E172
Visualization of Atrioventricular Nodal Reentry Tachycardia Slow Pathways Using Voltage Mapping for Pediatric Catheter Ablation David W. Bearl, MD,*† LuAnn Mill, RN,*† John D. Kugler, MD,*† John L. Prusmack, BA, CEPS,‡ and Christopher C. Erickson, MD*† *Department of Pediatrics, Children’s Hospital and Medical Center, †University of Nebraska Medical Center, Omaha, Neb and ‡St. Jude’s Medical, S.C., Inc., Austin, Tex, USA ABSTRACT
Background. Catheter ablation of the slow atrioventricular (AV) pathway has been shown to be safe and effective in pediatric patients with atrioventricular nodal reentrant tachycardia (AVNRT). Despite that, acute success is not guaranteed, and safety of ablating near the AV node remains a concern. Methods. A retrospective analysis was performed of all AVNRT ablations using the Ensite NavX voltage mapping technique at our institution. Each map was reviewed with patient and NavX computer data recorded. To account for a learning curve, each map was idealized and compared with the original map. Procedure and fluoroscopy time were compared with a control group. Results. Twenty-eight patients underwent catheter ablation for AVNRT from September 2011 until December 2012 using the voltage mapping technique. The historical control group comprised 24 patients who underwent catheter ablation using the electroanatomic approach. There was 96% acute success with one recurrence in the voltage mapping group, at a mean follow-up of 24 months. The slow pathway was visualized in 86% of patients at the time of ablation, while three of four without could be found on idealization of the voltage map. Mean high- and low-voltage parameters increased with idealization, but showed no correlation with age, gender, or weight. Estimated pathway size had significant inter-patient variability. Procedure and fluoroscopy times did not vary significantly compared with controls. Conclusion. Visualization of the AV nodal slow pathway in a pediatric population is possible using voltage mapping technique with the potential to increase safety and efficacy. Variability exists in the voltage parameters needed to visualize individual slow pathways, which leads to a distinct learning curve. Key Words. Pediatric AVNRT; AVNRT; Catheter Ablation; Voltage Mapping; Slow Pathway Visualization
Introduction
A
trioventricular nodal reentrant tachycardia (AVNRT) is a relatively common mechanism of supraventricular tachycardia in children. Catheter ablation has become a common patient/family management choice.1,2 The ablation target for AVNRT is the slow atrioventricular (AV) nodal pathway. More than one slow pathway (SP) can be present, but the overall incidence of multiple SPs, even in adults, is ill defined—ranging from 5 to 40%.3,4 Catheter ablation traditionally has relied Disclosure of grants or other funding: None. Congenit Heart Dis. 2015;10:E172–E179
on fluoroscopy to position catheters in the heart up until the late 1990s, when three-dimensional (3D)3 electroanatomic mapping was introduced and validated.5,6 The electroanatomic approach involves both fluoroscopic anatomic mapping using catheters as landmarks and electrical activation mapping searching for SP potentials in the posteroseptal region. Now computerized 3D mapping has become standard and has allowed for a substantial reduction in fluoroscopy during catheter ablation.7 The challenge of ablating the SP is ensuring that lesions are placed a safe distance from the compact AV node while achieving permanent SP ablation. Injury to the AV node can © 2015 Wiley Periodicals, Inc.
Voltage Mapping in Pediatric AVNRT cause AV block with subsequent need for a permanent pacemaker. While the electroanatomic approach is associated with a high rate of success, if the targeted SP is not in the safe posteroseptal region, lesions typically are successively placed closer and closer to the AV node until success is achieved or the operator chooses not to risk closer lesions to the AV node. Because anatomic variability, atypical SP locations, and multiple SPs can complicate ablation procedures, visualization of the SP could increase safety as well as enhance success. Voltage mapping is a technique that allows for visualization of voltage gradients in relation to anatomic locations and has been previously validated.8 Based on a report by Bailin et al.9 that described use of voltage mapping for AVNRT ablations with a St. Jude NavX 3D mapping system (St. Jude Medical, St. Paul, MN, USA), the use of this technique was begun at our institution. This technique involves the simultaneous collection of anatomic points and atrial electrograms. After data collection is complete, the EnSite NavX system creates both voltage and propagation maps of the region. In the posteroseptal region, there are low-voltage bridges or narrow myocardial isthmuses that connect high-voltage areas. These bridging isthmuses of low voltage are predicted to be SPs. Catheter ablation lesions are then directed at the SPs. The purpose of this study is to retrospectively assess the utility and effectiveness of voltage mapping described by Bailin et al. to visualize the SP in AVNRT for a pediatric population. A secondary purpose is to assess the characteristics of the visualized SPs including the voltage parameters and dimensions. Methods
Data Collection This protocol was approved by the University of Nebraska Medical Center and Children’s Hospital and Medical Center Institutional Review Board (Omaha, NE, USA). Patients with documented AVNRT who underwent SP ablation using the voltage mapping technique were identified retrospectively from an internal database at Children’s Hospital and Medical Center. Twenty-eight patients underwent SP ablation using the voltage mapping technique between November 2011 and December of 2012. Patient data were reviewed including date of procedure, ablation technique (radiofrequency ablation [RF], cryoablation, or both), procedure time, fluoroscopy time, number
E173 of lesions (including extra lesions or lesions placed after success was achieved), success and complications both at the time of procedure, and follow-up. Data were also collected from the EnSite NavX computer, which included interpolation, exterior and interior projection, final high- and lowvoltage parameters, number of anatomic/ electrographic points collected as well as number of points unused. Number of points used for each map was calculated. All cases were reviewed retrospectively including the NavX maps and objective parameters. An “idealized” map was created by adjusting the upper and lower voltage limit filters to subjectively better define the SP. In the four cases where voltage mapping did not identify an SP during the procedure, the upper and lower voltage limit filters were adjusted to see if a lowvoltage bridge could retrospectively be identified. Mean high- and low-voltage parameters, mean range between high- and low-voltage parameters, and differences between parameters from original voltage maps and idealized maps were plotted against several patient characteristics including age and weight, as well as chronologically. The SP(s) visualized were measured for length and width in millimeters using the His catheter electrodes as the calibration reference. A control group was created using the same internal database by identifying children who had undergone AVNRT ablation prior to the use of voltage mapping. A total of 24 patients that had complete comparable data were identified sequentially, working retrospectively from the initial patient who received ablation with the voltage mapping technique. Data collected from December 2008 to June 2011 included procedure time, fluoroscopy time, and total number of lesions (including extra lesions). Data from the control group were compared with the study group using a Students’ nonpaired t-test.
