Abstract
Maps obtained by means of electroanatomic high-density mapping (HDM) systems have shown their use in the identification of conduction gaps in experimental atrial linear lesion models. The objective of this study was to assess the use of HDM in the recognition of reconnection gaps in pulmonary veins (PV) in redo atrial fibrillation (AF) ablation procedures.
One hundred and eight patients were included in a non-randomized study that assessed the recognition of reconnection gaps in PV by means of HDM compared to a control group that received conventional non-fluoroscopic guidance with a circular multipolar catheter (CMC). Among the HDM group, adequate recognition of reconnection gaps was obtained in 60.99% of the reconnected PVs (86 of 141), a figure significantly higher than that achieved with analysis of CMC recorded signals (39.66%, 48 of 121; P = 0.001). The number of applications and total radiofrequency time were also significantly lower in the HDM group (12.46 ± 6.1 vs. 15.63 ± 7.7 and 7.61 ± 3 vs. 9.29 ± 5; P = 0.02, and P = 0.03, respectively). At the 6-month follow-up, no statistically significant differences were found in recurrence of AF or any other atrial tachycardia between the HDM group (8 patients, 14.8%) and the control group in (16 patients, 29.6%; P = 0.104).
An analysis of the high-density activation maps allows greater precision in the identification of reconnection gaps in PV, which results in lower radiofrequency time for the new isolation.
Description of gap analysis using high density mapping in redo atrial fibrillation (AF) procedures.
Comparison between high density mapping system and conventional approach using circular mapping catheter in the identification of pulmonary vein reconnection gaps in redo AF procedures.
Introduction
The electrical isolation of pulmonary veins (PV) is the cornerstone of atrial fibrillation (AF) ablation procedures. However, despite nearly 100% procedural success in achieving this objective, its clinical expression is, to a large extent, inferior.1,2 A large part of this limitation may be explained by the phenomenon of electric reconnection, which is observed in up to 59% of the PV of patients who undergo a 2nd ablation procedure for arrhythmia recurrence.3 Approximately 30% of patients undergoing ablation will require a redo procedure throughout their course.4 Regardless of the ablation technique employed in the 1st procedure, the habitual method for a redo is a point-by-point radiofrequency ablation, with non-fluoroscopic guidance and circular multipolar catheters (CMC) with a variable radius to diagnose reconnection in PV, estimate the reconnection site, and check the new isolation after ablation. New systems of electroanatomic high-density mapping (HDM) could be of use to verify and anatomically identify conduction gaps by means of an analysis of the activation maps and of the signals obtained from basket-type multi-electrode catheters, thus providing greater precision in redo procedures. The RhythmiaTM system has experimentally shown precision in the identification of conduction gaps in canine atrial linear lesion models.5 In this study, we compare performance in the identification and ablation of conduction gaps of a HDM system with that of a matched cohort of patients who underwent a redo procedure with a traditional guidance system and CMC.
Methods
This study consecutively included all patients undergoing PV re-isolation for recurrence of arrhythmia using a HDM system (Rhythmia Mapping System, Boston Scientific Corporation) in our institution between November 2015 and September 2016. The outcomes were compared with those of a control group composed of a cohort of consecutive patients who underwent a repeat ablation procedure for AF performed by the same principal operator, using a conventional guidance system (Ensite Velocity, St Jude Medical, St Paul, MN, USA).
The study was approved by our centre’s ethics committee and conducted in accordance with the ethical principles of the Declaration of Helsinki. All patients gave their informed consent prior to the procedure.
Description of the high-density mapping system-guided ablation procedure
The ablation was performed following our habitual protocol.6 Specifically, unfractionated heparin was administered after the transseptal punctures to maintain an activated clotting time of 300–350 s for the duration of the procedure, according to the manufactureŕs recommendations.
Strategy for map building
If the patient was in AF, electrical cardioversion was performed before starting the procedure.
A conventional decapolar diagnostic catheter was placed in the coronary sinus for pacing, as an electric and anatomical reference for impedance tracking. Access to the left atrium was performed by means of double transseptal puncture for the basket-type mapping catheter (IntellaMap Orion, Boston Scientific Corporation) and for the ablation catheter (Intella Nav OI or Blazer OI, Boston Scientific Corporation). Both catheters were positioned with the help of the corresponding steerable sheaths (Agilis; St Jude Medical, St Paul, MN, USA). Mapping of the left atrium was conducted with an IntellaMap Orion catheter during atrial pacing from proximal coronary sinus with a cycle length of 550 ms. The appropriate beats and electrograms were automatically selected by the system in accordance with predefined criteria. The initial criteria were: stability of cycle length with a tolerance of ±10 ms; a propagation reference with a tolerance of ±5 ms; a respiratory cycle accepting only beats during the exhalation phase; Motion = 1 mm; Stability = 0.25; Tracking = 3.
