Heart Failure — Types & Treatment Evidence

A clinician's reference to heart-failure phenotypes and the evidence behind current treatment, spanning adults and pediatrics, with links to the landmark trials.

Heart failure is a clinical syndrome, not a single disease, and its treatment is now among the most evidence-dense areas of medicine. This page organizes the phenotypes and the trial evidence behind current therapy across the age spectrum. Every landmark study links to its primary source (PubMed permalink or DOI). Recommendation classes reflect the 2022 AHA/ACC/HFSA and 2021/2023 ESC guidelines plus developments through mid-2026; always confirm against the primary source for an individual patient.

Use the tabs below to move between topics. The Landmark Trials tab collects every cited study in one place. This is an educational reference, not a substitute for clinical judgment.

Overview & ClassificationAdult · HFrEFAdult · HFpEF / HFmrEFAcute & ShockPediatric HFLandmark TrialsGuidelines & Tools

Definition & classification

Heart failure (HF) is a clinical syndrome — symptoms and/or signs produced by structural and/or functional cardiac dysfunction, corroborated by objective evidence (elevated natriuretic peptides and/or congestion). Two consensus efforts frame contemporary practice: the 2021 Universal Definition and the newly published 2026 Second Universal Definition, alongside the region-specific 2022 AHA/ACC/HFSA and 2021 ESC guidelines.

Universal Definition

The 2021 Universal Definition (Bozkurt et al., HFSA/HFA-ESC/JHFS) requires both (a) symptoms/signs and (b) at least one objective criterion (PMID 33662581). Natriuretic-peptide thresholds used as corroboration:

Setting	BNP	NT-proBNP
Ambulatory / outpatient	≥ 35 pg/mL	≥ 125 pg/mL
Hospitalized / acute	≥ 100 pg/mL	≥ 300 pg/mL

Thresholds run higher in atrial fibrillation and lower in obesity; a normal NP does not exclude HFpEF.

2026 update — Second Universal Definition of Heart Failure (AHA/ACC/ESC/WHF with HFSA/HFA/JHFS; published 29 June 2026, simultaneously in Circulation, JACC, European Heart Journal, and open-access Global Heart: AHA summary · JACC · Global Heart, open access). It reaffirms the A→D stage continuum and centers Stage B (pre-HF) as the critical intervention window, moves away from rigid EF cutoffs toward clinically actionable phenotypes, adds a trajectory framework (improvement / remission / recovery) and a standardized etiologic taxonomy, and redefines decompensation by need for therapeutic escalation rather than acuity.

EF-based phenotypes

The 2022 AHA/ACC/HFSA and 2021 ESC guidelines share the same LVEF bands:

Phenotype	LVEF	Notes
HFrEF	≤ 40%	HF with reduced EF
HFmrEF	41–49%	HF with mildly reduced EF (ESC renamed “mid-range” → “mildly reduced”); often a waypoint toward HFrEF or recovery
HFpEF	≥ 50%	HF with preserved EF (+ elevated filling pressures / diastolic dysfunction)
HFimpEF	baseline ≤ 40%, ≥ 10-point rise, follow-up > 40%	HF with improved EF

HFimpEF is a formal category only in the 2022 AHA/ACC/HFSA guideline (absent from 2021 ESC). “Improved” ≠ “recovered”: GDMT should continue, since withdrawal frequently precipitates relapse (the TRED-HF principle). The 2026 definition drops fixed cutoffs — all three vocabularies remain in active use.

ACC/AHA stages A–D

Stages describe disease progression and are unidirectional (a patient never regresses to an earlier stage):

Stage A — At risk: risk factors (HTN, diabetes, obesity, CAD, cardiotoxins, family history) without structural disease or symptoms. → primary prevention.
Stage B — Pre-HF: structural/functional abnormality or elevated NP/troponin without symptoms (reduced EF, LVH, valve disease, prior MI). → the “critical window” for early therapy.
Stage C — Symptomatic HF: structural disease with current or prior symptoms. → full GDMT.
Stage D — Advanced HF: refractory symptoms despite optimal therapy. → transplant, mechanical support, inotropes, or palliative care.

NYHA functional class

Captures current symptom burden and can move in either direction:

Class	Limitation
I	No limitation; ordinary activity is asymptomatic
II	Slight limitation; symptoms with ordinary activity
III	Marked limitation; symptoms with less-than-ordinary activity
IV	Symptoms at rest or with any activity

Stage (trajectory) and NYHA class (severity) are complementary — a Stage C patient may be NYHA II today and NYHA IV when decompensated.

Temporal / anatomic / physiologic descriptors

Acute vs chronic · de novo vs decompensated (the 2026 definition reframes decompensation as need for escalated/rescue therapy).
Left (pulmonary congestion) vs right (systemic congestion — edema, JVD, ascites; usually caused by left HF or pulmonary hypertension/RV infarct) vs biventricular.
Backward (congestion behind the failing chamber) vs forward (low output — fatigue, hypotension, renal hypoperfusion) failure.
High-output failure: HF despite elevated output. Modern leading causes: obesity, liver disease, and arteriovenous shunts (incl. dialysis fistulae); classic causes include severe anemia, thyrotoxicosis, thiamine deficiency (beriberi), sepsis, and Paget disease (JACC series).

