Dual-energy lattice-tip ablation system for persistent atrial fibrillation: a randomized trial

Atrial fibrillation (AF) is the most common cardiac arrhythmia and a leading cause of stroke. Catheter ablation with pulmonary vein isolation (PVI) is standard therapy, but outcomes in persistent AF are less favorable than in paroxysmal AF due to a broader, more complex arrhythmogenic substrate. Conventional focal radiofrequency ablation requires sequential point-by-point lesions and often leaves conduction gaps, with risks including atrio-esophageal fistula, phrenic nerve paralysis, and pulmonary vein narrowing, and typically requires separate catheters for mapping and ablation. Single-shot PVI tools ease vein isolation but are limited in creating additional linear lesions needed in persistent AF. The investigational platform integrates high-density electro-anatomical mapping and dual-energy ablation (radiofrequency and pulsed field ablation, PFA) in a single large-footprint lattice-tip catheter designed to create more contiguous, wider lesions with potential myocardial selectivity of PFA to minimize collateral injury. The SPHERE Per-AF trial tested whether this dual-energy lattice-tip system is non-inferior in safety and effectiveness compared with a conventional radiofrequency system in drug-refractory persistent AF.

The study builds on prior evidence that outcomes of ablation for persistent AF are historically suboptimal (approximately 45–62% single-procedure success at 1 year in multicenter trials, including PRECEPT). Conventional focal RF ablation is challenged by achieving durable, contiguous lesion sets, with risks of thermal injury to adjacent structures. Pulsed field ablation has shown myocardial selectivity and reduced collateral tissue injury in preclinical and first-in-human studies. Early studies of the lattice-tip catheter demonstrated rapid, durable PVI with fewer applications and full-thickness lesions, suggesting potential improvements in durability and efficiency over conventional approaches. The trial also references comparative safety data (e.g., silent cerebral lesions across technologies) and randomized evaluations of additional linear lesion sets in persistent AF, which have generally not demonstrated clear benefit.

