Extracorporeal CO2 removal (ECCO2R) in ICU · ICU Topics

Cinematic ICU scene of an ECCO2R circuit with a small low-flow pump and a compact membrane lung connected via a single dual-lumen jugular cannula, a sweep gas line, an ABG showing a falling PaCO2, a spontaneously breathing patient on NIV, clinical-blue lighting, no faces, no text — FigureECCO2R — respiratory dialysis for CO2. Low flow (200-500 mL/min), small cannulae, less anticoagulation than VV-ECMO. It removes CO2 efficiently but cannot oxygenate — use it for hypercapnia (COPD failing NIV, ultra-protective ventilation), never for severe hypoxaemia.

SAQ — ECCO2R for COPD exacerbation failing NIV

10 minutes · 10 marks

A 68-year-old man with severe COPD (FEV1 35% predicted) is admitted with an infective exacerbation. Despite optimal NIV (IPAP 18, EPAP 6, FiO2 0.4), his ABG shows pH 7.22, PaCO2 78 mmHg, PaO2 64. He is becoming somnolent and tachypnoeic at RR 32. The team is considering intubation. The examiners ask you to discuss the role of ECCO2R as an alternative.

[7]

SAQ — ECCO2R to facilitate ultra-protective ventilation in ARDS

10 minutes · 10 marks

A 50-year-old woman with severe ARDS (PaO2/FiO2 95 on FiO2 0.7, PEEP 14) is ventilated with Vt 6 mL/kg predicted body weight; her plateau pressure is 28 cmH2O and driving pressure 16 cmH2O. The examiners ask whether reducing the tidal volume to 3–4 mL/kg ('ultra-protective') — facilitated by ECCO2R — would reduce lung injury, and how you would assess the evidence.

[11]

Clinical pearls

High-yield ECCO2R points for CICM/FFICM exam

ECCO2R removes CO2 efficiently — but NOT oxygenation. CO2 is MORE diffusible than O2 (20x more soluble in water — diffuses across membrane faster). At LOW flow (200-500 mL/min): enough blood passes through membrane to remove significant CO2 (CO2 gradient blood→gas is large: PaCO2 ~50 vs sweep gas 0 → rapid diffusion). BUT: oxygenation requires HIGH flow (to deliver enough O2 to body — need 2-4 L/min of blood flow through oxygenator). Therefore: ECCO2R = CO2 removal. VV-ECMO = CO2 + O2.[4]
Indication: COPD with hypercapnia failing NIV. COPD exacerbation → CO2 retention → acidosis. NIV (BiPAP) is first-line (90% success). But 10% fail NIV (intolerance, mask leak, severe hypercapnia). TRADITIONAL: intubation + ventilation (with risks: VAP, ICUAW, prolonged stay). ECCO2R ALTERNATIVE: remove CO2 via extracorporeal circuit → PaCO2 falls → patient stabilises on NIV → AVOIDS intubation. Evidence: ~50-70% of ECCO2R COPD patients avoid intubation.[6]
Ultra-protective ventilation with ECCO2R — theoretical benefit. ARDS: low Vt (6 mL/kg) reduces mortality (ARMA trial). EVEN LOWER Vt (3-4 mL/kg) → LESS volutrauma (theoretical — animal models show less injury). PROBLEM: Vt 3-4 → very low minute ventilation → CO2 accumulates → acidosis. ECCO2R: removes CO2 → allows Vt 3-4 mL/kg without hypercapnia → ULTRA-PROTECTIVE ventilation. ECLAIR study: feasible (80% achieved Vt 4 mL/kg). But: no mortality benefit yet proven (SUPERNOVA ongoing).[3]
Single dual-lumen cannula — simpler than VV-ECMO. ECCO2R uses ONE catheter (Avalon Elite — dual-lumen): drainage port (venous blood out) + return port (oxygenated blood back) — both through ONE jugular vein. ADVANTAGE: one cannulation site (vs two for standard VV-ECMO — femoral drainage + jugular return). SMALLER: 15-19 Fr (vs 23-29 Fr drainage for VV-ECMO). CAUTION: correct positioning critical (return flow must enter RA — not IVC or SVC — recirculation if mispositioned).[4]
Lower anticoagulation than VV-ECMO. ECCO2R: lower flow (200-500 vs 2,000-4,000 mL/min) → less shear stress → less clotting tendency → LESS anticoagulation needed (ACT 150-180 vs 180-220 for ECMO). ADVANTAGE: LESS BLEEDING (especially important in: COPD patients — some on antiplatelets, or ARDS — often coagulopathic). LIMITATION: even lower anticoagulation → some clotting in circuit (may need circuit change — but less frequent than ECMO).[2]
Complications — fewer than VV-ECMO but not zero. (1) CANNULATION: bleeding (jugular vein — compressible, manageable), pneumothorax (rare — ultrasound-guided). (2) CIRCUIT: clotting (even with anticoagulation — may need circuit change), haemolysis (mechanical shear on RBCs — monitor LDH, free haemoglobin, haemoglobin drop). (3) ANTICOAGULATION: bleeding (from heparin — monitor ACT, aPTT). (4) INFECTION: line-related (from cannula — aseptic technique). (5) RECIRCULATION (dual-lumen cannula): some blood from return port immediately enters drainage port → less effective CO2 removal (check: if PaCO2 not falling → reposition cannula).[2]
When to CHOOSE ECCO2R over VV-ECMO. CHOOSE ECCO2R: if main problem is HYPERCAPNIA (COPD, asthma, or ARDS needing ultra-protective ventilation) — and oxygenation is adequate (PaO2 >60 on FiO2 ≤60%). CHOOSE VV-ECMO: if main problem is HYPOXAEMIA (severe ARDS — PaO2/FiO2 <80) — ECCO2R cannot provide adequate oxygenation. If BOTH hypoxaemia + hypercapnia: VV-ECMO (covers both). KEY: ECCO2R is for CO2 problems. VV-ECMO is for O2 problems (and CO2).[2]
ECCO2R for COPD — evidence emerging. (1) Sklar (2015): 20 COPD patients on ECCO2R → 80% avoided intubation. (2) Fanelli (2020, ICM): larger study — ~60% intubation avoidance with ECCO2R. (3) LIMITATIONS: small studies, heterogeneous, no large RCT. (4) ECLIPSE trial (ongoing): ECCO2R vs standard care in COPD exacerbation with NIV failure — large RCT (may define role). (5) CURRENT: ECCO2R for COPD is INVESTIGATIONAL — may be considered in specialist centres for carefully selected patients (NIV failure, acidosis, otherwise stable).[1]
Blood flow and sweep gas — titration. (1) BLOOD FLOW: 200-500 mL/min (titrate to PaCO2 target). If PaCO2 not falling: increase blood flow (more blood through membrane → more CO2 removed). Maximum: ~500 mL/min (device-dependent — some newer devices up to 1,000 mL/min). (2) SWEEP GAS: oxygen (countercurrent — flows opposite direction to blood). CO2 diffuses from blood (PaCO2 ~50) to sweep gas (CO2 = 0) → removed. Increase sweep gas flow → more CO2 removed (faster gas → less CO2 buildup in membrane → larger gradient). TYPICAL: 5-15 L/min sweep (titrate). (3) MONITOR: serial ABG (PaCO2, pH — adjust flow/sweep to achieve target).[4]
Limitations — why ECCO2R is NOT standard of care (yet). (1) NO large RCT showing mortality benefit (small studies — promising but not definitive). (2) COST: still expensive (circuit, cannula, monitoring, specialist staff) — though less than VV-ECMO. (3) AVAILABILITY: not widely available (specialist centres only). (4) COMPLICATIONS: not zero (bleeding, clotting, haemolysis, cannulation). (5) EVIDENCE GAP: need: large RCTs for COPD (ECLIPSE), ARDS (SUPERNOVA), transplant bridge. Until then: ECCO2R is INVESTIGATIONAL (considered in specialist centres for carefully selected).[5]
ECCO2R vs high-flow nasal cannula for COPD. HFNC: provides FiO2 control + dead space washout → reduces CO2 modestly (2-5 mmHg reduction — by flushing CO2 from upper airway). ECCO2R: removes CO2 EXTRACORPOREAL → reduces PaCO2 significantly (20-40 mmHg reduction — by filtering blood). HFNC: milder (for moderate hypercapnia). ECCO2R: more powerful (for severe hypercapnia, acidosis, NIV failure). ESCALATION: nasal cannula → HFNC → NIV (BiPAP) → ECCO2R → intubation/ventilation.[6]
Technology — devices. (1) Hemolung RAS (ALung Technologies): integrated pump + membrane in one unit. Low flow (400-500 mL/min). Single dual-lumen cannula. Most studied in COPD. (2) PrismaLung (Baxter): membrane lung attached to CRRT platform. Can use CRRT circuit (familiar to ICU staff). Blood flow: 200-450 mL/min. (3) iLA activve (Novalung): pump-driven. Higher flow possible (up to 1,000 mL/min). More CO2 removal. Used in Europe. (4) PROLUNG (Estor): similar concept. (5) All: membrane lung (hollow fibre) + pump + cannula + circuit + sweep gas.[4]
ECCO2R for near-apnoeic oxygenation concept. THEORY: if ECCO2R removes ALL CO2 (complete CO2 removal): patient needs MINIMAL ventilation (apnoeic oxygenation — O2 diffuses into alveoli passively if airway patent and FiO2 100%). CO2 is the main stimulus for ventilation. If CO2 removed extracorporeally → respiratory drive suppressed → less work of breathing → lung rest. CONCEPT: 'apnoeic oxygenation with ECCO2R' — patient barely breathes (or ventilated at very low rate) → lungs REST (no mechanical stress) → healing. EXPERIMENTAL — not routine. May benefit: severe asthma (allow bronchodilator time to work without ventilation damage).[5]
Future of ECCO2R. (1) EXPANDING INDICATIONS: COPD (ECLIPSE trial), ARDS ultra-protective (SUPERNOVA), asthma, transplant bridge, weaning from ECMO (step-down: VV-ECMO → ECCO2R → extubation). (2) TECHNOLOGY: smaller, simpler, safer devices (bio-coated membranes — less anticoagulation). (3) COST: decreasing (if mass-produced). (4) IF positive RCTs: ECCO2R could become STANDARD for: COPD with NIV failure (avoid intubation), ARDS ultra-protective (reduce volutrauma). Currently: BRIDGE between NIV and intubation (or between VV-ECMO and extubation).[5]

