Breathing circuits for manual and mechanical ventilation

There are numerous permutations of ventilator circuits, and of course ICU trainees are not expected to become familiar with all of them. The vast majority of the variants are relevant more to the anaesthetic environment, as their design is characterised by various attempts to conserve anaesthetic gas while discarding expired CO₂. The breathing circuits used in Intensive Care are usually much more straightforward.

First, in this chapter the usual circuit of the ICU ventilator will be discussed, to maintain some ICUcentric focus. This is an important item, as it is a piece of equipment to which a fair proportion of your patients may be connected. It is therefore reasonably important to know a little bit about it and to be able to discuss the characteristics which distinguish good circuits from bad. After that has been dealt with, the chapter descends into the deep swamp of anaesthetic breathing circuit taxonomy, where the reader is unlikely to find anything of exam relevance. It is included largely because it allows the author to digress pointlessly on historical trivia, and because no discussion of breathing systems is complete without somebody mentioning the name “Mapleson”.

The circuit of the typical ICU ventilator

This thing can be schematically represented as a single piece of tubing along which a humidification apparatus and a ventilated patient are mounted.

One may be able to describe the circuit as an “open” circuit because there is no reservoir for gas accumulation and there is no rebreathing. The inspiratory limb of the circuit largely isolates the patient’s lungs from the external environment (i.e. all the inspired gas is fresh gas from the ventilator). The expiratory valve at the end is open to the room, and vents gas straight into the atmosphere (fortunately there is usually nothing noxious that would require scavenging). It is a non-rebreathing circuit because the circuit allows flow only in one direction, and there is no opportunity for the patient to re-use any of the exhaled gases. If one had to describe it formally using some sort of eponymous system, one would have to call it a parallel Lack modification of the Mapleson A system.

Design characteristics of the ICU ventilator circuit

There are several criteria which a long-term breathing circuit would need to satisfy in order for people to describe it as suitable for use.

• Simple
• Lightweight
• Biologically inert
• Single-use, disposable
• Gas-impermeable
• Cheap to manufacture
• Low resistance and low compliance

There are numerous manufacturers, and this is a highly lucrative market because virtually anybody undergoing any sort of general anaesthetic is going to require a breathing circuit of one sort or another.

A circuit is a circuit and generally the expectation is that they all do the same thing and that there is little to discriminate between them. The depicted model is a Fisher & Paykel Evaqua (RT380) system which is representative of its kind. The manufacturer describes it as a “Dual Limb Adult Breathing Circuit Kit”. The device comes pre-connected to a kettle of sorts which forms a part of the proprietary humidifier. The company has some excellent propaganda literature which describes the use of this circuit in diagrams, which have been adopted below.

In general, it is possible to describe a breathing circuit like this as a length of corrugated flexible kink-resistant tubing which forms a closed system with the humidifier and ventilator. The inspiratory limb has a slightly larger volume because of the humidifier attachment. All connections are standardised, as should be the case for all such circuits (there is an ISO-mandated international understanding that all respiratory circuit components and airway equipment will have standardised sizes of connectors).

Physical characteristics of the mechanical ventilator circuit

The tubing is 22mm in diameter. That gives it an internal volume of approximately 380ml per meter. Each limb of the circuit is usually about 1.5-1.8m long, which gives a total volume of approximately 1100-1400ml. In other words, there’s a good couple of tidal volumes sitting in the inspiratory limb at any given time.

The circuit is usually made of some inert plastic which is designed to neither absorb anything from the respiratory gas flow (eg. water) nor to exude any sort of toxic solvent into the air flow. They are supplied in sterile packaging, despite the fact that nobody has ever been able to demonstrate any protective benefit from this practice. The circuits are disposable, which prevents cross-contamination between patients, among a few other nasty surprises. A review by Parmley et al (1972) offers an excellent glimpse into the bygone era of reusable rubber tubing to illustrate the sort of problems one might have had back in those days. You’d have your rubber circuit sterilised between patients, and the circuit would come back from the cleaning suite with the possibility of hidden pockets of bleach, which you would then share with the patient’s airway. During the operation, the tubing would absorb “rubber-soluble agents”. The heavy rubber would drag on the endotracheal tube and create pressure areas on the patients’ face. The inherent elasticity of the rubber would allow the spontaneously breathing patient to re-inhale expired gas from the expiratory limb, which would collapse under extremely negative pressure. 

