Temperature-Controlled Lattice Ablation Catheter — Preclinical Biophysics (2019)

Authors, Journal, Affiliations, Type, DOI

• Authors: Ayelet Shapira-Daniels, Michael Barkagan, Hagai Yavin, Jakub Sroubek, Vivek Y. Reddy, Petr Neuzil, Elad Anter
• Journal: Circulation: Arrhythmia and Electrophysiology 2019;12(11):e007661
• Affiliations: Harvard-Thorndike Electrophysiology Institute, Beth Israel Deaconess Medical Center; Helmsley Center for Electrophysiology, Mount Sinai, New York; Na Homolce Hospital, Prague
• Type: Preclinical multimodality study — ex vivo ablation in 11 hearts, in vivo thigh-muscle thermometry, in vivo beating-ventricle ablation in swine
• DOI: 10.1161/CIRCEP.119.007661

Overview

The biophysical paper behind the lattice-tip concept, and the one that explains why the electrode works: a 9-mm expandable nitinol lattice raises the effective ablating surface area from 28 mm² to 275 mm², which permits roughly fifteen times more current to be delivered at about half the current density. In ex vivo bovine and porcine ventricles, in vivo swine ventricles and a perfused thigh-muscle preparation, this produced lesions 40–50% deeper and wider (300–700% greater volume), higher tissue temperature at 7 mm depth, and fewer steam pops than a 40 W irrigated linear catheter. Bipolar delivery between two lattice tips produced 15.8-mm lesions that were transmural in 80% of applications versus none with conventional catheters. The study also documents the conceptual limit that later mattered: surface thermocouples read the endocardial surface, not the subendocardium where maximal heating and steam pops occur.

Keywords

Catheter; ablation; radiofrequency; ventricular arrhythmia

Key Takeaways

Device and rationale

• 8-Fr bidirectional deflectable catheter with an expandable nitinol lattice forming a 9-mm sphere once out of the sheath; effective surface area 275 mm² versus 27–31 mm² for a standard linear catheter, with current distributed uniformly around the sphere.
• Nine temperature sensors that also function as electrodes (two rings of four plus one at the tip, ~5 mm apart), a central protected irrigation nozzle with micropores, and a proprietary high-current generator delivering in temperature-controlled mode (Tmax 60 °C), with rapid feedback titration of current.
• The problem being solved: increasing power through a 3.5-mm tip raises current density, subendocardial overheating and steam-pop risk. Irrigation shifts maximal heating to the subendocardium rather than preventing it; adjunctive techniques that increase lesion size by 10–25% (prolonged duration, lower baseline impedance, hypo-osmolar irrigant, bipolar/unipolar pairs) also increase steam pops.
• At 40 W and 110 Ω, current is ~600 mA over 28 mm² ≈ 20 mA/mm². The lattice delivers up to 3,700 mA but at a maximum current density of 13 mA/mm².

Ex vivo unipolar ablation (11 hearts, 101 lesions)

(P≤0.001 for all comparisons; contact force fixed at 10 g; comparator 3.5-mm ThermoCool SmartTouch SF at 40 W, same generator.)

• Current-squared was 15–19-fold higher with the lattice (5.7 vs 0.3 A² at 30 s), while current density was ≥50% lower (8.7 vs 20.1 mA/mm² at 30 s; 7.6 vs 20.2 at 120 s).
• The proportional gain was constant across durations: ~40–50% greater depth and width, 300–700% greater volume.

Ex vivo bipolar ablation (5 porcine ventricles, 20 lesions, 60 s)

• Lattice 15.8 ± 4.1 mm deep and 18.1 ± 2.8 mm wide versus linear 10.5 ± 1.4 mm and 10.1 ± 1.7 mm (P≤0.001).
• Transmurality 8/10 (80%) with the lattice versus 0/10 with the linear catheter (P<0.001); the lattice penetrated 92.4 ± 16.2% of wall thickness versus 56.9 ± 14.6%.
• Current-squared 10-fold higher (3.3 vs 0.3 A²) with density 3-fold lower (6.5 vs 19.1 mA/mm²).
• With either catheter, bipolar beat unipolar for lesion size.

