See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/8122155

Barotrauma during air travel: Predictions of a mathematical model

Article  in  Journal of Applied Physiology · June 2005

DOI: 10.1152/japplphysiol.00974.2004 · Source: PubMed

CITATIONS READS
30 239

2 authors:

Stephen Chad Kanick William J Doyle

Profusa, Inc / Dartmouth College University of Pittsburgh
95 PUBLICATIONS   1,125 CITATIONS    372 PUBLICATIONS   13,191 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Otolaryngology translational research View project

Eustachian tube function and middle ear pressure regulation View project

All content following this page was uploaded by Stephen Chad Kanick on 11 September 2014.

The user has requested enhancement of the downloaded file.

Articles in PresS. J Appl Physiol (December 17, 2004). doi:10.1152/japplphysiol.00974.2004

Barotrauma during Air Travel: Predictions of a Mathematical Model

Stephen Chad Kanick 1, 2

William J. Doyle 1,3

1
Department of Pediatric Otolaryngology at Children's Hospital of Pittsburgh, Pittsburgh, PA 15213;
2
Department of Chemical and Petroleum Engineering at University of Pittsburgh, Pittsburgh, PA 15213;
3
Department of Otolaryngology at University of Pittsburgh School of Medicine, Pittsburgh, PA 15213

Correspondence To:

Stephen Chad Kanick

Children’s Hospital of Pittsburgh

3705 Fifth Ave @ DeSoto St

Pittsburgh, PA 15213

Phone: 412 692-5536

Fax: 412 692-6074

[email protected]

Key words: Barotrauma, Middle ear pressure regulation, Eustachian

tube, Valsalva maneuver, Tympanic membrane

Copyright © 2004 by the American Physiological Society.

Abstract
Middle ear barotrauma during flight is a painful disorder experienced by passengers who

cannot properly regulate their middle ear pressure in response to the changing cabin pressures

during ascent and descent. Previous reports emphasized the important role of poor Eustachian

tube function in disease pathogenesis but paid little attention to other moderating factors. Here

we describe a mathematical model of middle ear pressure regulation and simulate the pressure

response to the changes in cabin pressure experienced over typical flights. The results document

buffering mechanisms that decrease the requisite efficiency of active, muscle-assisted Eustachian

tube opening for disease-free flight. These include the relative difference between destination

and departure elevations and the ratio of maximum tympanic membrane volume displacement to

middle ear volume, where greater absolute values require lesser efficiencies for disease-free

flight. Also, the specific type of functional deficit is important since ears with a completely

obstructed Eustachian tube can be less susceptible to barotrauma than those with a Eustachian

tube that passively opens but fails to dilate in response to muscle activity. These buffering

systems can explain why some children and adults with poor Eustachian tube function do not

experience middle ear barotrauma.

2
Introduction

Middle ear (ME) barotrauma, the most common medical disorder associated with modern

air travel, affects an estimated 5 percent of adult and 25 percent of child passengers (49). Two

primary expressions of barotrauma can be distinguished based on signs and pathophysiology;

barotitis media and baromyringitis. Barotitis media is ME mucosal (MEM) inflammation,

hemorrhage and leakage of transudate into the ME precipitated by moderate ME underpressures

relative to the surrounding MEM. Baromyringitis is structural damage to the tympanic

membrane (TM) with severe pain caused by large pressure differences between ME and cabin.

Sequelae of barotrauma can include dizziness, tinnitus and deafness (16, 45).

Previous studies describing the pathogenesis of ME barotrauma were done on divers or

on patients being treated in hyperbaric O2 chambers (3, 25, 32, 41), situations that do not share

the physiological conditions experienced during pressurized flight. Moreover, most publications

and reviews that specifically focused on ME barotrauma during flight lack empirical data and

described disease pathogenesis using broad generalizations (2, 6, 33, 45) with a primary focus on

the function of the Eustachian tube (ET). The inadequacy of this approach was recently

highlighted in study by Sade and colleagues who reported disease-free flights for children and

adults with presumably poor ET function (48). Here, we approach the pathogenesis of ME

barotrauma from the perspective of basic physiology using both descriptive and mathematical

formats. Our goal is to clarify the buffering mechanisms that protect the ME from barotrauma

during pressurized flight.

Middle Ear Pressure Regulation

Barotrauma is caused by an inability to maintain near pressure equivalence between the

ME and airplane cabin as the latter is changed rapidly during ascent and descent. Normally, the

pressure of the fluid-free ME is near ambient (PME ≈ PAMB ≈ PCabin) which ensures free vibration

of the TM and efficient transduction of sound energy to the inner ear. Because the ME is usually
3
a closed, relatively non-collapsible, temperature stable, mucosal lined bony cavity, its pressure is

a direct function of the contained gas volume, and gas transfers to or from the ME change its

pressure.

The ME consists of two functionally discrete but continuous airspaces: the anterior,

tympanum which contains the ossicles, ligaments and muscles of the sound transducer

mechanism, and the posterior, mastoid cavity which is subdivided into numerous

intercommunicating air-cells (5). While the variance among individuals and age groups in

tympanum volume is low (Vtym ≈ 1 ml), that of the mastoid is large (Vmas ≈ 0 – 15 ml) due to

contributions of age, gender and disease history effects (37, 46). The anterior wall of the

tympanum is continuous with the osseous portion of the Eustachian tube (ET), the lateral wall

includes the TM, the medial wall includes the round window membrane and the posterior wall

opens to the mastoid airspace by way of a large air-cell, the antrum (5).

Figure 1a shows the various gas exchange pathways for the ME when isolated within an

airplane cabin. The tympanum can exchange gas with the external environment via the TM and

with the inner ear via the round window, but experimental measurements show that transfers

across these pathways are negligible (18, 21). Therefore, in describing ME pressure regulation,

the physiologically relevant pathways are: tympanum-antrum-mastoid, ME-MEM-blood and

tympanum-ET-nasopharynx (NP). Because the tympanum and mastoid are continuous in the

airphase, total pressure differentials are rapidly equilibrated and established gas partial-pressure

differentials decay quickly (24). ME-MEM-blood gas exchange is a diffusive process whose rate

depends on the extant partial-pressure gradients and gas specific exchange constants (20, 22, 23).

At physiological partial-pressure gradients between ME and venous blood (VB), gas exchange

across this path is primarily attributable to the relatively slow exchange of N2 and, consequently

this exchange is expected to have a minimal effect on ME pressure over most flight durations. In

contrast, gas exchange across the ET is a rapid, gradient dependent, bolus exchange of mixed

4
gases between NP and tympanum. Under normal physiological conditions, this is the only

direct, potential communication between ME and ambient environment and the only exchange

pathway capable of reducing established positive, ambient-ME pressure gradients.

The functional anatomy of the ET has been described in many publications (2, 5, 17, 52).

