The CICM WCA document (“Ventilation”) does not make any mention of humidification techniques. However, the topic has some exam relevance. It has come up as half of Question 3 from the second Part I paper of 2012, where the trainees were expected to “describe the mechanisms of humidification used within Intensive Care practice”. A good answer expected them to “briefly describe… the mechanisms of achieving humidification, e.g. bubble systems, heat-moister exchange filters, heated water, ultrasonic, etc. mechanisms.” The pass rate was 0%. For a more practically-minded level of question, the merits and demerits of active and passive humidification systems were explored in the Part Two exam, in Question 15 from the first paper of 2005. For whatever reason, this respiratory equipment question ended up being parked in the already bloated Equipment and Procedures section, and it was answered much better (81% pass rate) as it has actual relevance to the bedside practice of Intensive Care Medicine.
- The objectives of active humidification is to match the normal humidification capacity of the human respiratory tract
- This decreases the risk from inadequately humidified gas (eg. inspissation of secretions)
- The main types of humidifier are bubble, passover, counter-flow and inline vapourisers.
- The main disadvantages of using an active humidifier are the risk of infection and the potential for water precipitation in the circuit tubing
In answer to Question 15 from the first paper of 2005, it is possible to expand on this brief blurb with a tabulated comparison which explores the two main forms of circuit humidification: heat/moisture exchangers and active humidifiers.
|Device description||A hygroscopic in-line air filter||A circuit which incorporates an inline heated water chamber, with an integrated thermostat-controlled heating element|
|Cost||Cheap||Expensive - both the device and the attached consumables|
|Reusability||Single-use||Reusable humidified, disposable circuit|
|Workload||Minimal||Requires attention to water replacement and occasional troubleshooting|
|Humidification efficiency||Low efficiency;
approximately 50% of the required humidity is achieved.
The devices are expected to produce a consistent level of humidity around 30mg/L; whereas 20mg/L is the more typical performance
|Highly efficient. Humidity achieved ranged from 33mg/L to 44mg/L, which is near to the humidity achieved by the human respiratory tract.|
|Lifespan||Should not be used for longer than 72-96 hrs||Provided the circuit is well maintained and regularly changed, humidified ventilation can continue indefinitely|
|Risks with use||Increases dead space;
Becomes progressively more waterlogged, increasing resistance to gas flow;
Potentially, can become a source of infection
|"Rain-out": evaporated water collects in the circuit, pooling and attracting bacteria. The water bath itself is a nice warm environment which acts as a good incubator for bacteria|
|Contraindications to use||Need to minimise dead space;
Large volumes of secretions
Decreased expiratory airflow
Large minute volume (>10L/min)
Long term ventilation
Frequent nebulised medications
|There are no contraindications to circuit humidification.|
For the time-poor exam candidate, this summary is sufficient. The rest of this chapter is a somewhat disorganised exploration of material which is unlikely to be examinable. Because the physics of humidity and the features of disposable Heat and Moisture Exchangers (HMEs) are discussed in greater detail elsewhere, this chapter will largely focus on the methods of actively forcing water molecules to join the inspired gas mixture. There are several excellent resources to refer to, but by far the best is probably the 2014 article by Al Ashry and Modrykamien.
Objectives of active humidification
Humidification for mechanical ventilation aims to match the normal humidification capacity of the human respiratory tract, i.e. to achieve the target temperature of 37°C and 100% relative humidity at the isothermic boundary (just past the carina). This is the expectation upon active humidification equipment. It is more than what is expected of HMEs, which are only expected to perform for brief periods and which are permitted to be less efficient. However, the ICU patient may be trapped on the ventilator for some time, and therefore will require a sustained and reliable source of airway humidity. The aforementioned ICU patient may also be breathing in some profoundly disturbed manner, which might have some relationship to the reason as to why they are mechanically ventilated in the first place. In short, an ideal active humidifier needs to satisfy the following performance characteristics:
- Optimal temperature exiting the ETT (37°C)
- Optimal humidity at that temperature (100% relative humidity)
- Stable delivery of a gas mixture with these characteristics, in the face of
- Erratic respiratory rate
- High or low tidal volumes
- High or low ambient temperature and humidity
- High or low respiratory gas mixture flow rates
- Safe operation (i.e. there should be no situation in which the patient might come to harm as a result of routine operation)
- Freedom from microbial contamination
Basic structure of an active heater/humidifier
There are some basic features which are shared by all such systems, and they vary only by the expense and sophistication of the ways in which they manage the problems of trying to keep the gas clean, warm and humid on the way to the patient.
