Impact of Topical Corticosteroid Irrigations on Sinonasal Wound Healing

Abstract

Background

Chronic rhinosinusitis with nasal polyps (CRSwNP) is a chronic inflammatory disease frequently requiring surgical intervention and the long-term use of topical corticosteroids to control patient symptoms. However, corticosteroids may delay postoperative recovery following sinus surgery by disrupting tissue inflammation, a key mediator of early wound healing. This study therefore seeks to assess the impact of topical corticosteroids on sinonasal epithelial healing following local injury.

Methods

Primary human nasal epithelial cells (HNECs) were collected from two patients with CRSwNP and differentiated at air–liquid interface (ALI). A linear scratch was introduced to each monolayer of differentiated HNECs, followed by apical treatment with or without budesonide (2 μg/mL) to simulate topical use. Live-cell imaging assessed time to wound closure. Epithelial barrier function was assessed with trans-epithelial electrical resistance (TEER) before the injury as a baseline, immediately post-injury, and at 12 and 24 hours.

Results

All wounds closed within 19 hours. The average time to wound closure was 15.1 hours for CRSwNP HNECs exposed to budesonide and 12.3 hours for those exposed to control media. This difference was statistically significant (P = .0033) despite a relatively small cohort. TEER values universally increased following an initial drop from baseline induced by scratch injury. No statistically significant difference in TEER recovery was observed between groups.

Conclusion

Wound closure was delayed in CRSwNP HNECs exposed to budesonide versus control; however, no wounds failed to close, and no difference was identified in the return of epithelial barrier function. Topical corticosteroids may be safely initiated 24 hours following ESS.

Introduction

Chronic rhinosinusitis with nasal polyps (CRSwNP) is a recalcitrant disorder characterized by the presence of nasal polyps and persistent sinonasal inflammation.^1–3 Topical corticosteroids via inhaled sprays and irrigation devices are a mainstay of treatment for CRSwNP. Current treatment guideline advocate for the use of high-volume, low pressure topical corticosteroids following endoscopic sinus surgery (ESS) to improve long-term disease control. ⁴

Benefits of topical corticosteroids following ESS include improved endoscopic scores and patient reported outcome measures.^5,6 While timely initiation of these medications is important to maximize benefits, the inflammatory response they suppress is intimately involved in the process of wound healing. Corticosteroids adversely affect wound healing and exert their strongest influence in the inflammation-dependent early phases of reepithelization. ⁷ Recognizing this, widely cited evidence-based recommendations note that the optimal time to initiate topical corticosteroid therapy following ESS remains uncertain. ⁸ Therefore, we aimed to characterize the effects of topical corticosteroids on wound healing using human derived, in vitro models of CRSwNP.

Materials and Methods

Tissue Collection

Appropriate Institutional Review Board (IRB00102406) approval from Emory University was obtained for this study. Nasal polyps were collected intraoperatively from the ethmoid sinuses of consecutive adult patients undergoing elective ESS for CRSwNP. Exclusion criteria included comorbid cystic fibrosis, autoimmune vasculitis or history of non-inflammatory disease affecting the sinonasal cavities.

Tissue Processing and Cell Culture

Tissue processing and cell culture followed established protocols, as previously described. ⁹ Briefly, collected nasal polyp tissues were morselized using dissecting scissors and added to DMEM/F12 media containing 0.5% w/v protease (Sigma-Aldrich, Missouri, USA, #P5147) and 10% w/v DNase (Sigma-Aldrich, Mannheim, Germany, #11284932001). Protease digestion was halted at 4 hours by adding 10% FBS (Corning, California, USA, #35-015-CV) and filtering the tissue via a 100-m sieve. The flow-through was collected, centrifuged at 450 x g for 8 min, and washed in Hanks’ Balanced Salt Solution (HBSS) (Thermo Fisher, New York, USA, #14175-095). Red cell lysis buffer (Sigma-Aldrich, Missouri, USA, #R7757) was added to the pellet which was then incubated at room temperature for 5 minutes. Cells were resuspended in conditional reprogramming culture medium. ⁹ For basal epithelial cell growth, the resulting cell suspension was seeded into cell culture flasks containing irradiated 3T3 fibroblast feeder cells (ATCC #SCRC-1010). Following growth, the cells were cultured for 14 days in E-ALI medium on collagen-coated 24-well transwell plates at air–liquid interface (ALI). ⁹ Culture media was changed every 48 hours during this period. The standard culture media contains hydrocortisone in solution; therefore, a final media change was performed on day 12 of ALI culture using a steroid-free media. All experiments were conducted on day 14 of ALI culture.

