Electrophysiological Characteristics of Septal Perforation During Left Bundle Branch Pacing

Authors, Journal, Affiliations, Type, DOI

• Authors: Shunmuga Sundaram Ponnusamy (corresponding); William Basil; Pugazhendhi Vijayaraman
• Journal: Heart Rhythm 2022;19(5):728–734. © 2022 Heart Rhythm Society
• Affiliations: Velammal Medical College, Madurai, India; Medtronic India Private Limited, Madurai; Geisinger Heart Institute, Wilkes-Barre
• Type: Retrospective single-centre observational study — 219 consecutive attempts, 212 successful (96.8%); 30 septal perforations (14.1%); mean follow-up 9.9 ± 6.7 months
• DOI: 10.1016/j.hrthm.2022.01.018
• Disclosures: Ponnusamy and Basil — consultants to Medtronic; Vijayaraman — speaker/consultant/research support from Medtronic, consultant to Abbott, Biotronik and Boston Scientific, and a patent for an HBP delivery tool

Overview

Septal perforation is the complication every LBBP operator fears, but its electrophysiological signature had not been characterised. This study analyses 30 perforations among 212 successful implants (14.1%) and describes what the lead sees as it enters the LV cavity: unipolar impedance falls to ~405 Ω, the current of injury collapses from ~15 mV to ~1 mV, and the unfiltered unipolar electrogram changes shape in one of two ways. A QS pattern (type I, 67%) indicates complete perforation, with the helix free in the cavity and capture only at high output. An RS/rS pattern (type II, 33%) indicates partial perforation with part of the helix still in contact — which still captures in 80%, but at a high threshold. A unipolar impedance cut-off of <450 Ω diagnosed perforation with sensitivity 100%, specificity 96.6%, PPV 83.3% and NPV 100%. All 30 leads were withdrawn and reimplanted at a different site, with no VSD, septal haematoma, coronary injury, embolism, or dislodgement during follow-up. The paper's central argument is that perforation recognised at the table and managed by re-siting the lead is benign — but its 14.1% rate is 4–40 times higher than previously reported, and the diagnosis is made by the same electrical parameters being evaluated.

Keywords

Current of injury; Left bundle branch pacing; Pacing impedance; Septal perforation; Template beat; Unfiltered unipolar electrogram

Key Takeaways

Cohort and incidence

• 219 attempted, 212 successful (96.8%); 7 without confirmed capture excluded
• Septal perforation in 30 (14.1%) — the Discussion says 13.9%, a minor internal inconsistency
• Patients were asymptomatic during perforation
• Indications: AV block 14, CRT alternative 13, sinus node dysfunction 2, AV-junction ablation 1
• Age 60.1 ± 11.2 vs 62.3 ± 11.3 years (P = .32), septal thickness 10.1 ± 1.6 vs 10.5 ± 1.4 mm (P = .15), LVEF 44.1 ± 16.2 vs 47.8 ± 15.7% (P = .23) — no baseline differences from the non-perforation group
• Baseline QRS 152.1 ± 31.6 ms; post-LBBP 116.2 ± 11.1 ms; RWPT 71.1 ± 11.2 ms
• Fluoroscopy 16.3 ± 4.7 min

The impedance signature

• During perforation: 404.6 ± 19.9 Ω (range 360–440); 400–450 Ω in 16 (53.3%), <400 Ω in 14 (46.7%)
• Without perforation: 612.1 ± 113.2 Ω; 450–500 Ω in 18 (9.9%), 400–450 Ω in 5 (2.7%), <400 Ω in 1 (0.5%)

• <450 Ω is the proposed optimum: it misses no perforation and has a 100% NPV, at the cost of a PPV of only 83.3% (6 non-perforated patients also fell below 450 Ω)
• The Discussion reports specificity 96.4% and PPV 82.3% for the same cut-off, slightly different from the Results/Table 2 (96.6%/83.3%)

The current-of-injury signature

• COI amplitude fell from 15.4 ± 11.6 mV just before perforation to 0.9 ± 0.6 mV after — a near-complete loss
• Fifteen patients (50%) still had capture during perforation, but at a threshold of 3.02 ± 0.7 V/0.5 ms, with V6 RWPT 75.9 ± 17.3 ms
• The recovery manoeuvre: gently withdraw the lead while watching the unfiltered unipolar EGM — COI reappears as the helix re-enters the septum, confirming that the loss was positional rather than tissue death
• Reimplantation at a new site: impedance 606.8 ± 101.1 Ω, threshold 0.4 ± 0.1 V, RWPT 71 ± 12.4 ms (vs 75.9 ± 17.3 during perforation, P = .38)

Two unfiltered unipolar electrogram patterns

• The mechanism proposed: a unipolar electrogram records a negative deflection when the wavefront moves away from the electrode. A lead free in the cavity sees a steep, slurred QS; a lead with partial contact first sees the wavefront approaching the proximal helix, producing an initial R before the negative deflection
• Type II captured significantly more often than type I (80% vs 35%, P = .02), consistent with residual septal contact
• The filtered (30–300 Hz) unipolar EGM did not change usefully at perforation — the diagnostic information is in the unfiltered (0.5–500 Hz) channel

