Acute decompensated heart failure and cardiogenic pulmonary oedema · ICU Topics

Cinematic ICU scene of flash pulmonary oedema with the patient upright on CPAP, frothy sputum in the suction tubing, an IV furosemide syringe and a nitroglycerin infusion on the trolley, bilateral fluffy infiltrates on a chest X-ray on the screen, clinical-blue lighting, medical educational, no text, no people — FigureThe acute heart failure — the pulmonary oedema and the systemic congestion. The upright posture, the high-flow oxygen, the IV furosemide, the CPAP that lowers the mortality, and the nitroglycerin if the systolic is above 110. Identify and treat the precipitant.

Epidemiology and pathophysiology

Educational diagram of acute heart failure pathophysiology: elevated left ventricular filling pressure leading to pulmonary oedema and forward failure hypoperfusion — FigurePathophysiology of AHF: high LV filling pressure drives pulmonary congestion; reduced forward flow drives hypoperfusion. Therapy targets both axes (decongestion ± inotrope/MCS).

Acute heart failure (AHF) is responsible for >1 million hospitalisations annually in the United States and is the leading cause of hospital admission in patients over 65 in the developed world. In-hospital mortality ranges from 4% (warm-wet presentations) to >40% (cardiogenic shock), and 30-day readmission rates approach 25%. Half of all heart failure patients die within 5 years of diagnosis — a prognosis comparable to many metastatic cancers.[1]

The pathophysiology is a vicious cycle of decompensation: [1]

Index insult (myocardial ischaemia, arrhythmia, volume load) reduces stroke volume.
Fall in cardiac output activates the sympathetic nervous system and renin-angiotensin-aldosterone system (RAAS) in a compensatory attempt to maintain perfusion pressure.
Maladaptive neurohormonal activation drives sodium/water retention (aldosterone, ADH), vasoconstriction (angiotensin II, noradrenaline, endothelin), tachycardia, and direct cardiotoxicity.
Increased preload and afterload raise ventricular wall stress → further reduction in stroke volume, mitral/tricuspid regurgitation, and myocardial injury (troponin leak from subendocardial ischaemia due to elevated end-diastolic pressure).
Venous congestion (raised left-sided filling pressures) transudates fluid into the pulmonary interstitium and alveoli → pulmonary oedema and hypoxaemia, which worsens myocardial ischaemia.
Right heart distension (ventricular interdependence) impairs LV filling and coronary perfusion.
Cardiorenal syndrome develops as venous congestion (not low output per se) reduces glomerular filtration, creating a self-perpetuating cycle of worsening renal function, more diuretic resistance, and escalating congestion. [1]

The central therapeutic insight is that congestion (raised filling pressures), not low cardiac output in isolation, drives symptoms and most organ injury in the typical warm-wet patient. Decongestion — not augmentation of contractility — is therefore the primary target for the majority of presentations.[1][3]

Clinical classification — Stevenson hemodynamic profiles

The Stevenson (Nohria-Stevenson) classification stratifies AHF at the bedside by two axes — perfusion (warm vs cold, reflecting cardiac output) and congestion (wet vs dry, reflecting filling pressures). It is the dominant clinical framework taught in CICM/FFICM because it is determined without invasive monitoring and directly guides therapy.[5]

Profile	Congestion	Perfusion	Clinical findings	Treatment	Mortality signal
Warm-wet	Wet (JVP raised, pulmonary crackles, oedema)	Warm (well perfused, SBP >90, normal lactate)	Most common presentation (~70%). Pulmonary oedema ± peripheral oedema, normal extremities, adequate urine output	IV loop diuretics (1-2.5x oral dose) ± vasodilator (GTN if SBP >110) ± CPAP/BiPAP	Moderate
Warm-dry	Dry (no JVP rise, clear chest)	Warm	Euvolaemic, well perfused — the target state after treatment, or a non-congested decompensation (e.g. isolated high-output state)	Continue/optimise oral GDMT; identify non-volume precipitant	Low
Cold-wet	Wet	Cold (clammy, mottled, oliguric, confused, SBP <90 or lactate >2)	Cardiogenic shock with congestion. SBP often <90, cold extremities, oliguria, altered mentation	Inotrope (dobutamine/milrinone/levosimendan) ± vasopressor (noradrenaline to MAP ≥65) + decongest; consider mechanical support (Impella/VA-ECMO) early	High
Cold-dry	Dry	Cold	Low output without congestion — over-diuresed, or pure forward failure. Often pre-load dependent	Cautious fluid challenge (250 mL crystalloid) ± inotrope; look for occult bleeding/sepsis	Moderate-high

Warm-wet (most common)

~70% of AHF

Adequate perfusion preserved (SBP >90, warm extremities, lactate <2 mmol/L)
Congestion dominates symptoms: orthopnoea, crackles, raised JVP, peripheral oedema
Therapy target: decongest with IV loop diuretic ± vasodilator; CPAP if pulmonary oedema
Avoid inotropes (no hypoperfusion — they add arrhythmia/ischaemia risk without benefit)
Beta-blocker continuation: continue if chronic stable; hold if decompensated

Cold-wet (cardiogenic shock)

~5-8% of AHF

Hypoperfusion + congestion: SBP <90, cold/clammy, oliguria <0.5 mL/kg/h, lactate >2
Different pathway: do NOT rely on diuretics alone — they fail in low-output state
Inotrope first-line: dobutamine, milrinone or levosimendan to augment cardiac output
Vasopressor (noradrenaline) if MAP <65 despite inotrope
Escalate EARLY to MCS (Impella/VA-ECMO) — late escalation has poor outcomes
Revascularise if ACS (CULPRIT-SHOCK: culprit-only PCI)

Cold-dry (over-diuresed / forward failure)

Rare

Hypoperfusion without congestion: tachycardia, narrow pulse pressure, oliguria
Often iatrogenic (over-diuresis) or occult hypovolaemia (bleeding, sepsis)
Cautious fluid bolus 250 mL crystalloid, reassess
Bedside echo: small hypercontractile LV with kissing septum suggests underfilled
Inotrope only if confirmed low output not corrected by volume

Warm-dry (target / compensated)

Discharge-ready state

No congestion, normal perfusion — goal of therapy
May represent non-congested decompensation: anaemia, thyrotoxicosis, tachyarrhythmia
Transition to oral GDMT (4 pillars: BB + MRA + ARNI/ACEi + SGLT2i)
Discharge planning begins here

[3] [5]

Forrester classification (invasive hemodynamic)

The earlier Forrester classification (1976) uses invasive Swan-Ganz catheter data — cardiac index (CI) and pulmonary capillary wedge pressure (PCWP) — and is largely of historical and exam interest now that echocardiography and clinical assessment dominate. It is conceptually identical to Stevenson but uses measured rather than inferred values.[1]

Forrester classification — ADHF by cardiac index and PCWP

CI <2.2, PCWP <18 (cold + dry)

Mortality Moderate-high

Low output without congestion. Hypovolaemic or over-diuresed. Cautious fluid challenge.

[1]

Clinical presentation

Pulmonary oedema (left-sided)

Acute backward failure

Acute breathlessness, orthopnoea, paroxysmal nocturnal dyspnoea
Pink frothy sputum, bilateral coarse crackles, wheeze ("cardiac asthma")
Tachypnoea (RR >25), tachycardia, accessory muscle use
Hypoxia, often hypertensive initially (sympathetic surge)
CXR: bat-wing perihilar infiltrates, Kerley B lines, upper lobe blood diversion, pleural effusions

Cardiogenic shock (forward failure)

Cold-wet or cold-dry

Hypotension (SBP <90 or >30 below baseline), narrow pulse pressure
Cold clammy mottled skin, prolonged capillary refill
Oliguria (<0.5 mL/kg/h), altered mental status
Lactate >2 mmol/L, metabolic acidosis
Often complicating large MI, fulminant myocarditis, end-stage cardiomyopathy

Right heart failure

Isolated backward RV failure

Raised JVP, hepatomegaly, ascites, marked peripheral/sacral oedema
Relatively clear lung fields (no pulmonary oedema)
Causes: RV infarct (inferior MI), pulmonary embolism, severe TR/PR, pulmonary hypertension, ARVC
Kussmaul sign (JVP rises on inspiration) suggests severe RV failure or constrictive pericarditis

Flash pulmonary oedema

Sudden onset, recurrent

Explosive pulmonary oedema over minutes — bilateral renal artery stenosis (Pickering syndrome) until proven otherwise
Other causes: severe mitral regurgitation (flail leaflet), bilateral RAS, phaeochromocytoma, renal failure + volume load
Clue: AKI after starting ACE inhibitor → suspect bilateral RAS
Often markedly hypertensive on presentation

[1]

Identify the precipitant

Every AHF admission has a precipitant that must be actively sought and treated — decongestion alone without addressing the trigger leads to rapid readmission. [1]

Common precipitants

Find and treat

Acute coronary syndrome (#1 — check ECG, troponin)
Arrhythmia (especially new AF — rate control)
Infection (pneumonia, UTI — check cultures, CRP)
Medication non-adherence (stopped diuretics/ACEi)
Anaemia (transfuse if Hb <70)
Thyroid dysfunction (thyroid storm or myxoedema)
Renal failure (worsening AKI → fluid overload)
Dietary indiscretion (high salt intake)
Drugs (NSAIDs, calcium channel blockers, steroids — sodium retention)

Less common / high-yield for exam

Don't miss

Negative inotropes: diltiazem, verapamil, most antiarrhythmics
Cardiotoxic drugs: anthracyclines, trastuzumab, tyrosine kinase inhibitors
Alcohol-induced (holiday heart), cocaine, amphetamines
Peripartum cardiomyopathy (last month of pregnancy to 5 months postpartum)
Takotsubo (stress) cardiomyopathy — apical ballooning
Acute myocarditis (viral — coxsackie, parvovirus; giant cell)
Pulmonary embolism (sudden RV strain)
Endocarditis with acute valvular regurgitation
Tamponade / constrictive pericarditis
Tachycardia-mediated cardiomyopathy (persistent SVT/AF)

[1] [4]

Investigations

ADHF investigations

ECG

ACS, arrhythmia, LVH

Within 10 min

NT-proBNP

Diagnostic biomarker

>300 excludes, >1800 confirms

Echo

Assess LV function

HFrEF vs HFpEF, valve function

CXR

Pulmonary oedema

Bat-wing appearance, Kerley B lines

[1]

A full panel includes: ECG (within 10 minutes — ACS, AF, LVH, old Q waves), chest X-ray (pulmonary oedema, cardiomegaly, effusions, alternative diagnosis), bloods — FBC (anaemia, leucocytosis), U&E (renal function, K+), LFTs (hepatic congestion), troponin (ACS vs type 2 from wall stress), BNP/NT-proBNP (rule-out if <300 ng/L; rule-in if >1800 age-adjusted), CRP, D-dimer (if PE suspected), TSH, lactate (perfusion marker), venous blood gas (acid-base, mixed venous saturation), and urinalysis. Echocardiography (early, ideally within 48 hours) is essential — defines HFrEF (EF <40%) vs HFmrEF (41-49%) vs HFpEF (≥50%), identifies regional wall motion abnormalities, valve lesions, pericardial effusion, and estimates filling pressures/right heart function. [1]

Bedside ultrasound and POCUS in AHF

Point-of-care ultrasound accelerates diagnosis and monitoring of decongestion: [1]

Lung ultrasound: B-lines (comet-tail artefacts) arising from pleural line indicate interstitial/alveolar fluid. Bilateral diffuse B-lines confirm cardiogenic pulmonary oedema; their resolution tracks successful diuresis. A-lines (horizontal reverberation) indicate dry lung.
Cardiac POCUS: identifies gross LV/RV dysfunction, pericardial effusion, gross valvular pathology, and estimates LV filling (E/e′). A small, hypercontractile, underfilled LV suggests over-diuresis (cold-dry); a dilated hypocontractile LV confirms cardiomyopathy.
IVC ultrasound: a dilated plethoric IVC (>2.1 cm with <50% collapse) supports high right atrial pressure and congestion; a small collapsible IVC suggests volume responsiveness.
** pleural effusion**: quantified and tapped for diagnostic/therapeutic thoracentesis. [1]

Management of cardiogenic pulmonary oedema

Educational management pathway for cardiogenic pulmonary oedema: upright posture, oxygen, early CPAP or NIV, IV loop diuretic, vasodilator if hypertensive, inotropes if cold and shocked — FigureCardiogenic pulmonary oedema pathway: sit upright → oxygen → **early CPAP/NIV** → IV loop diuretic → nitrate if SBP allows → escalate to inotropes/MCS if cold/shocked.

