Cardiovascular magnetic resonance imaging in patients with cardiac implantable electronic devices: a device-dependent...

To prospectively determine evaluability of routine cardiovascular magnetic resonance (CMR) diagnostic modules in a referral population of implanted rhythm device all-comers, and to establish a device-dependent CMR imaging strategy to achieve optimal image quality.

One hundred and twenty-eight patients with cardiac implantable electronic devices [insertable cardiac monitoring system, n = 14; implantable loop-recorder, n = 21; pacemaker, n = 31; implantable cardioverter-defibrillator (ICD), n = 50; and cardiac resynchronization therapy defibrillator (CRT-D), n = 12] underwent clinically indicated CMR at 1.5 T. CMR protocols were tailored to the clinical indication and consisted of cine, perfusion, T1-/T2-weighted, late-gadolinium enhancement (LGE), 3D angiographic, and post-contrast cine spoiled gradient echo (SGE) scans. Image quality was determined using a 4-grade visual score per myocardial segment. Segmental evaluability was strongly influenced by device type and location with the highest proportion of non-diagnostic images encountered in the presence of ICD/CRT-D systems. Cine steady-state free-precession (SSFP) imaging was found to be mostly non-diagnostic in ICD/CRT-D patients, but a significant improvement of image quality was demonstrated when using SGE sequences with a further incremental improvement post-contrast resulting in an overall four-fold higher likelihood of achieving good image quality. LGE scans were found to be non-diagnostic in about one-third of left-ventricular segments of ICD/CRT-D patients but were artefact-free in > 94% for all other device types.

Device type and location constitute the main independent predictors of CMR image quality and thus, need to be considered during protocol adaptation. Most notably, post-contrast SGE cine imaging proved superior to conventionally used SSFP sequences. Thus, following the proposed device-dependent CMR imaging strategy, diagnostic image quality can be achieved in the majority of device patients.