Chris Thompson  

Humidity and Humidification


The prevention of cellular dehydration is a primary human homeostatic requirement. Complex and reliable mechanisms exist to maintain overall fluid balance.

Skin and other integuments are relatively impermeable to moisture, reducing evaporative water loss and consequent cooling. Special problems occur with specialised tissues (ie, cornea, airways, lungs), and under abnormal conditions (eg, burns, surgical procedures, some illnesses).

Anaesthetists need to maintain both global fluid balance and local tissue needs during anesthesia. The purpose of this talk is to generally review evaporative water loss and consequent heat loss during anaesthesia, with particular emphasis on how this can be influenced by humidification of breathing circuit gases.



Gas phase of a substance which is normally a liquid at ambient temperature and pressure.

Dalton's Law

The pressure exerted by a mixture of gases or vapours enclosed in a given space equals the sum of the partial pressures that each gas or vapour would exert if it alone occupied the same space.

Saturated Vapour Pressure (SVP)

The saturated vapour pressure of a liquid is the partial pressure of the vapour above its liquid state at equilibrium, and is very dependent on the type of liquid and its temperature.

If a volatile liquid is introduced into a closed container, the pressure in the container will increase in proportion to the partial pressure of the vapour. In an expansile system the volume of gas increases (if the temperature is unchanged) because of the addition of the vapour and all constituents are proportionally diluted.

The SVP of water at 37°C is 47mmHg and contains 44mg/l of water whereas at room temperature (20°C) it is 20mmHg and contains only 18 mg/l).

SVP at




Temperature (°C)




SVP (mmHg)




Content (mg/l)




Table 1: Water Vapour, temperature, and SVP at 0, 20, and 37°C.

Absolute Humidity

The absolute amount of water vapour in a gas expressed in either mg/l of gas mixture or mmHg (partial pressure) .

Relative Humidity

Amount of water vapour in a gas expressed as a percentage of that which could be held by the gas if it were fully saturated at the same temperature, ie:

Relative Humidity. = Actual Water Content / Water Content Fully Saturated %, or
Absolute Humidity = Actual Vapour Pressure / Saturated Vapour Pressure %

Relative Humidity is the common desciption of humidity used in weather reports as it correlates best with our perception of dryness or moistness of the air.


Heat is a form of energy that can be transfered from a warmer object to a cooler one and is related to the kinetic energy of vibration of molecules. Its units are Joules. 4.18 Joules equal one calorie and will raise one gram of water one degree centigrade. Total body heat production is about 80 Watts (ie Joules per second) for the average person (or 288 kJ/hr, 7,000 kJ/day, or about1,700 kcal/day). Shivering can increase heat production up to four-fold, and exercise even more.

Specific Heat

Specific heat is the amount of heat required to raise the temperature of 1 kg of a substance by one degree Kelvin (the same as one degree centigrade); the units are Joules Kg-1 degK-1. The specific heat of water is 4.18 KJ per kg per °C, but it is often written as 1 kcal or 1 Cal kg-1 °C-1. Calories and calories should no longer be used. The specific heat of air is only 1.2 J litre-1 °C-1 ie 1/500th that of water.

Heat Capacity

This refers to the amount of heat required to raise the temperature of an object by one degree Kelvin. For a 70 kg patient this is about 245kJ per °C. As basal heat production is about 280kJ/hr it takes a little less than 1hour to passively raise core temperature of a patient 1°C even if all losses are prevented.

Specific Latent Heat of Evaporation

The heat required to convert 1 kg of a liquid to its vapour at a given temperature. For water this is about 2.4 MJoules per kg at body temperature and only slightly greater at room temperature.


Generally speaking this is very difficult.


A system using two thermometers, one with a wet and the other a dry bulb. Air movement over the wet bulb causes evapourative cooling generating a difference in temperature readings. This difference relates to the rate of airflow over the wet bulb and the relative humidity. Tables are used to look up the relative humidity from the two temperatures, however there is a high degree of inaccuracy. When the relative humidity is 100% there is no temperature difference.

Dewpoint Hygrometer

Using a precisely cooled shiny plate the user observes the temperature at which condensation first occurs. At this temperature the gas is fully saturated with water hence both the water content and the relative humidity at any other temperature can be ascertained from a vapour pressure table.

Weighing Water

By weighing anhydrous silica before and after exposure to a known volume of sample gas the water content of the sample can be determined. The silica must be kept away from other sources of moisture during weighing, and in practice this is a cumbersome and tedious technique.

The performance of active humidifiers is most simply tested by measuring the outlet temperature, the amount of water consumed per unit time, and the gas flow. The relative humidity at the outlet of the humidifier under test (at this temperature and gas flow) can be established from the number of milligrams of water taken up per litre of gas. For example, 10 l/min at 37°C results in 600 litres of gas flow over one hour, and if each takes up 44 mg/l (to be fully saturated at this temperature), 26.4g of water should have been used over the hour. If only 20mg was taken up, the relative humidity was 20/26.4 or 75.8%.

