Part II — VV ECMO · Chapter 6
Evidence search date: 6 September 2026. The relationships in this chapter are mechanistic and well established; the contested part is not the physiology but what to titrate it against, which is where the 2024 evidence sits.
Clinical Question
A patient is on VV ECMO. Four numbers determine their arterial saturation, and only two of them are on the console. What are they, how do they interact, and which control actually moves which one?
Why This Matters
Chapter 2 gave the mixing equation. This chapter makes it operational — and adds the term Chapter 2 deliberately left abstract: effective extracorporeal blood flow, which is what remains after recirculation.
The practical consequence is that the pump display is not a measure of support. A circuit running 4 L/min with 25% recirculation is delivering the physiology of 3 L/min while showing 4, and no alarm will say so. Understanding VV ECMO means being able to reconstruct what the patient is actually receiving from numbers that do not report it directly.
In Series, In Parallel, or Both?
Chapter 1 described ECMO as a circuit in parallel with the patient. Several sources describe VV ECMO as being in series with the native lung. Both are correct, and the distinction is worth being precise about because it explains the whole chapter.
Physiology — parallel to the venous return, in series with the lung
The circuit draws a fraction of venous return, oxygenates it, and returns it to the venous compartment. With respect to the venous pool, that is a parallel arrangement: some blood goes through the circuit, the rest does not.
But the return is upstream of the right heart, so the mixture of circuit blood and non-circuit blood then crosses the native lung in series before it reaches the systemic arteries.
The sequence is therefore fixed:
venous return → split → partial oxygenation in the membrane lung → re-mixing in the right atrium → the whole mixture crosses the native lung → systemic arteries.
Arterial oxygen content is the pulmonary arterial content plus whatever the residual native lung adds (ECMO: A Practical Guide to Management). This is why VV ECMO cannot produce differential hypoxaemia — there is no watershed, because everything passes through the same lung — and why a patient with some surviving lung function needs far less circuit flow than one with none.
The Four Determinants of Arterial Saturation
Determinant | What moves it | Displayed on the console? | How to see it |
Effective circuit flow (total flow minus recirculation) | Pump speed, drainage cannula size and position, volume status, cannula tip separation | Only the total. Recirculation is invisible | Pre-oxygenator saturation trend; echo for tip position |
Cardiac output | Sepsis, fever, agitation, inotropes, fluid, sedation depth | No | Echo, thermodilution (with the caveats in Chapter 2), clinical trend |
Mixed venous saturation — i.e. the balance of DO₂ and VO₂ | Haemoglobin, VO₂ (fever, work of breathing, shivering, seizures), cardiac output | No — and on VV ECMO the usual sampling sites are contaminated (Chapter 2) | Trend of pre-oxygenator saturation, lactate, the Fick reasoning in Chapter 2 |
Residual native lung function | The disease, ventilator settings, recruitment, HPV state | No | The difference between predicted and actual SaO₂; compliance; imaging |
A fifth term, post-oxygenator saturation, is normally 1.00 and is therefore usually a constant rather than a variable — but when it is not, the oxygenator is failing (Chapter 32).
The Ratio That Matters
With no residual native gas exchange, arterial saturation is set by the ratio of effective circuit flow to cardiac output.
where is the recirculation fraction. That effective flow, not the console number, is what belongs in the mixing equation from Chapter 2.
Evidence — the 60% figure, from two independent sources
Chapter 2 cited a 2024 physiological review deriving that an ECBF/CO ratio of about 0.6, with venous saturation 75% and a fully saturating oxygenator, yields roughly 90% arterial saturation in complete native lung dysfunction (Tomarchio E, et al. Perfusion 2024).
The ECPR literature states the same relationship independently: "In the absence of gas exchange in the lung, the ratio of effective ECMO flow to cardiac output is the primary determinant of arterial oxygen content, with a value of 60% roughly correlating with systemic SaO₂ of 90%" — and adds the limiting case: if effective circuit flow equals cardiac output, all blood is oxygenated before reaching the right ventricle and saturation is normal even with no native lung function at all (ECPR and Resuscitative ECMO).
Certainty: physiological rationale, corroborated. Two independent expert sources deriving the same relationship from the same arithmetic. It is a model, not an outcome-validated target.
Pitfall — 0.6 is a worst case, not a prescription
The 60% figure assumes no native gas exchange whatsoever. Real patients almost always have some, and the required ratio falls accordingly.