Procedures All procedures were performed under general anesthesia. Electrophysiology catheters were placed in the high right atrium, coronary sinus, AV junction/His bundle, and right ventricle. Baseline electrophysiology measurements were obtained, followed by atrial and ventricular pacing protocols. Using 3D NavX, geometry of the Triangle of Koch, including the posteroseptal right atrium and proximal coronary sinus, was created. Multiple data points were collected from the region and the computer automatically measured the atrial voltage amplitude. The measurements were Congenit Heart Dis. 2015;10:E172–E179
E174
Bearl et al. minute when accelerated junctional rhythm with retrograde conduction persisted during the application. Acute success at the end of the procedure was defined as absence of anterograde SP function (loss of preablation A-H jump and/or sustained SP conduction with atrial pacing), the presence of junctional acceleration during RF lesions, inability to reinduce AVNRT, and absence of consistent echo beats.
Figure 1. Example of simultaneous voltage gradient map on the left and propagation map on the right overlaid on an anatomic map. The view is from the left ventricle back toward the right atrium. The coronary sinus os and His catheter are landmarked. The predominantly purple areas of the voltage map represent high-voltage signal surrounding the red-orange low-voltage bridge, thought to represent a slow pathway (SP). On the far left is the voltage gradient, with the low-voltage parameter of 0.2 mV and high-voltage parameter of 1.2 mV. The blue arrows on the propagation map on the right show the zone of collision of electrical activation in the location of the SP.
verified by a St. Jude Medical representative (JLP) and then displayed on the geometry. After sufficient representative points were obtained, adjustment of the voltage high and low ranges was performed to visualize the likely location of the SP(s) through a low-voltage bridge within the region. (Figure 1) If an SP could not be identified by voltage mapping, then the electroanatomic approach was used for the ablation procedure.5 Cryoablation was used first in all patients when dual AV nodal physiology was reliably demonstrated (dual AV nodal pathways, sustained SP conduction, AV nodal echo beats, or inducible AVNRT). During application of cryo-lesions, repeated testing for change in these findings could be observed, indicating success. When cryoablation was used, a temperature of −70 degrees centigrade for 4 minutes was achieved while programmed electrical stimulation was used to demonstrate evidence of SP function, AV nodal echo beats, or inducible AVNRT. Radiofrequency approach was used in two situations: (1) when there was documented AVNRT, but no evidence of dual AV nodal physiology and an inability to induce AVNRT preablation; and (2) when cryoablation lesions were not successful in eliminating SP conduction (demonstrated by atrial programmed stimulation and/or inducible AVNRT). The RF was terminated when junctional acceleration was not seen within 10–15 seconds of onset of RF energy and/or if absence of retrograde conduction was demonstrated. The RF was continued for 1 Congenit Heart Dis. 2015;10:E172–E179
Results
Twenty-eight patients underwent SP ablation using the voltage mapping technique. Two patients were excluded. One patient did not have the voltage map saved for review after the procedure. Another was ultimately found to have atrial ectopic tachycardia after her initial procedure that included successful SP ablation. Therefore, voltage maps were available from 26 patients. The mean age of the patients was 15.3 years, with a range from 8.5 to 20.6 years. Sixteen (57%) were female. Acute success was achieved in all but one patient (96%). The singular acute failure occurred with a voltage map that lacked a distinct SP. Twenty-one (75%) were ablated with cryoablation alone, three (11%) were ablated with RF ablation alone, and four (14%) were ablated with both. Four (15%) initial voltage maps showed no distinct SP, but all had successful ablations using the computerized 3D mapping technique. Of those four, three (75%) were found to have an SP after idealization of the voltage parameters. One patient (4%) had recurrence of AVRNT. One (4%) patient showed evidence of two SPs. Table 1 shows the details of idealization of the voltage maps. Both the idealized low- and high-voltage limits were higher than the settings during the actual procedure. The absolute difference in range reflects a positive value between the idealized voltage range and the original voltage range. Overall, the mean absolute difference in range was 0.340 mV, a reflection of the average amount of adjustment needed to find the idealized voltage range. The cohort was divided between the early half and later half of procedures. For the first half of ablations performed, the mean absolute difference in range was larger at 0.464 mV, while the second half was 0.255 mV. No complications were encountered during the voltage mapping procedures. Idealized parameters, including low- and highvoltage parameters along with the range (difference between high- and low-voltage parameters),
E175
The mean, standard deviation (SD), 95% confidence interval (95% CI), maximum (Max), and minimum (Min) values are shown for both original voltage maps used during ablation and postprocedure idealized voltage maps. For each original and idealized voltage map, a voltage gradient was created by defining a low and high voltage measured in millivolts (mV). The range between low and high voltage was calculated. The absolute difference in range is the positive value of the idealized voltage range minus the original voltage range. Used points calculated by subtracting measured unused points from total points. Slow pathway (SP) length and width measured in millimeters (mm).
8.19 5.68 0.069 25 2 10.44 4.07 0.051 20 4 2.48 1.23 0.015 5 1 0.340 0.404 0.005 1.42 0.01 1.156 0.503 0.006 2.154 0.403 1.550 0.722 0.009 3.525 0.548 0.394 0.451 0.006 2.172 0.077 0.994 0.587 0.007 2.773 0.05 1.321 0.688 0.008 3.025 0.131 223 79 0.954 393 56
Used Points
0.327 0.270 0.003 1.203 0.081
Original Range (mV) Original High (mV) Original Low (mV)
Mean SD 95% CI Max Min
Table 1.
