Oral Corticosteroids Following Endoscopic Sinus Surgery for Chronic Rhinosinusitis Without Nasal Polyposis: A Randomized Clinical Trial

Key Points

Question  Are oral corticosteroids beneficial following endoscopic sinus surgery (ESS) in patients with chronic rhinosinusitis (CRS) without nasal polyps?

Findings  In this prospective double-blinded, placebo-controlled, randomized clinical trial of 81 adults with CRS without polyps undergoing ESS, comparing longitudinal differences between treatment groups showed no clinically meaningful differences in Sinonasal Outcome Test-22 (SNOT-22) total scores, SNOT-22 rhinologic subdomain scores, or Lund-Kennedy endoscopy scores at any postoperative time point up to 6 months. However, patients who received prednisone had worse postoperative SNOT-22 psychologic subdomain scores compared with placebo.

Meaning  Oral corticosteroids following ESS for CRS without polyps did not have a measurable benefit in sinonasal outcomes compared with placebo, and may be associated with worse psychologic outcomes.

Importance  Although oral corticosteroids are commonly prescribed following endoscopic sinus surgery (ESS) for chronic rhinosinusitis (CRS) without nasal polyposis, there are little data to suggest that this is a beneficial practice.

Objective  To assess the efficacy of oral corticosteroids following ESS in CRS without polyps.

Design, Setting, and Participants  This prospective double-blinded, placebo-controlled, randomized noninferiority clinical trial conducted in a single academic tertiary rhinology practice included adults with CRS without polyps undergoing ESS. Of 81 patients recruited, 72 completed the study.

Interventions  Patients were randomized into 2 treatment groups: a 12-day postoperative taper of oral prednisone vs matched placebo tablets. All study patients also received a uniform 2-week postoperative regimen of oral antibiotics, fluticasone nasal spray, and saline rinses.

Main Outcomes and Measures  The primary outcome measures were Sinonasal Outcome Test-22 (SNOT-22) scores and Lund-Kennedy endoscopy scores, collected preoperatively and postoperatively at 1 week, 1 month, 3 months, and 6 months. Scores were compared between treatment groups at each time point using longitudinal difference between treatment groups and analyzed using 2-way, repeated measures analysis of variance. Secondary outcome measures included treatment-related adverse effects.

Results  Overall, 72 patients (mean [SD] age, 49.4 [14.9] years; 36 men, 36 women) completed the study, with 33 in the prednisone arm and 39 in the placebo arm. When comparing longitudinal differences between treatment groups, there was no clinically meaningful difference observed in SNOT-22 total (F[4254] = 1.71, η² = 0.01 [95% CI, 0.00-0.05]) or Lund-Kennedy scores (F[4247] = 1.23, η² = 0.02 [95% CI, 0.00-0.50]). In SNOT-22 subdomain analyses, there was no clinically meaningful difference between treatment groups for rhinologic, extranasal rhinologic, ear/facial, or sleep subdomains. However, the prednisone group had worse longitudinal scores for psychological dysfunction compared with the placebo group (F[4254] = 3.18, η² = 0.05 [95% CI, 0.02-0.09]). Reported adverse effects were similar between the 2 treatment groups.

Conclusions and Relevance  In this randomized clinical trial of patients with CRS without polyps, oral prednisone following ESS conferred no additional benefit over placebo in terms of SNOT-22 total scores, SNOT-22 rhinologic subscores, or Lund-Kennedy endoscopy scores up to 6 months after surgery. Patients receiving prednisone, however, did demonstrate worse SNOT-22 psychologic subdomain scores. These results suggest that the risks of oral corticosteroids may outweigh the benefits; thus use of oral corticosteroids after ESS for CRS without polyps should be carefully considered.

