Keywords
chronic otitis media - cholesteatoma - hearing outcome - air-bone gap - tympanoplasty
Introduction
Chronic otitis media (COM) is defined as an inflammation of the middle ear and mastoid
mucosa with more than 3 months of duration. In case of an intact eardrum, the most
frequent entity is otitis media with effusion (OME), whereas a chronic inflammation
with a central perforation is termed as chronic suppurative otitis media (CSOM) or
otitis media chronica simplex (OMCS). Intermittent or persistent otorrhea may be a leading symptom besides a moderate
hearing loss. The prevalence of an intact ossicular chain in cases of COM has been
reported in the literature to range from 72% to 90%.[1]
[2] Size matters regarding the perforation and its impact on the preoperative air-bone
gap (ABG). An almost linear correlation between the size of the perforation and the
ABG in cases of an intact and mobile chain was documented in a recent publication.[3] Chronic otitis media with cholesteatoma (Chole) is defined as skin and retention
of keratin in the middle ear and/or temporal bone with surrounding inflammatory reaction
and progressive bone resorption. Intermittent or persistent foul smelling otorrhea
combined with progressive hearing impairment are the leading symptoms. Due to the
aggressiveness of the disease, the prevalence of an intact ossicular chain is markedly
reduced, and it has been reported to range between 5.5% and 30%.[4]
[5]
[6]
[7]
[8]
[9] Most often, the preoperative ABG is quite variable and cannot be predicted, since
the Chole mass itself may transmit sound to the stapes footplate even in severely
impaired ossicles. We have recently presented our Chole staging system, coding the
extent of the disease, the integrity of the ossicular chain, the type of complications
(by the Chole) and a rating of mastoid pneumatization and ventilation.[10]
In both disease entities (COM and Chole) with an intact ossicular chain, the final
steps of surgery are equal and consist in the preservation of the ossicles and reconstruction
of the tympanic membrane. Our study addressed three research questions: 1) Is the
final outcome “normal” hearing with closure of the ABG? 2) Do patients with Chole
present a worse result than patients with “simple” perforations? 3) What is the impact
of the mastoid pneumatization/ventilation on the hearing outcome in cholesteatomatous
patients?
Materials and Methods
All patients operated at our tertiary referral center are entered prospectively into
an otology database (InnoForce, Ruggell, Liechtenstein). We retrospectively analyzed
the data of patients operated by two senior surgeons using the same surgical technique
in the period from 2010 to 2019. We only selected patients with the diagnosis of COM
or Chole and an intraoperatively intact and mobile ossicular chain. Patients with
insufficient follow-up (shorter than 3 months) or residual/recurrent disease were
excluded. The pre- and latest postoperative pure tone audiograms were analyzed for
the individual frequencies as well as the average between 0.5 kHz and 4 kHz. The degree
of pneumatization and ventilation and the location of the disease in Chole patients
was coded by the surgeon by applying the criteria of the ChOLE classification (https://chole.surgery) and evaluating the preoperative computed tomography (CT) scans. Most patients with
COM did not have a preoperative CT scan or tympanometric volume measurement, and therefore
could not be analyzed regarding their “Eustachian tube” function. The ENT Statistics
software Innoforce-cerative solutions (Industriestrasse 56, 9491 Ruggeli, Liechtenstein)
was used for the statistical analysis. The Mann–Whitney U test was used to compare
the preoperative, postoperative and the improvement of the mean ABG and isolated frequencies
among the different groups. The ranges of the mean ABGs and frequencies between 0.5 kHz
and 4 kHz (0.5 kHz, 1 kHz, 2 kHz, 3 kHz, and 4 kHz) were all considered for comparison.
The present study was approved by the local Ethics Committee under protocol number
2019–00914.
Results
The initial COM group consisted of 210 surgeries in 182 patients who underwent primary
tympanoplasty with temporalis fascia or tragal cartilage. Out of this group, 162 ears
had an intact ossicular chain (80%) and 77 were excluded because they did not meet
the inclusion criteria or due to insufficient follow-up. Finally, the group with COM
with intact ossicular chain involved 85 ears (43 left and 42 right ears, 50% each).
Overall, 38 (45%) patients were male, and 47 (55%) were female. The average age was
34.3 years at the time of the surgery (range 6.4 to 84.3 years).
In the second group with Chole, 279 patients (283 operated ears) with primary closed
cavity tympanomastoid surgery and tympanoplasty using temporalis fascia or cartilage
were initially enrolled. All of them had been classified using the ChOLE staging system.
A subgroup analysis ([Table 1]) of these patients examined the impact of the location (Ch1a to Ch4b), the status
of the ossicular chain (O0 to O4b) and the impairment of pneumatization and ventilation
(E0 to Ex).
