Respiration Physiology 108 (1997) 241 246

Effects on pulmonary function of daily exposure to dry or

humidified hyperbaric oxygen
A. Shupak *, A. Abramovich, Y. Adir, I. Goldenberg, Y. Ramon, P. Halpern,
A. Ariel
Israel Na6al Medical Institute, IDF Medical Corps, P.O. Box 8040, Haifa 31080, Israel
Accepted 10 March 1997

Abstract
The purpose of this study was to examine the effects of breathing dry or humidified hyperbaric oxygen on pulmonary
function. Pulmonary function tests were performed before and after each of 10 hyperbaric oxygen exposures at 2.5
atmospheres absolute (ATA) for 95 min in a group of 13 patients treated daily by hyperbaric oxygen for problem wounds.
Patients breathed dry oxygen during five successive sessions and humidified oxygen during the remaining five. No
differences were found between forced vital capacities (FVC) and maximal expiratory flows before and after hyperbaric
oxygen exposure while breathing dry or humidified oxygen. Significant differences were found for the changes in the
percentage of FVC expired in 1 s (FEV 1%) and mean forced mid-expiratory flow rate during the middle half of the FVC
(FEF25 75%) on day 1 alone: decrements of 1.42 and 2.96%, respectively, under dry oxygen, vs. increments of 3.93 and
34.4%, respectively, for humidified oxygen. Day-to-day decrements in the percent changes in FEV 1% and FEF25 75%
were observed while breathing humidified hyperbaric oxygen. These results demonstrate that repeated daily exposure to
humidified hyperbaric oxygen abolishes the initial beneficial effect of humidification on peripheral airways flow
characteristics. 1997 Elsevier Science B.V.

Keywords: Function test; Dry vs. humidified hyperbaric oxygen; Humidification; Hyperbaric oxygen; Lung

1. Introduction
Repeated daily exposure to hyperbaric oxygen
(HBO) at 2.5 atmospheres absolute (ATA) for
* Corresponding author. Tel.: + 972 4 8693040; fax: + 972
4 8693240.

0034-5687/97/$17.00 1997 Elsevier Science B.V. All rights reserved.

PII S 0 0 4 - 5 6 8 7 ( 9 7 ) 0 0 0 2 2 - 4

90 120 min is widely employed in the combined

treatment of problem wounds (Thom, 1992). The
oxygen delivered to the patients is kept dry to
prevent corrosion and icing in the gas storage and
supply system. To reduce the risk of bronchoconstriction due to the airway hyper-reactivity, air-way
epithelial damage and inflammation that may

242

A. Shupak et al. / Respiration Physiology 108 (1997) 241 246

be produced by inhaling dry

respiratory gases, humidifiers can
be introduced on the ambient
pressure side of the oxygen supply
system (OCain et al., 1980;
Barbet et al., 1988; Thorsen et al.,
1992; Murchie et al., 1993).
Although overt pulmonary oxygen
toxicity is unlikely on this protocol
and the cumulative effects of HBO
are considered negligible for daily
exposure (Clark, 1993), no study
has been made of day-to-day
variability in pulmonary function
tests during the course of HBO
therapy. Contradic-tory pulmonary
effects of breathing humidified gas
have
also
been
reported.
Breathing humi-dified air during a
single air-dive prevented the flow
limitation in the small airways
associated with dry air, and was
considered by the divers to be
more comfortable (Thorsen et al.,
1992), while breathing humidified
oxygen during a chamber dive
significantly increases both the wet
and dry weights of rodent lungs,
indicating
exac-erbation
of
pulmonary oxygen toxicity (Lin and
Jamieson, 1993). The purpose of
this study was to examine the daily
and day-to-day effects of dry and
humid
HBO
breathing
on
pulmonary function tests in a
group of patients receiving routine
HBO therapy.

2. Patients and methods

2.1. Patients
Thirteen patients, 42 63 years
of age, receiv-ing daily HBO
treatment for problem wounds,
participated in the study after
giving their writ-ten informed
consent. All were non-smokers,
had a negative history of chronic
pulmonary
dis-ease,
no
pathological findings on physical
exami-nation of the respiratory

system, and normal anterior

posterior
and
lateral
chest
roentgeno-grams.
2.2. HBO administration
Subjects
were
randomly
assigned to the dry and humidified
oxygen treatment groups in a
cross-over design, breathing dry
oxygen during