Red flags

Prognosis

ECCO2R evidence summary (2023)

[11]

Principle and mechanism of action

Educational diagram of ECCO2R physiology: low blood flow through a membrane lung, sweep gas removing CO2, limited O2 transfer compared with high-flow VV-ECMO — FigureCO2 is highly diffusible — low flow clears CO2, but ECCO2R cannot oxygenate like VV-ECMO.

ECCO2R is an extracorporeal blood-cleansing technique that selectively removes carbon dioxide (CO2) while providing negligible oxygenation. It is conceptually "respiratory dialysis" for CO2. Venous blood is diverted out of the body, passed across a semipermeable membrane lung (hollow-fibre oxygenator) where CO2 diffuses out of the blood and into a ventilating "sweep" gas stream, and then returned to the patient. The fundamental insight that makes low-flow ECCO2R possible is that CO2 is carried in venous blood at a far higher content and diffuses across a membrane far more readily than oxygen, so clinically meaningful CO2 clearance can be achieved at blood flows that are an order of magnitude lower than those needed for ECMO (200–500 mL/min versus 2–4 L/min).[14]

The essential distinction from ECMO is the therapeutic goal, not the hardware. In veno-venous ECMO (VV-ECMO) the objective is to deliver oxygen: the circuit must move enough blood through the oxygenator each minute to supply the body's entire metabolic O2 demand (~250 mL O2/min in a resting adult), which mandates high blood flow. In ECCO2R the objective is to clear a metabolite (CO2): because venous blood is CO2-rich and the membrane is very efficient at stripping CO2, a much smaller fraction of cardiac output (10–15%) needs to be processed. ECCO2R therefore uses the same building blocks — drainage cannula, pump (or arterial driving pressure), membrane lung, sweep gas, return cannula, heat exchanger — but downsized.[12]

[4]

Physiology — why low-flow CO2 removal works

Carbon dioxide is the ideal molecule for low-flow extracorporeal clearance, while oxygen is not. Three physiological facts underpin ECCO2R:[14]

1. CO2 content of venous blood is high and the storage capacity is enormous. Mixed venous blood carries ~52 mL CO2/dL (versus ~15 mL O2/dL for arterial blood, of which only ~5 mL/dL is extractible). Most CO2 (~90%) is stored as bicarbonate which, via carbonic anhydrase on the membrane, continuously regenerates diffusible CO2. The body's total CO2 "pool" (120–150 L of CO2 equivalent) dwarfs the O2 pool (~1.5 L). This large, labile reservoir means removing even 50–100 mL/min of CO2 substantially lowers PaCO2 over 30–60 minutes.[7]

2. CO2 diffuses across a membrane ~20× more readily than O2. CO2 is ~20× more soluble in water/blood than O2 (solubility coefficient ~0.03 mmol/L/mmHg vs ~0.0015 for O2). Although the O2 partial-pressure gradient across the lung is larger (~60 mmHg) than the CO2 gradient (~6 mmHg) in vivo, the 20× solubility advantage means CO2 membrane permeability exceeds O2 permeability. In the ECCO2R membrane lung the gradient for CO2 is also large (~50 mmHg, venous PaCO2 ~50 vs sweep gas ~0), so CO2 transfer is highly efficient even at low blood flow.[14]

3. Oxygenation is FLOW-limited, CO2 removal is partly GRADIENT-limited. To deliver the body's ~250 mL O2/min demand, the circuit must extract nearly all O2 from a large blood volume (high flow): delivering 250 mL O2/min from blood donating 5 mL O2/dL requires ~5 L/min of blood flow. By contrast, CO2 removal of 100 mL/min is easily achieved by stripping ~10 mL CO2/dL from just 1 L/min of blood — and because sweep gas flow can be increased independently, CO2 clearance can be boosted without raising blood flow. This asymmetry is the entire physiological justification for ECCO2R.[14]

[14]

Clinical consequence: a patient whose problem is pure or predominant hypercapnia (COPD exacerbation, asthma, permissive hypercapnia on protective ventilation) can be supported by ECCO2R. A patient whose problem is predominant hypoxaemia (severe ARDS, pneumonia) cannot — they need the high blood flow of VV-ECMO to deliver oxygen. ECCO2R is a CO2 machine, not an O2 machine.[14]

Techniques and configurations

There are three broad technical approaches to ECCO2R, distinguished by how blood is driven through the membrane lung and by the cannulation strategy. The choice is dictated by the clinical scenario, the amount of CO2 to remove, the desired simplicity, and the patient's haemodynamics.[23]

1. Arterio-venous pumpless — iLA / PECLA

Pumpless Extracorporeal Lung Assist (PECLA), marketed as the iLA (Novalung), uses the patient's own arterial pressure to drive blood from a femoral artery through a low-resistance membrane lung and back into a femoral vein. No pump, no pump-related haemolysis or mechanical failure.

Driving pressure: mean arterial pressure (requires MAP >70 mmHg and good cardiac function). Flow 1–2 L/min.
Membrane: very low-resistance biocompatible (heparin-coated) polymethylpentene.
Cannulae: large-bore femoral arterial (15–17 Fr) and venous (17–19 Fr).
CO2 removal: very efficient (can remove 100–200 mL/min) — the highest of any ECCO2R configuration.
Trade-off: places the systemic arterial tree at risk (limb ischaemia from the arterial cannula, distal embolisation) and requires a well-functioning heart. No longer first-line for routine ECCO2R because of lower-flow pump-driven systems, but historically important and still used where pumpless simplicity is desired (e.g., awake bridge-to-transplant).[18][19]

2. Veno-venous pump-driven (modern standard)

A roller or centrifugal pump draws venous blood, pushes it through a small membrane lung, and returns it to the venous system. This is the dominant configuration in modern ECCO2R because it decouples blood flow from the patient's arterial pressure and avoids femoral arterial cannulation.

Driving mechanism: integrated pump (e.g., Hemolung RAS, iLA activve, PrismaLung on a CRRT platform).
Blood flow: 200–500 mL/min (up to ~1,000 mL/min on some newer devices).
Membrane surface: 0.3–0.8 m².
Anticoagulation: systemic heparin (target ACT 150–180, or aPTT ~1.5×), less than ECMO.[12]

3. Cannulation strategies — single (dual-lumen) vs dual (two-cannula)

Single dual-lumen cannula (e.g., Avalon Elite / Spectrum): one catheter inserted via the right internal jugular vein has two lumens — one draining blood from the IVC/SVC into the circuit, the other reinfusing oxygenated/decabarboxylated blood directly into the right atrium, oriented toward the tricuspid valve. Advantages: single insertion site, ambulation possible, less recirculation than two-cannula when correctly positioned, suitable for awake patients. Critical caveat: position MUST be confirmed by echocardiography/fluoroscopy — malposition causes recirculation and is the commonest cause of "ECCO2R not working."[12]
Dual (two-cannula) configuration: drainage cannula (femoral vein) + return cannula (IJ or contralateral femoral). Advantages: simpler to insert, less dependent on precise positioning, less recirculation risk if flow is well directed. Disadvantage: two puncture sites, harder to mobilise the patient.