Trial Registration  ClinicalTrials.gov Identifier: NCT02748070

Oral corticosteroids are commonly used anti-inflammatory medications for chronic rhinosinusitis (CRS).¹Quiz Ref ID Although there is evidence supporting the use of oral corticosteroids for treating CRS with polyps, there remains a considerable knowledge gap with respect to indications for oral corticosteroids in CRS without nasal polyps.² Oral corticosteroids are commonly prescribed as part of the immediate postoperative medical regimen after endoscopic sinus surgery (ESS) for CRS without polyps, yet there is currently little high-quality evidence supporting this practice. The theoretical anti-inflammatory benefit of postoperative oral corticosteroids is counterbalanced by the well characterized risk of systemic adverse effects associated with this class of medications.³ The 2016 International Consensus of Allergy and Rhinology statement⁴ on rhinosinusitis did not make a recommendation on the use of perioperative oral corticosteroids for CRS without polyps, citing a lack of evidence addressing this issue. In this study, we aimed to assess in a randomized and placebo-controlled fashion the efficacy of postoperative oral corticosteroids in patients with CRS without polyps.

Patients were recruited from the Stanford Sinus Center from August 2015 to May 2019. Inclusion criteria were age older than 18 years, ability to provide informed consent, a diagnosis of CRS without nasal polyps, and appropriate candidacy for ESS based on guidelines published in the International Consensus Statement on Allergy and Rhinology for Rhinosinusitis.⁴ The 2021 ICAR statement on rhinosinusitis⁵ states that oral corticosteroids are an option, but still cites a major lack of high-level evidence. A surgery was considered ESS if at least 1 sinus was surgically addressed; patients undergoing concurrent septoplasty and/or turbinate surgery were eligible. Patients were excluded if they had a diagnosis of aspirin-exacerbated respiratory disease, allergic fungal sinusitis, cystic fibrosis, or immunosuppression. In addition, patients were ineligible if they had any oral corticosteroid use within 30 days prior to their surgery date. Enrollment took place during the preoperative clinic visit.

Sample size was calculated based on a statistical model for a continuous outcome in a parallel group noninferiority trial.⁶ We used a noninferiority limit of 10, based on studies demonstrating that the minimal clinically important difference (MCID) for the 22-item Sinonasal Outcome Test (SNOT-22) is at least 8.9 for surgically treated CRS.⁷ We determined that 35 patients per treatment group, or 70 patients total, were required to achieve 80% certainty that the lower limit of a 1-sided 95% confidence interval would be above the noninferiority limit. We aimed to recruit 80 patients to account for loss to follow-up.

At the preoperative visit, we recorded patient demographics and comorbidities. Preoperative disease burden was measured by the SNOT-22, Lund-Kennedy endoscopy, and Lund-Mackay radiograph scores. Preoperative computed tomography (CT) scans of the sinuses were scored using the Lund-Mackay scoring system by 2 investigators (M.T.C. and J.N.), and discrepant scores were reviewed by a third investigator (N.F.A.), with the mean of all reviewers’ scores as the final assigned score.

Endoscopic sinus surgery was performed under general anesthesia per our routine institutional protocol. Surgery characteristics (primary vs revision, unilateral vs bilateral) and surgical procedures performed were recorded.

Randomization into treatment groups was performed using an online random number generator; patients were sorted based on randomly generated odd vs even numbers. Patients were assigned to 1 of 2 postoperative treatment groups: oral prednisone vs matched placebo tablets. Both medications were manufactured and packaged by a licensed, independent compounding pharmacy and provided to the patient at the preoperative visit by an impartial nurse practitioner. The identity of the treatment was known only to the nurse practitioner providing the medication and a research assistant recording study data. The patient and treating physicians were not made aware of the treatment identity during the entire study period. The database was not accessed by any member of the treatment team during the entire study period. In addition, the person performing statistical analysis was blinded as to which group was the treatment arm and which group was the placebo arm.

