Keywords
otosclerosis - skeeter drill - stapedotomy - teflon piston
Introduction
Otosclerosis is an exclusive disease of the otic capsule, characterized by disordered
resorption and deposition of bone leading to conductive hearing loss from 5 to 60
dB. The incidence of otosclerosis varies between regions, as it is race dependent,
besides depending on age and sex within the race. Management options include fluoride
supplementation, hearing aids (conventional and bone anchored), and stapes surgery.
The choice between surgery and using hearing aids depends on factors, such as the
patient's age, attitude to risk, consideration of aesthetics, and quality of life.
The outcomes of various treatment modalities for otosclerosis must be measured both
objectively and subjectively. In India, despite many studies on the hearing outcomes
of otosclerosis surgery, there exist few studies on small fenestra stapedotomy performed
using a microdrill. Hence, we designed this study with the objectives of examining
the demographic profile, hearing improvement after surgery, anatomical variations
encountered at surgery, effect of microdrill use on bone conduction (BC), and postoperative
complications of small fenestra stapedotomy.
Methods
Sample Size Estimation
Analysis of long-term hearing gains after stapes surgery has shown an average air
conduction (AC) hearing gain of 35 dB over the speech frequencies.[1]
[2]
[3]
[4] Similar to these studies, we expected an average postoperative AC gain of 30 dB
in our study population. Hence, the sample size was estimated to be 63 dB with the
following formula:
where Zα = 1.96 (at 5% significance); Z1−β = 0.8416 (at 80% power); σ = standard deviation = 1.5, Δ = effect size = [(40 − 10)/40]
× 100 = 75%
Patients presenting to the outpatient department (OPD) of KKR ENT hospital fulfilling
the following criteria were taken as study subjects.
Inclusion Criteria
-
Intact tympanic membrane with pure tone audiometry (PTA) showing a conductive hearing
loss with an air–bone gap (ABG) of at least 20 dB with impedance showing A or A's-type
curve.
-
Age between 18 and 60 years.
Exclusion Criteria
-
Patients not consenting to the study.
-
History of previous surgery in the study ear.
-
Patients with conductive deafness not attributable to otosclerosis.
-
Patients with associated vertigo or aural fullness (coexisting Meniere's or benign
paroxysmal positional vertigo [BPPV]).
-
Pregnant females.
Complete information regarding the treatment options was explained, and a written
informed consent was obtained for surgery. A proforma was prepared for the study,
which included demographic profile, clinical history, examination surgical details,
and postoperative follow-up on first, third, and sixth month. Preoperative and postoperative
PTA (GSI AudioStar Pro, Minnesota, United States) and impedance audiometry (Interacoustics
AT235, Assens, Denmark) were recorded for BC and AC over the frequencies of 250, 500,
1,000, 2000, and 4,000 Hz.
Stapedotomy was performed by the same surgeon on all patients by a transcanal approach.
To facilitate assessment of hearing improvement on table, surgery was performed under
local anesthesia (a solution of 1:50,000 adrenaline prepared from 5 mL of 2% Xylocaine,
5 mL of distilled water, and 0.2 mL of 1 in 1,000 adrenaline, 2–3 mL infiltrated with
a 1.5-in 26 G needle, in the posterior quadrant of the external auditory canal, approximately
1 cm from the annulus). A tympanomeatal flap was elevated, and any anatomical variations,
such as narrow oval window niche, partial prolapse of facial nerve, were observed
and recorded in the study proforma. Chorda tympani was identified and retracted anteriorly.
Bone was curetted from the scutum to expose the incudostapedial joint, the stapedius
tendon, the pyramid base, and the stapes footplate. The stapes footplate was inspected,
and the ossicles were gently palpated, thus confirming the diagnosis. The stapedius
tendon was cut with microscissors, and the incudostapedial joint was dislocated with
a right-angled pick. The stapes suprastructure was downfractured gently and was removed.
