Catheter ablation is a widely used treatment for atrial fibrillation (AF), with pulmonary vein isolation (PVI) at its core. Durable PVI depends on both adequate lesion depth and the continuity of radiofrequency lesions. The CLOSE protocol, first described by Duytschaever et al. in 2017,1 established a maximum intertag lesion distance (ITD) of 6.0mm combined with region-specific ablation index targets, achieving high rates of first-pass isolation and strong 1-year arrhythmia-free survival. Despite its widespread adoption as the gold standard in many electrophysiology laboratories, ITD is routinely assessed by visual inspection only rather than precise measurement. We aimed to assess the real-world impact of ITD on procedural outcomes using a dedicated semiautomatic measurement tool.

We conducted a retrospective, investigator-initiated study at 3 tertiary hospitals in Spain (2019-2022), enrolling 459 consecutive patients with paroxysmal or persistent AF undergoing first-time PVI under the CLOSE protocol. Ablation was performed using the CARTO 3 system with the VISITAG SURPOINT module (Biosense Webster, Inc., United States), targeting a maximum ITD of 6.0mm and a tag index> 380 posterior/> 500 anterior. To overcome the limitations of visual ITD assessment, we developed a semiautomatic Python-based algorithm, which was previously validated and publicly available.2 The tool automatically groups lesion sets, reconstructs the ablation sequence around each vein, and calculates the ITD between consecutive tags, flagging any distance exceeding 6.0mm as a gap. Given the small number of patients without gaps, 3 groups were defined: strict-CLOSE (0 gaps), near-CLOSE (1–2 gaps), and non-CLOSE (≥ 3 gaps). Patients were followed up at 3, 6, and 12 months, and most were additionally evaluated at 18 and 24 months during ambulatory visits. The primary endpoint was freedom from atrial arrhythmia recurrence, documented by ECG or symptoms, beyond a 3-month blanking period at 12 months. The study was conducted in accordance with the Declaration of Helsinki and was approved by the Comité Territorial Santiago-Lugo Ethics Committee, which waived the requirement for informed consent.

Acute procedural success (all 4 PVs isolated) was achieved in 99% of cases, and the overall arrhythmia-free rate at 1 year was 83.1%. The algorithm demonstrated very high sensitivity for detecting ITD gaps. Quantitative analysis showed that only 17.8% of procedures had 0 gaps at the 6.0mm threshold when measured precisely, compared with the near-universal assumption of compliance based on visual assessment alone. Categorically, 16.7% had no gaps, 32.3% had 1 to 2 gaps, and 51.0% had 3 or more gaps. The median gap size was 6.81mm, and 60% of gaps were less than 1mm beyond the target.

All patient-related variables that were significant on univariate analysis (age, hypertension, chronic kidney disease, CHA2DS2–VASc score, and AF type) were entered into a multivariable model, together with procedural variables.

In the multivariable analysis, clinical variables lost independent significance, while mean ITD (P<.001) and number of gaps (P=.009) remained highly significant predictors of recurrence. When the strict-CLOSE and near-CLOSE groups were combined and compared with the non-CLOSE group, recurrence rates were 6.98% vs 20.9%; P=.0006. Kaplan-Meier analysis (figure 1), confirming that the gap-based groups diverged progressively over time.

Figure 1.

Kaplan-Meier survival curves for clinical success stratified by number of gaps: near-CLOSE (1–2 gaps), non-CLOSE (≥ 3 gaps), strict-CLOSE (0 gaps). A gap is defined as an interlesion distance greater than 6mm.

(0.12MB).

This study demonstrates that visual assessment of ITD is systematically insufficient: more than 80% of procedures had at least 1 undetected gap when measured precisely. Most of these gaps were small, making them difficult to identify without dedicated tools (figure 2), but were nevertheless clinically consequential. Several mechanisms contributed to this discrepancy: tag final positions can drift up to 1.5mm from their initial placement, color-coding of subthreshold tags may create a false sense of contiguity, and first-pass isolation can lead operators to forgo a final gap check.

Figure 2.

Series of images of near-CLOSE (1-2 gaps), non-CLOSE (≥ 3 gaps) and strict-CLOSE group. AI, ablation index; LA, left atrium; WACA, wide area circumferential ablation.

(0.94MB).

These findings align with prior work showing the benefits of lower ITD by Bo et al.,3 who independently identified a similar high prevalence of gaps when ITD was systematically measured, prompting a comparable near-CLOSE categorization. Our results could also help explain the variability in radiofrequency ablation success rates among centers sometimes departing substantially from those reported in initial submissions: slight, visually undetected deviations from protocol can meaningfully reduce long-term efficacy. In agreement with our results, other large multicenter studies are beginning to report findings stratified by strict vs non-strict workflow adherence, such as the SmartFIRE study.4

This study has several limitations inherent to its retrospective, nonrandomized, observational design across 3 centers. The success measure was assessed solely through 12-lead ECG at scheduled ambulatory visits and hospital emergency admissions. While consistent with standard clinical practice, this approach likely underestimates the true recurrence rate. Once ITD was strictly measured, the number of gaps and mean ITD were the strongest independent predictors of recurrence in this cohort, outweighing traditional clinical risk factors. We encourage all centers performing radiofrequency ablation to incorporate systematic ITD measurement as a standard procedural checkpoint, taking advantage of the alert tools now available in modern navigation systems.

Even in the era of pulsed field ablation, focal tip or large footprint catheters are expected to be relevant for PVI and therefore lesion contiguity adherence will remain critical for long-term durability, making objective metrics worthy of dedicated evaluation.