Author + information
- Received August 30, 2004
- Revision received October 12, 2004
- Accepted October 26, 2004
- Published online February 15, 2005.
- Aya Miyazaki, MD,
- Andrew D. Blaufox, MD* (, )
- David L. Fairbrother, MD and
- J. Philip Saul, MD
- ↵*Reprint requests and correspondence:
Dr. Andrew D. Blaufox, Children's Heart Program of South Carolina-MUSC, 165 Ashley Avenue, PO Box 250915, Charleston, South Carolina 29425
Objectives The aim of this study was to evaluate the efficacy and safety of catheter-based cryo-therapy for septal tachycardia substrates in pediatric patients.
Background Cryo-therapy may be particularly useful for ablation of septal tachycardias, including atrioventricular nodal re-entry tachycardia (AVNRT), atrioventricular reciprocating tachycardia (AVRT), and ventricular tachycardia (VT) originating high in the conduction system.
Methods Thirty-one pediatric patients (median = 13.7 years, range 5.3 to 19.6 years) with septal tachycardia substrates underwent cryo-ablation (CA). Twenty-two had AVNRT, 8 AVRT, and 1 VT. Applications were considered cryo-maps (CMs) if the temperature set-point was −35°C or the application time was <120 s. Other lesions were considered CAs.
Results A total of 242 CMs (4 per patient, range 0 to 40 CMs) and 89 CAs (2 per patient, range 1 to 8 CMs) were performed, for a total cryo-therapy time of 689 s/patient (range 158 to 3,300 s). Procedural success with cryo-therapy was achieved in 27 of 31 patients (87.1%), including two procedures with a His potential at the CA location and three performed in tachycardia. The success rate for AVNRT was higher than for AVRT (95.5% vs. 62.5%, p < 0.05). For AVRT, a sustained effect on accessory pathway conduction occurred −3.3 ± 4.9 s after reaching −25°C, whereas for those sites at which the effect was transient, the effect took 24.8 ± 25.5 s (p = 0.07). Transient atrioventricular (AV) block occurred during eight cryo-applications (1 CA, 7 CMs) with immediate return of normal AV conduction upon cessation of application. There were no other complications.
Conclusions Cryo-therapy was used to effectively and safely ablate septal tachycardias in this group of 31 pediatric patients. Cryo-therapy may be more effective for AVNRT than septal AVRT.
Catheter-based cryo-therapy has been introduced recently for ablation of a variety of cardiac arrhythmias (1–8), but only limited reports exist in children (9). Cryo-therapy has several potential advantages over radiofrequency ablation (RFA), which include: 1) reversible cryo-mapping before the production of a permanent lesion (1,10,11); 2) adherence of the catheter tip to the endocardium upon freezing (2); 3) a well-defined edge of the cryo-lesion (10); 4) minimal effects on adjacent coronary arteries (1); and 5) a lower incidence of thrombus (12). The first four of these are particularly relevant to small children because of the close proximity of a variety of critical cardiac structures to the ablation target and the reported potential for radiofrequency lesion growth in immature myocardium (13). In fact, the most common major complication during RFA in pediatric patients is atrioventricular (AV) block (14–16). Acute coronary artery injury has also been reported in a variety of locations in children (17,18), including during slow pathway modification (17).
This report addresses the efficacy and safety of catheter-based cryo-therapy in pediatric patients for ablation of septal tachycardias, including atrioventricular nodal re-entry tachycardia (AVNRT), atrioventricular reciprocating tachycardia (AVRT) caused by a septal accessory pathway, and ventricular tachycardia (VT) originating high in the conduction system.
The pediatric ablation registry at the Medical University of South Carolina was searched for all patients who had undergone catheter-based cryo-ablation (CA) for tachycardias with septal substrates during the period from May 2003 until July 2004. Only patients with structurally normal hearts under 20 years of age were included in the analysis.