The tubing is not indestructible, and the intensivist may encounter its physical limits in certain select ventilation scenarios. Spiral reinforcement allows for kink resistance and some degree of rigidity, but in order to remain flexible, the tubing needs to remain reasonably soft. This limits its tolerance of pressure. For example, the abovementioned Fisher & Paykel model has a maximum chamber operating pressure of 8 kPa (81.m cmH₂O), which may actually be achievable during the ventilation of patients with extremely high airway resistance (it is no inconceivable that the airway pressure limits for such patients might be set to the machine maximum, usually 90-100 cmH₂O).

Compliance of the mechanical ventilator circuit

With pressure applied to the circuit in inspiration, the tubing may expand by several cubic centimetres, which can be observed as movement (the tubing tends to straighten). The more pressure is applied the more tube movement there will be, and this is often seen in patients with status asthmaticus where the peak airway pressure may be quite high.

The change in volume in response to distending pressure is described as circuit compliance. It is an undesirable feature; circuits are designed to minimise compliance. With high circuit compliance, the delivered tidal volume is reduced in mandatory ventilation because some of the flow and pressure are "wasted" on distending the tubing. In spontaneous ventilation with high respiratory effort, the patient may also inhale gas from the expiratory tubing if it is insufficiently rigid. Circuits designed for specialised high-pressure applications (for example HFOV) have the least compliance and are usually not corrugated.

International standards recommend circuit compliance ideally no greater than 1ml per every 1cm H₂O distending pressure for every 1m of tubing. Thus, when ventilated with a driving pressure of 10 cmH2O, a typical 3.6m ventilator circuit should change volume only by 30-40ml.  If we continue picking on the Fisher & Paykel product depicted above, the manufacturer claims a compliance of 2.0 ml/cmH₂O for the entire 3.0m circuit or a compliance of 0.66 ml/m/cmH₂O.

In addition to changes in compliance, the compressibility of gas needs to be taken into account. Using old-style rubber tubing,  Bushman et al (1967) determined that with a driving pressure of 20 cmH₂O  up to 150ml of gas is compressed in the tubing.

Resistance of the mechanical ventilator circuit

The tubing used in conventional adult ventilators is of a sufficiently wide diameter to minimise resistance. The resistance to flow through a typical ventilator circuit generates a pressure of less than 0.5 cmH2O at 30L/min flow rate. The depicted Fisher & Paykel model generates a pressure of 0.2 kPa (2 cm H₂O) at a flow of 40L/min .

Increasing the length of tubing increases the resistance to flow.  This has implications for transport of critically ill patients, where extra lengths of tubing need to be attached for logistics reasons (eg. transport to and from MRI or CT). Additional tubing will increase the resistance and compliance of the circuit.

An additional point of resistance is the expiratory filter. This is a fine mesh filter designed to dehumidify the gas entering the expiratory cassette, to decrease its bacterial colonisation and improve the longevity of its electronic components. In the presence of high circuit humidity (for example, in the context of continuous nebulised medications) this filter can rapidly become waterlogged, resulting in high circuit resistance. This will manifest as high peak inspiratory pressures.

Taxonomy of breathing circuits

One could go quite mad trying to classify the numerous varieties of breathing circuits, but the most widely accepted system separates them into open, closed, semi-open or semi-closed. “Open” in this context refers to what happens to the expired gas; an open system releases the CO₂ into the atmosphere, whereas a closed system attempts to contain the CO₂ and manage it locally. The reservoir referred to in this context refers to the floppy rubber bag used by anaesthetists; the bag exists because the inspiratory flow rates generated by the main circuit of the anaesthetic machine are always going to be lower than the flow rates required by even a calmly breathing patient(10-20L/min). 

Thus:

• Open system: no reservoir and no rebreathing
• Semi-open system: has a reservoir but no rebreathing
• Semi-closed system:  has a reservoir and partial rebreathing
• Closed system: has a reservoir and complete rebreathing.

The classification can be functionally improved by adding some element which describes how CO₂ is handled in the system. The table here is from McIntyre (1986). In this system, “washout circuits” allow the release of exhaled CO₂  into the atmosphere, and “absorption” circuits incorporate some sort of mechanism for absorption. By either of these definitions, the ICU ventilator circuit is an “open” circuit because there is no reservoir (wall or turbine pressure is enough) and the CO2 is vented directly into the room air.