Major etiologies

Ischemic (most common in high-income settings) · hypertensive · valvular.
Cardiomyopathies: dilated (idiopathic/genetic/toxic), hypertrophic (commonest inherited), restrictive, and infiltrative (amyloidosis, sarcoidosis, hemochromatosis).
Myocarditis (viral/autoimmune/checkpoint-inhibitor) · tachycardia-induced (reversible with rate/rhythm control) · peripartum (often partially reversible) · toxic/chemotherapy (anthracyclines, trastuzumab, alcohol) · genetic (titin, lamin, sarcomeric).

Diagnostic workup

Natriuretic peptides — high negative predictive value; adjust up for AF, down for obesity. A normal NP does not exclude HFpEF.
Transthoracic echocardiography — the cornerstone: EF (phenotype), chamber size, wall thickness, diastolic function, valves, filling pressures.
HFpEF scoring when EF ≥ 50% and diagnosis is uncertain: H₂FPEF (validation, PMID 33647474) is more sensitive; HFA-PEFF (Pieske, EHJ 2019) is more specific. Use them to structure the workup, not to replace hemodynamic judgment.
Adjuncts: ECG, chest radiograph, labs (renal/electrolytes, iron studies, TSH, HbA1c); cardiac MRI for tissue characterization (myocarditis, infiltration, scar); ischemia evaluation where suspected. Screen for cardiac amyloidosis in unexplained HFpEF (LVH, low-voltage ECG, older adults) — it now has disease-modifying therapy (see the Landmark Trials tab).

Adult chronic HFrEF — guideline-directed medical therapy

For adults with chronic HF and LVEF ≤ 40%, four foundational drug classes — the “four pillars” — reduce mortality and hospitalization on top of loop diuretics used for congestion. The 2022 AHA/ACC/HFSA guideline makes all four Class 1, and the 2024 ACC Expert Consensus Decision Pathway emphasizes rapid initiation of all four rather than sequential titration of one at a time.

Pillar	Example agents (target dose)	Key trials	Effect
ARNI (preferred RASi)	Sacubitril/valsartan (97/103 mg BID)	PARADIGM-HF	↓ CV death or HF hosp 20%; ↓ all-cause death 16%
ACEi/ARB (if ARNI not feasible)	Enalapril, lisinopril, candesartan	SOLVD, CHARM	↓ mortality vs placebo (foundational evidence)
Beta-blocker	Carvedilol (25 mg BID), metoprolol succinate (200 mg), bisoprolol (10 mg)	MERIT-HF, CIBIS-II, COPERNICUS	↓ all-cause mortality ~34%
MRA	Spironolactone (25–50 mg), eplerenone (50 mg)	RALES, EMPHASIS-HF	↓ mortality 30% (RALES); ↓ CV death/HF hosp 37%
SGLT2 inhibitor	Dapagliflozin (10 mg), empagliflozin (10 mg)	DAPA-HF, EMPEROR-Reduced	↓ CV death/worsening HF ~25%

The four pillars

1. ARNI. In PARADIGM-HF (n = 8,442), sacubitril/valsartan beat enalapril for CV death or HF hospitalization (HR 0.80) and all-cause death (HR 0.84), stopping early. ARNI is the preferred renin-angiotensin inhibitor (Class 1); an ACEi (or ARB if ACEi-intolerant) is used when ARNI is not feasible. Switching from an ACEi requires a 36-hour washout to avoid angioedema. In-hospital initiation is supported by PIONEER-HF, which showed a greater NT-proBNP fall than enalapril in stabilized ADHF.

2. Beta-blockers. Only three agents have HFrEF mortality evidence and should not be treated as a class effect: carvedilol, metoprolol succinate, bisoprolol. MERIT-HF cut all-cause mortality 34%; CIBIS-II reduced mortality (11.8% vs 17.3%); COPERNICUS extended benefit to severe/NYHA IV disease (35% RRR; 11.2% vs 16.8%). Start only when euvolemic and up-titrate gradually.

3. MRA. RALES (spironolactone, NYHA III–IV) reduced mortality 30%; EMPHASIS-HF (eplerenone, NYHA II) reduced CV death or HF hospitalization (HR 0.63). Class 1 when eGFR > 30 and K⁺ < 5.0; monitor potassium and renal function. A potassium binder (patiromer or sodium zirconium cyclosilicate) can enable RAASi/MRA tolerance in patients prone to hyperkalemia.

4. SGLT2 inhibitors. DAPA-HF (HR 0.74) and EMPEROR-Reduced (HR 0.75) reduced worsening HF or CV death independent of diabetes. Class 1; the 2023 ESC focused update later extended SGLT2i across the full EF spectrum.

Rapid, simultaneous initiation

Historical practice titrated one drug to target before adding the next, taking months. Because each pillar’s benefit is largely independent and additive, modern guidance starts all four quickly (days to weeks) at low doses, then up-titrates. STRONG-HF (n = 1,078, post-acute HF) randomized patients to rapid up-titration to full GDMT within ~2 weeks with close follow-up versus usual care, and cut 180-day death or HF readmission (15.2% vs 23.3%; absolute risk reduction 8.1%). Cross-trial modeling (Vaduganathan et al., Lancet 2020) estimates comprehensive four-drug therapy lowers mortality by roughly 73% versus no therapy and adds years of event-free survival in younger patients.

