ICD and CRT indications in heart failure

Overview and definitions

ESC 2026 groups ICDs and CRT under guideline-directed interventional therapies (GDIT) for HF.^[1] ESC 2026 states that ICDs are effective in preventing sudden cardiac death (SCD) and prolonging survival in patients with HFrEF who are at high risk of SCD.^[1]

Device decisions turn on two prevention settings. Secondary prevention means an ICD after a ventricular arrhythmia causing haemodynamic instability, which is how the ESC 2026 secondary-prevention row is labelled.^[1] For primary prevention, ESC 2026 bases patient selection on LVEF thresholds established in RCTs and on the underlying aetiology.^[1]

New York Heart Association functional classification (ESC 2026 Table 8)

^[1]

NYHA class matters because the rows are split by it. The ESC 2026 primary-prevention ICD rows cover NYHA class II/III.^[1] The ESC 2026 ICD row for NYHA class IV with severe symptoms refractory to pharmacological therapy is a Class III row unless the patient is a candidate for CRT, a ventricular assist device or heart transplantation.^[1]

Classification of devices

^{[1] [3] [4] [6]}

ESC 2021 explains HBP as a more physiological, simultaneous activation of the ventricles via the His–Purkinje system than right ventricular (RV) pacing.^[2] With LBBAP, ESC 2021 describes a lead implanted slightly distal to the His bundle and screwed deep in the LV septum, ideally to capture the left bundle branch.^[2]

^{[1] [3] [6]}

Epidemiology

^[1]

ESC 2026 notes that the risk of SCD has declined over time but calls the residual risk still high.^[1] It gives this as the reason for primary-prevention ICD implantation in patients with HFrEF deemed at high risk of SCD.^[1] In a systematic review of 26 studies (6 prospective) on nearly 58,000 patients, HRS 2023 reports a pooled prevalence of pacing-induced cardiomyopathy (PICM) of 12%, using 15 unique definitions from 23 publications.^[6]

Pathophysiology

On sudden arrhythmic death, ESC 2026 states that antiarrhythmic drugs such as amiodarone and dronedarone reduce ventricular arrhythmias but do not prolong survival and may shorten it.^[1] ESC 2026 adds that ICDs, unlike antiarrhythmic drugs such as amiodarone, reduce mortality in cardiac arrest survivors and in those with sustained ventricular arrhythmias, regardless of LVEF or presence of HF.^[1]

ESC 2026 defines HF associated with abnormal cardiac conduction as LV dysfunction due to abnormal electrical conduction.^[1] ESC 2026 states that abnormal conduction as the sole cause of HFrEF is most likely with LBBB, wide QRS duration and no myocardial scarring on CMR, and in RV pacing-induced HF.^[1]

ESC 2026 states that the effect of ICD treatment depends on the competing risk of non-sudden death, both cardiovascular (CV) and non-CV.^[1] Thus, in patients with many comorbidities and older age, the benefit of ICD therapy may be substantially lower (ESC 2026).^[1]

On electrical dyssynchrony, HRS 2023 describes how, during RV apical pacing and LBBB, early-activated regions contract early and late-activating LV segments contract late.^[6] HRS 2023 states that, as a result, a proportion of patients with long-term RV pacing or LBBB may develop dyssynchrony-induced cardiomyopathy and HF.^[6]

HRS 2023 says the limited role of physiologic pacing with non-LBBB and shorter QRS is most likely due to the fact that prolonged LBBB usually reflects delay within the conduction system.^[6] In prolonged LBBB, activation is usually latest in the posterolateral LV, which is more amenable to correction with CPP (HRS 2023).^[6] Shorter non-LBBB conduction abnormalities, by contrast, reflect intrinsic myocardial disease or variable sites of delayed LV activation, which are less amenable to correction (HRS 2023).^[6]

^{[1] [6]}

Clinical presentation: who should prompt a device assessment

• A survivor of a ventricular arrhythmia causing haemodynamic instability: the ESC 2026 secondary-prevention row applies if survival >1 year with good functional status is expected, there is no reversible cause and the arrhythmia did not occur <48 h after an MI.^[1]
• Symptomatic HFrEF (NYHA class II/III) with LVEF ≤35% despite ≥3 months of optimal FMT and expected survival longer than 1 year with good functional status: the ESC 2026 primary-prevention ICD rows (the ischaemic row also excludes an MI in the prior 40 days).^[1]
• LV dysfunction with QRS duration ≥130 ms and no evidence of other causes for HF: ESC 2026 says HF associated with abnormal cardiac conduction should be suspected.^[1]
• A pacemaker or ICD delivering RV pacing: ESC 2026 says HF associated with abnormal conduction should also be considered.^[1]
• A cardiac implantable electronic device (CIED) with de novo or worsening HFrEF: ESC 2026 Recommendation Table 4 says review of the proportion of RV pacing should be considered in all such patients (Class IIa, Level C).^[1]
• A patient admitted with ACS: ESC 2023 ACS recommends routine echocardiography during hospitalization to assess regional and global LV function, detect mechanical complications and exclude LV thrombus (Class I, Level C).^[9]
• Before discharge after ACS: the ESC 2023 ACS text recommends that the LVEF is determined before hospital discharge in all patients with ACS.^[9]
• ESC 2022 had given an earlier row for acute MI: early (before discharge) assessment of LVEF is recommended in all patients with acute MI (Class I, Level B).^[3]

