See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/284136515

Eustachian Tube Mucosal Inflammation Scale Validation Based on Digital

Video Images

Article in Otology & neurotology: official publication of the American Otological Society, American Neurotology Society [and] European Academy of Otology and Neurotology · November
2015
DOI: 10.1097/MAO.0000000000000895

CITATIONS READS

24 15,310

5 authors, including:

Ilkka Kivekäs Leena Pöyhönen

Tampere University Tampere University
61 PUBLICATIONS 831 CITATIONS 4 PUBLICATIONS 59 CITATIONS

SEE PROFILE SEE PROFILE

Antti Aarnisalo Markus Rautiainen

Helsinki University Central Hospital Tampere University
109 PUBLICATIONS 2,581 CITATIONS 152 PUBLICATIONS 2,664 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Markus Rautiainen on 07 December 2017.

The user has requested enhancement of the downloaded file.

CE: D.C.; ON-15-164; Total nos of Pages: 6;
ON-15-164

Otology & Neurotology

xx:xx–xx ß 2015, Otology & Neurotology, Inc.

Eustachian Tube Mucosal Inflammation Scale Validation Based

on Digital Video Images
yIlkka Kivekäs, Leena Pöyhönen, zAntti Aarnisalo,
Markus Rautiainen, and yDennis Poe
Department of Otorhinolaryngology, Tampere University Hospital and the University of Tampere, Tampere, Finland;
yDepartment of Otolaryngology, Boston Children’s Hospital and Harvard Medical School, Boston, Massachusetts, U.S.A.; and
zDepartment of Otorhinolaryngology, Helsinki University Hospital, Helsinki, Finland

Background: The most common cause for Eustachian tube Results: Twenty-six clinicians with various levels of experi-
dilatory dysfunction is mucosal inflammation. The aim of ence rated a total of 35 videos. Thirteen clinicians rated the
this study was to validate a scale for Eustachian tube videos twice. The overall correlation coefficient for the
mucosal inflammation, based on digital video clips obtained rating of inflammation severity was relatively good (0.74,
during diagnostic rigid endoscopy. 95% confidence interval, 0.72–0.76). The intralevel corre-
Methods: A previously described four-step scale for grading lation coefficient for intrarater reliability was high (0.86).
the degree of inflammation of the mucosa of the Eustachian For those who rated videos twice, the intralevel correlation
tube lumen was used for this validation study. A tutorial for coefficient improved after the first rating (0.73, to 0.76), but
use of the scale, including static images and 10 second video improvement was not statistically significant.
clips, was presented to 26 clinicians with various levels of Conclusion: The inflammation scale used for Eustachian
experience. Each clinician then reviewed 35 short digital tube mucosal inflammation is reliable and this scale can be
video samples of Eustachian tubes from patients and rated used with a high level of consistency by clinicians with
the degree of inflammation. A subset of the clinicians various levels of experience. Key Words: Dysfunction—
performed a second rating of the same video clips at a Eustachian Tube—Inflammation scale—Video images.
subsequent time. Statistical analysis of the ratings provided
inter- and intrarater reliability scores. Otol Neurotol 00:xxx–xxx, 2015.

Eustachian tube dilatory dysfunction is the most com- It has been observed that laser and balloon dilations
mon cause of middle ear disorders, including negative seemed to reduce the severity of mucosal inflammation
middle ear pressure and otitis media. Consequences of within the lumen of the Eustachian tube (3,4). It has been
chronic dilatory dysfunction can be tympanic membrane postulated that these therapies may strip off irreversibly
retraction or perforation, middle ear effusion or infection, injured or infected mucosa to allow for new healthy
conductive hearing loss, pain or injury with barochal- mucosa to regenerate and that the inflammatory tissue
lenges (e.g., flights or scuba diving), and even choles- within the submucosa may be reduced (5). It has become
teatoma (1). The etiology of Eustachian tube dilatory necessary to employ a validated rating scale for grading
dysfunction can be categorized into two groups: obstruc- the inflammation of the mucosa within the lumen of the
tive (usually functional and not anatomically blocked), or Eustachian tube to compare the outcomes of surgery on
dynamic (muscular dysfunction). Obstructive dysfunc- the Eustachian tube. An instrument for rating Eustachian
tion is much more common than dynamic dysfunction tube mucosal inflammation using a four-point grading
(2). Mucosal inflammation with edema seems to be the scale has been proposed and previously reported, but has
leading source of functional obstruction and has multiple not been validated up to this time (3,6).
possible etiologies including upper respiratory infection, The present study was designed to validate an Eusta-
allergies, laryngopharyngeal reflux, tobacco smoke chian tube mucosal rating scale based on the review of
exposure, chronic rhinosinusitis, and nasal polyps (1). digital video clips of clinical endoscopic examinations,
rather than static images of Eustachian tube orifice.
Address correspondence and reprint requests to Ilkka Kivekäs,
M.D., Ph.D., Department of Otorhinolaryngology, Tampere University MATERIALS AND METHODS
Hospital, Teiskontie 35, 33521 Tampere, Finland; E-mail: ilkka.kivekas@
fimnet.fi This study was approved by the head of the clinic in Tampere
AQ2 The authors disclose no conflicts of interest. University Hospital, Tampere, Finland. This type of survey study

