Interplay Between Rhinosinusitis and Eustachian Tube Dysfunction in Patients Undergoing Functional Endoscopic Sinus Surgery

Abstract

Objectives

To evaluate the association between chronic rhinosinusitis (CRS) and Eustachian tube dysfunction (ETD) using subjective and objective measures, and the relationship between disease laterality and Eustachian tube function (ETF).

Study Design

Prospective cohort study.

Setting

Single tertiary medical institution.

Methods

Eighty‐five CRS patients who were candidates for FESS underwent comprehensive evaluation. Subjective assessments utilized the Eustachian Tube Dysfunction Questionnaire‐7 (ETDQ‐7) and Sino‐Nasal Outcome Test‐22 (SNOT‐22). Objective evaluations comprised a 9‐step inflation–deflation tympanometric test scored by maximal peak pressure difference (MPD), nasal endoscopy using the modified Lund‐Kennedy (MLK) system, sinus CT scored via the Lund‐Mackay (LM) system, and acoustic rhinometry measuring the second minimal cross‐sectional area (MCA₂). Patients were stratified into fair (<14) and poor (≥14) ETF groups based on ETDQ‐7 scores. Correlations were analyzed using Pearson coefficients, receiver operating characteristic (ROC) curves, and Bland‐Altman plots, with additional assessment of CRS lateralization effects.

Results

Poor ETF (ETDQ‐7 ≥ 14) occurred in 36.5% of patients and was associated with higher SNOT‐22 (median 48.0 vs 31.5, P < .001). ETDQ‐7 correlated moderately with SNOT‐22 total (r = 0.65). No significant differences in MPD, MLK, LM, or MCA₂ were found between fair and poor ET function groups. ROC analysis identified optimal cut‐offs of ETDQ‐7 > 15 (AUC 0.761) and SNOT‐22 > 45 (AUC 0.802). Laterality analysis showed no significant side‐specific associations.

Conclusion

In CRS, sinonasal symptom burden shows only a moderate association with patient‐reported ETD. Higher ETDQ‐7 scores often parallel greater SNOT‐22 severity, suggesting that PROMs may help identify patients in whom ETD symptoms are partly influenced by CRS.

The Eustachian tube (ET) connects the middle ear to the nasopharynx, playing a key role in middle ear ventilation, pressure equalization, mucociliary clearance, and protection from nasopharyngeal reflux. Eustachian tube dysfunction (ETD) can lead to symptoms such as negative middle ear pressure, aural fullness, muffled hearing, hearing loss, popping sensations, tinnitus, ear pain, sharp otalgia, autophony, disequilibrium, and reduced quality of life. ¹ , ² , ³

ETD is a common comorbidity in chronic rhinosinusitis (CRS). Research shows that 15% to 68% of CRS patients experience ETD symptoms, highlighting the close anatomical and functional connection between the sinonasal cavity and the ET. ⁴ , ⁵ Mucosal edema and secretions associated with CRS are thought to impair the ET's ability to regulate pressure and facilitate clearance, thereby contributing to secondary ETD. ² , ⁶ A multicenter prospective study evaluated the effects of endoscopic sinus surgery on Eustachian tube function (ETF) in CRS patients, revealing significant improvements in ETF and alleviation of symptoms like ear fullness and pressure imbalance. ⁷

Despite numerous studies investigating the link between CRS and ETD, their relationship remains poorly understood. The absence of diagnostic consensus, standardized assessment tools, and a clear understanding of secondary ETD pathophysiology underscores the need for targeted research. ³ , ⁴ Patient‐reported outcomes, like the Eustachian Tube Dysfunction Questionnaire (ETDQ‐7), offer valuable insight into symptom severity but need to be complemented with objective tools such as tympanometry and nasal endoscopy for a more comprehensive assessment. ³ , ⁸