Data from Original and Idealized Voltage Maps
Idealized Low (mV)
Idealized High (mV)
Idealized Range (mV)
Absolute Difference in Ranges (mV)
SP Width (mm)
SP Length (mm)
Number of Lesions
Voltage Mapping in Pediatric AVNRT
were stratified by age and weight. The lowest idealized voltage parameters were seen at 11.9 years and 51 kg for idealized low at 0.077 mV and 8.5 years and 30.5 kg for idealized high at 0.548 mV, with the narrowest range seen at 8.5 years and 30.5 kg at 0.403 mV and widest range at 16.2 years and 77.6 kg at 2.154 mV (Figure 2). The measurement of the low-voltage bridges selected for ablation yielded an average width of 2.5 mm with a range from 1 to 5 mm and an average length of 10.5 mm with a range from 4 to 20 mm. When comparing the voltage mapping group to the traditional technique control group, number of lesions did not achieve statistical significance, but trended in that direction (Table 2). Fluoroscopy time and procedure time also were not significantly different. Two cases are of particular interest within this cohort, which illustrate the learning curve experienced as well as the utility of voltage mapping. One patient had an acutely successful ablation with lesions placed in the location of what appeared to be a low-voltage bridge (Figure 3). On follow-up 1 month later, she was found to have a recurrence of her symptoms and documentation of her AVNRT. With idealization of her voltage map, a separate and more convincing low-voltage bridge was found in a different location from where her cryoablation lesions had been placed. The initial low-voltage bridge seen on the original voltage map is thought to have been falsely generated because of inadequate adjustment of highand low-voltage filters. The utility of voltage mapping to delineate complex underlying SPs is illustrated in another patient. During creation of the voltage map, it was noted that the patient appeared to have two low-voltage bridges, possibly representing two SPs (Figure 4). Ablation was undertaken on the more prominent pathway with acute success initially. On isoproterenol testing, there were consistent and persistent echo beats. With that data in addition to the voltage map containing a second low-voltage bridge, ablation of the second pathway was undertaken. Overall success was then achieved with no recurrence of AVNRT. Discussion
The study by Bailin et al.9 in adults demonstrated that the voltage mapping technique using a 3D electroanatomic computer could be used to visualize the putative SP. The technique involves creation of a detailed anatomic model of the Triangle Congenit Heart Dis. 2015;10:E172–E179
E176
Bearl et al.
Figure 2. Graph of the idealized voltage parameters used for each patient stratified by either (A) age in years or (B) weight in kilograms (kg). The darker line represents the idealized low-voltage parameter, expressed in millivolts (mV), and the lighter line represents the idealized high-voltage parameter, expressed in millivolts (mV).
Table 2.
Comparison of Secondary Outcomes between Ablation Techniques
Fluoroscopy time (minutes) Procedure time (minutes) Number of lesions (including extra lesions)
Traditional Technique (n = 24)
Voltage Mapping Technique (n = 28)
Mean
SD
Mean
SD
P
11.2 243.6 11.13
6.80 86.6 9.02
11.97 225.9 8.19
6.84 48.4 5.68
.690 .379 .178
Mean and standard deviation (SD) for fluoroscopy time and procedure time, both expressed in minutes, as well as number of lesions including extra lesions are shown between the traditional ablation technique group and the voltage mapping technique group. P value was obtained through a nonpaired t-test, and failed to reach significance in all three comparisons.
of Koch, including the AV node, coronary sinus, and tricuspid valve as landmarks.10 A superimposed atrial electrogram voltage map is then used to visualize a low-voltage bridge, which most likely represents the SP in AVNRT as ablation lesions placed on these bridges resulted in changes in AV conduction consistent with SP ablation. This study, as well as the recently reported series by Malloy et al.,11 demonstrates that visualization of probable SPs via low-voltage bridges through voltage mapping can also be successful in a pediatric population. The technique used in this cohort was similar to that used by Bailin et al. and Malloy et al. As expected, we experienced a learning curve with the technique, with a trend toward more stability and reproducibility in the second Congenit Heart Dis. 2015;10:E172–E179
half of the cases. Another outcome of this study was that we were able to elucidate low-voltage bridge parameters of voltage and physical dimensions. The ideal high- and low-voltage parameters were stratified by age, gender, and weight. There appears to be variability in age and weight that may affect the overall voltage of the lowvoltage SP (Figure 2), but no specific trends or associations were identified. Dimensions were also attributed to the low-voltage bridges, but no associations emerged. There are inherent advantages of visualization of low-voltage bridges or putative SPs. First, voltage mapping has the potential to increase safety by visualizing low-voltage bridges in anatomic relationship to the fast pathway/compact
Voltage Mapping in Pediatric AVNRT
E177
Figure 3. Example of a slow pathway (SP) revealed through the process of idealization. On the left is the voltage map generated during the procedure with coronary sinus os and bundle of His/atrioventricular (AV) node landmarked, but no low-voltage bridge evident. The cryoablation lesions made during the procedure are noted as green circles. On the right is the voltage map after idealization. Notice the increase in both low-voltage and high-voltage parameters seen on the voltage gradient, which revealed a low-voltage bridge and the discrepancy with regard to the lesions placed.
Figure 4. Example of a subject with two slow pathways (SPs) identified. On the left is the idealized voltage map identifying the coronary sinus os and bundle of His/atrioventricular (AV) node landmarks along with two low-voltage bridges. On the right is the same voltage map with location of the successful cryoablation lesions placed denoted as green circles. The cryoablation lesions were successful at eliminating alternation of atrioventricular nodal reentrant tachycardia (AVNRT) cycle length with ablation of the first pathway followed by no evidence of any SP function or inducible AVNRT with ablation of the second SP.
