GLIMPSE — the GLP-1 receptor agonist Management in the Peri-operative Setting project — is a national, prospective, multicentre cohort study delivered through the UK resident research networks across all four nations, and it is the largest thing anyone has done on this question. Mike and Calum take it apart the way they did ITACS: the question, the methods, the results, the critique. Then the half that wasn't in the room — what it means for cardiothoracic anaesthesia and intensive care.

Two numbers to start. One in 36 patients presenting for anaesthesia in the UK is taking a GLP-1 receptor agonist. And in those patients, the incidence of pulmonary aspiration or regurgitation was 1 in 71, against 1 in 802 in everyone else — an odds ratio of about 11.

Then the work of deciding how much of that to believe.

The framing that governs everything: the primary outcome is a prevalence, not a comparison. GLIMPSE was designed to answer "how many?" The aspiration finding is a secondary outcome, and every conclusion about the odds ratio has to sit inside that. We cover the design — each site choosing its own fortnight, screening every patient's drug history — the exclusions, the deliberate discarding of twelve sites that screened fewer than three quarters of their eligible patients, and the 97% screening rate that makes this a remarkably complete national snapshot.

The prevalence findings are more interesting than the headline. Between institutions, use ranged from 0.4% to 7.8% — a twenty-fold difference — so the national average is close to useless for planning your own list. Tirzepatide accounts for 70% and semaglutide 26%, and this is now predominantly a weight-loss population rather than a diabetes one. And the number that should change your practice this week: 41% obtained the drug from somewhere that is neither a GP nor a hospital specialist. It will not be on the GP summary, the discharge letter or the drug chart. It is only there if somebody asks the patient directly, by name.

Management was highly variable. Thirty per cent were asked to stop, most for 8–14 days — consistent with "stop it for a week", which the paper quietly demolishes: with half-lives of five and seven days, five half-lives means 25 and 35 days. A week does not eliminate the drug, and national consensus guidance says to continue it anyway. Meanwhile gastric ultrasound — the one bedside test that answers precisely the question being asked — was used in 1.3% of patients.

The safety finding gets the full appraisal, including the qualifier that changes how you read it: of the 19 events in the exposed group, all 19 involved regurgitation and only two reached the lung, which suggests the intubation, head-up positioning and pre-oxygenation were doing their job. And the finding we'd most like to change practice: more than half the events happened at emergence, not induction.

The critique covers association without any ability to adjust for confounding, observer bias inflating the ratio from both ends (with the NAP7 cross-check that partly defends it), small event numbers driving a large effect, and generalisability. The strengths get their due too.

Then the cardiothoracic half, which the paper does not address. Why our prevalence is probably higher than the national figure. Why the months between assessment and surgery make the drug-history gap worse for us. Why emergence happening on the intensive care unit, hours later, makes this a handover problem as much as an airway problem.What it means that we put a probe into the stomach of nearly every patient we anaesthetise — and that we already own the skill that was used in 1% of cases nationally. The double-lumen tube versus tube-exchange dilemma, taken from a real case in the paper. The conflict between a rapid sequence induction and haemodynamic stability in critical aortic stenosis. Why stopping the drug costs our patients more, through hyperglycaemia and sternal wound infection. And the exclusion that matters most to us: GLIMPSE excluded patients whose airway management began elsewhere — ICU transfers explicitly — so our sickest patients are entirely outside the dataset.

Chapters

• (00:00) Cold open — outside Hall 1A, minutes after the launch
• (00:50) The two headline numbers
• (01:30) Why the question needed asking
• (02:40) Methods: design, sites and the fortnight
• (03:30) The exclusions — including ICU transfers
• (04:10) The primary outcome is a prevalence, not a comparison
• (05:00) Powering, and discarding twelve sites on purpose
• (05:50) One in 36 — and the twenty-fold variation between hospitals
• (07:00) Which drugs, and why people were taking them
• (07:40) Where they got them: the 41% that isn't on any record
• (08:40) What clinicians did: stopping for a week
• (09:40) Why a week is pharmacologically meaningless
• (10:30) Airway management — and gastric ultrasound at 1%
• (11:40) The safety outcome: 1 in 71 against 1 in 802
• (12:40) Nineteen events, two aspirations — the mitigation worked
• (13:30) Emergence, not induction
• (14:20) The critique, in four parts
• (16:20) The strengths it deserves
• (17:00) Cardiothoracic: our prevalence, and the pre-assessment gap
• (18:00) Emergence on the unit — a handover problem
• (18:50) The probe we already own
• (19:30) Double-lumen tube, or exchange?
• (20:00) Rapid sequence versus haemodynamics in critical AS
• (20:30) Whether to stop the drug at all
• (21:00) The exclusion that leaves out our sickest patients
• (21:30) Wrap-up

Key takeaways

• 1 in 36 patients presenting for anaesthesia in the UK is taking a GLP-1 receptor agonist — but between hospitals the rate ranged from 0.4% to 7.8%, so find out your own number
• The primary outcome was prevalence; the aspiration comparison is a secondary outcome and should be read as such
• 41% obtained the drug from neither a GP nor a hospital specialist — ask patients by drug name, at assessment and again on the day
• Tirzepatide 70%, semaglutide 26%; 97% on weekly injections; 59% taking it for weight loss
• Aspiration and/or regurgitation occurred in 1 in 71 of exposed patients against 1 in 802 of the unexposed, odds ratio 11.4 (95%CI 6.5–20)
• All 19 events in the exposed group involved regurgitation; only two were pulmonary aspiration, giving an aspiration rate of about 1 in 674 — the mitigation appears to have worked
• More than half the events occurred at emergence, not induction — the extubation plan matters at least as much as the intubation plan
• Stopping for a week is pharmacologically meaningless against half-lives of five and seven days; national consensus guidance recommends continuing
• Gastric ultrasound was used in 1.3% of patients — a skills gap, and an open goal for a specialty that already images the stomach
• This is an association from a study that could not adjust for confounding, in which observers knew exposure status — hold it firmly but not tightly
• For cardiac patients, emergence usually happens on the ICU hours later, which makes GLP-1 use a handover item
• Stopping the drug in cardiac surgical patients risks hyperglycaemia and its associated sternal wound infection, without reliably emptying the stomach
• GLIMPSE excluded patients whose airway management began outside the procedure location, so the critical care population is not represented

References / further reading

• Potter TE, Cronin JN, Kua J, et al., on behalf of the GLIMPSE Resident Research Networks and Collaborators. Peri-operative glucagon-like peptide-1 receptor agonist use and outcomes: a national prospective multicentre cohort study. Anaesthesia 2026. doi:10.1111/anae.70380
• Annual Congress 2026, Association of Anaesthetists. Session T8a, Peri-operative medicine — "GLIMPSE: … live launch and results", Dr Tom Potter, Thursday 17 September 2026, 16:35–16:55
• Association of Anaesthetists and colleagues. Multidisciplinary consensus statement on the peri-operative management of patients receiving GLP-1 receptor agonists
• Royal College of Anaesthetists. 4th National Audit Project (NAP4)
• Royal College of Anaesthetists. 7th National Audit Project (NAP7): activity survey and peri-operative cardiac arrest

Follow the podcast

• Bluesky: @cardiacoutput.bsky.social
• X: @CardiacOutputMC

If you've found this useful, a follow or a share genuinely helps other trainees find it — and do get in touch if there's a paper you'd like taken apart.

This podcast is for medical education for healthcare professionals. It is not clinical advice. Discussion of published research reflects our own reading and interpretation — always read the primary source and follow your own centre's guidelines.

So: three questions, and every device answers all three. Which ventricle? How long is it meant to stay in? Surgical or percutaneous?

This is part two of two, following the structure of the mechanical support chapter from the fourth edition of Practical Perioperative Transoesophageal Echocardiography. Part one covered balloon pumps and ECMO.

Every assist device has the same four components — a pump providing continuous flow, an inflow cannula, an outflow cannula, and an external controller connected by a driveline — and all of them sit in parallel with the native circulation. An LVAD drains from the left ventricular apex and returns to the ascending aorta end to side; an RVAD drains from the right atrium and returns to the pulmonary artery. Temporary means days to weeks, durable means weeks to years, and those are genuinely different machines rather than the same one left in longer. The predominant durable device worldwide is now the HeartMate 3.

Then the word that causes the most trouble: BiVAD. Two quite different situations get called the same thing. A true biventricular assist device means two devices implanted for long-term biventricular support, and that is occasional. Whereas a patient receiving a durable LVAD not infrequently needs a temporary RVAD at the same operation, because the right heart cannot cope — two devices, but one durable and one coming out. BiVAD describes the anatomy and tells you nothing about the timescale, which is precisely why people get confused.

With that grid in place, the rest follows the chapter. Before implantation: ventricular assessment, intracardiac thrombus, the valves, the aorta, and intracardiac shunts. Shunts matter because of a pressure change you are about to create — once support starts, left atrial pressure falls below right atrial pressure, so a patent foramen that has been silent for a lifetime can shunt right to left, giving arterial hypoxaemia or systemic embolisation of right-sided gas or thrombus. And because a foramen can be genuinely hard to detect before support starts, you re-examine the atrial septum afterwards.

Aortic regurgitation gets the attention it deserves, because it is the lesion that quietly destroys LVAD efficacy: blood leaves the outflow cannula, flows back across the incompetent valve, and returns straight into the inflow cannula — a circuit inside the chest that never reaches the patient. How you correct it depends on the intention of the device, why a bioprosthesis is preferred if replacement is needed, why colour Doppler underestimates it in end-stage failure and you should therefore assess on bypass, which measures work and which do not — and a free clue that costs nothing, in the left ventricular vent flows during implantation. Plus the exception on aortic stenosis that matters only for partial-support devices, and the mitral and tricuspid lesions corrected at the same operation.

After implantation: de-airing, and then the single most useful structure on the screen. The interventricular septum should be flat and neutral. Marked rightward displacement means the ventricle is inadequately decompressed — underpumping. Marked leftward means it has collapsed — the suction event, or ventricular suckdown. That is how you run a ramp study, and it is the same logic you will use later for the Impella.

Right ventricular function afterwards gets four reasons why it does not always improve despite reduced afterload, the echo signs of acute right ventricular failure, and then — in its proper place — temporary right ventricular support: either a surgically grafted CentriMag, or the Protek Duo, a percutaneous dual-lumen cannula from the right internal jugular with its inflow in the right atrium and its outflow in the proximal main pulmonary artery.

We then cover aortic valve opening and the HeartMate 3's programmed rhythmic flow changes, cannula assessment including the biplane tip for confirming inflow orientation, early and late causes of obstruction, why you reinterrogate both cannulas at chest closure, how to read the echo alongside the console, and the surveillance schedule.

The episode closes with temporary ventricular assist devices as their own topic — the Impella. Indications, the models and their licensed durations, the axial Archimedes screw, the left- and right-sided configurations, the contraindications to rule out first, the insertion sequence including the measurement trap of the pigtail tip, and the ongoing assessment, which comes back to exactly the same septum.

Chapters

• (00:00) Cold open — sorting out the vocabulary first
• (00:50) The four components every device shares
• (01:20) Which ventricle, and for how long
• (02:30) BiVAD — two situations, one word
• (03:20) Surgical or percutaneous
• (04:00) The rule that runs underneath, and where this comes from
• (04:40) Before implantation: the five headings
• (05:10) Intracardiac shunts, and the pressure change you create
• (06:30) Aortic regurgitation, and recirculation inside the chest
• (07:40) Why you grade it on bypass
• (08:40) The vent flow clue
• (09:10) Aortic stenosis, mitral stenosis and tricuspid regurgitation
• (10:00) Intracardiac thrombus
• (10:30) Right ventricular function, predictors and the aorta
• (11:40) De-airing
• (12:20) The septum: flat, rightward, leftward
• (13:40) Ramp studies
• (14:30) Why the right ventricle doesn't always improve
• (15:40) Signs of acute right ventricular failure
• (16:20) Temporary RVAD: the CentriMag and the Protek Duo
• (17:30) Aortic valve opening, and the HeartMate 3 artificial pulse
• (18:10) Assessing the cannulas, and the biplane tip
• (19:20) Obstruction, early and late
• (20:00) Reading the console alongside the echo
• (20:40) Complications, tamponade and surveillance
• (21:30) Temporary devices: the Impella
• (22:10) Contraindications and insertion
• (22:50) Ongoing assessment and what goes wrong
• (23:20) Wrap-up

Key takeaways

• Classify every device by three questions: which ventricle, how long, surgical or percutaneous
• All assist devices share four components — pump, inflow cannula, outflow cannula, and an external controller on a driveline — and sit in parallel with the native circulation
• LVAD: LV apex to ascending aorta, end to side. RVAD: right atrium to pulmonary artery
• Temporary means days to weeks; durable means weeks to years. The HeartMate 3 is the predominant durable device worldwide
• A true BiVAD is two devices for long-term biventricular support and is occasional — distinct from the common situation of a temporary RVAD at the time of durable LVAD implantation
• Once LVAD support starts, left atrial pressure falls below right atrial pressure, so a silent PFO or ASD can shunt right to left — re-examine the atrial septum after support begins
• Aortic regurgitation causes LVAD recirculation: outflow cannula, back across the valve, straight into the inflow cannula
• Correction depends on intention — repair for bridge to recovery, suture closure for transplant or destination therapy, and a bioprosthesis if replacement is needed
• Colour Doppler underestimates AR in end-stage failure because diastolic pressures equalise; assess on cardiopulmonary bypass and again at different pump speeds
• Use vena contracta width and jet-to-outflow-tract ratio; pressure half-time and holodiastolic flow reversal are unreliable here
• High LV vent flows on bypass may point to clinically significant AR
• Aortic stenosis usually doesn't matter — except for partial-support devices requiring ongoing ejection
• The interventricular septum should be flat and neutral: rightward means underpumping, leftward means suckdown
• Ramp from minimum speed, aiming for a flat septum, a filled but not overdistended ventricle, and the aortic valve opening at least once every 3–5 beats
• RV function doesn't always improve after LVAD: bypass, elevated PVR, a suddenly normalised preload, and altered LV geometry all work against it
• Temporary right ventricular support is either a surgically grafted CentriMag or a percutaneous Protek Duo — right internal jugular, inflow in the right atrium, outflow in the proximal main pulmonary artery, up to about 4.5 L/min
• Image the inflow cannula in two orthogonal long-axis views with biplane imaging, and reinterrogate both cannulas at chest closure
• Low cardiac output with low device flows and a high central venous pressure is tamponade until echo says otherwise
• The Impella is a temporary percutaneous device supporting either ventricle, with an inlet in the ventricle and outlet beyond the valve; optimal inlet position about 3.5 cm from the aortic valve, and the pigtail tip is not included in that measurement
• Any deterioration in a patient on mechanical support gets a TOE, looking specifically for cannula malposition or occlusion

References / further reading

• Charlesworth M, Allen SJ. Echocardiography for mechanical support (Chapter 20). In: Lambert AS, Allen SJ, Sidhu S, eds. Practical Perioperative Transoesophageal Echocardiography, 4th edn. Oxford: Oxford University Press, 2025. ISBN 9780198873686
• Stainback RF, Estep JD, Agler DA, et al. Echocardiography in the management of patients with left ventricular assist devices: recommendations from the American Society of Echocardiography. J Am Soc Echocardiogr 2015; 28: 853–909
• Mehra MR, Uriel N, Naka Y, et al. A fully magnetically levitated left ventricular assist device — final report. N Engl J Med 2019; 380: 1618–27
• Crowley J, Cronin B, Essandoh M, D'Alessandro D, Shelton K, Dalia AA. Transesophageal echocardiography for Impella placement and management. J Cardiothorac Vasc Anesth 2019; 33: 2663–8
• Potapov EV, Stepanenko A, Dandel M, et al. Tricuspid incompetence and geometry of the right ventricle as predictors of right ventricular function after implantation of a left ventricular assist device. J Heart Lung Transplant2008; 27: 1275–81
• Chumnanvej S, Wood MJ, MacGillivray TE, Melo MF. Perioperative echocardiographic examination for ventricular assist device implantation. Anesth Analg 2007; 105: 583–601

Follow the podcast

• Bluesky: @cardiacoutput.bsky.social
• X: @CardiacOutputMC

If you've found this useful, a follow or a share genuinely helps other trainees find it — and do get in touch if there's a topic you'd like covered.

