Molecular detection of pulmonary oxygen toxicity using exhaled breath analysis after hyperbaric hyperoxic exposure.

Exposure to hyperoxia is common in military oxygen diving and in hyperbaric
oxygen therapy (HBOT). Breathing oxygen at a partial pressure (PO2) of more
than 50 kPa for a longer duration can lead to pulmonary oxygen toxicity (POT).
(Klein 1990, Miller 1981) The most mentioned changes which can be found are
atelectasis, interstitial oedema and inflammation. (Sackner 1975) These changes
are reversible. (Winter 1972) However, when the administration of oxygen is
continued, this will eventually lead to irreversible lung fibrosis. (van Ooij
2013, Kapanci 1972)

The current standard for determining POT in diving and hyperbaric medicine, is
a decrease in vital capacity (VC). (Clark 1970) Bardin & Lambertsen related the
decrease in VC to the PO2 and time exposed to oxygen and introduced the unit of
pulmonary toxicity dose (UPTD). (Bardin 1970) To cope with the wide range of
inter- and intrapersonal variability, the limits of acceptable oxygen exposure
are based on median decreases in VC. For instance; 450 UPTD gives a 2% decrease
in VC in 50% of the cases. The decrease in VC was derived from dry-dives (in a
recompression chamber), not from actual hyperbaric oxygen in an immersed
setting. At the time of publication, the authors recognized the limitations of
the model and suggested that more advanced research techniques would probably
increase the validity of the UPTD model.

Recent publications indicate that more advanced parameters such as diffusion
capacity of carbon monoxide (DLCO) and nitric oxide (DLNO), could more
accurately determine POT. (van Ooij 2014) However, these measurements are quite
difficult to perform and require specialised equipment. Therefore, these
methods cannot be used by clinicians or divers as a measurement of POT in an
outward setting. In combination with the recent findings that immersion affects
the rate at which POT develops and the high intra- and interpersonal variance,
the diving industry and the field of hyperbaric medicine needs a new and valid
model which allows correction for individual susceptibility.

In an earlier study we found volatile organic compounds (VOCs) detected in a
single exhaled breath (EB) four hours after a hyperbaric exposure (with Gas
Chromatography Mass Spectrometry [GCMS] analysis). (van Ooij 2014) The
conclusion if this study was that more accurate EB measurement should be
performed less than four hours post-dive, however the exact moment is unknown.
Also, the GCMS-analysis requires an external laboratory. Therefore, the
traditional method analysing EB does not meet the requirements of point-of-care
testing.

With the recent development of the SpiroNose® by the department of respiratory
medicine in the Academic Medical Center a highly advanced technique became
available to overcome these difficulties. The SpiroNose allows analysis of EB
and compare it to an online database. However, no research has been conducted
to validate VOCs detected by the SpiroNose are just as valid as GCMS to detect
POT after (immersed) hyperbaric oxygen exposure.

Our hypothesis is that VOCs detected with the SpiroNose in a single exhaled
breath are just as valid as DLNO/CO and are a patient-friendly and easy to use
method to detect POT after hyperbaric oxygen exposure.

Primary Objective:
To study the differences between EB post oxygen-dive, post air-dive (control),
dry-dive (recompression chamber) and pre-dive (baseline).
1. Will an exposure to a PO2 of 190 kPa (100% oxygen at 9 msw) during 60
minutes in an immersed setting lead to changes in EB compared a PO2 of 40 kPa
(21% oxygen at 9 msw)?
2. Will a repetitive exposure to a PO2 of 250 kPa (100% oxygen at 15 msw,
similar to HBOT) during 90 minutes in a dry setting lead to changes in EB?

Secondary Objective(s):
1. Is the SpiroNose just as sensitive in detecting VOCs in EB associated with
POT as GCMS?
2. Which time-interval post-dive give the most reliable result to detect POT?

Experiment I * *Wet dives*
This is a randomized cross-over trial which has three measuring days per
subject. All measurements and experiments will be performed at the Royal
Netherlands Navy Diving Medical Center at the Naval harbour in Den Helder, the
Netherlands.

Study day one (Monday): baseline measurements in which we measure DLNO/CO and
EB without being exposed to either oxygen or pressure in the preceding
twenty-four hours. These baseline measurements will take place at least 48
hours before study day two.
Study day two (Thursday): the subject will make a dive to nine meters during
sixty minutes in which he either breathes 100% oxygen (active exposure) or air
(control, PO2 of 0,40 kPa) in random order. Once pre-dive and five times
post-dive EB will be measured (*, 1, 2, 3 and 4 hours after exposure). DLNO/CO
will be measured once at four hours post-dive. A sample of the inspired and
ambient air will be taken for reference purposes.
Study day three (Thursday +1 week): this day is analogous to day two, but the
subjects will breathe oxygen if they breathed air on the previous dive and vice
versa. Measurements will be the same. This day will be at least one week after
the previous dive to ensure all physiological parameters have been normalised.