Briefly, the posterior portion of the ET is a mucosa lined, bony tube continuous with the anterior

tympanum while the anterior portion is cartilaginous medially and membranous laterally (Figure

2). The cartilaginous portion is usually closed by a tissue pressure, PET that equals the sum of the

ambient pressure (PAMB - a consequence of the incompressibility of body fluids) and a vascular

pressure (PVAS) (17, 27). A muscle, the tensor veli palatini (mTVP) takes origin from the

membranous wall of the ET and terminates on the hamular process and within the palatine

aponeurosis (5). Activation of the muscle during swallowing exerts an anterior-lateral-inferior

vector force (FTVP) on the membranous wall of the ET (52).

Figure 1b depicts these functional relationships. There, the ET is represented as a balloon

pressure valve that is normally closed by the pressure difference between PET and both PNP and

PME. ET opening can be effected by passive, pressure-driven processes or by active, pressure-

driven or muscle-assisted mechanisms (36, 51). Passive, pressure-driven ET opening occurs

when the force, F(P), associated with either PME or PNP exceeds that of PET. Active, pressure-

driven ET opening occurs when PNP is increased by Valsalva or other maneuvers so that the

applied F(PNP) exceeds F(PET), or for some individuals, when F(PET) is reduced by yawning or

mandibular repositioning (5, 13, 17, 36). Active, muscle-assisted ET opening occurs when the

mTVP contracts with sufficient force (FTVP) to overcome the F(PET) (5, 27, 28). The teleological

effect of these “normal” ET openings is to allow NP-ME gas exchange so as to maintain an

approximate equilibrium between PME and PAMB as PME is decreased by transmucosal gas

exchange and PAMB fluctuates with barometric conditions; i.e. normal ME pressure regulation (5,

31, 43).

5
 Movements of the TM in response to ME-ambient pressure differentials are an important

exception to the assumed fixed ME volume. There, small fluctuations in that pressure gradient

can be absorbed by ME volume changes in response to pressure-driven TM movements (46, 47).

This is illustrated in Figure 1c which shows the TM response to a ME-Cabin pressure gradient.

As given by Boyles law, the magnitude of this pressure buffering effect is a function of the ratio

of TM volume displacement to ME volume. In healthy ears, the maximum TM displacement

volume is approximately 1% of the ME (i.e. tympanum + mastoid) volume (46) and the

buffering effect of TM displacement on ME pressure is limited. However, persistent ME disease

causes a significantly reduced mastoid volume and can cause a hyper-compliant TM (19, 26),

changes that will increase the determinate ratio for TM buffering and may reduce the affected

ME’s susceptibility to barotrauma.

Normal ME Pressure Regulation during Flight

During airplane ascent, PCabin (=PAMB) decreases which causes decreasing PNP, PET and

PMEM while PME is relatively unchanged (with the exception of a minor decrease associated with

TM bulging) vis a vis takeoff. This results in the development of positive ME-ambient, ME-NP

and ME-ET pressure gradients. At times when F(PME) exceeds F(PET), the ET passively opens,

gas of ME composition flows from the ME to NP and PME is reset to the extant value of PET. The

residual ME-Cabin pressure gradient representing PVAS (i.e. PET -PAMB) as well as any gradients

that develop by transMEM gas exchange (PAMB-PME- PME) or by minor changes in elevation

during flight (PAMB+ PAMB-PME) are reduced by directional gas flows when the ET is actively

opened by the mTVP.

On descent, PAMB increases causing increases in PNP, PET and PMEM, while PME is

relatively unchanged vis a vis cruising altitude. This causes a rapidly developing, positive

ambient-ME (and ET-ME) pressure gradient, and a relative PMEM overpressure with respect to

PME. Under such conditions, neither F(PNP) or F(PME) will exceed F(PET) and passive ET
6
openings are not possible. Consequently, during descent, the passenger must periodically open

the ET actively by swallowing to induce mTVP activity or by other maneuvers that cause F(PNP)

to transiently exceed the extant F(PET) or cause F(PET) to decrease to less than F(PNP). Of the

latter, Valsalva is the most commonly used wherein air is forcibly expelled from the lungs while

keeping the mouth closed and pinching the nose (6, 17, 33, 36, 45). This greatly increases the

PNP and can passively open the ET to allow for NP gas transfer to the ME.

Pathogenesis of Barotrauma
The rapid changes in cabin (ambient) pressure during airplane ascent and descent can

overtax the ME pressure-regulating system and provoke barotrauma. For passengers with

excellent active ET opening function, ME pressure regulation during flight is a nominal task but

for those with less efficient ET function, infants and children and those with concurrent nasal

inflammation caused by colds or allergy, the task may be impossible (6, 7, 33). If transET gas

flow does not reestablish a near zero ME-ambient pressure gradient during descent, PET will

exert its force over a larger collapsible section of the ET lumen which can exceed the maximal

force exerted by either the mTVP or active NP pressurization (17). This phenomenon, known as

ET “locking”, occurs at an individual-specific ME-ambient pressure gradient and effectively

obstructs the ET to any further gas flow.

In the absence of adequate pressure regulation, the large ME-ambient pressure gradients

that develop during ascent and descent cause maximal extension of the TM with stretching and

tearing of its structural elements. The TM can develop focal hemorrhages, local pocket

formation and may perforate (17, 45). At submaximal extension, this is perceived as a feeling of

“fullness” in the ear and at maximal extension as severe pain (16, 55). These are signs and

symptoms of baromyringitis. Alternatively, at a specified value of approximately 200-300

mmH2O, the positive PAMB-PME gradient that develops during descent will cause a larger PMEM-

PME gradient resulting in MEM swelling, capillary dilatation, transudative leakage and

7
accumulation of fluid in the ME via “hydrops ex vacuo” (50). This set of signs presents as

barotitis media.

An issue often faced by Otolaryngologists is the assignment of individual patients to risk

groups for barotrauma, i.e. which patients can fly safely and which should take precautions prior

to airflight (54). Currently, such assignments are based on history, clinical observations and, in

some centers, ET function test. We believe that these assessments may not account for all

influential factors that determine barotrauma risk. Here, we take a unique approach to

addressing this issue by first formulating a mathematical model of ME pressure regulation during

flight based on the physiological considerations outlined above and then studying the effects on

barotrauma risk of varying physiological parameters included within the model.

8
Methods

Definition of Disease Sates

We use the ME-Cabin pressure gradient ( PME-Cabin) as an index measure of barotrauma,

or

∆PME −Cabin (t ) = PME (t ) − PCabin (t ) (1)

where PME and PCabin are absolute pressures within the ME and cabin at a time step (t). Based on

the results of previous studies, we assigned PME-Cabin < -250 mmH2O as the threshold for onset

of barotitis media (50) and │ PME-Cabin│> 1300 mmH2O as the threshold for onset of

baromyringitis with severe pain (6).

Gas Exchange Model

The model compartments and linkages shown in Figure 1a depict the gas exchange

components of the ME system. All compartments are assumed to be well-mixed and isothermal

with inter-compartmental communication defined as the transfer of gas moles down pressure

gradients along the linkages. Model compartments include the ME (tympanum + mastoid),

MEM, NP, VB and Cabin. The ME is linked periodically to the NP during ET openings and

continuously with the VB via the MEM. The Cabin acts as the ambient environment for the

system, directly affects PET and PMEM (assumed to be nearly instantaneous and linear based on

the results of pressure chamber experiments (30)), exerts a mechanical force on the TM and

exchanges gas with the NP. The Cabin is assumed to be an infinite gas source/sink and the

volumes of the NP and VP are assumed to be finite but much greater than that of the ME.