The relationship between sinusitis and ETD laterality is a complex and underexplored area in otolaryngology. While sinus pathologies are known to contribute to ETD, the extent of lateralization—whether one side is more affected than the other—remains unclear. This prompts important questions about whether unilateral sinus disease aligns with ipsilateral ETD or if bilateral symptoms suggest a more systemic or diffuse pathology. ⁹ Understanding this relationship is vital for improving diagnostic accuracy and tailoring treatment approaches based on the laterality of disease presentation. ¹⁰ Additionally, better comprehension of this relationship could refine the use of diagnostic tools, such as imaging and endoscopy, to identify and address the underlying causes of ETD more precisely. ¹¹ , ¹² , ¹³

This study investigates the link between ETD and CRS by assessing the prevalence and clinical features of ETD in CRS patients using both subjective and objective diagnostic tools. By exploring the interaction between nasal/sinus inflammation and ETF, this research aims to fill knowledge gaps and guide future diagnostic approaches.

Methods

Study Population Selection and Protocol

This prospective cohort study was conducted at a single tertiary medical center in Taiwan between January 2020 and January 2024. A total of 85 consecutive adult patients (aged ≥18 years) with a confirmed diagnosis of CRS scheduled for functional endoscopic sinus surgery (FESS) were enrolled. Patients with a prior history of middle‐ear disease (eg, chronic otitis media, cholesteatoma, tympanic membrane perforation) were excluded to avoid confounding the assessment of CRS‐related ETD.

Baseline measurements of all subjective and objective parameters were obtained during the standard preoperative assessment for FESS at our institution, prior to any surgical intervention or perioperative medical therapy adjustments. Subjective evaluations included ETDQ‐7 and the Sino‐Nasal Outcome Test‐22 (SNOT‐22). Objective assessments were performed in the following sequence: Nine‐step inflation‐deflation tympanometry (also referred to as the Pressure‐Swallow Test or ETF Testing) was performed using the GSI TympStar Pro system to measure the maximal peak pressure difference (MPD). Nasal endoscopy was graded according to the modified Lund‐Kennedy (MLK) scoring system, while sinus CT scans were evaluated using the Lund‐Mackay (LM) system. In addition, acoustic rhinometry (ARM) was conducted to determine the second minimal cross‐sectional area (MCA₂), which reflects nasal airway patency. Severity grading was applied uniformly across subjective and objective measures (Figure 1). For patient‐reported outcomes, the SNOT‐22 total score was classified as mild (8‐20), moderate (21‐50), or severe (>50). Endoscopic findings based on the MLK score were categorized as mild (0‐4), moderate (5‐8), or severe (9‐12). Radiologic severity according to the LM score was defined as mild (0‐4), moderate (5‐12), or severe (13‐24). For ARM, MCA₂ was graded as mild (>0.5 cm²), moderate (0.3‐0.5 cm²), or severe (<0.3 cm²).

Figure 1.

Patient flow diagram. This flow diagram illustrates the enrollment and evaluation process of 85 adult patients with chronic rhinosinusitis undergoing preoperative assessment for functional endoscopic sinus surgery.

ETDQ‐7

The ETDQ‐7 assesses seven symptoms: ear pressure, ear pain, ear fullness or “underwater” sensation, ear symptoms during cold or sinusitis, crackling or popping, tinnitus, and muffled hearing. Each item is rated on a 1‐7 scale, where 1 denotes no issues and 7 represents the most severe symptoms over the past month. The ETDQ‐7 provides a total score ranging from 7 to 49. In this study, we used the validated Chinese version and adopted a score ≥14 as the threshold for symptomatic ETD, consistent with the cutoff reported in a Taiwanese validation study by Lin et al. ¹⁴ , ¹⁵

Sino‐Nasal Outcome Test

The SNOT‐22 addresses symptoms and functional impairments such as nasal obstruction, facial pain, olfactory dysfunction, sleep disturbances, and emotional well‐being. ¹⁶ , ¹⁷ Each item is rated on a 5‐point Likert scale (1 = very mild, 2 = mild, 3 = moderate, 4 = severe, 5 = worst possible), yielding a total score of 0 to 110.

