AJP - Heart Calcium Transients and Cell-Sarcomere
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Am J Physiol Heart Circ Physiol 289: H1898-H1907, 2005; doi:10.1152/ajpheart.00041.2005
0363-6135/05 $8.00
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
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 ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (17)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Chen, J.
Right arrow Articles by Yu, X.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Chen, J.
Right arrow Articles by Yu, X.

Regional ventricular wall thickening reflects changes in cardiac fiber and sheet structure during contraction: quantification with diffusion tensor MRI

Junjie Chen,1,2 Wei Liu,1,2 Huiying Zhang,1 Liz Lacy,1 Xiaoxia Yang,1 Sheng-Kwei Song,3 Samuel A. Wickline,1,2 and Xin Yu1,2

1Cardiovascular Magnetic Resonance Laboratories, Cardiovascular Division, Department of Medicine; 2Department of Biomedical Engineering; and 3Departments of Chemistry and Radiology, Washington University, St. Louis, Missouri

Submitted 14 January 2005 ; accepted in final form 21 June 2005


   ABSTRACT
TOP
 ABSTRACT
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 
Dynamic changes of myocardial fiber and sheet structure are key determinants of regional ventricular function. However, quantitative characterization of the contraction-related changes in fiber and sheet structure has not been reported. The objective of this study was to quantify cardiac fiber and sheet structure at selected phases of the cardiac cycle. Diffusion tensor MRI was performed on isolated, perfused Sprague-Dawley rat hearts arrested or fixed in three states as follows: 1) potassium arrested (PA), which represents end diastole; 2) barium-induced contracture with volume (BV+), which represents isovolumic contraction or early systole; and 3) barium-induced contracture without volume (BV–), which represents end systole. Myocardial fiber orientations at the base, midventricle, and apex were determined from the primary eigenvectors of the diffusion tensor. Sheet structure was determined from the secondary and tertiary eigenvectors at the same locations. We observed that the transmural distribution of the myofiber helix angle remained unchanged as contraction proceeded from PA to BV+, but endocardial and epicardial fibers became more longitudinally orientated in the BV– group. Although sheet structure exhibited significant regional variations, changes in sheet structure during myocardial contraction were relatively uniform across regions. The magnitude of the sheet angle, which is an index of local sheet slope, decreased by 23 and 44% in BV+ and BV– groups, respectively, which suggests more radial orientation of the sheet. In summary, we have shown for the first time that geometric changes in both sheet and fiber orientation provide a substantial mechanism for radial wall thickening independent of active components due to myofiber shortening. Our results provide direct evidence that sheet reorientation is a primary determinant of myocardial wall thickening.

magnetic resonance imaging; potassium arrest; barium contracture; myofiber; reorientation; myocyte shortening

THE VENTRICULAR EXCITATION-contraction process elicits transmural wall thickening that serves as the mechanism for ejection of blood. Although myocyte contraction provides the cellular basis for regional myocardial wall thickening, previous studies of dog hearts (27, 32) suggest that the typical 14% myocyte shortening leads to only an ~8% increase in myocyte diameter, which cannot fully account for the observed 28–50% increase in average wall thickness (4, 28, 36). Apex-to-base (longitudinal) shortening represents another important mechanism that has recently been recognized as a significant contributor to the ejection fraction that necessitates complex myofiber rearrangements during systole. However, the microscopic structural changes that occur during myocardial contraction remain poorly defined because of the technical difficulties in making such measurements.

The organization of myocardial fibers can be described in part by individual fiber orientations and by multiple myocyte "sheet" arrangements separated by extensive "sheet cleavage" planes. Using histological methods, several investigators documented the helical patterns of the myocardial fibers that shifted continuously from a right-handed helix at the endocardium to a left-handed helix at the epicardium (1, 23, 36). Several studies have demonstrated significant transverse shear along the sheet planes (7, 20), which suggests that the laminar organization of the myocytes may provide a structural basis for systolic wall thickening. However, the destructive nature of conventional histological analysis employed in such meticulous studies complicates direct evaluation of any changes in myocardial fiber and sheet structure that accompany cardiac contraction.

Recently, diffusion tensor MRI (DTMRI) has been validated as an alternative method for rapid and nondestructive analysis of 3-D myocardial structure in both normal (13, 14, 29) and diseased (6) hearts. This method was applied previously to depict fiber tracts in the central nervous system to detect stroke and other fiber-disrupting pathologies (18, 40, 43). The recent work of Tseng et al. (41) confirmed that quantitative measurements derived from DTMRI (e.g., the secondary and tertiary eigenvectors) are highly aligned with myocardial "fiber sheet" and "sheet-normal" directions, respectively. Accordingly, DTMRI represents an ideal candidate method for delineating changes in myocardial fiber and sheet architecture from diastole to systole.

The goal of this study was to measure changes in 3-D myofiber and sheet structure at selected phases of the cardiac cycle to elucidate alternative mechanisms of myocardial wall thickening beyond simple myocyte shortening. DTMRI of myofiber structure first was performed on isolated, perfused heart arrested at end diastole with potassium. The same heart was then induced to undergo either an isometric or isotonic contraction after BaCl2-induced cardiac contracture. The heart was rapidly fixed either in systole with volume or in systole without volume to preserve fiber and sheet structures at their contracting states for subsequent DTMRI. Our results demonstrate that the initial event of cardiac contraction comprises the reduction of sheet angles without significant changes in fiber angles, which suggests the reorientation of myocardial sheets toward a more radial direction accompanied by minimal changes in myofiber orientation. At end systole, both myofiber and fiber sheet orientations have changed substantially. These observations indicate that the geometric alterations in fiber and sheet structures represent a fundamental mechanism of circumferential shortening and regional wall thickening in systole.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 
Isolated heart model. All procedures conformed to the guidelines set forth by Washington University. Sprague-Dawley rats (2–4 mo old, 240–360 g, n = 21) were heparinized (100 U) and anesthetized with 5% isoflurane. The heart was excised and cannulated for retrograde perfusion at 100 cm hydrostatic pressure with a modified Krebs-Henseleit buffer equilibrated with 95% O2-5% CO2 (6). The left atrium was opened, and a water-filled latex balloon was inserted into the left ventricle (LV) through the mitral valve to record the left ventricular developed pressure (LVDP) and the heart rate (HR). The rate-pressure product (HR x LVDP) was calculated as an index of cardiac work. The perfused heart was placed within a sample tube in a 2-cm radio-frequency coil for MRI data collection.