All treatment regimens began on the first day following ESS. Oral prednisone was administered in a 12-day postoperative taper in daily doses as follows: 30 mg, 4 days; 20 mg, 4 days; and 10 mg 4 days. The oral placebo tablets were administered in a similar pattern. All study patients also received the same 2-week postoperative regimen of oral antibiotics, fluticasone intranasal spray, and saline rinses. The oral antibiotic was amoxicillin, 825 mg, with clavulanate, 125 mg; if the patient had a penicillin allergy then a different antibiotic was prescribed at the discretion of the physician. After the first 2 postoperative weeks, the nasal medication regimen was tailored based on the patient’s symptoms and clinician judgment.

Medication usage and changes were tracked throughout the study period. Medication lists were also reviewed at each clinic visit, and again at the end of the study by retrospectively reviewing the medical record for prescriptions administered during the 6-month study period. Medications captured included corticosteroid nasal spray, corticosteroid nasal rinse, additional courses of oral corticosteroids, saline rinse, topical nasal antihistamine, topical antibiotic irrigation, and oral antibiotics.

Two primary outcomes were measured: SNOT-22 and Lund-Kennedy endoscopy score. The SNOT-22 is a validated patient-reported sinonasal outcomes measure in which 22 items are rated on a scale of 0 to 5 for a maximum total score of 110, with higher scores representing greater symptom severity. The SNOT-22 questionnaires were completed by patients at the following postoperative clinic visits: 1 week, 1 month, 3 months, and 6 months. Owing to patient travel and physician schedules, the follow-up times ranged as follows: the 1 week visit was conducted between 1 and 2 weeks, 1 month visit between 3 and 6 weeks; 3 months visit between 3 and 4 months, and 6 months visit between 5 and 7 months. In addition to SNOT-22 total scores, scores were recorded in each of the 5 SNOT-22 subdomains that have been shown to be differentially affected by treatment: rhinologic symptoms, extranasal rhinologic symptoms, ear-facial symptoms, sleep dysfunction, and psychological dysfunction.⁸

The Lund-Kennedy score was used to rate 5 endoscopic parameters on a scale of 0 to 2 for each side of the nose (polyps, edema, drainage, crusting, scarring), for a maximum total score of 20 points.⁹ Clinicians performed nasal endoscopy on study patients at their preoperative visit and postoperatively at approximately 1 week, 1 month, 3 months, and 6 months.

Corticosteroid-related adverse effects were studied as a secondary outcome measure. At postoperative visits 1 week, 1 month, and 3 months, an itemized questionnaire was completed by patients to assess for symptoms potentially related to adverse effects of systemic corticosteroid use. The questionnaire was not administered at the 6-month postoperative visit as symptoms reported at this time point would likely have been unrelated to the medication.

All statistical analyses were performed using Stata statistical software (version 15; StataCorp, LP). Baseline characteristics of the prednisone and placebo groups were compared using t test for continuous variables and χ² test for categorical variables. The SNOT-22 and Lund-Kennedy endoscopic scores were compared between prednisone and placebo groups at each time point using t tests. In addition, longitudinal performance of a cohort was analyzed using 2-way, repeated measures ANOVAs; the within-subjects factor (time) was used to report changes in performance over time for the overall cohort, while the between-subjects factor (group) was used to report differences between prednisone and placebo groups over time. The model accounted and adjusted for multiple comparisons. The Greenhouse-Geisser correction was used when sphericity violations were indicated by Mauchly’s test. The ANOVA results (F statistic) were reported with effect size (η²) with 95% CIs, where equality was rejected when the lower estimate of the 95% CI was larger than 0.00. We interpreted η² values of 0.01 as small, 0.05 as medium, and greater than 0.15 as large effect sizes. For medications other than the treatment drug, rates of usage between the two groups were compared using the t test, with the assumption of a normal distribution. Given the variability in medication regimen following the uniform postoperative regimen given in the initial 2 weeks, all results were adjusted for medication usage. Additionally, to assess if other medications were potential confounders of our primary outcome, we conducted a multivariate regression analysis to assess whether any particular medication had a significant influence on SNOT-22 scores. The protocol (Supplement 1) was approved by Stanford University institutional review board (protocol #33096). All participants gave written informed consent.