For the voltage maps, low voltage areas were considered those that presented electrograms with an amplitude less than 0.3 mV and normal voltage as those that presented electrograms with an amplitude greater than 0.5 mV.
In all cases, particular care was taken to produce a very high-density map of points at the level of venous antrum and the left crest by means of slow movement and rotation of the Orion catheter.
Gap analysis with high-density mapping
Once the map was finished, each pulmonary vein was separately analysed using a propagation and voltage map (set by default at 0.1–0.3 mV, adjusted when necessary). The propagation bar was adjusted to 5 ms to see a clearer spreading through the gaps. When a gap was detected (Figure 1 and see Supplementary material online, Video S1), a virtual roving probe was used to show the electrogram of the gap as well as the propagation across the spline of the Orion catheter, from proximal to distal electrodes (Figure 2), to avoid far field signals. In addition, when analysing left PV (in a propagation map performed during pacing from proximal coronary sinus), the window of interest was adjusted to take left atrial appendage far field signals out of the system´s analysis. Finally, with the virtual roving probe, we checked whether there was an electrogram behind the ventricular window because by default it would have been hidden by the system’s ventricular-overlap algorithm and thus not included in the propagation map.
Activation map during recognition of the reconnection gap of the left superior pulmonary vein from superior (left panel) and left anterior oblique (right panel) views. A reconnection gap can be observed in the superior region (white arrow). The map also shows that the left inferior pulmonary vein remains isolated. Map time: 13 min and 35 s. LAA Os, left atrial appendage ostium; LSPV, left superior pulmonary vein; LIPV, left inferior pulmonary vein; MA, mitral annulus.
The same map as Figure 1. After recognition of a possible gap, the virtual roving probe is positioned over it, so that not only can the local electrogram be observed but the position of the Orion catheter when the point was recorded is shown, along with the electrograms recorded for the rest of the electrodes in this exact beat. Right panel: electrograms of the purple spline (H1–2 a H7–8) in which the impulse propagation from proximal to distal (red arrow) is observed. LAA Os, left atrial appendage ostium; LSPV, left superior pulmonary vein; LIPV, left inferior pulmonary vein; RP, virtual roving probe.
Ablation strategy
After mapping, conduction gaps in the reconnected veins were defined by means of propagation of the activation maps. Focal ablation was subsequently performed in the gaps observed. All catheters used in this series were open-irrigated. For each lesion, ablation was performed for 60 s. Maximum ablation powers were 30 W for posterior gaps and 35 W for other locations. The temperature was limited to 43 °C, and the irrigation rate was between 17 and 30 mL/min for both groups. During application of the radiofrequency energy, the Orion catheter was positioned in the proximal portion of the corresponding vein and expanded to its nominal position. In the event of mechanical or electric interference with the ablation catheter, the Orion catheter was advanced to a more distal position within the vein. In all cases, isolation was confirmed with the insertion of an Orion catheter within the PVs and its subsequent expansion until maximum diameter was reached or until deformation of the catheter basket due to contact with venous walls was observed. In the event of confirmation of entrance block, pacing was performed from the equatorial electrodes of the Orion catheter to confirm exit block, defined as local capture of the pulmonary vein without exit to the atrium. Left atrial stimulation manoeuvres were performed from the left atrial appendage and from the right atrium to uncover far-field potentials if necessary. In all cases and for each of the isolated veins, an intravenous bolus of adenosine was administered with the Orion catheter inside the vein and a focal ablation was performed in the event of an observed reconnection. Isolation was confirmed in all cases a minimum 20 min after the last radiofrequency application.
Description of the standard ablation procedure
Ablation was performed on the control group following our habitual protocol and a previously published methodology.7,8 In brief, a double transseptal puncture was performed, with an electroanatomic map guided by Ensite VelocityTM (St Jude Medical, St Paul, MN, USA), using a TactiCathTM (St Jude Medical) irrigated ablation catheter and a Reflexion SpiralTM variable radius circular diagnostic catheter (St Jude Medical). Unfractionated heparin was administered after the transseptal punctures to maintain an activated clotting time of 300–350 s for the duration of the procedure.