This podcast is for medical education for healthcare professionals. It is not clinical advice. Practice described reflects local Wythenshawe practice at the time of recording — always follow your own centre's guidelines and current local policy.

The argument is that in mechanical circulatory support, transoesophageal echo is not an investigation you order. It is part of the device. It confirms the indication, guides the cannulas in, optimises the support once it is running, and finds the complication before the patient tells you about it. And one rule runs through the whole topic: any deterioration in a patient on mechanical support gets a TOE, with specific attention to cannula malposition or occlusion.

This is part one of two, covering balloon pumps and ECMO. Part two takes on assist devices and the Impella.

We start with the balloon pump, where TOE has real advantages over the chest film and fluoroscopy — real-time structures, portable, no ionising radiation. Scan before you insert, for two reasons: to confirm the indication, and to exclude the absolute contraindications, which are severe aortic regurgitation, severe atheroma in the descending thoracic aorta, aortic dissection, and severe distal occlusive disease. We explain why regurgitation is so absolute — the balloon inflates in diastole to augment diastolic pressure, so a leaking valve means you drive that augmented pressure straight back into the ventricle, worsening the regurgitation without improving coronary perfusion pressure. The device fails at its own job. Then the relative contraindications, including dynamic left ventricular outflow tract obstruction, where reducing afterload makes systolic anterior motion worse.

Positioning gets done properly: the tip about two centimetres distal to the origin of the left subclavian, with the proximal balloon above the diaphragm — too high and you occlude the subclavian, too low and you lose augmentation and may intermittently occlude the mesenteric or renal arteries. How to find the subclavian origin in the upper oesophageal arch short axis, how to tell the echo-dense tip and lucent balloon body from the wire, and what to use as a landmark when you genuinely cannot see the origin. Plus a technique that needs no technology at all: identify the balloon tip in the descending aorta short axis, put your fingertips on the probe at the level of the teeth, withdraw until the subclavian origin appears — and the distance your fingers have travelled is the distance from tip to subclavian. You have measured it with your hand.

The ECMO section opens with the decision that matters most, which is mode — and the most useful piece of thinking in the episode. When a severe respiratory failure patient becomes haemodynamically unstable, it is usually one of four things: right ventricular dysfunction from aggressive ventilation and acutely raised pulmonary vascular resistance, ventricular dysfunction from sepsis or underlying cardiac disease, vasodilation from sepsis, or hypovolaemia. Most of those do not need arterial support. Ventilator-induced right ventricular dysfunction usually resolves once ECMO lets you drop to rest settings, and vasodilation is usually managed with VV plus vasopressors. So instability alone does not mean VA — it means work out the mechanism first, and echo is how you do that.

Then what to look for before you commit: significant aortic or mitral regurgitation, which can cause acute left ventricular distension in a VA patient who is barely ejecting; severe aortic disease as a contraindication to VA; severe atheroma as a relative one, because of the sandblasting effect of the return jet; and the precise position on a PFO, which may complicate VV but is not a contraindication.

Cannula positioning is laid out by configuration, because the rules are clean once you separate them — confirming the guidewire before you dilate, where the return tip belongs in VV, where the drainage tip goes for jugular–femoral versus femoral–femoral versus a double-lumen cannula, and why VA is simpler. Then recirculation: how to diagnose it with 2D and colour Doppler, the two findings that confirm it, and how to fix it under echo guidance.

We finish with the complications that are hard to see coming. The specific screen appearance of left ventricular distension on VA and the two families of solution. Circuit and intracardiac thrombus. And tamponade after cardiac surgery, which is genuinely difficult because the circuit largely bypasses flow through the heart, so it is initially well tolerated and you never get the classic picture — until the right atrium collapses and your circuit flow falls. Raised central venous pressures with low circuit flows on central VA means tamponade until proven otherwise. Plus the TOE predictors associated with successful weaning from VA.

Chapters

• (00:00) Cold open — the tool that makes the machines safe
• (00:50) Any deterioration on support gets a TOE
• (01:20) The balloon pump: why scan before you insert
• (02:00) Absolute contraindications, and why AR is absolute
• (03:10) Relative contraindications, including dynamic LVOT obstruction
• (04:00) The pre-insertion study
• (04:50) Positioning the tip — and what happens if it's wrong
• (05:40) Finding the subclavian origin
• (06:30) Measuring the distance with your fingers
• (07:20) Post-insertion complications, and is it working?
• (08:00) ECMO: choosing the mode
• (08:40) The four causes of instability in respiratory failure
• (09:50) Why instability alone doesn't mean VA
• (10:30) Lesions that will complicate things — and the PFO
• (11:30) Using the probe during cannulation
• (12:10) Cannula positions by configuration
• (13:10) Recirculation: diagnosis and fix
• (14:10) Left ventricular distension on VA
• (15:10) Thrombus
• (15:40) Tamponade — why it hides on central VA
• (16:30) Weaning, and the TOE predictors
• (17:20) Wrap-up

Key takeaways

• In mechanical circulatory support, TOE is part of the device rather than an investigation you order
• Any deterioration in a patient on mechanical support gets a TOE, looking specifically for cannula malposition or occlusion
• Absolute contraindications to an IABP on echo: severe aortic regurgitation, severe descending aortic atheroma, dissection, and severe distal occlusive disease
• Aortic regurgitation is absolute because augmenting diastolic pressure across a leaking valve worsens the regurgitation without improving coronary perfusion pressure
• Reducing afterload worsens dynamic LVOT obstruction and systolic anterior motion, so that is a relative contraindication
• Position the balloon tip about 2 cm distal to the left subclavian origin, with the proximal balloon above the diaphragm
• You can measure tip-to-subclavian distance by walking the probe back with your fingers at the level of the teeth
• ECMO mode follows the mechanism: instability alone does not mean VA
• Ventilator-induced RV dysfunction and septic vasodilation are usually managed on VV, with rest settings and vasopressors
• Assess both the aortic and mitral valves before VA, because regurgitation predisposes to acute LV distension in a barely ejecting ventricle
• Severe aortic dissection or aneurysm contraindicates VA; severe atheroma is relative, because of the sandblasting effect of the return jet
• A PFO may complicate VV ECMO but is not a contraindication to ECMO
• Confirm the guidewire with the probe before dilating and passing a large-bore cannula
• In VV, the return tip sits in the right atrium with the jet aimed at the tricuspid valve, and the tips must be kept apart to avoid recirculation
• Diagnose recirculation by finding both tips on 2D, then looking for a high-velocity jet down the IVC or a jet that includes the drainage tip
• LV distension on VA looks like a distended ventricle, minimal contractility, a closed aortic valve, a distended left atrium and significant mitral regurgitation, with a non-pulsatile arterial trace
• Severe distension causes pulmonary oedema, delays recovery and predisposes to intracavity thrombus
• Tamponade is poorly detected on central VA because the circuit bypasses the heart — raised venous pressures with low circuit flows means tamponade until proven otherwise
• Have a low threshold for rescanning after starting ECMO: complications are common and hard to detect early

References / further reading

• Charlesworth M, Allen SJ. Echocardiography for mechanical support (Chapter 20). In: Lambert AS, Allen SJ, Sidhu S, eds. Practical Perioperative Transoesophageal Echocardiography, 4th edn. Oxford: Oxford University Press, 2025. ISBN 9780198873686
• Douflé G, Roscoe A, Billia F, Fan E. Echocardiography for adult patients supported with extracorporeal membrane oxygenation. Crit Care 2015
• Platts DG et al. The role of echocardiography in the management of patients supported by extracorporeal membrane oxygenation. J Am Soc Echocardiogr 2012
• Hahn RT et al. Guidelines for performing a comprehensive transesophageal echocardiographic examination. J Am Soc Echocardiogr 2013
• Truby LK et al. Left ventricular distension and venoarterial extracorporeal membrane oxygenation. ASAIO J 2017

Follow the podcast

• Bluesky: @cardiacoutput.bsky.social
• X: @CardiacOutputMC

If you've found this useful, a follow or a share genuinely helps other trainees find it — and do get in touch if there's a topic you'd like covered.

This podcast is for medical education for healthcare professionals. It is not clinical advice. Practice described reflects local Wythenshawe practice at the time of recording — always follow your own centre's guidelines and current local policy.

This is cardiopulmonary bypass top to bottom, pitched at FRCA level and a little beyond, and built around the questions that have actually appeared in past papers. Almost everything else we discuss on a cardiac unit assumes this episode.

Please note: the thresholds, ACT practice and doses are Wythenshawe-specific local practice. Take the principles, and check your own guidelines.

We start with the circuit, answering the examiner's real question component by component. The reservoir is a volume buffer — somewhere to hold blood so flow is maintained when the surgeon lifts the heart and venous return suddenly falls — and it is where air gets trapped rather than delivered. The heat exchanger is the patient's entire thermoregulation for the duration, not an accessory. The arterial filter is about the brain, because embolic load is one of the two mechanisms behind neurocognitive injury after bypass, and hypoperfusion is the other.

Then cannulation, including the one absolute rule in the whole chapter: never site a right radial arterial line if the axillary artery is going to be cannulated, because clamping it takes out your trace for a prolonged period and you will be flying blind at the worst possible moment. Plus the circuit variants worth knowing — hybrid circuits switchable between full bypass and VA-ECMO for lung transplantation, haemoadsorption for the septic patient, and ultrafiltration, which must be handed over explicitly to intensive care rather than buried in the chart. And a question most people are never taught properly: how is the patient anaesthetised on bypass? An isoflurane vaporiser on the circuit, capnography switched to the oxygenator — and depth monitoring throughout, because the transitions on and off pump are exactly where awareness happens.

Haemodilution gets a proper answer rather than a hand-wave. The haematocrit falls the moment bypass starts because you have connected the patient to a crystalloid prime, and it is tolerated because haematocrit is two competing things at once: oxygen-carrying capacity, which you want high, and viscosity, which you want low — especially when cold. Hence a target somewhere around 21–24%, and retrograde autologous priming to reduce the crystalloid load in selected patients.

Anticoagulation is given both ways. Heparin 300–400 units/kg and an ACT over 480 seconds is the version the exam wants; locally the threshold is over 400, because our machines double-count. Then the scenario: the ACT is 210 after a full dose. Don't give more heparin. Confirm it reached the circulation — go back to the lumen you used and aspirate it, because a surprising amount of this is a delivery problem rather than a pharmacology one. Then think heparin resistance, which almost always means acquired antithrombin deficiency, and give antithrombin III 500 units.

Cardioplegia covers the mechanism precisely — a high extracellular potassium depolarises the membrane, inactivating the fast inward sodium channels so there is no action potential upstroke — the five things cardioplegia actually achieves, the six advantages of blood over crystalloid (the exam wants three), warm versus cold, and the three indications for retrograde delivery with the 40 mmHg coronary sinus limit. Then the material that takes you past Fellowship: del Nido, and why Custodiol arrests the heart by hyperpolarisation rather than depolarisation. Opposite direction, same destination.

Hypothermia is delivered in the shape the question asks for — four advantages and three disadvantages — including the paradox worth saying out loud, that you are cooling to reduce oxygen demand while simultaneously making oxygen harder to offload. Plus flow and pressure targets, and alpha-stat versus pH-stat.

We finish with an eight-step checklist for separation from bypass, the vasoplegia that commonly follows, and protamine: the electrostatic reversal, why you must account for the heparin in the prime as well as the dose you gave, the four groups of adverse reaction with the Horrow type III as the one to fear, and why too much protamine is itself an anticoagulant.

Chapters

• (00:00) Cold open — "draw me a bypass circuit"
• (01:00) What each component does to the patient
• (03:10) Cannulation, and the right radial rule
• (04:20) Circuit variants: hybrid ECMO, haemoadsorption, ultrafiltration
• (05:40) How is the patient anaesthetised on bypass?
• (06:50) Priming, and why the haematocrit falls
• (08:20) Heparin, the ACT, and what the ACT is actually for
• (09:50) The ACT is 210 — what do you do, in order?
• (11:20) Cardioplegia: how potassium arrests the heart
• (12:30) Blood versus crystalloid, warm versus cold
• (13:20) Antegrade and retrograde delivery
• (14:10) del Nido, Custodiol and hyperpolarising arrest
• (15:20) Laplace, and why we vent
• (15:50) Hypothermia: four advantages, three disadvantages
• (17:00) Flow, pressure, alpha-stat and pH-stat
• (17:40) Coming off bypass — the eight-step checklist
• (19:10) Vasoplegia
• (19:40) Protamine: pharmacology, reactions and technique
• (20:40) What arrives on the unit afterwards
• (21:40) Wrap-up

Key takeaways

• Name the parts, then say what each does to the patient: the reservoir is a volume buffer and an air trap, the heat exchanger is the patient's thermoregulation, the arterial filter protects the brain
• Neurocognitive injury after bypass comes from embolic load and hypoperfusion
• Never site a right radial arterial line if the axillary artery is to be cannulated
• Hybrid circuits can switch between full bypass and VA-ECMO — used for lung transplantation, because it avoids full-dose anticoagulation and lets the heart keep ejecting
• If ultrafiltration has been used, hand that over explicitly — over-filtration makes postoperative fluid and electrolyte management difficult
• Monitor depth of anaesthesia throughout: going on and coming off pump disrupts both volatile and intravenous delivery, and that is where awareness happens
• The haematocrit falls because of the crystalloid prime, and it is tolerated because haematocrit is oxygen carriage and viscosity at the same time — target around 21–24%
• Heparin 300–400 units/kg with an ACT over 480 seconds for the exam; over 400 locally because the machines double-count
• The ACT's genuinely useful role is confirming it is safe to turn the suckers on and go onto bypass
• If the ACT won't rise, don't give more heparin — confirm it reached the patient, then give antithrombin III 500 units for acquired antithrombin deficiency
• Potassium cardioplegia depolarises the myocyte membrane and inactivates the fast inward sodium channels, abolishing the action potential upstroke
• Blood cardioplegia adds oxygen carriage, hydrogen ion buffering, free-radical scavenging, improved microvascular flow, reduced myocardial oedema and nutrient delivery
• Retrograde delivery is needed for significant aortic regurgitation, root surgery and severely diseased coronaries — keep coronary sinus pressure below 40 mmHg
• Custodiol arrests the heart by hyperpolarisation through sodium depletion, the opposite mechanism to potassium depolarisation
• Laplace's law is the basis of every unloading strategy: venting, the IABP, and LV decompression on VA-ECMO
• Hypothermia reduces cerebral and myocardial oxygen consumption but shifts the oxyhaemoglobin curve left, so offloading is impaired at the same time as demand falls
• Separation from bypass in eight steps: warm, rhythm and AV synchrony, electrolytes and acid–base, lungs, de-air under TOE, rate to suit the lesion, titrate support, separate incrementally
• Protamine 300 mg reverses roughly 30,000 units of heparin — and you must account for the heparin in the pump prime too
• The Horrow type III reaction is pulmonary hypertension with right heart strain, and excess protamine is itself an anticoagulant

References / further reading

• Charlesworth M. CTCCU Handbook, 2nd edition, sections 2b and 2d
• Wahba A et al. EACTS/EACTAIC/EBCP Guidelines on cardiopulmonary bypass in adult cardiac surgery. Eur J Cardiothorac Surg
• Murphy GS, Hessel EA, Groom RC. Optimal perfusion during cardiopulmonary bypass: an evidence-based approach. Anesth Analg 2009
• Ltaief Z et al. Pathophysiology and clinical implications of vasoplegic syndrome after cardiopulmonary bypass. J Clin Med 2022
• Finley A, Greenberg C. Heparin sensitivity and resistance: management during cardiopulmonary bypass. Anesth Analg 2013
• Boer C et al. 2017 EACTS/EACTA Guidelines on patient blood management for adult cardiac surgery. J Cardiothorac Vasc Anesth 2018

Follow the podcast

• Bluesky: @cardiacoutput.bsky.social
• X: @CardiacOutputMC

If you've found this useful, a follow or a share genuinely helps other trainees find it — and do get in touch if there's a topic you'd like covered.