[11]

[7]

Devices in current clinical use

Hemolung RAS (ALung Technologies): integrated pump-oxygenator in one cartridge; single dual-lumen cannula (15 Fr); ~350–450 mL/min flow; most-studied in COPD (the Burki and Del Sorbo cohorts used it).[8][7]
PrismaLung (Baxter): membrane lung that bolts onto the familiar Prismaflex CRRT platform — leverages ICU staff expertise; 200–450 mL/min.[11]
iLA activve (Xenios/Novalung): pump-driven, higher flows possible (up to ~1,000 mL/min); also offers the pumpless iLA; widely used in Europe.[18]
ProLUNG (Estor): similar low-flow pump-driven concept.
All share: membrane lung + pump + cannula(e) + circuit tubing + sweep-gas source + (optional) heat exchanger.

Indications (in detail)

Clinical pathway for ECCO2R: hypercapnic failure or ultra-protective ARDS ventilation, exclude refractory hypoxaemia needing VV-ECMO, anticoagulate, wean as native ventilation recovers — FigureUse ECCO2R for CO2 problems with adequate oxygenation — escalate to VV-ECMO if hypoxaemia dominates.

ECCO2R has three established-or-emerging indications; all rest on the principle that the dominant problem is CO2 retention, not hypoxaemia.[4]

Indication 1 — Hypercapnic respiratory failure (COPD exacerbation) avoiding intubation

The prototypical and best-studied use. A COPD patient with an acute exacerbation develops worsening hypercapnia and acidosis (PaCO2 >70 mmHg, pH <7.25–7.30). NIV (BiPAP) is first-line and succeeds in ~80–90%, but the 10–20% who fail NIV face invasive mechanical ventilation — with its attendant risks (ventilator-associated pneumonia, ICU-acquired weakness, prolonged stay, difficult weaning, mortality). ECCO2R offers a "rescue" to avoid intubation: by removing CO2 extracorporeally, it lowers PaCO2 and corrects pH, reducing the ventilatory load so that the patient can be maintained on (or weaned to) NIV or even HFNC.

Evidence: Del Sorbo (2015, matched cohort): ECCO2R in hypercapnic patients at risk of NIV failure was feasible; intubation was avoided in a meaningful proportion.[7] Burki (2013, Chest pilot): 20 COPD patients with hypercapnic failure on the Hemolung — most improved their PaCO2 and avoided intubation.[8] Cummins (2018, UK register): real-world ECCO2R outcomes across UK centres.[16]
Caveat: no definitive large RCT yet; ECLIPSE and similar trials are ongoing/being designed. Current guidance: investigational, specialist centres, carefully selected patients.

Indication 2 — Facilitating ultra-protective ventilation in ARDS

Standard ARDS ventilation (Vt 6 mL/kg PBW, plateau <30) already reduces mortality (ARMA). Ultra-protective ventilation (Vt 3–4 mL/kg PBW) should further reduce volutrauma and strain, but at that Vt minute ventilation is too low and the patient develops permissive hypercapnia and acidosis — which is poorly tolerated (pulmonary hypertension, arrhythmia, raised ICP). ECCO2R removes the excess CO2, allowing the lung to be ventilated at extremely low Vt (or very low driving pressure) without paying the hypercapnia penalty.

Evidence: Schmidt (2018, Crit Care): low-flow ECCO2R on a CRRT platform enabled reduction of Vt to ~4 mL/kg in mild-moderate ARDS.[11] Combes/SUPERNOVA (2019, ICM): feasibility/safety of ECCO2R to facilitate protective ventilation in moderate ARDS — the device successfully lowered PaCO2 and permitted tidal-volume reduction.[9] Goligher (2019, ICM): SUPERNOVA sub-study showing CO2-removal capacity varied widely between patients and devices — directly informing the design of a definitive trial.[10] Fitzgerald (2014, systematic review): early evidence insufficient for mortality benefit.[17] Worku (2023, meta-analysis): VV-ECCO2R enables ultra-protective ventilation but no proven survival benefit to date.[21]
Caveat: feasible and safe, but a mortality benefit is unproven; the SUPERNOVA programme is the pivotal trial informing design of larger studies.[13]

Indication 3 — Bridge to lung transplantation

End-stage lung disease (COPD, pulmonary fibrosis, cystic fibrosis) with an acute hypercapnic decompensation can be bridged to transplant with ECCO2R. Because the dominant problem is ventilatory failure (CO2 retention) rather than refractory hypoxaemia, low-flow ECCO2R often suffices and — crucially — can be used in an awake, spontaneously breathing patient via a single dual-lumen cannula, avoiding the deconditioning and complications of intubation while awaiting a donor organ. (When hypoxaemia dominates, VV-ECMO is used instead.)

Emerging / niche indications

Severe asthma with refractory hypercapnia/acidosis to avoid injurious ventilation (small case series).
Near-apnoeic oxygenation concept: near-total CO2 removal markedly suppresses respiratory drive, allowing minimal ventilation and "lung rest" (experimental).
Weaning/liberation from VV-ECMO: step-down from ECMO to ECCO2R before decannulation.
Peri-operative/peri-procedural support in severe lung disease.

Contraindications

Absolute:

Severe refractory hypoxaemia as the dominant problem (ECCO2R cannot oxygenate adequately — use VV-ECMO).
Inability to anticoagulate (active uncontrollable bleeding; absolute heparin contraindication without alternative anticoagulant) — although ECCO2R needs less anticoagulation than ECMO, some anticoagulation is still required.
No venous access feasible, or uncorrected severe coagulopathy (INR >2.0, platelets <50) not amenable to correction.
Patient not a candidate for intensive therapy (futility, advance directive declining life support).[20]

Relative:

Haemodynamic instability / shock for the pumpless AV configuration (needs MAP >70 and good cardiac function — use pump-driven VV instead).
Severe peripheral vascular disease (femoral cannula risk; favour IJ dual-lumen).
Severe anaemia (low haemoglobin reduces CO2 carriage and oxygen reserve).
Morbid obesity or local infection at intended cannulation site.
Lack of experienced centre/staff — ECCO2R should not be initiated where expertise and circuit-management capability are absent.
Recovery expected within hours (transient hypercapnia correctable by NIV optimisation).[7]

Complications

ECCO2R complications are fewer and less severe than VV-ECMO (lower flow, smaller cannulae, less anticoagulation) but are not zero.[12][16]

Bleeding — the commonest significant complication. Sources: cannulation site (jugular is compressible and manageable; femoral arterial worse), heparin anticoagulation, circuit-related consumption of coagulation factors and platelets. Monitor ACT/aPTT, haemoglobin; have low threshold for cannula-site compression and transfusion.
Haemolysis — mechanical shear on red cells across the pump and membrane. Monitor plasma free haemoglobin, LDH, haemoglobin drop, haptoglobin. Newer centrifugal pumps reduce (but do not eliminate) this; the pumpless iLA has the least haemolysis.
Circuit clotting / thrombosis — despite anticoagulation, the membrane and tubing can clot, reducing efficiency and requiring circuit exchange. Visible "clot in chamber," rising transmembrane pressure, falling CO2 removal all signal this.
Access / cannulation complications — pneumothorax (rare with ultrasound guidance), arterial puncture, vessel injury, haematoma, limb ischaemia (especially femoral arterial in pumpless iLA — always use a distal perfusion catheter).
Recirculation (especially dual-lumen cannula) — reinfused blood is immediately re-drained, halving effective CO2 removal. Suspect when PaCO2 fails to fall; confirm with high pre-/post-membrane CO2 difference; fix by repositioning under imaging.
Infection — line/catheter-related bloodstream infection; strict asepsis and daily line review.
Air embolism — risk at any negative-pressure segment of the circuit; modern devices have air detectors and traps.
Fluid/electrolyte shifts, hypothermia (heat loss across the circuit — use a heat exchanger).
Device failure (pump stop, membrane rupture) — emergency clamp-and-disconnect protocol must be rehearsed.[4]

Comparison with conventional mechanical ventilation

[7]

The fundamental trade-off: conventional ventilation substitutes the lung entirely (oxygenation AND ventilation) at the cost of an artificial airway and positive pressure; ECCO2R substitutes only the ventilatory (CO2-clearance) function, sparing the airway and the lung from pressure, at the cost of an extracorporeal circuit and its bleeding/clotting risks. They are complementary, not interchangeable — ECCO2R fits the gap between failing NIV and intubation.[13]

Additional clinical pearls (advanced)

