Characteristics associated with rapid chronic rhinosinusitis with nasal polyp recurrence following endoscopic sinus surgery in NSAID-exacerbated respiratory disease

Abstract

Background

Patients with nonsteroidal anti-inflammatory drug–exacerbated respiratory disease (NSAID-ERD) or aspirin-exacerbated respiratory disease are prone to rapid recurrence of chronic rhinosinusitis with nasal polyps (CRSwNP) after endoscopic sinus surgery (ESS), often resulting in persistent symptoms and reduced quality of life.

Objective

We sought to determine patient-reported timing and associated symptoms of CRSwNP recurrence after ESS, as well as to assess whether current disease burden differs by prior recurrence timing.

Methods

We analyzed data from 325 participants in our NSAID-ERD registry who reported a history of prior ESS. At registry enrollment, patients were asked to recall the time to recurrence of CRSwNP following their most recent ESS (≥2 years before enrollment). At enrollment, participants also completed the 22-Item Sino-Nasal Outcome Test (SNOT-22) and Asthma Control Test (ACT) questionnaires, reflecting current sinonasal and asthma disease burden. In a separate follow-up survey, a larger cohort of participants were asked to report the symptoms that they experienced with CRSwNP recurrence.

Results

The median patient-reported time to post-ESS CRSwNP recurrence was 6 months, with 51.1% reporting recurrence in 6 months or less, 18.2% reporting recurrence in 6 to 12 months, 15.4% reporting recurrence in 12 to 24 months, and 15.4% reporting no recurrence at 24 months. The most common symptoms reported with CRSwNP recurrence were nasal congestion (91.3%) and hyposmia/anosmia (86.6%). At registry enrollment, those with fast recurrence (≤6 months) had higher total SNOT-22 scores (52.4 ± 23.6 vs 39.7 ± 20.3 [P < .0001]) and lower ACT scores (18.2 ± 5.7 vs 20 ± 4.9 [P < .01]) than those with slow or no recurrence (>6 months). Scores in all SNOT-22 domains were higher in patients who reported fast recurrence.

Conclusions

Post-ESS CRSwNP recurrence in patients with NSAID-ERD was most often signaled by worsening nasal congestion and loss of smell. Patients with fast CRSwNP recurrence later exhibited greater overall respiratory disease burden.

Introduction

Aspirin-exacerbated respiratory disease (AERD), also known as nonsteroidal anti-inflammatory drug–exacerbated respiratory disease (NSAID-ERD), is characterized by chronic rhinosinusitis with nasal polyps (CRSwNP), eosinophilic asthma, and hypersensitivity to COX-1 inhibitors. Endoscopic sinus surgery (ESS) is often used to treat CRSwNP in patients with NSAID-ERD; however, post-ESS CRSwNP recurrence is common in this group unless they receive substantial adjuvant therapy, such as biologic therapy or aspirin therapy after desensitization (ATAD).^1,2 NSAID-ERD and CRSwNP are associated with increased health care utilization and impairments in quality of life (QoL), including depression and lost productivity.3, 4, 5 Data on the patient-reported post-ESS rate of CRSwNP recurrence, symptom recognition of recurrence, and QoL following ESS in NSAID-ERD remain limited. We sought to examine the patient-reported rates of CRSwNP recurrence and associated clinical features to characterize patient perceptions and provide insights that will inform management strategies.

Data were collected from the Brigham and Women’s Hospital (BWH) AERD registry, which includes more than 2500 patients with physician-diagnosed NSAID-ERD who either consented to enrollment when they were first seen as patients at our NSAID-ERD center, or who self-enrolled into the registry online without being followed clinically at our center. The primary data set used to examine patient-reported CRSwNP recurrence rates comprised existing participant information from the NSAID-ERD registry. It was narrowed from the entire cohort to include only participants who enrolled between 2013 and 2024, reported an ESS at least 2 years before enrollment, responded to the question “How long did it take for your nasal polyps to regrow after surgery?”, had no prior use of biologics, and had sufficient demographic and clinical data available to us. When available, the clinical data included the following: patient-reported outcomes at time of enrollment (≥2 years after ESS), as measured by the Asthma Control Test (ACT) and 22-Item Sino-Nasal Outcome Test (SNOT-22) as validated measures of disease severity, use of oral corticosteroids in the past 12 months, data on lung function (as measured by FEV₁ percent predicted), body mass index, and absolute eosinophil count. All clinical data used for this data set were collected at the participant’s enrollment visit, which occurred after the historical ESS for which timing and symptoms of recurrence were determined. A total of 325 registry participants met these criteria, and their data were included in the primary analysis.

