Introduction
Eosinophilic chronic rhinosinusitis is a refractory inflammatory subtype of chronic rhinosinusitis with nasal polyps, characterised by type 2 inflammation, frequent recurrence and significant impairment of quality of life (QoL).Reference Tokunaga, Sakashita, Haruna, Asaka, Takeno and Ikeda1–Reference Stevens, Schleimer and Kern4 Patients with eosinophilic chronic rhinosinusitis often experience persistent nasal obstruction, sticky rhinorrhoea and loss of smell. In particular, olfactory dysfunction is one of the most debilitating symptoms affecting daily activities and patients’ satisfaction with treatment.Reference DeConde and Soler5, Reference Hummel, Whitcroft, Andrews, Altundag, Cinghi and Costanzo6
Current international guidelines and consensus statements recommend a stepwise management strategy for chronic rhinosinusitis with nasal polyps, integrating topical corticosteroids as baseline therapy, short courses of systemic corticosteroids when indicated and functional endoscopic sinus surgery (FESS) for patients with insufficient disease control.Reference Fokkens, Lund, Hopkins, Hellings, Kern and Reitsma7–Reference Rosenfeld, Piccirillo, Chandrasekhar, Brook, Ashok Kumar and Kramper9 More recently, biologic therapies targeting type 2 inflammation have been introduced for patients with severe, uncontrolled disease, particularly in those with post-operative recurrence, systemic type 2 inflammatory burden or steroid dependency.Reference Orlandi, Kingdom, Smith, Bleier, DeConde and Luong8–Reference Fokkens, Lund, Bachert, Mullol, Bjermer and Bousquet10
Type 2 inflammation in eosinophilic chronic rhinosinusitis is driven by cytokines such as interleukin (IL)-4, IL-5 and IL-13, which promote eosinophilic inflammation and mucosal dysfunction.Reference Fujieda, Imoto, Kato, Ninomiya, Tokunaga and Tsutsumiuchi2, Reference Kato3, Reference Fokkens, Lund, Hopkins, Hellings, Kern and Reitsma7 In Japan, dupilumab and mepolizumab have recently become available for the treatment of refractory chronic rhinosinusitis with nasal polyps.Reference Fujieda, Matsune, Takeno, Asako, Takeuchi and Fujita11, Reference Fujieda, Wang, Yoshikawa, Asako, Suzaki and Bachert12
Dupilumab blocks the IL-4 receptor alpha subunit, thereby inhibiting both IL-4 and IL-13 signalling and suppressing type 2 inflammation, with consequent improvement in sinonasal inflammation and olfactory function.Reference Bachert, Mannent, Naclerio, Mullol, Ferguson and Gevaert13–Reference Jonstam, Swanson, Mannent, Cardell, Tian and Wang15 Type 2 inflammation driven by IL-4 and IL-13 has been strongly associated with olfactory dysfunction in patients with chronic rhinosinusitis.Reference Stevens, Peters, Tan, Klingler, Poposki and Hulse16 Accordingly, blockade of IL-4 and IL-13 signalling with dupilumab provides a biological mechanism for olfactory improvement by directly suppressing the type 2 inflammatory pathways implicated in olfactory dysfunction.
Olfactory dysfunction has gained increasing attention as a patient-centred outcome and is often emphasised during discussions regarding treatment escalation.Reference DeConde and Soler5, Reference Whitcroft and Hummel17 Clinical trials of biologic therapies have demonstrated significant improvements in olfactory outcomes, leading to a perception that biologics may be particularly advantageous for smell recovery.Reference Fujieda, Matsune, Takeno, Asako, Takeuchi and Fujita11, Reference Fujieda, Wang, Yoshikawa, Asako, Suzaki and Bachert12 Surgical treatment also plays a key role in interrupting the vicious cycle of inflammation. Functional endoscopic sinus surgery has certainly been shown to improve olfactory function by reducing local inflammatory load and restoring sinonasal ventilation.Reference Saito, Fushimi, Hirose, Haruna and Tsuzuki18 However, the magnitude and durability of olfactory improvement may vary among patients with eosinophilic chronic rhinosinusitis.Reference Oka, Tsuzuki, Takebayashi, Kojima, Daimon and Sakagami19–Reference Zhao, Chen and Tang21
In real-world clinical practice, biologic therapy is rarely used as a primary treatment in isolation. Instead, it is typically introduced after primary FESS in patients who develop post-operative recurrence or exhibit persistent disease activity despite standard post-operative medical management.Reference Orlandi, Kingdom, Smith, Bleier, DeConde and Luong8, Reference Fokkens, Lund, Bachert, Mullol, Bjermer and Bousquet10, Reference Ramkumar, Lal and Miglani22 Nevertheless, direct comparisons of olfactory outcomes between patients with well-controlled post-operative courses after FESS and those who require biologic therapy for post-operative recurrence still remain limited.