Acute pulmonary oedema management

Position + oxygen

Sit UPRIGHT (reduces preload, improves ventilation). High-flow oxygen via non-rebreather mask (15 L/min). Target SpO2 >92% (or >88% in COPD).

Non-invasive ventilation (CPAP/BiPAP) — EARLY

CPAP 5-10 cmH2O or BiPAP (IPAP 10, EPAP 5). Reduces work of breathing, improves oxygenation, decreases preload and afterload (positive intrathoracic pressure). Evidence: reduces intubation rates and mortality. Start within 15-30 min of presentation. Continue until patient comfortable, SpO2 >92%, RR <25.<Cite id="2" />

IV furosemide

Give 1-2.5x the patient usual oral dose IV. For new-onset HF: furosemide 40-80 mg IV. For chronic HF on diuretics: give 2-2.5x usual oral dose IV. Mechanism: initially venodilation (within 5-10 min, before diuretic effect), then diuresis. Monitor urine output (target >100 mL/h initially).<Cite id="6" />

Vasodilator (nitroglycerin) — if SBP >110

IV nitroglycerin (start 10 mcg/min, titrate up) or sublingual GTN spray (2 puffs). Reduces preload (venodilation) and afterload (arteriolar dilation). Reduces mitral regurgitation. Do NOT use if SBP <110 (risk of hypotension). Contraindicated if phosphodiesterase inhibitor (sildenafil) within 24h.

Morphine — USE CAUTIOUSLY

Morphine 2.5-5 mg IV (reduces anxiety, preload via venodilation, breathlessness). BUT associated with increased ICU admission and mortality in some studies (observational — confounding likely). Use selectively for severe anxiety/distress. Not routinely recommended.

Do NOT give beta-blockers acutely

Acute beta-blockade has negative inotropic effects → may precipitate/worsen cardiogenic shock. Start beta-blockers only once the patient is stabilised and euvolaemic (usually 24-48h after presentation). Continue pre-existing beta-blocker if patient is stable and not in shock.

[1] [3]

Oxygen therapy detail

Indications and targets

Saturate judiciously

Target SpO2 92-96% in most patients; 88-92% in COPD retainers
Supplemental O2 only if hypoxaemic — routine O2 in non-hypoxaemic patients is NOT beneficial (and may be harmful via coronary vasoconstriction and ROS)
High-flow nasal cannula (HFNC) is an alternative to NIV in cooperative patients with moderate hypoxaemia
Reserve endotracheal intubation for: respiratory failure despite NIV, reduced GCS, inability to protect airway, exhaustion (rising CO2)

NIV / CPAP — the evidence

Mortality benefit

Mechanism: positive intrathoracic pressure → ↑FRC, ↓work of breathing, ↓LV preload and afterload, redistributes alveolar fluid
CPAP vs BiPAP: 3CPO trial found no mortality difference; either is acceptable — start with whichever is available
Cochrane meta-analysis: NIV reduces mortality (NNT ~13) and need for intubation vs standard O2 therapy
Wean when: SpO2 >92% on FiO2 <0.4, RR <25, comfortable, ABG normalized — typically after 2-6 hours
Contraindications: cardiac arrest, GCS <8, facial trauma, inability to protect airway, severe agitation, vomiting

[2] [7]

3CPO (Gray et al, NEJM 2008)

Multicentre RCT: standard O2 vs CPAP vs NIV (BiPAP)

Population: 1069 patients with severe acute cardiogenic pulmonary oedema

Key finding

No significant difference in 7-day mortality between groups (~9.5% all arms). NIV groups had faster improvement in symptoms, HR, BP, and acidosis. Reduced need for intubation/escalation.

[7]

Diuretic therapy — DOSE-AHF and practical strategy

Diuretic strategy in ADHF

Initial IV bolus dose

Total daily IV furosemide = 1-2.5x the patient usual total daily oral dose. New-onset HF: furosemide 40-80 mg IV. Chronic HF on 80 mg PO daily: give 80-200 mg IV. Maximum single bolus ~200 mg furosemide (or 5 mg torsemide/bumetanide).

Assess response at 2-6 hours

Target urine output 100-150 mL/h (3-4 L/24h) for effective decongestion. If inadequate (UO <100 mL/h after 2h), double the dose.

Bolus vs continuous infusion

DOSE-AHF: no difference in symptoms or renal function between q12h bolus and continuous infusion. Either strategy is acceptable; boluses allow titration to response.

High-dose vs low-dose

DOSE-AHF: high-dose (2.5x oral) achieved greater fluid loss and weight reduction vs low-dose (1x oral), with transiently worse renal function but no difference in outcomes. Favour higher doses in chronic HF.

Add thiazide if diuretic resistance

If response inadequate after stepwise doubling, add oral metolazone 2.5-5 mg daily OR IV chlorothiazide 500-1000 mg (sequential nephron blockade). Caution: hypokalaemia, hyponatraemia, severe volume depletion.

Add potassium-sparing agent

Spironolactone 25-50 mg daily (also proven mortality benefit in chronic HFrEF). Monitor K+ closely when combined with ACEi/ARB/ARNI.

[6]

DOSE-AHF (Felker et al, NEJM 2011)

2x2 factorial RCT: (a) high-dose (2.5x oral) vs low-dose (1x oral) furosemide; (b) bolus q12h vs continuous infusion

Population: 308 patients with acute decompensated heart failure and ≥1 sign of congestion

Key finding

No difference in primary endpoint (global symptom assessment at 72h) or 60-day death/rehospitalisation between any arms. High-dose produced greater net fluid loss and weight loss but transiently more renal dysfunction.

[6]

Vasodilators — when and what

Nitroglycerin (GTN)

First-line if SBP >110

Mechanism: venodilation (↓ preload) at low dose; arteriolar dilation (↓ afterload) at higher doses
Dose: sublingual spray 400 mcg (1-2 puffs); IV infusion 10-200 mcg/min titrated
Indications: AHF with SBP >110 and significant congestion/pulmonary oedema, concomitant ischaemia, or acute MR
Cautions: avoid if SBP <110, right ventricular infarction, PDE5 inhibitor in last 24-48h, severe aortic stenosis
Tachyphylaxis after 16-24h — useful for the first 24h of acute decompensation

Nesiritide (BNP)

No mortality benefit

Recombinant B-type natriuretic peptide — vasodilator, natriuretic, RAAS inhibitor
ASCEND-HF (OConnor 2011): modest dyspnoea improvement, NO mortality or rehospitalisation benefit, no excess renal harm
Not routinely used in ANZ/Europe; largely abandoned due to cost and lack of hard outcomes benefit

Sodium nitroprusside

Severe hypertension / valve regurg

Potent arterial + venous vasodilator; rapid onset/offset
Indications: hypertensive emergency with severe pulmonary oedema, acute severe mitral/aortic regurgitation
Dose: 0.3-3 mcg/kg/min IV; requires arterial line monitoring
Caution: cyanide/thiocyanate toxicity with prolonged use (>24-48h) or renal failure; raises ICP

ACEi / ARB / ARNI

Start AFTER stabilisation

NOT for acute phase — risk of hypotension and AKI in volume-depleted/hypotensive patient
Initiate once euvolaemic, SBP stable, renal function stable (usually 24-48h post-admission)
ARNI (sacubitril/valsartan) preferred over ACEi in HFrEF (PARADIGM-HF mortality benefit)

[1]

ASCEND-HF (O'Connor et al, NEJM 2011)

RCT: nesiritide (recombinant BNP) vs placebo

Population: 7141 patients with acute decompensated heart failure

Key finding

Nesiritide did not improve 30-day death/rehospitalisation or mortality. Small, non-clinically-significant improvement in dyspnoea. No excess renal dysfunction or mortality (addressing earlier safety concerns).

[13]

Inotropes for the cold/shocked patient

Inotropes are NOT indicated in warm-wet AHF — they add arrhythmia, ischaemia, and tachyphylaxis risk without benefit. Reserve for cold-wet or cold-dry profiles (hypoperfusion) where augmentation of cardiac output is required pending definitive therapy.[20]

Dobutamine

β1 agonist

Dose: 2-20 mcg/kg/min IV
↑ cAMP → ↑ contractility and HR (chronotropy); mild β2 vasodilation
First-line inotrope in most units; cheap, familiar, titratable
Side effects: tachyarrhythmia, ↑ myocardial O2 demand, hypotension (β2), tolerance

Milrinone

PDE-3 inhibitor

Dose: 0.125-0.75 mcg/kg/min IV (loading dose often omitted in AHF)
↑ cAMP via PDE-3 inhibition → ↑ contractility + vasodilation (pulmonary and systemic)
Less tachyarrhythmia than dobutamine; favourable in pulmonary hypertension/RV failure
Caution: renally cleared — halve dose in CKD; significant hypotension on initiation
OPTIME-CHF: routine milrinone in ADHF did NOT improve outcomes and increased atrial arrhythmias/hypotension

Levosimendan

Calcium sensitiser

Binds cardiac troponin C → ↑ myofilament Ca2+ sensitivity without ↑ intracellular Ca2+ (less arrhythmia/ischaemia)
Also opens ATP-sensitive K+ channels → vasodilation (including coronary)
SURVIVE: no mortality benefit vs dobutamine in ADHF (subgroup signal in prior β-blocker use)
Used in some European units; not available in Australia; expensive

Noradrenaline (norepinephrine)

First-line vasopressor

α1 + β1 agonist → vasoconstriction + modest inotropy
Use when MAP <65 despite inotrope, or as first-line vasoconstrictor in shock
Preferred over adrenaline (epinephrine) — less lactataemia and arrhythmia
Dose: 0.05-1 mcg/kg/min IV titrated to MAP ≥65 mmHg

Adrenaline (epinephrine)

Refractory shock

α + β agonist; potent inotrope and vasoconstrictor
Reserved for refractory shock not responding to noradrenaline + inotrope
Causes lactataemia (β2 glycolysis), arrhythmia, worsening renal/splanchnic perfusion

Dopamine

Largely abandoned

Dose-dependent: low dose "renal" (1-3 mcg/kg/min) — ROSE-AHF showed NO renal benefit
Higher doses: tachyarrhythmia excess vs noradrenaline (SOAP-II)
Not recommended for routine use in AHF

[20]

ROSE-AHF (Chen et al, JAMA 2013)

2x2 factorial RCT: low-dose dopamine (2 mcg/kg/min) vs placebo, AND low-dose nesiritide vs placebo

Population: 360 patients with acute heart failure and renal dysfunction (eGFR 15-60)

Key finding

Neither low-dose dopamine nor low-dose nesiritide improved the primary endpoints of 72-hour cumulative urine output or cystatin C change. No renal protection.

