Keywords eustachian tube - validation study - translation
Introduction
The Eustachian tube (ET), or pharyngotympanic tube, is an important canal located
in the middle ear, linking it with the nasopharynx.[1 ] Normally, the ET remains passively closed with periodical openings.[2 ] The main three functions of the Eustachian tube are to equalize the middle ear pressure
with that of the external environment and to provide ventilation and protection.[3 ] Any disruption in these functions can result in Eustachian tube dysfunction (ETD).
This condition is quite common, affecting up to 5% of the adults seen in otorhinolaryngology
practice, and can have a significant impact on the quality of life of the patient
if it becomes chronic.[4 ]
The main symptom of ETD is a feeling of pressure in the ears or the inability to rapidly
equilibrate middle ear pressure. Other symptoms can include muffled hearing, tinnitus
or a popping sound in the ear, and a sensation of the ears being clogged or being
underwater. The clinical evaluation of these symptoms is subjective; diagnosis of
this condition usually emerges as a combination of symptoms and clinical evaluation
conducted to estimate the severity of the condition and treatment outcomes.[5 ] However, application of ET function tests, such as sonotubometry,[6 ]
[7 ] forced response test,[8 ]
[9 ] tubomanometry,[10 ]
[11 ] videoendoscopy,[9 ]
[12 ] pressure chamber test,[13 ] and inflation-deflation test,[7 ] is limited by the need for high-cost equipment and trained staff, which are primarily
available in specialized centers.
For this reason, a simple tool, such as a questionnaire, can help the physician to
reach a diagnosis of ETD.
McCoul et al.[14 ] developed the Eustachian Tube Dysfunction Questionnaire-7 (ETDQ-7), a reliable and
valid method for clinical applications, as a way to assess ETD severity and treatment
outcomes. The ETDQ-7 has been translated into German,[15 ] Dutch,[16 ] Turkish,[17 ] European and Brazilian Portuguese,[3 ]
[18 ] and traditional Chinese.[19 ] It has been used for the clinical evaluation of ETD,[20 ] and as an objective measure of surgical procedure outcomes, such as balloon dilation
of the ET.[21 ]
The aim of the present study was to develop an Arabic version of the ETDQ-7 scale
and to examine its validity and reliability.
Methods
Study Design
A multicenter prospective cross-sectional study was conducted in Riyadh between August
2019 and October 2020 to develop an Arabic version of the ETDQ-7 and assess its validity
and reliability. The ETDQ-7 includes seven items, responses to which are ranked on
a 7-point Linkert scale, ranging from 1 (no problem) to 7 (severe problem). The total
and average scores are then calculated for each respondent. All the participants filled
out the questionnaires twice: first at the time of diagnosis and then 6 weeks later
for test-retest reliability.
Questionnaire Translation and Adaptation
The English version of the ETDQ-7 ([Fig. 1 ]) was translated into Arabic by two independent certified translators. These Arabic
versions ([Fig. 2 ]) were then checked by two otolaryngologists and compared with the original English
version. The two Arabic versions were merged into one version, which was then back
translated into English by another independent translator who was not aware of the
original English version of the ETDQ-7. There were no significant differences between
the back translated version and the original English version.
Fig. 1 Original ETDQ-7 questionnaire, by McCoul et al.
Fig. 2 Validated Arabic version of the ETDQ-7.
Subjects
All patients included in the present study were at least 18 years old and native Arabic
speakers. The subjects were divided into two groups: an ETD group and a control group.
Patients in the former group had been diagnosed with ETD due to their history of at
least two of the following symptoms in one or both ears over the preceding month:
a sensation of clogged or muffled hearing, aural fullness or pressure, and inability
to rapidly equilibrate middle ear pressure. Patients also exhibited abnormal tympanic
membrane examination on otoscopy. A pure tone audiogram was performed for all patients
at the time of diagnosis
The exclusion criteria included any evidence of recent acute upper respiratory infection,
history of ear diseases (such as chronic supportive otitis media; cholesteatoma; or
ear, nose, throat, head, or neck surgery within the previous 3 months), evidence of
adenoid hypertrophy, nasal polyposis, or cleft palate.
For the control group, patients who did not meet the ETD group inclusion criteria
and who had presented with medical complaints unrelated to ETD were included. All
of these patients had a normal tympanic membrane, middle ear, nasal cavity, and nasopharynx
on examination. Written informed consent was obtained from each subject, and institutional
review board approval was obtained in advance from the Medical Research Center, College
of Medicine, Imam Mohammad Ibn Saud Islamic University, Riyadh, Saudi Arabia (Project
Number: 38–2019).
