Abstract 17984: Validating the Gold-standard: Steady-State Free Precession Sequence for Assessment of Left Ventricular and Right Ventricular Mass

Abstract

Background: Left ventricular (LV) mass (LVM) predicts cardiovascular morbidity and mortality, thus accurate quantification of LVM is essential. Cardiac magnetic resonance (CMR) is considered the gold-standard for LVM and right ventricular (RV) mass (RVM) quantification. The steady-state free precession (SSFP) is the most widely used sequence nowadays; however, SSFP-derived LVM/RVM has only been validated in normal animals, while CMR is usually performed under pathological conditions. Therefore, our objective was to validate in vivo SSFP-derived LVM/RVM with the gold-standard autopsy weights of experimental animals with myocardial infarction (MI), LV remodeling, and pulmonary hypertension (PH). Methods: MI was induced in 11 pigs by balloon occlusion of the proximal LAD for 60 min, and MRIs were obtained 2 months post-MI. PH was induced in 11 pigs by surgical ligation of three pulmonary veins and animals underwent CMR 4 months post-surgery. Animals were euthanized immediately after CMR. Each ventricle was separately weighted using a high-fidelity scale. MRI studies were performed with a 3.0 Tesla magnet. Results: In the MI model, infarct size (IS) was 29±6% of LV, LVEF 34±8%, RVEF 62±149%, mean pulmonary artery pressure (mPAP) 16±4mmHg, and pulmonary vascular resistance (PVR) 2.3±1.1 Wood units. In the PH model, IS was 0%, LVEF 64±5%, RVEF 55±10%, mPAP 36±16mmHg and PVR was 7.2±5.5 Wood units. All animals provided images of diagnostic quality. Excellent correlations were obtained between SSFP-calculated LV mass (86.6±12.9g) and autopsy-measured LV mass (91.1±15.2g, r=0.97, p<0.001). For LV, the correlation was not different in both groups of animals (r=0.98, p=0.01 for post-MI animals; r=0.96, p=0.01 for PH animals). There was also a strong correlation between RV mass obtained from CMR (37.9±14.1g) and from autopsy (41.6±13.1g r=0.9, p=0.01). For RV, the correlation was higher in PH animals (r=0.92, p<0.001) than in post-MI pigs (r=0.8, p=0.01). Conclusions: In vivo SSFP-CMR sequences determine LVM and RVM accurately and reliably compared with autopsy. Therefore, our study provides further validation for the clinical use of SSFP sequence derived LVM and RVM.