Clinical efficacy of an extended local treatment in nasal polyps with different endotypes: A prospective randomized controlled study☆ Authors: Ziqi Huang, Zhaofeng Xu, Wufuerjiang Yiming, Tingyan Wang, Xiaolin Chen, Jianbo Shi, Yinyan Lai Published: 2026 ABSTRACT PurposeTo compare the efficacy of 3 corticosteroid delivery methods and assess whether extending treatment duration enhances patient outcomes.MethodsThis prospective randomized controlled trial enrolled 138 patients with chronic rhinosinusitis with nasal polyps (CRSwNP) from FA Hospital between September 2019 and February 2025. Patients were randomly allocated to 3 groups receiving a two-week regimen: Nasal Spray (budesonide nasal spray 256 μg/day), Nasal Drop (budesonide nasal spray 256 μg/day & budesonide nasal drop 1 mg/day) and Oral Steroids (budesonide nasal spray 256 μg/day & oral prednisone 30 mg/day). Visual-analogue scale (VAS), the 22-item Sinonasal Outcome Test (SNOT-22), and nasal polyps endoscopic scores (NP scores) were assessed pre- and post-treatment. The two-week data were compared to one-week time points from separate cohorts in our prior work.ResultsCompared to baseline (Day 0), the nasal drop group showed significantly greater VAS reduction than the nasal spray group after 2 weeks (P = 0.02), particularly in nasal obstruction (P = 0.04) and purulent rhinorrhea (P < 0.01), and was comparable to the oral group (P = 0.96). Regarding SNOT-22, the nasal drop group demonstrated superior improvement in extranasal (P = 0.02) and sleep disorder symptoms (P = 0.03) compared to the nasal spray group from baseline to 2 weeks, with no significant difference versus the oral group (P = 0.48). Additionally, the 2 weeks use of oral corticosteroids was associated with notable polyp size reduction (P = 0.02), especially in type-2 CRSwNP (P = 0.01). Extending treatment duration from 1 to 2 weeks yielded significant subjective improvements in the nasal drop group (P = 0.04), and significant endoscopic improvement only in the nasal spray group (P < 0.01).ConclusionsAdding intranasal drops to nasal spray enhances short-term symptom relief, whereas oral steroids reduce polyp size more effectively, particularly in type 2 CRS. Introduction Chronic rhinosinusitis (CRS) is a prevalent disease worldwide, with a prevalence ranging from 3.0% to 6.4% in Europe,1 2.1%–28.4% in Asia2 and approximately 8% in China.3 Characterized by chronic inflammation of the nasal and paranasal sinus mucosa persisting for over 3 months, CRS is divided into CRS with nasal polyps (CRSwNP) and without nasal polyps (CRSsNP). Patients with CRSwNP always experience more severe symptoms and a complex treatment process,4 alongside distinct pathological mechanisms, necessitating precise and individualized therapeutic approaches. Current guidelines5 recommend nasal corticosteroids as first-line therapy, reserving surgery for those unresponsive to adequate medical management. However, the definition of "adequate" therapy, especially the optimal delivery method and duration, remains contentious. As delayed surgery may impair symptom control and reduce patient benefit,6 it is critical to establish criteria for adequate pharmacological treatment. Oral corticosteroids are systemically effective but carry higher risks of adverse effects such as hypertension, hyperglycemia and gastric ulcers, especially with prolonged use.7,8 Topical options include nasal sprays, nasal drops,9 irrigations,10 and nebulizers,11 all with demonstrated efficacy,12 and nasal drops and sprays offer greater convenience. Prior studies indicate that adding drops to the regimen of nasal sprays for 1 week improves subjective symptoms and polyp size compared to nasal sprays alone, especially in patients with eosinophilic CRSwNP.9 Whether extending treatment enhances outcomes and which modality offers the greatest benefit across CRS endotypes is uncertain. The aim of this study is to compare the efficacy of 3 short-course corticosteroid regimens over 2 weeks and evaluate the impact of extending therapy duration. Materials and methods Patients and study design This study is a randomized controlled