Differential diagnosis: what changes the device decision

A low LVEF opens the device question, and the situations in the table can change or defer the answer. 2022 AHA/ACC/HFSA states that GDMT is optimized before ICD and CRT implantation to assess whether the LVEF improves.^[5]

Situations that change or defer a device decision

^{[1] [9] [5] [3] [4]}

Clinical and bedside assessment

For the QRS, ESC 2021 cautions that ECG criteria of intraventricular conduction disturbance, LBBB and non-LBBB have not been consistently defined and reported in any of the past CRT studies.^[2] ESC 2021 strongly discourages selecting CRT patients solely on cardiac imaging data when the QRS is narrow (<130 ms), after the Echo-CRT trial suggested possible harm in that setting.^[2]

For aetiology, ESC 2026 notes that CMR may be useful in identifying myocardial scar and morphological abnormalities pointing to other aetiologies of HF.^[1] ESC 2026 also states that the presence and extent of myocardial fibrosis, genotypes and additional factors may help arrhythmic risk stratification in patients with cardiomyopathies.^[1]

Selected assessment rows (ESC 2026 HF and ESC 2023 cardiomyopathy)

^{[1] [4]}

ESC 2022 and ESC 2023 cardiomyopathy both recommend an ICD only in patients who have an expectation of good quality survival >1 year (Class I, Level C).^[3][4] 2022 AHA/ACC/HFSA states that ICD and CRT-D are not indicated when comorbidities or frailty limit survival with good functional capacity to <1 year (COR 3: No Benefit, LOE C-LD).^[5] ESC 2026 calls shared decision-making essential when considering an ICD in HF, with individualized information on the potential risks and benefits to support an informed decision.^[1]

Management: after a life-threatening ventricular arrhythmia

ESC 2026 refers secondary-prevention ICD recommendations to the 2022 ESC ventricular arrhythmia guideline, and cardiomyopathy-specific ones to the 2023 ESC cardiomyopathy guideline, while printing its own secondary-prevention row for HF.^[1]

Secondary-prevention rows (in patients with HF, the later ESC 2026 row applies; the ESC 2022 and ESC 2023 rows give the further detail it refers to)

^{[1] [3] [4]}

In patients with HF, the later ESC 2026 secondary-prevention row is the ESC row that applies (Class I, Level A).^[1] It applies with its own conditions: expected survival >1 year with good functional status, no reversible cause, and no ventricular arrhythmia <48 h after an MI.^[1] ESC 2026 refers to the ESC 2022 and ESC 2023 guidelines for further secondary-prevention and cardiomyopathy-specific recommendations, so their rows are given beside it.^[1] The ESC 2022 secondary-prevention table points to its section 7 for primary prevention and specific aspects of secondary prevention.^[3] Adults with a secondary-prevention ICD indication who are temporarily not candidates for ICD implantation are covered by the wearable defibrillator rows below, which include the later ESC 2026 HF WCD row for patients with HF.^[3][1]

Management: primary-prevention ICD

ESC 2026 bases selection on LVEF thresholds established in RCTs and on aetiology, because individuals with non-ischaemic HF may derive less benefit than those with ischaemic heart disease.^[1] ESC 2026 adds that a meta-analysis of RCTs demonstrated an overall survival benefit, even in patients with non-ischaemic HF.^[1]

ESC 2026 HF, Recommendation Table 6: primary-prevention rows

^[1]

Coronary artery disease outside the ESC 2026 rows

ESC 2026 Recommendation Table 6 has no row for NYHA class I or for the inducible-VT pathway, so among the guidelines checked for this topic the ESC 2022 chronic coronary artery disease (CAD) rows still apply there.^[1][3] The NSVT group still overlaps the ESC 2026 ischaemic row in part.^[1][3] With CAD, NSVT and symptomatic HFrEF (NYHA class II/III) with LVEF ≤35% despite ≥3 months of optimal FMT, the later ESC 2026 ischaemic row applies (Class I, Level B1).^[1][3] That row recommends an ICD to reduce the risk of sudden death and all-cause death.^[1] It excludes an MI in the prior 40 days and applies provided survival longer than 1 year with good functional status is expected.^[1] Both ESC 2022 rows are given below with their conditions.