In vivo beating ventricle (swine)

• Lattice (Tmax 60 °C, 60 s) versus linear (40 W, 60 s, target CF 10–20 g), on separate animals because the mapping systems are incompatible: depth 10.5 ± 1.4 vs 6.5 ± 0.8 mm; width 27.3 ± 5.2 vs 9.2 ± 1.3 mm; volume 4.3 ± 1.8 vs 0.3 ± 0.1 mL (P<0.001).
• Current 2,582 ± 149 mA versus 566 ± 28 mA (22-fold greater current-squared) at half the current density (9.4 vs 20.2 mA/mm²). Baseline impedance is lower with the lattice (37 vs 131 Ω) because of the different mapping systems, so impedance values were not compared.
• Electrograms attenuated within ~3 seconds of energy delivery at the ablation site.

Intramuscular thermometry (perfused swine thigh)

• At 3 mm depth the temperature profiles and maxima were similar between catheters.
• At 7 mm depth the lattice produced +15.1 ± 2.4 °C higher tissue temperature throughout the application — the direct evidence for deeper heating rather than simply hotter surfaces.
• Steam pops occurred in 92% of linear-catheter applications in this preparation versus none with the lattice (P<0.001).

Catheter–tissue footprint (ink imprinting)

• Surface area at 0°/45°/90°: lattice 101.5 / 92.6 / 130.7 mm² versus linear 27.5 / 18.5 / 10.8 mm² — the sphere's contact area is larger and, crucially, almost angle-independent, reducing the risk of "hot spots" when the catheter approaches tissue obliquely.

Safety

• Steam pops: ex vivo 3/52 (5.7%) lattice vs 9/49 (18.4%) linear (P=0.02); in vivo 0/15 vs 2/14 (P=0.13). No char or coagulum with either catheter.

Mapping implications

• Nine microelectrodes of ~0.5 mm² over 9 mm give high-resolution data; the central electrode is by design never in contact with tissue, which helps separate near-field from far-field potentials; fixed interelectrode distances allow surface conduction velocity to be estimated from a single beat.

The authors' own caveat

• Surface temperature is not tissue temperature. The thermocouples sit on the endocardium, whereas maximal heating and steam-pop formation occur in the subendocardium, so surface readings will underestimate maximal tissue temperature. The surface–tissue relationship is non-linear and depends on duration, contact force, current density, tissue conductivity and microstructure — so the 60 °C limit may need adjustment for different durations and tissues.

Limitations of the document

• Normal myocardium only, acute endpoints. No scarred or heterogenous substrate, and no chronic healing; the authors state that clinical scar may behave differently and call for human study.
• Ex vivo and animal models. Bovine and porcine ventricular geometry, perfusibility and conductivity differ from human scar, and the thigh-muscle preparation is muscle, not myocardium.
• In vivo comparison was not randomised within animals — the two systems are mutually incompatible, so each technology was used in separate swine of similar age and weight, confounding lesion size with animal and operator.
• Industry funding and equity. Funded in part by an Affera research grant; Anter and Reddy hold stock options in the company.
• Proprietary data not shared, as stated in the methods.
• Thermal safety of higher power is not fully characterised. The authors acknowledge that the benefit of higher-power technology may be offset by greater collateral damage, and the study did not systematically assess coronary, phrenic or oesophageal effects.
• Internal inconsistencies in reporting. The abstract describes "11 bovine hearts" for the unipolar phase, whereas the methods describe 11 ex vivo hearts comprising 6 bovine and 5 porcine; the abstract cites 9 swine whereas the methods describe 10 in vivo Yorkshire swine. Lesion numbers (101 ex vivo, 34 in vivo) do not reconcile exactly with the per-group counts reported.

Key Concepts Mentioned

• concepts/VT-Ablation — the lesion-depth ceiling this technology targets
• concepts/Ventricular-PFA — the pulsed-field mode of the same platform
• concepts/Epicardial-VT-Ablation — the transmurality problem
• concepts/VT-Ablation-Nonischemic-Cardiomyopathy — septal and intramural substrate
• concepts/PVC-Mapping-Ablation — intramural escalation strategies for comparison

Key Entities Mentioned

• Sphere-9 lattice catheter, HexaGen generator, HexaPump, HexaMap, Prism-1 mapping system (Affera, Inc.)
• ThermoCool SmartTouch SF, SmartAblate generator, CARTO 3 (Biosense Webster)
• Beth Israel Deaconess Medical Center, Boston

Wiki Pages Updated

• Created: wiki/sources/lattice-tip-rf-lesion-circep-2019.md
• Created: wiki/entities/Sphere-9.md
• Updated: wiki/concepts/VT-Ablation.md, wiki/concepts/Ventricular-PFA.md, wiki/concepts/Epicardial-VT-Ablation.md
• Updated: wiki/sourceindex.md, wiki/wikiindex.md, log.md