Consequently, species-gas exchange between the ME and larger compartments does not affect

the partial/total pressures of those compartments, but does have a significant effect on ME

partial/total pressures.

Cabin Pressurization

9
 During ascent, the airplane rises to a cruising altitude of approximately 30,000 ft above

sea level. In order to protect passengers from the adverse affects of these extreme, low-

pressures, the cabin is pressurized to an effective cruising altitude of approximately 8000 ft (35,

45, 55). Cabin pressurization was modeled by increasing cabin altitude at a constant rate of 90

m/min (approximately that of a Boeing 747) from departure elevation to the effective cruising

altitude (45). Cabin pressure is a function of cabin elevation and, assuming ideal compressible

gas behavior, is given by:

⎛ − mgz ( t ) ⎞
⎜ ⎟
⎜ BT ⎟
P tot
Cabin (t ) = PAMB e ⎝ o ⎠
, (2)

where t is time, g is acceleration due to gravity, m is the average mass of an air molecule, B is

boltzman’s constant, To is the cabin temperature, PAMB is ambient pressure (ref. sea level), and

z (t ) is the effective altitude of cabin pressurization (ref. sea level). Because gas species mole

fractions are constant during flight, cabin N2 and O2 partial pressures are calculated using:

N2
PCabin (t ) = 0.79 PCabin
tot
(t ) , (3)

O2
PCabin (t ) = 0.21PCabin
tot
(t ) . (4)

Similarly, airplane descent was modeled as a linear decrease from effective cruising to

destination altitudes at -90 m/min, while calculating the increases in PCabin(t).

Pulmonary Exchange

Total NP pressure is assumed to be equal to that of the Cabin or,

PNP (t ) = PCabin (t ) , (5)

while NP gas species pressures are assumed to be an average of the respective Cabin

(experienced during inhalation) and alveolar (experienced during exhalation) values (34). Total

VB pressure is linked to total Cabin pressure via nasopharyngal-pulmonary gas exchange as,

PVB (t ) = PNP (t ) (6)

10
under the assumptions that the pulmonary-blood gas exchange is very rapid and blood-gases

stored in fatty tissues would contribute minimally to the ME arterial supply. Throughout flight,

VB partial-pressures of O2 and CO2 are assumed to be buffered at constant values by hemoglobin

and bicarbonate reactions, the VB remains saturated at a constant H2O pressure, and VB N2

pressure is a function of nasopharyngeal N2 pressure, calculated as:

PVBN 2 (t ) = PNP (t ) − PVBO2 (t ) − PVBCO2 (t ) − PVBH 2O (t ) . (7)

Middle Ear Pressure Dynamics during Flight

The driving mechanisms included in the model that affect ME pressure dynamics during

flight are transET and transMEM gas exchanges, and the pressure effects of ME volume changes

due to TM displacement.

Eustachian Tube Opening

During ET openings, gas flows between the ME and NP in response to the total extant

pressure gradient. The ET opens when a force applied to the ET lumen overcomes the ET

closing force equal to the sum of the force of the mucosal tissue pressure (PETAc) and that

attributable to intraluminal surface tension (FST) or:

FET (t ) = PET (t ) AET + FST (8)

where AET is the surface area of mucosal contact.

Pressure-driven ET opening occurs when ME (passive) or NP pressure (active or passive)

exerts a force (PMEAME’ or PNPANP’) on the ET lumen greater than FET such that,

FET (t ) F (t )
PME (t ) > = PME
O
− ET (t ) or PNP (t ) > ET = PNP
O
− ET (t ) (9)
AME ' ANP '

where AME’ and ANP’ are the effective ET surface areas exposed to the ME and NP, respectively,

and POME-ET and PONP-ET are the ME and NP opening pressures of the ET. These opening

pressures have been measured empirically and were reported as pressure differentials referenced

11
 − ET = PME (t ) − PAMB (t ) ; PNP − ET = PNP (t ) − PAMB (t ) ) (13, 51). We used
O' O O' O
to ambient (i.e. PME

representative values from those data sets in this model (See Table I).

When relative ME overpressures cause the ET to passively open, gas exchange continues

until the intraluminal airphase pressure of the ET (PIET) equals the tissue pressure (PET) of the

ET, where PIET = PME. This results in a residual ME overpressure with respect to the NP (PC’)

that is usually referred to as the ET closing pressure and can be written expressed as:

P C ' = PET (t ) − PNP (t ) = PVAS . (10)

Because gas flows from ME to NP, ME species gas fractions are not affected by this transfer and

these were calculated by multiplying the preexisting gas fractions by the revised total ME

pressure. As with PO’ME-ET, PC’ has been measured empirically (51) and representative values are

used in this model (See Table I).

For ET openings caused by relative NP overpressures, gas exchange first occurs between

NP and ME wherein those pressures are equilibrated, and then between ME and NP as ME

pressure is reduced to the ET closing pressure. The effect of the NP to ME gas transfers on ME

partial pressures at a timestep (dt) was modeled as the weighted average of NP and ME species

pressures as given by:

i
dPME (t ) = y i ( P tot (t ) − P tot (t )) . (11)
NP NP ME
dt

where yNPi equals the species mole fraction in the NP. These partial pressures were then adjusted

for the ME to NP gas exchange as described above.

Active muscle-assisted ET opening occurs when the force of mTVP contraction surpasses

FET, where

FmTVP (t ) > FET (t ). (12)

For all FmTVP satisfying this condition, the magnitude of that muscular force determines the ET

lateral wall displacement as described by Hooke’s law:

12
 X ET (t ) = FmTVP (t )C ET . (13)

where CET is the compliance of the ET lumen and XET is the lumen wall displacement distance.

Figure 2 provides a detailed representation of the forces acting on the ET during mTVP activity.

Assuming that transET gas exchange follows Hagen-Poiseuille flow between two parallel plates

(11), then

⎛ 2 ⎞ ∆P (t )X ET (t ) W 3

QET (t ) = ⎜ ⎟ ME − NP . (14)
⎝3⎠ µL

where: QET is the volume of gas transferred, L is the ET length, W is the superior-inferior height

of the tube lumen, XET is the mediolateral lumen width, is the viscosity of air and PME-NP is

the driving force for transfer. Because W, and L are constants for a given ET and XET is a

defined function of FmTVP, we can extract from this equation an analytical expression for the

active resistance to gas flow (RA) that is conditioned on FmTVP, or:

∆PME − NP (t ) ⎛ 3 ⎞ µL
R A (t ) = =⎜ ⎟ . (15)
QET (t ) ⎝ 2 ⎠ W (FmTVP (t )C ET )
3

While FmTVP is not measurable in vivo, RA is an outcome measure of the forced-response test of

ET function which has been used in both clinical evaluations and experimental studies in humans

(11, 12, 23, 51). In the model, RA is an inputted parameter used to describe mTVP effectiveness

with respect to active ET openings with representative values selected from existing data sets

(See Table I). Lacking measured values of FmTVP , we did not include ET “locking” in the model

description.