Overall, 81 patients were randomized (Figure 1). Six study patients (5 in prednisone, 1 in placebo) did not show up to their postoperative appointments. One patient was randomized into the prednisone group but never received their study drug. One patient in the prednisone group did not complete the entire medication course due to a communication error. One patient in the prednisone group withdrew from the study due to postoperative epistaxis necessitating control in the operating room. A total of 72 patients (mean [SD] age, 49.4 [14.9] years; 36 men, 36 women) completed the study, with 33 in the prednisone arm and 39 in the placebo arm.

Characteristics of participants who completed the study are shown in Table 1. An adequate randomization was achieved, as there were no differences between the 2 treatment groups in terms of age, sex, or medical comorbidities. In addition, there was a similar preoperative disease burden as measured by the baseline SNOT-22 score, Lund-Kennedy score, and Lund-Mackay score. Overall, 38 of 72 (52%) patients had undergone ESS prior to participation in this study, and there was no difference in rates of prior ESS between the 2 treatment groups. There were similar rates of procedure types or procedure laterality between the 2 groups.

Primary outcome results are shown in Table 2. The number of each patient cohort achieving a clinically significant improvement, as defined by a change of 8.9 points or greater in SNOT-22 total from baseline, was 11 of 33 (33.3%) for the prednisone group vs 14 of 39 (35.9%) for the placebo group at 1 week, 18 of 31 (58.1%) for the prednisone group vs 21 of 36 (58.3%) for the placebo group at 1 month, 17 of 29 (58.6%) for the prednisone group vs 38 of 35 (80.0%) for the placebo at 3 months, and 15 of 25 (60.0%) for the prednisone group vs 25 of 34 (73.5%) for the placebo group at 6 months.Quiz Ref ID Comparison of overall SNOT-22 scores at each time point revealed no clinically meaningful difference between treatment groups. When longitudinal SNOT-22 score data was considered, Mauchly’s test indicated that sphericity was violated (χ²₉ = 26.29), so the Greenhouse-Geisser correction was applied to all longitudinal SNOT-22 analyses. For the overall study cohort, there was a large effect size observed in longitudinal SNOT-22 scores after surgery (Time, F[4254] = 28.21; η² = 0.31 [95% CI, 0.21-0.38]). However, the effect size (η² = 0.01) and upper bound of the 95% CI (0.05) of the difference in longitudinal SNOT-22 scores observed between prednisone and placebo groups suggest a small effect.

In longitudinal analyses of the SNOT-22 subdomains, there were only small effect sizes (η² <0.05 with all upper bounds of 95% CIs <0.05) observed between the prednisone and placebo groups for rhinologic, extranasal rhinologic, ear/facial, and sleep subdomains. Quiz Ref IDHowever, the effect size for the difference in longitudinal scores for psychological dysfunction between the prednisone and placebo group suggest a medium deleterious effect of prednisone (η² = 0.05).

When longitudinal Lund-Kennedy endoscopic score data was considered, Mauchly’s test indicated that sphericity was violated (χ²₉ = 35.71), so the Greenhouse-Geisser correction was applied to all longitudinal Lund-Kennedy endoscopic score analyses. For the overall study cohort, there was a large effect size observed in longitudinal Lund-Kennedy endoscopic scores following surgery (Time, F[4196] = 24.96; η² = 0.33 [95% CI, 0.22-0.42]). However, the effect size (η² = 0.02) of the difference in longitudinal Lund-Kennedy scores observed between prednisone and placebo groups suggest a small effect, although the 95% CI, (0.00-0.50) did not necessarily rule out a medium or large effect.