As the main NSAID-ERD registry enrollment visit questionnaires did not provide detailed insight into the symptoms that patients first noticed with CRSwNP recurrence or the methods by which their recurrence was recognized or confirmed, we developed an additional survey to further elucidate these patient-reported findings. It was emailed to all participants in the BWH AERD registry in 2024 (for the full survey questions, see the Supplementary Text in the Online Repository at www.jaci-global.org). Participants were able to select as many indicators and symptoms as applicable to their experience and were also asked to identify which symptom had been their initial indicator of CRSwNP recurrence.

Results and discussion

The median patient-reported time to post-ESS CRSwNP recurrence for the 325 registry participants was 6 months. On this basis, we divided participants into 1 of 2 dichotomized groups: (1) “fast recurrence” (defined as reported CRSwNP recurrence ≤6 months after ESS [n = 166 (51.1%)]) or (2) “slow or no recurrence” (defined as reported CRSwNP recurrence >6 months after ESS [n = 159 (48.9%)]).⁶ To compare findings between the 2 groups, clinical data were analyzed using the Fisher exact test or an unpaired t test, as appropriate. ACT and SNOT-22 scores were analyzed with the Mann-Whitney U test or unpaired t test, as appropriate.

Of the 159 participants in the “slow recurrence” group, 59 (18.2% of the 325) reported CRSwNP recurrence between 6 and 12 months after ESS, 50 (15.4%) reported recurrence between 12 and 24 months, and 50 (15.4%) reported that they still had no CRSwNP recurrence after 24 months. Notably, 4 of the participants with no recurrence at 24 months were noted to have had aspirin desensitization followed by ATAD after ESS. For 4 patients, utilization of ATAD was unknown; of these 4 patients, 3 were in the fast recurrence group. No other patients reported ATAD after ESS.

There were no significant differences between the fast recurrence and slow recurrence groups in terms of sex, current body mass index, baseline FEV₁ percent predicted, oral corticosteroid use in the past year, or current peripheral blood eosinophil count at registry enrollment (Table I); nor was there any significant difference in the mean reported number of lifetime ESSs (2.5 for those with fast recurrence vs 2.68 for those with slow recurrence). However, participants in the slow recurrence group were older (mean age 53.5 years vs 49.6 years [P < .01]) and had a longer average lifetime CRSwNP duration in years (17.7 vs 12.9 [P < .0001]) than did the participants in the fast recurrence group. Additionally, participants in the fast recurrence group had significantly higher total SNOT-22 scores at their registry enrollment visit than did those in the slow or no recurrence recurrence group (52.4 ± 23.6 vs 39.7 ± 20.3 [P < .0001]). Further, the scores in all of the individual SNOT-22 domains (nasal [P < .01], ear and/or facial [P < .001]), sleep [P < .001], and emotional [P < .0001]) were significantly higher in the fast recurrence group, suggesting that more severe sinonasal disease burden later developed in patients with early recurrence. Participants in the fast recurrence group also had lower ACT scores than did participants in the slow recurrence group (18.2 ± 5.7 vs 20 ± 4.9 [P < .01]), indicating development of worse asthma control after their recurrence. SNOT-22 scores correlated with ACT scores, suggesting a link between sinus disease severity and asthma control (r = –0.55; P < .0001).

Table I.