Using the empirical measures of RA and ET opening time (TA) reported by Cantekin and

colleagues (10-12), transET volume gas exchange can be then be described as the following:

∆PME − NP (t )T A
QET (t ) = . (16)
RA

13
Regular tubal opening by mTVP activity occurs during rhythmic swallowing as reported by

Tideholm (53). In this model, we used a normal swallowing frequency (Sf) of 5.2 openings/hr

during cruising and an increased value of 31 openings/hr during descent.

Volume gas flow during mTVP induced tubal openings (at timestep t) represents the

directional movement of a proportional number of gas moles (N) between compartments, with

the relationship formalized as:

PNP (δt )QET (δt )

∆N ET (δt ) = , (17)
K

where K is the product of ME temperature and the gas constant. Assuming an ideal gas, ME

pressure after the swallow is calculated from the sum of NET( t) and the NME(t) before the

swallow. This value is then used to calculate a new ME volume, VME(t+ t) and ME pressure,

PME(t+ t) (see “TM displacement” section below). The effect of these transfers on ME gas

species pressures was modeled as described above for the directional transfers caused by passive

ET openings.

Middle Ear-Mucosal Gas Exchange

The ME exchanges gas with the local VB by diffusion across the MEM. Here, the MEM

was modeled as the VB gas source/sink for this exchange, such that ME gas species pressures,

PMEi are calculated as

= k i (PME (t ) − PVBi (t )),

i
dPME (t ) i
(18)
dt

where ki is an empirical species exchange constant, PMEi is ME species pressure, and PVBi is VB

species pressure. Equation 18 was applied for N2, O2, CO2 and H2O and total ME pressure was

equal to the summation:

tot
PME (t ) = ∑ PME
i
(t ) . (19)

14
Table II lists the initial gas-species pressures for these compartments and the transMEM time-

constants measured by experiment (22). The resultant PME(t+ t) value following transMEM

exchange is calculated after the ME volume (VME(t+ t)) is adjusted for V(t+ t) (see “TM

displacement” section below).

TM Displacement
Figure 1c illustrates TM displacements in response to a pressure gradient across the

membrane, PME-cabin. TM deformation is a function of its compliance and the force applied to

the TM (equal to transTM pressure gradient multiplied by TM surface area). The deformation is

governed by Hooke’s law:

X TM (t + δt ) = ∆PME −Cabin (t + δt )ATM CTM (20)

where XTM is the TM displacement distance, ATM is the TM surface area and CTM is the TM

compliance. TM volume displacement is calculated as:

∆VTM = X TM (t + δt )ATM , (21)

with displacements constrained to the range,

− ∆VTM
max
< ∆VTM (t + δt ) < ∆VTM
max
. (22)

ME volume is calculated as the sum of the system ME volume and the TM volume displacement

as,

VTM (t + δt ) = VME
sys
+ ∆VTM (t + δt ) . (23)

VMEsys is the value of the closed system (i.e. the “initial” starting point for TM displacement

calculation), equal to either the initial ME value ( VME (t = 0 ) ) or the value following the previous

transET or transMEM transfer. From Boyle’s Law (i.e. PMEVME = constant ) PME is then

calculated for varying TM displacements, as

tot
PME (t )VME (t )
P (t + δt ) =
tot
. (24)
VME (t + δt )
ME

15
Simulation Package

The above listed equations allow for the calculation of the time-dependent changes in the

ME-ambient pressure gradient during simulated flights. The required input parameters for the

model are listed in Table I. The relevant equations were coded into a MatLab v.6.1 m-file and

entered into a loop which was iterated using a timestep ( t) of 0.001 minutes. Durations of all

flights were obtained from published flight schedules, with domestic flights averaging ~170

minutes in length. The order of sequential operations at each time step was the calculation of:

cabin pressurization, gas species-pressures and total pressure for each compartment (PCabin, PNP

and PVB); gas species-pressures and total pressure (PME adjusted for VTM) for the ME after

transMEM exchange, and gas species-pressures and total pressure for the ME after conditional

gas transfers through the ET based on inputted swallowing rhythm (QET adjusted for VTM)

and/or passive openings (PME adjusted for VTM).

16
Results

Model Validation

To evaluate the predictive accuracy of the model, we simulated the ME pressure

dynamics for a pressure chamber experiment by Groth and colleagues (29) who described ME

pressure change (measured as TM volume displacements) in pilots exposed to high rates of

pressurization (1920 ft/min) over short time periods (25 sec). Model parameters were estimated

from the experimental data (PO’ = 292 mmH2O, PC’ = 136 mmH2O, RA= 7.5 mmHg/cc/min,

CTM= 425 mmHg/mL, TA = 250 msec, and Sf = 33 swallows/ min). A comparison of model and

experimental results is shown in Figure 3. During ascent, ambient pressure decreased and the

resulting ME overpressures caused outward TM displacement. At a relative ME overpressure of

292 mmH2O, the ET passively opened and the ME-ambient pressure gradient was partly

dissipated as gas was transferred from ME to NP, a process interrupted when the ET passively

closed at PME = PET. This was associated with TM repositioning to a lesser volume

displacement. During simulated descent, ambient pressure increased causing inward

displacement of the TM. At all times, PET exceeded PME and PNP and passive ET openings did

not occur. Rather, at semi-regular intervals, swallowing caused mTVP contraction and active ET

openings. Each opening was associated with a transfer of gas from NP to ME, a consequent

reduction in the ME-ambient pressure gradient and reduced TM volume displacement.

Sequential swallows caused a progressive lessening of the residual ME underpressure. This

comparison shows that our model can accurately reproduce experimental data for ME pressure

behavior during simulated flights.

Flight Simulations

Figure 4 shows PCabin as a function of time during three simulated 170 min “flights”, each

departing from Pittsburgh, PA (PIT) and arriving at: PIT, Denver, CO (DEN) and Miami, FL

(MIA). For all “flights”, PCabin decreased during airplane ascent, remained relatively constant
17
during cruising and increased on descent. The magnitude of pressure change experienced by

passengers depends on the relative pressure differences between departure/cruising/destination

elevations. Table III lists the elevation and ambient pressures for these airports and for the

airplane cabin at the effective cruising altitude. Using these three flight paths, we simulated the

ME pressure dynamics for a “normal” ME (See Table I) and for ears with “abnormal” structural

(e.g. ME volume, TM displacement) or functional (e.g. PO’ ME-ET, RA) parameters.

1. ET function measurements in “normal”, disease-free ears document passive ET openings

at moderate ME-ambient overpressures (300 to 500 mmH2O), passive ET closing at above-

ambient ME pressures (100-200 mmH2O), and the ability of the mTVP to open the ET over a

large range of applied ME-ambient pressure gradients (12). The ME pressure changes during a

simulated flight for such an ear (all parameters equal to normal) are shown in Figure 5a that

demonstrates the development of relatively low magnitude ME-ambient pressure gradients

throughout the duration of the flight, with none of those gradients exceeding the threshold for

either expression of ME barotrauma.

