PO023 Cardiovascular Magnetic Resonance Imaging Compared To Echocardiography For Detecting Doxorubicin-Induced Cardiotoxicity

Abstract

Strain imaging is a robust clinical tool in cardiac surveillance of patients with breast cancer. However, image quality and therefore ability to accurately measure strain are often limited in this patient group because of tissue expanders, implants, and/or flap reconstruction. The aim of the present study was to evaluate the feasibility of measuring left ventricular longitudinal strain (LVLS) in the echocardiographic subcostal view in female patients with breast cancer.A total of 110 studies from 68 female patients with breast cancer were included. The feasibility of LVLS speckle-tracking measurements in the apical three-chamber (3C) and four-chamber (4C) views and in the subcostal 3C and 4C views was evaluated. The LVLS speckle-tracking measurements obtained in these two echocardiographic views were compared using intraclass correlation coefficients and Bland-Altman analyses.The feasibility of LVLS in the apical 3C and 4C views was 98.1% and 98.1%, respectively. In comparison, the feasibility of LVLS in the subcostal 3C and 4C views was 93.6% and 96.3%, respectively. A high degree of reliability was found between apical and subcostal LVLS 3C and 4C measurements. For the 3C view, the average measure intraclass correlation coefficient was 0.81 (95% CI, 0.72–0.88). For the 4C view, the average measure intraclass correlation coefficient was 0.80 (95% CI, 0.70–0.87). Bland-Altman analysis showed good agreement between apical and subcostal measurements in both the 3C and 4C views.Subcostal 3C and 4C LVLS can be reliably measured, with good agreement with conventional LVLS from the apical views, in female patients with breast cancer. Importantly, the subcostal view may provide a novel alternative for trending LVLS in patients with breast cancer who have technically limited apical windows.