Identifying patients who benefit from the S3 protocol in ventricular tachycardia ablation
T Rosseel, E Ayauja Lopez, C Paires-Ramis, R Pittorru, P Bhagirath, R Borras, A Victor-Baldoma, M Regany-Closa, J Reventos-Presmanes, P Rodriguez, J Llevadot, J B Guichard, A Porta-Sanchez, I Roca-LuqueAbstract
Background and purpose
Using two extrastimuli during substrate mapping (the S3 protocol) has been shown to unmask concealed arrhythmogenic substrate and better identify the critical sites for VT compared with S1 mapping alone. However, not all patients undergoing VT ablation exhibit hidden substrate. This study aimed to identify which subset of patients particularly benefits from the S3 protocol, in order to optimise mapping efficiency and procedural strategy.
Methods
87 consecutive patients undergoing ventricular tachycardia (VT) ablation in whom the S3 protocol was applied were included in the analysis. A successful S3 protocol was defined as the detection of (1) additional DZs, (2) an increased number of regions with LPs, or (3) a larger overall LP surface area compared with S1 mapping. Clinical characteristics and pre-procedural cardiac MRI parameters were evaluated in relation to each of these outcomes. Scar characteristics and conduction channel (CC) properties were quantified using late gadolinium-enhanced MRI processed with ADAS 3D software.
Results
In 15 patients the S3 protocol was not feasible due to repetitive induction of VT during the stimulation protocol. Seventy-two patients were analysed (5.6% female; 69% ischaemic cardiomyopathy; mean LVEF 35 ± 11%). Overall, 60 of 72 patients (83%) demonstrated previously hidden substrate during S3 compared with S1. In univariate analysis, lower LVEF (OR 0.95, 95% CI 0.91–1.00, p = 0.034), a higher number of CCs longer than 10 mm (OR 1.68, 95% CI 1.07–2.63, p = 0.025), and few DZs during S1 mapping (≤ 1 DZ; OR 3.12, 95% CI 1.16–8.41, p = 0.024) were significantly associated with the detection of additional DZs in S3. The use of antiarrhythmic drugs other than beta-blockers showed a trend (OR 3.18, p = 0.058). No significant predictors were identified for the other S3 outcomes. In multivariate analysis, the number of CCs > 10 mm remained the only independent predictor (OR 1.81, 95% CI 1.05–3.12, p = 0.033), while the number of DZs in S1 showed a non-significant trend (p = 0.079). LVEF was not independently associated with S3 positivity. ROC analysis identified ≥ 4 CCs > 10 mm as the optimal empirical cut-off (AUC 0.67), yielding a positive predictive value of 92.9% and specificity of 95.5% for additional substrate during S3.
Conclusion
Patients with multiple conduction corridors on MRI are most likely to benefit from the S3 protocol, as S1 mapping alone may miss relevant hidden substrate. When initial S1 mapping reveals minimal functional substrate (≤ 1 DZ), adding S3 often unmasks additional delayed zones. Since 83% of patients in this cohort exhibited extra substrate with S3, these results support routine use of the S3 protocol in VT ablation to improve substrate identification.