ESC 2022 chronic CAD rows still current for these groups

^{[3] [1]}

ESC 2022 also had a CAD row recommending ICD therapy with symptomatic HF (NYHA class II–III) and LVEF ≤35% despite ≥3 months of OMT (Class I, Level A).^[3] The ESC 2022 row was replaced for these patients by the ESC 2026 ischaemic row (Class I, Level B1), which adds the 40-day and survival conditions.^[1][3] The 2023 ESC ACS guideline had given a similar Class I, Level A row, which is likewise dated by ESC 2026.^[9] The 2023 ESC ACS row had recommended ICD therapy to reduce sudden cardiac death in symptomatic HF (NYHA Class II–III) with LVEF ≤35%.^[9] Its conditions were LVEF ≤35% despite optimal medical therapy for >3 months and at least 6 weeks after MI.^[9] That dated row applied to patients expected to survive for at least 1 year with good functional status.^[9]

Dilated and non-dilated cardiomyopathy

The ESC 2023 cardiomyopathy rows split patients by LVEF and genotype.^[4] For symptomatic HFrEF (NYHA class II/III) with LVEF ≤35% despite ≥3 months of optimal FMT and expected survival longer than 1 year with good functional status, the ESC 2026 non-ischaemic row applies.^[1] The ESC 2023 DCM and NDLVC rows for LVEF ≤35% despite >3 months of OMT (Class IIa, Level A) carry no NYHA class or survival condition of their own.^[4] ESC 2023 still recommends an ICD in patients with cardiomyopathy only when there is an expectation of good quality survival >1 year (Recommendation Table 12, Class I, Level C).^[4] The later ESC 2026 HF guideline adds a Class III, Level C row for NYHA class IV.^[1] It does not recommend an ICD in NYHA class IV with severe symptoms refractory to pharmacological therapy unless the patient is a candidate for CRT, a ventricular assist device or heart transplantation.^[1]

ESC 2023 cardiomyopathy primary-prevention rows (Recommendation Tables 24 and 26)

^{[4] [1]}

The footnotes of Recommendation Tables 24 and 26 state that additional risk factors include syncope and late gadolinium enhancement (LGE) presence on CMR.^[4] ESC 2023 defines the NDLVC phenotype by non-ischaemic LV scarring or fatty replacement in the absence of LV dilatation, with or without global or regional wall motion abnormalities.^[4] The phenotype also covers isolated global LV hypokinesia without scarring (as assessed by LGE on CMR) that is unexplained solely by abnormal loading conditions (hypertension, valve disease) or CAD.^[4]

North American rows

2022 AHA/ACC/HFSA ICD rows

^{[5] [8]}

2022 AHA/ACC/HFSA adds a value statement: a transvenous ICD provides high economic value in primary prevention of SCD.^[5] The statement applies particularly when the risk of death caused by ventricular arrhythmia is deemed high and the risk of nonarrhythmic death (either cardiac or noncardiac) is deemed low based on comorbidities and functional status.^[5] The 2022 AHA/ACC/HFSA stage B table had carried the same NYHA class I post-MI row (COR 1, LOE B-R); for post-MI patients the newer 2025 ACC/AHA ACS row now applies.^[5][8]

For patients after MI, the 2025 ACC/AHA ACS guideline is newer.^[8] The 2025 ACC/AHA ACS guideline recommends an ICD in selected patients with LVEF ≤40% (its Table 17) at least 40 days post MI and at least 90 days postrevascularization to reduce death (COR 1, LOE A).^[8] The guideline marks this row as adapted from the 2017 AHA/ACC/HRS ventricular arrhythmia guideline.^[8]

2025 ACC/AHA ACS Table 17: LVEF and patient characteristics defining an indication for prophylactic ICD implantation for primary prevention in ischemic heart disease

^[8]

For post-MI patients, these 2025 rows are newer than the earlier 2022 AHA/ACC/HFSA post-MI ICD rows and add, as a row condition, at least 90 days postrevascularization.^[8][5] The 2022 AHA/ACC/HFSA row remains the North American row for nonischemic DCM.^[5]

^{[1] [5] [8]}

Timing: the 40-day, 3-month and 90-day rules

ESC 2026 states that two studies showed primary-prevention ICD implantation within 40 days of an MI does not improve outcomes.^[1] ESC 2022 states that early routine prophylactic ICD implantation in the first 40 days after MI did not reduce mortality in post-MI patients with reduced LVEF in two randomized trials, DINAMIT and IRIS.^[3]