Nine‐Step Inflation‐Deflation Tympanometric ETF Test

ETF was evaluated using the nine‐step inflation‐deflation tympanometric test with the GSI TympStar Pro system (Grason‐Stadler Inc.), also known as the Pressure‐Swallow Test. This is a device‐enabled modification of Bluestone's original nine‐step test (1975), which assessed ETF by applying positive and negative ear canal pressures and recording middle‐ear pressure changes during swallowing via tympanometry. ¹⁸ , ¹⁹ , ²⁰ In the modified protocol, the TympStar Pro automates pressure modulation, applies standardized positive (+400 daPa) and negative (−400 daPa) pressures, and captures tympanograms in real time under 3 conditions: resting, pressurized, and depressurized states. During the nonatmospheric phases, participants were instructed to swallow repeatedly to promote Eustachian tube opening and middle‐ear pressure equilibration. The primary metric was the MPD, with higher MPD values indicating better ETF. The test remains grounded in the same physiological principle: a normally functioning Eustachian tube will open during swallowing, producing a measurable pressure shift (typically 10‐15 daPa), ⁶ , ²⁰ , ²¹ , ²² whereas reduced or absent shifts indicate dysfunction. Participants failing postinflation or postdeflation step were classified as having poor ETF, while those passing both steps were deemed to have good results. An MPD value of 10 daPa was used as the cutoff, as suggested by McBride et al and Fernau et al. ⁶ , ²¹

Lund‐Kennedy Endoscopic Scoring System

The MLK endoscopic scoring system assesses 3 critical factors during nasal endoscopy ²³ , ²⁴ : Polyps: 0 = none; 1 = middle meatus only; 2 = beyond middle meatus. Edema: 0 = absent; 1 = mild; 2 = severe. Discharge: 0 = none; 1 = clear, thin; 2 = thick, purulent. Scoring: Each nasal cavity is scored separately, with a maximum of 6 points per side (2 points per category × 3 categories). Maximum Bilateral Score: 12 points (6 points per side × 2 sides).

Lund‐Mackay CT Scoring System

The LM CT system quantifies opacification in the paranasal sinuses and osteomeatal complex. Each sinus is scored 0 (none), 1 (partial), or 2 (complete), while the osteomeatal complex is rated 0 (patent) or 2 (obstructed). Unilateral scores range from 0 to 12 and bilateral from 0 to 24, comprehensively reflecting disease burden. ²⁵

Statistical Analysis

A priori power calculations were conducted to justify the adequacy of the sample size (n = 85) for detecting the expected correlations. Based on previous literature, an effect size of r = 0.50‐0.65 was anticipated. Priori analysis using a two‐tailed test, α = 0.05, and n = 85 demonstrated that the achieved power (1–β) was 1.000 for r = 0.65 and 0.926 for r = 0.35, indicating sufficient power (>80%) to detect correlations within the expected range.

Demographics (age, sex), SNOT‐22 scores, and objective measures (MPD by nine‐step tympanometry, MLK endoscopic score, LM CT score, and MCA₂) were recorded. Scores were categorized as mild, moderate, or severe (SNOT‐22: 8‐20, 21‐50, >50; MLK: 0‐4, 5‐8, 9‐12; LM: 0‐4, 5‐12, 13‐24; MCA₂: >0.5 cm², 0.3‐0.5 cm², <0.3 cm²).

Statistical analyses were performed using SPSS (version 22.0; IBM Corp.), and G*Power (version 3.1). Continuous variables were assessed for normality and reported as median (IQR), whereas categorical variables were summarized as n (%). Between‐group differences were examined using χ², Fisher's exact, or Mann‐Whitney U tests, and repeated binary outcomes were analyzed with Cochran's Q test. Pearson correlation coefficients with 95% confidence intervals were calculated to assess associations between ETDQ‐7 and SNOT‐22 scores, while Bland‐Altman analysis was employed to evaluate agreement and identify potential proportional bias, providing complementary information to correlation analysis. Receiver operating characteristic (ROC) curves were generated to determine the discriminatory performance of ETDQ‐7 and SNOT‐22, with the area under the curve (AUC), optimal cut‐off values, sensitivity, specificity, and likelihood ratios reported.