[8]

SURVIVE (Mebazaa et al, JAMA 2007)

RCT: levosimendan vs dobutamine

Population: 1327 patients with AHF requiring inotropic support

Key finding

No significant difference in primary endpoint (180-day all-cause mortality: 26% vs 28%). Subgroup suggested possible benefit of levosimendan in patients with prior β-blocker therapy.

[12]

OPTIME-CHF (Cuffe et al, JAMA 2002)

RCT: 48-72h infusion of milrinone vs placebo (not selected by clinician for inotrope indication)

Population: 949 patients with acute exacerbation of chronic systolic heart failure

Key finding

No improvement in 60-day hospitalisation/mortality. Increased adverse events: atrial arrhythmias (HR 1.7), sustained hypotension (HR 1.6). Subgroup trend to worse outcomes in ischaemic cardiomyopathy.

[17]

Mechanical circulatory support (MCS)

For refractory cardiogenic shock despite inotropes/vasopressors, MCS bridges to recovery, decision, durable device, or transplant. [1]

IABP (intra-aortic balloon pump)

Counterpulsation

Mechanism: balloon inflates in diastole (↑ coronary perfusion), deflates in systole (↓ afterload)
Indications: historically first-line; IABP-SHOCK II showed NO mortality benefit in MI-shock
Contraindications: aortic regurgitation, aortic dissection, severe PAD, severe sepsis
Now reserved for mechanical complications of MI (acute MR, VSD) or as bridge to definitive support

Impella (axial flow micro-pump)

LV / RV support

Mechanism: transvalvular axial pump aspirates LV blood, ejects into aorta (LV unloading)
Impella 2.5 (2.5 L/min), CP (4.0 L/min), 5.0 (5.0 L/min); RP for RV support
Active unloading reduces LV wall stress and improves forward flow more than IABP
Complications: limb ischaemia, haemolysis, device malposition, bleeding at insertion site

VA-ECMO

Venoarterial ECMO

Mechanism: drains RA blood, returns oxygenated blood to femoral artery — supports both circulation and gas exchange
Indications: refractory cardiogenic shock, massive PE, fulminant myocarditis, post-cardiotomy, severe ARDS + shock
Flows 3-6 L/min; provides complete cardiopulmonary support
Complications: differential hypoxaemia (Harlequin syndrome — femoral configuration), LV distension (increased afterload — consider Impella or IABP venting), limb ischaemia, bleeding, stroke, haemolysis
Femoral vs central cannulation: central preferred post-cardiotomy; femoral for rapid percutaneous insertion

TandemHeart

Atrial-to-femoral LA→femoral artery

Drains left atrium via transseptal puncture, returns to femoral artery
Flows up to 5 L/min; potent LV unloading
Requires transseptal expertise — less commonly used than Impella/ECMO

IABP-SHOCK II (Thiele et al, NEJM 2012)

RCT: IABP vs no IABP

Population: 600 patients with acute MI and cardiogenic shock undergoing early revascularisation

Key finding

No difference in 30-day mortality (39.7% vs 41.3%) or 12-month mortality. No improvement in secondary endpoints (bleeding, renal function, inflammatory markers, ICU stay).

[15]

CULPRIT-SHOCK (Thiele et al, NEJM 2017)

RCT: culprit-lesion-only PCI vs immediate multivessel PCI

Population: 555 patients with acute MI and multivessel disease presenting in cardiogenic shock

Key finding

Culprit-only PCI reduced primary endpoint (death or severe renal failure requiring RRT at 30 days: 45.9% vs 55.4%) and reduced mortality (43.3% vs 51.6%).

[16]

Cardiogenic shock (SBP <90)

See dedicated topic: cardiogenic-shock-mechanical-support. Key points:

Cold + wet = cardiogenic shock → different pathway
Inotropes: dobutamine or milrinone (improve cardiac output)
Vasopressor: noradrenaline (maintain MAP >65)
Mechanical support: IABP, Impella, VA-ECMO
Urgent revascularisation if ACS (CULPRIT-SHOCK: culprit-only PCI)[16]
Diurese cautiously (may worsen hypotension)

The SCAI shock stages provide a granular framework: A (at risk), B (beginning — clinical suspicion), C (classic — hypoperfusion meeting criteria), D (deteriorating — worsening despite support), E (extremis — refractory cardiac arrest). Early MCS escalation at stage C/D (before lactate >10, before cardiac arrest) is associated with markedly better outcomes than late "salvage" escalation at stage E. [1]

Diuretic resistance

Defined as failure to achieve euvolaemia despite adequate diuretic dosing. Common in chronic HF, CKD, hypoalbuminaemia, and high sodium intake. [1]

Approach to diuretic resistance

Confirm adherence and intake

Review medication history (adherence, stopped diuretics, NSAIDs). Restrict sodium to <2 g/day and fluid to 1-1.5 L/day.

Stepwise IV furosemide up-titration

Start with 1-2.5x oral dose; reassess UO at 2h. Double each subsequent dose until UO >100 mL/h or 200 mg single bolus reached.

Switch to continuous infusion

May improve efficacy once high doses are needed; titrate infusion to UO 100-150 mL/h.

Add thiazide-type diuretic (sequential nephron blockade)

Oral metolazone 2.5-10 mg daily, IV chlorothiazide 500-1000 mg, or IV hydrochlorothiazide. Acts at distal tubule; synergistic with loop diuretic. Major risk: hypokalaemia, hyponatraemia, severe volume depletion — monitor 6-hourly.

Add acetazolamide

Carbonic anhydrase inhibitor proximal tubule — synergistic in metabolic alkalosis with diuretic resistance (ADVOR trial 2022 showed improved decongestion).

Consider ultrafiltration (rarely)

Reserved for refractory overload NOT responsive to pharmacological therapy. CARRESS-HF showed pharmacologic therapy superior to ultrafiltration.

Address hypoalbuminaemia / nephrotic

In low-albumin states, diuretics enter tubule poorly; consider albumin-furosemide co-infusion (controversial).

Reassess for occult shock / over-diuresis

A "non-responder" may be cold-dry (underfilled) — over-diuresis causes AKI without relieving symptoms. Bedside echo/IVC assessment guides therapy.

[6]

CARRESS-HF (Bart et al, NEJM 2012)

RCT: stepped pharmacological therapy (high-dose IV loop + thiazide + dopamine/nitroglycerin) vs ultrafiltration

Population: 188 patients with ADHF, worsened renal function, and persistent congestion

Key finding

Pharmacological therapy was superior: less weight gain (primary endpoint: -1.7 kg worse in UF arm), fewer serious adverse events (72% in UF vs 57% in pharmacologic), and fewer creatinine rises.

[9]

UNLOAD (Costanzo et al, JACC 2007)

RCT: ultrafiltration vs standard IV diuretic therapy

Population: 200 patients hospitalised with ADHF

Key finding

Ultrafiltration produced greater weight loss and net fluid loss at 48h and fewer 90-day rehospitalisations. No difference in symptoms or renal function. Significant industry sponsorship bias.

[14] [21]

Failed disease-modifying therapies — a recurring pattern

Multiple promising AHF therapies have failed in large RCTs — exam candidates should know these to avoid reaching for unproven agents. [1]

RELAX-AHF (Teerlink et al, Lancet 2013)

RCT: serelaxin (recombinant human relaxin-2) vs placebo for 48h

Population: 1161 patients with acute heart failure, dyspnoea, mild-moderate renal dysfunction, SBP >125

Key finding

Serelaxin improved dyspnoea (visual analogue scale) and reduced 180-day mortality in the original trial (a secondary endpoint). Subsequent RELAX-AHF-2 (2017) FAILED to confirm the mortality signal.

[10]

VERITAS (McMurray et al, JAMA 2007)

RCT: tezosentan (endothelin receptor antagonist) vs placebo

Population: 1435 patients with acute heart failure

Key finding

No improvement in symptoms or 30-day death/heart-failure worsening. Increased adverse events (headache, hypotension).

[11]

The recurring lesson: decongestion with diuretics + NIV + addressing the precipitant remains the cornerstone; no acute medical therapy (vasodilator, natriuretic peptide, inodilator, endothelin antagonist, relaxin) has convincingly modified AHF mortality. Disease modification happens through chronic GDMT, not acute interventions. [1]

Guideline-directed medical therapy (GDMT) and the 4 pillars

The four pillars of HFrEF therapy, ideally initiated/up-titrated before discharge: [1]

Beta-blocker (bisoprolol, carvedilol, metoprolol succinate, nebivolol) — start once euvolaemic and stable (NOT in acute decompensation)
MRA (spironolactone/eplerenone) — K+ <5.0, eGFR >30
ARNI (sacubitril/valsartan) preferred over ACEi/ARB (PARADIGM-HF); SGLT2 inhibitor (dapagliflozin, empagliflozin) — both HFrEF and HFpEF [1]

STRONG-HF (Mebazaa et al, Lancet 2022)

RCT: intensive (rapid up-titration of GDMT to target doses within 2 weeks pre-discharge and at 2-week follow-up) vs usual care

Population: 1078 patients hospitalised with acute heart failure

Key finding

Intensive up-titration reduced 180-day all-cause death or heart failure readmission (15.2% vs 23.3%, HR 0.64). No excess serious adverse events despite rapid titration.

[19]

DAPA-HF (McMurray et al, NEJM 2019)

RCT: dapagliflozin 10 mg daily vs placebo, added to standard GDMT

Population: 4744 patients with HFrEF (EF ≤40%), NYHA II-IV

Key finding

Reduced primary composite (worsening HF or CV death) by 26% (HR 0.74); reduced CV death and HF hospitalisation; reduced all-cause mortality. Benefit across all subgroups including diabetic and non-diabetic.

[18]

Discharge criteria

Discharge preparedness is as important as acute stabilisation — premature discharge drives the 25% 30-day readmission rate. The patient should meet ALL of the following: [1]

Clinical — congestion resolved

Required

Euvolaemic: no orthopnoea, JVP not raised, no new crackles, no ascites
Stable dry weight for ≥24h (target 2 kg below admission for chronic HF)
No IV diuretics required — stable on oral diuretic regimen
Off IV vasoactive drugs for ≥24h; SBP stable on oral medications
Symptom improvement: NYHA improved by ≥1 class; patient ambulating without dyspnoea

Investigations — stable

Required

Renal function stable/improving (no ongoing rise in creatinine; K+ 4.0-5.0 mmol/L)
Haemodynamics stable (HR <100, SBP >100, no orthostatic drop)
No evidence of ongoing ischaemia (troponin falling, pain-free)
Echo performed and LV/RV/valve function documented

Medications — GDMT optimised

Required

On at least low dose of each eligible GDMT pillar (BB + MRA + ARNI/ACEi + SGLT2i)
Stable oral diuretic dose defined
NSAIDs, nephrotoxins, and contraindicated drugs stopped
Vaccinations offered (influenza, pneumococcal, COVID-19)

Education and follow-up

Required

Patient taught daily weights (call if >2 kg in 3 days or 2.5 kg in a week)
Sodium (≤2 g/day) and fluid (1.5-2 L/day) restriction counselled
Medication adherence plan; device therapy (ICD/CRT) referral if eligible
Cardiology / HF clinic follow-up booked within 7-14 days
GP follow-up within 1 week
Cardiac rehabilitation referral

[1] [19]

SaqBlocks — fellowship exam practice

SAQ — Acute cardiogenic pulmonary oedema in a warm-wet HFrEF patient

10 minutes · 10 marks

A 72-year-old man with known HFrEF (ejection fraction 30 per cent) on bisoprolol 5 mg, ramipril 5 mg, spironolactone 25 mg and furosemide 80 mg daily presents with two hours of acute severe breathlessness. He is sitting upright, distressed, RR 34, SpO2 84 per cent on room air, BP 168/95, HR 116 (atrial fibrillation), pink frothy sputum, bilateral coarse crackles to the apices, marked accessory-muscle use, JVP raised to the angle of the jaw. ECG shows fast atrial fibrillation with no ischaemic changes. Chest X-ray: bat-wing perihilar infiltrates, Kerley B lines and upper-lobe blood diversion.

