Content uploaded by Emmanuel Gempp
Author content
All content in this area was uploaded by Emmanuel Gempp on Oct 16, 2017
Content may be subject to copyright.
MILITARY MEDICINE, 176, 4:446, 2011
446 MILITARY MEDICINE, Vol. 176, April 2011
INTRODUCTION
Rebreather diving has become popular in the past decade and
offers many advantages in comparison to open-circuit scuba
diving, such as economy of gas consumption and reduction of
decompression procedure by using oxygen-enriched breath-
ing gas mixtures. However, this equipment requires special-
ized training and techniques more complex than scuba air
diving, leading to more malfunctions and procedural errors
with subsequent potential risk of drowning.
For more than 50 years, French military divers routinely
use rebreather diving capabilities for their underwater opera-
tional activities and very often at the limits of their physiologi-
cal conditions with respect to gas toxicity, dive duration, or
the effort required. The rebreather systems used in the French
army are currently all mechanically controlled devices made
with nonmagnetic materials, which are selected according to
the depth to be reached and the type of underwater military
action. They include both closed-circuit rebreathers (CCR)
and semi-closed circuit rebreathers (SCR).
CCR produce no bubbles and use pure oxygen appropriate
for use at depths shallower than 7 msw due to concerns for
CNS oxygen toxicity. They are worn in a prone position with
a gas supply on demand mode: fresh oxygen coming from a
small high-pressure cylinder is injected in the breathing bag
through a fl ow valve to refi ll it when the volume of gas into
the loop falls due to oxygen consumption. Exhaled breathing
gases are purifi ed by a scrubber canister that contains soda
lime (Divesorb, Dräger, Luebeck, Germany), which fi xes the
CO
2 before the gas is returned to the breathing bag. Replacing
the OXYGERS 57 (La spirotechnique, Carros, France) with
the Full Range Oxygen Gas System (FROGS; Aqualung,
Carros, France) in 2002 has increased the endurance of this
equipment (up to 4 hours of use compared with 3 hours
previously) and optimized breathing comfort.
SCR incorporate a counterlung designed as a bellows sys-
tem with 2 concentric bags allowing the breathing gas to be
periodically vented into the water through a relief valve in
proportion to the volumetric ratio of the 2 bags (the working
principle is discussed in greater detail elsewhere
1 ). Additional
breathing mixture is taken in from the cylinders by the con-
traction of the breathing bag via a fl ow injector that balances
the gas leak. After several ventilatory cycles, the partial pres-
sure of oxygen in the breathing bag is constant, but with a con-
centration in the gas mixture below that of the cylinders due
to dilution of the breathing gas in the counterlung by exhaled
gas. These devices operate with predefi ned gas mixtures that
determine the maximum depth at which they can be used.
The DC 55 (La spirotechnique) allows the diver to go down
to 55 msw with oxygen-enriched mixtures (nitrox), and the
MIXGERS (La spirotechnique) uses a trimix mixture contain-
ing 23% O
2 , 37% N
2 , and 40% He for deep dives between 55
and 80 msw. Both breathing apparatus were replaced in 2009
by the Complete Range Autonomous Breathing Equipment
(CRABE; Aqualung), a new version of mechanically con-
trolled device designed for the same specifi c tasks and capa-
ble of using nitrox or trimix gas mixtures to the same depths
range.
A mixed device, the OXYMIXGERS (La spirotechnique),
can be used in closed or semi-closed mode via a manual switch.
The system is supplied with pure oxygen in closed mode to a
depth limit of 7 msw or with nitrox 60% O
2 in semi-closed
Descriptive Epidemiology of 153 Diving Injuries With Rebreathers
Among French Military Divers From 1979 to 2009
LTC Emmanuel Gempp , French Armed Forces Health Service , MC* ;
COL Pierre Louge , French Armed Forces Health Service , MC * ;
COL Jean-Eric Blatteau , French Armed Forces Health Service , MC † ;
BG Michel Hugon , French Armed Forces Health Service , MC *
ABSTRACT Introduction: Rebreathers are routinely used by military divers, which lead to specifi c diving injuries.