AV node. Permanent AV block has been reported in 1–2% for RF ablation,2 although a more recent case series showed no permanent AV block with 3% of patients having brief transient AV block.12 Cryoablation has had a better, albeit shorter record, with no permanent AV block having been reported. Transient AV block has been reported in 6–23% of cryoablation procedures.13 Safety may also be enhanced with the voltage technique by a decreased overall proce-
dure time. Our study showed a trend toward as well as less variability (standard deviation 86.6 vs. 48.4) compared with the traditional technique. With more experience, we also believe that fluoroscopy time can be reduced further, even though we showed no statistical difference between our groups. In a larger pediatric study, a significant drop in overall fluoroscopy time was seen when procedures were performed with NavX electroanatomic mapping.14 Congenit Heart Dis. 2015;10:E172–E179
E178 Second, the data trended toward fewer lesions required to ablate the low-voltage bridges. With increased experience of visualizing and therefore directly targeting the low-voltage bridges, fewer lesions may be necessary to achieve success. In addition, the later cases in this cohort included interpretation of a propagation map (Figure 1) that also helps to define the probable SP by providing a temporal element to mapping. If the target is in an unusual location, or there are multiple functional SPs, then more lesions may be necessary but even in this situation, it is reasonable to speculate that fewer lesions would be placed compared to the electroanatomic approach. With a larger sample size, it is possible that the trend toward fewer ablative lesions may become statistically significant. Third, success may be greater for patients with complicated underlying conditions including those with abnormal anatomy and multiple SPs. There is anecdotal evidence of this benefit even in this small sample size, as illustrated with the two example cases in the Results section. Idealizing the voltage map at the time of ablation is an important aspect to this technique. These cases exemplify the value of idealizing the voltage map because atypically located SP or multiple SPs were delineated. More experience and research are needed to better systematically visualize SPs in these complex cases. The recent paper published by Malloy et al. demonstrated the first results of voltage mapping in a pediatric population.11 They also demonstrated the feasibility, efficacy, and safety of this approach in the pediatric-age group. Our study confirms those findings, but with several important additions. The recurrence rate was only 4% in our study, while their recurrence rate was almost double at 7% in the voltage mapping group. Additionally, we report on the variability and measurements of the likely SPs. Lastly, two complex cases were added to illustrate both the experiential process of using the voltage mapping approach and its benefits. There are several limitations to this study. The retrospective nature of the study limits the ability to control for confounding circumstances pre- and post-procedure. The sample size was small and therefore we did not compare total procedure success to the procedure success of our control group. A much larger cohort and control group would have been necessary to achieve statistical power given the overwhelming success in AVNRT ablations by the conventional electroanatomic approach. Another limitation is the nonCongenit Heart Dis. 2015;10:E172–E179
Bearl et al. standardization of EnSite NavX parameters (interpolation, interior and exterior projection) in this study, which may play a role in the variability of voltage ranges observed. With the experience gained from this cohort and patients in the future, knowledge of EnSite NavX parameters during AVNRT cases will be gained so that voltage mapping settings will need less adjusting from case to case. This study provides evidence that the voltage mapping technique for AVNRT ablation is feasible and associated with high success rates and no complications in a pediatric population. A learning curve can be expected when starting to use this technique, but with increased experience, there is potential for safer and shorter procedures, while maintaining the efficacy of the traditional electroanatomic technique. Further research is needed to better define functional voltage parameters for pediatrics as a whole as well as subgroups such as gender and age to better reach the potentials of voltage mapping. Authors’ Contributions David W. Bearl, MD—Data collection and interpretation, drafting of article, critical revision of article, statistics. LuAnn Mill, RN—Data collection and administrative support. John D. Kugler, MD—Concept/design, statistics, critical revision of article. John L. Prusmack—Data collection and technical support. Christopher C. Erickson, MD—Concept/design, data collection and interpretation, drafting of article, critical revision of article.
Corresponding Author: Christopher C. Erickson, MD, Children’s Hospital and Medical Center, 8200 Dodge St, Omaha, NE 68114, USA. Tel: 402-9554339; Fax: 402-955-4356; E-mail: cerickson@ childrensomaha.org Conflict of interest: Coauthor John L. Prusmack is employed by St. Jude’s Medical S.C., Inc. Accepted in final form: December 29, 2014. References
1 Kugler JD, Danford DA, Houston KA, Felix G. Pediatric radiofrequency catheter ablation registry success, fluoroscopy time, and complication rate for supraventricular tachycardia: comparison of early and recent eras. J Cardiovasc Electrophysiol. 2002;13: 336–341.
E179
Voltage Mapping in Pediatric AVNRT 2 Van Hare GF, Javitz H, Carmelli D, et al. Prospective assessment after pediatric cardiac ablation: demographics, medical profiles and initial outcomes. J Cardiovasc Electrophysiol. 2004;15:759–770. 3 Heinroth KM, Kattenbeck K, Stabenow I, Trappe HJ, Weismuller P. Multiple AV nodal pathways in patients with AV nodal reentrant tachycardia—more common than expected? Europace. 2002;4:375–382. 4 Tai C-T, Chen S-A, Chiang C-E, et al. Multiple anterograde atrioventricular node pathways in patients with atrioventricular node reentract tachycardia. J Am Coll Cardiol. 1996;28:725–731. 5 Varanasi S, Dhala A, Blanck Z, Deshpande S, Akhtar M, Sra J. Electroanatomic mapping for radiofrequency ablation of cardiac arrhythmias. J Cardiovasc Electrophysiol. 1999;10:538–544. 6 Wittkampf FH, Wever EF, Derksen R, et al. LocaLisa: new technique for real-time 3dimensional localization of regular intracardiac electrodes. Circulation. 1999;99:1312–1317. 7 Smith G, Clark JM. Elimination of fluoroscopy use in a pediatric electrophysiology laboratory utilizing three-dimensional mapping. Pacing Clin Electrophysiol. 2007;30:510–518. 8 Casella M, Perna F, Russo AD, et al. Right ventricular substrate mapping using the Ensite Navx system: accuracy of high-density voltage map obtained by
9
10 11
12
13 14
automatic point acquisition during geometry reconstruction. Heart Rhythm. 2009;6:1598–1605. Bailin SJ, Korthas MA, Weers NJ, Hoffman CJ. Direct visualization of the slow pathway using voltage gradient mapping: a novel approach for successful ablation of atrioventricular nodal reentry tachycardia. Europace. 2011;13:1188–1194. Koch W. Weiter mitteilungen uber den sinusknoten des Herzens. Verh Dtsch Ges Pathol. 1909;13:85– 92. Malloy L, Law IH, Von Bergen NH. Voltage mapping for slow-pathway visualization and ablation of atrioventricular nodal reentry tachycardia in pediatric and young adult patients. Pediatr Cardiol. 2014;35:103–107. Fishberger SB, Whalen R, Zahn EM, Welch EM, Rossi AF. Radiofrequency ablation of pediatric AV nodal reentrant tachycardia during the ice age: a single center experience in the cryoablation era. Pacing Clin Electrophysiol. 2010;33:6–10. De Sisti A, Tonet J. Cryoablation of atrioventricular nodal reentrant tachycardia: a clinical review. Pacing Clin Electrophysiol. 2012;35:233–240. Kwong W, Neilson AL, Chiu CC, et al. The effect of NavX on fluoroscopy times in pediatric catheter ablation. J Interv Card Electrophysiol. 2012;33: 123–126.