Mass Spectroscopy

Mass Spectroscopy can be very accurate but only if condensation (rain-out) does not occur in the sample line. This is the best technique for assessing "in-circuit" humidity as it can assess breath by breath changes.

Humidity Transducers

Special transducers are available in which the electical conductivity of a membrane changes with water vapour pressure are available.


Normal humidity in the airways

While nose breathing at rest, inspired gases become heated to 36°C and are about 80% to 90% saturated with water vapour by the time they reach the carina, largely due to heat transfer in the nose. Mouth breathing reduces this to 60% to 70% relative humidity. Heat and moisture content falls from carina to nares, so that the nose is typically at 30°C. A countercurrent mechanism of heat and moisture exhange in the aiways maximises efficiency, with nasal cooling on inspiration and warming on exhalation. On very cold days while exercising a dripping nose is evidence of this. Tracheal temperature and humidity fall with increased ventilation particularly when the inspired gases are cold and dry.

Heat and Water Loss

If totally dry gases were inspired and fully saturated gases exhaled the total water loss from ventilation at rest would be about 300 ml/day in the average adult. Normally about half is retained thanks to the efficiency of the nose (30% saving) and the humidity of inspired room air (25% saving). Bypassing the nose with an ETT and not humidifying gases cases maximal losses.

Non-respiratory water losses are typically 300 to 600 ml/day but are increased if warm moist surafces are exposed (ie burns, open abdomen) particularly if the operating theatre is cold and has high flow airconditioning.

Heat losses are the result of four primary processes:

  • Radiation 40% (depends on clothing; 4th power of absolute temp diff 10°C = 18%)
  • Convection 30% (increased in windy environments)
  • Evaporation 20% (ie, from skin)
  • Respiration 10% NB: 8% water evaporation, 2% heating Air

Respiratory losses of both heat and water increase with increased ventilation, hyperthermia, and dry inspired gases.


Reduced heat loss

Heat loss from warming inspired air from 20 deg.C to 37 °C:

=   Ventilation x Specific Heat Air x Temp rise
=   7 litre min-1 x 1.2 Joule litre-1 deg.C x 17 °C
=   142.8 Joule min-1 (or 8.6 kJoule per hour or 2.38 watts)

Thus warming the inspired air from 20 °C requires only about 3% of the body's normal heat production and at maximum could only cool a non-heat producing body mass by 0.035 °C per hour.

Heat loss from humidifying exhaled air fully saturated at 37 °C:

=   Ventilation x Water required x Specific latent heat of vapourisation
=   5 litre min-1 x 44 mg litre-1 x 2.4 MJoule kg-1
=   528 Joule min-1 (or 31.7 kJoule per hour or 8.8 watts)

Hence in the worst case (completely dry inspired gases and, no countercurrent benefits), humdification about 11.3% of the body's heat production. Thus fully humidifying completely dry warm air requires four times as much heat as warming it alone. Even so, in the worst case at rest, the total rate of body cooling would be only 0.16 °C per hour from ventilation. If a 70% efficient HME was used the total heat loss would only be 0.05 °C per hour and the total heat loss from ventilation over a 10 hour period would be only half a degree. In contrast, rapidly infusing one litre of water at 20 °C would cool the patient by 0.3 °C almost immediately.

Note that respiratory heat loss increases if the inspired air is very cold or ventilation is increased considerably, for example exercise at high altitudes.

Very importantly neonates and infants have metabolic rates (and hence ventilatory requirements) approximately 2 to 3 times that of adults on a weight basis. They stand to lose a lot more heat relative to their heat capacity (body mass) and hence will cool down 2 to 3 times quicker from ventilation. Neonates stand to lose 0.3 to 0.5 °C per hour from respiratory heatloss (more if hyperventilated) unless gases are humidified.

Consequently in adults either 70% humidification of inspired gases for 10 hours or no humidification for 3 hours result in possible total temperature falls of only 0.5 °C. Radiant heat loss and heat loss from room temperature fluid infusions far exceeds this is clinical practice. Unless there are other potential or actual problems with hypothermia, or to avoid excessive drying of secretions, using any form of humidification in adults for procedures of 2 to 3 hours duration is unjustified, in my opinion.

In neonates, however, respiratory heat loss may be significant even during short periods of ventilation.

A small but important practical point is that body temperature should be measures fom a wedged nasal rather than oesophageal temperature probe, as oesophageal probes tend to measure endotracheal tube rather than patient temperature.

Reduced water loss

During anaesthesia it is easy to replace the small respiratory water loss with iv fluids. Prolonged extreme exertion as in fun runs and mountaineering may cause significant dehyration from respiration and adequate fluid intake is essential.

Prevention of cilial damage and reduced drying of secretions

Cilial paralysis and reduced rates of mucus flow occur below 50% relative humidity at 37 °C, but how long it takes for irreversible and/or significant changes is not known. During brief general anaesthetics this is not a problem. Prolonged severe dehydration of the bronchial tree leads to encrustation of mucus and bronchial or endotracheal obstruction, particularly in neonates and patients with respiratory infection.

Recommendations for humidification of inspired gases on this basis are generally anecdotal.