This podcast is for medical education for healthcare professionals. It is not clinical advice. All thresholds, doses and practice described reflect local Wythenshawe practice at the time of recording — always follow your own centre's guidelines and current local policy.

This is the intra-aortic balloon pump, top to bottom, pitched at Final FRCA and a bit past it — it has appeared in past papers three separate ways, as principles, as indications and contraindications, and as complications. But the version worth having is the one where you can look at a trace on the unit and know what's wrong with it.

Please note: the weaning approach and anticoagulation discussed are Wythenshawe-specific local practice. Take the principles, and check your own guidelines.

We start with helium, and the two reasons to give rather than one — it's low density, so it shuttles down a long narrow catheter fast enough to work inside a fraction of a cardiac cycle, and it's highly soluble in blood, so rupture is far more forgiving than air would be. Then counterpulsation, and the sentence the whole device hangs on: the balloon pump increases myocardial oxygen supply and reduces demand at the same time, which almost nothing else does. Give adrenaline to an ischaemic ventricle and coronary perfusion may improve, but rate, contractility and wall stress have all gone up, so you've bought a little flow at a large metabolic price. Diastolic augmentation raises coronary perfusion pressure, presystolic deflation drops aortic end-diastolic pressure and therefore afterload and wall stress — supply up, demand down.

Then the trace, as a learnable set piece. The rule first: put the pump on 1:2, so every other beat is unassisted and you have a control sitting next to your test. Then three comparisons in order — augmented diastolic higher than unassisted systolic, assisted end-diastolic lower than unassisted end-diastolic, and assisted systolic lower than unassisted systolic. That last one catches people every time: a lower assisted systolic pressure is not the pump failing, it's direct evidence you have unloaded the ventricle.

The four timing errors get sorted by harm rather than by name, which is the distinction that shows understanding rather than recall. Early inflation and late deflation both load the ventricle during ejection and are the dangerous pair — late deflation worst of all, because the ventricle is ejecting against an inflated balloon. Late inflation and early deflation merely waste benefit, although early deflation can drive retrograde coronary and carotid flow and cause angina. Then triggers, the asynchronous mode and why you'd ever want it, and why arrhythmia is the balloon pump's great enemy.

Indications and contraindications follow, including the most satisfying piece of physiology in the episode: why a balloon pump helps in acute severe mitral regurgitation. The ventricle has two exits, and how much blood goes each way depends on the relative resistance of the two routes — so dropping aortic end-diastolic pressure makes the forward path easier, the regurgitant fraction falls, and forward output rises. You're not fixing the valve, you're changing the arithmetic while somebody organises theatre. And on the other side, why aortic regurgitation is an absolute hard stop: everything the balloon does in diastole raises aortic root pressure, so in an incompetent valve you are augmenting the leak straight back into a failing ventricle.

Then placement and the landmarks that matter, what TOE adds, the daily chest film, complications split into insertion, use and removal, and back to the 2am patient — falling urine output means think down, a lost left radial pulse means think up, and helium or blood in the tubing means rupture and it comes out now.

We finish on the trap. IABP-SHOCK II was negative, and Altshock-2 in 2025 was stopped for futility in heart failure–related shock. So why is there one running in bed four? The answer isn't to ignore the evidence — it's to notice what those trials actually studied, which was routine, unselected use in two specific shock populations. That is a different question from the patient who cannot come off bypass, the one with acute severe mitral regurgitation waiting for theatre, or the one who needs to survive four hours until the cath lab.

Chapters

• (00:00) Cold open — the augmented pressure has dropped
• (01:00) Why this episode, and what level we're pitching at
• (01:40) What it actually is, and why helium — two reasons
• (02:45) Counterpulsation: supply up and demand down at once
• (03:45) Inflation, and why the left ventricle is perfused in diastole
• (04:15) Deflation, afterload and wall stress
• (05:15) The trace — and why you put it on 1:2
• (06:05) Reading a pair of beats
• (07:35) Lower is better: the comparison everyone misreads
• (08:05) The four timing errors, sorted by harm
• (10:35) Triggers, and the asynchronous mode
• (11:15) Arrhythmia and tachycardia
• (12:05) Indications
• (12:35) Acute mitral regurgitation, and post-infarct VSD
• (13:25) Contraindications — and why aortic regurgitation is absolute
• (15:15) Insertion, TOE, and the daily film
• (16:55) Complications: insertion, use, removal
• (18:35) Back to the 2am patient — migration up and down
• (19:55) Balloon rupture
• (20:15) Anticoagulation and weaning
• (21:15) IABP-SHOCK II, Altshock-2, and the trap
• (24:15) Wrap-up

Key takeaways

• Helium for two reasons: low density so it shuttles fast, and high solubility so rupture is survivable
• The balloon occupies 80–90% of the aortic cross-section, inflating in diastole and deflating before systole
• It raises myocardial oxygen supply and lowers demand simultaneously — almost nothing else does
• Diastolic augmentation raises coronary perfusion pressure; presystolic deflation drops aortic end-diastolic pressure, and therefore afterload, wall stress and demand
• Read the trace on 1:2 so you have an unassisted control beat next to your assisted one
• Augmented diastolic should exceed unassisted systolic — the tallest thing on the screen should be the balloon, not the heart
• A lower assisted systolic pressure is good: it's evidence of unloading, not of pump failure
• Early inflation and late deflation load the ventricle during ejection and are the dangerous errors; late inflation and early deflation waste benefit, though early deflation can cause retrograde coronary flow and angina
• Arrhythmia is the balloon pump's great enemy; above a rate of about 120 you may get better support at 1:2 than 1:1
• In acute severe mitral regurgitation, reducing aortic end-diastolic pressure shifts flow from the regurgitant route to the forward one
• Aortic regurgitation is an absolute contraindication because you would be augmenting the leak into an already failing ventricle — along with dissection, aneurysm, a prosthetic aorta, and futility
• Tip at the level of the carina and at least 2 cm below the aortic knob, just distal to the left subclavian, balloon above the diaphragm — checked every single day
• Falling urine output or a distended abdomen means think downward migration; a lost or damped left radial trace means think upward
• Helium or blood in the tubing means rupture: stop and remove urgently, before clot inside the balloon entraps it
• Wean by ratio or augmentation percentage, judged on haemodynamics, urine output and lactate rather than on the machine
• IABP-SHOCK II and Altshock-2 were negative for routine use in specific shock populations — which is not the question you are usually asking on a cardiac unit

References / further reading

• Thiele H et al. Intraaortic balloon support for myocardial infarction with cardiogenic shock (IABP-SHOCK II). N Engl J Med 2012; 367: 1287–96
• Thiele H et al. Intra-aortic balloon pump in cardiogenic shock complicating acute myocardial infarction: long-term 6-year outcome of the randomised IABP-SHOCK II trial. Circulation 2019
• Early intra-aortic balloon support for heart failure-related cardiogenic shock (Altshock-2): a randomised clinical trial. J Am Coll Cardiol 2025
• Møller JE et al. Microaxial flow pump or standard care in infarct-related cardiogenic shock (DanGer Shock). N Engl J Med 2024
• Thiele H et al. Extracorporeal life support in infarct-related cardiogenic shock (ECLS-SHOCK). N Engl J Med2023
• Byrne RA et al. 2023 ESC Guidelines for the management of acute coronary syndromes. Eur Heart J 2023
• Krishna M, Zacharowski K. Principles of intra-aortic balloon pump counterpulsation. Contin Educ Anaesth Crit Care Pain 2009

Follow the podcast

• Bluesky: @cardiacoutput.bsky.social
• X: @CardiacOutputMC

If you've found this useful, a follow or a share genuinely helps other trainees find it — and do get in touch if there's a topic you'd like covered.

This podcast is for medical education for healthcare professionals. It is not clinical advice. Weaning and anticoagulation practice described reflects local Wythenshawe practice at the time of recording — always follow your own centre's guidelines and current local policy.

ITACS, published in The BMJ in August 2026, is the best answer we have ever had to that question. It is also a trial with a great deal to teach about how to read a paper properly — so this is a journal club, and we go through the methods slowly, because that is where the meaning lives.

The question first. A third of patients coming for cardiac surgery are anaemic, twenty to fifty per cent are transfused, and both are powerfully associated with complications, longer stays and death — in a specialty that consumes around ten per cent of the entire NHS blood supply. Correcting the anaemia in clinic attacks both problems at once. But anaemia may be a marker as well as a mechanism: if patients do badly because of the kidney disease or inflammation that made them anaemic, then fixing the number fixes the screen and changes nothing about the patient. A strong prognostic marker is not automatically a treatment target — and that idea runs through the whole episode.

Then the methods. Thirty-three hospitals across ten countries, 955 anaemic adults for elective cardiac surgery, a single 1,000 mg dose of intravenous iron or placebo one to twenty-six weeks beforehand — and, deliberately, no requirement to prove iron deficiency. We look at how you blind a brown drug (a black syringe and an opaque line), and at the detail that separates a good trial from a trial that merely says "double blind" in its abstract: they checked whether the masking had worked by asking patients to guess their allocation.

The primary outcome is days alive and at home at 90 days — what it captures, and the two things it hides. And then the finding that isn't in the abstract at all: the primary outcome was originally days at home at thirty days, and was amended to ninety in December 2020. We give the defence and the concern, and then we look at what the original outcome showed. It was null. Had the investigators kept it, this would be a negative trial, and that belongs in any honest summary of the paper.

The results deserve care. The iron worked biochemically — ferritin rose from around 110 to over 500 — but haemoglobin rose by under 4 g/L, and three quarters of treated patients were still anaemic on the day of surgery, which explains the size of everything that follows. The primary result is one day, with a confidence interval touching zero, in a trial powered for a day and a half. Transfusion is the solid finding: 68% down to 61%, about fifteen patients treated to prevent one transfusion, and roughly 44 units of blood saved per hundred patients. But length of stay was identical, complications were identical — so where did the extra day come from? The paper answers it, and the answer is a smaller claim than the headline sounds.

The best thing in the trial isn't in the abstract either. Rather than only reporting the median, the investigators reported the treatment effect across the whole distribution — and it turns out that patients who recovered well gained a fraction of a day, while at the 25th centile the difference was 6.1 days. All the benefit sits with the patients who did badly. It is post hoc, it was requested at peer review, and one of its confidence intervals is enormous — but it is biologically coherent, and it may be the most important idea to come out of this trial.

We also correct an argument we would have made before reading the full paper. The obvious criticism — that enrolling patients without proven iron deficiency dilutes the effect — was pre-specified, tested, and not supported. Calum then supplies the more sophisticated objection, which is that an interaction test in a trial this size cannot exclude a subgroup difference. Plus the safety signal that matters to us specifically: a sixteen-fold increase in hypophosphataemia.

We finish with the four-point critique, the strengths the trial genuinely deserves, and what to do in clinic on Monday.

Chapters

• (00:00) Cold open — a haemoglobin of 118 and six weeks to go
• (01:00) Why the question matters, and the ten per cent of the blood supply
• (02:10) Marker or mechanism? The idea that runs through everything
• (03:00) What we knew before ITACS
• (03:40) The methods: 33 hospitals, 955 patients, no iron deficiency required
• (04:40) How do you blind a brown drug — and how do you prove it worked?
• (05:40) Days alive and at home: what it captures and what it hides
• (07:10) The primary outcome was changed mid-trial
• (08:30) What the original outcome showed
• (09:20) Sample size, two interim analyses, and the 95.4% interval
• (10:30) Did the iron actually do anything? Under 4 g/L
• (11:40) The primary result — one day, and an interval touching zero
• (12:50) Transfusion: the solid finding
• (13:50) Same length of stay, same complications — so where did the day come from?
• (15:00) The result the headline misses: who actually benefited
• (16:30) Iron deficiency — an argument corrected, and a better objection
• (18:00) Safety, and the phosphate
• (18:40) The critique in four parts
• (19:30) The strengths, and Monday morning
• (20:30) Wrap-up

Key takeaways

• Anaemia is a marker as well as a mechanism — a strong prognostic marker is not automatically a treatment target
• ITACS: 33 hospitals, 10 countries, 955 anaemic adults, a single 1,000 mg dose of IV iron or placebo, 1–26 weeks before elective cardiac surgery
• The primary outcome was changed from days at home at 30 days to 90 days in December 2020 — and the original outcome was null
• Always ask what a trial's original primary outcome was, and what it showed
• The iron filled the stores but raised haemoglobin by under 4 g/L, and three quarters of treated patients were still anaemic on the day of surgery
• Primary result: one extra day at home in 90, with a confidence interval touching zero, in a trial powered for 1.5 days
• Transfusion is the robust finding: 68% to 61%, NNT about 15, and roughly 44 units of blood saved per 100 patients treated
• Length of stay and complications were identical — the extra day accrued after discharge as shorter readmissions, not fewer of them
• A post hoc quantile analysis suggests all the benefit sits with the poor recoverers: 6.1 days at the 25th centile against a fraction of a day at the 75th
• Benefit was the same with or without demonstrable iron deficiency — but that interaction test is underpowered, so read it as no reason to restrict treatment rather than proof that deficiency is irrelevant
• Intravenous iron caused a sixteen-fold increase in hypophosphataemia, which is not a footnote in a patient about to be weaned
• The 95.4% confidence interval is the fingerprint of pre-planned interim analyses and alpha spending
• Find the anaemia early: the biggest barrier in this trial was time, not biology
• Read the methods before the abstract, and convert every relative number into an absolute one

References / further reading

• Myles PS, Klein AA, Smith JA, et al. Intravenous iron to treat anaemia before cardiac surgery (ITACS): international, double blind, placebo controlled randomised trial. BMJ 2026; 394: e100407
• Myles PS, Richards T, Klein A, et al. Rationale and design of the intravenous iron for treatment of anemia before cardiac surgery trial. Am Heart J 2021; 239: 64–72
• Boer C et al. 2017 EACTS/EACTA Guidelines on patient blood management for adult cardiac surgery. J Cardiothorac Vasc Anesth 2018
• Agarwal S, Choi SW, Fletcher SN, Klein AA, Gill R. The incidence and effect of resternotomy following cardiac surgery on morbidity and mortality: a 1-year national audit on behalf of the Association of Cardiothoracic Anaesthesia and Critical Care. Anaesthesia 2021; 76: 19–26
• Myles PS et al. Validation of days at home as an outcome measure after surgery. BMJ Open 2017

Follow the podcast

• Bluesky: @cardiacoutput.bsky.social
• X: @CardiacOutputMC

If you've found this useful, a follow or a share genuinely helps other trainees find it — and do get in touch if there's a paper you'd like taken apart.

This podcast is for medical education for healthcare professionals. It is not clinical advice. Discussion of published research reflects our own reading and interpretation — always read the primary source and follow your own centre's guidelines.

This is the theatre half of the bleeding week — the drugs, the traps in the tests, and a salvage case at the far end of what's possible. Last episode was the unit.

Please note: the doses and drug choices are Wythenshawe-specific local practice. Take the principles, and check your own guidelines.