Agreement between subjective (ETDQ‐7, SNOT‐22) and objective (MPD, MLK, LM, MCA₂) measures was further evaluated using Bland‐Altman plots. Although traditionally applied for comparing two different measurement methods, Bland‐Altman analysis was used here to complement correlation analysis by quantifying the degree of agreement and identifying systematic bias or limits of agreement between scales. This approach avoids the limitation of correlation coefficients, which measure association strength but not interchangeability.

This study was approved by the Institutional Review Board of Taichung Veterans General Hospital (IRB No. CE23522C), and written informed consent was obtained from all participants.

Results

Figure 1 illustrates the enrollment and evaluation process of 85 adult CRS patients undergoing preoperative assessment for FESS. Patients were stratified into two groups according to ETDQ‐7 scores: a fair ETF group (ETDQ‐7 < 14, n = 54) and a poor ETF group (ETDQ‐7 ≥ 14, n = 31). ¹⁵ Both groups underwent standardized evaluations, including nasal endoscopy (MLK), sinus CT (LM), ARM (MCA₂), and nine‐step tympanometry (MDP). Subjective disease burden was also quantified using SNOT‐22, with uniform severity grading applied across all measures. Median age was comparable between groups (49.5 vs 53.0 years, P = .340), as was the female distribution (37.04% vs 41.94%, P = .656) (Table 1).

Table 1.

Comparison of Demographic Characteristics Between Fair and Poor ETF Groups

		ETDQ‐7 score
	Total (N = 85)	<14 (n = 54)	≥14 (n = 31)	P value
Age, median (IQR)	51.00 (39.00‐59.5)	49.50 (36.75‐59.00)	53.00 (43.00‐63.00)	.340
Female, n (%)	33 (38.82%)	20 (37.04%)	13 (41.94%)	.656
SNOT‐22 score, median (IQR)	35.00 (26.00‐51.5)	31.5 (24‐46.25)	48.00 (34.00‐68.00)	<.001**
Sinonasal component score of SNOT‐22, median (IQR)	16.00 (10.00‐20.00)	13.00 (7.00‐18.25)	17.00 (14.00‐23.00)	.002**
Nine‐step test MPD value, median (IQR)	8.50 (3.00‐12.50)	9.25 (4.25‐14.00)	7.50 (2.50‐12.00)	.187
Modified Lund‐Kennedy score, median (IQR)	5.00 (4.00‐6.00)	5.00 (3.75‐6.00)	5.00 (4.00‐6.00)	.806
Lund‐Mackay score, median (IQR)	10.00 (6.00‐14.50)	10.00 (6.00‐15.00)	11.00 (4.00‐14.00)	.615
MCA2 value, median (IQR)	0.22 (0.15‐0.41)	0.23 (0.17‐0.41)	0.22 (0.13‐0.41)	.805
Subgroup characteristics by score‐defined severities
SNOT‐22, n (%)				.004**
Mild	10 (11.76%)	9 (16.67%)	1 (3.23%)
Moderate	53 (62.35%)	37 (68.52%)	16 (51.61%)
Severe	22 (25.88%)	8 (14.81%)	14 (45.16%)
SNOT‐22_Severe, n (%)	22 (25.88%)	8 (14.81%)	14 (45.16%)	.002**
SNOT‐22_Moderate/Severe, n (%)	75 (88.24%)	45 (83.33%)	30 (96.77%)	.085
Modified Lund‐Kennedy score, n (%)				.128
Mild	37 (43.53%)	24 (44.44%)	13 (41.94%)
Moderate	42 (49.41%)	24 (44.44%)	18 (58.06%)
Severe	6 (7.06%)	6 (11.11%)	0 (0%)
Modified Lund‐Kennedy score, Severe, n (%)	6 (7.06%)	6 (11.11%)	0 (0%)	.082
Modified Lund‐Kennedy score, Moderate/Severe, n (%)	48 (56.47%)	30 (55.56%)	18 (58.06%)	.822
Modified Lund‐Mackay score, n (%)				.365
Mild	16 (18.82%)	8 (14.81%)	8 (25.81%)
Moderate	40 (47.06%)	28 (51.85%)	12 (38.71%)
Severe	29 (34.12%)	18 (33.33%)	11 (35.48%)
Lund‐Mackay score, Severe, n (%)	29 (34.12%)	18 (33.33%)	11 (35.48%)	.840
Lund‐Mackay score, Moderate/Severe, n (%)	69 (81.18%)	46 (85.19%)	23 (74.19%)	.212
MCA2 value, n (%)				.964
Mild	10 (11.76%)	6 (11.11%)	4 (12.9%)
Moderate	20 (23.53%)	13 (24.07%)	7 (22.58%)
Severe	55 (64.71%)	35 (64.81%)	20 (64.52%)
MCA2 value, severe, n (%)	55 (64.71%)	35 (64.81%)	20 (64.52%)	.978
MCA2 value, moderate/severe, n (%)	75 (88.24%)	48 (88.89%)	27 (87.1%)	1.000
ETDQ‐7 score, median (IQR)	11 (7‐17)	8 (7‐11)	18 (16‐33)	<.001**
ETDQ‐7 score ≥ 14, n (%)	31 (36.47%)