Unfortunately, the nomenclature of anaesthetic circuits is swarming with pointlessly eponymous designations, invoked mainly during exam time to bedevil the candidates. The need for an unambiguous descriptive naming system is unfortunately sacrificed to tradition. This anaesthetic folklore has deep roots stretching back to an ancient era when “anaesthesia” was spelled with an “æ”, and would therefore be quite difficult to eradicate. Grey-haired elders revisit the ordeals of their own education on the junior staff, believing that they are sharing with them something intricate and genuinely difficult, apparently unaware that most of the complexity is man-made. Deliberate efforts to create a better classification have been frustrated by the massive proliferation of slightly different systems which have sufficient functional differences to be patented separately and therefore carry the name of different authors. In short, it’s a mess.

Consistent to form, the author of these notes, instead of trying to develop a sensible set of categories to classify or organise this information, instead proposes to discuss the breathing in a completely disorganised order. They are vaguely ordered from simplest and oldest systems to more complex and more recent. Unfortunately, this leaves the Mapleson classification in complete alphabetic disarray (the first of the discussed circuits is the Mapleson E). This is of course perfectly fine, as there is nothing in Mapleson’s original article to suggest that the A to E ordering was according to any sort of internal rules.

Ayre T-piece

First described by T. Phillip Ayre (that’s right, T.P Ayre)  in 1937, the T-piece is a fairly unassuming thing. It is literally T-shaped, and was described by the original author in the following manner:

“…the T-piece (fig. 1) consists of a light metal tube 1 cm in diameter, into which nitrous oxide - oxygen supplemented by minimal ether is "injected" through a small inlet tube at right angles to the main limb. One end of the T-piece is connected to the endotracheal tube, while the other end is left open to the air: a length of rubber tubing attached to the open end constitutes a small reservoir for the anaesthetic gases, most of which would otherwise escape into the outside air.”

The description is from a somewhat older article by Ayre (1956) which discusses the technique in some detail. In short, this thing is basically a rigid T-shaped connector with a variable length of tubing attached, which represents either a reservoir for fresh gas, or a useless volume of dead space, depending on how much fresh gas flow there is. The author went on to describe the precise dimensions, which were very carefully thought out. For instance, the rubber tubing for adults should be 12.5 mm in internal diameter to supply 3ml per every inch of length, according to Ayre.

Notable features of this circuit and its function are:

• With fresh gas flow turned off, this circuit allows spontaneous respiration on room air. Exhaled air stays in the tube, i.e. the reservoir tube forms an extension of the anatomic dead space. In order to decrease the amount of CO2 rebreathing, Ayre recommended that the volume of reservoir tubing volume be restricted to no more than 1/3rd  of the tidal volume. For a rubber tube with  3ml/inch capacity, 138cm of tubing would satisfy this criterion.
• With sufficient fresh gas flow, the tubing would be completely flushed through with fresh gas, removing both expired CO2 and air from the dead space. By allowing a small amount of gas to build up, the tubing allowed an inspiratory reservoir to form. With sufficient fresh gas flow, the circuit would be completely devoid of rebreathed air, and the tidal volume would consist only of fresh gas (i.e there would be no dilution with room air). Ayre wrote that a fresh gas flow around 1.52.5 times the minute volume was enough to achieve this.
• By offering some resistance to the flow of gas out of the circuit, the reservoir tubing also produced a small amount of PEEP effect.

 There is also a version of Ayre’s T-piece with no reservoir tubing. By occlusion of the open end, the anaesthetist was able to ventilate the patient (i.e. gas straight from the tank of nitrous and oxygen was directed into the lungs). In this manner, Ayer was able to ventilate his patients. The positive pressure produced by this practice greatly diminished the degree of postoperative atelectasis. One might imagine how one might be able to do a great deal of harm with this method, eg. if the tubing were to accidentally kink and produce insanely high intrathoracic pressures.

The T-piece system remains in use even today. It is certainly useless for automated mechanical ventilation (the ventilator wastes its energy on blowing gas through the open tube). However, in paediatric applications,  for short-term anaesthetic purposes and in resource-poor environments it can’t be beaten for simplicity and reliability. 