Design: SPHERE Per-AF (NCT05120193) was a pivotal, multicenter, randomized, single-blind, non-inferiority trial across 23 centers in the United States, Czech Republic, and Israel. An independent data and safety monitoring board oversaw safety; an independent, blinded clinical events committee adjudicated safety events; and core laboratories, blinded to assignment, adjudicated rhythm monitoring and MRI endpoints. The trial was sponsored by Affera, Inc. (acquired by Medtronic). FDA and institutional approvals were obtained. Participants: Adults 18–80 years with symptomatic persistent AF, refractory or intolerant to at least one class I/III anti-arrhythmic drug, eligible for catheter ablation. Key exclusions included continuous AF ≥12 months, prior left atrial ablation, significant structural heart disease constraints, LVEF <35%, LA diameter >55 mm, BMI >40 kg/m2, recent cardiac procedures, and thromboembolic disease. Written informed consent was obtained. Randomization and blinding: After a roll-in phase (up to two patients per site), patients were randomized 1:1 to investigational dual-energy lattice-tip system or control conventional RF system, blocked and stratified by site and neurological substudy enrollment. Patients were blinded to assignment; operators were not. Interventions: Investigational arm used the Sphere-9 lattice-tip catheter with the Affera Mapping and Ablation System, capable of both mapping and ablation. RF applications were temperature-controlled (target 73 °C, 5 s duration; current limit 80–90%); PFA applications were microsecond pulses delivered for 4 s. Operators were advised to use PFA on the posterior wall, around the left inferior PV, and near the phrenic nerve; elsewhere, energy selection was at discretion. Control arm used a standard electro-anatomical mapping system (Carto3), multi-electrode mapping catheter, and contact force-sensing RF ablation catheter (THERMOCOOL SMARTTOUCH SF). All procedures required high-density mapping. Protocol mandated wide-area circumferential PVI with entrance block after a ≥20-min observation or adenosine/isoproterenol. Cavo-tricuspid isthmus ablation was required if typical flutter was documented. Additional linear lesions were permitted for documented macro-reentrant tachycardias; empiric lines were generally discouraged. Block across lines was confirmed by pacing maneuvers and activation mapping per standard practice. Follow-up: Patients were followed for 12 months with visits at 1, 3, 6, and 12 months. After a 90-day blanking period, monthly trans-telephonic ECG monitoring was required, with symptom-triggered transmissions. A 24-hour Holter was performed at 6 and 12 months; 12-lead ECGs at 3, 6, and 12 months. QOL was assessed by AFEQT and SF-12v2 at baseline and 12 months. A neurological substudy obtained brain MRI (DWI and FLAIR) within 72 hours post-ablation and at 90 days if acute ischemia was detected. Endpoints: Primary effectiveness was freedom from a composite of acute failure to isolate all targeted PVs or complete left atrial ablation with the assigned device, any repeat ablation at any time, and, after a 3-month blanking period, documented atrial tachyarrhythmia (AF/AT/AFL) ≥30 s (or entire 10 s 12-lead ECG), escalation/initiation of class I/III anti-arrhythmic drugs, or cardioversion. Primary safety was a composite of device- or procedure-related serious adverse events: death, atrio-esophageal fistula, stroke/TIA, MI, cardiac tamponade/perforation, PV stenosis, phrenic nerve paralysis, thromboembolism, major vascular complications/bleeding, heart block, gastroparesis, severe pericarditis, or new/extended hospitalization for cardiovascular or pulmonary adverse events. Secondary pre-specified outcomes (tested sequentially upon trial success) included energy application time, elapsed ablation time, total procedure time, and superiority of primary effectiveness; additional endpoints included fluoroscopy time, lesion sets, anti-arrhythmic drug use, and QOL changes. Neurological substudy assessed silent ischemic lesions. Statistics: Non-inferiority was tested by Farrington-Manning method with one-sided alpha 0.05 (safety, margin 8%) and 0.025 (effectiveness, margin 15%). A sample size of 350 evaluable patients (175/arm) provided >80% power; 410 randomized planned to allow for attrition. Kaplan–Meier analyses and log-rank tests assessed time-to-event effectiveness. SAS v9.4 was used. Primary analysis cohort included randomized and treated patients with available outcome data.