Abbreviations: ETDQ‐7, seven‐item Eustachian Tube Dysfunction Questionnaire; ETF, Eustachian tube function; IQD, interquartile range; MCA₂, the second minimal cross‐sectional area; MPD, maximal peak pressure difference; Nine‐step test, nine‐step inflation‐deflation tympanometric Eustachian tube function test; SNOT‐22, twenty‐two item Sino‐Nasal Outcome Test. Statistical analyses were two‐sided, with significance thresholds set at

^** P < .01.

Thirty‐one patients with poor ETF exhibited significantly higher SNOT‐22 scores (median 48.0 vs 31.5, P < .001) and elevated sinonasal component score of SNOT‐22 (median 17.0 vs 13.0, P = .002). Conversely, objective measures—including MPD (median 7.50 vs 9.25, P = .187), MLK score (median 5.0 in both groups, P = .806), LM score (median 11.0 vs 10.0, P = .615), and MCA₂ (median 0.22 in both groups, P = .805)—did not differ significantly (Table 1).

CRS patients were divided into fair and poor ETF subgroups according to ETDQ‐7‐based stratification. The median ETDQ‐7 score was 8 in the fair ETF group and 18 in the poor ETF group. This clear separation in symptom scores indicates that ETDQ‐7 effectively distinguishes different levels of patient‐reported ET symptom burden within the CRS cohort, enabling subsequent analyses of the relationship between sinonasal disease severity and subjective ETD manifestations. A severity subgroup analysis revealed a higher proportion of participants with poor ETF in the severe SNOT‐22 category compared to those with fair ETF (45.16% vs 14.81%, P = .002), indicating more pronounced sinonasal symptoms in the poor ETF group. Conversely, MLK system severity distribution showed no significant difference between groups, with moderate/severe categories present in 58.06% of the poor ETF group and 55.56% of the fair ETF group (P = .822). Similarly, the severe LM system category proportions were comparable (35.48% vs 33.33%, P = .840). These results suggest that while subjective sinonasal symptom severity is markedly worse in the poor ETF group, objective sinonasal assessments and anatomical metrics remain similar between fair and poor ETF groups (Table 1).

ETDQ‐7 scores showed a moderate correlation with the SNOT‐22 total score (r = 0.65, P < .001), suggesting an association between subjective ET function and sinonasal symptom severity.

