Introduction
Background
Chronic rhinosinusitis is defined as chronic inflammation of the nasal and sinus mucosa lasting over 12 weeks, characterised by core symptoms including nasal obstruction, mucoid and/or mucopurulent rhinorrhoea, cephalofacial distension and/or pain, and hyposmia or anosmia. With an incidence of approximately 4.8 per cent to 9.7 per cent and an annual increasing trend,Reference Goshtasbi, Abouzari, Abiri, Yasaka, Sahyouni and Bitner1 chronic rhinosinusitis severely impairs patients’ work capacity and quality of life.
Functional endoscopic sinus surgery (FESS) is currently the first-line surgical treatment for chronic rhinosinusitis, allowing thorough resection of nasal and sinus lesions under direct endoscopic visualisation and restoration of nasal-sinus ventilation and drainage functions.Reference Fokkens, Lund, Hopkins, Hellings, Kern and Reitsma2 However, standalone FESS is associated with high rates of post-operative mucosal oedema, scar adhesion and disease recurrence,Reference Amedee3 requiring some patients to undergo long-term medication or even re-operation.Reference Stankiewicz, Lal, Connor and Welch4
In recent years, intra-operative implantation of middle meatus drug-eluting stents has gained scholarly recognition. Implanted in the surgical cavity, these stents not only maintain sinus ostium patency but also sustainably release anti-inflammatory agents to mitigate local inflammation.Reference Huang, Hwang, Sun and Zhou5 Existing research on stent efficacy mostly relies on patients’ subjective symptom feedback, while their long-term effectiveness awaits multidimensional verification.
Objectives
This study aimed to comprehensively assess the clinical value of middle meatus drug-eluting stents from the perspectives of subjective symptoms, mucosal recovery and inflammatory factor levels to provide evidence for clinicians to use in selecting optimal treatment regimens.
Materials and methods
Study design
This was a single-centre, parallel-group, non-randomised controlled study conducted at Changsha Central Hospital from January 2023 to June 2024. Eligible patients were assigned to either the stent group or the control group based on their treatment preference after full informed consent, rather than random allocation. The study was approved by the Institutional Review Board (Approval No. 2024-232) and performed in accordance with the Declaration of Helsinki and Consolidated Standards of Reporting Trials 2010 guidelines. All participants and their families provided written informed consent.
Outcome assessors and statistical analysts were blinded to group assignment, while patients and the operating surgeon were unblinded because of the nature of stent implantation. A Consolidated Standards of Reporting Trials flow diagram illustrating participant screening, enrolment, allocation and follow up is shown in Figure 1.
The Consolidated Standards of Reporting Trials (CONSORT) flow diagram illustrates participant screening, enrollment, randomization and follow-up analysis.
Figure 1 Long description
The process of participant allocation and follow-up in a study begins with ′Assessed for eligibility n equal 72′. From there, participants are divided into two groups: ′Excluded n equal 12′ and ′Enrolled, allocated by patient choice n equal 60′. The ′Excluded′ group is further divided into ′Unmet inclusion/exclusion criteria n equal 8′ and ′Declined to participate n equal 4′. The ′Enrolled′ group splits into ′Experimental group n equal 30′ and ′Control group n equal 30′. Both groups received assigned treatment, with ′Received assigned treatment n equal 30′ for each. Both groups experienced ′Lost to follow-up n equal 0′, followed by ′Completed 12m follow-up n equal 30′ and finally ′Included in final analysis n equal 30′.
A total of 72 patients with CRSwNP were initially screened. Among them, 8 patients were excluded because they met the exclusion criteria (3 with allergies to stent materials, 2 with severe heart disease, 1 with acute-phase sinusitis, 1 pregnant woman and 1 intolerant to general anaesthesia) and 4 patients refused to participate. Finally, 60 eligible patients were enrolled and assigned to the experimental group and the control group according to their treatment preference, with 30 patients in each group. All 60 patients completed the 12-month follow up.
