meritmedscript.blogspot.com · Clinical exam notes · September 2026

One-line viva summary:

CMR LGE now rivals LVEF for NICM selection.

S-ICD is first-line when no pacing/ATP/CRT is needed (PRAETORIAN-XL).

EV-ICD (substernal Aurora) adds ATP without veins.

APPRAISE-ATP: a single burst delays first shock but does not cut shock burden and may increase VT storm.

Infected leads are extracted, never abandoned.

Part 1. Why the paradigm is shifting

The old rule that is being rewritten

Primary-prevention ICD allocation was built on MADIT-II / SCD-HeFT / DEFINITE: LVEF ≤35% after GDMT. That rule is under pressure because contemporary GDMT (ARNI, SGLT2i, MRA, quadruple therapy) has lowered SCD rates, transvenous hardware still produces endocarditis and extraction morbidity, and scar on CMR identifies arrhythmic risk better than pump function alone.

Three hardware families the examiner expects

Platform	Lead / can	What it can and cannot do
TV-ICD / CRT-D	Endovascular RV ± LV lead; pectoral can	ATP, bradycardia pacing, CRT, DFTs with 30–40 J. Pays the price of veins, tricuspid valve, extraction.
S-ICD (Emblem)	Parasternal subcutaneous coil + left mid-axillary can; 80 J	Shock only. No ATP, no bradycardia pacing, no CRT. Needs surface ECG screening (3 vectors).
EV-ICD (Aurora)	Substernal (anterior mediastinum) lead + mid-axillary can; 40 J. FDA 2023.	ATP + pause-prevention / post-shock pacing without a vein. Not CRT.

Original core concept (kept): Delivering defibrillation, ATP and backup bradycardia pacing without invading the vascular space or the heart itself — this is the EV-ICD, not the S-ICD.

Part 2. Selection work-up before the generator is chosen

Original investigations retained in full. Extra exam cut-offs added in the right-hand column.

Test	Original teaching (unchanged)	Exam extras / 2023–2026 numbers
CMR with LGE	Absolute gold standard for risk stratification before ICD, especially NICM. LGE identifies arrhythmogenic scar and supersedes LVEF alone for predicting SCD.	Mid-wall septal LGE in dilated NICM is the classic arrhythmic signature. Absence of LGE with LVEF <35% predicts pump-failure death more than SCD (DANISH-era teaching). CMR GUIDE (LVEF 36–50% + scar) missed its primary composite; SCD signal only — do not quote as a Class I change.
12-lead ECG	Mandatory. QRS duration and morphology. QRS >130 ms with LBBB redirects from a standard ICD to CRT-D or a conduction-system pacing defibrillator.	Also read: QTc (LQTS), type-1 Brugada, epsilon / TWI V1–V3 (ACM), fragmented QRS, and resting HR >80 (PRAETORIAN IAT predictor in TV-ICD). Screening ECG for S-ICD: ≥1 of 3 sensing vectors must pass supine and standing.
Ambulatory ECG	Documents NSVT burden — a major risk modifier in HCM and ARVC/ACM guidelines.	Also used after unexplained syncope and to quantify PVC burden before a primary-prevention decision in ACM and myocarditis.
Genetic testing	Class I when channelopathy (LQTS, Brugada) or high-risk cardiomyopathy genes (LMNA, FLNC, PLN, TMEM43, DSP) are suspected. Early prophylactic ICD may be indicated before the echo looks terrible.	Add RBM20, DES, TNNT2 in selected ACM/DCM panels. Pathogenic LMNA or FLNC plus NSVT or LVEF <45% is an ICD conversation even if LVEF is not ≤35%.

