Valvular Heart Disease
Siva Sreedhar
Medical Student
Northwestern Feinberg School of Medicine, Illinois, United States
Siva Sreedhar
Medical Student
Northwestern Feinberg School of Medicine, Illinois, United States
Aakash Gupta, MD
Radiology Resident Physician
Stanford University
Chicago, Illinois, United States
Teodora Chitiboi, PhD
Research Scientist
Siemens Healthcare GmbH, Hamburg, Germany
Hamburg, Germany
Maurice Pradella, MD
Deputy Section Head Cardiothoracic Radiology
University Hospital of Basel, Switzerland
Mohammed Elbaz, PhD
Assistant Professor
Northwestern University
Chicago, Illinois, United States
Mitral regurgitation (MR) affects over 10% of the US population above the age 751. Monitoring its course is crucial for treatment in both primary and secondary MR types2. Left atrial (LA) strain from echocardiography has been shown to be associated with MR outcomes3. Currently, CMR LA strain by feature tracking (FT) is measured through long axis views, which can assess longitudinal strain but not circumferential strain, which requires quantifying LA deformation over cine short-axis (SAX) cross-section. In this study, we sought to quantify circumferential LA strain by FT from LA SAX and test LA circumferential strain in terms of a) feasibility for quantifying differences in MR patients versus healthy controls; b) association with MR severity and cardiac function in primary and secondary MR patients.
Methods:
Forty-three MR patients (23 male, age: 62.0±10.3) – n=22 primary MR and n=21 secondary MR patients – and 12 healthy controls (6 male, age: 44.0±16.2 years) were retrospectively analyzed [Fig. 1A]. MR severity was graded by CMR4 [Fig. 1A]. Each participant underwent a cine SAX SSFP scan with a cross-sectional slice at the mid-LA level [Fig. 1B]. Circumferential strain was computed as the strain component tangential to the LA contour deformation using a FT software (TrufiStrain, Siemens) [Fig. 1B]. Strain values were averaged over the slice at each time point to create a strain time curve to calculate phasic strain parameters: reservoir, conduit, and booster5 [Fig. 1B]. If a strain curve did not have a clear shape, only peak (reservoir) strain was calculated. Per MR type (primary or secondary), we tested the association of circumferential strain parameters with ordinal MR severity (CMR grading) and LV cardiac function: stroke volume (SV), cardiac output (CO), and ejection fraction (EF).
Results: Circumferential strain was significantly lower in the total MR cohort versus healthy controls at every phase: reservoir (p< 0.01), conduit (p=0.02), and booster (p< 0.01) [Fig. 2A-C]. In primary MR patients, circumferential strain was inversely correlated with MR severity (ρbooster = -0.55), CO (ρbooster = -0.63), and SV (ρbooster = -0.51) [Fig. 3A-C]. In secondary MR patients, circumferential strain was inversely correlated with MR severity (ρbooster = -0.54) but directly correlated with EF (ρbooster = 0.63, ρreservoir = 0.48) [Fig. 3D-F].
Conclusion: To our knowledge, this is the first study to assess CMR SAX LA circumferential strain. The results showed the feasibility of LA circumferential strain by FT from cine in assessing differences in MR patients versus healthy controls. Circumferential booster strain was inversely correlated with MR severity, irrespective of MR type. Circumferential strain associated with cardiac function: SV & CO in the primary and EF in the secondary MR cohort. These results suggest that CMR LA circumferential strain may help assess impaired mechanics in MR patients. Future studies are needed to assess potential prognostic value of LA circumferential strain.