Anda di halaman 1dari 5

The Laryngoscope

Lippincott Williams & Wilkins, Inc.

© 2005 The American Laryngological,
Rhinological and Otological Society, Inc.

Aural Barotrauma in Submarine Escape:

Is Mastoid Pneumatization of Significance?
Akin S. Toklu, MD; Avi Shupak, MD; Senol Yildiz, MD; Samil Aktas, MD; Offir Ertracht, MSc;
Hakan Ay, MD; Yochai Adir, MD; Maide Cimsit, MD

Objective: Submarine escape training is carried the normal distribution of variation in organ size and
out by preselected, healthy young men under strictly is not related to the ability to equalize pressure in the
controlled conditions regarding exposure to pressure middle ear. Key Words: Middle ear, barotrauma, div-
and the rate of pressure change. This provides a ing, mastoid pneumatization, submarine escape.
unique opportunity to investigate the relations be- Laryngoscope, 115:1305–1309, 2005
tween middle ear characteristics and susceptibility to
barotrauma while avoiding possible confounding pa-
rameters. We examined a possible association be- INTRODUCTION
tween mastoid pneumatization and middle ear baro- Barotrauma is pressure-induced tissue damage re-
trauma (MEB) in submarine escape trainees. Study sulting from a failure to equalize the pressure in any
Design: Cross-sectional, parallel-group design. Meth- gas-filled body cavity with that of the external atmo-
ods: Sixty-six subjects aged 19 to 28 participated in sphere. Middle ear barotrauma (MEB) is the main injury
the study. The escape simulation included pressuriza- in sport and professional diving, involving 10% and 30% of
tion to 30 or 60 feet followed by a buoyant ascent to experienced and novice divers, respectively.1 It also affects
the surface. Subjects were evaluated for MEB after an estimated 5% of adult and 25% of child airline passen-
each ascent. A Schuller’s mastoid radiograph was gers.2 MEB results from the inability of the eustachian
taken for the evaluation of mastoid pneumatization.
tube (ET), which is the only efficient conduit available for
Results: Fifteen (23%) of the subjects suffered from
MEB, and 6 (40%) of them had bilateral involvement. pressure regulation during rapid changes in ambient pres-
Repeated impedance audiometry after the comple- sure, to maintain an equal pressure in the tympanic cleft
tion of a successful ascent revealed a significant in- and the ambient atmosphere.2 MEB during the descent on
crease in middle ear compliance. Schuller’s radio- a dive, when pressure changes are measured in atmo-
graphs were obtained from 49 (74%) of the subjects. Of spheres and rates of change reach atmospheres per
these radiographs, 16 (16%) were of ears that had suf- minute, is a direct consequence of the extreme negative
fered MEB. Mastoid pneumatization for all ears ap- middle ear pressures that evolve in such situation. Be-
proached a normal Gaussian distribution, with a cause the middle ear is a closed, noncollapsible, mucosa-
mean area of 9.58 cm2. The mastoid areas and the lined bony cavity, the increased negative pressure will
proportion of ears with mastoid pneumatization at
result in tympanic membrane retraction and focal hemor-
the extremes of the study population did not differ
between barotrauma and no-barotrauma ears. Con- rhages, middle ear mucosal swelling, capillary dilatation,
clusion: In a population with no history of recurrent transudate leakage, hemotympanum, and finally inward
or chronic otitis media and normal tympanic mem- rupture of the tympanic membrane pars tensa.3
brane morphology and compliance, the amount of Submarine escape training simulates the egress from
mastoid pneumatization probably represents merely a sunken or disabled submarine through a hatch into the
sea followed by a buoyant ascent to the surface. The pro-
cedure is carried out in purpose-built, water-filled tower
From the Department of Underwater and Hyperbaric Medicine
(A.S.T., S.A., M.C.), Istanbul University, Istanbul Faculty of Medicine, Istan- under strictly controlled conditions and meticulous super-
bul, Turkey; the Israel Naval Medical Institute (A.S., O.E., Y.A.), Haifa, vision to avoid the potential dangers of drowning, pulmo-
Israel; the Department of Otolaryngology—Head and Neck Surgery (A.S.),
Carmel Medical Center and the Otoneurology Unit, Lin Medical Center
nary barotrauma, cerebral gas embolization, paranasal
Haifa, Israel; and the Department of Underwater and Hyperbaric Medicine sinus, and aural barotrauma.4 Research into middle ear
(S.Y., H.A.), Gülhane Military Medical Academy, Haydarpasa Training Hos- parameters, which might be related to the development of
pital, Kadikoy, Istanbul, Turkey.
MEB, may be of benefit not only for the prediction and
Editor’s Note: This Manuscript was accepted for publication April 1,
2005. prevention of MEB under unusual conditions of rapid
Send Correspondence to Dr. Avi Shupak, Department of Otolaryn- pressure change, as experienced in diving and air-travel,
gology—Head and Neck Surgery, Carmel Medical Center, 7 Michal Street, it may also contribute to our understanding of the patho-
Haifa, Israel. E-mail:
genesis of the chronic negative middle ear pressure fun-
DOI: 10.1097/01.MLG.0000165804.09586.B6 damental to the common pathologies of tympanic mem-

