Executive Summary
The evidence base for advanced heart failure has a structural defect: the trials that define guideline-directed medical therapy largely did not enroll advanced heart failure patients, and the trials that did enroll them have frequently been negative. Any treatment plan for a stage D or C2D patient is therefore built partly on extrapolation, and honesty about which parts are extrapolated is the difference between good and careless practice.
Four findings organize this document.
GDMT benefit does not automatically extend. LIFE randomized 335 patients with advanced HFrEF and recent NYHA class IV symptoms to sacubitril/valsartan or valsartan; the NT-proBNP AUC ratio was 0.95 (95% CI 0.84–1.08, P = 0.45) — no benefit — and 18% could not tolerate the run-in while 29% discontinued [1]. Against this, a post hoc GALACTIC-HF analysis found omecamtiv mecarbil benefit concentrated in severe heart failure (HR 0.80, 95% CI 0.71–0.90) with none in the rest (HR 0.99, 95% CI 0.91–1.08; P interaction 0.005) [2]. Both directions are real. Class effects cannot be assumed either way.
Decongestion is where most advanced heart failure is actually managed, and the best evidence is about efficiency, not survival. ADVOR showed IV acetazolamide added to loop diuretics improved successful decongestion at 3 days (42.2% vs 30.5%; ARR 11.7%, NNT 9) with no effect on death or rehospitalization at 3 months (HR 1.07, 95% CI 0.78–1.48) [3]. CARRESS-HF showed that in acute heart failure with worsening renal function and persistent congestion, stepped pharmacologic therapy was superior to ultrafiltration for preserving renal function, with more serious adverse events on ultrafiltration (72% vs 57%, P = 0.03) [4].
Structured, intensive follow-up outperforms most drugs. STRONG-HF up-titrated GDMT to full doses within two weeks of discharge with four monitored visits over two months: 180-day heart failure readmission or death fell from 23.3% to 15.2% (adjusted risk difference 8.1%, 95% CI 2.9–13.2; RR 0.66, 95% CI 0.50–0.86; NNT approximately 12) [5]. The trial was stopped early for benefit. More adverse events occurred in the intensive arm (41% vs 29%), which is the honest price.
When medical therapy is exhausted, mechanical circulatory support delivers the largest absolute effect in this document, and palliative care delivers the most reliable symptom effect. In MOMENTUM 3 at five years, the fully magnetically levitated centrifugal-flow LVAD beat the axial-flow device on the composite of survival to transplant, recovery, or support free of disabling stroke or pump replacement (54.0% vs 29.7%; ARR 24.3 points) with overall survival 58.4% vs 43.7% [6]. PAL-HF improved KCCQ by 9.49 points (95% CI 0.94–18.05) and FACIT-Pal by 11.77 (95% CI 0.84–22.71) without changing rehospitalization or mortality [7].
In stage C, the question is "what else can I add." In advanced heart failure, the question is "what is this patient's dominant physiology, and what does this patient want." Adding drugs to a patient with a cardiac index of 1.5 and an RA pressure of 22 is not therapy — it is postponement.
1. Why the Evidence Base Is Thin Where It Matters Most
Advanced heart failure patients are systematically under-represented in the trials that generate guidelines, for reasons that are individually defensible and collectively crippling: hypotension excludes them, renal dysfunction excludes them, inotrope dependence excludes them, and short life expectancy makes event-driven trials logistically fraught.
The result is a benefit-extrapolation problem. The 2018 HFA-ESC statement states the situation directly: standard treatment is, by definition, insufficient in these patients [8]. When a patient no longer responds to the therapies a trial proved, the trial's estimate no longer straightforwardly applies.
Two empirical checks on extrapolation exist, and they disagree:
| Trial | Population | Result | Implication |
|---|---|---|---|
| LIFE [1] | 335 pts, advanced HFrEF, recent NYHA IV | Sacubitril/valsartan vs valsartan: NT-proBNP AUC ratio 0.95 (0.84–1.08), P = 0.45. No clinical composite benefit. Hyperkalemia 17% vs 9%, P = 0.04 | Benefit proven in stage C did not extend |
| GALACTIC-HF severe subgroup [2] | 2,258/8,232 (27.4%) with NYHA III–IV and EF ≤30% and HF hospitalization within 6 mo | Omecamtiv mecarbil HR 0.80 (0.71–0.90) in severe vs 0.99 (0.91–1.08) in non-severe; P interaction 0.005 | Benefit was concentrated in the advanced group |
LIFE is a properly randomized negative trial in advanced heart failure. The GALACTIC-HF severe analysis is a post hoc subgroup of an industry-sponsored trial with a modest overall result. They carry different evidentiary weight. The correct summary is: advanced heart failure changes drug effects unpredictably, sometimes toward null and possibly sometimes toward benefit, and each agent must be assessed on its own data.
