Non-invasive characterization of the area-at-risk using magnetic resonance imaging in chronic ischaemia

Abstract

Aims We investigated the performance of quantitative stress perfusion magnetic resonance imaging (MRI) as a basis for identifying and characterizing the area-at-risk subtending a chronic coronary artery (CA) stenosis.Methods and results Pigs underwent a percutaneous copper-coated stent implantation in the circumflex CA (n 11) or a sham operation (n 5). After 6 weeks, angiography and MRI were performed including cine (rest, low- and high-dose dobutamine stress), dual-bolus first-pass perfusion (rest and adenosine stress), and contrast-enhanced imaging to quantify myocardial infarction (MI). Myocardial blood flow (MBF) was quantified based on Fermi-model deconvolution and compared with microsphere measurements. On the basis of Evans blue staining, MBF thresholds to define the area-at-risk were determined by receiver-operating characteristic (ROC) analysis. CA stenosis was 94 ± 7 and infarct size (IS) 7.3 ± 3.1 of left ventricular mass. Segmental thresholds of hyperaemic MBF yielded the best performance for detecting area-at-risk. There was a good correlation between MRI and microsphere perfusion (r2 0.84, P <50 infarcted (MI), and >50 infarcted (MI) segments. MBF was reduced in at-risk vs. remote segments at rest (non-MI, 0.50 ± 0.21; MI, 0.47 ± 0.14; MI, 0.42 ± 0.14; remote, 0.84 ± 0.25 mL/min/g) and during stress (non-MI, 0.69 ± 0.09; MI, 0.66 ± 0.14; MI, 0.51 ± 0.11; remote, 1.70 ± 0.36 mL/min/g). Segmental wall thickening showed different responses to stress (remote, progressive increase during incremental stress; non-MI, increase at low-dose and discontinued at high-dose; MI, initial increase and decrease at high-dose; MI, progressive decrease). Conclusion Quantitative hyperaemic perfusion MRI accurately defines segments in the area-at-risk in chronic ischaemia, which present with different functional response to stress related to segmental IS. © 2010 The Author.