Left atrial haemodynamic evaluation using 4D flow cardiac magnetic resonance imaging: application to hypertension and hypertrophic cardiomyopathy

Abstract Introduction Changes in left atrial (LA) hemodynamics have been suggested reflective of left ventricular (LV) disease progression. Kinetic energy (KE), viscous energy loss, and vorticity could be promising indices for the assessment of left atrial dynamics by time resolved four-dimensional cardiac magnetic resonance (4D flow CMR). Hypertrophic cardiomyopathy (HCM) patients are at risk of developing mitral regurgitation and left ventricular outflow tract obstruction, with poor outcome (1). Hypertensive myocardial remodelling has been suggested to be due to complex flow and tissue interactions leading to hypertrophy and diastolic dysfunction (2). Early detection of altered LA hemodynamic abnormalities may help non-invasive assessment, improving their long-term outcome (3). Purpose Quantitative and qualitative evaluation of left atrial hemodynamic metrics by 4D flow CMR in a cohort of HCM patients, hypertensive patients, and healthy controls with no known cardiovascular disease. Methods Eighteen HCM patients (50.0 [±18.2] years, 8 females), twenty-one hypertensive patients (55.2 [±6.2] years, 11 females), and seventeen controls (36.7 [±12.8] years, 1 female) underwent 4D flow CMR. Both patient groups had significantly higher LV-indexed mass, end-diastolic volume, and end-systolic volumes (P < 0.05). A semi-automated pipeline was used to segment the LA including pulmonary inflow and mitral outflow tracts using phase contrast magnetic resonance images derived from the 4D flow acquisition's phase and magnitude images. LA KE, viscous energy loss and vorticity were computed throughout the cardiac cycle. Results The HCM and hypertensive groups demonstrated significantly higher average KE integrated across the cardiac cycle (1.58 ± 0.79 and 1.32 ± 0.50 mJ respectively) compared to controls (1.20 ± 0.47 mJ) (P < 0.05), with the systolic curve in HCM showing the highest peak KE. Qualitatively, KE in the diastolic portion of the cardiac cycle appeared less variable between groups (Figure 1). Vorticity in HCM showed different configurations with loss of the central atrial vortex core (Figure 2). Viscous energy loss was also significantly higher in HCM patients confirming the increase in energy loss (Figure 2). Average KE across the cardiac cycle highly correlated with LA volume and pulmonary capillary wedge pressure calculated by a previously published regression method (4) (Pearson correlation: r= 0.76 and 0.71 respectively P < 0.05). There were also significant correlations with cardiac metrics such as indexed end-diastolic and indexed end-systolic volumes (Pearson correlation: r= 0.54 and 0.52 respectively P < 0.05) as well as indexed LV mass (r= 0.59 P < 0.05). Conclusion Hemodynamic metrics derived from 4D flow CMR may be valuable in establishing valuable predictive metrics of ventricular function. Integration of 4D flow MRI in clinical workflows could provide information that may be indicative of pathological onset before abnormal remodelling.Kinetic energy and viscous energy lossParticle pathlines and vorticity