Subsequently, the map obtained was fused with the three-dimensional reconstruction of the left atrium segmented from a contrast enhanced computerized tomography (CT), with Ensite VerismoTM software (St Jude Medical). As is customary, the ablation was performed with the application of radiofrequency at the anatomical location of the venous antrum that presented the highest tendency in the pulmonary vein electrogram, which was identified by the positioning and deployment of the circular mapping catheter at the junction of the left atrium with the mouth of the corresponding pulmonary vein. The rest of the ablation procedure, confirmation of vein entrance and exit block, waiting times, and verification of the dormant conduction with adenosine were the same as the steps performed on the HDM group, all of which form part of our centre’s practice protocol. Likewise, the sedation and anticoagulation treatment protocols (both intra- and peri-procedural) were the same in all cases.
All patients of both groups were given a three-dimensional reconstruction of the left atrium, obtained from the segmentation of a CT or cardiovascular magnetic resonance, which was used to ascertain the number, position, and size of the PV.
Definition of the conduction gap
Successful delimitation of a gap was predefined as the onset of any of the following criteria after a single focal application of radiofrequency: (i) the electrical isolation of the vein or (ii) a delay of the pulmonary vein electrograms equal to or greater than 10 ms with a change in the activation pattern of the circular catheter’s electrograms, or of the electrograms of the Orion catheter’s equatorial electrodes. Since there is no reference for the definition of gap recognition, the cut-off point was selected arbitrarily, considering that it is large enough to avoid simple variability in the measurement. To support the value of this definition the number of veins in which single radiofrequency application resulted in complete isolation, the total number of applications and radiofrequency time were also analysed. The readings were examined blind by three doctors separately, none of whom knew the opinion of the other two. Cases where one of the reviewers disagreed with the diagnosis of successful gap recognition were classified as unsuccessful.
Follow-up
Patients were followed up in the outpatient clinic of our institution according to our standard practice. Follow-up visits including 12-lead ECG and 48-h Holter ECG were scheduled 3 and 6 months after discharge and additionally non-scheduled visits were performed if patients presented with any symptoms suggestive of AF. Episodes of AF or other atrial arrhythmias lasting longer than 30 s were considered for analysis. Episodes that occurred after ablation with a blanking period of the first 3 months were considered to indicate a recurrence of AF. Antiarrhythmic drugs were usually continued for the first 3 months after ablation and then systematically discontinued.
Statistical analysis
Quantitative variables are presented as a mean ± standard deviation or as the median and 25 and 75 percentile. Categorical variables are presented as a number and a percentage. The intervals between the two groups of patients were analysed by means of Student's t-test for independent samples. Fisher's exact test or the χ2 test were used to compare qualitative variables. Statistical significance was defined as a value of P < 0.05 in two-tailed tests. The statistical analysis was performed using version 22 of SPSS for Windows (SPSS Statistics, IBM Software Group).
Results
A total number of 56 patients underwent re-isolation of PV guided by the HDM system; two patients were excluded from the study as they did not present any PV reconnection, so finally 54 patients were analysed. The results were compared with a cohort of 54 consecutive patients who underwent pulmonary vein re-isolation performed by the same principal operator with a conventional non-fluoroscopic guidance system. No significant differences were observed in the baseline characteristics of the two groups regarding age, sex, CHA2DS2VASc score or type of index procedure (Table 1). In one patient of the HDM group, because there was frequent induction of AF, partial maps of the pair of ipsilateral veins (instead of the whole atrium) followed by ablation during sinus rhythm were performed.
