Dual-Venc 5D flow for Assessment of Congenital Heart Disease in Pediatrics

项目摘要/摘要 在美国，2%的活产婴儿患有先天性心脏病(CHO)。其中最严重的CHO是单脑室疾病(SVD)，即婴儿出生时只有一个足够功能的脑室。SVD通常需要一系列手术，这些手术是通过Fontan程序完成的。这会导致循环绕过右心驱动的肺循环，而是将全身静脉直接返回到肺动脉。虽然这是一个挽救生命的过程，但在生命后期的一个常见并发症是Fontan生理学失败，这通常会导致心脏移植、开始姑息治疗或死亡。尽管对这些患者进行了定期监测，但导致Fontan衰竭的潜在机制尚不清楚。 有几种Fontan衰竭模式被认为是Fontan连接内3D血流动力学改变的继发性改变。目前的标准护理成像方法(超声心动图和2D相位对比MRI)只能捕获一维血流动力学，并且具有很大的用户可变性。由我们实验室首创的4D Flow MRI技术就是为了解决这些挑战而开发的，并能够在整个心脏周期内进行体积三维速度测量。虽然这项技术已经被用于研究Fontan血流动力学，但在其他CHD中，它并不是彻底的、终身监测的最佳选择。4D Flow的扫描时间很长，不可预测，这不适合在心脏MRI研究期间需要麻醉的小孩子。此外，4D血流的速度动态范围有限，缺乏针对不同时间和空间分辨率的灵活重建，并且不能捕捉呼吸驱动的血流，这是已知在Fontan连接中重要的血流动力学特征。为了解决这些局限性，我们开发了一种名为5D Flow MRI的方法，这是一种自由运行的技术，具有灵活的压缩感知重建，可以捕获心脏和呼吸周期的3D速度。 这项提议的第一个目标是开发一种不到10分钟的双速编码(Venc)5D流动方法以及一条简化的处理和分析管道。这将使准确、同时测量静脉和动脉中的血流，同时保持足够短的时间，供需要全身麻醉的患者使用。该方案的第二个目的是验证该方法，并优化重建和采集参数。这对于了解这些指标与临床标准相比的可靠性以及确保呼吸解析流量是准确的至关重要。这项建议的第三个目标是将开发的序列应用于Fontan连接患者的队列。我假设：1.双静脉5D血流将比临床标准采集更快和更简单，2.呼吸分辨血流测量揭示了额外的血流动力学信息，3.从5D血流中获得的体素参数，如呼吸驱动的峰值速度，与肺血管阻力相关，这是一种有创测量。

Project Abstract/Summary Congenital heart disease (CHO) afflicts up to 2% of live births in the United States. Among the most severe CHO is single ventricle disease (SVD) where infants are born with only one sufficiently functional ventricle. SVD often requires a series of surgeries that is completed with the Fontan procedure. This results in a circulation that bypasses a right-heart-driven pulmonary circulation, instead, routing systemic venous return directly to the pulmonary arteries. While it is a life-saving procedure, a common complication later in life is failing Fontan physiology which often results in heart transplant, initiation of palliation, or death. Despite regular monitoring of these patients, the underlying mechanisms leading to Fontan failure are poorly understood. Several modes of Fontan failure are hypothesized to be secondary to changes in 3D blood flow dynamics inside the Fontan connection. Current standard-of care imaging methods (echocardiography and 2D phase contrast MRI) capture only 1D hemodynamics and have substantial user variability. 4D flow MRI, a technique pioneered by our lab, was developed to address these challenges, and enable volumetric, 3D velocity measurement over the cardiac cycle. While this technique has been used to study Fontan hemodynamics, among other CHDs, it is not optimal for thorough, life-long monitoring. 4D flow has long, unpredictable scan times which are not amenable to small children requiring anesthesia during cardiac MRI studies. Additionally, 4D flow has a limited velocity dynamic range, lacks flexible reconstruction for different temporal and spatial resolution, and does not capture respiratory driven flow, a hemodynamic feature known to be important in the Fontan connection. To address these limitations, we have developed a method termed 5D flow MRI, a free-running technique with flexible, compressed sensing reconstruction that captures 3D velocities along the cardiac and respiratory cycles. The first aim of this proposal is to develop an under 10 minutes dual-velocity encode (venc) 5D flow method along with a streamlined processing and analysis pipeline. This will enable accurate, simultaneous measurement of flow in veins and arteries while remaining sufficiently short for patients requiring general anesthesia. The second aim of this proposal is to validate this method and to optimize the reconstruction and acquisition parameters. This will be crucial for understanding the reliability of the metrics compared to clinical standards and to ensure that respiratory resolved flow is accurate. The third aim of this proposal is to apply the developed sequence in a cohort of patients with the Fontan connection. I hypothesize that: 1. Dual-venc 5D flow will be faster and simpler than clinically standard acquisitions, 2. Respiratory resolved flow measurements reveal additional hemodynamic information, 3. Voxel-wise parameters derived from 5D flow, such as respiratory driven peak velocities, correlate with pulmonary vascular resistance, an invasive measurement.