MINERVA ANESTESIOL 1998;64:75-81

Comparison

A
of three different humidification systems

IC
during prolonged mechanical ventilation
M. LUCHETTI, A. STUANI*, G. CASTELLI*, G. MARRARO

T® D
H E
IG M
Background. An efficient humidification
system is expected to maintain fluid and
Azienda Ospedaliera
«Fatebenefratelli e Oftalmico» - Milano
easily drainable airway secretions. This study Servizio di Anestesia e Rianimazione
aims to compare the efficiency and safety of *Azienda Ospedaliera «Carlo Poma» - Mantova
three humidification systems during pro- Servizio di Anestesia e Rianimazione
R A
longed mechanical ventilation.
Design. Two-center, prospective, randomized
study.
Y V

Methods. 45 critically ill patients undergoing Conclusions. Passive humidification systems

mechanical ventilation were included in the provided low degrees of humidity and tem-
study and allocated to receive one of three perature and could not maintain good secre-
humidification techniques: 1) Bennett Cas- tions. Active systems appeared to satisfy the
P R

cade water-bath humidifier (Bennett group); recommended standards and to allow fluid
2) Fisher & Paykel servocontrolled humidi- and easily drainable secretions.
fier (F & P group); 3) HME Hygrobac DAR Key words: Respiration artificial, instrumenta-
O E

(HME group). Clinical and experimental tion - Intensive Care - Humidity - Ventilators
observations were conducted for 3 to 7 con- mechanical.
secutive days and included: body T°, room T°,
C IN

inspired gas T°, tracheal T°, relative and abso-

lute humidity, heat and water loss, airway
secretion score, need for endotracheal saline
instillation and incidence of ETT occlusion.
Results. The HME group showed a lower tem-
T he upper respiratory tract functions as a
heat and moisture exchanger. Under
normal physiologic conditions inhaled air
M

perature of inspired gases compared to the F
& P group (p<0.05); it also showed a lower
absolute humidity compared to both Bennett
and F & P groups (p<0.05). A better airway
secretion score was obtained in Bennett and F
& P groups compared to the HME group
(p<0.01).
This work was presented in part as an abstract at the
APICE Congress held in Trieste, Italy, November 13-19,
1995.
Received September 11, 1997.
Accepted for publication February 3, 1998.
Address reprint requests to: M. Luchetti - Via Varese, 4 -
20121 Milan, Italy.
reaches the lungs saturated with water
vapor at a temperature corresponding to
the body temperature.
The absolute humidity of inhaled air in
the lungs, if completely saturated with
water vapor at 37 °C, is 44 mg/l. In the
spontaneously breathing patient inspired
gases are usually warmed in the upper
airway to 34 °C.1
During nasotracheal intubation and me-
chanical ventilation a substantial part of the
upper respiratory tract is by-passed, elimi-
nating at least one-fifth of the air condi-
tioning surface area. In this case adequate

