Main Article

The Journal of Laryngology & Otology, 1 of 4.

# JLO (1984) Limited, 2009
doi:10.1017/S0022215109990806

Epithelial migration in the atelectatic tympanic membrane

I P TANG, N PREPAGERAN, R RAMAN, T SHARIZHAL

Abstract
Objective: To determine whether epithelial migration in the atelectatic tympanic membrane (secondary to
any pathology) occurs in a similar fashion to that in the normal (non-pathological) tympanic membrane,
by calculating and comparing the epithelial migration rate and pattern.
Study design: Prospective, non-randomised, casecontrol study. This study was a pilot study. We included
patients with an atelectatic pars tensa region of the tympanic membrane and a healthy contralateral
tympanic membrane (the latter used as the control).
Setting: Otorhinolaryngology out-patient clinic. The study was conducted from January 2006 to January
2008.
Intervention: A total of 40 patients (80 ears) were included based on their otoscopic appearance and
tympanography findings. All patients ears were examined under an operating microscope. A dot of
methylene blue dye was applied in the immediate vicinity of the umbo of the atelectatic tympanic
membrane, and in the same position on the tympanic membrane of the control ear. Patients were then
followed up weekly until the dye dot had migrated to reach the annulus.
Main outcome measures: Direction and rate of epithelial migration.
Results: In the atelectatic tympanic membranes, epithelial migration proceeded from the centre of the
retraction pocket toward the annulus in a lateral, radiating manner. Thirty-three (82.5 per cent) of the 40
patients showed a similar migratory pattern in both the study and control ears. Sixty per cent of the
retracted tympanic membranes showed no migration in the first week after methylene blue staining. After
the first week, the migration rate was nearly constant from the umbo towards the periphery, in both the
study and control ears. The mean daily epithelial migration rate in the study and control ears was 62.6
and 64.7 mm/day, respectively; however, this difference was statistically insignificant ( p 0.202,
independent samples t-test).
Conclusion: There is no difference in the epithelial migration rate or pattern, comparing atelectatic
tympanic membranes and normal tympanic membranes. However, there may be a significant difference
in the epithelial migration rate with increased grades of tympanic membrane atelectasis.
Key words: Tympanic Membrane; Epithelium; Migration; Retraction Pocket

similarly or differently to that in the normal TM

remains unanswered. This study was conducted in
an attempt to answer this question.
The objectives of this study were to characterise
epithelial migration in the atelectatic TM, and to
compare the rate and pattern of epithelial migration
in the atelectatic TM versus the normal TM.

Introduction
Migration of tympanic membrane (TM) epithelia has
been known for more than 100 years, and its contribution to the physiological requirements of the ear
canal is generally accepted.1 This process is important to remove keratin debris and foreign bodies
from the TM and external auditory canal, in order
to propagate the sound wave more efficiently.
However, all previous studies of TM epithelial
migration rate and pattern have focused on the
normal TM, perforated TM, open mastoidectomy
cavity or postmyringoplasty TM.
To the best of our knowledge, there have been no
previous studies of epithelial migration in the atelectatic TM. Therefore, the question of whether epithelial migration in the atelectatic TM proceeds

Methods and materials

This was a prospective, non-randomised, case
control study. This study was a pilot study. Patients
were selected from the otorhinolaryngology clinic
of the University of Malaya Medical Centre. The
study was conducted from January 2006 to January
2008. The study group comprised patients with an

From the Department of Otorhinolaryngology, Faculty of Medicine, University of Malaya, Kuala Lumpur, Malaysia.
Accepted for publication: 20 April 2009.
1