^{[9] [3] [1] [8]}

The 2022 AHA/ACC/HFSA supportive text puts the three intervals together: re-evaluation of EF >40 days after MI, >90 days after revascularization and >90 days after GDMT is useful to determine candidacy for ICD or CRT.^[5]

Waiting has a cost. ESC 2026 states that delaying ICD implantation carries a risk of SCD of 2%–4% by 6 months and ∼7% by 1 year.^[1]

^{[9] [1] [8] [5]}

Management: cardiac resynchronization therapy

ESC 2026 states that, in carefully selected patients with HFrEF, CRT improves cardiac function and quality of life (QoL), and reduces hospitalizations and death.^[1] In SR, the ESC 2026 CRT rows are split by QRS duration and QRS morphology.^[1]

ESC 2026 HF, Recommendation Table 7 (all eight rows)

^[1]

Two ESC 2026 points sit outside the table. The ESC 2026 Figure 10 legend lists what to weigh in any device decision: patient preference, likely benefit of ICD/CRT at the patient’s age, frailty and comorbidities, risk of complications and venous access issues.^[1] The ESC 2026 timing text gives no class or level.^[1] It says it may be beneficial to implant CRT early in those most likely to benefit (LVEF ≤35%, LBBB with QRS ≥150 ms), ideally before a HF hospitalization has occurred.^[1]

How the ESC rows changed from 2021

The 2021 ESC pacing guideline gave the sinus-rhythm rows with Level B for the intermediate groups; ESC 2026 now gives the same groups Level C.^[2][1] ESC 2026 lists the HF and AV block row among its revised recommendations.^[1] The 2021 version (CRT rather than RV pacing recommended for HFrEF, including patients with AF; Class I, Level A) has become a Class IIa, Level B1 row.^[1] The 2021 rows are dated history and are given so that older sources can be read correctly.

ESC CRT grades: 2021 pacing guideline (dated history) and 2026 HF guideline

^{[2] [1]}

ESC 2021 explained the morphology split with MADIT-CRT, REVERSE and RAFT.^[2] These trials suggest that there is likely to be potential benefit in all patients with LBBB regardless of QRS duration, whereas any benefit with non-LBBB is evident mostly with a QRS duration ≥150 ms (ESC 2021).^[2] ESC 2021 adds that, in the MADIT-CRT long-term study and RAFT, the benefit with QRS <150 ms appeared later during follow-up.^[2]

North American CRT rows

2022 AHA/ACC/HFSA CRT rows

^[5]

2022 AHA/ACC/HFSA also gives CRT a value statement: with LVEF ≤35%, SR, LBBB with QRS ≥150 ms and NYHA class II, III or ambulatory IV symptoms on GDMT, CRT implantation provides high economic value.^[5] The 2022 AHA/ACC/HFSA supportive text cautions that subgroup analyses of the device trials were not their primary endpoints and should be interpreted with caution.^[5]

^{[1] [5]}

HRS 2023 cardiac physiologic pacing rows

The 2023 HRS/Asia Pacific Heart Rhythm Society/Latin American Heart Rhythm Society (HRS/APHRS/LAHRS) guideline has a section headed “Indications for CPP in patients with HF”.^[6] That section covers patients with HF (NYHA class I–IV) who have no a priori indication for pacing due to bradycardia.^[6]

Selected HRS 2023 rows on CRT with BiV pacing and CPP (the CSP rows are given in the conduction system pacing section)

^[6]

Management: CRT-P or CRT-D

ESC 2026 cites the trial evidence for both devices.^[1] In COMPANION, CRT-P and CRT-D reduced the composite of all-cause death and hospitalizations for any cause by 20%, and a reduction in all-cause death was shown with CRT-D but not with CRT-P.^[1] In CARE-HF, which randomized participants to medical therapy with or without CRT-P, CRT-P reduced all-cause death and HFH.^[1]

Choosing between CRT-P and CRT-D

^{[2] [3] [4] [5]}

ESC 2021 concluded that prospective randomized trials are lacking and that the data are insufficient to firmly prove a superiority of CRT-D over CRT-P.^[2] ESC 2021 added that CRT trials in mild HF almost exclusively included patients with an ICD, and that survival benefit of CRT without an ICD is uncertain in this group.^[2] It also reported that observational data point towards significant survival benefits by CRT-D over CRT-P in ischaemic cardiomyopathy, while no clear benefit has been shown in non-ischaemic cardiomyopathy.^[2]

