Histologic validation of stress cardiac magnetic resonance T1-mapping techniques for detection of coronary microvascular dysfunction in rabbits

Highlights

• The T1 response to ATP stress defined by stress ΔT1 (%) was lower in CMD rabbits compared to matched reference controls.

• The stress ΔT1 (%) was found to be robustly correlated with histopathologic evidence of MVD and CVF.

• Stress CMR T1-Mapping Technique can accurately reflect changes in microvascular structure

Abstract

Background

To investigate the diagnostic performance of stress cardiac magnetic resonance (CMR) T1-mapping for the detection of coronary microvascular dysfunction (CMD) by correlating microvascular density (MVD) and collagen volume fraction (CVF) with T1 response to adenosine triphosphate (ATP) stress (stress ΔT1) in rabbits.

Methods

Twenty-four New Zealand white rabbits were randomly divided into the CMD group induced by microembolization spheres (n = 10), sham-operated group (n = 5), and control group (n = 9). All rabbits underwent 3.0 T CMR, both rest and ATP stress T1-maps were obtained, and first-pass perfusion imaging was performed. Stress ΔT1 and myocardial perfusion reserve index (MPRI) were calculated. For the histologic study, each rabbit was sacrificed after CMR scanning. Left ventricular myocardial tissue was stained with Hematoxylin-eosin (H&E), Masson, and CD31, from which MVD and CVF were extracted. Pearson correlation analyses were performed to determine the strength of the association between the stress ΔT1 and both MVD and CVF.

Results

The stress ΔT1 values (CMD, 2.53 ± 0.37% vs. control, 6.00 ± 0.64% vs. Sham, 6.07 ± 0.97%, p < 0.001) and MPRI (CMD, 1.45 ± 0.13 vs. control, 1.94 ± 0.23, vs. sham, 1.89 ± 0.15, p < 0.001) were both lower in CMD rabbits compared with sham-operated and control rabbits. Further, the stress ΔT1 showed a high correlation with CVF (r = −0.806, p < 0.001) and MVD (r = 0.920, p < 0.001).

Conclusions

Stress T1 response strongly correlates with pathological MVD and CVF, indicating that stress CMR T1 mapping can accurately detect microvascular dysfunction.

Introduction

More than 40% of patients with angina have normal-appearing coronary arteries or non-obstructive coronary artery disease (NOCAD) on angiography [1]. Coronary microvascular dysfunction (CMD), defined as elevated microvascular resistance as evidenced by reduced coronary flow reserve (CFR) in response to pharmacological vasodilators, is the primary cause of asymptomatic angina when no evidence of CAD exists [2], [3], [4], [5]. Since evidence has shown that CMD could be a worthwhile and modifiable therapeutic target with prognostic significance [6], an accurate diagnosis of this condition is imperative. Currently, the most common method used to diagnose CMD is percutaneous coronary angiography (PCI), which detects microcirculatory resistance indices (IMR) and CFR [7]. However, PCI is an invasive procedure [8], [9], and therefore it is essential to identify an equally accurate but less invasive diagnostic approach for this condition.

Stress T1 mapping, a non-contrast cardiac magnetic resonance (CMR) method, shows considerable promise for the non-invasive assessment of myocardial ischemia and microvascular function during vasodilation stress [10], [11], [12]. Myocardial T1 values, quantified by T1-mapping, are the sum of longitudinal relaxation times of cardiomyocytes and extracellular interstitial components such as myocardial fibers and vessels [13], [14], [15]. Typically, vasodilator stress leads to an increase in myocardial blood volume, which extends relaxation times by increasing free water and produces increased myocardial T1 values [16]. Therefore, the myocardial stress T1 response (stress ∆T1) represents the percentage increase in myocardial blood volumes during vasodilator stress [17]. CMD is associated with reduced myocardial perfusion reserve that is caused by a blunted capacity for microvascular dilatation capacity during stress [18], resulting in a lower stress ΔT1 [10], [19], [20], [21]. However, due to the complexity of microvascular structures, current on the diagnosis of CMD determined by stress T1 mapping only looked at the functional and not structural changes. Thus, animal experiments are required for a deeper exploration.

In this study, we developed a well-controlled rabbit model of CMD by injecting microembolization spheres into the coronary artery. Then, using CMR at 3 Tesla, we assessed T1 response to adenosine triphosphate (ATP) stress and myocardial perfusion reserve indices (MPRI) in rabbits with CMD but without obstructive CAD and in healthy age-matched control rabbits. In addition, we measured the strength of the correlation between stress ∆T1 values and histopathologic markers, including the microvascular density (MVD) and collagen volume fraction (CVF) measurements.