Scratch Assay and Budesonide-Exposure

Epithelial wounding was manually performed using a 3-μL pipette tip. A single longitudinal scratch was created across the diameter of each well. Wells were washed twice with DPBS to remove detached cells, and media with or without budesonide was added according to the experimental group. Stock solution of budesonide was generated by reconstituting 20 mg budesonide (Biovison CAS# 51333-22-3) in 1 mL dimethyl sulfoxide (DMSO). Budesonide containing media was prepared by adding the stock solution to steroid-free ALI media to a concentration of 2 μg/mL. This is the same concentration used clinically for nasal rinses at Emory Sinus, Nasal, and Allergy Center. Control media contained hydrocortisone-free ALI media with 20 μL/mL DMSO vehicle, matching the DMSO concentration used in the treatment media.

Confocal Imaging and Analysis

Images were taken on the CSU-W1 SoRa spinning disk laser scanning confocal microscope (Nikon, Tokyo, Japan). 24-well plates were placed in a stage-top incubator and coordinates were defined along the scratch to automate image capture over time. Scratches were imaged immediately after loading the plates and at 30-minute intervals thereafter for 24 hours (Supplemental Video S1). Collection time was based on preliminary experiments demonstrating wound closure within 24 hours (data not shown). Images were analyzed using an ImageJ/Fiji plugin, Wound Healing Size Tool, ¹⁰ to determine wound areas in microns.

Epithelial Barrier Function

Trans-epithelial electrical resistance (TEER) was measured across CRSwNP human nasal epithelial cells (HNECs) as a proxy to evaluate the influence of corticosteroid exposure on return of epithelial barrier function during wound healing in vitro. On day 14 of ALI culture, prior to wounding, TEER measurements were obtained using the CellZscope 2 (NanoAnalytics, Münster, Germany, CZS0421918) to establish baseline TEER values. Measurements were then repeated in both control and budesonide treated cohorts immediately after scratching, and at 12 and 24 hours after wounding. Transwell culture inserts were placed into the CellZscope and media was changed for each measurement according to the device manufacturer; 350μl of control or budesonide media was added apically and 500 μL was added basolaterally. The CellZscope was placed in an incubator for the duration of data collection.

Statistical Analysis

Ad hoc power analysis was not completed for this pilot study. Experiments from each participant were run in parallel with three replicates per assay. For initial wound area and TEER measurements, control and budesonide treated cohorts were analyzed using unpaired, two-tailed Student's t-tests. Mixed-effects analysis was used to examine differences in wound area over time between control and budesonide treated cohorts. Statistical significance was defined as P ≤ .05. Data were analyzed and graphics generated in GraphPad Prism version 9 (GraphPad Software, Boston, MA).

Results

Tissue Samples

Two nasal polyp samples obtained from different patients with CRSwNP were used in this study. This included a 54-year-old Asian male and a 45-year-old Caucasian male. Technical replicates were plated in multiple wells for TEER measurements and live-cell imaging of wound closure.

Wound Closure in CRSwNP HNECs

Epithelial wound closure was defined as the earliest image interval where a gap could no longer be visually identified between the migrating wound edges (Figure 1). One control well demonstrated progressive cell death without migration of wound edges and this replicate was excluded from statistical analyses. All other wounds were closed after 19 hours. Wound area was measured as the space unoccupied by cells in each image (Supplemental Video S1). Initial wound areas did not differ between groups (P = .47) suggesting no difference in the size of the wound generated by each scratch (Figure 2). The average time to wound closure for cohorts exposed to budesonide containing media (n = 12) was 15.1 hours compared to 12.3 hours for the media-only control (n = 11) population; this difference was statistically significant (P = .0033) (Figure 3A). Analysis of residual wound area over time supported these findings (P < .0001) (Figure 3B).

Figure 1.

Wound closure in CRSwNP HNECs after scratch. Time-lapsed images of control (top) and budesonide (bottom) treated CRSwNP HNECs. Images are immediately after the scratch (0H), during wound healing (5H), and right before wound closure (10H). Wound edges are highlighted in red. Scale bar represents 200 μm.

Figure 2.

Initial wound area in CRSwNP HNECs. Bar graph showing the wound area immediately after scratch in CRSwNP HNECs exposed to both steroid-free (control) and budesonide containing media. Data represented as mean ± standard error. Student’s t-test. n = 11–12 wells.