Laryngoscope 115: July 2005 Toklu et al.: Ear Barotrauma in Submarine Escape
brane atelectasis and otitis media, both serous and participants gave their informed consent for the radiography
chronic.5 Mastoid pneumatization has previously been after receiving a full explanation of the study. The institutional
suggested as an important factor in middle ear pressure human research committees approved the protocol and testing
regulation.6 – 8 However, contradictory results have been procedures.
Mastoid pneumatization was measured in a blinded manner
reported, mainly because of confounding parameters such
by one of the authors (O.E.) to avoid possible bias. For that
as different pressure exposures and rates of change, ex- purpose, all radiographs were digitized, the extent of pneumati-
isting middle ear pathology, and bias in the selection of zation was delineated, and the area was calculated using image-
the study group.2,5,9 The submarine escape training pro- analysis software (AnalySIS 3.0, Hamburg, Germany). A good
cedure offers a unique opportunity to study the relation correlation has previously been reported between the pneuma-
between mastoid aeration and susceptibility to MEB while tized area of the mastoid as measured from Schuller’s radiograph
avoiding most of the confounding parameters. Partici- and the mastoid volume found by computerized tomography.10
pants are preselected, the conditions of acceptance being Data were expressed as the mean value ⫾ SD for the mas-
no history of active or previous ear pathology, normal toid areas of normal ears and those who had barotrauma. These
tympanic membrane morphology, normal hearing and im- areas were compared using the simple Student’s t test. The pro-
portions of ears with pneumatized areas of the mastoid at the
pedance audiometry, and the ability to equalize middle
extremes of the study population distribution were compared
ear pressure on land. The maximal pressure and the rates between the barotrauma and no-barotrauma groups using Fish-
of pressure increase and decrease are carefully controlled, er’s exact test. The middle ear pressure equivalents and compli-
avoiding previous research limitations on the simulation ance values derived from impedance audiometry were compared
of rapid pressure changes while immersed. The purpose of using the simple Student’s t test for the baseline measurements
the present study was to prospectively investigate the and repeated measures one-way analysis of variance (ANOVA)
relation between mastoid pneumatization and MEB in for within subject longitudinal comparisons. When applicable, the
submarine escape trainees. Tukey multiple comparisons test was used to delineate the source
of the significant differences. Statistical analysis was performed
using SAS software (SAS Institute, Inc., Cary, NC) on a personal
The study population included 66 healthy male submari-
ners aged 19 to 28 who were participating in a submarine escape RESULTS
training program. None of the subjects had a history of middle or Fifteen of the 66 (23%) subjects suffered from MEB
inner ear disease or experienced difficulty equalizing pressure in during the escape training, and 6 (40%) of them had bi-
the middle ear. Pneumatic otoscopy was carried out before the lateral involvement (Table I). On the basis of the subjects’
study to ensure that all subjects had normal external auditory