Advanced ECCO2R pearls — mechanism, physiology, selection, management

ECCO2R is "respiratory dialysis for CO2," not a small ECMO. The conceptual frame matters for exam answers: ECMO replaces OXYGENATION (a flow-limited function needing high blood flow); ECCO2R replaces VENTILATION/CO2 clearance (a gradient-limited function achievable at low blood flow because CO2 is 20× more soluble and venous CO2 content is ~3× venous O2 content).[14]
Why CO2 — and not O2 — can be cleared at low flow, in one sentence. CO2 is ~20× more soluble than O2 and venous blood carries ~52 mL CO2/dL vs ~15 mL O2/dL, so stripping ~10 mL CO2/dL from 1 L/min of blood removes 100 mL/min — enough to clear half of metabolic VCO2 — whereas delivering 250 mL/min of O2 needs ~5 L/min because each dL of blood donates only ~5 mL O2.[14]
Bicarbonate is the hidden CO2 reservoir the membrane exploits. ~90% of venous CO2 is bicarbonate; as the membrane strips dissolved CO2, carbonic anhydrase on/around the membrane converts HCO3⁻ + H⁺ → CO2 + H2O, continuously replenishing the diffusible pool. This is why a small membrane at modest flow can still clear a large CO2 load.[14]
Two independent knobs control CO2 removal: blood flow and sweep gas flow — use both. At low blood flows, CO2 clearance rises steeply with sweep gas flow; at high sweep flows, blood-side delivery limits and clearance plateaus. Practically: increase SWEEP first (cheap, no haemolysis risk); increase BLOOD FLOW only if sweep-gas titration is insufficient.[15]
AV pumpless iLA (PECLA) gives the most CO2 removal but bites the arterial tree. Driven by the patient's own MAP, it clears 150–250 mL CO2/min — the highest of any configuration — but requires MAP >70 and a healthy heart, and risks femoral-artery limb ischaemia (always use a distal perfusion catheter). Reserve for niche cases (awake bridge-to-transplant) where pumpless simplicity and maximal clearance are wanted.[18][19]
Pump-driven VV-ECCO2R is the modern default because it works in any haemodynamics. Decoupling blood flow from arterial pressure means you can run ECCO2R in a shocked patient (on vasopressors) — impossible with pumpless iLA. The price is pump-related haemolysis and clot, both manageable with modern centrifugal pumps and heparin-coated circuits.[12]
The single dual-lumen cannula (Avalon) is the key to "awake ECCO2R." One right-IJ puncture, the patient can sit up, eat, and do physiotherapy — invaluable for COPD (avoid deconditioning) and bridge-to-transplant. The price: positioning is critical — malposition causes recirculation and is the commonest reason ECCO2R "doesn't work." Always confirm with echocardiography/fluoroscopy.[12]
Recirculation is the silent killer of ECCO2R efficacy — know how to detect it. If reinfused blood is immediately re-drained, effective CO2 clearance halves. Clues: PaCO2 not falling despite "adequate" flow; small pre-/post-membrane PaCO2 difference; rising sweep-gas requirement. Fix: reposition the dual-lumen cannula so the return jet points at the tricuspid valve, or switch to a two-cannula femoral-IJ configuration.[10]
Selection rule of thumb: ECCO2R for the CO2-problem, VV-ECMO for the O2-problem. If PaO2/FiO2 is >150 (oxygenation acceptable on NIV/HFNC) but PaCO2 is climbing and pH falling → ECCO2R. If PaO2/FiO2 <80 with refractory hypoxaemia → VV-ECMO (ECCO2R cannot oxygenate). If BOTH are failing → VV-ECMO covers both; ECCO2R does not.[12][13]
Ultra-protective ventilation + ECCO2R: the rationale is volutrauma reduction, not oxygenation. Dropping Vt from 6 to 3–4 mL/kg PBW reduces driving pressure, strain, and tidal recruitment — plausible mechanisms for less lung injury — but produces dangerous hypercapnia/acidosis that ECCO2R is recruited to clear. SUPERNOVA proved this is feasible and safe; a mortality benefit awaits a definitive trial.[9][10][21]
Anticoagulation is LESS than ECMO but NOT optional. Target ACT 150–180 (vs 180–220 for ECMO) reflects lower flow and surface area, reducing bleeding. Bein showed adding low-dose aspirin to heparin in pumpless iLA reduced circuit clotting without raising bleeding — a useful trick in pumpless configurations.[20]
Haemolysis monitoring is mandatory even on "gentle" devices. Check plasma free haemoglobin, LDH, haptoglobin, and trend haemoglobin. A rising free-Hb with falling platelets and circuit consumption mimics TMA/HUS — distinguish circuit-induced mechanical haemolysis (membrane/pump change) from true thrombotic microangiopathy.[12]
The COPD evidence is promising but unproven — sell it as "rescue, not routine." Del Sorbo (matched cohort) and Burki (pilot) showed meaningful intubation avoidance in NIV-failing hypercapnic COPD, and the UK register (Cummins) confirmed real-world feasibility — but no large RCT yet demonstrates improved survival. ECLIPSE-type trials are the inflection point.[7][8][16]
Weaning ECCO2R is the mirror of starting it — turn down sweep first, then flow. As the native lung improves (bronchodilation, antibiotics, diuresis), reduce sweep gas flow to hold PaCO2 at target, then reduce blood flow, then decannulate when PaCO2/pH are stable on minimal/no sweep for 6–12 hours. Avoid abrupt stoppage — rebound hypercapnia can occur.[15]
ECCO2R can be a step-DOWN from VV-ECMO. A patient recovering on VV-ECMO can transition to ECCO2R (lower flow, single cannula) once oxygenation is no longer the problem but CO2 clearance is still needed — a gentler bridge to extubation and decannulation, reducing ECMO-days and anticoagulation exposure.[13]
Know the devices and their niches for viva. Hemolung RAS (integrated pump-oxygenator, single 15 Fr cannula, COPD heritage); PrismaLung (bolts onto CRRT platform — leverages ICU skills, Schmidt ARDS study); iLA activve (pump-driven, higher flow possible; also offers pumpless iLA for maximal clearance/awake bridge); ProLUNG (similar concept). All = membrane + pump + cannula + sweep gas.[8][11][18]
Cost and access are the real-world limits. ECCO2R is cheaper and simpler than VV-ECMO but still expensive (circuit, cannula, specialist staff, monitoring) and available only in specialist centres — a practical reason it remains investigational even where the physiology is compelling.[13]
The exam answer on ECCO2R vs ECMO in one breath. "ECCO2R is a low-flow extracorporeal circuit that removes CO2 — not oxygen — by exploiting CO2's 20-fold higher solubility and high venous content; it uses smaller cannulae, lower flow (200–500 mL/min), and less anticoagulation than VV-ECMO, and is used for hypercapnic failure (COPD) and to facilitate ultra-protective ventilation in ARDS, but cannot treat severe hypoxaemia."[12]

Evidence summary and ongoing trials

Key ECCO2R studies (verified PMIDs)

Where the evidence is heading

Red flags (consolidated)

ECCO2R red flags — must-not-miss

ECCO2R removes CO2, NOT oxygen — never deploy it as sole support for severe hypoxaemia (use VV-ECMO).[12]
Lower flow (200–500 mL/min), smaller cannulae, less anticoagulation than VV-ECMO → fewer but not zero complications.[13]
Ultra-protective ventilation (Vt 3–4 mL/kg) is enabled by ECCO2R removing the excess CO2 — the rationale in ARDS.[9][10]
No large RCT yet — ECCO2R is investigational, specialist centres, carefully selected patients.[13]
PaCO2 not falling? think recirculation, circuit clot, cannula malposition, inadequate sweep gas flow.[15]
Femoral arterial cannula + cold pulseless leg → limb ischaemia (pumpless iLA) — distal perfusion catheter now.[18]
Rising free Hb/LDH, falling haptoglobin → haemolysis; inspect circuit, consider membrane/pump change.[12]
Dual-lumen cannula must be echo/fluoroscopy-positioned — malposition = recirculation = ineffective therapy.[12]
Monitor: PaCO2/pH (serial ABG), plasma free Hb, LDH, ACT/aPTT, haemoglobin, cannulation site, limb perfusion.[13]

Prognosis and current status

ECCO2R sits between NIV/HFNC and invasive ventilation (or between VV-ECMO and liberation) as a CO2-clearance tool. Its physiological rationale is robust and its complication profile is favourable compared with ECMO, but it has not yet been shown in a large RCT to improve survival in any single indication. Current practice: investigational, deployed in specialist ECMO-capable centres for carefully selected patients with predominant hypercapnia — COPD failing NIV, ARDS requiring ultra-protective ventilation, and awake bridge-to-transplant or step-down from VV-ECMO. The trajectory of evidence (SUPERNOVA, ECLIPSE-type trials) will determine whether ECCO2R becomes standard respiratory support or remains a niche rescue.[13][16]