trial, approved by the Clinical Trial Registry (ChiCTR1900028585). A total of 138 patients with chronic sinusitis with bilateral nasal polyps were enrolled at the First Affiliated Hospital of Sun Yat-sen University in Guangzhou, China, from September 2019 to February 2025. The trial design is illustrated in Fig. 1. The inclusion criteria are as follows: (a) age range from 18 to 75 years. (b) bilateral nasal polyps meeting EPOS2020 criteria. The exclusion criteria include: (a) oral corticosteroids or antibiotics use within 4 weeks before. (b) upper respiratory tract infection in 2 weeks. (c) pregnant or breastfeeding. (d) uncontrolled comorbidities (eg, diabetes, autoimmune diseases requiring long-term corticosteroids). (e) other nasal disorders such as fungal sinusitis or malignant tumor. According to Chinese Expert Consensus,13 type 2 CRS was defined by: (a) peripheral blood eosinophil (EOS) count ≥0.3 × 109/L. (b) serum total IgE ≥100 IU/ml, (c) tissue EOS ≥10/HPF, or (d) comorbid asthma.Fig. 1Workflow diagram of participants enrolled in this study Workflow diagram of participants enrolled in this study Randomization and intervention Using a random number table (generated by Microsoft Office Excel), 138 patients were assigned to: (a) Nasal Spray: budesonide spray 128 μg BID. (b) Nasal Drop: budesonide spray 128 μg BID & budesonide drop 1 mg QD (Mygind position, lie down in supine position with the head off the bed in hyperextension, keep the chin as the highest point of the head for 10 min14). (c) Oral Steroids: budesonide spray 128 μg BID & prednisone 30 mg QD (administered at 8:00 and discontinued after 2 weeks). Outcomes were assessed at baseline and 2 weeks. The recording of results and the data analysis were performed by 2 researchers independently. Outcomes and measurements The primary outcome was the change in VAS score, including nasal obstruction, rhinorrhea, olfactory loss and pain. Secondary outcomes included SNOT-22, NP scores, and EOS counts. Demographic and clinical data, including allergen testing, Lund-Mackay score, and pulmonary function, were recorded at enrollment. One-week outcome data was derived from our previous study,8 which meant the data of the 1-week and 2-week time points was collected from separate cohorts of participants, rather than through longitudinal follow-up of the same subjects. We compared the baseline data of the same treatment groups between the 2 populations and found no significant differences, indicating that these groups were comparable. Statistical analysis The sample size was estimated based on the previous finding of our preliminary study by using a reduction in VAS score as the primary outcome. Using a target statistical power of 0.9 and a two-tailed α value of 0.05, we calculated that 36 subjects for each group would be able to detect a 1-point difference (standard deviation = 1.8). Accounting for a 10% dropout, we ultimately recruited 40–41 participants in each group. Subgroup analyses classified patients into the following categories: (a) total IgE levels of ≥100 IU/mL (high) versus <100 IU/mL (low), (b) Lund-Mackay scores of ≥14 (high) versus <14 (low), and (c) EOS counts of ≥0.3 × 109/L (high) versus <0.3 × 109/L (low). Treatment response was defined as a reduction in VAS severity grade post-treatment according to the classification of VAS subjective symptom scores in EPOS2020:5 VAS score of 0–3 is considered mild discomfort, 3 to 7 is moderate discomfort, and 7 to 10 is severe discomfort. We defined treatment-sensitive patients as those whose VAS severity grade decreased (eg, from severe to moderate) after 2 weeks, and the ratio of the number of people with treatment sensitivity to the number of people treated in that group was defined as the treatment-sensitivity rate. The significance among each group was calculated using the chi-square test. The statistical software used in this study included SPSS Statistics 25 and GraphPad Prism 9.5. Analysis of Variance (ANOVA) with Tukey’s test was applied for normally distributed continuous variables. Kruskal-Wallis and Dunn’s test were used for non-parametric data. Chi-square test