Device-choice matrix (viva table)

Clinical need	Preferred platform	Avoid / rethink
Primary or secondary prevention, no pacing, no planned ATP, no CRT	S-ICD first-line (PRAETORIAN-XL). Strongest in the young.	TV-ICD only if anatomy, screening failure, or future pacing is likely.
Monomorphic VT substrate that may be pace-terminable, but veins should be spared	EV-ICD (substernal). Only extra-vascular system with ATP.	S-ICD cannot deliver ATP. Do not pick S-ICD 'because it is extra-vascular' if ATP is the reason.
Bradycardia pacing, pause-prevention beyond brief post-shock pacing, or CRT	TV-ICD or CRT-D. CSP-D is the emerging alternative when LBBB + QRS >130 ms.	S-ICD. EV-ICD offers only limited substernal pacing — not a CRT substitute.
Prior endocarditis, chronic dialysis access, or extracted infected system	S-ICD or EV-ICD after cultures are negative ≥72 h.	Re-implanting a transvenous lead into a recently infected vein.

Part 3. Risk stratification that no longer stops at LVEF 35%

Secondary prevention — still Class I

Aborted SCA due to VT/VF, or haemodynamically unstable sustained VT, remains a Class I ICD indication across ESC 2022 VA/SCD and AHA/ACC/HRS documents. Platform choice then follows the matrix above.

Primary prevention — the contested zone

Scenario	Exam position in 2026
Ischaemic HFrEF, LVEF ≤35%, ≥40 days post-MI (or ≥90 days post-revascularisation), NYHA II–III on GDMT	Still the MADIT-II / SCD-HeFT backbone. PROFID EHRA (recruiting; NCT05665608) is testing whether OMT alone is non-inferior to OMT + ICD for all-cause death in post-MI LVEF ≤35%. Do not pre-empt the result.
NICM, LVEF ≤35% on GDMT	DANISH original: no all-cause mortality benefit. DANISH 13.2-year extension (JACC 2025): all-cause death HR 0.96 (0.82–1.13); sudden CV death HR 0.54 (0.36–0.80), benefit on SCD concentrated at age ≤70. Tissue-guided add-on: LGE-positive NICM is the group most likely to have an arrhythmic mode of death.
LVEF >35% but heavy LGE / high-risk gene / unexplained syncope	Do not use LVEF as a veto. Channelopathy and LMNA/FLNC/PLN phenotypes can justify an ICD with a near-normal ventricle.
LVEF <35% and zero LGE	Original teaching (kept): ICD survival benefit is statistically insignificant — mode of death is pump failure. LGE can still predict sudden death when LVEF >35%. 2024/2025 consensus language: tissue-guided ICD strategy in NICM.

High-risk modifiers the examiner will add after you quote LVEF

Unexplained syncope, especially exertional.
NSVT on Holter in HCM, ACM, LMNA, or myocarditis.
Pathogenic LMNA, FLNC, PLN, TMEM43, DSP.
Extensive mid-wall or ring-like LGE.
Age ≤70 in NICM when the question is SCD, not all-cause death (DANISH-13y).