At present, there are no published epidemiologic data in this fi eld of study. Methods: Diving disorders with rebreathers
used in the French army were retrospectively analyzed since 1979 using military and medical reports. Results: One hun-
dred and fi fty-three accidents have been reported, with an estimated incidence rate of 1 event per 3,500 to 4,000 dives.
Gas toxicities were the main disorders (68%). Loss of consciousness was present in 54 cases, but only 3 lethal drowning
were recorded. Decompression sicknesses (13%) were exclusively observed using 30 and 40% nitrox mixtures for depth
greater than 35 msw. Eleven cases of immersion pulmonary edema were also noted. Conclusion: Gas toxicities are fre-
quently encountered by French military divers using rebreathers, but the very low incidence of fatalities over 30 years can
be explained by the strict application of safety diving procedures.
*Department of Hyperbaric and Diving Medicine, Sainte Anne’s Military
Hospital, BP 20545, 83041 Toulon cedex 9, France.
†Institute of Biomedical Research in French Armed Forces, Toulon cedex 9,
France.
The opinions and assertions contained herein are those of the authors
and are not to be construed as the offi cial or as refl ecting the views of the
Department of Defense.
Downloaded from publications.amsus.org: AMSUS - Association of Military Surgeons of the U.S. IP: 088.168.200.104 on Oct 16, 2017.
Copyright (c) Association of Military Surgeons of the U.S. All rights reserved.
Rebreather Incidents in French Military Divers
MILITARY MEDICINE, Vol. 176, April 2011 447
mode to a depth of 25 msw. It is designed exclusively for
combat swimmers in their delivery submarine vehicle.
Although the number of diving fatalities increases each
year with rebreathers, data regarding the source and specifi cs
of accidents analyzed are scarce and often incomplete,
2,3 and
to date, there are no published reports gathering information
on the potential risks other than death. The intent of this study
was, therefore, to determine the distribution of diving injuries
involving rebreathers reported in French military divers over
30 years and to identify the triggering circumstances associ-
ated with the different adverse events that justify the keeping
of strict safety procedures imposed by military diving regula-
tions since many years.
MATERIALS AND METHODS
The study population consisted of French military divers
certifi ed for the use of rebreathers and divided into 2 sub-
populations that are mine-clearance divers (MCD) and com-
bat swimmers (CS). MCD (total number = 200) are trained for
identifying, neutralizing, or destroying underwater explosive
devices. They are qualifi ed to work at depths of as much as
80 msw in “square” diving profi les. Nevertheless, the majority
of their dives are performed with SCR using a nitrox mixture
of 60% O
2 up to 25 msw for 3 hours or a nitrox mixture of
40% O
2 from 25 to 45 msw. Deep dives using nitrox 32 %
O
2 between 45 and 55 msw and trimix mixture from 60 to
80 msw are relatively rare (4–8 dives per year per diver) and
include extra safety precautions (support diver and decom-
pression stops using surface-supplied oxygen). CS (total
number = 100) conduct clandestine operations for counter-ter-
rorism missions and for ground or amphibious reconnaissance
purposes. They perform most of their dives at shallow depths
in so-called horizontal diving profi les with CCR, but are also
qualifi ed to dive with SCR using nitrox mixtures and SCUBA
air up to 60 msw.
Since 1979, each diving disorder was subject to a military
and medical statutory declaration to identify potential failure
equipment after careful investigation and to improve diving
procedures. The information collected has given rise to a data-
base used to produce a detailed retrospective analysis of reg-
istered cases. Each datasheet was subsequently reviewed by 1
of the authors (P.L.) to complete and update the initial decla-
ration forms. For the purposes of this study, diving injury with
rebreather was defi ned as a clinical adverse event that led to
a hazardous behavior during diving or an emergency ascent
assisted by the buddy if necessary.
