Heating and cooling the ECMO circuit

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, and the waterbath heater is such an unglamorous and stable part of the ECMO experience that it seems almost a pity to dedicate an entire chapter to it. There's not even enough to summarise for a little "grey box" breakdown of the chapter. But on the other hand the absence of this device, or its sudden failure, would be spectacularly terrible in a way that would immeditely remind you of its importance. 

Fairly, the HU-35 brochure and user manual is the best resource for this device (a "widely used" system), but if you want to refer to a more generic discussion of the heating or cooling needs of an ECMO patient, the discussion section from Bilodeau et al (2024) is an easy to read dissertation on the decisonmaking about which unit to use by the article authors, as well a comparison of three commonly used appliances.

 The ECMO heating unit

To be unfairly simplistic, this thing is usually a simple kettle with a pump and thermostat, functionally indistinguishable from enthusiast-grade aquarium equipment. The local HU-35  is capable of delivering a maximum of 500W of heat, and is said to pump with a 10 W pump at a maximum flow rate of 5L/min, which makes it suitable for a 350-400L tropical tank full of sensitive Cichlids. The water pipes connect to the oxygenator with attractive blue tubing:

It is important to note that this device cannot cool. But considering how frivolously an ECMO patient squanders their joules, this may not be a problem, even if rapid cooling is desired.  Mojoli et al (2015) measured the power consumption of a heater and equated that to patient heat dispersion (total  heat of the system remaining stable) and concluded that the circuit must lose heat at a rate of 58 ± 12 W when everything is stable and the flow is 3L/min. In other words, when a human body produces 100W of heat at idle rest, it remains at a stable temperature, and when you add an ECMO circuit into the picture, an additional 58W needs to be contributed to maintain a stable temperature; ergo the unheated circuit would be cooling the patient with an energy deficit of -58W. Considering the human body has a specific heat of about 2.98 kJ per kg per 1°C, it would take the dispersion of 208,600J of heat energy (70× 2980) to cool a 70kg human by 1°C, which would occur over about 60 minutes (as 1W = 1J/s, i.e we are losing 58J per second, and it would take 3596.5 seconds, or 59.94 minutes, to completelty burn through 208,600J). In short, when asked "how fast can the ECMO circuit cool a patient", one could confidently blurt the factoid,  "by 1 degree per hour".

Potential problems with the ECMO heating unit

What could go wrong, you ask, looking at a piece of industrial-looking machinery with a dependable heft (about 15kg) and reassuringly 198os scifi-looking glowy numbers. Surely this could never fail in a million years. Indeed, reader, these are not particularly temperamental, as far as ICU equipment goes, but they can still cause problems, some of which are listed here along with solutions thereto:

• It runs out of water and fails to work. The water level indicator at the front of the device is literally a water level same as what one might see in a household kettle. Similarly to a kettle, the tank capacity is usually 1-2L (1.4L for the HU-35). An alarm will inform you when it is empty, and then it needs to be filled with sterile water.
• It leaks. Like all plumbing, age-related changes lead to dribbling and hesitancy. The solution is often to replace the rubbery inner seal ring behind the coupling valve of each connector. One may acquire a large number of these and perform this service confidently without having to call out a company representative, even without a solid background in plumbing or urology.
• It claims to be overheating. The persistent report of a temperature in a disturbingly high range will cause most of these units to cut the power to the heating element in a way which cannot be overridden. A common reason for this is a lack of water circulation, i.e. the tubing is kinked somewhere.
• It is infected.   By far the worst possible thing that could happen to a water heating unit is a contamination of the heater water with some kind of organism. This organism does not need to be especially pathogenic to humans, and in fact often is not; Bilodeau et al (2024) found Cupriavidus growing in their heaters, which is a Gram-negative of the Burkholderia family which typically hangs out in soil and stagnant waterways. But then they also found multiple other organisms which could really cause trouble, such as Stenotrophomonas and Pseudomonas species, amid others. Obviously this is concerning, as the circuit would be very effective at dispersing these organisms throughout the patient's circulation.  Mycobacterium chimaera was isolated from a number of bypass circuit heater/coolers in NSW and Queensland some years ago, a contamination which originated with the manufacturer and which resulted in a series of embarrasing product recalls.