Oral Presentation 46th Annual Meeting of the Fetal and Neonatal Physiological Society 2019

Utilising MRI to measure cardiac function in the sheep fetus (#1)

Kyung Sik S Cho 1 , Jack RT Darby 2 , Brahmdeep S Saini 3 , Mitchell C Lock 2 , Stacey L Holman 2 , Sunthara R Perumal 4 , Christopher K Macgowan 5 6 , Janna L Morrison* 2 , Mike Seed* 7 8
  1. Department of Physiology, University of Toronto, Toronto, Ontario, Canada
  2. Early Origins of Adult Health Research Group, School of Pharmacy & Medical Sciences, University of South Australia, Adelaide, SA, Australia
  3. Institute of Medical Science, Faculty of Medicine, University of Toronto, Toronto, Ontario, Canada
  4. South Australian Health & Medical Research Institute, Preclinical, Imaging & Research Laboratories, Adelaide, SA, Australia
  5. Translational Medicine, Hospital for Sick Children, Toronto, Ontario, Canada
  6. Department of Medical Biophysics, University of Toronto, Toronto, Ontario, Canada
  7. Division of Cardiology, Hospital for Sick Children, Toronto, Ontario, Canada
  8. Department of Paediatrics, University of Toronto, Toronto, Ontario, Canada

The assessment of fetal cardiac function by cardiac magnetic resonance (CMR) has always remained challenging. Although cine phase-contrast (PC), as the gold-standard in blood flow measurements, can provide cardiac output (CO) by analysis of the ascending aorta (AAo) and main pulmonary artery (MPA) flow, it does not allow comprehensive functional assessment of the fetal heart including ejection fraction (EF). Therefore, to validate this approach we compared the CO of both the left (LV) and right ventricle (RV) as measured by CMR with the CO of the great arteries as measured by PC blood flow analysis in the fetal sheep.

Merino ewes (n=7) underwent surgery at 112-120 days (d; term, 150d) to catheterize the fetal femoral artery. At 139-140d, ewes underwent fetal CMR using femoral arterial pressure for cardiac gating. Short-axis cine imaging of the fetal heart was acquired, and LV and RV were segmented by ventricular volumetry (VV) to measure EF, stroke volume (SV), LV-CO, RV-CO and combined ventricular output (CVO). LV-CO and RV-CO were also measured by cine PC acquisitions of AAo and MPA flow, respectively. Cardiac measurements were indexed to fetal weight. The VV and PC CO measurements were compared by linear regression and Bland-Altman analysis.

The fetal heart rate during MRI was 134± 6bpm (mean±SEM). EF was comparable between the ventricles: LV-EF = 54.4±1.9% and RV-EF=52.5±2.8% (P=0.5717). The LV-SV of 6.8±0.4ml was lower than RV-SV of 8.4±0.4ml (P=0.0112). The LV-CO and RV-CO by VV were 194±10ml/min/kg and 242±11ml/min/kg, respectively, compared to CO by PC-AAo of 192±8ml/min/kg and PC-MPA of 238±19ml/min/kg (Figure 1). We found good agreement between VV and PC CO measurements with a bias of -6.2ml/min/kg.

Although challenging, imaging of the fetal heart was feasible. This data suggests that VV is a valid technique to assess cardiac function, chamber sizes, and CO in the fetal sheep.5d2881fdafe09-Figure1.png