This is why observed clinical ratios are often much lower. In the data discussed in a 2024 AJRCCM editorial, the mean extracorporeal-flow-to-cardiac-output ratio ranged from 20% to 44% across low and high venous saturation targets (Douflé G, Katira BH. Am J Respir Crit Care Med 2024). Those are not contradictory numbers — they answer a different question, in patients whose lungs still work and whose venous saturation is being targeted directly.
Use 0.6 to reason about the patient with no lung. Do not use it as a flow target for the patient in front of you.
Recirculation
Recirculation is oxygenated return blood drawn straight back into the drainage cannula. It does no harm to the patient directly — it simply does nothing — but it degrades the circuit in two separate ways at once.
Why it costs twice
- It reduces effective flow, shrinking the ratio above.
- It reduces oxygen transfer per litre of flow, because blood arriving at the membrane is already partly saturated and the membrane can only add what is missing.
The second mechanism is the less intuitive one and it is worth seeing in numbers.
Situation | Pre-oxygenator O₂ content | Post-oxygenator O₂ content | Oxygen added per dL |
Baseline (pre-oxygenator saturation 60%, PO₂ 40) | 8.3 mL/dL | 15.2 mL/dL | 6.9 mL/dL |
Recirculation (pre-oxygenator saturation 90%, PO₂ 60) | 12.4 mL/dL | 15.2 mL/dL | 2.8 mL/dL |
Haemoglobin 10 g/dL; post-oxygenator saturation 100% (PO₂ 500 mmHg) in both. Table as published in ECMO: A Practical Guide to Management, Chapter 5.
Clinical interpretation. Recirculation cut oxygen transfer per unit of blood flow by roughly 60% — with the pump display unchanged. The circuit is working exactly as hard and delivering far less.
The diagnostic rule
Clinical Pearl — the one-line test
The most practical way to evaluate recirculation is to trend pre-oxygenator saturation (SpreO₂) against arterial saturation (SaO₂). A rising SpreO₂ with a falling SaO₂ points to recirculation — and the decisive form of the rule is:
SpreO₂ can only exceed SaO₂ in the setting of recirculation. (ECMO: A Practical Guide to Management)
Nothing else produces that pattern. It is the cleanest single diagnostic sign in VV ECMO.
Worked example — the flow that isn't there
Circuit flow 4.0 L/min. Cardiac output 6 L/min. No native gas exchange. Venous saturation 75%, post-oxygenator saturation 100%.
Step | Calculation | Result |
Ratio implied by the console | 4.0 ÷ 6 | 0.67 |
Predicted SaO₂ on that basis | (0.67 × 1.00) + (0.33 × 0.75) | 92% |
Effective flow with 25% recirculation | 4.0 × (1 − 0.25) | 3.0 L/min |
True ratio | 3.0 ÷ 6 | 0.50 |
Actual SaO₂ | (0.50 × 1.00) + (0.50 × 0.75) | 88% |
Clinical interpretation. Four saturation points, invisible on every displayed parameter. The patient looks like a partial responder; the circuit looks fine. This is the gap that makes recirculation worth thinking about before escalating flow — because escalating pump speed in a recirculating circuit often increases the recirculation fraction and buys nothing.
Determinants, and what to do
Baseline recirculation varies with cannulation configuration, cannula size and positioning, the direction and speed of ECMO flow, and intracavitary pressures. Some recirculation is expected in every VV configuration and "should only be addressed when it is clinically relevant" — that is, when oxygen delivery is compromised.
When it is relevant, the levers are the determinants themselves: reposition (Chapter 3's evidence says insertion depth first for a dual-lumen cannula), reduce pump speed, correct hypovolaemia, or change configuration. Chapter 9 handles the full workup.
Pitfall — recirculation cannot be measured routinely, and the arithmetic is circular
The textbook recirculation fraction requires true mixed venous saturation — the saturation the drainage blood would have had without recirculation. On VV ECMO that value is exactly what recirculation makes unmeasurable (Chapter 2). Dedicated methods exist — ultrasound dilution and thermodilution techniques have both been described — but neither is in routine adult clinical use.
In practice recirculation is inferred, not measured: from the SpreO₂–SaO₂ relationship, from a saturation that does not respond to increased flow, and from echo. Treat any quoted recirculation percentage at the bedside as an estimate.
The Membrane Lung's Ceiling
The oxygenator is not an unlimited source of oxygen. Its capacity is described by rated flow.