ROC analysis demonstrated that both ETDQ‐7 and SNOT‐22 scores effectively discriminated clinical outcomes, with optimal thresholds determined by maximizing the Youden Index. For ETDQ‐7, using severe SNOT‐22 (>50) as the classification variable, the AUC was 0.761 (95% CI: 0.656‐0.847, P = .0001), indicating fair diagnostic accuracy; a cut‐off >15 yielded 63.64% sensitivity and 80.95% specificity (+LR 3.34, −LR 0.45). For SNOT‐22, classified by ETDQ‐7 > 15, the AUC was 0.802 (95% CI: 0.701‐0.880, P < .0001), indicating good diagnostic accuracy; a cut‐off >45 provided 73.08% sensitivity and 76.27% specificity (+LR 3.08, −LR 0.35). These cutoffs align with prior CRS–ETD literature, in which SNOT‐22 scores exceeding approximately 45 points and ETDQ‐7 scores exceeding about 15 points have been associated with clinically significant disease burden. ¹⁴ , ²⁶ Both scores demonstrated relatively strong negative predictive values, supporting their utility in excluding severe disease (Figure 2).

Figure 2.

Stratification of SNOT‐22 and ETDQ‐7 scores for severe and nonsevere classification by Receiver Operating Characteristic curve analysis. ETDQ‐7, seven‐item Eustachian Tube Dysfunction Questionnaire; SNOT‐22, twenty‐two item Sino‐Nasal Outcome Test.

Correlations between ETF metrics and sinonasal outcomes varied. The MPD from the nine‐step test did not significantly correlate with subjective scores (ETDQ‐7: r = −0.08, P = .45) (SNOT‐22 total: r = −0.15, P = .167; sinonasal component: r = −0.15, P = .174), indicating limited reflection of self‐reported symptom severity. Conversely, the SNOT‐22 sinonasal component showed a modest correlation with LM scores (r = 0.26, P = .015), indicating a limited association with radiologic disease severity. LM scores were moderately correlated with MLK scores (r = 0.53, P < .001), reflecting a moderate level of consistency between radiologic and endoscopic assessments of sinonasal pathology. Although LM and ETDQ‐7 scores showed a statistically significant but weak negative linear correlation (r = –0.26, P = .017), the nonnormal distribution and low effect size indicate minimal explanatory value, consistent with the absence of between‐group differences observed in Table 1 (Table 2).

Table 2.

Pearson Correlation Between Eustachian Tube Function Test and Sinonasal Measurements

	ETDQ‐7 score		Nine‐step test MPD value		SNOT‐22 score		Sinonasal component score of SNOT‐22 score		Modified Lund‐Kennedy score		Lund‐Mackay score		MCA2 value
	r	P value	r	P value	r	P value	r	P value	r	P value	r	P value	r	P value
ETDQ‐7 score	1.00	‐	−0.08	.450	0.65	<.001**	0.50	<.001**	−0.15	.169	−0.26	.017*	0.04	.722
Nine‐step test MPD value	−0.08	.450	1.00	‐	−0.15	.167	‐0.15	.174	0.13	.234	0.05	.664	−0.13	.248
SNOT‐22 score	0.65	<.001**	−0.15	.167	1.00	‐	0.79	<.001**	0.03	.788	0.08	.459	−0.16	.138
Sinonasal component of SNOT‐22 score	0.50	<.001**	−0.15	.174	0.79	<.001**	1.00	‐	0.16	.153	0.26	.015*	−0.09	.438
Lund‐Kennedyscore	−0.15	.169	0.13	.234	0.03	.788	0.16	.153	1.00	‐	0.53	<.001**	−0.06	.616
Lund‐Mackay score	−0.26	.017*	0.05	.664	0.08	.459	0.26	.015*	0.53	<.001**	1.00	‐	−0.06	.570
MCA2 value	0.04	.722	−0.13	.248	−0.16	.138	−0.09	.438	−0.06	.616	−0.06	.570	1.00	‐

Abbreviations: ETDQ‐7, seven‐item Eustachian Tube Dysfunction Questionnaire; MCA₂, the second minimal cross‐sectional area; MPD, maximal peak pressure difference; Nine‐step test, nine‐step inflation‐deflation tympanometric Eustachian tube function test; SNOT‐22, twenty‐two item Sino‐Nasal Outcome Test.