Vol. 64, N. 3 MINERVA ANESTESIOLOGICA 75

LUCHETTI COMPARISON OF THREE DIFFERENT HUMIDIFICATION SYSTEMS DURING PROLONGED MECHANICAL VENTILATION
humidification and warming should be pro-
vided by external devices. To such purpose
active systems (humidifiers) or passive
systems (heat and moisture exchangers,
HMEs) are available.
A
The active humidifiers supply the patient
with heat and humidity by warming and
IC
saturating the ventilated gas with water
vapor.
The HMEs achieve humidification similar
to normal upper tract humidification. Du-
T® D
ring expiration, the warm and wet gases
pass through plastic fibers which retain heat
and water. During the following inspiration
H E
this heat and humidity are released, thus
warming and humidifying the cold and dry
inspired gas.2 3
IG M
Basically, an efficient humidification
system is expected to maintain fluid and
easily drainable airway secretions. Exces-
sive or poor humidification have to be
R A
avoided, since both expose the patient to
related side-effects, such as impairment of
the mucociliary function,4 5 damage to the
Y V
tracheal mucosa, atelectasis, hypothermia,
hyperthermia, water intoxication, bron-
chopneumonia.
P R
There is no agreement as to which is the
best system for use during prolonged
mechanical ventilation. Both passive and
O E
active systems have advantages and disad-
vantages and even among active systems the
C IN
differences in performance are important.
The present study aims to assess which,
among three different humidification
systems, two active and one passive, is
M
more efficient and safer during long-term
mechanical ventilation, in terms of the heat
and humidity they are able to provide and
of their effect on airway secretions.
Materials and methods
After local ethical committee approval
and informed consent, 45 patients, 30 male
and 15 female, aged between 18 and 84
years, were included in the study. They
were affected by three different diseases:
post-anoxic coma, complicated COPD and
polytrauma and were all nasotracheally
76 MINERVA ANESTESIOLOGICA
intubated and artificially ventilated. The
patients were prospectively and randomly
allocated to one of 3 different humidifica-
tion systems.
In 15 of them the humidification of venti-
lated gases was obtained by means of a
Bennett Cascade II water-bath humidifier
and a normal inspiratory line (Bennett
group). This type of humidifier does not
allow precise temperature settings but only
approximate ones, on a numeric scale
ranging from 0 to 12. In this study it was set
at 8, as in our routine practice.
In 15 patients a Fisher & Paykel MR600
servocontrolled humidifier with heater wire
and a coated inspiratory line were emplo-
yed (F & P group). In this case the temper-
ature of the humidifier was set at 37°C, in
accordance with our routine practice.
In other 15 patients (HME group) humid-
ification was provided by passive systems,
consisting of filters/HMEs (DAR Hygrobac)
and a normal inspiratory line.
The Hygrobac DAR consists of two water-
repellent electrostatic pre-filters, a micro-
porous membrane performing mechanical
filtration, and a layer of cellulose sponge
acting as the heat and moisture exchanging
element, the four layers all housed in a
transparent plastic casing. The HME is
claimed to allow an inspired relative
humidity of 95% at 32 °C and a moisture
output of at least 30 mg/L.6 The Hygrobac
DAR is employed in the majority of ICUs in
Italy and is reported to be one of the best
HMEs available. This was the reason for
selecting this particular brand.
The HME was used from the moment of
intubation and was changed every 24
hours, according to the manufacturer’s
advice.
Fluid balance was maintained according
to the individual needs of each patient. In
general, attention was paid that neither
hyperhydration nor hypohydration oc-
curred.
All the measurements and recordings
were performed every day for 7 days or
until the extubation of the patient if this
occurred earlier. Patients with fewer than 3
recordings were excluded from the study.
Marzo 1998
COMPARISON OF THREE DIFFERENT HUMIDIFICATION SYSTEMS DURING PROLONGED MECHANICAL VENTILATION LUCHETTI
At each daily observation, ventilatory set-
tings, room and body temperatures, and
blood gases were also recorded.
Humidification and warming data were
obtained by using a chart recorder (Yoko-
A
gawa MR 100, Japan) and 4 silica-titanium
fast-response thermal probes with 0.002%
IC
precision.
A rotatory circuit with unidirectional val-
ves, interposed between the Y piece or the
filter/HME and the ETT allowed separation
of the inspired flow from the expired flow.
T® D
The inspired gas temperature was meas-
ured by means of two probes inserted on
H E
the inspiratory flow, one of these naked
(dry probe, T1) and the other drapped in a
thin wet piece of cloth (wet probe, T2). The
IG M
third probe (T 3 ), inserted into the ETT
down to its tip, measured the tracheal tem-
perature. When active systems were used, a
fourth probe (T4) was positioned at the exit
R A
of the humidifier in order to measure gas
temperature at that level.
Continuous recordings were carried out
Y V
for at least 30 minutes, the time required for
the system to “warm up”. The sum of the
daily measurements in each patient was
P R
averaged. The values obtained were then
averaged for each group.
Relative humidity was obtained from dry
O E
and wet probe temperatures by means of a
psychrometric method. Absolute humidity
C IN
at saturation point (AHs) was derived
according to the following formula:
AHs=16.41563-0.731 T+0.03987 T2,
where: AHs=absolute humidity at 100% RH
M
(mgH2O/L), and T=dry probe t° (°C).
Absolute humidity was then calculated
from the following formula:
AH=(AHs×RH)/100 (in mgH2O/L).
The heat loss along the inspiratory line
was also calculated, as the difference bet-
ween T4 and T1. The difference between
absolute humidity values derived from T4
and T1 gave the water loss along the inspi-
ratory line.
The quantity, the quality and the viscosity
of airway secretions were evaluated accor-
ding to the rating scale described in Table I.
This is a scale created by us and routinely
Vol. 64, N. 3 MINERVA ANESTESIOLOGICA
TABLE I.—Airway secretion score (AWSS).
Points 0 1 2
Quantity Poor Normal Abundant
Quality Purulent Brown White
Viscosity Thick Fluid —
The final score is given by the sum of the scores of each
parameter and ranges from 0 to 5 (worst - best secretions).
used for the daily evaluation of airway secre-
tions in our intensive care unit. Due to the
lack in the literature of any similar scale, it
was decided to employ the scale for the pur-
pose of the study. The sum of the 3 ratings
gives a score (AWSS - Airway Secretion Score)
ranging between 0 (worst secretions) and 5
(best secretions). The AWSS is calculated
once a day at the same time (8 a.m,). For
each patient the average of all the determina-
tions all along the study was obtained. Those
values were then averaged for each group.
The need for ETT instillation with saline
was reported as well as the incidence of ETT
occlusion. This latter was defined as the
nurse feeling of obstacle while suctioning
together with an increase of insufflation pres-
sure, and confirmed by observation of the
inside of the ETT once it had been removed.
Statistical analysis was performed by one
way ANOVA for comparison between
means, and by χ2 test for the other data. A
p value <0.05 was considered significant.
Results
The groups were comparable with regard
to patients’ age, weight and body tempera-
ture, room temperature, ventilatory settings,
humidifier temperature and blood gases.
Different pathologies were homogeneously
distributed in the groups (Table II).
The other relevant results are summar-
ized in Table III.
The HME group showed a lower temper-
ature of inspired gases compared to the F &
P group (p<0.05); it also showed a lower
absolute humidity compared to both Ben-
nett and F & P groups (p<0.05).
Relative humidity was found to be lower
in the F & P group (p<0.05).
77
LUCHETTI COMPARISON OF THREE DIFFERENT HUMIDIFICATION SYSTEMS DURING PROLONGED MECHANICAL VENTILATION