Physiology — rated flow
At low blood flow there is time for full equilibration and outlet blood leaves 100% saturated. As flow rises, a point is reached where blood passes through too fast for every red cell to be oxygenated, and outlet saturation falls below 100%. Rated flow is the flow of standardised venous blood that exits the membrane lung at 95% saturation.
The amount of mixing produced by secondary flows within the fibre bundle is one of the most important determinants of maximal oxygenating capacity — which is why rated flow is an empirical device property, not something calculable from surface area alone (ELSO Red Book, 6th ed.).
A source discrepancy worth knowing. The ELSO Red Book and the ISCCM manual define standardised venous blood as Hb 12 g/dL, saturation 70%; ECMO: A Practical Guide to Management states Hb 12 g/dL, saturation 75%. The definition is otherwise identical. Use the manufacturer's stated test conditions when comparing devices, and treat cross-device comparisons made under different assumptions with caution.
Device | Surface area | Rated flow |
Getinge HLS 5.0 set advanced | 1.3 m² | 5 L/min |
Getinge HLS 7.0 set advanced | 1.8 m² | 7 L/min |
Xenios XLUNG kit 230 | 1.9 m² | 7 L/min |
Xenios iLA membrane ventilator | 1.3 m² | 4.5 L/min |
Eurosets adult | 1.81 m² | 7 L/min |
Device figures as tabulated in ECMO: A Practical Guide to Management, Chapter 5. Confirm against the current instructions for use before relying on them.
Below rated flow, oxygen delivered by the membrane is simply the outlet-minus-inlet content difference multiplied by flow. The normal outlet–inlet difference is about 5 mL/dL (ELSO Red Book, 6th ed.) — which puts the recirculation table above in context: a fall from 6.9 to 2.8 mL/dL is a fall from better-than-normal to roughly half of normal transfer.
Three things reduce oxygen transfer for the same pump speed, and only one of them is the device:
- Anaemia — fewer carriers presented to the membrane (Chapter 2, Chapter 3)
- Recirculation — inlet blood already partly saturated
- Oxygenator failure — the device itself (Chapter 32)
Clinical Pearl — the outlet gas tells you which
Post-oxygenator blood should be fully saturated. If it is not, the problem is the membrane. A rising PCO₂ in the sample taken from the oxygenator outlet indicates loss of membrane function (ISCCM Manual) — and Chapter 2 gave the one exception in which CO₂ transfer fails before oxygenation: water vapour accumulating on the gas side. That specific failure has a specific remedy — briefly increasing sweep gas flow for under a minute to clear condensed vapour through the exhaust port, without raising the gas-to-blood pressure gradient.
What Each Control Actually Does
Control | Directly changes | Does not change | Typical adult setting |
Pump speed (RPM) → blood flow | Effective circuit flow, and therefore oxygenation | CO₂ clearance, in any useful way, unless sweep is changed with it | 3.5–5 L/min in severe ARDS; 4–6 L/min for full support; 2–3 L/min if the goal is CO₂ clearance alone |
FdO₂ (oxygen fraction of sweep gas) | Post-oxygenator PO₂ — but only matters if outlet saturation is below 100% | CO₂ clearance | 1.0 for all emergent VV cases; weaned later |
Sweep gas flow | CO₂ clearance — the analogue of minute ventilation | Oxygenation, essentially not at all | Commonly initiated at about half the blood flow (e.g. 2 L/min sweep with 4 L/min blood flow), then titrated to PaCO₂ |
Ventilator settings | Native lung contribution; RV afterload; lung injury | Circuit performance | Ultra-protective — Chapter 53 |
Transfusion | Oxygen content and membrane transfer capacity | The ratio | Chapter 46 — the threshold is genuinely uncertain |
Sedation, cooling, antipyresis, NMB | VO₂, and therefore venous saturation and the second term of the mixing equation | Circuit performance | Often the fastest available intervention — Chapter 2 |
Danger — correct the CO₂ slowly
The membrane lung can normalise a PaCO₂ of 100 mmHg within minutes, and rapid falls are associated with neurological complications through abrupt changes in cerebral blood flow. Hypercapnia should be corrected over roughly 4–8 hours after ECMO initiation (ECMO: A Practical Guide to Management).
The practical expression of this is the low starting sweep. Beginning at about half the blood flow, rather than matching it, is a deliberately submaximal setting — and the reason to resist the urge to "fix" the gas on the first blood sample.
