OTOLOGY

Transmeatal excision of pars tensa retraction pocketswith simultaneous ventilation tube insertion in children:a prospective study

Gabor Rath • Imre Gerlinger • Zsuzsanna Csakanyi •

Monika Sultesz • Veronika Gaal • Gabor Katona

Received: 16 August 2010 / Accepted: 1 February 2011 / Published online: 18 February 2011

� Springer-Verlag 2011

Abstract Grade II and III (according to Sade’s classifi-

cation) retraction pockets (RPs) in 40 ears of 30 children

were excised transmeatally with simultaneous ventilation

tube insertion. At a mean follow-up of 16.1 months, an

intact tympanic membrane and mild, grade I retractions not

requiring treatment were observed in 32 ears (80%). The

mean air-bone gap had decreased from 22.4 to 9.7 dB. Two

residual perforations and six recurrent grade II RPs were

detected, all these eight ears undergoing further surgery.

The second surgery solved both of the residual perforations

and four of the six grade II recurrences. The age of the

children did not influence the success rate, whereas bilat-

eral pathology, a higher grade and more extensive RPs

were associated with a higher rate of unsuccessful cases. A

staging system is introduced to ease the selection of can-

didates with a good prognosis for this procedure.

Keywords Retraction pocket � Tympanic membrane �Pars tensa � Ventilation tube � Children

Introduction

Retraction pockets (RPs) of the tympanic membrane (TM)

are frequently encountered by otologists. The underlying

disorder is a dysfunction of the Eustachian tube (ET),

leading to retraction and atrophy of the TM with loss of the

organized collagenous layer and effusion formation. Chil-

dren lacking this pressure-equalizing function are prone to

recurrent episodes of acute suppurative otitis media, which

additionally results in histological degeneration and atro-

phy of the lamina propria of the TM [1]. Habitual snifflers

may repetitively evacuate the middle ear through a patent

ET, which may end in the development of a RP [2].

In childhood, the tubal malfunction is caused by adenoid

vegetation in most of the cases and is likely to resolve

spontaneously over time. The RP can retract onto the incus

and the medial wall of the tympanic cavity, and may adhere

to the underlying structure. Such adhesion gives rise to

chronic pressure and may lead to erosion of the incus.

Some RPs are stable and self-cleansing. In other cases, the

produced keratin accumulates in the RP and cannot migrate

toward the outer ear canal through the mouth of the RP,

and a cholesteatoma develops.

Pars tensa RPs have been classified by several authors

(Table 1) [3–6]. The symptomatology includes fullness of

the ear, otorrhoea, a hearing loss and recurrent episodes of

acute purulent otitis media. Some cases of RPs are

asymptomatic. Surgical procedures with the aim of

improvement of the ET function, such as adenoidectomy,

may be helpful. The treatment of choice can include

watchful waiting, nasal decongestants, forced middle ear

insufflation and various surgical procedures [4]. Creation

of a perforation by excision of the RP and appropriate

aeration of the middle ear cleft via ventilation tube (VT)

insertion, first described by Sharp and Robinson [7], is

G. Rath (&)

Pediatric ENT Department, Medical School, Pecs University,

Pecs, Jozsef A. u. 7, Pecs 7623, Hungary

e-mail: gaborrath@freemail.hu

I. Gerlinger � V. Gaal

Department of Otorhinolaryngology and Head and Neck

Surgery, Medical School, Pecs University, Pecs, Hungary

Z. Csakanyi � M. Sultesz � G. Katona

Pediatric ENT Department, Heim Pal Hospital, Budapest,

Hungary

123

Eur Arch Otorhinolaryngol (2011) 268:1549–1556

DOI 10.1007/s00405-011-1521-2

considered to promote spontaneous TM repair. Our aim

was to evaluate the impact of different factors (age, bilat-

erality, gravity, and extent of the RPs) on the efficacy of

transmeatal excision of pars tensa RPs with simultaneous

VT insertion.

Methods

The classification of Sade was used to determine the

gravity of the RPs [3]. Thirty children presenting with

grade II or III pars tensa RPs during a 30-month period

were included in this prospective study. Bilateral RPs,

present in 10 children, were operated upon simultaneously.