The inclusion criteria were: (1) met the diagnostic criteria of the European Position Paper on Rhinosinusitis and Nasal Polyps 20202 for CRSwNP; (2) pre-operative standardised medication with poor efficacy; (3) age over 18 years old; (4) able to understand and complete the relevant scales; and (5) signed the informed consent form. The exclusion criteria were: (1) allergy or contra-indication to middle meatus drug-eluting stents or their degradation products; (2) complicated with major systemic diseases (heart, brain, liver, kidney, etc.); (3) intolerance to general anaesthesia; (4) acute-phase sinusitis; (5) pregnant or peripartum women; (6) participating in other clinical trials simultaneously; and (7) previous history of aspirin-exacerbated respiratory disease, bronchial asthma, fungal rhinosinusitis or prior treatment with biologics.
Baseline demographic data were uniformly recorded for all participants, including age, sex and body mass index. Per the predefined exclusion criteria, all enrolled patients were free of co-morbid asthma, aspirin-exacerbated respiratory disease (NSAID‑exacerbated respiratory disease), allergic fungal rhinosinusitis and prior biological agent therapy, which eliminated the potential intergroup imbalance caused by these major confounding factors.
Pre-operative standardised medication and definition of poor efficacy
Pre-operative standardised medication
Patients received a 4-week standardised medical treatment before enrolment, including mometasone furoate nasal spray, 2 sprays per nostril, once daily; montelukast sodium chewable tablets, 10 mg once daily; and saline nasal irrigation, twice daily, morning and evening. Patients with moderate to severe symptoms received oral glucocorticoids, prednisone acetate tablets, 30 mg once daily for 1 week, then gradually reduced to 10 mg once daily for 3 weeks.
Poor efficacy definition
After 4 weeks of standardised medication, patients still met one of the following criteria: Sino-Nasal Outcome Test 22 (SNOT-22) score more than 30 points, Lund–Kennedy score more than 5 points, persistent nasal obstruction, purulent nasal discharge and other main symptoms, with no significant improvement compared with before treatment.
Surgical methods
After admission, pre-operative sinus computed tomography (CT) was evaluated with Lund–Mackay scoring, and statistical comparison confirmed comparable pre-operative lesion severity between two groups (p > 0.05). The sinus opening range was individually customised strictly according to the actual lesion extent on pre-operative CT, including selective ethmoidectomy, maxillary antrostomy, frontal sinusotomy and sphenoidotomy as required. The core surgical principle and mucosal dissection standard were unified for all enrolled patients to minimise operative heterogeneity.
Patients completed relevant pre-operative examinations to rule out contra-indications and all underwent an endoscopic Messerklinger surgical procedure performed by the same surgeon: nasal polyps, uncinate processes and ethmoid bullae were resected first, followed by sequential opening of the maxillary, ethmoid, frontal and sphenoid sinuses according to lesion locations to remove pathological tissues and abnormal anatomical structures for adequate drainage.
Following thorough haemostasis, the experimental group had mometasone furoate drug-eluting stents implanted in the ethmoid sinuses with degradable haemostatic sponges packed inside the stents, while the control group only used the same type and size of degradable haemostatic sponges for packing, and both groups had expandable sponges packed in the common nasal meatus.
Post-operative management included anti-infection treatment with intravenous cefuroxime sodium (1.5 g twice daily) for 3 days followed by oral cefuroxime axetil tablets (0.5 g twice daily) for 4 days, oral mucolytics with eucalyptol, limonene and pinene enteric soft capsules (0.3 g three times daily) for 2 weeks and nasal corticosteroid spray with mometasone furoate nasal spray (2 sprays per nostril once daily) for 6 months.
For follow up, nasal debridement was performed as needed at 2 weeks and 1 month post-operatively, and patients were followed up at 3 months and 12 months to collect the SNOT-22 score, Lund–Kennedy score and MMP-9 detection results in nasal secretions (Figure 2).
Typical Case.