2. Guideline-Directed Medical Therapy at the Ceiling
2.1 What the Four Pillars Show, and Where the Extrapolation Begins
| Therapy | Trial | Primary outcome | Effect | ARR / NNT | Sponsor |
|---|---|---|---|---|---|
| Dapagliflozin | DAPA-HF [9] | Worsening HF or CV death | 16.3% vs 21.2%; HR 0.74 (0.65–0.85) | ARR 4.9%, NNT 21 over median 18.2 mo | AstraZeneca |
| Empagliflozin | EMPEROR-Reduced [10] | CV death or HF hospitalization | 19.4% vs 24.7%; HR 0.75 (0.65–0.86) | ARR 5.3%, NNT 19 over median 16 mo | Boehringer Ingelheim / Eli Lilly |
| Vericiguat | VICTORIA [11] | CV death or first HF hospitalization | 35.5% vs 38.5%; HR 0.90 (0.82–0.98) | ARR 3.0%, NNT 34 over median 10.8 mo | Merck / Bayer |
| Sacubitril/valsartan | LIFE [1] | NT-proBNP AUC, advanced HF | Ratio 0.95 (0.84–1.08) | No benefit | Novartis-supported |
Three observations a careful reader should hold:
- VICTORIA is the most advanced-heart-failure-like of the positive trials. It enrolled patients with recent hospitalization or recent IV diuretic therapy, and the placebo-arm event rate of 38.5% over a median 10.8 months reflects that. Its NNT of 34 is the weakest of the three, which is what one expects when treating sicker patients with a therapy whose relative effect is modest.
- Neither SGLT2 inhibitor trial was designed to test SGLT2 inhibition specifically in advanced heart failure. Applying DAPA-HF and EMPEROR-Reduced to a stage D patient is extrapolation. It is reasonable extrapolation — the agents are hemodynamically well tolerated, and empagliflozin slowed eGFR decline (−0.55 vs −2.28 mL/min/1.73 m²/year, P < 0.001) [10], which matters in a population whose renal decline accelerates [12] — but it is extrapolation.
- Renal protection is a reason to keep SGLT2 inhibitors when other agents are withdrawn. In a phenotype where eGFR slope is itself a prognostic marker, the empagliflozin eGFR-slope finding is a concrete argument for retention.
2.2 The GDMT Ceiling — and What to Do About It
Inability to up-titrate or the need to withdraw GDMT is itself an advanced heart failure criterion — the "P" of I NEED HELP [13] and the HFSA "circulatory–renal limitation to RAAS inhibition or beta-blocker therapy" trigger [8]. The ceiling is usually one of three things:
| Ceiling | Typical trigger | Practical approach |
|---|---|---|
| Hypotension | SBP consistently <90–100 mmHg | Separate dosing times; reduce or stop non-prognostic vasodilators (nitrates, amlodipine, alpha-blockers) before reducing GDMT; reassess volume — hypotension from over-diuresis is reversible |
| Renal | Creatinine rise, falling eGFR on RAASi/ARNI | Distinguish hemodynamic dip from injury; a stable rise without congestion is often tolerable. Check for congestion first — worsening renal function with high RA pressure is a decongestion problem, not a RAASi problem |
| Hyperkalemia | K⁺ >5.5 mmol/L on MRA/ARNI | Potassium binders to preserve GDMT; review dietary and drug contributors. See MRA vs nsMRA vs Aldosterone Synthase Inhibitors |
Venous congestion raises renal venous pressure, reduces filtration fraction, and impairs tubular diuretic secretion — see RHC Comprehensive Interpretation Guide §12. A creatinine that rises because the patient is congested will not improve by stopping the ARNI. Decongest first, then re-challenge.
2.3 Intensity of Follow-Up Is a Therapy
STRONG-HF up-titrated to 100% of recommended doses within two weeks of discharge with four scheduled visits over two months, monitoring clinical status, laboratories, and NT-proBNP [5]:
- Full-dose achievement at day 90: RAAS blockers 55% vs 2%; beta-blockers 49% vs 4%; MRAs 84% vs 46%
- 180-day HF readmission or all-cause death: 15.2% vs 23.3% (adjusted risk difference 8.1%, 95% CI 2.9–13.2; RR 0.66, 95% CI 0.50–0.86; NNT approximately 12)
- Stopped early by the DSMB for greater-than-expected benefit
- Adverse events by day 90: 41% vs 29% (serious adverse events similar)
STRONG-HF's NNT of roughly 12 over 180 days is better than any individual drug in §2.1. The active ingredient is structured, frequent, laboratory-monitored follow-up — which is exactly what an advanced heart failure program provides and what a patient bouncing between an ED and a primary care office does not. This is the single strongest argument for referral.
