Introduction
Eustachian tube dysfunction (ETD) is common in chronic rhinosinusitis with nasal polyps (CRSwNP) and is prevalent in about 1 per cent of the general population, while Chen et al. have reported 47.2 per cent of chronic rhinosinusitis (CRS) patients have ETD.Reference Chen, Shih, Edwards, Nguyen, Meyer and Soler1 The Eustachian tube (ET) has three functions: pressure equalisation and ventilation of the middle ear, mucociliary clearance of secretions from the middle ear and protection of the middle ear from sounds, pathogens and secretions from the nasopharynx.Reference Schilder, Bhutta, Butler, Holy, Levine and Kvaerner2, Reference Chen, Yang, Lin, Fang, Wu and Luo3 ETD usually refers to a problem with the ventilatory function of the ET and is defined by symptoms and signs of pressure dysregulation in the middle ear.Reference Schilder, Bhutta, Butler, Holy, Levine and Kvaerner2, Reference Chen, Yang, Lin, Fang, Wu and Luo3
In patients with CRS, symptoms of ETD may occur when mucosal oedema and secretions are present around the ET orifice, extending from sino-nasal cavities, disrupting the pressure equilibrium in the ET and middle ear cavity.Reference Chen, Shih, Edwards, Nguyen, Meyer and Soler1, Reference Chen, Yang, Lin, Fang, Wu and Luo3, Reference Adams, Derbarsegian, Sedaghat and Phillips4 ETD symptoms may further worsen patients’ quality of life by affecting their auditory and vestibular function.Reference Wuraola, Afolabi and Ologe5 The signs of ETD are either a retracted ear drum, negative pressure on tympanogram or tympanic membrane excursion with breathing depending on the subtype of ETD.Reference Schilder, Bhutta, Butler, Holy, Levine and Kvaerner2
The seven-item Eustachian Tube Dysfunction Questionnaire (ETDQ-7) and 22-item sino-nasal outcome test (SNOT-22) questionnaires are validated tools for the quantitative evaluation of ETD and CRS, respectively.Reference Chen, Yang, Lin, Fang, Wu and Luo3, Reference Maniakas, Desrosiers, Asmar, Al Falasi, Endam and Hopkins6–Reference McCoul, Anand and Christos8 A positive correlation between ETDQ-7 and SNOT-22 scores suggests that sino-nasal symptoms may have a causative effect on ETD.Reference McCoul, Anand and Christos8
We aimed to identify the prevalence and severity of ETD-associated symptoms in CRSwNP and compared it to a control group presenting with CRS and ETD-related symptoms but found to have no evidence of nasal polyps at endoscopy or mucosal thickening in sinuses on computed tomography (CT) scan.
Methods
This was a retrospective single-centre study of adult patients. Caldicott Guardian approval was granted for accessing and collecting the data (IGTCAL10373+). The patients were seen in the rhinology clinic during a 14-month period and had completed SNOT-22 and ETDQ-7 questionnaires. The study group were patients with bilateral CRSwNP while the control group were patients presenting with symptoms of CRS and ETD-related symptoms with normal endoscopy and CT sinus scan based on EPOS2020 criteria.Reference Fokkens, Lund, Hopkins, Hellings, Kern and Reitsma9 Patients with known otologic disease were excluded, as well as if they had undergone endoscopic sinus surgery (ESS) within the past year or were receiving regular oral corticosteroids (≥5 mg of prednisolone or equivalent daily on a maintenance basis) or biologics.
They had type 2 (T2) biomarkers, including peripheral blood eosinophils (PBE) and total immunoglobulin E (IgE), tested as part of routine investigations. The average values of the T2 biomarkers over the previous 6 months were compared with the normal reference range PBE less than 300 cells/µL and total IgE less than 100 kU/L.Reference Misirovs, Chan and Lipworth10 The results of PBE and total IgE were excluded when the blood tests were obtained during acute hospital admissions and taken within a 3-month period after receiving oral corticosteroids.
Information on comorbid asthma, allergy to common aeroallergens, previous ESS history, smoking history and body mass index (BMI) was accessed from electronic and paper medical records and blood test results.
NPS was obtained from clinic letters reported by two ENT consultants. The Nasal Polyp Score grading system was used, with the grading in each nasal passage being scored from 0 to 4 (0, no nasal polyps; 1, small polyps confined to the middle meatus; 2, moderate-sized polyps reaching below the lower border of the middle turbinate; 3, large polyps reaching the lower edge of the inferior turbinate or polyps medial to the middle turbinate; 4, large polyps touching the floor of the nasal passage) with a total maximum score of 8.
The severity of sinus inflammation was quantified by assessing the degree of sinus mucosal thickening on CT sinus imaging using Lund-Mackay Score (LMS) system. In the LMS system, each paranasal sinus is graded separately from 0 to 2 (0, no opacification; 1, partial opacification; 2, complete opacification), and the ostiomeatal complex is graded with 0 (not obstructed) or 2 (obstructed). The combined score ranges from 0 to 24. Only CT sinus scans performed within three months before or after the outpatient clinic appointment having not had medical polypectomy during the time were included in the data analysis.