Efficacy evaluation
All patients underwent pre-operative Lund–Mackay scoring based on sinus CT to assess the extent of lesions. The scoring was applied to bilateral maxillary sinuses, anterior ethmoid sinuses, posterior ethmoid sinuses, sphenoid sinuses, frontal sinuses and ostiomeatal complexes. Pre-operatively, and at 3 and 12 months post-operatively, patients underwent SNOT-22 scoring. Pre-operatively, and at 3 and 12 months post-operatively, patients underwent nasal endoscopic Lund–Kennedy scoring. Scores were recorded bilaterally, with lower values reflecting better symptom improvement. Nasal secretions were collected at the same time points and MMP-9 levels were measured via Enzyme-Linked Immunosorbent Assay (ELISA).
Statistical analysis
All data were analysed using SPSS 27.0 software, Armonk, New York, USA. For continuous data (height, weight, SNOT-22, MMP-9, Lund–Kennedy and Lund–Mackay scores), the Shapiro–Wilk test was first used for normality testing. Data conforming to a normal distribution (age) were described as mean plus or minus the standard deviation, and inter-group comparisons were performed using the t-test. Data not conforming to a normal distribution were described as median (interquartile range) and inter-group comparisons were conducted using the non-parametric rank-sum test.
Bonferroni correction was applied to all intergroup pairwise comparisons of MMP-9, SNOT-22 and Lund–Kennedy scores across pre-operative, and 3- and 12-month follow-up time points to control the inflation of type I error caused by multiple statistical testing. All results were considered statistically significant at p less than 0.05.
No formal prospective sample size calculation was performed prior to patient enrolment because participants were consecutively recruited and allocated based on voluntary treatment preference rather than predefined randomisation. The total sample size of 60 patients was determined by the available annual clinical volume of CRSwNP patients meeting the inclusion criteria during the study period (January 2023 to June 2024). Post hoc power analysis was not conducted because post hoc power is widely recognised to produce biased and unreliable results.
Results
General situation
Summarizes the baseline demographic and clinical characteristics of the two groups of patients with CRSwNP. There were no significant between-group differences in gender distribution, age, height, weight, preoperative SNOT-22 score, nasal secretion MMP–9 level, Lund–Kennedy endoscopic score and Lund–Mackay radiological score, indicating that the two groups were comparable at baseline (Table 1).
Comparison of general data between two groups of patients with CRSwNP
Table 1 Long description
General characteristics are listed for a control group and an experimental group, with columns for t and p. Age in years is 43.5 with values 32.75 and 50.25 for the control group, and 38.5 with values 28.25 and 50.25 for the experimental group, with p 0.515 and no t shown. Height in centimeters is 169.83 plus or minus 7.02 for the control group and 167.80 plus or minus 6.60 for the experimental group, with t 1.156 and p 0.253. Weight in kilograms is 68.50 plus or minus 7.66 for the control group and 65.70 plus or minus 9.41 for the experimental group, with t 1.264 and p 0.211. Sino Nasal Outcome Test 22 score is 60.7 plus or minus 20.37 for the control group and 56.97 plus or minus 18.26 for the experimental group, with t 0.748 and p 0.458. MMP 9 level is 3860.88 plus or minus 1119.03 for the control group and 4029.75 plus or minus 1141.77 for the experimental group, with t 0.579 and p 0.565. Lund Kennedy score is 7.33 plus or minus 2.98 for the control group and 7.67 plus or minus 2.81 for the experimental group, with t 0.446 and p 0.657. Lund Mackay score is 12.77 plus or minus 7.66 for the control group and 14.00 plus or minus 7.74 for the experimental group, with t 0.621 and p 0.537.
SD = standard deviation
Comparison of MMP-9, Sino-Nasal Outcome Test 22 and Lund–Kennedy scores in all CRSwNP patients
At both 3 and 12 months post-operatively, patient SNOT-22, Lund–Kennedy and MMP-9 levels were lower than their pre-operative values (all p < 0.05) (Figure 3).
(a) Matrix Metalloproteinase-9 (MMP-9), (b) Sino-Nasal Outcome Test 22 score and (c) Lund–Kennedy score.