2.4 The 2026 C2D Adaptation Algorithm
The JACC: Heart Failure position statement on ambulatory stage C2D publishes a practical sequence (its Table 2) for what to do when guideline-directed therapy cannot be tolerated [27]. It is the most operational guidance available for this population, and it is worth following in order rather than jumping to dose reduction.
Step 1 — exclude the reversible before touching GDMT [27]:
| Action | |
|---|---|
| A | Evaluate volume status and address clinical hypervolemia or hypovolemia |
| B | Check for and address low serum albumin, new or severe anemia, thyroid dysfunction |
| C | Address other reversible factors such as arrhythmias that may limit GDMT tolerance |
| D | Review need for unrelated medications that may decrease blood pressure or renal function |
| E | Consider right heart catheterization to resolve uncertainty regarding relative right heart–left heart filling pressures, contribution of pulmonary vascular or parenchymal disease, or unexpected systemic vasodilation or vasoconstriction |
Step 2 — for symptomatic hypotension, substitute before subtracting [27]:
| Pillar | Substitution ladder |
|---|---|
| Renin-angiotensin system | Replace ARNI with ACEI or ARB → replace ACEI with ARB → replace the RAS inhibitor with hydralazine/nitrates |
| Sympathetic system | Replace carvedilol with metoprolol succinate or bisoprolol (selective β₁ blockade minimizes hypotension) → adjust the HF beta-blocker when sotalol is prescribed for arrhythmia or during acute amiodarone loading → down-titrate or stop → supplement or replace with ivabradine when sinus heart rate >70/min |
| Adjunct | Digitalis glycosides — may improve tolerability of neurohormonal antagonists when GDMT is limited by hypotension; may decrease HF hospitalizations added to triple or quadruple therapy; a digitalis load may facilitate weaning from IV inotropes and can acutely slow rate without hypotension in AF with rapid ventricular response |
Step 3 — for postural hypotension with optimal volume status [27]: replace carvedilol with metoprolol or bisoprolol; switch ARNI to ARB or consider stopping all RAS antagonists; re-evaluate chronic tamsulosin/finasteride; review chronic medications for depression, anxiety, and pain that lower blood pressure.
In progressive renal dysfunction, the statement directs review of ACEI, ARB, ARNI, and MRA with consideration of a lower dose or discontinuation — and for diuretic resistance specifically, switching to a loop diuretic with better oral bioavailability (torsemide or bumetanide) and/or adding intermittent thiazide-like agents or acetazolamide for sequential nephron blockade [27].
Notice what comes first. Volume status, albumin, anemia, thyroid, arrhythmia, and non-cardiac medications are all assessed before any GDMT is reduced, and a right heart catheterization is offered before dose reduction rather than after it fails. The commonest error in this population is reflexively halving the ARNI for a blood pressure of 92 without first asking whether the patient is dry, anemic, or on tamsulosin.
What the statement says about the pillars themselves [27]:
- MRAs and SGLT2 inhibitors are the pair most likely to be tolerated — minimal to no effect on perfusion or blood pressure after initiation. A pooled analysis of more severe heart failure patients in DAPA-HF and DELIVER found the reduction in cardiovascular death or HF hospitalization with dapagliflozin was similar in patients with more severe HF compared with others enrolled.
- The remaining pillars "require finesse to assemble when cardiac output and blood pressure are less robust." A modest blood pressure reduction is often tolerated without symptoms, and hypotension has less adverse impact on outcomes when it results from GDMT up-titration.
- Previous intolerance is a risk marker, not a contraindication — it "should not preclude attempts to reinitiate or up-titrate in selected stable patients."
The position statement reads the LIFE trial as a C2D population and lists its defining features: mean LVEF 20%, KCCQ 53, mean systolic BP 113 mm Hg, and furosemide equivalents over 130 mg/day [27]. Even after restricting randomization to the 82% of patients who tolerated low-dose ARNI, by 24 weeks sacubitril/valsartan had been discontinued by 29% and valsartan by 21%, usually for symptomatic hypotension — which the statement interprets as evidence that tolerance even for an ARB diminishes as heart failure advances [27]. Compare those furosemide equivalents with the >160 mg/day advanced heart failure threshold in Classification, Staging, and Referral Criteria §4.
3. Decongestion — The Core Business of Advanced Heart Failure
Most advanced heart failure management is decongestion, and most decongestion failure is physiologic rather than behavioral.