five successive sessions and

humidified oxygen during the
remaining five sessions. The study
protocol was approved by the local
Helsinki committee. HBO was
administered once daily in a
multilock chamber compressed by
air to an absolute pressure of 2.5
ATA for 95 min. Each treatment
session was composed of two 40
min intervals of oxygen breathing,
separated by 5 min of air
breathing. The temperature inside
the chamber was maintained at
approximately 23C. Oxygen was
inhaled via a full face mask fitted
with a demand valve and an
expiratory over-board dumping
system. To establish humid conditions, the inspired oxygen was
passed
through
a
bubble
humidifier. Relative humidity was
mea-sured
with
a
humidity
thermistor (Tecnologic, Vigevano,
Italy) at the output of the
humidifier, and was maintained at
70 80%.
2.3. Pulmonary function tests
Pulmonary function tests were
performed at ambient pressure
breathing room air, immedi-ately

before and 5 20 min after each

HBO ex-posure on the first 10
treatment days. On the day
preceding the first HBO session,
each sub-ject was repeatedly trained
in the performance of pulmonary
function tests until reproducibility of
the results was achieved, according
to the Amer-ican Thoracic Society
recommendations
(Ameri-can
Thoracic Society, 1995). Forced vital
capacities (FVC) and maximal
expiratory flows were measured
immediately after calibration of the
spirometer
(ST-250,
Fukuda
Sangyo, Japan), and validated
according to the American Tho-racic
Society
recommendations
(American Tho-racic Society, 1995).
The respiratory parameters recorded
included forced vital capacity (FVC),
volume expired in 1 s (FEV 1),
percentage of FVC expired in 1 s
(FEV1%), peak expiratory flow rate
(PEF), and forced mid-expiratory
flow rate (FEF25 75%). A minimum of
three attempts was made for each
spirometry session. Only the results
of the best test, defined according to
the highest value obtained by the
summation of FEV1 and FVC, were
taken into consideration. All values
were corrected to BTPS conditions.

A. Shupak et al. / Respiration Physiology 108 (1997) 241 246

2.4. Symptoms of oxygen toxicity

After each HBO exposure, the
study
partici-pants
were
specifically asked about symptoms
that have previously been related
to pulmonary oxygen toxicity.
These
symptoms
include
retrosternal
irri-tation
on
inspiration, chest tightness or pain,
cough, and dyspnoea (Clark and
Lambertsen, 1971).
2.5. Statistical analysis
Pulmonary
function
values
obtained before and after HBO
exposure were compared separately
for the dry and humidified oxygen
groups using the paired t -test.
Differences in daily measurements
between the groups were evaluated
by the two-sample t -test. The
significance of day-to-day vari-ations
in the test results was calculated by
repeated
measures
one-way
analysis of variance (ANOVA), and
the source of significant differ-ences
between the days by the Tukey test
for the pairwise comparison of
means. Statistical analysis was
carried out using SAS software
(SAS Insti-tute, Cary, NC) on an
IBM-compatible personal computer.

3. Results
3.1. Symptoms of oxygen toxicity
No
symptoms
related
to
pulmonary oxygen toxicity were
reported by the patients during the
study.
3.2. Changes in pulmonary
function tests
No differences were found on any
of the days between the average
forced vital capacities (FVC) and
maximal expiratory flows measured
before or after HBO exposure while

243

breathing dry oxygen. The only

significant change observed after
humi-dified HBO breathing was an
increase in the
average absolute value for FEF 25
75% on day 1: 3.43 l/min before and
4.59 l/min after the HBO

session (p = 0.03, paired t -test).

When the percent changes in

pulmonary func-tion tests after HBO
exposure were compared, significant
differences were noted in the
changes
in FEV1% and FEF25 75% on day 1
alone; a decre-ment of 1.42 9 1.63
(SE) and 2.96 9 6.31 in

FEV1%
and
FEF25

75%,
respectively, under dry oxygen vs.
an increment of 3.93 9 2.29 (SE)
and
34.4 9 12.5 in FEV1% and FEF25
respectively, for humidified
75%,
oxygen (p = 0.03, p = 0.006, respectively, two-sample t -test) (Table 1).
In the dry HBO group, no
significant daily variations were
found in the percent change in
pulmonary function tests following
HBO
expo-sure.
However,
significant decrements in the per-

cent changes in FEV1% and FEF25

were observed while

75%
breathing humidified HBO (p =
0.03 and p = 0.005, respectively,
one-way
repeated
measures
ANOVA) (Figs. 1 and 2, Table 1).
These differences between the
days were at-tributed to the
variance between day 1 and day 4
(Tukey test).
4. Discussion