Baseline characteristics
| Baseline characteristics | High density mapping | Control group | P-value |
|---|---|---|---|
| (n = 54) | (n = 54) | ||
| Age (years) | 66.24 | 64.48 | 0.392 |
| Sex (M/F) | 40/14 | 37/17 | 0.671 |
| AF type (paroxismal/persistent) | 40/14 | 37/17 | 0.523 |
| CHA2DS2-VASc Score (median) | 2 | 2 | 0.836 |
| Left ventricle ejection fraction (%) | 63.33 | 64.65 | 0.358 |
| Left atrium diameter (milimetres) | 43.09 | 42.10 | 0.189 |
| Method used in initial procedure | |||
| Cryoablation (%) | 7.4 | 9.2 | 0.728 |
| Radiofrequency (%) | 92.6 | 90.8 | 0.526 |
| More than one previous ablation | 3 | 2 | 0.647 |
| Baseline characteristics | High density mapping | Control group | P-value |
|---|---|---|---|
| (n = 54) | (n = 54) | ||
| Age (years) | 66.24 | 64.48 | 0.392 |
| Sex (M/F) | 40/14 | 37/17 | 0.671 |
| AF type (paroxismal/persistent) | 40/14 | 37/17 | 0.523 |
| CHA2DS2-VASc Score (median) | 2 | 2 | 0.836 |
| Left ventricle ejection fraction (%) | 63.33 | 64.65 | 0.358 |
| Left atrium diameter (milimetres) | 43.09 | 42.10 | 0.189 |
| Method used in initial procedure | |||
| Cryoablation (%) | 7.4 | 9.2 | 0.728 |
| Radiofrequency (%) | 92.6 | 90.8 | 0.526 |
| More than one previous ablation | 3 | 2 | 0.647 |
Baseline characteristics
| Baseline characteristics | High density mapping | Control group | P-value |
|---|---|---|---|
| (n = 54) | (n = 54) | ||
| Age (years) | 66.24 | 64.48 | 0.392 |
| Sex (M/F) | 40/14 | 37/17 | 0.671 |
| AF type (paroxismal/persistent) | 40/14 | 37/17 | 0.523 |
| CHA2DS2-VASc Score (median) | 2 | 2 | 0.836 |
| Left ventricle ejection fraction (%) | 63.33 | 64.65 | 0.358 |
| Left atrium diameter (milimetres) | 43.09 | 42.10 | 0.189 |
| Method used in initial procedure | |||
| Cryoablation (%) | 7.4 | 9.2 | 0.728 |
| Radiofrequency (%) | 92.6 | 90.8 | 0.526 |
| More than one previous ablation | 3 | 2 | 0.647 |
| Baseline characteristics | High density mapping | Control group | P-value |
|---|---|---|---|
| (n = 54) | (n = 54) | ||
| Age (years) | 66.24 | 64.48 | 0.392 |
| Sex (M/F) | 40/14 | 37/17 | 0.671 |
| AF type (paroxismal/persistent) | 40/14 | 37/17 | 0.523 |
| CHA2DS2-VASc Score (median) | 2 | 2 | 0.836 |
| Left ventricle ejection fraction (%) | 63.33 | 64.65 | 0.358 |
| Left atrium diameter (milimetres) | 43.09 | 42.10 | 0.189 |
| Method used in initial procedure | |||
| Cryoablation (%) | 7.4 | 9.2 | 0.728 |
| Radiofrequency (%) | 92.6 | 90.8 | 0.526 |
| More than one previous ablation | 3 | 2 | 0.647 |
The results are detailed in Table 2. In the HDM group, mean valid points per map were 13 425 ± 5422 and of the 221 PV explored, 141 (63.8%) presented reconnection. In 86 of the 141 reconnected PVs (60.99%), isolation of the vein or an evident delay in the electrograms and a change in the entrance pattern was achieved after a single focal application in the conduction gap defined in the activation map propagation. For the remaining reconnected veins (55 of 141, 39%), analysis of the activation map could not predict a successful ablation point, either because no clear gap was observed (39 of 141, 27.65%) or due to the fact that the focal ablation in the supposed gap did not achieve vein isolation (16 of 141, 11.34%). Among the control group, of the 216 veins explored, 121 presented reconnection (56%), and in 48 of the 121 reconnected PVs (39.66%), the initial pattern of electric activation of the circular catheter defined a conduction gap according to the previously stated criteria (P = 0.001, compared to the HDM group) (Figure 3). Isolation of the vein was achieved after a single focal application in 35.4% (50 of 141) in the HDM group and in 21.4% (26 of 121) in the control group (P = 0.014).
