P334 Heart model method: could it be a new echocardiographic gold-standard for the assessment of the left ventricular systolic function in a population under chemotherapy treatment?

Abstract

Background Echocardiographic assessment of the left ventricular systolic function is essential in diagnosis and during the follow up of cardiovascular diseases. Although subjective visual approach method is easily applied, quantitative systems give more objective information about systolic function analyses. The purpose of this study is to evaluate the different quantitative methods of estimating systolic function basal in non-invasive techniques Methods We used a group of 40 patients, prospectively collected, under chemotherapy treatment with preserved systolic function. Same echocardiography device (Philips EPIQ-7) has been used in all studies (acquiring apical 4 and 2 chambers and 3 D of apical volume by an experimented operator). We compare three standard methods with impact in the literature (Speckle tracking and 3D Heartmodel system) to the echocardiographic gold-standard (Simpson’s biplane method). The Bland-Altman method has been used for the graphic comparison of the values of the resulting measures while the statistical comparison was made by a T-student method. Results Three quantitative methods were used to compare left ventricular systolic function assessment (Heart Model 3D (60.4% ± 5.2%), Strain (60.50% ± 7.1%), global longitudinal strain (-19.7 ± 3.15%) to Simpson’s biplane (mean 62.10% ± 5.75%). Values of differential means (2.73 with Heart model 3D and 2.08 with Strain) compared to Simpson’s biplane were translated to a Bland-Altman plot and means were compared with a T-student method. A statistically significant difference was found in case of the Heartmodel method compared with Simpson’s biplane (p < 0.05), though it does not imply any clinical difference. Less time consuming and better segmentation of the cardiac cavities in just one beat with the 3D-Heart Model technique was a magnificent point compared to the Strain method that needed a postprocessing modification Conclusions Heartmodel method is probably the most time-saving and with a good accuracy of left ventricular systolic function assesment and it is not inferior compared to the echocardiographic gold-standard Simpson’s biplane method. Abstract P334 Figure 1. Bland-Altman analysis