Congenit Heart Dis. 2015;10:E172–E179
Visualization of Atrioventricular Nodal Reentry Tachycardia Slow Pathways Using Voltage Mapping for Pediatric Catheter Ablation David W. Bearl, MD,*† LuAnn Mill, RN,*† John D. Kugler, MD,*† John L. Prusmack, BA, CEPS,‡ and Christopher C. Erickson, MD*† *Department of Pediatrics, Children’s Hospital and Medical Center, †University of Nebraska Medical Center, Omaha, Neb and ‡St. Jude’s Medical, S.C., Inc., Austin, Tex, USA ABSTRACT
Background. Catheter ablation of the slow atrioventricular (AV) pathway has been shown to be safe and effective in pediatric patients with atrioventricular nodal reentrant tachycardia (AVNRT). Despite that, acute success is not guaranteed, and safety of ablating near the AV node remains a concern. Methods. A retrospective analysis was performed of all AVNRT ablations using the Ensite NavX voltage mapping technique at our institution. Each map was reviewed with patient and NavX computer data recorded. To account for a learning curve, each map was idealized and compared with the original map. Procedure and fluoroscopy time were compared with a control group. Results. Twenty-eight patients underwent catheter ablation for AVNRT from September 2011 until December 2012 using the voltage mapping technique. The historical control group comprised 24 patients who underwent catheter ablation using the electroanatomic approach. There was 96% acute success with one recurrence in the voltage mapping group, at a mean follow-up of 24 months. The slow pathway was visualized in 86% of patients at the time of ablation, while three of four without could be found on idealization of the voltage map. Mean high- and low-voltage parameters increased with idealization, but showed no correlation with age, gender, or weight. Estimated pathway size had significant inter-patient variability. Procedure and fluoroscopy times did not vary significantly compared with controls. Conclusion. Visualization of the AV nodal slow pathway in a pediatric population is possible using voltage mapping technique with the potential to increase safety and efficacy. Variability exists in the voltage parameters needed to visualize individual slow pathways, which leads to a distinct learning curve. Key Words. Pediatric AVNRT; AVNRT; Catheter Ablation; Voltage Mapping; Slow Pathway Visualization
Introduction
A
trioventricular nodal reentrant tachycardia (AVNRT) is a relatively common mechanism of supraventricular tachycardia in children. Catheter ablation has become a common patient/family management choice.1,2 The ablation target for AVNRT is the slow atrioventricular (AV) nodal pathway. More than one slow pathway (SP) can be present, but the overall incidence of multiple SPs, even in adults, is ill defined—ranging from 5 to 40%.3,4 Catheter ablation traditionally has relied Disclosure of grants or other funding: None. Congenit Heart Dis. 2015;10:E172–E179
on fluoroscopy to position catheters in the heart up until the late 1990s, when three-dimensional (3D)3 electroanatomic mapping was introduced and validated.5,6 The electroanatomic approach involves both fluoroscopic anatomic mapping using catheters as landmarks and electrical activation mapping searching for SP potentials in the posteroseptal region. Now computerized 3D mapping has become standard and has allowed for a substantial reduction in fluoroscopy during catheter ablation.7 The challenge of ablating the SP is ensuring that lesions are placed a safe distance from the compact AV node while achieving permanent SP ablation. Injury to the AV node can © 2015 Wiley Periodicals, Inc.
Voltage Mapping in Pediatric AVNRT cause AV block with subsequent need for a permanent pacemaker. While the electroanatomic approach is associated with a high rate of success, if the targeted SP is not in the safe posteroseptal region, lesions typically are successively placed closer and closer to the AV node until success is achieved or the operator chooses not to risk closer lesions to the AV node. Because anatomic variability, atypical SP locations, and multiple SPs can complicate ablation procedures, visualization of the SP could increase safety as well as enhance success. Voltage mapping is a technique that allows for visualization of voltage gradients in relation to anatomic locations and has been previously validated.8 Based on a report by Bailin et al.9 that described use of voltage mapping for AVNRT ablations with a St. Jude NavX 3D mapping system (St. Jude Medical, St. Paul, MN, USA), the use of this technique was begun at our institution. This technique involves the simultaneous collection of anatomic points and atrial electrograms. After data collection is complete, the EnSite NavX system creates both voltage and propagation maps of the region. In the posteroseptal region, there are low-voltage bridges or narrow myocardial isthmuses that connect high-voltage areas. These bridging isthmuses of low voltage are predicted to be SPs. Catheter ablation lesions are then directed at the SPs. The purpose of this study is to retrospectively assess the utility and effectiveness of voltage mapping described by Bailin et al. to visualize the SP in AVNRT for a pediatric population. A secondary purpose is to assess the characteristics of the visualized SPs including the voltage parameters and dimensions. Methods
Data Collection This protocol was approved by the University of Nebraska Medical Center and Children’s Hospital and Medical Center Institutional Review Board (Omaha, NE, USA). Patients with documented AVNRT who underwent SP ablation using the voltage mapping technique were identified retrospectively from an internal database at Children’s Hospital and Medical Center. Twenty-eight patients underwent SP ablation using the voltage mapping technique between November 2011 and December of 2012. Patient data were reviewed including date of procedure, ablation technique (radiofrequency ablation [RF], cryoablation, or both), procedure time, fluoroscopy time, number
E173 of lesions (including extra lesions or lesions placed after success was achieved), success and complications both at the time of procedure, and follow-up. Data were also collected from the EnSite NavX computer, which included interpolation, exterior and interior projection, final high- and lowvoltage parameters, number of anatomic/ electrographic points collected as well as number of points unused. Number of points used for each map was calculated. All cases were reviewed retrospectively including the NavX maps and objective parameters. An “idealized” map was created by adjusting the upper and lower voltage limit filters to subjectively better define the SP. In the four cases where voltage mapping did not identify an SP during the procedure, the upper and lower voltage limit filters were adjusted to see if a lowvoltage bridge could retrospectively be identified. Mean high- and low-voltage parameters, mean range between high- and low-voltage parameters, and differences between parameters from original voltage maps and idealized maps were plotted against several patient characteristics including age and weight, as well as chronologically. The SP(s) visualized were measured for length and width in millimeters using the His catheter electrodes as the calibration reference. A control group was created using the same internal database by identifying children who had undergone AVNRT ablation prior to the use of voltage mapping. A total of 24 patients that had complete comparable data were identified sequentially, working retrospectively from the initial patient who received ablation with the voltage mapping technique. Data collected from December 2008 to June 2011 included procedure time, fluoroscopy time, and total number of lesions (including extra lesions). Data from the control group were compared with the study group using a Students’ nonpaired t-test.