3.1 The Stepwise Approach
- Confirm congestion and quantify it. Weight, JVP, orthopnea, edema, ascites, natriuretic peptides. In the ambulatory C2D patient, congestion is often abdominal rather than pulmonary.
- Fix delivery before raising the dose. Gut edema makes oral furosemide absorption variable and delayed. Switching to IV, or to oral bumetanide whose oral and parenteral responses are nearly equal, improves delivery predictability — see Loop Diuretic Route and Equivalence.
- Double the loop dose rather than increase frequency when response is inadequate; loop diuretics have a threshold.
- Measure the response, do not assume it. Spot urine sodium at 2 hours post-dose and urine output at 6 hours are the HFA-recommended early assessments of diuretic response [14].
- Add a second nephron segment when the loop alone fails (§3.2).
- Escalate to mechanical fluid removal only with a specific indication (§3.4).
TRANSFORM-HF randomized 2,859 patients discharged after HF hospitalization to torsemide or furosemide. All-cause death over a median 17.4 months: 26.1% vs 26.2% (HR 1.02, 95% CI 0.89–1.18), with no difference across EF subgroups [15]. Choose the loop agent for pharmacokinetic reasons — bioavailability, duration, gut edema — not for a survival claim.
Thirty to fifty percent of patients are discharged with residual congestion, carrying 30% to 50% higher rates of rehospitalization and mortality — and decongestion is "often limited by inadequate escalation of diuretics." In RELAX-AHF-2 the average daily IV furosemide dose over the first five hospital days was only 77 mg in patients who were "wet and sent home." The most frequently reported reason for discharging a congested patient is kidney dysfunction [27].
3.2 Sequential Nephron Blockade
Adding a thiazide-type agent to a loop diuretic blocks distal sodium reabsorption that has been upregulated by chronic loop exposure. Two points of evidence discipline this common maneuver:
- Metolazone is not uniquely privileged at low eGFR. The premise is not directly evidenced, and chlorthalidone has shown efficacy at eGFR levels once thought prohibitive — see Water Pills and Diuretics: Evidence Review.
- Metolazone's metabolic cost is measurable. In 61 patients with heart failure resistant to IV furosemide randomized to dapagliflozin 10 mg or metolazone 5–10 mg for 3 days, weight fell 3.0 ± 2.5 kg vs 3.6 ± 2.0 kg (mean difference 0.65 kg, 95% CI −0.12 to 1.41, P = 0.11), with smaller decreases in sodium and potassium and smaller rises in urea and creatinine on dapagliflozin [16].
And remember that reaching for metolazone at all satisfies an ACCF/AHA advanced heart failure criterion [8,17] — see Advanced Heart Failure: Classification, Staging, and Referral Criteria §4.
3.3 Acetazolamide — The Proximal Tubule
ADVOR randomized 519 patients with acute decompensated heart failure and volume overload to IV acetazolamide 500 mg daily or placebo added to standardized IV loop diuretics at twice the oral maintenance dose [3]:
| Endpoint | Acetazolamide | Placebo | Effect |
|---|---|---|---|
| Successful decongestion at 3 days | 108/256 (42.2%) | 79/259 (30.5%) | RR 1.46 (95% CI 1.17–1.82), P < 0.001; ARR 11.7%, NNT 9 |
| Death or HF rehospitalization at 3 months | 76/256 (29.7%) | 72/259 (27.8%) | HR 1.07 (95% CI 0.78–1.48) — no benefit |
Acetazolamide produced higher cumulative urine output and natriuresis, with similar rates of worsening kidney function, hypokalemia, and hypotension [3]. Funded by the Belgian Health Care Knowledge Centre — not industry-sponsored, which strengthens the finding.
ADVOR proved faster and more complete decongestion, not improved survival or fewer readmissions. Use it as a decongestion tool in the congested patient, and do not represent it as a disease-modifying therapy. Watch the bicarbonate and the potassium.
3.4 Ultrafiltration — A Narrow Indication
CARRESS-HF randomized 188 patients with acute decompensated heart failure, worsened renal function, and persistent congestion to stepped pharmacologic therapy or ultrafiltration [4]:
- Ultrafiltration was inferior on the bivariate endpoint of creatinine change and weight change at 96 hours (P = 0.003), driven by a creatinine rise on ultrafiltration (+0.23 ± 0.70 vs −0.04 ± 0.53 mg/dL, P = 0.003)
- Weight loss was equivalent (5.7 ± 3.9 kg vs 5.5 ± 5.1 kg, P = 0.58)
- Serious adverse events were more frequent with ultrafiltration (72% vs 57%, P = 0.03; absolute increase 15 points, NNH approximately 7)
- NHLBI-funded — not industry-sponsored
Ultrafiltration removed the same fluid with worse renal outcomes and more harm than a protocolized, escalating diuretic algorithm. The comparator was not lazy care — it was a stepped algorithm. Reserve ultrafiltration for genuine diuretic failure or an established renal replacement indication, and treat "we tried Lasix and it did not work" as a claim to be verified with a urine sodium before it justifies a circuit.