Statistical analyses were two‐sided, with significance thresholds set at

^* P < .05 and

^** P < .01.

MPD values did not significantly correlate with MLK scores (r = 0.13, P = .234), LM scores (r = 0.05, P = .664), or MCA₂ (r = −0.13, P = .248), suggesting that MPD may capture unique ETF aspects not reflected by these sinonasal metrics. Similarly, MCA₂ showed no significant correlation with any sinonasal measurement, underscoring the independent evaluation of sinonasal and ETF function (Table 2).

The Bland‐Altman analysis comparing ETDQ‐7 and SNOT‐22 scores (Figure 3) showed a mean difference of 0.00 (95% CI, –0.1806 to 0.1806; P = 1.000), indicating no systematic bias and relatively narrow limits of agreement (±1.6). Most values were within these limits, suggesting that the two questionnaires yield comparable average scores in patients with CRS.

Figure 3.

Comparison of the SNOT‐22 and ETDQ‐7 scores by Bland‐Altman plots assessment. ETDQ‐7, seven‐item Eustachian Tube Dysfunction Questionnaire; SNOT‐22, twenty‐two item Sino‐Nasal Outcome Test.

The distribution of laterality for objective ETF and unilateral sinonasal measurements is summarized in Table 3. The proportions of left‐ and right‐sided findings were comparable across the nine‐step test MPD values (47.06% vs 52.94%), MLK scores (50.59% vs 49.41%), and LM scores (47.06% vs 52.94%). For MCA₂ values, a slightly higher proportion was observed on the left side compared with the right (56.47% vs 43.53%). No statistically significant difference in laterality was detected for any of these measures (P = .461). Further laterality analysis revealed no significant association between the nine‐step test MPD value and sinonasal parameters, including the MLK score, LM score, and MCA₂ (P = .660).

Table 3.

Association Between the Objective Eustachian Tube Function Test and Unilateral Sinonasal Measurements

	Nine‐step test MPD value	Modified Lund‐Kennedy score	Lund‐Mackay score	MCA2 value	P value
Laterality, n (%)					.660
Left	40 (47.06%)	43 (50.59%)	40 (47.06%)	48 (56.47%)
Right	45 (52.94%)	42 (49.41%)	45 (52.94%)	37 (43.53%)

Abbreviations: MCA₂, the second minimal cross‐sectional area; MPD, maximal peak pressure difference; Nine‐step test, nine‐step inflation‐deflation tympanometric Eustachian tube function test.

Statistical analyses were two‐sided, with significance thresholds set at *P < .05.

Discussion

This study evaluated the association between CRS severity and ETD through a combination of validated PROMs and objective diagnostic assessments. Consistent with the meta‐analysis by Chen et al, ⁴ which reported a pooled ETD prevalence of 47.2% among CRS patients, we observed a similarly high burden of sinonasal and aural symptoms. Notably, more severe sinonasal symptom scores demonstrated a moderate‐to‐strong correlation with concurrent ETD, suggesting a clinically relevant link that warrants further investigation.

While consensus exists that CRS may contribute to ETD, this association is largely supported by observational data, including the high ETD prevalence in CRS populations and the symptomatic relief of ETD following CRS treatment. ⁴ , ⁵ , ⁷ , ¹⁰ , ¹³ In our study, a key observation was the robust correlation between ETDQ‐7 and SNOT‐22 scores (r = 0.65, as shown in Tables 1 and 2, consistent with earlier reports (r = 0.46‐0.57), reinforcing the utility of PROMs in reflecting sinonasal symptom burden. Patients classified as having poor ETF based on ETDQ‐7 scores showed higher total and sinonasal SNOT‐22 scores, suggesting that greater CRS‐related symptom burden may be associated with increased perceived ETD. This strong association aligns with previous studies, emphasizing the utility of subjective assessments in gauging the effect of sinonasal disease on ETF. ²⁷