Figure 3.

Assessment of wound closure after scratch injury in CRSwNP HNECs. Bar graph showing time to wound closure in control and budesonide treated cells (A). Line graph depicting changes in wound area over time in CRSwNP HNECs exposed to budesonide or steroid-free (control) culture media (B). Student’s t-test and mixed-effects analysis. n = 11 control, 12 budesonide treated. **P ≤ .01; ****P ≤ .0001.

Epithelial Barrier Function

TEER measurements were performed on CRSwNP HNEC models to evaluate the impact of topical corticosteroids on the recovery of epithelial barrier function following local wounding. TEER values were first measured on the intact barrier prior to scratch and used for normalization (Figure 4). Immediately following scratch injury, as expected, TEER values in both treatment groups uniformly plummeted compared to pre-scratch values (∼1.2 Ω cm² pre-scratch vs ∼0.2 Ω cm² at 0H, P ≤ .001) demonstrating the disruption of the epithelial barrier due to the scratch (Figure 4). However, measurements at 12 hours revealed increased resistance compared to values recorded immediately after the scratch, with TEER values rising to an average of nearly 0.5 Ω cm² for the control group and an average of 0.6 Ω cm² for budesonide group (P ≤ .05). At 24 hours, TEER measurements further increased (∼0.8 Ω cm² for control and 1 Ω cm² for budesonide group) compared to the 12-hour values, indicating a return to baseline barrier function. Importantly, there was no difference in TEER values between control and budesonide treated cell lines at any timepoint. A summary of the statistical comparisons is provided in Supplemental Table S1.

Figure 4.

Recovery of epithelial barrier function after scratch injury. Bar graph depicting the normalized TEER (Ω·cm²) before injury (pre-scratch), immediately after injury (0), 12 and 24 hours post-injury. Data represents mean with standard error (± SEM). Student’s t-test. n = 6 wells. *P ≤ .05; ***P ≤ .001.

Discussion

This study examined the effects of topically applied corticosteroids on early wound healing using an in vitro model of human-derived CRSwNP epithelium. Specifically, we examined the influence of topical budesonide at a clinical concentration of 2 μg/mL on the time to wound closure, as well as the recovery of baseline epithelial barrier function. The closure of mucosal wounds represents an early stage in the postoperative recovery following ESS. Scratch assay was therefore utilized to model this stage of recovery. Primary findings include the prolonged closure of epithelial wounds when exposed to budesonide, without evidence of a difference in recovery in epithelial barrier function.

Historically, investigations of the effects of corticosteroids on wound healing in sinonasal mucosa have been conducted in animal models, where outcomes are evaluated using histopathology.^11–16 The lack of a commonly available CRSwNP-phenotype animal model has restricted the study of sinonasal wound repair to healthy sinus tissues, limiting the ability to extrapolate findings to patients with inflammatory disease. In turn, this lack of data limits the creation of evidence-based recommendations for when to begin these treatments following ESS.

In vitro models of wound healing in sinonasal tissues have been successfully employed to investigate the effects of a range of agents on wound repair in primary HNECs and primary human nasal fibroblasts (HNF).^17–20 To date, similar experiments have not been conducted to evaluate effects of corticosteroid exposure in early phases of wound healing. The initial closure of sinonasal wounds occurs via the migration of epithelial cells, with the highest rates of epithelial cell migration occurring within the initial 24 hours after wounding. ⁷ This phase of wound healing coincides with the acute inflammatory response in injured tissues and represents a critical period during which steroid medications may influence wound repair. Our study found that topical corticosteroid exposure prolonged initial wound closure in primary HNECs. Closure of scratch wounds in cells exposed to budesonide occurred in 15.1 hours compared to 12.3 hours in controls (P = .0033). While this result is statistically significant, the clinical significance of these findings is uncertain. Consequently, we sought to determine if corticosteroid exposure also influenced the recovery of epithelial barrier function as risk of postoperative infection with delayed wound healing has been cited as a rationale for delayed initiation of topical corticosteroids after ESS. ²¹