Table 1
Overall classification of all cholesteatomas
|
Ch stage
|
N
|
%
|
E stage
|
n
|
%
|
|
Ch1a
|
112
|
39%
|
E0
|
75
|
26%
|
|
Ch1b
|
48
|
17%
|
E1
|
72
|
26%
|
|
Ch2a
|
28
|
10%
|
E2
|
126
|
44%
|
|
Ch2b
|
17
|
6%
|
Ex
|
10
|
4%
|
|
Ch3
|
34
|
12%
|
|
|
|
|
Ch4a
|
42
|
15%
|
|
Ch4b
|
2
|
1%
|
|
Total
|
283
|
100%
|
Total
|
283
|
100%
|
Only 53 operated ears presented with an intact ossicular chain, classified as O0,
and 9 patients were excluded due to insufficient follow-up. The final Chole group
consisted of 44 ears (21 left and 23 right ears) of 22 male (50%) and 22 (50%) female
patients. The average age at the time of surgery was 38.3 years (range 4.1 to 81.7
years) [Tables 1] & [2]. For consistency a minor change could be made: The extension of the disease among
the cholesteatomatous patients in this group was limited (Ch1 in 90%, versus 57% out
of the overall group), and they had better pneumatization and ventilation (E0 in 55%
versus 26%) than the patients with ossicular destruction ([Tables 1] and [2]).
Table 2
Classification of operated ears with cholesteatoma and an intact ossicular chain
|
Ch stage
|
N
|
%
|
E stage
|
n
|
%
|
|
Ch1a
|
30
|
68%
|
E0
|
24
|
55%
|
|
Ch1b
|
10
|
23%
|
E1
|
7
|
16%
|
|
Ch2a
|
1
|
2%
|
E2
|
9
|
20%
|
|
Ch2b
|
2
|
5%
|
Ex
|
4
|
9%
|
|
Ch3
|
1
|
2%
|
|
|
|
|
Total
|
44
|
100%
|
Total
|
44
|
100%
|
[Table 3] summarizes the differences between the two groups for the pre- and postoperative
audiograms. The mean preoperative ABG of 12 dB in the Chole group was significantly
better than the almost 20 dB of the COM group. The final ABG of the two groups was
11.2 dB and 10.8 dB respectively, and did not differ significantly anymore. The final
air conduction for the mean frequencies between 0.5 kHz and 4kHz was 25 dB for both
entities. Therefore, the overall hearing improvement was higher in COM patients. Analyzing
individual frequencies ([Table 4]), the Chole group had a better preoperative ABG in each one in comparison with the
COM group (p< 0.05). Postoperatively, they no longer differed from each other. The lowest ABG
was consistently found at 2 kHz, whereas the predominant ABG was at 0.5 kHz and 4 kHz.
A postoperative ABG < 20 dB is considered a successful outcome in most studies. In
the Chole group, 40 ears (93%) reached this goal, and 19 (43%) of those had an ABG < 10 dB.
In total, 2 (5%) patients remained with a postoperative ABG between 20 dB and 30 dB,
and 1 patient presented with a poor result (ABG > 30 dB). In the COM group 78 ears
(92%) remained within < 20 dB, and 45 (53%) had results < 10 dB. A total of 7 (8%)
patients had ABGs between 20 dB and 30 dB, and none of the patients in this group
had an ABG > 30 dB. The results are presented in the [Table 5].
Table 3
Mean pre- and postoperative hearing outcome
|
Chole group (n = 44)
|
COM group (n = 85)
|
p Value
|
|
Preoperative ABG (dB)
|
11.8
|
20
|
p
= 0.0000
|
|
Postoperative ABG (dB)
|
11.2
|
10.8
|
p = 0.8935
|
|
Improvement of ABG (dB)
|
0.7
|
9.2
|
p
= 0.0001
|
|
Preoperative AC (dB)
|
24.8
|
33.1
|
p
= 0,0007
|
|
Postoperative AC (dB)
|
25.4
|
25.7
|
p = 0.8597
|
|
Improvement of AC (dB)
|
-0.5
|
7.4
|
p
= 0.0004
|
Abbreviations: ABG, air-bone gap; AC, air conduction; dB, decibels.
Table 4
Pre- and postoperative air-bone gap single frequencies for both groups
|
Frequencies (kHz)
|
Chole group (dB)
|
COM group (dB)
|
p Value
|
Chole group (dB)
|
COM group (dB)
|
p Value
|
|
0.5
|
13.1
|
21.1
|
p
= 0.0015
|
12.6
|
10.6
|
NS
|
|
1
|
13.5
|
20.4
|
p
= 0.0016
|
11.9
|
10.3
|
NS
|
|
2
|
7.1
|
16.6
|
p
= 0.0000
|
4.4
|
5.0
|
NS
|
|
3
|
9.6
|
19.7
|
p
= 0.0000
|
9.8
|
9.4
|
NS
|
|
4
|
13.9
|
22.3
|
p
= 0.0021
|
13.9
|
17.6
|
NS
|
|
Preoperative air-bone gap
|
Postoperative air-bone gap
|
Abbreviations: Chole, cholesteatoma; COM, chronic otitis media; dB, decibel; kHz,
kilohertz; NS, not significant.