A detailed preoperative history was taken, including

symptomatology, predisposing factors and previous medi-

cal or surgical treatment. The ears were observed under an

operating microscope. The site and grade of the RPs, the

state of the underlying incudostapedial joint, and the

presence of any TM or middle ear abnormality were noted.

In the event of a RP in the attic region, large RPs inac-

cessible by a transmeatal approach and the presence of a

cholesteatoma, we did not perform this procedure and the

children were excluded from the study.

Ventilation tubes (10 Aesculap and 30 Shah grommets)

were inserted into the anterior quadrants of the TM. Gentle

suction was applied to the retracted TM area to lift off the

medial wall of the tympanic cavity and/or the incudostape-

dial joint, and the pars tensa RPs were then excised with a

sickle knife and microscissors through a transmeatal

approach (Figs. 1, 2, 3). The operations were performed

under general anesthesia by four surgeons, in two tertiary

paediatric otolaryngology centres. The procedure was car-

ried out on a one-day basis. The postoperative healing of the

eardrum was followed and any persistent perforations, scars,

tympanosclerosis, cholesteatoma formation or further

developing RPs were noted. When necessary, repeat exci-

sion of the retracted areas was considered. Pre- and

postoperative pure tone audiometry (air and bone conduction

thresholds at 0.5, 1, 2, and 3 kHz) and preoperative tympa-

nometry were performed on all 30 children. The results were

assessed by means of microscopic examination, together

with audiometric evaluation, 1, 3, and 6 months postopera-

tively and then at 6-monthly intervals. Statistical analysis

(Mann–Whitney test, Fischer’s exact test) was carried out to

elicit whether age, or the extent and severity of the RP had

any influence on the final outcome of the procedure.

A staging system was introduced for patient selection

for this procedure: Sade grade II and grade III situations

were awarded scores of 1 and 2, respectively. An additional

score was given if the RP affected more than one quadrant.

The rate of success was evaluated in different stages.

Results

RPs were excised in a total of 40 ears (Table 2) in 30

children (14 male and 16 female). The average age at

operation was 7.2 years (range 3–14 years). The follow-up

ranged from 6 to 29 months, with a mean of 16.1 months.

The presenting symptoms were a hearing loss in 34 ears

and recurrent episodes of acute suppurative otitis media in

25 ears.

The severity of the RP was grade II in 28 and grade III in

12 cases. In 22 ears, the postero-superior quadrant alone, in

one case the postero-inferior quadrant alone, and in 15 cases

both posterior sites were affected. In two ears, both the

anterior-inferior and posterior quadrants were affected. The

long process of the incus was partially eroded, but still in

contact with the head of the stapes in four cases. No ossicular

discontinuity was noted. In 25 ears, concomitant middle ear

fluid was aspirated. The retracted area was successfully

excised and a grommet was inserted in all the ears. No

ossicular reconstruction procedures were attempted.

The average pre- and postoperative air-bone gaps

(averaged at 0.5, 1, 2, and 3 kHz) were 22.4 and 9.7 dB,

Table 1 Classification of the severity of retraction pockets by different authors

Grade Sade [3] Borgstein et al. [4] Characon et al. [5] Yung [6]

I Mild retraction TM atrophic, but not adherent

to middle ear structures

Mobile RP Mobile RP

II Retraction onto the

incudostapedial joint with or

without erosion

TM adherent to the promontory

only

Fixed and controllable RP

(totally visible)

Posterior RP adherent

to incus/stapes

III Retraction onto the promontory

without adhesion

TM adherent to incus and/or stapes Fixed and uncontrollable RP

(the deepest part cannot be seen)

Fixed and completely

atelectatic TM

IV Adhesion to the promontory TM adherent to ossicles with deep

RP, but without cholesteatoma

V Spontaneous perforation RP with cholesteatoma and/or

breakthrough

1550 Eur Arch Otorhinolaryngol (2011) 268:1549–1556

123

respectively (Table 3). The bone conduction thresholds

remained in the normal range in all children.

