Integrated cardiac shape and flow analysis of repaired Transposition of great arteries (TGA) patients (RF_FR_416)

Presenting Author(s)

• JB

  Jose Maria Lorenzo Bunag

  Research Trainee
  Children's National Hospital

Primary Author(s)

• JB

  Jose Maria Lorenzo Bunag

  Research Trainee
  Children's National Hospital

Co-Author(s)

• 

  Yue-Hin Loke, MD

  Associate Professor
  Children's National Medical Center
• MA

  Mithra Agamy, BSc

  Research Trainee
  Children's National Hospital
• FC

  Francesco Capuano, PhD

  Assistant Professor
  Universitat Politecnica deCatalunya Barcelona Tech, Spain
• TC

  Thomas Casto, MD

  Cardiology Fellow
  Children's National Hospital
• TP

  Tin T. Phan, MD

  Pediatrics Resident
  Children's National Medical Center

Background: Patients with repaired Transposition of Great Arteries (TGA) often develop abnormal growth of the branch pulmonary arteries (PA) and aorta (Ao) after the Lecompte maneuver. TGA patients typically have preserved right and left ventricular (RV/LV) function, however the relationship between abnormal vessel shape and intracardiac flow energetics has not been investigated. This study utilizes 4D flow MRI post-processing and statistical shape modeling (SSM) to compare the flow energetics in TGA and the potential relationship with vessel shape.

Methods: We conducted a retrospective study on patients with TGA and normal controls with cardiac magnetic resonance (CMR) imaging, including 4D flow and 3D SSFP images. CMR datasets with any stent artifact were excluded. Clinical endpoints include the most recent exercise stress test results and proximal PA pressure by catheterization. 3D models of the PA, Ao, RV and LV were segmented using 3D Slicer. 4D flow post-processing was performed via IT Flow (Cardioflow, Japan) to measure systolic/diastolic flow energy loss (FEL) and kinetic energy (KE) in PA, Ao, RV and LV. SSM was performed via ShapeWorks Studio (NIH/NIGMS CIBC) to derive mean PA/Ao shape template and scores. T-test was conducted to compare FEL and KE between the two cohorts; Pearson’s correlation coefficient was used to assess the relationship between FEL/KE, shape scores and clinical endpoints.

Results: Seventy-five TGA patients (18.9 ± 14.8 years; 1.6 ± 0.4 BSA) and forty-nine normal controls (14.9 ± 4.7 years; 1.6 ± 0.4 BSA) were analyzed. In systole, TGA patients had elevation in total FEL and average KE of PA (Figure 1A). Furthermore, TGA patients also had elevation in total FEL and average KE of PA/Ao and RV/LV during diastole (Figure 1B). Several PA shape modes were identified, including Shape Mode 5 (rounder/wider bend at the bifurcation) that correlated with KE in the PAs (Figure 2) and lower main PA pressure (r = -0.469, p=0.032). Finally, the main Ao shape mode (Gothic arch configuration) was associated with lower exercise capacity (Figure 3).

Conclusion: The intracardiac flow energetics in TGA patients are altered compared to normal controls. The systolic energetics are correlated with PA shape modes. RV/LV flow energetics indicate that ventricular filling is altered in TGA patients. Finally, the shape bends in both Ao and PA are associated with pulmonary afterload and overall exercise capacity. Future research is needed to determine the clinical significance of these correlations in TGA patients.  

(Figure 1) FEL and KE comparison between TGA and normal controls.

(Figure 2) Correlation between Shape Mode 5 (rounder/wider bend at bifurcation) and systolic KE in PA.

(Figure 3) Correlation between Shape Mode 1 (Gothic arch shape) and lower exercise capacity.