Aims: 2D echocardiography is limited for identifying chemotherapy-related cardiotoxicity. This study compared standard echo, 2D, and 3D speckle tracking echocardiography (STE) for detection of subclinical anthracycline (ANT) cardiotoxicity in breast cancer patients.
Methods and results: One-hundred consecutive breast cancer patients free of cardiac symptoms were treated by multiple protocols including ANT and cyclophosphamide and/or 5-fluorouracil for 3-4 cycles. Both before and after treatment, patients underwent standard echo, 2D STE-derived left ventricular (LV) global longitudinal strain (GLS), 3D volumetric echo and 3D STE with measurements of GLS, global circumferential strain (GCS), global area strain (GAS), and global radial strain (GRS). The follow-up period from the beginning of cancer therapy was 129 ± 18 days. All patients completed the chemotherapy cycles, without experiencing symptoms/signs of heart failure. 2D ejection fraction (EF) was not significantly changed after treatment, whereas E/e' ratio was higher than baseline (from 6.9 ± 2.2 to 7.3 ± 2.1, P = 0.006). 2D GLS was reduced after treatment (from -22.2 ± 2.3% to - 20.1 ± 6.6%, P = 0.004). 3D derived LV end-systolic volume was increased (P < 0.01) and 3D EF (from 62.4 ± 6.5% to 60.3 ± 7.3%, P < 0.01), 3D GLS (P < 0.001) GRS (P < 0.002), GCS, and GAS (both P < 0.0001) were reduced after ANT. Post-ANT 2D GLS was feasible in 90%, 3D EF in 66%, and 3D STE in 60% of patients.
Conclusions: Our study demonstrates potential superiority but also suboptimal feasibility of 3D EF and 3D strain in diagnosing subclinical ANT cardiotoxicity in breast cancer patients. 2D GLS is superior to standard echo and presents a good feasibility. E/e' ratio also offers advantages in revealing cardiotoxicity.
Keywords: anthracycline; ejection fraction; global longitudinal strain; left ventricular filling pressure; real-time 3D echocardiography; breast cancer.
Published on behalf of the European Society of Cardiology. All rights reserved. © The Author 2017. For permissions, please email: journals.permissions@oup.com.