Informed written consent was obtained before each procedure. All anti-arrhythmic drugs were discontinued for at least five half-lives. Under either general anesthesia or conscious sedation, all patients underwent an electrophysiologic study using standard pacing and mapping techniques to determine the mechanism of the tachycardia and the location of the ablation target. Patients were administered an initial heparin bolus of 100 IU/kg up to 5,000 IU. Activated clotting times were checked every 30 min, and additional heparin boluses were given to keep the activated clotting time >200 s (right) or >250 s (left). Isoproterenol was administered when the tachycardia was not inducible at baseline.
Once a target had been established, cryo-mapping and CA were undertaken using a microprocessor-based electromechanical refrigeration console controlling a 7-F, 4-mm or 6-mm tipped ablation catheter (CryoConsole, Freezor, Freezor Xtra, CryoCath Technologies Inc., Montreal, Canada). For AVNRT, both anatomical and electrogram criteria were used to choose initial application sites (19). For AVRT, sites for cryo-applications were determined on the basis of previously reported standard mapping criteria (20). For the patient with VT, the site of earliest ventricular activation was used as an ablation target. Cryo-therapy was performed in sinus rhythm, with atrial or ventricular pacing, or in slow junctional rhythm. However, in contrast to RFA, cryo-therapy was also used both during active tachycardia (AVNRT or AVRT) and in the presence of isoproterenol, because junctional acceleration is not generally present during cryo-therapy, and because catheter adherence during cryo-therapy allows for catheter stability during tachycardia or isoproterenol infusion. Cryo-therapy was initiated with either a preset minimal temperature of −35°C for up to 1 min, or a preset temperature of −70°C for up to 8 min. For the purposes of this study, a “cryo-map” (CM) was defined as any application with a set-point of −35°C or a colder application of <120 s duration. A CA was defined as an application with a set-point of −70°C that was ≥120 s. Each CM or CA was considered a separate cryo-application. The number of “cryo-sites” was defined as the number of anatomically separate locations where either a CM or CA was delivered. If an application at a single site was initiated as a CM but then extended to a CA, it was counted as two cryo-applications, but one cryo-site. “Application time” was defined as the time of each CA or CM. “Cryo-therapy time” was defined as the time from <−25°C until the termination of cryo-therapy at each cryo-site.
The decision to proceed from CM to CA was made using one or more of the following criteria: 1) AVNRT termination in the slow pathway; 2) modification or elimination of the slow pathway during atrial extra-stimulus testing; 3) accessory pathway block during sinus rhythm, ventricular pacing, or AVRT; and 4) for this VT case, a site of early ventricular activation not associated with AV block. For any CM or CA, the application was immediately discontinued for any degree of AV block.
Each cryo-site was classified by the outcome at that location, as success, transient success, failure, or post-success (additional application). A cryo-site was considered successful if it resulted in elimination of the slow pathway, the accessory pathway, or the disappearance of VT. A cryo-site was considered a transient success if there was recurrence of the arrhythmia or its substrate within 30 min of cryo-therapy termination. A cryo-site was considered a failure if it had no effect on the substrate. A cryo-site was considered post-success if it was performed very near the location of a previously successful cryo-site.
Procedural success was defined using RFA standards, including a minimum of 30 min of post-ablation testing. No inducible tachycardia was mandatory for any substrate. For AVNRT, a procedure was deemed successful if there was elimination of the slow pathway or not more than a single echo beat. For AVRT, loss of accessory pathway conduction was also required. If isoproterenol was required for the assessment of slow pathway or accessory pathway conduction or for the induction of tachycardia before ablation, the use of isoproterenol was required during post-ablation testing to deem the procedure a success. If cryo-therapy was considered a failure, a switch was generally made to RFA; however, there was no preset protocol for making this determination. The investigators tended to switch more readily when the results of the cryo-therapy indicated that the pathway location was one where there was a lower risk of AV block.
Values are shown as median and range unless otherwise specified. Comparisons were made with the two-tailed unpaired Student ttest or by Fisher exact test, whenever appropriate. A p value ≤0.05 was considered statistically significant.