2. A common treatment for otitis media is the insertion of tympanostomy tubes, small tubes

placed within the TM that allow for constant communication between ME and ambient

environment (42). An abnormal physiological condition referred to as a “patulous” ET also

allows for constant NP-ME communication (4). There, function tests show that the PET is less

than PAMB resulting in a continuously open ET (5). Simulated flights for ears with either of these

conditions yield the trivial result of a zero mmH2O ME-cabin pressure gradient throughout flight

and consequently protection from barotrauma.

3. Rarely, clinical tests document an ET that is physically obstructed by enlarged adenoids

or by nasopharyngeal carcinoma (40, 44). More frequently, the ET is intrinsically blocked by

intraluminal swelling and venous engorgement caused by posterior extension of NP

inflammation that accompanies viral infections or allergy (5). ET function tests for both

18
conditions document a failure of applied ME overpressures to passively open the ET and an

inability of the mTVP to effect ME-NP gas transfers (5). We modeled this condition by

inputting high PO’values (PO’ME-ET = 2500 mmH2O, PO’NP-ET > 2500 mmH2O) and a high RA

(1/RA ≈ 0) value (other parameters equal normal). The results for the three simulated flights are

shown in figure 5b. During ascent, the lack of passive ET openings leads to a positive ME-cabin

gradient of 2020 mmH2O, a pressure that exceeds the threshold for pain and baromyringitis.

During cruising, that gradient is slightly reduced by the slow, transMEM N2 exchange, and

during descent, the gradient is decreased as PCabin increases. On landing, the ME-cabin gradient

(terminal pressure gradient = TPG) depends almost exclusively on the difference in elevation

between departure and arrival; the TPG for a flight departing and arriving at PIT was -202

mmH20, for a flight arriving in DEN was 1070 mmH20 and for a flight arriving in MIA was -612

mmH2O. Only the MIA destination was associated with the expression of barotitis media.

4. The most common cause of ET dysfunction is a constitutively impaired, active ET

opening mechanism. There, function tests document “normal” passive ET opening and closing

pressures, but an inability of the mTVP muscle to dilate the ET during swallowing (5). To

model these ears, we inputted normal values for the opening and closing pressures (and other

variables), but constrained the activity of the mTVP muscle by inputting a high RA value (1/RA ≈

0). Note that 1/RA is the airflow conductance of the ET during a swallow (i.e. the extent to

which the ET dilates during mTVP contraction) and does not necessarily reflect the airflow

conductance resulting from applied pressure differentials or the other passive properties of the

ET. Figure 5c shows the dynamics of the ME-cabin pressure gradient for the three simulated

flights. During ascent, the developing positive ME-cabin pressure gradient is repeatedly reduced

to the value of PC’ as the ET is passively opened at PO’. No barotrauma is experienced during this

phase of flight. The residual gradient ( PME-Cabin = PC’) is slowly reduced during flight by

transMEM N2 exchange. However, the developing negative ME-cabin gradient during descent
19
cannot be alleviated by muscle assisted ET openings leading to TPGs of –1731, -2226, and -486

mmH2O for landings at PIT, MIA and DEN, respectively. All underpressures are of sufficient

magnitude to provoke barotitis media and the former two are expected to provoke

baromyringitis.

The results of this simulation is not applicable to ears that are test positive for the

Valsalva maneuver wherein large NP pressure gradients are generated by closed nose/mouth

forced exhalations. If the generated NP-ET pressure gradient is sufficient to passively open the

ET, NP gas is transferred to the ME and the ME pressure is increased (See EQ 9). On descent,

repetition of this maneuver can, like the effect of swallowing for the “normal” ET, maintain near

ambient ME pressures, establish near zero mmH2O TPGs and prevent barotrauma.

5. The majority of persons who fly do not exhibit these extreme forms of ET dysfunction

but rather exhibit a graded series of active ET opening efficiencies. For example, studies

comparing children with adults or persons with and without a history of otitis media document

similar passive ET properties among all groups, but less efficient active ET openings in the

former groups (5, 8, 12). In our model, this variability in active opening efficiency can be

represented by varying RA. Figure 6a shows the simulated ME-cabin pressure gradient during

the course of a PIT-MIA flight for an ear with normal and one with compromised mTVP induced

ET openings (RA = 2 and 20 mmHg/cc/min; other parameters = “normal” values). The larger RA

value limits transET flow at each opening, compromises the ability of the ET to regulate ME

pressure and leads to a negative TPG sufficient to precipitate barotitis on landing.

6. ME volume shows a growth related increase (attributable primarily to expansion of

mastoid volume), an effect that is stunted or delayed in ears with poor ET function and/or a

history of otitis media (14, 37, 38). This observation causally links poor ET function to small

ME volumes. Figure 6b shows the simulated ME-cabin pressure gradient for ears with

compromised mTVP assisted ET openings (RA = 20 mmHg/cc/min) and large and small ME

20
volumes (+/- 50% baseline VME; other parameters = “normal” values). The smaller ME volume

(VME = 4.4 mL) buffers the effect of the compromised mTVP function on ME-Cabin pressure

deviations and prevented the barotitis documented for the larger volume ME (VME = 13.1 mL).

From these observations, the ability to maintain a near zero mmH2O ME-cabin pressure

gradient depends on the relative magnitudes of volume gas supply and demand. In the absence

of active, pressure-driven ET openings (e.g. Valsalva maneuver), supply is a function of mTVP

ET opening efficiency (proportional to SfTA/RA) while demand is a function of both the

difference in cabin pressure at effective cruising and landing altitudes (maximum ∆P to be

equilibrated) and ME volume (moles of gas required to equilibrate that ∆P). Figure 7a

summarizes this relationship for simulated PIT-MIA flights by plotting the TPGs for ears with

constant Sf and TA but different RA and VME values. There, low RA (< 4 mmHg/cc/min) allows

for the exchange of sufficient gas volumes to prevent both expressions of barotrauma over all

reasonable VME (< 16mL). In contrast, buffering against barotrauma for increasing ME volumes

was decreased with increasing values of RA.

7. In ears with a history of disease, changes in the TM are observed frequently (16). These

include increases in TM compliance which is termed atelectasis. In the extreme, TM retraction

can displace the total volume of the tympanum resulting in a ME volume restricted to that of the

mastoid. As noted, the magnitude of ME-ambient pressure deviations in ears with compromised

mTVP function can be buffered by TM displacement volume (See EQ 24). Figure 7b shows the

simulated TPG values for a ME with compromised mTVP function (RA = 8, other parameter

values=“normal”) as a function of both TM stiffness ( =1/CTM) and VME (with ∆VTM max <

tympanum volume = 1mL) after a PIT-MIA flight. The plot demonstrates the expected effect of

changing the VTM/VME ratio on ME-cabin pressure gradients. Specifically, greater values

are associated with lesser TPG values and the magnitude of this effect is greater for larger VME.

