In Vivo Microstructural Dynamics of the Hypoplastic Left Heart

Presenting Author(s)

• 

  Pedro Ferreira, PhD

  Senior Physicist
  Royal Brompton Hospital
  London, England, United Kingdom

Primary Author(s)

• 

  Pedro Ferreira, PhD

  Senior Physicist
  Royal Brompton Hospital
  London, England, United Kingdom

Co-Author(s)

• SK

  Sylvia Krupickova

  Consultant paediatric cardiologist
  Royal Brompton Hospital and Imperial College, London, United Kingdom
• 

  Julieta Rancati, MD

  Paediatric Cardiologist
  Royal Brompton Hospital, Hospital de Niños Pedro Elizalde & Sanatorio Franchin
  Buenos Aires, Ciudad Autonoma de Buenos Aires, Argentina
• MJ

  Melonie Johns

  Paediatric Cardiology Doctor
  Royal Brompton Hospital, United Kingdom
• AF

  Andrea Freixa-Benavente

  Paediatric Cardiology Doctor
  Royal Brompton Hospital, United Kingdom
• PK

  Pekin Kaan

  Paediatric Cardiology Doctor
  Royal Brompton Hospital, United Kingdom
• 

  Alessandra Frigiola, MD

  ACHD Consultant Cardiologist
  Guy's and St Thomas' NHS Foundation Trust
  London, England, United Kingdom
• HB

  Hannah Bellsham-Revell, MD

  Pedeatric Cardiology Consultant
  Evelina London Children’s Hospital, England, United Kingdom
• KP

  Kuberan Pushparajah, MD

  Paediatric cardiology consultant
  Evelina London Children’s Hospital/ King's College London
  London, England, United Kingdom
• EP

  Elettra Pomiato

  Paediatric Cardiology Doctor
  Royal Brompton Hospital, United Kingdom
• AD

  Alberto Di Biase, MSc

  Research Assistant
  Imperial College London
  Lonodn, England, United Kingdom
• RW

  Ricardo Wage

  CMR Radiographer
  Royal Brompton & Harefield NHS Foundation Trust, England, United Kingdom
• CM

  Camila Munoz

  Research Associate
  Royal Brompton Hospital and Imperial College, London, United Kingdom
• 

  Andrew D. Scott, PhD, FSCMR

  Associate Professor
  Imperial College London and Royal Brompton Hospital
  London, England, United Kingdom
• SN

  Sonia Nielles-Vallespin, PhD

  Senior Lecturer
  Royal Brompton Hospital and National Heart and Lung Institute, Imperial College London, United Kingdom
• DP

  Dudley Pennell

  Professor of cardiology
  Royal Brompton Hospital and Imperial College, London, United Kingdom

Background: Hypoplastic Left Heart Syndrome (HLHS) is a severe congenital heart defect characterised by underdevelopment of left cardiac structures, requiring urgent neonatal intervention for survival[1,2].

In-vivo Cardiac Diffusion Tensor Imaging (cDTI) is an emerging non-invasive technique that enables detailed characterisation of the myocardial microstructure, providing insights into the complex and dynamic organisation of cardiac muscle architecture[3].

In this study, we investigate the in-vivo myocardial microstructural dynamics in HLHS using cDTI, and compare the cardiomyocyte architecture of the HLHS heart with that of a normal heart LV.

Methods: Stimulated-echo cDTI was performed in vivo at 3T during end-systole and late-diastole in 18 patients with HLHS (age 20 ± 3 years; all post total cavopulmonary connection) and 14 healthy controls (age 25 ± 3 years). One to three matching short-axis slices were acquired per participant in both cardiac-phases. Voxel-wise analysis of diffusion tensor orientation relative to ventricular coordinates was performed for the myocardial free wall. The left ventricular free wall (LVFW) in healthy controls was compared to the HLHS right ventricular free wall (RVFW). When present, the LVFW in HLHS hearts was also assessed (Fig 1).

Tensor orientation metrics were compared: the helix angle (HA), reflecting the predominant cardiomyocyte orientation; and the secondary eigenvector angle (E2A), representing the orientation of the sheetlet structure.

Results: In the HLHS cohort, 7 patients had no discernible LV, while 11 had a hypoplastic LV, with only 2 supplying to the systemic circulation.

Helix-angle (HA) and sheetlet-angle (E2A) distributions are shown in Fig 2. HA distributions in both cohorts were centred near zero, indicating a predominance of circumferential cardiomyocytes. Patterns were similar across cohorts and cardiac phases, except for a narrower diastolic distribution in healthy hearts.

In contrast, E2A distributions demonstrated marked phase-dependent changes in the healthy LVFW and HLHS RVFW, but only small changes in the HLHS LVFW. Sheetlet mobility (change in absolute E2A between diastole and systole) (Fig 3), was conspicuous in the healthy LVFW and HLHS RVFW, with a small but statistically significant E2A mobility in the HLHS LVFW (p = 0.040).

Healthy vs HLHS comparisons revealed significant differences in diastole (p < 0.001) but not systole (p = 0.065).

Conclusion: In HLHS, the RV, the primary contributor to systemic circulation, exhibits a microstructural organisation comparable to that of the healthy LV. It demonstrates marked sheetlet reorientation between diastole and systole, although mobility is reduced compared with the healthy left ventricle, mainly in diastole. Therefore although the RV adopts a near-LV type microstructure, the dynamics of diastolic relaxation and filling are impaired, suggesting that long-term maladaption may occur. When present, the HLHS LV sheetlet mobility is markedly reduced compared to the RV.

Diagram representing the segmentation regions. A: magnitude images from left to right: normal heart; HLH with single ventricle; HLH with left ventricle. B: healthy heart segmentation region corresponding to the left ventricle free-wall (yellow). C: HLHS hearts. For single ventricle the entire wall was analysed (yellow), for hearts with a left ventricle present, the free-wall was divided in RV (yellow) and LV free-wall (orange).
Left: Helix-angle (HA) and sheetlet-angle (E2A) distribution of values for the myocardium in the healthy LVFW (green), HLHS RVFW (blue) and LVFW (orange). Right: cDTI map examples for HA and absolute E2A in two HLHS hearts at diastole and systole.
Sheetlet mobility plots: Median absolute E2A values for each subject between diastole and systole. From left to right: HLHS RVFW, HLHS LVFW, Healthy LVFW. The p-values were obtained with Mann-Whitney rank-sum and signed-rank tests.