- Enrollment and follow-up: 469 enrolled; 432 randomized (219 investigational, 213 control); 420 treated (212 investigational, 208 control) across 20 centers and 40 operators. Overall visit adherence 97%; Holter 84%; 12-lead ECG 85%; trans-telephonic monitoring 92%. 408/420 (97.1%) completed 12-month follow-up. - Primary effectiveness: Success in 73.8% (155/210) investigational vs 65.8% (133/202) control; difference 8.0% (95% CI -0.9% to 16.8%); non-inferiority met (P<0.0001). KM estimates at 1 year: 73.5% investigational vs 65.2% control. Post hoc freedom from atrial arrhythmias: 76.7% (161/210) investigational vs 72.8% (147/202) control; non-inferiority met (P<0.0001). Failure components included documented AF/AT/AFL recurrence (23.3% vs 27.2%), AAD escalation/initiation (3.8% vs 7.4%), repeat ablation (4.8% vs 8.4%). All patients had PVI; the control arm required adjunctive cryoballoon for one PV in one patient and vein of Marshall alcohol ablation to complete one mitral line. - Primary safety: Events in 1.4% (3/212) investigational vs 1.0% (2/208) control; difference 0.4% (90% CI -2.8% to 3.7%); non-inferiority met (P<0.0001). Events were hospitalizations for pulmonary/cardiovascular issues (e.g., pulmonary edema, COPD exacerbation, hemoptysis); no deaths, stroke/TIA, tamponade, atrio-esophageal fistula, PV stenosis, or permanent phrenic nerve paralysis within the primary safety windows. Five deaths occurred post-ablation but were adjudicated not related to device/procedure (2 investigational; 3 control). - Procedural efficiency (superiority): • Total energy application time: 7.1 ± 2.0 min vs 36.4 ± 17.7 min (difference -29.2 min; 95% CI -31.7 to -26.8; P<0.0001). • Transpired ablation time (first-to-last application): 46.7 ± 20.0 min vs 73.5 ± 34.4 min (difference -26.8 min; 95% CI -32.2 to -21.4; P<0.0001). • Skin-to-skin procedure time: 100.9 ± 30.8 min vs 126.1 ± 49.2 min (difference -25.1 min; 95% CI -33.0 to -17.3; P<0.0001). • Fluoroscopy: 4.9 ± 6.6 min vs 6.3 ± 9.1 min. PVI time: 25.9 ± 10.7 min vs 53.6 ± 28.8 min. Lower catheter irrigation volume and less esophageal temperature probe/deviation use in investigational arm. Single transseptal access used in 95.3% of investigational vs 62.0% of control procedures; one catheter used for mapping/ablation in 97.2% investigational vs typically two in control. - Superiority of primary effectiveness was not achieved (two-sided P=0.078). - PVI durability at redo: Among 26 patients with repeat ablation (10 investigational, 16 control; plus 1 control surgical redo), PVI durability was 50% per patient and 66.7% per vein in investigational vs 18.8% per patient and 48.4% per vein in control. - Neurological substudy: Silent cerebral ischemic lesions with FLAIR hyperintensity detected in 8.1% (3/37) investigational vs 5.7% (2/35) control; most resolved by 90 days. - Quality of life: AFEQT improved from 66.7 ± 21.5 to 89.0 ± 13.7 (change 22.3 ± 19.5) in investigational and from 68.7 ± 20.2 to 90.9 ± 13.3 (change 22.2 ± 19.3) in control; SF-12v2 physical and mental scores improved similarly in both arms.

The dual-energy lattice-tip catheter system achieved non-inferior safety and effectiveness versus a conventional RF mapping/ablation platform in patients with drug-refractory persistent AF, with marked improvements in procedural efficiency. The separation of Kaplan–Meier curves after the 90-day blanking period suggests a potential effectiveness advantage, though formal superiority was not reached. Safety event rates were low in both arms with no major complications such as atrio-esophageal fistula, tamponade, PV stenosis, permanent phrenic nerve paralysis, or adjudicated device/procedure-related death. The integrated mapping-ablation approach, large-footprint lattice tip, lower tissue pressure, and the ability to toggle to PFA—particularly near sensitive structures—may contribute to fewer catheter exchanges, fewer transseptal punctures, and shorter procedures. Redo procedure data suggested higher PVI durability in the investigational arm. These findings support the investigational system as an effective alternative to conventional RF systems for persistent AF ablation, with workflow and efficiency benefits that may translate into broader clinical value.

In a randomized, single-blind non-inferiority trial of persistent AF ablation, the lattice-tip dual-energy (RF/PFA) catheter integrated with high-density mapping demonstrated non-inferior effectiveness and safety compared to a conventional RF system and provided superior procedural efficiency (shorter energy application, ablation, and total procedure times). PVI durability at redo favored the investigational device. The results support the dual-energy lattice-tip platform as a viable and efficient option for treating persistent AF. Future work should include larger, longer-term, and real-world studies to confirm durability, assess outcomes across broader patient populations, evaluate resource utilization, and clarify the role of additional linear ablation strategies.

- Potential under-detection of asymptomatic atrial tachyarrhythmias due to absence of continuous invasive monitoring, although randomized design likely balanced this between groups. - Heterogeneity in additional ablation lines, with relatively high rates of linear lesions—particularly in the investigational arm—though post hoc analyses did not show impact on the primary outcome; the study was not powered to assess empiric lines. - Operators were highly experienced with conventional systems but had limited prior experience with the investigational device; learning curve effects may influence outcomes. - Follow-up limited to 12 months; longer-term durability and safety require further study. - Sponsor involvement in data collection and analysis, although independent committees and core labs were used for adjudication and blinding of key endpoints.