ECMO circuit tubing

ECMO is half of "Mechanical supports: ECMO/IABP", an L1 topic of Section 2.1.4 in the second edition of the CICM Syllabus for the Second Part Examination. But ECMO tubing is <1% of ecmology, the most minor player and the least interesting aspect of circuit design, albeit as visually spectacular as a shiny blood-filled tube can be. 

In summary,

Apart from adding a dash of colour to the otherwise drab critical care work environment, the ECMO circuit tube really has only one job, which is not to offend the blood as it passes through. Most of the other design specifications are sufficiently predictable and boring that one does not need to list them (eg. "don't collapse under negative pressure, kink or randomly fracture"), and moreover these were already met by the tubing that the early pioneers of bypass technology purchased from their local garden centre. Well, one might also say that "not contaminated with excess manufacturing compounds of unknown toxicity" should be another desirable characteristic, and this is the main way in which modern tubing varies from the legendary tubing of the elders, who had to manually scrub it clean before use. These days the tubing is still oozing dangerous chemicals (eg. DHEP and DOA) but at least these kill you much more slowly than the 1950s benzenesulfonamide would have (eg. DHEP is probably carcinogenic but otherwise it mostly just wants to interfere with your androgens). 

Flexible polyvinyl chloride tubing for ECMO

It's still PVC, same as your bathroom plumbing, but with agents such as di(2-ethylhexyl) phthalate (DEHP) or TEHTM (tri-(2-ethylhexyl) trimellitate) to make it more transparent and less brittle. For the nerds, this is a gazillion vinyl chloride monomers (H₂C=CHCl) all arranged head to tail into long linear molecules of 20-30 kDa, which requires it to be mixed with water and heated in the presence of an initiator.  It is born as a fragile white powder and has no charm or game until combined with a plasticiser to make it soft, flexible and transparent. The favoured plasticisers are DOP,  DEHP and TEHTM - colourless oily viscous liquids which are added to molten PVC. They do not bind to PVC in any way - they just infiltrate the tangle of its polymer threads, occupying space between them and reducing the number of the tight dipole-dipole bonds that hold the polymner together and make it tough but brittle. This makes PVC suitable for medical use, and good transparent elastic tubing may be up to 40% DEHP by weight. 

Why this PVC/plasticiser mix? Polyvinyl chloride is a standard material for medical tubing, because:

• cheap
• soft and flexible
• can be recycled
• mostly ignored by your immune system
• cheap
• mostly because cheap

The other reason is "available on a massive scale", i.e. no alternative exists that could satisfy the worldwide demand (Blass et al, 1992, estimated 170,000 tons per year in Western Europe). Why not other flexible plastics, one might ask? The reason is mostly related to cost, as the alternatives (polyurethane, polyethylene, silicone polymers) are even more expensive and often either too permeable or insufficiently transparent.

Biocompatibility of ECMO circuit tubing

It's surprisingly benign, Mulholland et al (2000) discovered when they tested different variables to determine which parts of the cardiopulmonary bypass circuit did all the damage to the patient blood components. The conclusion they reached suggested that the air interface and negative circuit pressure (especially in combination) were by far the most damaging, and rubbing against the nonendothelial surfaces was a minor influence. 

Radius, length, area, volume and resistance of ECMO tubing

ECMO circuit tubing has the following character stats:

• About 2m in length per limb, for a total of 4-5 m: this lets the device remain at a comfortable distance from the patient, with enough slack on the lines to do pressure area care, but not enough loose tubing to risk the heavy blood-filled lines dragging on the insertion site sutures, risking self-dislodgement. 
• Internal diameter 3/8 inch (0.95 cm)
• External diameter is about 11.0 mm
• Wall thickness is about 3/32 inch, or 2.5mm
• About 50-60 ml per every metre of tubing (about 50% of the total prime volume in total, whereas the oxygenator has 200-350 ml of prime volume)
• Total surface area of 1000-1500 cm² (because 2πrh), or about (0.15 m²)
• Minimal flow resistance, with a pressure difference of only 30-40 mmHg per linear metre required to generate a flow rate of ~5L/min

Everybody seems to use tubing which is 3/8th of an inch in internal diameter (9.5 mm), and most authors use the phrase "by convention" to explain the reasoning behind this choice, as if to suggest that it was purely arbitrary and completely unrelated to some desirable combination of flexibility and kink resistance. Indeed this appears to be the longest-running unquestioned tradition in critical care, going back as far as Jamison et al (1954), who used 3/8 Tygon PVC tubing in their apparatus because that was one of the standard sizes for medical use. That company still lists the events that followed on their development timeline, and still manufactures the same tubing.

Insofar as internal diameter factors into consideration as the main determinant of resistance, the decision to go with big fat tubing is a good one, as the resistance is related to the fourth power of the radius. With tubing as wide as this, length matters much less (only a factor of 8).  Okura & Thind (2024) experimented with different lengths of tubing and found that doubling the length from 2.1 to 

4.2m increased the negative pressure from -76 to - 121 mmHg in the access line, which translated into a difference of 1.3L/min flow when the pressure was kept constant at -100 mmHg.

Wall thickness needs to be considered as well. Hermann & Keller (2022) mention that the ELSO registry reports an 0.3% incidence of circuit rupture, which seems like a lot (considering the possible consequences). A thicker tubing is likely to protect against this. Another benefit is the (slow, but not zero) leak of precious joules of patient body heat out of the circuit. 

The total surface area of 1000-1500 cm² (because 2πrh), or about 0.15 m², needs to be put into perspective, we can compare it to the oxygenator membrane, which often features a gas exchange surface in excess of 2.5 m². As such, the main origin of the adverse drug effects and blood-plastic interface problems is the ECMO oxygenator, and the PVC tubing is less of a problem.

 Speaking of which: this tubing is often blamed for much of the weird pharmacokinetic challenges of ECMO, such as the phenomenon of drug adsorption onto the ECMO circuit (for example, midazolam). 

Tubing connectors for ECMO circuits

Yes, they are just dumb spigotts. 

But they deserve a mention anyway, largely because of their importance to the circuit. Fairly so; they hold the tubes together and prevent them from disconnecting accidentaly during turns and suchlike. The paranoid perfusionist and nursings staff of the author's own institution don't even trust them as far as that, and tighten cable ties around their bayonette connections to make sure the tubing does not slip off. 

The main reason to bring this up is that the integrity of the circuit tubing is an essential safety feature. The venous side of the circulation is particularly at risk. Being exposed to -80 mmHg pressure means that if even a small communication between the bllodstream and the atmosphere is created, the circuit will entrain air bubbles, and the pump can potentially fail as the result.