The scale first: cardiac surgery consumes around ten per cent of the entire NHS blood supply, and ninety per cent of those products go into ten per cent of patients. Which makes anticipating who that will be most of the job. Bypass is hostile to clotting in five ways at once — a large heparin dose, haemodilution, cooling, platelet activation and consumption, and contact with the circuit — so know the cases where you will predictably need products (deep hypothermic circulatory arrest, redos, long pump runs, transplants and VAD explants, endocarditis, and anyone arriving anticoagulated) and act on it while they're still on bypass. Order them, get them thawed, get them physically in the room. The worst position is realising you need four units of something that's still in a freezer twenty minutes away.

Then the antifibrinolytics. Tranexamic acid at three to four grams is the modern default; aprotinin — Trasylol — was withdrawn in 2008, reintroduced in 2012 with very narrow licensing, and is now used off-license for high bleeding risk, with the honest admission that there is very little evidence directly comparing the two. And a trap the crib sheet asks about directly: aprotinin artificially prolongs the R time on your TEG, so the trace suggests a coagulopathy that isn't there and you can go chasing it with products the patient doesn't need.

We do the acquired von Willebrand story properly, because it's lovely physiology: turbulent flow across a tight aortic valve shears platelets and cleaves the large von Willebrand multimers, so the valve lesion itself causes a bleeding disorder — and explains the patient who oozes at the end of an aortic valve replacement for no other obvious reason. DDAVP, twenty micrograms in a hundred millilitres, given slowly, because giving it fast to a marginal patient at the end of a case will drop their pressure.

The centrepiece is protamine. What it actually does (electrostatic neutralisation, not a receptor effect), the three types of reaction, and why type three — profound pulmonary vasoconstriction, pulmonary hypertension and right ventricular failure — is the one that empties the ventricle and stops the heart. Then the technique that follows from the principle that when you decide to stop, no more protamine must reach the patient: not through a running flush with a column of drug behind it, but through the central line with the tap turned between patient and syringe. And the counterintuitive part — protamine in excess is itself an anticoagulant, so the generous dose given "to be sure" makes the patient oozier, not drier.

Heparin rebound gets its own section, because it's the four-in-the-morning diagnosis that saves a patient a lot of unnecessary products: protein-bound and extravascular heparin redistributing into a circulation whose protamine has already been cleared. Unexplained oozing in a patient who left theatre dry needs a small further dose of protamine, not factors. We also cover heparin resistance as an antithrombin problem, and recombinant factor seven — including a real case where it arrived from another hospital by ambulance and stopped the bleeding immediately, with proper caveats about cost, availability, thrombosis and haematology authorisation.

We finish at the salvage end: a patient with acute HIT and thrombotic complications who needs an emergency BiVAD, with no time for plasma exchange or immunoglobulin. Heparin is out, argatroban is a unit drug rather than a theatre one, and almost nobody is comfortable with bivalirudin. The answer starts with getting the right surgeon — because some will go onto cardiopulmonary bypass first, and a bypass circuit demands full anticoagulation. Instead: VA-ECMO first as a heparin-free run, then the BiVAD from there, decided at a full MDT that is honest about what this is. And for the patient with HIT who needs surgery but isn't an emergency, the first and best intervention is simply time.

Chapters

• (00:00) Cold open — the most dangerous drug we give
• (00:50) The scale: ten per cent of the blood supply
• (01:40) Why bypass is hostile to clotting
• (02:30) Who predictably needs products — and acting while still on bypass
• (03:40) Tranexamic acid, and aprotinin
• (05:20) Aprotinin prolongs the R time — don't chase it
• (06:10) Acquired von Willebrand disorder in aortic stenosis
• (07:40) DDAVP, and why you give it slowly
• (08:50) Protamine — what it does and why it's dangerous
• (10:00) The three types of reaction
• (11:00) Giving it so that stopping means stopping
• (12:20) Too much protamine is an anticoagulant
• (13:00) Heparin rebound — the 4am diagnosis
• (14:40) Heparin resistance and antithrombin
• (15:40) Recombinant factor seven, and the ambulance
• (17:20) Acute HIT needing an emergency BiVAD
• (19:40) After the salvage: no platelets, and buying time
• (20:30) Elective HIT: delay is the treatment
• (21:40) Bivalirudin, if you had to
• (22:40) Wrap-up

Key takeaways

• Cardiac surgery uses around 10% of the NHS blood supply, and 90% of those products go to 10% of patients — anticipate, and get products thawed and in the room while the patient is still on bypass
• Bypass impairs clotting five ways: heparin, haemodilution, cooling, platelet activation and consumption, and circuit contact
• Tranexamic acid 3–4 g is the default; aprotinin is off-license for redos and high bleeding risk, with little comparative evidence
• Aprotinin prolongs the R time on the TEG — don't chase a coagulopathy that isn't there
• Severe aortic stenosis causes acquired von Willebrand disorder through shear; DDAVP 20 micrograms in 100 mL, given slowly, and cryoprecipitate is a rational choice
• Protamine neutralises heparin by charge, and can cause pulmonary hypertension, RV failure and cardiac arrest
• Type 1 is rate-related hypotension, type 2 anaphylactoid, type 3 the catastrophic pulmonary vasoconstriction — type 3 is what kills
• Give protamine so that stopping means stopping: no flush behind it, tap turned between patient and syringe
• Excess protamine is itself an anticoagulant — titrate it, don't pour it in
• Unexplained oozing hours later on the unit is heparin rebound: give more protamine, not more products
• Heparin resistance is an antithrombin problem — antithrombin III around 500 units, not more heparin
• Recombinant factor seven works but is expensive, hard to locate, prothrombotic and needs haematology sign-off
• For emergency surgery in acute HIT: the right surgeon, VA-ECMO before the BiVAD, a heparin-free run, no platelets, and an MDT honest about it being salvage
• Where there is time, time is the treatment — delay towards remote HIT, plasma exchange, IVIG, and a single intra-operative heparin exposure only

References / further reading

• Boer C et al. 2017 EACTS/EACTA Guidelines on patient blood management for adult cardiac surgery. J Cardiothorac Vasc Anesth 2018
• Task Force et al. 2024 EACTS/EACTAIC Guidelines on patient blood management in adult cardiac surgery
• Myles PS et al. Tranexamic acid in patients undergoing coronary-artery surgery (ATACAS). N Engl J Med 2017
• Fergusson DA et al. A comparison of aprotinin and lysine analogues in high-risk cardiac surgery (BART). N Engl J Med 2008
• Nybo M, Madsen JS. Serious anaphylactic reactions due to protamine sulfate: a systematic literature review. Basic Clin Pharmacol Toxicol 2008
• Vincentelli A et al. Acquired von Willebrand syndrome in aortic stenosis. N Engl J Med 2003
• Cuker A et al. American Society of Hematology 2018 guidelines: heparin-induced thrombocytopenia. Blood Adv2018
• Koster A et al. Anticoagulation during cardiopulmonary bypass in patients with heparin-induced thrombocytopenia. Ann Thorac Surg

Follow the podcast

• Bluesky: @cardiacoutput.bsky.social
• X: @CardiacOutputMC

If you've found this useful, a follow or a share genuinely helps other trainees find it — and do get in touch if there's a topic you'd like covered.

This podcast is for medical education for healthcare professionals. It is not clinical advice. All drugs and doses discussed reflect local Wythenshawe practice at the time of recording — always follow your own centre's guidelines.

This is the intensive care half of the bleeding week: resternotomy, blood products, point-of-care testing and heparin-induced thrombocytopenia. Mike takes a deliberately unfashionable position on the TEG and defends it, and Calum pushes back.

Please note: the doses and product choices are Wythenshawe-specific local practice. Take the principles, and check your own guidelines and transfusion policy.

We start with the thing everyone is trying to avoid — going back. Resternotomy for bleeding happens perhaps a couple of times a week, and once a patient needs a second procedure the odds of a poor outcome climb sharply: more red cells, more acute kidney injury and filtration, prolonged intubation, tracheostomy and a longer intensive care stay. Some of that is confounding, since you don't get taken back unless something has already gone wrong, but the UK audit data is clear enough. Which makes prevention one of the real jobs of the cardiac anaesthetist: stop the dual antiplatelets a week or two beforehand, insist on a proper period of haemostasis at the end of the case, and don't let surgeons close wet chests. If the patient is oozing, the chest gets packed, you give products, you give it time, you get them warm — and if they still aren't dry, you encourage the surgeon to leave the chest open. That is a good decision, not a failure of the operation, and the conversation needs diplomacy.

Then the products, in the two pairs that get confused constantly. FFP is human plasma, so it contains everything — all the factors plus fibrinogen, albumin, protein C and S and antithrombin III — but it comes at high volume, and pushing four units in fast can cause real problems for the right ventricle. Octaplex is four-factor prothrombin complex — factors two, seven, nine and ten — made up at the bedside, low volume, titratable, with around three thousand units to reverse warfarin or a DOAC, and a thousand units often enough to dry up needle-hole oozing that isn't responding to protamine. Then cryoprecipitate versus fibrinogen concentrate, where the newer, more expensive product isn't automatically the better one: cryo also carries von Willebrand factor and factor XIII, and in a patient with the acquired von Willebrand disorder that severe aortic stenosis gives you, the "inferior" product may be exactly the right one.

The centrepiece is the argument about point-of-care testing. Algorithms beat individual judgement and point-of-care testing produces more restrictive transfusion than laboratory testing — but it hasn't yet shown better outcomes, and in a patient with very high drain output the danger is waiting for information before doing anything to stop the bleeding. In a genuinely bleeding cardiac patient the TEG is usually hard to interpret and normally tells you to give everything. Use it, but never let it delay treatment — and have a low threshold for a TOE, because tamponade is the diagnosis you cannot afford to miss while you're chasing numbers.

The second half is HIT, taken well past "platelets dropped, send a screen". A rare, immune-mediated, severe drug reaction with antibodies against the platelet factor four–heparin complex, causing platelet activation, thrombin generation and life-threatening thrombosis — a clotting disease with a low platelet count, and around six per cent mortality per day untreated. We cover the 4T score and why pre-test probability still matters when you're sending a test anyway, how to separate it from the far more common heparin-associated thrombocytopenia, the biphasic platelet pattern that should make you sit up after cardiac surgery, and why a count above 150 doesn't exclude it. Then the in-house immunoassay that rules out versus the functional assay that confirms, the acute and subacute categories, why platelet transfusion is contraindicated, and what plasma exchange and IVIG buy you. We finish with argatroban — a direct thrombin inhibitor that needs no cofactor — and a case where a patient was bridged straight onto warfarin instead, and thrombosed.

Chapters

• (00:00) Cold open — the drains are filling, and your instinct may harm this patient
• (01:10) Resternotomy, and what it costs the patient
• (02:40) Preventing it: antiplatelets, haemostasis, and the wet chest
• (04:20) Leaving the chest open, and how that conversation goes
• (05:20) FFP versus Octaplex
• (07:30) Doses, and a thousand units for oozing
• (08:50) Cryoprecipitate versus fibrinogen concentrate
• (10:40) More of an art than a science
• (11:40) The unfashionable view on TEGs
• (13:40) Don't miss the tamponade
• (14:20) When does the patient go back?
• (15:40) HIT — what it actually is
• (17:00) The 4T score, and why pre-test probability matters
• (18:20) The biphasic pattern, and what else confounds it
• (20:00) In-house screen versus the functional assay
• (21:40) Why platelets are contraindicated
• (22:40) Plasma exchange and IVIG
• (23:40) Argatroban — and the warfarin bridge that went wrong
• (25:20) Wrap-up

Key takeaways

• Going back to theatre for bleeding is a serious marker of poor outcome — more transfusion, more AKI, longer ventilation and longer stay
• Prevention starts in theatre: stop dual antiplatelets, take a proper haemostatic pause, and never let a surgeon close a wet chest
• If they still aren't dry, leaving the chest open is a good decision rather than a failure
• FFP contains everything but at high volume — pushing it in fast can be very bad for the right ventricle
• Octaplex is four-factor PCC, made up at the bedside, low volume and titratable; around 3,000 units reverses warfarin or a DOAC
• A thousand units of Octaplex often settles needle-hole oozing that protamine hasn't fixed
• Cryoprecipitate isn't simply the inferior product — it carries von Willebrand factor and factor XIII, which may be exactly what an aortic stenosis patient needs
• Use the TEG, but never wait for it in the acutely bleeding patient — give products, tell the surgeon, and go back if you can't stop it
• Have a low threshold for TOE, specifically to exclude tamponade
• HIT is a prothrombotic disease with a low platelet count, and roughly 6% mortality per day untreated
• Use the 4T score for pre-test probability, look for the biphasic platelet pattern, and don't be reassured by a count above 150
• The in-house immunoassay rules out; the functional assay confirms — and both can be falsely negative
• Platelet transfusion in acute HIT is contraindicated; plasma exchange and IVIG are what buy you heparin
• Stopping heparin is necessary but not sufficient — argatroban is a direct thrombin inhibitor needing no cofactor, and you must never bridge a HIT patient onto warfarin alone

References / further reading

• Agarwal S, Choi SW, Fletcher SN, Klein AA, Gill R. The incidence and effect of resternotomy following cardiac surgery on morbidity and mortality: a 1-year national audit on behalf of the Association of Cardiothoracic Anaesthesia and Critical Care. Anaesthesia 2021; 76: 19–26
• Boer C et al. 2017 EACTS/EACTA Guidelines on patient blood management for adult cardiac surgery. J Cardiothorac Vasc Anesth 2018
• Task Force et al. 2024 EACTS/EACTAIC Guidelines on patient blood management in adult cardiac surgery
• Wikkelsø A et al. Thromboelastography or thromboelastometry to monitor haemostatic treatment. Cochrane Database Syst Rev 2016
• Greinacher A. Heparin-induced thrombocytopenia. N Engl J Med 2015
• Cuker A et al. American Society of Hematology 2018 guidelines for management of venous thromboembolism: heparin-induced thrombocytopenia. Blood Adv 2018
• Warkentin TE, Greinacher A. Management of heparin-induced thrombocytopenia. Curr Opin Hematol 2016

Follow the podcast

• Bluesky: @cardiacoutput.bsky.social
• X: @CardiacOutputMC

If you've found this useful, a follow or a share genuinely helps other trainees find it — and do get in touch if there's a topic you'd like covered.

This podcast is for medical education for healthcare professionals. It is not clinical advice. All drugs, doses and product choices discussed reflect local Wythenshawe practice at the time of recording — always follow your own centre's guidelines and transfusion policy.

Half of them go through one of the biggest operations a human being can have, while septic, followed by a long intensive care stay and very often a tracheostomy. And the outcomes remain poor — prosthetic valve endocarditis carries around a 27% one-year mortality, with device-related disease closer to a one in four risk of not getting home at all. As Mike puts it, we don't really tell people that when we consent them for a heart valve.

Please note: the drugs and doses discussed are Wythenshawe-specific local practice. Take the principles, and check your own guidelines.

We start with the endocarditis team — a couple of mornings a week, regional cases discussed, antibiotics and imaging reviewed, and surgery arranged when indicated. If you're the registrar in a district general hospital with a patient you think has endocarditis, there is a route in: ring your cardiac centre and ask for the endocarditis MDT. Don't sit on them.

Then the indications for surgery — prosthetic material, severe regurgitation, uncontrolled infection and embolic risk, with the vegetation sizes worth memorising — and the argument that matters most: timing. Guidance says emergency or urgent, and some read that as "theatre now". The case made here is that there is almost always a twenty-four hour window to bring that patient to intensive care first: filter them, take fluid off, start inotropes, fix the antibiotics, correct the metabolic mess. Because if you take a grossly overloaded septic patient, put them on bypass and repair their valve, they will not come off bypass. That day isn't a delay. It's what makes the operation survivable.

In theatre, the thing you must not forget: send tissue for 16S PCR. For a patient whose blood was sterilised by antibiotics before anyone took proper cultures, the valve in the surgeon's hand may be the only remaining chance to name the organism — and it decides their treatment for the next six weeks.