A microdrill (Skeeter system, Xomed-Treace, Jacksonville, Florida, United States;
in all but one case, where the footplate was removed inadvertently while fracturing
the crura) with a 0.5-mm diamond burr was used to make the fenestra (of approximately
0.6 mm) in posterior part of stapes footplate. Drilling was usually done for 30 to
40 seconds at 7,000 to 12,000 rpm depending on the type of footplate.[5]
[6] In patients who had a narrow oval window niche, the bone of the promontory was drilled
with 0.7-mm Skeeter burr until footplate was visualized. A measuring jig was used,
and a Teflon piston of appropriate length was used, supported with a small piece of
blood-soaked Gelfoam. The loop of the piston was gently crimped to the long process
of the incus. The tympanomeatal flap was repositioned, and the external ear canal
was packed with absorbable Gelfoam soaked in antibiotic drops. Patients were reviewed
on the fourth postoperative day for canal pack removal. Subsequently, patients were
reviewed on the first, third, and sixth postoperative months. PTA was performed at
each visit, and findings were recorded.
Results
Distribution of study population by footplate type, drill duration and speed, piston
diameter, and length was done by chi-square test. The changes in mean values of PTA
findings were assessed and compared between before and after surgery using paired
t-test. Comparison of hearing outcomes in patients who underwent drilling for narrow
niche with those with normal anatomy (hence no drilling of niche) was done using unpaired
t-test (comparing means between two groups). Similar change in the mean value of PTA
findings for 1, 3, and 6 months was assessed and compared using repeated measures
analysis of variance (RMANOVA).
Demographic Profile and Examination Findings
Patients’ age in the study population ranged from 18 to 67 years with a mean of 44.51
years (±11.989) and a median of 47 years. Of the total 63 patients, 37 were males
(58.7%) and 26 were females (41.3%). Majority of patients were older than 35 years
([Table 1]).
Table 1
Distribution of the study population by age and sex
|
Age groups
|
Sex
|
Total
|
|
Male
|
Female
|
|
< 35 y (%)
|
8 (21.6)
|
4 (15.4)
|
12 (19.0)
|
|
35–49 y (%)
|
12 (32.4)
|
13 (50.0)
|
25 (39.7)
|
|
> 50 y (%)
|
17 (45.9)
|
9 (34.6)
|
26 (41.3)
|
|
Total
|
37 (58.7)
|
26 (41.3)
|
63 (100.0)
|
All patients had hearing impairment ranging from 5 to 360 months with a median of
36 months; 29 patients complained of tinnitus (46%), 20 (31.7%) had a family history
of otosclerosis, and 9 (14.28%) had a history of measles ([Fig. 1]). Thirteen (50%) women had a history of aggravation of the symptoms during pregnancy.
In 53 (84.13%) patients, the disease was bilateral.
Fig. 1 Proportion of the study population with family history of otosclerosis and history
of measles.
Air conduction was measured at 250, 500, 1,000, 2,000, 4,000, and 8,000 Hz, whereas
BC was measured at 250, 500, 1,000, 2,000, and 4,000 Hz ([Table 2], [Fig. 2]). Impedance audiometry was A type in 45 (71.4%) patients and A's type in 18 (28.6%)
patients.
Table 2
Baseline mean (±SD) air conduction, bone conduction, and air–bone gap values at different
frequencies
|
Frequency (Hz)
|
Mean (±SD) AC at baseline
|
Mean (±SD) BC at baseline
|
Mean (±SD) ABG at baseline
|
|
Abbreviations: ABG, air–bone gap; AC, air conduction; BC, bone conduction; SD, standard
deviation.
|
|
250
|
49.37 (±12.427)
|
8.33 (±5.750)
|
41.03 (±9.165)
|
|
500
|
52.06 (±12.529)
|
9.21 (±7.252)
|
42.86 (±9.277)
|
|
1,000
|
55.63 (±14.298)
|
14.52 (±9.702)
|
41.11 (±10.098)
|
|
2,000
|
53.73 (±15.912)
|
17.46 (±12.044)
|
36.27 (±11.708)
|
|
4,000
|
56.93 (±19.98)
|
18.63 (±13.58)
|
37.70 (±12.915)
|
|
8,000
|
58.05 (±18.98)
|
–
|
–
|
Fig. 2 Baseline mean air conduction, bone conduction, and air–bone gap values at different
frequencies.
