Is magnetocardiography a breakthrough medical device ready to support first-line clinical diagnosis of myocardial ischemia?
D Brisinda, M Picerni, P Fenici, R FeniciAbstract
Abstract
The first evidence that magnetocardiography (MCG) detects ischemia-induced electrophysiological abnormalities undetected by ECG was reported in 1975 [1]. Subsequent SQUID-based MCG studies suggested higher diagnostic accuracy of resting MCG compared with ECG [2]. Since 2020, a novel MCG device based on optical magnetometry (OPM) has received two breakthrough device designations from the US FDA and was approved by Health Canada as a supportive method to facilitate the diagnosis or exclusion of ischemic heart disease (IHD), with or without coronary obstruction, particularly in patients (pts) presenting to the ED with chest pain of uncertain origin and normal or non-diagnostic troponin and ECG findings [3]. However, despite an increasing number of clinical studies (now including thousands of patients) further supporting its potential diagnostic utility and socio-economic benefits [3,4], clinical adoption of MCG remains uncertain.
Objective
To analyze MCG diagnostic accuracy for IHD reported in the most recent studies and in a selection of previous (2005–2022) studies, comparing the performance of SQUID-based and OPM-based devices, and to examine the reasons for the persistent limitation of large-scale clinical MCG application despite growing evidence and recent regulatory recognition.
Methods
Thirty MCG studies [20 SQUID-MCG (15 unshielded) and 10 OPM-MCG] were selected to compare their statistical results and evaluate potential differences related to MCG technologies, signal processing, and analytical parameters used.
Results
Overall, statistical results obtained from MCG data of 6,871 IHD pts and 3,259 control subjects were analyzed. For IHD diagnosis, pooled MCG sensitivity (Se), specificity (Sp), positive predictive value (PPV), negative predictive value (NPV), and predictive accuracy (PA) were (%): 83.1, 82.2, 80.9, 84.6, and 83.6, respectively, and the area under the ROC curve (AUC) was 0.86 (Table 1). Cumulative statistics of SQUID-MCG studies were 85.6, 83.5, 80.8, 85.7, 84.6, and 0.90, respectively, with significantly lower OPM-MCG Se and AUC (p<0.01) (Table 2). Considerable heterogeneity in technology, signal processing, analytical methods, and statistical results was observed across studies.
Conclusions
The cumulative results of 30 MCG studies confirm that the average PA of MCG in diagnosing IHD exceeds 83% (with higher Se for SQUID-MCG), comparable to that of established noninvasive functional and imaging stress tests [5]. This may support the use of MCG as an effective, noninvasive, radiation-free tool for first-line IHD screening and more targeted referral to second-level stress tests or costly imaging methods involving radiation exposure. However, the observed heterogeneity highlights the need for a coordinated interdisciplinary working group to define MCG standardization criteria and consensus-based technical and clinical guidelinesTable 1.Pooled MCG statistics 2005-2026Table 2.SQUID-MCG vs OPM-MCG statistics