Questionnaire Validation and Statistical Analysis
For the reliability of the questionnaire items, Cronbach α was used. A minimum value
of 0.7 was considered a measure of good internal consistency. The Pearson correlation
was used to assess the test-retest reliability of the items. A receiver operating
curve (ROC) was used to assess the predictive power of the total score in predicting
the disease condition. The area under the curve (AUC) and the corresponding 95% confidence
interval (CI) were calculated. A minimal threshold of 0.7 was deemed acceptable. The
Youden index was used to estimate the optimal cutoff point for the total score. Bootstrapping
(using 100 bootstrapped samples) was used to validate the cutoff point and performance
metrics. The test-retest reliability of the ETDQ items and total score were assessed
using the intraclass correlation coefficient (ICC) for the seven individual items
and the total ETDQ score.
Statistical analysis was performed using R v 3.6.3. (RStudio, Inc.: Boston, MA, USA).
Counts and percentages were used to summarize the distribution of categorical variables,
while the mean ± standard deviation (SD) were used to summarize the distribution of
the questionnaire items. An unpaired t -test was used to compare the distribution of questionnaire items between the ETD
and control groups. The Epi and cutpointr packages were used for ROC analysis. Hypothesis
testing was performed at a 5% level of significance.
Results
A total of 51 patients with ETD and 45 controls completed the questionnaire ([Table 1 ]). Cases and groups were balanced regarding age (p = 0.314) and gender (p = 1 ). There was a statistically significant difference in the severity of symptoms between
both groups (p < 0.001).
Table 1
Descriptive statistics for the study sample
All
Case
Control
Overall p-value
n = 96
n = 51
n = 45
Gender
Female
41 (42.7%)
22 (43.1%)
19 (42.2%)
1.000
Male
55 (57.3%)
29 (56.9%)
26 (57.8%)
Age (years old)
31.1 (9.30)
32.0 (9.26)
30.1 (9.33)
0.314
Location of the symptoms
Both ears
46 (47.9%)
33 (64.7%)
13 (28.9%)
< 0.001
Right ear
15 (15.6%)
7 (13.7%)
8 (17.8%)
Left ear
14 (14.6%)
11 (21.6%)
3 (6.67%)
No symptoms
21 (21.9%)
0 (0.00%)
21 (46.7%)
The results showed that the average scores for all ETDQ-7 items (except item 7) were
significantly higher in the ETD group compared with controls (p < 0.001). Similarly, the average and total ETDQ scores were significantly higher
in the ETD group than in the control group (p < 0.001, [Table 2 ]). All Arabic ETDQ items were significantly correlated with the total ETDQ score
(p < 0.001, [Table 3 ]). The overall internal consistency of the Arabic ETDQ items, as measured by Cronbach
α, was satisfactory (Cronbach α = 0.803).
Table 2
Comparison of Arabic ETDQ scores between cases and controls
[ALL]
Case
Control
p-value
n = 96
n = 51
n = 45
Ear pressure
2.11 (1.29)
2.84
1.29
< 0.001
Ear pain
1.73 (1.14)
2.25
1.13
< 0.001
Ear clogged
2.41 (1.68)
3.10
1.62
< 0.001
Ear symptoms after cold or sinusitis
2.38 (1.62)
2.90
1.78
< 0.001
Crackling sound in ear
1.78 (1.23)
2.27
1.22
<0.001
Ringing ears
1.96 (1.26)
2.51
1.33
<0.001
Muffled hearing
1.60 (1.16)
1.71
1.49
0.356
Total score
14.0 (6.41)
17.6
9.87
<0.001
Mean score
2.00 (0.92)
2.51
1.41
<0.001
Table 3
Correlation between the ETDQ total score and scores for individual items
Item 1
Item 2
Item 3
Item 4
Item 5
Item 6
Item 7
Ear pressure
Ear pain
0.474***
Ear clogged
0.576***
0.386***
Ear symptoms after cold or sinusitis
0.478***
0.408***
0.470***
Crackling sound in ear
0.503***
0.453***
0.367***
0.230*
Ringing ears
0.505***
0.438***
0.395***
0.362***
0.510***
Muffled hearing
0.275**
0.162
0.331***
0.227*
0.158
0.228*
Mean score
0.791***
0.630***
0.777***
0.709***
0.602***
0.665***
0.444***
Computed correlation used spearman-method with listwise-deletion.
* p < 0.05.
*** p < 0.001.
The results showed that the self-reported severity of symptoms was higher in cases
than in controls ([Fig. 3 ]). The prevalence of symptoms grades 5, 6, 7 (higher severity) in cases was 12, 22,
and 22% for items 1, 3, and 4, respectively. In the controls the percentages were
0, 2, and 0%, respectively. The proportion of respondents who reported moderate symptoms
was also higher in cases than in controls, specifically for item 1 (22 versus 4%,
respectively). Almost all respondents in the control group responded with 1, 2, and
3.
Fig. 3 Likert scale plot for the ETDQ items.