Additional / second-line agents

Ivabradine (SHIFT): sinus rhythm, HR ≥ 70 bpm on maximal beta-blocker; reduced HF hospitalization/CV death. Class 2a.
Hydralazine + isosorbide dinitrate (A-HeFT): in self-identified Black patients with NYHA III–IV on GDMT, reduced the rate of death by 43% (10.2% → 6.2%). Class 1 for this group.
Vericiguat (VICTORIA): after a recent worsening-HF event, HR 0.90; Class 2b. The 2025 VICTOR trial in stable ambulatory HFrEF was neutral for its primary endpoint (CV death HR 0.83), refining vericiguat’s niche toward recently-decompensated patients.
Digoxin (DIG): neutral on mortality, reduces HF hospitalization; Class 2b.
IV iron for iron deficiency: AFFIRM-AHF reduced total HF hospitalizations (RR 0.74) but the primary composite narrowly missed significance (RR 0.79; P = 0.059); IRONMAN trended similarly (RR 0.82; P = 0.070). Class 2a for symptoms/functional status.
Omecamtiv mecarbil (GALACTIC-HF): small composite reduction (HR 0.92) but no mortality benefit; not approved, limited role.
Loop diuretics: Class 1 for congestion; no mortality claim.

Device therapy & structural interventions

ICD (primary prevention, Class 1): LVEF ≤ 35%, NYHA II–III on ≥ 3 months GDMT, survival > 1 year. SCD-HeFT and MADIT-II. Defer if EF may recover.
CRT (Class 1): LVEF ≤ 35%, sinus rhythm, LBBB with QRS ≥ 150 ms, NYHA II–IV on GDMT — COMPANION, CARE-HF, MADIT-CRT. Weaker with non-LBBB or QRS 120–149 ms.
Transcatheter mitral repair (TEER): for persistent NYHA II–IV symptoms with severe secondary MR despite GDMT, COAPT showed fewer HF hospitalizations and lower mortality (Class 2a).
AF catheter ablation: in HFrEF with AF, CASTLE-AF reduced death or HF hospitalization (HR 0.62) versus medical rate/rhythm control.

Advanced heart failure

Refer to an advanced-HF center for the “I-NEED-HELP” triggers (repeated hospitalizations, escalating diuretics, GDMT intolerance from hypotension/renal decline, inotrope dependence, end-organ dysfunction). Durable LVAD — the magnetically levitated HeartMate 3 in MOMENTUM 3 was superior to HeartMate II (77% vs 65% survival free of disabling stroke/reoperation) and is now standard. Transplantation remains the definitive therapy for eligible patients.

Adult HFpEF & HFmrEF

HF with preserved (HFpEF, EF ≥ 50%) and mildly reduced (HFmrEF, EF 41–49%) ejection fraction were, for two decades, a graveyard of neutral trials. That changed with the SGLT2 inhibitors, and the evidence base has expanded rapidly through 2024–2025.

1. SGLT2 inhibitors — the first robust benefit

EMPEROR-Preserved (empagliflozin, ~5,988, EF > 40%): CV death or HF hospitalization ↓ 21% (HR 0.79), driven by fewer hospitalizations; attenuated but significant at EF ≥ 50% (PMID 34449189).
DELIVER (dapagliflozin, 6,263, EF > 40%): worsening HF or CV death ↓ 18% (HR 0.82), consistent above and below EF 60% (PMID 36027570).

The 2023 ESC focused update made SGLT2 inhibitors Class I (LOE A) across HFmrEF and HFpEF; the 2022 AHA/ACC guideline (predating DELIVER) rated them Class 2a. These are now first-line disease-modifying therapy across the EF spectrum.

2. MRA — from TOPCAT to finerenone

TOPCAT (spironolactone, 3,445): neutral overall (HR 0.89); a post-hoc regional analysis showed benefit in the Americas (HR ~0.82) but not Russia/Georgia, where enrollment/adherence signals were problematic — hypothesis-generating, not definitive (PMID 24716680).
FINEARTS-HF (finerenone, nonsteroidal MRA, 6,001, EF ≥ 40%): total worsening-HF events + CV death ↓ 16% (RR 0.84; P = 0.007); CV death alone not significantly reduced; more hyperkalemia (PMID 39225278).

On the strength of FINEARTS-HF, the FDA approved finerenone (Kerendia) for HF with LVEF ≥ 40% on 14 July 2025 — the first nonsteroidal MRA across HFmrEF/HFpEF. This post-dates the 2022/2023 US/ESC guidelines, so formal class assignments are pending there; the 2025 Canadian (CCS/CHFS) update and 2025 Japanese guideline (Class IIa) already incorporate it. Steroidal spironolactone retains only weak support (2022 AHA/ACC Class 2b).

