Real-time 3-Dimensional Echocardiography for Assessment of Cardiac Function and Congenital Heart Disease

I. SPECIFIC AIMS

1. To measure and calculate Z Scores of cardiac structures and function using 3-DE in a
large cohort of healthy infants, children, and adolescents.

2. To investigate the utility of 3-DE in infants, children and adolescents with congenital
and acquired heart disease.

II. BACKGROUND AND SIGNIFICANCE

Congenital heart disease (CHD)is one of the most common congenital malformations and is
associated with significant mortality and morbidity. Acquired heart diseases are also common
and have significant impact on the health of the young. As the therapeutic strategies are
evolving, reliable non-invasive methods to image the cardiovascular system and
quantitatively assess cardiac function with accuracy and reproducibility are pivotal to
define the natural history, to evaluate the results of intervention and to assess long-term
outcome.Three-dimensional echocardiography (3-DE) has important clinical applications to
CHD. Because of growth throughout childhood,interpretation of these measurements requires
normalization of the dimensions of cardiac structures to the size of the body. This is
commonly done by the calculation of Z scores for the measurement of interest. Currently
available normative data in the medical literature are limited to one-dimensional and
two-dimensional echocardiography measurements only.

In this study, we will generate normative data of the 3-DE measurements for cardiac
structure and function in a large cohort of healthy infants, children,and adolescents. We
will also investigate the utility of 3DE for assessing the cardiovascular structures and
cardiac function for congenital and acquired heart disease.

III. DESIGN AND METHODS

A. Subjects:

A total of 800 study subjects will be enrolled in the study after approval by the
Institutional Review Board. Echocardiography data will be prospectively collected. The
subjects will be divided into 5 groups:

Group 1. (n=300) Patients with normal cardiac anatomy and function by echocardiography.

Group 2. (n=300) Patients with various congenital heart diseases.

Group 3. (n=100) Patients with sickle cell disease and at risk for ventricular remodeling
and pulmonary hypertension.

Group 4. (n=50) Patients with Duchenne and muscular dystrophy at risk for ventricular
remodeling and dysfunction.

Group 5. (n=50) Patients with Marfan syndrome or the aortic disease.

B. Procedure:

Before the study, heart rate and blood pressures will be obtained for each study subject.
The subjects will be placed in a left recumbent position and ECG attached. After a complete
conventional echocardiographic assessment of each subject, real-time 3-DE /Doppler study
will be performed with 3-DE system. The 3-DE study will take about 30 minutes. After each
study, all data will be saved and processed off line. Indication for the clinical
echocardiogram will include cardiac murmur, syncope or chest pain but with a normal
echocardiogram. Informed consent will be obtained from each patient and legal
representatives after a full explanation of the procedure.

1. Conventional echocardiography All patients will undergo a complete 2- dimensional,
color flow Doppler, and spectral Doppler examination. Examination will be performed
using a commercially available ultrasound system (IE33 system, Philips, Andover, Mass).
All measurements will be made offline using a computer workstation(Xcelera,Philips
Medical Systems,Andover,MA). All measurements will be made according to American
Society of Echocardiography protocol. For any given structure, measurement will be made
if excellent and unambiguous views were available.

2. Transthoracic 3-DE The 3-DE/Doppler studies will be performed using the IE33 (Philips,
Andover, MA,USA) ultrasound machine with X3-1 or X7-2 matrix-array transducers. The
X7-2 matrix-array transducer is a new tool and particularly well suited for small
children. The patient will be in a decubitus position. Image acquisition will be
performed from an apical window with the LV as the region of interest based on
international recommendations. To encompass the complete LV into the 3D dataset, a full
volume scan will acquire in harmonic mode from four R-wave triggered subvolumes. This
scan will carried out during an end-expiratory breath hold a few seconds when possible.
The 3D dataset will be stored in a DICOM format on a CD-ROM and transferred to separate
workstations for off-line data analysis.

C. Image off-line analysis

1. QLAB Semiautomated Border Detection Method:A semiautomated border detection biplane LV
volume analysis was performed using off-line QLAB version 8.0 software (Philips). The
LV end-diastolic volume (EDV) and end-systolic volume (ESV) (smallest LV volume) frames
are identified.On both of these end-diastolic and end-systolic frames,five
identification points are marked: the septal, lateral, anterior, and inferior mitral
annulus and the apex (from either one of the views). Then the software automatically
delineates the LV endocardial border and, by sequential analysis, creates a LV
mathematical model or "cast" that represents the LV cavity.

2. TomTec Semiautomated Method:LV volume analysis will be performed off-line using TomTec
4D LV-Analysis software. TomTec requires triplane manual tracing of the endocardial
border after which the software selects a fitting geometric model and the total LV
endocardial border is automatically delineated with possible manual corrections. Based
on the six initial contours, a spatiotemporal interpolation model (like a pulsating
balloon) is created by rotational and temporal interpolation of these contours.
According to the initial balloon, the algorithm starts to detect the endocardial border
continuously in the entire 4D dataset (without large gaps due to interpolation), like
deforming the balloon in the LV until it best fits the walls in each frame. Adjustments
can be made manually after this step in oblique, sagittal, and coronal planes.
Subsequently, a final reconstruction of the LV model and a time-volume curve are
created. LV-EDV and LV-ESV are the largest and smallest volume, respectively, from this
time-volume curve.

D. Data Analysis

All numerical values are expressed as mean +/- standard deviation. The real- time 3-DE
measurements will be compared with other clinical data and measurements, such as cardiac
catheterization or magnetic resonance imaging. Statistical significance will be defined as
P<0.05. All statistical analyzes will be done using SAS for Windows version9.1 and Microsoft
Excel (Microsoft Office 2007) for all analysis.

Inclusion Criteria:

- Subjects with diagnosis that fits into Group 1 to 5 and who are willing to
participate (consent/assent obtained)

- Do not meet the exclusion criteria (as follows).

Exclusion Criteria:

- Unwilling to participate or give consent.

- Clinically unstable.

- Unable to co-operate.

- Poor acoustic windows.