ESC 2021 text (no class or level given) says the addition of an ICD to CRT should be considered especially in younger patients with a good survival prognosis, ischaemic aetiology, and a favourable comorbidity profile or presence of myocardial fibrosis.^[2] ESC 2021 reports that the DANISH trial assigned 1116 patients with HF and non-ischaemic cardiomyopathy to a primary prophylactic ICD or usual clinical care alone, and that 58% in both groups also had CRT.^[2] In its subgroup analysis, CRT-D was not superior to CRT-P for the primary outcome of all-cause mortality after a median follow-up of 67.6 months.^[2] ESC 2021 asks that the choice between CRT-P and CRT-D be guided by shared decision-making that takes into account both medical facts and patient values.^[2] The dated NHFA/CSANZ 2018 text says it would be reasonable to consider CRT-P over CRT-D in other circumstances, to improve quality of life and longevity.^[7] Its examples are patients who do not wish to have the potential for defibrillation, those whose left ventricle is likely to improve, the very elderly, and those who retain a poorer prognosis but remain symptomatic.^[7]

Management: pacing need, high RV pacing and upgrade

A patient who needs ventricular pacing, or who is already being paced from the RV, is at risk of pacing-induced HF.^[1][6] ESC 2026 reports that in the BUDAPEST-CRT Upgrade trial, the first RCT, which included only 360 patients, upgrading to CRT-D reduced death and HFH in patients with ≥20% RV pacing burden and QRS ≥150 ms.^[1] ESC 2026 also lists RV pacing burden and intrinsic QRS duration as an important focus at follow-up when deciding whether to upgrade from an ICD to a CRT-D device.^[1]

Rows for patients who need or already have ventricular pacing

^{[1] [5] [6]}

ESC 2021, now dated, defined its upgrade threshold in a footnote: a limit of 20% RV pacing is supported by observational data, but no data support any percentage as a true limit below which RV pacing is safe.^[2] HRS 2023 notes that most PICM studies used a decline in LVEF of ≥10% from a baseline above 50%, pacing ≥20%, and no alternative explanation, while stating there is no single definition.^[6]

Management: CRT in atrial fibrillation

ESC 2021 states that a high rate of biventricular pacing is not reached in two-thirds of patients with persistent or permanent AF, and that in most AF patients with intact AV conduction adequate biventricular pacing can be achieved only by AV junction (AVJ) ablation.^[2]

ESC 2021 rows for CRT in persistent or permanent AF (2021 grades and definitions: HFrEF <40%, HFmrEF 40–49%, HFpEF ≥50%; the later ESC 2026 row on AV node ablation combined with CRT applies in its own population, set out below)

^{[2] [1]}

ESC 2021 rows 2A–2D set the device in symptomatic AF with an uncontrolled heart rate when the patient is a candidate for AVJ ablation.^[2] ESC 2026 HF has a later row on AV node ablation combined with CRT in its AF table.^[1] AV node ablation combined with CRT should be considered in severely symptomatic permanent AF with poor rate control despite medical therapy and at least one HFH, to reduce symptoms, physical limitations, recurrent HFH and death (Class IIa, Level B1).^[1] ESC 2026 states that AV node ablation combined with CRT was found superior to pharmacological therapy in severely symptomatic permanent AF and HF, regardless of LVEF.^[1] The 2024 ESC AF guideline gives the same question Class IIa, Level B, in severely symptomatic patients with permanent AF and at least one HF hospitalization, without the poor-rate-control condition that ESC 2026 adds.^[13][1] For CRT in HFrEF and AF more widely, the 2024 ESC AF guideline refers to the 2021 ESC pacing guideline.^[13] It puts an important focus on ensuring effective biventricular pacing, with a low threshold for considering atrioventricular node ablation.^[13] ESC 2026 lists the 2021 HF and AV block row among its revised recommendations: the 2021 row recommended CRT rather than RV pacing for HFrEF with an indication for ventricular pacing for high-degree AV block, including patients with AF (Class I, Level A).^[1] The 2026 version covers HFrEF regardless of NYHA class or QRS width with an indication for ventricular pacing for high-degree AV block.^[1] It says CRT, rather than RV pacing, should be considered in order to reduce the risk of HFH and death (Class IIa, Level B1).^[1]

2023 ACC/AHA/ACCP/HRS AF guideline rows on AF with HF

^[10]

2022 AHA/ACC/HFSA adds that, in AF with LVEF ≤35% on GDMT, CRT can be useful to reduce total mortality, improve symptoms and QOL, and increase LVEF (COR 2a, LOE B-NR).^[5] That row applies if the patient requires ventricular pacing or otherwise meets CRT criteria and AV nodal ablation or pharmacological rate control will allow near 100% ventricular pacing with CRT.^[5] ESC 2026 lists the percentage of effective biventricular pacing (without fusion or pseudofusion) as an important focus of follow-up in patients with AF.^[1]