Results
| High density mapping | Control group | P-value | |
|---|---|---|---|
| (n = 54) | (n = 54) | ||
| Reconnected PV | 141/221 (63.8%) | 121/216 (56%) | 0.09 |
| Successful gap recognition | 60.99% | 39.66 | 0.001 |
| PV isolation after single focal application | 35.4% | 21.4% | 0.014 |
| Number of RF applications | 12.46 ± 6.1 | 15.63 ± 7.7 | 0.02 |
| RF time (minutes) | 7.61 ± 3 | 9.29 ± 5 | 0.03 |
| Procedure time (minutes) | 142.2 ± 42 | 138 ± 44 | 0.64 |
| Complications (%) | 3 (5.5%) | 3 (5.5%) | 1 |
| Pericardial effusion | 0 | 1 | |
| Groin haematoma | 1 | 1 | |
| AV femoral fistula | 1 | 0 | |
| Pericarditis | 1 | 1 | |
| Catheter entrapment | 0 | 0 |
| High density mapping | Control group | P-value | |
|---|---|---|---|
| (n = 54) | (n = 54) | ||
| Reconnected PV | 141/221 (63.8%) | 121/216 (56%) | 0.09 |
| Successful gap recognition | 60.99% | 39.66 | 0.001 |
| PV isolation after single focal application | 35.4% | 21.4% | 0.014 |
| Number of RF applications | 12.46 ± 6.1 | 15.63 ± 7.7 | 0.02 |
| RF time (minutes) | 7.61 ± 3 | 9.29 ± 5 | 0.03 |
| Procedure time (minutes) | 142.2 ± 42 | 138 ± 44 | 0.64 |
| Complications (%) | 3 (5.5%) | 3 (5.5%) | 1 |
| Pericardial effusion | 0 | 1 | |
| Groin haematoma | 1 | 1 | |
| AV femoral fistula | 1 | 0 | |
| Pericarditis | 1 | 1 | |
| Catheter entrapment | 0 | 0 |
RF, radiofrequency; PV, Pulmonary vein; AV, Arteriovenous.
Results
| High density mapping | Control group | P-value | |
|---|---|---|---|
| (n = 54) | (n = 54) | ||
| Reconnected PV | 141/221 (63.8%) | 121/216 (56%) | 0.09 |
| Successful gap recognition | 60.99% | 39.66 | 0.001 |
| PV isolation after single focal application | 35.4% | 21.4% | 0.014 |
| Number of RF applications | 12.46 ± 6.1 | 15.63 ± 7.7 | 0.02 |
| RF time (minutes) | 7.61 ± 3 | 9.29 ± 5 | 0.03 |
| Procedure time (minutes) | 142.2 ± 42 | 138 ± 44 | 0.64 |
| Complications (%) | 3 (5.5%) | 3 (5.5%) | 1 |
| Pericardial effusion | 0 | 1 | |
| Groin haematoma | 1 | 1 | |
| AV femoral fistula | 1 | 0 | |
| Pericarditis | 1 | 1 | |
| Catheter entrapment | 0 | 0 |
| High density mapping | Control group | P-value | |
|---|---|---|---|
| (n = 54) | (n = 54) | ||
| Reconnected PV | 141/221 (63.8%) | 121/216 (56%) | 0.09 |
| Successful gap recognition | 60.99% | 39.66 | 0.001 |
| PV isolation after single focal application | 35.4% | 21.4% | 0.014 |
| Number of RF applications | 12.46 ± 6.1 | 15.63 ± 7.7 | 0.02 |
| RF time (minutes) | 7.61 ± 3 | 9.29 ± 5 | 0.03 |
| Procedure time (minutes) | 142.2 ± 42 | 138 ± 44 | 0.64 |
| Complications (%) | 3 (5.5%) | 3 (5.5%) | 1 |
| Pericardial effusion | 0 | 1 | |
| Groin haematoma | 1 | 1 | |
| AV femoral fistula | 1 | 0 | |
| Pericarditis | 1 | 1 | |
| Catheter entrapment | 0 | 0 |
RF, radiofrequency; PV, Pulmonary vein; AV, Arteriovenous.
Successful recognition of the reconnection gap for the HDM group (86 of 141; 60.99%) and the control group (48 of 121; 39.66%; P = 0.001).
The total number of applications and radiofrequency time were significantly lower in the HDM group (12.46 ± 6.1 and 7.61 ± 3.) than in the control group (15.63 ± 7.7 and 9.29 ± 5; P = 0.02 y P = 0.03, respectively) (Figure 4).
Comparison of mean and confidence interval of 95% of the total number of radiofrequency applications (left panel) and total radiofrequency time (right panel) between the HDM group and the control group. The total number of applications and radiofrequency time were significantly lower in the HDM group (12.46 ± 6.1 and 7.61 ± 3) than in the control group (15.63 ± 7.7 and 9.29 ± 5; P = 0.02 y P = 0.03, respectively).