atelectatic pars tensa region of the TM in one ear and

a normal TM in the other ear (these latter ears were
used as the control group), as determined by
otoscopic examination and tympanography findings.
Tympanic membrane atelectasis was graded using
Sades system, based on the severity of retraction.2
Thus, the study group was graded as: grade I (i.e.
slight retraction of the TM over the annulus); grade
II (the TM touches the long process of the incus);
grade III (the TM touches the promontory); or
grade IV (the TM is adherent to the promontory).
The purpose and procedure of the study were
explained thoroughly to the patient, to ensure their
full understanding. Following this, written consent
was obtained from all patients. The study was
approved by the medical ethics committee of the
University of Malaya Medical Centre.
Inclusion criteria
The inclusion criteria of the study were as follows: (1)
atelectatic pars tensa region of the TM, secondary to
any pathology, to any grade (i.e. grades I IV, Sade
classification), with a normal contralateral TM, as
determined by otoscopy and tympanography; (2)
normal, dry external auditory canals bilaterally; (3)
a type C tympanogram in the atelectatic TM and a
type A tympanogram in the normal TM; and (4) no
planned surgical intervention, and no history of
allergy to methylene blue dye.
Study instruments
The following instruments were used for the study:
basic demographic data record forms; diagrammatic
records of epithelium migration pattern and rate
measurement; an operating microscope; a micrometer;
methylene blue dye; and an angled, malleable wire.
Procedures
The selected patients ears were examined using an
operating microscope. A malleable wire was used
as an applicator; this wire was dipped in methylene
blue and then touched onto the epithelium in the
immediate vicinity of the umbo, to create a dot
stain. The patients were then followed up by the
investigator at weekly intervals until the dye dot
had migrated to reach the annulus. The distance of
migration was measured with a micrometer inserted
into the eyepiece of the operating microscope. A
schematic diagram with serial recording was made
to document the migratory pattern for each TM.
The daily migratory rate was calculated as the distance between the umbo and the annulus, divided by
the total days taken for migration.
Results
Forty patients from the otorhinolaryngology clinic
were included in the study, with 40 retracted TMs
and 40 normal TMs. Patients ages ranged from 11
to 71 years, with a mean of 36.58 years. There were
21 (52.5 per cent) males and 19 (47.5 per cent)
females. The ethnic distribution of the group was

I P TANG, N PREPAGERAN, R RAMAN et al.

45 per cent Malay, 30 per cent Chinese and 25 per

cent Indian.
Tympanic membrane retraction was grade I in nine
(22.5 per cent) TMs, grade II in 21 (52.5 per cent),
grade III in nine (22.5 per cent) and IV in one (2.5
per cent). Tympanic membrane retraction was due
either to eustachian tube dysfunction secondary to rhinosinusitis (72.5 per cent), nasopharyngeal carcinoma
(12.5 per cent) or unknown causes (15 per cent).
Of the atelectatic TMs, epithelial migration was posterosuperior in 18 (45 per cent), anterosuperior in 12
(30 per cent), posteroinferior in four (10 per cent),
anteroinferior in three (7.5 per cent) and anterior in
three (7.5 per cent) (Table I). Of the control TMs, epithelial migration was posterosuperior in 17 (42.5 per
cent), anterosuperior in nine (22.5 per cent), anterior
in seven (17.5 per cent), anteroinferior in three (7.5
per cent), posteroinferior in three (7.5 per cent) and
posterior in one (2.5 per cent) (Table I). Thirty-three
(82.5 per cent) of the 40 patients showed a similar
direction of migration in their study and control TMs.
Both study and control TMs showed a migration
pattern from the umbo towards the annulus and
canal wall. Sixty per cent of the retracted TMs and
35 per cent of the control TMs showed no migration
during the first week after methylene blue staining.
Thereafter, the rate of migration from the umbo
towards the periphery was nearly constant in both
the study and control TMs, assessed two to three
weeks after dye application.
In the retracted TMs, the mean daily migration
rate was 62.6 mm/day (range 50 78.3 mm/day). In
the control TMs, the mean daily migration rate was
64.7 mm/day (range 50 80.5 mm/day) (Table II).
However, this difference in mean daily migration
rate was not statistically significant ( p 0.202, independent samples t-test).
The mean daily migration rates for the different
grades of TM atelectasis were: grade I, 60.48 mm/
day; grade II, 63.66 mm/day; grade III, 63.17 mm/
day; and grade IV, 54.60 mm/day (Table III).
However, due to the small numbers in the different
grading groups, it was difficult to make an accurate
statistic comparison of the migration rate differences
between the different grades.
Discussion
In 1877, Burnett was the first to suggest the existence
of epidermal migration in the human ear drum.3
TABLE I
MIGRATION PATTERN IN STUDY AND CONTROL GROUPS

Pattern

Study
(n (%))

Control
(n (%))

Similar
(n (%))

Anterior
Anterosuperior
Superior
Posterosuperior
Posterior
Posteroinferior
Inferior
Anteroinferior
Total

3
12
0
18
0
4
0
3
40

7 (17.5)
9 (22.5)
0
17 (42.5)
1 (2.5)
3 (7.5)
0
3 (7.5)
40 (100)

3
9
0
16
0
2
0
3
33

(7.5)
(30)
(45)
(10)
(7.5)
(100)

(9.1)
(27.3)
(48.5)
(6)
(9.1)
(82.5)

EPITHELIAL MIGRATION IN ATELECTATIC TYMPANIC MEMBRANE

TABLE II
DAILY EPITHELIAL MIGRATION RATE IN STUDY AND CONTROL GROUPS

Migration rate (mm/day)

Range
Mean


Study

Control

50.0 78.3
62.6

50.0 80.5
64.7

2.1 mm/day difference.