Experimental protocol. Cardiac structure was characterized with DTMRI at three different states that are representative of different phases of a cardiac cycle including 1) potassium arrested (PA), which represents end diastole; 2) barium-induced contracture with volume (BV+), which represents isovolumic contraction or early systole; and 3) barium-induced contracture without volume (BV–), which represents end systole. Diastolic arrest was induced by perfusing the heart with a modified St. Thomas' cardioplegic solution that contained excessive potassium [(in mmol/l) 118 NaCl, 16 KCl, 16 MgCl2, 1.2 CaCl2, 10 NaHCO3, and 10 glucose]. LV pressure was maintained at 5–10 mmHg by adjusting the volume of the intraventricular balloon. Diffusion-weighted images of the arrested viable heart were acquired.

Once image acquisition of the potassium-arrested heart was completed, the perfusate was switched to regular Krebs-Henseleit buffer for the heart to resume normal cardiac work. After cardiac work was stabilized (~10 min), BaCl2 was introduced to achieve cardiac contracture (22). Specifically, the heart was first perfused with a modified Tyrode solution that contained (in mmol/l) 140 NaCl, 5.4 KCl, 1 MgCl2, 0.078 CaCl2, and 10 HEPES for 1.5 min to reduce the calcium content in the myocardium and was then perfused with the modified Tyrode solution that contained 2.5 mmol/l BaCl2. Both solutions contained adenosine (1 mg/min) to dilate the coronary vessels. Perfusion with BaCl2 lasted for 5 min to achieve maximal contracture. The heart was then rapidly fixed with 5% formalin.

The contracture-arrested hearts comprised two groups at different phases of ventricular contraction. In the first group (BV+; n = 10), the volume of the intraventricular balloon was kept constant during perfusion with BaCl2. Hence, LV volume change and wall thickening were small. This state was representative of the heart at isovolumic contraction or early systole. In the second group (BV–; n = 11), the balloon was replaced by a polyethylene-50 tubing inserted from the left atrium before BaCl2 perfusion. Perfusate containing BaCl2 was then introduced to the heart to induce contracture. The tubing allowed the residual fluid to be drained from the left ventricle. Significant wall thickening and volume change occurred in this group. This state was representative of the heart at end systole.

All of the solutions were equilibrated with 95% O2-5% CO2. Bovine serum albumin (BSA; 0.4% wt/wt) was also added to the perfusate to reduce interstitial edema (10, 42). The cardioplegic solution was kept at room temperature, and the rest of the solutions were maintained at 37°C.

Magnetic resonance imaging. Diffusion-weighted magnetic resonance (MR) images of both perfused (potassium-arrested) and fixed (barium-arrested) hearts were acquired on a 4.7-T Varian INOVA system (Varian Associates; Palo Alto, CA) using a custom-built solenoid coil. Long-axis scout images were acquired as previously described (6). A multislice spin-echo sequence with diffusion-sensitizing bipolar gradient was used to acquire short-axis, diffusion-weighted images. Diffusion-sensitizing gradients were applied in six noncollinear directions (6). Imaging parameters were as follows: echo time, 36 ms; time interval between gradient pulses, 20 ms; gradient pulse duration, 6 ms; gradient factor, 948 s/mm2; and in-plane resolution, 156 x 156 µm. Seven slices that covered the heart from base to apex were acquired from the potassium-arrested heart. The slice thickness was 1 mm, and gaps between adjacent slices ranged from 0.3 to 0.5 mm. For barium-arrested and fixed hearts, 11 contiguous slices were acquired. The slice thickness (0.8–0.9 mm) was adjusted according to the longitudinal shortening that occurred during barium-induced contracture. Repetition time for potassium-arrested heart was 1.3 s, and total image-acquisition time was ~1 h. Repetition time for barium-arrested heart was 1.7 s, and total image-acquisition time was ~2 h.

Data analysis. The effective diffusion tensor (Deff) was calculated from diffusion-weighted images as described previously (6). Subsequently, the primary, secondary, and tertiary eigenvalues ({lambda}1, {lambda}2, and {lambda}3, respectively) of the diffusion tensor were calculated. Diffusion anisotropy was analyzed with fractional anisotropy (FA; 3). The primary eigenvector of the diffusion tensor is considered to be the myofiber orientation (11, 13, 14, 29), and the secondary and tertiary eigenvectors are considered to be the sheet and sheet-normal orientations, respectively (41).

Epicardial and endocardial borders of the heart were traced manually on both short- and long-axis images (6). LV wall thickness was calculated as the mean distance between epicardial and endocardial borders. Interstitial edema was assessed by measuring changes in wall thickness before and after DTMRI of potassium-arrested hearts.

The local wall-bound myocardial coordinates were used to quantify myofiber structure (34). The LV long axis was determined as the line that best fit the centers of the epicardial borders. A prolate spheroid was fit to the epicardial borders as the epicardial surface. Subsequently, the three principal axes of the wall-bound coordinates, i.e., the radial, circumferential, and longitudinal axes, were determined. Specifically, the radial axis was defined as normal to the local epicardial surface. The circumferential axis was tangent to the local epicardial surface and perpendicular to the LV long axis. The longitudinal axis was perpendicular to both the radial and circumferential axes.

The DTMRI-determined myofiber helix angle ({alpha}h), myofiber transverse angle ({alpha}t), and sheet angle ({beta}s) were calculated to quantitatively describe the cardiac structure in the local wall-bound coordinates (7, 29). The {alpha}h was defined as the angle between the circumferential axis and the projection of the myofiber onto the circumferential-longitudinal plane; {alpha}h for a right-handed helix at the subendocardial region was set as positive. The {alpha}t was defined as the angle between the circumferential axis and the projection of myofiber orientation onto the radial-circumferential plane. The {beta}s was defined as the angle between the radial axis and the secondary eigenvector (7); {beta}s for a sheet orientated toward the base from endocardium to epicardium was set as positive. The {alpha}h, {alpha}t, and {beta}s were characterized in basal, midventricular, and apical slices located at 25, 50, and 75% of the distance from the mitral valve to the apex.

