Main content area

A comparison of retrospectively self-gated magnetic resonance imaging and high-frequency echocardiography for characterization of left ventricular function in mice

Amundsen, Brage Høyem, Ericsson, Madelene, Seland, John Georg, Pavlin, Tina, Ellingsen, Øyvind, Brekken, Christian
Laboratory animals 2011 v.45 no.1 pp. 31-37
anesthesia, cardiac output, echocardiography, image analysis, isoflurane, magnetic resonance imaging, mice
Non-invasive imaging methods like echocardiography and magnetic resonance imaging (MRI) are very valuable in longitudinal follow-up studies of cardiac function in small animals. To be able to compare results from studies using different methods, and explain possible differences, it is important to know the agreement between these methods. As both self-gated high-field MRI and high-frequency echocardiography (hf-echo) M-mode are potential methods for evaluation of left ventricular (LV) function in healthy mice, our aim was to assess the agreement between these two methods. Fifteen healthy female C57BL/6J mice underwent both self-gated MRI and hf-echo during the same session of light isoflurane anaesthesia. LV dimensions were estimated offline, and agreement between the methods and reproducibility for the two methods assessed using Bland–Altman methods. In summary, hf-echo M-mode had better inter-observer repeatability than self-gated MRI for all measured parameters. Compared with hf-echo, systolic posterior wall thicknesses were significantly higher when measured by MRI, while diastolic anterior wall thicknesses were found to be significantly smaller. MRI measurements of diastolic LV diameter were also higher using MRI, resulting in larger fractional shortening values compared with the values obtained by hf-echo. In conclusion, hf-echo M-mode is easy to apply, has high temporal and spatial resolution, and good reproducibility. Self-gated MRI might be advantageous in cases of abnormal LV geometry and heterogeneous regional myocardial function, especially with improvements in spatial resolution. The moderate agreement between the methods must be taken into account when comparing studies using the two modalities.