Lawrence M. Boxt

Estimation of right ventricular mass in normal subjects and in patients with primary pulmonary hypertension by nuclear magnetic resonance imaging

OBJECTIVES This study was designed to test the accuracy of nuclear magnetic resonance (NMR) imaging as a noninvasive technique for estimating right ventricular mass in normal subjects and in patients with primary pulmonary hypertension. BACKGROUND An accurate means of noninvasively estimating right ventricular mass may allow better characterization of the degree of right-sided pressure or volume overload caused by underlying cardiac or pulmonary diseases. METHODS End-diastolic short-axis electrocardiogram (ECG)-gated spin echo NMR images of the heart were obtained in vivo in 13 patients with primary pulmonary hypertension and 10 normal adult volunteers. Both right and left ventricular mass were computed by summing the myocardial slice volumes over all slices spanning the myocardium and multiplying by myocardial density. This technique of myocardial mass determination was verified by imaging 10 calf hearts and comparing the NMR-determined right and left myocardial mass with the actual mass determined by weighing the right and left ventricles. RESULTS In the calf heart study, an excellent correlation was obtained between the directly measured ventricular mass and the NMR-calculated mass, for both the right and the left ventricle. Patients with primary pulmonary hypertension had an elevated right ventricular mass index compared with that of normal subjects (62.69 +/- 8.72 g/m2 vs. 23.32 +/- 1.36 g/m2, p < 0.0005). There was no significant difference in left ventricular mass index between the two groups. Both mean intraobserver and inter-observer variability in myocardial mass determination were low. Linear regression analysis between right ventricular mass index and mean pulmonary artery pressure was significant (r = 0.75, p < 0.003). CONCLUSIONS Electrocardiogram-gated spin echo NMR imaging of the heart may be used for quantitating right ventricular mass in normal subjects and in patients with primary pulmonary hypertension, in whom it may also provide an alternative noninvasive technique for estimating mean pulmonary artery pressure.

Direct quantitation of right and left ventricular volumes with nuclear magnetic resonance imaging in patients with primary pulmonary hypertension

To test the utility of electrocardiographically gated spin echo nuclear magnetic resonance (NMR) imaging in quantitating right and left ventricular volumes and function in patients with primary pulmonary hypertension, right and left ventricular end-diastolic and end-systolic volumes, stroke volumes and ejection fractions were determined in 11 patients with primary pulmonary hypertension and in 10 subjects with normal echocardiographic findings. Ventricular chamber volumes were computed by summing the ventricular chamber volumes of each NMR slice at end-diastole and end-systole. This technique was verified by comparison of results obtained by this method and with the water displacement volumes of eight water-filled latex balloons and ventricular casts of eight excised bovine hearts. In the patients with primary pulmonary hypertension, right ventricular volume indexes were 121 +/- 45 ml/m2 at end-diastole and 70.1 +/- 41.6 ml/m2 at end-systole; both values were significantly greater than values in the normal subjects (67.9 +/- 13.4 and 27.9 +/- 7.5 ml/m2, respectively). Left ventricular end-diastolic volume index was significantly less in the patients (44.9 +/- 9.7 ml/m2) than in the normal subjects (68.9 +/- 13.1 ml/m2). There was no significant difference in left ventricular end-systolic volume between the two groups (24.4 +/- 8.6 and 27.1 +/- 7.8 ml/m2, respectively). Right and left ventricular ejection fractions in the patients with primary pulmonary hypertension (0.43 +/- 0.21 and 0.46 +/- 0.15, respectively) were significantly less than values in normal subjects (0.59 +/- 0.09 and 0.6 +/- 0.11, respectively).(ABSTRACT TRUNCATED AT 250 WORDS)

Three-dimensional Echocardiographic Volume Computation by Polyhedral Surface Reconstruction: In Vitro Validation and Comparison to Magnetic Resonance Imaging

Two-dimensional echocardiographic methods of left ventricular volume computation are limited by geometric assumptions and image plane positioning error in the nonvisualized dimension. We evaluated a three-dimensional (3D echocardiographic method that addresses these limitations. Our method uses a volume computation algorithm based on polyhedral surface reconstruction (PSR) and nonparallel, unequally spaced, nonintersecting short-axis planes. Seventeen balloon phantoms were subjected to volume computation by the 3D echocardiography-PSR method and by magnetic resonance imaging (MRI) and compared to true volumes determined by water displacement. The results for 3D echocardiography-PSR were: accuracy = 2.27%, interobserver variability = 4.33%, r = 0.999, SEE = 2.45 ml, and p less than 0.001. Results for MRI were 8.01%, 13.78%, r = 0.995, SEE = 7.01 ml, and p less than 0.001. There was no statistically significant difference between the methods. We conclude that precise image plane positioning and use of the 3D echocardiographic-PSR volume computation method achieves high accuracy and reproducibility in vitro. The excellent in vitro correlation between 3D echocardiography-PSR and MRI indicates that MRI may also serve as an in vivo standard of comparison.