Procedures All procedures were performed under general anesthesia. Electrophysiology catheters were placed in the high right atrium, coronary sinus, AV junction/His bundle, and right ventricle. Baseline electrophysiology measurements were obtained, followed by atrial and ventricular pacing protocols. Using 3D NavX, geometry of the Triangle of Koch, including the posteroseptal right atrium and proximal coronary sinus, was created. Multiple data points were collected from the region and the computer automatically measured the atrial voltage amplitude. The measurements were Congenit Heart Dis. 2015;10:E172–E179
E174
Bearl et al. minute when accelerated junctional rhythm with retrograde conduction persisted during the application. Acute success at the end of the procedure was defined as absence of anterograde SP function (loss of preablation A-H jump and/or sustained SP conduction with atrial pacing), the presence of junctional acceleration during RF lesions, inability to reinduce AVNRT, and absence of consistent echo beats.
Figure 1. Example of simultaneous voltage gradient map on the left and propagation map on the right overlaid on an anatomic map. The view is from the left ventricle back toward the right atrium. The coronary sinus os and His catheter are landmarked. The predominantly purple areas of the voltage map represent high-voltage signal surrounding the red-orange low-voltage bridge, thought to represent a slow pathway (SP). On the far left is the voltage gradient, with the low-voltage parameter of 0.2 mV and high-voltage parameter of 1.2 mV. The blue arrows on the propagation map on the right show the zone of collision of electrical activation in the location of the SP.
verified by a St. Jude Medical representative (JLP) and then displayed on the geometry. After sufficient representative points were obtained, adjustment of the voltage high and low ranges was performed to visualize the likely location of the SP(s) through a low-voltage bridge within the region. (Figure 1) If an SP could not be identified by voltage mapping, then the electroanatomic approach was used for the ablation procedure.5 Cryoablation was used first in all patients when dual AV nodal physiology was reliably demonstrated (dual AV nodal pathways, sustained SP conduction, AV nodal echo beats, or inducible AVNRT). During application of cryo-lesions, repeated testing for change in these findings could be observed, indicating success. When cryoablation was used, a temperature of −70 degrees centigrade for 4 minutes was achieved while programmed electrical stimulation was used to demonstrate evidence of SP function, AV nodal echo beats, or inducible AVNRT. Radiofrequency approach was used in two situations: (1) when there was documented AVNRT, but no evidence of dual AV nodal physiology and an inability to induce AVNRT preablation; and (2) when cryoablation lesions were not successful in eliminating SP conduction (demonstrated by atrial programmed stimulation and/or inducible AVNRT). The RF was terminated when junctional acceleration was not seen within 10–15 seconds of onset of RF energy and/or if absence of retrograde conduction was demonstrated. The RF was continued for 1 Congenit Heart Dis. 2015;10:E172–E179
Results
Twenty-eight patients underwent SP ablation using the voltage mapping technique. Two patients were excluded. One patient did not have the voltage map saved for review after the procedure. Another was ultimately found to have atrial ectopic tachycardia after her initial procedure that included successful SP ablation. Therefore, voltage maps were available from 26 patients. The mean age of the patients was 15.3 years, with a range from 8.5 to 20.6 years. Sixteen (57%) were female. Acute success was achieved in all but one patient (96%). The singular acute failure occurred with a voltage map that lacked a distinct SP. Twenty-one (75%) were ablated with cryoablation alone, three (11%) were ablated with RF ablation alone, and four (14%) were ablated with both. Four (15%) initial voltage maps showed no distinct SP, but all had successful ablations using the computerized 3D mapping technique. Of those four, three (75%) were found to have an SP after idealization of the voltage parameters. One patient (4%) had recurrence of AVRNT. One (4%) patient showed evidence of two SPs. Table 1 shows the details of idealization of the voltage maps. Both the idealized low- and high-voltage limits were higher than the settings during the actual procedure. The absolute difference in range reflects a positive value between the idealized voltage range and the original voltage range. Overall, the mean absolute difference in range was 0.340 mV, a reflection of the average amount of adjustment needed to find the idealized voltage range. The cohort was divided between the early half and later half of procedures. For the first half of ablations performed, the mean absolute difference in range was larger at 0.464 mV, while the second half was 0.255 mV. No complications were encountered during the voltage mapping procedures. Idealized parameters, including low- and highvoltage parameters along with the range (difference between high- and low-voltage parameters),
E175
The mean, standard deviation (SD), 95% confidence interval (95% CI), maximum (Max), and minimum (Min) values are shown for both original voltage maps used during ablation and postprocedure idealized voltage maps. For each original and idealized voltage map, a voltage gradient was created by defining a low and high voltage measured in millivolts (mV). The range between low and high voltage was calculated. The absolute difference in range is the positive value of the idealized voltage range minus the original voltage range. Used points calculated by subtracting measured unused points from total points. Slow pathway (SP) length and width measured in millimeters (mm).
8.19 5.68 0.069 25 2 10.44 4.07 0.051 20 4 2.48 1.23 0.015 5 1 0.340 0.404 0.005 1.42 0.01 1.156 0.503 0.006 2.154 0.403 1.550 0.722 0.009 3.525 0.548 0.394 0.451 0.006 2.172 0.077 0.994 0.587 0.007 2.773 0.05 1.321 0.688 0.008 3.025 0.131 223 79 0.954 393 56
Used Points
0.327 0.270 0.003 1.203 0.081
Original Range (mV) Original High (mV) Original Low (mV)
Mean SD 95% CI Max Min
Table 1.
