Epicardial VT Ablation
Definition
Epicardial VT ablation is catheter mapping and energy delivery on the epicardial surface of the ventricles after percutaneous or surgical access to the pericardial space. It is used when the arrhythmogenic substrate is epicardial or midmyocardial and therefore inaccessible or insufficiently reachable from the endocardium, or when endocardial ablation has failed. It is not a routine first approach: most patients undergo endocardial mapping and ablation first, and epicardial access is reserved for substrates with a high pre-test probability of epicardial involvement.
Key Concepts
Patient selection by cardiomyopathy
| Condition |
Likelihood and pattern of epicardial substrate |
| Ischaemic cardiomyopathy |
19–39% of patients referred for epicardial ablation; epicardial substrate required in 13–29%; scar is usually subendocardial but may extend transmurally |
| Non-ischaemic cardiomyopathy |
Epicardial ablation required in 32–63%; basal perivalvular, midmyocardial and epicardial fibrosis |
| ARVC |
Fibrofatty replacement progresses from epicardium inward; epicardial substrate exceeds endocardial; peritricuspid and RVOT/infundibular regions, RV anterior and inferior free walls; apex less often involved |
| Chagas cardiomyopathy |
Epicardial scar area about twice the endocardial area; combined endo-epicardial ablation improves freedom from VT |
| Brugada syndrome |
Epicardial RVOT substrate; ablation eliminates the type 1 ECG pattern and reduces arrhythmia more than endocardial-only ablation |
| Long QT syndrome |
Epicardial RV fragmented, low-voltage, prolonged electrograms reported in high-risk patients; ablation suppressed arrhythmia and reduced QTc over 12 months |
| Cardiac sarcoidosis |
Epicardial mapping/ablation beneficial in 18–25% |
| Viral myocarditis |
Drug-refractory VT may require epicardial ablation in about 30% |
(sources/epicardial-vt-ablation-jaccep-2024, rating: high)
ECG criteria for epicardial origin
- Morphology. An endocardial exit produces an initial R wave in the lead overlying the region; an epicardial exit produces a Q wave because activation travels away from that lead.
- Q wave in lead I → superior left lateral LV or anterior RV epicardium.
- Q waves in II, III and aVF → inferior LV or inferior RV epicardium.
- QS in V2 → anterior epicardial exit near the anterior interventricular sulcus.
- Absence of inferior Q waves has a high negative predictive value for inferior epicardial origin.
- Intervals. Slow initial conduction through myocardium before the conduction system is engaged widens the initial QRS.
| Criterion |
Threshold |
Reported sensitivity / specificity |
| Precordial pseudo-delta wave (PdW) |
≥34 ms |
83% / 95% (Berruezo); 73% / 63% (Valles) |
| V2 intrinsicoid deflection time |
≥85 ms |
87% / 90% (Berruezo); 83% / 70% (Valles) |
| Shortest precordial RS interval |
≥121 ms |
76% / 85% (Berruezo); 74% / 57% (Valles) |
| Myocardial deflection index (MDI) |
≥0.55 |
100% / 99% (Daniels); 30% / 95% (Valles) |
| Q wave in lead I |
Present |
91% / 96% for basal superior/lateral LV (Valles) |
- Valles' four-step algorithm for basal superior/lateral LV epicardial origin — absence of inferior Q waves, PdW ≥75 ms, MDI ≥0.59, Q wave in lead I — achieved sensitivity 96% and specificity 93%.