Thirty-eight of the 40 surgically created perforations

healed completely in a mean of 1.9 months (range 2 weeks

to 6 months); the remaining two decreased in size, but

failed to close (Fig. 4). The edges of the residual perfora-

tions were carefully wounded with microinstruments and

VTs were inserted into the holes under general anaesthesia

7 and 15 months postoperatively. One of the tubes remains

in place, whereas the other was extruded 6 months later,

leaving a continuous eardrum behind. Six severe RP

recurrences were noted postoperatively, all in the same

quadrants as before surgery. A second excision of the RPs

with grommet (5 Shah and 1 Aesculap) insertion was

performed in all six cases. No residual perforations were

detected and four TMs became free of RPs. The other two

patients underwent a tympanoplasty procedure with rein-

forcement of the newly retracted posterior quadrants. No

cholesteatoma formation, scarring or tympanosclerosis was

noted.

A successful outcome was defined as an intact TM

following the initial surgery, with no RP or merely the

development of a stable, grade I mild retraction of the pars

tensa that required only observation. With these criteria,

the procedures proved successful in 32 ears (80%). The

success rate was 75% (15 of 20 ears) in the bilateral and

85% (17 of 20) in the unilateral group. The eight unsuc-

cessful cases comprised two residual perforations and 6

grade II RP recurrences in the pars tensa region.

Adenoidectomy ? VT insertion, adenoidectomy alone

and VT insertion alone had been carried out as previous

surgical procedures in 24, 10, and 4 ears, respectively.

Similar success rates were achieved in ears with previously

undertaken adenoidectomy procedure (27/34 ears, 79.4%)

and in ears on which no adenoidectomy had been carried

out in the past (5/6 ears, 83.3%).

The average ages of the patients who exhibited grade II

recurrence of the RP and those with a successful outcome

were 6.75 years (4–10 years) and 7.0 years (3–14 years),

respectively. There was no significant difference between

the two groups (P = 0.842; Mann–Whitney test). Signifi-

cantly higher success rates were achieved in the grade II

than in the grade III group (26/28 vs. 6/12 ears; Fischer’s

exact test, P \ 0.01) and in the ears with RPs involving

only one quadrant as compared with those with RPs

extending to more than one quadrant (22/23 vs. 10/17 ears;

Fischer’s exact test, P \ 0.01). In ears with scores of 1, 2,

and 3, the success rate was 100% (21/21), 66.7% (6/9), and

50% (5/10), respectively (Table 4).

The overall success rate increased from 80% (32 of 40)

to 95% (38 of 40) after the second surgery.

Fig. 1 Grade II retraction pocket in the right postero-superior

quadrant, in contact with the long process of the incus

Fig. 2 The ventilation tube is placed into the antero-inferior quadrant

Fig. 3 The retraction pocket is excised

Eur Arch Otorhinolaryngol (2011) 268:1549–1556 1551

123

Discussion

In contrast with the pressure of the ambient air, the pres-

sure in the middle ear continuously decreases because of

the absorption of the gases by the mucous membrane of the

middle ear cleft. Meanwhile, equalization of this difference

through the ET takes place in stages. Low pressure may

develop due to the elastic resistance of the TM. In the event

Table 2 Sex, age in years at the time of the initial operation, severity

grade of retraction, number of quadrants affected, length of follow-up,

presence of intact TM postoperatively, presence of residual

perforation, grade of recurrent RP after initial surgery, performance

of second surgery and grade of recurrent RP after second surgery are

demonstrated

Patient Sex Age

(years)

Grade Quadrants

involved

Follow-up

(months)

Intact TM

postop

Res.

perf.

Postop

grade

Sec.

surg.

Grade after

Sec. surg.