Thirty-three pediatric patients with septal tachycardia substrates underwent catheter-based cryo-therapy. Two patients with AVNRT and congenital heart disease were excluded because of variation in AV node anatomy. Thus, 31 patients (19 male; age 13.7 years, range 5.3 to 19.6 years; weight 51.4 kg, range 19.7 to 130.2 kg) are included in this study (Table 1).Twenty-two patients had AVNRT, 8 AVRT, and 1 VT. Among patients with AVRT, accessory pathway locations were as follows: five right anteroseptal (RAS), one right mid-septal, and two right posteroseptal.
The RFA had been performed previously in four patients. Two of these patients had recurred after a previous RFA attempt (1 AVNRT, 1 AVRT), and one had undergone successful RFA of permanent junctional reciprocating tachycardia but subsequently developed AVNRT. The one with VT had failed RFA twice at another institution. Another two patients (1 AVNRT, 1 AVRT with RAS pathway) had undergone an electrophysiologic study at another institution, but did not undergo RFA because of unstable fast pathway conduction or a perceived high risk for AV block.
Cryo-sites and cryo-applications
In all but two patients, the 4-mm tipped ablation catheter was used (Table 1), with the 6-mm tipped catheter also used in the remaining patients. The ablation site was accessed via the inferior vena cava in 29 patients, the superior vena cava in 1 patient, and the aorta in the patient with VT (Table 1).
A total of 290 cryo-sites were analyzed (median 5.0 per patient), with a wide range in the number of sites (1 to 42) (Table 1). A His signal was seen on the cryo-catheter just before and after cryo-therapy in four patients at 15 sites (median 3.5 sites, range 1 to 7 sites). The rhythm at the beginning of cryo-therapy was sinus at 194 sites, tachycardia at 45 sites, atrial or ventricular pacing at 39 sites, and slow junctional rhythm at 12 sites. Cryo-therapy was performed during isoproterenol infusion at 137 sites.
Two hundred forty-two CMs (median 4 per patient, range 0 to 40 CMs) and 89 CAs (median 2 per patient, range 1 to 8 CMs) were performed. There was a wide range in the total cryo-therapy time for each patient (range 158 to 3,300 s, median 689 s).
Procedural success was achieved in 27 of 31 patients (AVNRT 21 of 22, AVRT 5 of 8, VT 1 of 1). For the 21 patients with AVNRT procedural success, slow pathway conduction was completely eliminated in 9 patients and modified in the remaining 12 patients, with no echo beats in 8 patients and one echo beat in 4 patients . The 12 patients with a modified slow pathway were classified as having undergone a successful procedure, because only cryo-site success was defined as complete elimination of the slow pathway, whereas procedural success was defined as having no more than a single AV nodal echo. The three patients who failed cryo-therapy were successfully ablated with radiofrequency energy. One patient with a RAS pathway did not go on to RFA because accessory pathway conduction was intermittent after cryo-therapy led to transient AV block at the same cryo-site. Cryo-therapy was performed during tachycardia at 45 cryo-sites, resulting in three permanent successes and five transient successes. Of the 137 cryo-sites where an application was made during isoproterenol infusion, permanent success was achieved at 4 sites and transient success was achieved at 14 sites.
For AVRT, where loss of pathway conduction could be immediately assessed, a sustained effect on accessory pathway conduction occurred −3.3 ± 4.9 s after reaching −25°C, whereas for those sites at which the effect was transient, the effect took 24.8 ± 25.5 s (p = 0.07) (Fig. 1).
Successful lesions, which resulted in complete substrate elimination without return of substrate conduction before the end of the procedure, occurred at 15 sites. These sites were the nine slow pathways, five accessory pathways, and the VT that were completely eliminated as stated previously. Transient success, where complete elimination of the substrate during an application was only temporary, was seen at 35 sites. Lesion failure, where there was no elimination of substrate during the application, was seen at 229 sites. Post-success applications were performed at 11 sites. Two successes were achieved in the presence of a large His potential on the ablation catheter (Figs. 2and 3).
