1Uniklinik RWTH Aachen Med. Klinik I - Kardiologie, Angiologie und Internistische Intensivmedizin Aachen, Deutschland
Background: Incongruent findings of mitral regurgitation (MR) severity are frequently observed and discordance of quantification by echocardiographic proximal isovelocity surface area (PISA) method and cardiac magnetic resonance (CMR) has been described. Similar to CMR, echocardiographic volumetric method for quantification of MR has been proposed but there are concerns about the ability of accurate assessment of left ventricular (LV) volumes using manual endocardial tracings with biplane method of discs.
Aim: This study sought to evaluate MR quantification by transthoracic echocardiography with a volumetric approach (volumetric ECHO) using a semi-automatic method to overcome interobserver variability of measurements of LV volumes and to compare these analyses to the standard echocardiographic PISA method as well as to CMR analysis as the gold standard.
Methods: 30 consecutive patients (67±16 years) who were assigned for CMR analyses for different indications were included. Echocardiographic apical four- and two-chamber views were acquired for measurements of left ventricular end-diastolic volume (LVEDV) and end-systolic volume (LVESV) with a semi-automatic software (AUTO-EF). Furthermore, forward stroke volume (SV) was calculated by multiplying left ventricular outflow tract (LVOT) area calculated by measurements in parasternal long axis view and velocity-time-integral by pulsed-wave Doppler in LVOT in apical long-axis view. MR regurgitant volume (RVol) by volumetric ECHO approach was calculated as: RVol = (LVEDV - LVESV) - forward SV in LVOT (Figure 1). Additionally, MR severity was also assessed by standard echocardiographic PISA method.
CMR image acquisition was performed on a 1.5-T magnetic resonance scanner. LVEDV and LVEDV were derived from the short axis stack from cine sequences and through-plane velocity-encoded phase-contrast sequences were planned orthogonal to the proximal ascending aorta for calculation of forward SV. RVol by CMR was calculated as: RVol = (LVEDV - LVESV) - forward aortic SV (Figure 2).
Results: RVol calculated by volumetric ECHO was lower than RVol assessed by CMR (ΔRVol=-10.0±13.7 ml; P=0.013) indicating a remarkable bias, while there was less difference of RVol by the PISA method and CMR (ΔRVol=4.2±14.4 ml; P=0.2794) (Table). The difference of RVol between the volumetric ECHO method and CMR was mainly caused by the lower values of LVEDV (ΔLVEDV=-33,1%) and LVESV (ΔLVESV=-33,4%) measured by AUTO-EF resulting in lower total left ventricular SV (ΔLVSV=-32,7%). However, correlation of RVol measurements by volumetric ECHO and CMR, although only being moderate, was stronger (r=0.416) than by the PISA method and CMR (r=0.280).
Conclusions: Although quantification of MR by echocardiography with a volumetric approach using AUTO-EF resulted in notably smaller RVol values compared to the PISA-method, correlation of volumetric ECHO with CMR was stronger indicating better accordance because of the equivalent methodological approach. The absolute bias resulting in lower RVol by volumetric echocardiography compared to CMR was mainly caused by lower LVEDV and LVESV values assessed by echocardiographic AUTO-EF.