TABLE II.—Comparison of demographic data, underlying diseases, temperatures, ventilatory settings and blood
gases.
Bennet F&P HME

Patients (n°) 15 15 15

A
Age (years) 056.1±23.7 058.5±20.4 056.3±18.8
Weight (kg) 071.4±16.4 73.1±7.7 072.8±14.2
Underlying diseases

IC
Post-anoxic coma (n° pts) 4 3 5
Complicated COPD (n° pts) 6 7 6
Politrauma (n° pts) 5 5 4
Temparatures
Body (°C) 37.1±0.8 37.0±0.6 37.1±0.6

T® D
Room (°C) 24.8±1.5 23.4±1.7 24.2±1.8
Humidifier (T4) (°C) 37.5±0.4 37.3±0.3 —
Ventilatory settings
RR (breaths/min) 17.8±1.2 18.0±0.8 17.5±1.1

H E
TV (ml) 559.0±44.3 566.6±50.4 594.2±58.6
FiO2 00.42±0.05 00.45±0.05 00.46±0.06
PEEP (cmH2O)
IG M 08.2±0.3 08.8±0.3 08.7±0.4
Blood gases
PaO2 (kPa) 11.4±1.4 11.9±1.6 12.2±1.6
PaCO2 (kPa) 05.6±8.8 05.3±0.7 05.4±0.5
All values are expressed as mean ±SD except for “underlying diseases” (number of patients). No significant difference was
found; RR=respiratory rate; TV=tidal volume; FiO2=fraction of inspired oxygen; PEEP=positive end expiratory pressure; PaO2=arte-
R A
rial tension of oxygen; PaCO2=arterial tension of carbon dioxide.
Y V

TABLE III.—Comparison of temperature, humidity, No statistically significant difference bet-

heat and water loss, secretion, and incidence of ween groups was found with regard to tra-
saline instillation and ETT ostruction.
P R

cheal temperature, although the HME

Bennet F&P HME group showed a lower mean temperature.
In the F & P group lesser heat and water
O E

Patients (n°) 15 15 15
Measurements (n°) 48 50 59 loss was found compared to the Bennett
Insp gas dry T° (T1) group (p<0.05).
C IN

(°C) 32.8±1.4 34.9±1.3 30.5±1.6° A better airway secretions score was ob-
Insp gas wet T° (T2) tained in Bennett and F & P groups com-
(°C) 32.5±1.3 33.0±1.5 30.3±1.20
Tracheal T° (T3)
pared to HME group (p<0.01).
(°C) 35.6±1.4 36.0±1.2 33.4±2.40 Compared to the active humidification
Insp rel. hum groups, the HME group also showed a
M