interviews after each ascent, all barotraumas occurred
canals free of cerumen, normal morphology and movement of the
tympanic membranes, and good ability to equalize middle ear
during the descent to the escape depth. In the majority of
pressure, as concluded from outward bulging of the eardrum ears, the barotrauma was of mild severity. In 5 of the 21
during the Valsalva maneuver. All had normal hearing, middle (24%) barotrauma ears, a moderate grade III barotrauma
ear pressure, and compliance on pure-tone, speech, and imped- was diagnosed involving significant hemorrhage within
ance audiometry. Bilateral Schuller’s projection lateral mastoid the tympanic membrane.
radiographs were taken for the evaluation of mastoid Baseline tympanometry showed middle ear pressure
pneumatization. equivalents of ⫺24 ⫾ 43 mm H2O (mean ⫾ SD) and ⫺11
The escape training included two free buoyant ascents from ⫾ 30 mm H2O and middle ear compliance of 0.97 ⫾ 0.65
30 feet followed by two hooded ascents from 60 feet. For each mL (mean ⫾ SD) and 1.05 ⫾ 0.62 mL, for ears that later
ascent, the trainees entered the appropriate escape tank com-
had barotrauma and noninvolved ears, respectively.
partment, which was pressurized at the rate of 75 feet/minute to
30 or 60 feet. The subjects were instructed to clear their ears at
These differences were not statistically significant (simple
will during the descent by using the Valsalva maneuver, and if Student’s t test). Repeated impedance audiometry after
this failed, to use the Toynbee maneuver. When the pressure in the completion of successful ascents from 30 and 60 feet
the compartment matched the ambient pressure in the tank, it revealed a significant increase in compliance (P ⫽ .04,
was flooded with water, the hatch was opened, and the submari- repeated measures ANOVA) (Table II).
ner moved into the tank to perform the ascent. The ascent rates Schuller’s projection radiographs were obtained from
for the escapes from 30 and 60 feet averaged 375 feet/minute and 49 (74%) of the submarine escape trainees. Sixteen of the
590 feet/minute, respectively. total 98 (16%) radiographs were of ears that had MEBs.
The subjects were interviewed after each ascent and were Mastoid pneumatization for all ears approached a normal
specifically questioned regarding middle ear clearing problems,
Gaussian bell-shaped distribution (Fig. 1), with an aver-
ear pain, reduced hearing, tinnitus, dizziness, or vertigo. Re-
peated pneumatic otoscopy and tympanometry were carried out
age pneumatized mastoid area of 9.58 ⫾ 5.42 cm2 (mean ⫾
immediately before the first descent and after each ascent to
verify fitness for the next escape drill. MEB was diagnosed by the
findings on pneumatic otoscopy, and its severity was categorized TABLE I.
according to Edmonds’ scale.3 If barotrauma was identified, treat- Barotrauma Events among the Study Participants.
ment was begun with topical and systemic decongestants, and the
No. of Subjects (%) No. of Ears (%)
individual concerned was temporarily disqualified from further
training until complete resolution of symptoms and signs. Barotrauma 15 (23) 21 (16)
Apart from the Schuller’s mastoid radiograph, the subma- No barotrauma 51 (77) 116 (84)
rine escape training procedures, medical examinations, and su- Total 66 132
pervision were all part of the routine training program. The