Part 4. Management algorithm

Step	Action
1. Confirm the indication	Secondary prevention, or primary prevention after ≥3 months GDMT (post-MI wait as per guideline). Re-measure LVEF. Obtain CMR in NICM before listing. Screen for pacing/CRT need (QRS, LBBB, PR, sinus-node disease).
2. Choose the platform	No pacing + no ATP planned → S-ICD (PRAETORIAN-XL). ATP desired without veins → EV-ICD. Pacing or CRT needed → TV-ICD / CRT-D / CSP-D. Failed S-ICD screen → TV or EV.
3. Programme for fewer shocks	TV-ICD: MADIT-RIT logic — high rate cut-off (≥200 bpm) and long detection (≈6–12 s) in primary prevention. S-ICD: UNTOUCHED recipe — conditional zone 200–250 bpm, shock zone ≥250 bpm, SMART Pass ON. EV-ICD: ATP was nominally OFF at discharge in the pivotal study; operators turned it on over time as ATP success accumulated.
4. ATP policy (keep original + published addendum)	Original note (kept): APPRAISE-ATP asked whether empirical ATP for fast VT >200 bpm reduces first appropriate shocks versus shock-only; the source note concluded ATP did not significantly reduce first appropriate shocks or all-cause mortality and that ineffective ATP delayed defibrillation — hence 'simplified high-rate shock-only for primary prevention, reserve aggressive ATP for secondary prevention with known pace-terminable MMVT.' Published JAMA 2024 addendum (do not delete the line above): n=2595; single 8-beat burst at ~88% CL before shock delayed first all-cause shock (HR 0.72; 14.6% vs 19.4% at 5 years) but did not reduce total shock burden (12.3 vs 14.9 /100 pt-y) and increased VT/VF storm (HR 2.26). Mortality HR 1.15 (0.94–1.41), not significant. Viva answer: ATP can postpone the first shock; it is not a free lunch.
5. Drugs that stay first-line	Beta-blocker to the highest tolerated dose. Amiodarone or sotalol for recurrent VT / ICD shocks. Full HFrEF GDMT if LVEF is reduced. Quinidine discussion in Brugada with recurrent shocks. Do not use the ICD as a substitute for substrate ablation when VT is recurrent.
6. Complications — original protocol kept	Inappropriate shocks: AF with RVR or T-wave oversensing. Interrogate immediately. Extend detection to ~30 intervals or raise the primary-prevention cut-off (e.g. >220 bpm). Escalate rate-control. S-ICD: check SMART Pass and the sensing vector. Transvenous lead failure / endocarditis: pocket infection or Duke-positive lead endocarditis mandates complete hardware removal. Percutaneous mechanical or laser extraction in a high-volume centre. Never abandon an infected lead. Re-implant after blood cultures are negative 72 h, preferably S-ICD or EV-ICD. Electrical storm: ≥3 appropriate shocks in 24 h. Deep sedation (propofol ± intubation), IV propranolol, amiodarone load, urgent ablation transfer.

Part 5. Trial cards the examiner actually uses

Domain	Trial (years)	Original paradigm line + extra numbers
Non-transvenous ATP	EV-ICD Pivotal NEJM 2022; long-term Circulation; Enlighten 2026	Original: substernal Aurora lead + mid-axillary can met safety and defibrillation endpoints; 40 J output; painless ATP and post-shock pacing. Extra: implant DFT success 98.7%; 6-month major-complication-free 92.6%; long-term ATP success 77% (37/48); discrete spontaneous shock success 100%; 3-year IAS 17.5% driven by P-wave oversensing. Enlighten 1-year: appropriate therapy 7.3%, GEE ATP success 74.2%, IAS 7.1% after Smart Sense. FDA October 2023.
First-line ICD choice	PRAETORIAN NEJM 2020; PRAETORIAN-XL Circulation 2025	Original: 4–5 year extension — S-ICD non-inferior for SCD prevention; eliminated severe lead infection and high-risk extraction; slightly more minor pocket events. Extra: median 87.5 months; all complications 8.0% vs 11.6% (HR 0.73, NS); major complications 5.7% vs 10.2% (HR 0.58); lead-related 2.4% vs 8.3% (HR 0.33). IAS rates similar; S-ICD IAS = oversensing, TV-ICD IAS = SVT/AF. Conclusion language: consider S-ICD in every patient without a pacing indication.
Programming ATP	APPRAISE-ATP JAMA 2024	Original line kept: empirical ATP for fast VT was judged not to reduce first appropriate shocks or mortality; ineffective ATP can delay shock; reserve aggressive ATP for secondary-prevention MMVT. Extra published primary result: ATP-first delayed first all-cause shock (HR 0.72) without reducing shock burden and with more VT storm. Quote both layers.
NICM selection	DANISH + 13.2-y extension 2025; PROFID data stream	Original: LVEF <35% without LGE — ICD benefit insignificant; death is pump failure. Extra: DANISH-13y all-cause HR 0.96; SCD HR 0.54, age ≤70. PROFID EHRA is a post-MI LVEF ≤35% OMT ± ICD non-inferiority trial still recruiting — not a published outcome paper.
S-ICD programming	UNTOUCHED; SMART Pass	1-year IAS 3.1% with conditional 200 / shock 250 and Gen-3 devices. SMART Pass halves T-wave oversensing; it auto-off if R-wave is tiny — turn it back on.
Lead management	Extraction guidelines	Original: abandoning failed or infected leads is obsolete; complete percutaneous laser / mechanical extraction is required for infection.

