25 Mitral inflow velocity encoded imaging by CMR for the assessment of left ventricular haemodynamics

Background Heart failure is common and is associated with high mortality and morbidity. Haemodynamic assessment is critical to diagnosis and therapeutic monitoring. Non-invasive Doppler echocardiography is commonly used to assess peak early and late mitral inflow velocities to predict left ventricular (LV) haemodynamics. Cardiovascular magnetic resonance imaging offers two-dimensional through plane flow assessment using phase contrast acquisition. It remains unclear which through plane mitral inflow velocity (averaged or peak) best predicts LV haemodynamics better. Purpose The objective of this study is to investigate which through plane mitral inflow velocities (averaged or peak) is better associated with invasive left heart haemodynamic parameters. Methods In this observational study, 16 patients were prospectively recruited at a large tertiary pulmonary hypertension unit. These patients had invasive right heart haemodynamic catheter-based study (RHC). In addition, all patients received CMR on the 1.5 T scanner (HDx scanner, GE Healthcare, Waukesha, Wisconsin, USA), using an 8-channel cardiac coil. CMR protocol included long and short axis cines and mitral inflow phase contrast acquisition. The LV haemodynamic parameters assessed by RHC included the following: pulmonary capillary wedge pressure (PCWP), cardiac output, cardiac index, systemic vascular resistance (SVR). Cardiac output assessment was done by standard thermodilution method. Results Mean age of the 16 patients analysed was 68.7±7.9 years and 56.3% were female. Of the 16 patients, 5 patients had non-analysable data. Mean PCWP was 11.1±4.7 mmHg, mean cardiac output was 4.4±1.3 L/min, mean cardiac index was 2.4±0.8 L/min/m2 and mean SVR was 1827±589 dynes/seconds/cm-5. Peak EA ratio correlated with PCWP (r=0.68 p=0.02) and with cardiac index (r=0.68 p=0.02). There was a notable correlation between mean EA ratio and cardiac index (r=0.66 p=0.03). Mean A showed a significant correlation with SVR (r=0.64, p=0.03). Conclusions Peak mitral inflow velocities are more closely associated with LV haemodynamics. This may be due to reduced variation in through-plane peak velocity at different inflow levels versus the mean velocity. Future work is required to determine the effect of addressing the valvular through-plane motion, which may make the averaged mitral inflow assessment a better tool for LV hemodynamics.