References

[1]Braune S, Kluge S ECCO(2)R in COPD exacerbation: response to comments by Del Sorbo et al. Intensive care medicine, 2016.PMID 27586995
[2]Bein T, Aubron C, Papazian L Focus on ECMO and ECCO(2)R in ARDS patients. Intensive care medicine, 2017.PMID 28717835
[3]Combes A, Fanelli V, Pham T, et al. Feasibility and safety of extracorporeal CO(2) removal to enhance protective ventilation in acute respiratory distress syndrome: the SUPERNOVA study. Intensive care medicine, 2019.PMID 30790030
[4]Staudinger T Update on extracorporeal carbon dioxide removal: a comprehensive review on principles, indications, efficiency, and complications. Perfusion, 2020.PMID 32156179
[5]Combes A, Fanelli V, Pham T, et al. Feasibility and safety of extracorporeal CO(2) removal to enhance protective ventilation in acute respiratory distress syndrome: the SUPERNOVA study. Intensive care medicine, 2019.PMID 30790030
[7]Del Sorbo L, Pisani L, Filippini C, et al. Extracorporeal Co2 removal in hypercapnic patients at risk of noninvasive ventilation failure: a matched cohort study with historical control. Critical care medicine, 2015.PMID 25230375
[8]Burki NK, Mani RK, Herth FJF, et al. A novel extracorporeal CO(2) removal system: results of a pilot study of hypercapnic respiratory failure in patients with COPD. Chest, 2013.PMID 23460154
[9]Combes A, Fanelli V, Pham T, Ranieri VM Feasibility and safety of extracorporeal CO(2) removal to enhance protective ventilation in acute respiratory distress syndrome: the SUPERNOVA study. Intensive care medicine, 2019.PMID 30790030
[10]Goligher EC, Combes A, Brodie D, et al. Determinants of the effect of extracorporeal carbon dioxide removal in the SUPERNOVA trial: implications for trial design. Intensive care medicine, 2019.PMID 31432216
[11]Schmidt M, Herisse M, Ponthus C, et al. Feasibility and safety of low-flow extracorporeal CO(2) removal managed with a renal replacement platform to enhance lung-protective ventilation of patients with mild-to-moderate ARDS. Critical care, 2018.PMID 29743094
[12]Morelli A, Del Sorbo L, Pesenti A, Ranieri VM, Fan E Extracorporeal carbon dioxide removal (ECCO(2)R) in patients with acute respiratory failure. Intensive care medicine, 2017.PMID 28132075
[13]Boyle AJ, Sklar M, McNamee JJ, et al. Extracorporeal carbon dioxide removal for lowering the risk of mechanical ventilation: research questions and clinical potential for the future. Lancet Respiratory Medicine, 2018.PMID 30484429
[14]Terragni P, Maiolo G, Ranieri VM Role and potentials of low-flow CO(2) removal system in mechanical ventilation. Current opinion in critical care, 2012.PMID 22186219
[15]Strassmann S, Mertens MC, Bickenbach J, et al. Impact of sweep gas flow on extracorporeal CO(2) removal (ECCO(2)R). Intensive care medicine experimental, 2019.PMID 30911910
[16]Cummins C, McFadyen A, Rafferty I, Poenaru D, Kinsella J A United Kingdom Register study of in-hospital outcomes of patients receiving extracorporeal carbon dioxide removal. Journal of intensive care society, 2018.PMID 29796067
[17]Fitzgerald M, Millar J, Blackwood B, et al. Extracorporeal carbon dioxide removal for patients with acute respiratory failure secondary to the acute respiratory distress syndrome: a systematic review. Critical care, 2014.PMID 25033302
[18]Zimmermann M, Bein T, Arlt M, et al. Pumpless extracorporeal interventional lung assist in patients with acute respiratory distress syndrome: a prospective pilot study. Critical care, 2009.PMID 19183475
[19]Muller T, Lubnow M, Philipp A, et al. Extracorporeal pumpless interventional lung assist in clinical practice: determinants of efficacy. European respiratory journal, 2009.PMID 19010979
[20]Bein T, Zimmermann M, Hergeth K, et al. Addition of acetylsalicylic acid to heparin for anticoagulation management during pumpless extracorporeal lung assist. ASAIO journal, 2011.PMID 21427564
[21]Worku E, Okeke C, Patel R, Goyal R, Henry M Venovenous extracorporeal CO(2) removal to support ultraprotective ventilation in moderate-severe acute respiratory distress syndrome: a systematic review and meta-analysis of the literature. Perfusion, 2023.PMID 35656595
[22]Karagiannidis C, Brodie D, Strassmann S, et al. Extracorporeal membrane oxygenation: evolving epidemiology and mortality. Intensive care medicine, 2016.PMID 26942446
[23]Westhoff M, Schonhofer B, Neumann P, et al. Severe hypercapnic respiratory failure in acute exacerbation of COPD: significance of ventilation and extracorporeal CO2 removal. Deutsche medizinische Wochenschrift, 2016.PMID 27903026
[6]Kryvenko V, Husain-Syed F, Schnell E, et al. Minimally invasive extracorporeal CO(2) removal in hypercapnic respiratory failure: a prospective observational study. Critical care (London, England), 2026.PMID 42057044

[11] [23]

SAQ — ECCO2R for COPD exacerbation failing NIV

10 minutes · 10 marks

[7]

SAQ — ECCO2R to facilitate ultra-protective ventilation in ARDS

10 minutes · 10 marks

[11]

