collagen concentration replacing the extracellular matrix after MI.30 In addition, we found that at 26 days after acute MI, the trace ADC and FA in each viability zone showed significant inverse correlation (Figure 4C).

We found that the proportion of helix angle distribution of the myocardial fibers was significantly altered across the 3 zones (see Data Supplement sections 3 and 4). As shown in Table 2 and Figure 6, RHF% decreased and LHF% increased in the infarct zone; inversely, RHF% increased and LHF% decreased in the remote zone. Although this particular pattern of fiber architecture remodeling has never been reported, we proposed a hypothesis involving a cascade of mechanisms. First, the area most susceptible to ischemia is the subendocardial area. Therefore, RHF had the most severe loss in the infarct zone. Second, an increase in LHF% in the infarct zone may represent a remodeling process triggered to compensate for the loss of RHF in the infarct zone and to keep the balanced ratio of (LHF%+RHF%)/CF%. Third, the RHF% increase in remote compensatory hypertrophy may be an adaptive remodeling response to the increase in wall stress.


Figure 6177604
View larger version (21K):
[in this window]
[in a new window]

  		Figure 6. Zonal distribution of the 3 types of myocardial fibers (A) and fiber distribution in the 3 viability zones (B). Side bars are mean±SD. Probability values were attained after square-root transformation of data and Bonferroni correction.

Several lines of evidence indirectly support our hypothesis. First, from computer simulation and animal model, Arts et al6 found that helical fiber orientation was a result of mechanical feedback making the fiber stress and/or strain more uniform across the wall; therefore, it would be of mechanical benefit to restore the (LHF% + RHF%)/CF% ratio during the healing process of MI. Second, it is known that wall stress is greater at the subendocardial surface than the subepicardial surface of the heart.31 Conservation of this gradient may be the reason for preferential subendocardial hypertrophy secondary to various stimulations such as heart failure or exercise training in animal heart, which is most likely an adaptation to the increased wall stress,32,33 and may be cause of remote zone hypertrophy in our patients. Third, a particular figure-of-eight pattern of myofibers was disclosed early in the 19th century3,6,34 and later confirmed by Streeter and Hanna,26,27 using the micrometric method. According to the model, the subepicardial fibers (LHF) in one wall connect histologically to the subendocardial fibers (RHF) of the opposite wall through the apex and the base. Therefore, we postulated that a decrease in RHF in the free lateral wall may correspond to a decrease in LHF in the septal wall. Our recent 3-dimensoinal tractography of DT-MRI in a normal bovine heart ex vivo showed direct visualization of this particular pattern of fiber architecture35 and provided a histological basis to support our postulation.

Study Limitations
The adjacent zone we propose is relatively conceptual, defined as one fourth the length of the infarct zone bilaterally. It is still controversial and difficult to precisely identify the "functional/histological border zone" in human living heart such as applying different experimental contrast agent to differentiate the infarct zone from the border zone.36 However, our simple definition of the adjacent zone served to support the hypothesis of transitional change of myocardium remodeling.

Although we found interesting fiber architecture remodeling after MI, we have no direct histopathological correlation to validate our findings. Study on a living human heart is impossible, however.

Conclusions
In the present study, we used DT-MRI to study living human heart after MI. We found that the alterations in tissue integrity and fiber architecture of the injured myocardium were zone dependent, which also correlated with functional impairment. Our findings were consistent with the knowledge of myocardial fiber mechanics and might provide insight into the structure mechanism of remodeling of the myocardium after MI.


*    Acknowledgments
 
We are grateful for the statistical consultation by Yi-Hsin Connie Yang, PhD, Director of Statistical Analysis Laboratory, Department of Clinical Research, Kaohsiung Medical University, Chung-Ho Memorial Hospital. We thank MR technologists Chia-Chi Hsiao, Mei-Hwa Chang, and Wen-Chuan Huang and research assistants Chin-Ying Tsai and Chih-Feng Cheng for their assistance.

Sources of Funding

This work was supported in part by the Medical Research and Advancement Foundation in memory of Dr Chi-Shuen Tsou under grants VTY 90-G3-03, VTY91-G3-03, and VTY92-G3-03 and by Kaohsiung Veterans General Hospital Research Program No. VGHKS89-127.

Disclosures

None.


