One of the major advantages of MCG over 12-lead ECG is the ability to study the three-dimensional localization of cardiac electrical activity. This makes MCG a potentially useful tool in the determination of source parameters to evaluate the presence and localization of changes in myocardial de- and repolarization. As described in Section 22.214.171.124, current density estimation (CDE) is one approach in the description of source parameters in order to identify differences between electrophysiolog-
ical properties in healthy subjects and patients in the state of myocardial ischemia. On the basis of a computer simulation study, Killmann et al. (1995) investigated the method of CDE to localize myocardial ischaemia. A computer model of the entire human heart was used to simulate the excitation and repolarization process in eight topographically different cases of myocardial ischemia. At the S point of the cardiac cycle, myocardial ischemia could be localized on the basis of the current density reconstruction method for the injury current, and the authors concluded that magnetocardiography might be a suitable method for the noninvasive diagnosis of ischemia in daily clinical routine.
In the field of chronic ischemia and myocardial viability, current densities for the endocardial surface of the left ventricle were calculated and applied to predefined myocardial geometry acquired from MR images (Leder et al., 1998). The areas corresponding to the infarcted segments of the heart defined by the clinical reference methods revealed markedly decreased regional current densities. CDE was also tested in the localization of ischemia in patients with single- and three-vessel disease under stress. Areas of low CDE amplitude were found to match with the scar regions, whereas a high CDE amplitude was found to correlate with areas of viable, ischemic cardiac tissue (Pesola et al., 1999a).
Using a high-resolution DC SQUID gradiometer, Uchida et al. (2001) studied changes of the current source distribution before and after coronary artery occlusion in seven rats on the basis of minimum norm estimation. The current distribution increased significantly at the ischemic area within the ST segment, whereas in the T wave the direction of the currents changed. These results supported the simulations of the infarction model by Killmann et al. (1995) and Czapski et al. (1996).
Clinically relevant results have also been obtained using a one-channel system in an unshielded setting. On the basis of CDE within the ST-T interval at rest, differences could be shown between normals and a group of 52 CAD patients (Hailer et al., 2003). The results were based on a classification system derived from quantification of the symmetries or asymmetries of the current density vector (CDV) maps in the phase of ventricular repolarization. During normal repolarization, the underlying electrical activity should be coordinated and the current distributions represented in the CDV maps should be primarily characterized by currents in a left- and downward direction. Disturbances in repolarization ought to affect the symmetry of the maps, and these asymmetries were quantified on the basis of the weighted sums of the directions of the vectors. Normal CDV maps were classified as 0 and with the classes 1-4 indicating an increasing deviation from the normal direction and from a dipolar pattern. The results confirmed the initial hypothesis: in healthy subjects most maps were classified as category 0, 1 or 2, whereas in CAD patients the categories 3 and 4 (i.e., the asymmetrical map patterns) prevailed. The same approach showed that, after successful coronary interventional therapy in CAD patients, the classification tended toward normal maps (Hailer et al., 2005a). Confirmation was provided by one of the largest MCG studies published to date, which examined 177 patients with angiographically documented CAD (stenoses > 50%), 123 symptomatic patients without hemodynamically relevant stenosis, and 117 healthy subjects (Hailer et al., 2005b). Under resting condi tions, CAD patients could be identified on the basis of CDV map classification with a sensitivity of 73% and a specificity of 70%.
An alternative approach which makes no ECD assumptions is the visualization of magnetic field gradients as so-called pseudo current density maps (Kosch et al., 2001). Using this method, Sato and coworkers created vector arrow maps during ventricular de- and repolarization in 25 patients with IHD and in 25 healthy subjects (Sato et al., 2001). On that basis they were able to identify 88% of the patients with severe coronary lesions during repolarization. A further correlation could be found between certain map patterns (such as a rightward shift of current arrows, a multipolar pattern and a decreased time integral value) and various heart diseases, for example left ventricular overloading or prior MI. Kandori et al. (2001a) recorded 64-channel MCG in eight patients with angina and four healthy subjects, and analyzed the relationship between currents recorded before and after exercise test as well as after interventional therapy. These authors calculated the exercise-induced current for each channel as the ratio of current after exercise to that during rest. CAD patients displayed three distinct patterns in current ratio maps, depending on the affected coronary artery. After successful interventional therapy the pattern returned to that of healthy subjects.
Although single ECDs in a homogenous medium are not necessarily accurate enough for 3D localization of focal activity (see Section 2.3.3), their properties might nonetheless aid in the differentiation of ischemia. It has been shown that the orientation of ECDs calculated in a homogenous sphere model are similar in healthy subjects at selected times during the QT interval, and that the orientation in patients with CAD, with and without MI, deviated from the normal (sensitivity 80%, specificity 90%) (Van Leeuwen et al., 1999a). Particularly at T-wave apex, depending on infarct localization, the ECD rotated away from the normal direction in diverging directions.
Was this article helpful?