Clinical pearls

High-yield ECCO2R points for CICM/FFICM exam

ECCO2R removes CO2 efficiently — but NOT oxygenation. CO2 is MORE diffusible than O2 (20x more soluble in water — diffuses across membrane faster). At LOW flow (200-500 mL/min): enough blood passes through membrane to remove significant CO2 (CO2 gradient blood→gas is large: PaCO2 ~50 vs sweep gas 0 → rapid diffusion). BUT: oxygenation requires HIGH flow (to deliver enough O2 to body — need 2-4 L/min of blood flow through oxygenator). Therefore: ECCO2R = CO2 removal. VV-ECMO = CO2 + O2.[4]
Indication: COPD with hypercapnia failing NIV. COPD exacerbation → CO2 retention → acidosis. NIV (BiPAP) is first-line (90% success). But 10% fail NIV (intolerance, mask leak, severe hypercapnia). TRADITIONAL: intubation + ventilation (with risks: VAP, ICUAW, prolonged stay). ECCO2R ALTERNATIVE: remove CO2 via extracorporeal circuit → PaCO2 falls → patient stabilises on NIV → AVOIDS intubation. Evidence: ~50-70% of ECCO2R COPD patients avoid intubation.[6]
Ultra-protective ventilation with ECCO2R — theoretical benefit. ARDS: low Vt (6 mL/kg) reduces mortality (ARMA trial). EVEN LOWER Vt (3-4 mL/kg) → LESS volutrauma (theoretical — animal models show less injury). PROBLEM: Vt 3-4 → very low minute ventilation → CO2 accumulates → acidosis. ECCO2R: removes CO2 → allows Vt 3-4 mL/kg without hypercapnia → ULTRA-PROTECTIVE ventilation. ECLAIR study: feasible (80% achieved Vt 4 mL/kg). But: no mortality benefit yet proven (SUPERNOVA ongoing).[3]
Single dual-lumen cannula — simpler than VV-ECMO. ECCO2R uses ONE catheter (Avalon Elite — dual-lumen): drainage port (venous blood out) + return port (oxygenated blood back) — both through ONE jugular vein. ADVANTAGE: one cannulation site (vs two for standard VV-ECMO — femoral drainage + jugular return). SMALLER: 15-19 Fr (vs 23-29 Fr drainage for VV-ECMO). CAUTION: correct positioning critical (return flow must enter RA — not IVC or SVC — recirculation if mispositioned).[4]
Lower anticoagulation than VV-ECMO. ECCO2R: lower flow (200-500 vs 2,000-4,000 mL/min) → less shear stress → less clotting tendency → LESS anticoagulation needed (ACT 150-180 vs 180-220 for ECMO). ADVANTAGE: LESS BLEEDING (especially important in: COPD patients — some on antiplatelets, or ARDS — often coagulopathic). LIMITATION: even lower anticoagulation → some clotting in circuit (may need circuit change — but less frequent than ECMO).[2]
Complications — fewer than VV-ECMO but not zero. (1) CANNULATION: bleeding (jugular vein — compressible, manageable), pneumothorax (rare — ultrasound-guided). (2) CIRCUIT: clotting (even with anticoagulation — may need circuit change), haemolysis (mechanical shear on RBCs — monitor LDH, free haemoglobin, haemoglobin drop). (3) ANTICOAGULATION: bleeding (from heparin — monitor ACT, aPTT). (4) INFECTION: line-related (from cannula — aseptic technique). (5) RECIRCULATION (dual-lumen cannula): some blood from return port immediately enters drainage port → less effective CO2 removal (check: if PaCO2 not falling → reposition cannula).[2]
When to CHOOSE ECCO2R over VV-ECMO. CHOOSE ECCO2R: if main problem is HYPERCAPNIA (COPD, asthma, or ARDS needing ultra-protective ventilation) — and oxygenation is adequate (PaO2 >60 on FiO2 ≤60%). CHOOSE VV-ECMO: if main problem is HYPOXAEMIA (severe ARDS — PaO2/FiO2 <80) — ECCO2R cannot provide adequate oxygenation. If BOTH hypoxaemia + hypercapnia: VV-ECMO (covers both). KEY: ECCO2R is for CO2 problems. VV-ECMO is for O2 problems (and CO2).[2]
ECCO2R for COPD — evidence emerging. (1) Sklar (2015): 20 COPD patients on ECCO2R → 80% avoided intubation. (2) Fanelli (2020, ICM): larger study — ~60% intubation avoidance with ECCO2R. (3) LIMITATIONS: small studies, heterogeneous, no large RCT. (4) ECLIPSE trial (ongoing): ECCO2R vs standard care in COPD exacerbation with NIV failure — large RCT (may define role). (5) CURRENT: ECCO2R for COPD is INVESTIGATIONAL — may be considered in specialist centres for carefully selected patients (NIV failure, acidosis, otherwise stable).[1]
Blood flow and sweep gas — titration. (1) BLOOD FLOW: 200-500 mL/min (titrate to PaCO2 target). If PaCO2 not falling: increase blood flow (more blood through membrane → more CO2 removed). Maximum: ~500 mL/min (device-dependent — some newer devices up to 1,000 mL/min). (2) SWEEP GAS: oxygen (countercurrent — flows opposite direction to blood). CO2 diffuses from blood (PaCO2 ~50) to sweep gas (CO2 = 0) → removed. Increase sweep gas flow → more CO2 removed (faster gas → less CO2 buildup in membrane → larger gradient). TYPICAL: 5-15 L/min sweep (titrate). (3) MONITOR: serial ABG (PaCO2, pH — adjust flow/sweep to achieve target).[4]
Limitations — why ECCO2R is NOT standard of care (yet). (1) NO large RCT showing mortality benefit (small studies — promising but not definitive). (2) COST: still expensive (circuit, cannula, monitoring, specialist staff) — though less than VV-ECMO. (3) AVAILABILITY: not widely available (specialist centres only). (4) COMPLICATIONS: not zero (bleeding, clotting, haemolysis, cannulation). (5) EVIDENCE GAP: need: large RCTs for COPD (ECLIPSE), ARDS (SUPERNOVA), transplant bridge. Until then: ECCO2R is INVESTIGATIONAL (considered in specialist centres for carefully selected).[5]
ECCO2R vs high-flow nasal cannula for COPD. HFNC: provides FiO2 control + dead space washout → reduces CO2 modestly (2-5 mmHg reduction — by flushing CO2 from upper airway). ECCO2R: removes CO2 EXTRACORPOREAL → reduces PaCO2 significantly (20-40 mmHg reduction — by filtering blood). HFNC: milder (for moderate hypercapnia). ECCO2R: more powerful (for severe hypercapnia, acidosis, NIV failure). ESCALATION: nasal cannula → HFNC → NIV (BiPAP) → ECCO2R → intubation/ventilation.[6]
Technology — devices. (1) Hemolung RAS (ALung Technologies): integrated pump + membrane in one unit. Low flow (400-500 mL/min). Single dual-lumen cannula. Most studied in COPD. (2) PrismaLung (Baxter): membrane lung attached to CRRT platform. Can use CRRT circuit (familiar to ICU staff). Blood flow: 200-450 mL/min. (3) iLA activve (Novalung): pump-driven. Higher flow possible (up to 1,000 mL/min). More CO2 removal. Used in Europe. (4) PROLUNG (Estor): similar concept. (5) All: membrane lung (hollow fibre) + pump + cannula + circuit + sweep gas.[4]
ECCO2R for near-apnoeic oxygenation concept. THEORY: if ECCO2R removes ALL CO2 (complete CO2 removal): patient needs MINIMAL ventilation (apnoeic oxygenation — O2 diffuses into alveoli passively if airway patent and FiO2 100%). CO2 is the main stimulus for ventilation. If CO2 removed extracorporeally → respiratory drive suppressed → less work of breathing → lung rest. CONCEPT: 'apnoeic oxygenation with ECCO2R' — patient barely breathes (or ventilated at very low rate) → lungs REST (no mechanical stress) → healing. EXPERIMENTAL — not routine. May benefit: severe asthma (allow bronchodilator time to work without ventilation damage).[5]
Future of ECCO2R. (1) EXPANDING INDICATIONS: COPD (ECLIPSE trial), ARDS ultra-protective (SUPERNOVA), asthma, transplant bridge, weaning from ECMO (step-down: VV-ECMO → ECCO2R → extubation). (2) TECHNOLOGY: smaller, simpler, safer devices (bio-coated membranes — less anticoagulation). (3) COST: decreasing (if mass-produced). (4) IF positive RCTs: ECCO2R could become STANDARD for: COPD with NIV failure (avoid intubation), ARDS ultra-protective (reduce volutrauma). Currently: BRIDGE between NIV and intubation (or between VV-ECMO and extubation).[5]

Red flags

Prognosis

ECCO2R evidence summary (2023)

[11]

Principle and mechanism of action

[4]

Physiology — why low-flow CO2 removal works

Carbon dioxide is the ideal molecule for low-flow extracorporeal clearance, while oxygen is not. Three physiological facts underpin ECCO2R:[14]

[14]

Techniques and configurations

1. Arterio-venous pumpless — iLA / PECLA

Driving pressure: mean arterial pressure (requires MAP >70 mmHg and good cardiac function). Flow 1–2 L/min.
Membrane: very low-resistance biocompatible (heparin-coated) polymethylpentene.
Cannulae: large-bore femoral arterial (15–17 Fr) and venous (17–19 Fr).
CO2 removal: very efficient (can remove 100–200 mL/min) — the highest of any ECCO2R configuration.
Trade-off: places the systemic arterial tree at risk (limb ischaemia from the arterial cannula, distal embolisation) and requires a well-functioning heart. No longer first-line for routine ECCO2R because of lower-flow pump-driven systems, but historically important and still used where pumpless simplicity is desired (e.g., awake bridge-to-transplant).[18][19]

2. Veno-venous pump-driven (modern standard)

Driving mechanism: integrated pump (e.g., Hemolung RAS, iLA activve, PrismaLung on a CRRT platform).
Blood flow: 200–500 mL/min (up to ~1,000 mL/min on some newer devices).
Membrane surface: 0.3–0.8 m².
Anticoagulation: systemic heparin (target ACT 150–180, or aPTT ~1.5×), less than ECMO.[12]

3. Cannulation strategies — single (dual-lumen) vs dual (two-cannula)

Single dual-lumen cannula (e.g., Avalon Elite / Spectrum): one catheter inserted via the right internal jugular vein has two lumens — one draining blood from the IVC/SVC into the circuit, the other reinfusing oxygenated/decabarboxylated blood directly into the right atrium, oriented toward the tricuspid valve. Advantages: single insertion site, ambulation possible, less recirculation than two-cannula when correctly positioned, suitable for awake patients. Critical caveat: position MUST be confirmed by echocardiography/fluoroscopy — malposition causes recirculation and is the commonest cause of "ECCO2R not working."[12]
Dual (two-cannula) configuration: drainage cannula (femoral vein) + return cannula (IJ or contralateral femoral). Advantages: simpler to insert, less dependent on precise positioning, less recirculation risk if flow is well directed. Disadvantage: two puncture sites, harder to mobilise the patient.

[11]

[7]

Devices in current clinical use

Hemolung RAS (ALung Technologies): integrated pump-oxygenator in one cartridge; single dual-lumen cannula (15 Fr); ~350–450 mL/min flow; most-studied in COPD (the Burki and Del Sorbo cohorts used it).[8][7]
PrismaLung (Baxter): membrane lung that bolts onto the familiar Prismaflex CRRT platform — leverages ICU staff expertise; 200–450 mL/min.[11]
iLA activve (Xenios/Novalung): pump-driven, higher flows possible (up to ~1,000 mL/min); also offers the pumpless iLA; widely used in Europe.[18]
ProLUNG (Estor): similar low-flow pump-driven concept.
All share: membrane lung + pump + cannula(e) + circuit tubing + sweep-gas source + (optional) heat exchanger.

Indications (in detail)

ECCO2R has three established-or-emerging indications; all rest on the principle that the dominant problem is CO2 retention, not hypoxaemia.[4]

Indication 1 — Hypercapnic respiratory failure (COPD exacerbation) avoiding intubation

Evidence: Del Sorbo (2015, matched cohort): ECCO2R in hypercapnic patients at risk of NIV failure was feasible; intubation was avoided in a meaningful proportion.[7] Burki (2013, Chest pilot): 20 COPD patients with hypercapnic failure on the Hemolung — most improved their PaCO2 and avoided intubation.[8] Cummins (2018, UK register): real-world ECCO2R outcomes across UK centres.[16]
Caveat: no definitive large RCT yet; ECLIPSE and similar trials are ongoing/being designed. Current guidance: investigational, specialist centres, carefully selected patients.