In both groups, entrance block was achieved in all of the reconnected PVs. No significant differences were observed between the two groups in mean procedure time (142.2 ± 42 minutes in the HDM group, 138 ± 44 in the control group, P = 0.64). Although the absence of atrial capture during pacing within the pulmonary vein was achieved in 100% of isolated veins, confirmed exit block, defined as a clear observation of local electrogram capture within the pulmonary vein without conduction to the atrium, was shown in 68.79% of the HDM group and in 47.93% of the control group (P = 0.001).
There were no thromboembolic complications or differences in the global incidence of complications between the two groups.
At the 6-month follow-up, no statistically significant differences were found in recurrence of AF or any other atrial tachycardia between the HDM group (8 patients, 14.8%) and the control group in (16 patients, 29.6%; P = 0.104).
Discussion
Pulmonary vein isolation redo procedures are increasingly habitual in clinical practice. They are, however, of limited effectiveness,9 which suggests the need to improve this approach. In this article, for the first time in the literature, we describe the peri-procedural results of employing a new technology, HDM, in the identification and anatomical location of conduction gaps in patients with AF recurrence after an initial procedure to isolate PV.
The findings of this study suggest that an analysis of the activation patterns obtained from HDM is useful guidance for PV isolation redo procedures. Specifically, in our series, the detailed analysis of cardiac activation fronts based on a study of the high-density maps allowed a more precise identification of reconnection gaps, facilitating greater precision in planning the ablation strategy to adopt, which resulted in a reduction of the number of applications and total time of radiofrequency, without increasing procedural time or the number of complications. In addition, and despite discrepancies in the clinical expression of this fact, exit block with a basket-type catheter was shown to be more frequent than with the CMC, probably due to the better apposition of the electrodes of the former on muscle tissues within the PV.
Although the difference did not reach statistical significance, a higher percentage of reconnected PV was observed among the HDM group (63.8% vs. 56%, P = 0.09); this could simply be a random incidence or due to a greater sensitivity of the IntellaMap Orion catheter when detecting remnants of unisolated pulmonary vein tissue that cannot be detected with conventional CMC, a fact that has already been observed in previous studies.10 This phenomenon could be explained by an incomplete apposition of CMC in different venous anatomies. Due to the different take offs of the PVs, CMC are usually obliquely oriented with the anterior part of the mapping catheter being positioned more profoundly within the vein and the posterior part closer to the atrium. The inferior part of the circular mapping catheter is typically deeper within the superior PVs, whereas the opposite is true for the inferior PVs. Additionally, in small PV, as frequently occurs with the intermediate right vein, or in those that rapidly branch, as frequently occurs with the lower right PV, it may be difficult to position the deployed circular catheter in the vein antrum, on occasion causing loss of the benefits of verification with a multipolar catheter. In cases of common trunks, the possibility to map all the subsidiary branches and thereafter, to obtain a detailed electroanatomical map of the whole antrum, identifying the previous ablation line with the voltage map could be also advantageous. On the other hand, a basket-type catheter allows simultaneous positioning of the electrode wires in longitudinal and radial directions that provides information about the activation on several transversal planes at different levels which, added to a better placement of the electrodes on the wall as a consequence of the possibility of expanding the Orion catheter, could help in the more thorough demonstration of real disconnection of the PV (Figure 5).
Reconnection of right superior pulmonary vein in a patient with a prior cryoablation procedure. Left panel: electrograms observed within the vein using the IntellaMap Orion catheter. Right panel: electrograms observed within the vein using a CMC. An atrial pacing spike (yellow arrows) can be seen in both. With the Orion catheter it is possible to observe pulmonary vein electrograms that could not be appreciated with a circular mapping catheter, even after several attempts at repositioning the catheter. A new isolation was achieved after a focal application of radiofrequency.