Rates of adverse effects are shown in Figure 2. The most commonly reported symptoms overall were headache, difficulty sleeping, and stomach discomfort. However, there were no differences in the rates of any of the 14 symptoms assessed between the prednisone and placebo group at all surveyed time points of 1 week, 1 month, and 3 months. We did not observe any difference in rates of adverse effects in patients of different age groups.

Comparison of rates of postoperative medication use is shown in Table 3. Rates for usage of other medications were comparable between the 2 treatment groups. To assess if any of these other medications confounded the primary outcome, a multivariable regression analysis was performed and demonstrated that no postoperative medications had a meaningful effect on SNOT-22 scores (eTable in Supplement 2).

The findings of this study suggest that post-ESS oral corticosteroids offer no greater benefit than placebo with regard to Lund-Kennedy endoscopy scores, SNOT-22 scores, or SNOT-22 rhinologic subdomain scores after ESS. The lack of observed benefit of postoperative oral corticosteroids in this study alters the risk-benefit calculation associated with prescribing oral corticosteroids in the postoperative period. Quiz Ref IDSystemic corticosteroids have a well characterized potential for multiple adverse effects, including hypertension, hyperglycemia and diabetes, adrenal suppression, weight gain, glaucoma, osteoporosis, fluid retention, and immunosuppression.^1,3,10-14 The risk of complications can be considerably heightened with as little as 0.5 to 1.0 g cumulative dose of oral corticosteroids, equivalent to about 4 lifetime bursts (3-7-day course of daily 40 mg of prednisolone).¹⁵

In addition, oral corticosteroids can provoke psychiatric disturbances in 13% to 62% of patients, with symptoms such as agitation, distractibility, anxiety, and insomnia.^3,16,17 Studies have shown that most of these psychiatric disturbances occur primarily in the first week of drug administration and resolve with cessation.^17,18 There is little data on longer-term psychiatric adverse effects of corticosteroids. Although this study did not show differences in psychological function in the immediate postoperative period, patients receiving prednisone had worse longitudinal SNOT-22 psychological domain subscores, suggesting the potential for longer-term negative effect of oral corticosteroids. Our study suggests that post-ESS oral corticosteroids may subject patients to the risk of creating or exacerbating psychological dysfunction without a measurable benefit in sinonasal outcome.

Corticosteroids are a widely used class of anti-inflammatory therapies that act by repressing proinflammatory mediators and cellular machinery for tissue remodeling.^19,20 In CRS, corticosteroids are thought to act through several mechanisms that decrease inflammation in the sinonasal mucosa, thereby improving mucociliary clearance of the sinuses.¹ Specifically, corticosteroids reduce the synthesis of several proinflammatory cytokines such as interleukin (IL) 1, IL-2, IL-3, IL-4, IL-6, tumor necrosis factor α, interferon γ, and granulocyte/macrophage colony-stimulating factor.²⁰ In addition, corticosteroids downregulate expression of adhesion molecules such as intercellular adhesion molecule 1, and vascular cell adhesion molecule 1, which decreases the recruitment of inflammatory cells.²¹

Oral corticosteroids are commonly used in multiple facets of CRS treatment, including initial medical therapy, treatment of acute exacerbations, and in the perioperative period for patients undergoing sinus surgery. In the literature studying CRS with nasal polyposis, there are multiple high-quality studies supporting the use of oral corticosteroids in medical treatment because they have been shown to improve sinonasal symptoms, reduce polyp size, and lower expression of inflammatory markers.^22-24 For CRS with nasal polyps, preoperative oral corticosteroids may reduce the technical difficulty of the surgery and improve Lund-Kennedy endoscopy scores postoperatively up to 6 months compared with placebo.²⁴