3. Incretin therapy in obese HFpEF

STEP-HFpEF (semaglutide, 529, EF ≥ 45%, BMI ≥ 30, no diabetes): large KCCQ and weight benefit (PMID 37622681).
STEP-HFpEF DM (with type 2 diabetes, 616): replicated the symptom/weight benefit (PMID 38587233).
SUMMIT (tirzepatide, 731, EF ≥ 50%, BMI ≥ 30): first incretin trial powered for events — CV death or worsening HF ↓ (9.9% vs 15.3%; HR 0.62), driven by worsening-HF events, with large KCCQ gains (PMID 39555826).

4. ARNI / ARB — mostly neutral

PARAGON-HF (sacubitril/valsartan, 4,796 of 4,822 randomized, EF ≥ 45%): narrowly missed (rate ratio 0.87; 95% CI 0.75–1.01; P = 0.059); signals in women and lower-EF patients (PMID 31475794).
PARAGLIDE-HF (after a worsening-HF event, EF > 40%): greater NT-proBNP reduction with ARNI; pooled with PARAGON supports a possible role at the lower-preserved EF range (PMID 37212758).
CHARM-Preserved (PMID 13678871) and I-PRESERVE (PMID 19001508): neutral. ARNI/ARB carry only weak (Class 2b) symptom/hospitalization support.

5. HFmrEF (EF 41–49%)

No dedicated HFmrEF outcome trial exists; evidence is from subgroups and pooled analyses showing HFrEF therapies retain benefit that attenuates as EF rises. SGLT2 inhibitors have direct RCT support (Class I ESC 2023 / 2a AHA 2022) and finerenone now adds Class I-level evidence spanning HFmrEF. For ARNI/ACEi/ARB, beta-blockers, and steroidal MRA, the 2022 AHA/ACC guideline gives Class 2b (“may be considered”).

6. Comorbidities & congestion

Loop diuretics to euvolemia (no mortality claim); aggressive blood-pressure control (a core HFpEF driver); atrial fibrillation rate/rhythm management, anticoagulation, and selective ablation; weight/cardiometabolic management (now with a pharmacologic tool in the obese phenotype). Always screen for specific etiologies — cardiac amyloidosis, ischemia — before defaulting to a generic HFpEF label.

Bottom line: SGLT2 inhibitors are foundational across HFmrEF/HFpEF. Finerenone is now approved for LVEF ≥ 40%. Incretins are a phenotype-specific option for obese HFpEF. ARNI/ARB and steroidal MRA remain weak options. Diuretics plus management of hypertension, AF, obesity, and specific etiologies complete the approach.

Acute / decompensated HF & cardiogenic shock

Clinical profiles

Bedside classification of acute decompensated HF (ADHF) uses two axes — congestion (“wet” vs “dry”) and perfusion (“warm” vs “cold”). The Stevenson profiles (warm-dry A, warm-wet B, cold-wet C, cold-dry L) can be assigned in a two-minute exam and guide initial therapy (Nohria/Stevenson, JACC 2003); the older Forrester quadrants map the same logic invasively. Most admissions are warm-and-wet (congestion over low output). Common precipitants: ischemia, tachy-/brady-arrhythmia (esp. AF), dietary/medication nonadherence, uncontrolled hypertension, infection, renal dysfunction, and NSAID or negative-inotrope exposure.

Decongestion

IV loop diuretics are the cornerstone. DOSE (n = 308) found no difference between bolus and continuous infusion; high-dose (2.5× oral) gave greater fluid loss and dyspnea relief with a transient, non-clinically-important creatinine rise. Practically: dose adequately (≈ 2–2.5× the home oral dose IV), reassess natriuresis early, and escalate for resistance.

For sequential nephron blockade, ADVOR added IV acetazolamide and improved decongestion (42.2% vs 30.5%; RR 1.46); CLOROTIC added oral hydrochlorothiazide (more weight loss, more transient renal impairment). SGLT2 inhibitors should be started in-hospital once stable — EMPULSE showed net clinical benefit (win ratio 1.36). On discharge diuretic choice, TRANSFORM-HF found no difference between torsemide and furosemide for mortality. None of these were powered for mortality; benefits are decongestion/symptom/event endpoints.

Vasodilators

IV nitroglycerin (preload reduction; useful in hypertensive/flash pulmonary edema) and nitroprusside (balanced, for severe afterload excess) relieve symptoms but no vasodilator has shown a mortality benefit in ADHF: nesiritide (ASCEND-HF — neutral, dyspnea effect did not meet its prespecified significance threshold, NEJM 2011), ularitide (TRUE-AHF — no CV-mortality benefit, NEJM 2017), and an early intensive vasodilation strategy (GALACTIC — neutral, JAMA 2019). Vasodilators are symptom tools, not disease-modifiers.

Inotropes

Dobutamine (β-agonist) and milrinone (PDE-3 inhibitor) are reserved for cold (low-output) profiles to bridge to recovery, decongestion, or advanced therapies. DOREMI found no difference between milrinone and dobutamine in cardiogenic shock (49% vs 54%; RR 0.90). Milrinone is renally cleared and a more potent vasodilator; dobutamine is blunted by chronic beta-blockade. Avoid inotropes in warm profiles; use the lowest effective dose for the shortest time — chronic exposure increases mortality.