Table 5
Postoperative air-bone gap for both groups
|
Postoperative ABG (dB)
|
Chole group (n = 44)
|
COM group (n = 85)
|
|
0–10 dB
|
19 (43%; < 20dB)
|
45 (53%; < 20dB)
|
|
< 20 dB
|
40 (93%)
|
78 (92%)
|
|
20–30 dB
|
2 (5%)
|
7 (8%)
|
|
>30 dB
|
1 (2%)
|
0 (0%)
|
Abbreviations: ABG, air-bone gap; Chole, cholesteatoma; COM, chronic otitis media;
dB, decibel.
As part of the study, we evaluated the hearing outcome in the Chole group considering
the ChOLE classification. The individual stages are presented in [Table 2]. The size and location of the Chole had no impact on the final hearing outcome (p> 0.05), and there was no patient with a large extension of Ch4a or Ch4b. On the other
hand, the group with moderate to good pneumatization and ventilation (E0) achieved
a significantly better (p = 0.0413) postoperative ABG of 9.4 dB ([Table 6]) compared with the patients with reduced ventilation (E1) and a mean ABG of 14.7 dB.
The hearing outcome of the patients with sclerotic mastoids (E2) was almost to the
same as that of the E0 subgroup and did not show a significant difference to the E1
subgroup, due to the limited number of patients in both groups.
Table 6
Degree of pneumatization and hearing outcome in cholesteatoma patients
|
E0 (n = 24)
|
E1 (n = 7)
|
E2 (n = 9)
|
Ex (n = 4)
|
|
Preoperative ABG (dB)
|
10.7
|
12.5
|
11.1
|
15.4
|
|
Postoperative ABG (dB)
|
9.4
|
14.7
|
9.4
|
16.9
|
|
Improvement (dB)
|
1.3
|
-2.2
|
1.7
|
-1.5
|
Abbreviations: ABG, air-bone gap; dB, decibel.
Discussion
There are various forms of COM, and all of them – to a different degree – impair sound
conduction and, therefore, lead to a conductive hearing loss. Patients with COM with
and without suppurative episodes (CSOM) contact their physician to get a dry and stable
ear with an intact eardrum, and ask for improvement of their hearing hoping for “normal
hearing.” In our series, 80% of all COM patients present at surgery with an intact
and mobile ossicular chain. In a previous publication,[3] we confirmed that the size (and not the location) of the tympanic-membrane perforation
had a direct and almost linear impact on the preoperative hearing impairment in these
patients. The least severe effect was found at the middle-ear resonance frequency
of 2 kHz, and the most pronounced impact, at the low (0.5 kHz) and high (4 kHz) frequencies.
Even in cases of subtotal perforations, the preoperative ABG did not exceed 35 dB.
On the other hand, patients with middle-ear Choles contact their physician to completely
remove the disease, get rid of the foul smelling otorrhea, and hope for some hearing
improvement. Choles are much more aggressive. Considering the ChOLE classification,[10] the size and extension of the disease is generally correlated with a higher surrounding
bone reabsorption and ossicular erosion. Indeed, in our series, only 20% of all patients
presented with an intact ossicular chain. In comparison to the other cholesteatoma
patients, their extent of the disease was rather limited (stage-1 Chole in 80% or
35 out of 44 cases), and they had better pneumatization and ventilation on the preoperative
CT scans. Contrary to the COM patients, the extent of the Chole did not have a direct
impact on the preoperative ABG. The mean preoperative ABG of 12 dB in the Chole group
was significantly better than the almost 20 dB of the COM group, and this was also
true for the individual frequencies. In summary, the chance to encounter an intact
and mobile chain in cases of COM is of 80%, and it drops to less than 20% in patients
with limited cholesteatomatous ear disease, whereas the preoperative hearing may be
better in the Chole patients and is unpredictable.
At the end of the surgery, once the eardrum has been reconstructed, surgeons and patients
hope for optimal hearing improvement, and may even expect “normal” hearing in cases
of an intact and mobile ossicular chain and closed cavity setting. One might expect
that patients with Chole present with a worse outcome due to the severity of the inflammatory
disease. However, our results reveal some interesting findings. First, there was no
difference between the groups regarding the final postoperative ABG. More than 90%
had an ABG (0.5–4 kHz) lower than 20 dB (and 19 (43%) in the Chole and 45 (53%) in
the COM group within 10 dB) and only 3 patients in the Chole group had an ABG higher
than 20 dB. Reviewing the charts of these 3 patients, they had minor erosion of the
incus (but still with an intact incudostapedial joint at the first surgery) and 1
patient underwent an ossiculoplasty 2 years later, with hearing improvement. In the
COM group, 7 patients (8%) ended with an ABG >20 dB. Specifically analyzing these
cases, we did not find a consistent factor, but the presence of glue, a more severely
inflamed middle-ear mucosa or a postoperative eardrum retraction and atelectasis as
reasonable explanations. Secondly, the relative hearing improvement was better among
the COM patients, but this is due to the fact that the preoperative hearing was worse
and the final outcome, the same. Thirdly, we looked at the impact of pneumatization
and ventilation, which could be encoded for the Chole group using the ChOLE classification.