Comparison of three-dimensional echocardiographic assessment of volume, mass, and function in children with functionally single left ventricles with two-dimensional echocardiography and magnetic resonance imaging

Diminished systolic function or inappropriate hypertrophy are considered risk factors for outcome following the Fontan procedure. These parameters are difficult to assess in univentricular hearts that do not conform to the uniform shapes prescribed by conventional 2-dimensional imaging volume algorithms. Three-dimensional echocardiography requires no geometric assumptions and has been validated in both normal and distorted left ventricles. To assess the feasibility and accuracy of this technique in patients with univentricular hearts, we compared 2- and 3-dimensional echocardiographic estimates of ventricular volume, ejection fraction, and mass in patients with functionally single left ventricles with results obtained by magnetic resonance imaging (MRI). Twelve patients with functionally single left ventricles (6 months to 22 years) underwent examination by all 3 modalities. Correlation and agreement with MRI were calculated for volumes, ejection fraction, and mass. Three-dimensional echocardiographic comparison with MRI yielded a bias of 3.4 +/- 5.5 ml and 14.2 +/- 8.3 ml for systolic and diastolic volumes, respectively. Agreement analysis for mass showed a bias of 5.8 +/- 8.4 grams. Two-dimensional echocardiography showed less agreement for both volumes and mass (bias of -2.9 +/- 8.1, 2.9 +/- 10.4 ml and -8.3 +/- 12.0 g for volume and mass, respectively, p >0.05). Ejection fraction by 3-dimensional echocardiography showed significantly closer agreement with MRI (bias of 4.4 +/- 5.3%) than 2-dimensional echocardiography (bias of 8.5 +/- 10.3%, p = 0.04). Thus, 3-dimensional echocardiography provides estimates of ventricular volumes, ejection fraction, and mass that are comparable to MRI in this select group of patients with single ventricles of left ventricular morphology.

Estimation of myocardial water content using transverse relaxation time from dual spin-echo magnetic resonance imaging

Dual spin-echo magnetic resonance imaging may be used for calculation of transverse myocardial relaxation time from the signal intensity of the echoes considered. In this study, the ability of myocardial transverse relaxation time (T2) to quantitate myocardial edema of the right ventricle (RV) and left ventricle (LV) was tested. Dual spin-echo magnetic resonance images of the entire hearts were obtained and T2 of the RV and LV myocardium calculated from the signal intensities within multiple regions of interest distributed over the myocardium. Six hearts were intermittently perfused through an aortic cannula with three perfusates of decreasing osmolality. Biopsies were obtained for water content (WC) analysis both before and after imaging the hearts at baseline and post-perfusion. A seventh (control) heart was not perfused; instead dual spin-echo imaging was performed at the same time intervals as in the perfused hearts. Prior to any intervention, there was no significant difference between baseline RV (79.49 +/- 2.10%) and LV (77.99 +/- 2.44%, p = .2) myocardial water content; RV myocardial T2 (59.9 +/- 5.8 msec) was slightly but not significantly longer than that of the LV (54.6 +/- 5.7 msec, p = .1). After induction of edema, strong correlation was found between right ventricular myocardial water content measurements and right ventricular T2 (RV WC = 68.5 + 0.19 x RV T2; N = 27, R = 0.92, p < .0001, SEE = 1.56%). Similarly, strong correlation was found between left ventricular myocardial water content and T2 (LV WC = 62.1 + 0.29 x LV T2; N = 27, R = 0.92, p < .0001, SEE = 1.80%).(ABSTRACT TRUNCATED AT 250 WORDS)

Feasibility of a two-dimensional echocardiographic method for the clinical assessment of right ventricular volume and function in children.