TEER is universally disrupted with a mechanically injured cell layer, recovering after epithelial resealing has occurred. Additionally, inflammation also lowers epithelial barrier function and is considered a pathognomonic finding in several forms of CRSwNP. ⁴ Recovery of TEER after a mechanical injury therefore depends not only on epithelial resealing, but also on modulation of the inflammatory response driving the wound healing processes. When considering the appropriate time to initiate postoperative corticosteroid irrigations, it is important to note that corticosteroid exposure may initially slow epithelial resealing, then paradoxically increase TEER values by decreasing baseline inflammation. In this study, we demonstrated a trending elevation of normalized TEER values in both cohorts following scratch injury, without a significant difference between control and corticosteroid exposed cohorts (Figure 4). This suggests that there may be no functional difference in healing of epithelial wounds exposed to topical steroids at clinically utilized concentrations despite a relative delay in the rate of epithelial cell migration during healing. There were no wounds in either exposure group which persisted longer than 16 hours except for one control well which exhibited nonviable cells. This further suggests that the initial phase of wound healing is completed rapidly and the window during which topical corticosteroid exposure may exert negative influences on healing passes quickly. This correlates with the findings of Hosemann et al in early studies of nasal mucosal healing in animal models. ¹³ We acknowledge that our findings are based on the specific dose used in this study, and it is possible that different doses could have varying effects on barrier function. However, we selected this dose as it reflects the clinical concentration used at our institution. Ramezanpour et al conducted a study examining the effects of topical treatments on the sinonasal epithelial barrier in vitro. ²² They used budesonide inhalation suspension at a concentration of 1 mg/2 mL and observed a significant increase in TEER in CRS patients after 24 hours of exposure. Additionally, they reported no significant effect on paracellular permeability in HNEC monolayers or alterations in the localization of ZO-1. ²² To date, no studies have systematically investigated the dose–response effects of budesonide on HNECs in vitro. Future research exploring whether different doses could compromise epithelial barrier structure and function would be valuable in further understanding the impact of corticosteroid exposure on sinonasal epithelial integrity.

A unique strength of our in vitro approach is that HNECs at ALI have been shown to retain their inflammatory phenotype. ¹⁷ Utilizing a model system with characteristic inflammation consistent with CRSwNP is important to demonstrate the impact of topical corticosteroids in this inflammatory condition. Notably, most data associating corticosteroid exposure with delayed epithelial healing have been generated in non-inflamed tissues. This includes many studies of mucosal healing in animal models.^14,23 After finding delayed re-epithelialization with topical mometasone after maxillary sinus stripping in a rabbit model, Jain et al postulated that this may be attenuated in tissues with chronic inflammation. Our study indicates that while topical corticosteroids slow epithelial wound closure in an in vitro CRSwNP model, they do not hinder barrier recovery. The clinical relevance of this delay remains unclear, as wound healing in vivo is a complex process influenced by various cellular and molecular factors not captured in our model. Additionally, our findings do not define the optimal timing for postoperative corticosteroid rinses, given the differences between our model and the postoperative nasal mucosal environment. Since epithelial migration is most active within the first 24 hours after injury, initiating corticosteroids on postoperative day 1 may offer a balance between supporting wound repair and providing early anti-inflammatory effects. However, this should be interpreted with caution. Further in vivo studies and clinical trials are necessary to determine the safest and most effective timing for corticosteroid use after ESS.

This study has several limitations that must be acknowledged. First, epithelial wound repair in vivo is a complex process involving multiple cell types and mediators that are not captured in our in vitro HNEC model. Second, our investigation focused on the effects of topical budesonide and did not include other corticosteroids such as mometasone that are currently used in clinical practice. Additionally, only male HNECs were utilized, which may limit the applicability of our findings, as biological sex can influence wound healing and inflammatory responses. Finally, the small cohort of patients and lack of diverse representation of CRSwNP further limit the generalizability of our findings. Future study incorporating a diverse population of patients with CRSwNP, as well as different inflammatory CRSwNP endotypes, is needed.

Conclusions

This study evaluates the impact of topical budesonide exposure in CRSwNP human nasal epithelia during the initial phase of wound healing. Exposure to budesonide prolonged closure of wounds in epithelial monolayers, but the clinical significance of this delay is yet to be established. No significant differences with budesonide exposure were observed in the recovery of epithelial barrier function following epithelial wounding. Since epithelial migration is most active within the first 24 hours after injury, starting corticosteroid-containing nasal rinses on postoperative day 1 may provide a balance between supporting wound healing and reducing inflammation. However, additional in vivo studies are necessary to determine the safest and most effective timing for corticosteroid use after surgery.

Supplemental Material

Video 1. Example video of wound closure after scratch. Time-lapsed images of control (left) and budesonide-treated (right) CRSwNP HNECs after scratch. Time indicated at the top in hours. Video stopped at 14 hours when wounds were fully closed. The wound area in each field of view is outlined in yellow.