Data from Original and Idealized Voltage Maps
Idealized Low (mV)
Idealized High (mV)
Idealized Range (mV)
Absolute Difference in Ranges (mV)
SP Width (mm)
SP Length (mm)
Number of Lesions
Voltage Mapping in Pediatric AVNRT
were stratified by age and weight. The lowest idealized voltage parameters were seen at 11.9 years and 51 kg for idealized low at 0.077 mV and 8.5 years and 30.5 kg for idealized high at 0.548 mV, with the narrowest range seen at 8.5 years and 30.5 kg at 0.403 mV and widest range at 16.2 years and 77.6 kg at 2.154 mV (Figure 2). The measurement of the low-voltage bridges selected for ablation yielded an average width of 2.5 mm with a range from 1 to 5 mm and an average length of 10.5 mm with a range from 4 to 20 mm. When comparing the voltage mapping group to the traditional technique control group, number of lesions did not achieve statistical significance, but trended in that direction (Table 2). Fluoroscopy time and procedure time also were not significantly different. Two cases are of particular interest within this cohort, which illustrate the learning curve experienced as well as the utility of voltage mapping. One patient had an acutely successful ablation with lesions placed in the location of what appeared to be a low-voltage bridge (Figure 3). On follow-up 1 month later, she was found to have a recurrence of her symptoms and documentation of her AVNRT. With idealization of her voltage map, a separate and more convincing low-voltage bridge was found in a different location from where her cryoablation lesions had been placed. The initial low-voltage bridge seen on the original voltage map is thought to have been falsely generated because of inadequate adjustment of highand low-voltage filters. The utility of voltage mapping to delineate complex underlying SPs is illustrated in another patient. During creation of the voltage map, it was noted that the patient appeared to have two low-voltage bridges, possibly representing two SPs (Figure 4). Ablation was undertaken on the more prominent pathway with acute success initially. On isoproterenol testing, there were consistent and persistent echo beats. With that data in addition to the voltage map containing a second low-voltage bridge, ablation of the second pathway was undertaken. Overall success was then achieved with no recurrence of AVNRT. Discussion
The study by Bailin et al.9 in adults demonstrated that the voltage mapping technique using a 3D electroanatomic computer could be used to visualize the putative SP. The technique involves creation of a detailed anatomic model of the Triangle Congenit Heart Dis. 2015;10:E172–E179
E176
Bearl et al.
Figure 2. Graph of the idealized voltage parameters used for each patient stratified by either (A) age in years or (B) weight in kilograms (kg). The darker line represents the idealized low-voltage parameter, expressed in millivolts (mV), and the lighter line represents the idealized high-voltage parameter, expressed in millivolts (mV).
Table 2.
Comparison of Secondary Outcomes between Ablation Techniques
Fluoroscopy time (minutes) Procedure time (minutes) Number of lesions (including extra lesions)
Traditional Technique (n = 24)
Voltage Mapping Technique (n = 28)
Mean
SD
Mean
SD
P
11.2 243.6 11.13
6.80 86.6 9.02
11.97 225.9 8.19
6.84 48.4 5.68
.690 .379 .178
Mean and standard deviation (SD) for fluoroscopy time and procedure time, both expressed in minutes, as well as number of lesions including extra lesions are shown between the traditional ablation technique group and the voltage mapping technique group. P value was obtained through a nonpaired t-test, and failed to reach significance in all three comparisons.
of Koch, including the AV node, coronary sinus, and tricuspid valve as landmarks.10 A superimposed atrial electrogram voltage map is then used to visualize a low-voltage bridge, which most likely represents the SP in AVNRT as ablation lesions placed on these bridges resulted in changes in AV conduction consistent with SP ablation. This study, as well as the recently reported series by Malloy et al.,11 demonstrates that visualization of probable SPs via low-voltage bridges through voltage mapping can also be successful in a pediatric population. The technique used in this cohort was similar to that used by Bailin et al. and Malloy et al. As expected, we experienced a learning curve with the technique, with a trend toward more stability and reproducibility in the second Congenit Heart Dis. 2015;10:E172–E179
half of the cases. Another outcome of this study was that we were able to elucidate low-voltage bridge parameters of voltage and physical dimensions. The ideal high- and low-voltage parameters were stratified by age, gender, and weight. There appears to be variability in age and weight that may affect the overall voltage of the lowvoltage SP (Figure 2), but no specific trends or associations were identified. Dimensions were also attributed to the low-voltage bridges, but no associations emerged. There are inherent advantages of visualization of low-voltage bridges or putative SPs. First, voltage mapping has the potential to increase safety by visualizing low-voltage bridges in anatomic relationship to the fast pathway/compact
Voltage Mapping in Pediatric AVNRT
E177
Figure 3. Example of a slow pathway (SP) revealed through the process of idealization. On the left is the voltage map generated during the procedure with coronary sinus os and bundle of His/atrioventricular (AV) node landmarked, but no low-voltage bridge evident. The cryoablation lesions made during the procedure are noted as green circles. On the right is the voltage map after idealization. Notice the increase in both low-voltage and high-voltage parameters seen on the voltage gradient, which revealed a low-voltage bridge and the discrepancy with regard to the lesions placed.
Figure 4. Example of a subject with two slow pathways (SPs) identified. On the left is the idealized voltage map identifying the coronary sinus os and bundle of His/atrioventricular (AV) node landmarks along with two low-voltage bridges. On the right is the same voltage map with location of the successful cryoablation lesions placed denoted as green circles. The cryoablation lesions were successful at eliminating alternation of atrioventricular nodal reentrant tachycardia (AVNRT) cycle length with ablation of the first pathway followed by no evidence of any SP function or inducible AVNRT with ablation of the second SP.