None of the four objective measures (MPD, MLK, LM, MCA₂) significantly differentiated between fair and poor ETF groups or demonstrated meaningful correlations with PROMs. In particular, MPD values showed no significant association with objective sinonasal parameters, including MLK scores, LM scores, and MCA₂ values (Table 2). This discrepancy suggests that although MPD objectively assesses ETF, it may not adequately capture the complex and dynamic physiological mechanisms governing ETF regulation and function. ETD represents a multifactorial and dynamic disease spectrum influenced by factors beyond rhinosinusitis, such as acid reflux and pterygoid muscle hypertrophy. ¹³ , ²⁸ , ²⁹ , ³⁰ This aligns with Bluestone's description of ETD as a multifactorial and temporally variable process influenced by mucosal inflammation, neuromuscular control, and extrinsic factors such as allergy or reflux. ¹ , ³¹ These findings underscore the limitations of relying on a single static tympanometric measurement and highlight the need for innovative diagnostic tools that integrate subjective symptom evaluation with dynamic, objective metrics to more comprehensively assess ETD. ³ , ³²

As detailed in Table 3, laterality analysis revealed no significant side‐specific association between CRS disease metrics and ETF, arguing against a simple ipsilateral effect. Clinically, this underscores that even in unilateral CRS, ETD screening should encompass both ears. While patients with higher sinonasal symptom scores were more likely to report ETD symptoms, the lack of strong correlations between objective sinonasal parameters and objective ETF measures tempers the inference that sinonasal disease severity directly causes measurable ET dysfunction. Although our analysis demonstrated only modest associations between sinonasal pathology and ETD, these findings provide a solid basis for future investigations into their complex relationship.

From a clinical standpoint, these findings suggest that routine ETD screening in CRS—particularly in patients with high SNOT‐22 scores—may help identify patients who could benefit from targeted evaluation or concurrent management. However, given the limitations of current objective tools, reliance on a single physiologic test such as MPD is insufficient. We adhered to the STROBE reporting guidelines, which strengthens the transparency and reproducibility of this study. (Supplemental S1, available online).

Limitations

Several methodological considerations temper our conclusions. First, Patients with prior middle‐ear disease were excluded to reduce confounding from pre‐existing ET pathology, which enhanced internal validity but may have introduced selection bias by omitting those with more severe or complex ETD. Second, potential confounders—such as allergic rhinitis, ³³ laryngopharyngeal reflux, craniofacial variation, and other comorbid conditions—were not systematically assessed and may have influenced outcomes. Third, the modest sample size (n = 85), absence of a formal power calculation, and lack of an age‐ and sex‐matched control group limit the precision of subgroup analyses, including those for CRSwNP versus CRSsNP, and constrain causal inferences regarding the CRS–ETD relationship.

Conclusion

This prospective study supports the value of incorporating routine CRS assessment in patients with ETD using validated PROMs. The moderate association between ETDQ‐7 and SNOT‐22 indicates that elevated ETDQ‐7 scores frequently coincide with a higher sinonasal symptom burden, suggesting that PROM‐based evaluation may help clinicians identify cases in which ETD symptoms are partly driven by underlying CRS. Nevertheless, the limited concordance between subjective symptom scores and objective ET function measures underscores the need for a multimodal diagnostic approach in this population. ROC analysis demonstrated fair to good diagnostic accuracy, with clinically meaningful thresholds of ETDQ‐7 > 15 and SNOT‐22 > 45, consistent with prior CRS–ETD literature. These conclusions should be interpreted in light of certain limitations, including exclusion of patients with prior middle‐ear disease, restriction to surgical candidates, and a modest sample size.

Author Contributions

Jing‐Jie Wang, design, interpretation of the research, manuscript writing; Chien‐Hsiang Weng, interpretation of the research, manuscript writing; Yi‐Ting Yao, data processing; Rong‐San Jiang, design, conduct.

Disclosures

Competing interests

None.

Funding source

None.

Supporting information

Supplement 1. STROBE Statement Checklist.

We adhered to the STROBE reporting guidelines, which strengthens the transparency and reproducibility of this study.