3D echocardiographic quantification of ejection fraction and cardio-toxicity onset

The aim of this study was to evaluate if variability in EF estimate from echocardiographic data acquired with two dimensional (2DE) and three-dimensional (3DE) systems and analyzed using different software packages could affect cardio-toxicity assessment. We analyzed 2DE and 3DE datasets in 94 patients treated for breast cancer with anthracycline and trastuzumab. EF was computed from 2DE and 3DE data using two software packages (EchoPAC, GE Healthcare and TomTec 4D LV analysis). Corresponding estimates were compared. In addition, in a subgroup of 20 patients 3DE data were re-analyzed and intra-observer and inter-observer variability by three investigators were computed, using both software packages. As expected 2DE-based estimates significantly underestimated 3DE-based estimates. Intra-observer and inter-observer variability using both analysis packages showed a huge variability, due to significant differences in end systolic volume and EF. Following clinical definition of cardio-toxicity onset, these variability results could be a confounding factor since variations in EF measurement are in the range of EF decrease due to cardiac adverse effects from cancer therapeutic drugs.