We cover repair versus replacement (aortic valves tend to be replaced; up to 80% of mitral valves can be repaired), and then two traps. First, these patients are thrombotic rather than coagulopathic — until they bleed, when it can be catastrophic. Second, heparin resistance: many arrive on a heparin infusion having depleted their antithrombin, so the answer is antithrombin III, not more heparin. Treat the cofactor, not the drug.

Then cytokine absorbers — what they are, why an endocarditis patient in particular gets one, and an honest account of the evidence, which is essentially "it probably won't harm and the rationale is strong". Vasoplegia gets the full stepwise ladder, with the warning that matters: be certain it isn't a low cardiac output state before you give methylene blue. We finish with post-operative ECMO, and a patient the team had agreed wasn't for mechanical support — until the decision was reversed on the table, and they went home.

Chapters

• (00:00) Cold open — more than half need surgery
• (01:20) The outcomes nobody mentions at consent
• (02:30) The endocarditis team, and how to get a patient discussed
• (04:00) Indications for surgery, and the vegetation sizes
• (05:40) When not to operate
• (06:40) The timing argument, from both sides
• (08:40) Why the twenty-four hours makes the operation survivable
• (10:00) Send the tissue: 16S PCR
• (11:20) Repair or replace?
• (12:30) Thrombotic, not coagulopathic
• (13:40) Heparin resistance — treat the cofactor
• (15:00) Cytokine absorbers, and honest evidence
• (16:30) Vasoplegia, and the methylene blue warning
• (18:00) Post-operative ECMO, and a decision reversed
• (19:30) Prophylaxis for non-cardiac surgery
• (20:40) Wrap-up

Key takeaways

• More than half of patients with endocarditis meet the criteria for surgery, and the outcomes remain poor
• Every suspected case should be discussed with an endocarditis team — there is a route in from any hospital
• Operate for prosthetic material, severe regurgitation, uncontrolled infection or embolic risk; vegetations over 10 mm after an embolus, or over 15 mm in isolation
• Surgery is contraindicated with intracranial haemorrhage or coma, but the restrictions around stroke have relaxed
• There is almost always a twenty-four hour window to optimise — filter, inotropes, antibiotics, fluid off — and that isn't a delay, it's what lets them come off bypass
• Send tissue for 16S PCR: it may be the only chance to identify the organism
• Aortic valves tend to be replaced; up to 80% of mitral valves can be repaired
• These patients are thrombotic rather than coagulopathic — but when they bleed it can be catastrophic
• Heparin resistance is an antithrombin problem: give antithrombin III, not more heparin
• Cytokine absorbers are used on rationale rather than randomised evidence — and they adsorb drugs too
• Two arterial lines, and climb the vasoplegia ladder — but be certain it isn't low cardiac output before giving methylene blue
• Needing ECMO after endocarditis surgery is a poor prognostic sign — but stay humble, because patients surprise us

References / further reading

• Charlesworth M, Williams BG, Ray S. Infective endocarditis. BJA Education 2023
• Delgado V et al. 2023 ESC Guidelines for the management of endocarditis. Eur Heart J 2023
• Pettersson GB, Hussain ST. Current AATS guidelines on surgical treatment of infective endocarditis. Ann Cardiothorac Surg 2019
• Kang DH et al. Early surgery versus conventional treatment for infective endocarditis. NEJM 2012
• Boer C et al. EACTS/EACTA Guidelines on patient blood management for adult cardiac surgery. 2017
• Levin RL et al. Methylene blue reduces mortality and morbidity in vasoplegic patients after cardiac surgery. Ann Thorac Surg 2004
• National Institute for Health and Care Excellence. Prophylaxis against infective endocarditis (CG64)

Follow the podcast

• Bluesky: @cardiacoutput.bsky.social
• X: @CardiacOutputMC

If you've found this useful, a follow or a share genuinely helps other trainees find it — and do get in touch if there's a topic you'd like covered.

This podcast is for medical education for healthcare professionals. It is not clinical advice. All drugs and doses discussed reflect local Wythenshawe practice at the time of recording — always follow your own centre's guidelines and current local policy.

Not because we've stopped looking — because the disease has changed underneath us. In this episode Mike and Calum work through infective endocarditis: why it presents so differently now, how it gets diagnosed, and the traps that make it harder than it needs to be. This is the diagnosis and intensive care half; the theatre half is the companion episode.

Please note: antibiotic choices vary enormously between centres and even between regions. Everything here is illustrative — follow your own microbiology advice.

Endocarditis used to be a community disease, brewing over weeks, with time to develop immune-complex phenomena in the fingernails. It's now much more of an acute, hospital-associated disorder — roughly a third of cases are healthcare-associated, and those carry a higher mortality. And we generate the risk factors ourselves, with valves, pacemakers, indwelling catheters and dialysis lines.

We cover the organisms and how they differ by continent, HACEK (and why the lab needs warning about slow growers), and fungal endocarditis — rare, bulky and grim. Then the pathophysiology, which is more elegant than it first appears: a healthy valve is remarkably resistant to infection, so something has to damage it first. Turbulence injures the endothelium, a sterile vegetation forms from platelets and the clotting cascade, and that bland thrombus becomes the landing pad a later bacteraemia colonises. That two-hit sequence explains where vegetations sit and why any valve, device or structural disease makes a patient high risk.

Then the two pillars of the Duke criteria, both of which get routinely mishandled. Blood cultures are frequently done badly — one set, then Tazocin, and by the time the microbiologist rings the antibiotics are already in and everything comes back negative. You've manufactured your own culture-negative endocarditis. Echo access is the second problem, though a new generation of intensivists and anaesthetists confident with bedside scanning is changing the timeline. We give clean rules: transthoracic first, isolated right heart disease can stop there, but a prosthetic valve or any implantable device always needs a TOE.

The criteria themselves have grown up — surgical inspection of the valve is now a major criterion, and gated CT and nuclear imaging rescue the uncertain case. We're honest about the antibiotic evidence, which amounts to no high-quality randomised trials at all, and work through exactly why that trial is so hard to design. Then the specifics worth knowing: aminoglycosides are out for staphylococcal native valve disease, daptomycin and fosfomycin are in for MRSA, and rifampicin waits until the bacteraemia has cleared.

We finish with what gets mistaken for endocarditis on echo, built around a real diagnostic argument — a lesion that turned out to be a fibroelastoma, because it was on the wrong side of the valve, too smooth and too round, and the patient was far too well.

Chapters

• (00:00) Cold open — what you were taught to look for at the hands
• (01:10) Why the classical signs have vanished
• (02:20) The organisms, and why they've changed
• (03:40) HACEK, and warning the lab
• (04:50) Fungal endocarditis
• (05:50) The two-hit pathophysiology
• (07:10) Risk factors: any valve, any device
• (08:00) Blood cultures done badly
• (09:20) Transthoracic, transoesophageal, and who must have one
• (10:40) The criteria grow up: surgery, CT and nuclear imaging
• (12:00) The antibiotic evidence — and how you'd design the trial
• (13:30) Aminoglycosides, daptomycin and the rifampicin timing rule
• (14:40) What gets mistaken for endocarditis on echo
• (16:00) Wrap-up

Key takeaways

• The disease has moved from the community into hospital — stop waiting for splinter haemorrhages
• About a third of cases are healthcare-associated, and we create the risk factors with valves, devices and lines
• HACEK organisms are slow-growing, so the lab needs warning or you'll get a falsely negative culture
• Endocarditis rarely affects a normal valve: turbulence damages endothelium, a sterile vegetation forms, and bacteraemia colonises it
• Vegetations sit on the upstream side of the valve, where the shear stress is
• Take adequate blood for culture before the antibiotics, or you invent your own culture-negative case
• Transthoracic first; isolated right heart disease can stop there; a prosthetic valve or device always needs a TOE
• Surgical inspection of the valve is now a major criterion, and gated CT and nuclear imaging rescue uncertain cases
• There is no high-quality randomised evidence behind the antibiotic regimens — and there are good reasons why
• Aminoglycosides are out for staphylococcal native valve disease; daptomycin and fosfomycin are in for MRSA; rifampicin waits until the bacteraemia clears
• Not everything shaggy is endocarditis — check the side, the shape, and the patient

References / further reading

• Charlesworth M, Williams BG, Ray S. Infective endocarditis. BJA Education 2023
• Delgado V et al. 2023 ESC Guidelines for the management of endocarditis. Eur Heart J 2023
• Fowler VG et al. The 2023 Duke–ISCVID criteria for infective endocarditis. Clin Infect Dis 2023
• Habib G et al. Recommendations for the practice of echocardiography in infective endocarditis. Eur J Echocardiogr
• Cahill TJ, Prendergast BD. Infective endocarditis. Lancet 2016
• National Institute for Health and Care Excellence. Prophylaxis against infective endocarditis (CG64)

Follow the podcast

• Bluesky: @cardiacoutput.bsky.social
• X: @CardiacOutputMC

If you've found this useful, a follow or a share genuinely helps other trainees find it — and do get in touch if there's a topic you'd like covered.

This podcast is for medical education for healthcare professionals. It is not clinical advice. Antibiotic regimens and practice described are illustrative and vary by centre — always follow your own microbiology advice, local guidelines and current policy.

In ECMO, the hard part is almost never the cannulation. Putting cannulas in and troubleshooting hypoxia are learnable. What's hard is the decision — and there's no textbook for it.

In this episode Mike and Calum work through four invented scenarios. In each one, Mike takes Calum to the point where a decision has to be made, asks what he'd do, and then reveals what happened. Every scenario has more than one entirely defensible answer, and in every one a cognitive bias is quietly doing the deciding.

We start with a young man who needs ECMO for trauma-related lung injury — and also has a traumatic brain injury. The reflex is that you can't anticoagulate intracranial contusions, but that reflex rests on an assumption modern circuits no longer require, and it crowds out one of the best prognostic groups we ever see. That's base rate neglect.

Then a patient at day forty, where a phrase appears in the notes almost daily: "he's been on for forty days." The number has become the argument. Duration isn't a diagnosis — and asking a different question ("what specifically is stopping him weaning?") turns up something entirely fixable. That's anchoring, with availability bias underneath it.

The third scenario is the subtlest, because nobody does anything wrong. A young woman is treated for severe pneumonia, correctly. Then she starts bleeding, and it's blamed on the anticoagulation — also a completely satisfying explanation. Two plausible answers in a row, each of which stops anyone completing the diagnosis. That's search satisficing, and it ends with a thirty-second rule you can use tomorrow.

The last scenario ends badly, deliberately. A single word in a CT report — "fibrosis" — carries a certainty the imaging doesn't support. A time-limited trial of steroids is agreed, with a review date set in advance. It doesn't work, and the patient dies. And the decision was still reasonable. Judging it by the result would be outcome bias — which leads to the most uncomfortable idea in the episode: when you decline, you almost never find out you were wrong, so the only errors you can see are the ones where you acted.

We finish with the full list — availability, conjunction, overconfidence, representativeness, diagnostic momentum, commission bias, the IKEA effect — and the one to end on, the GI Joe fallacy: the tendency to think that knowing about cognitive bias is enough to overcome it. It isn't. So we close with four things that genuinely help, none of which are clever.

Chapters

• (00:00) Cold open — the hard part isn't the cannula
• (00:50) Why these cases are fictional, and why the reasoning still isn't
• (01:40) The referee problem: making calls you can't verify
• (02:20) Case one: trauma lung injury, and a head injury
• (04:50) Base rate neglect
• (05:30) Case two: "he's been on for forty days"
• (07:40) Anchoring to a number rather than a trajectory
• (08:30) Case three: the bleeding everyone blames on the circuit
• (11:00) Search satisficing — and a thirty-second rule
• (11:50) Case four: one word in a CT report
• (13:40) Outcome bias, and the asymmetry of declining
• (14:40) The full list of biases
• (15:30) The GI Joe fallacy, and what actually helps
• (16:10) Wrap-up

Key takeaways

• The hard part of ECMO is the decision, not the cannula — and there's no textbook for it
• Base rate neglect: one alarming feature can crowd out a favourable underlying picture
• A contraindication that feels absolute may be a modifiable risk — ECMO does not obligatorily mean full anticoagulation
• Duration is not a diagnosis; ask what specifically is preventing weaning
• A satisfying diagnosis stops the search — and two satisfying explanations in a row are worse
• In a bleeding ECMO patient, ask whether it's the circuit bleeding or the disease bleeding. Dip the urine
• A single word in a report can carry more certainty than the evidence behind it
• A reasonable decision can produce a bad outcome; don't judge the decision by the result
• When you decline, you rarely learn you were wrong — which should make everyone humbler about saying no
• Time-limited trials turn an impossible decision into a manageable one
• Knowing about cognitive bias does not protect you from it

References / further reading

• Croskerry P. From mindless to mindful practice — cognitive bias and clinical decision making. NEJM 2013
• Croskerry P. The importance of cognitive errors in diagnosis and strategies to minimise them. Acad Med 2003
• Saposnik G et al. Cognitive biases associated with medical decisions: a systematic review. BMC Med Inform Decis Mak 2016
• Baron J, Hershey JC. Outcome bias in decision evaluation. J Pers Soc Psychol 1988
• Chang DW et al. Evaluation of time-limited trials among critically ill patients. JAMA Intern Med 2021
• Kahneman D. Thinking, Fast and Slow. 2011
• Extracorporeal Life Support Organization (ELSO) General Guidelines for Adult ECMO

This podcast is for medical education for healthcare professionals. It is not clinical advice. All cases are entirely fictitious and were created for teaching purposes only; they do not describe real patients. Always follow your own centre's guidelines and current local policy.

Mike and Calum work through thoracic and cardiac airway management: one-lung ventilation, the difficult double-lumen tube, the shared airway, and a couple of scenarios that will catch you out badly if you haven't thought about them first.

Please note: this reflects local Wythenshawe practice, and a fair amount of personal preference — flagged as such where it isn't gospel. Check your own guidelines.

We start with how one-lung ventilation used to be taught — volume control, 500 mL, rate of 16, and it'll all be fine — and how completely that has changed. Bronchoscope before you turn them and again afterwards, because that tube will move. Then rather than picking numbers, find where the dependent lung actually sits on its compliance curve: start with low PEEP, hold the driving pressure constant, ramp the PEEP up a couple of centimetres at a time, and watch compliance. Most patients land around a PEEP of 6–8 with a driving pressure of 14–16 — and above 16 you're risking ventilator-associated lung injury. Driving pressure is the number to watch, not tidal volume.

Then hypoxia on one lung, where the textbook answer and the real answer differ slightly. Check the tube. Then check it again — are you actually on one lung and not down one lobe, and in the correct lumen? Most of the time that's the answer. Optimise the dependent lung before you touch the other side.

For the difficult double-lumen tube we cover a lubricated bougie inside the bronchial lumen with video laryngoscopy (a personal preference, and not uncontroversial), awake intubation for a truly predicted difficult airway, and exchanging a single-lumen tube over an airway exchange catheter — plus the answer that isn't a technique at all, which is asking for a second pair of hands early. Then rigid bronchoscopy and tracheal stenting: TIVA, rocuronium and sugammadex, depth of anaesthesia monitoring, and a Sanders jet ventilator on a genuinely shared airway.

The centrepiece is a scenario that can trick anyone: high airway pressures coming off bypass. The answer you must hold in mind is anaphylaxis — and every classic sign works against you. It may be something the perfusionist gave, a surgical dye, a cleaning solution or a coated line. The patient is completely covered, so you won't see a rash. They're hypotensive, but hypotension coming off bypass is expected anyway. And if you then try to extubate after a lot of fluid, you may find oedematous cords. Look at the patient, look for a rash, consider anaphylaxis, and give adrenaline.

We finish with anaesthetising a patient already on VV-ECMO (and why you must not do a tracheostomy on someone who isn't properly anaesthetised), restrictive pericarditis, the cardiac difficult airway, and ERAS — designing the whole anaesthetic backwards from the patient walking out.