Some form of transient AV block was seen during six applications and soon after termination of two applications (7 of 242 CMs vs. 1 of 89 CAs, p = NS). All eight episodes were during sinus rhythm, of which six occurred during isoproterenol infusion. An AV block occurred in 6 of 137 sites without isoproterenol and 2 of 153 sites with isoproterenol (p = NS). Fast pathway block, manifest by atriohisian prolongation, was seen in five patients, Wenckebach block in two patients, and high-grade second-degree AV block in one patient (Table 2).A His potential was present on the ablation catheter at the beginning of one application that resulted in first-degree AV block. Transient AV block occurred a median of 52.5 s (range 15 to 189 s) after reaching −25°C at a median minimum temperature of −73.5°C (−31°C to −81°C). All applications were terminated immediately upon recognition of AV block, and in all cases AV node function recovered quickly. Junctional acceleration was not observed in any patient during cryo-therapy. Mechanical catheter manipulation caused transient fast pathway block in two patients with AVNRT, not associated with cryo-therapy. There were no other complications.
During follow up of 8.2 months (range 0.8 to 14.4 months), 3 of 27 patients (11.1%) had a recurrence, all within four weeks of the procedure. One patient had AVNRT (Patient #12), and two had AVRT (Patients #24 and #26). Before the ablation, Patient #24 had pre-excitation and frequent episodes of AVRT. During the ablation procedure, the accessory pathway had intermittent conduction, which was eliminated with catheter manipulation during mapping. Retrograde pathway conduction was then cryo-ablated using an approach from the superior vena cava to the ventricular side of the AV annulus, owing to a large His potential at the site of earliest activation on the atrial side. Pre-excitation was not seen during the procedure but recurred within two weeks. However, the patient has not had AVRT recurrence, suggesting that retrograde conduction was successfully ablated. If this patient were counted as a clinical success, the recurrence rate would be reduced to 9.0%. No variable was found to be associated with recurrence. Both Patients #12 and #26 subsequently went on to a successful RFA procedure.
AVNRT versus AVRT
Table 3compares demographic variables and results for patients with AVNRT and AVRT. Patients with AVNRT were statistically older and heavier than patients with AVRT. The number of cryo-sites, CAs, CMs, total cryo-applications, and total cryo-therapy time were all greater for AVNRT than AVRT patients. The success rate was higher for AVNRT than AVRT (95.5% vs. 62.5%, p < 0.05); however, the recurrence rate was not statistically different (4.8% vs. 40.0%, p = NS).
This study demonstrates several important points about cryo-ablation of septal tachycardia substrates in pediatric patients. First and foremost, despite patient size as small as 20 kg, cryo-ablation was performed safely around the AV conduction system in this group of pediatric patients, even in the presence of a His potential. Second, cryo-ablation is an effective alternative to RFA for most septal substrates. Third, in contrast to RFA, cryo-therapy was safely and effectively delivered near the AV conduction system both during tachycardia and during isoproterenol infusion. Finally, cryo-therapy may be the only technique where avoidance of AV conduction damage can be nearly ensured during ablation attempts of tachycardia substrates such as VT originating high in the conduction system and some accessory pathways.
Definition of CM versus CA
Before discussing the implications of this study, it is important to understand why the definitions of CM and CA were modified compared with previous studies. In this study, a CM was defined as an application at any temperature with a duration of <120 s, whereas in previous studies, a CM was defined as a cryo-application with a preset minimum temperature between −15°C and −40°C (2–4,6,10,11,12). Our definition was based both on these previous clinical reports and previous animal studies (1,3,10,11), suggesting that permanent electrophysiologic effects require longer application times at lower temperatures. Dubuc et al. (11) reported that transient high-degree AV block was produced in dogs at temperatures near −40°C. However, permanent AV block only occurred with temperatures below −50°C applied for more than 5 min. Because the goal of a CA is to produce permanent loss of tissue function, it did not make sense to define all applications at lower temperatures or that were discontinued early as CAs. Thus, 120 s at any temperature was chosen as the cutoff.
