Abstract Body (Do not enter title and authors here): Background
Electrocardiographic (ECG) signals collected inside the magnetic resonance (MR) scanner bore are distorted by gradient artefacts and magnetohydrodynamic effects, making them nondiagnostic except for R-peak detection. We developed, implemented and validated a novel hardware/software system that collects diagnostic 12-lead ECGs inside the 3 Tesla(T) MR scanner bore and is capable of measuring ischemic changes during adenosine.
Methods
Reference standard 12-lead ECGs were first recorded outside the MR-environment before the CMR scan. Using 3 MR-safe electrode patches connected to signal modules and fibre-optic cables (Figure 1), in-bore ECGs were then recorded for 30 seconds prior to and during adenosine infusion (preceding image acquisition). Post-processing filters removed artefact and a subject-specific matrix was applied to reconstruct 12-lead ECGs from the raw signal. In-bore and reference ECGs were compared using Pearson’s correlation coefficient on the PQRST waveform and by manual measurements of QRS and QT durations, with repeat measurements to assess intra- and inter-observer variability. New ischemic changes during the adenosine infusion were compared with corresponding anatomical territories on quantitative and qualitative in-line stress perfusion maps.
Results
26 participants were prospectively recruited (70.8±13.0years; 34.6% female). In-bore ECGs were successfully recorded with no additional scan time. ECG reconstruction took <2 minutes per participant.
In-bore and reference ECGs (Figure 2) were strongly correlated (r=0.81, interquartile range 0.72–0.88), and mean differences in QRS and QT durations were small (12.5ms [95% limits of agreement -42.5, 17.5ms], and -5.5ms [-54.6, 43.6ms] respectively).
Ischemic ECG changes (Figure 3) during adenosine predicted inducible ischemia (stress myocardial blood flow <1.75ml/g/min) with sensitivity 91.7% (95% confidence interval (61.5, 99.8), specificity 59.3% (45.0, 72.4), positive predictive value 33.3% (18.0, 51.8), negative predictive value 97.0% (84.0, 99.9) and area under curve 0.74 (0.61, 0.86).
Conclusion
Diagnostic-quality ECGs can be reconstructed in the bore of a clinical 3T CMR scanner. In this proof-of-concept study, in-bore ECGs correlated closely with reference ECGs and showed high specificity and NPV for predicting inducible ischemia. Diagnostic in-bore ECGs have the potential to improve R-peak detection, patient monitoring and the interpretation of quantitative stress perfusion maps.