Maximum Sustainable Flow
Pump speed and delivered flow are not linearly related in the centrifugal pumps used in modern circuits. Flow is set by resistance across the whole circuit — dominated, as Chapter 3 established, by the drainage side.
For a hypoxaemic patient, blood flow is increased until arterial saturation is adequate or the maximum sustainable flow is reached. That maximum has been exceeded when:
- flows become erratic, or
- there is physical movement of the venous limb — chatter or chugging — which reflects collapse of the vena cava around the drainage cannula, or
- drainage pressures become highly negative (though no absolute threshold is established).
Where none of those signs is present, maximum achievable flow can be estimated as the point at which further increases in RPM produce little or no additional flow (ECPR and Resuscitative ECMO).
Pitfall — filling the patient to buy flow
Fluid administration may allow additional flow, and it is the reflex response to chatter. It should be used judiciously: volume overload has its own costs, and higher-than-necessary flow requires greater volume expansion, worsening extravascular lung water and complicating later diuresis and sedation reduction (Shekar K, et al. J Thorac Dis 2020).
Chatter is a signal that drainage is at its limit. The durable answers are the ones in Chapters 3 and 5 — cannula size and tip position — not a litre of crystalloid.
What Should Flow Be Titrated Against?
Evidence — target the physiology, not the flow number
A 2024 AJRCCM editorial argues explicitly against targeting raw extracorporeal blood flow rate, and for titrating support against mixed venous oxygen saturation and its physiological consequences, including right ventricular workload. Its stated priority is the oxygen delivery-to-consumption ratio: maintain DO₂:VO₂ of at least the critical threshold of 2:1, with DO₂ close to or above half of normal — around 300 mL/min/m². In the data discussed, the mean extracorporeal-flow-to-cardiac-output ratio ranged from 20% to 44% across low and high venous saturation targets. (Douflé G, Katira BH. Am J Respir Crit Care Med 2024;210(5):539–541)
Certainty: expert opinion built on physiological reasoning. An editorial, not a trial. No randomised comparison of titration targets on VV ECMO exists.
Note how exactly this lands on Chapter 2's framework. The 2:1 DO₂:VO₂ floor is the same threshold Chapter 2 derived from the critical-delivery literature, and the febrile patient in that chapter's worked example — who reached 2.2:1 with no change in any setting — is precisely the patient this editorial is asking us to notice.
A reasonable initial oxygenation goal after starting VV ECMO is an arterial saturation of 88–90%, with lower values tolerated where there is no evidence of a delivery–consumption imbalance (ECMO: A Practical Guide to Management). Flow should initially be titrated to the highest sustainable value to establish maximal delivery capacity, then reduced as the physiology allows — the daily down-titration paradigm developed in Chapter 10.
Titrating at Initiation
Clinical Pearls
- The console shows total flow; the patient experiences effective flow. The difference is recirculation, and nothing on the machine reports it.
- SpreO₂ above SaO₂ means recirculation. Nothing else produces that pattern.
- Escalating pump speed in a recirculating circuit often increases recirculation and buys no oxygenation. Reposition before you accelerate.
- The 0.6 ratio is the no-lung worst case. A patient with residual gas exchange needs far less, and observed clinical ratios are frequently half that.
- Post-oxygenator saturation should be 100%. When it is not, stop reasoning about the patient and look at the device.
- Start the sweep low. Half the blood flow, then titrate — because the membrane can correct a PaCO₂ faster than the brain can tolerate.
- Chatter is a drainage message, not a fluid prescription.
Pitfalls
- Treating pump flow as a measure of support delivered.
- Using 0.6 as a flow target rather than as a worst-case model.
- Raising FdO₂ for hypercapnia, or sweep for hypoxaemia — Chapter 1's inversion, and it survives contact with real bedsides.
- Filling the patient repeatedly to chase flow, then struggling to de-resuscitate later.
- Chasing arterial saturation above 90% when delivery is already adequate.
- Forgetting that a rising cardiac output lowers the ratio, so a patient who becomes septic on VV ECMO desaturates without anything in the circuit changing.
- Comparing rated flows between manufacturers without checking that the test conditions match.
Controversies
Controversy — what should extracorporeal blood flow be titrated against?
Clinical question: Should VV ECMO flow be set to an arterial saturation target, or to a delivery-and-consumption target?
Evidence supporting a saturation target: It is directly measured, continuously available, and understood by everyone at the bedside. Established practice initiates flow at the highest sustainable value and accepts an initial goal of 88–90%.