Figure 3 Long description
MMP-9. The horizontal axis labels are Preoperative, 3 Month Postoperative, 12 Month Postoperative. The vertical axis shows tick labels 0, 2000, 4000, 6000, 8000. Each polyline represents one subject, with a point at each of the three timepoints connected by straight segments. The polylines use the same line style and there is no legend. Most polylines slope downward from Preoperative to 3 Month Postoperative. Many polylines continue downward from 3 Month Postoperative to 12 Month Postoperative and some are nearly level between these two timepoints. SNOT-22. The horizontal axis labels are Preoperative, 3 Month Postoperative, 12 Month Postoperative. The vertical axis shows tick labels 0, 20, 40, 60, 80, 100. Each polyline represents one subject, with a point at each of the three timepoints connected by straight segments. The polylines use the same line style and there is no legend. Most polylines slope downward from Preoperative to 3 Month Postoperative. Many polylines are lower at 12 Month Postoperative than at 3 Month Postoperative and some are nearly level between these two timepoints. Lund-Kennedy. The horizontal axis labels are Preoperative, 3 Month Postoperative, 12 Month Postoperative. The vertical axis shows tick labels 0, 5, 10, 15. Each polyline represents one subject, with a point at each of the three timepoints connected by straight segments. The polylines use the same line style and there is no legend. Most polylines slope downward from Preoperative to 3 Month Postoperative. Many polylines are similar or lower at 12 Month Postoperative compared with 3 Month Postoperative, with several lines showing small changes between these two timepoints.
Comparison of MMP-9, Sino-Nasal Outcome Test 22 and Lund–Kennedy scores between the two groups at different times
At 3 months post-operatively, MMP-9 level reductions were comparable between the two groups, with no significant difference (p > 0.05). At 12 months, the experimental group had a significantly greater reduction in MMP-9 levels than the control group (p < 0.05). Additionally, the experimental group showed significantly larger reductions in both SNOT-22 and Lund–Kennedy scores than the control group at both 3 and 12 months post-operatively (p < 0.05) (Figure 4).
(a) MMP-9, (b) Sino-Nasal Outcome Test 22 score and (c) Lund–Kennedy score.
Figure 4 Long description
MMP 9. The horizontal axis labels are Preoperative, 3 Months Postoperative, 12 Months Postoperative and 12 Months Postoperative minus Preoperative. The vertical axis shows 0 to 6000. Each category has two bars with error bars, one for Control Group and one for Experimental Group. The bar heights decrease from Preoperative to 3 Months Postoperative and to 12 Months Postoperative for both groups. In 12 Months Postoperative minus Preoperative, the Control Group bar is higher than the Experimental Group bar. A bracket with an asterisk is shown over the two bars for 12 Months Postoperative minus Preoperative. SNOT 22. The horizontal axis labels are Preoperative, 3 Months Postoperative, 12 Months Postoperative and 12 Months Postoperative minus Preoperative. The vertical axis shows 0 to 100. Each category has two bars with error bars, one for Control Group and one for Experimental Group. The bar heights decrease from Preoperative to 3 Months Postoperative and to 12 Months Postoperative for both groups. At 12 Months Postoperative, the Control Group bar is higher than the Experimental Group bar. In 12 Months Postoperative minus Preoperative, the Control Group bar is higher than the Experimental Group bar. A bracket with an asterisk is shown over the two bars for 12 Months Postoperative and another bracket with an asterisk is shown over the two bars for 12 Months Postoperative minus Preoperative. Lund Kennedy. The horizontal axis labels are Preoperative, 3 Months Postoperative, 12 Months Postoperative and 12 Months Postoperative minus Preoperative. The vertical axis shows 0 to 15. Each category has two bars with error bars, one for Control Group and one for Experimental Group. The bar heights decrease from Preoperative to 3 Months Postoperative and to 12 Months Postoperative for both groups. At 12 Months Postoperative, the Control Group bar is higher than the Experimental Group bar. In 12 Months Postoperative minus Preoperative, the Control Group bar is higher than the Experimental Group bar. A bracket with an asterisk is shown over the two bars for 12 Months Postoperative and another bracket with an asterisk is shown over the two bars for 12 Months Postoperative minus Preoperative.