This study aimed to compare subjective and psychophysical olfactory outcomes between patients whose disease was controlled after primary FESS alone and those who were administered dupilumab for post-operative recurrence following primary FESS. By focusing on post-treatment olfactory outcomes rather than primary treatment efficacy, we sought to clarify the clinical positioning of biologic therapy in the management of olfactory dysfunction in eosinophilic chronic rhinosinusitis.
Materials and methods
Study design and patients
This retrospective cohort study compared subjective and psychophysical olfactory outcomes between patients whose disease was well-controlled after primary endoscopic sinus surgery and those treated with dupilumab for post-operative recurrence. A total of 357 adult patients diagnosed with eosinophilic chronic rhinosinusitis who underwent primary bilateral FESS in our department between March 2011 and October 2024 were included. There were 202 men and 155 women. Median age was 50 years (range, 18–82 years). The diagnosis of eosinophilic chronic rhinosinusitis was established using the Japanese Epidemiological Survey of Refractory Eosinophilic Chronic Rhinosinusitis criteria.Reference Tokunaga, Sakashita, Haruna, Asaka, Takeno and Ikeda1
Patients were excluded if they had a history of prior sinonasal surgery, including revision endoscopic sinus surgery, sinonasal tumours, paranasal cystic lesions, major craniofacial trauma or incomplete olfactory assessment data. This study was approved by the institutional regulations of the Ethics Committee of Hyogo Medical University (approval numbers: 1512 and 2819). The study was conducted in accordance with the Declaration of Helsinki.
Treatment procedure and group definitions
All patients underwent primary FESS followed by standard post-operative medical management as baseline therapy, in accordance with international guidelines.Reference Fokkens, Lund, Hopkins, Hellings, Kern and Reitsma7–Reference Rosenfeld, Piccirillo, Chandrasekhar, Brook, Ashok Kumar and Kramper9 All procedures were performed under general anaesthesia. The surgical procedure aimed to remove inflammatory and polypoid tissue, and to improve sufficient ventilation and drainage of the paranasal sinuses. The extent of surgery was tailored to disease distribution and severity, but generally included opening of the ethmoid, maxillary, frontal and sphenoid sinuses as indicated.Reference Fokkens, Lund, Hopkins, Hellings, Kern and Reitsma7, Reference Orlandi, Kingdom, Smith, Bleier, DeConde and Luong8
Standard post-operative management included routine nasal saline irrigation and intranasal corticosteroid spray therapy. Post-operative recurrence of eosinophilic chronic rhinosinusitis was defined as endoscopically confirmed reappearance or progression of nasal polyps accompanied by clinically relevant worsening of sinonasal symptoms despite standard post-operative treatment. Additional medical treatments, including local and/or systemic corticosteroid therapy and dupilumab, were prescribed according to disease activity and clinical judgement. Systemic corticosteroids were prescribed selectively for disease exacerbation for a short period, based on symptom severity and endoscopic findings.Reference Fokkens, Lund, Hopkins, Hellings, Kern and Reitsma7–Reference Rosenfeld, Piccirillo, Chandrasekhar, Brook, Ashok Kumar and Kramper9
Dupilumab was introduced when disease control remained insufficient despite these conventional treatments, based on a shared decision-making process considering endoscopic disease activity, symptom burden, co-morbidities and systemic type 2 inflammatory status. Dupilumab was administered subcutaneously at a dose of 300 mg every 2 weeks, in accordance with approved clinical practice and pivotal clinical trials.Reference Bachert, Han, Desrosiers, Hellings, Amin and Lee14, Reference Jonstam, Swanson, Mannent, Cardell, Tian and Wang15 In some patients, the administration interval was extended to every four weeks after stabilisation of sinonasal symptoms, whereas in patients with co-morbid severe asthma, the two-week dosing regimen was maintained.
Patients were classified according to their post-operative disease course into two groups of FESS and dupilumab. The FESS group consisted of patients whose symptoms and intranasal condition were well controlled after primary FESS and standard post-operative management, defined as not requiring any additional medical treatment, including local or systemic corticosteroids or dupilumab. The dupilumab group consisted of patients who were administrated dupilumab because of deteriorated symptoms and polyposis recurrence after primary FESS, defined as endoscopically confirmed reappearance or progression of nasal polyps accompanied by worsening sinonasal symptoms that required additional medical therapy.