Evidence supporting a physiological target: Chapter 1's arithmetic showed that the last few saturation points contribute almost nothing to delivery, while haemoglobin and cardiac output dominate. The 2024 AJRCCM editorial argues for titrating against mixed venous saturation and a DO₂:VO₂ ratio of at least 2:1, noting that this can be achieved at extracorporeal-to-cardiac-output ratios as low as 20–44%. Lower flow means smaller cannulae, less volume expansion, less haemolysis and less right ventricular strain.
Current consensus: Both are used. Saturation is the operational signal; delivery is the physiological objective. Most experienced practice titrates to saturation initially and to delivery thereafter.
Practical approach: Set flow to establish adequate delivery, confirm it with lactate and venous saturation trends with a known sampling site, then reduce flow as the physiology allows rather than leaving it where it started.
Knowledge gap: No randomised comparison of titration targets on VV ECMO. The editorial position is expert reasoning, not trial evidence.
Controversy — should recirculation be measured?
Clinical question: Is it worth quantifying the recirculation fraction rather than inferring it?
Evidence supporting measurement: Recirculation systematically inflates apparent support, and the worked example above shows a 25% fraction costing four saturation points invisibly. Ultrasound dilution and thermodilution methods for quantifying it have both been described.
Evidence supporting inference: Neither method is in routine adult clinical use; the textbook calculation requires a true mixed venous saturation that recirculation itself makes unmeasurable; and the practical signal — SpreO₂ rising as SaO₂ falls — is available continuously and free. Sources also caution that some recirculation is expected in every VV configuration and should be addressed only when it is clinically relevant.
Current consensus: Infer, do not measure. Act on it when oxygen delivery is compromised, not because a number is non-zero.
Practical approach: Trend the two saturations, and when they diverge, look at cannula position before anything else.
Knowledge gap: No validated bedside method, and no evidence that any specific recirculation threshold should trigger any specific action.
Evidence Summary
Statement | Certainty | Basis |
Arterial saturation on VV ECMO is determined by effective circuit flow, cardiac output, venous saturation and residual native lung function | High — physiological rationale | Mixing arithmetic; Chapter 2; multiple concordant expert sources |
Effective flow = total flow × (1 − recirculation fraction), and effective flow is what belongs in the mixing equation | High — definitional | Tomarchio 2024; ECPR and Resuscitative ECMO |
With no native gas exchange, an effective-flow-to-cardiac-output ratio of about 0.6 corresponds to SaO₂ around 90% | Physiological rationale, independently corroborated | Tomarchio 2024 and ECPR and Resuscitative ECMO, derived separately |
SpreO₂ can exceed SaO₂ only in the presence of recirculation | High — necessary consequence of the arrangement | ECMO: A Practical Guide to Management |
Recirculation reduces oxygen transfer per unit flow as well as effective flow | High — arithmetic | Worked table reproduced from ECMO: A Practical Guide to Management |
Sweep gas governs CO₂ clearance; blood flow and haemoglobin govern oxygenation | High — physiological rationale, uncontested | ELSO Red Book; ISCCM manual; Chapter 2 |
Hypercapnia should be corrected gradually, over roughly 4–8 hours | Moderate — mechanism established, interval is expert practice | ECMO: A Practical Guide to Management; cerebral blood flow physiology |
Chatter reflects vena caval collapse around the drainage cannula and marks maximum sustainable flow | High — mechanistic | ECPR and Resuscitative ECMO |
Flow should be titrated against DO₂:VO₂ and venous saturation rather than raw flow | Expert opinion | Douflé and Katira 2024 editorial; no randomised comparison |
Any specific recirculation fraction should trigger a specific intervention | Not established | No validated bedside measurement and no threshold evidence |
Key Takeaways
- VV ECMO is parallel to the venous return and in series with the native lung. Everything mixes in the right atrium, then crosses the lung once.
- Four determinants set arterial saturation: effective circuit flow, cardiac output, venous saturation, and residual lung function. Only total flow is displayed.
- Effective flow = total flow × (1 − recirculation fraction).
- A ratio of about 0.6 gives roughly 90% saturation when the lung contributes nothing — corroborated by two independent sources, and a worst case rather than a target.
- Recirculation costs twice: less effective flow, and less oxygen added per litre. A 25% fraction can cost four saturation points invisibly.
- SpreO₂ rising while SaO₂ falls is recirculation until proven otherwise.