The incidence rates (IR) of diving rebreather injuries were
calculated using the number of injuries sustained by each
population of divers as the numerator and the total number of
dives performed in each subgroup over the study period as the
denominator. Incidence rate ratios (IRR) were determined by
dividing the rate of diving injuries in MCD by the rate in CS to
quantify the increased or decreased risk of rebreather disorders
associated with 1 subgroup of divers. Signifi cant difference
between IR was noted where the 95% confi dence interval (95%
CI) of the IRR excluded unity. The 95% CIs for rates and rate
ratios were computed using standard large-sample formulas.
4
RESULTS
Study Population and Diving Injury Incidence
Of the 362 declaration forms reported over 30 years, 153 (42%)
concerned diving injuries involving a rebreather, with an aver-
age of 5 accidents per year (range 2–16). Almost two-thirds
of the injured divers were MCD (69%, n = 106), whereas the
remaining were CS (31%, n = 47). All subjects were men (but
the advent of women as MCDs is extremely recent and only 2
women have been certifi ed since 2006) with a mean age (SD)
of 29 ± 5 years (range 20–47 years). In 76 cases (50%), the
disorders took place during the intensive qualifi cation training
course that every MCD or CS candidate must complete at the
diving school ( Fig. 1 ).
The number of dives using rebreathers per year was esti-
mated at 14,000 dives for MCD population (70 dives per year
for each diver) and 5,000 dives for CS population (50 dives
per year for each diver), leading to an overall incidence rate
of diving rebreather injuries of 2.5 per 10,000 diver-exposures
(95% CI: 2.1, 3.0) for MCD and 3.1 per 10,000 diver-
exposures (95% CI: 2.2, 4.0) for CS. The IRR for MCD vs. CS
was 0.8 (95% CI: 0.6, 1.1).
Distribution Frequencies of Diving Injuries
The distribution of rebreather diving-related injuries is summa-
rized in Figures 2 and 3 . Gas toxicities with a prevalence rate
of 68% ( n = 104) were the main causal factor associated with
rebreather disorders. Hypercapnia during diving was listed in
the majority of cases (62 out of 104, ie, 59%) and resulted in
apparent breathing discomfort (air-hunger and breathlessness),
FIGURE 1. Diagram presenting the process of reviewing the declaration
forms of diving accidents in French military divers from 1979 to 2009.
Downloaded from publications.amsus.org: AMSUS - Association of Military Surgeons of the U.S. IP: 088.168.200.104 on Oct 16, 2017.
Copyright (c) Association of Military Surgeons of the U.S. All rights reserved.
Rebreather Incidents in French Military Divers
448 MILITARY MEDICINE, Vol. 176, April 2011
intense headache, and in some cases altered mental status
and unconsciousness, leading to the dive being stopped and
the injured diver being assisted during ascent by the buddy.
CO
2 retention was determined as the common cause due to
strenuous diving in 36 cases, whereas exogenous gas toxic-
ity linked to inadequate CO
2 elimination (soda lime saturated
at the end of a long dive, scrubber canister not renewed after
each dive, or inoperative after water leaking into the circuit)
was suspected in other cases. Acute hyperoxia was observed in
26 cases (25%) invariably revealed by a loss of consciousness
followed by tonic–clonic generalized seizure, but rarely pre-
ceded by warning symptoms. This occurred either in water or
on the surface, following rescue, on removing the mouthpiece.