3.5 What Does Not Work
ROSE-AHF randomized patients with acute heart failure and renal dysfunction to low-dose dopamine, low-dose nesiritide, or placebo added to diuretics [18]. Neither agent improved 72-hour cumulative urine volume (dopamine difference 229 mL, 95% CI −714 to 1,171; nesiritide difference 279 mL, 95% CI −618 to 1,176) nor cystatin C, nor any secondary endpoint of decongestion, renal function, or clinical outcome [18]. NIH-funded.
ROSE-AHF was adequately powered, placebo-controlled, publicly funded, and unambiguously negative for the exact clinical scenario in which renal-dose dopamine is still occasionally requested [18].
3.6 Iron Deficiency
IRONMAN randomized 1,137 patients with heart failure, reduced EF, and iron deficiency to IV ferric derisomaltose or usual care, median follow-up 2.7 years [19]:
- Primary endpoint (HF hospitalization or CV death): 22.4 vs 27.5 per 100 patient-years, RR 0.82 (95% CI 0.66–1.02), P = 0.070 — did not reach significance
- COVID-adjusted sensitivity analysis: RR 0.76 (95% CI 0.58–1.00), P = 0.047
- Fewer cardiac serious adverse events: 36% vs 43% (difference −7.0%, 95% CI −12.69 to −1.32, P = 0.016)
- Funded by the British Heart Foundation and Pharmacosmos (partial industry)
Iron repletion is a reasonable symptomatic and event-reducing adjunct with a genuinely borderline primary result. Report it as borderline.
4. Hemodynamic Assessment and Monitoring
4.1 Invasive Hemodynamics — Diagnostic, Not Routine
The 2018 HFA-ESC position is precise: invasive hemodynamic assessment does not improve the accuracy of heart failure prognostication, but it is a critical component of the work-up for potential heart transplantation or long-term MCS recipients, allowing accurate estimation of pulmonary capillary wedge pressure, pulmonary vascular resistance, and transpulmonary gradient, and adding to the assessment of right ventricular function [8]. Invasive hemodynamic monitoring is not routinely recommended for in-hospital management of advanced heart failure, but is useful in critical conditions such as cardiogenic shock not responding to standard treatment [8].
The empirical basis for that restraint is ESCAPE, which randomized patients hospitalized with severe heart failure (average LVEF 19%, SBP 106 mmHg, creatinine 1.5 mg/dL) to therapy guided by clinical assessment alone or with a pulmonary artery catheter [20]:
- Days alive and out of hospital at 6 months: 133 vs 135 (HR 1.00, 95% CI 0.82–1.21, P = 0.99)
- Mortality: 10% vs 9% (OR 1.26, 95% CI 0.78–2.03, P = 0.35)
- In-hospital adverse events: 21.9% vs 11.5%, P = 0.04
- No deaths related to PAC use
ESCAPE did not show that hemodynamics are uninformative. It showed that routine catheterization to guide decongestion in patients whose congestion is clinically obvious adds risk without benefit. The indications that survive ESCAPE are the ones where the clinical assessment is genuinely uncertain, where perfusion is in question, or where a number is needed for a decision — transplant candidacy, MCS evaluation, pulmonary vascular resistance, phenotyping the C2D patient. See RHC Comprehensive Interpretation Guide.
4.2 Implanted Pulmonary Artery Pressure Monitoring
| Trial | Population | Result |
|---|---|---|
| CHAMPION [21] | NYHA class III, prior HF hospitalization | HF hospitalizations at 6 months 0.31 vs 0.44 per patient-year (HR 0.70, 95% CI 0.60–0.84, P < 0.0001); 39% reduction over mean 15 months (HR 0.64, 95% CI 0.55–0.75). Funded by CardioMEMS |
| GUIDE-HF [22] | Broader: NYHA II–IV with hospitalization or elevated natriuretic peptides | Primary composite not met overall: HR 0.88 (95% CI 0.74–1.05, P = 0.16). Pre-COVID impact analysis HR 0.81 (95% CI 0.66–1.00, P = 0.049). Funded by Abbott |
CHAMPION was strongly positive in NYHA class III patients. GUIDE-HF, in a broader population and disrupted by the pandemic, missed its primary endpoint, with benefit appearing only in a pre-specified pre-COVID sensitivity analysis. Both trials were funded by the device manufacturer. The defensible summary is that hemodynamic-guided management has good evidence in the NYHA III, previously hospitalized patient and unproven evidence outside it.