*    References
up arrowTop
 up arrowAbstract
 up arrowIntroduction
 up arrowMethods
 up arrowResults
 up arrowDiscussion
 *References
 

  1. Streeter DD Jr, Spotnitz HM, Patel DP, Ross J Jr, Sonnenblick EH. Fiber orientation in the canine left ventricle during diastole and systole. Circ Res. 1969; 24: 339–347.[Abstract/Free Full Text]
  2. Armour JA, Randall WC. Structural basis for cardiac function. Am J Physiol. 1970; 218: 1517–1523.[Free Full Text]
  3. Arts T, Reneman RS, Veenstra PC. A model of the mechanics of the left ventricle. Ann Biomed Eng. 1979; 7: 299–318.[CrossRef][Medline] [Order article via Infotrieve]
  4. Waldman LK, Nosan D, Villarreal F, Covell JW. Relation between transmural deformation and local myofiber direction in canine left ventricle. Circ Res. 1988; 63: 550–562.[Abstract/Free Full Text]
  5. Weisman HF, Bush DE, Mannisi JA, Bulkley BH. Global cardiac remodeling after acute myocardial infarction: a study in the rat model. J Am Coll Cardiol. 1985; 5: 1355–1362.[Abstract]
  6. Arts T, Prinzen FW, Snoeckx LH, Rijcken JM, Reneman RS. Adaptation of cardiac structure by mechanical feedback in the environment of the cell: a model study. Biophys J. 1994; 66: 953–961.[Abstract/Free Full Text]
  7. 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.[Abstract/Free Full Text]
  8. Holmes AA, Scollan DF, Winslow RL. Direct histological validation of diffusion tensor MRI in formaldehyde-fixed myocardium. Magn Reson Med. 2000; 44: 157–161.[CrossRef][Medline] [Order article via Infotrieve]
  9. 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] [Order article via Infotrieve]
  10. 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.[Abstract/Free Full Text]
  11. Scollan DF, Holmes A, Zhang J, Winslow RL. Reconstruction of cardiac ventricular geometry and fiber orientation using magnetic resonance imaging. Ann Biomed Eng. 2000; 28: 934–944.[CrossRef][Medline] [Order article via Infotrieve]
  12. Edelman RR, Gaa J, Wedeen VJ, Loh E, Hare JM, Prasad P, Li W. In vivo measurement of water diffusion in the human heart. Magn Reson Med. 1994; 32: 423–428.[Medline] [Order article via Infotrieve]
  13. Pearlman JD, Brookeman JR. Diffusion drift: a new magnetic resonance imaging parameter. Clin Res. 1985; 33: 217A. Abstract.
  14. 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] [Order article via Infotrieve]
  15. Reese TG, Wedeen VJ, Weisskoff RM. Measuring diffusion in the presence of material strain. J Magn Reson B. 1996; 112: 253–258.[CrossRef][Medline] [Order article via Infotrieve]
  16. 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] [Order article via Infotrieve]
  17. 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.[Abstract/Free Full Text]
  18. 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] [Order article via Infotrieve]
  19. 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.[Abstract/Free Full Text]
  20. Le BD. Looking into the functional architecture of the brain with diffusion MRI. Nat Rev Neurosci. 2003; 4: 469–480.[CrossRef][Medline] [Order article via Infotrieve]
  21. 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.[Abstract/Free Full Text]
  22. Fieno DS, Kim RJ, Chen EL, Lomasney JW, Klocke FJ, Judd RM. Contrast-enhanced magnetic resonance imaging of myocardium at risk: distinction between reversible and irreversible injury throughout infarct healing. J Am Coll Cardiol. 2000; 36: 1985–1991.[Abstract/Free Full Text]
  23. 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.[Abstract/Free Full Text]
  24. Holman ER, Buller VG, de Roos A, van der Geest RJ, Baur LH, van der Laarse A, 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.[Abstract/Free Full Text]
  25. Basser PJ, Pierpaoli C. A simplified method to measure the diffusion tensor from seven MR images. Magn Reson Med. 1998; 39: 928–934.[Medline] [Order article via Infotrieve]
  26. 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.[Abstract/Free Full Text]
  27. Streeter DD Jr, Hanna WT. Engineering mechanics for successive states in canine left ventricular myocardium, II: fiber angle and sarcomere length. Circ Res. 1973; 33: 656–664.[Abstract/Free Full Text]
  28. Kleinbaum DG, Kupper LL, Muller KE, Nizam A. Regression diagnostics. In: Applied Regression Analysis and Other Multivariable Methods. 3rd ed. Pacific Grove, Calif: Brooks/Cole Publishing Co; 1998: 251–252.
  29. Pierpaoli C, Righini A, Linfante I, Tao-Cheng JH, Alger JR, Di Chiro G. Histopathologic correlates of abnormal water diffusion in cerebral ischemia: diffusion-weighted MR imaging and light and electron microscopic study. Radiology. 1993; 189: 439–448.[Abstract/Free Full Text]
  30. Whittaker P. Wound healing after myocardial infarction. In: Armonk KTW, ed. Wound Healing in Cardiovascular Disease. New York, NY: Futura; 1995: 33–48.
  31. Yin FC. Ventricular wall stress. Circ Res. 1981; 49: 829–842.[Free Full Text]
  32. 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] [Order article via Infotrieve]
  33. 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.[Abstract/Free Full Text]
  34. Sutton MG, Sharpe N. Left ventricular remodeling after myocardial infarction: pathophysiology and therapy. Circulation. 2000; 101: 2981–2988.[Free Full Text]
  35. Tseng W-Y I, Magnussen L, Gilbert R, Benner T, Wang R, Reese T, Wedeen V. Global myocardial fiber and sheet architecture in diffusion spectrum imaging tractography. Proc Int Soc Mag Reson Med. 2005; No. 1663: 334.
  36. Jeong AK, Choi SI, Kim DH, Park SB, Lee SS, Choi SH, Lim TH. Evaluation by contrast-enhanced MR imaging of the lateral border zone in reperfused myocardial infarction in a cat model. Korean J Radiol. 2001; 2: 21–27.[Medline] [Order article via Infotrieve]
 