These disorders were particularly described before 2002 with the
OXYGERS 57 ( n = 16) after prolonged exposure (2–3 hours) at
oxygen partial pressures, where effects of CNS oxygen toxicity
are dominant (160–170 kPa) in combination with in a sustained
physical exercise of fi n-swimming. Some cases of acute hyper-
oxia were also demonstrated using the DC 55 with Nitrox 32%
( n = 2) or the OXYMIXGERS ( n = 3). In these cases, an equip-
ment malfunction or human error in using the device was noted,
with ventilation of hyperoxic gas mixture well beyond the oxy-
gen exposure tolerance limit (accidental opening of the oxygen
cylinder valve at 50 msw with the DC 55, gas switch from 60%
O
2 to 100% O
2 at 25 msw with the OXYMIXGERS). Finally,
16 divers (15%) sustained an acute hypoxia during rebreather
diving. These accidents were characterized by an insidious loss
of consciousness, sometimes followed by convulsive move-
ments and agitation with recovery of consciousness on the sur-
face. They particularly concerned devices in which the breathing
mixture became hypoxic by dilution on reaching the surface
in the 0 to 15 msw depth ranges (3 cases involving the DC
55 with 32% O
2 and 5 cases involving the MIXGERS system
with trimix mixture). In 6 other cases, an equipment problem
with the breathing loop was identifi ed. In 1 case, a poor rins-
ing of the closed-circuit breathing equipment (OXYGERS 57)
was reported, and in another case the cylinders were supposed
to contain a mixture with 60% O
2 but contained only nitrogen
(cylinders not rinsed on return from retesting).
Overall, these gas toxicities–related disorders were pre-
dominantly observed in student divers (67 cases out 104,
ie, 64%) and resulted in an impairment of consciousness in
54 cases out of 104 (52%), with 11 cases related to severe
hypercapnia. Outcome was always favorable once the injured
diver was rescued by his buddy to be brought back to the sur-
face and disconnected from his breathing apparatus. However,
2 cases with moderate water aspiration were noted follow-
ing a case of hypercapnia and 1 of hyperoxia. Regarding the
specifi c disorders encountered by each sub-population of
divers, the IRR for MCD vs. CS were as follows: 0.6 (95%
CI: 0.3, 1.1) for hypercapnia, 0.2 (95% CI: 0.1, 0.5) for hyper-
oxia and 1.5 (95% CI: 0.4, 5.3) for hypoxia.
Rebreather injuries attributed to decompression sickness
(DCS) were relatively rare ( n = 20, 13%), with 17 cases of
neurological DCS, 1 case presenting with inner ear DCS and
2 cases of musculoskeletal DCS. All the cases recovered with-
out sequelae after prompt recompression with hyperbaric oxy-
gen breathing (time to treatment less than 1 hour). They were
never observed for devices using 100% O
2 or mixtures with
60% O
2 . On the other hand, they were frequently documented
with the DC 55 system using a mixture of 40% O
2 ( n = 17, ie,
80% of DCS cases) used between 35 and 45 msw.
Pulmonary disorders were assigned as disabling injuries
in 11 cases (7%). These symptoms characterized by dyspnea,
FIGURE 2. Distribution of rebreather diving-related specifi c disorders
among the 153 injuries reported over 30 years.
FIGURE 3. Specifi c disorders according to the 2 categories of French military divers using rebreathers.
Downloaded from publications.amsus.org: AMSUS - Association of Military Surgeons of the U.S. IP: 088.168.200.104 on Oct 16, 2017.
Copyright (c) Association of Military Surgeons of the U.S. All rights reserved.
Rebreather Incidents in French Military Divers
MILITARY MEDICINE, Vol. 176, April 2011 449
cough, and expectoration of blood-tinged sputum were initially
classifi ed as pulmonary barotrauma. However, case review of
the datasheets did not show a history of breath-hold or a rapid
ascent during the dive, suggesting an etiology of immersion
pulmonary edema. These pathologic events were predomi-
nantly associated with SCR worn on the back ( n = 10). Only
1 case was observed in a MCD during a CCR dive to a depth of
6 msw for 10 min, but close questioning revealed that before
the dive he had performed a strenuous swimming. In all cases,
outcome was favorable with a complete recovery in the hours
following the initial manifestations.