Significant differences were found

between the
average changes in FEV1% and
FEF25 75% under dry HBO relative to
baseline measurements, and
the average changes in the same
parameters
while
breathing
humidified HBO. This observation
was documented only for the first
session. While no significant day-today variability could be found in any
of the pulmonary function tests
measured for the dry oxygen group,
repeated HBO hu-midification was
associated with a decrease in the
relative post-exposure increments in
FEV1% and

FEF25 75% observed for the first

humidified HBO exposure.
Previous animal models and
human studies in-vestigating the
pulmonary effects of breathing
humidified normobaric oxygen or
hyperbaric air during a single
exposure support our findings
regarding
the
short-term
advantage
of
oxygen
humidification. Rats exposed to 0.94
ATA dry nor-mobaric oxygen for 48
h showed significant bronchial
epithelial thickening, was prevented by oxygen humidification
(Murchie
et
al.,
1993).
Anaesthetized dogs ventilated at 1
ATA for

244

A. Shupak et al. / Respiration Physiology 108 (1997) 241 246

Table 1
Pulmonary function test changes (%) after HBO exposure: humidified vs. dry oxygen breathing (mean 9 SE)
FCV
Day 1
Humidified
Dry
Day 2
Humidified
Dry
Day 3
Humidified
Dry
Day 4
Humidified
Dry
Day 5
Humidified
Dry

FEV1

FEV1%a

PEF

FEF25b 75%

2.02 9 3.10
4.58 9 6.04

1.10 9 2.29
2.61 9 4.93

3.93 9 2.29
c
1.42 9 1.63

8.39 9 8.25
15.44 9 16.2

34.429 12.6
d
2.969 6.31

0.79 9 2.08
1.18 9 1.90

2.55 9 3.56
0.91 9 1.66

1.48 9 1.66
0.92 9 0.92

1.09 9 8.07
0.3 9 4.05

1.799 9.36
5.269 8.97

2.52 9 2.35
0.38 9 2.41

2.10 9 2.62
1.38 9 2.44

0.41 9 0.98
1.18 9 1.78

0.12 9 5.39
0.03 9 4.54

2.559 4.00
10.829 8.21

5.86 9 5.39
1.23 9 2.70

1.12 9 0.72
1.06 9 2.58

3.24 9 1.45
0.62 9 1.42

0.66 9 8.32
4.28 9 3.02

8.589 6.42
4.569 6.93

1.93 9 1.82
1.20 9 3.32

1.67 9 2.18
0.27 9 2.14

0.53 9 1.31
0.99 9 1.69

1.37 9 11.0
6.11 9 4.37

2.879 2.51
7.179 6.86

P = 0.03 (repeated measures ANOVA) for day-to-day variations when breathing humidified oxygen.
P = 0.005 (repeated measures ANOVA) for day-to-day variations when breathing humidified oxygen.
c
P = 0.03 (two-sample t -test).
d
P = 0.006 (two-sample t -test).
b

12 h showed less of a decrease in surfactant

activity when breathing humidified oxygen compared with animals ventilated with dry oxygen
(Motlagh et al., 1969). However, in another
study using similar animals breathing oxygen at
1 ATA for 7 h, potentiation of pulmonary toxicity,
as evaluated by pulmonary vein PO2 values, was
re-ported to result from oxygen humidification
(Ching et al., 1973).

Fig. 1. Day-to-day variations in FEV1% after dry and humidified HBO exposure. Data are presented as mean 9 SE.

Two previous human studies compared the

changes in pulmonary function tests measured
before and after dry or humidified compressed airdives (Thorsen et al., 1992; Ronnestad et al.,
1994). Dry air and oxy-helium dives to pressures of
1.16 5.94 ATA for 30 min to 4 h were associ-ated
with significant reductions in FEV1 and FEF25 75%.
Such changes were not found for dives to the
same depths and bottom times on humi-

Fig. 2. Day-to-day variations in FEF25 75% after dry and

humidified HBO exposure. Data are presented as mean 9 SE.