was used for categorical variables, Wilcoxon signed-rank test for paired comparisons, and Mann-Whitney test for time-point comparisons. A p-value of <0.05 was considered statistically significant. Safety assessment All adverse events were documented based on patient self-reports or physician observations. Results Patient clinical characteristics A total of 138 patients were enrolled, and 122 completed the two-week follow-up. Baseline demographics and clinical features are shown in Table 1.Table 1Demographic and clinical characteristics of participants. SD, standard deviation. ESS, endoscopic surgery. VAS score, visual-analogue scale. SNOT-22, the 22-item Sinonasal Outcome Test. IQR, interquartile range. EOS, eosinophil. NEU, neutrophiliaTable 1Nasal sprayNasal dropOral steroidsp valueGender (female,%)17 (41.46%)16 (39.02%)18 (45%)0.872Age (mean,SD)39.19 (13.23)43.47 (13.09)44.66 (10.79)0.449Allergic rhinitis (n,%)16 (39.02%)9 (21.95%)14 (34.15%)0.197asthma (n,%)8 (19.51%)11 (26.83%)16 (39.02%)0.139Type2 (n,%)26 (63.41%)26 (63.41%)27 (65.85%)0.916ESS History (n, %)12 (29.3%)14 (34.1%)14 (35.9%)0.74QuestionnaireVAS score (mean,SD)20.28 (6.25)20.17 (7.55)19.16 (7.93)0.991SNOT-22 (mean,SD)40.33 (19.83)38.94 (19.05)38.1 (22.8)0.795Endoscopic score (mdedian,IQR)6 (2)5.75 (2)6 (2)0.084Lund-Mackay score (mdedian,IQR)22 (10)22.5 (12)24 (10)0.198Blood testTotal IgE (mdedian,IQR)110.6 (106.98)116.8 (190.1)132.77 (236.87)0.998EOS count in peripheral blood (median∗106,IQR)0.3 (0.41)0.18 (0.27)0.27 (0.38)0.237Percentage of EOS in peripheral blood (median,IQR)0.043 (0.06)0.0275 (0.04)0.038 (0.05)0.069NEU count in peripheral blood (mean∗106,SD)4.07 (1.57)4.55 (1.47)4.42 (1.56)0.225Percentage of NEU in peripheral blood (mean,SD)0.59 (0.11)0.61 (0.08)0.61 (0.08)0.128 Table 1Demographic and clinical characteristics of participants. SD, standard deviation. ESS, endoscopic surgery. VAS score, visual-analogue scale. SNOT-22, the 22-item Sinonasal Outcome Test. IQR, interquartile range. EOS, eosinophil. NEU, neutrophiliaTable 1Nasal sprayNasal dropOral steroidsp valueGender (female,%)17 (41.46%)16 (39.02%)18 (45%)0.872Age (mean,SD)39.19 (13.23)43.47 (13.09)44.66 (10.79)0.449Allergic rhinitis (n,%)16 (39.02%)9 (21.95%)14 (34.15%)0.197asthma (n,%)8 (19.51%)11 (26.83%)16 (39.02%)0.139Type2 (n,%)26 (63.41%)26 (63.41%)27 (65.85%)0.916ESS History (n, %)12 (29.3%)14 (34.1%)14 (35.9%)0.74QuestionnaireVAS score (mean,SD)20.28 (6.25)20.17 (7.55)19.16 (7.93)0.991SNOT-22 (mean,SD)40.33 (19.83)38.94 (19.05)38.1 (22.8)0.795Endoscopic score (mdedian,IQR)6 (2)5.75 (2)6 (2)0.084Lund-Mackay score (mdedian,IQR)22 (10)22.5 (12)24 (10)0.198Blood testTotal IgE (mdedian,IQR)110.6 (106.98)116.8 (190.1)132.77 (236.87)0.998EOS count in peripheral blood (median∗106,IQR)0.3 (0.41)0.18 (0.27)0.27 (0.38)0.237Percentage of EOS in peripheral blood (median,IQR)0.043 (0.06)0.0275 (0.04)0.038 (0.05)0.069NEU count in peripheral blood (mean∗106,SD)4.07 (1.57)4.55 (1.47)4.42 (1.56)0.225Percentage of NEU in peripheral blood (mean,SD)0.59 (0.11)0.61 (0.08)0.61 (0.08)0.128 Demographic and clinical characteristics of participants. SD, standard deviation. ESS, endoscopic surgery. VAS score, visual-analogue scale. SNOT-22, the 22-item Sinonasal Outcome Test. IQR, interquartile range. EOS, eosinophil. NEU, neutrophilia Comparison of efficacy among different corticosteroid delivery methods After 14 days, all groups showed significant reductions in VAS score. Comparing to the nasal spray group, the nasal drop group (the comparison between groups, −9.34&-5.41, P = 0.02) and the oral steroids group (the comparison between groups, −8.99&-5.41 P = 0.04) achieved greater reduction than the nasal spray group, while no significant difference was found between the nasal drop and oral steroids (P = 0.96) (Fig. 2a). Further analysis indicated that nasal obstruction (the comparison between groups, −4.10&-2.46, P = 0.04) and purulent rhinorrhea (the comparison between groups, −2.38&-0.80, P < 0.01) showed the most significant improvements (Fig. 2b–c). SNOT-22 scores improved more in the nasal drop versus nasal spray (the comparison between groups, −18.08&-8.67, P = 0.03) (Fig. 3a), particularly in extra-nasal symptoms (the comparison between groups, −2.63&-1.05, P = 0.02) (Fig. 3b), including pyorrhea, cough and post-nasal drip syndrome, and sleep-related symptoms (the comparison between groups, −4.08&-1.26, P = 0.04) (Fig. 3c), including difficulty falling asleep, frequent nighttime awakenings, poor sleep quality and feeling tired upon waking. Additionally, based on previous studies, the Minimal Clinically Important Difference (MCID) for SNOT-22 is 8.9.15,16 We calculated the proportion of patients in each group who achieved this threshold, with results presented in Supplementary Fig. 1. There was a significant difference among the 3 groups in the proportion of participants achieving a clinically meaningful difference (χ2 = 7.24, P = 0.02). Multiple comparisons revealed that the proportion of participants achieving a clinically meaningful difference was significantly different between the nasal spray group and the nasal drop group (χ2 = 7.23, P = 0.01). NP scores decreased significantly only between the oral steroids group and the nasal spray group (the comparison between groups, −1.34&-0.51, P < 0.01) (Fig. 4).Fig. 2Comparison of VAS score between baseline (Day 0) and 2-weeks among 3 different administration methods: (a) the change in VAS score of total nasal scores. (b) The change of VAS score in nasal obstruction. (c) The change of VAS score in rhinorrhea. (d) The change of VAS score in olfaction. (e) The change of VAS score in facial pain. Data were shown as mean and SD. VAS, visual analogue scale. SNOT22, Sinonasal Outcome Test 22. NP, endoscopic nasal polyps. SD, standard deviation. ns, no significant. ∗Significant at P < 0.05. ∗∗Significant at P < 0.01Fig. 3The comparison of the reduction of the SNOT-22 among 3 different administration methods: (a) the change in VAS score. (b) The change in extranasal symptoms. (c) The change in sleep disorder symptoms. (d) The change in neuropsychiatric symptoms. (e) The change in nasal symptoms. (f) The change in facial/ear symptoms. Data are shown as mean and SD. SNOT22, The 22-item Sinonasal Outcome Test. SD, standard deviation. ns, no significant. ∗Significant at P < 0.05Fig. 4The comparison among 3 different administration methods in endoscopic nasal polyps scores. Data are shown as mean and SD. SD, standard deviation. ns, no significant. ∗Significant at P < 0.05 Comparison of VAS score between baseline (Day 0) and 2-weeks among 3 different administration methods: (a) the change in VAS score of total nasal scores. (b) The change of VAS score in nasal obstruction. (c) The change of VAS score in rhinorrhea. (d) The change of VAS score in olfaction. (e) The change of VAS score in facial pain. Data were shown as mean and SD. VAS, visual analogue scale. SNOT22, Sinonasal Outcome Test 22. NP, endoscopic nasal polyps. SD, standard deviation. ns, no significant. ∗Significant at P < 0.05. ∗∗Significant at P < 0.01 The comparison of the reduction of the SNOT-22 among 3 different administration methods: (a) the change in VAS score. (b) The change in extranasal symptoms. (c) The change in sleep disorder symptoms. (d) The change in neuropsychiatric symptoms. (e) The change in nasal symptoms. (f) The change in facial/ear symptoms. Data are shown as mean and SD. SNOT22, The 22-item Sinonasal Outcome Test. SD, standard deviation. ns, no significant. ∗Significant at P < 0.05 The comparison among 3 different administration methods in endoscopic nasal polyps scores. Data are shown as mean and SD. SD, standard deviation. ns, no significant. ∗Significant at P < 0.05 Comparison of efficacy differences among CRS endotypes In type 2 CRS, oral steroids showed a significantly greater reduction in NP scores compared to both nasal spray (the comparison between groups, −1.5&-0.56, P = 0.01) and nasal drop (the comparison between groups, −1.50&-0.48, P < 0.01) (Fig. 5c), with no significant difference in subjective symptom score reductions (Fig. 5a–b). In non-type 2 CRS, no significant difference was observed in subjective or objective outcomes (Fig. 5d–f), though the nasal drop showed a trend toward greater NP