Mapleson E (a modification of the Ayre T-piece)

The keen-eyed critical care trainee will immediately recognise that the Mapleson E circuit is essentially the same circuit as the original Ayre T-piece. The only real difference that it the tubing length is increased to be greater than the patient’s tidal volume. 

The point of this long tube is the expectation that following expiration, the fresh gas flow will push the expired gas up the tube, far enough that the patient – next time they take a breath- will neither be re-breathing any expired gas, nor inhaling any atmospheric air. This works as long as the fresh gas flow rate during the expiratory pause is at least equal to the tidal volume. For a stereotypical patient breathing at a rate of 12 breaths per minute with 500ml tidal volumes and an I:E ratio of 1:2, the expiratory pause might be 2 seconds – which means the fresh gas flow rate would have to be at least 15 L/min to make this circuit work as intended, i.e. 2.5 times the minute volume.

The advantages and disadvantages of this circuit are:

Advantages:

• Minimal dead space (if flow rate is high enough)
• No valves: therefore, no resistance to airflow, and no points of possible  mechanical failure
• Suitable for paediatric patients

Disadvantages:

• No valves means no PEEP. Positive pressure would have to be applied by obstructing the tube, with all the potential complications thereof.
• Wasteful (high gas flow is required), and scavenging is impossible. You’re essentially blowing anaesthetic gas all over the room.

Mapleson F (also known as Jackson Rees modification)

In answer to some of its disadvantages, the Mapleson E circuit was modified by Gordon Jackson Rees into the doubly eponymous Mapleson F, or Jackson Rees (or Jackson-Rees as the author’s middle name is occasionally incorrectly hyphenated with his surname). He got  this thing named after him and is credited with its creation purely because of a few lines in his 1950 paper on anaesthesia for the newborn:

“Artificial ventilation may be carried out by attaching a double-ended bag (B.L.B. type) to the exhaust tube, the open end of which is fitted with a vulcanite tap. This tap can be adjusted so that the intermittent pressure applied to the bag expels the amount of gas required to maintain the equilibrium of the system.”

This is the modernised version of the Mapleson E, and generally the “final word” in T-pieces of all sorts. When one asks for a T-piece circuit, one may instead get a Jackson Rees system which probably looks a bit like the depicted product from Armstrong Medical. 

This thing is furnished with a flexible bag for two reasons:

• It is possible to occlude the tail, thereby directing gas flow into the patient’s lungs
• It is possible (in some models) to put a PEEP valve on the end of the bag, thereby producing PEEP (i.e. converting a famously valveless system into a valvy one, if that’s a word).
• The elasticity of the bag takes  some of the pressure off the lungs, reducing the risk of barotrauma
• As the bag inflates, it is possible to “bag” breaths into the patient.

It also has some disadvantages:

• You still end up having to have high fresh gas flows. In order for rebreathing to be avoided, the length of tubing between the T-piece and the bag needs to be larger than the tidal volume, and completely flushed through with fresh gas.
• Scavenging is still impossible; the room still fills with whatever halogenated greenhouse gas you’re using.
• The act of incompletely occluding the “nipple” while at the same time squeezing the bag is something of an art, particularly where one is expected to do so onehanded (the other hand being used to do operate the anaesthetic mask). This raises the question: surely the generation of positive pressure in this circuit could be outsourced to some sort of dumb valve? That way, the anaesthetist could focus on more important things. This last point offers a nice segue into a discussion of the Mapleson D circuit.

Mapleson D circuit

In many ways, this seems like an evolution of the Jackson Rees system.  Instead of a partially occluded rubbery bag tail, the positive pressure in the circuit is maintained by an expiratory valve.
 

The circuit operates similarly to the Jackson Rees circuit:

• When the patient inhales, the inhaled gas comes from the length of tubing as well as from the fresh gas flow.
• Upon expiration, the patient fills some of the tubing with expired gas.
• The fresh gas flow pushes this expired gas up the tubing and into the bag, inflating the elastic bag
• When the bag distends sufficiently, the pressure in the circuit rises high enough to open the expiratory valve
• The expiratory valve then vents gas, which – provided the expiratory pause is not too prolonged – will mainly be expired gas.

