Dynamic three-dimensional echocardiography combined with semi-automated border detection offers advantages for assessment of resynchronization therapy

Abstract

Simultaneous electrical stimulation of both ventricles in patients with interventricular conduction disturbance and advanced heart failure improves hemodynamics and results in increased exercise tolerance, quality of life. We have developed a novel technique for the assessment and optimization of resynchronization therapy. Our approach is based on transthoracic dynamic three-dimensional (3D) echocardiography and allows determination of the most delayed contraction site of the left ventricle (LV) together with global LV function data. Our initial results suggest that fast reconstruction of the LV is feasible for the selection of the optimal pacing site and allows identifying LV segments with dyssynchrony.

Background

Simultaneous electrical stimulation of both ventricles in patients with interventricular conduction disturbance and advanced heart failure improves hemodynamics and results in increased exercise tolerance, quality of life [1–3]. Reduction in morbidity and mortality was recently reported.

Case presentation

We have developed a novel technique for the assessment and optimization of resynchronization therapy. Our approach is based on transthoracic dynamic three-dimensional (3D) echocardiography and allows determination of the most delayed contraction site of the left ventricle (LV) together with global LV function data. We use a self-developed fast rotating transducer, which is connected to a commercially available ultrasound system (GE Vingmed Vivid FiVe, Horton, Norway). The 64-element phased-array transducer has a center frequency of 3 MHz, second harmonic capabilities and continuously rotates inside the transducer assembly at 8-revolutions/ sec. During resynchronization device implantation, images are obtained with the patient in the left lateral decubitus position with the transducer in the apical position and the image plane rotating around the LV long axis. Images are acquired in sinus rhythm and in two different pacing modes: during right ventricular apical pacing and biventricular pacing. With a self-developed software, using MatLab (The MathWorks, Inc, Natick, MA, USA), the original 2D images are post-processed by placing them in their correct spatial and temporal (ECG reference) position using multi beat data fusion. All the cross-sectional images re-sampled from each dataset are subsequently imported into the TomTec^® 4D LV-analysis software (TomTec^® Imaging Systems GmbH, Germany) for automated endocardial border detection.

Subsequently, the program performs a dynamic surface rendered endocardial reconstruction of the LV in sinus rhythm and in the different pacing modes. For each pacing mode, a time volume curve (TVC) is plotted from which global end-diastolic (LVEDV), end-systolic volumes (LVESV) and ejection fraction (EF) are calculated. The LV endocardial surface is subdivided in 16 segments, which are color-coded for orientation. The difference in time to maximal myocardial contraction between segments is used to assess and measure regional mechanical delay and dyssynergy (Movies 1, 2 and 3 – see additional file 1, additional file 2 and additional file 3 respectively).

Our initial results suggest that fast reconstruction of the LV is feasible for the selection of the optimal pacing site and allows identifying LV segments with dyssynchrony. Additional hemodynamic evaluation is also possible (Movies 1, 2 and 3). Currently, the LV reconstruction is still off-line. However, real time three-dimensional echocardiography is now available, and the optimal pacing site can be determined on-line.

Conclusions

Our major finding is that transthoracic dynamic 3D echocardiography performed by a fast rotating transducer and combined with automated contour analysis is feasible to determine mechanical asynchrony during LV contraction. Additionally, it is capable of hemodynamic evaluation and therefore may be advantageous as compared to TDI. Our data suggest that this method may be useful for evaluation and guiding resynchronisation device implantation. Tissue Doppler Imaging (TDI) is the imaging modality, which is currently used to assess the most delayed ventricular site before implantation of a resynchronisation device[4, 5]. Although the method is proven to be feasible, it is uncomfortable to use online and, most importantly, assessment of the hemodynamic effect requires additional and separate studies. The aim of our project was to develop and test the feasibility of an alternative approach for determining the optimal pacing site. Our proposed method, based on 3D echocardiographic time-volume curves (TVC), includes additional features such as hemodynamic evaluation by measuring true ventricular volumes. Previous studies demonstrated that generation of TVCs, representing LV volume changes over the cardiac cycle, is feasible and accurate and allows detailed quantitative analysis of LV performance, even in patients with LV dysfunction.