5. Inotropes
Inotropes occupy a specific, narrow, and honestly palliative place. The HFA-ESC statement is explicit: inotropic therapy may be used as a bridge strategy, but it is only a palliative measure when used on its own, because of the lack of outcomes data [8].
| Use | Rationale | Evidence status |
|---|---|---|
| Bridge to decision / transplant / MCS | Maintain perfusion and end-organ function while a definitive plan is executed | Accepted practice; supported by the MCS evaluation pathway [8] |
| Bridge to recovery | Selected acute myocarditis, peripartum, post-cardiotomy | Case-based |
| Palliative / symptom-directed ambulatory inotropes | Symptom relief and reduced hospitalization in patients declining or ineligible for advanced therapies | No mortality benefit; used with explicit goals-of-care framing [8] |
| "Renal-dose" dopamine to improve diuresis | — | Refuted by ROSE-AHF [18] |
Requirement of inotropes defines INTERMACS Profiles 1–3 and is the "I" of I NEED HELP [8,13]. Starting an inotrope is therefore simultaneously a treatment and a staging event.
6. Advanced Therapies
6.1 Durable Mechanical Circulatory Support
MOMENTUM 3 five-year outcomes, in the per-protocol population still receiving LVAD support at 2 years (289 centrifugal-flow, 247 axial-flow; median age 62, 18% women) [6]:
| Endpoint at 5 years | Centrifugal-flow (HeartMate 3) | Axial-flow (HeartMate II) | Effect |
|---|---|---|---|
| Survival to transplant, recovery, or support free of disabling stroke or pump replacement | 54.0% | 29.7% | HR 0.55 (95% CI 0.45–0.67), P < 0.001; ARR 24.3 points, NNT ≈ 5 |
| Overall survival | 58.4% | 43.7% | HR 0.72 (95% CI 0.58–0.89), P = 0.003; ARR 14.7 points, NNT ≈ 7 |
Serious adverse events of stroke, bleeding, and pump thrombosis were less frequent with the centrifugal-flow pump [6].
This analysis is restricted to patients still on support at two years and is an observational extended-phase follow-up of a randomized trial, not a fresh randomization. It answers "among durable LVAD patients who reach two years, which pump does better over five" — not "what happens to an advanced heart failure patient who gets an LVAD." The effect size is nonetheless the largest in this document.
Two population facts constrain how often this option applies: REVIVAL found robust survival in ambulatory advanced patients not selected for VAD or transplant, which its authors state challenges advocacy for earlier VAD implantation [23]; and 43.4% of advanced heart failure is HFpEF [24], for whom durable LVAD is not an option.
6.2 Transplantation
Pre-transplant evaluation has three components per the HFA-ESC statement: establishing candidacy, estimating prognosis (greatest benefit accrues to patients at high mortality risk without transplant who also have good expected post-transplant survival), and evaluating comorbidities that affect surgical or post-transplant outcomes [8]. The required workup explicitly includes complete history and examination, cardiopulmonary exercise testing, right heart catheterization, peripheral vascular assessment, frailty and nutritional assessment, organ function determination (lung, liver, kidney), neoplasm and infection screening, prognostic scores (HFSS, SHFM, IMPACT), blood group compatibility, HLA antibody assessment, and complete psychosocial evaluation [8].
Access is inequitable in measured ways: Black patients with advanced heart failure were more likely to receive an LVAD (P = 0.010) and less likely to receive a transplant than White patients (P = 0.034), and advanced therapies were used less in older patients and in women [25].
7. Palliative Care — Concurrent, Not Terminal
PAL-HF randomized 150 patients with advanced heart failure to usual care or usual care plus an interdisciplinary palliative care intervention, with quality of life at 6 months as the primary endpoint [7]:
| Outcome at 6 months | Effect |
|---|---|
| KCCQ overall summary | +9.49 points (95% CI 0.94–18.05), P = 0.030 |
| FACIT-Pal | +11.77 points (95% CI 0.84–22.71), P = 0.035 |
| HADS-depression | −1.94 points, P = 0.020 |
| HADS-anxiety | −1.83 points, P = 0.048 |
| FACIT-Spiritual Well-Being | +3.98 points, P = 0.027 |
| Rehospitalization, mortality | No effect |
PAL-HF delivered a clinically meaningful KCCQ improvement — comparable in magnitude to what device therapies are celebrated for — with no survival cost and no survival benefit. Concurrent palliative care is not what happens when treatment stops. In a disease with a median survival of 12.2 months from diagnosis of the advanced stage [24], it is part of treatment. Note the wide confidence intervals: this was a 150-patient single-center trial.