Laryngoscope 115: July 2005 Toklu et al.: Ear Barotrauma in Submarine Escape
Impedance Audiometry Results in No-Barotrauma Subjects.
Baseline After 30 Feet Ascent After 60 Feet Ascent P Value*

Pressure (mm H2O) ⫺11 ⫾ 30 ⫺6 ⫾ 31 ⫺17 ⫾ 53 0.12

Compliance (mL) 1.05 ⫾ 0.62 1.17 ⫾ 0.64 1.29 ⫾ 0.78 0.04†
Results are mean ⫾ SD.
* P value of the repeated measures analysis of variance comparison between the examinations.
† The source of the significant variance was the difference between the baseline and post-60 feet ascent tests.

SD) and a median of 9.81 cm2. The mastoid areas did not the number of ears with mastoid areas greater than 7 cm2
differ between barotrauma and no-barotrauma ears (Ta- or smaller than 7 cm2 or greater than 20.42 cm2 (Fisher’s
ble III) (Fig. 2). exact test).
It has previously been suggested that ears with mas-
toid pneumatization of 7 cm2 and greater seldom develop DISCUSSION
significant middle ear negative pressure or other sequelae The physiologic pathways for gas exchange between
of ET dysfunction.8 On the other hand, it has been re- the tympanic cleft and the external environment are gas
ported that airline passengers with larger mastoids are diffusion through the middle ear mucosa and pressure
more susceptible to MEB.5 We compared the proportion of equilibration by way of the ET. Gas exchange between the
ears with mastoid areas smaller than 7 cm2 with larger middle ear and the blood by way of the middle ear mucosa
mastoids in the barotrauma and no-barotrauma groups. depends mainly on the slow diffusion of nitrogen at a rate
Mastoids with pneumatized areas greater than the mean of 0.0008 mm Hg/minute.11 Consequently, this pathway
for the study population plus two standard deviations, may be expected to have a minimal effect on middle ear
calculated as 20.4 cm2, represent the upper 3% percentile pressure during the specific conditions of submarine es-
and were considered as large mastoids for the purpose of cape training when the ambient pressure during descent
this comparison. No significant differences were found and ascent changes at the extreme rates of 1,900 and
between the barotrauma and no-barotrauma groups for 9,500 to 15,000 mm Hg/minute, respectively. In contrast,

Fig. 1. Distribution of pneumatized mastoid area in the study population. The area is shown in 2 cm2 intervals on the x-axis. The number of
ears for each interval is depicted separately for all ears (black bars), barotraumas (gray bars), and no-barotrauma (white bars) ears. Gaussian
approximation for all ears (heavy black line).

Laryngoscope 115: July 2005 Toklu et al.: Ear Barotrauma in Submarine Escape
Descriptive Statistics of Mastoid Aeration in the Barotrauma, No-Barotrauma, and All Ears.
Barotrauma No-Barotrauma All Ears

Number of ears 16 82 98
Range (minimum–maximum) 0.13–23.15 0.28–22.77 0.13–23.15
Mean ⫾ SD 9.51 ⫾ 1.49 9.59 ⫾ 5.23 9.58 ⫾ 5.42
Median 9.02 9.88 9.81
The area of mastoid pneumatization is expressed in cm2.
The average mastoid area did not differ between the barotrauma and no-barotrauma groups (simple Student’s t test).

gas flow across the ET is a brisk, gradient-dependent, by itself is known to induce elevation of central venous
bolus exchange of gases between the nasopharynx and the pressure, resulting in reduced patency of the ET.17 Fi-
tympanic cavity and is the main mechanism for middle nally, as dictated by Dalton’s law, the elevated ambient
ear pressure equilibration in the event of rapid changes in pressure is associated with a proportional increase in tym-
ambient pressure. According to Boyle’s law, the volume of panic cavity and systemic oxygen partial pressures. The
the tympanic cavity will increase during the ascent from a resulting hyperoxia causes considerable difficulty in ET
dive, exerting increasing force on the ET. Passive opening ventilatory function, with higher opening, closing, and
of the ET will take place at middle ear-to-ambient over- steady-state pressures.18
pressures of 23 to 38 mm Hg.12 On the other hand, equal- Limitations on the simulation of rapid pressure
ization of the middle ear negative pressure gradients that changes while immersed and simultaneous testing of ET
develop during descent requires active opening of the ET function and the fact that not all individuals with poor ET
by the tensor veli palatine muscle.13 function are susceptible to barotrauma5 prompted the
Previous studies have shown that normal tympanic search for other parameters that might be correlated with
membrane morphology, successful active pressure equili- the ability to equalize middle ear pressure. A recent the-
bration under terrestrial conditions, and normal audiom- oretical model of middle ear pressure regulation during
etry and tympanometry do not necessarily preclude the air travel demonstrated the importance of the relative
occurrence of MEB during a dive.14 Moreover, the results difference between the height above sea level at the point
of ET function tests are inconsistent when it comes to of departure and the destination, the ratio of tympanic
predicting diving-related barotrauma. Whereas the nine- membrane volume displacement to middle ear volume,
step inflation/deflation test was found to be highly effi- and the magnitude of ET obstruction.2 Although this
cient for this purpose,15 the swallow test was of no prac- mathematical model provides an important insight into
tical value in screening prospective divers.16 The the pathogenesis of MEB, there was no investigation of
discrepancy between apparently normal middle ear func- possible interactions between the various parameters.
tion and ability to equalize pressure in the middle ear Specifically, although high tympanic membrane compli-
when there is a rapid change in external pressure reflects ance may buffer pressure changes in middle ears having a
the increased physiologic demand on the ET.14 Immersion low volume, these middle ear characteristics result mainly