There were 13 cases of barotraumas involving ENT area
(8%) registered in the database. They did not have any specifi c
symptom pattern compared to scuba diving and only the most
severe forms were declared. In particular, 6 cases resulted in
inner ear barotrauma with cochlear and vestibular impairment,
whereas an isolated rupture of tympanic membrane was cited
in 4 other cases.
Finally, 4 cases of ingestion of caustic liquid (3%), a mix-
ture of water-contaminated soda lime leading to corrosive
injury of the oropharynx and/or esophagus were observed. In
each case, it was due to an error in manipulating the mouth-
piece with water entering the circuit.
Three fatalities were ultimately reported; these concerned
2 CS students using the OXYGERS 57 who were unable to
return to the surface, caught under a barge, during an attack
swim and another accident in which an MCD was trapped in a
deep wreck with no visibility during a dive using a MIXGERS
apparatus. CNS oxygen toxicity for the CS and insuffi cient
gas associated with panic for the last decedent were identifi ed
as the disabling agents in post-mortem investigations.
DISCUSSION
This descriptive study is unique in analyzing the distribu-
tion of nonfatal diving injuries resulting from rebreathers in
a large community of divers, hence making direct compari-
son with other studies diffi cult. The comparison of disorders
among our military divers showed that the overall incidence
did not differ between MCD and CS, but specifi c pathologic
conditions were observed more in 1 of the 2 categories due
to the difference in usage patterns of diving equipment. For
instance, the higher rate of hyperoxia among CS can largely
be explained by more dives performed with CCR using
pure oxygen. Similarly, DCS cases were more identifi ed in
the category of MCD because their operational tasks imply
that they dive predominantly down to 55 msw with Nitrox
mixtures.
Not surprisingly, the most common diving disorders involv-
ing rebreathers were linked to gas toxicity in the majority of
cases, supporting the limited quantifi able data available on
rebreather fatalities investigated by root cause analysis.
3 I n
half of the cases, loss of consciousness occurred with the
potential risk of drowning due to releasing the mouthpiece and
being unable to reach the surface alone. Current diving proce-
dures using rebreathers for military tasks in the French army
do, however, ensure that these complications are exceptional,
as demonstrated by the very low number of fatalities registered
over 30 years. Indeed, 2 preventive measures are mandatory:
(1) systematic linking of divers in pairs, so that a diver can fi nd
his buddy regardless of diving conditions (particularly if vis-
ibility is poor) and can lend assistance in the event of rescue;
(2) using a strap to hold the mouthpiece in position, along with
a lip guard, so that an unconscious diver can still breathe with-
out risk of drowning. The rescuer can then concentrate on the
quality of assistance and respecting the diving parameters for
regaining the surface.
A great majority of these biochemical troubles were encoun-
tered by student divers in whom the intensity of physical exer-
tion (sustained fi nning), diffi culty to adapt breathing to the
device at the start of the course, and the long duration of training
dives led to an imbalance between CO
2 production and elimina-
tion with subsequent arterial CO
2 build-up. Additionally, exces-
sive work of breathing and exercising for any length of time
are prone to make a diver less tolerant to high oxygen levels,
5
and consequently, acute hyperoxia was also predominant in CS
student during the initial part of their course. The association
between hypercapnia and susceptibility to CNS-O
2 toxicity was
ascribed mainly to the vasodilatory effect of CO
2 antagonizing
the protective O
2 -induced cerebral vasoconstriction,
6,7 thereby
increasing delivery of O
2 to neural tissue, resulting in increased
production of deleterious reactive oxygen species.
It is noteworthy that cases of carbonarcosis with loss of con-
sciousness without any warning symptoms were relatively fre-
quently described in our series (18% of hypercapnia cases). A
possible explanation might be related to a decrease in the per-
ception of sensation of CO
2 -related symptoms because of using
hyperoxic gas mixture that depresses peripheral chemorecep-
tors, but the inability to detect suggestive symptoms of hyper-
capnia by novice oxygen divers who did not experience a
training period is another presumed mechanism.