[1]

SAQ — Cardiogenic shock complicating anterior STEMI

10 minutes · 10 marks

A 58-year-old woman presents with six hours of central crushing chest pain and progressive breathlessness. She is diaphoretic, cold and clammy. BP 78/48 (MAP 58), HR 124 (sinus), RR 32, SpO2 90 per cent on 15 L via non-rebreather, JVP raised, bilateral crackles, urine output 15 mL/h, GCS 14. Lactate 5.2 mmol/L, creatinine 170 (baseline 80). ECG shows 3 mm ST elevation in V1-V4 with reciprocal depression inferiorly. Bedside echo: severe LV dysfunction with anterior and septal akinesis, estimated ejection fraction 25 per cent, no pericardial effusion.

[1]

Clinical pearls

High-yield ADHF points for the CICM/FFICM exam

CPAP/BiPAP reduces mortality in cardiogenic pulmonary oedema — use EARLY.[2]
IV furosemide: 1-2.5x usual oral dose. Mechanism: initial venodilation, then diuresis.[6]
Nitroglycerin if SBP >110 (venodilation reduces preload). NOT if hypotensive.
Do NOT give beta-blockers acutely (negative inotropy — may worsen shock).
SBP <90 = cardiogenic shock → inotropes/mechanical support (different pathway).
Forrester classification: warm/cold (perfusion) x wet/dry (congestion).
NT-proBNP: >300 excludes HF, >1800 confirms (age-adjusted thresholds).
Flash pulmonary oedema: consider bilateral renal artery stenosis (Pickering syndrome).
ACE inhibitor/ARB: start once stable and euvolaemic (not acutely).
ACS is #1 precipitant — check ECG and troponin in ALL patients.
Morphine: use cautiously (associated with worse outcomes in observational studies).
HFrEF vs HFpEF: echo differentiates (ejection fraction <40% = HFrEF).
NSAIDs worsen HF — stop all NSAIDs. Sodium retention + vasoconstriction.
Atrial fibrillation: rate control urgently (digoxin, beta-blocker once stable, amiodarone).

Advanced pearls — trials, traps, and exam distinction

Stevenson vs Forrester: Stevenson uses clinical signs (warm/cold × wet/dry); Forrester uses invasive CI and PCWP. Same concept, different measurement.[5]
DOSE-AHF: no difference between bolus vs continuous infusion OR high vs low dose for hard outcomes — titrate to urine output (target 100-150 mL/h).[6]
3CPO: no mortality difference between CPAP and BiPAP — use whichever is available; both beat standard O2.[7]
CARRESS-HF: pharmacologic stepped therapy BEATS ultrafiltration for cardiorenal ADHF. Ultrafiltration is not first-line.[9]
IABP-SHOCK II: IABP provides no benefit in MI-shock — guideline class III. Use Impella/VA-ECMO for refractory shock.[15]
CULPRIT-SHOCK: culprit-only PCI beats multivessel PCI in MI-shock.[16]
SURVIVE & OPTIME-CHF: routine inotropes (levosimendan, milrinone) do NOT improve outcomes in warm-wet ADHF — use ONLY in cold profiles.[12][17]
ROSE-AHF: low-dose "renal" dopamine does NOT protect the kidney in ADHF — abandon it.[8]
STRONG-HF: rapid GDMT up-titration before/at discharge is SAFE and reduces readmission.[19]
SGLT2 inhibitors are now the 4th pillar of HFrEF — start during/after admission, not just for diabetics.[18]
Cardiorenal syndrome is driven by venous congestion, not low output — decongestion (not inotropes or dopamine) is the renal-sparing therapy.
Continue pre-existing beta-blocker if the patient is stable (warm-wet, not in shock); holding it acutely worsens outcomes. STOP only if hypoperfused/cold.
Over-diuresis is a common iatrogenic trap: rising creatinine with cleared congestion may mean underfilling (cold-dry), not diuretic failure — assess volume status with echo/IVC before escalating diuretics.
BNP falls with decongestion — a ≥30% fall in NT-proBNP from admission to discharge is associated with lower readmission (useful discharge readiness marker).
Thiazide + loop = sequential nephron blockade — powerful but dangerous (monitor K+, Na+ q6h); reserve for true resistance.
Echo before inotropes: in apparent cardiogenic shock, confirm structurally — massive PE, tamponade, and severe valvular lesions mimic shock but need different therapy.
The "decongestion paradox": serum creatinine often rises 0.1-0.3 mg/dL with effective decongestion (haemoconcentration) — this is NOT a reason to stop diuresis if the patient remains congested.
Iron deficiency is present in 50% of HF patients and worsens symptoms independent of anaemia — IV ferric carboxymaltose (AFFIRM-AHF) reduces HF rehospitalisation. Check ferritin + TSAT in every admission.

Pharmacology pearls — doses, mechanisms, and exam pitfalls

Furosemide initial venodilation within 5-10 min (prostaglandin-mediated) precedes diuresis — gives early symptom relief before urine flows.
GTN infusion is tachyphylactic after 16-24h; useful acutely, not for prolonged courses.
Dobutamine tolerance develops within 24-48h via β-receptor downregulation — don't leave patients on it indefinitely; bridge to MCS or recovery.
Milrinone is renally cleared — halve the dose if eGFR <30; significant first-dose hypotension (start low, no bolus in AHF).
Levosimendan works downstream of the β-receptor — favoured (where available) in patients already β-blocked who need inotropy.
Noradrenaline (not adrenaline) is first-line vasopressor in cardiogenic shock — less lactataemia, less arrhythmia.
Adrenaline causes β2-driven lactataemia that confounds lactate-guided resuscitation — avoid as first-line shock pressor.
Spironolactone hyperkalaemia risk is highest with concomitant ACEi/ARB/ARNI and CKD — check K+ within 3 days and weekly.
Sacubitril/valsartan must not be started within 36h of ACEi (angioedema risk); switch ACEi → ARNI as a clean interval.
SGLT2 inhibitors cause genital mycotic infections and euglycaemic DKA (rare); transient eGFR dip in first 2 weeks is expected and benign.
Ivabradine (If channel inhibitor) reduces HR without lowering BP — for HFrEF in sinus rhythm with HR >75 despite max-tolerated β-blocker.
Digoxin has no mortality benefit but reduces hospitalisations; useful for rate control in AF with low-output HF (doesn't drop BP). Narrow therapeutic index — toxicity worsens in hypokalaemia (common with loop diuretics).
Hydralazine + nitrate is the evidence-based alternative for HFrEF in African-American patients or those intolerant of ACEi/ARNI (A-HeFT).
Vericiguat (soluble guanylate cyclase stimulator) reduces HF hospitalisation in high-risk chronic HF (VICTORIA) — newer addition to the armamentarium.

[1]

Investigation and monitoring pearls

Lung ultrasound B-lines track decongestion in real time — more sensitive than CXR for interstitial oedema and bedside-friendly.
IVC collapse <50% in a spontaneously breathing patient suggests RA pressure >10 mmHg (congested); >50% collapse with small IVC suggests volume responsiveness.
Mixed venous saturation (SvO2) <60% on a central line suggests low cardiac output — supports inotrope use even if SBP preserved.
Central venous O2 from a ScvO2 (superior vena cava) sample approximates SvO2 with a ~5% offset — usable at the bedside.
Pulse pressure narrows in low-output states (PP/ systolic <25% suggests severe low output) — a quick clinical marker.
Cold extremities, mottling, capillary refill >3s are bedside markers of hypoperfusion independent of BP — use to detect cold-wet/cold-dry.
Troponin elevation is common in ADHF even without ACS (type 2 from wall stress) — only treat as ACS if dynamic rise/fall or ischaemic features.
NT-proBNP <300 ng/L effectively excludes AHF at the emergency presentation — high negative predictive value.
Bilateral pleural effusions that are transudative (LDH low, protein <25 g/L) support a cardiac cause; unilateral raises suspicion of alternative diagnosis.
Hepatic congestion (raised bilirubin, transaminases, PT) from passive venous congestion mimics acute hepatitis and reverses with decongestion.
Spot urine sodium <30 mmol/L in a fluid-restricted patient on loop diuretic suggests high sodium avidity (RAAS activation) — may need higher diuretic doses.
Body weight is the cheapest, most underused monitoring tool — daily weights (same scale, same time, same clothes) guide diuretic titration and catch early fluid retention pre-discharge.

[1]

Red flags

Traps and pitfalls

Routine inotropes in warm-wet ADHF are harmful (OPTIME-CHF) — they add arrhythmia and ischaemia without benefit. Reserve for cold/hypoperfused profiles.[17]
Ultrafiltration is NOT first-line for cardiorenal ADHF — CARRESS-HF showed stepped pharmacologic therapy is superior.[9]
IABP has no role in routine MI-shock (IABP-SHOCK II, ESC class III) — escalate to Impella/VA-ECMO for refractory cases.[15]
Low-dose "renal" dopamine does not protect the kidney in ADHF (ROSE-AHF) — abandon the practice.[8]
Rising creatinine during decongestion may be haemoconcentration (good), not AKI — distinguish by clinical context and stop only if creatinine rise is severe or symptoms of underperfusion appear.
Continuing pre-existing beta-blocker in stable warm-wet ADHF is generally correct — STOP only if cold/hypoperfused.
CULPRIT-SHOCK: do NOT perform multivessel PCI in MI-shock — culprit-only is safer.[16]
VA-ECMO can worsen LV distension and pulmonary oedema (increased afterload) — consider concurrent LV venting (Impella or IABP) in severe LV failure.
ACEi → ARNI switch needs 36h washout to avoid angioedema.
Don't discharge until GDMT optimised — STRONG-HF shows rapid up-titration before discharge is safe and reduces readmission.[19]

When to escalate / call for help

Cold-wet patient not improving on inotrope within 1-2h → call MCS/ECMO team EARLY (before lactate >10 or arrest).
Recurrent ventricular arrhythmia in AHF → exclude ischaemia, electrolyte disturbance (K+, Mg2+), and inotrope toxicity.
New murmur + flash pulmonary oedema → echocardiogram urgently for acute MR (papillary muscle rupture), VSD, or aortic regurgitation — surgical emergency.
Bilateral RAS suspected → arrange renal artery Doppler/CTA; do NOT give ACEi until excluded.
Right heart failure disproportionate to LV failure → consider massive PE, RV infarct (inferior MI), or severe pulmonary hypertension — different therapy.
Postpartum dyspnoea + HF → peripartum cardiomyopathy; bromocriptine + standard HF therapy.
Young patient with new HF → screen for myocarditis (viral serology, CMR, sometimes biopsy), cocaine, thyrotoxicosis, autoimmune.