Evaluation of general symptoms and quality of life
General symptom burden and health-related QoL due to eosinophilic chronic rhinosinusitis were assessed using the nasal symptoms questionnaireReference Saito, Tsuzuki, Nishikawa, Okazaki, Hashimoto and Sakagami23 and a visual analogue scale (VAS) in the dupilumab group.
Nasal symptoms questionnaire and visual analogue scale
The nasal symptom questionnaire was a self-administered questionnaire composed of 10 items organised into two sections (Parts I and II).Reference Saito, Tsuzuki, Nishikawa, Okazaki, Hashimoto and Sakagami23 Part I included eight items addressing nasal symptoms: (1) sneezing and/or nasal itching; (2) nasal discharge; (3) nasal obstruction; (4) post-nasal drip and/or sputum; (5) olfactory loss; (6) pain (including toothache, buccal pain, facial pain and/or headache); (7) eye itching and/or epiphora; and (8) cough and/or throat irritation. Part II consisted of two items related to QoL: (9) impairment in daily activities such as work, school, outdoor activities or social functioning; and (10) sleep disturbances or general physical and emotional problems. Each item was scored on a 4-point ordinal scale: 0 (no symptoms), 1 (mild), 2 (moderate) and 3 (severe). The total nasal symptoms questionnaire score ranged from 0 to 30, with higher scores indicating more severe symptoms.
In addition, overall nasal symptom severity was also assessed using a VAS. The VAS consisted of a 10-cm horizontal line with anchors labelled ‘normal’ (0 per cent) at the left end and ‘most severe’ (100 per cent) at the right end.
Olfactory evaluation
Olfactory function was evaluated using both patient-reported subjective assessment tools and standardised psychophysical olfactory tests. Subjective olfaction was assessed using the self-administered odour questionnaire,Reference Tsuzuki, Hamamoto, Iijima, Kobayashi, Kondo and Mori24 while psychophysical olfactory function was evaluated using the Takagi and Toyota olfactometer. Olfactory assessments were performed pre-operatively in both groups, post-operatively in the FESS group, and before and after initiation of biologic therapy in the dupilumab group.
Self-administered odour questionnaire
The self-administered odour questionnaire consisted of 20 odour items, as recommended by the Japan Rhinologic SocietyReference Tsuzuki, Hamamoto, Iijima, Kobayashi, Kondo and Mori24: steamed rice, miso, seaweed, soy sauce, baked bread, butter, curry, garlic, orange, strawberry, green tea, coffee, chocolate, household gas, garbage, timber, stercus, sweat, flowers and perfume. For each item, patients selected one of four responses: (i) ‘I can understand the smell’ (2 points); (ii) ‘I can sometimes understand the smell’ (1 point); (iii) ‘I cannot understand the smell at all’ (0 points); or (iv) ‘I have not smelled it recently or before’ (classified as unevaluable). The self-administered odour questionnaire score was derived by summing the points from evaluable responses (i)–(iii), dividing by the number of evaluable items, multiplying by two and converting the value to a percentage, with lower percentages indicating poorer olfactory status. Patients who selected response (iv) for 11 or more items were regarded as invalid cases.
Olfactory threshold test
Olfactory function was assessed using the Takagi and Toyota olfactometer, a standardised olfactory test reimbursed by the Japanese health insurance system.Reference Oka, Tsuzuki, Takebayashi, Kojima, Daimon and Sakagami19, Reference Tsuzuki, Hamamoto, Iijima, Kobayashi, Kondo and Mori24 The Takagi and Toyota olfactometer comprises five odorants: β-phenyl ethyl alcohol (rose-like odour), methyl cyclopentenolone (burnt odour), isovaleric acid (sweaty odour), γ-undecalactone (fruity odour) and skatole (putrid odour) (Daiichi Yakuhin Sangyo Co., Ltd., Tokyo, Japan). Odorants A, C, D and E were prepared in eight graded concentrations (−2 to 5), whereas odorant B was prepared in seven concentrations (−2 to 4). The highest concentrations of odorants A5, B4, C5, D5 and E5 were 631, 25.1, 100, 795 and 79.5 mg/ml, respectively. Each odorant was serially diluted in 10-fold steps down to the lowest concentration (−2).
