This was a retrospective single-center study assessing the safety and efficacy of cryoablation for AVNRT or right-sided septal APs. The medical records of all patients undergoing focal cryoablation between January 2007 and May 2025 were reviewed. Patients were included if they had a confirmed diagnosis of AVNRT or AP based on electrophysiological (EP) criteria (described below) and underwent focal cryoablation. The choice to perform cryoablation was made at the operator's discretion.

Exclusion criteria were: incomplete procedural or follow-up data, and cases in which the cryocatheter was used exclusively for diagnostic purposes or for cryomapping without subsequent ablation. The study reporting follows the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement, as illustrated in the flow chart (Figure 1). Out of 61 patients assessed, 49 were included in the study. Among the excluded cases, one patient presented with a manifest parahisian accessory pathway in whom cryomapping failed to eliminate pathway conduction, leading to conversion to radiofrequency ablation. Clinical, electrophysiological, and follow-up data were systematically collected from hospital electronic medical records.

Figure 1.

STROBE flow chart. STROBE: Strengthening the Reporting of Observational Studies in Epidemiology.

Antiarrhythmic drugs were discontinued for at least five half-lives prior to the ablation procedure. Written informed consent was obtained from all patients. Procedures were performed in a dedicated electrophysiology laboratory under local anesthesia with conscious sedation, or under general anesthesia in pediatric or selected cases.

All procedures were performed by three experienced electrophysiologists. Over time, procedural experience progressively increased, accompanied by more widespread use of electroanatomical mapping systems – which are now employed in all cases – and the routine adoption of 6–8 mm cryocatheters, which improved catheter stability during ablation.

Catheter placement was made via femoral venous access. Three diagnostic catheters were positioned alternately in the coronary sinus (CS), His bundle region, right ventricular apex, and high right atrium. Procedures were guided by fluoroscopy, intracardiac signs and, when available, a three-dimensional electroanatomic mapping system enabling the cryocatheter visualization (EnSite Precision™, Abbott).

Typical AVNRT was diagnosed by demonstrating dual AV nodal physiology, usually evidenced by an AH jump during atrial extrastimulus testing; tachycardia initiation by atrial pacing showing an atrial–His–atrial (A–H–A) response and a short septal VA time. Atypical AVNRT was diagnosed when dual AV nodal physiology was present and tachycardia exhibited features of slow-slow or fast-slow AV nodal reentrant circuit, characterized by a longer septal VA time and a H–A–H during ventricular entrainment or tachycardia initiation. Diagnosis was further supported by ventricular entrainment maneuvers with earliest atrial activation recorded at the His region. During ventricular overdrive pacing, a V–A–V pattern and a corrected post-pacing interval minus tachycardia cycle length (cPPI–TCL) >110 ms supported the diagnosis.

An AP was diagnosed when manifest pre-excitation was present on the baseline electrocardiogram (ECG), pre-excitation appeared during atrial pacing, or concealed pathway conduction was observed during ventricular pacing or tachycardia, with earliest atrial activation recorded at the suspected AP site. AP location was classified according to standard electrophysiological criteria. Right-septal pathways were classified based on the site of earliest atrial or ventricular activation and the position of the mapping/ablation catheter relative to the His bundle and the tricuspid annulus. Parahisian pathways were defined as those with earliest activation recorded at, or immediately adjacent to, the His bundle region. A His electrogram on the ablation catheter was considered supportive for this classification. Anteroseptal pathways were identified when the earliest activation occurred superiorly along the tricuspid annulus without a His signal, and midseptal pathways when earliest activation was located between the His region and the CS ostium. These criteria were applied systematically to ensure reproducible classification of septal pathway subtypes.

Electroanatomical mapping was used whenever available and systematically in all procedures performed from 2020 onwards. Mapping included activation mapping combined with anatomical reference points to improve localization of the accessory pathway.

Cryoablation was performed using the Freezor™ cryocatheter (Medtronic™), with a 6 mm tip, delivering energy at −70 to −80°C. For parahisian or slow-pathway modification near the His bundle, cryomapping was performed at −30°C prior to full ablation. A cryoablation lesion consisted of 4 minutes of application at −80°C with stable catheter contact (cryoadhesion). Energy delivery was immediately interrupted if transient AV block or right bundle branch block occurred, until recovery. In all cases, a minimum 20-minute waiting period was observed, followed by systematic repeat programmed stimulation to assess for recurrence of conduction or inducibility. In cases of APs, adenosine was administered.