Indication 2 — Facilitating ultra-protective ventilation in ARDS

Evidence: Schmidt (2018, Crit Care): low-flow ECCO2R on a CRRT platform enabled reduction of Vt to ~4 mL/kg in mild-moderate ARDS.[11] Combes/SUPERNOVA (2019, ICM): feasibility/safety of ECCO2R to facilitate protective ventilation in moderate ARDS — the device successfully lowered PaCO2 and permitted tidal-volume reduction.[9] Goligher (2019, ICM): SUPERNOVA sub-study showing CO2-removal capacity varied widely between patients and devices — directly informing the design of a definitive trial.[10] Fitzgerald (2014, systematic review): early evidence insufficient for mortality benefit.[17] Worku (2023, meta-analysis): VV-ECCO2R enables ultra-protective ventilation but no proven survival benefit to date.[21]
Caveat: feasible and safe, but a mortality benefit is unproven; the SUPERNOVA programme is the pivotal trial informing design of larger studies.[13]

Indication 3 — Bridge to lung transplantation

Emerging / niche indications

Severe asthma with refractory hypercapnia/acidosis to avoid injurious ventilation (small case series).
Near-apnoeic oxygenation concept: near-total CO2 removal markedly suppresses respiratory drive, allowing minimal ventilation and "lung rest" (experimental).
Weaning/liberation from VV-ECMO: step-down from ECMO to ECCO2R before decannulation.
Peri-operative/peri-procedural support in severe lung disease.

Contraindications

Absolute:

Severe refractory hypoxaemia as the dominant problem (ECCO2R cannot oxygenate adequately — use VV-ECMO).
Inability to anticoagulate (active uncontrollable bleeding; absolute heparin contraindication without alternative anticoagulant) — although ECCO2R needs less anticoagulation than ECMO, some anticoagulation is still required.
No venous access feasible, or uncorrected severe coagulopathy (INR >2.0, platelets <50) not amenable to correction.
Patient not a candidate for intensive therapy (futility, advance directive declining life support).[20]

Relative:

Haemodynamic instability / shock for the pumpless AV configuration (needs MAP >70 and good cardiac function — use pump-driven VV instead).
Severe peripheral vascular disease (femoral cannula risk; favour IJ dual-lumen).
Severe anaemia (low haemoglobin reduces CO2 carriage and oxygen reserve).
Morbid obesity or local infection at intended cannulation site.
Lack of experienced centre/staff — ECCO2R should not be initiated where expertise and circuit-management capability are absent.
Recovery expected within hours (transient hypercapnia correctable by NIV optimisation).[7]

Complications

ECCO2R complications are fewer and less severe than VV-ECMO (lower flow, smaller cannulae, less anticoagulation) but are not zero.[12][16]

Bleeding — the commonest significant complication. Sources: cannulation site (jugular is compressible and manageable; femoral arterial worse), heparin anticoagulation, circuit-related consumption of coagulation factors and platelets. Monitor ACT/aPTT, haemoglobin; have low threshold for cannula-site compression and transfusion.
Haemolysis — mechanical shear on red cells across the pump and membrane. Monitor plasma free haemoglobin, LDH, haemoglobin drop, haptoglobin. Newer centrifugal pumps reduce (but do not eliminate) this; the pumpless iLA has the least haemolysis.
Circuit clotting / thrombosis — despite anticoagulation, the membrane and tubing can clot, reducing efficiency and requiring circuit exchange. Visible "clot in chamber," rising transmembrane pressure, falling CO2 removal all signal this.
Access / cannulation complications — pneumothorax (rare with ultrasound guidance), arterial puncture, vessel injury, haematoma, limb ischaemia (especially femoral arterial in pumpless iLA — always use a distal perfusion catheter).
Recirculation (especially dual-lumen cannula) — reinfused blood is immediately re-drained, halving effective CO2 removal. Suspect when PaCO2 fails to fall; confirm with high pre-/post-membrane CO2 difference; fix by repositioning under imaging.
Infection — line/catheter-related bloodstream infection; strict asepsis and daily line review.
Air embolism — risk at any negative-pressure segment of the circuit; modern devices have air detectors and traps.
Fluid/electrolyte shifts, hypothermia (heat loss across the circuit — use a heat exchanger).
Device failure (pump stop, membrane rupture) — emergency clamp-and-disconnect protocol must be rehearsed.[4]

Comparison with conventional mechanical ventilation

[7]

Additional clinical pearls (advanced)

Advanced ECCO2R pearls — mechanism, physiology, selection, management

ECCO2R is "respiratory dialysis for CO2," not a small ECMO. The conceptual frame matters for exam answers: ECMO replaces OXYGENATION (a flow-limited function needing high blood flow); ECCO2R replaces VENTILATION/CO2 clearance (a gradient-limited function achievable at low blood flow because CO2 is 20× more soluble and venous CO2 content is ~3× venous O2 content).[14]
Why CO2 — and not O2 — can be cleared at low flow, in one sentence. CO2 is ~20× more soluble than O2 and venous blood carries ~52 mL CO2/dL vs ~15 mL O2/dL, so stripping ~10 mL CO2/dL from 1 L/min of blood removes 100 mL/min — enough to clear half of metabolic VCO2 — whereas delivering 250 mL/min of O2 needs ~5 L/min because each dL of blood donates only ~5 mL O2.[14]
Bicarbonate is the hidden CO2 reservoir the membrane exploits. ~90% of venous CO2 is bicarbonate; as the membrane strips dissolved CO2, carbonic anhydrase on/around the membrane converts HCO3⁻ + H⁺ → CO2 + H2O, continuously replenishing the diffusible pool. This is why a small membrane at modest flow can still clear a large CO2 load.[14]
Two independent knobs control CO2 removal: blood flow and sweep gas flow — use both. At low blood flows, CO2 clearance rises steeply with sweep gas flow; at high sweep flows, blood-side delivery limits and clearance plateaus. Practically: increase SWEEP first (cheap, no haemolysis risk); increase BLOOD FLOW only if sweep-gas titration is insufficient.[15]
AV pumpless iLA (PECLA) gives the most CO2 removal but bites the arterial tree. Driven by the patient's own MAP, it clears 150–250 mL CO2/min — the highest of any configuration — but requires MAP >70 and a healthy heart, and risks femoral-artery limb ischaemia (always use a distal perfusion catheter). Reserve for niche cases (awake bridge-to-transplant) where pumpless simplicity and maximal clearance are wanted.[18][19]
Pump-driven VV-ECCO2R is the modern default because it works in any haemodynamics. Decoupling blood flow from arterial pressure means you can run ECCO2R in a shocked patient (on vasopressors) — impossible with pumpless iLA. The price is pump-related haemolysis and clot, both manageable with modern centrifugal pumps and heparin-coated circuits.[12]
The single dual-lumen cannula (Avalon) is the key to "awake ECCO2R." One right-IJ puncture, the patient can sit up, eat, and do physiotherapy — invaluable for COPD (avoid deconditioning) and bridge-to-transplant. The price: positioning is critical — malposition causes recirculation and is the commonest reason ECCO2R "doesn't work." Always confirm with echocardiography/fluoroscopy.[12]
Recirculation is the silent killer of ECCO2R efficacy — know how to detect it. If reinfused blood is immediately re-drained, effective CO2 clearance halves. Clues: PaCO2 not falling despite "adequate" flow; small pre-/post-membrane PaCO2 difference; rising sweep-gas requirement. Fix: reposition the dual-lumen cannula so the return jet points at the tricuspid valve, or switch to a two-cannula femoral-IJ configuration.[10]
Selection rule of thumb: ECCO2R for the CO2-problem, VV-ECMO for the O2-problem. If PaO2/FiO2 is >150 (oxygenation acceptable on NIV/HFNC) but PaCO2 is climbing and pH falling → ECCO2R. If PaO2/FiO2 <80 with refractory hypoxaemia → VV-ECMO (ECCO2R cannot oxygenate). If BOTH are failing → VV-ECMO covers both; ECCO2R does not.[12][13]
Ultra-protective ventilation + ECCO2R: the rationale is volutrauma reduction, not oxygenation. Dropping Vt from 6 to 3–4 mL/kg PBW reduces driving pressure, strain, and tidal recruitment — plausible mechanisms for less lung injury — but produces dangerous hypercapnia/acidosis that ECCO2R is recruited to clear. SUPERNOVA proved this is feasible and safe; a mortality benefit awaits a definitive trial.[9][10][21]
Anticoagulation is LESS than ECMO but NOT optional. Target ACT 150–180 (vs 180–220 for ECMO) reflects lower flow and surface area, reducing bleeding. Bein showed adding low-dose aspirin to heparin in pumpless iLA reduced circuit clotting without raising bleeding — a useful trick in pumpless configurations.[20]
Haemolysis monitoring is mandatory even on "gentle" devices. Check plasma free haemoglobin, LDH, haptoglobin, and trend haemoglobin. A rising free-Hb with falling platelets and circuit consumption mimics TMA/HUS — distinguish circuit-induced mechanical haemolysis (membrane/pump change) from true thrombotic microangiopathy.[12]
The COPD evidence is promising but unproven — sell it as "rescue, not routine." Del Sorbo (matched cohort) and Burki (pilot) showed meaningful intubation avoidance in NIV-failing hypercapnic COPD, and the UK register (Cummins) confirmed real-world feasibility — but no large RCT yet demonstrates improved survival. ECLIPSE-type trials are the inflection point.[7][8][16]
Weaning ECCO2R is the mirror of starting it — turn down sweep first, then flow. As the native lung improves (bronchodilation, antibiotics, diuresis), reduce sweep gas flow to hold PaCO2 at target, then reduce blood flow, then decannulate when PaCO2/pH are stable on minimal/no sweep for 6–12 hours. Avoid abrupt stoppage — rebound hypercapnia can occur.[15]
ECCO2R can be a step-DOWN from VV-ECMO. A patient recovering on VV-ECMO can transition to ECCO2R (lower flow, single cannula) once oxygenation is no longer the problem but CO2 clearance is still needed — a gentler bridge to extubation and decannulation, reducing ECMO-days and anticoagulation exposure.[13]
Know the devices and their niches for viva. Hemolung RAS (integrated pump-oxygenator, single 15 Fr cannula, COPD heritage); PrismaLung (bolts onto CRRT platform — leverages ICU skills, Schmidt ARDS study); iLA activve (pump-driven, higher flow possible; also offers pumpless iLA for maximal clearance/awake bridge); ProLUNG (similar concept). All = membrane + pump + cannula + sweep gas.[8][11][18]
Cost and access are the real-world limits. ECCO2R is cheaper and simpler than VV-ECMO but still expensive (circuit, cannula, specialist staff, monitoring) and available only in specialist centres — a practical reason it remains investigational even where the physiology is compelling.[13]
The exam answer on ECCO2R vs ECMO in one breath. "ECCO2R is a low-flow extracorporeal circuit that removes CO2 — not oxygen — by exploiting CO2's 20-fold higher solubility and high venous content; it uses smaller cannulae, lower flow (200–500 mL/min), and less anticoagulation than VV-ECMO, and is used for hypercapnic failure (COPD) and to facilitate ultra-protective ventilation in ARDS, but cannot treat severe hypoxaemia."[12]