Intraoperative Findings
The chorda tympani nerve was minimally manipulated in 59 (93.6%) patients, completely
sectioned in 3 (4.8%), and partially injured in 1 (1.6%) patient.
Other anatomical variations included narrow oval window niche in seven (9.5%) patients,
partial facial nerve overhang in two (3.2%) patients, persistent stapedial artery
in one (1.6%), and exposed horizontal facial nerve in one (1.6%) patient.
Footplate was type I in 33 (52.4%), type II in 22 (34.9%), type III in 6 (9.5%), and
type IV in 2 (3.2%) patients. However, the association between sex and the type of
footplate was statistically not significant (χ2 = 2.676, p = 0.444; [Table 3]).
Table 3
Distribution of the study population by sex and the type of footplate
|
Type of footplate
|
Sex
|
Total
|
Statistical significance
|
|
Male
|
Female
|
|
Type 1 (%)
|
22 (59.5)
|
11 (42.3)
|
33 (52.4)
|
χ2 = 2.676
p = 0.444
|
|
Type 2 (%)
|
12 (32.4)
|
10 (38.5)
|
22 (34.9)
|
|
Type 3 (%)
|
2 (5.4)
|
4 (15.4)
|
6 (9.5)
|
|
Type 4 (%)
|
1 (2.7)
|
1 (3.8)
|
2 (3.2)
|
|
Total
|
37 (58.7)
|
26 (41.3)
|
63 (100.0)
|
The piston diameter varied from 0.4 to 0.6 mm. In 61 (96.8%) patients, the diameter
of piston was 0.4 mm. The length of the piston varied from 3.75 to 4.75 mm, with majority
of patients (40 [63.5%]) requiring piston of 4.5 or 4.25 mm (22.2%).
Postoperative Hearing Outcomes
The mean values for AC, BC, and ABG at each frequency were compared before and 1 month
after surgery using paired t-test ([Tables 4]
[5]
[6]).
Table 4
Mean air conduction values at baseline (before surgery) and at 1 month after surgery:
comparison using paired t-test
|
Frequency (Hz)
|
Baseline
|
1 mo after the surgery
|
t-Value
|
Statistical significance
|
|
250
|
49.37 (±12.427)
|
22.62 (±10.075)
|
17.563
|
0.000
|
|
500
|
52.06 (±12.529)
|
23.65 (±11.043)
|
19.802
|
0.000
|
|
1,000
|
55.63 (±14.298)
|
27.62 (±11.943)
|
17.753
|
0.000
|
|
2,000
|
53.73 (±15.912)
|
30.56 (±13.384)
|
16.128
|
0.000
|
|
4,000
|
56.93 (±19.98)
|
39.76 (±19.353)
|
11.968
|
0.000
|
Table 5
Mean bone conduction values at baseline (before surgery) and at 1 month after surgery:
comparison using paired t-test
|
Frequency (Hz)
|
Baseline
|
1 mo after the surgery
|
t-Value
|
Statistical significance
|
|
250
|
8.33 (±5.750)
|
9.37 (±5.922)
|
–1.496
|
0.140
|
|
500
|
9.21 (±7.252)
|
11.03 (±7.681)
|
–1.924
|
0.059
|
|
1,000
|
14.52 (±9.702)
|
14.68 (±9.196)
|
–0.174
|
0.862
|
|
2,000
|
17.46 (±12.044)
|
18.33 (±10.999)
|
–0.737
|
0.464
|
|
4,000
|
18.63 (±13.58)
|
21.31 (±13.475)
|
–3.245
|
0.002
|
Table 6
Mean ABG values at baseline (before surgery) and at 1 month after surgery: comparison
using paired t-test
|
Frequency (Hz)
|
Baseline
|
1 mo after the surgery
|
t-Value
|
Statistical significance
|
|
Abbreviation: ABG, air–bone gap.