(%) 98.5±0.5 99.0±0.4 98.9±0.20 greater need for endotracheal tube saline
Insp abs. hum.
(mgH2O/L) 34.9±1.5 35.6±1.6 31.0±2.5* instillation (p<0.05) and a higher incidence
Heat loss of ETT obstruction (p<0.05).
(°C) 04.7±1.7 002.4±1.3* — The three patients of the HME group who
Water loss
(°C) 07.8±1.6 003.6±1.5* — had their ETT occluded were then switched
AWSS over to active humidification.
(0-5) 03.5±0.8 03.8±1.0 01.9±1.3§
Saline instillation
(n° pts) 0 0 3*
ETT ostruction Discussion
(n° pts) 0 0 3*
All values are expressed as mean ±SD except for “saline Recently, a few works have addressed
instillation” and “ETT obstruction” (number of patients); the topic of humidification during long-
°p<0.05 compared to F&P group; *p>0.05 compared to the
other groups; § p<0.01 compared to the other groups; term mechanical ventilation with different
AWSS=Airway Secretions Score; ETT = endotracheal tube.
results.7-10

78 MINERVA ANESTESIOLOGICA Marzo 1998

COMPARISON OF THREE DIFFERENT HUMIDIFICATION SYSTEMS DURING PROLONGED MECHANICAL VENTILATION LUCHETTI
The first thing a humidification system is
asked is to maintain the airway secretions
and the mucociliary clearance as similar as
possible to physiologic. The present study
aimed to compare the efficiency and safety
A
of three different humidification systems,
by assessing their capability of providing
IC
given degrees of temperature and humidity
and, more important, their capability of
maintaining fluid and easily drainable
airway secretions, thus avoiding complica-
T® D
tions such as obstruction of the ETT.
The optimal humidity degree of the
inspired gases during prolonged mechan-
H E
ical ventilation is not well established yet,
neither is the minimum accepted level.
Actually, values published in the litera-
IG M
ture range from 17 to 44 mgH2O/L.11-14
In 1970 the British Standards Institution
recommended a tracheal temperature of
35 °C and an absolute humidity of at least
R A
33 mgH2O/L to maintain good secretions
and avoid complications during mechan-
ical ventilation.15
Y V
In a recent paper Williams et al. 16 analyze
the evidence from the literature and suggest
that warming and humidifying gas to its nat-
P R
ural target of core temperature (37 °C) and
100% relative humidity (44 mg/L) achieves
O E
full gas conditioning, maintains the rhe-
ology and normal volume of airway secre-
tions, maximizes mucociliary clearance,
C IN
prevents inflammatory reactions, sustains
airway patency and preserves lung compli-
ance. Any deviation of the inspired gas
from the core temperature and 100% rela-
M
tive humidity condition would result in
both thermal and water mass challenges to
the mucosa. The damage of the mucosa
would be greater with increasing exposure
time.
Some previous works, which considered
lower values of humidity as sufficient,
reported apparently good performance for
several kinds of HMEs,2 17-19 although with
many limitations, such as a relevant warm-
up time (45-60 minutes), a short activity
time (max 24 hours), a tidal volume-depen-
dant efficiency (large tidal volumes cause
high water loss) and an inadequacy for use
in children (high dead space).
Vol. 64, N. 3 MINERVA ANESTESIOLOGICA
Compared to active humidifiers, heat and
moisture exchangers are reported to have
also some advantages, such as ease of han-
dling, low weight, small dead space, no risk
of water intoxication, bacterial filtering
capability and reduced nursing require-
ments.6 HMEs’ performance and safety in
recent years have greatly improved and
their use is now widespread. Given the sim-
plicity of the devices, many intensivists may
find that the level of humidity they are able
to provide is acceptable.
Undoubtedly, active humidifiers require
more nursing and do not offer a bacterial
filtering capability. When not properly
used, they can even increase the risk of
bacterial contamination, and represent a
potential for water-drowning. These are
probably the major reasons for someone
avoiding their use.
In the present study HMEs appeared not
to meet the aforementioned requirements,
providing lower degrees of humidity and
temperature than recommended.
