This work presents the development and validation of a new cardiac stress imaging modality combining treadmill exercise stress testing with cardiac magnetic resonance imaging (exercise CMR). CMR offers several distinct advantages over standard stress imaging modalities. It has higher spatial resolution compared to nuclear scintigraphy without exposure to ionizing radiation, provides better contrast-to-noise than echocardiography, and offers comprehensive diagnostic information by combining stress cardiac function, stress myocardial perfusion, and myocardial viability. Although pharmacologic stress CMR has important drawbacks compared to exercise stress because of the additional physiologic information derived from exercise, technical challenges with MRI-compatible exercise and patient monitoring equipment as well as imaging in the presence of deep breathing and high heart rates have previously prevented the realization of exercise CMR.
The exercise CMR concept feasibility was successfully demonstrated in 20 healthy subjects using a partially MRI-compatible treadmill located in the corner of the MRI room. In order to reduce the image artifact levels, the TSENSE and TGRAPPA methods of parallel imaging were compared under the condition of deep breathing, and both quantitative and qualitative results showed that TGRAPPA was superior. In addition, it was shown that using a 32-channel cardiac array coil resulted in improved temporal resolution.
Accurate 12-lead ECG monitoring is required during treadmill exercise and recovery, but may be distorted by the magnetohydrodynamic effect. It was determined that reliable ECG measurements could be obtained at magnetic field strengths below 70 mT measured at the aortic arch, corresponding to approximately 80 cm from the bore entrance for the Siemens 1.5T Avanto. Based on this threshold, it was shown that accurate 12-lead ECG monitoring is feasible during treadmill exercise immediately adjacent to the magnet and during supine recovery from exercise on the MRI patient table.
Next, the feasibility of exercise CMR for accurate diagnosis of ischemia was investigated in 43 patients with known or suspected coronary artery disease who were referred for treadmill nuclear imaging. Both exercise CMR and nuclear data were obtained while exercising the patient only once. It was found that exercise CMR could accurately detect coronary artery disease compared to coronary angiography as the gold standard, and had favorable accuracy compared to nuclear stress.
Although these preliminary studies were conducted using the partially MRI-compatible treadmill in the corner of the MRI room, it was necessary to minimize the time between exercise and imaging in order to detect rapidly resolving exercise-induced cardiac wall motion abnormalities, necessitating treadmill placement beside the MRI table. Therefore, a fully MRI-compatible water hydraulic treadmill was developed. This thesis presented the development and testing of the feedback control system for continuous control of treadmill speed and elevation inside the MRI room. Component selection, design of speed and elevation control, LabVIEW software implementation, failure modes and effects analysis, and implementation of safety features are described. The results of speed and elevation performance testing are presented, followed by testing in healthy subjects and cardiac patients undergoing the standard Bruce treadmill protocol to reach peak stress immediately adjacent to the MRI table.