For patients with CRS without polyps, post-ESS medication regimens often involve a nonstandardized combination of oral corticosteroids, intranasal corticosteroids, and/or oral antibiotic, driven largely by personal preference, anecdotal data, and expert opinion. In most existing studies pertaining to post-ESS medical therapy for CRS without polyps, oral corticosteroids have been evaluated as 1 part of a multiagent sinonasal regimen, limiting the ability to discern the degree of improvement attributable to oral corticosteroids. For example, studies have shown oral corticosteroids to be effective in improving symptoms, endoscopic scores, and radiographic scores, but the potential benefit in these studies was confounded by concurrent treatment with saline rinses, oral antibiotics, and/or intranasal corticosteroids.^25-28

Although this study is strengthened by its double-blind placebo-controlled design, it is important to recognize limitations when interpreting the results. Quiz Ref IDEight of the 41 patients randomized to the prednisone treatment group did not complete the study, compared with only 1 of 40 patients in the placebo group, potentially introducing attrition bias. Although the study was blinded to participant, statistician, and physician, the research assistant recording study data was not blinded. In addition, there was variability in the sinonasal medication regimen over the 6-month study period following the uniform postoperative regimen given in the initial 2 weeks. This may diminish the ability to discern the effect of the initial corticosteroid regimen on longer-term sinonasal outcomes. This study was limited in its ability to further elucidate the relationship between oral corticosteroid use and longer-term psychiatric effects because the SNOT-22 psychologic subdomain is not a validated clinical tool for assessing such symptoms. Although there were similar rates of diagnosed depression between the 2 treatment groups, we did not screen for additional psychiatric comorbidities that may have predisposed 1 treatment arm toward worse psychological scores.

Nonetheless, the findings of this study suggest the need to reconsider the therapeutic indication for oral corticosteroids following ESS. For patients with CRS without polyps, more investigation is needed to determine how best to use corticosteroids in a judicious and clinically effective manner.

For patients with CRS without polyps undergoing ESS, postoperative oral prednisone does not improve sinonasal symptoms or endoscopic appearance through 6 months compared with placebo. Oral corticosteroids may, however, contribute to worse psychological symptoms. The potential benefit of oral corticosteroids appears limited and must be weighed against the risk of complications. Oral corticosteroids should be prescribed judiciously in this patient population.

Accepted for Publication: January 6, 2021.

Published Online: March 4, 2021. doi:10.1001/jamaoto.2021.0011

Corresponding Author: Peter H. Hwang, MD, Professor and Vice Chair, Stanford University School of Medicine, Department of Otolaryngology–Head & Neck Surgery, 801 Welch Rd, Stanford, CA 94305 ([email protected]).

Author Contributions: Drs Chang and Noel had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Co-first authors: Drs Chang and Noel.

Concept and design: Noel, Ayoub, Nayak, Patel, Hwang.

Acquisition, analysis, or interpretation of data: All authors.

Drafting of the manuscript: Chang, Noel, Qian, Hwang.

Critical revision of the manuscript for important intellectual content: Chang, Noel, Qian, Nayak, Patel, Hwang.

Statistical analysis: Chang, Qian, Hwang.

Administrative, technical, or material support: Chang, Noel, Ayoub, Dholakia, Nayak, Hwang.

Supervision: Noel, Nayak, Patel, Hwang.

Conflict of Interest Disclosures: All authors declare no conflicts of interests relevant to this study. Dr Hwang. is a consultant for Lyra Therapeutics, Third Wave Therapeutics, and Slate Therapeutics. Dr Nayak is a consultant for Medtronic, Olympus America, Cook Medical, Hydravascular, and Tissium. Dr Patel is a consultant for Medtronic, Stryker, and Intersect, and on the advisory board for Optinose.

Funding/Support: This study was funded by the Division of Rhinology within the Department of Otolaryngology–Head & Neck Surgery at Stanford University School of Medicine.

Role of the Funder/Sponsor: The Stanford University School of Medicine had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Data Sharing Statement: See Supplement 3.

Additional Contributions: We thank Aakanksha Rathor MD, Nour Ibrahim MD, Ximena Maul MD, Nicole Borchard BA, and Jane Wang NP for their contributions to this study. They were not compensated.