8. A Practical Approach
8.1 Sequence
- Confirm the stage. Apply the HFA-ESC four criteria and assign an INTERMACS profile with modifiers — Classification, Staging, and Referral Criteria.
- Name the dominant physiology. Low output, right heart failure, cardiorenal, congestive hepatopathy, frailty — Stage C2D Heart Failure §2. Use RHC when clinical assessment is genuinely uncertain or a number will change a decision.
- Decongest to a defined endpoint, verifying response with urine sodium and output rather than assuming it (§3).
- Re-attempt GDMT after decongestion, not before. Congestion is the commonest reversible cause of apparent intolerance (§2.2).
- Escalate follow-up intensity, which has the best NNT in this document (§2.3).
- Refer while options remain open. Referral is not a commitment to transplant; it is an evaluation.
- Have the goals-of-care conversation early and concurrently (§7).
8.2 The Nephrology-Specific Contribution
| Task | Why it is ours |
|---|---|
| Quantify diuretic burden in furosemide equivalents | The >160 mg/day threshold is an advanced HF criterion [8,17] |
| Verify diuretic response objectively | Urine sodium at 2 h and output at 6 h [14] — prevents escalation based on assumption |
| Distinguish congestion-driven from injury-driven creatinine rise | Determines whether GDMT can be preserved |
| Track eGFR slope | Decline accelerates from 7.6% to 10.9% per year across the transition [12] |
| Gatekeep ultrafiltration | CARRESS-HF puts the burden of proof on the circuit [4] |
| Manage hyperkalemia to preserve GDMT | The "P" of I NEED HELP [13] |
| Recognize the HFpEF majority has no device pathway | 43.4% of advanced HF [24] |
9. Evidence Gaps
- No trial has prospectively enrolled by advanced heart failure or C2D status and tested a GDMT strategy.
- Nothing in this document addresses the HFpEF majority of advanced heart failure with device-level evidence.
- Natriuresis-guided diuretic strategies [14] are validated in hospitalized patients; their ambulatory role is untested.
- Ambulatory inotropic therapy has no randomized outcome data.
- The optimal timing of referral remains undefined — REVIVAL's survival in non-selected ambulatory patients argues against reflexive early VAD [23].
References
- Mann DL, Givertz MM, Vader JM, et al. Effect of treatment with sacubitril/valsartan in patients with advanced heart failure and reduced ejection fraction: a randomized clinical trial. JAMA Cardiol. 2022;7(1):17-25. PMID: 34730769
- Felker GM, Solomon SD, Claggett B, et al. Assessment of omecamtiv mecarbil for the treatment of patients with severe heart failure: a post hoc analysis of data from the GALACTIC-HF randomized clinical trial. JAMA Cardiol. 2022;7(1):26-34. PMID: 34643642
- Mullens W, Dauw J, Martens P, et al. Acetazolamide in acute decompensated heart failure with volume overload. N Engl J Med. 2022;387(13):1185-1195. PMID: 36027559
- Bart BA, Goldsmith SR, Lee KL, et al. Ultrafiltration in decompensated heart failure with cardiorenal syndrome. N Engl J Med. 2012;367(24):2296-2304. PMID: 23131078
- Mebazaa A, Davison B, Chioncel O, et al. Safety, tolerability and efficacy of up-titration of guideline-directed medical therapies for acute heart failure (STRONG-HF): a multinational, open-label, randomised, trial. Lancet. 2022;400(10367):1938-1952. PMID: 36356631
- Mehra MR, Goldstein DJ, Cleveland JC, et al. Five-year outcomes in patients with fully magnetically levitated vs axial-flow left ventricular assist devices in the MOMENTUM 3 randomized trial. JAMA. 2022;328(12):1233-1242. PMID: 36074476
- Rogers JG, Patel CB, Mentz RJ, et al. Palliative care in heart failure: the PAL-HF randomized, controlled clinical trial. J Am Coll Cardiol. 2017;70(3):331-341. PMID: 28705314
- Crespo-Leiro MG, Metra M, Lund LH, et al. Advanced heart failure: a position statement of the Heart Failure Association of the European Society of Cardiology. Eur J Heart Fail. 2018;20(11):1505-1535. PMID: 29806100
- McMurray JJV, Solomon SD, Inzucchi SE, et al. Dapagliflozin in patients with heart failure and reduced ejection fraction. N Engl J Med. 2019;381(21):1995-2008. PMID: 31535829