Cardiogenic shock

Grade severity with the SCAI SHOCK stages (A at-risk → E extremis) plus a cardiac-arrest modifier (SCAI 2022). The 2025 ACC Concise Clinical Guidance on cardiogenic shock is the current US expert-consensus document on evaluation and MCS device selection (JACC 2025, PMID 40100174). In AMI shock, emergent culprit-lesion revascularization is the one intervention with a clear survival benefit (culprit-only PCI per CULPRIT-SHOCK). Norepinephrine is first-line: in SOAP II, dopamine caused more arrhythmias and overall mortality did not differ; a prespecified cardiogenic-shock subgroup suggested higher mortality with dopamine (interaction test not significant — hypothesis-generating).

Mechanical circulatory support (MCS) evidence is heterogeneous:

IABP — neutral in IABP-SHOCK II (no 30-day or long-term benefit), prompting a guideline downgrade.
Routine early VA-ECMO — neutral and harmful in ECLS-SHOCK (more bleeding/limb ischemia).
Microaxial flow pump (Impella CP) — DanGer Shock, the first positive MCS RCT, reduced 180-day death in STEMI shock (45.8% vs 58.5%; HR 0.74; NNT ~8) at the cost of more complications (bleeding, limb ischemia, renal replacement).

The lesson is selective, not routine early MCS: DanGer’s benefit was confined to a carefully selected STEMI population, whereas routine ECMO/IABP failed. Device choice, timing, and patient selection remain the active controversy.

Before discharge

Initiate and up-titrate the four GDMT pillars before discharge in HFrEF. STRONG-HF tested rapid pre-discharge optimization to full dose within ~2 weeks and cut 180-day death or HF readmission (15.2% vs 23.3%). The vulnerable early post-discharge phase is where rapid optimization pays off. (Full specifications for every trial above are in the Landmark Trials tab.)

Pediatric heart failure

Pediatric HF is a clinical syndrome with a strikingly thin evidence base. Almost every drug is prescribed by extrapolation from adult trials; the handful of dedicated pediatric RCTs have been neutral or negative. Two therapies (ivabradine, sacubitril/valsartan) reached FDA pediatric labeling largely on surrogate or non-inferiority grounds rather than demonstrated mortality benefit. Treat the recommendations below as low-certainty and physiology-driven.

Etiology & dominant physiology

Two physiologies dominate and demand opposite management:

CHD with a large systemic-to-pulmonary shunt (VSD, AVSD, PDA, truncus, post-stage-I single ventricle) → volume-overload “over-circulation,” pulmonary edema, and failure to thrive despite preserved contractility; presents in the first weeks–months as pulmonary vascular resistance falls.
Ventricular dysfunction / cardiomyopathy → impaired contractility (dilated) or filling (restrictive/hypertrophic); this is the physiology adult HF drugs were designed for.

By age: neonate/infant — ductal-dependent and shunt lesions, critical obstruction (coarctation, aortic stenosis), ALCAPA, metabolic/mitochondrial disease, SVT-induced cardiomyopathy; older child/adolescent — dilated cardiomyopathy (~50% idiopathic), myocarditis, HCM/RCM, anthracycline cardiotoxicity, neuromuscular/genetic (Duchenne, sarcomeric), and the failing single ventricle/systemic RV. Framework: AHA statements on cardiomyopathy classification (2019) and treatment (2023).

Classification systems

Modified Ross classification — the infant analogue of NYHA, grading feeding volume/time, diaphoresis, tachypnea, growth, and hepatomegaly (I asymptomatic → IV symptoms at rest). Original: Ross, grading CHF in infants (Pediatr Cardiol 1992, PMID 1614922); NT-proBNP-incorporating revision (Pediatr Cardiol 2012, PMID 22476605).
NYHA — used unchanged in older children able to report exertional limitation.
ISHLT pediatric classification and ACC/AHA-style Stages A–D adapted for pediatrics; no single system is validated as superior for outcome prediction.

Treatment evidence — what is and isn’t proven

Therapy	Key pediatric evidence	Result	Bottom line
Diuretics (furosemide)	No RCT	—	Symptomatic mainstay; zero mortality data — expert consensus.
ACE inhibitor (enalapril)	Infant single-ventricle trial, Circulation 2010	Negative	230 single-ventricle infants; no benefit on growth, Ross class, BNP, or EF.
Beta-blocker (carvedilol)	Shaddy, JAMA 2007	Negative	161 children; no benefit vs placebo (high ~56% placebo response; heterogeneous population).
Digoxin	Interstage single ventricle, JAHA 2016	Observational	Lower interstage mortality (12.3% → 2.9%; adj. HR 3.5) — registry only, no RCT.
Ivabradine	Bonnet, JACC 2017	Positive (surrogate)	116 children, DCM; ≥ 20% HR reduction in 70% vs 12% placebo → FDA peds approval (≥ 6 mo, DCM).
ARNI (sac/val)	PANORAMA-HF, Circulation 2024	Not superior to enalapril	375 children; both arms improved, no superiority on the global-rank endpoint. FDA approved (≥ 1 yr) on earlier 12-week NT-proBNP data.
MRA (spironolactone)	No pediatric RCT	—	Extrapolated from adult data; used adjunctively.
Berlin Heart EXCOR (VAD)	Fraser, NEJM 2012	Positive (survival)	Bridge-to-transplant; survival greater than ECMO historical control, at cost of stroke/bleeding.