Fig. 2. Box-whisker plots of the pneuma-

tized mastoid area for all ears, barotrauma,
and no-barotrauma groups. Median (black
line inside box), average (broken black line),
upper and lower limits of the 75% and 25%
percentiles, respectively (box). The 90%
and 10% percentiles (right and left fine hor-
izontal lines), the 95% and 5% percentiles,
respectively (black dots).

Laryngoscope 115: July 2005 Toklu et al.: Ear Barotrauma in Submarine Escape
from poor ET function or a history of otitis media.6 – 8 tion. In a healthy male population that has normal tym-
Thus, the net effect of mastoid aeration on middle ear panic membrane morphology and compliance and no his-
pressure equalization cannot be concluded from a theoret- tory of recurrent or chronic otitis media, mastoid
ical model unless all possible confounding parameters are pneumatization probably represents merely the normal
eliminated. The findings of two previous studies that ex- distribution of the variation in organ size and is not re-
amined the relation between MEB and mastoid area are lated to the ability to equalize pressure in the middle ear.
contradictory. Less mastoid pneumatization was found
among sport divers who were prone to MEB.9 On the other
hand, significantly larger areas of the mastoids were doc- Acknowledgment
umented among commercial airline passengers who had The authors thank Mr. Richard Lincoln of the Israel
experienced MEB.5 Unfortunately, both studies had con- Naval Medical Institute for his skillful editing of the text.
siderable design limitations. Only those divers who com-
plained of ear symptoms were tested for signs of baro-
trauma, whereas the remainders were diagnosed as BIBLIOGRAPHY
normal without being examined at all. Furthermore, the 1. Clenney TL, Lassen LF. Recreational scuba diving injuries.
pressure gradients varied significantly among the divers Am Fam Physician 1996;53:1761–1774.
because both shallow dives to 6 feet and deep dives up to 2. Kanick SC, Doyle WJ. Barotrauma during air travel: predic-
183 feet were included in the evaluation. In the airline tions of a mathematical model. J Appl Physiol 2004;
passengers study, the barotrauma group did not undergo 3. Edmonds C, Lowry C, Pennefather J. Diving and Subaquatic
otolaryngologic evaluation before the flight, the study par- Medicine, 3rd ed. Oxford: Butterworth-Heinemann, 1992:
ticipants’ ability to perform a satisfactory middle ear pres- 115–139.
sure equalization maneuver was not verified, and the 4. Yildiz S, Ay H, Gunay A, et al. Submarine escape from depths
of 30 and 60 feet: 41,183 training ascents without serious
pressure gradients were probably not controlled because injury. Aviat Space Environ Med 2004;75:269 –271.
no information is provided regarding possible differences 5. Sade J, Ar A, Fuchs C. Barotrauma vis-à-vis the “chronic
in height above sea level between the point of departure otitis media syndrome”: two conditions with middle ear gas
and the destination. deficiency. Is secretory otitis media a contraindication to
Submarine escape training provided a unique oppor- air travel? Ann Otol Rhinol Laryngol 2003;112:230 –235.
6. Tos M, Stangerup SE, Hvid G. Mastoid pneumatization. Ev-
tunity to study the role of mastoid pneumatization in the idence of the environmental theory. Arch Otolaryngol
pathogenesis of MEB under strictly controlled conditions 1984;110:502–507.