Management and outcome

• All 30 patients were repositioned at a new site rather than simply withdrawn, to reduce the risk of subsequent dislodgement
• Confirmation of a genuinely new site: fluoroscopic landmark; pacemapping on the right septum showing a different inferior-lead QRS; template/fixation beats during rapid deployment; LAO lead movement; and absence of template beats with hypertransmission of rotations suggesting the old site
• Peak troponin 188 ± 162 pg/mL
• No coronary artery injury, septal haematoma, or systemic embolism; post-procedure echo showed no lead perforation or VSD
• Follow-up echo: LVEF 44.1 → 49.6% and LVEDD 54.5 → 51.2 mm (both P > .15, not significant)
• No dislodgement in the perforation group during 9.9 ± 6.7 months; 2 dislodgements (1.09%) occurred in the non-perforation group

Limitations of the document

Authors' own stated limitations

• Retrospective, single-centre study using intracardiac electrograms and pacing parameters to identify perforation
• Intracardiac echocardiography was not used to demonstrate perforation during implantation — "a major limitation"
• Electroanatomic mapping during native rhythm was not performed, so the "RS" partial-perforation hypothesis requires further validation

Additional limitations evident in the study

• ⚠ The 14.1% incidence is an outlier by an order of magnitude. Published intraprocedural perforation rates cited in the same paper are 0.3–3.2%; the authors attribute their figure to deliberate subendocardial deployment for selective capture, a thin septum (10.1 mm), and ethnicity — all plausible but unverified. If the rate is real, it describes a specific technique and population; if the diagnostic criteria are over-sensitive, the "rate" is partly a definition artefact
• The diagnosis is circular. Perforation was defined by impedance <500 Ω, threshold rise and COI loss — the very parameters then evaluated as diagnostic tests. There was no imaging gold standard (no ICE, no contrast confirmation in this study). The reported sensitivity of 100% is therefore guaranteed by the definition, and the specificity is the only informative number
• The cut-offs were derived and tested in the same 30 events. No external validation, no confidence intervals, and the PPV of 83.3% means roughly one in six <450 Ω readings was not a perforation
• The two EGM patterns are a hypothesis, not an observed anatomy. Type II ("partial perforation") is inferred from the initial R wave; without ICE or electroanatomic mapping, it could equally represent a lead in a septal cleft, a different tissue plane, or far-field recording
• The cohort is not a typical bradycardia cohort: 13/30 (43%) were CRT alternatives and mean LVEF was 44%, so the technique (deep subendocardial deployment) and the perforation risk may not generalise
• The "benign" conclusion rests on 30 patients followed ~10 months, with no systematic imaging for VSD or thrombus, no CT, and no comparison group. The absence of events in a small, short follow-up cannot establish safety
• The counterintuitive dislodgement finding — none in the perforation group, 2 in the non-perforation group — is reported without comment, and the groups are not comparable in size or selection
• The troponin rise (188 pg/mL) is modest and not related to perforation type or outcome, so it does not help adjudicate severity
• The management recommendation (re-site rather than withdraw) is based on reasoning about dislodgement risk, not on a comparison of the two strategies
• The paper excludes 7 patients without confirmed capture, so the denominator for "attempted" perforation is the successful-implant cohort, which may bias the rate downward
• Rating rationale: rated medium. It is the largest dedicated perforation series and provides the practical electrical signature — <450 Ω, COI collapse, and the QS versus RS distinction — that the wiki now uses. It is not high because it is retrospective, single-centre, without an imaging gold standard, its diagnostic criteria are circular, its 14.1% rate is a striking outlier, its pattern classification is hypothesis-generating, and its safety conclusion rests on a small short-follow-up cohort

Key Concepts Mentioned

• concepts/Left-Bundle-Branch-Area-Pacing — the technique; this paper characterises its principal mechanical complication
• concepts/Conduction-System-Pacing — the umbrella category
• Septal perforation — complete (helix in LV cavity) or partial (part of helix in septal contact)
• Unipolar pacing impedance — falls to ~405 Ω at perforation; <450 Ω is the proposed diagnostic cut-off
• Current of injury (COI) — collapses from ~15 mV to ~1 mV at perforation and returns on withdrawal
• Unfiltered unipolar electrogram — 0.5–500 Hz recording that shows QS (complete) or RS/rS (partial) patterns
• Filtered unipolar electrogram — 30–300 Hz; does not show a useful change at perforation
• Template/fixation beats — used both to confirm a new implantation site and to detect returning to the old one
• Deep septal deployment — the deliberate technique that may raise perforation risk
• Partial perforation — residual contact allowing high-threshold capture

Key Entities Mentioned

• concepts/Left-Bundle-Branch-Area-Pacing — primary concept page
• Velammal Medical College, Madurai, India — the investigating centre
• Shunmuga Sundaram Ponnusamy; Pugazhendhi Vijayaraman — the investigators
• Medtronic — 3830 lead and C315 sheath; disclosed relationships of all three authors
• Abbott Workmate Claris — the EP recording system
• Jastrzębski et al. — the cadaver simulation describing lead behaviours and the LV endocardial barrier
• Heart Rhythm Society / Elsevier — publisher

Wiki Pages Updated

• Created: wiki/sources/septal-perforation-hrs-2022.md
• Updated: wiki/concepts/Left-Bundle-Branch-Area-Pacing.md
• Updated: wiki/concepts/Conduction-System-Pacing.md
• Updated: wiki/sourceindex.md
• Updated: wiki/wikiindex.md