1
CE: D.C.; ON-15-164; Total nos of Pages: 6;
ON-15-164

2 I. KIVEKA¨S ET AL.

does not require approval of the ethical board of the hospital in (Grade 4) Severely inflamed mucosa, unable to dilate open,
Finland, according to statements from the ethical board. Figure 4.
Deidentified clinical videos clips taken from transnasal
endoscopic examinations of Eustachian tubes were assembled We selected 35 videos to use in the study on the basis of high-
to be presented to the volunteer reviewers. Rigid and fiberoptic quality video recording and excellent images of the Eustachian
endoscopic videos were used from adult patients seen in the tube. The two PIs independently rated the videos using the same
clinic. Examinations were performed after topical decongestant methodology for inter- and intrarater assessment of the scores.
and anesthetic spray (Oxymetaxoline þ Lidocaine 4% 1:1 There was 100% correlation between the PI’s scores. It should
mixture) were given. Patients were examined in the sitting be acknowledged that the senior PI was not blinded to the
position and the endoscopes were maneuvered to the tubal source of the videos, which could be considered a limitation in
orifice to provide an optimal view into the lumen along its the study design. The distributions of the videos were the
longitudinal axis. Thirty or 45 degree angled rigid endoscopes following: Grade 1: 10 videos; Grade 2: 8 videos; Grade 3:
were employed or fiberoptic scopes were passed through the 11 videos, and Grade 4: 6 videos.
ipsilateral or contralateral nasal cavities to optimize the view. Statistical analyses were calculated using SPSS (SPSS, Inc.,
Volunteer reviewers were recruited from the Department an IBM Company, Chicago, IL, U.S.A.). Spearman’s corre-
of Otorhinolaryngology at Tampere University Hospital and lations were calculated using a p < 0.01 as the significant limit.
Helsinki University Hospital, including 26 raters with varying A two-tailed paired t test was performed using a p < 0.05 as the
levels of experience. The participants were first presented with significance limit for Eustachian tube inflammation scores.
a short tutorial that explained the mucosal inflammation rating The inter-rater Spearman’s correlation coefficient was cal-
scale. The educational tutorial presented the criteria for the culated as a correlation between the participant assessment and
grading system with static images and approximately 10-second PIs assessment. The intrarater correlation was calculated as a
duration video clips of normal mucosa or of mild, moderate, and Spearman’s correlation coefficient between the first and the
severe mucosal inflammation. The reviewers were then pre- second assessment. The standard error values of each Grade
sented with 35 additional 10-second video clips in a randomized group (Grade 1 videos, Grade 2 videos, Grade 3 videos, and
order that varied with every viewing session. The participants Grade 4 videos) were measured as a difference between the PIs’
were instructed to grade the inflammation appropriately accord- value and the mean value of all participants.
ing to the tutorial and enter their rating using an online data
collection tool (SurveyGizmo, Widgix, U.S.A.). All reviewers RESULTS
evaluated the sample video clips independently, at their own
desktop computers. These evaluations were done at separate Twenty-six participants were enrolled into the study.
locations, with no opportunity for the raters to discuss results. Thirteen participants rated 35 video clips once and 13
The participants were asked to perform a second rating of the participants rated the same video clips twice (arranged in
same videos (35 videos) at least a week later; Intrarater different orders), so video clips were assessed totally
reliability was assessed on the basis of this second rating. 39 times.
The Eustachian tube inflammation scale (3,6) is based on The participants ranged in experience from Otolar-
assessment of mucosal inflammation within the nasopharyngeal yngology residents to senior physicians. For all raters, the
orifice and lumen (edema, erythema, mucus quantity and
intralevel Spearman’s correlation coefficient was rela-
viscosity, lymphoid hyperplasia with ‘‘cobblestoning’’, etc.)
and Eustachian tube function, as follows: tively high 0.74 (95% confidence interval, 0.72–0.76).
For those responders, (13 raters), who rated the videos
(Grade 1) Normal mucosa and normal dilation, Figure 1. twice, the intraindividual correlation between two events
(Grade 2) Mildly inflamed mucosa, no apparent compromise was high at 0.86 (95% confidence interval 0.69–0.91). At
of dilation, Figure 2. the second response time, the intralevel correlation
(Grade 3) Moderately inflamed mucosa, some compromise improved from 0.73 to 0.76, although the improvement
of dilation, Figure 3. was not statistically significant (paired t test, p ¼ 0.16).

