A novel solid-state magnetocardiography (MCG) system outperformed electrocardiographic imaging (ECGi) in pinpointing arrhythmia sites of origin in nine swine with implanted pacing leads. Across 17 pacing datasets, median site-of-origin error was 19.6 mm vs. 31.2 mm (p=0.023) for atrial sites and 12.0 mm vs. 26.1 mm (p=0.074) for ventricular sites, benchmarked against MRI-confirmed lead-tip positions. Local activation time correlation with invasive epicardial mapping reached r=0.63–0.68 in the ventricle, with conduction velocity agreement modest but numerically favoring MCG.
The significance here extends well beyond incremental improvement. Traditional MCG required superconducting sensors cooled with liquid helium — expensive, bulky, and clinically impractical. Solid-state sensors eliminate cryogenic infrastructure entirely, potentially democratizing a technology that has languished in research labs for decades. For patients with arrhythmias, today's standard-of-care EP mapping demands vascular access, fluoroscopy radiation, and sedation — all avoidable if non-invasive pre-procedural mapping could guide ablation strategy upfront.
Critical limitations must temper enthusiasm. This is a small animal study (n=9 swine, 17 datasets), and paced rhythms in healthy myocardium poorly simulate the scarred, heterogeneous substrates seen in clinical arrhythmias like atrial fibrillation or ventricular tachycardia. Atrial performance was noticeably weaker than ventricular (r=0.40 vs. 0.63). Crucially, this is a preprint not yet peer-reviewed, and results may change materially. Still, if solid-state MCG's accuracy holds in human trials, this could meaningfully reduce procedural risk for millions undergoing EP studies annually.
