Transthoracic 3D echocardiographic left heart chamber quantification in patients with bicuspid aortic valve disease

Allard T. van den Hoven; Jackie S. McGhie; Raluca G. Chelu; Anthonie L. Duijnhouwer; Vivan J.M. Baggen; Adriaan Coenen; Wim B. Vletter; Marcel L. Dijkshoorn; Annemien E. van den Bosch; Jolien W Roos-Hesselink

doi:10.1007/s10554-017-1192-1

June 19, 2017International journal of cardiac imagingOpen Access

Transthoracic 3D echocardiographic left heart chamber quantification in patients with bicuspid aortic valve disease

View Full Paper

Ask AI

Bookmark

Share

Why the study?

Does a fully automated 3D echocardiography algorithm (Left Heart Model) accurately quantify left heart chamber volumes compared to conventional 2D/3D echocardiography and CT in patients with bicuspid aortic valve disease?

Structured PICO

Does a fully automated 3D echocardiography algorithm (Left Heart Model) accurately quantify left heart chamber volumes compared to conventional 2D/3D echocardiography and CT in patients with bicuspid aortic valve disease?

P

Population

53 prospectively included patients with bicuspid aortic valve (BAV) disease (37 analyzed after exclusions for poor image quality or missing datasets), median age 35.2, 68% male, based in the Netherlands.

I

Intervention

Fully automated 3D echocardiography algorithm (Left Heart Model, LHM) for left atrial and left ventricular volume and ejection fraction quantification.

C

Comparator

Manual and semi-automated measurements using conventional 2D echocardiography (bi-plane, xPlane), 3D echocardiography, and cardiac computed tomography (CT).

O

Outcome

Feasibility and accuracy (Pearson's correlation) of LHM in quantifying left ventricular end-diastolic volume (LVEDV), end-systolic volume (LVESV), ejection fraction (LVEF), and left atrial volume (LAV).surrogate

The fully automated Left Heart Model software provides feasible, accurate, and highly reproducible 3D echocardiographic quantification of left heart chambers in patients with bicuspid aortic valve disease, overcoming major hurdles of manual 3D analysis.

Limitations

16 patients (30%) had to be excluded due to poor imaging quality
No cardiac magnetic resonance imaging (MRI) available for comparison
CT systematically estimated larger EDV compared to LHM and other echo modalities

Implication

May support automated 3D echo quantification in bicuspid aortic valve; leaves open broader validation before routine adoption.

Abstract

Integration of volumetric heart chamber quantification by 3D echocardiography into clinical practice has been hampered by several factors which a new fully automated algorithm (Left Heart Model, (LHM)) may help overcome. This study therefore aims to evaluate the feasibility and accuracy of the LHM software in quantifying left atrial and left ventricular volumes and left ventricular ejection fraction in a cohort of patients with a bicuspid aortic valve. Patients with a bicuspid aortic valve were prospectively included. All patients underwent 2D and 3D transthoracic echocardiography and computed tomography. Left atrial and ventricular volumes were obtained using the automated program, which did not require manual contour detection. For comparison manual and semi-automated measurements were performed using conventional 2D and 3D datasets. 53 patients were included, in four of those patients no 3D dataset could be acquired. Additionally, 12 patients were excluded based on poor imaging quality. Left ventricular end-diastolic and end-systolic volumes and ejection fraction calculated by the LHM correlated well with manual 2D and 3D measurements (Pearson's r between 0.43 and 0.97, p < 0.05). Left atrial volume (LAV) also correlated significantly although LHM did estimate larger LAV compared to both 2DE and 3DE (Pearson's r between 0.61 and 0.81, p < 0.01). The fully automated software works well in a real-world setting and helps to overcome some of the major hurdles in integrating 3D analysis into daily practice, as it is user-independent and highly reproducible in a group of patients with a clearly defined and well-studied valvular abnormality.