FIG. 1. Grade 1. (A) Normal left Eustachian tube in rest position with healthy mucosa. (B) The Eustachian tube opens normally when
swallowed.

Otology & Neurotology, Vol. xx, No. xx, 2015

CE: D.C.; ON-15-164; Total nos of Pages: 6;
ON-15-164

AQ1 EUSTACHIAN TUBE MUCOSAL INFLAMMATION SCALE 3

FIG. 2. Grade 2. (A) Mildly inflamed mucosa. (B) No apparent compromise of dilation.

The standard error value was calculated as the differ- prevented opening of the Eustachian tube lumen.
ence between the real value (as determined by the PIs) Inability to open was evidently a very recognizable
and the rater’s response. Grade 4 inflammation videos phenomenon. In the other grades, there seemed to
showed the best consistency with the PIs (mean error be more subjective variance in the assessment of the
value ¼ 0.24). The mean error value was similar in all the degree of mucosa inflammation and determining whether
other grades; Grade 1 (0.47), Grade 2 (0.52), and Grade 3 the opening of the Eustachian tube was partially com-
(0.49) videos. The difference between the real value and promised or not. Therefore, for Grades 1 through 3 there
rater’s response was 0 (meaning identical assessment) in were the bigger differences between the PIs’ values and
58%, 1 in 38%, 2 in 3%, and 3 in none of cases. the participants’ mean values. In reviewing the inaccur-
ate values, the most common difference between PIs’ and
DISCUSSION the participants’ assessment was most commonly by one
grade, in a few cases two grades, but never three grades.
Eustachian tube dilatory dysfunction is relatively com- Evaluation of Eustachian tube mucosal inflammation
mon finding in patients presenting to Otolaryngology trended toward improving over time, presumably as
clinics (7). As clinicians try to better understand the long- the physician gained more experience. None of the
term outcomes of Eustachian tube dilatory dysfunction, participants were familiar with the inflammation scale
the demand for a validated and user-friendly scale to before and they got only a short tutorial slide show
describe the severity of the mucosal inflammation of the before the first evaluation. Despite the limited training,
Eustachian tube lumen has risen. the scale was very reproducible within raters as seen a
The standard error value was lowest in Grade 4 videos, high correlation between the first and the second assess-
meaning that it was the easiest grade to identify. The ment. In this study, the second evaluation was frequently
standard error in Grade 1, Grade 2, and Grade 3 was more accurate (i.e., more closely aligned with the PIs)
relatively similar, roughly average 0.5 error. Grade 4 than the first one, although the improvement was not
inflammation was likely easier to recognize, because of statistically significant. The main issue was the errors
the severe degree of mucosal inflammation that totally compared with the PIs, The trend toward improvement

FIG. 3. Grade 3. (A) Left side with moderately inflamed mucosa. (B) Obvious compromise of dilation.

Otology & Neurotology, Vol. xx, No. xx, 2015

CE: D.C.; ON-15-164; Total nos of Pages: 6;
ON-15-164

4 I. KIVEKA¨S ET AL.

FIG. 4. Grade 4. (A) Right side with severely inflamed mucosa. (B) Unable to dilate the Eustachian tube lumen.