This concept was also noted by Burk in 1880 and

Blake in 1882.3,4
Stinson was the first to study epithelial migration in
the TM, in 1936, after the incidental observation that
a piece of reed accidentally embedded in the TM was
gradually carried from the drumhead to the posterior
canal wall and hence to the exterior.5 Later, Stinson
defined five laws of epithelial escalation. Briefly
summarised, these were: (1) the epidermis migrates
on its original plane from the anterior to the posterior margin of the membrane and then along the
canal wall to the exterior; (2) the epidermis over
the pars flaccida fans out over the canal wall; (3)
the epidermis over the processus brevis proliferates
in whorls to the superior part of the canal; (4) the
progress is most rapid in the posterior inferior quadrant; and (5) the epidermis does not proliferate from
canal to drumhead.
Magnoni concluded in 1938 that epithelial growth
was radial from the umbo, but agreed that epithelium
never migrated from canal wall to drumhead.
However, Magnonis results were neglected in the
English language literature.3
In 1963, Litton was the first to estimate the rate of
epithelial migration in the human TM, at approximately 0.05 mm/day, and also noted that an ink dot
placed at the umbo disintegrated and proceeded
outward in a radial fashion.6
Alberti repeated Littons work in 1964, in a study
of 62 human TMs, estimating an epithelial migration
rate of 0.07 mm/day.3 Both Litton and Alberti postulated that the rate of migration reduced steadily as
the distance from the umbo increased, and also that
the migration rate was proportional to the thickness
of the epithelial layer. The usual pattern of migration
was found to be centrifugal from the umbo, spreading out to all quadrants of the TM. There was no
bar to epithelial movement between the pars tensa
and the pars flaccida. The overall migration rate
was equivalent to the growth rate of a finger nail.3
Further studies have been carried out by Boedts
and Kuijpers in 1978, ODonoghue in 1983, Michaels
and Soucek in 1989, Ong et al. in 2007, and Deong
TABLE III

et al. in 2006; these authors have confirmed the

existence of TM epithelial migration and elucidated
its pathways.1,7 10
The current, pilot study demonstrates that
epithelial migration does occur in the atelectatic
TM. The marker dye dot was clearly observed to
migrate from the umbo of the TM to the annulus
and ear canal wall.
A very similar migratory pattern was observed in
both the study and control TMs. The most common
direction of epithelial migration was posterosuperior
(occurring in 45 per cent of the study TMs and 42.5
per cent of the control TMs), followed by anterosuperior migration (occurring in 30 per cent of study
TMs and 22.5 per cent of control TMs). None of
the study or control TMs showed any evidence of epithelial migration in the superior and inferior
directions.
The migration patterns observed in this study were
slightly different from those reported by Litton and
Alberti. These authors reported that the ink dot disintegrated and proceeded outwards in a radial
fashion, whereas, in the current study, the dye dot
disintegrated from the umbo and proceeded in a
particular direction and segment.3,6 This was most
probably because the dye dot had not been placed
in the exact centre of the umbo.
The observed epithelial migration rate was slower
near the umbo region (i.e. at the beginning of the
observation period). Some TMs showed no movement of the dye dot during the first week. In almost
all cases (in both study and control TMs), the epithelial migration rate increased once the dye dot
had migrated further out towards the periphery.
The observed mean daily migration rates for the
study and control TMs were 62.6 and 64.7 mm/day,
respectively. The mean rate difference was 2.1 mm/
day; however, this difference was statistically insignificant ( p . 0.05). The observed mean daily migration
rate was within the range documented by Litton and
Alberti.3,6

. This study aimed to determine whether

epithelial migration in the atelectatic tympanic
membrane (TM) (secondary to any pathology)
behaves in a similar fashion to that in the
normal, non-pathological TM
. Migration of dye dots was used to assess
epithelial migration
. There was no difference in epithelial migration
rate or pattern, comparing atelectatic and
normal TMs