For testing, one end of a paper strip (7 mm in width and 140 mm in length) was immersed in the odorant solution and participants were instructed to sniff the strip from approximately 10–20 mm from the nostrils. Using a standardised test kit, thresholds were determined by presenting odorants in ascending order of concentration. The detection threshold was defined as the lowest concentration at which the participant perceived the presence of an odour, regardless of correct identification. The recognition threshold was defined as the lowest concentration at which the participant correctly identified the odour. The mean detection threshold across the five odorants was calculated and used as an index of overall olfactory sensitivity. The mean recognition threshold across the five odorants was calculated and used as an index of overall olfactory acuity. Higher threshold values indicated worse olfactory function.
Although the Takagi and Toyota olfactometer is often referred to as an objective olfactory test in clinical practice, it is technically a psychophysical assessment that relies on patient perception and response. In this study, the term ‘psychophysical’ is used to distinguish standardised threshold-based testing from patient-reported subjective questionnaires.
Statistical analysis
Continuous variables are presented as medians with interquartile ranges. Within-group comparisons of pre- and post-treatment olfactory outcomes were performed using the Wilcoxon signed-rank test. Comparisons between the groups were conducted using the Mann–Whitney U test. Categorical variables were compared between groups using Fisher’s exact test. Data are expressed as medians with ranges unless otherwise specified. All p values were calculated using two-sided tests, and a p value of less than 0.05 was regarded as statistically significant. Statistical analyses were conducted using Stat Flex software, version 7.0 (Osaka, Japan). This study was not designed as an equivalence or non-inferiority trial, therefore the absence of statistically significant differences in post-treatment outcomes should not be interpreted as evidence of equivalence.
Results
Patient characteristics
Of the 357 patients diagnosed with eosinophilic chronic rhinosinusitis who underwent primary FESS, 334 patients achieved sufficient disease control without post-operative recurrence and were classified into the FESS group at a median post-operative follow up of 6 months (range, 2–72 months) (Figure 1). The remaining 23 patients developed post-operative recurrence, showed an insufficient response to local and/or systemic corticosteroid therapy, and were subsequently treated with dupilumab without undergoing revision FESS. These patients were classified into the dupilumab group.
Treatment algorithm showing patient selection and classification. A total of 357 patients with eosinophilic chronic rhinosinusitis underwent primary functional endoscopic sinus surgery (FESS). Patients without post-operative recurrence were classified as the FESS group (n = 334), whereas those who developed post-operative recurrence and were treated with dupilumab without undergoing revision FESS were classified as the dupilumab group (n = 23). ECRS = eosinophilic chronic rhinosinusitis.
There were no significant differences between the groups in age, sex, prevalence of asthma and aspirin-exacerbated respiratory disease, or baseline olfactory detection and recognition thresholds in the pre-treatment stage (Table 1). In contrast, the dupilumab group showed significantly higher Japanese Epidemiological Survey of Refractory Eosinophilic Chronic Rhinosinusitis scores and peripheral blood eosinophil levels, and baseline self-administered odour questionnaire scores were significantly worse in the dupilumab group than in the FESS group. In the dupilumab group, the median interval from primary FESS to initiation of dupilumab was 49 months (range, 6–128 months) and the median follow-up period after initiation of dupilumab was 12 months (range, 6–62 months).
Patient characteristics in the pre-treatment stages
Table 1 Long description
The table compares baseline patient characteristics between a FESS group of 334 people and a dupilumab group of 23 people, with p values for between-group differences. Median age was lower in the dupilumab group at 45 years versus 51 years in the FESS group, and this difference was statistically significant. Sex distribution was similar, with 187 men and 147 women in the FESS group and 15 men and 8 women in the dupilumab group, with no significant difference. Asthma prevalence was comparable at 46 percent in the FESS group and 52 percent in the dupilumab group, and aspirin-exacerbated respiratory disease was also similar at 7 percent versus 9 percent; neither comparison was significant. Peripheral blood eosinophils were higher in the dupilumab group, 10.1 percent versus 7.1 percent, and this difference was statistically significant. The JESREC score medians were both 15, but the p value indicates a statistically significant difference despite the same median and range shown. Smell-related measures favored the FESS group: the self-administered odour questionnaire median was 34.7 percent in FESS versus 0 percent in dupilumab, a statistically significant difference, while detection and recognition thresholds were higher in the dupilumab group but not significantly different. Because the dupilumab group is much smaller, results may be more sensitive to variability, and p values reflect statistical tests rather than clinical importance.
p values indicate differences between groups.
* Statistically significant differences (p < 0.05).
† Mann–Whitney U test.