Evidence summary and ongoing trials

Key ECCO2R studies (verified PMIDs)

Where the evidence is heading

Red flags (consolidated)

ECCO2R red flags — must-not-miss

ECCO2R removes CO2, NOT oxygen — never deploy it as sole support for severe hypoxaemia (use VV-ECMO).[12]
Lower flow (200–500 mL/min), smaller cannulae, less anticoagulation than VV-ECMO → fewer but not zero complications.[13]
Ultra-protective ventilation (Vt 3–4 mL/kg) is enabled by ECCO2R removing the excess CO2 — the rationale in ARDS.[9][10]
No large RCT yet — ECCO2R is investigational, specialist centres, carefully selected patients.[13]
PaCO2 not falling? think recirculation, circuit clot, cannula malposition, inadequate sweep gas flow.[15]
Femoral arterial cannula + cold pulseless leg → limb ischaemia (pumpless iLA) — distal perfusion catheter now.[18]
Rising free Hb/LDH, falling haptoglobin → haemolysis; inspect circuit, consider membrane/pump change.[12]
Dual-lumen cannula must be echo/fluoroscopy-positioned — malposition = recirculation = ineffective therapy.[12]
Monitor: PaCO2/pH (serial ABG), plasma free Hb, LDH, ACT/aPTT, haemoglobin, cannulation site, limb perfusion.[13]

Prognosis and current status

References

[1]Braune S, Kluge S ECCO(2)R in COPD exacerbation: response to comments by Del Sorbo et al. Intensive care medicine, 2016.PMID 27586995
[2]Bein T, Aubron C, Papazian L Focus on ECMO and ECCO(2)R in ARDS patients. Intensive care medicine, 2017.PMID 28717835
[3]Combes A, Fanelli V, Pham T, et al. Feasibility and safety of extracorporeal CO(2) removal to enhance protective ventilation in acute respiratory distress syndrome: the SUPERNOVA study. Intensive care medicine, 2019.PMID 30790030
[4]Staudinger T Update on extracorporeal carbon dioxide removal: a comprehensive review on principles, indications, efficiency, and complications. Perfusion, 2020.PMID 32156179
[5]Combes A, Fanelli V, Pham T, et al. Feasibility and safety of extracorporeal CO(2) removal to enhance protective ventilation in acute respiratory distress syndrome: the SUPERNOVA study. Intensive care medicine, 2019.PMID 30790030
[7]Del Sorbo L, Pisani L, Filippini C, et al. Extracorporeal Co2 removal in hypercapnic patients at risk of noninvasive ventilation failure: a matched cohort study with historical control. Critical care medicine, 2015.PMID 25230375
[8]Burki NK, Mani RK, Herth FJF, et al. A novel extracorporeal CO(2) removal system: results of a pilot study of hypercapnic respiratory failure in patients with COPD. Chest, 2013.PMID 23460154
[9]Combes A, Fanelli V, Pham T, Ranieri VM Feasibility and safety of extracorporeal CO(2) removal to enhance protective ventilation in acute respiratory distress syndrome: the SUPERNOVA study. Intensive care medicine, 2019.PMID 30790030
[10]Goligher EC, Combes A, Brodie D, et al. Determinants of the effect of extracorporeal carbon dioxide removal in the SUPERNOVA trial: implications for trial design. Intensive care medicine, 2019.PMID 31432216
[11]Schmidt M, Herisse M, Ponthus C, et al. Feasibility and safety of low-flow extracorporeal CO(2) removal managed with a renal replacement platform to enhance lung-protective ventilation of patients with mild-to-moderate ARDS. Critical care, 2018.PMID 29743094
[12]Morelli A, Del Sorbo L, Pesenti A, Ranieri VM, Fan E Extracorporeal carbon dioxide removal (ECCO(2)R) in patients with acute respiratory failure. Intensive care medicine, 2017.PMID 28132075
[13]Boyle AJ, Sklar M, McNamee JJ, et al. Extracorporeal carbon dioxide removal for lowering the risk of mechanical ventilation: research questions and clinical potential for the future. Lancet Respiratory Medicine, 2018.PMID 30484429
[14]Terragni P, Maiolo G, Ranieri VM Role and potentials of low-flow CO(2) removal system in mechanical ventilation. Current opinion in critical care, 2012.PMID 22186219
[15]Strassmann S, Mertens MC, Bickenbach J, et al. Impact of sweep gas flow on extracorporeal CO(2) removal (ECCO(2)R). Intensive care medicine experimental, 2019.PMID 30911910
[16]Cummins C, McFadyen A, Rafferty I, Poenaru D, Kinsella J A United Kingdom Register study of in-hospital outcomes of patients receiving extracorporeal carbon dioxide removal. Journal of intensive care society, 2018.PMID 29796067
[17]Fitzgerald M, Millar J, Blackwood B, et al. Extracorporeal carbon dioxide removal for patients with acute respiratory failure secondary to the acute respiratory distress syndrome: a systematic review. Critical care, 2014.PMID 25033302
[18]Zimmermann M, Bein T, Arlt M, et al. Pumpless extracorporeal interventional lung assist in patients with acute respiratory distress syndrome: a prospective pilot study. Critical care, 2009.PMID 19183475
[19]Muller T, Lubnow M, Philipp A, et al. Extracorporeal pumpless interventional lung assist in clinical practice: determinants of efficacy. European respiratory journal, 2009.PMID 19010979
[20]Bein T, Zimmermann M, Hergeth K, et al. Addition of acetylsalicylic acid to heparin for anticoagulation management during pumpless extracorporeal lung assist. ASAIO journal, 2011.PMID 21427564
[21]Worku E, Okeke C, Patel R, Goyal R, Henry M Venovenous extracorporeal CO(2) removal to support ultraprotective ventilation in moderate-severe acute respiratory distress syndrome: a systematic review and meta-analysis of the literature. Perfusion, 2023.PMID 35656595
[22]Karagiannidis C, Brodie D, Strassmann S, et al. Extracorporeal membrane oxygenation: evolving epidemiology and mortality. Intensive care medicine, 2016.PMID 26942446
[23]Westhoff M, Schonhofer B, Neumann P, et al. Severe hypercapnic respiratory failure in acute exacerbation of COPD: significance of ventilation and extracorporeal CO2 removal. Deutsche medizinische Wochenschrift, 2016.PMID 27903026
[6]Kryvenko V, Husain-Syed F, Schnell E, et al. Minimally invasive extracorporeal CO(2) removal in hypercapnic respiratory failure: a prospective observational study. Critical care (London, England), 2026.PMID 42057044