Validation of QwikStarCatheter for Left Ventricular Electromechanical Mapping with NOGA XP System

Abstract

Left ventricular electromechanical mapping (LVEM) is a method for mapping the left ventricular cavity in 3 dimensions by use of a catheter that samples points on the endocardial surface. These points provide data on unipolar voltage and linear local shortening, which can then be used to evaluate myocardial ischemia and viability. The new QwikStar multi-electrode catheter, which acquires data from multiple points simultaneously, potentially improves map quality and decreases mapping time in comparison with the single-point NogaStar catheter. Our study sought to validate the QwikStar catheter's LVEM capabilities in a porcine model of chronic ischemia.

Eight pigs underwent ameroid placement over the proximal left circumflex artery, to induce chronic ischemia. In 60 days, LVEM was performed on each animal with the NogaStar and QwikStar catheters. Unipolar voltage and linear local shortening results were displayed in 9-segment polar maps. The unipolar voltage data from both maps were then correlated by means of linear regression.

There were no adverse events during LVEM. Mapping time was similar for both groups (QwikStar, 44.6 ± 25.62 min; NogaStar, 65.75 ± 25.33 min; P = 0.13). Results of mean unipolar voltage maps acquired with the 2 catheters showed a moderate correlation (r =0.59, P <0.001). Selecting segments with more than 6 point samples increased the Pearson coefficient to 0.69 (P <0.001).

Our findings show that the QwikStar catheter enables the reproducible performance of LVEM by sampling fewer points, which shortens procedure time, decreases manipulation of the left ventricular cavity, and might increase procedural safety.

Cell therapy is a promising approach for treating patients who have chronic ischemic heart diseases^1,2 and left ventricular (LV) dysfunction after acute myocardial infarction.^3,4 Among the techniques available for delivering cells to the heart,^5,6 transendocardial injection has shown the best retention rates,⁷ obviating the inherent risks of microvascular plugging and lack of reflow that are associated with intracoronary injections.⁸ Left ventricular electromechanical mapping (LVEM), a recently developed method for evaluating the LV cavity in 3 dimensions (3D), can be used to target the transendocardial injections to ischemic and viable myocardium. This method may improve the efficacy of cell delivery and engraftment, thereby increasing the cell-mediated recovery of LV function.^9,10

The standard NOGA® 4.0 mapping system (Biosense Webster [Israel], Ltd., a subsidiary of Biosense Webster, Inc., a Johnson & Johnson company; Diamond Bar, Calif) enables 3D reconstruction of the LV cavity via the consecutive sampling of points in the endocardium with use of the NogaStar catheter. Each point sample contains information about local electrical activity: unipolar voltage (UniV), expressed as mV; and local contractility or linear local shortening (LLS), expressed in percentages. This information can be used to differentiate ischemic and infarcted myocardium from normal tissue¹¹ and to predict myocardial viability.¹²

In conjunction with the new NOGA® XP mapping system, the new multi-electrode QwikStar® catheter (Biosense Webster) enables the sampling of several points simultaneously. This feature may permit the construction of maps with more points while decreasing manipulation of the LV cavity, resulting in shorter mapping times and greater mapping accuracy.

In this study, we used a porcine model of chronic ischemia to compare LVEM performance of the new QwikStar catheter with that of the NogaStar catheter.

Materials and Methods

The Texas Heart Institute's Institutional Animal Care and Use Committee approved the study protocol. The protocol adhered to the National Institutes of Health's criteria and the American Heart Association's guidelines for animal research.

Ischemic Model. Eight domestic pigs (weight, 30–60 kg) were used in this study. General anesthesia was induced with intravenous pentothal (17 mg/kg) and was maintained with isoflurane (0.5%–3%). The animals were orally intubated and were supported by mechanical ventilation throughout the procedure. A left thoracotomy was performed on each, and an ameroid constrictor was placed over the proximal left circumflex coronary artery (Fig. 1A) to induce chronic ischemia by progressive arterial narrowing. At 60-day follow-up, each animal underwent LVEM with the NogaStar and then the QwikStar catheter. Both maps were obtained by the same operator in order to avoid variability.

Fig. 1 A) Ameroid placement (arrow) over the left circumflex artery induces chronic myocardial ischemia via progressive coronary flow reduction. Left ventricular electromechanical map shows (B) results of decreased local linear shortening and (C) preserved unipolar voltage which are compatible with myocardial ischemia of the lateral wall.

AP = anteroposterior; LLS = linear local shortening

The NOGA Mapping System. In the NOGA mapping system, maps are constructed by sampling multiple points on the endocardial surface. Data from the endocardial sampling are gated with the surface electrocardiogram, enabling consistent evaluation at the same phase of the cardiac cycle. An ultra-low magnetic field (5 × 10^–6 to 5 × 10⁻⁵) is generated by a triangular magnetic pad that is positioned underneath the patient, which permits 3D orientation with 1-mm accuracy, for navigation inside the LV cavity. The result is a map of the entire endocardial surface, which contains accurate segmental information regarding contractility (Fig. 1B) and unipolar voltage (Fig. 1C). The integration of these results provides important information about myocardial ischemia and viability.^13,14

Left Ventricular Electromechanical Mapping Technique. The LVEM was performed as previously described.¹⁵ In brief, the animals received heparin, and biplane LV angiography was performed to ensure segmental wall-motion compromise after implantation of the ameroid. The mapping catheter was selected on the basis of the size of each LV cavity. The mapping catheter was advanced into the ascending aorta under fluoroscopic guidance. With tip deflected, the catheter was pushed against the aortic valve to prolapse into the LV. Once inside the LV, the catheter tip was straightened and oriented toward the LV apex. The first 3 points were acquired from the apex, the septal basal wall, and the lateral basal wall. Subsequent points were acquired until all endocardial segments were uniformly sampled, with at least 3 points per segment. Each data point was filtered online, immediately after acquisition and at the postprocessing analysis.