score reduction versus nasal sprays.Fig. 5The comparison of efficacy among 3 different administration methods between type 2 and non-type 2 CRS: (a) the change of VAS score in type 2 CRS. (b) The change of the SNOT 22 score in type 2 CRS. (c) The change of endoscopic nasal polyps scores in type 2 CRS. (d) The change of VAS score in non-type 2 CRS. (e) The change of the SNOT 22 score in non-type 2 CRS. (f) The change of endoscopic scores in non-type 2 CRS. Data are shown as mean and SD. CRS, chronic rhinosinusitis. VAS, visual analogue scale. SNOT22, The 22-item Sinonasal Outcome Test. SD, standard deviation. ns, no significant. ∗Significant at P < 0.05. ∗∗Significant at P < 0.01 The comparison of efficacy among 3 different administration methods between type 2 and non-type 2 CRS: (a) the change of VAS score in type 2 CRS. (b) The change of the SNOT 22 score in type 2 CRS. (c) The change of endoscopic nasal polyps scores in type 2 CRS. (d) The change of VAS score in non-type 2 CRS. (e) The change of the SNOT 22 score in non-type 2 CRS. (f) The change of endoscopic scores in non-type 2 CRS. Data are shown as mean and SD. CRS, chronic rhinosinusitis. VAS, visual analogue scale. SNOT22, The 22-item Sinonasal Outcome Test. SD, standard deviation. ns, no significant. ∗Significant at P < 0.05. ∗∗Significant at P < 0.01 Comparison of efficacy with different treatment durations We found that extending treatment from 1 week to 2 weeks significantly improved SNOT-22 scores in the nasal drop group (the comparison between groups, −10.74&-18.07, P < 0.01) (Fig. 6b). NP scores improved only in the nasal spray group after extended use (the comparison between groups, −0.10&-0.53, P < 0.01) (Fig. 6c). We further explored the efficacy across different CRSwNP endotypes (Fig. 7). In patients with type 2 CRSwNP, extending nasal spray therapy from 1 week (n = 31) to 2 weeks (n = 25) yielded significant efficacy in SNOT22 (−5.23 vs −12.28, P < 0.01) and endoscopic score (−0.09 vs −0.56, P < 0.01). Besides, significant efficacy in SNOT22 (−9.75 vs −18.03, P < 0.01) was also observed in extending nasal drop therapy from 1 week (n = 30) to 2 weeks (n = 29). In the non-type 2 cohort, the significant improvement in VAS score from 1 week (n = 8) to 2 weeks (n = 15) was observed exclusively in the spray group (−0.75 vs −4.46, P = 0.04). Notably, no significant differences were detected between the one-week and two-week regimens in the oral steroid group across both endotypes, suggesting a one-week course of oral steroid therapy may be sufficient to achieve maximal benefit. However, due to the fact that the 1-week and 2-week data were derived from different cohort, potential biases in clinical characteristics may exist.Fig. 6The comparison of the efficacy between using 3 different short-course steroids for 1 week and 2 weeks: (a) the reduction of VAS score. (b) The reduction of SNOT 22. (c) The reduction of endoscopic nasal polyps scores. VAS = visual analogue scale. SNOT22, Sinonasal Outcome Test 22. Data are shown as mean and SD. SD, standard deviation. ns, no significant. ∗Significant at P < 0.05. ∗∗Significant at P < 0.01. ∗∗∗Significant at P < 0.001Fig. 7The comparison of the efficacy between type 2 and non-type 2 CRSwNP patients while using 3 different short-course steroids for 1 week and 2 weeks: (a) VAS score in type 2 CRSwNP patients. (b) SNOT 22 in type 2 CRSwNP patients. (c) Endoscopic nasal polyps scores in type 2 CRSwNP patients. (d) VAS score in non-type 2 CRSwNP patients. (e) SNOT 22 in non-type 2 CRSwNP patients. (f) Edoscopic nasal polyps scores in non-type 2 CRSwNP patients. VAS = visual analogue scale. SNOT22, Sinonasal Outcome Test 22. Data are shown as mean and SD. SD, standard deviation. ns, no significant. ∗Significant at P < 0.05. ∗∗Significant at P < 0.01. ∗∗∗Significant at P < 0.001 The comparison of the efficacy between using 3 different short-course steroids for 1 week and 2 weeks: (a) the reduction of VAS score. (b) The reduction of SNOT 22. (c) The reduction of endoscopic nasal polyps scores. VAS = visual analogue scale. SNOT22, Sinonasal Outcome Test 22. Data are shown as mean