We decided to primarily use the information from the activation map instead of the voltage map in the HDM group because we consider that the exclusive analysis of the voltage maps usually shows the sites where there may be reconnection, but is less specific than analysis of the activation map. The rationale for this strategy is that activation maps show not only how the activation front is moving around the mapped chamber but also where the lines of block are located on the activation wavefront. Moreover, voltage maps are influenced by the amplitude of the recorded electrodes. The amplitude of a bipolar electrogram is influenced by the direction of wavefront propagation, with signal amplitude largest when the wavefront is propagating parallel to the axis of the recording electrodes, and reduced when the propagation wavefront is perpendicular to the electrode.11 In addition, when there is a gap in a line of block voltage maps, a higher amplitude is usually shown where there is a block (particularly if the wavefront propagation is parallel to the axis of the recording electrode) than where there is a gap. This is due to the fact that quite often gap electrograms are fractionated and with low amplitude. In this series, we retrospectively reviewed the 50 PV maps in which isolation was achieved after a single focal application. In only 34% of cases would the gap have been adequately recognized using the voltage map alone. In the rest, the gap was undetectable in 48.4%, overdetection (a wider gap than that which finally existed, or more than one gap when in fact there was only one) was observed in 36.3% (Figure 6 and see Supplementary material online, Video S1), and imprecision in the exact position of the gap was found in 15.1%. In our experience, analysis of the voltage map was particularly unhelpful in atria with diffuse fibrosis and wide scar areas.
Comparison between voltage and activation maps of right superior pulmonary vein gap. Left: voltage map from an anterior view which indicates a wide gap in the anterosuperior area (white arrows). Centre: activation map which indicates a line of conduction block (white arrows) and an anterosuperior focal reconnection gap on which a single radiofrequency application isolated the vein (blue arrow and black dot). Right: local bipolar electrogram observed during revision with the virtual roving probe.
With respect to catheter contact, the Rhythmia system provides an indicator of closeness of the tip of the catheter to the created surface. In our experience, once a good-quality map is created, this indicator is highly reliable. Nevertheless, it is possible that the addition of contact information could have improved the results of the HDM group.
Finally, one could argue that, clinically, the importance of clear gap detection in re-do procedures and therefore, a more precise ablation of the PV antrum is completely unclear, and that maybe even a more extensive approach could be associated with a higher success rate due to a lower reconnection rate. However, our results suggest that that a better mapping and understanding of gaps and channel detection is safe and has potential clinical advantages. In any event, after achieving this mandatory endpoint, the operator can choose whether to perform a more extensive ablation in this area or in another of his choosing.
As a conclusion of this study, an analysis of the high-density activation maps allows greater precision in the identification of conduction gaps in AF ablation re-do procedures, which allows a reduction in the number of radiofrequency lesions necessary during the procedure. Although we observed a trend to better outcome in the HDM group, the clinical expression of this method should be tested in future studies.
Limitations of the study
The obvious limitations of this study are the lack of randomization of the control group and the lack of long-term follow-up.
The definition of successful delimitation of a gap and the electrogram delay cut-off point were selected arbitrarily and may be questionable. Against that, the fact that there were also significant differences in the total number of radiofrequency applications and in the number of isolated veins after a single focal application, two indicators of unquestionable value, suggest an appropriate validity of the endpoint used. Moreover, because in the control group no comparing mapping protocol was used, we do not claim that this is an exclusive benefit of a specific non-fluoroscopic navigation system; rather we propose a potential benefit of HDM by comparing the standard practice (ablation guided primarily by CMC electrograms) against a novel approach (focal ablation guided exclusively by the analysis of the activation pattern obtained by HDM).
Although a reduction in the applications of radiofrequency was observed, this study does not have sufficient statistical power to show differences in the rate of complications. It also has no statistical power to demonstrate differences in the rate of recurrence in such a short follow-up period, so that the lack of differences should be interpreted with caution. In the future, randomized studies with a wider sample will be necessary to assess whether a significant long-term clinical benefit exists.
Finally, this study does not analyse the cost of the procedure. Given the additional cost that this system currently implies, it would be advisable to have specific studies available in order to select the groups of patients for whom this system would be most cost effective.
Conclusions
The analysis of high-density activation maps allows greater precision in the identification of pulmonary vein reconnection gaps compared to the usual approach guided primarily by the analysis of PV electrograms from the CMC.
The clinical expression of this potential advantage should be explored in studies designed for this purpose.
Conflict of interest: Dr García-Bolao has received consultant, proctoring and speaker's fees from Boston Scientific Corporation and St. Jude Medical. Ane Erkiaga is an employee of Boston Scientific Corporation.
References
Author notes
The Ignacio García-Bolao and Gabriel Ballesteros authors contributed equally to the study.