- Rated flow is the ceiling of the membrane lung; anaemia and recirculation reduce transfer below it without any device fault.
- Blood flow and FdO₂ for oxygen; sweep for CO₂; sedation, cooling and transfusion for the patient side of the equation.
- Start the sweep low and correct PaCO₂ over hours, not minutes.
- Titrate to delivery, not to a flow number — and then take flow back down as the lung recovers.
Key References
- Tomarchio E, Momigliano F, Giosa L, Collins PD, Barrett NA, Camporota L. The intricate physiology of veno-venous extracorporeal membrane oxygenation: an overview for clinicians. Perfusion. 2024;39(1_suppl):49S–65S. DOI: 10.1177/02676591241238156
- Douflé G, Katira BH. Extracorporeal blood flow rate: target the right thing! Am J Respir Crit Care Med. 2024;210(5):539–541. DOI: 10.1164/rccm.202403-0654ED
- Shekar K, Buscher H, Brodie D. Protocol-driven daily optimisation of venovenous extracorporeal membrane oxygenation blood flows: an alternate paradigm? J Thorac Dis. 2020;12(11):6854–6860. DOI: 10.21037/jtd-20-1515
- Parker LP, Svensson Marcial A, Brismar TB, Broman LM, Prahl Wittberg L. Hemodynamic and recirculation performance of dual lumen cannulas for venovenous extracorporeal membrane oxygenation. Scientific Reports. 2023;13:7472. DOI: 10.1038/s41598-023-34655-1
- Brodie D, Peek G, MacLaren G, et al. (eds). Extracorporeal Life Support: The ELSO Red Book, 6th edition — Chapter 5 (rated flow; outlet–inlet content difference; CO₂ transfer and gas:blood ratio)
- Taha AR, Caridi-Scheible M, Leiendecker E, et al. ECMO: A Practical Guide to Management — Chapter 5, Physiology I: Venovenous ECMO (rated flow and device table; the recirculation worked example; the SpreO₂/SaO₂ rule; cardiac output effects; oxygenation goals; rate of PaCO₂ correction)
- Shinar Z, Badulak J (eds). ECPR and Resuscitative ECMO — Chapter 15 (effective flow to cardiac output ratio and the 60% figure; pump flow non-linearity; chatter and maximum sustainable flow; initial FdO₂ and sweep settings)
- Indian Society of Critical Care Medicine. ISCCM Manual of RRT and ECMO in ICU — Chapter 26 (rated flow; oxygenator outlet PCO₂ as a failure signal; the water-vapour flush manoeuvre)
- Cross-references: Chapter 1 (the five quantities; the parallel-circuit framing), Chapter 2 (Fick, DO₂–VO₂ threshold, the mixing equation, sampling-site traps), Chapter 3 (configuration and recirculation geometry; cannula flow physics), Chapter 5 (cannulation and depth), Chapter 7 (initial management), Chapter 9 (persistent hypoxaemia — the troubleshooting counterpart to this chapter), Chapter 10 (weaning and down-titration), Chapter 32 (oxygenator failure), Chapter 53 (ventilation on ECMO)
Citation verification status. The Douflé and Katira 2024 editorial (authors, journal, volume, pages, DOI, and its stated DO₂:VO₂ and flow-ratio figures) and Shekar 2020 (authors, journal, volume, pages, DOI, and the down-titration protocol) were verified against source records during the 6 September 2026 search cycle. Tomarchio 2024 and Parker 2023 were verified in earlier cycles for Chapters 2 and 3.
All device figures, worked tables and technique statements are attributed to the named project texts and are expert-synthesis sources, not primary evidence.
Numerical audit note. The recirculation table is reproduced as published. Recalculating it with the Hüfner constant of 1.34 used elsewhere in this book gives values about 0.15 mL/dL lower; the source appears to have used approximately 1.36. The difference does not affect the conclusion.
Source discrepancy flagged. Standardised venous blood for the rated-flow definition is given as saturation 70% by the ELSO Red Book and the ISCCM manual, and as 75% by ECMO: A Practical Guide to Management. Chapter 3 quoted the 75% figure from the latter source; this chapter records both. Confirm the manufacturer's test conditions before comparing devices.
Not retrieved this cycle: the primary literature on ultrasound-dilution and thermodilution measurement of recirculation. These methods are named but no performance figures are quoted.
Educational use only. This chapter does not replace institutional ECMO protocols, local policy, specialist consultation, current guidelines or patient-specific clinical judgement.