Unshielded magnetocardiography contributes to non-invasively rule-out ischemic heart disease in patients with chest pain of uncertain origin

Abstract

Abstract Background Out of millions of patients (pts) screened every year for chest pain in emergency departments (EDs) only 10% to 20% are affected by an acute coronary syndrome (ACS) or angina due to ischemic heart disease (IHD). Unless ECG recording and serum markers are clearly positive, most of them must undergo complete clinical workout to exclude IHD prior to discharge. Moreover, since even effort-ECG and echocardiography don't exhaustively rule-out IHD, second-level exams are frequently needed, enhancing hospitalization time and related costs. Magnetocardiography (MCG) is increasingly reported as a sensitive, non-invasive, radiation-free method with high negative predictive value (PV) to rule out IHD. Purpose Retrospective cohort study to evaluate the reliability likelihood of unshielded MCG (uMCG) as a tool for rapid rule-out of IHD. Method Rest uMCG data of 263 patients (pts), all chest-pain free at the moment of MCG scan, were retrospectively analyzed. uMCG was recorded with a 36-channel SQUID system (intrinsic sensitivity 30 fT/√Hz, above 1Hz), for 90 seconds. With proprietary software, 13 MCG parameters were automatically calculated from T-wave magnetic field (T-MF): 5 parameters quantifying the dynamics of the MF extrema and eight, obtained from inverse solution with the equivalent magnetic dipole (EMD) model, quantifying the spatial dynamics of the T-wave effective magnetic vector (T-EMV). Pts were stratified as follows: – Group A, chest pain pts, with IHD confirmed by subsequent SPECT and coronary angiography (mean age 66±11 years, 36.7% female); – Group B, chest pain pts, with IHD excluded by subsequent SPECT or coronary angiography (mean age 59±10 years, 28.8% female); – Group C, healthy control subjects (mean age 25±7 years, 48.6% female). Only 60 pts in each group were randomly selected for the analysis. Non-parametric Kruskal-Wallis tests were used to compare the 13 MCG predictors among groups. Linear discriminant analysis was employed to assess likelihood ratios (LHR) and PVs, considering a mean IHD prevalence of 13% (range 10–20%) in chest-pain pts. Results Positive LHR was 9, while the negative LHR was 0.3, resulting in a helpful reduction of the pre-test probability of IHD. Whereas positive PV ranges from 51 to 70%, hardly confirming IHD diagnosis, negative PV ranges from 92 to 96%, giving confidence for a safe IHD rule-out. Conclusions Automatic analysis of T-MF and of T-EMV dynamics from rest MCG provided impressive negative LHR and PV values that, if confirmed in prospective multicenter trials, would rank MCG as a useful tool for quick and early triage of pts with acute chest pain and still non-diagnostic ECG and enzyme patterns, to rule-out those whose symptoms are of non-ischemic origin. Interestingly, a similar high negative PV has been recently reported using MCG QRS predictors. Thus, even better results could be expected by combining algorithms for automatic assessment of both T-wave and QRS predictors. Funding Acknowledgement Type of funding sources: None.