The valve therefore intermittently (or possibly, constantly) vents circuit gas into the atmosphere, which is not ideal from a scavenging and environmental perspective. However, the valve can be set up in a way which allows the gas to vent into some sort of scavenging system.

Again, because the fresh gas is required to push the expired gas out, the fresh gas flow rate ends up having to be 2-3 times the minute volume. Additionally, when you squeeze the bag you increase pressure in the circuit and thus some of the gas you expel from the bag will end up venting out of the valve. 

For a variety of practical reasons, the Mapleson D circuit was modified by J.A Bain and W.E Spoerel in 1972, to create a “more streamlined system”.

Bain circuit

What problems did Bain and Spoerel have with the Mapleson D circuit? Well, they complained that there were too many heavy tubes dragging all around the place. The Bain modification puts tube into tube to minimise the tubing and connections. The fresh gas flow is delivered through a thin (6mm) tube, which opens close to the patient’s airway.

The function of this system is virtually identical to the function of the Mapleson D system.

• During inspiration, the patient inhales the content of both the tubes – which should contain mainly fresh gas
• During expiration, the patient fills the external tubing with expired gas
• During the expiratory pause, the fresh gas flow pushes the expired gas up the tubing
• If the fresh gas  flow is sufficiently high, the next tidal volume worth of tube gas will be composed entirely of fresh gas (i.e. same as the Mapleson D, this thing needs about 2-3 times the minute volume of fresh gas flow to function properly, although some textbooks report that it is not essential to eliminate rebreathing completely and at fresh gas flows around 1.5-2 times the minute volume rebreathing is at an acceptable low level)

Mapleson A, otherwise known as the Magill system

This circuit is essentially a reversal of the Mapleson D systems. The position of the valve and gas supply are reversed. Fresh gas flow enters the long tubing at the machine end, and the expiratory valve is mounted on the patient’s end.

The system features a one-way valve which closes when the patient produces any sort of spontaneous inspiratory effort. Thus:

• Fresh gas flow is through the machine end of the tube; fresh gas flows from the supply to the patient
• The patient, when inhaling, can only inhale fresh gas from the tube
• Provided the tube is long enough the patient should only ever get fresh gas. Generally, these circuits tend to have about 1.6 m of tubing.
• Upon exhaling, the expired gas flows through the reservoir tubing until the bag fills and the pressure opens the expiratory valve. At that stage, the patient’s expired gas is vented to the atmosphere via the expiratory valve. 
• Then, during the expiratory pause fresh gas flow pushes more of the expired gas out of the expiratory valve, minimising rebreathing.

This system has some substantial economic advantages over the previously discussed systems. Because there is no need for fresh gas flow to push expired gas out of the circuit or keep up with tidal volume, the fresh gas flow can be equal to the minute volume. The gas flow can actually be reduced even further because of the expired gas, some is the contents of the anatomical dead space, and is therefore indistinguishable from fresh gas (i.e. because no gas exchange took place, there is no CO₂ in it).  The contents of the anatomical dead space is exhaled first, and propelled along the reservoir tubing. The portion of the expired gas which ends up being vented is the last postion, only at the very end of expiration when pressure rises sufficiently to open the expiratory valve. Thus, in the Mapleson A circuit, the first 150ml of every subsequent breath is the anatomical dead space contents from the previous breath. This increases the efficiency of the system such that one only needs to supply approximately 0.7 of the total minute volume in fresh gas flow, and there will still be no rebreathing.

For J.A. Lack (1976), the main disadvantage for this system was the fact that the expiratory valve was very close to the patient, inaccessible during many surgeries, and constantly spewing expired anaesthetic gases into the theatre environment. The author’s modification to address these concerns was the Lack breathing circuit.

Lack circuit: a coaxial modification of the Mapleson A system

The Lack circuit is a coaxial modification which places the expiratory valve at the machine end of the system. The tubing has to be enlarged somewhat from the 6mm coaxial tube of the Bain circuit, because expiratory gas flow is passive and tubing which is too narrow would offer a prohibitively high level of resistance to expiratory flow. Thus, Lack’s coaxial tube had to be 30mm in diameter, in order to accommodate the wider 14mm tubing for the expiratory limb. The main virtue of this system was the ability to place expiratory valve apparatus and scavenging equipment well away from the patient, where its weight does not drag on the airway.