[1]

References

[1]McDonagh TA, Metra M, Adamo M, et al. 2021 ESC Guidelines for the diagnosis and treatment of acute and chronic heart failure Eur Heart J, 2021.PMID 34447992
[2]Vital FMR, Ladeira MT, Atallah AN. Non-invasive positive pressure ventilation (CPAP or bilevel NPPV) for cardiogenic pulmonary oedema Cochrane Database Syst Rev, 2013.PMID 23728654
[3]Chioncel O, Mebazaa A, Harjola VP, et al. Acute heart failure congestion and perfusion status - impact of the clinical classification on in-hospital and long-term outcomes; insights from the ESC-EORP-HFA Heart Failure Long-Term Registry Eur J Heart Fail, 2019.PMID 31127678
[4]Seferovic PM, Ponikowski P, Anker SD, et al. Clinical practice update on heart failure 2019: pharmacotherapy, procedures, devices and patient management. An expert consensus meeting report of the Heart Failure Association of the European Society of Cardiology Eur J Heart Fail, 2019.PMID 31129923
[5]Nohria A, Tsang SW, Fang JC, et al. Clinical assessment identifies hemodynamic profiles that predict outcomes in patients admitted with heart failure J Am Coll Cardiol, 2003.PMID 12767667
[6]Felker GM, Lee KL, Bull DA, et al. Diuretic strategies in patients with acute decompensated heart failure N Engl J Med, 2011.PMID 21366472
[7]Gray A, Goodacre S, Newby DE, Masson M, Sampson F, Nicholl J (3CPO Trialists). Noninvasive ventilation in acute cardiogenic pulmonary edema N Engl J Med, 2008.PMID 18614781
[8]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.PMID 24247300
[9]Bart BA, Goldsmith SR, Lee KL, et al. Ultrafiltration in decompensated heart failure with cardiorenal syndrome N Engl J Med, 2012.PMID 23131078
[10]Teerlink JR, Cotter G, Davison BA, et al. Serelaxin, recombinant human relaxin-2, for treatment of acute heart failure (RELAX-AHF): a randomised, placebo-controlled trial Lancet, 2013.PMID 23141816
[11]McMurray JJ, Teerlink JR, Cotter G, et al. Effects of tezosentan on symptoms and clinical outcomes in patients with acute heart failure: the VERITAS randomized controlled trials JAMA, 2007.PMID 17986694
[12]Mebazaa A, Nieminen MS, Packer M, et al. Levosimendan vs dobutamine for patients with acute decompensated heart failure: the SURVIVE Randomized Trial JAMA, 2007.PMID 17473298
[13]O'Connor CM, Starling RC, Hernandez AF, et al. Effect of nesiritide in patients with acute decompensated heart failure N Engl J Med, 2011.PMID 21732835
[14]Costanzo MR, Guglin ME, Saltzberg MT, et al. Ultrafiltration versus intravenous diuretics for patients hospitalized for acute decompensated heart failure J Am Coll Cardiol, 2007.PMID 17291932
[15]Thiele H, Zeymer U, Neumann FJ, et al. Intraaortic balloon support for myocardial infarction with cardiogenic shock N Engl J Med, 2012.PMID 22920912
[16]Thiele H, Akin I, Sandri M, et al. PCI Strategies in Patients with Acute Myocardial Infarction and Cardiogenic Shock N Engl J Med, 2017.PMID 29083953
[17]Cuffe MS, Califf RM, Adams KF Jr, et al. Short-term intravenous milrinone for acute exacerbation of chronic heart failure: a randomized controlled trial JAMA, 2002.PMID 11911756
[18]McMurray JJV, Solomon SD, Inzucchi SE, et al. Dapagliflozin in Patients with Heart Failure and Reduced Ejection Fraction N Engl J Med, 2019.PMID 31535829
[19]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.PMID 36356631
[20]Scheeren TWL, Wicentowski M, Gan TJ, et al. Current use of inotropes in circulatory shock Ann Intensive Care, 2021.PMID 33512597
[21]Srivastava M, Safi L, Pipil A, et al. Ultrafiltration for acute heart failure Cochrane Database Syst Rev, 2022.PMID 35061249

[1]

Epidemiology and pathophysiology

The pathophysiology is a vicious cycle of decompensation: [1]

Index insult (myocardial ischaemia, arrhythmia, volume load) reduces stroke volume.
Fall in cardiac output activates the sympathetic nervous system and renin-angiotensin-aldosterone system (RAAS) in a compensatory attempt to maintain perfusion pressure.
Maladaptive neurohormonal activation drives sodium/water retention (aldosterone, ADH), vasoconstriction (angiotensin II, noradrenaline, endothelin), tachycardia, and direct cardiotoxicity.
Increased preload and afterload raise ventricular wall stress → further reduction in stroke volume, mitral/tricuspid regurgitation, and myocardial injury (troponin leak from subendocardial ischaemia due to elevated end-diastolic pressure).
Venous congestion (raised left-sided filling pressures) transudates fluid into the pulmonary interstitium and alveoli → pulmonary oedema and hypoxaemia, which worsens myocardial ischaemia.
Right heart distension (ventricular interdependence) impairs LV filling and coronary perfusion.
Cardiorenal syndrome develops as venous congestion (not low output per se) reduces glomerular filtration, creating a self-perpetuating cycle of worsening renal function, more diuretic resistance, and escalating congestion. [1]

Clinical classification — Stevenson hemodynamic profiles

Profile	Congestion	Perfusion	Clinical findings	Treatment	Mortality signal
Warm-wet	Wet (JVP raised, pulmonary crackles, oedema)	Warm (well perfused, SBP >90, normal lactate)	Most common presentation (~70%). Pulmonary oedema ± peripheral oedema, normal extremities, adequate urine output	IV loop diuretics (1-2.5x oral dose) ± vasodilator (GTN if SBP >110) ± CPAP/BiPAP	Moderate
Warm-dry	Dry (no JVP rise, clear chest)	Warm	Euvolaemic, well perfused — the target state after treatment, or a non-congested decompensation (e.g. isolated high-output state)	Continue/optimise oral GDMT; identify non-volume precipitant	Low
Cold-wet	Wet	Cold (clammy, mottled, oliguric, confused, SBP <90 or lactate >2)	Cardiogenic shock with congestion. SBP often <90, cold extremities, oliguria, altered mentation	Inotrope (dobutamine/milrinone/levosimendan) ± vasopressor (noradrenaline to MAP ≥65) + decongest; consider mechanical support (Impella/VA-ECMO) early	High
Cold-dry	Dry	Cold	Low output without congestion — over-diuresed, or pure forward failure. Often pre-load dependent	Cautious fluid challenge (250 mL crystalloid) ± inotrope; look for occult bleeding/sepsis	Moderate-high

Warm-wet (most common)

~70% of AHF

Adequate perfusion preserved (SBP >90, warm extremities, lactate <2 mmol/L)
Congestion dominates symptoms: orthopnoea, crackles, raised JVP, peripheral oedema
Therapy target: decongest with IV loop diuretic ± vasodilator; CPAP if pulmonary oedema
Avoid inotropes (no hypoperfusion — they add arrhythmia/ischaemia risk without benefit)
Beta-blocker continuation: continue if chronic stable; hold if decompensated

Cold-wet (cardiogenic shock)

~5-8% of AHF

Hypoperfusion + congestion: SBP <90, cold/clammy, oliguria <0.5 mL/kg/h, lactate >2
Different pathway: do NOT rely on diuretics alone — they fail in low-output state
Inotrope first-line: dobutamine, milrinone or levosimendan to augment cardiac output
Vasopressor (noradrenaline) if MAP <65 despite inotrope
Escalate EARLY to MCS (Impella/VA-ECMO) — late escalation has poor outcomes
Revascularise if ACS (CULPRIT-SHOCK: culprit-only PCI)

Cold-dry (over-diuresed / forward failure)

Rare

Hypoperfusion without congestion: tachycardia, narrow pulse pressure, oliguria
Often iatrogenic (over-diuresis) or occult hypovolaemia (bleeding, sepsis)
Cautious fluid bolus 250 mL crystalloid, reassess
Bedside echo: small hypercontractile LV with kissing septum suggests underfilled
Inotrope only if confirmed low output not corrected by volume

Warm-dry (target / compensated)

Discharge-ready state

No congestion, normal perfusion — goal of therapy
May represent non-congested decompensation: anaemia, thyrotoxicosis, tachyarrhythmia
Transition to oral GDMT (4 pillars: BB + MRA + ARNI/ACEi + SGLT2i)
Discharge planning begins here

[3] [5]

Forrester classification (invasive hemodynamic)

Forrester classification — ADHF by cardiac index and PCWP

CI <2.2, PCWP <18 (cold + dry)

Mortality Moderate-high

Low output without congestion. Hypovolaemic or over-diuresed. Cautious fluid challenge.

[1]

Clinical presentation

Pulmonary oedema (left-sided)

Acute backward failure

Acute breathlessness, orthopnoea, paroxysmal nocturnal dyspnoea
Pink frothy sputum, bilateral coarse crackles, wheeze ("cardiac asthma")
Tachypnoea (RR >25), tachycardia, accessory muscle use
Hypoxia, often hypertensive initially (sympathetic surge)
CXR: bat-wing perihilar infiltrates, Kerley B lines, upper lobe blood diversion, pleural effusions

Cardiogenic shock (forward failure)

Cold-wet or cold-dry

Hypotension (SBP <90 or >30 below baseline), narrow pulse pressure
Cold clammy mottled skin, prolonged capillary refill
Oliguria (<0.5 mL/kg/h), altered mental status
Lactate >2 mmol/L, metabolic acidosis
Often complicating large MI, fulminant myocarditis, end-stage cardiomyopathy

Right heart failure

Isolated backward RV failure

Raised JVP, hepatomegaly, ascites, marked peripheral/sacral oedema
Relatively clear lung fields (no pulmonary oedema)
Causes: RV infarct (inferior MI), pulmonary embolism, severe TR/PR, pulmonary hypertension, ARVC
Kussmaul sign (JVP rises on inspiration) suggests severe RV failure or constrictive pericarditis

Flash pulmonary oedema

Sudden onset, recurrent

Explosive pulmonary oedema over minutes — bilateral renal artery stenosis (Pickering syndrome) until proven otherwise
Other causes: severe mitral regurgitation (flail leaflet), bilateral RAS, phaeochromocytoma, renal failure + volume load
Clue: AKI after starting ACE inhibitor → suspect bilateral RAS
Often markedly hypertensive on presentation

[1]

Identify the precipitant

Every AHF admission has a precipitant that must be actively sought and treated — decongestion alone without addressing the trigger leads to rapid readmission. [1]

Common precipitants

Find and treat

Acute coronary syndrome (#1 — check ECG, troponin)
Arrhythmia (especially new AF — rate control)
Infection (pneumonia, UTI — check cultures, CRP)
Medication non-adherence (stopped diuretics/ACEi)
Anaemia (transfuse if Hb <70)
Thyroid dysfunction (thyroid storm or myxoedema)
Renal failure (worsening AKI → fluid overload)
Dietary indiscretion (high salt intake)
Drugs (NSAIDs, calcium channel blockers, steroids — sodium retention)