A. Shupak et al. / Respiration Physiology 108 (1997) 241 246

dified air or oxy-helium. For the shallower humidified air dives, relative increments were reported
in FEV1 and FEF25 75%, matching our observa-tion
regarding the relative changes in FEV1% and
FEF25 75% after the first humidified HBO session.
Although the subjects of the above mentioned
studies were, like our patients, all non-smokers,
with no history, physical or baseline pulmonary
function findings indicating bronchoconstriction, in
their case as well, humidification of the breath-ing
gas improved airway flow characteristics rela-tive
to baseline measurements.
The mechanism of improved pulmonary airflow
characteristics following a single exposure to a
humidified hyperoxic gas mixture has not yet been
investigated or elucidated. Breathing dry air results in bronchoconstriction, probably due to osmolarity changes affecting the fluid lining of the
airways (Anderson, 1992). It is believed that these
changes cause stimulation of autonomic nerve
endings or the release of an as yet unidentified
mediator, inducing a bronchomotor response, because the latter may be prevented by the administration of cholinergic antagonists (Anderson et al.,
1979; Wilson et al., 1984). One may speculate that
contrasting osmolarity changes facilitated by
breathing highly humidified oxygen produced the
bronchodilatation observed in the present study.
One mediator which might be involved in the
suggested response is nitric oxide, which has a
well documented smooth muscle relaxation effect
(Gaston et al., 1994). Repeated daily HBO exposures may result in the accumulation of oxygenderived free radicals, which overwhelm existing
scavenger systems. The reaction of these free
radi-cals with nitric oxide to produce various
nitrogen oxides would lead to the disappearance of
the temporary bronchodilatory effect observed
after the first HBO session.
The results of a recently published rodent model
suggest that oxygen humidification might affect the
development of pulmonary oxygen toxi-city to
different degrees according to the pressure applied
and the duration of the exposure (Lin and
Jamieson, 1993). Mice which breathed humidified
oxygen for 30 min at pressures of 5.1 5.8 ATA
had significant increases in wet and dry lung
weights compared with animals breathing dry

245

oxygen. This observation was interpreted as a sign

of enhanced pulmonary oxygen toxicity under
humid conditions. In contrast to this, rats exposed
to normobaric hyperoxia for 68 72 h had a decreased mortality rate, lower lung weights, and
less pleural effusion under humid conditions when
compared with a group of similar animals breathing dry oxygen (Lin and Jamieson, 1993).
In spite of the fact that in the present study no
symptoms related to pulmonary oxygen toxicity
were reported by the patients in either group,
gradual but significant decrements in FEV 1% and
FEF25 75% normalised values were found after
repeated humidified HBO exposures. It is known
that despite the absence of clinical symptomatology, the biochemical effects of oxygen toxicity are
initiated concurrently with the elevation of PO2
(Lambertsen, 1978). Deterioration in lung mechanical function precedes the clinical symptoms
of pulmonary oxygen toxicity, and abnormal pulmonary function has been reported to persist for as
long as 11 days after the disappearance of such
symptoms (Clark and Lambertsen, 1971). The
initial exposure interval represents an asymptomatic period of slowly developing toxicity, from
which recovery is considered to be rapid and
complete on return to normoxia (Clark, 1993).
However, our results imply that when a course of
daily 95 min sessions of 2.5 ATA humidified HBO is
in question, cumulative effects of pulmonary
oxygen toxicity might persist. These may abolish
the initial benefits of oxygen humidification on
pulmonary flow characteristics observed after the
first HBO session. We were unable to find any
previous study comparing the pulmonary effects of
dry vs. humidified oxygen in patients receiving
daily HBO treatments on one of the protocols
recommended by the Undersea and Hyperbaric
Medical Society for problem wounds (Thom, 1992).
The only study examining changes in pul-monary
function tests after humidified oxygen breathing at
2.5 ATA employed a single continu-ous oxygen
exposure lasting 5.7 h. The greatest changes
found were an average drop of 30% in

FEF25 75% and 20% in maximal expiratory flow

rate at 50% of FVC (MEF50%), indicating increased resistance of the peripheral airways
(Clark, 1988, 1993). Although the daily oxygen

246

A. Shupak et al. / Respiration Physiology 108 (1997) 241 246

breathing period employed in our

study was much shorter (two
intervals of 40 min separated by a 5
min air break), these observations
are comparable to the gradual
decline observed in FEV1% and
FEF25 75% in the course of our
study when the patients were
breathing humidified oxygen, reaching maximal decrements of 3.2 and
8.6%, respec-tively. FEV1% contains
both the effort-dependent and effortindependent parts of the flow
volume loop, whereas FEF25 75% is
effort-independent, and a decrease
in its value relative to the baseline
measurement
represents
flow
limitation in the pe-ripheral airways
(McFadden and Linden, 1972).

Our results show that although

patients might benefit from oxygen
humidification during a sin-gle
HBO session with regard to airway
flow
char-acteristics,
this
advantage might disappear during
successive HBO treatments. We
suggest that the practice of
breathing humidified gas during
rou-tine daily HBO therapy
protocols over a pro-longed period
should be reconsidered.