Acute success in this study was achieved in 87.1% of all patients, including 95.5% of those with AVNRT and 62.5% of those with AVRT. These results are not as satisfactory as the results reported in recent large series of RFA for similar substrates in pediatric patients (AVNRT 99%, septal AVRT 90%) (21). However, a number of factors must be taken into account. The present report constitutes the initial experience with a new technology (22–25). Two initial reports on the use of RFA for supraventricular tachycardia substrates in small pediatric groups revealed acute success rates of 68% and 82% (22,23). Further, Danford et al. (25) showed that the learning curve for RFA in the Pediatric Ablation Registry was dependent upon the tachycardia mechanism and substrate location, and that the steady state for efficacy was not reached until approximately 50 previous procedures. Because the physicians in the present study have the advantage of previous experience with RFA, one might not expect such a long learning curve, but a number of differences in the techniques suggest that some learning curve will be present. For instance, the current generation of cryo-catheters is not nearly as flexible as the current generation of RFA catheters. The lack of familiarity with the cryo-ablation catheter's handling characteristics could have translated into a lower procedural success rate (3). In addition, the time to effect for a successful cryo-application in the present study was much longer than that for RFA (26), making it difficult to judge when to terminate an application. The notion of a learning curve suggests that success rates will increase over time. Fortunately, this study suggest that RFA can still be safely applied after failure of cryo-therapy, indicating that cryo-therapy can be used as a safe initial approach for septal substrates, reducing the number of patients exposed to RFA.
In the present study, cryo-therapy was more effective for AVNRT than AVRT with septal accessory pathway. The AVNRT success rate in this report is similar to that using cryo-therapy for AVNRT in adults (94%, 91%) (2,3). Although the success for AVRT in this small group of subjects was not as good as early reports for cryo-therapy in adults with septal pathways (100%) (4,5), it is comparable to a recent study involving AVRT in all locations (69%) (3). A number of factors may have contributed to these results. Because cryo-lesions are more delineated (9) and smaller than radiofrequency lesions (12), cryo-ablation may require more precise positioning, particularly for the relatively discrete accessory pathways. The learning curve for AVRT may have been less steep because of early abandonment of cryo-therapy in favor of radiofrequency energy at locations with electrographic characteristics considered adequate for the well-known effects of radiofrequency therapy. Also a possibility, though difficult to quantify, safe application of cryo-therapy in a particular location may have provided some level of confidence for applying radiofrequency energy in the same location, again lowering the threshold for changing therapies earlier in the procedure.
Time to accessory pathway block
After reaching −25°C, accessory pathway block occurred earlier when the delivered cryo-therapy was permanently successful than when it was transiently successful (Fig. 3). This finding did not reach statistical significance, needing to be substantiated in a larger group of pediatric patients, but it is in agreement with previous reports. Gaita et al. (4) reported that accessory pathway conduction block occurred during cryo-mapping a mean of 14 s after a temperature of −25°C was consistently achieved, and that sudden conduction block over the accessory pathway during CM was predictive of permanent success. A rapid loss of electrical conduction during higher temperature mapping should indicate that the target is well within the cooling zone and that with lower temperature and more time cell death will occur. Alternatively, later loss of conduction suggests the target is more distant and even lower temperatures at the electrode may not cause cell death at the target. This concept is similar to that with heating for RFA, but the time to effect for RFA is compressed compared with cryo-ablation (26).