The relative ease of acquisition and safety of two-dimensional echocardiography has established it as the mainstay for routine cardiac imaging. Translation of imaging data into useful quantitative information, however, requires fitting the ventricle to a specific geometric model. Because of its complex shape and anterior position, many attempts at right ventricular quantitation by two-dimensional echocardiography have been criticized as impractical and not reproducible. A simple method incorporating subcostal and apical imaging was introduced in 1984. This approach appeared to combine accuracy and practicability but was never validated in a clinical setting because of the difficulties of subcostal imaging in adults. This study assessed the feasibility and accuracy of this technique in the pediatric population. Results of volume comparison to values derived by magnetic resonance imaging were r = 0.96, standard error of the estimate (SEE) = 19.3 ml, and mean difference = 15 +/- 19.4 ml and r = 0.97, SEE = 12.3 ml, and bias = 5 +/- 11.8 ml for diastolic and systolic volumes, respectively. Comparison of estimates of ejection fraction with magnetic resonance imaging demonstrated r = 0.90, SEE = 5.9%, and bias = 3% +/- 5.7%. Interobserver and intraobserver variability was 9.9% and 8.2%, respectively, for systolic volumes and 11.5% and 8.9%, respectively, for diastolic volumes. Evaluation of right ventricular size and function by this approach is comparable to determinations by magnetic resonance imaging and may be clinically useful in the management of pediatric patients.

Double-quantum-filtered sodium imaging

Abstract Double-quantum-filtered (DQ) images of 23Na at 7.05 T are reported for phantoms containing concentric volumes of 145 mM sodium chloride and 145 mM sodium chloride entrapped in a 4% (w/v) agarose gel. Excellent single-quantum suppression was obtained. DQ image profiles are also reported for 46 and 4.6 mM sodium chloride in agarose gels. The sensitivity of the DQ-filtered signal to magnetic field inhomogeneity and to susceptibility effects is explicitly demonstrated. The DQ filter can be used to introduce an additional source of contrast into Na images. We describe the considerations necessary for obtaining DQ-filtered images with good signal-to-noise in short imaging times and quantify the losses encountered when using a DQ filter. Possible extensions to in vivo and in vitro biological specimens and to nonbiological materials are discussed.

Intravascular ultrasound imaging of human cerebral arteries

The aim of this study was to assess the feasibility of imaging cerebral arteries in vitro with intravascular ultrasound and to establish a correlation between echographic images and corresponding histological architecture. Intravascular ultrasound imaging was performed using a 30‐MHz, 4.3F ultrasound probe. Twenty‐two arterial segments were obtained at autopsy from 6 patients and were imaged fresh. Arteries were then processed for histological examination and comparisons were made between echographic and histological findings. The correlation between luminal area measurements as determined histologically and by intravascular ultrasound was tested by linear regression analysis. Intravascular ultrasound demonstrated a three‐layered appearance in normal cerebral arteries but not in those affected by severe atherosclerosis. Overall, ultrasound correctly identified the presence of a plaque in 83% of patients. Intravascular ultrasound sensitivity and specificity, respectively, were 100 and 80% for calcium deposits and 83 and 75% for fibrous tissue. Intravascular ultrasound and histological measurements correlated well for the determination of luminal area (r = 0.89). Intravascular ultrasound provides accurate characterization of the arterial lumen and geometry, as well as the presence and histological features of atherosclerotic plaque. Thus, it appears to have a great potential for an earlier and more accurate diagnosis of atherosclerosis and may serve to guide new interventional techniques being utilized in the treatment of cerebrovascular diseases.

Motion dependence of myocardial transverse relaxation time in magnetic resonance imaging

We discuss the effects of motion on the computation of the myocardial transverse relaxation time by use of magnetic resonance imaging. Equations describing its behavior are derived and illustrated graphically under different conditions. It is shown that the myocardial transverse relaxation time calculated from magnetic resonance images depends on the actual myocardial transverse relaxation time ex vivo (T2) as well as the phase of the cardiac cycle in which it is computed, heart rate, cardiac wall velocity, choice of spin-echoes used in the calculation, and the spin-echo times employed. In particular, the error in T2 decreases when both the first and third echoes are employed in the calculation, rather than only the first two echoes. However, the myocardial transverse relaxation time is more strongly dependent on heart rate in the former case rather than in the latter. Furthermore, the error in T2, when both the first and second spin echoes are used in the calculation, is seen to increase as the spin-echo time shortens. On the other hand, the error in T2 decreases for shorter spin-echo times when both the first and third spin echoes are used instead. The results are relevant to the noninvasive assessment of ischemia, cardiac transplantation rejection, and other myocardial disorders.