Less common / high-yield for exam

Don't miss

Negative inotropes: diltiazem, verapamil, most antiarrhythmics
Cardiotoxic drugs: anthracyclines, trastuzumab, tyrosine kinase inhibitors
Alcohol-induced (holiday heart), cocaine, amphetamines
Peripartum cardiomyopathy (last month of pregnancy to 5 months postpartum)
Takotsubo (stress) cardiomyopathy — apical ballooning
Acute myocarditis (viral — coxsackie, parvovirus; giant cell)
Pulmonary embolism (sudden RV strain)
Endocarditis with acute valvular regurgitation
Tamponade / constrictive pericarditis
Tachycardia-mediated cardiomyopathy (persistent SVT/AF)

[1] [4]

Investigations

ADHF investigations

ECG

ACS, arrhythmia, LVH

Within 10 min

NT-proBNP

Diagnostic biomarker

>300 excludes, >1800 confirms

Echo

Assess LV function

HFrEF vs HFpEF, valve function

CXR

Pulmonary oedema

Bat-wing appearance, Kerley B lines

[1]

Bedside ultrasound and POCUS in AHF

Point-of-care ultrasound accelerates diagnosis and monitoring of decongestion: [1]

Lung ultrasound: B-lines (comet-tail artefacts) arising from pleural line indicate interstitial/alveolar fluid. Bilateral diffuse B-lines confirm cardiogenic pulmonary oedema; their resolution tracks successful diuresis. A-lines (horizontal reverberation) indicate dry lung.
Cardiac POCUS: identifies gross LV/RV dysfunction, pericardial effusion, gross valvular pathology, and estimates LV filling (E/e′). A small, hypercontractile, underfilled LV suggests over-diuresis (cold-dry); a dilated hypocontractile LV confirms cardiomyopathy.
IVC ultrasound: a dilated plethoric IVC (>2.1 cm with <50% collapse) supports high right atrial pressure and congestion; a small collapsible IVC suggests volume responsiveness.
** pleural effusion**: quantified and tapped for diagnostic/therapeutic thoracentesis. [1]

Management of cardiogenic pulmonary oedema

Acute pulmonary oedema management

Position + oxygen

Sit UPRIGHT (reduces preload, improves ventilation). High-flow oxygen via non-rebreather mask (15 L/min). Target SpO2 >92% (or >88% in COPD).

Non-invasive ventilation (CPAP/BiPAP) — EARLY

IV furosemide

Vasodilator (nitroglycerin) — if SBP >110

Morphine — USE CAUTIOUSLY

Do NOT give beta-blockers acutely

[1] [3]

Oxygen therapy detail

Indications and targets

Saturate judiciously

Target SpO2 92-96% in most patients; 88-92% in COPD retainers
Supplemental O2 only if hypoxaemic — routine O2 in non-hypoxaemic patients is NOT beneficial (and may be harmful via coronary vasoconstriction and ROS)
High-flow nasal cannula (HFNC) is an alternative to NIV in cooperative patients with moderate hypoxaemia
Reserve endotracheal intubation for: respiratory failure despite NIV, reduced GCS, inability to protect airway, exhaustion (rising CO2)

NIV / CPAP — the evidence

Mortality benefit

Mechanism: positive intrathoracic pressure → ↑FRC, ↓work of breathing, ↓LV preload and afterload, redistributes alveolar fluid
CPAP vs BiPAP: 3CPO trial found no mortality difference; either is acceptable — start with whichever is available
Cochrane meta-analysis: NIV reduces mortality (NNT ~13) and need for intubation vs standard O2 therapy
Wean when: SpO2 >92% on FiO2 <0.4, RR <25, comfortable, ABG normalized — typically after 2-6 hours
Contraindications: cardiac arrest, GCS <8, facial trauma, inability to protect airway, severe agitation, vomiting

[2] [7]

3CPO (Gray et al, NEJM 2008)

Multicentre RCT: standard O2 vs CPAP vs NIV (BiPAP)

Population: 1069 patients with severe acute cardiogenic pulmonary oedema

Key finding

No significant difference in 7-day mortality between groups (~9.5% all arms). NIV groups had faster improvement in symptoms, HR, BP, and acidosis. Reduced need for intubation/escalation.

[7]

Diuretic therapy — DOSE-AHF and practical strategy

Diuretic strategy in ADHF

Initial IV bolus dose

Assess response at 2-6 hours

Target urine output 100-150 mL/h (3-4 L/24h) for effective decongestion. If inadequate (UO <100 mL/h after 2h), double the dose.

Bolus vs continuous infusion

DOSE-AHF: no difference in symptoms or renal function between q12h bolus and continuous infusion. Either strategy is acceptable; boluses allow titration to response.

High-dose vs low-dose

Add thiazide if diuretic resistance

Add potassium-sparing agent

Spironolactone 25-50 mg daily (also proven mortality benefit in chronic HFrEF). Monitor K+ closely when combined with ACEi/ARB/ARNI.

[6]

DOSE-AHF (Felker et al, NEJM 2011)

2x2 factorial RCT: (a) high-dose (2.5x oral) vs low-dose (1x oral) furosemide; (b) bolus q12h vs continuous infusion

Population: 308 patients with acute decompensated heart failure and ≥1 sign of congestion

Key finding

[6]

Vasodilators — when and what

Nitroglycerin (GTN)

First-line if SBP >110

Mechanism: venodilation (↓ preload) at low dose; arteriolar dilation (↓ afterload) at higher doses
Dose: sublingual spray 400 mcg (1-2 puffs); IV infusion 10-200 mcg/min titrated
Indications: AHF with SBP >110 and significant congestion/pulmonary oedema, concomitant ischaemia, or acute MR
Cautions: avoid if SBP <110, right ventricular infarction, PDE5 inhibitor in last 24-48h, severe aortic stenosis
Tachyphylaxis after 16-24h — useful for the first 24h of acute decompensation

Nesiritide (BNP)

No mortality benefit

Recombinant B-type natriuretic peptide — vasodilator, natriuretic, RAAS inhibitor
ASCEND-HF (OConnor 2011): modest dyspnoea improvement, NO mortality or rehospitalisation benefit, no excess renal harm
Not routinely used in ANZ/Europe; largely abandoned due to cost and lack of hard outcomes benefit

Sodium nitroprusside

Severe hypertension / valve regurg

Potent arterial + venous vasodilator; rapid onset/offset
Indications: hypertensive emergency with severe pulmonary oedema, acute severe mitral/aortic regurgitation
Dose: 0.3-3 mcg/kg/min IV; requires arterial line monitoring
Caution: cyanide/thiocyanate toxicity with prolonged use (>24-48h) or renal failure; raises ICP

ACEi / ARB / ARNI

Start AFTER stabilisation

NOT for acute phase — risk of hypotension and AKI in volume-depleted/hypotensive patient
Initiate once euvolaemic, SBP stable, renal function stable (usually 24-48h post-admission)
ARNI (sacubitril/valsartan) preferred over ACEi in HFrEF (PARADIGM-HF mortality benefit)

[1]

ASCEND-HF (O'Connor et al, NEJM 2011)

RCT: nesiritide (recombinant BNP) vs placebo

Population: 7141 patients with acute decompensated heart failure

Key finding

[13]

Inotropes for the cold/shocked patient

Dobutamine

β1 agonist

Dose: 2-20 mcg/kg/min IV
↑ cAMP → ↑ contractility and HR (chronotropy); mild β2 vasodilation
First-line inotrope in most units; cheap, familiar, titratable
Side effects: tachyarrhythmia, ↑ myocardial O2 demand, hypotension (β2), tolerance

Milrinone

PDE-3 inhibitor

Dose: 0.125-0.75 mcg/kg/min IV (loading dose often omitted in AHF)
↑ cAMP via PDE-3 inhibition → ↑ contractility + vasodilation (pulmonary and systemic)
Less tachyarrhythmia than dobutamine; favourable in pulmonary hypertension/RV failure
Caution: renally cleared — halve dose in CKD; significant hypotension on initiation
OPTIME-CHF: routine milrinone in ADHF did NOT improve outcomes and increased atrial arrhythmias/hypotension

Levosimendan

Calcium sensitiser

Binds cardiac troponin C → ↑ myofilament Ca2+ sensitivity without ↑ intracellular Ca2+ (less arrhythmia/ischaemia)
Also opens ATP-sensitive K+ channels → vasodilation (including coronary)
SURVIVE: no mortality benefit vs dobutamine in ADHF (subgroup signal in prior β-blocker use)
Used in some European units; not available in Australia; expensive

Noradrenaline (norepinephrine)

First-line vasopressor

α1 + β1 agonist → vasoconstriction + modest inotropy
Use when MAP <65 despite inotrope, or as first-line vasoconstrictor in shock
Preferred over adrenaline (epinephrine) — less lactataemia and arrhythmia
Dose: 0.05-1 mcg/kg/min IV titrated to MAP ≥65 mmHg

Adrenaline (epinephrine)

Refractory shock

α + β agonist; potent inotrope and vasoconstrictor
Reserved for refractory shock not responding to noradrenaline + inotrope
Causes lactataemia (β2 glycolysis), arrhythmia, worsening renal/splanchnic perfusion

Dopamine

Largely abandoned

Dose-dependent: low dose "renal" (1-3 mcg/kg/min) — ROSE-AHF showed NO renal benefit
Higher doses: tachyarrhythmia excess vs noradrenaline (SOAP-II)
Not recommended for routine use in AHF

[20]

ROSE-AHF (Chen et al, JAMA 2013)

2x2 factorial RCT: low-dose dopamine (2 mcg/kg/min) vs placebo, AND low-dose nesiritide vs placebo

Population: 360 patients with acute heart failure and renal dysfunction (eGFR 15-60)

Key finding

Neither low-dose dopamine nor low-dose nesiritide improved the primary endpoints of 72-hour cumulative urine output or cystatin C change. No renal protection.

[8]

SURVIVE (Mebazaa et al, JAMA 2007)

RCT: levosimendan vs dobutamine

Population: 1327 patients with AHF requiring inotropic support

Key finding

No significant difference in primary endpoint (180-day all-cause mortality: 26% vs 28%). Subgroup suggested possible benefit of levosimendan in patients with prior β-blocker therapy.