The poor humidification appeared to
increase the need for endotracheal instilla-
tion with saline prior to endotracheal suc-
tioning. This is a ritual that persists despite
a lack of benefit and a potential for infec-
tion. In fact, saline instillation may dislodge
viable bacteria from a colonized endotra-
cheal tube into the lower airway, over-
whelming the defense mechanism of immu-
nocompromised patients.20
In three patients of the HME group the
poor humidification led to endotracheal
tube occlusion, requiring the ETT change.
In these patients, the decision was taken to
switch over to active humidification.
Improved airway secretions were ob-
tained within a few hours of active humidifi-
cation. Other A. have previously docu-
mented the risk of endotracheal tube occlu-
sion associated with the use of HMEs during
prolonged mechanical ventilation.21-23
Active humidification systems, in this
study, behaved much better than passive
systems. Both types of active humidifiers
provided mean tracheal temperature and
absolute humidity well above the minimum
recommended; this resulted in fluid and
79
LUCHETTI COMPARISON OF THREE DIFFERENT HUMIDIFICATION SYSTEMS DURING PROLONGED MECHANICAL VENTILATION
easily drainable secretions with no inci-
dence of endotracheal tube obstruction.
Servocontrolled humidifiers with heater
wire inside a coated inspiratory limb ap-
peared to be more efficient and safer. The
A
other active systems, although capable of
maintaining good secretions, showed a sub-
IC
stantial heat and water loss along the inspir-
atory line. This caused abundant condensa-
tion in the circuit, which could possibly
increase the risk of infection and water-
drowning. However, these latter were not
T® D
demonstrated in this study, as we did not
look into the incidence of either infectious
H E
complications or water-drowning.
IG M Conclusions
Surely, more prospective, randomized,
controlled studies are needed on the topic
R A
of humidification during prolonged
mechanical ventilation, but according to the
data obtained in this study two conclusions
Y V
can be drawn:
1) both types of active humidification
P R
systems tested in this study appear to be
able to maintain good airway secretions
even during prolonged mechanical ventila-
O E
tion by providing an adequate humidity
degree, very close to core temperature;
2) heat and moisture exchangers, at least
C IN
those which have been employed in this
study, seem to be inadequate for humidifi-
cation during prolonged mechanical venti-
lation and, in our opinion, should be
M
employed only for short-time ventilation.
Riassunto
Tre differenti sistemi di umidificazione a confronto
durante ventilazione meccanica prolungata.
Obiettivo. Un efficiente sistema di umidificazione
dovrebbe mantenere le secrezioni delle vie aeree
fluide e facilmente drenabili. Lo scopo di questo
studio è quello di confrontare l’efficacia e la sicu-
rezza di tre differenti sistemi di umidificazione
durante ventilazione meccanica prolungata
Disegno sperimentale. Studio comparativo, pro-
spettico, randomizzato.
Pazienti e metodi. Quarantacinque pazienti
80 MINERVA ANESTESIOLOGICA
ricoverati in Terapia Intensiva e sottoposti a venti-
lazione meccanica sono stati inclusi nello studio e
suddivisi in maniera randomizzata in tre gruppi a
seconda della tecnica di umidificazione adottata:
1) umidificatore attivo a bagno d’acqua Bennett
Cascade (gruppo Bennett);
2) umidificatore attivo servocontrollato Fisher &
Paykel (gruppo F & P);
3) scambiatore di calore e umidità HME Hygro-
bac DAR (gruppo HME). Per 3-7 giorni consecutivi
sono state condotte osservazioni cliniche e speri-
mentali consistenti in: temperatura (T°) corporea, T°
ambiente, T° dei gas inspirati, T° tracheale, umidità
relativa e assoluta, perdita di calore e umidità,
secrezioni delle vie aeree, necessità di lavaggi con
soluzione fisiologica del tubo endotracheale e inci-
denza di occlusione del tubo endotracheale.
Risultati. Il gruppo HME ha mostrato una T° dei gas
inspirati più bassa rispetto al gruppo F & P (p<0,05),
nonché un’umidità assoluta inferiore rispetto sia al
gruppo Bennett che F & P (p<0,05). Nei gruppi Ben-
