Quantitative assessment of magnetic resonance derived myocardial perfusion measurements using advanced techniques: microsphere validation in an explanted pig heart system

Journal of Cardiovascular Magnetic Resonance

Research

Pepijn van Horssen¹  Maria Siebes¹  Jeroen PHM van den Wijngaard¹  Jos AE Spaan¹  Niloufar Zarinabad²  Nicolas Smith²  Matthew Sinclair²  Eike Nagel²  Amedeo Chiribiri²  Masaki Ishida²  Geraint Morton²  Andreas Schuster³  Boris Bigalke⁴  Gilion LTF Hautvast⁵ 

[1] Department of Biomedical Engineering & Physics, Academic Medical Centre, Amsterdam, The Netherlands;Division of Imaging Sciences and Biomedical Engineering; King’s College London British Heart Foundation (BHF) Centre of Excellence; National Institute of Health Research (NIHR) Biomedical Research Centre at Guy’s and St. Thomas’ NHS Foundation Trust; Wellcome Trust and Engineering and Physical Sciences Research Council (EPSRC) Medical Engineering Centre, The Rayne Institute, St. Thomas’ Hospital, London, UK;Division of Imaging Sciences and Biomedical Engineering; King’s College London British Heart Foundation (BHF) Centre of Excellence; National Institute of Health Research (NIHR) Biomedical Research Centre at Guy’s and St. Thomas’ NHS Foundation Trust; Wellcome Trust and Engineering and Physical Sciences Research Council (EPSRC) Medical Engineering Centre, The Rayne Institute, St. Thomas’ Hospital, London, UK;Department of Cardiology and Pneumology and German Centre for Cardiovascular Research (DZHK, Partner Site Göttingen), Georg-August-University, Göttingen, Germany;Division of Imaging Sciences and Biomedical Engineering; King’s College London British Heart Foundation (BHF) Centre of Excellence; National Institute of Health Research (NIHR) Biomedical Research Centre at Guy’s and St. Thomas’ NHS Foundation Trust; Wellcome Trust and Engineering and Physical Sciences Research Council (EPSRC) Medical Engineering Centre, The Rayne Institute, St. Thomas’ Hospital, London, UK;Medizinische Klinik III, Kardiologie und Kreislauferkrankungen, Eberhard-Karls-Universitüt Tübingen, Tübingen, Germany;Philips Healthcare, Imaging Systems - MR, Best, The Netherlands;

关键词: Cardiovascular magnetic resonance;    Myocardial perfusion imaging;    Quantification;    Deconvolution;    Microspheres;    Isolated heart perfusion;   

DOI  :  10.1186/s12968-014-0082-0

 received in 2014-05-13, accepted in 2014-09-11,  发布年份 2014

来源: Springer

【 摘 要 】

BackgroundCardiovascular Magnetic Resonance (CMR) myocardial perfusion imaging has the potential to evolve into a method allowing full quantification of myocardial blood flow (MBF) in clinical routine. Multiple quantification pathways have been proposed. However at present it remains unclear which algorithm is the most accurate. An isolated perfused, magnetic resonance (MR) compatible pig heart model allows very accurate titration of MBF and in combination with high-resolution assessment of fluorescently-labeled microspheres represents a near optimal platform for validation. We sought to investigate which algorithm is most suited to quantify myocardial perfusion by CMR at 1.5 and 3 Tesla using state of the art CMR perfusion techniques and quantification algorithms.MethodsFirst-pass perfusion CMR was performed in an MR compatible blood perfused pig heart model. We acquired perfusion images at physiological flow (“rest”), reduced flow (“ischaemia”) and during adenosine-induced hyperaemia (“hyperaemia”) as well as during coronary occlusion. Perfusion CMR was performed at 1.5 Tesla (n = 4 animals) and at 3 Tesla (n = 4 animals). Fluorescently-labeled microspheres and externally controlled coronary blood flow served as reference standards for comparison of different quantification strategies, namely Fermi function deconvolution (Fermi), autoregressive moving average modelling (ARMA), exponential basis deconvolution (Exponential) and B-spline basis deconvolution (B-spline).ResultsAll CMR derived MBF estimates significantly correlated with microsphere results. The best correlation was achieved with Fermi function deconvolution both at 1.5 Tesla (r = 0.93, p < 0.001) and at 3 Tesla (r = 0.9, p < 0.001). Fermi correlated significantly better with the microspheres than all other methods at 3 Tesla (p < 0.002). B-spline performed worse than Fermi and Exponential at 1.5 Tesla and showed the weakest correlation to microspheres (r = 0.74, p < 0.001). All other comparisons were not significant. At 3 Tesla exponential deconvolution performed worst (r = 0.49, p < 0.001).ConclusionsCMR derived quantitative blood flow estimates correlate with true myocardial blood flow in a controlled animal model. Amongst the different techniques, Fermi function deconvolution was the most accurate technique at both field strengths. Perfusion CMR based on Fermi function deconvolution may therefore emerge as a useful clinical tool providing accurate quantitative blood flow assessment.

【 授权许可】

CC BY   
© Schuster et al.; licensee BioMed Central Ltd. 2014