The Lack circuit shares the advantages and disadvantages of the Magill / Mapleson A circuit. They both offer a high level of efficiency for the spontaneously breathing patient, requiring flow rates around 1.0-0.7 times the minute volume, and with minimal rebreathing (the circuit dead space volume quoted in Lack’s article was 4ml).  However, both the Magill and Lack circuits are inefficient when positive pressure ventilation is required. Additionally, there is one other disadvantage. Imagine if the expiratory tube ruptures or becomes disconnected. The entire length of 30mm tubing becomes dead space, filled with rebreathed gas. In order to prevent this outcome, the system’s unique coaxial design would need to be abandoned.

Parallel Lack

By taking the expiratory inner tube out and putting it in parallel with the inspiratory tube, one dodges the possibility of the entire thing becoming filled with expired gas. Gas flow is now unidirectional, preventing rebreathing. The tubing can be a reasonable diameter. And there is no valve near the face. This is the most basic form of the ICU ventilator circuit. It was described by Ooi et al (1993).

Mapleson B system

There is no convenient way to smoothly transition from a discussion of the relatively popular Magill  / Lack circuits to talking about the Mapleson B, an inefficient system which has fallen into disuse.

During the function of this circuit:

• Fresh gas flow fills the reservoir tubing
• The patient inhales the gas mixture from the tubing and bag
• Upon expiration, the fresh gas and expired gas mix in the reservoir tubing
• During the expiratory pause, the continuing fresh gas flow contributes enough volume to the elastic bag that the pressure inside the circuit rises, and the expiratory valve opens
• Mixed fresh gas and expired gas vent through the expiratory valve
• When the patient takes the next breath, it is of mixed gas, consisting of both fresh gas and expired gas

Therefore, the “freshness” of the gas mixture just before the next inspiration depends on the fresh gas flow, and will always contain some admixture of CO₂. The fact that high fresh gas flows are required, of which a proportion is wasted, is a major disadvantage. Additionally, both the expiratory valve and the fresh gas tube attach close to the patients’ face, dragging on the endotracheal tube. Combined with these disadvantages, the inevitable rebreathing of CO₂ makes this circuit a total loser.

In order to overcome the rebreathing limitation, one solution would be to ensure that the expired gas has nowhere to go. If the expiratory reservoir is minimal (i.e. equal to approximately 1 tidal volume) then fresh gas at even a low-ish flow should be able to rinse out much of the expired CO₂-rich gas, so that the breath consists of mainly fresh gas. This theory is behind the Mapleson C circuit.

Mapleson C circuit

Yes, this is basically a greatly shortened Mapleson B system. Because of the fact that the reservoir space is smaller, the mixing of expired gas and fresh gas is more efficient

However, this does not have much of an advantage. If the expiratory pause is 2 seconds, the gas flow still has to supply about 1 tidal volume worth during that time in order to dilute the expired gas. The inevitable venting of gases from the valve makes this at least as inefficient as the Mapleson B circuit. Its only advantage is that it is lightweight and compact, which makes it a viable alternative to the self-inflating resuscitator bag. 

The closed circle circuit

The discussion of these ancient obsolete breathing systems now brings us to a point where we can consider the closed breathing circuit of the modern anaesthetic machine. Though that makes it sound like something relatively recent, in actual fact Brian Sword first described the system in 1930 and so by all rights this thing should be called Swords’ circuit.

Technically, one can only describe this thing as a closed breathing system if the free gas flow rate is exactly the same as the gas uptake by the patient, but in reality the fresh gas flow is slightly higher. For one, the need to inflate the bag or bellows calls for some additional gas.

The main advantage of this circuit is gas economy. Unused anaesthetic gas and unmetabolised oxygen will continue to circulate until they are used or metabolised. CO2 exhaled by the patient is not rebreathed because the soda lime canister in the CO2 absorber constantly removes it from the expired gas. The system only wastes gas when it vents via the APL valve or the ventilator expiratory pressure pop-off valve, i.e when the airway pressure reaches some unacceptable maximum.

Other advantages include:

• Reduced pollution of the operating theatre
• Conservation of heat and moisture
• Decreased risk of soda lime inhalation (the canister is far from the airway)
• Minimal dead space (really, it’s just the Y-piece at the connection to the endotracheal tube)