AV node. Permanent AV block has been reported in 1–2% for RF ablation,2 although a more recent case series showed no permanent AV block with 3% of patients having brief transient AV block.12 Cryoablation has had a better, albeit shorter record, with no permanent AV block having been reported. Transient AV block has been reported in 6–23% of cryoablation procedures.13 Safety may also be enhanced with the voltage technique by a decreased overall proce-
dure time. Our study showed a trend toward as well as less variability (standard deviation 86.6 vs. 48.4) compared with the traditional technique. With more experience, we also believe that fluoroscopy time can be reduced further, even though we showed no statistical difference between our groups. In a larger pediatric study, a significant drop in overall fluoroscopy time was seen when procedures were performed with NavX electroanatomic mapping.14 Congenit Heart Dis. 2015;10:E172–E179
E178 Second, the data trended toward fewer lesions required to ablate the low-voltage bridges. With increased experience of visualizing and therefore directly targeting the low-voltage bridges, fewer lesions may be necessary to achieve success. In addition, the later cases in this cohort included interpretation of a propagation map (Figure 1) that also helps to define the probable SP by providing a temporal element to mapping. If the target is in an unusual location, or there are multiple functional SPs, then more lesions may be necessary but even in this situation, it is reasonable to speculate that fewer lesions would be placed compared to the electroanatomic approach. With a larger sample size, it is possible that the trend toward fewer ablative lesions may become statistically significant. Third, success may be greater for patients with complicated underlying conditions including those with abnormal anatomy and multiple SPs. There is anecdotal evidence of this benefit even in this small sample size, as illustrated with the two example cases in the Results section. Idealizing the voltage map at the time of ablation is an important aspect to this technique. These cases exemplify the value of idealizing the voltage map because atypically located SP or multiple SPs were delineated. More experience and research are needed to better systematically visualize SPs in these complex cases. The recent paper published by Malloy et al. demonstrated the first results of voltage mapping in a pediatric population.11 They also demonstrated the feasibility, efficacy, and safety of this approach in the pediatric-age group. Our study confirms those findings, but with several important additions. The recurrence rate was only 4% in our study, while their recurrence rate was almost double at 7% in the voltage mapping group. Additionally, we report on the variability and measurements of the likely SPs. Lastly, two complex cases were added to illustrate both the experiential process of using the voltage mapping approach and its benefits. There are several limitations to this study. The retrospective nature of the study limits the ability to control for confounding circumstances pre- and post-procedure. The sample size was small and therefore we did not compare total procedure success to the procedure success of our control group. A much larger cohort and control group would have been necessary to achieve statistical power given the overwhelming success in AVNRT ablations by the conventional electroanatomic approach. Another limitation is the nonCongenit Heart Dis. 2015;10:E172–E179
Bearl et al. standardization of EnSite NavX parameters (interpolation, interior and exterior projection) in this study, which may play a role in the variability of voltage ranges observed. With the experience gained from this cohort and patients in the future, knowledge of EnSite NavX parameters during AVNRT cases will be gained so that voltage mapping settings will need less adjusting from case to case. This study provides evidence that the voltage mapping technique for AVNRT ablation is feasible and associated with high success rates and no complications in a pediatric population. A learning curve can be expected when starting to use this technique, but with increased experience, there is potential for safer and shorter procedures, while maintaining the efficacy of the traditional electroanatomic technique. Further research is needed to better define functional voltage parameters for pediatrics as a whole as well as subgroups such as gender and age to better reach the potentials of voltage mapping. Authors’ Contributions David W. Bearl, MD—Data collection and interpretation, drafting of article, critical revision of article, statistics. LuAnn Mill, RN—Data collection and administrative support. John D. Kugler, MD—Concept/design, statistics, critical revision of article. John L. Prusmack—Data collection and technical support. Christopher C. Erickson, MD—Concept/design, data collection and interpretation, drafting of article, critical revision of article.
Corresponding Author: Christopher C. Erickson, MD, Children’s Hospital and Medical Center, 8200 Dodge St, Omaha, NE 68114, USA. Tel: 402-9554339; Fax: 402-955-4356; E-mail: cerickson@ childrensomaha.org Conflict of interest: Coauthor John L. Prusmack is employed by St. Jude’s Medical S.C., Inc. Accepted in final form: December 29, 2014. References
1 Kugler JD, Danford DA, Houston KA, Felix G. Pediatric radiofrequency catheter ablation registry success, fluoroscopy time, and complication rate for supraventricular tachycardia: comparison of early and recent eras. J Cardiovasc Electrophysiol. 2002;13: 336–341.
E179
Voltage Mapping in Pediatric AVNRT 2 Van Hare GF, Javitz H, Carmelli D, et al. Prospective assessment after pediatric cardiac ablation: demographics, medical profiles and initial outcomes. J Cardiovasc Electrophysiol. 2004;15:759–770. 3 Heinroth KM, Kattenbeck K, Stabenow I, Trappe HJ, Weismuller P. Multiple AV nodal pathways in patients with AV nodal reentrant tachycardia—more common than expected? Europace. 2002;4:375–382. 4 Tai C-T, Chen S-A, Chiang C-E, et al. Multiple anterograde atrioventricular node pathways in patients with atrioventricular node reentract tachycardia. J Am Coll Cardiol. 1996;28:725–731. 5 Varanasi S, Dhala A, Blanck Z, Deshpande S, Akhtar M, Sra J. Electroanatomic mapping for radiofrequency ablation of cardiac arrhythmias. J Cardiovasc Electrophysiol. 1999;10:538–544. 6 Wittkampf FH, Wever EF, Derksen R, et al. LocaLisa: new technique for real-time 3dimensional localization of regular intracardiac electrodes. Circulation. 1999;99:1312–1317. 7 Smith G, Clark JM. Elimination of fluoroscopy use in a pediatric electrophysiology laboratory utilizing three-dimensional mapping. Pacing Clin Electrophysiol. 2007;30:510–518. 8 Casella M, Perna F, Russo AD, et al. Right ventricular substrate mapping using the Ensite Navx system: accuracy of high-density voltage map obtained by
9
10 11
12
13 14
automatic point acquisition during geometry reconstruction. Heart Rhythm. 2009;6:1598–1605. Bailin SJ, Korthas MA, Weers NJ, Hoffman CJ. Direct visualization of the slow pathway using voltage gradient mapping: a novel approach for successful ablation of atrioventricular nodal reentry tachycardia. Europace. 2011;13:1188–1194. Koch W. Weiter mitteilungen uber den sinusknoten des Herzens. Verh Dtsch Ges Pathol. 1909;13:85– 92. Malloy L, Law IH, Von Bergen NH. Voltage mapping for slow-pathway visualization and ablation of atrioventricular nodal reentry tachycardia in pediatric and young adult patients. Pediatr Cardiol. 2014;35:103–107. Fishberger SB, Whalen R, Zahn EM, Welch EM, Rossi AF. Radiofrequency ablation of pediatric AV nodal reentrant tachycardia during the ice age: a single center experience in the cryoablation era. Pacing Clin Electrophysiol. 2010;33:6–10. De Sisti A, Tonet J. Cryoablation of atrioventricular nodal reentrant tachycardia: a clinical review. Pacing Clin Electrophysiol. 2012;35:233–240. Kwong W, Neilson AL, Chiu CC, et al. The effect of NavX on fluoroscopy times in pediatric catheter ablation. J Interv Card Electrophysiol. 2012;33: 123–126.
Congenit Heart Dis. 2015;10:E172–E179