The carvedilol failure is widely read as a trial-design problem (heterogeneous population, high placebo response) rather than proof of no effect, so beta-blockade remains common in stable pediatric DCM. SGLT2 inhibitors and sGC stimulators are newly endorsed by the 2025 ISHLT update but rest on adult data and small pediatric experience — no completed dedicated pediatric RCT.

Acute pediatric HF & myocarditis

Inotropes: milrinone (inodilator, preferred with adequate BP) and dobutamine first-line; epinephrine/dopamine for hypotensive shock. Levosimendan is positioned by ISHLT 2025 as a rescue option in refractory low output. No pediatric RCT establishes a mortality-superior agent.
Acute myocarditis: supportive care is the backbone. IVIG is controversial and not evidence-supported — the Cochrane review found uncertain evidence and does not recommend routine use (Robinson 2020). Fulminant myocarditis warrants early escalation to MCS (ECMO/VAD); paradoxically it carries a better long-term recovery rate than non-fulminant myocarditis if the patient survives the acute phase.
Endpoints: VAD (Berlin Heart EXCOR for smaller children; continuous-flow devices for larger) and heart transplantation for refractory disease.

Overall: pediatric HF pharmacotherapy is a field of negative flagship trials (enalapril in single ventricle, carvedilol) and non-superiority approvals (ARNI). The strongest positive pediatric-specific signals are a VAD (survival) and, on surrogate endpoints, ivabradine. Everything else is extrapolated — manage by physiology and be candid with families about the evidence gap.

Landmark trials

Every study cited on this page, grouped by therapy area and ordered by year. Each acronym links to its primary publication (PubMed permalink or DOI). Effect sizes are the trials’ primary-endpoint results as published; “NEUTRAL” marks trials that did not meet their primary endpoint.

ARNI

Trial	Year	Population	Key result
PARADIGM-HF	2014	HFrEF, EF ≤ 40%, NYHA II-IV	20% ↓ CV death or HF hosp (HR 0.80, 95% CI 0.73-0.87); all-cause death HR 0.84

Beta-blocker

Trial	Year	Population	Key result
US Carvedilol	1996	Chronic HFrEF, EF ≤ 35%	65% ↓ mortality (7.8% vs 3.2%); stopped early for benefit
MERIT-HF	1999	HFrEF, EF ≤ 40%, NYHA II-IV	34% ↓ all-cause mortality (RR 0.66, 95% CI 0.53-0.81); stopped early
CIBIS-II	1999	HFrEF, EF ≤ 35%, NYHA III-IV	34% ↓ mortality (HR 0.66; 11.8% vs 17.3%); fewer sudden deaths; stopped early
COPERNICUS	2001	Severe HFrEF, EF < 25%	35% ↓ all-cause mortality (11.2% vs 16.8%); stopped early
REDUCE-AMI	2024	Post-MI, preserved EF (≥ 50%)	NEUTRAL: death or MI 7.9% vs 8.3% (HR 0.96); no benefit of routine beta-blockade after MI with preserved EF

MRA (steroidal)

Trial	Year	Population	Key result
RALES	1999	Severe HFrEF, EF ≤ 35%, NYHA III-IV	30% ↓ mortality (RR 0.70, 95% CI 0.60-0.82); stopped early
EMPHASIS-HF	2011	HFrEF, EF ≤ 35%, NYHA II	37% ↓ CV death or HF hosp (HR 0.63, 95% CI 0.54-0.74)

SGLT2 inhibitor

Trial	Year	Population	Key result
DAPA-HF	2019	HFrEF, EF ≤ 40%, ± diabetes	26% ↓ worsening HF or CV death: 16.3% vs 21.2% (HR 0.74)
EMPEROR-Reduced	2020	HFrEF, EF ≤ 40%, ± diabetes	25% ↓ CV death or HF hosp: 19.4% vs 24.7% (HR 0.75)

ACEi / ARB (foundational)

Trial	Year	Population	Key result
CONSENSUS	1987	Severe HF, NYHA IV	27% ↓ mortality at 6 mo; 40% at 1 yr; stopped early
SOLVD-Treatment	1991	Symptomatic HFrEF, EF ≤ 35%	16% ↓ mortality (35.2% vs 39.7%; RR 0.84)
Val-HeFT	2001	HFrEF, NYHA II-IV, on background therapy	Neutral mortality; 13% ↓ combined morbidity/mortality (fewer HF hosp)
CHARM-Alternative	2003	HFrEF, EF ≤ 40%, ACEi-intolerant	23% ↓ primary endpoint (HR 0.77, 95% CI 0.67-0.89)
CHARM-Added	2003	HFrEF, EF ≤ 40%, already on ACEi	15% ↓ primary endpoint (HR 0.85, 95% CI 0.75-0.96)