‡ Fisher’s exact test. FESS = functional endoscopic sinus surgery; JESREC = Japanese Epidemiological Survey of Refractory Eosinophilic Chronic Rhinosinusitis
Changes in subjective olfaction (self-administered odour questionnaire)
Self-administered odour questionnaire scores improved significantly after treatment in both groups (Figure 2). In the FESS group, pre-operative self-administered odour questionnaire score (34.7 per cent; range, 0–100 per cent) showed marked improvement to 81.6 per cent (range, 0–100 per cent) after surgery (p < 0.0001) (Figure 2a). Similarly, patients in the dupilumab group also demonstrated significant improvement in self-administered odour questionnaire scores from 0 per cent (range, 0–100 per cent) to 50 per cent (range, 10–100 per cent) after initiation of dupilumab (p < 0.001) (Figure 2b).
Changes in self-administered odour questionnaire (SAOQ) scores before (pre) and after (post) treatment in the functional endoscopic sinus surgery (FESS) group (a) and the dupilumab group (b). Both groups showed significant improvement after treatment. Comparisons were performed using the Wilcoxon signed-rank test. Asterisks indicate statistically significant differences (*p < 0.05). ‘Pre’ represents pre-operative baseline data in both groups and ‘post’ represents post-treatment data.
Figure 2 Long description
The image A showing a box-and-whisker plot with overlaid points. The horizontal axis shows Pre and Post. The vertical axis label is SAOQ (percent). The vertical axis ranges from 0 to 100 in increments of 10. A bracket above Pre and Post is labeled p less than 0.0001 asterisk. Text below the axis reads FESS group (n equals 334). Pre: The box is positioned low on the scale, with the median line near about 0. The box spans roughly from about 0 to about 10. The whiskers extend from about 0 to about 40. Multiple points extend above the upper whisker, reaching to about 90. Post: The box is positioned higher on the scale, with the median line near about 80. The box spans roughly from about 60 to about 90. The whiskers extend from about 0 to about 100. Several points appear below the lower whisker, down to about 0. The image B showing a box-and-whisker plot with overlaid points. The horizontal axis shows Pre and Post. The vertical axis label is SAOQ (percent). The vertical axis ranges from 0 to 100 in increments of 10. A bracket above Pre and Post is labeled p less than 0.001 asterisk. Text below the axis reads Dupilumab group (n equals 23). Pre: The box is positioned near the bottom of the scale, with the median line near about 0. The box spans roughly from about 0 to about 10. The whiskers extend from about 0 to about 20. Points extend above the upper whisker to about 30. Post: The box is positioned mid to high on the scale, with the median line near about 50. The box spans roughly from about 40 to about 80. The whiskers extend from about 30 to about 90. Points extend below the lower whisker to about 20. Across both images, the Post boxes are positioned higher than the Pre boxes.
Changes in psychophysical olfactory thresholds (Takagi and Toyota olfactometer)
Mean olfactory detection and recognition thresholds improved significantly after treatment in both groups (Figure 3). In the FESS group, post-operative olfactory assessments were performed at a median of 6 months after primary FESS (range, 2–72 months). Both mean detection and recognition thresholds showed significant improvement when comparing pre- and post-operative values. The median detection threshold improved from 4.2 (range, −0.8–5.8) pre-operatively to 1.2 (range, −0.6–5.8) post-operatively, and the median recognition threshold improved from 5.0 (range, 0.4–5.8) to 3.4 (range, 0–5.8) (both p < 0.0001) (Figure 3a,b).
Changes in mean olfactory detection and recognition thresholds assessed using the Takagi and Toyota olfactometer. Comparisons of pre-operative (pre-op) and post-operative (post-op) values are shown for the functional endoscopic sinus surgery (FESS) group (a, b) and comparisons of values before (pre-bio) and after dupilumab treatment (post-bio) are shown for the dupilumab group (c, d). Statistical comparisons were performed using the Wilcoxon signed-rank test. Asterisks indicate statistically significant differences (*p < 0.05).