Procedural success was defined as non-inducibility of AVNRT post-ablation, and as elimination of AP conduction (anterograde and/or retrograde) for accessory pathway ablation.

All patients were followed in the outpatient clinic by an electrophysiologist for at least 12 months post-procedure. At each visit, a standard ECG was obtained, and a 24-hour Holter monitor was performed during the first year. Arrhythmia recurrence was defined as: documentation of AVNRT or its inducibility during a repeat electrophysiologic (EP) study; recurrence of manifest pre-excitation; documentation of atrioventricular reentrant tachycardia mediated by an accessory pathway.

Descriptive statistics were used to summarize baseline characteristics and procedural outcomes. Continuous variables are presented as mean±standard deviation (SD) or median [interquartile range], depending on data distribution. Categorical variables are expressed as absolute numbers and percentages. Comparisons between subgroups were performed using the Student's t-test or Mann–Whitney U test for continuous variables, and the chi-square or Fisher's exact test for categorical variables. A two-tailed p-value <0.05 was considered statistically significant. All statistical analyses were performed using SPSS version 26.

A total of 49 patients were included, 23 of whom were male (47%). The median age was 28 years (IQR 17–50). Six patients were >65 years and had pre-existing conduction system abnormalities, which contributed to the selection of cryoablation as the preferred strategy.

The indication for the procedure was Wolff–Parkinson–White syndrome in 27 patients (55%); documented tachycardia in 18 patients (36.7%) and recurrent palpitations in the remaining patients. Seven patients had previously undergone RF ablation with subsequent arrhythmia recurrence.

Six patients (12%) presented with baseline conduction system abnormalities, namely 1st degree AV block (n=3), left bundle branch block (n=1), bifascicular block (n=1) and intermittent Mobitz I AV block (n=1). The arrhythmia substrate in this subgroup consisted of AVNRT (n=5) and intermittent WPW with palpitations (n=1). These patients were older, with a median age of 72 years (IQR 51–84). Baseline patient characteristics are summarized in Table 1.

The electrophysiologic evaluation revealed dual nodal physiology with AVNRT in 15 patients (30.6%), including two cases of atypical (slow–slow) AVNRT and one patient with triple node physiology. The remaining 34 patients (69.4%) had evidence of an AP, localized as follows: parahisian (n=24), anteroseptal (n=6), midseptal (n=3), and posteroseptal (n=1). The median tachycardia cycle length was 380 ms (IQR 310–475).

Among the patients with AVNRT, cryoablation was selected at the operator's discretion. Reasons included pre-existing conduction system disease (n=5), previous failed or recurrent ablation (n=3), atypical AVNRT (n=2), and perceived high-risk anatomy or young age in the remaining patients – 40% of this group were <40 years.

For AP, the ablation target was identified based on the earliest ventricular activation in the manifest accessory pathway or by the earliest atrial activation during ventricular pacing or orthodromic tachycardia.

Cryomapping at −30°C was used to confirm target sites when pathway conduction was abolished without affecting AV nodal conduction. Full cryoablation was delivered when these conditions were met, with cryoadhesion ensuring catheter stability.

For AVNRT, the target was the posteroinferior right atrial septum near the CS ostium (slow pathway region). Cryomapping and cryoablation were performed while pacing from the proximal CS catheter (drive cycle length of 600–550 ms and a S2 beat) to demonstrate fast pathway block and slow pathway conduction. The site was considered suitable when cryomapping suppressed slow pathway conduction without impairing the AV nodal conduction and rendered tachycardia non-inducible. Cryomapping was used in approximately two-thirds of all the procedures. The median duration of cryoenergy delivery was 5 minutes (IQR 4–6). For AVNRT, there was acute procedural success in 100% (15/15 patients) of the cases.

For AP, the overall acute success rate was 76% (26/34 patients). Eight procedures were unsuccessful. One case required conversion to radiofrequency ablation during the same session. The remaining failed ablations involved electrophysiologically low-risk pathways with an effective refractory period >240 ms in patients with mild or infrequent episodes of palpitations. One patient developed transient complete atrioventricular block during multiple applications. Notably, none of the patients with unsuccessful accessory pathway ablation experienced tachycardia recurrence during follow-up.

All patients with a history of prior unsuccessful RF ablation or with post-RF recurrence achieved acute success with cryoenergy, and none of them experienced further recurrences during follow-up. Zero-fluoroscopy procedures were performed in 24% of cases. Representative cases of accessory pathway and AVNRT ablation are shown in Figures 2 and 3, respectively.

Figure 2.