For each slice, data were analyzed in four 20°-wide sectors at anterior, lateral (between the papillary muscles), inferior, and septal regions, respectively (Fig. 1A). The through-wall difference of {alpha}h, defined as the difference in endocardial and epicardial helix angles, i.e., {Delta}{alpha}h = {alpha}h(endocardial) – {alpha}h(epicardial), was used to quantify transmural changes of fiber orientation. The magnitude of the sheet angle, {beta}s, was used as a measure of sheet slope from base to apex. Finally, changes of the 3-D fiber and sheet structure in a cardiac cycle were reconstructed and visualized using measured {alpha}h, {alpha}t, and {beta}s values from the PA, BV+, and BV– groups.



View larger version (52K):
[in this window]
[in a new window]
  		Fig. 1. A: selected left ventricular (LV) sectors (shaded areas) for data analysis. Anterior and inferior fusion sites of the right ventricle (RVA and RVI, respectively) are shown. B and C: helix-angle ({alpha}h) maps of a basal slice in potassium-arrested (PA; B) and barium-induced contracture without volume (BV–; C) hearts, which represent end diastole and end systole, respectively. D and E: sheet-angle ({beta}s) maps of the same basal slice in PA (D) and BV– (E) hearts.

 
Histological analysis. Three additional hearts were arrested and fixed in diastole for histological evaluation of myocardial structure. All hearts were sliced along the LV long axis that bisected both the right and left ventricles. Each slice was embedded in paraffin and sectioned at 4 µm. The tissue sections were stained with hematoxylin and eosin to evaluate the transmural sheet structure. In each section, two lateral regions corresponding to the size and location of basal and apical MR slices were selected. The orientation of the sheet cleavage plane in each region was automatically measured using an intensity-gradient technique proposed by Karlon et al. (17). Subsequently, sheet cleavage angle ({beta}s) was determined as the angle between the sheet cleavage plane and the direction normal to the local epicardial border.

Statistical analysis. All data are expressed as means ± SD. The {alpha}h and {beta}s values were characterized from 5% (endocardial surface) to 95% (epicardial surface) of transmural depth at 10% increments. Paired Student's t-tests were used to compare changes in fiber and sheet angles between PA and BV+ groups as well as between PA and BV– groups. A two-tailed value of P < 0.05 was considered significant. Two-way ANOVA and subsequent Freeman-Tukey tests were performed for comparisons of {Delta}{alpha}h and {beta}s at the base, midventricle, and apex among PA, BV+, and BV– groups. A value of P < 0.05 was considered significant.


   RESULTS
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 GRANTS
 REFERENCES
 
Myocardial function and morphology. Cardiac function of the isolated, perfused heart was preserved during the experiment. Rate-pressure product values before and after potassium arrest were 43,711 ± 10,890 and 41,368 ± 10,496 beats/min x mmHg, respectively [P = not significant (NS)]. LVDP in barium-perfused hearts was 80 ± 15 mmHg (BV+ group), whereas LVDP before barium perfusion was 146 ± 25 mmHg.

Average LV wall thickness at midventricle in the PA group was the same before and after DTMRI of potassium-arrested hearts (2.3 ± 0.2 mm), which suggests that interstitial edema was not significant. No significant wall thickening was observed in the BV+ group (2.7 ± 0.4 mm; P = NS compared with PA). Wall thickness increased by 46% in the BV– group (3.5 ± 0.1 mm; P < 0.001 compared with PA). Longitudinal shortening occurred in both the BV+ and BV– groups. LV length (in cm) was 1.49 ± 0.04 in the PA group, 1.38 ± 0.08 in the BV+ group (P < 0.001 compared with PA), and 1.28 ± 0.03 in the BV– group (P < 0.001 compared with PA).

Diffusion characteristics. FA in all three groups was well above that of the surrounding media (0.07 ± 0.02), which indicates significant diffusion anisotropy in all three stages of ventricular contraction. FA was 0.36 ± 0.03 and 0.32 ± 0.05 in the PA and BV+ groups, respectively, and was slightly decreased in the BV– group (0.30 ± 0.04; P < 0.001 compared with PA). The {lambda}1/{lambda}2 values were 1.56 ± 0.07, 1.50 ± 0.10, and 1.47 ± 0.10 in the PA, BV+, and BV– groups, respectively. The {lambda}2/{lambda}3 values were 1.40 ± 0.10, 1.33 ± 0.16, and 1.30 ± 0.06 in the PA, BV+, and BV– groups, respectively.

Myocardial fiber orientation. Representative helix angle ({alpha}h) maps of a rat heart in diastole and systole are shown in Fig. 1, B and C, respectively. Transmural courses of {alpha}h were quantified on the whole short-axis slice at the base, midventricle, and apex to characterize changes of myofiber orientation from diastole (PA) to early systole (BV+; Fig. 2A) and from diastole (PA) to end systole (BV–; Fig. 2B). The transmural distribution of {alpha}h in BV+ hearts was similar to that in PA hearts with only a slight decrease at the epicardial region. In the BV– group, {alpha}h increased from the endocardial to midwall region (5–65% transmural depth; P < 0.05) and decreased at the subepicardial region (85–95% transmural depth; P < 0.05). The lateral region demonstrated the greatest changes in fiber angles. At the midventricular level, lateral {alpha}h changed from 47 ± 9° (endocardium) to –50 ± 10° (epicardium) in the PA group, from 48 ± 9° (endocardium) to –60 ± 7° (epicardium) in the BV+ group, and from 65 ± 10° (endocardium) to –66 ± 7° (epicardium) in the BV– group. Changes in {alpha}h at the anterior, inferior, and septal regions were less pronounced but similar to the changes in the lateral region.