Differences between the NogaStar–NOGA 4.0 and the QwikStar–NOGA XP Systems. The NogaStar catheter has only 1 sensor at its tip. Besides a tip-point sensor, the QwikStar catheter has 6 sensors along its shaft that capture additional UniV information (Fig. 2), which enables simultaneous multiple sampling. Offline, the UniV map can be constructed from data gathered by the QuikStar's tip alone, or by adding data gathered by the shaft points. To ensure homogeneous quality of the maps that were created by use of the 2 different catheters, we obtained data from at least 3 points from each catheter in each of the 9 endocardial segments.

Fig. 2 QwikStar catheter with tip (white arrow) and shaft sensors (red arrows).

Map Analysis. A total of 72 polar map segments, 9 per pig, were used for comparison between the 2 catheters. The final electromechanical maps contained information on LLS and UniV and were displayed in a 9-segment polar map in a bull's-eye format (Fig. 3). Individual results from each segment were used to first compare the NogaStar map with the map that had been constructed from the QwikStar's tip data only. The NogaStar UniV map was then compared with the QwikStar map that had been constructed from both tip-point and shaft-point data. Data on mapping times, complications, and feasibility were also collected and analyzed at the end of the protocol.

Fig. 3 Polar maps show A) linear local shortening and B) unipolar voltage.

LLS = linear local shortening

Statistical Analysis. Individual UniV and LLS results from each polar map segment were recorded as mean ± SD. Linear regression was used to correlate the UniV and LLS data from the maps that were constructed as a result of the use of the 2 different catheters. The Pearson correlation indices were reported, and α values of <0.05 were considered significant.

Results

The mapping times for the NogaStar and the QwikStar catheters were similar (65.75 ± 25.33 vs 44.6 ± 25.62 min; P = 0.13). The pigs experienced no procedural sequelae, including death, cardiac perforation, ventricular fibrillation, and ventricular tachycardia.

Comparison of the Catheters: Tip Reconstruction Only. The total number of tip points obtained with the NogaStar was higher than that obtained with the QwikStar (102.12 ± 27.50 vs 63.125 ± 29.17 points; P = 0.01). This difference was expected, because the shaft of the QwikStar catheter is less flexible than that of the NogaStar and therefore less maneuverable, making it slightly more difficult to obtain tip data. Nevertheless, linear regression analysis revealed moderate correlation of mean UniV per segment between the QwikStar and NogaStar maps (r =0.41, P <0.001). In contrast, there was no correlation between the respective maps in regard to LLS data (r = 0.06, P = 0.6). When only segments that contained more than 6 tip-only points were selected, the correlation improved between UniV maps, but not LLS (Fig. 4).

Fig. 4 A Pearson correlation index for unipolar voltage shows NogaStar and tip-only QwikStar reconstruction, with more than 6 points per segment.

Comparison of the Standard NogaStar and the New QwikStar Reconstructions. As expected, the total number of points in the QuikStar shaft-point reconstruction was significantly higher than that obtained with the NogaStar (266.42 ± 71.28 vs 102.12 ± 29.44 points; P <0.001). Figure 5A shows an example of a QwikStar tip-only UniV map compared with a QwikStar map that includes shaft points (Fig. 5B). Linear regression analysis revealed moderate correlation in mean UniV per segment between the 2 maps (r = 0.59, P < 0.001). Selecting NogaStar segments with more than 6 points improved the level of correlation between NogaStar and QuikStar (Fig. 6).

Fig. 5 QwikStar reconstruction with A) tip only and B) the addition of shaft points. The addition of shaft points resulted in a more detailed map.

Fig. 6 A Pearson correlation index for unipolar voltage shows NogaStar and QwikStar (tip plus shaft) reconstruction, with more than 6 points per segment.

Discussion

Left ventricular electromechanical mapping is approved in the United States for invasive evaluation of myocardial ischemia while the patient is in the catheterization laboratory. Results from UniV maps, which correlate with the transmurality of scars found on magnetic resonance imaging, can be used for online detection of areas of myocardial viability.¹³ Numerous studies have used LVEM to target the delivery of therapeutic stem cells.1,2,16–18

Despite the acquisition of data from fewer tip points and a shorter acquisition time, the QwikStar catheter obtained data for UniV maps that correlated well with maps obtained by use of the NogaStar catheter. (This result was not reproducible for the LLS maps, because those data could be acquired only by the QwikStar's tip.) When UniV data acquired by the QuikStar's shaft were added to the data from its tip, correlation with the NogaStar UniV maps increased. These findings suggest that the QwikStar catheter enables the construction of reliable electromechanical UniV maps with the acquisition of fewer points.

Selecting segments with more than 6 points increased the level of correlation between the maps, regardless of the type of reconstruction used (with or without QwikStar shaft points). This should be taken into consideration in defining the criteria for performance of reproducible, accurate, and reliable LVEM.

Conclusion

The QuikStar catheter, although slightly less maneuverable than the standard NogaStar, enables the reproducible performance of LVEM while sampling fewer points in the LV cavity. The result is shorter procedure time with less manipulation of the heart, which might increase procedural safety.

Acknowledgment

The authors thank Pierrette Lo, MS, ELS, for editorial assistance.