Echocardiographic patterns of myocardial fibrosis in hypertensive patients: endomyocardial biopsy versus ultrasonic tissue characterization.

Echocardiographic image texture has been demonstrated to reflect the physical properties of the tissue under examination. To evaluate the role of collagen in determining the echo pattern of the left ventricular wall, we studied nine hypertensive patients with left ventricular hypertrophy (left ventricular mass index > 125 gm/m2) and biopsy-proven different degrees of myocardial fibrosis by analyzing the echocardiographic examinations performed before the biopsy. Myocardial tissue was sampled under fluoroscopy and two-dimensional echo guidance in the interventricular septum. Collagen volume fraction (CVF; normal range up to 2%) was taken as an index of fibrosis. The echo patterns were assessed by analyzing standard two-dimensional parasternal long-axis echocardiograms recorded on videotape. Images were color-coded at 256 levels (0 = yellow, 256 = black) and digitized off-line onto a personal computer. The region of analysis was set using a selection tool (20 x 10 mm) in the general area of septum where the specimen was taken. For each selection a color-level histogram, representing the frequency distribution, was derived with estimates of the average pixel intensity (mCS), skewness (SK), kurtosis (K), and the broad band (Bb) of the echoes about the distribution. Echo-derived parameters in each patient were compared with corresponding CVF values. CVF was out of range in all patients, ranging from 2.6% to 7.6% (mean 4.3% +/- 1.6%). No correlation was found between CVF and mCS, whereas a significant correlation was found at end diastole between CVF and the parameters describing histogram morphology, respectively, SK (r = 0.73), K (r = 0.69), Bb (r = 0.72). These findings for the first time demonstrate in vivo in hypertensive patients with left ventricular hypertrophy an agreement between echo amplitude and histologically assessed collagen volume. Thus in our studied patients collagen content appears to be the major determinant of regional echo intensity, its increase resulting in a significant and progressive wider asymmetrical left shift (yellow) of the color histogram.

Matrix metalloproteinase activity expression in infarcted, noninfarcted and dilated cardiomyopathic human hearts.

In the normal myocardium matrix metalloproteinases (MMP) are present in the latent form. To examine whether MMP are activated following infarction or idiopathic dilated cardiomyopathy (DCM), we extracted and measured MMP activity in tissue derived from 7 explanted, failing human hearts due to either previous myocardial infarction (MI) or DCM. MMP activity in infarcted left ventricle (LV), noninfarcted LV and right ventricle (RV) from MI patients, as well as tissue from either ventricle of DCM patients, were compared to the activity of donor heart tissue. SDS-PAGE and dye-binding assays were used to determine total protein concentration, while collagenase activity was measured by SDS-PAGE type substrate gels embedded with type I gelatin (zymography). Accuracy of the zymographic technique was shown for tissue samples as small as 0.05 mg and was comparable to results obtained by a spectrophotometric method. After normalization for total protein concentration, we found 3 +/- 1% collagenase activity in normal atrial tissue which could be activated to 80-90% by trypsin or plasmin, indicating that collagenase is normally inactive or in a latent form in human heart. In endo- and epimyocardium of infarcted LV, on the other hand, collagenase activity was 85-95% and 10-20%, respectively, while 5-10% and 3-5%, respectively, in noninfarcted LV. In DCM, collagenolytic activity in the endo and epimyocardium was 75 +/- 5 and 35 +/- 5% in the LV and 35 +/- 7 and 20 +/- 5% in the RV, respectively. Thus, in dilated failing human hearts secondary to previous MI or DCM, MMP activity is increased. This is particularly the case within the endomyocardium of the infarcted and noninfarcted portions of either ventricle with MI and in both ventricles in DCM. This suggests that an activation of collagenase throughout the myocardium may contribute to its remodeling that includes ventricular dilatation and wall thinning.