Electroanatomical maps of the right atrium. (A) Local activation map during sinus rhythm and pacing from proximal CS electrodes (left anterior oblique view). (B) Peak frequency and emphasis maps (left anterior oblique view) obtained during sinus rhythm and CS pacing. The peak frequency map displays the distribution of high-frequency electrogram components, where higher values (white areas) indicate sharper, near-field local activations. The emphasis map quantifies the stability and organization of the local signals, with higher values (purple areas) representing more regular and well-defined electrograms. The frequency cutoff value was set to 300 Hz. The area of higher peak frequency was more focused on the right anteroseptal region. (C) Atrioventricular electrogram separation showing disappearance of pre-excitation during cryoenergy application. (D) Electrocardiogram at the end of the procedure.

Figure 3.

Slow-pathway cryoablation. In panel A, left lateral view of virtual three-dimensional geometry of right atrium showing a catheter in CS (yellow), orange dots in the area of His bundle and cryoablation lesions in slow-pathway region (blue). In panel B, intracavitary signals showing slow-pathway potential in ablation catheter.

No major complications were observed. Minor complications included: transient advanced AV block in three patients (one with AVNRT and two with an anteroseptal AP), right bundle branch block in three patients (all with parahisian AP), Mobitz I AV block in one patient (with AVNRT) and AH prolongation in two patients (both with AVNRT). All events resulted in immediate interruption of energy delivery and resolved completely. No cases of permanent AV block occurred in this cohort.

When the population was stratified according to the date of ablation, relevant differences became apparent. Before 2016, 18 patients underwent cryoablation: four with AVNRT and 14 with AP. Acute success was 100% for AVNRT and 50% for AP (seven unsuccessful cases). No zero-fluoroscopy procedures were performed during this period. Minor complications consisted of two cases of transient advanced AV block (both in patients with anteroseptal AP) and one case of right bundle branch block. One patient who underwent AP ablation experienced recurrence eight months after the procedure.

After 2016, 31 patients were included: 11 with AVNRT and 20 with AP. Acute success was 100% for AVNRT and 95% for AP (one unsuccessful case). Zero-fluoroscopy ablation was performed in 39% of cases. Acute complications included one episode of transient advanced AV block (AVNRT), two cases of right bundle branch block, one Mobitz I AV block, and two cases of AH prolongation. During follow-up, four recurrences were documented (overall recurrence rate 13%): two in patients with AVNRT and two in patients with AP.

Over a median follow-up (FUP) of 13 months (IQR 1–60), arrhythmia recurrence occurred in 5 of 41 patients (12.2%).

The recurrence rate was 13.3% for AVNRT (2/15 patients) and 11.5% for AP (3/26 patients).

The median time to recurrence was 3 months (IQR 1–12), with a 6-month recurrence rate of 7.3% (Figure 4).

Figure 4.

Kaplan–Meier curve of survival free of recurrence after 5 years of follow-up comparing AVNRT and AP patients. AVNRT: atrioventricular nodal reentry tachycardia (AVNRT); AP: accessory pathway.

During FUP, three patients with previous conduction disturbances required permanent pacemaker implantation more than 1 year after ablation, due to progressive conduction disease unrelated to the procedure. During the index procedure, one of these patients developed transitory Mobitz I AV block and another experienced complete AV block during cryomapping (cryoenergy was subsequently delivered at an alternative site). All three patients had a normal HV interval at the end of the procedure and no additional conduction abnormalities on 24-hour Holter monitoring at the one-year follow-up.

This study has several limitations that should be acknowledged. First, it is a retrospective, single-center analysis with a relatively small sample size, which may limit the generalizability of the findings. In addition, patient selection was highly specific: most unsuccessful procedures involved parahisian APs with long refractory periods and minimal symptom burden, which may have introduced bias to the acute success rates in this subgroup.

Although the duration of follow-up was relatively long, the subclinical recurrence of arrhythmia may have been underestimated, as systematic Holter monitoring beyond the first year was not routinely performed. Furthermore, the absence of a direct comparison group undergoing RF ablation prevents definitive conclusions regarding the relative efficacy of cryoablation versus RF in this clinical setting.

Finally, advances in catheter technology and increasing operator experience over the long study period (2007–2025) may have influenced outcomes and introduced heterogeneity that cannot be fully accounted for. In the early phase of the study, limited cryocatheter availability and the operators’ initial learning curve restricted patient inclusion. Moreover, concerns regarding higher recurrence historically associated with cryoablation further limited its use to a highly selected patient population.