Comparison of Sino-Nasal Outcome Test 22 symptom scores between the two groups at different times
At 3 and 12 months post-operatively, both groups exhibited significantly lower scores in post-nasal drip, purulent nasal discharge, clear nasal discharge, nasal obstruction and olfactory symptoms compared with pre-operative values (p < 0.05). The experimental group showed significantly lower scores in post-nasal drip, purulent nasal discharge, nasal obstruction and olfactory symptoms than the control group at the same time points (p < 0.05), whereas no significant intergroup difference was observed in clear nasal discharge scores (p> 0.05) (Figure 5).
(a) Nasal obstruction symptom, (b) post-nasal drip symptom, (c) olfactory symptom, (d) clear nasal discharge and (e) purulent nasal discharge.
Figure 5 Long description
Five violin plots display symptom scores for nasal obstruction, postnasal drip, olfactory, clear nasal discharge and purulent nasal discharge. Each plot compares control and experimental groups at preoperative, 3 months post-operation and 12 months post-operation. The y-axis represents symptom scores, ranging from 0 to 8. The legend indicates control group and experimental group. 1. Nasal Obstruction Symptom: The experimental group shows lower scores at 3 and 12 months post-operation, indicated by asterisks for statistical significance. 2. Postnasal Drip Symptom: The experimental group scores lower at 3 and 12 months post-operation, with asterisks marking significant differences. 3. Olfactory Symptom: Significant improvement in the experimental group at 3 and 12 months post-operation, marked by asterisks. 4. Clear Nasal Discharge: Both groups show similar scores across all time points, with no significant differences. 5. Purulent Nasal Discharge: The experimental group scores lower at 3 and 12 months post-operation, with asterisks indicating significance. Overall, the experimental group exhibits significant improvement in most symptoms compared to the control group, except for clear nasal discharge, where no significant difference is observed.
Conclusion
This study showed that the SNOT-22 and Lund–Kennedy scores of both groups at 3 and 12 months post-operatively were significantly lower than pre-operatively. The experimental group demonstrated more prominent improvements in all post-operative follow-up indicators and exhibited more favourable trends in core symptom relief compared with the control group in core symptoms at 3 and 12 months post-operatively, which is consistent with the findings of Goshtasbi and Hamour et al.Reference Goshtasbi, Abouzari, Abiri, Yasaka, Sahyouni and Bitner1, Reference Arancibia, Langdon, Mullol and Alobid6 This is probably because middle meatus drug-eluting stents slowly release local glucocorticoids to continuously inhibit mucosal inflammation during the critical post-operative repair period.Reference Hamour, Lee, Wasilewski, Monteiro, Lee and Vescan7, Reference Fieux, Noel, Roozdar, Neves, Yan and Tyler8
MMP-9 is a biomarker for extracellular matrix degradation and can serve as a marker for chronic inflammation and oxidative stress. It plays a key role in the pathogenesis of chronic rhinosinusitis: it can degrade the nasal mucosal basement membrane, damage the epithelial barrier, aggravate inflammatory responses and tissue damage, and its level is positively correlated with the severity of inflammation.Reference Franczak, Skomro, Bil-Lula, Fenton, Lawson and Sawicki9 The results of this study showed that the MMP-9 levels in both groups at 3 and 12 months post-operatively were significantly lower than the pre-operative levels. The MMP-9 level in the experimental group was significantly lower than that in the control group at 12 months post-operatively.
By comparing nasal polyp specimens with healthy nasal mucosa specimens, Lygeros et al. found that the mRNA and protein expression levels of MMP-9 were significantly increased in patients with nasal polyps, and MMP-9 may be involved in the progression of CRSsNP to CRSwNP.Reference Lygeros, Danielides, Kyriakopoulos, Sapardoni, Grafanaki and Stathopoulos10 Both groups removed substantial MMP-9-containing nasal polyp tissue via surgery, slowing its production and leading to rapid, comparable reductions in MMP-9 levels within three post-operative months. In the experimental group, glucocorticoids released by middle meatus drug-eluting stents inhibited inflammatory cell activation during the early mucosal repair phase.