- The criteria are site-specific and were derived in small cohorts (N=12–67); they perform poorly in transmural infarction, where infarct Q waves obscure the origin, and QRS duration does not reliably distinguish epicardial from endocardial RV VT. (sources/epicardial-vt-ablation-jaccep-2024, rating: high)
Imaging
Pericardial access
- Subxiphoid Tuohy technique (Sosa). General anaesthesia is preferred; anticoagulation should be discontinued and reversed; blood products and surgical backup should be available. Access is obtained one finger-width below and left of the xiphoid, through the foramen of Morgagni (space of Larrey). A 6-inch 17-gauge Tuohy needle with a curved blunt tip is used, with a 0.035-inch J wire kept extended past the tip except during advancement. Entry is confirmed by loss of resistance, "dye hanging" with contrast, and free wire passage into the pericardial space. (sources/epicardial-vt-ablation-jaccep-2024, rating: high)
- Anterior versus posterior approach. The anterior approach reduces intra-abdominal organ injury and is preferred by the authors; the posterior approach may be preferable when anterior adhesions are severe. Access at the mid-apical RV avoids the RV marginal and posterior descending arteries. (sources/epicardial-vt-ablation-jaccep-2024, rating: high)
- Needle-in-needle technique. An 18-gauge Cook needle creates the track, a 21-gauge micropuncture or Chiba needle punctures the pericardium, and an 0.018-inch wire confirms access. (sources/epicardial-vt-ablation-jaccep-2024, rating: high)
- CO2 insufflation via the coronary sinus. A deflectable sheath cannulates the coronary sinus, a distal posterolateral branch is punctured with a 0.014-inch high-tip-force wire, and 150–250 mL of CO2 separates the layers. Reported success is higher and complications lower than with traditional subxiphoid access; CO2 is rapidly resorbed, but pneumopericardium can raise the defibrillation threshold. (sources/epicardial-vt-ablation-jaccep-2024, rating: high)
- Surgical window. Indicated for severe adhesions, prior cardiac surgery or epicardial leads, or when direct visualisation of coronaries is needed. Options are a subxiphoid window or limited anterior thoracotomy; a perfusionist and bypass standby should be available. (sources/epicardial-vt-ablation-jaccep-2024, rating: high)
Epicardial mapping
- Clues that the circuit is epicardial or intramural: no sharp early pre-QRS signal, failure to entrain the critical isthmus, or failed ablation despite an apparent exit-site entrainment. (sources/epicardial-vt-ablation-jaccep-2024, rating: high)
- Unipolar voltage <8.3 mV in the RV and <5.5 mV in the LV suggests epicardial scar even when endocardial bipolar voltage is normal. (sources/epicardial-vt-ablation-jaccep-2024, rating: high)
- Normal epicardial bipolar voltage is >0.94 mV. Fat over the coronary sulcus and interventricular grooves can mimic scar; late, fractionated or split potentials are more reliable markers, and CT/MRI fat registration helps. (sources/epicardial-vt-ablation-jaccep-2024, rating: high)
- Thin epicardial scar over healthy myocardium can produce normal voltage; look for fractionation, late potentials and long stimulus-to-QRS. (sources/epicardial-vt-ablation-jaccep-2024, rating: high)
- High-density multipolar catheters with small bipoles are preferred because the epicardial surface is smooth and untrabeculated; linear/uniplanar catheters oppose the surface better than basket or PentaRay-type catheters. (sources/epicardial-vt-ablation-jaccep-2024, rating: high)
- Contact force: signal amplitude increases up to about 10 g; 4 g is the threshold for normal signal in the epicardium and 9 g for adequate LV epicardial contact. Median adequate CF is lower epicardially (8 g) than endocardially (13–15 g). (sources/epicardial-vt-ablation-jaccep-2024, rating: high)
Ablation technique
- Irrigated RF is preferred because the pericardial space lacks convective cooling; irrigation prevents tip overheating and creates larger lesions through fat. (sources/epicardial-vt-ablation-jaccep-2024, rating: high)
- Fluid accumulates in the pericardial space and can impair filling or cause tamponade, so intermittent aspiration through the sheath side port is used; continuous suction is avoided because it limits catheter movement. (sources/epicardial-vt-ablation-jaccep-2024, rating: high)