and SD. SD, standard deviation. ns, no significant. ∗Significant at P < 0.05. ∗∗Significant at P < 0.01. ∗∗∗Significant at P < 0.001 The comparison of the efficacy between type 2 and non-type 2 CRSwNP patients while using 3 different short-course steroids for 1 week and 2 weeks: (a) VAS score in type 2 CRSwNP patients. (b) SNOT 22 in type 2 CRSwNP patients. (c) Endoscopic nasal polyps scores in type 2 CRSwNP patients. (d) VAS score in non-type 2 CRSwNP patients. (e) SNOT 22 in non-type 2 CRSwNP patients. (f) Edoscopic nasal polyps scores in non-type 2 CRSwNP patients. VAS = visual analogue scale. SNOT22, Sinonasal Outcome Test 22. Data are shown as mean and SD. SD, standard deviation. ns, no significant. ∗Significant at P < 0.05. ∗∗Significant at P < 0.01. ∗∗∗Significant at P < 0.001 Response rate in subpopulations To precisely identify the optimal treatment strategies for the specific characteristics of type 2 CRS, we categorized EOS count, IgE levels, and the presence of comorbid asthma and allergic rhinitis (AR) for subgroup analysis, followed by chi-square testing. The results demonstrated that patients with low peripheral EOS (<0.3 × 109/L) had significantly different response rates among the 3 treatment groups (P = 0.04) (Fig. 8). Although there is no significant difference (P = 0.27), patients with comorbid AR exhibited higher sensitivity across all treatments, with oral steroids being the most effective.Fig. 8The treatment-sensitive distribution heat maps of different clinical characteristics under different administration regimens. Red indicates the treatment sensitivity rate is greater than 50%, while blue indicates less than 50%. LM score, Lund-Mackey score. AR, allergic rhinitis. EOS, eosinophils. ∗Significant at P < 0.05 The treatment-sensitive distribution heat maps of different clinical characteristics under different administration regimens. Red indicates the treatment sensitivity rate is greater than 50%, while blue indicates less than 50%. LM score, Lund-Mackey score. AR, allergic rhinitis. EOS, eosinophils. ∗Significant at P < 0.05 Safety analysis No serious adverse events occurred. One nasal spray user reported mild epistaxis, which resolved after dose adjustment. Two oral group patients experienced transient gastrointestinal discomfort. One oral group participant withdrew due to an acute upper respiratory infection. Discussion This study demonstrated that the addition of nasal glucocorticoid drop to a nasal spray regimen significantly alleviated nasal obstruction and rhinorrhea symptoms in CRSwNP patients, showing superior efficacy compared to nasal spray alone. Oral glucocorticoids provided a more pronounced reduction of polyp size, particularly in type 2 CRS patients. Extended therapy with nasal drop provided symptom alleviation, while the nasal spray group saw a decrease in polyp size. This suggests that while prolonged use of nasal drops may benefit specific subgroups, a one-week course of nasal spray alone appears insufficient for optimal therapeutic outcomes. Glucocorticoids are widely used in the treatment of CRS due to their potent anti-inflammatory effects. Nasal glucocorticoids, in particular, are extensively applied as they achieve effective local concentrations while exhibiting fewer systemic side effects compared to oral glucocorticoids. In recent years, efforts have been focused on optimizing glucocorticoid regimens to maximize patient benefits in the shortest possible time and to identify candidates who might require surgical intervention despite maximal medical therapy. Numerous studies have explored various glucocorticoid administration strategies for rhinosinusitis patients. However, there remains insufficient evidence to identify the optimal regimen.17 Nasal spray has been widely utilized for decades, with a 2016 meta-analysis of 18 randomized controlled trials (RCTs) confirming the efficacy in improving subjective symptoms.12 However, nasal spray alone is often insufficient, particularly in olfactory improvement, likely due to the difficulty of delivering the medication to the olfactory region.17 Nasal drop represents another effective mode