21
Conversely, hyper-compliant TMs ( < .14 “normal” ) protected the ME from barotitis

media over all reasonable VME.

Finally, we examined the effect of flight duration on TPG by comparing the predicted

TPG values for PIT-MIA (170 min) and PIT to London, UK (533 min), destinations with similar

elevations (Table III). For all ME function/structure configurations, the TPGs for the two flights

were similar. Because the major difference between these flights is the duration of cruising at

fixed altitude, any effect of flight duration will be driven by the rate of transMEM N2 exchange,

a process that was previously measured to be extremely slow (20, 22).

22
Discussion
Unlike previous descriptions that focused only on a categorical representation of ET

function (poor/good ET opening), our model of ME pressure regulation during flight is founded

on mathematical descriptions of the physiology underlying gas transfers between the ME and all

adjacent compartments. Calibration of the model parameters was done using published data for

disease-free ME’s, and thus, this description is not applicable to the ME with extant otitis media

or MEM inflammation. Specifically, those conditions: 1) introduce additional system

compartments (e.g. effusion), 2) change the capacitances of existing compartments (e.g. increase

MEM volume at the expense of ME volume), and 3) affect the exchange parameters for

transMEM gas transfers (e.g. increase MEM blood flow) (1). None-the-less, our model does

have broad applicability to the “disease-free” ME and to MEs expressing the predispositions

(e.g. poor mTVP function) and/or sequelae (e.g. altered TM compliance, reduced mastoid

volume) of those conditions. Moreover, by including continuous measures of relevant

parameters, our model realistically maps disease expression onto the known continuum of

underlying ET dysfunctions.

An important test of any model is its predictive accuracy with respect to describing and

explaining well-established observations. For ME barotrauma, these include the previously

documented increased risk associated with young age and nasal inflammation (concurrent colds

or nasal allergy). Our model is capable of representing and explaining these effects by

incorporating the changes in the contributing parameters measured by experiment. For example,

the age effect is explicable by the established improvement in mTVP functional efficiency

(modeled as progressively decreasing RA) with advancing age (8, 9) and the effect of nasal

inflammation is mediated by intraluminal venous engorgement (modeled as a greater PET) (17).

These explanatory analyses can be extended to include the effects of preventative treatments

23
such as nasal decongestants (17, 39) that act by decreasing tissue inflammation (decreased PET)

or of less well established interventions such as bottle-feeding of infants during descent (7)

where the associated jaw movements initiate mTVP activity (greater Sf) and/or reduce ET tissue

pressure (lesser PET).

Earlier descriptions of barotrauma during airflight usually did not discriminate between

barotitis media and baromyringitis in reporting results. As discussed above, these expressions

have different underlying causes with the former resulting from a moderate, positive MEM-ME

pressure gradient and the latter resulting from large positive or negative ME-cabin pressure

gradients. Consequently, baromyringitis can be experienced throughout flight and is usually

associated with signs of TM damage and symptoms of ear-fullness and pain, but barotitis media

develops during descent and, in the absence of baromyringitis, is often unrecognized by the

traveler. By considering both expressions, our model predicts post-flight ME barotrauma that is

and is not perceived by the traveler and, by consequence, recorded as an event in the compilation

of prevalence reports (49).

Perhaps the most important feature of our model is the demonstration of potential

buffering mechanisms that modify or prevent disease expression in ears with constitutively or

situationally impaired ET function. For example, we showed that for ears with a blocked ET (by

enlarged adenoids, nasopharyngeal carcinoma, nasal inflammation due to a cold/allergy or other

conditions), high positive pressures and baromyringitis will develop on ascent to cruising altitude

for all flights but the development of barotitis media on descent will depend on the difference

between departure and destination altitudes. Likewise, for ears with poor ET function, a

protective effect is provided by a high TM volume displacement/ME volume ratio. Support for

the physiological relevance of these buffering mechanisms was provided in a recent paper (48)

that reported a low frequency of barotrauma in ears that were expected to have poor ET function

24
but also had pre-existing conditions that favored a hypermobile TM and small ME (mastoid)

volume.

Earlier descriptions of the pathogenesis of barotrauma focused primarily on ET function

and did not include these nuances. In that regard, tests of ET function were used to screen

candidates for service as pilots (29) and attempts have been made by industry to extend these

tests to the professional flight crews of commercial airlines. Our results suggest that, while good

ET function is highly predictive of disease free flight, poor function only defines an increased

risk of flight-induced barotrauma. This distinction has important implications to interpreting the

results of ET function screening where failure to repeatedly open the ET during swallowing or to

transfer NP gas to the ME during Valsalva can be career limiting.

In conclusion, we present a physiological model of barotrauma development for “normal”

MEs during flight. The presented model simulates the empirical data for experiments conducted

on pilots in a pressure chamber and explains past observations with respect to risk assessments.

Also, our results identified diverse physiological and anatomical parameters that interact in

effecting adequate and abnormal ME pressure regulation during flight. This underscores the

importance of considering contextual relationships in predicting the susceptibility of a given ME

to barotrauma.

25
Acknowledgements

This study was funded in part by a Grant the NIH (NIDCD 01250) and by support from

the 2003 NASA Space Grant Fellowship. We would like to thank our many colleagues at the

University of Pittsburgh for their clinical insights and technical comments.

26
List of Figures
Figure 1. A cartoon illustrating the pathways for ME gas exchange (a), the factors contributing

to passive (via pressure induced flow past a compressing balloon) and active (via mTVP muscle

contraction) ET function (b), and the effect of ME-ambient pressure gradients on TM

displacement (c). See text for details.

Figure 2. Component forces acting on the ET during muscle-assisted openings. The lumen

remains open until the force of the surrounding tissue (a function of tissue pressure, PET and

contact area, AET) exceeds that exerted by the mTVP. Airflow through the ET is a function of

the cross-sectional area of the lumen, related to the width opened, shown as XET, and is

determined in part by the of mechanical stiffness of the lumen and surrounding tissue, shown as

KET.

Figure 3. Ambient pressure changes for the pressure chamber experiment described by Groth

(a), the corresponding experimental TM displacement data for a pilot during compression and

decompression and the model predictions for TM displacement (b). Model parameters, fitted to

the data were: PO’ = 292 mmH2O, PC’ = 136 mmH2O, RA= 7.5 mmHg/cc/min, CTM= 425

mmHg/mL, TA = 250 msec, and Sf = 33 swallows/ min.

Figure 4. Change in PCabin for 170 minute flights departing from Pittsburgh, PA and arriving at

Miami, FL (MIA), Pittsburgh (PIT), and Denver, CO (DEN)

Figure 5. Predicted ∆PME-cabin for a “normal” ME with parameters listed in Table II (a), a ME

with an obstructed ET (PO’ME-Tissue =2500, PO’NP-Tissue>2500 mmH2O) (b) and for a ME with poor

mTVP function (1/RA ≈ 0) (c). Barotrauma onset is specified by the dashed and dotted indicator

lines.