The ETDQ-7 consists of seven questions, each scored 1 to 7, ranging from 7 to 49; a score of greater than or equal to 14.5 is indicative of ETD.Reference Chen, Yang, Lin, Fang, Wu and Luo3, Reference McCoul, Anand and Christos8, Reference Chang, Roozdar, Lin, Lee, Nayak and Patel11 Two out of three SNOT-22 questionnaire ear domains (ear fullness and ear pressure/pain) were used to compare it to ETDQ-7.Reference Maniakas, Desrosiers, Asmar, Al Falasi, Endam and Hopkins6 The third ear domain (dizziness) does not correspond to any of the ETDQ-7 domains and the consensus statement on ETD.Reference Schilder, Bhutta, Butler, Holy, Levine and Kvaerner2 The rhinology-focused, abbreviated SNOT-5 score, as previously described, consists of the need to blow nose, runny nose, thick nasal discharge, sense of smell/taste and blockage/congestion domains of the SNOT-22 questionnaire.Reference Misirovs, Chan and Lipworth12
Statistical analysis
Analysis was done using Jamovi (Version 2.3) and IBM SPSS® Statistics software version 27.0 statistical programmes. Continuous variables were assessed for the distribution of normality with normality plots. Student’s t-test was used to compare means between groups for continuous variables with normally distributed data. Mann–Whitney was used to compare medians between groups for continuous variables with non-normally distributed data. Chi-squared test was used to compare groups for categorical variables. Spearman correlation test was used for non-normally distributed data. Binomial logistic regression test was used to analyse the odds ratio for ETD based on the SNOT-22 ear domain score. Youden index was used to interpret and evaluate the receiver operator curve (ROC) for the SNOT-22 ear domain as a predictor for ETD based on ETDQ score of greater than or equal to 14.5. Statistical significance was determined with a two-tailed alpha error of 0.05.
Results
A total of 94 patients were included in the study – 53 in the CRSwNP group and 41 in the control group. The difference in various demographic variables, T2 markers, SNOT-22 and ETDQ-7 scores and proportion of ETD between the groups is reported in Table 1. CRSwNP patients were significantly older, had a higher proportion of males and had higher levels of T2 markers. There was no significant difference in SNOT-22, ETDQ-7 and the proportion of patients with ETD between the groups.
Demographics between CRSwNP and control group
CRSwNP = chronic rhinosinusitis with nasal polyps; ETDQ-7 = seven-item Eustachian Tube Dysfunction Questionnaire; IgE = immunoglobulin E; IQR = interquartile range; NA = not applicable; PBE = peripheral blood eosinophils; SNOT-22 = 22-item Sino-Nasal Outcome Test.
Correlations between ETDQ-7 and SNOT-22, SNOT-5 and SNOT-22 ear domain were statistically significant in all CRSwNP patients and CRSwNP patients with ETD, according to an ETDQ-7 cutoff of greater than or equal to 14.5 (Table 2).
ETDQ-7 correlation with different variables in CRSwNP
CRSwNP = chronic rhinosinusitis with nasal polyps; ETDQ-7 = seven-item Eustachian Tube Dysfunction Questionnaire; IgE = immunoglobulin E; LMS = Lund-Mackay Score; NPS = Nasal Polyp Score; PBE = peripheral blood eosinophils; SNOT-5 = five-item Sino-Nasal Outcome Test; SNOT-22 = 22-item Sino-Nasal Outcome Test.
CRSwNP patients with ETD had a higher median SNOT-22 score of 55 (interquartile range [IQR] 44–67.5) vs 28.5 (IQR 20–43.8), p less than 0.001; a higher median SNOT-5 score of 15 (IQR 11 – 19.5) vs 10.5 (IQR 8.25–15.0), p less than 0.05; a higher median SNOT-22 ear domain score of 4 (IQR 2-6) vs 1 (IQR 0-2.75), p less than 0.001, compared with CRSwNP patients without ETD. There were no significant differences in LMS (p = 0.49), NPS (p = 0.13), PBE (p = 0.38) and total IgE (p = 0.77) between CRSwNP patients with and without ETD.
No differences were observed in CRSwNP patients regarding the proportion of patients with ETD based on different NPS or LMS cutoff points or PBE greater than or equal to 300/µL and IgE greater than or equal to 100 kU/L.
ETDQ-7 was not significantly different between CRSwNP patients based on asthma co-morbidity, T2 biomarkers (PBE ≥ 300 or IgE ≥ 100), history of previous ESS, allergy defined by a raised specific IgE to at least one aeroallergen, smoking history, BMI or LMS; but it was significantly different based on an NPS cutoff greater than or equal to 6: 22 (IQR 13–29) vs 14 (IQR 9–21), p less than 0.05.
There was an increased likelihood for the presence of CRSwNP when PBE was greater than or equal to 300/µL: odds ratio (OR) = 8.38 (95 per cent confidence interval [CI] 2.34-29.95), p = 0.001 and when the total IgE was ≥100 kU/L: OR = 10.74 (95 per cent CI 2.42-47.72), p = 0.002.