- Packer M, Anker SD, Butler J, et al. Cardiovascular and renal outcomes with empagliflozin in heart failure. N Engl J Med. 2020;383(15):1413-1424. PMID: 32865377
- Armstrong PW, Pieske B, Anstrom KJ, et al. Vericiguat in patients with heart failure and reduced ejection fraction. N Engl J Med. 2020;382(20):1883-1893. PMID: 32222134
- Quiram BJ, Killian JM, Redfield MM, et al. Changes in kidney function after diagnosis of advanced heart failure. J Card Fail. 2023;29(12):1617-1625. PMID: 37451601
- Baumwol J. "I Need Help" — a mnemonic to aid timely referral in advanced heart failure. J Heart Lung Transplant. 2017;36(5):593-594. PMID: 28258792
- Mullens W, Damman K, Harjola VP, et al. The use of diuretics in heart failure with congestion — a position statement from the Heart Failure Association of the European Society of Cardiology. Eur J Heart Fail. 2019;21(2):137-155. PMID: 30600580
- Mentz RJ, Anstrom KJ, Eisenstein EL, et al. Effect of torsemide vs furosemide after discharge on all-cause mortality in patients hospitalized with heart failure: the TRANSFORM-HF randomized clinical trial. JAMA. 2023;329(3):214-223. PMID: 36648467
- Yeoh SE, Osmanska J, Petrie MC, et al. Dapagliflozin vs. metolazone in heart failure resistant to loop diuretics. Eur Heart J. 2023;44(31):2966-2977. PMID: 37210742
- Yancy CW, Jessup M, Bozkurt B, et al. 2013 ACCF/AHA guideline for the management of heart failure: executive summary. Circulation. 2013;128(16):1810-1852. PMID: 23741057
- Chen HH, Anstrom KJ, Givertz MM, et al. Low-dose dopamine or low-dose nesiritide in acute heart failure with renal dysfunction: the ROSE acute heart failure randomized trial. JAMA. 2013;310(23):2533-2543. PMID: 24247300
- Kalra PR, Cleland JGF, Petrie MC, et al. Intravenous ferric derisomaltose in patients with heart failure and iron deficiency in the UK (IRONMAN): an investigator-initiated, prospective, randomised, open-label, blinded-endpoint trial. Lancet. 2022;400(10369):2199-2209. PMID: 36347265
- Binanay C, Califf RM, Hasselblad V, et al. Evaluation study of congestive heart failure and pulmonary artery catheterization effectiveness: the ESCAPE trial. JAMA. 2005;294(13):1625-1633. PMID: 16204662
- Abraham WT, Adamson PB, Bourge RC, et al. Wireless pulmonary artery haemodynamic monitoring in chronic heart failure: a randomised controlled trial. Lancet. 2011;377(9766):658-666. PMID: 21315441
- Lindenfeld J, Zile MR, Desai AS, et al. Haemodynamic-guided management of heart failure (GUIDE-HF): a randomised controlled trial. Lancet. 2021;398(10304):991-1001. PMID: 34461042
- Aaronson KD, Stewart GC, Stevenson LW, et al. Optimizing triage of ambulatory patients with advanced heart failure: 2-year outcomes from REVIVAL. JACC Heart Fail. 2024;12(10):1734-1746. PMID: 38970587
- Dunlay SM, Roger VL, Killian JM, et al. Advanced heart failure epidemiology and outcomes: a population-based study. JACC Heart Fail. 2021;9(10):722-732. PMID: 34391736
- Subramaniam A, van Houten H, Redfield MM, et al. Advanced heart failure characteristics and outcomes in commercially insured U.S. adults. JACC Heart Fail. 2023;11(11):1595-1606. PMID: 37589611
- Dunlay SM, Colvin M, Cowger JA, et al. Ambulatory stage C2D heart failure definitions and current therapeutic approaches: JACC: Heart Failure position statement. JACC Heart Fail. 2026;14(6):103101. PMID: 42230068
- Dunlay SM, Colvin M, Cowger JA, et al. Ambulatory stage C2D heart failure definitions and current therapeutic approaches: JACC: Heart Failure position statement — Table 2 (GDMT adaptations) and section Key Points. JACC Heart Fail. 2026;14(6):103101. PMID: 42230068
Related Reading
- Advanced Heart Failure: Classification, Staging, and Referral Criteria — The seven INTERMACS profiles, the formal definitions of advanced heart failure, and the diuretic thresholds that should trigger referral.
- Stage C2D Heart Failure — The ambulatory transition from stage C to stage D: the 2026 JACC Heart Failure position statement, its criteria, and its prognosis.
- HFpEF Diagnostic Criteria 2025 — relevant because most advanced heart failure is not HFrEF.
- Cardiorenal Module index