The retest questionnaire was completed by 48 of the 51 ETD patients ∼ 6 weeks after
the first questionnaire. The results indicate a good to excellent correlation (> 0.7),
which indicates good test-retest reliability. All ICCs were statistically significant
(p ≤ 0.05), which indicates that the pretest scores are significantly associated with
the post-test scores ([Table 4 ]). The results showed an excellent predictive power for the ETDQ total score, with
an AUC ∼ 0.9 ([Fig. 4 ]), which suggests that the Arabic ETDQ-7 can be used to differentiate between patients
with and without ETD.
Fig. 4 ROC curve for the ETDQ total score.
Table 4
Test re-test reliability of the ETDQ items
Item
ICC
95% CI
P
Ear pressure
0.93
0.87–0.96
< 0.001
Ear pain
0.87
0.78 - 0.92
< 0.001
Ear clogged
0.91
0.84 - 0.95
< 0.001
Ear symptoms after cold or sinusitis
0.91
0.85–0.95
< 0.001
Crackling sound in ear
0.90
0.83–0.94
< 0.001
Ringing ears
0.94
0.89 - 0.96
< 0.001
Muffled hearing
0.92
0.86–0.95
< 0.001
total
0.95
0.92 - 0.97
< 0.001
The cutoff value and performance of the ETDQ total score were validated using 100
bootstrapped samples ([Fig. 5 ]). Model performance was assessed using sensitivity, specificity, positive and negative
predictive values (PPV and NPV), accuracy, and AUC ([Table 5 ]). The results showed that a cutoff value of 13 resulted in the best model performance.
The AUC for the total ETDQ score was 88.6% (95%CI: 80.3–96.8%).
Table 5
Validated cut-off value for the total ETDQ score
Cut-off value
Sensitivity
Specificity
PPV
NPV
Accuracy
AUC
≥ 13
71.4%
84.4%
83.3%
73.1%
77.9%
88.6%
[80.3% - 96.8%]
Fig. 5 Validation of the model using bootstrapping (B = 100).
Discussion
The diagnosis and treatment of ET diseases have been gaining considerable interest
worldwide. Various diagnostic tools have been suggested in an attempt to test ET function.[16 ] Several studies have focused on translating and testing the validity and reliability
of the ETDQ-7 in regions like Europe, where various languages are spoken, including
German,[15 ] Dutch,[16 ] Turkish,[17 ] and European Portuguese.[18 ] The EDTQ-7 has also been translated into Brazilian Portuguese[3 ] and traditional Chinese.[19 ] Since no validated and reliable Arabic version has yet been published in the literature,
the present study aimed to develop an Arabic version of the ETDQ-7 scale and to examine
the validity and reliability of the Arabic version among Arabic speakers living in
Saudi Arabia.
The sample was divided into an EDT group (those who met the inclusion criteria) and
a control group. Three patients in the ETD group failed to attend their follow-up
appointments and, therefore, did not complete the test-retest questionnaire. The results
indicate a good to excellent correlation (> 0.7) between measurements taken at the
time of the first and second questionnaires, which indicates good test-retest reliability.
This is in line with the findings of the study originally conducted to develop the
ETDQ-7 (Spearman rank correlation coefficient = 0.78).[14 ]
All ETDQ items (except item 7) were found to be significantly higher in the ETD group
compared with the control group (p < 0.001), which is similar to the original study.[14 ] Furthermore, a study conducted in 2015 found that the mean ETDQ-7 total score was
9.91 in a control group and 25.77 in patients with obstructive ET dysfunction, in
comparison with our results, which were 17.6 and 9.87, respectively.[16 ] Item 3 (feeling clogged or “underwater”) had the highest mean score (3.06 ± 1.78),
which is consistent with another study.[17 ] Item 7 (feeling muffled) had the lowest mean score (1.71 ± 1.27). We believe that
the average age in the ETD group (32 ± 9.39 years old) possibly affected the results
of all items, especially item 7.
The overall internal consistency in the original study was α = 0.711 (95%CI: 0.570–0.818).
All items were close to measuring the same underlying construct, and no additional
items were deleted from the instrument.[14 ] In the present study, the internal consistency was sufficiently good (Cronbach α = 0.803).
In addition, all Arabic questions were significantly correlated with the total score
(p < 0.001).
Previous studies have reported the AUC to be between 0.95 and 1.0.[14 ]
[15 ] The results of our study were similar and showed an excellent predictive power for
the total score (AUC ∼ 0.9), which suggests that the Arabic version of ETDQ-7 can
be used to differentiate between patients with and without ETD.
Our current findings suggest that the Arabic version of the ETDQ-7 is a reliable instrument
with reproducibility and validity that can be used to assess any native Arabic speaker
despite their nation since we used standard Arabic to translate the questionnaire.
The limitations of our study include its small population size. In addition, our study
did not test pediatric age groups. Therefore, we recommend that future studies include
patients in the pediatric age group because of the high incidence of ETD in children.
Conclusion
In conclusion, the Arabic version of the ETDQ-7 is a reliable and valid scale that
can be used as an adjunct to the patient history-taking process as a method to evaluate
and diagnose ETD. It can also be used as an important tool for patient follow-up.