Studies have demonstrated that these glucocorticoids suppress the Nuclear factor kappa-light-chain-enhancer of activated B cells (NF-κB) pathway regulating MMP-9 production by binding the p65 subunit via glucocorticoid receptors to block nuclear translocation, reducing NF‑kappa‑B inhibitor alpha (IκBα) degradation while promoting its expression to sequester NF-κB in the cytoplasm, and decreasing NF-κB binding to the MMP-9 promoter, thereby inhibiting MMP-9 synthesis.Reference Eberhardt, Schulze, Engels, Klasmeier and Pfeilschifter11 Consequently, the experimental group sustained low MMP-9 levels at 12 months post-operatively, confirming that middle meatus drug-eluting stents enable long-term inflammatory factor regulation via early post-operative targeted drug delivery, providing molecular evidence for the long-term stability of chronic rhinosinusitis.
Serial detection of MMP-9 in non-invasive nasal secretions can serve as a convenient post-operative biomarker to dynamically monitor mucosal inflammatory remodelling, identify patients at high risk of post-operative adhesion and polyp recurrence, and guide individualised long-term anti-inflammatory nasal spray intervention without repeated endoscopic examination.
This study has certain limitations. Firstly, group assignment depended on patients’ voluntary treatment preference rather than randomisation, which may introduce inherent selection bias despite balanced baseline lesion severity (Lund–Mackay scores). Secondly, no prospective sample size calculation was performed before enrolment. The sample size of 60 patients was limited by the annual clinical volume of eligible cases within the research period, potentially increasing the risk of type II error and limiting detection of subtle clinical differences. Post hoc power analysis was not conducted because of its well-documented statistical bias.
Thirdly, the single-centre design restricts the external generalisability of our findings. Fourthly, the follow-up period was only 12 months and therefore lacked long-term (≥2 years) data on recurrence, sustained mucosal status and stent safety. Further large-sample, multicentre randomised controlled trials with pre-specified sample size are required to validate our conclusions. In addition, smoking history was not recorded at enrolment. Smoking status is a known confounding factor for CRSwNP inflammation and relevant data should be collected in future studies.
Clinically, the sustained reduction in MMP-9 levels observed in the experimental group at 12 months provides a molecular rationale for improved mucosal healing and reduced inflammation. MMP-9 may serve as a post-operative biomarker to monitor mucosal recovery and identify patients at high risk of recurrence, potentially guiding targeted anti-inflammatory therapy or early intervention. The long-term suppression of MMP-9 by drug-eluting stents supports their role in maintaining sinus mucosal integrity and preventing post-operative adhesion and polyp recurrence.
Regarding safety, no stent-related adverse events of any grade (including bleeding, infection, allergic reaction or stent migration) occurred in the experimental group. The cost of a single drug-eluting stent is approximately 1500–2000 RMB, which may limit widespread use in resource-constrained settings. Future studies should evaluate the cost-effectiveness of stents relative to standard FESS, particularly in high-risk patients (e.g. severe CRSwNP, recurrent disease).
This study suggests that FESS combined with sinus drug stent implantation can effectively improve subjective symptoms, promote sinus mucosal repair in patients with chronic rhinosinusitis and maintain better anti-inflammatory effects in the long term (12 months post-operatively). It has significant advantages, especially in relieving core symptoms such as post-nasal drip, purulent nasal discharge, nasal obstruction and olfactory dysfunction. As a safe and effective treatment regimen, it is worthy of clinical promotion and application. While the stent appears safe with no severe adverse events observed, the small sample size, single-centre design and lack of long-term follow up limit the generalisability of these findings. Large-scale, multicentre randomised, controlled trials with longer follow up and cost-effectiveness analysis are warranted to confirm the clinical value and optimal indications of drug-eluting stents in CRSwNP.
Acknowledgements
This work was supported by the Natural Science Foundation of Changsha City (grant number kq2403176), the Scientific Research Program Project of the Changsha Health Commission (grant number KJ-B2023034) and the Joint Fund of Medical and Health Industry of the Natural Science Foundation of Hunan Province (grant number 2025JJ80548).
Data availability statement
All data are available from the corresponding author.
Competing interests
None declared