Figure 6. ∆PME-cabin as a function of flight time from PIT to MIA for MEs (Table II) with two

different RAs and fixed volume = “normal” (a) and for two different volumes at fixed RA

27
=20mmHg/cc/min (b). Barotrauma onset is designated by the dashed indicator lines. Other

parameters were set to “normal”.

Figure 7. PIT-MIA TPG as a function of RA with fixed TM stiffness coefficient = 179

mmH2O/mL (a) and as a function of TM stiffness coefficient ( ) with fixed RA = equal 8

mmHg/cc/min (b) over a range of ME volumes. Barotrauma onset is designated by the dashed

indicator lines. Sf was set to 1.1/hr during cruising and 20/hr during descent, other parameters

were set to “normal”.

28
Figure 1

Figure 2

29
 a

Figure 3

30
Figure 4

31
Figure 5

32
Figure 6

33
Figure 7

34
Table I: Average values of model parameters for “normal” MEs used in simulation

Parameter Description Mean Units Reference

VME ME volume 8.75 mL (14, 38)

VTMmax TM displacement volume 0.025 mL (14, 46)

ATM TM surface area 0.6 cm2 (15)

TM stiffness coefficient 179 mmH2O/mL (26)

Sf Active ET opening rate 5.2 openings/hr (53)

RA ET Active resistance 2 mmHg/cc/min (11)

TA ET active opening duration 0.25 sec (11)

PO’ME-ET ME Opening pressure 350 mmH2O (51)

PO’NP-ET NP Opening pressure 600 mmH2O (13)

PC’ Closing pressure 100 mmH2O (51)

35
Table II: Initial gas species partial pressure in the ME (24), NP (34), VB and Trans-MEM

Exchange Constants (21)

Gas Species Partial Pressure (mmHg) Time-Constant

ME NP VB Rate (min-1)

Oxygen 40 112 45 0.008

Carbon Dioxide 46 32 46 0.16

Water Vapor 47 47 47 0.32

Nitrogen Balance Balance Balance 0.0008

36
Table III: Elevations and ambient pressures for airports and airplane cabin (35).

Location Elevation (ft) Ambient Pressure (mmHg)

Pittsburgh, Pa 1204 730

Miami, FL 8 760

Denver, CO 5431 630

London UK 80 758

Location Equivalent Elevation (ft) Ambient Pressure (mmHg)

Airplane Cabin 8000 577

37
Table IV: Glossary
Symbols Descriptions
Pressures
PME Total ME pressure
PMEi ME partial pressure (O2, CO2, N2, H2O)
PCabin Total Cabin pressure
PCabini Cabin partial pressure (O2, CO2, N2, H2O)
PNP Total NP pressure
PNPi NP partial pressure (O2, CO2, N2, H2O)
PET Tissue pressure surrounding ET lumen
PIET ET intra-lumen pressure
PAMB Ambient pressure
PVAS Vascular pressure
PME-Cabin ME –Cabin pressure differential
PME-NP ME- NP pressure differential
Volumes
VME Middle ear airspace volume
Vmas Mastoid volume
Vtyp Tympanum volume
ET passive opening
Po’ ET opening pressure (ref AMB)
PME-ETo ME-side opening pressure (absolute)
PME-ETo’ ME-side opening pressure (ref AMB)
PNP-ETo NP-side opening pressure (absolute)
PNP-ETo’ ME-side opening pressure (ref AMB)
Pc’ ET closing pressure (ref AMB)
AME’ ET contact area from ME
ANP’ ET contact area from NP
TM displacement
ATM TM cross-sectional area
CTM TM compliance
XTM TM linear displacement
ET active opening
FET ET “closing” force
FST ET Intra-lumen surface tension force
FTVP Force exerted by mTVP on ET lumen
CET ET compliance
XET ET mediolateral lumen width
QET transET volume gas flow
RA ET active resistance
TA ET opening time
Sf Swallowing frequency
Misc
ki Species specific transMEM exchange time-
constants (O2, CO2, N2, H2O)

38
References
1. Alper CM, Doyle WJ, and Seroky JT. Higher rates of pressure decrease in inflamed
compared with noninflamed middle ears. Otolaryngol Head Neck Surg 121: 98-102, 1999.
2. Armstrong H and Heim J. The effect of flight on the middle ear. JAMA 109: 417-421,
1934.
3. Becker GD and Parell GJ. Barotrauma of the ears and sinuses after scuba diving. Eur
Arch Otorhinolaryngol 258: 159-163, 2001.
4. Bluestone CD. Eustachian tube function: physiology, pathophysiology, and role of
allergy in pathogenesis of otitis media. J Allergy Clin Immunol 72: 242-251, 1983.
5. Bluestone CD and Doyle WJ. Anatomy and physiology of eustachian tube and middle
ear related to otitis media. J Allergy Clin Immunol 81: 997-1003, 1988.
6. Brown TP. Middle ear symptoms while flying. Ways to prevent a severe outcome.
Postgrad Med 96: 135-137, 141-132, 1994.
7. Byers PH. Infant crying during aircraft descent. Nurs Res 35: 260-262, 1986.
8. Bylander A. Comparison of eustachian tube function in children and adults with normal
ears. Ann Otol Rhinol Laryngol Suppl 89: 20-24, 1980.
9. Bylander A. Function and dysfunction of the eustachian tube in children. Acta
Otorhinolaryngol Belg 38: 238-245, 1984.
10. Cantekin EI, Bluestone CD, Saez CA, Doyle WJ, and Phillips DC. Normal and
abnormal middle ear ventilation. Ann Otol Rhinol Laryngol Suppl 86: 1-15, 1977.
11. Cantekin EI, Doyle WJ, and Bluestone CD. Comparison of normal eustachian tube
function in the rhesus monkey and man. Ann Otol Rhinol Laryngol 91: 179-184, 1982.
12. Cantekin EI, Saez CA, Bluestone CD, and Bern SA. Airflow through the eustachian
tube. Ann Otol Rhinol Laryngol 88: 603-612, 1979.
13. Chan KH, Cantekin EI, Karnavas WJ, and Bluestone CD. Autoinflation of eustachian
tube in young children. Laryngoscope 97: 668-674, 1987.
14. Cinamon U and Sade J. Mastoid and tympanic membrane as pressure buffers: a
quantitative study in a middle ear cleft model. Otol Neurotol 24: 839-842, 2003.
15. Cinamon U and Sade J. Tympanometry versus direct middle ear pressure measurement
in an artificial model: is tympanometry an accurate method to measure middle ear pressure? Otol
Neurotol 24: 850-853, 2003.
16. Devine JA, Forte VA, Jr., Rock PB, and Cymerman A. The use of tympanometry to
detect aerotitis media in hypobaric chamber operations. Aviat Space Environ Med 61: 251-255,
1990.
17. Dickson EDD, McGibbon JEG, and Campbell ACP. Acute Otitic Barotrauma -
Clinical findings, mechanism and relationship to the pathological changes produced
experimentally in the middle ears of cats by variations in pressure. J Laryngol Otol, 1943.
18. Doyle WJ. Mathematical model explaining the sources of error in certain estimates of the
gas exchange constants for the middle ear. Ann Otol Rhinol Laryngol 109: 533-541, 2000.
19. Doyle WJ. Mucosal surface area determines the middle ear pressure response following
establishment of sniff-induced underpressures. Acta Otolaryngol 119: 695-702, 1999.
20. Doyle WJ, Alper CM, and Seroky JT. Trans-mucosal inert gas exchange constants for
the monkey middle ear. Auris Nasus Larynx 26: 5-12, 1999.
21. Doyle WJ, Alper CM, Seroky JT, and Karnavas WJ. Exchange rates of gases across
the tympanic membrane in rhesus monkeys. Acta Otolaryngol 118: 567-573, 1998.
22. Doyle WJ, Seroky JT, and Alper CM. Gas exchange across the middle ear mucosa in
monkeys. Estimation of exchange rate. Arch Otolaryngol Head Neck Surg 121: 887-892, 1995.