In this small series of patients, there were no instances of permanent AV block despite the large number of septal applications and seven instances of transient first-degree and second-degree AV block. Both the CM and CA modes allowed for reversible loss of tissue function, similar to that previously reported in animals and humans during CM (1,3,10,11). These transient effects occurred for both the normal AV conduction fibers and the targeted tachycardia substrate. Despite the fact that AV block was observed in this study at temperatures below −50°C in five of the seven cases (Table 2), all instances of block fully recovered after termination of cryo-therapy, probably because cryo-therapy was immediately terminated at the first sign of AV block. Friedman et al. (3) also reported 12 instances of transient AV block, 11 of which occurred during cryo-ablation modes, and all of them resolved completely. Because the leading edge of the ice ball during cryo-therapy is by definition near 0°C and warmer than the temperature measured at the catheter tip, it is likely that discontinuation of cryo-therapy at the first signs of an electrophysiologic effect will reverse that effect. Conversely, because AV block occurred after the temperature had been set to −70°C, careful attention, including intermittent programmed stimulation, should be used to evaluate AV conduction in both the mapping and ablation modes, whether or not isoproterenol is being used. Another potential safety feature of cryo-therapy is catheter stability at the point of tissue freeze. This lack of tip movement should both improve success when the catheter is in the correct place (2) and prevent lesion spread to undesirable locations through the sliding movement seen with RFA. Catheter stability is particularly helpful when programmed stimulation is performed, isoproterenol is administered during ablation, or when tachycardia stops during the application.
As reported previously (2), junctional acceleration did not occur in the present study, avoiding the masking of AV node conduction by junctional acceleration seen during RFA, and allowing for a more thorough assessment of fast pathway conduction during ablation. This advantage probably offsets the fact that junctional acceleration cannot be used as a marker of the slow pathway location as it can during RFA.
Finally, the results of previous animal studies suggest that some advantages of cryo-therapy may be particularly important for children. Cryo-ablation sites are histologically well delineated, discrete, and show homogeneous dense fibrous tissue without viable myocardium interspersed (10). Cryo-lesions are smaller than RFA lesions (12), contributing to the ability to safely create cryo-lesions even adjacent to the His bundle, as was done for three of the patients in this study. Presumably because the size of Koch's triangle is smaller in children (27), the risk of causing AV node injury with any technique is higher in children (14–16).
The small number of subjects in this study limited the ability to reach statistical significance for some findings. The lack of a consistent protocol of when to abandon cryo-ablation in this retrospective study clearly allowed for user bias, making it difficult to draw conclusions from the failure rate, particularly for AVRT. Another limitation was that the criteria for success were based upon those developed for RFA and may have been either too stringent or too lenient for cryo-ablation. Clearly, future prospective studies with pre-defined criteria for success and failure should be undertaken.
Cryo-ablation can effectively be used to ablate septal tachycardias in pediatric patients. Cryo-ablation was used safely in this group of pediatric patients, even in very close proximity to the His bundle and during tachycardia. In this series, cryo-therapy was more effective for ablation of AVNRT than for ablation of septal accessory pathways.
Dr. Saul is a consultant for CryoCath Technologies Inc., Montreal, Quebec, Canada.
- Abbreviations and acronyms
- atrioventricular nodal re-entry tachycardia
- atrioventricular reciprocating tachycardia
- right anteroseptal
- radiofrequency ablation
- ventricular tachycardia
- Received August 30, 2004.
- Revision received October 12, 2004.
- Accepted October 26, 2004.
- American College of Cardiology Foundation
- Skanes A.C.,
- Dubuc M.,
- Klein G.J.,
- et al.
- Rodriguez L.,
- Geller J.C.,
- Tse H.,
- et al.
- Kirsh J.A.,
- Gross G.J.,
- Ponti R.D.,
- et al.
- Khairy P.,
- Chauvt P.,
- Lehmann J.,
- et al.
- Saul J.P.,
- Hulse J.E.,
- Papgiannis J.,
- Van Praagh R.,
- Walsh E.P.
- Schaffer M.S.,
- Silka M.J.,
- Ross B.A.,
- Kugler J.D.
- Blaufox A.D.,
- Felix G.L.,
- Saul J.P.
- Bertram H.,
- Bokenkamp R.,
- Peuster M.,
- Hausdorf G.,
- Paul T.
- Taylor G.S.,
- Kay G.N.
- Wood M.A.,
- Swartz, J.F.
- Dick M. II.,
- O'Connor B.K.,
- Werwer G.A.,
- LeRoy S.,
- Armstrong B.
- Van Hare G.F.,
- Lesh M.D.,
- Scheinman M.,
- Langberg J.J.