with experience could suggest that a more thorough the Eustachian tube lumen, there was significant lym-
tutorial and longer experience with the assessments could phocytic infiltration and even follicles within the sub-
improve rater’s accuracy. mucosa as well as loss of the cilia in areas, indicating
It is important to assess the dynamic nature of the chronic inflammation. A small sample of patients with
Eustachian tube (8). Therefore, the rating of the degree of postoperative histological examinations showed that the
inflammation in the mucosa within the lumen of the lymphocytic infiltration in the submucosa had become
Eustachian tube should be evaluated not only in the markedly reduced and was replaced, in part, by thinner
resting position, but also in the active dilation phase. fibrotic scar. The previously injured epithelium had
In some circumstances, the mucosa of the Eustachian regenerated to healthy pseudocolumnar, ciliated epi-
tube may look like it is severely swollen at the orifice, but thelium (5). The significant improvement in the ET
the mucosa within the lumen may become abruptly mucosal inflammation scores in these same patients
normal, allowing for adequate dilation. Furthermore, supports the hypothesis that inflammation within the
the mucosa may look like normal, but there can be an cartilaginous ET plays an important role in ET dilatory
underlying weakness in the dilatory muscles that can dysfunction.
cause inadequate dilation. The present scale does not The study was limited by the number of video clips and
grade dynamic dilatory dysfunction. This is a limitation by the duration (from 8 to 16 s) of each of the video clips
of the proposed mucosal scale, which should be used in that can be presented to raters without creating fatigue.
conjunction with a description of the quality of muscular Additionally, only half of the raters reviewed the clips a
efforts of the tensor veli palatine (TVP) and levator second time. The scale was new to all of the raters so it is
veli palatine (LVP) muscles involved in dilation of the possible that the standard error could have been reduced
Eustachian Tube. with additional experience with the scale or additional
Previous studies have demonstrated that in most cases sessions repeating the evaluations. The scale only scored
of Eustachian tube dilatory dysfunction, there is a sig- mucosal inflammation and did not assess the problems of
nificant association with mucosal inflammation that may dynamic dysfunction (muscular) or anatomical obstruc-
play an etiological role in the dysfunction (4,6,9). Kujaw- tion (e.g., a mass lesion). However, most patients with
ski et al. described this with 18 pre and postoperative dilatory dysfunction are found to have mucosal inflam-
histological samples that displayed improvements in matory changes that are now thought to cause compro-
endoscopic observations, which correlated with mucosal mise of dilation.
healing after laser-assisted Eustachian tuboplasty (9).
With balloon Eustachian tuboplasty, mucosal inflam- CONCLUSION
mation was observed to decrease by a statistically sig-
nificant amount and there was correlation between The four grade Eustachian tube inflammation rating
reduction of the mucosal inflammation and clinical post- scale has been presented as an instrument for the clinical
operative outcomes (4,6). evaluation of the degree of inflammatory changes in the
The need for a validated instrument to grade inflam- mucosa of the lumen of the cartilaginous Eustachian
mation of the Eustachian tube mucosa has grown with the tube. It demonstrated validity in this study and it can
increasing use of surgery of the Eustachian tube in the serve as a tool for clinicians in their assessment of
treatment of dilatory dysfunction. Currently, there is no mucosal inflammation to facilitate future clinical visits
generally accepted mechanism to explain the benefits for and studies.
Eustachian tube balloon dilation. Observations taken
postoperatively have shown that the lumen of the ET Acknowledgments: The authors wish to gratefully acknowl-
seemed to be wider, with reduced inflammation and edge Tali Rasooly, BA, for her assistance in writing and editing
thinner mucosa. In preballoon histological samples from this manuscript.

Otology & Neurotology, Vol. xx, No. xx, 2015

CE: D.C.; ON-15-164; Total nos of Pages: 6;
ON-15-164

EUSTACHIAN TUBE MUCOSAL INFLAMMATION SCALE 5

REFERENCES 5. Kivekäs I, Chao W, Faquin W, et al. Histopathology of

Balloon dilation Eustachian Tuboplasty. Laryngoscope 2015;125:
1. Seibert JW, Danner CJ. Eustachian tube function and the middle ear. 436–41.
Otolaryngol Clin North Am 2006;39:1221–35. 6. Poe DS, Silvola J, Pyykkö I. Balloon dilation of the cartilaginous
2. Poe DS, Abou-Halawa A, Abdel-Razek O. Analysis of the dysfunc- Eustachian tube. Otolaryngol Head Neck Surg 2011;144:563–9.
tional eustachian tube by video endoscopy. Otol Neurotol 2001;22: 7. Browning GG, Gatehouse S. The prevalence of middle ear disease in
590–5. the adult British population. Clin Otolaryngol Allied Sci
3. Poe DS, Metson RB, Kujawski O. Laser eustachian tuboplasty: A 1992;17:317–21.
preliminary report. Laryngoscope 2003;113:583–91. 8. Poe DS, Pyykkö I, Valtonen H, et al. Analysis of Eustachian tube
4. Silvola J, Kivekäs I, Poe DS. Balloon dilation of the cartilaginous function by video endoscopy. Am J Otol 2000;21:602–7.
portion of the eustachian tube. Otolaryngol Head Neck Surg 9. Kujawski OB, Poe DS. Laser Eustachian tuboplasty. Otol Neurotol
2014;151:125–30. 2004;25:1–8.

Otology & Neurotology, Vol. xx, No. xx, 2015

CE: D.C.; ON-15-164; Total nos of Pages: 6;
ON-15-164

MAO Otology & Neurotology

Manuscript No. 15-164

Dear Author,

During the preparation of your manuscript for typesetting, some queries have arisen. These are listed
below. Please check your typeset proof carefully and mark any corrections in the margin as neatly as
possible or compile them as a separate list. This form should then be returned with your marked proof/
list of corrections to the Production Editor.

QUERIES: to be answered by AUTHOR

QUERY NO. QUERY DETAILS RESPONSE
<AQ1> Please check the running head as
suggested. OK
<AQ2> Please check the accuracy of the
conflicts of interest statement. OK

View publication stats