|
|
250
|
41.03 (±9.165)
|
13.25 (±7.249)
|
20. 061
|
0.000
|
|
500
|
42.86 (±9.277)
|
12.62 (±6.467)
|
23.402
|
0.000
|
|
1,000
|
41.11 (±10.098)
|
12.94 (±7.050)
|
20.178
|
0.000
|
|
2,000
|
36.27 (±11.708)
|
12.22 (±6.207)
|
15.507
|
0.000
|
|
4,000
|
37.70 (±12.915)
|
19.13 (±15.620)
|
8.394
|
0.000
|
Comparison of mean values of BC by the duration of drilling is shown in ([Table 7]). The mean worsening in postoperative BC was only 3.035 dB. Comparison of mean values
of BC by the speed of drilling is shown in ([Table 8]).
Table 7
Mean bone conduction values distributed by the duration of drilling
|
Drill duration
|
Total no. of patients
|
Mean preoperative BC
|
Mean postoperative BC
|
|
Abbreviation: BC, bone conduction.
|
|
< 30 s
|
55 (87.3 %)
|
13.56 (±8.690)
|
16.77 (±10.878)
|
|
30–60 s
|
5 (7.9%)
|
12.40 (±7.829)
|
17.00 (±7.213)
|
|
> 60 s
|
2 (3.2%)
|
14.00 (±11.314)
|
22.50 (±15.910)
|
|
No drill
|
1 (1.6%)
|
37.00
|
51.25
|
Table 8
Mean values distributed by the speed of drilling
|
Speed of drilling
|
Total no. of patients
|
Mean preoperative BC
|
Mean postoperative BC
|
|
Abbreviation: BC, bone conduction.
|
|
No drill
|
1 (1.59%)
|
37.00
|
51.25
|
|
Less than full (around 7,000 rpm)
|
54 (85.71%)
|
13.56 (±8.690)
|
16.77 (±10.878)
|
|
Full speed (around 12,000 rpm)
|
8 (12.70%)
|
12.86 (±7.925)
|
18.57 (±9.169)
|
Comparison of the hearing parameters for seven patients who had a narrow oval window
niche (the bone of the promontory was drilled with 0.7-mm Skeeter burr until footplate
was visualized) with seven randomly selected controls (with normal anatomy, no drilling
of niche done) from the study group was performed ([Table 9]). The unpaired t-test revealed no difference in hearing outcomes between these two groups.
Table 9
Comparison of hearing outcomes in patients who underwent drilling for narrow niche
with randomly selected control group
|
Total no. of cases
|
Average gain in AC at 0.5, 1, 2, and 4 kHz
|
Postoperative ABG
|
Mean increase in postoperative BC
|
|
Abbreviations: ABG, air–bone gap; AC, air conduction; BC, bone conduction.
|
|
Narrow niche (promontory bone drilled with 0.7-mm burr and Skeeter drill)
|
7
|
25 (±12.22) dB
|
12.63 (±4.54) dB
|
3.035 (±5.81) dB
|
|
Normal anatomy (drilling not done)
|
7
|
20.71 (±7.93) dB
|
11.38 (±4.329) dB
|
1.428 (±4.86) dB
|
|
t-Value using unpaired t-test
|
–
|
0.7780
|
0.5610
|
0.5275
|
|
p-Value
|
–
|
0.456
|
0.585
|
0.6075
|
Postoperative Complications
([Table 10]) enlists the complaints patients had at different postoperative visits. One patient
developed sensorineural hearing loss as a complication of surgery. One patient developed
facial paresis (House–Brackmann grade 4) on the sixth postoperative day and recovered
completely with conservative treatment. ([Table 11]) details the intraoperative handling of the chorda and its relation to the postoperative
taste sensation.