Other HFrEF therapies

Trial	Year	Population	Key result
DIG	1997	HFrEF, EF ≤ 45%, sinus rhythm	No mortality difference (RR 0.99); reduced HF and all-cause hospitalizations
A-HeFT	2004	Self-identified Black patients, HFrEF, NYHA III-IV	43% ↓ rate of death (HR 0.57); improved QoL; stopped early; basis for BiDil approval
SHIFT	2010	HFrEF, EF ≤ 35%, sinus rhythm, HR ≥ 70, on beta-blocker	18% ↓ primary endpoint (HR 0.82); driven by HF hospitalizations
VICTORIA	2020	HFrEF, EF < 45%, recent worsening HF (high-risk)	35.5% vs 38.5% (HR 0.90, 95% CI 0.82-0.98); ARR ~4.2/100 patient-yr
AFFIRM-AHF	2020	Stabilized acute HF, EF < 50%, iron deficiency	Primary composite missed significance (RR 0.79; P = 0.059); total HF hosp ↓ (RR 0.74)
GALACTIC-HF	2021	HFrEF, EF ≤ 35%, NYHA II-IV	8% ↓ composite (HR 0.92); no mortality benefit; not FDA-approved
IRONMAN	2022	HF, EF ≤ 45%, iron deficiency	Trend to benefit, not significant (RR 0.82; P = 0.070)

HFpEF / HFmrEF

Trial	Year	Population	Key result
CHARM-Preserved	2003	HFpEF, EF > 40%, NYHA II-IV	Neutral (HR 0.89, P = 0.118); modest ↓ HF hospitalizations
TOPCAT	2014	HFpEF, EF ≥ 45%	Neutral overall (HR 0.89); regional Americas subgroup signal
PARAGON-HF	2019	HFpEF, EF ≥ 45% (4,796 of 4,822 rand.)	Narrowly missed (rate ratio 0.87, 95% CI 0.75-1.01; P = 0.059); signals in women/lower EF
EMPEROR-Preserved	2021	HFpEF, EF > 40%	21% ↓ CV death or HF hosp (HR 0.79); first positive outcomes trial in HFpEF
DELIVER	2022	HFmrEF/HFpEF, EF > 40%	18% ↓ worsening HF or CV death (HR 0.82); consistent above/below EF 60%
PARAGLIDE-HF	2023	HFmrEF/HFpEF, EF > 40%, after worsening HF	Greater NT-proBNP reduction with ARNI; more hypotension
STEP-HFpEF	2023	HFpEF, EF ≥ 45%, BMI ≥ 30, no diabetes	KCCQ +7.8 and ~10.7% weight loss (both P < 0.001); +20 m 6MWD
FINEARTS-HF	2024	HFmrEF/HFpEF, EF ≥ 40%	16% ↓ worsening HF + CV death (RR 0.84, P = 0.007); first MRA benefit in HFpEF → 2025 FDA approval
STEP-HFpEF DM	2024	HFpEF, EF ≥ 45%, BMI ≥ 30, type 2 diabetes	KCCQ +7.3 and −6.4% weight vs placebo (both P < 0.001)
SUMMIT	2025	HFpEF, EF ≥ 50%, BMI ≥ 30 (obesity phenotype)	CV death or worsening HF 9.9% vs 15.3% (HR 0.62); KCCQ +6.9

Devices & procedures

Trial	Year	Population	Key result
MADIT-II	2002	Prior MI, LVEF ≤ 30%	31% ↓ all-cause mortality with prophylactic ICD (HR 0.69)
COMPANION	2004	Advanced HFrEF, EF ≤ 35%, QRS ≥ 120 ms	CRT ↓ death/hospitalization ~20%; CRT-D ↓ all-cause mortality 36%
SCD-HeFT	2005	HFrEF, EF ≤ 35%, NYHA II-III	Shock-only ICD ↓ mortality 23% (HR 0.77); amiodarone no benefit
CARE-HF	2005	NYHA III-IV, EF ≤ 35%, QRS prolongation	CRT-P ↓ primary endpoint 37% (HR 0.63); ↓ mortality 36%
MADIT-CRT	2009	Mild HFrEF (NYHA I-II), EF ≤ 30%, QRS ≥ 130 ms	34% ↓ HF events (HR 0.66); benefit greatest with LBBB
COAPT	2018	HF + severe secondary MR despite GDMT, EF 20-50%	TEER ↓ HF hosp (HR 0.53) and mortality (HR 0.62) vs GDMT alone
CASTLE-AF	2018	AF + HFrEF, EF ≤ 35%, with ICD/CRT-D	Ablation ↓ death or HF hosp (28.5% vs 44.6%; HR 0.62)
MOMENTUM 3	2019	Advanced HFrEF requiring durable LVAD	HeartMate 3 superior (76.9% vs 64.8%); far fewer pump thromboses/strokes