[12]

OPTIME-CHF (Cuffe et al, JAMA 2002)

RCT: 48-72h infusion of milrinone vs placebo (not selected by clinician for inotrope indication)

Population: 949 patients with acute exacerbation of chronic systolic heart failure

Key finding

[17]

Mechanical circulatory support (MCS)

For refractory cardiogenic shock despite inotropes/vasopressors, MCS bridges to recovery, decision, durable device, or transplant. [1]

IABP (intra-aortic balloon pump)

Counterpulsation

Mechanism: balloon inflates in diastole (↑ coronary perfusion), deflates in systole (↓ afterload)
Indications: historically first-line; IABP-SHOCK II showed NO mortality benefit in MI-shock
Contraindications: aortic regurgitation, aortic dissection, severe PAD, severe sepsis
Now reserved for mechanical complications of MI (acute MR, VSD) or as bridge to definitive support

Impella (axial flow micro-pump)

LV / RV support

Mechanism: transvalvular axial pump aspirates LV blood, ejects into aorta (LV unloading)
Impella 2.5 (2.5 L/min), CP (4.0 L/min), 5.0 (5.0 L/min); RP for RV support
Active unloading reduces LV wall stress and improves forward flow more than IABP
Complications: limb ischaemia, haemolysis, device malposition, bleeding at insertion site

VA-ECMO

Venoarterial ECMO

Mechanism: drains RA blood, returns oxygenated blood to femoral artery — supports both circulation and gas exchange
Indications: refractory cardiogenic shock, massive PE, fulminant myocarditis, post-cardiotomy, severe ARDS + shock
Flows 3-6 L/min; provides complete cardiopulmonary support
Complications: differential hypoxaemia (Harlequin syndrome — femoral configuration), LV distension (increased afterload — consider Impella or IABP venting), limb ischaemia, bleeding, stroke, haemolysis
Femoral vs central cannulation: central preferred post-cardiotomy; femoral for rapid percutaneous insertion

TandemHeart

Atrial-to-femoral LA→femoral artery

Drains left atrium via transseptal puncture, returns to femoral artery
Flows up to 5 L/min; potent LV unloading
Requires transseptal expertise — less commonly used than Impella/ECMO

IABP-SHOCK II (Thiele et al, NEJM 2012)

RCT: IABP vs no IABP

Population: 600 patients with acute MI and cardiogenic shock undergoing early revascularisation

Key finding

No difference in 30-day mortality (39.7% vs 41.3%) or 12-month mortality. No improvement in secondary endpoints (bleeding, renal function, inflammatory markers, ICU stay).

[15]

CULPRIT-SHOCK (Thiele et al, NEJM 2017)

RCT: culprit-lesion-only PCI vs immediate multivessel PCI

Population: 555 patients with acute MI and multivessel disease presenting in cardiogenic shock

Key finding

Culprit-only PCI reduced primary endpoint (death or severe renal failure requiring RRT at 30 days: 45.9% vs 55.4%) and reduced mortality (43.3% vs 51.6%).

[16]

Cardiogenic shock (SBP <90)

See dedicated topic: cardiogenic-shock-mechanical-support. Key points:

Cold + wet = cardiogenic shock → different pathway
Inotropes: dobutamine or milrinone (improve cardiac output)
Vasopressor: noradrenaline (maintain MAP >65)
Mechanical support: IABP, Impella, VA-ECMO
Urgent revascularisation if ACS (CULPRIT-SHOCK: culprit-only PCI)[16]
Diurese cautiously (may worsen hypotension)

Diuretic resistance

Defined as failure to achieve euvolaemia despite adequate diuretic dosing. Common in chronic HF, CKD, hypoalbuminaemia, and high sodium intake. [1]

Approach to diuretic resistance

Confirm adherence and intake

Review medication history (adherence, stopped diuretics, NSAIDs). Restrict sodium to <2 g/day and fluid to 1-1.5 L/day.

Stepwise IV furosemide up-titration

Start with 1-2.5x oral dose; reassess UO at 2h. Double each subsequent dose until UO >100 mL/h or 200 mg single bolus reached.

Switch to continuous infusion

May improve efficacy once high doses are needed; titrate infusion to UO 100-150 mL/h.

Add thiazide-type diuretic (sequential nephron blockade)

Add acetazolamide

Carbonic anhydrase inhibitor proximal tubule — synergistic in metabolic alkalosis with diuretic resistance (ADVOR trial 2022 showed improved decongestion).

Consider ultrafiltration (rarely)

Reserved for refractory overload NOT responsive to pharmacological therapy. CARRESS-HF showed pharmacologic therapy superior to ultrafiltration.

Address hypoalbuminaemia / nephrotic

In low-albumin states, diuretics enter tubule poorly; consider albumin-furosemide co-infusion (controversial).

Reassess for occult shock / over-diuresis

A "non-responder" may be cold-dry (underfilled) — over-diuresis causes AKI without relieving symptoms. Bedside echo/IVC assessment guides therapy.

[6]

CARRESS-HF (Bart et al, NEJM 2012)

RCT: stepped pharmacological therapy (high-dose IV loop + thiazide + dopamine/nitroglycerin) vs ultrafiltration

Population: 188 patients with ADHF, worsened renal function, and persistent congestion

Key finding

Pharmacological therapy was superior: less weight gain (primary endpoint: -1.7 kg worse in UF arm), fewer serious adverse events (72% in UF vs 57% in pharmacologic), and fewer creatinine rises.

[9]

UNLOAD (Costanzo et al, JACC 2007)

RCT: ultrafiltration vs standard IV diuretic therapy

Population: 200 patients hospitalised with ADHF

Key finding

Ultrafiltration produced greater weight loss and net fluid loss at 48h and fewer 90-day rehospitalisations. No difference in symptoms or renal function. Significant industry sponsorship bias.

[14] [21]

Failed disease-modifying therapies — a recurring pattern

Multiple promising AHF therapies have failed in large RCTs — exam candidates should know these to avoid reaching for unproven agents. [1]

RELAX-AHF (Teerlink et al, Lancet 2013)

RCT: serelaxin (recombinant human relaxin-2) vs placebo for 48h

Population: 1161 patients with acute heart failure, dyspnoea, mild-moderate renal dysfunction, SBP >125

Key finding

Serelaxin improved dyspnoea (visual analogue scale) and reduced 180-day mortality in the original trial (a secondary endpoint). Subsequent RELAX-AHF-2 (2017) FAILED to confirm the mortality signal.

[10]

VERITAS (McMurray et al, JAMA 2007)

RCT: tezosentan (endothelin receptor antagonist) vs placebo

Population: 1435 patients with acute heart failure

Key finding

No improvement in symptoms or 30-day death/heart-failure worsening. Increased adverse events (headache, hypotension).

[11]

Guideline-directed medical therapy (GDMT) and the 4 pillars

The four pillars of HFrEF therapy, ideally initiated/up-titrated before discharge: [1]

Beta-blocker (bisoprolol, carvedilol, metoprolol succinate, nebivolol) — start once euvolaemic and stable (NOT in acute decompensation)
MRA (spironolactone/eplerenone) — K+ <5.0, eGFR >30
ARNI (sacubitril/valsartan) preferred over ACEi/ARB (PARADIGM-HF); SGLT2 inhibitor (dapagliflozin, empagliflozin) — both HFrEF and HFpEF [1]

STRONG-HF (Mebazaa et al, Lancet 2022)

RCT: intensive (rapid up-titration of GDMT to target doses within 2 weeks pre-discharge and at 2-week follow-up) vs usual care

Population: 1078 patients hospitalised with acute heart failure

Key finding

Intensive up-titration reduced 180-day all-cause death or heart failure readmission (15.2% vs 23.3%, HR 0.64). No excess serious adverse events despite rapid titration.

[19]

DAPA-HF (McMurray et al, NEJM 2019)

RCT: dapagliflozin 10 mg daily vs placebo, added to standard GDMT

Population: 4744 patients with HFrEF (EF ≤40%), NYHA II-IV

Key finding

[18]

Discharge criteria

Discharge preparedness is as important as acute stabilisation — premature discharge drives the 25% 30-day readmission rate. The patient should meet ALL of the following: [1]

Clinical — congestion resolved

Required

Euvolaemic: no orthopnoea, JVP not raised, no new crackles, no ascites
Stable dry weight for ≥24h (target 2 kg below admission for chronic HF)
No IV diuretics required — stable on oral diuretic regimen
Off IV vasoactive drugs for ≥24h; SBP stable on oral medications
Symptom improvement: NYHA improved by ≥1 class; patient ambulating without dyspnoea

Investigations — stable

Required

Renal function stable/improving (no ongoing rise in creatinine; K+ 4.0-5.0 mmol/L)
Haemodynamics stable (HR <100, SBP >100, no orthostatic drop)
No evidence of ongoing ischaemia (troponin falling, pain-free)
Echo performed and LV/RV/valve function documented

Medications — GDMT optimised

Required

On at least low dose of each eligible GDMT pillar (BB + MRA + ARNI/ACEi + SGLT2i)
Stable oral diuretic dose defined
NSAIDs, nephrotoxins, and contraindicated drugs stopped
Vaccinations offered (influenza, pneumococcal, COVID-19)

Education and follow-up

Required

Patient taught daily weights (call if >2 kg in 3 days or 2.5 kg in a week)
Sodium (≤2 g/day) and fluid (1.5-2 L/day) restriction counselled
Medication adherence plan; device therapy (ICD/CRT) referral if eligible
Cardiology / HF clinic follow-up booked within 7-14 days
GP follow-up within 1 week
Cardiac rehabilitation referral

[1] [19]

SaqBlocks — fellowship exam practice

SAQ — Acute cardiogenic pulmonary oedema in a warm-wet HFrEF patient

10 minutes · 10 marks

[1]

SAQ — Cardiogenic shock complicating anterior STEMI

10 minutes · 10 marks

[1]

Clinical pearls

High-yield ADHF points for the CICM/FFICM exam

CPAP/BiPAP reduces mortality in cardiogenic pulmonary oedema — use EARLY.[2]
IV furosemide: 1-2.5x usual oral dose. Mechanism: initial venodilation, then diuresis.[6]
Nitroglycerin if SBP >110 (venodilation reduces preload). NOT if hypotensive.
Do NOT give beta-blockers acutely (negative inotropy — may worsen shock).
SBP <90 = cardiogenic shock → inotropes/mechanical support (different pathway).
Forrester classification: warm/cold (perfusion) x wet/dry (congestion).
NT-proBNP: >300 excludes HF, >1800 confirms (age-adjusted thresholds).
Flash pulmonary oedema: consider bilateral renal artery stenosis (Pickering syndrome).
ACE inhibitor/ARB: start once stable and euvolaemic (not acutely).
ACS is #1 precipitant — check ECG and troponin in ALL patients.
Morphine: use cautiously (associated with worse outcomes in observational studies).
HFrEF vs HFpEF: echo differentiates (ejection fraction <40% = HFrEF).
NSAIDs worsen HF — stop all NSAIDs. Sodium retention + vasoconstriction.
Atrial fibrillation: rate control urgently (digoxin, beta-blocker once stable, amiodarone).