View larger version (21K):
[in this window]
[in a new window]
  		Fig. 2. Transmural distribution of myofiber helix angle ({alpha}h) in PA, BV+, and BV– groups. Changes in {alpha}h were evaluated at base, midventricle (Mid), and apex. A: transmural distribution of {alpha}h in PA and BV+ groups. B: transmural distribution of {alpha}h in PA and BV– groups. TD, transmural depth; Endo, endocardial surface (TD = 0%); Epi, epicardial surface (TD = 100%). *P < 0.05; {dagger}P < 0.001 compared with PA group.

 
The through-wall difference of the helix angle, {Delta}{alpha}h, increased as a result of the changes in {alpha}h during cardiac contraction (Fig. 3). Maximal increase in {Delta}{alpha}h was observed in the lateral region. At the midventricular level, {Delta}{alpha}h was 96 ± 10° in the PA group, 108 ± 8° in the BV+ group (P < 0.05 compared with PA), and 131 ± 10° in the BV– group (P < 0.001 compared with PA). When averaged over the whole slice, {Delta}{alpha}h values at the base, midventricle, and apex were 102 ± 10, 105 ± 6, and 102 ± 7°, respectively, in PA hearts; 109 ± 7, 111 ± 8, and 101 ± 12°, respectively (P < 0.05 at midventricle), in BV+ hearts; and 121 ± 6, 134 ± 4, and 113 ± 6°, respectively (P < 0.001 at all three levels), in BV– hearts.



View larger version (29K):
[in this window]
[in a new window]
  		Fig. 3. Through-wall difference of myofiber helix angle ({Delta}{alpha}h) in PA (open bars), BV+ (gray bars), and BV– (solid bars) groups. A: lateral. B: septum. C: anterior. D: inferior. *P < 0.05; {dagger}P < 0.001 for intergroup comparisons.

 
The transverse angle, {alpha}t, ranged between –20 and 20°. No significant differences in {alpha}t were observed for PA, BV+, and BV– hearts.

Laminar fiber sheet orientation. Representative color maps of the sheet angle, {beta}s, in diastole and systole are shown in Fig. 1, D and E, respectively. The hematoxylin and eosin staining of potassium- and barium-arrested (without volume) hearts are shown in Fig. 4, A and B, respectively, where sheet structure can be identified as fine separations of myofiber bundles that transit from positive sheet angles at the base to negative sheet angles at the apex. DTMRI-determined sheet structures (lateral region) in PA, BV+, and BV– hearts were also reconstructed (Fig. 4). Qualitative agreement was evident between DTMRI-measured sheet structure and that revealed by histological staining, which is similar to the corroboration provided in a recent study by Tseng et al. (41). Visual inspection of the reconstructed sheets suggested that reorientation of the sheet toward the radial direction occurred in both the BV+ and BV– groups.



View larger version (66K):
[in this window]
[in a new window]
  		Fig. 4. A and B: hematoxylin and eosin-stained, long-axis slices of rat hearts fixed at diastole (A) and systole (B), respectively. C: quantification of diastolic sheet angle of the boxed areas in A. D: quantification of systolic sheet angle of the boxed areas in B. Measured orientation of sheet cleavage plane was overlaid. E: reconstructed sheet structure from diffusion tensor MRI (DTMRI) data at base, midventricle, and apex. Sheet surfaces are color coded with local myofiber helix angle.

 
Transmural courses of {beta}s are shown in Fig. 5. In accordance with histology, {beta}s changed from predominantly positive at the base to predominantly negative at the apex. The magnitude of {beta}s (|{beta}s|), which is an index of local sheet slope, showed a trend to decrease from diastole (PA hearts) to early systole (BV+ hearts), and was significantly decreased by end systole (BV– hearts). Quantitative analysis of DTMRI data revealed 23 and 44% decreases in |{beta}s| in all sectors in BV+ and BV– hearts, respectively (Fig. 6). The |{beta}s| values measured from histological analysis exhibited similar changes, as follows: from 34 ± 4° in diastole to 18 ± 7° in systole (P < 0.05) at the base and from 35 ± 6° in diastole to 21 ± 3° in systole (P < 0.05) at the apex. These data provide direct evidence of reorientation of the sheet toward the radial direction during cardiac contraction.



View larger version (24K):
[in this window]
[in a new window]
  		Fig. 5. Transmural distribution of lateral sheet angle ({beta}s) at base, midventricle, and apex. A: from PA to BV+. B: from PA to BV–. *P < 0.05; {dagger}P < 0.001 compared with PA group.

 


View larger version (28K):
[in this window]
[in a new window]
  		Fig. 6. Average magnitude of the sheet angle (|{beta}s|) across the wall at base, midventricle, and apex in PA (open bars), BV+ (gray bars), and BV– (solid bars) groups. A: lateral. B: septum. C: anterior. D: inferior. *P < 0.05; {dagger}P < 0.001 for intergroup comparisons.

 

   DISCUSSION
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
GRANTS
 REFERENCES
 
The present study was designed to quantify the changes of cardiac fiber and sheet structure from diastole to systole with respect to their independent contributions to local wall thickening. Despite previous detailed investigations of the fiber and laminar structure of the left ventricle, quantitative descriptions of fiber structural changes during ventricular contraction are lacking. Early studies that compared diastolic and systolic myocyte orientations were conducted by Streeter et al. (36) with the use of histological methods. Because of large intersubject variations, histological analyses of hearts fixed separately in diastole and systole have not succeeded in demonstrating statistically significant changes in myofiber angles with contraction. However, calculations based on diastolic fiber structure and ventricular geometry in both diastole and systole have predicted a small increase in the endocardial fiber angles in systole (35). Takayama et al. (39) have also calculated a small decrease in fiber angle (<5°) with increased end-diastolic pressure based on measured myocardial deformation.

Compared with histology, DTMRI allows direct quantification of structural changes during cardiac contraction in the same heart. In a recent in vivo study on a normal human subject, Dou et al. (8) observed the broadening of the histogram of myofiber helix angles at end systole, which suggests that fibers become more longitudinally oriented in systole. However, spatial changes in the helix angle could not be evaluated because of low spatial imaging resolution. The present study provides the first quantitative measurements of the myofiber and sheet structural changes during active cardiac contraction. These observations both confirm and quantify the alteration in myofiber helix angles at end systole.