1 M 8 II 1 15 I

2 M 8 II 1 6 ?

3 F 8 III 3 18 ? ?

4 M 10 II 2 6 ?

5 M 11 II 1 6 ?

6 F 6 III 3 6 ?

7 F 3 II 1 25 ?

8 M 9 III 2 6 ?

9 M 11 III 2 12 ?

10 M 8 III 1 18 ?

11 F 3 II 2 29 I

12 F 10 III 2 28 II ? II

13 M 4 II 1 28 ?

14 F 6 II 2 27 ?

15 F 7 II 1 27 ?

16 F 6 II 2 27 I

17 R M 4 III 1 26 II ?

17 L III 2 I

18 F 7 II 1 26 ?

19 R F 6 II 1 26 ?

19 L II 1 ?

20 M 7 II 1 25 ?

21 M 5 II 2 20 II ?

22 R F 8 III 2 16 II ? II

22 L III 2 I

23 R M 6 II 1 16 I

23 L II 1 I

24 R M 14 II 2 15 ? ? ? I

24 L II 2

25 F 5 II 1 11 ?

26 R M 7 II 1 9 ?

26 L II 1

27 R M 8 III 2 9 ? II ?

27 L III 2 II ?

28 R F 7 II 1 8 ?

28 L II 1 ?

29 R F 7 II 1 7 ?

29 L II 1 ?

30 R F 7 II 1 7 ?

30 L II 1 ?

R right, L left ear, Res. perf. residual perforation, Postop. grade grade of recurrent RP after initial surgery, Sec. surg. second surgery, grade after sec.surgery grade of recurrent RP after second surgery

1552 Eur Arch Otorhinolaryngol (2011) 268:1549–1556

123

of a partially or totally atrophied, relaxed TM, the

absorption of the gases results only in a reduction of the

volume of the air in the middle ear cavity, without the

development of a lower pressure, since the relaxed mem-

brane moves in a medial direction without resistance. As

the air pressure is equal on the two sides of the relaxed

drum, the membrane retains its medial position. As the

absorption of the gases is a continuous process and opening

of the tube does not alter the position of the relaxed TM,

the retraction goes on with all its well-known conse-

quences. Atrophy of the TM is caused mainly by a pro-

longed ET malfunction, which may improve. The period of

improvement is critical. After its development, atrophy

itself will cause a progression of the pathological process,

independently of the normal function of the ET [8]. This

explains why patients with a RP may give a normal

tympanogram.

The RP may disappear, remain unchanged, continue to

grow or, in a minority of the cases, form a cholesteatoma.

There does not appear to be any predictive sign as to which

of these possibilities will occur. A progressive process can

be distinguished from a stable one only through regular

follow-ups.

We achieved similar success rates in ears with previ-

ously undertaken adenoidectomy and in those on which

adenoidectomy had not been performed earlier. Adenoid-

ectomy has a beneficial effect on the ET function in a

majority of children with adenoid hypertrophy. Disap-

pearance of a RP after adenoidectomy is not uncommon.

However, if adenoidectomy is performed when the TM is

already atrophied, the atrophy itself causes progression of

the pathological process independently of the normal

function of the ET. The volume reduction caused by the

resorption of gases is not followed by a reduction of the

pressure in the tympanic cavity, as the atrophied TM has

already lost its elasticity. The deterioration of the RP in the

atrophied region continues.

It may be speculated that the improvement in ET

function caused by the adenoidectomy in a well-selected

Table 3 Number and percentages of ears in different air-bone gap

decibel (dB) ranges before and after initial surgery

Average air-bone

gap (dB)

Number of ears (%)

Preoperative Postoperative

0–10 0 19 (47.5)

10\ 16 (40) 18 (45)

B20

20\ 19 (47.5) 2 (5)

B30

30– 5 (12.5) 1 (2.5)

Fig. 4 The procedures

performed, the numbers of RPs,

residual perforations,

recurrences of RPs and intact

TMs. RP retraction pocket, TMtympanic membrane, VTventilation tube

Table 4 Numbers and percentages of successful cases and the

prognosis in relation to the given scores

Score Number of ears Successful cases Prognosis

Number Percentage

1 21 21 100 Excellent

2 9 6 66.7 Good

3 10 5 50 Acceptable

Grade II and Grade III situations correspond to scores of 1 and 2,

respectively. An additional score was awarded if the RP extended to

more than one quadrant

Eur Arch Otorhinolaryngol (2011) 268:1549–1556 1553

123

child with a RP and partial TM atrophy may indicate a

turning point and the weakened, yet still existing elasticity

leads to an intratympanic pressure decrease, air enters the

middle ear as the ET opens and the TM hopefully regains

its original normal position. Hence, any surgery aimed at

restoration of the ET function should be performed before

the TM loses its elasticity.