The SNOT-22 ear domain score is a good predictor for ETD (ETDQ ≥ 14.5) AUC = 0.76 (95 per cent CI 0.63-0.89), sensitivity of 0.67, specificity of 0.73 (Figure 1). The SNOT-22 ear domain score of greater than or equal to 4 was associated with a greater likelihood for ETDQ score greater than or equal to 14.5: OR = 1.57 (95 per cent CI 1.18-2.09), p = 0.002.
Receiver operating characteristics (ROC) curves.
Discussion
The ETDQ-7 is a questionnaire which, validated by McCoul et al., enables clinicians to identify patients with or without ETD.Reference McCoul, Anand and Christos8 Other investigators have reported a prevalence of ETD in 25 per cent and 48.5 per cent of CRS patients.Reference Chen, Yang, Lin, Fang, Wu and Luo3, Reference Tangbumrungtham, Patel, Thamboo, Patel, Nayak and Ma7 In our cohort of CRSwNP the prevalence of ETD was 50.9 per cent, and, of interest, 63.4 per cent in the control group who presented with CRS symptoms but had no underlying sinus disease on imaging along with no polyps at endoscopy.
Statistically significant correlations between ETDQ-7 and SNOT-22 have previously been reported, but not in relation to CRS disease severity as evaluated by the LMS.Reference Chen, Shih, Edwards, Nguyen, Meyer and Soler1, Reference Chen, Yang, Lin, Fang, Wu and Luo3 We report similar findings here but in addition a correlation between ETDQ-7 and SNOT-5.
Despite SNOT-22 incorporating ear domains and having moderate specificity and sensitivity for detecting ETD, the ETDQ-7 performs better as a disease-specific tool.Reference McCoul, Anand and Christos8 Nevertheless, we have shown that the SNOT-22 ear domain score can be used as a guide to suspect ETD.
ETDQ-7 scores may improve with medical and surgical management of CRS.Reference Adams, Derbarsegian, Sedaghat and Phillips4, Reference Chang, Roozdar, Lin, Lee, Nayak and Patel11 Wuraola et al. reported that types B and C tympanograms were significantly more common in CRS compared to healthy controls who had no tympanograms suggestive of ETD.Reference Wuraola, Afolabi and Ologe5 Unfortunately, our cohort did not have tympanograms done as, anecdotally, in our cohort there were more patients with ETD symptoms in the control group compared to the CRSwNP group. The presence of ETD symptoms in our control group could be attributable to allergic or non-allergic rhinitis.
Chen et al. reported no difference in SNOT-22 and ETDQ-7 between CRS patients with and without nasal polyps.Reference Chen, Shih, Edwards, Nguyen, Meyer and Soler1 Of interest, our results show no difference in SNOT-22 and ETDQ-7 between the CRSwNP and non-CRS control group presenting with CRS symptoms. Chen et al. suggested the degree of ETD was independent of nasal polyp status, while our data show a correlation between and difference in ETDQ-7 based on NPS in CRSwNP. We suggest that the degree of ETD may in part be independent of CRS status, especially since our non-CRS patients presented with high CRS symptom burden in conjunction with high score ETDQ-7 scores. This in turn points to the need to always perform sino-nasal endoscopy and CT scan in patients presenting with high SNOT-22 and ETDQ-7 scores to identify underlying disease pathology to guide optimal medical and surgical management strategies.
We appreciate the limitations of our data, in particular, the patients did not undergo otoscopy or tympanometry as they were seen in a dedicated rhinology clinic and, therefore, we do not have information on the objective findings of ETD in the study population which are required to diagnose negative pressure in the middle ear.Reference Schilder, Bhutta, Butler, Holy, Levine and Kvaerner2
• Eustachian tube dysfunction (ETD) is prevalent in patients with chronic rhinosinusitis with nasal polyps (CRSwNP)
• The seven-item Eustachian Tube Dysfunction Questionnaire (ETDQ-7) and 22-item Sino Nasal Outcome Test (SNOT-22) questionnaires are validated tools for the quantitative evaluation of ETD and chronic rhinosinusitis, respectively
• An ETDQ-7 score of greater than or equal to 14.5 is indicative of ETD
• ETDQ-7 correlates with SNOT-22, SNOT-5 and SNOT-22 ear domain scores
• The clinical relevance of the ETDQ-7 questionnaire in CRSwNP is rather debatable as our findings reveal the presence of elevated scores not only in CRSwNP but also in the control group who had no sinus or polyp disease
Conclusion
The clinical relevance of the ETDQ-7 questionnaire in CRSwNP is rather debatable as our findings reveal the presence of elevated scores not only in CRSwNP but also in the control group who had no sinus or polyp disease. Likely ETDQ-7 is not particularly helpful in evaluating ETD symptoms in patients with high SNOT-22 scores. Further studies might evaluate ETD with ETDQ-7 in relation to tympanograms in patients with CRSwNP including response to treatment.
Acknowledgements
We acknowledge the administrative staff helping us to obtain notes from medical records: Andrew Low, Aneliya Velinova and Stuart Davidson.
Funding
This research received no specific grant from any funding agency, commercial or not-for-profit sectors.