Chapters

• (00:00) Cold open — the lung you're deliberately collapsing
• (00:50) How one-lung ventilation used to be taught
• (02:00) Bronchoscope before and after you turn them
• (02:50) Titrating PEEP against compliance
• (04:20) When they don't tolerate one lung
• (05:10) Hypoxia on one-lung ventilation — check the tube first
• (06:30) The difficult double-lumen tube
• (08:00) Rigid bronchoscopy and jet ventilation
• (09:40) High airway pressures off bypass — think anaphylaxis
• (11:30) Anaesthetising a patient already on ECMO
• (12:40) Restrictive pericarditis and the cardiac difficult airway
• (13:50) ERAS: designing the anaesthetic backwards
• (15:00) Wrap-up

Key takeaways

• One-lung ventilation has moved on: bronchoscope before and after turning them, because the tube moves
• Titrate PEEP against compliance rather than picking numbers — usually 6–8, with a driving pressure of 14–16
• Driving pressure is the number to watch; above 16 you're risking lung injury
• If they desaturate on one lung, check the tube first — that's usually the answer
• Optimise the dependent lung before doing anything to the non-dependent one
• For a difficult double-lumen tube: bougie and video laryngoscopy, awake intubation, or exchange over a catheter — and ask for help early
• Rigid bronchoscopy needs neuromuscular blockade; airway trauma is reported without it
• High airway pressures off bypass should make you think anaphylaxis, because every classic sign is hidden or explained away
• A patient on ECMO having a tracheostomy needs full anaesthesia and depth of anaesthesia monitoring
• Turn the PEEP off in restrictive pericarditis, and have vasopressors ready at induction
• In the cardiac difficult airway, protect the stomach for the TOE probe and bail out to awake intubation early

References / further reading

• Lohser J, Slinger P. Lung injury after one-lung ventilation. Anesth Analg 2015
• Amato MBP et al. Driving pressure and survival in the acute respiratory distress syndrome. NEJM 2015
• Campos JH. Lung isolation techniques for patients with difficult airway. Curr Opin Anaesthesiol 2010
• Ahmad I et al. Difficult Airway Society 2025 guidelines for management of unanticipated difficult intubation in adults. Br J Anaesth 2025
• Royal College of Anaesthetists. 6th National Audit Project (NAP6): perioperative anaphylaxis. 2018
• Levy JH, Adkinson NF. Anaphylaxis during cardiac surgery. Anesth Analg 2008
• Engelman DT et al. Guidelines for perioperative care in cardiac surgery: ERAS Society recommendations. JAMA Surg 2019

Which raises the obvious question: why do a whole episode on it? Because the fact that the anaesthetic is trivial does not make this a low-risk procedure — and our value here has almost nothing to do with giving an anaesthetic. That's the thesis of the episode.

Please note: the drugs and doses discussed are Wythenshawe-specific local practice. Take the principles, and check your own guidelines.

We start with scale. TAVI is arguably the biggest disruptor in medical practice of the last twenty years — from perhaps forty cases in the UK in the early years to several thousand annually now. While anaesthesia debated video laryngoscopy and TIVA, the cardiologists took a brand new procedure and generated randomised trial after randomised trial.

Then diagnosis done properly. How you derive the aortic valve area on TOE using the continuity equation — and why three separate measurements plus a geometric assumption, with the LVOT diameter squared, is a lot of places to be wrong. Hence the velocity ratio, the dimensionless index that cancels the LVOT area entirely and doesn't need the ventricle to generate a big gradient. Plus why gated CT with a calcium score now answers anatomy, feasibility, access and sizing in a single scan.

We cover who gets TAVI over surgery — and the point that surprises people, which is that almost anybody can have a surgical AVR, while TAVI is the fussy one, ruled out by access and anatomy. Then the evidence arc from PARTNER through NOTION, SURTAVI, PARTNER 3 and Evolut Low Risk, a defence of non-inferiority as exactly the right question here, and two honest problems: the patients we actually treat would never have met the trial inclusion criteria, and the trials are funded by the people selling the valves. Plus the warning signal for younger patients, where surgical explant of a TAVI valve carries a high mortality.

Finally the practical half: minimalist TAVI and the fall in mortality from over 5% to under 2%; why rapid ventricular pacing at over 200 for a few seconds is needed and how to avoid it altogether with a self-expanding valve; the complications, including a roughly one-in-five permanent pacemaker rate; what the published reports into a struggling centre actually identify (selection, expertise and governance — never the anaesthetic technique); conversion planning and a patient who dissected in the cath lab and did well anyway; the anaesthetic and sedation techniques when they are needed; the principles for the severe aortic stenosis patient, ending on patience; and the rest of the cath lab, including MitraClip.

Chapters

• (00:00) Cold open — "it's lidocaine into the groin"
• (01:10) Why a trivial anaesthetic doesn't mean a low-risk procedure
• (02:10) TAVI as the biggest disruption in twenty years
• (03:40) Grading the valve: the continuity equation and its errors
• (05:30) Why the velocity ratio is the better number
• (06:40) The gated CT that answers everything at once
• (07:40) Who gets TAVI — and why TAVI is the fussy option
• (09:10) PARTNER to Evolut: the evidence arc
• (11:00) Two problems: external validity, and who funds the trials
• (12:40) The warning signal for younger patients
• (13:50) Minimalist TAVI, and the fall in mortality
• (15:20) Rapid ventricular pacing — and when to avoid it
• (16:40) Complications, and the one-in-five pacemaker
• (17:50) When it goes wrong: selection, expertise, governance
• (18:40) Conversion planning, and a patient who surprised us
• (19:40) Anaesthesia, sedation, and the principle of patience
• (20:40) MitraClip and the rest of the cath lab
• (21:20) Wrap-up

Key takeaways

• For over 95% of patients TAVI is local anaesthetic and light sedation — and that does not make it low risk
• The continuity equation needs three measurements and a geometric assumption; the velocity ratio needs neither
• Gated CT with a calcium score answers grading, anatomy, feasibility and access in one scan
• Almost anyone can have a surgical AVR; TAVI is the fussy option, ruled out by access and anatomy
• Non-inferiority is the right question, because the recovery benefits are so large
• The evidence is excellent but generated in patients unlike ours — and funded by the valve manufacturers
• Surgical explant of a TAVI valve carries a high mortality, which matters for younger patients
• Minimalist TAVI has taken mortality from over 5% to under 2%, with fewer steps and less stroke
• Rapid pacing stops the ventricle fighting the balloon; if they won't tolerate it, use a self-expanding valve
• Roughly one in five need a permanent pacemaker — a genuine complication, though the rate is falling
• Plan the conversion before you start: pericardial window, bypass, or neither
• For the aortic stenosis patient: minimise demand, optimise supply — and above all, be patient

References / further reading

• Charlesworth M et al. Anaesthesia support for transcatheter heart valve interventions: a narrative review. Anaesthesia 2025
• Leon MB et al. PARTNER: TAVI in patients who cannot undergo surgery. NEJM 2010
• Thyregod HGH et al. NOTION: TAVI versus surgery in lower-risk patients. J Am Coll Cardiol 2015
• Reardon MJ et al. SURTAVI: TAVI versus surgery in intermediate-risk patients. NEJM 2017
• Mack MJ et al. PARTNER 3: TAVI with a balloon-expandable valve in low-risk patients. NEJM 2019
• Popma JJ et al. Evolut Low Risk: TAVI with a self-expanding valve in low-risk patients. NEJM 2019
• Fukuhara S et al. Surgical explantation of transcatheter aortic bioprostheses. J Thorac Cardiovasc Surg 2021
• Vahanian A et al. 2021 ESC/EACTS Guidelines for the management of valvular heart disease. Eur Heart J 2021

This podcast is for medical education for healthcare professionals. It is not clinical advice. All drugs and doses discussed reflect local Wythenshawe practice at the time of recording — always follow your own centre's guidelines and current local policy.

In this episode Mike and Calum work through aortic stenosis and the problems it creates on the cardiac unit. This is the intensive care half; the cath lab half, on TAVI and sedation, is the next episode.

Please note: the fluids and practice described are Wythenshawe-specific. Take the principles, and check your own guidelines.

We start with why that patient is empty. Aortic stenosis is mechanically an outflow tract obstruction, so the ventricle hypertrophies against it — and this afternoon the surgeon fixed the obstruction, but the ventricle didn't get the memo. A thick, stiff, non-compliant ventricle is exquisitely preload-dependent, and reaching for the vasopressor instead of the fluid will have you chasing that patient all night. Expect to give five litres in twenty-four hours, expect them to be overloaded a few days later, and don't reach for albumin on day zero — because by the time it arrives, your patient is another litre behind.

Then the disease itself. Why symptoms are the prognostic trigger — a one-year mortality of fifty percent from the moment they appear — and the murky world of low-gradient aortic stenosis, where the gradient depends on a ventricle that can still generate it. The sickest patients are precisely the ones the measurement fails on, underestimating severity and overestimating valve area. We cover the velocity ratio, which is dimensionless and sidesteps the problem, the four classes of severe aortic stenosis, dobutamine stress echo for pseudo-severe disease, and why gated CT with a calcium score has displaced echo as the most important scan for structural heart disease.

We also cover the heart team and why anaesthetists and intensivists belong on it (we're generalists, and we're good at assessing risk — where once we were simply the brake at the end of the process), the causes of aortic stenosis, and what a right heart catheter actually tells you.

Finally, two things that land on the reg overnight. Out-of-hospital cardiac arrest — where the principles are the same wherever the patient is, and the cardinal rule is not to prognosticate early, because the picture genuinely changes and the story matters as much as the tests. And pacing: AAI is fine after grafts but dangerous after valve surgery, VVI backup at 30–40 is your safety net, interrogate the box daily, restore community settings before discharge — and if a patient arrests, look at the pacing box before you open the chest.

Chapters

• (00:00) Cold open — the pressors that keep climbing
• (01:20) Why the post-AVR ventricle is empty, not vasoplegic
• (03:10) Five litres in twenty-four hours
• (04:40) Why not albumin on day zero
• (06:00) AS pathophysiology, and why symptoms change everything
• (07:40) Low-gradient AS: when the gradient lies
• (09:20) The velocity ratio and the calcium score
• (11:00) The heart team, and why we're on it
• (12:40) Causes of AS, and the right heart catheter
• (14:00) Out-of-hospital arrest: don't prognosticate early
• (16:00) Pacing: AAI after grafts, VVI backup after valves
• (18:00) When TAVI patients come to ICU
• (19:10) Wrap-up

Key takeaways

• The post-AVR patient with a hypertrophied ventricle is preload-dependent — when the pressure sags they're usually empty, not vasoplegic
• Five litres in 24 hours is normal here; expect overload and diuretics a few days later
• Don't use albumin on day-zero hearts — it arrives too late to help, and there's no evidence one fluid beats another
• Symptoms are the trigger in aortic stenosis: one-year mortality of 50% from the moment they appear
• In low-gradient disease the gradient lies — it underestimates severity and overestimates valve area
• The velocity ratio is dimensionless, so it sidesteps the geometric assumptions and the need for a high gradient
• Gated CT with a calcium score has displaced echo as the key scan in structural heart disease
• Anaesthetists and intensivists belong on the heart team because we're generalists and we assess risk
• Never prognosticate early after an out-of-hospital arrest — go multimodal, and weigh the story alongside the tests
• AAI is fine after grafts but dangerous after valve surgery; VVI backup at 30–40 is the safety net
• If a patient arrests on the unit, look at the pacing box first

References / further reading

• Vahanian A et al. 2021 ESC/EACTS Guidelines for the management of valvular heart disease. Eur Heart J 2021
• Otto CM et al. 2020 ACC/AHA Guideline for the Management of Patients With Valvular Heart Disease. Circulation 2021
• Pibarot P, Dumesnil JG. Low-flow, low-gradient aortic stenosis with normal and depressed left ventricular ejection fraction. J Am Coll Cardiol 2012
• Clavel MA et al. Aortic valve calcification by computed tomography in aortic stenosis. J Am Coll Cardiol 2013
• Nolan JP et al. European Resuscitation Council and ESICM guidelines 2021: post-resuscitation care. Intensive Care Med 2021
• Dankiewicz J et al. TTM2: hypothermia versus normothermia after cardiac arrest. NEJM 2021
• Royal College of Anaesthetists. 7th National Audit Project (NAP7): perioperative cardiac arrest. 2023
• Charlesworth M et al. Anaesthesia support for transcatheter heart valve interventions: a narrative review. Anaesthesia 2025

This podcast is for medical education for healthcare professionals. It is not clinical advice. Fluids, drugs and practice described reflect local Wythenshawe protocol at the time of recording — always follow your own centre's guidelines and current local policy.

Please note: this reflects local Wythenshawe practice. Take the principles, and check your own guidelines.

We start with the history, because it explains the present. The first patient was a road traffic accident victim in 1960s America — go and look at the photograph of the machinery. Then years in which ECMO was essentially associated with death, and genuine doubt that it solved anything. Then, around 2008 to 2010, everything arrived at once: swine flu, a patient in Scotland who had to be transferred to Sweden because we couldn't offer ECMO here, the political question that followed, CESAR out of Leicester, and observational data from Australia and New Zealand — who were ahead of us because their lung transplant organs travelled further and arrived with longer ischaemic times. NICE looked at all of it, concluded equipoise had been lost, and Wythenshawe won one of the bids.

Then candidacy, which is far less formulaic than it used to be. It's now an MDT decision with two, three or more consultants, and the ideal patient doesn't really exist anymore. We work through two contrasting referrals that show why: a patient with a BMI of 50–60 and acute asthma may be an easier yes than an older patient with a bad pneumonia — because reversibility and expected run length matter more than any single exclusion criterion.

We cover what COVID changed (bifemoral cannulation, awake patients, less sedation, better steroid timing), the actual rest settings for a newly cannulated patient, and the three levers that shorten runs: early tracheostomy, negative diuresis and sedation weaning — with the caveat that each is harmful at the wrong moment. Plus an honest answer to why tracheostomy on ECMO stays a consultant procedure.

Then the part that changes practice most: a failing oxygenator doesn't just impair gas exchange, it causes coagulopathy. A D-dimer in the tens of thousands, a fibrinogen refractory to daily transfusion, unexplained platelet drops with negative HIT screens. It looks like DIC and isn't — and the treatment is changing a membrane that may be oxygenating perfectly well. We finish with hypoxia troubleshooting (including the classic mistake of turning up the sweep gas), decannulation and the microbiology plan, and exactly how you lay out an ECMO patient to get them through a CT scanner.