Management: conduction system pacing

For CSP in HFrEF, the 2026 ESC HF guideline is the later ESC position.^[1] The 2023 HRS guideline grades CSP in its own rows, and the 2023 ACC/AHA/ACCP/HRS AF guideline grades it for AF with HFrEF undergoing AV nodal ablation (COR 2b, LOE C-LD).^[6][10] ESC 2026 states that left bundle branch area pacing may prove a feasible alternative to CRT when implantation is challenging, that no RCT with patient-centred outcomes has evaluated CSP efficacy or long-term safety in HFrEF, and that no recommendations can currently be made.^[1] The ESC 2021 His bundle pacing rows below are earlier rows.^[2] Where they apply to HFrEF, now LVEF <50%, the later ESC 2026 statement that no recommendation can currently be made is the ESC position.^[2][1]

Conduction system pacing rows (all ESC 2021 His bundle pacing rows, dated, with the later ESC 2026 position where they apply to HFrEF; selected HRS 2023 rows)

^{[2] [6] [1]}

ESC 2021 said that, especially in CRT candidates with LBBB, biventricular pacing has more solid evidence of efficacy and safety and therefore remains first-line therapy.^[2] ESC 2021 stated that recommendations for LBBAP could not be formulated at that stage.^[2] ESC 2021 reports mid-term HBP lead revision of ∼7%, higher than the 2–3% for RV pacing, and 9% generator changes at 5 years with HBP against 1% with RV pacing.^[2]

Management: subcutaneous and wearable defibrillators

Subcutaneous ICD and wearable defibrillator rows (for patients with HF, the later ESC 2026 HF WCD row applies beside the ESC 2022 and ESC 2023 WCD rows)

^{[1] [3] [4] [9] [8] [12]}

ESC 2022 had said the WCD may be considered in the early phase after MI in selected patients (Class IIb, Level B).^[3] The later 2023 ESC ACS row on ICD or WCD use in selected patients within 40 days of MI (Class IIb, Level C) is given in the table above.^[9] For a primary-prevention ICD, the later ESC 2026 HF row is a Class III row (not recommended within 40 days of an MI).^[1] For patients with HF, the later ESC 2026 HF WCD row in the same table also applies, with its own conditions.^[1]

The trial behind the caution is VEST.^[3] ESC 2022 reports that VEST randomized 2302 patients with acute MI and LVEF ≤35% in a 2:1 fashion to a WCD or not, and found no difference at 90 days in the primary endpoint of arrhythmic death (1.6 vs. 2.4%).^[3] ESC 2022 notes concern about a low median wear time of 18 h, and its task force does not recommend routine WCD use in the early post-MI phase.^[3]

ESC 2026 text adds that no mortality benefit of the wearable defibrillator has been shown in RCTs.^[1] On the S-ICD, ESC 2026 states that it is comparable to transvenous ICDs in efficacy, complication rates and inappropriate shock rates, and is particularly attractive with difficult venous access or infection risk.^[1]

Follow-up, programming and generator replacement

• ESC 2026: routine defibrillation testing of shock efficacy does not reduce arrhythmic death and is therefore no longer recommended.^[1]
• ESC 2026: programming a high-rate cut-off and long detection intervals dramatically reduces inappropriate shocks and unnecessary ICD therapy.^[1]
• ESC 2026: in patients with a CRT device, regular follow-up and optimization should be implemented, as CRT programming and FMT may require further optimization after implantation.^[1]
• ESC 2026: implanting an ICD commits the patient to lifelong monitoring and generator replacements, and the lifelong risk of complications should be considered in the decision to implant.^[1]
• ESC 2026: before generator replacement for CIEDs, a shared decision-making process between the patient and a cardiologist experienced in CIEDs and HF should be considered to ensure that the balance between risks and benefits still favours a CIED (Class IIa, Level C).^[1]

ESC 2026 reports that ICD shocks occur in more than 15% of patients after generator replacement in those who had no appropriate shock during the lifetime of their first generator.^[1] ESC 2026 adds that a considerable residual risk of SCD remains even after LVEF improvement with FMT and CRT.^[1] ESC 2026 asks that the need for an ICD be reconsidered over time and before every generator replacement, weighing SCD risk, remaining lifespan, complications, competing risks of death, frailty and patient choice.^[1]

