This study was performed to evaluate the feasibility of intra-procedural visualization of optimal pacing sites and image-guided left ventricular (LV) lead placement in cardiac resynchronization therapy (CRT). In fifteen patients (10 males, 68 ± 11 years, 7 with ischemic cardiomyopathy and ejection fraction of 26 ± 5%), optimal pacing sites were identified pre-procedurally using cardiac imaging. Cardiac magnetic resonance (CMR) derived scar and dyssynchrony maps were created for all patients. In six patients the anatomy of the left phrenic nerve (LPN) and coronary sinus ostium was assessed via a computed tomography (CT) scan. By overlaying the CMR and CT dataset onto live fluoroscopy, aforementioned structures were visualized during LV lead implantation. In the first nine patients, the platform was tested, yet, no real-time image-guidance was implemented. In the last six patients real-time image-guided LV lead placement was successfully executed. CRT implant and fluoroscopy times were similar to previous procedures and all leads were placed close to the target area but away from scarred myocardium and the LPN. Patients that received real-time image-guided LV lead implantation were paced closer to the target area compared to patients that did not receive real-time image-guidance (8 mm [IQR 0–22] vs 26 mm [IQR 17–46], p = 0.04), and displayed marked LV reverse remodeling at 6 months follow up with a mean LVESV change of −30 ± 10% and a mean LVEF improvement of 15 ± 5%. Real-time image-guided LV lead implantation is feasible and may prove useful for achieving the optimal LV lead position.

Comparison of the targeting accuracy of a new software method for MRI-fluoroscopy guided endomyocardial interventions with a clinically available 3D endocardial electromechanical mapping system. The new CARTBox2 software enables therapy target selection based on infarction transmurality and local myocardial wall thickness deduced from preoperative MRI scans. The selected targets are stored in standard DICOM datasets. Fusion of these datasets with live fluoroscopy enables real-time visualization of MRI defined targets during fluoroscopy guided interventions without the need for external hardware. In ten pigs (60–75 kg), late gadolinium enhanced (LGE) MRI scans were performed 4 weeks after a 90-min LAD occlusion. Subsequently, 10–16 targeted fluorescent biomaterial injections were delivered in the infarct border zone (IBZ) using either the NOGA 3D-mapping system or CARTBox2. The primary endpoint was the distance of the injections to the IBZ on histology. Secondary endpoints were total procedure time, fluoroscopy time and dose, and the number of ventricular arrhythmias. The average distance of the injections to the IBZ was similar for CARTBox2 (0.5 ± 3.2 mm) and NOGA (− 0.7 ± 2.2 mm; p = 0.52). Injection procedures with CARTBox2 and NOGA required 69 ± 12 and 60 ± 17 min, respectively (p = 0.36). The required endocardial mapping procedure with NOGA prior to injections, leads to a significantly longer total procedure time (p < 0.001) with NOGA. Fluoroscopy time with NOGA (18.7 ± 11.0 min) was significantly lower than with CARTBox2 (43.4 ± 6.5 min; p = 0.0003). Procedures with CARTBox2 show a trend towards less ventricular arrhythmias compared to NOGA. CARTBox2 is an accurate and fast software-only system to facilitate cardiac catheter therapy based on gold standard MRI imaging and live fluoroscopy.Electronic supplementary materialThe online version of this article (10.1007/s10554-019-01541-9) contains supplementary material, which is available to authorized users.

Cardiac regenerative therapies aim to protect and repair the injured heart in patients with ischemic heart disease. By injecting stem cells or other biologicals that enhance angio- or vasculogenesis into the infarct border zone (IBZ), tissue perfusion is improved, and the myocardium can be protected from further damage. For maximum therapeutic effect, it is hypothesized that the regenerative substance is best delivered to the IBZ. This requires accurate injections and has led to the development of new injection techniques. To validate these new techniques, we have designed a validation protocol based on myocardial tissue analysis. This protocol includes whole-heart myocardial tissue processing that enables detailed two-dimensional (2D) and three-dimensional (3D) analysis of the cardiac anatomy and intramyocardial injections. In a pig, myocardial infarction was created by a 90-min occlusion of the left anterior descending coronary artery. Four weeks later, a mixture of a hydrogel with superparamagnetic iron oxide particles (SPIOs) and fluorescent beads was injected in the IBZ using a minimally-invasive endocardial approach. 1 h after the injection procedure, the pig was euthanized, and the heart was excised and embedded in agarose (agar). After the solidification of the agar, magnetic resonance imaging (MRI), slicing of the heart, and fluorescence imaging were performed. After image post-processing, 3D analysis was performed to assess the IBZ targeting accuracy. This protocol provides a structured and reproducible method for the assessment of the targeting accuracy of intramyocardial injections into the IBZ. The protocol can be easily used when the processing of scar tissue and/or validation of the injection accuracy of the whole heart is desired.

Many cardiac catheter interventions require accurate discrimination between healthy and infarcted myocardia. The gold standard for infarct imaging is late gadolinium–enhanced MRI (LGE-MRI), but during cardiac procedures electroanatomical or electromechanical mapping (EAM or EMM, respectively) is usually employed. We aimed to improve the ability of EMM to identify myocardial infarction by combining multiple EMM parameters in a statistical model. From a porcine infarction model, 3D electromechanical maps were 3D registered to LGE-MRI. A multivariable mixed-effects logistic regression model was fitted to predict the presence of infarct based on EMM parameters. Furthermore, we correlated feature-tracking strain parameters to EMM measures of local mechanical deformation. We registered 787 EMM points from 13 animals to the corresponding MRI locations. The mean registration error was 2.5 ± 1.16 mm. Our model showed a strong ability to predict the presence of infarction (C-statistic = 0.85). Strain parameters were only weakly correlated to EMM measures. The model is accurate in discriminating infarcted from healthy myocardium. Unipolar and bipolar voltages were the strongest predictors.Electronic supplementary materialThe online version of this article (10.1007/s12265-019-09899-w) contains supplementary material, which is available to authorized users.

No abstract