Microbial Filtration

Some HME's incorporate viral/bacterial filters.



All humidifiers increase the component count in the breathing circuit and increase the risk of disconnection. Heated humidifiers commonly require actively heated delivery tubes and these may be heavy and bulky or use non-standard connectors. Much less of a problem with HME's although some have non-standard connectors or are quite bulky.


An uncommon event with modern servo controlled active humidifiers, although burns from the delivery hose have been recently reported. Impossible with HME's.


Usually only a problem with nebuliser type devices. Impossible with HME's.

Circuit resistance, deadspace, and circuit compliance changes

Most modern HME's cause a small increase in resistance to gas flow, typically 2 cm H20 at 40 l/min (a typical inspiratory flow rate during anaesthesia). Obstruction of HME's with mucus or as a result of expansion of saturated heat exchanging material may occur and can result in dangerous increases in resistance. Heater humidifiers also increase circuit resistance but usually to a lesser extent (provicded that tubing of adequate diameter is used). Bubble-through humidifiers cause obvious increases in resistance. Rain-out may cause obstuction of breathing tubes.

Deadspace considerations in HME's limit their performance but not their clinical utility. Greater mass of heat exchanging material improves performance (especially with larger tidal volumes) but the deadspace increases as well, so usually it is necessary to choose the right size HME to suit the patient.

Increased circuit compliance is important to consider when ventilating neonates.


Not a problem with disposable HME's, but can occur in ICU with the water bath of heated humidifiers.


Possible on tilting the water bath of some heater humidifiers, particularly for neonates on continuous flow circuits. Can't occur with HME's.

Interference with other devices

Excessive humidity in the proximal breathing circuit from heated humidifiers may interfere with sampling (side-stream) type CO2 analysers and condensation may affect the reading on some tidal volume meters. Rain-out from active humidifiers may be a considerable problem, particularly in ICU's.

Inadequate humidification

HME's are probably inadequate for:

  • Prolonged ICU use, ie more than 2 to 3 days
  • More than 6 hours or so where respiratory secretions are a problem
  • Warming cold patients
  • Hyperventilation.

Active heater humidifiers can provide 100% relative humidity at 37 °C or more for prologed periods and are preferable in the above situations


Disposable HME's cost $2.00 to $7.00 per patient. A F&P type dome costs $25.00, and as well as the capital cost of the base, the delivery tube will have to be sterilised at the end of the case.


Anaesthetic circuits

Water's to and fro type systems generate warm moist gases but are little used these days. Bain type circuits result in both countercurrent heating of the inspired gases and rebreathing of exhaled gas for some humidification, but they probably only are about 10% to 20% efficient on IPPV and even less when used for spontaneous breathing (because of the high fresh gas flows required). Circle circuits warm up after a period of time and do generate water but again are relatively little help.


Rarely used these days as they nebulisation process results in an inhaled aerosol of 100% relative humidity at lower than room temperature, which increases heatloss rather than reducing it. Water uptake may be excessive and there is a real potential for infection from the water bath. Used only to liquefy secretions, but these days active humidification is far preferable.

Heat and Moisture Exchangers (HME's)

Initally made for tracheostomied patients from copper mesh ("Swedish nose"). Now most are cheap and disposable and made from modified hygroscopic paper filters encased in a plastic case.

HME's are usually 50% to 80% efficient at best (depending on inspired humidity), consequently heat and water loss and dehydration of respiratory secretions still occur, particularly oif the inspired gases are completely dry. Efficiency is reduced further by large tidal volumes or by failure to place the HME right on the endotracheal tube, permitting rain-out. Full function is not immediate, typically taking 5 to 20 minutes to achieve steady state.

None the less they are almost as effective as the human nose and provide enough humidification to maintain ciliary action and mucus flow and to reduce heat and moisture losses during anaesthesia to insignificant levels. The Pall filter is also one of the best bacterial filters on the market. They are simple to use, cheap, act as a macroscopic particle trap and avoid many of the problems associated with active humidifiers.

Active Heated Humidifiers

These electrically powered heated water bath devices are capable of fully saturating and heating inspired gases to 37 °C at high flow rates flow rates (ie to 60 l/min) and in this regard are far superior to HME's. A heated delivery hose is required to prevent cooling and loss of humidity in the inspiratory limb of the circuit. Sterile single use water chambers with wicks are common, and some are automatically filled from a water reservoir.

They have many disadvantantages, including cost, storage requirements, servicing, circuit complexity, water rain-out leading to monitor malfunction or tubing occlusion, extra risk of disconnection, infection hazard, potential for burns and drowning, and altered circuit compliance.

However I think they are definitely required for ICU patients who need hyperventilation for prolonged periods, particularly for neonated and small children. In my opinion the rationale has little to do with heat loss, or even prevention of ciliary dysfunction. The main benefit is to ensure that there is no net evaporation of water from secretions that lodge in the lumen of the endotracheal tube, so that they do not become dehydrated and encrusted and lead to obstruction.


Last updated Tuesday, April 13, 2010
Comments? please use this form.