We deduce from these data that the contraction-induced changes in myofiber orientation per se will not contribute directly to the wall thickening for the following reason. The cross-section of an obliquely oriented myocyte is typically an ellipse on the short-axis plane, and the diameter of the ellipse is determined by the diameter of the myocyte and the helix angle of the myofiber. Changes in myofiber orientation alone do not change the diameter of such ellipses in the radial direction, which is the direction of wall thickening.

In contrast, circumferential shortening could result in part from systolic changes in fiber orientation. The projection of the myocyte diameter in the circumferential direction can be significantly reduced when the myocytes become more longitudinally orientated (Fig. 7). A simple calculation suggests that even with the 8% increase in myocyte diameter by end systole, the observed increase in helix angle can lead to up to 12% decrease in circumferential diameter at 30% transmural depth. Therefore, changes in fiber orientation may provide a significant mechanism for circumferential shortening.



View larger version (12K):
[in this window]
[in a new window]
  		Fig. 7. Contribution of myofiber reorientation to circumferential shortening during cardiac contraction. Value of dc, the cross-sectional length of myocyte in the circumferential direction (Xc), is determined by the myocyte diameter (d) and the helix angle ({alpha}h). Myofiber becomes more longitudinally oriented (increased {alpha}h) from diastole to systole, which leads to a decrease in dc.

 
The mechanical role of the laminar sheet structures of heart with respect to myocardial wall thickening was investigated in a series of studies by analyzing diastolic sheet structure and systolic wall strain. LeGrice et al. (20) previously reported that for the inner third of the myocardium, maximum sheet sliding was coplanar with the orientation of local myocardial sheets. Costa et al. (7) also reported large sheet-normal shear strain in the inner half of the myocardial wall. In addition, they also observed negative sheet-normal strain and positive sheet strain, indicating sheet thinning and extension. Furthermore, it was demonstrated that both sheet extension and sheet-normal shear contributed significantly to the radial wall thickening. Based on these observations, it was proposed that the shearing of adjacent muscle layers along the cleavage planes between them might provide a mechanism of systolic wall thickening. However, direct measurements of the changes of sheet structure have not been reported.

Dou et al. (9) recently combined diffusion and strain rate measurements in an MRI study involving human subjects. By expressing the strain tensor in the local fiber and sheet coordinates, the fractional contribution of each strain component to the overall radial thickening was calculated. It was demonstrated that a major contribution to radial thickening was associated with sheet-related strains, but that the three components of fiber-related strains contributed less to radial thickening. Their observation of sheet-normal thickening (positive sheet-normal strain) deviates from Costa's observation of sheet thinning. Nevertheless, the significant contribution from sheet-normal shear strain provides additional support that transverse sheet shear contributes to systolic wall thickening. However, quantitative evaluation of the geometric changes in the sheet structure induced by sheet-shear strain could not be evaluated because of low imaging resolution.

The observed sheet-normal shear will lead to the reorientation of myocardial sheets toward the radial direction during systole (7). An earlier study by Spotnitz et al. (33) also indicated that an increase in wall thickness was associated with a decreased sheet slope and an increased number of fibers across the wall, rather than with an increase in fiber size. The progressive decrease of the sheet slope from diastole to systole observed in the present study for the first time provides direct evidence of the changes in sheet orientation during ventricular contraction. These observed changes in sheet angle were uniform in the whole left ventricle despite the considerable regional variations in sheet structure. On average, sheet angle changed from 36° at end diastole to 20° at end systole. Thus sheet reorientation alone may increase wall thickness by 16%. Accordingly, for an average wall thickening of 40%, reorientation of the sheet toward the radial direction may contribute up to 40% of the overall wall thickening. This observation is in agreement with the estimation from strain measurements provided by Costa et al. (7). However, our present study cannot quantify the contribution from sheet extension, i.e., positive sheet strain measured by Costa et al. (7). Given the small increase in myofiber diameter and the fact that changes in fiber orientation do not directly contribute to radial wall thickening, it is very likely that sheet extension is another important mechanism for wall thickening.

Longitudinal shortening is another important component of ventricular contraction. It is directly associated with the contraction of longitudinally and obliquely oriented fibers. Previous studies have suggested that longitudinal shortening may occur before radial shortening, so that the LV cavity initially becomes more spherical (15). Longitudinal shortening has been observed during the isovolumic phase of contraction and can reach 10–12% at end systole in dog heart (12). In the present study, the measured longitudinal shortening in contracture-arrested hearts was 14%, which is similar to that observed in vivo. Interestingly, longitudinal shortening and reorientation of the sheet toward the radial direction occurred progressively from PA to BV+ to BV–, which suggests that sheet reorientation may be directly related to the longitudinal contraction of myocardial fibers.

The parallel relationship between the axes of myofiber sheet structure and the eigenvectors of the diffusion tensor is supported by the results of previous studies and the present study. The work of Scollan et al. (30) showed a qualitative agreement between the helix angle of the sheet normal and that of the tertiary eigenvector. Recently, Tseng et al. (41) validated this relationship in a quantitative study. In the present study, DTMRI-determined changes in sheet angle from diastole (PA) to systole (BV–) were comparable to those measured from histology. Our data provide additional evidence that DTMRI can be used as a nondestructive method to characterize alterations in myocardial fiber and sheet structure.

Several factors may have contributed to the large standard deviation of the DTMRI-determined sheet angle. First, the sheet structure of the myocardium demonstrates significant heterogeneity. Our present imaging resolution was not sufficient to depict the finer structure of myocardial laminae. The slice thickness of the MR images ranged from 0.8 to 1.0 mm, whereas the average thickness of each individual sheet is ~50 µm (19). Thus the measured sheet angle in each image voxel represents the average value from ~20 sheets. Second, sheet branching was observed in the wall (2, 19), which will contribute to the measurement error for the sheet angle. Finally, although the three eigenvalues are statistically distinct from one another, the measured {lambda}2/{lambda}3 value was ~17% lower than {lambda}1/{lambda}2. Thus somewhat higher sorting errors for the secondary (sheet direction) and tertiary (sheet-normal direction) eigenvectors are anticipated.

