Page 36 - Revista SPORL - Vol 58. Nº1

TABLE 2
Results of the ETDQ7

OETD Group Control Group p-value
15 (IQR = 13) < 0.001*
Median (IQR) 34 (IQR = 8.5)
40.7% (n=11) 0.001**
Score 59.3% (n=16)

< 14.5 5.6% (n=2)

> 14.5 94.4% (n=34)

*Independent-Samples Mann-Whitney U Test, **Pearson Chi-square

TABLE 3
Results in the objective exams: tympanometry and TMM

Tympanometry OETD Group Control Group p-value
Type A 0.001*
Type C 38.9% (n=14) 92.6% (n=25) < 0.001*
Tubomanometry 61.1% (n=22) 7.4% (n=2)
R<1
R>1 8.3% (n=3) 66.7% (n=18)
*Pearson Chi-square 91.7% (n=33) 33.3% (n=9)

TABLE 4
Sensitivity, specificity, positive predictive value, negative predictive value and Area-under the ROC curve of the three exams
considering clinical diagnosis the reference

Sensitivity Specificity PPV NPV AUC
68% 84.6% 0.68 ± 0.07
ETDQ-7 94.4% 40.7% 81.5% 61.1% 0.71 ± 0.07
78.6% 85.7% 0.8 ± 0.06
Tympanometry 61.1% 92.6%

Tubomanometry 91.7% 67.7%

PPV: positive predictive value; NPV: negative predictive value

FIGURE 1 DISCUSSION
Receiver operating characteristic (ROC) curve of the ETDQ7, Despite OETD being a common condition and existing
tympanometry, and TMM in relation to the reference clinical several tests, there is no gold-standard to evaluate ET
diagnosis function.
Several investigators argued that given the absence
of widely accepted PROMs, functional tests or scoring
systems, the diagnosis of ETD should rely on the
clinical evaluation, namely symptoms and signs.4
Nonetheless, our results showed that there was an
important percentage (10%) of the evaluated patients
in whom the diagnosis of OETD based on the clinical
evaluation was not consistent between the evaluating
Otorhinolaryngologists. Other studies had also high
rates of disagreement using the same diagnostic criteria,
varying from 1.7 to 25%.6 This emphasizes the need
for an objective exam, repeatable, able to standardize
findings.
Despite being quantitative, easing the process of clinical
history-taking regarding symptoms of ETD, making it
repeatable and comparable between patients, the clear
disadvantage of ETDQ7 was confirmed by our results,
namely the poor specificity. Our study showed good
sensitivity and an AUC of 0.68 ± 0.07. Other authors

34 REVISTA PORTUGUESA DE OTORRINOLARINGOLOGIA E CIRURGIA DE CABEÇA E PESCOÇO