of nasal glucocorticoid delivery. When combined with proper positioning techniques, such as the Mygind method, it can enhance medication delivery to the middle meatus and improve frontal sinus drainage.14,18,19 Our study was designed as an add-on model, in which additional interventions were evaluated on top of baseline nasal spray therapy. This design reflects real-world clinical practice, where treatment is often intensified in patients with inadequate response to single nasal spray therapy. Our study also found that the addition of drop-form intranasal corticosteroids to spray-based corticosteroids significantly improved extranasal symptoms, including cough, postnasal drip, and purulent discharge. Moreover, a small amount of the medication may reach the throat via the posterior nasal apertures, potentially alleviating certain throat discomfort. Demirel et al compared different nasal drop formulations and found that using 400 mg/day of fluticasone drops for 12 weeks significantly increased nasal cavity volume. Notably, administering 400 mg twice daily resulted in significant volume improvement by week 4 and was the only regimen that demonstrated olfactory improvement.20 Xu et al reported that adding budesonide drops (1 mg/day) to nasal spray for 1 week significantly improved symptoms and polyp score, with outcomes comparable to oral glucocorticoids. However, no significant olfactory improvement was observed across groups.9 Building upon the study by Xu et al , our research extended the therapy across groups and observed significant symptom improvement only in the nasal drops, while no similar gains were noted in the nasal spray or oral glucocorticoid groups. These findings suggest that one-week treatment may be sufficient for most patients, and prolonged therapy may only benefit selected individuals. Although no statistically significant difference was found between nasal-drop corticosteroids and oral corticosteroids in terms of symptomatic improvement, a tendency toward better overall response was observed in the nasal drop group. This may be attributed to the adverse effects associated with systemic corticosteroids, which could offset the symptomatic relief they provide. In contrast, topical agents, owing to their favorable safety profile, may offer more net clinical benefit by avoiding systemic side effects. However, it should be noted that the 1-week and 2-week data originated from different groups, which might disrupt the continuity of efficacy assessment and compromise the validity of the study. Therefore, longitudinal follow-up of a single cohort would be necessary to substantiate these findings. Chronic rhinosinusitis is categorized into type 1, type 2, and type 3 based on immunological endotypes, with type 2 inflammation, characterized by EOS infiltration, being predominant in most nasal polyps.21 Type 2 CRSwNP patients are generally more responsive to glucocorticoid therapy. Our study found that oral glucocorticoids significantly reduced polyp size in type 2 CRSwNP patients. This aligns with findings by Wen et al,22 which suggest that increased neutrophilic infiltration in polyp tissue may be associated with reduced responsiveness to oral glucocorticoids. Additionally, Milara et al reported that glucocorticoid resistance in nasal polyp patients may be linked to overexpression of MUC423 and downregulation of MUC1.24 Although the precise mechanisms remain unclear, our findings support the recommendation of nasal drops for symptom relief in type 2 CRSwNP patients, while individualized treatment strategies should be explored for non-type 2 patients. However, the study included a limited number of patients with non-type 2 chronic rhinosinusitis. As a result, the statistical power for analyzing this subgroup was low. A larger cohort of non-type 2 patients is required to draw more reliable conclusions. Future studies should further investigate the relationship between nasal polyp endotypes and treatment efficacy to achieve precision medicine. When nasal glucocorticoids prove ineffective, clinicians often resort to oral glucocorticoids. However, oral administration can