Acknowledgements
The authors are indebted to
Esther Eilender and Richard
Lincoln for their assistance in the
preparation of the manuscript.

References
American
Thoracic
Society,
1995.
Standardization of Spirome-try, 1994
Update. Am. J. Resp. Crit. Care Med.
152, 1107 1136.
Anderson, S.D., Seale, J.P., Ferris, L.,
Schoeffel, R., Lindsay, D.A., 1979. An
evaluation of pharmacotherapy for
exer-cise-induced asthma. J. Allergy
Clin. Immunol. 64, 612 624.
Anderson, S.D., 1992. Asthma provoked by
exercise, hyper-ventilation, and the
inhalation of non-isotonic aerosols. In:
Barnes, P.J., Rodger, I.W., Thomson,
N.C.
(Eds.),
Asthma:
Basic

Mechanisms
and
Clinical
Management, 2nd ed. Academic
Press, London, pp. 473 490.
Barbet, J.P., Chauveau, M., Labbe, S.,
Lockhart, A., 1988. Breathing dry air
causes acute epithelial damage and
infl-

ammation of the guinea pig trachea. J.

Appl. Physiol. 64, 1851 1857.
Ching, N., Kazigo, J.M., Hicks, R.G., Nealon,
T.F. Jr., 1973. Potentiation of oxygen
toxicity by excessive levels of humidification. Surg. Forum 24, 222 223.
Clark, J.M., Lambertsen, C.J., 1971.
Pulmonary oxygen toxic-ity: a review.
Pharmacol. Rev. 23, 37 133.
Clark, J.M., 1988. Extension of oxygen tolerance.
Proceedings of a symposium in honor of
Christian J. Lambertsen, M.D.. Exp. Lung
Res. 14 (Suppl.), 863 1058.

Clark, J.M., 1993. Oxygen toxicity. In: Bennett,

P.B., Elliott, D.H. (Eds.), The Physiology
and Medicine of Diving, 4th ed. W.B.
Saunders, London, pp. 121 169.
Gaston, B., Drazen, J.M., Loscalzo, J.,
Stamler, J.S., 1994. The biology of nitrogen
oxides in the airways. Am. J. Resp. Crit.
Care Med. 149, 538 551.
Lambertsen, C.J., 1978. Effects of hyperoxia
on organs and their tissues. In: Robin, E.D.
(Ed.), Extrapulmonary Man-ifestations of
Respiratory Disease (Lung Biology in
Health and Disease, vol. 8), Marcel
Dekker, New York, pp. 239 303.
Lin, Y., Jamieson, D., 1993. Effect of humidity
on hyperoxic toxicity. J. Appl. Physiol. 75,
1980 1983.
McFadden, E.R. Jr., Linden, D.A., 1972. A
reduction in maximum mid-expiratory flow
rate. A spirographic mani-festation of small
airway disease. Am. J. Med. 52, 725 737.
Motlagh, F.A., Kaufman, S.Z., Giusti, R.,

Cramer, M., Gar-zon, A.A., Karlson,

K.E., 1969. Electron microscopic appearance
and
surface
tensions
properties of the lungs ventilated with
dry or humid air or oxygen. Surg. Forum
20, 219 220.
Murchie, P., Johnston, P.W., Ross, J.A.S.,
Godden, D.J., 1993. Effects of hyperoxia
on bronchial wall dimensions and lung
mechanics in rats. Acta Physiol. Scand.
148, 363 370.
OCain, C.F., Dowling, N.B., Slutsky, A.S.,
Hensley, M.J., Strohl, K.P., McFadden,
E.R. Jr., Ingram, R.H. Jr., 1980. Airway
effects of respiratory heat loss in normal
subjects. J. Appl. Physiol. 49, 875 880.
Ronnestad, I., Thorsen, E., Segadal, K.,
Hope, A., 1994. Bronchial response to
breathing dry gas at 3.7 MPa ambi-ent
pressure. Eur. J. Appl. Physiol. 69, 32
35.
Thom, S.R., 1992. Hyperbaric Oxygen
Therapy: a Committee Report. Undersea
and Hyperbaric Medical Society, UHMS
publication no. 30 CR, HBO, Bethesda,
MD.

Thorsen, E., Ronnestad, I., Segadal, K.,

Hope, A., 1992. Respiratory effects of
warm and dry air at increased ambi-ent
pressure. Undersea Biomed. Res. 19,
73 83.
Wilson, N., Dixon, C., Silverman, M., 1984.
Bronchial
respon-siveness
to
hyperventilation in children asthma: inhibition by ipratropium bromide. Thorax
39, 588 593.