In this material, the postero-superior site was affected in

all but one ear (98%). Similar distributions were reported

by Blaney et al. (97%) and by Walsh et al. (91%) whereas

that described by Ars was lower (60%) [9–11]. The his-

topathological studies carried out by Ruah et al. [12]

demonstrated the persistence of mesenchyme (normally

resorbed at 2 months of age) in the posterior quadrants for

up to 3 years. There is merely poor elastin in this area,

leaving a weaker TM once this mesenchyme finally dis-

appears. The postero-superior is the largest of the quad-

rants, which justifies its greater deflection due to the lower

intratympanic pressure. In this region, the fibre distribution

of the lamina propria is less able to withstand pressure

differences, the tympanic sulcus decreases in depth until it

practically no longer exists and the annulus is also reduced

in calibre, which explains a less resistant insertion of the

TM [13]. Unfortunately, due to the close proximity to the

long process of the incus, which is the most vulnerable part

of the ossicular chain, postero-superior RPs may result in

bone erosion [14].

There are wide variety of management options for RPs.

The usual treatment modalities include watchful waiting,

nasal decongestants, forced middle ear insufflation, VT

insertion with or without excision of the rectracted area,

laser myringoplasty, tympanoplasty with or without rein-

forcement of the TM and cortical mastoidectomy [4, 9, 10,

15–19]. A VT may temporarily successfully substitute the

function of the ET until it falls out. Tympanoplasty pro-

cedures can be combined with VT insertion. CO2 [16],

holmium [17], and KTP [18] laser beams contract and

tighten the atelectatic segment of the TM. Couloigner et al.

[20] performed RP excision and reinforcement tympano-

plasty in 60 severe postero-superior RPs in children with a

tragal or conchal cartilage graft. During the mean follow-

up of 27 months, retractions requiring additional surgery

recurred in only 8% of the ears.

Nevertheless, cartilage reinforcement of the TM leads to

obscuring opacity, which may conceal any evidence of

cholesteatoma pearls should it develop behind the eardrum.

In cases of grade I–III RPs, reinforcement tympanoplasty

appears to be unnecessarily aggressive in childhood, when

middle ear aeration tends to become more efficient spon-

taneously over time. Previously reported closure rates after

tympanoplasty and permeatal excision were reasonably

comparable, in the intervals 54–96% [20–22] and 87–97%

[9, 23]. Transmeatal excision has several advantages over

tympanoplasty: it is quicker, simpler, less invasive, is

associated with less morbidity and can be performed on a

1-day basis. Bilateral RPs can be excised transmeatally in

the same sitting, while a bilateral major ear surgery with

extensive manipulations in close proximity to the ossicules

is not recommended. Postponement of the operation on the

other ear is time-consuming or may even be depressive for

the patient and, not irrelevantly, the additional general

anaesthesia contributes to the risks. The healed eardrums

showed no sign of scarring or tympanosclerosis in the

regenerated area. This suggests that even large traumatic

perforations extending to two or even three quadrants can

be treated conservatively. It is not uncommon that the

transparency of the drum decreases after tympanoplasty in

the grafted region. If an epithelial pearl develops from

squamous remnants in the tympanic cavity due to incom-

plete removal of the RP, it would be easier to identify it

through a transparent TM. Should the excision fail, tym-

panoplasty can still be performed afterwards. Nevertheless,

RPs adhering to the medial wall are extremely difficult to

remove completely transmeatally; there is a high risk of

tearing of the exceedingly thin TM that may lead to epi-

thelial pearl formation. We do not consider transmeatal

excision suitable for grade IV or V situations.

In principle, the excision of RPs in paediatric ears does

not require grafting, as the vast majority of the cases will

heal spontaneously [23]. Orji an Ayu reported that the rate

of spontaneous healing of a traumatic TM perforation

varies inversely with the age of the patient and the size of

the perforation [24]. Jeo concludes that the management of

a fresh TM perforation should be limited to cleaning the

traumatized ear and preventing infection [25]. However, it

is optional to cover the excised region with a sheet of

silicone or other material to guide the repair growth of the

epithelial layer of the eardrum. A group of 86 ears of 62

patients who had undergone simple excision of the atrophic

eardrum segment without any grafting procedure were

followed by Borgstein et al. [23] to determine the incidence

of spontaneous healing; 94.2% of the eardrums healed

spontaneously within 7 weeks. In 17 ears (19.8%), there

was a recurrence of the atelectasis. Yung [6] reported a

recurrence rate of 19% for smaller RPs. Very similarly, in

our series, 95% of the ears healed without residual perfo-

ration after the initial surgery, and the rate of recurrence of

higher than grade I retractions was 15%.