Table 10
Patient complaints at different follow-up visits
|
First wk (N)
|
First mo (N)
|
Third mo (N)
|
Sixth mo (N)
|
|
N = 63, but the numbers in each category are not mutually exclusive, and hence the
total will not come up to 63.
|
|
Tinnitus (%)
|
8 (12.7)
|
10 (15.9)
|
7 (11.1)
|
3 (4.8)
|
|
Altered taste (%)
|
10 (15.9)
|
11 (17.5)
|
5 (7.9)
|
3 (4.8)
|
|
Giddiness (%)
|
4 (6.4)
|
1 (1.6)
|
1 (1.6)
|
0 (0.0)
|
|
Discomfort in loud noise (%)
|
4 (4.8)
|
12 (19.0)
|
11 (17.5)
|
9 (14.3)
|
|
Hard of hearing (%)
|
1 (1.6)
|
4 (6.4)
|
5 (8.0)
|
3 (4.8)
|
|
Ear-blocked sensation (%)
|
0 (0.0)
|
0 (0.0)
|
0 (0.0)
|
1 (1.6)
|
|
Facial palsy (%)
|
1 (1.6)
|
0 (0.0)
|
0 (0.0)
|
0 (0.0)
|
|
Patients without any complaints (%)
|
39 (61.9)
|
31 (49.2)
|
37 (58.73)
|
44 (69.84)
|
Discussion
The distribution of otosclerosis in western studies among men and women has been found
to be in the ratio of 1:1.4. In our study, this ratio was found to be 1.4:1.[7] This apparent difference in distribution may be indicative of the poorer health-seeking
behavior among Indian women, in comparison with western women. It may also be due
to the relatively smaller sample size in our study. Familial cases of otosclerosis
constituted 31.7% of the study population, supporting the strong genetic association.[8]
[9] In addition, 14.28% had a history of measles. The incidence of tinnitus in our study
population was 46%, as compared with other studies reporting incidence from 65 to
74%.[10]
[11]
Table 11
Chorda tympani handling and taste sensation
|
Type of chorda handling
|
No. and % of patients
|
No. and % with altered taste at 1 mo
|
No. and % with altered taste at 3 mo
|
No. and % with altered taste at 6 mo
|
|
No or minimal handling (%)
|
59 (93.6)
|
8 (13.5)
|
5 (8.4)
|
3 (5)
|
|
Injury with Skeeter burr (%)
|
1 (1.6)
|
0
|
0
|
0
|
|
Chorda sectioned (%)
|
3 (4.8)
|
1 (33.33)
|
0
|
0
|
|
Total
|
63
|
9
|
5
|
3
|
A narrow oval window niche was encountered in 9.5% of patients. Drilling was done
with a 0.7-mm Skeeter drill burr to access the footplate. Theoretically, this procedure
may result in damage to the cochlear endosteum and thus cause sensorineural hearing
loss.[12] The worsening in BC thresholds was 3.5 dB that was comparable to that of other patients
who did not have this anatomical variation, and no significant deleterious effect
was observed in the BC of these patients.
Facial nerve variations (4.8%) included partial overhang (3.2%) and exposed tympanic
segment (1.6%). Prosthesis was placed in the segment of the footplate that was away
from the facial nerve. Ballestar et al[13] found that 6.7% of the cases of stapedotomies had variations of the facial nerve
including prolapse and exposed tympanic segment. No audiologic or neurologic complications
were encountered among these patients in the postoperative period.
A persistent stapedial artery is a rare surgical finding and was seen in one patient.
This artery can cause troublesome bleeding, if violated. It is advised to coagulate
or cauterize the vessel, if not too large. Another option is to move the artery to
one side, which was the technique used in our study.
Hearing outcomes showed a statistically significant improvement in the mean AC values
at all the frequencies. The gain was highest in the low frequencies followed by that
in the speech frequencies and at 4,000 Hz. Overall four frequency AC gains were found
to be very similar to the four frequency AC gain seen in other studies.[3]
[14]
[15] Similar statistically significant improvement was observed in the mean ABG. In our
study, 90.48% of the patients demonstrated a postoperative ABG of 41.27% in the 0
to 10 dB group and 49.21% in the 11 to 20 dB group. This indicates that small fenestra
stapedotomy gives good closure of ABG in a larger number of patients.