Figure 3 Long description
The image A showing a box plot with the vertical axis labeled “Mean detection threshold” and tick labels 0, 1, 2, 3, 4, 5, 6. The horizontal axis shows “Pre-op” and “Post-op” with “FESS group (n = 334)” below. Text above the plot reads “p < 0.0001 asterisk”. Many individual points are plotted for both categories. The box for “Pre-op” spans roughly from 3 to 5 with a median line near 4. The box for “Post-op” spans roughly from 1 to 3 with a median line near 2. The image B showing a box plot with the vertical axis labeled “Mean recognition threshold” and tick labels 0, 1, 2, 3, 4, 5, 6. The horizontal axis shows “Pre-op” and “Post-op” with “FESS group (n = 334)” below. Text above the plot reads “p < 0.0001 asterisk”. Many individual points are plotted for both categories. The box for “Pre-op” spans roughly from 4 to 6 with a median line near 5. The box for “Post-op” spans roughly from 3 to 5 with a median line near 4. The image C showing a box plot with the vertical axis labeled “Mean detection threshold” and tick labels 0, 1, 2, 3, 4, 5, 6. The horizontal axis shows “Pre-op” and “Post-bio” with “Dupilumab group (n = 23)” below. Text above the plot reads “p = 0.0322 asterisk”, “p = 0.1250” and “p = 0.8125”. Individual points are plotted for both categories. The “Pre-op” category shows a narrow box near the upper part of the scale with a median line near 5. The “Post-bio” category shows a taller box spanning a wider range, with a median line near 3. The image D showing a box plot with the vertical axis labeled “Mean recognition threshold” and tick labels 0, 1, 2, 3, 4, 5, 6. The horizontal axis shows “Pre-op” and “Post-bio” with “Dupilumab group (n = 23)” below. Text above the plot reads “p = 0.2301” and “p = 0.3125”. Individual points are plotted for both categories. The “Pre-op” category shows a box near the upper part of the scale with a median line near 5. The “Post-bio” category shows a box spanning a lower range with a median line near 3.
In the dupilumab group, median detection and recognition thresholds were comparable between pre-operative and pre-dupilumab assessments (detection, 5.8 (range, 0.8–5.8) vs 3.4 (range, 0.4–5.8); recognition, 5.8 (range, 0.8–5.8) vs 5.6 (range, 2.0–5.8)). No significant differences were observed between before and after dupilumab treatment assessments (detection, 3.4 (range, 0.4–5.8) vs 1.9 (range, 0.4–5.8); recognition, 5.6 (range, 2.0–5.8) vs 3.1 (range, 1.6–5.8)). However, both detection and recognition thresholds improved significantly when comparing pre-operative and after dupilumab treatment values (detection, 5.8 (range, 0.8–5.8) vs 1.9 (range, 0.4–5.8); recognition, 5.8 (range, 0.8–5.8) vs 3.1 (range, 1.6–5.8)) (Figure 3c,d).
Post-treatment comparison of olfactory outcomes between groups
In the post-treatment stages, no significant differences were observed between groups (post-operative and after dupilumab treatment) in subjective olfaction assessed using the self-administered odour questionnaire (median self-administered odour questionnaire score post-operative, 81.6 per cent (range, 0–100 per cent) vs after dupilumab treatment, 50.0 per cent (range, 10–100 per cent); p = 0.1853) (Figure 4a). Similarly, no significant differences were observed between groups in post-treatment mean detection and recognition thresholds (median detection threshold post-operative, 1.2 (range, −0.6–5.8) vs after dupilumab treatment, 1.9 (range, 0.4–5.8); p = 0.1119; median recognition threshold post-operative, 3.4 (range, 0–5.8) vs after dupilumab treatment, 3.1 (range, 1.6–5.8); p = 0.6023) (Figure 4b,c). These findings indicate that olfactory outcomes after FESS alone and after dupilumab for post-operative recurrence were comparable in both subjective and psychophysical measures.
Comparison of post-treatment subjective and psychophysical olfactory outcomes between two groups, including self-administered odour questionnaire (SAOQ) scores (a), mean detection thresholds (b) and mean recognition thresholds (c). Comparisons were performed using the Mann–Whitney U test. No significant differences were observed between groups. FESS = functional endoscopic sinus surgery.
Figure 4 Long description
Three sub-images labeled a, b and c. The image A showing a box and scatter plot with two categories on the horizontal axis: FESS group (n equals 34) and Dupilumab group (n equals 23). The vertical axis is labeled SAOQ score, with tick labels from 0 to 100 in increments of 10. A bracket above the two categories is labeled p equals 0.1853. The FESS group shows many individual points spanning from near 0 up to near 100, with a box centered around the midrange and a horizontal median line. The Dupilumab group shows fewer points, with a box spanning roughly the midrange to upper range and a horizontal median line. The image B showing a box and scatter plot with two categories on the horizontal axis: FESS group (n equals 34) and Dupilumab group (n equals 23). The vertical axis is labeled Mean detection threshold, with tick labels from 0 to 6 in increments of 1. A bracket above the two categories is labeled p equals 0.1119. The FESS group shows many points concentrated between about 0 and 3, with a box in the lower range and a horizontal median line. The Dupilumab group shows fewer points, with a box positioned higher than the FESS group box and a horizontal median line. The image C showing a box and scatter plot with two categories on the horizontal axis: FESS group (n equals 34) and Dupilumab group (n equals 23). The vertical axis is labeled Mean recognition threshold, with tick labels from 0 to 6 in increments of 1. A bracket above the two categories is labeled p equals 0.6023. The FESS group shows many points concentrated between about 1 and 5, with a box around the midrange and a horizontal median line. The Dupilumab group shows fewer points, with a box around the midrange and a horizontal median line.