Intact TMs following initial surgery with no retraction

or with development of a stable, grade I retraction of the

pars tensa were found in 32 of 40 ears (80%), which is

comparable to the findings of others (67–91%) [9, 10]. We

achieved a mildly higher success rate in cases of unilateral

versus bilateral RPs (85 vs. 75%), and significantly higher

rates in ears with lower versus higher grade RPs (93 vs.

50%) and in cases involving one quadrant versus those

1554 Eur Arch Otorhinolaryngol (2011) 268:1549–1556

123

extending to more than one quadrant (96 vs. 59%). Thus, a

higher grade of RP, and the involvement of more than one

quadrant of the TM are presumed to indicate a more severe

pathology in the background. Srinavasan et al. [19]

reported that the age and the severity of the RP do not

significantly influence the final outcome, and they attained

a higher success rate in cases of bilateral RPs. Sharp and

Robinson found that recurrence was no more likely in

patients with bilateral disease [7], whereas Walsh et al. [10]

observed a higher recurrence rate in bilateral cases. A ET

dysfunction is presumed to resolve in time due to stiffening

of the tubal cartilage, widening and lengthening of the ET

itself and development of the palatal muscles. In our study,

however, similarly as experienced by other authors [7, 9,

23], the rate of a successful outcome after the initial sur-

gery was independent of the average age of the children.

Consequently, it may be speculated that the eardrum might

respond similarly in adulthood.

For lack of histological evidence, it is only hoped that

the newly formed TM in the previously excised area has

the classical three-layer structure. After transmeatal exci-

sion, the rate of recurrence of RPs is between 15 and 20%

[6, 23]. This suggests that a majority of the eardrums

regenerates to at least a certain degree, with a more or less

stable membrane. Nevertheless, this part of the TM com-

prises a weaker point; all six RP recurrences that we

observed were in the same quadrant as the original RP. It

may be speculated that, in the long run, the rate of RP

recurrence may be higher in this weaker area.

The hearing improvement was acceptable, as the aver-

age pre- and postoperative air-bone gaps (averaged at four

frequencies) were 22.4 and 9.7 dB, respectively. Walsh

et al. [10] experienced that the average air conduction

threshold measured at three frequencies decreased from 16

to 7 dB.

Though unsuccessful in many cases, re-establishment of

the function of the ET is of key importance, as a persistent

ET malfunction may lead to recurrence of the disease

postoperatively. After a period of malfunction, the re-

improved ET function itself will not cause the atrophied,

retracted TM to resume the normal position, but with

autoinsufflation the children may achieve this aim. In the

case of a tubal malfunction, however, autoinsufflation can

furnish only a temporary solution and does not have a long-

term beneficial effect on the ventilation of the middle ear

cleft as the absorption of gases remains continuous. A

properly inserted VT successfully substitutes the missing

ET function.

With our staging system children with grade II RPs

localized within one quadrant may have an excellent

prognosis, while those with grade II RPs extending to more

than one quadrant or grade III RPs extending to only one

quadrant may have a good prognosis after the transmeatal

excision and simultaneuous VT insertion (Table 4). The

treatment of pars tensa RPs remains a controversial area,

but in well-selected cases this procedure can be performed

with high chances of a good surgical and functional

outcome.

Acknowledgments The authors thank Laszlo Poto (Institute of

Bioanalysis, Pecs University, Hungary) for his kind help with the

statistical analysis.

Conflict of interest None.