Humidity in the circuit degrades as the gas flows along the tubing, because the walls of the tubing are generally thin and corrugated. This means they offer a relatively large surface area for heat exchange with the ambient atmosphere. The ambient atmosphere of an ICU is usually very carefully controlled, and well below 37ºC; and because the solubility of water vapour in gas is dependent on temperature the humidity of the gas decreases as it cools on the way to the patient. The duration of exposure to this ambient temperature must therefore also influence this loss in temperature. In short, the effectiveness of the humidifier in terms of providing a stable 37 ºC / 100% humidity is determined by the following factors:
- Ambient temperature and pressure
- Ambient humidity of the supplied gas
- Temperature of the moisture source
- Surface area available for the evaporation
- Length of circuit tubing between patient and humidifier
- Insulating properties of the circuit tubing
- Gas flow rate
The different designs of ventilators differ in their handling of these factors, as will be discussed below. The four major design types are:
- Bubble humidifiers
- Passover humidifiers, which have a few sub-variants
- Counter-flow humidifiers
- Inline vapourisers
Actively heated circuit tubing
Because the circuit precipitation can become a major issue, many humidifier companies bundle their “kettles” with heated circuit components to prevent this “rain-out” phenomenon and to keep the temperature of the circuit stable throughout the circuit length. These tend to come with a heated wire which maintains a certain circuit temperature.
- They are flow-limited: heating efficiency drops with increasing flow (obviously, more gas means more heat exchange). The depicted unit is rated to be maximally efficient only up to 60L/min, which works fine because the circuit is designed to fit into a HFNP system which is limited to 60L/min flow. In the case of non-invasive (or even invasive) ventilation, there may be much leak and flow rates may exceed 100L/min (remember that the ventilators generally max out at about 200-250L/min flow rates). This is much more than most circuit heating systems can deal with; most ventilator-attached systems (like the Drager VentStar Helix for example) are still only rated to guarantee over 33mg/L of absolute humidity at 60L/min, and no more than 10mg/L in the case of NIV.
- They are volume-limited; apart from leak, the practical limitation of this flow limitation is on minute volume, where many circuits lose their efficiency. Lellouche et al (2004) found that with a minute volume of around 20L/min, two models they tested only provided an absolute humidity of around 20mg/L.
- They can cause burns. Lee et al (2014) reported on this happening to an unconscious surgical patient to whose thigh the circuit was clamped (for over 3 hours).
- They do not prevent precipitation in the expiratory limb, which is not usually heated. Expiratory flow may be impeded by water collecting in a loop of circuit tubing.
- They cannot be extended, or if they are extended than the extension tubing often does not benefit from active heating because there is nowhere to connect the heating element.
- The heated wire circuit may not be MRI-compatible.
Bubble humidifier technology
They were fairly popular in the 1980s, but have subsequently lost out to pass-over humidifiers because of notable limitations:
- Slow flow: generally, the slower the flow, the better the humidification (though heated humidifiers overcome this by increasing the water bath temperature, thereby increasing the rate of evaporation and allowing flow rates in excess of 100L/min)
- Flow resistance: the ventilator or gas source has to blow gas into an underwater outlet, i.e. the gas pressure has to overcome the water pressure at a depth equal to the level of the outlet. This is less of an issue for mindless blowers like the high flow nasal prongs, which do not require sophisticated inspiratory flow and pressure control.
- Depth reliance: the deeper the outlet is submerged, the greater the travel distance for the bubbles, and therefore the greater the transition of water vapour into the bubbles. This means, an ideal bubble humidifier would be very deep, and offer a high level of resistance. Most designs made for use with mechanical ventilators tend to compromise with something of a noodle bowl shape, whereas humidifiers expected to work with wall gas pressure and without heating tend to be deeper and more resembling a milkshake glass.
- Noise could be an issue with devices which have large bubbles, but is largely eliminated with the humidifiers which have fine bubble diffusers.