Advanced pearls — trials, traps, and exam distinction

Stevenson vs Forrester: Stevenson uses clinical signs (warm/cold × wet/dry); Forrester uses invasive CI and PCWP. Same concept, different measurement.[5]
DOSE-AHF: no difference between bolus vs continuous infusion OR high vs low dose for hard outcomes — titrate to urine output (target 100-150 mL/h).[6]
3CPO: no mortality difference between CPAP and BiPAP — use whichever is available; both beat standard O2.[7]
CARRESS-HF: pharmacologic stepped therapy BEATS ultrafiltration for cardiorenal ADHF. Ultrafiltration is not first-line.[9]
IABP-SHOCK II: IABP provides no benefit in MI-shock — guideline class III. Use Impella/VA-ECMO for refractory shock.[15]
CULPRIT-SHOCK: culprit-only PCI beats multivessel PCI in MI-shock.[16]
SURVIVE & OPTIME-CHF: routine inotropes (levosimendan, milrinone) do NOT improve outcomes in warm-wet ADHF — use ONLY in cold profiles.[12][17]
ROSE-AHF: low-dose "renal" dopamine does NOT protect the kidney in ADHF — abandon it.[8]
STRONG-HF: rapid GDMT up-titration before/at discharge is SAFE and reduces readmission.[19]
SGLT2 inhibitors are now the 4th pillar of HFrEF — start during/after admission, not just for diabetics.[18]
Cardiorenal syndrome is driven by venous congestion, not low output — decongestion (not inotropes or dopamine) is the renal-sparing therapy.
Continue pre-existing beta-blocker if the patient is stable (warm-wet, not in shock); holding it acutely worsens outcomes. STOP only if hypoperfused/cold.
Over-diuresis is a common iatrogenic trap: rising creatinine with cleared congestion may mean underfilling (cold-dry), not diuretic failure — assess volume status with echo/IVC before escalating diuretics.
BNP falls with decongestion — a ≥30% fall in NT-proBNP from admission to discharge is associated with lower readmission (useful discharge readiness marker).
Thiazide + loop = sequential nephron blockade — powerful but dangerous (monitor K+, Na+ q6h); reserve for true resistance.
Echo before inotropes: in apparent cardiogenic shock, confirm structurally — massive PE, tamponade, and severe valvular lesions mimic shock but need different therapy.
The "decongestion paradox": serum creatinine often rises 0.1-0.3 mg/dL with effective decongestion (haemoconcentration) — this is NOT a reason to stop diuresis if the patient remains congested.
Iron deficiency is present in 50% of HF patients and worsens symptoms independent of anaemia — IV ferric carboxymaltose (AFFIRM-AHF) reduces HF rehospitalisation. Check ferritin + TSAT in every admission.

Pharmacology pearls — doses, mechanisms, and exam pitfalls

Furosemide initial venodilation within 5-10 min (prostaglandin-mediated) precedes diuresis — gives early symptom relief before urine flows.
GTN infusion is tachyphylactic after 16-24h; useful acutely, not for prolonged courses.
Dobutamine tolerance develops within 24-48h via β-receptor downregulation — don't leave patients on it indefinitely; bridge to MCS or recovery.
Milrinone is renally cleared — halve the dose if eGFR <30; significant first-dose hypotension (start low, no bolus in AHF).
Levosimendan works downstream of the β-receptor — favoured (where available) in patients already β-blocked who need inotropy.
Noradrenaline (not adrenaline) is first-line vasopressor in cardiogenic shock — less lactataemia, less arrhythmia.
Adrenaline causes β2-driven lactataemia that confounds lactate-guided resuscitation — avoid as first-line shock pressor.
Spironolactone hyperkalaemia risk is highest with concomitant ACEi/ARB/ARNI and CKD — check K+ within 3 days and weekly.
Sacubitril/valsartan must not be started within 36h of ACEi (angioedema risk); switch ACEi → ARNI as a clean interval.
SGLT2 inhibitors cause genital mycotic infections and euglycaemic DKA (rare); transient eGFR dip in first 2 weeks is expected and benign.
Ivabradine (If channel inhibitor) reduces HR without lowering BP — for HFrEF in sinus rhythm with HR >75 despite max-tolerated β-blocker.
Digoxin has no mortality benefit but reduces hospitalisations; useful for rate control in AF with low-output HF (doesn't drop BP). Narrow therapeutic index — toxicity worsens in hypokalaemia (common with loop diuretics).
Hydralazine + nitrate is the evidence-based alternative for HFrEF in African-American patients or those intolerant of ACEi/ARNI (A-HeFT).
Vericiguat (soluble guanylate cyclase stimulator) reduces HF hospitalisation in high-risk chronic HF (VICTORIA) — newer addition to the armamentarium.

[1]

Investigation and monitoring pearls

Lung ultrasound B-lines track decongestion in real time — more sensitive than CXR for interstitial oedema and bedside-friendly.
IVC collapse <50% in a spontaneously breathing patient suggests RA pressure >10 mmHg (congested); >50% collapse with small IVC suggests volume responsiveness.
Mixed venous saturation (SvO2) <60% on a central line suggests low cardiac output — supports inotrope use even if SBP preserved.
Central venous O2 from a ScvO2 (superior vena cava) sample approximates SvO2 with a ~5% offset — usable at the bedside.
Pulse pressure narrows in low-output states (PP/ systolic <25% suggests severe low output) — a quick clinical marker.
Cold extremities, mottling, capillary refill >3s are bedside markers of hypoperfusion independent of BP — use to detect cold-wet/cold-dry.
Troponin elevation is common in ADHF even without ACS (type 2 from wall stress) — only treat as ACS if dynamic rise/fall or ischaemic features.
NT-proBNP <300 ng/L effectively excludes AHF at the emergency presentation — high negative predictive value.
Bilateral pleural effusions that are transudative (LDH low, protein <25 g/L) support a cardiac cause; unilateral raises suspicion of alternative diagnosis.
Hepatic congestion (raised bilirubin, transaminases, PT) from passive venous congestion mimics acute hepatitis and reverses with decongestion.
Spot urine sodium <30 mmol/L in a fluid-restricted patient on loop diuretic suggests high sodium avidity (RAAS activation) — may need higher diuretic doses.
Body weight is the cheapest, most underused monitoring tool — daily weights (same scale, same time, same clothes) guide diuretic titration and catch early fluid retention pre-discharge.

[1]

Red flags

Traps and pitfalls

Routine inotropes in warm-wet ADHF are harmful (OPTIME-CHF) — they add arrhythmia and ischaemia without benefit. Reserve for cold/hypoperfused profiles.[17]
Ultrafiltration is NOT first-line for cardiorenal ADHF — CARRESS-HF showed stepped pharmacologic therapy is superior.[9]
IABP has no role in routine MI-shock (IABP-SHOCK II, ESC class III) — escalate to Impella/VA-ECMO for refractory cases.[15]
Low-dose "renal" dopamine does not protect the kidney in ADHF (ROSE-AHF) — abandon the practice.[8]
Rising creatinine during decongestion may be haemoconcentration (good), not AKI — distinguish by clinical context and stop only if creatinine rise is severe or symptoms of underperfusion appear.
Continuing pre-existing beta-blocker in stable warm-wet ADHF is generally correct — STOP only if cold/hypoperfused.
CULPRIT-SHOCK: do NOT perform multivessel PCI in MI-shock — culprit-only is safer.[16]
VA-ECMO can worsen LV distension and pulmonary oedema (increased afterload) — consider concurrent LV venting (Impella or IABP) in severe LV failure.
ACEi → ARNI switch needs 36h washout to avoid angioedema.
Don't discharge until GDMT optimised — STRONG-HF shows rapid up-titration before discharge is safe and reduces readmission.[19]

When to escalate / call for help

Cold-wet patient not improving on inotrope within 1-2h → call MCS/ECMO team EARLY (before lactate >10 or arrest).
Recurrent ventricular arrhythmia in AHF → exclude ischaemia, electrolyte disturbance (K+, Mg2+), and inotrope toxicity.
New murmur + flash pulmonary oedema → echocardiogram urgently for acute MR (papillary muscle rupture), VSD, or aortic regurgitation — surgical emergency.
Bilateral RAS suspected → arrange renal artery Doppler/CTA; do NOT give ACEi until excluded.
Right heart failure disproportionate to LV failure → consider massive PE, RV infarct (inferior MI), or severe pulmonary hypertension — different therapy.
Postpartum dyspnoea + HF → peripartum cardiomyopathy; bromocriptine + standard HF therapy.
Young patient with new HF → screen for myocarditis (viral serology, CMR, sometimes biopsy), cocaine, thyrotoxicosis, autoimmune.

[1]

References

[1]McDonagh TA, Metra M, Adamo M, et al. 2021 ESC Guidelines for the diagnosis and treatment of acute and chronic heart failure Eur Heart J, 2021.PMID 34447992
[2]Vital FMR, Ladeira MT, Atallah AN. Non-invasive positive pressure ventilation (CPAP or bilevel NPPV) for cardiogenic pulmonary oedema Cochrane Database Syst Rev, 2013.PMID 23728654
[3]Chioncel O, Mebazaa A, Harjola VP, et al. Acute heart failure congestion and perfusion status - impact of the clinical classification on in-hospital and long-term outcomes; insights from the ESC-EORP-HFA Heart Failure Long-Term Registry Eur J Heart Fail, 2019.PMID 31127678
[4]Seferovic PM, Ponikowski P, Anker SD, et al. Clinical practice update on heart failure 2019: pharmacotherapy, procedures, devices and patient management. An expert consensus meeting report of the Heart Failure Association of the European Society of Cardiology Eur J Heart Fail, 2019.PMID 31129923
[5]Nohria A, Tsang SW, Fang JC, et al. Clinical assessment identifies hemodynamic profiles that predict outcomes in patients admitted with heart failure J Am Coll Cardiol, 2003.PMID 12767667
[6]Felker GM, Lee KL, Bull DA, et al. Diuretic strategies in patients with acute decompensated heart failure N Engl J Med, 2011.PMID 21366472
[7]Gray A, Goodacre S, Newby DE, Masson M, Sampson F, Nicholl J (3CPO Trialists). Noninvasive ventilation in acute cardiogenic pulmonary edema N Engl J Med, 2008.PMID 18614781
[8]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.PMID 24247300
[9]Bart BA, Goldsmith SR, Lee KL, et al. Ultrafiltration in decompensated heart failure with cardiorenal syndrome N Engl J Med, 2012.PMID 23131078
[10]Teerlink JR, Cotter G, Davison BA, et al. Serelaxin, recombinant human relaxin-2, for treatment of acute heart failure (RELAX-AHF): a randomised, placebo-controlled trial Lancet, 2013.PMID 23141816
[11]McMurray JJ, Teerlink JR, Cotter G, et al. Effects of tezosentan on symptoms and clinical outcomes in patients with acute heart failure: the VERITAS randomized controlled trials JAMA, 2007.PMID 17986694
[12]Mebazaa A, Nieminen MS, Packer M, et al. Levosimendan vs dobutamine for patients with acute decompensated heart failure: the SURVIVE Randomized Trial JAMA, 2007.PMID 17473298
[13]O'Connor CM, Starling RC, Hernandez AF, et al. Effect of nesiritide in patients with acute decompensated heart failure N Engl J Med, 2011.PMID 21732835
[14]Costanzo MR, Guglin ME, Saltzberg MT, et al. Ultrafiltration versus intravenous diuretics for patients hospitalized for acute decompensated heart failure J Am Coll Cardiol, 2007.PMID 17291932
[15]Thiele H, Zeymer U, Neumann FJ, et al. Intraaortic balloon support for myocardial infarction with cardiogenic shock N Engl J Med, 2012.PMID 22920912
[16]Thiele H, Akin I, Sandri M, et al. PCI Strategies in Patients with Acute Myocardial Infarction and Cardiogenic Shock N Engl J Med, 2017.PMID 29083953
[17]Cuffe MS, Califf RM, Adams KF Jr, et al. Short-term intravenous milrinone for acute exacerbation of chronic heart failure: a randomized controlled trial JAMA, 2002.PMID 11911756
[18]McMurray JJV, Solomon SD, Inzucchi SE, et al. Dapagliflozin in Patients with Heart Failure and Reduced Ejection Fraction N Engl J Med, 2019.PMID 31535829
[19]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.PMID 36356631
[20]Scheeren TWL, Wicentowski M, Gan TJ, et al. Current use of inotropes in circulatory shock Ann Intensive Care, 2021.PMID 33512597
[21]Srivastava M, Safi L, Pipil A, et al. Ultrafiltration for acute heart failure Cochrane Database Syst Rev, 2022.PMID 35061249