CLINICAL PERSPECTIVE

The unique 3-dimensional organization of myocardial fibers of mammalian ventricle is composed of obliquely running fibers changing smoothly from left-handed helical fibers in the subepicardium to circumferential fibers in the mid wall to right-handed helical fibers in the subendocardium. Diffusion tensor magnetic resonance imaging (DT-MRI) provides a nondestructive means to characterize myocardial architecture. DT-MRI can measure diffusion property of the myocardium, ie, mean diffusivity and fractional anisotropy, to probe alterations in tissue integrity and can determine the fiber orientation on the basis of the principal direction of the diffusivity. In this study, we used in vivo DT-MRI to define structural correlates of functional remodeling after myocardial infarction in 37 patients. DT-MRI revealed a significant increase in mean diffusivity and a decrease in fractional anisotropy in the infarct zone, indicating altered tissue integrity. Moreover, the fiber architecture showed a progressive increase in the percentage of left-handed helical fibers from the remote to the adjacent to the infarct zone and a progressive decrease in right-handed helical fibers from the remote to the adjacent to the infarct zone. The changes in myocardial architecture correlated with infarct size and left ventricular function. This capability of DT-MRI might have potential clinical application in monitoring the evolution of myocardial architecture after infarct in response to various therapeutics, including rehabilitation, stem cell therapy, and ventriculoplasty.


*    Footnotes
 
*The first 2 authors contributed equally to this work. Back

The online-only Data Supplement is available with this article at http://circ.ahajournals.org/cgi/content/full/CIRCULATIONAHA.105.545863/DC1.




This article has been cited by other articles:


Home page
 HeartHome page
A T Burns, I G McDonald, J D Thomas, A MacIsaac, and D Prior
Doin' the twist: new tools for an old concept of myocardial function
Heart, August 1, 2008; 94(8): 978 - 983.
[Abstract] [Full Text] [PDF]

Home page
 J Am Coll Cardiol ImgHome page
N. M. Rofsky
Nephrogenic Systemic Fibrosis: Considerations for the Cardiologist
J. Am. Coll. Cardiol. Img., July 1, 2008; 1(4): 457 - 459.
[Full Text] [PDF]

Home page
 Am. J. Physiol. Heart Circ. Physiol.Home page
Y. Jiang, J. M. Guccione, M. B. Ratcliffe, and E. W. Hsu
Transmural heterogeneity of diffusion anisotropy in the sheep myocardium characterized by MR diffusion tensor imaging
Am J Physiol Heart Circ Physiol, October 1, 2007; 293(4): H2377 - H2384.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Data Supplement
Right arrow All Versions of this Article:
114/10/1036    most recent
CIRCULATIONAHA.105.545863v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Wu, M.-T.
Right arrow Articles by Yang, C.-F.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Wu, M.-T.
Right arrow Articles by Yang, C.-F.
Right arrowPubmed/NCBI databases
Medline Plus Health Information
*Heart Attack
*MRI Scans
Related Collections
Right arrow Structure
Right arrow Contractile function
Right arrow CT and MRI
Right arrow Imaging

Circulation Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search
Copyright © 2006 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited.