First 36-Channel Magnetocardiographic Study of Patients with Coronary Artery Disease in an Unshielded Laboratory

Multichannel Magnetocardiography (MMCG) is a contactless technique, which is increasingly used for non-invasive study of patients (pts) with coronary artery disease (CAD) and exercise¿induced myocardial ischemia. So far the majority of MMCG studies of CAD pts had been carried out in magnetically shielded rooms. On the contrary we have recently installed the first 36-channel instrumentation for MMCG, working in an unshielded hospital laboratory equipped for intensive and interventional care of critical pts. This pilot study was aimed to evaluate if unshielded MMCG could be sensitive enough to detect ventricular repolarization (VR) abnormalities due to myocardial ischemia, and the predictive value, for the diagnosis of CAD, of several magnetocardiographic parameters. Method: A 36-channel DC-SQUID MMCG system (CMI) [1] has been used (sensitivity: 20 fT/Hz, above 1 Hz). Typical mapping time was 90 seconds. 20 patients (14 male, 6 female, age 63+/-10.5) with CAD documented by abnormal coronary angiography and SPECT, were investigated at rest. Magnetic field gradient orientation of the integral of the second quarter of the ST interval from the J-point (ST angle) and that of the T wave apex (T angle) were computed according to Hänninen et al [2], and compared with an automatic ST-T score analysis (ST-ASA) provided by the CMI system [1]. For comparison, the same parameters were calculated in 18 age-matched normal subjects. Results: ST angle (176.7 ± 119) was abnormal in 10/19 pts and in one control; T angle (97.5 ± 105.9) was normal in all but 3 pts. Control values were 55.7 ± 20.3 and 57.4 ± 11.6, respectively (sensitivity 50%; specificity 94.4%). CMI ST-ASA was abnormal in 19/20 pts and in 6/18 controls (sensitivity 95%; specificity 66.6%). Discussion: In our laboratory, MMCG sensitivity was good enough to detect VR abnormalities in pts with CAD. However, although the median values of the quantitative parameters clearly separated pts from normal controls, a certain degree of overlap was observed for some parameters, which were also affected by the baseline selection, in several cases. In this study, the sensitivity and specificity of the two methods used for quantitative analysis of VR gave different results. Furthermore VR abnormalities similar to those detected in CAD pts can be observed also in pts with other kinds of cardiomyopathy. Therefore we believe that, further work is needed to define which are the most appropriate parameters for the diagnosis of CAD with MMCG. In our selected patient population, the integration of two methods of analysis provided a high level of predictivity in differentiating pts from normals. However the specificity of MMCG to differentiate between CAD and other cardiomyopathies has not been defined yet.