nett e F & P è stato ottenuto un punteggio delle
secrezioni superiore rispetto al gruppo HME (p<0,01).
Conclusioni. I sistemi di umidificazione passiva
hanno fornito bassi livelli di umidità e temperatura
e non sono stati in grado di mantenere fluide le
secrezioni. I sistemi attivi sono sembrati soddisfare
gli standard raccomandati e hanno permesso di
mantenere fluide le secrezioni.
Parole chiave: Respirazione artificiale - Terapia
Intensiva - Ventilatori meccanici - Umidificatori.
References
1. Mc Fadden ER, Pichurko BM, Bowman FH, Ingenito
E, Burns S, Dowling N et al. Thermal mapping of the
airways in humans. J Appl Physiol 1985;58:564.
2. Ogino M, Kopotic R, Mannino FL. Moisture con-
serving efficiency of condenser humidifier. Anaesthe-
sia 1985;40:990-5.
3. Haslam KR, Nielsen CH. Do passive heat and mois-
ture exchangers keep the patient warm? Anesthesio-
logy 1986;64:379-81.
4. Forbes AR. Humidification and mucus flow in the
intubated trachea. Br J Anaesth 1973;45:874-8.
5. Dick W. Aspects of humidification: requirements and
techniques. Int Anesthesiol Clin 1974;4:217-39.
6. Department of Health, Medical Devices Directorate,
London, UK. Heat and moisture exchanger (HME):
DAR Hygrobac/Hygroboy. Evaluation nº 103, Feb-
ruary 1992.
7. Martin C, Papazian L, Perrin G, Saux P, Gouin F. Pres-
ervation of humidity and heat of respiratory gases in
patients with a minute ventilation greater than
10L/min. Crit Care Med 1994;22:1871-5.
8. Misset B, Escudier B, Rivara D, Leclercq B, Nitemberg
G. Heat and moisture exchangers vs heated humidi-
fier during long-term mechanical ventilation. Chest
1991;100:160-3.
9. Sottiaux T. Use of a heat and moisture exchanger
during long-term mechanical ventilation. Chest
1992;102:979.
Marzo 1998
COMPARISON OF THREE DIFFERENT HUMIDIFICATION SYSTEMS DURING PROLONGED MECHANICAL VENTILATION LUCHETTI
10. Tenaillon A, Cholley G, Boiteau R, Perrin Gachadoat
D, Burdin M. Heat and moisture exchanging bacterial
filter versus heated humidifier in long-term mechani-
cal ventilation. Care Crit Ill 1991;7:56-61.
11. Weeks DB. A laboratory evaluation of recently avail-
able heat and moisture exchangers. Anesthesiol Rev
A
1986;13:33-6.
12. Lowe GD. Efficiency of heat and moisture exchan-
gers: a reply. Anaesthesia 1987;42:1118.
13. Epstein RA. Humidification during positive pressure
IC
ventilation of infants. Anesthesiology 1971;35:532-6.
14. Cigada M, Elena A, Solca M, Damia G. The efficiency
of 12 heat and moisture exchangers: an in vitro eval-
uation. Intens Care World 1990;7:98-101.
15. British Standards Institution. Specifications for
humidifiers for use with breathing machines. BS
T® D
4494. London, 1970.
16. Williams R, Rankin N, Smith T, Galler D, Seakins P.
Relationship between the humidity and temperature
H E
of inspired gas and the function of the airway
mucosa. Crit Care Med 1996;24:1920-9.
17. Shelly M, Bethune DW, Latimer RD. A comparison of
five HME’s. Anaesthesia 1986;41:527-32.
IG M
R A
Y V
P R
O E
C IN
M
Vol. 64, N. 3 MINERVA ANESTESIOLOGICA
18. Eckerbom B, Lindholm CE. Performance evaluation
of six heat and moisture exchangers according to the
draft international standard (ISO/DIS 9360). Acta
Anaesthesiol Scand 1990;31:404-9.
19. Martin C, Papazian L, Perrin G, Bantz P, Gouin F. Per-
formance evaluation of three vaporizing humidifiers
and two heat and moisture exchangers in patients
with minute ventilation >10 L/min. Chest 1992;
102:1347-50.
20. Hagler DA, Traver GA. Endotracheal saline and suc-
tion catheters: sources of lower airway contamina-
tion. Am J Crit Care 1994;3:444-7.
21. Cohen IL, Weinberg PF, Fein IA, Rowinsky GS. Endo-
tracheal tube occlusion associated with the use of
heat and moisture exchangers in the Intensive Care
Unit. Crit Care Med 1988;16:277-9.
22. Roustan JP, Kienlen J, Aubas S, DuCailar J. Compar-
ison of hydrophobic heat and moisture exchangers
with heated humidifier during prolonged mechanical
ventilation. Intens Care Med 1992;18:97-100.
23. Martin C, Perrin G, Gevaudan MJ, Saux P, Gouin F.
Heat and moisture exchangers and vaporizing humidi-
fiers in the Intensive Care Unit. Chest 1990;97:144-9.
81