Specific scenarios

^{[1] [9] [8] [4] [5] [3]}

Complications and pitfalls

• Inappropriate shocks: 2018 NHFA/CSANZ balances the benefits of an ICD against complications of implantation and psychological sequelae of inappropriate shocks for atrial tachyarrhythmias or lead fracture.^[7]
• Upgrade procedures: ESC 2021 reports complication rates of 6.8% to 20.9% in several studies, higher than de novo implantation, which a later registry analysis did not confirm.^[2]
• S-ICD limits: it cannot treat bradyarrhythmia (other than post-shock pacing) and does not provide anti-tachycardia pacing (ESC 2026).^[1]
• HBP leads: higher mid-term revision and earlier generator changes than RV pacing (ESC 2021).^[2]
• Selecting CRT on imaging alone with a narrow QRS: strongly discouraged after Echo-CRT (ESC 2021).^[2]
• Assuming CRT works in AF without checking capture: ESC 2021 notes that a high rate of biventricular pacing is not reached in two-thirds of patients with persistent or permanent AF.^[2]

Applying a dated grade is an exam pitfall. ESC 2021 had recommended CRT rather than RV pacing for HFrEF (<40%) regardless of NYHA class with an indication for ventricular pacing and high-degree AV block, in order to reduce morbidity (Class I, Level A).^[2] ESC 2026 now says CRT should be considered in that setting for HFrEF, now defined as LVEF <50%, regardless of NYHA class or QRS width, to reduce the risk of HFH and death (Class IIa, Level B1).^[1]

Prognosis and disposition

Landmark device trials as reported in the held guidelines

^{[1] [5] [8] [3]}

The 2026 ESC cardiac rehabilitation guideline recommends cardiac rehabilitation (CR) after ICD or CRT device implantation to improve physical functioning (VO₂ peak, 6-minute walk distance) (Class I, Level B1).^[11] The ESC 2026 rehabilitation exercise-training table recommends exercise training, as part of CR, in ICD recipients to improve physical functioning and reduce the incidence of shocks, and in CRT recipients to improve physical functioning and health-related QoL (both Class I, Level B1).^[11]

Special populations

• Older patients and multimorbidity: ESC 2026 states the benefit of an ICD may be substantially lower in patients with many comorbidities and older age, because of the competing risk of non-sudden death.^[1]
• Women: ESC 2021 reports that a US Food and Drug Administration meta-analysis of patient-level data found the main difference in patients with LBBB and QRS 130–149 ms, where women had a 76% reduction in HF or death.^[2]
• Women (HRS 2023): HRS recommends CRT with BiV pacing in patients with select characteristics (eg, female sex) with LVEF ≤35%, SR, LBBB with QRS 120–149 ms and NYHA class II-IV on GDMT, to reduce mortality and HF events and to improve LVEF (COR 1, LOE A), and says women appear to derive more benefit from CRT across QRS durations.^[6]
• Pregnancy: the 2025 ESC pregnancy guideline states that an ICD is indicated in pregnant women at high risk of SCD and that implantation can be performed safely, particularly after 8 weeks of gestation; it states that a WCD may be considered when pregnant women have an ICD indication from reversible conditions.^[12]
• Peripartum cardiomyopathy: ESC 2022 text notes higher thresholds for early ICD implantation because of a high rate of spontaneous recovery after delivery, and the 2025 ESC pregnancy guideline says a WCD may be considered with PPCM and LVEF <35% (Class IIb, Level C).^[3][12]
• Limited life expectancy: HRS 2023 states that, with a life expectancy of <1 year, the decision to implant a CPP device should incorporate shared decision-making, taking into account the potential improvement in QoL balanced against the risk of procedural complications (COR 1, LOE C-EO).^[6]

Evidence, guidelines and regional differences

^{[1] [5] [6]}

ESC 2026 HF supplies the ICD and CRT recommendation tables for HF (Recommendation Tables 6 and 7), and refers secondary-prevention ICD recommendations to ESC 2022 and cardiomyopathy-specific ones to ESC 2023.^[1] The ESC guidelines checked for this topic are listed in the evidence pack with the date of the check. Five ESC guidelines in the register were not held with their recommendation tables as text and were not checked: 2022 pulmonary hypertension, 2022 cardio-oncology, 2023 diabetes, 2025 myocarditis and pericarditis, and 2026 chronic kidney disease. Where a row is called current on this page, that claim refers to the checked guidelines only.