BaCl2 was used to induce myocardial contracture. Although barium is paramagnetic, the work of Litt and Brody (21) suggests that a low concentration of barium (<25 mM) in agarose gel does not substantially alter the relaxation time of the gel. Nevertheless, this approach has several restrictions. First, the heart must be fixed in the systolic phases to avoid degradation of barium-induced contracture during image acquisition (31). Although formalin fixation did not substantially change fiber structure in physiological tissues (13, 29, 37), the irreversible fixation process prevented us from analyzing fiber and sheet structure at multiple phases of ventricular contraction from the same heart. Second, a small amount of perfusate can be observed between the balloon and the endocardial surface from MR images, which indicates that the intraventricular balloon did not occupy the whole LV cavity. As a result, the wall thickness was slightly increased in the BV+ group compared with the PA group. However, these changes were very small, and the BV group still should be directly informative of events that accompany isovolumic contraction.

Diffusion-encoding gradients were applied only in six noncollinear directions to minimize image-acquisition time. Although the accuracy of DTMRI-determined fiber architecture could be improved by increasing the directions of the applied diffusion-encoding gradients and using the optimized diffusion-encoding scheme (16), the present diffusion-encoding scheme, which required minimal image-acquisition time, was optimized to maximally preserve myocardial contractility of the isolated perfused heart. The DTMRI-determined myofiber and sheet orientations were consistent in each group of hearts.

Direct evaluation of myocardial fiber and sheet structure may have important clinical implications. Present clinical approaches to evaluate regional ventricular function are based mainly on wall thickening. However, recent studies indicate that developed ventricular wall stress and strain are very sensitive to changes in fiber and sheet structure over the cardiac cycle (5, 24–26, 38). Thus it is conceivable that evaluation of preclinical abnormalities of ventricular function can be improved by incorporating quantitative data on fiber and sheet structure in systole (44). Furthermore, characterization of such structural changes in diseased hearts may facilitate the investigation of the mechanisms of structural and functional adaptations in the ventricular remodeling process.

In summary, we have shown for the first time that geometric changes in both fiber and sheet orientations provide a substantial mechanism for radial wall thickening independent of active components due to myofiber shortening. In other words, myocyte contraction contributes to radial wall thickening and ventricular ejection both by myocyte shortening and by the related secondary induction of changes in fiber and sheet organization. This mechanism depends critically on the interaction of the myocytes and the extracellular matrix throughout the ventricular wall during myocardial contraction. The complexity of this mechanism of wall thickening suggests that abnormalities in either the contractile apparatus itself (myocyte) or the infrastructure (extracellular matrix and cellular syncytium) can dramatically affect wall thickening, and as such, both require concordant assessment for comprehensive and accurate analysis of contractile abnormalities.


   GRANTS
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
REFERENCES
 
The authors acknowledge the support of the Washington University Small Animal Imaging Resource, which is funded in part through National Cancer Institute Small Animal Imaging Research Program Grant R24 CA-83060. This work was also supported by National Institutes of Health Grants R01 HL-73315 (to X. Yu) and R01 HL-42950 (to S. A. Wickline). J. Chen was supported by Predoctoral Fellowship 0315249Z from the Heartland Affiliate of the American Heart Association.


   ACKNOWLEDGMENTS
 
The authors wish to thank Dr. Joseph J. H. Ackerman, Director of the Biomedical MR Laboratory at Washington University, for advice on DTMRI methods, and Ms. Jian Wang for assistance with statistical analysis.

Present address of Xin Yu: Department of Biomedical Engineering, Case Western Reserve University, Cleveland, Ohio.


   FOOTNOTES
 

Address for reprint requests and other correspondence: Xin Yu, Dept. of Biomedical Engineering, Case Western Reserve Univ., Wickenden 319, 10900 Euclid Ave., Cleveland, OH 44106 (E-mail: Xin.Yu{at}case.edu)

The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.


   REFERENCES
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 GRANTS
 REFERENCES
 