lead to various side effects, including gastrointestinal disturbances, insomnia, and mood changes, particularly with prolonged use.8,25 In contrast, long-term use of nasal glucocorticoid is considered safe and effective.5 Zhang et al suggested that maintaining oral prednisone doses below 50 mg/day can maximize therapeutic benefits while minimizing adverse effects.26 In our study, 2 cases of adverse effects were observed in the oral glucocorticoid group, presenting as mild abdominal pain, which resolved with symptomatic treatment. One case developed facial and back erythema within 1 week after 2 weeks’ use of oral steroids. This finding underscores the overall safety of short-term oral glucocorticoids but highlights the need for strict dosage control and close monitoring to ensure safety and efficacy in clinical practice. Meanwhile, although glucocorticoid therapy lasting less than 3–4 weeks is less likely to cause HPA axis suppression, any dose greater than the physiologic daily equivalent dose can still potentially lead to HPA axis suppression.27 Therefore, it is necessary to measure morning serum cortisol for patients who have been prescribed glucocorticoids for a relatively short duration or patients who develop symptoms indicative of adrenal insufficiency as the glucocorticoid dose is reduced.28 In recent years, the rapid development of biologics has provided new options for the precision treatment of chronic rhinosinusitis (CRS), particularly in patients with nasal polyps. Targeting molecules such as IgE, IL-4, IL-5, IL-13, IL-33, and TSLP, drugs like omalizumab, dupilumab, tezepelumab, and mepolizumab had demonstrated significant efficacy.29, 30, 31 A retrospective analysis by Gurnell et al revealed that dupilumab, by blocking the IL-4Rα receptor and reducing Th2 inflammatory cytokines, not only significantly improves nasal symptoms but also markedly reduces the frequency of systemic corticosteroid use.32 This effect is likely associated with its ability to control inflammation in both the upper and lower airways, offering an alternative for patients dependent on systemic corticosteroids. Conclusion This study demonstrated that adding nasal corticosteroid drop significantly enhances nasal obstruction and rhinorrhea in CRSwNP patients, while oral steroids excel in reducing polyp size, particularly in type 2 CRS. Extended therapy with nasal drops may offer additional benefits to certain subgroups, whereas a 1-week nasal spray regimen alone is inadequate to achieve optimal outcomes. These findings underscore the importance of personalized treatment strategies based on CRS endotypes and symptom profiles. Data sharing statement The data that supports the findings of this study can be obtained from the corresponding author upon reasonable request. Author contributions Yinyan Lai and Jianbo Shi conceptualized and supervised the study. Ziqi Huang and Zhaofeng Xu developed the methodology of the study, performed the visualization and drafted the manuscript. Ziqi Huang analyzed the data. Tingyan Wang and Xiaolin Chen performed the enrollment of patients and clinical interventions. Wufuerjiang Yiming performed the specific outcome assessments. Yinyan Lai and Jianbo Shi reviewed and revised the content. Ethics statement The study was approved by the Ethics Committee for Clinical Research and Animal Trials of The First Affiliated Hospital of Sun Yat-sen University (2023879). Disclosure of the use of generative AI and AI-assisted technologies The author employed the DeepSeek and the ChatGPT with the aim of improving textual fluency and optimizing lexical choices. Upon utilizing these tools, the author conducted a critical review and made necessary revisions to the content. The ultimate responsibility for the published work rests with the author. Funding This study was supported by grants from the 10.13039/501100001809National Natural Science Foundation of China (81970854, 82301282, 82471145) and the 10.13039/501100003453Natural Science Foundation of Guangdong Province (2025A1515012577). Declaration of competing interest The authors declare that they have no relevant conflicts of interest.