Australia and New Zealand: NHFA/CSANZ 2018 (dated)

The National Heart Foundation of Australia and Cardiac Society of Australia and New Zealand (NHFA/CSANZ) 2018 guideline is the Australian heart failure guideline listed in the register. It predates every other guideline on this page, so its rows are dated and are given for local context. The dated NHFA/CSANZ 2018 device rows carry a GRADE strength of recommendation and a quality of evidence.^[7]

NHFA/CSANZ 2018 device rows (section 9.1; dated)

^[7]

NHFA/CSANZ 2018 practice advice adds that measures are required to ensure at least 92% biventricular capture when CRT is performed in AF, and that CRT is not beneficial, and may be harmful, with QRS duration of less than 130 ms.^[7] NHFA/CSANZ 2018 practice advice also says that subcutaneous ICDs may be considered in younger people for primary prevention, but do not provide antitachycardia pacing or bradyarrhythmia pacing support.^[7] NHFA/CSANZ reports that a DANISH substudy with the prespecified endpoint of age found a significant reduction in mortality in patients under 70 years.^[7]

Exam pearls

• ESC 2026 primary-prevention ICD rows need all of: NYHA class II/III, LVEF ≤35% despite ≥3 months of optimal FMT, and expected survival longer than 1 year with good functional status; the ischaemic row also excludes an MI in the prior 40 days.^[1]
• Ischaemic is Class I, Level B1; non-ischaemic is Class IIa, Level B1 (ESC 2026).^[1]
• The 90-day rule is North American: at least 90 days postrevascularization in the 2025 ACC/AHA ACS row for selected post-MI patients with LVEF ≤40%, and >90 days after revascularization or GDMT in the 2022 AHA/ACC/HFSA supportive text.^[8][5]
• In ESC 2026 Recommendation Table 7, the only Class I row is CRT in symptomatic HFrEF with LVEF ≤35% despite optimal FMT, in SR with LBBB and QRS ≥150 ms, in order to improve symptoms and reduce the risk of hospitalizations and death (Class I, Level A).^[1]
• ESC 2026 HFrEF is LVEF <50%, so its row favouring CRT over RV pacing to reduce the risk of HFH and death, with an indication for ventricular pacing for high-degree AV block regardless of NYHA class or QRS width (Class IIa, Level B1), is not limited to LVEF ≤35%.^[1]
• A WCD has no mortality benefit shown in RCTs (ESC 2026), and ESC 2026 grades it Class IIb, Level C for HF patients at risk of SCD for a limited period, as a bridge to decision for a permanent ICD or while listed for heart transplantation.^[1]

References13ShowHide

1. [1]Køber L, et al. 2026 ESC Guidelines for the management of heart failure. Eur Heart J, 2026.PMID 42661420
2. [2]Glikson M, et al. 2021 ESC Guidelines on cardiac pacing and cardiac resynchronization therapy. Eur Heart J, 2021.PMID 34455430
3. [3]Zeppenfeld K, et al. 2022 ESC Guidelines for the management of patients with ventricular arrhythmias and the prevention of sudden cardiac death. Eur Heart J, 2022.PMID 36017572
4. [4]Arbelo E, et al. 2023 ESC Guidelines for the management of cardiomyopathies. Eur Heart J, 2023.PMID 37622657
5. [5]Heidenreich PA, et al. 2022 AHA/ACC/HFSA Guideline for the Management of Heart Failure: A Report of the American College of Cardiology/American Heart Association Joint Committee on Clinical Practice Guidelines. Circulation, 2022.PMID 35363499
6. [6]Chung MK, et al. 2023 HRS/APHRS/LAHRS guideline on cardiac physiologic pacing for the avoidance and mitigation of heart failure. Heart Rhythm, 2023.PMID 37283271
7. [7]Atherton JJ, et al. National Heart Foundation of Australia and Cardiac Society of Australia and New Zealand: Guidelines for the Prevention, Detection, and Management of Heart Failure in Australia 2018. Heart Lung Circ, 2018.PMID 30077227
8. [8]Rao SV, et al. 2025 ACC/AHA/ACEP/NAEMSP/SCAI Guideline for the Management of Patients With Acute Coronary Syndromes: A Report of the American College of Cardiology/American Heart Association Joint Committee on Clinical Practice Guidelines. J Am Coll Cardiol, 2025.PMID 40013746
9. [9]Byrne RA, et al. 2023 ESC Guidelines for the management of acute coronary syndromes. Eur Heart J, 2023.PMID 37622654
10. [10]Joglar JA, et al. 2023 ACC/AHA/ACCP/HRS Guideline for the Diagnosis and Management of Atrial Fibrillation: A Report of the American College of Cardiology/American Heart Association Joint Committee on Clinical Practice Guidelines. Circulation, 2024.PMID 38033089
11. [11]Bäck M, et al. 2026 ESC Guidelines on cardiac rehabilitation. Eur Heart J, 2026.PMID 42661418
12. [12]De Backer J, et al. 2025 ESC Guidelines for the management of cardiovascular disease and pregnancy. Eur Heart J, 2025.PMID 40878294
13. [13]Van Gelder IC, et al. 2024 ESC Guidelines for the management of atrial fibrillation developed in collaboration with the European Association for Cardio-Thoracic Surgery (EACTS). Eur Heart J, 2024.PMID 39210723