Changes in symptoms and quality of life after dupilumab treatment
In the dupilumab group, changes in sinonasal symptoms and health-related QoL were additionally evaluated using the nasal symptoms questionnaire and VAS. Although the total nasal symptoms questionnaire score showed a trend towards improvement after dupilumab treatment, the change did not reach statistical significance. Among individual nasal symptoms questionnaire items, nasal obstruction and olfactory loss showed significant improvement after treatment (p = 0.0488 and p = 0.0371, respectively), whereas the other items did not. In addition, VAS for overall nasal symptoms significantly improved after dupilumab treatment compared with before treatment (p = 0.0020).
Discussion
This study demonstrated that patients with eosinophilic chronic rhinosinusitis who achieved good disease control after primary FESS experienced significant improvements in olfactory dysfunction. Notably, these improvements were comparable to those observed in patients who developed post-operative recurrence and subsequently received dupilumab therapy. Although baseline disease severity differed between the two groups, post-treatment olfactory outcomes were likewise favourable. These findings suggest that both surgical and biologic approaches can contribute meaningfully to olfactory recovery in eosinophilic chronic rhinosinusitis.
As shown in Figure 1, most patients with eosinophilic chronic rhinosinusitis in the present cohort (93.6 per cent, 334 of 357) achieved well-controlled disease with primary FESS alone, without requiring additional treatment. This highlights the substantial effectiveness of surgery as an initial therapeutic intervention, even in a disease subtype traditionally regarded as highly recurrent. Our findings indicate that, despite its reputation for frequent recurrence, most cases of eosinophilic chronic rhinosinusitis can be adequately managed with surgery and standard post-operative medical therapy.Reference Mattos, Soler, Schlosser, Mace, Alt and Ramakrishnan25 This is consistent with the Japanese clinical practice guidelines 2025, which recommend FESS as an effective treatment option for olfactory dysfunction associated with eosinophilic chronic rhinosinusitis.Reference Tsuzuki, Hamamoto, Iijima, Kobayashi, Kondo and Mori24
Biologic therapy targeting type 2 inflammation has emerged as an effective treatment option for patients with severe and uncontrolled chronic rhinosinusitis with nasal polyps, particularly intractable eosinophilic chronic rhinosinusitis. Dupilumab has been shown to improve multiple disease domains, including nasal polyp burden, nasal obstruction and olfactory dysfunction. Accordingly, olfactory dysfunction is often regarded as a key symptom prompting consideration of biologic therapy. The similar improvements in olfactory function observed after FESS and dupilumab treatment in the present study may reflect distinct mechanisms of action ultimately converging on comparable clinical outcomes.
In this study, dupilumab treatment was also associated with improvement in other sinonasal symptoms, particularly nasal obstruction, as assessed by the nasal symptoms questionnaire. These findings suggest that dupilumab treatment may occur with improvements in conductive olfactory dysfunction. Given that experimental studies have demonstrated that IL-4 and IL-13 can impair olfactory sensory neurons,Reference Hara, Jha, Huang, Han, Langohr and Gaglia26–Reference Saraswathula, Liu, Kulaga and Lane28 and that direct evidence in humans remains limited, inhibition of these cytokine pathways may also contribute to the recovery of sensorineural olfactory dysfunction. However, biologic therapy is associated with high cost and the need for long-term administration, making careful patient selection essential.Reference Bachert, Zhang, Cavaliere, Weiping, Gevaert and Krysko29, Reference Scangas, Wu, Ting, Metson, Walgama and Shrime30
Although dupilumab is highly effective in patients with refractory disease and co-morbid severe asthma, only a small proportion of patients in this cohort required biologic therapy in real-world practice (6.4 per cent, 23 of 357). This perspective aligns with recent otolaryngology-focused reviews emphasising that, given the high cost and long-term nature of biologic therapy, surgery should remain the initial intervention, with biologics reserved for cases that remain uncontrolled despite appropriate post-operative management.Reference Armstrong and Senior31–Reference Codispoti and Mahdavinia34
We previously demonstrated that the post-operative endoscopic appearance score can serve as a practical index for guiding optimal treatment strategies in recurrent eosinophilic chronic rhinosinusitis.Reference Saito, Fushimi, Hirose, Haruna and Tsuzuki18 Specifically, additional corticosteroid therapy was required when the post-operative endoscopic score exceeded 30 per cent, whereas biologic therapy and/or revision FESS became necessary in patients with scores of 65 per cent or higher. This endoscopy-based algorithm supports a rational and stepwise escalation of treatment following primary FESS. Functional endoscopic sinus surgery reduces local inflammatory burden, restores sinonasal ventilation and improves access of topical therapies to the olfactory cleft.Reference Fokkens, Lund, Hopkins, Hellings, Kern and Reitsma7, Reference Orlandi, Kingdom, Smith, Bleier, DeConde and Luong8 Dupilumab suppresses type 2 inflammation at a systemic level, including within sinonasal tissues.Reference Fokkens, Lund, Hopkins, Hellings, Kern and Reitsma7, Reference Bachert, Mannent, Naclerio, Mullol, Ferguson and Gevaert13, Reference Bachert, Zhang, Cavaliere, Weiping, Gevaert and Krysko29 Through these different pathways, both approaches may facilitate recovery of olfactory mucosal function, ultimately resulting in similar post-treatment olfactory outcomes.