of pressure exposure and rate of pressure change. The 7. Cinamon U, Sade J. Mastoid and tympanic membrane as
study population was homogenous in terms of sex and age pressure buffers: a quantitative study in a middle ear cleft
group, normal tympanic membrane morphology, good ac- model. Otol Neurotol 2003;24:839 – 842.
8. Sade J, Fuchs C. Secretory otitis media in adults. I. The role
tive middle ear clearing technique, normal hearing and of mastoid pneumatization as a risk factor. Ann Otol Rhi-
tympanometry, and a lack of any significant history of nol Laryngol 1996;105:643– 647.
otologic disease or current pathology that might influence 9. Uzun C, Adali MK, Koten M, et al. Relationship between
their ability to equalize pressure in the middle ear. mastoid pneumatization and middle ear barotrauma in
divers. Laryngoscope 2002;112:287–291.
The main finding of this study is that the magnitude
10. Todd NW, Pitts RB, Braun IF, Heindel H. Mastoid size de-
of mastoid pneumatization per se is not a critical factor in termined with lateral radiographs and computerized to-
the pathogenesis of MEB. This supports the notion that mography. Acta Otolaryngol 1987;103:226 –231.
mastoid hypopneumatization does not necessarily reflect 11. Doyle WJ, Alper CM, Seroky JT, Karnavas WJ. Exchange
poor ability to equalize middle ear pressure, and a well- rates of gases across the tympanic membrane in rhesus
monkeys. Acta Otolaryngol 1998;118:567–573.
developed mastoid air cell system is not always associated 12. Cantekin EI, Saez CA, Bluestone CD, Bern SA. Airflow
with good ET function.6 Retraction of the pars flaccida through the eustachian tube. Ann Otol Rhinol Laryngol
first, and then of the pars tensa, can compensate for some 1979;88:603– 612.
of the volume reduction that is required when ambient 13. Bluestone CD, Doyle WJ. Anatomy and physiology of the
pressure increases. This mechanism can displace up to 0.2 eustachian tube and middle ear related to otitis media. J
Allergy Clin Immunol 1988;81:997–1003.
to 0.3 mL,5 and buffers negative pressures of up to about 14. Shupak A, Sharoni Z, Ostfeld E, Doweck I. Pressure chamber
23 mm Hg in a middle ear having an average volume of 10 tympanometry in diving candidates. Ann Otol Rhinol La-
mL. The small yet significant increase in middle ear com- ryngol 1991;100:658 – 660.
pliance found in our subjects after repeated submarine 15. Uzun C, Adali MK, Tas A, et al. Use of the nine-step inflation/
deflation test as a predictor of middle ear barotrauma in
escape simulations probably reflects the strain exerted on
sports scuba divers. Br J Audiol 2000;34:153–163.
the middle elastic layer of the tympanic membrane. How- 16. Schuchman G, Joachims HZ. Tympanometric assessment of
ever, significant attenuation of the large middle ear pres- eustachian tube function of divers. Ear Hear 1985;6:
sure fluctuations during descent and ascent may not be 325–328.
anticipated in such situation. 17. Andreasson L, Ingelstedt S, Ivarsson A, et al. Pressure-
dependent variation in volume of mucosal lining of the
middle ear. Acta Otolaryngol 1976;81:442– 449.
CONCLUSION 18. Shupak A, Tabari R, Swarts JD, et al. Effects of systemic
The occurrence of MEB in submarine escape trainees hyperoxia on eustachian tube ventilatory function. Laryn-
cannot be predicted by the size of mastoid pneumatiza- goscope 1997;107:1409 –1413.

Laryngoscope 115: July 2005 Toklu et al.: Ear Barotrauma in Submarine Escape