Experiment II * *Dry dives*
This is a prospective cohort study which has seven measuring days per subject.

Study day one (Friday): baseline measurements in which we measure DLNO/CO and
EB without being exposed to either oxygen or pressure in the preceding
twenty-four hours. These baseline measurements will take place at least 48
hours before day two.
Study day two (Monday): Measurement of EB pre-dive. Afterwards the subject will
make a dive to fifteen meters for ninety minutes in a hyperbaric chamber. In
the four hours post-dive we will measure DLNO/CO once and EB three times (*, 2
and 4 hours after exposure).
Study day three (Tuesday): As day two.
Study day four (Wednesday): As day two.
Study day five (Thursday): As day two.
Study day six (Friday): As day two.
Study day seven (Monday): After two days of non-diving (comparable to regular
hyperbaric oxygen treatment schedules). Exposure and measurements similar to
day two.

Benefits: For military oxygen diving, as well as hyperbaric oxygen treatment,
it is of vital importance to know at which level and duration (hyperbaric)
oxygen can be breathed before it will lead to POT. In addition, with this
research, we hope to be able to develop a *bed-side* point of care measurement
to help divers and clinicians determine safe limits for hyperbaric oxygen
therapy.

Risk assessment:
Decompression sickness: Any well performed in-water dive has some risk of DCS.
The risk of DCS associated with the wet dive (60 minutes at 9 msw with
compressed air) can be considered as very small (0.1%). The dry dive has no
risk for DCS, because the subjects only breath oxygen.
Oxygen toxicity: There is a risk for both cerebral as well as pulmonary
toxicity. The latter is subject of this study. The risk or cerebral oxygen
toxicity during wet dives is about 4.7% as determined by Arieli and Butler.16,
17 This estimation is based on oxygen divers who are subjected to physical
exertion. As our subjects will not perform any exercise we expect the risk of
cerebral oxygen toxicity to be less than 4.7%. In 85 similar wet exposures in
previous experiments we had no cases of cerebral oxygen toxicity. In dry dives
(recompression chamber) this risk is even lower. Cases of cerebral oxygen
toxicity are very rare, with incidences of 1 in 40.000 exposures not being
uncommon.18,19
If nevertheless a convulsion occurs the proper action will be taken according
to *Noodplan Neurologische Zuurstof Vergiftiging* (see section K6). Overall, we
think the burden of these wet and dry dives can be considered low, and less
than an oxygen dive in *open water* which is part of their regular work.
Barotrauma: Hyperbaric exposure requires equalisation of the middle-ear during
ascends and descents to avoid tympanic membrane ruptures. All subjects are
medically evaluated according to standards of the European Diving Technology
Committee (EDTC), which includes evaluation of the ability to equalize.20
Additionally, all subjects trained and certified in diving or recompression
procedures. Finally, ascents and descents in both experiments are artificially,
with the possibility to temporarily halt the change of depth to give the
subject more time to equalize. We feel this risk is negligible.
Fire hazard: Although fire in recompression chambers are extremely rare,
hyperoxic conditions increase the risk of fire when combined with high
temperature and fuel. To avoid any contamination (which could act as fuel) all
material exposed to 100% oxygen are *oxy clean* and regularly maintained
according to both international and Dutch standards. Oxygen is administered to
subjects via a breathing apparatus (wet experiments) or breathing mask (dry
experiments). To ensure no oxygen is leaking, the recompression chamber is
fitted with oxygen monitoring sensors. If ambient oxygen levels rise above 23%
the oxygen supply is limited and the chamber starts *flushing* oxygen to the
exterior. We are confident this risk is properly addressed.

Group relatedness:
Oxygen diving is only possible with adequate training and material. Navy divers
are the only group in the Netherlands who are trained and certified to work
with this equipment. As oxygen diving is part of their daily work, the divers
will benefit of all research being conducted to increase the safety of oxygen
diving.
For the dry dives, recompression personnel is medically evaluated and trained
to recognize any dangers associated with hyperbaric exposure. This training
consists of four weeks of education and hands-on experience. Because the
Ministry of Defence considers participation of this study as *regular work*,
all safety and certification standards apply. We consider it an
disproportionate burden to use untrained volunteers who require 4 weeks of
training before being able to participate in this study.