- Bacterial contamination: water baths make an excellent bacterial incubator, and authors such as Rhame et al (1986) have observed the production of aerosols during their operation – a Cascade 1 humidifier produced roughly 500-1000 water droplets per litre of humidified gas in the course of routine operation (10-80 L/min flow rates).
As one might guess from the name, these humidify the inspiratory gas mixture by letting it drift passively over a steaming lake of heated water. Because the outlet is not submerged underwater, this thing only offers the most minimal resistance to gas flow, often even less than the circuit tubing. Because it does not bubble, it is a quiet and polite device.
Clearly, this shares many disadvantages with the bubble humidifier. There is going to be an obvious dependence on flow, i.e. the higher the flow rate and tidal volume the less efficient the humidifier. This is because without bubbles, the surface of the water pot is not large enough to be an efficient surface for water evaporation. One way to get around this is to introduce a wick, which sits in the water and ends up with all its tiny pores coated with water. The surface area of the gas-water interface is greatly increased by this.
Another wickless method of humidification is to use a hydrophobic membrane. This method relies on the fact that the membrane is hydrophobic and therefore will not permit migration of water droplets – only molecular water vapour. Hannsler et al (1992) described the use of a system which used several membrane layers with 10 Angstrom size pores. The membrane, by having tiny pores, also ends up acting as a bacterial filter (though one supposes there is nothing stopping the opposite side of the membrane from becoming contaminated. Obviously there is still going to be a flow dependent problem: the surface area is still only pot-sized, and on top of that there is an increase in the resistance to water flow (proportional to the thickness and permeability of the membrane). Hannsler’s group tested these devices in neonatal ventilation, using flow rates ranging from 4 to 16L/min.
Obviously, all these devices give the impression that their relatively small surface area will render them vulnerable to underperforming at high flow rates. The answer to this may be to increase the efficiency of evaporation by using a counter-current gas flow circuit. Water is heated and drizzled lightly over an evaporator grate which increases the surface area. Gas is pumped from beneath, moving counter to the water’s current. The water which is not evaporated is collected on the bottom of the chamber and recycled.
The idea was that counter-current exchange of water molecules with the gas would be so efficient that gas flow could vary over a wide range but humidification would remain stable (namely, stable at 100%). Schumann et al (2007) were able to demonstrate that these models perform exactly in that way – whether the flow was 3L/min or 120L/min, the total moisture deliver per respiratory cycle remained around 33mg/L (700ml tidal volumes). Unfortunately, Schumann et al also demonstrated that conventional pass-over models also were unaffected by changing the flow rate. The pass-over models ended up humidifying the gas mixture at a slightly less efficient rate only at high minute volumes, where they delivered only 30mg/cycle – a difference which reached statistical significance but is probably clinically meaningless. The design is additionally more expensive and cumbersome because to recycle the water which collects at the bottom of the chamber this device has to use an electrical pump, pumping it back to the top.
These seem to be virtually unknown, and there is little literature on their use since 2011, when Tiffin et al (2011) described their use in the setting of high-frequency oscillatory ventilation. The weird circuits required by these machines do not permit the attachment of conventional humidifiers and require an in-line system. Moreover, the high bias flow of the circuits (30L or so) tend to make normal humidifiers ineffective.
There appears to be only one company making these (Hydrate Inc., who make the Hydrate OMNI™) and their supporting literature consists mainly of abstracts and posters from conferences. The device itself is a peristaltic pump drives water into the heater capsule, which then passes it through a ceramic disk full of capillary channels and into range of a small heating disk, which blasts the water in the capillaries with such fierce heat that the jet of steam squirts out of the device and into the respiratory gas mixture.
It’s the ultimate way of divorcing the water vapour flow rate from the ventilator circuit flow rate. You let the ventilator blow as much as it wants and you just add whatever amount of hot steam you need, on demand. The limits really become related to the rate at which you can deliver water vapour to the heating element. This specific model can apparently humidify gas flows up to 40L/min.
This method should be mentioned as a legitimate means for increasing the humidity of inspired gases, largely because of its major advantage: it disconnects the temperature of the gas from its humidity. By filling the inspiratory gas mixture with nanoscale water particles (smaller than an aerosol) the ultrasonic nebuliser can double the water content of the inhaled gas mixture. Klein et al (1973) found that this does not need to produce any sort of visible fog.