Unfortunately, we did not perform routine CT scans in the COM patients, and did not
record routinely the preoperative volume on tympanometry. Patients with reasonably
good ventilation (E0 on the ChOLE score) had a higher chance of having an intact chain
and a better hearing outcome, suggesting a positive effect of the function of the
Eustachian tube. However, a final statement cannot be made yet.
Interestingly, not many previous publications have addressed this research question.
A comparison is also limited, since some journals require the presentation of the
hearing data for the frequencies between 0.5 kHz and 3 kHz, and do not ask for data
regarding 4 kHz. It is known that the high frequencies (including 4 kHz) are more
important for speech understanding and the overall benefit to the patients. Our data
reveal that the closure of the ABG was rather limited at 4 kHz. Therefore, excluding
this important frequency would improve the surgeon's success rates, but does not honestly
summarize the benefit for the patient. So far we do not routinely ask for postoperative
speech audiograms, and, therefore, did not have enough data for comparison. [Table 7] summarizes the results of previous studies in the literature. They are quite similar
to our findings, with reported postoperative ABGs of 10 dB to 17 dB for COM and 10 dB
to 13 dB in Chole patients. As more and more surgeons switch to an endoscopic approach,
it will be very interesting to see the results, since during the endoscopy, most ear
surgeons completely detach the remaining eardrum from the malleus handle to access
the anterior extensions of the middle-ear disease. The impact of lifting and reinstating
the drum from the umbo and malleus handle has not yet been properly addressed. One
study by Ohki et al[11] ended with a favorable postoperative ABG for the endoscopic as well as microscopic
groups ([Tabel 7]), with comparable hearing outcomes and no inner ear damage. The use of lasers to
detach choles from an intact chain has been advocated to preserve the integrity of
the chain, and it revealed better and more reliable hearing outcomes in comparison
to the more conventional approach of dismantling the chain and reconstructing the
gap.[12]
Table 7
Literature overview
|
Thresholds (kHz)
|
Author
|
Year
|
Disease
|
n
|
Preop
ABG
|
Postop
ABG
|
Preop
AC
|
Postop
AC
|
Remarks
|
|
0.5–4
|
Our study
|
2020
|
COM
|
85
|
20
|
10.8
|
33.1
|
25.7
|
|
|
0.5–4
|
Our study
|
2020
|
Chole
|
44
|
11.8
|
11.2
|
24.8
|
25.4
|
|
|
0.5–3
|
Horvath et al[1]
|
2019
|
COM
|
147
|
22.1
|
17.2
|
|
|
|
|
0.5–3
|
Ohki et al[11]
|
2019
|
COM
|
122
|
19.7
20.3
|
10.1
13.1
|
43.4
42.3
|
32.7
33.9
|
Endoscopic
Microscopic
|
|
0.5–4
|
Pontillo et al[4]
|
2018
|
Chole
|
65
|
11,7
17,4
|
7.7
12.9
|
27.7
41.2
|
23.1
35.1
|
Canal wall up
Canal wall down
|
|
0.5–3
|
Hamilton[12]
|
2010
|
Chole
|
80
|
|
11.4
|
|
|
Use of KTP laser
|
Abbreviations: ABG, air-bone gap; AC, air conduction; Chole, cholesteatoma; COM, chronic
otitis media; dB, decibel; kHz, kilohertz; KTP, potassium titanyl phosphate; Postop,
postoperative; Preop, preoperative.
Conclusion
Statistically, 80% of the patients with COM will have an intact and mobile chain at
surgery, whereas less than 20% retain their intact chain in cases of Chole. Preoperatively,
patients with eardrum perforations (COM) present with worse hearing thresholds than
patients with Chole and an intact ossicular chain. Postoperatively, their performance
is almost the same, with a mean ABG of 10 dB to 12 dB, which enables a higher hearing
gain for COM patients. Roughly, 50% will end up with an ABG lower than 10 dB, and
90%, within 20 dB. In chole patients, the extent of the disease is small, and it seems
that better pneumatization and ventilation leads to improved outcomes. It remains
difficult to close the ABG at 4 kHz, a frequency that should not be neglected when
presenting outcome data.