- Excess contact force is associated with steam pops, coronary injury and phrenic nerve injury. (sources/epicardial-vt-ablation-jaccep-2024, rating: high)
- Cryoablation is an alternative when the target is near a coronary artery or the phrenic nerve; animal data show epicardial lesions similar in size to endocardial lesions, but cryo can cause neointimal proliferation in adjacent epicardial vessels. (sources/epicardial-vt-ablation-jaccep-2024, rating: high)
- Midmyocardial substrate is approached from the endocardium with low starting power (~25 W), gradual titration to a 10–15-unit impedance drop, and 2–3-minute applications. (sources/epicardial-vt-ablation-jaccep-2024, rating: high)
Complications
| Complication |
Frequency |
Main causes |
Management |
| Early haemopericardium (within 10 min) |
5–14% |
RV puncture, epicardial vessel injury, adhesion disruption |
Coronary angiography before access, mid-ventricular access, echocardiography, RV contrast; drain, window or thoracotomy |
| Haemopericardium during the procedure |
0–1% |
Slow bleeding from multiple punctures, steam pops |
Usually self-limiting; drain if needed |
| Late haemopericardium (end of procedure) |
~1% |
Access-tract vessel injury, double RV puncture |
Advance a long wire before sheath removal; re-advance sheath to tamponade; surgical repair |
| Delayed haemopericardium (post-procedure) |
~1% |
Anticoagulation resumption, pericarditis/Dressler |
Pericarditis treatment; drainage; brief anticoagulation hold |
| Phrenic nerve injury |
0–1% |
Ablation over the nerve |
Identify by pacing; displace nerve with CO2/air/saline or a balloon; cryoablation; diaphragmatic plication for severe cases |
| Pericarditis |
Up to 30% |
Parietal layer irritation; extent of ablation |
NSAIDs and colchicine; intrapericardial triamcinolone 2–3 mg/kg; early drain removal |
| Haemoperitoneum/intra-abdominal injury |
0–1% |
Diaphragm, liver or bowel puncture |
Anterior approach, inspiration during puncture, imaging; supportive care, reverse anticoagulation, surgery for severe cases |
Relation to transmural energy delivery
- The claim that transmural endocardial lesions could make epicardial access unnecessary predates PFA. A 9-mm expandable lattice tip created 10.3 ± 2.9 mm ovine ventricular lesions versus 5.5 ± 2.4 mm with a conventional irrigated tip, and the authors argued that such depth would allow transmural ablation "in virtually any area of the LV" with preserved wall thickness, rendering epicardial ablation "anecdotal" and even permitting ablation of epicardial substrate beneath coronary arteries. (sources/lattice-tip-rf-lesion-jce-2019, rating: medium)
- That hypothesis has never been tested against epicardial ablation, and the same platform's epicardial pulsed-field lesions were only 20% transmural — so the depth argument supports an endocardial-first strategy, not the abandonment of pericardial access. See entities/Sphere-9. (sources/lattice-tip-preclinical-circep-2024, rating: medium)
- Epicardial access is one solution to intramural/epicardial substrate, but it adds access risk. High-voltage focal PFA delivered endocardially abolished adjacent epicardial abnormal potentials in 9/9 patients in a first-in-human study, suggesting that a transmural endocardial lesion could sometimes replace epicardial mapping and ablation. (sources/pfa-vt-circ-2025, rating: high)
- This is proof-of-concept only: the cohort was small, the follow-up short, and epicardial PFA lesions were avoided in most patients. (sources/pfa-vt-circ-2025, rating: high)
- Conventional pulsed field ablation does not solve the epicardial circuit. In a 3-case series, twelve endocardial applications with a pentaspline catheter created dense scar (<0.1 mV) but VT was re-induced and only epicardial RF (40 W) terminated it; the lesion-depth ceiling was 6.0 ± 1.2 mm. (sources/pfa-vt-case-series-jacep-2023, rating: low)
- Epicardial PFA is feasible but shallow and spasm-prone. In swine, epicardial monopolar lattice-tip lesions measured 30.4 × 23.5 × 9.1 mm with only 20% transmurality, sparing epicardial coronary branches with a patent lumen; in patients, epicardial PF was not transmural by endocardial signal assessment and coronary spasm occurred in 3 of 4 patients (tip-to-artery distance 7 ± 3 mm, none at 21 mm), all resolving with intracoronary nitrates. (sources/lattice-tip-preclinical-circep-2024, rating: medium) (sources/lattice-tip-vt-europace-2024, rating: medium)