39
23. Elner A. Normal gas exchange in the human middle ear. Ann Otol Rhinol Laryngol 85:
161-164, 1976.
24. Felding JU, Rasmussen JB, and Lildholdt T. Gas composition of the normal and the
ventilated middle ear cavity. Scand J Clin Lab Invest Suppl 186: 31-41, 1987.
25. Fitzpatrick DT, Franck BA, Mason KT, and Shannon SG. Risk factors for
symptomatic otic and sinus barotrauma in a multiplace hyperbaric chamber. Undersea Hyperb
Med 26: 243-247, 1999.
26. Gaihede M, Lildholdt T, and Lunding J. Sequelae of secretory otitis media: changes in
middle ear biomechanics. Acta Otolaryngol 117: 382-389, 1997.
27. Ghadiali SN, Banks J, and Swarts JD. Effect of surface tension and surfactant
administration on Eustachian tube mechanics. J Appl Physiol 93: 1007-1014, 2002.
28. Ghadiali SN, Banks J, and Swarts JD. Finite Element Analysis of Active Eustachian
Tube Function. J Appl Physiol, 2004.
29. Groth P, Ivarsson A, Nettmark A, and Tjernstrom O. Eustachian tube function in
selection of airmen. Aviat Space Environ Med 51: 11-17, 1980.
30. Groth P, Ivarsson A, Tjernstrom O, and White P. The effect of pressure change rate
on the eustachian tube function in pressure chamber tests. Acta Otolaryngol 99: 67-73, 1985.
31. Hamada Y, Utahashi H, and Aoki K. Physiological gas exchange in the middle ear
cavity. Int J Pediatr Otorhinolaryngol 64: 41-49, 2002.
32. Hamilton-Farrell M and Bhattacharyya A. Barotrauma. Injury 35: 359-370, 2004.
33. Hanna HH. Aeromedical aspects of otolaryngology. Aviat Space Environ Med 50: 280-
283, 1979.
34. Harell M, Mover-Lev H, Levy D, and Sade J. Gas composition of the human nose and
nasopharyngeal space. Acta Otolaryngol 116: 82-84, 1996.
35. Hunt EH and Space DR. The Airplane Cabin Environment. 1994.
36. Ingelstedt S. Physiology of the Eustachian tube. Ann Otol Rhinol Laryngol 85: 156-160,
1976.
37. Isono M, Murata K, Azuma H, Ishikawa M, and Ito A. Computerized assessment of
the mastoid air cell system. Auris Nasus Larynx 26: 139-145, 1999.
38. Koc A, Ekinci G, Bilgili AM, Akpinar IN, Yakut H, and Han T. Evaluation of the
mastoid air cell system by high resolution computed tomography: three-dimensional multiplanar
volume rendering technique. J Laryngol Otol 117: 595-598, 2003.
39. Lildholdt T, Cantekin EI, Bluestone CD, and Rockette HE. Effect of a topical nasal
decongestant on Eustachian tube function in children with tympanostomy tubes. Acta
Otolaryngol 94: 93-97, 1982.
40. Low WK, Lim TA, Fan YF, and Balakrishnan A. Pathogenesis of middle-ear effusion
in nasopharyngeal carcinoma: a new perspective. J Laryngol Otol 111: 431-434, 1997.
41. Miyazawa T, Ueda H, and Yanagita N. Eustachian tube function and middle ear
barotrauma associated with extremes in atmospheric pressure. Ann Otol Rhinol Laryngol 105:
887-892, 1996.
42. Morris MS. Tympanostomy tubes: types, indications, techniques, and complications.
Otolaryngol Clin North Am 32: 385-390, 1999.
43. Mover-Lev H, Priner-Barenholtz R, Ar A, and Sade J. Quantitative analysis of gas
losses and gains in the middle ear. Respir Physiol 114: 143-151, 1998.
44. Nguyen LH, Manoukian JJ, Yoskovitch A, and Al-Sebeih KH. Adenoidectomy:
selection criteria for surgical cases of otitis media. Laryngoscope 114: 863-866, 2004.
45. Parris C and Frenkiel S. Effects and management of barometric change on cavities in
the head and neck. J Otolaryngol 24: 46-50, 1995.
46. Sade J. The buffering effect of middle ear negative pressure by retraction of the pars
tensa. Am J Otol 21: 20-23, 2000.
40
 47. Sade J. On the function of the pars flaccida: retraction of the pars flaccida and buffering
of negative middle ear pressure. Acta Otolaryngol 117: 289-292, 1997.
48. Sade J, Ar A, and Fuchs C. Barotrauma vis-a-vis the "chronic otitis media syndrome":
two conditions with middle ear gas deficiency Is secretory otitis media a contraindication to air
travel? Ann Otol Rhinol Laryngol 112: 230-235, 2003.
49. Stangerup SE, Tjernstrom O, Harcourt J, Klokker M, and Stokholm J. Barotitis in
children after aviation; prevalence and treatment with Otovent. J Laryngol Otol 110: 625-628,
1996.
50. Swarts JD, Alper CM, Seroky JT, Chan KH, and Doyle WJ. In vivo observation with
magnetic resonance imaging of middle ear effusion in response to experimental underpressures.
Ann Otol Rhinol Laryngol 104: 522-528, 1995.
51. Swarts JD and Bluestone CD. Eustachian tube function in older children and adults
with persistent otitis media. Int J Pediatr Otorhinolaryngol 67: 853-859, 2003.
52. Swarts JD and Rood SR. The morphometry and three-dimensional structure of the adult
eustachian tube: implications for function. Cleft Palate J 27: 374-381, 1990.
53. Tideholm B, Carlborg B, and Brattmo M. Continuous long-term measurements of the
middle ear pressure in subjects with symptoms of patulous eustachian tube. Acta Otolaryngol
119: 809-815, 1999.
54. Weiss MH and Frost JO. May children with otitis media with effusion safely fly? Clin
Pediatr (Phila) 26: 567-568, 1987.
55. Westerman ST, Fine MB, and Gilbert L. Aerotitis: cause, prevention, and treatment. J
Am Osteopath Assoc 90: 926-928, 1990.

41

View publication stats