8
From 2002, the replacement of the OXYGERS 57 with
the FROGS, a new generation closed-circuit rebreather, has
increased soda lime capacity, with improved ergonomics and
greater safety in use, leading to a reduction in cases of hyper-
capnia and hyperoxic seizures at the end of long dives.
9
A further preventive measure is applied at the diving
school for trainee divers who carry a higher risk of biochemi-
cal disorders: Each pair of divers is accompanied by at least
1 instructor with open circuit, down to 80 msw (use of a
2-cylinder unit with 18% O
2 , 41% N
2 , and 41% He between
60 and 80 msw). The students only dive in pairs on their own
on the condition that they have validated the rescue procedures
with this type of equipment or are marked on the surface by a
buoy.
It should be noted that accidents caused by equipment fail-
ure were extremely rare (9 cases out of 153, ie, 6%), which
confi rms the reliability of these mechanical devices and the
fact that military divers are trained in a rigorous maintenance
and testing of their equipment, specifi cally with the conserva-
tive management of the scrubber.
Downloaded from publications.amsus.org: AMSUS - Association of Military Surgeons of the U.S. IP: 088.168.200.104 on Oct 16, 2017.
Copyright (c) Association of Military Surgeons of the U.S. All rights reserved.
Rebreather Incidents in French Military Divers
450 MILITARY MEDICINE, Vol. 176, April 2011
Out of the whole series concerning all the rebreathers, DCS
were observed only with the DC 55. Clinical presentations did
not seem to differ from those encountered in air diving with
a predominance of neurological DCS. It is interesting to note
the absence of DCS development with the nitrox 60% O
2 ,
which delivers high oxygen partial pressure in the 25 msw
depth range. On the other hand, the nitrox 40% O
2 mixture
seemed to be more hazardous with an increased propensity for
DCS in dive depths between 30 and 40 msw without decom-
pression stops provided by the decompression tables routinely
used in the Navy since 1965. No DCS cases were noted for
deep dives below 45 msw with nitrox 32.5% and trimix mix-
tures, but this was certainly linked to the protocol imposed
(static dive, limited work time, and in-water decompression
with oxygen), which also led to a low number of dives per-
formed within these depth ranges. Since 2009, 2 neurological
DCS with nitrox 40% have been observed with the new SCR
“CRABE” during validation dives performed at the limit of
physiological conditions. Analysis of PpO2 measured with a
“black box” in this device allowed to highlight lower values
than the theoretical values, thus requiring to change the decom-
pression procedures with the nitrox 40% mixture. Recently,
new decompression procedures with addition of oxygen dur-
ing decompression stops were validated in the hyperbaric cen-
tre of the French navy (CEPHISMER).
Since recent years, there is a growing interest for the occur-
rence of pulmonary edema in endurance swimmers, breath-
hold divers, and SCUBA divers. Various stressors have
been put forward including cold water immersion, exertion,
and increase in breathing resistance from diving equipment,
thereby leading to increased pulmonary vascular pressures
with subsequent capillary stress failure.
10,11 Conversely, pulmo-
nary edema during rebreather diving is an uncommon disorder
that has been reported once only in the literature.
12 Potential
pathophysiological mechanisms that might have contributed
to the development of diving-induced pulmonary edema in
our military divers include the pressure difference between the
lung centroid and the breathing bag of SCR positioned on the
back. These could have potentiated the transmural pulmonary
hydrostatic forces, hence favoring a fl uid shift from the pulmo-
nary capillaries vasculature into the alveoli. Oxygen breath-
ing effects on systemic vascular resistance, cardiac output, and
pulmonary infl ammation might also promote an already exist-
ing impairment of the pulmonary blood–gas barrier.
13
In summary, diving disorders with rebreathers in the French
army are relatively common (around 1 accident per 3,500–4,000
dives) similar to the incidence of DCS with SCUBA air diving
reported in the same population (estimated risk of 1 DCS event
per 3,000 dives).