Part 6. High-yield pitfalls

S-ICD vs EV-ICD. S-ICD is subcutaneous and shock-only. EV-ICD is substernal and can ATP. Calling every extra-vascular device an S-ICD loses marks.
LVEF 35% as a religion. NICM without LGE is not the same disease as ischaemic scar with LVEF 30%. Quote DANISH + tissue, not SCD-HeFT alone.
APPRAISE-ATP sound-bite. Do not say 'ATP is useless' or 'ATP is mandatory'. Say: one burst delays the first shock; burden and storm are the cost; secondary-prevention MMVT still deserves ATP.
Programming the S-ICD like a 2014 device. UNTOUCHED 200/250 plus SMART Pass is the modern default. Low-rate shock zones recreate the old IAS problem.
Abandoning an infected lead. Complete extraction. Re-implant extra-vascular after 72-hour negative cultures.
Treating electrical storm with another shock. Sympathetic blockade first (sedation + IV propranolol), then amiodarone, then ablation.
Skipping S-ICD screening. If no vector passes supine and standing, do not implant an S-ICD and hope.
PROFID as settled science. It is an ongoing strategy trial in post-MI HFrEF. CMR GUIDE missed its primary endpoint. Do not invent a Class I downgrade of the ischaemic primary-prevention ICD.
QRS >130 ms + LBBB and a 'simple' S-ICD. That patient may need CRT-D or CSP-D, not a shock-only can.

Part 7. Quick viva cards

Question	Model answer
Preferred ICD if no pacing need?	S-ICD. PRAETORIAN-XL: fewer major and lead complications at 8 years.
Only extra-vascular ICD that can ATP?	EV-ICD (Aurora). Substernal lead, 40 J, mid-axillary can. FDA 2023.
EV-ICD pivotal headline numbers?	DFT 98.7%; 6-month complication-free 92.6%; long-term ATP 77%; discrete shock 100%.
Why not LVEF alone in NICM?	DANISH: no all-cause benefit. Add LGE, age ≤70, gene, NSVT.
DANISH 13-year extension?	All-cause HR 0.96. Sudden CV death HR 0.54, mainly ≤70 years.
APPRAISE-ATP — one sentence?	One ATP burst delays first shock (HR 0.72) but does not cut shock burden and doubles VT storm.
S-ICD programming to quote?	UNTOUCHED: conditional 200, shock 250, SMART Pass on. IAS ~3% at 1 year.
Commonest IAS mechanisms?	TV-ICD: AF/SVT. S-ICD: T-wave / myopotential oversensing. EV-ICD pivotal: P-wave oversensing.
Infected ICD system?	Complete extraction. No abandoned leads. Re-implant S-ICD/EV-ICD after 72 h negative cultures.
Electrical storm definition and first drug?	≥3 appropriate shocks / 24 h. IV propranolol plus deep sedation; then amiodarone; then ablation.
When is a gene enough?	LMNA, FLNC, PLN, TMEM43, DSP, and channelopathies — ICD may precede a 35% LVEF.
QRS >130 ms + LBBB?	Do not default to S-ICD. Think CRT-D or conduction-system pacing defibrillator.
What is PROFID EHRA?	Ongoing post-MI LVEF ≤35% trial of OMT ± ICD. Not yet an outcome paper.
S-ICD screening rule?	At least one of three vectors must pass supine and standing before implant.

Part 8. Sources

ESC 2022 ventricular arrhythmia and sudden-death guidelines; AHA/ACC/HRS ICD / VA documents; Friedman et al., EV-ICD Pivotal, N Engl J Med 2022 and long-term Circulation follow-up; Enlighten registry, Circulation 2026; Knops / Olde Nordkamp, PRAETORIAN NEJM 2020 and PRAETORIAN-XL Circulation 2025; Schuger et al., APPRAISE-ATP, JAMA 2024; Køber / DANISH original and JACC 2025 13.2-year extension; PROFID EHRA design, Am Heart J 2026 (NCT05665608); UNTOUCHED; MADIT-RIT programming standard.