Chapters

• (00:00) Cold open — everything episode one didn't say
• (00:50) A road accident in the 1960s, and the years ECMO meant death
• (02:40) Swine flu, a patient sent to Sweden, and the political case
• (04:10) CESAR, Australia, and why NICE decided equipoise was lost
• (05:40) Why there won't be more UK trials
• (06:40) Who gets ECMO now: the MDT, and two contrasting patients
• (09:00) What COVID changed: bifemoral, awake, steroids
• (10:30) Rest settings for a newly cannulated patient
• (11:50) The three levers that shorten a run
• (13:10) Why tracheostomy stays a consultant procedure
• (14:20) The failing oxygenator that isn't failing
• (16:00) Hypoxia on ECMO — and the sweep gas mistake
• (17:30) Decannulation and the microbiology plan
• (18:20) Taking an ECMO patient to CT
• (19:10) Wrap-up

Key takeaways

• ECMO went from a therapy associated with death to a commissioned national service because swine flu, a patient sent abroad and CESAR all arrived together
• Equipoise has been lost in the UK, so don't expect further randomised trials here
• Candidacy is an MDT decision — a very obese asthmatic may be an easier yes than an older patient with pneumonia, because reversibility and run length outrank single exclusions
• Since COVID: bifemoral cannulation, awake patients, less sedation, better steroid timing
• Ventilate gently — peak around 20, PEEP around 10, FiO₂ 0.5, rate 12 — because not thrashing the lung is where the benefit lives
• Early tracheostomy, negative diuresis and sedation weaning shorten runs, but each is harmful at the wrong time
• A failing oxygenator causes coagulopathy, not just poor gas exchange — change it even if gas exchange looks fine
• For hypoxia: check cannula position, match ECMO flow to cardiac output, consider sedation — and don't turn up the sweep
• Accept saturations around 92%; chasing a normal number leads to harm
• Plan decannulation with microbiology, and recognise the post-decannulation SIRS response for what it is

References / further reading

• Peek GJ et al. CESAR trial. Lancet 2009
• Australia and New Zealand ECMO Influenza Investigators. ECMO for 2009 influenza A(H1N1) ARDS. JAMA2009
• National Institute for Health and Care Excellence. Extracorporeal membrane oxygenation for severe acute respiratory failure in adults (IPG391), 2011
• Combes A et al. EOLIA trial. NEJM 2018
• Goligher EC et al. Bayesian re-analysis of EOLIA. JAMA 2018
• ELSO Guidelines: Management of Adult Patients Supported with VV-ECMO, 2021
• Camporota L et al. Outcomes of the NHS England National ECMO Service. BJA 2021
• NHS England Adult Respiratory ECMO Service Specification

This podcast is for medical education for healthcare professionals. It is not clinical advice. Practice described reflects local Wythenshawe protocol at the time of recording — always follow your own centre's guidance and your regional ECMO centre.

The answer is that you cool the patient right down and stop the circulation altogether. In this episode Mike and Calum work through major aortic surgery and deep hypothermic circulatory arrest — the theatre half of the topic. The ICU half, on hypertensive emergencies and acute aortic syndromes, is the previous episode.

Please note: the drugs and doses discussed are Wythenshawe-specific local practice. Take the principles, and check your own guidelines for the numbers.

We start with the arrest itself: why the anaesthetist stays in the room, cooling to 18°C with ice packed round the head, and thiopentone to drop the cerebral metabolic rate. Then cannulation — femoral or right axillary — and the trap that follows, because if they clamp the right axillary artery your arterial line and saturation probe need to be on the left. We cover selective antegrade cerebral perfusion, why cerebral oximetry is your window on the only organ you can't afford to lose, and what the anaesthetist, perfusionist and surgeon can each do about low cerebral saturations.

Then rewarming, and a strong opinion: after circulatory arrest, the nasopharyngeal probe tells you a comfortable lie. Believe it and you'll come off bypass cold, and hand over a patient who then cools further, drops their cardiac output and becomes vasoplegic and coagulopathic. Wait for the bladder temperature.

The centrepiece is an echo walkthrough of the aortic root, because it directly decides the operation. The four levels of the root, the two workhorse views, and the question that actually matters — not how badly the valve leaks, but why. We work through the three mechanisms of aortic regurgitation in dissection (a dilated root pulling normal cusps apart, a commissure stripped off the wall by the flap, and the flap itself prolapsing through), all of which mean the valve is a victim rather than the culprit and can potentially be saved. Then what forces replacement, why an unstable patient may be better served by a quicker operation, and how to scan the finished repair.

We finish with DOAC reversal and why certainty at the bedside beats elegance in a paper, washing jets versus paravalvular leaks, postoperative goals, and where CSF drainage fits for descending aortic work.

Chapters

• (00:00) Cold open — the aorta you can't clamp
• (01:30) Cooling to 18°C: how the arrest protects the brain
• (03:20) Cannulation, and why your art line goes on the left
• (05:00) Cerebral perfusion, oximetry and low saturations
• (06:40) Knowing the surgeon's plan — the essence of cardiac anaesthesia
• (07:40) The cross-clamp, and why there isn't one during arrest
• (08:40) Rewarming: the temperature probe that lies to you
• (10:30) Echo deep dive: the four levels of the aortic root
• (12:00) The views — and asking why it leaks, not how badly
• (13:30) Three mechanisms of AR: the valve as victim
• (15:00) When to replace, and why time changes the operation
• (16:30) What else to report: ostia, tamponade, entry tear
• (17:40) Scanning the finished repair
• (18:40) They've arrived on a DOAC
• (20:00) Washing jets versus paravalvular leaks
• (21:00) Postoperative goals and CSF drainage
• (22:00) Wrap-up

Key takeaways

• You cannot cannulate or clamp an aorta that is both dissected and the operative site — hence circulatory arrest
• Cool to 18°C, ice the head, and give thiopentone: both cooling and barbiturate reduce cerebral oxygen demand
• If the return goes to the right axillary artery, put your arterial line and saturation probe on the left
• During the arrest itself there is no cross-clamp at all — nothing is flowing
• Cerebral oximetry is your window on the brain, and low saturations are a three-way conversation
• Do not trust the nasopharyngeal temperature — wait for a bladder temperature of 36.5–37°C
• On echo, the surgeon needs to know why the valve leaks, not how badly
• A dilated root, a stripped commissure or a prolapsing flap all mean the valve is repairable
• Intrinsically diseased cusps or a destroyed annulus mean replacement — and an unstable patient may need the quicker operation
• Octaplex for DOAC reversal: we know it works, and it's at the bedside
• Washing jets are normal; if they're coagulopathic, leave the chest open

References / further reading

• Isselbacher EM et al. 2022 ACC/AHA Guideline for the Diagnosis and Management of Aortic Disease. Circulation2022
• Erbel R et al. 2014 ESC Guidelines on the diagnosis and treatment of aortic diseases. Eur Heart J 2014
• Boodhwani M et al. Repair-oriented classification of aortic insufficiency. J Thorac Cardiovasc Surg 2009
• Schäfers HJ et al. Cusp geometry and effective height in aortic valve repair. J Thorac Cardiovasc Surg 2006
• Hahn RT et al. Guidelines for performing a comprehensive transoesophageal echocardiographic examination. J Am Soc Echocardiogr 2013
• Davies EA, Charlesworth M, Agarwal S. Hypertensive emergencies. BJA Education 2024
• Boer C et al. EACTS/EACTA Guidelines on patient blood management for adult cardiac surgery. 2017

This podcast is for medical education for healthcare professionals. It is not clinical advice. All drugs and doses discussed reflect local Wythenshawe practice at the time of recording — always follow your own centre's guidelines and current local policy.

In this episode Mike and Calum work through hypertension and the acute aortic syndromes as they actually present — the intensive care half of the topic. The theatre half, on deep hypothermic circulatory arrest, is the next episode.

Please note: the drugs and doses discussed are either local Wythenshawe practice or suggestions quoted from international guidance. Take the principles, and check your own guidelines and local policy for the numbers.

We start with a distinction people use interchangeably and shouldn't: emergency versus urgency. It hangs entirely on hypertension-mediated organ damage — acute damage to the heart, retina, brain, kidneys or large arteries — and the counterintuitive consequence is that severe hypertension without organ damage does not mandate emergency treatment. Meanwhile HMOD can occur at normal blood pressure readings, because the rate of change matters more than any threshold.

Then the disease itself. The syndromes that should raise your suspicion — Marfan, Turner, bicuspid aortic valve, Ehlers-Danlos — and the trap that these patients are usually not under surveillance, because nobody has ever diagnosed them. Stanford and DeBakey classification, roughly one percent mortality per hour for the first forty-eight hours, and the contrasting logic for a chronic aneurysm, where you watch and wait until 5.5 cm in the ascending aorta or 6.5 cm descending.

We're fair about why it gets missed, and honest about the consequence: treat it as an ACS or a PE and you anticoagulate a patient who was going to bleed anyway. What they need is a CT chest and a phone call to a cardiac centre.

Finally, the pharmacology done properly. Why heart rate matters as much as pressure — shear stress relates to dP/dt, so you're blunting the impulse, not just the number. Labetalol versus esmolol, the personalities and downsides of GTN, magnesium and hydralazine, and the two ways to hurt someone with treatment: overtreatment and overshoot, causing watershed infarction, mesenteric ischaemia and acute kidney injury. Plus the thing to do before any antihypertensive at all — treat the pain. We close with phaeochromocytoma (alpha blockade first, always) and PRES, the one that needs the scan as well as the clinical picture.

Chapters

• (00:00) Cold open — the chest pain that gets sent home
• (01:20) Emergency vs urgency: it's the organ damage, not the number
• (03:30) The syndromes — and why nobody has diagnosed them
• (05:30) Stanford, DeBakey, and one percent an hour
• (07:20) The chronic aneurysm: watch, or operate?
• (09:00) Why type A dissections get missed
• (10:40) Preoperative goals and anti-impulse therapy
• (12:30) The drugs and their personalities
• (14:20) Overtreatment, overshoot — and treating pain first
• (15:50) Phaeochromocytoma: alpha before beta, always
• (17:00) PRES — the one that needs the scan
• (18:00) Wrap-up

Key takeaways

• A hypertensive emergency is defined by acute hypertension-mediated organ damage, not by the number — and HMOD can occur at normal blood pressures
• Hypertensive urgency, without organ damage, does not mandate emergency treatment
• Patients presenting with dissection are often undiagnosed — it may be the first time anyone has looked at them properly
• Type A is any dissection involving the ascending aorta; roughly 1% mortality per hour for the first 48 hours
• Chronic aneurysm is different logic: surveillance, then operate above 5.5 cm ascending or 6.5 cm descending
• It gets missed as ACS, PE, acute abdomen or anxiety — and those patients arrive anticoagulated
• Treat pain first, then anti-impulse therapy: shear stress is about dP/dt, not just peak pressure
• Target systolic 100–120 and a rate of 60–70, using an infusion rather than boluses to avoid overshoot
• Alpha blockade before beta in phaeochromocytoma — beta blockade alone is contraindicated
• PRES is a clinical and radiological diagnosis, and can occur in normotensive patients

References / further reading

• Davies EA, Charlesworth M, Agarwal S. Hypertensive emergencies. BJA Education 2024
• Isselbacher EM et al. 2022 ACC/AHA Guideline for the Diagnosis and Management of Aortic Disease. Circulation2022
• Erbel R et al. 2014 ESC Guidelines on the diagnosis and treatment of aortic diseases. Eur Heart J 2014
• van den Born BJ et al. ESC Council on Hypertension position document on the management of hypertensive emergencies. Eur Heart J Cardiovasc Pharmacother 2019
• Williams B et al. 2018 ESC/ESH Guidelines for the management of arterial hypertension. Eur Heart J 2018
• Connor D, Boumphrey S. Perioperative care of phaeochromocytoma. BJA Education 2016
• Fischer M, Schmutzhard E. Posterior reversible encephalopathy syndrome. J Neurol 2017

This podcast is for medical education for healthcare professionals. It is not clinical advice. Drugs and doses discussed reflect local practice or quoted international guidance at the time of recording — always follow your own centre's guidelines and current local policy.

In this episode Mike and Calum work through the first forty-eight hours after a heart or lung transplant — and almost everything comes back to one structure: the right ventricle. This is the postoperative ICU half; the theatre half is the previous episode.

Please note: the drugs and doses discussed are Wythenshawe-specific local protocol, and are given as a worked example of how one centre does it. Take the principles, and check your own guidelines for the numbers.

We start with why the RV dominates everything — because when these patients fail, it happens almost without you noticing, and by the time you've understood the trajectory you're reopening the chest. That drives how you ventilate them: keep intrathoracic pressures low, but hold normocarbia, because letting the CO₂ drift causes pulmonary vasoconstriction and starts the spiral. We call it the Goldilocks zone.

Then the mechanical support decisions. Why a lung transplant with wet lungs or a long ischaemic time comes out on veno-arterial rather than veno-venous ECMO — a practical answer, not a physiological one — and the real cost of a long VA run, including patients returning for serial tracheal dilatations years later. We cover a piece of institutional learning worth hearing: heart transplants were once cannulated on the strength of a damped radial trace, when a femoral line would have shown straightforward vasoplegia that fluid and time would fix. Hence two arterial lines, always, and the radiofemoral difference.

We also make the case for leaving the chest open when things haven't been straightforward, the deliberately soft threshold for filtration, and the counterintuitive reason you might run low-dose adrenaline on VA-ECMO — to keep the heart ejecting, so the bypassed pulmonary circulation doesn't go stagnant and clot.

Finally: why a transplanted heart is paced at 110 (a fixed stroke volume makes cardiac output almost entirely rate-dependent), milrinone as a first choice with the honest admission that inotropes are an art rather than a science, immunosuppression and whether the patient is actually absorbing it, the BiVAD and VV-ECMO bridges and why we want those patients awake and off cardioactive drugs — and how to manage acute RV dysfunction before the spiral rather than during it.

Chapters

• (00:00) Cold open — the hard bit isn't the operation
• (01:00) Why everything comes back to the right ventricle
• (02:30) Ventilating for the RV: the Goldilocks zone
• (05:00) Why lung transplants come out on VA rather than VV ECMO
• (07:00) The bronchial anastomosis cost of a long ECMO run
• (08:20) Heart transplants and the damped radial trace
• (10:30) Two arterial lines and the radiofemoral difference
• (12:00) Open or closed chest — being realistic with the surgeons
• (14:00) Renal replacement: much softer criteria than you'd expect
• (15:40) Adrenaline on VA-ECMO — keeping the heart ejecting
• (17:30) Pacing at 110 and the fixed stroke volume
• (19:30) Milrinone, and why inotropes are an art
• (21:00) Immunosuppression — and whether they're absorbing it
• (22:30) The BiVAD bridge: awake, rehabbed, off cardioactive drugs
• (24:00) VV-ECMO as a bridge to lung transplant
• (25:30) Acute RV dysfunction — deciding before the spiral
• (26:40) Wrap-up

Key takeaways

• Everything after a heart or lung transplant comes back to the right ventricle — and RV failure arrives fast enough that you have to be ahead of it
• Ventilate for the RV: low intrathoracic pressures but normocarbia, normoxia and a normal pH
• Lung transplants come out on VA-ECMO because the cannulas are already there — but a long run risks the bronchial anastomosis
• Two arterial lines, always: the radiofemoral difference distinguishes vasoplegia from low output and can spare someone an unnecessary cannulation
• If it's been anything other than straightforward, leave the chest open
• Filter early — much softer criteria than a general ICU — because overload and acidosis tip the RV over
• Low-dose adrenaline on VA-ECMO keeps the heart ejecting so the pulmonary circulation doesn't go stagnant
• Pace a transplanted heart at 110: stroke volume is fixed, so output is all rate
• Make the VA-ECMO decision before the patient spirals, not during

References / further reading

• Velleca A et al. ISHLT Guidelines for the Care of Heart Transplant Recipients. J Heart Lung Transplant 2023
• Snell GI et al. ISHLT Working Group report on primary graft dysfunction: definition and grading. J Heart Lung Transplant 2017
• Hoetzenecker K et al. Extracorporeal support in lung transplantation: intraoperative and postoperative strategies. J Thorac Cardiovasc Surg 2020
• Royal College of Anaesthetists. 7th National Audit Project (NAP7): perioperative cardiac arrest. 2023
• Reade MC. Temporary epicardial pacing after cardiac surgery: a practical review. Anaesthesia 2007 (parts 1 and 2)
• Mathew R et al. DOREMI: milrinone vs dobutamine in cardiogenic shock. NEJM 2021
• NHS England. Adult Extracorporeal Membrane Oxygenation Service Specification

This podcast is for medical education for healthcare professionals. It is not clinical advice. All drugs and doses discussed reflect local Wythenshawe protocol at the time of recording — always follow your own centre's guidelines and current local policy.

In this episode Mike and Calum work through heart and lung transplantation from the listing call to the handover on the unit — the twelve hours in the middle. This is the theatre half; the postoperative ICU episode is separate.