14 There is a distinct predominance of adverse
events associated with gas toxicities, particularly in student
divers. However, the diving procedures imposed by military
regulations (mouthpiece strap, buddy team with link, and div-
ing instructor with open circuit to lend assistance if necessary
during training) have greatly limited life-threatening complica-
tions, ie, drowning, which are too often recorded in recreational
technical diving.
REFERENCES
1. Williams S : Underwater breathing apparatus . In: The Physiology and
Medicine of Diving and Compressed Air Work , pp 17 – 35 . Edited by
Bennett PB and Elliott DH . London , Baillière Tindall & Cassell , 1969 .
2. Denoble PJ , Ellis J , Vann RD : Fatalities in divers using rebreathers .
Undersea Hyperb Med 2007 ; 34: H10 .
3. Vann RD , Pollock NW , Denoble PJ : Rebreather fatality investiga-
tion . In: Diving for Science 2007 . Edited by Pollock NW , Godfrey JM .
Proceedings of the American Academy of Underwater Sciences 26th
Symposium, pp 101 – 111 . Dauphin Island, AL , AAUS , 2007 .
4. Knowles SB , Marshall SW , Guskiewicz KM : Issues in estimating risks
and rates in sports injury research . J Athl Train 2006 ; 41: 207 – 15 .
5. Butler FK Jr , Thalmann ED : Central nervous system oxygen toxicity in
closed circuit scuba divers . Undersea Biomed Res 1986 ; 13: 193 – 223 .
6. Clark JM , Gelfand R , Lambertsen CJ : Ventilatory, arterial PCO 2 , and
cerebral circulatory responses to incremental exercise during O
2 breath-
ing at 2.0 ATA . Undersea Hyperb Med 1995 ; 22 (Suppl) : 69 .
7. Cantais E , Louge P , Goutorbe P , Méliet JL : Flow velocity changes in the
cerebral arteries during a long period of oxygen breathing . In: Proceedings
of the 26th Annual Meeting of European Underwater and Baromedical
Society (EUBS) , pp 111 – 116 . Edited by Carli-Corleo R. Malta , 2000 .
8. Eynan M , Daskalovic YI , Arieli R , Shupak A , Eilender E , Kerem DH :
Training improves divers’ ability to detect increased CO
2 . Aviat Space
Environ Med 2003 ; 74: 537 – 45 .
9. Blatteau JE , Guigues JM , Pontier JM, et al: [Diving with rebreathing
apparatus. Retrospective study about 96 accidents from 1979 to 2002 in
the French navy] [In French] . Médecine & armées 2006 ; 34: 143 – 50 .
10. Edmonds C : Scuba divers’ pulmonary oedema. A review . Diving and
Hyperbaric Medicine 2009 ; 39: 226 – 31 .
11. Koehle MS , Lepawsky M , Mc Kenzie DC : Pulmonary oedema of immer-
sion . Sports Med 2005 ; 35: 183 – 90 .
12. Shupak A , Guralnik L , Keynan Y , Yanir Y , Adir Y : Pulmonary edema
following closed-circuit oxygen diving and strenuous swimming . Aviat
Space Environ Med 2003 ; 74: 1201 – 4 .
13. Coulange M , Rossi P , Gargne O , et al : Pulmonary oedema in healthy
SCUBA divers: new physiopathological pathways . Clin Physiol Funct
Imaging 2010 ; 30: 181 – 6 .
14. Blatteau JE , Guigues JM , Hugon M , et al : Air diving with decompression
table MN 90: 12 years of use by the French Navy: study about 61 decom-
pression sicknesses from 1990–2002 [In French] . Science & Sports 2005 ;
20: 119 – 23 .
Downloaded from publications.amsus.org: AMSUS - Association of Military Surgeons of the U.S. IP: 088.168.200.104 on Oct 16, 2017.
Copyright (c) Association of Military Surgeons of the U.S. All rights reserved.