DAILY MIGRATION RATE FOR DIFFERENT GRADES OF TM ATELECTASIS

Grade

Mean

Min

Max

Difference

I
II
III
IV
Total

60.48
63.66
63.17
54.60
62.60

53.40
50.00
50.00
54.60
50.00

64.10
78.30
74.00
54.60
78.30

10.70
28.30
24.00
0.00
28.30

Data represent mm/day. TM tympanic membrane; min

minimum; max maximum

Clinical observation indicates that TM atelectasis

and retraction pockets are unstable, dynamic
phenomena. The degree of retraction of the atelectatic drum fluctuates, over weeks and months and also
over hours during sleep.11 In a prospective study by
Luntz and Sade, 84 human, atelectatic ears were
examined at different times of day.12 They noticed
that over one-third of adolescent and adult TMs

were found to be inflated (usually even hyperinflated) following a nights sleep. It took these hyperinflated TMs an average period of 54 minutes to
return to their original, atelectatic state. In our
study, most of the cases had reversible atelectasis
(i.e. Sade grade I to III), and it could be argued
that many of these TMs reverted back to normal
during part of the day. Therefore, epithelial
migration may not be significantly impeded.
This study found an increased mean daily
migration rate with increased grading of TM atelectasis, especially grade IV. This difference may become
statistically significant if the number of TMs of
different grades is increased. Hence, further study
is required.
We acknowledge that the study had a number of
limitations. Firstly, it was difficult to apply the dye
dot in the immediate vicinity of the umbo. Secondly,
measurement was performed only at weekly intervals. Thirdly, there was a certain amount of unavoidable objective and subjective measurement error.
Finally, the number of TMs in each atelectasis
grade group was small, preventing accurate statistic
analysis.

I P TANG, N PREPAGERAN, R RAMAN et al.

2 Sade J. The atelectatic ear. In: Sade J, ed. Secretory Otitis

Media and its Sequelae. London: Churchill Livingstone,
1979;64 8
3 Alberti PWRM. Epithelial migration on the tympanic
membrane. J Laryngol Otol 1964;78:808 29
4 Blake CJ. The progressive growth of the dermoid coat of
TM. Am J Otolaryngol 1882;4:266 7
5 Stinson WD. Reparative process in membrane tympani.
Arch Otolaryngol 1936;24:600 2
6 Litton WB. Epithelial migration over tympanic
membrane and external canal. Arch Otolaryngol 1963;77:
2547
7 Boedts D, Kuijpers W. Epithelial migration on tympanic
membrane. An experimental study. Acta Otolaryngol
1978;85:248 52
8 ODonoghue GM. Epithelial migration on the tympanic
membrane of children. Int J Pediatr Otorhinolaryngol
1983;8:297 303
9 Michaels L, Soucek S. Development of the stratified squamous epithelium of the human TM and external canal: the
origin of auditory epithelial migration. Am J Anat 1989;
184:334 44
10 Ong CA, Prepageran N, Godbole S, Raman R. Epithelial
migration in open mastoidectomy cavities. Asian J Surg
2007;30:57 9
11 Sade J, Avraham S, Brown M. Atelectasis, retraction
pockets and cholesteatoma. Acta Otolaryngol 1981;92:
50112
12 Luntz M, Sade J. Daily fluctuations of middle ear pressure
in atelectatic ears. Ann Otol Rhinol Laryngol 1990;99:
2014

Conclusion
Based on the above results, we conclude that there is
no statistically significant difference in epithelial
migration rates and patterns, comparing atelectatic
and normal TMs. However, there may be a significant difference in epithelial migration rate with
increased atelectasis grading. As we only have one
patient with Grade IV TM. Therefore, we are unable
to analyse statistically. Further study is required.

Address for correspondence:

Dr Ing Ping Tang,
66, Lorong Setia Raja 12A1,
Taman Stutong Indah,
93350 Kuching,
Sarawak, Malaysia.

References
1 Deong KK, Prepageran N, Raman R. Epithelial migration
of the postmyringoplasty tympanic membrane. Otol Neurotol 2006;27:855 8

Dr I P Tang takes responsibility for the integrity of the

content of the paper.
Competing interests: None declared

Fax: 602422564
E-mail: ingptang@yahoo.com