  1. Armour JA and Randall WC. Structural basis for cardiac function. Am J Physiol 218: 1517–1523, 1970.[Free Full Text]
  2. Arts T, Costa KD, Covell JW, and McCulloch AD. Relating myocardial laminar architecture to shear strain and muscle fiber orientation. Am J Physiol Heart Circ Physiol 280: H2222–H2229, 2001.[Abstract/Free Full Text]
  3. Basser PJ and Pierpaoli C. Microstructural and physiological features of tissues elucidated by quantitative-diffusion-tensor MRI. J Magn Reson 111: 209–219, 1996.[CrossRef][ISI][Medline]
  4. Bogaert J and Rademakers FE. Regional nonuniformity of normal adult human left ventricle. Am J Physiol Heart Circ Physiol 280: H610–H620, 2001.[Abstract/Free Full Text]
  5. Bovendeerd PH, Arts T, Huyghe JM, van Campen DH, and Reneman RS. Dependence of local left ventricular wall mechanics on myocardial fiber orientation: a model study. J Biomech 25: 1129–1140, 1992.[CrossRef][ISI][Medline]
  6. Chen J, Song SK, Liu W, McLean M, Allen JS, Tan J, Wickline SA, and Yu X. Remodeling of cardiac fiber structure after infarction in rats quantified with diffusion tensor MRI. Am J Physiol Heart Circ Physiol 285: H946–H954, 2003.[Abstract/Free Full Text]
  7. Costa KD, Takayama Y, McCulloch AD, and Covell JW. Laminar fiber architecture and three-dimensional systolic mechanics in canine ventricular myocardium. Am J Physiol Heart Circ Physiol 276: H595–H607, 1999.[Abstract/Free Full Text]
  8. Dou J, Reese TG, Tseng WY, and Wedeen VJ. Cardiac diffusion MRI without motion effects. Magn Reson Med 48: 105–114, 2002.[CrossRef][ISI][Medline]
  9. Dou J, Tseng WY, Reese TG, and Wedeen VJ. Combined diffusion and strain MRI reveals structure and function of human myocardial laminar sheets in vivo. Magn Reson Med 50: 107–113, 2003.[CrossRef][ISI][Medline]
  10. Dunphy G, Richter HW, Azodi M, Weigand J, Sadri F, Sellke F, and Ely D. The effects of mannitol, albumin, and cardioplegia enhancers on 24-h rat heart preservation. Am J Physiol Heart Circ Physiol 276: H1591–H1598, 1999.[Abstract/Free Full Text]
  11. Geerts L, Bovendeerd P, Nicolay K, and Arts T. Characterization of the normal cardiac myofiber field in goat measured with MR-diffusion tensor imaging. Am J Physiol Heart Circ Physiol 283: H139–H145, 2002.[Abstract/Free Full Text]
  12. Hawthorne E. Instantaneous dimensional changes of the left ventricle in dogs. Circ Res 9: 110–119, 1961.[Abstract/Free Full Text]
  13. Holmes AA, Scollan DF, and Winslow RL. Direct histological validation of diffusion tensor MRI in formaldehyde-fixed myocardium. Magn Reson Med 44: 157–161, 2000.[CrossRef][ISI][Medline]
  14. Hsu EW, Muzikant AL, Matulevicius SA, Penland RC, and Henriquez CS. Magnetic resonance myocardial fiber-orientation mapping with direct histological correlation. Am J Physiol Heart Circ Physiol 274: H1627–H1634, 1998.[Abstract/Free Full Text]
  15. Jones CJ, Raposo L, and Gibson DG. Functional importance of the long axis dynamics of the human left ventricle. Br Heart J 63: 215–220, 1990.[Abstract/Free Full Text]
  16. Jones DK, Horsfield MA, and Simmons A. Optimal strategies for measuring diffusion in anisotropic systems by magnetic resonance imaging. Magn Reson Med 42: 515–525, 1999.[CrossRef][ISI][Medline]
  17. Karlon WJ, Covell JW, McCulloch AD, Hunter JJ, and Omens JH. Automated measurement of myofiber disarray in transgenic mice with ventricular expression of ras. Anat Rec 252: 612–625, 1998.[CrossRef][Medline]
  18. Kealey SM, Kim Y, and Provenzale JM. Redefinition of multiple sclerosis plaque size using diffusion tensor MRI. Am J Roentgenol Radium Ther 183: 497–503, 2004.
  19. LeGrice IJ, Smaill BH, Chai LZ, Edgar SG, Gavin JB, and Hunter PJ. Laminar structure of the heart: ventricular myocyte arrangement and connective tissue architecture in the dog. Am J Physiol Heart Circ Physiol 269: H571–H582, 1995.[Abstract/Free Full Text]
  20. LeGrice IJ, Takayama Y, and Covell JW. Transverse shear along myocardial cleavage planes provides a mechanism for normal systolic wall thickening. Circ Res 77: 182–193, 1995.[Abstract/Free Full Text]
  21. Litt HI and Brody AS. BaSO4-loaded agarose: a construction material for multimodality imaging phantoms. Acad Radiol 8: 377–383, 2001.[CrossRef][ISI][Medline]
  22. Munch DF, Comer HT, and Downey JM. Barium contracture: a model for systole. Am J Physiol Heart Circ Physiol 239: H438–H442, 1980.[Abstract/Free Full Text]
  23. Nielsen PM, Le Grice IJ, Smaill BH, and Hunter PJ. Mathematical model of geometry and fibrous structure of the heart. Am J Physiol Heart Circ Physiol 260: H1365–H1378, 1991.[Abstract/Free Full Text]
  24. Rijcken J, Arts T, Bovendeerd P, Schoofs B, and van Campen D. Optimization of left ventricular fibre orientation of the normal heart for homogeneous sarcomere length during ejection. Eur J Morphol 34: 39–46, 1996.[CrossRef][ISI][Medline]
  25. Rijcken J, Bovendeerd PH, Schoofs AJ, van Campen DH, and Arts T. Optimization of cardiac fiber orientation for homogeneous fiber strain at beginning of ejection. J Biomech 30: 1041–1049, 1997.[CrossRef][ISI][Medline]
  26. Rijcken J, Bovendeerd PH, Schoofs AJ, van Campen DH, and Arts T. Optimization of cardiac fiber orientation for homogeneous fiber strain during ejection. Ann Biomed Eng 27: 289–297, 1999.[CrossRef][ISI][Medline]
  27. Rodriguez EK, Hunter WC, Royce MJ, Leppo MK, Douglas AS, and Weisman HF. A method to reconstruct myocardial sarcomere lengths and orientations at transmural sites in beating canine hearts. Am J Physiol Heart Circ Physiol 263: H293–H306, 1992.[Abstract/Free Full Text]
  28. Ross J Jr, Sonnenblick EH, Covell JW, Kaiser G, and Spiro D. The architecture of the heart in systole and diastole. Technique of rapid fixation and analysis of left ventricular geometry. Circ Res 21: 409–421, 1967.[Abstract/Free Full Text]
  29. Scollan DF, Holmes A, Winslow R, and Forder J. Histological validation of myocardial microstructure obtained from diffusion tensor magnetic resonance imaging. Am J Physiol Heart Circ Physiol 275: H2308–H2318, 1998.[Abstract/Free Full Text]
  30. Scollan DF, Holmes A, Zhang J, and Winslow RL. Reconstruction of cardiac ventricular geometry and fiber orientation using magnetic resonance imaging. Ann Biomed Eng 28: 934–944, 2000.[CrossRef][ISI][Medline]
  31. Shibata T, Berman MR, Hunter WC, and Jacobus WE. Metabolic and functional consequences of barium-induced contracture in rabbit myocardium. Am J Physiol Heart Circ Physiol 259: H1566–H1574, 1990.[Abstract/Free Full Text]
  32. Sonnenblick EH, Ross J Jr, Covell JW, Spotnitz HM, and Spiro D. The ultrastructure of the heart in systole and diastole. Changes in sarcomere length. Circ Res 21: 423–431, 1967.[Abstract/Free Full Text]
  33. Spotnitz HM, Spotnitz WD, Cottrell TS, Spiro D, and Sonnenblick EH. Cellular basis for volume related wall thickness changes in the rat left ventricle. J Mol Cell Cardiol 6: 317–331, 1974.[CrossRef][ISI][Medline]
  34. Streeter DD Jr. Gross morphology and fiber geometry of the heart. In: Handbook of Physiology. The Cardiovascular System. The Heart. Bethesda, MD: Am. Physiol. Soc., 1979, sect. 2, vol. I, chapt. 4, p. 61–112.
  35. Streeter DD Jr and Hanna WT. Engineering mechanics for successive states in canine left ventricular myocardium. II. Fiber angle and sarcomere length. Circ Res 33: 656–664, 1973.[Abstract/Free Full Text]
  36. Streeter DD Jr, Spotnitz HM, Patel DP, Ross J Jr, and Sonnenblick EH. Fiber orientation in the canine left ventricle during diastole and systole. Circ Res 24: 339–347, 1969.[Abstract/Free Full Text]
  37. Sun SW, Neil JJ, and Song SK. Relative indices of water diffusion anisotropy are equivalent in live and formalin-fixed mouse brains. Magn Reson Med 50: 743–748, 2003.[CrossRef][ISI][Medline]
  38. Taber LA, Yang M, and Podszus WW. Mechanics of ventricular torsion. J Biomech 29: 745–752, 1996.[CrossRef][ISI][Medline]
  39. Takayama Y, Costa KD, and Covell JW. Contribution of laminar myofiber architecture to load-dependent changes in mechanics of LV myocardium. Am J Physiol Heart Circ Physiol 282: H1510–H1520, 2002.[Abstract/Free Full Text]
  40. Thomalla G, Glauche V, Koch MA, Beaulieu C, Weiller C, and Rother J. Diffusion tensor imaging detects early Wallerian degeneration of the pyramidal tract after ischemic stroke. Neuroimage 22: 1767–1774, 2004.[CrossRef][ISI][Medline]
  41. Tseng WY, Wedeen VJ, Reese TG, Smith RN, and Halpern EF. Diffusion tensor MRI of myocardial fibers and sheets: correspondence with visible cut-face texture. J Magn Reson Imaging 17: 31–42, 2003.[CrossRef][ISI][Medline]
  42. Watts JA and Maiorano PC. Trace amounts of albumin protect against ischemia and reperfusion injury in isolated rat hearts. J Mol Cell Cardiol 31: 1653–1662, 1999.[CrossRef][ISI][Medline]
  43. Werring DJ, Toosy AT, Clark CA, Parker GJ, Barker GJ, Miller DH, and Thompson AJ. Diffusion tensor imaging can detect and quantify corticospinal tract degeneration after stroke. J Neurol Neurosurg Psychiatry 69: 269–272, 2000.[Abstract/Free Full Text]
  44. Yin FC. Ventricular wall stress. Circ Res 49: 829–842, 1981.[Free Full Text]