- No phrenic nerve palsy occurred in the three patients with epicardial PF delivered on or within 1–2 cm of the phrenic nerve, consistent with the 0.06% rate in the 17,000-patient AF registry — a genuine advantage over RF and cryoablation in the pericardium. (sources/lattice-tip-vt-europace-2024, rating: medium)
- Bipolar PFA across the wall (endocardial tip to epicardial tip) achieved 87% transmurality at 14.3 mm in swine, which is the only PFA strategy that has reproduced transmural left ventricular ablation; it requires two catheters and carries a heart-block risk at basal septal sites. (sources/lattice-tip-preclinical-circep-2024, rating: medium)
- Transatrial CO₂/saline-assisted access (right atrial appendage puncture with a 1.9 Fr microcatheter, insufflation to separate the layers, then standard subxiphoid puncture) succeeded in 7/7 swine on full heparin with only minor pericardial staining — a technique that could reduce bleeding risk at access. (sources/lattice-tip-preclinical-circep-2024, rating: medium)
- Epicardial access complications in the largest ventricular PFA registry (AVAAR, n=126, epicardial in 17%): coronary artery laceration during access causing tamponade and infarction with subsequent death in one patient; intrapericardial saline (40–50 mL) used to ease manipulation; tissue found on the lattice tip after epicardial ablation in one patient before returning endocardially. (sources/avaar-registry-europace-2025, rating: high)
Contradictions / Open Questions
- ECG criteria are modest and site-specific. Sensitivity and specificity vary widely between derivation and validation cohorts, cut-offs differ, and the criteria fail in transmural infarction and ischaemic cardiomyopathy; no prospectively validated algorithm exists. (sources/epicardial-vt-ablation-jaccep-2024, rating: high)
- Epicardial fat versus scar. Fat dampens bipolar voltage and can be misclassified as scar; late/fractionated potentials and CT/MRI fat registration are more reliable, but the optimal mapping strategy is not standardised. (sources/epicardial-vt-ablation-jaccep-2024, rating: high)
- Staged versus upfront epicardial access. Most centres perform endocardial ablation first and proceed to epicardium selectively; whether a pre-specified upfront epicardial strategy improves outcomes in high-probability substrates has not been tested. (sources/epicardial-vt-ablation-jaccep-2024, rating: high)
- Complication rates are imprecise. Frequencies come from heterogeneous registries with different access techniques and definitions; rare but catastrophic events such as double RV puncture are likely underreported. (sources/epicardial-vt-ablation-jaccep-2024, rating: high)
- No pooled efficacy data. The review does not meta-analyse freedom from VT or mortality, and comparative effectiveness versus endocardial-only ablation is not established. (sources/epicardial-vt-ablation-jaccep-2024, rating: high)
- Transmurality may be undesirable. More extensive epicardial ablation can damage viable myocardium and coronary vessels; the balance between substrate elimination and preservation of contractility is unresolved. (sources/pfa-vt-circ-2025, rating: high)
- Whether PFA replaces epicardial access or merely changes it is unresolved. High-voltage focal PFA abolished epicardial potentials endocardially in 9/9 patients, but pentaspline PFA failed to reach an epicardial circuit in arrhythmogenic cardiomyopathy, and epicardial PFA itself is only 20% transmural; the answer appears to be device- and depth-dependent rather than a property of the energy. (sources/pfa-vt-circ-2025, rating: high) (sources/pfa-vt-case-series-jacep-2023, rating: low) (sources/lattice-tip-preclinical-circep-2024, rating: medium)
- Coronary spasm after epicardial PFA has no agreed prophylaxis. Spasm occurred at a tip-to-artery distance of 7 ± 3 mm and not at 21 mm, but nitrates were given only reactively; whether prophylactic intracoronary or intravenous nitroglycerine should be routine, and what the long-term vascular consequence is, remain open. (sources/lattice-tip-vt-europace-2024, rating: medium)
Connections
Sources