This study has several limitations that should be acknowledged. Firstly, its retrospective design may introduce selection bias, particularly given the differences in baseline disease severity and the limited sample size in the dupilumab group. The lack of significant differences in QoL items measured using the nasal symptoms questionnaire in the dupilumab group may also be caused by the limited number of patients. Secondly, olfactory outcomes were assessed using psychophysical threshold testing, which may not fully capture multidimensional aspects of olfactory function or patient-reported impairment.Reference Prasad, Fong and Ooi35 Thirdly, the study was not designed as an equivalence or non-inferiority trial, therefore the absence of statistically significant differences should not be interpreted as evidence of equivalence. Fourthly, dupilumab was introduced only after post-operative recurrence following primary FESS, and thus the two groups differed in baseline characteristics, disease severity and timing of outcome assessment.Reference Hernandez, Wendler, Mayr, Iro, Hummel and Mueller36 Fifthly, the follow-up period varied widely among patients, which may have influenced the observed rates of post-operative disease control.Reference Hsu, Wang, Shen, Weitzel, Lai and Wormald37 Finally, this study was conducted at a single tertiary centre, potentially limiting the generalisability of the findings. Future prospective, multicentre studies with standardised olfactory assessments are warranted.
Taken together, these findings position FESS and biologic therapy as complementary components within a stepwise treatment strategy for eosinophilic chronic rhinosinusitis. Primary FESS remains a highly effective first-line intervention capable of achieving olfactory improvement in most patients. Biologic therapy should be introduced selectively, based on post-operative disease trajectory, co-morbidities and inflammatory burden. This integrated approach may help optimise long-term outcomes while avoiding unnecessary escalation of therapy, thereby supporting a more personalised and rational management framework for this challenging disease.
• Olfactory dysfunction is one of the major symptoms affecting quality of life in patients with eosinophilic chronic rhinosinusitis and is often emphasised when considering escalation of treatment
• The treatment approach is stepwise, consisting of primary functional endoscopic sinus surgery (FESS), post-operative medical management with local and/or systemic corticosteroids, biologic therapy for refractory postoperative recurrence and revision FESS for adhesions and residual disease
• Despite greater baseline disease severity in patients treated with dupilumab for post-operative recurrence, both primary FESS alone and dupilumab resulted in significant improvement in olfaction
• No statistically significant differences were observed in post-treatment subjective or psychophysical olfactory outcomes between patients who remained well controlled after primary FESS and those treated with dupilumab for post-operative recurrence, supporting dupilumab as a viable option for improving olfaction in post-operative recurrence
Conclusion
In patients with eosinophilic chronic rhinosinusitis, both primary FESS and dupilumab administered for post-operative recurrence yielded favourable olfactory outcomes. Post-treatment subjective olfactory scores and psychophysical thresholds were similar between patients who remained well controlled after FESS alone and those who required dupilumab. These findings support dupilumab as an effective option for managing post-operative recurrence within a stepwise treatment strategy.
Acknowledgements
The authors gratefully acknowledge the help of their technical assistant, Ms Yumi Kida. This work was supported by Grants-in-Aid for Scientific Research (Grant numbers: 20K09700 and 24K12687) from the Japan Society for the Promotion of Science and a Health Labor Sciences Research Grant (21FC1013). The funder had no role in the design, data collection, data analysis and reporting of this study.
Competing interests
None declared
