Closure of atrial septal defect normalizes left and right ventricular hemodynamic forces in children and adolescents

Figures

• 

  Fig. 1. Method overview for hemodynamic force analysis. Left, cine loops are acquired in the short-axis views and biventricular volumes contoured over the full cardiac cycle. Center, four-dimensional flow is acquired with coverage of both ventricles. Right, pressure gradients may be computed using the Navier-Stokes equations and visualized as a time-varying relative pressure field. Integrating the sum of all regional pressure gradients over the entirety of the ventricle produces an instantaneous global hemodynamic force, which may then be decomposed into three orthogonal components for reproducible quantification

• 

  Fig. 2. Inclusion flowchart

• 

  Fig. 3. Example of hemodynamic forces (HDF) in control (left), patient before ASD closure (center), and patient after surgery (right). Top row: left ventricular HDF, bottom row: right ventricular HDF. Different curve rendering styles are used to facilitate visual separation of LV and RV force traces. Note the relatively unaltered LV forces after closure (top right), and the changes to both systolic and diastolic forces in the RV, indicating both altered loading conditions (red and green dotted lines) and their effects on the role of the interventricular septum (blue dotted line). ASD atrial septal defect, HDF hemodynamic forces, LV left ventricular, RV left ventricular

• 

  Fig. 4. Biventricular HDF in controls and in patients before (Pre) and after (Post) ASD closure. Top, left ventricular HDF; bottom, right ventricular HDF. RMS (root mean square) values shown for systole and diastole separately. Squares denote a selected individual with persistently elevated HDF post closure, discussed separately. Controls shown as Tukey plots for visual clarity. Note different scales on Y axes. *, p<0.05; **, p<0.01. ASD atrial septal defect, HDF hemodynamic forces, RMS root mean square

• 

  Fig. 5. Pearson correlations of volumetric parameters associated with longitudinal HDF at baseline, with RMS values for systole and diastole separately. Pearson R and p values from non-log transformed data. Filled circles, systolic HDF; open circles, diastolic HDF. HDF hemodynamic forces, RMS root mean square

• 

  Fig. 6. Linear regression analysis of parameters significantly associated with changes in HDF after ASD closure. Top row: Increasing LV end-diastolic volume was significantly associated with increased systolic LV HDF in the lateral-septal direction, and conversely with decreased systolic RV HDF in the septum-free wall direction. EDV augmentation was also inversely correlated with diastolic RV HDF. Bottom row: Decreasing RV end-diastolic volume was associated with decreased systolic RV HDF in the apex-base and diaphragm-RVOT directions. Furthermore, RV HDF was also associated with reductions in shunt fraction (ΔQp/Qs). ASD atrial septal defect, HDF hemodynamic forces, LV left ventricular, RV left ventricular, EDV end-diastolic volume