This article has been cited by other articles:


Home page
 Am. J. Physiol. Heart Circ. Physiol.Home page
A. L. Baggish, K. Yared, F. Wang, R. B. Weiner, A. M. Hutter Jr., M. H. Picard, and M. J. Wood
The impact of endurance exercise training on left ventricular systolic mechanics
Am J Physiol Heart Circ Physiol, September 1, 2008; 295(3): H1109 - H1116.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Physiol. Heart Circ. Physiol.Home page
A. J. Pope, G. B. Sands, B. H. Smaill, and I. J. LeGrice
Three-dimensional transmural organization of perimysial collagen in the heart
Am J Physiol Heart Circ Physiol, September 1, 2008; 295(3): H1243 - H1252.
[Abstract] [Full Text] [PDF]

Home page
 CirculationHome page
A. Cheng, T. C. Nguyen, M. Malinowski, G. T. Daughters, D. C. Miller, and N. B. Ingels Jr
Heterogeneity of Left Ventricular Wall Thickening Mechanisms
Circulation, August 12, 2008; 118(7): 713 - 721.
[Abstract] [Full Text] [PDF]

Home page
 J Am Coll CardiolHome page
S. J. Lester, A. J. Tajik, R. A. Nishimura, J. K. Oh, B. K. Khandheria, and J. B. Seward
Unlocking the Mysteries of Diastolic Function Deciphering the Rosetta Stone 10 Years Later.
J. Am. Coll. Cardiol., February 19, 2008; 51(7): 679 - 689.
[Abstract] [Full Text] [PDF]

Home page
 Circ. Res.Home page
C. M. Ripplinger, W. Li, J. Hadley, J. Chen, F. Rothenberg, R. Lombardi, S. A. Wickline, A. J. Marian, and I. R. Efimov
Enhanced Transmural Fiber Rotation and Connexin 43 Heterogeneity Are Associated With an Increased Upper Limit of Vulnerability in a Transgenic Rabbit Model of Human Hypertrophic Cardiomyopathy
Circ. Res., November 9, 2007; 101(10): 1049 - 1057.
[Abstract] [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]

Home page
 Eur. J. Cardiothorac. Surg.Home page
S. H. Gilbert, A. P. Benson, P. Li, and A. V. Holden
Regional localisation of left ventricular sheet structure: integration with current models of cardiac fibre, sheet and band structure
Eur. J. Cardiothorac. Surg., August 1, 2007; 32(2): 231 - 249.
[Abstract] [Full Text] [PDF]

Home page
 J Am Coll CardiolHome page
J. D. Thomas and Z. B. Popovic
Assessment of Left Ventricular Function by Cardiac Ultrasound
J. Am. Coll. Cardiol., November 21, 2006; 48(10): 2012 - 2025.
[Abstract] [Full Text] [PDF]

Home page
 J Am Coll CardiolHome page
P. P. Sengupta, J. Korinek, M. Belohlavek, J. Narula, M. A. Vannan, A. Jahangir, and B. K. Khandheria
Left Ventricular Structure and Function: Basic Science for Cardiac Imaging
J. Am. Coll. Cardiol., November 21, 2006; 48(10): 1988 - 2001.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services