Inline Quantitative Myocardial Perfusion by CMR: Coming Online Soon?
Inline Quantitative Myocardial Perfusion by CMR: Coming Online Soon?
复制标题
CMR 在线定量心肌灌注：即将上线？
DOI：
10.1016/j.jcmg.2019.06.011
复制
发表时间：
2019
期刊：
影响因子：
--
通讯作者：
Salerno,Michael
中科院分区：
文献类型：
--
作者：
Salerno,Michael
心肌灌注定量在缺血性心脏病的评价中提供重要的诊断和预后信息。定量正电子发射断层扫描（PET）负荷心肌血流量（MBF）和心肌灌注储备（MPR）检测心外膜冠状动脉狭窄的能力已在文献中得到充分证实。多项研究表明，与相对灌注差异的视觉评估相比，负荷MBF和MPR的诊断性能有所改善（1-3）。此外，PET研究表明，即使在调整心外膜动脉粥样硬化的严重程度后，微血管疾病（MVD）导致的MPR降低与不良心血管结局独立相关（4）。尽管PET的保留MPR对于排除高危疾病（定义为左主干、3支血管疾病或2支血管疾病伴近端左前降支狭窄）具有较高的阴性预测值（NPV），但PET的低MPR对于检测高危冠状动脉疾病（CAD）仅具有中等阳性预测值（PPV）。在1项研究中，PET全局MPR检测高危CAD的最佳截止值的灵敏度为89%，但特异性仅为36%（5）。PET定量心肌灌注评估的临床应用受益于美国食品药品监督管理局（FDA）批准的用于负荷心肌灌注成像的试剂（13-N氨和82-铷）以及用于执行MBF定量的市售软件。尽管有这些优点，PET仍然受到缺乏广泛可用示踪剂的限制，因为需要回旋加速器或82-铷发生器、电离辐射和有限的空间分辨率（6）。心脏磁共振（CMR）心肌灌注脉冲序列于20世纪90年代初首次引入（7），CMR的定量于20世纪90年代末首次进行（8）。该方法的诊断和预后有用性随后在文献中得到了很好的描述（9，10）。通过应力CMR进行的MPR定量显示与使用PET成像进行的MPR定量相关（11）。研究表明，MPR的定量分析可以区分3支血管疾病和单支血管疾病，而视觉分析低估了心肌缺血负荷（12）。最近通过CMR进行的定量灌注研究提供了MBF和MPR的逐像素评估，证明了定量优于单独视觉分析的诊断有用性（13，14）。CMR灌注的更高空间分辨率可允许评估灌注的跨壁梯度，有助于区分梗阻性CAD和MVD（15）。CMR灌注成像降低了无阻塞性CAD的MVD高风险患者的MPR和负荷MBF（16）。以前，心肌灌注的量化需要大量的用户交互和处理时间，这限制了其广泛应用;然而，最近出现了用于评估心肌灌注的自动化管道，这是由首过灌注期间运动校正技术的改进驱动的（17，18）。心肌灌注自动在线评估的临床评价，其中生成灌注图
Quantification of myocardial perfusion provides important diagnostic and prognostic information in the evaluation of ischemic heart disease. The ability of quantitative positron emission tomography (PET) stress myocardial blood flow (MBF) and myocardial perfusion reserve (MPR) to detect epicardial coronary stenosis has been well established in the literature. Multiple studies have demonstrated improved diagnostic performance of stress MBF and MPR compared with visual assessment of relative perfusion differences (1–3). Furthermore, PET studies have demonstrated that reduced MPR resulting from microvascular disease (MVD) is independently associated with adverse cardiovascular outcomes, even after adjustment for severity of epicardial atherosclerosis (4). Although a preserved MPR by PET has a high negative predictive value (NPV) for excluding high-risk disease, defined as left main, 3-vessel disease, or 2-vessel disease with a proximal left anterior descending stenosis, a low MPR by PET only has moderate positive predictive value (PPV) for the detection of high-risk coronary artery disease (CAD). In 1 study, the optimal cutoff for global MPR by PET to detect high-risk CAD resulted in a sensitivity of 89% but a specificity of only 36%(5). Clinical application of quantitative myocardial perfusion assessment by PET has benefitted by having Food and Drug Administration (FDA) Àapproved agents (13-N ammonia and 82-rubidium) for stress myocardial perfusion imaging, and commercially available software for performing quantification of MBF. Despite these advantages, PET is still limited by the lack of widespread availability of tracers due to the need for a cyclotron or 82-rubidium generator, ionizing radiation, and limited spatial resolution (6).Cardiac magnetic resonance (CMR) myocardial perfusion pulse sequences were first introduced in the early 1990s (7), and quantification of CMR was first performed in the late 1990s (8). The diagnostic and prognostic usefulness of this method has subsequently been well described in the literature (9, 10). Quantification of MPR by stress CMR has been shown to correlate with MPR quantified using PET imaging (11). Studies have demonstrated that quantitative analysis of MPR can differentiate 3-vessel disease from single-vessel disease, whereas visual analysis underestimates the myocardial ischemic burden (12). Recent studies of quantitative perfusion by CMR have provided pixel-wise assessments of MBF and MPR that demonstrated improved diagnostic usefulness of quantification over visual analysis alone (13, 14). The higher spatial resolution of CMR perfusion may allow assessments in transmural gradients of perfusion, aiding in the differentiation of obstructive CAD from MVD (15). Both MPR and stress MBF are reduced by CMR perfusion imaging in patients at high risk for MVD without obstructive CAD (16). Previously, quantification of myocardial perfusion required significant user interaction and processing time, which limited its widespread application; however, automated pipelines for assessment of myocardial perfusion recently emerged, driven by improvements in techniques for motion correction during first-pass perfusion (17, 18). Clinical evaluation of automated inline assessment of myocardial perfusion, in which perfusion maps are generated