Please note: the drugs and doses discussed are Wythenshawe-specific local protocol, and are given as a worked example of how one centre does it. Take the principles, and check your own guidelines for the numbers.

We start upstream of theatre — candidacy as a gift of life, why the patients being transplanted now are sicker and more borderline than a decade ago, and why compliance and mental health belong in that conversation. Then the middle-of-the-night listing assessment: what the reg is actually there to check, and what's already been done for you.

Then into theatre. Preparation that has to happen before the patient arrives (products in the fridge, six units of red cells from the start, pacemaker to fixed mode, immunosuppression before they leave the ward). Induction drugs, including the vitamin K that stops the rebound phenomenon in a warfarinised LVAD explant. And a proper walk through the lines — why the femoral venous sheath is an escape route for a balloon pump wire, why the right-sided line is shorter to dodge the caval snare, and the Swan you must remember to withdraw before bicaval cannulation.

For lungs we cover risk stratification up to "extreme high risk" — where you cannulate the groin before you induce, with a primed circuit and two consultants plus an ECMO consultant scrubbed — lung isolation, and the elegant argument for VA-ECMO over full bypass. Then the first implant: inflating to 15–20 cmH₂O, loosening the PA clamp, the three causes of hypotension at that moment, and the deliberately austere protective strategy for a new lung (FiO₂ 0.21, 3–4 mL/kg, under 20 cmH₂O) because hyperoxia drives primary graft dysfunction.

Finally the sharp end: the written escalation ladder coming off bypass — milrinone, dopamine and noradrenaline, then sequential pacing, then the balloon pump, then nitric at 20 ppm, then conversion to VA-ECMO — and why you do not re-heparinise. Plus vasoplegia and its most dangerous trap: give methylene blue to a patient who is actually in a low cardiac output state and you have put a brick wall in front of the heart. We finish with the TOE numbers for the pulmonary vein and PA anastomoses, and an end-of-case checklist that includes putting the vascath in yourself.

Chapters

• (00:00) Cold open — a donor heart is coming
• (01:00) Candidacy: who gets a transplant, and the MDT
• (03:20) The middle-of-the-night listing assessment
• (05:10) Preparation before the patient arrives
• (06:40) Induction drugs — and vitamin K for LVAD explants
• (08:20) Lines, and the traps that catch people out
• (11:00) The extreme high-risk lung induction
• (12:40) Lung isolation and positioning
• (13:40) Baseline TOE and metabolic management
• (14:40) Why transplants need so much insulin
• (15:40) Antifibrinolytics: tranexamic acid and aprotinin
• (16:30) Why VA-ECMO beats bypass for lungs
• (18:00) The first lung in: de-airing and protective ventilation
• (20:00) ECMO flow problems on the table
• (20:50) Coming off bypass: the escalation ladder
• (23:40) Vasoplegia — and the methylene blue brick wall
• (25:40) Low-volume blood products
• (26:40) TOE after implantation: the PV and PA numbers
• (28:00) End of case, vascaths and handover

Key takeaways

• A transplant is a gift of life — candidacy is an MDT decision, not a 3am one
• Build your escape routes at the start: a femoral sheath that will take a balloon pump wire, a short right-sided line to clear the caval snare, and a Swan withdrawn before bicaval cannulation
• For the sickest lungs, cannulate the groin before you induce — with the room already full of the right people
• VA-ECMO beats bypass for lung transplant: less anticoagulation, less bleeding, and the heart keeps beating
• Protect the new lung — room air, 3–4 mL/kg, under 20 cmH₂O — because hyperoxia drives primary graft dysfunction
• Coming off, follow the ladder: inotropes → pacing → IABP → nitric → VA-ECMO, and don't re-heparinise
• Be certain it's vasoplegia before giving methylene blue; in a low output state it is a brick wall in front of the heart
• Put the vascath in yourself in theatre rather than leaving it to the unit

References / further reading

• Velleca A et al. ISHLT Guidelines for the Care of Heart Transplant Recipients. J Heart Lung Transplant 2023
• Leard LE et al. ISHLT consensus document for the selection of lung transplant candidates. J Heart Lung Transplant2021
• Snell GI et al. ISHLT Working Group report on primary graft dysfunction: definition and grading. J Heart Lung Transplant 2017
• Ius F et al. Lung transplantation on cardiopulmonary support: VA-ECMO versus cardiopulmonary bypass. J Thorac Cardiovasc Surg 2012
• Diamond JM et al. Clinical risk factors for primary graft dysfunction after lung transplantation. Am J Respir Crit Care Med 2013
• Levin RL et al. Methylene blue reduces mortality and morbidity in vasoplegic patients after cardiac surgery. Ann Thorac Surg 2004
• Boer C et al. EACTS/EACTA Guidelines on patient blood management for adult cardiac surgery. 2017

This podcast is for medical education for healthcare professionals. It is not clinical advice. All drugs and doses discussed reflect local Wythenshawe protocol at the time of recording — always follow your own centre's guidelines and current local policy.

In this episode Mike and Calum get scrubbed for the theatre half of the cardiac topic: the problems that show up on the table rather than on the unit, and the ones that catch people out. This is the companion to the ICU episode on heart failure and mechanical circulatory support.

We start with vasoplegia — when to add vasopressin, why a femoral arterial line earns its place (the radiofemoral difference tracks the severity, and it takes days rather than hours to close), and the second-line agents: methylene blue at 1 mg/kg, why you warn the surgeon before you give it, and high-dose hydroxocobalamin when that isn't enough. Then a quick tour of the inotropes — and the honest admission that there is no perfect one, and essentially no evidence on which is best in acute heart failure.

Then the moment everyone dreads: coming off bypass with a failing right ventricle. Get the rate up, get atrial wires in, nitric on, inotropes running — and choose Octaplex and fibrinogen over big volumes of FFP and cryo, because bleeding in the face of RV failure is one of the hardest balancing acts in the building. We cover SAM — systolic anterior motion — which is easy to miss, hard to manage, and can have you telling a surgeon there's a problem with a valve that's perfectly fine.

Finally, a proper deep dive on pacing, from the epicardial wires up: why you nag for atrial wires, how to read the three-letter code, what AAI, DDD, VVI and the asynchronous modes actually do, and when diathermy forces your hand. We finish on the two dials everyone muddles — output and sensitivity — and the counterintuitive trap at the heart of it: turning the millivolt number up makes the box less sensitive, which is how you end up pacing onto a T wave. R-on-T is the commonest cause of cardiac arrest on a cardiac ICU, and NAP7 says so. Plus a closing bugbear about vascaths.

Chapters

• (00:00) Cold open — the noradrenaline that won't hold
• (00:50) Vasoplegia: vasopressin, and the radiofemoral difference
• (02:20) Methylene blue and hydroxocobalamin
• (03:50) The inotropes — and why there's no perfect one
• (05:00) Cardiac output monitoring in theatre
• (06:00) Coming off bypass with a failing RV
• (07:40) SAM — systolic anterior motion
• (09:00) Pacing deep dive: the wires and the three-letter code
• (10:30) AAI, DDD, VVI — and the atrial kick
• (12:00) Asynchronous modes and diathermy
• (13:00) Output vs sensitivity — the trap that causes R-on-T
• (14:40) NAP7, and the daily bedside discipline
• (15:40) Permanent pacemakers: checks and mode agreement
• (16:20) The vascath bugbear
• (17:00) Wrap-up

Key takeaways

• Escalate vasoplegia in order: noradrenaline → vasopressin → methylene blue → hydroxocobalamin
• A femoral arterial line earns its place — the radiofemoral difference tracks the degree of vasoplegia
• There is no perfect inotrope, and almost no evidence on which is best in acute heart failure
• A failing RV coming off bypass wants rate, atrial wires, nitric and inotropes — and low-volume factor concentrates rather than FFP and cryo
• SAM is dynamic LVOT obstruction — manage it by filling and pacing, not by blaming the valve
• Nag for atrial wires: no atrial wire means you're stuck in VVI
• Set output at 2–3× the capture threshold, and re-check daily as the wires fibrose
• Sensitivity is inverted — a higher millivolt setting makes the box less sensitive; undersensing causes R-on-T, the commonest cause of arrest on a cardiac ICU (NAP7)

References / further reading

• Royal College of Anaesthetists. 7th National Audit Project (NAP7): perioperative cardiac arrest. 2023
• Reade MC. Temporary epicardial pacing after cardiac surgery: a practical review. Anaesthesia 2007 (parts 1 and 2)
• Levin RL et al. Methylene blue reduces mortality and morbidity in vasoplegic patients after cardiac surgery. Ann Thorac Surg 2004
• Shaefi S et al. Vasoplegia after cardiovascular procedures: pathophysiology and targeted therapy. J Cardiothorac Vasc Anesth 2018
• Mathew R et al. DOREMI: milrinone vs dobutamine in cardiogenic shock. NEJM 2021
• Mehta RH et al. LEVO-CTS: levosimendan in patients with reduced EF undergoing cardiac surgery. NEJM 2017
• Ibrahim M et al. Modern management of systolic anterior motion of the mitral valve. Eur J Cardiothorac Surg 2012
• Boer C et al. EACTS/EACTA Guidelines on patient blood management for adult cardiac surgery. 2017

This podcast is for medical education for healthcare professionals. It is not clinical advice — always follow your local protocols and your own centre's guidance.

In this episode Mike and Calum work through heart failure and mechanical circulatory support the way it actually plays out on a cardiothoracic ICU: how you know the output is low, what the devices really do, and how to escalate before it's too late. This is the intensive care half of the topic — the theatre half is a separate episode.

We start with the idea that cardiac output is a clinical diagnosis, not a single figure — why venous sats and the Fick principle are clues rather than answers, and the picture that actually tells you the pump is failing (the tachycardia they're leaning on, the deranged liver enzymes and kidneys, the rising lactate, and the low-output gut that gets mistaken for an acute abdomen). We cover the pulmonary artery catheter — still the gold standard against a field of derived monitors — and how to float one, trace by trace.

Then the machines. We give you the single most important habit on the unit — on any patient on support, always work out where the cannulas are and where the blood is going — and use it to walk through VA-ECMO (and why bypassing the heart and lungs means stagnation, clot, and a patient you usually can't wake), BiVADs (both sides supported, awake for weeks), and the ambulatory LVAD (why its flow is derived from power, why it's preload-dependent and afterload-sensitive, and why the shocked HeartMate patient in Emergency Department needs fluid before you pick up the phone).

Finally, the practical half: escalating heart failure up the ladder — milrinone, then dopamine, then the balloon pump, and by then you're knocking on the door of a mechanical support assessment — plus inhaled nitric, a monitoring trap with the axillary return line, an honest look at ECMO-CPR, and why Harlequin syndrome is the signature limitation of peripheral VA-ECMO.

Chapters

• (00:00) Cold open — 3am, and "the cardiac output's low"
• (01:00) Low cardiac output: why it's not a single number
• (03:30) The pulmonary artery catheter — gold standard, and how to float one
• (06:30) The golden rule: always find the cannulas — and VA-ECMO
• (08:30) BiVADs: supporting both sides, awake for weeks
• (10:00) LVADs and the HeartMate 3: flow is derived, not measured
• (12:10) Escalating heart failure: milrinone, dopamine, the balloon pump
• (14:40) The axillary return line — a monitoring trap
• (15:30) ECMO-CPR: the evidence and the criteria
• (17:20) Harlequin — the VA payoff
• (18:10) Wrap-up

Key takeaways

• Cardiac output is a clinical diagnosis, not a number — venous sats and monitors are clues, not answers
• On any support device, always work out where the cannulas are and where the blood is flowing
• VA-ECMO bypasses heart and lungs — it clots, it's temporary, and you usually can't wake the patient; the elegant fix is femoral drainage with an axillary return
• LVAD flow is derived from power — it's preload-dependent and afterload-sensitive, so when in doubt, give fluid
• Milrinone → dopamine → balloon pump is the escalation ladder; a balloon pump means an MCS assessment is close
• Harlequin (differential hypoxaemia) is the signature limitation of peripheral VA-ECMO

References / further reading

• Binanay C et al. ESCAPE trial: pulmonary artery catheter in advanced heart failure. JAMA 2005
• Mathew R et al. DOREMI: milrinone vs dobutamine in cardiogenic shock. NEJM 2021
• Mehra MR et al. MOMENTUM 3: HeartMate 3 (fully magnetically levitated LVAD). NEJM 2019
• Ostadal P et al. ECMO-CS: early VA-ECMO in cardiogenic shock. Circulation 2023
• Stub D et al. CHEER trial: refractory arrest, ECMO and cooling. Resuscitation 2015
• Suverein MM et al. INCEPTION: ECPR vs conventional CPR in refractory out-of-hospital arrest. NEJM 2023
• McDonagh TA et al. ESC Guidelines for the diagnosis and treatment of acute and chronic heart failure. Eur Heart J 2021

This podcast is for medical education for healthcare professionals. It is not clinical advice — always follow your local protocols and your regional cardiac / ECMO centre's guidance.

In this episode Mike and Calum work through veno-venous ECMO the way it actually gets used on an ICU: what it does, who it's for, and the things that catch people out at the bedside.

We cover the single most important idea in ECMO — that it doesn't treat anything, it's a bridge — and why that reframes every decision you make. We get into the physiology consultants love to quiz you on (sweep gas for CO2, blood flow for oxygen, and why a septic, hyperdynamic patient can desaturate with a perfectly functioning circuit). We walk through who actually qualifies, using the EOLIA and CESAR thresholds, the Murray score, and the harder question of whether the lungs have a plausible route back.

Then we take an honest look at the evidence. CESAR randomised to referral, not to ECMO. EOLIA was stopped for futility with an 11% mortality gap — and we unpick why the 28% crossover in the control arm makes "negative trial" the wrong conclusion, and what the individual-patient-data meta-analysis and the COVID-era NHS England data added.

Finally, the practical half: the nationally commissioned UK centres (now including Barts, Bristol and Newcastle) and how to make a referral that gets your patient assessed fast; cannulation configurations; lung-rest ventilation and the bleeding-versus-clotting balancing act; and a rapid-fire troubleshooting round — recirculation, the suddenly desaturating patient, line chatter, and the failing oxygenator. Plus why Harlequin syndrome is a VA problem, not a VV one.

Chapters

• (00:00) Cold open — 3am, and a P/F of 70
• (01:16) What ECMO actually is (and isn't)
• (03:22) Sweep vs flow — the physiology you'll be quizzed on
• (05:22) Who actually gets cannulated: EOLIA, CESAR, Murray, RESP
• (07:02) Optimise first — and prone them
• (07:40) The evidence, honestly: CESAR and EOLIA
• (10:15) The UK service: 8 commissioned centres
• (11:11) Cannulation and configurations
• (12:31) Day-to-day: lung rest and anticoagulation
• (13:54) Troubleshooting at 2am
• (16:07) Harlequin — why it's a VA problem
• (16:49) Weaning and the sweep-off trial
• (17:49) Wrap-up

Key takeaways

• ECMO is a bridge, not a treatment — no bridgeable destination, no bridge
• Sweep gas controls CO2; blood flow controls oxygenation
• VV-ECMO provides no haemodynamic support — preserved cardiac function is a prerequisite
• Optimise and prone before you refer
• Recirculation is the VV gremlin; Harlequin is a VA phenomenon

References

• Peek GJ et al. CESAR trial. Lancet 2009
• Combes A et al. EOLIA trial. NEJM 2018
• Goligher EC et al. Bayesian re-analysis of EOLIA. JAMA 2018
• Combes A et al. CESAR/EOLIA individual patient data meta-analysis
• ELSO Guidelines: Management of Adult Patients Supported with VV-ECMO (2021)
• NHS England Adult ECMO Service Specification
• Camporota L et al. Outcomes of the NHS England National ECMO Service. BJA 2021

This podcast is for medical education for healthcare professionals. It is not clinical advice — always follow your local protocols and your regional ECMO centre's guidance.