References

1. Sade J (1993) Atelectatic tympanic membrane: histological study.

Ann Otol Rhinol Laryngol 102:712–716

2. Magnuson B (1981) The atelectatic ear. Int J Pediatr Otorhinol

3:25–35

3. Sade J (1979) The atelectatic ear. In: Sade J (ed) Secretory otitis

media and its sequelae. Churchill-Livingstone, New York,

pp 64–88

4. Borgstein J, Gerritsma T, Bruce I, Feenstra L (2009) Atelectasis

of the middle ear in pediatric patients: safety of surgical inter-

vention. Int J Ped Otorhinol 73:257–261

5. Characon R, Barthez M, Lejeune JM (1992) Spontaneous

retraction pockets in chronic otitis media medical and surgical

therapy. Ear Nose Throat J 71:578–583

6. Yung MW (1997) Retraction of the pars tensa—long-term results

of surgical treatment. Clin Otolaryngol 22:323–326

7. Sharp JF, Robinson JM (1992) Treatment of tympanic membrane

retraction pockets by excision. A prospective study. J Laryngol

Otol 106:882–886

8. Mink A, Bauer M (1985) Physiological and clinical significance

of atrophy of the tympanic membrane. Ful-orr-gegegyogyaszat

31:1–9

9. Blaney SPA, Tierney P, Bowdler DA (1999) The surgical man-

agement of the pars tensa retraction pockets in the child–results

following simple excision and ventilation tube insertion. Int J Ped

Otorhinol 50:133–137

10. Walsh RM, Pracy JP, Harding L, Bowdler DA (1995) Manage-

ment of retraction pockets of the pars tensa in children by exci-

sion and ventilation tube insertion. J Laryng Otol 109:817–820

11. Ars BM (1991) Tympanic membrane retraction pockets. Etiol-

ogy, pathogeny, treatment. Acta Otorhinolaryngol Belg

45:265–277

12. Ruah CB, Schachern PA, Papparella MM, Zelterman D (1992)

Mechanisms of retraction pocket formation in the pediatric

tympanic membrane. Arch Otolaryngol Head Neck Surg

118:1298–1305

13. Paco J, Branco C, Estibeiro H, Carmo DE (2009) The postero-

superior quadrant of the tympanic membrane. Otolaryngol Head

Neck Surg 140:884–888

14. Rath G, Bauer M, Pytel J, Vona I, Szanyi I, Lujber L, Gerlinger I

(2008) Ionomer cement for reconstruction of the long process of

the incus: the Pecs experience. Clin Otolaryngol 33:116–120

15. Cassano M, Cassano P (2010) Retraction pockets of pars tensa in

pediatric patients: Clinical evolution and treatment. Int J Ped

Otorhinol 74:178–182

16. Ostrowski VB, Bojrab DI (2003) Minimally invasive laser con-

traction myringoplasty for tympanic membrane atelectasis. Oto-

laryngol Head Neck Surg 128:711–718

Eur Arch Otorhinolaryngol (2011) 268:1549–1556 1555

123

17. Kurokawa H, Goode RL (1995) Treatment of tympanic mem-

brane retraction with the holmium laser. Otolaryngol Head Neck

Surg 112:512–519

18. Brawner JT, Saunders JE, Berryhill WE (2008) Laser myringo-

plasty for tympanic membrane atelectasis. Otolaryngol Head

Neck Surg 139:47–50

19. Srinivasan V, Banhegyi G, O’Sullivan G, Sherman IW (2000)

Pars tensa retraction pockets: treatment by excision and ventila-

tion tube insertion. Clin Otolaryngol 25:253–256

20. Couloigner V, Molony N, Viala P, Contencin P, Narcy P, Van

Den Abbeele T (2003) Cartilage tympanoplasty for poste-

rosuperior retraction pockets of the pars tensa in children. Otol

Neurotol 24:264–269

21. Halik JJ, Smyth GD (1988) Long-term results of tympanic

membrane repair. Otolaryngol Head Neck Surg 98:162–169

22. Rath G, Balazs K, Gerlinger I, Moricz P, Jarai T, Bauer M, Pytel

J (2004) Long-term audiologic follow-up of paediatric tym-

panoplasties. Ful-orr-gegegyogyaszat 50:367–372

23. Borgstein J, Stoop E, Halim A, Feenstra L (2008) The extraor-

dinary healing properties of the pediatric tympanic membrane: a

study of atelectasis in the pediatric ear. Int J Ped Otorhinol

72:1789–1793

24. Orji FT, Agu CC (2008) Determinants of spontaneous healing in

traumatic perforations of the tympanic membrane. Clin Otolar-

yngol 33:420–426

25. Jeo A (2002) An observational study of the management of

traumatic tympanic membrane perforations. J Laryngol Otol

116:181–184

1556 Eur Arch Otorhinolaryngol (2011) 268:1549–1556

123