Acute HF & cardiogenic shock

Trial	Year	Population	Key result
SOAP II	2010	Shock requiring vasopressors (mixed etiology)	No overall mortality difference (52.5% dopamine vs 48.5% NE); more arrhythmias with dopamine
DOSE	2011	Acute decompensated HF	No difference bolus vs infusion; high-dose → more decongestion, transient ↑Cr
IABP-SHOCK II	2012	AMI with cardiogenic shock, early revascularization	Neutral (39.7% vs 41.3%; RR 0.96); neutral long-term
PIONEER-HF	2019	Stabilized acute decompensated HFrEF, EF ≤ 40%	Greater NT-proBNP reduction with in-hospital ARNI (ratio 0.71); fewer rehospitalizations
SOLOIST-WHF	2021	Type 2 diabetes + recent worsening HF	33% ↓ total events (HR 0.67); ended early (funding)
DOREMI	2021	Cardiogenic shock (SCAI B-E)	No difference milrinone vs dobutamine (49% vs 54%; RR 0.90)
EMPULSE	2022	Hospitalized for acute HF, any EF	Net clinical benefit with empagliflozin (win ratio 1.36)
STRONG-HF	2022	Post-acute HF, pre-discharge	Rapid GDMT ↓ death/HF readmission (15.2% vs 23.3%; ARR 8.1%); stopped early
ADVOR	2022	Acute HF with volume overload	Add-on acetazolamide ↑ decongestion (42.2% vs 30.5%; RR 1.46)
CLOROTIC	2023	Acute HF on oral loop diuretic	Add-on HCTZ ↑ weight loss; more renal impairment
ECLS-SHOCK	2023	AMI with cardiogenic shock	Routine VA-ECMO neutral (47.8% vs 49.0%); more bleeding/vascular harm
TRANSFORM-HF	2023	Hospitalized HF, any EF	NEUTRAL: torsemide vs furosemide, mortality 26.1% vs 26.2% (HR 1.02)
DanGer Shock	2024	STEMI-related cardiogenic shock	Impella CP ↓ 180-day death (45.8% vs 58.5%; HR 0.74; NNT ~8); more complications

ATTR-CM (cardiac amyloidosis)

Trial	Year	Population	Key result
ATTR-ACT	2018	Transthyretin amyloid cardiomyopathy, NYHA I-III	Tafamidis ↓ mortality (29.5% vs 42.9%; HR 0.70) and CV hospitalizations
ATTRibute-CM	2024	ATTR-CM, NYHA I-III	Acoramidis win ratio 1.8 (P < 0.001); ~50% ↓ CV hospitalization
HELIOS-B	2025	ATTR-CM (± background tafamidis)	Vutrisiran ↓ death + CV events (HR 0.72); all-cause mortality ↓ ~35%

Pediatric HF

Trial	Year	Population	Key result
Pediatric Carvedilol	2007	Children with symptomatic systolic dysfunction	NEUTRAL: no benefit vs placebo (high placebo response; heterogeneous population)
Infant Single Ventricle	2010	Single-ventricle infants (enalapril)	NEUTRAL: no benefit on growth, Ross class, BNP, or EF
Pediatric Ivabradine	2017	Children (6 mo-18 yr), DCM + chronic HF	≥ 20% HR reduction 70% vs 12% (OR 17.2, P < 0.0001) → FDA peds approval
PANORAMA-HF	2024	Children 1 mo-<18 yr, systemic LV dysfunction	Sacubitril/valsartan NOT superior to enalapril; both improved over 52 wk

Guidelines, key documents & calculators

Recommendation classes and drug approvals reflect the published documents below; consult the primary source before applying to an individual patient.

Adult guidelines

2022 AHA/ACC/HFSA HF Guideline (Heidenreich et al., Circulation 2022) — current US comprehensive guideline; establishes the four-pillar GDMT. DOI · PubMed
2021 ESC HF Guidelines (McDonagh et al., Eur Heart J). DOI
2023 ESC Focused Update — extended SGLT2 inhibitors to HFmrEF/HFpEF. PubMed
2024 ACC Expert Consensus Decision Pathway for HFrEF — practical initiation/titration sequencing. DOI
2025 ACC Concise Clinical Guidance: Cardiogenic Shock — evaluation and MCS device selection. PubMed
2026 Second Universal Definition of HF (AHA/ACC/ESC/WHF; co-published Circulation/JACC/EHJ/Global Heart). AHA summary · Global Heart, open access

Recent therapy note: finerenone (nonsteroidal MRA) was FDA-approved in July 2025 for HF with LVEF ≥ 40%, based on FINEARTS-HF.

Pediatric guidelines & statements

ISHLT 2025 update — Management of Pediatric Heart Failure — supersedes 2014; now endorses ARNI, SGLT2 inhibitors, and sGC stimulators for pediatric chronic HFrEF. Article
ISHLT 2014 (Executive Summary) (Kirk et al.) — the prior foundational edition. PubMed
CCS 2013 — Medical Management of HF in Children (Kantor et al.). PubMed
AHA scientific statements — cardiomyopathy classification (2019) and treatment (2023) in children.

Clinical calculators

MAGGIC Risk Score — 1- and 3-year mortality (HFrEF and HFpEF).
Seattle Heart Failure Model — survival estimate incorporating meds/devices.
GWTG-HF Risk Score — in-hospital mortality for admitted patients.
H₂FPEF Score — probability of HFpEF in unexplained dyspnea.
NYHA Functional Classification — symptom-based severity (I–IV).
HFA-PEFF algorithm — ESC/HFA stepwise HFpEF workup.
Pediatric — Ross classification: functional HF severity in infants (the pediatric NYHA analogue); described in the ISHLT/CCS pediatric guidelines above.