There was a slight worsening in the BC values at all frequencies, which was not statistically
significant. The worsening was higher when the drilling was done at full speed (12,000
rpm).[6] However, the number of patients in the groups in which drill duration was between
30 and 60 seconds and more than 60 seconds was too small to perform statistical analysis.
Perhaps prolonged duration and higher speed of drill could be more deleterious to
the delicate inner ear structures. To our knowledge, no study has analyzed the effect
of microdrill duration and effect of drill speed on BC. Further studies in a larger
number of patients, distributed evenly in the various subgroups of drill duration
and drill speed, may provide more clarity on the effect of these particular aspects
of drill use.
We compared our overall hearing results based on piston diameter versus another study
done by Marchese et al,[16] in which Schuknecht Teflon wire prosthesis was used at two diameters: 0.6 and 0.4
mm. Based on hearing outcomes, the study concluded that the 0.6-mm piston gave better
results. In an overwhelming majority of our study population (96.8%), a Teflon piston
of diameter of 0.4 mm was used. The values of ABG and AC gains (both AC PTA and at
individual frequencies) from our study (0.4-mm Teflon piston) remain better than those
obtained with a 0.6-mm Teflon wire prosthesis in the study by Marchese et al. Apart
from the difference in piston diameter, use of different types of prostheses and the
variations in surgical techniques employed in each of these studies could have influenced
the hearing outcome.
Postoperative Complications
In one (1.6%) patient, part of the footplate was removed while fracturing the crura,
and the patient developed sensorineural hearing loss after surgery. This number is
comparable to the incidence of sensorineural hearing loss of 0.5 to 3% in other series.[3]
[17]
One patient developed facial paresis (House–Brackmann grade 4) on the sixth postoperative
day. He did not have any significant intraoperative findings and had an uneventful
surgery. He was treated with a tapering course of oral corticosteroids and recovered
completely in a few weeks. This rare complication has been reported to occur in approximately
0.5% of the population between the 5th and 20th day. It resolved within 1 to 2 months
of treatment on the lines of Bell's palsy management.[18]
Four patients complained of giddiness within the first week. They were treated with
labyrinthine sedatives. One of these patients complained of mild giddiness occasionally
for up to the third month and improved thereafter; 14.3% patients complained of improvement
in hearing but discomfort in noisy environments. This is a known complication of cutting
the stapedius tendon.
Damage to the chorda tympani occurred in around 30% of the patients and caused dry
mouth, soreness of tongue, and metallic taste.[19] In our study, the most common complaints included altered taste, metallic taste,
and numbness over the tongue. Chorda tympani underwent minimal manipulation, that
is, it was gently moved away from the field without stretching or injuring it in 93.6%
patients. In this subgroup, by the end of the first month, 13.5% had taste-related
complaints. By 3 and 6 months, the percentage reduced to 8.4 and 5%, respectively,
in this group. Chorda was injured with the Skeeter drill in one patient. However,
the patient did not have any taste-related complaints postoperatively. Chorda was
sectioned in 4.8% of patients who underwent surgery, and 25% of patients in this subgroup
had altered taste sensation in the first month. The complaint did not persist in the
subsequent follow-ups.
In a study by Just and colleagues,[20] the postoperative taste sensation was assessed subjectively and also by regional
chemical taste test and electrogustometry in which the percentage of patients diagnosed
to have taste dysfunction was found to be higher. Electrogustometry was found to be
superior to regional chemical taste tests. In our study, the taste sensation was assessed
only subjectively. In both our study and the above-mentioned study, the complaints
were found to decline over time, although a small percent of patients continued to
have troublesome symptoms.
Conclusion
Microdrill-assisted small fenestra stapedotomy, performed under local anesthesia,
with placement of a 0.4-mm Teflon piston, causes good closure of ABG with minimum
complication or long-term sequelae in patients with otosclerosis. Microdrill used
for fenestration is safe when used for short duration and at lower speed. The complication
rates are low, and the surgery has a positive impact on the patient's hearing.