Baseline characteristics and symptom scores

Characteristic	Rate of post-ESS recurrence		P value
	≤6 mo	>6 mo
Female sex, no. (%)	94 (57)	91 (57)	NS∗
Age (y), mean ± SD	49.6 ± 13.6	53.5 ± 12.6	<.01†
FEV1 % predicted, mean ± SD	84.3 ± 19.7	81.9 ± 16.1	NS†
Lifetime prior ESSs (no.), mean ± SD	2.5 ± 1.7	2.7 ± 2.0	NS†
CRSwNP duration (y), mean ± SD	12.9 ± 9.4	17.7 ± 11.3	<.0001†
BMI (kg/m2), mean ± SD	28 ± 5.5	27 ± 4.7	NS†
Absolute eosinophil count (cells/μL), mean ± SD	600 ± 480	520 ± 280	NS†
SNOT-22 score (0-110), mean ± SD	52.34 ± 23.6	39.7 ± 20.3	<.0001†
Nasal domain (0-40)	23.8 ± 8.5	19.9 ± 9.2	<.01†
Ear/facial domain (0-20)	6.8 ± 5.1	4.6 ± 4.0	<.001†
Sleep domain (0-20)	18.9 ± 10.7	13.6 ± 9.7	<.001†
Emotional domain (0-30)	5.0 ± 4.1	2.7± 3.2	<.0001†
ACT score (5-25), mean ± SD	18.2 ± 5.7	20.0 ± 4.9	<.01†

BMI, Body mass index; NS, not significant.

^∗Fisher exact test.

^†Unpaired t test.

A total of 528 participants with a history of CRSwNP recurrence responded to the additional survey, including 61 who had also been a part of the primary registry analysis of 325 patients. The most commonly reported indicator of CRSwNP recurrence was an increase in sinonasal symptom burden (76.1% of respondents), followed by CRSwNP noted on physical examination (70.1%), and computed tomography or magnetic resonance imaging evidence of recurrence (33.9%). Specific symptoms of recurrence included nasal congestion (91.3%), hyposmia/anosmia (86.8%), increased mucous production (78.2%), facial pain or pressure (52.8%), and insomnia (30.9%) (Fig 1). In all, 21% of patients reported experiencing all 5 symptoms, 57% reported experiencing 3 or 4 symptoms, and 22% reported experiencing 2 or fewer symptoms with CRSwNP recurrence. The most common first symptom of CRSwNP recurrence was nasal congestion or blockage (44.7%), followed by reduced sense of smell (30.5%) and increased mucus production (16.9%) (Fig 2).

Fig 1.

Follow-up survey responses indicating initial symptom associated with CRSwNP recurrence.

Fig 2.

Follow-up survey responses indicating symptoms associated with CRSwNP recurrence. Participants selected as many responses as were applicable.

Our study highlights the significant impact of severe CRSwNP in patients with NSAID-ERD, with rapid recurrence following ESS leading to impairment in QoL, as evidenced by higher SNOT-22 and lower ACT scores. There is no significant difference in age of onset of CRSwNP or asthma between the groups; however, patients in the fast recurrence group were younger and had a shorter lifetime duration of CRSwNP at registry enrollment. This may be reflective of less severe disease burden in older patients or normalization of symptoms over time.⁷ The significantly elevated baseline SNOT-22 scores in patients with fast recurrence reflect more severe symptoms across established domains,⁸ including sleep, nasal, ear and facial, and emotional. The lower ACT scores also reflect a greater spectrum of overall disease severity and correlation between upper and lower airway disease burden. Although the elevated SNOT-22 scores likely reflect the post-ESS recurrence symptoms, the elevation in the extranasal SNOT-22 domains and ACT scores support overall impairments in QoL for patients with rapid CRSwNP recurrence. Future studies assessing pre-ESS baseline patient-reported outcome measures as predictors of postsurgical treatment outcomes are needed to determine which patient-reported outcome measures are most predictive of early recurrence in patients with NSAID-ERD.

The limitations to our study include reliance on patient-reported outcomes and patient-perceived rates of CRSwNP recurrence, which may introduce recall bias and skewing of severity of symptoms. Additionally, we do not have information about other treatments that patients may have used in the perioperative setting (intranasal steroid rinses or sprays) or objective measurements of CRSwNP recurrence (nasal endoscopy, imaging). Given that the BWH AERD Registry is open to patients who can self-enroll from all over the world, there is no standardized regimen of post-ESS therapy that can be evaluated, as many of our registry participants had their care outside of our institution. Furthermore, the data were collected at time of enrollment into the registry (which may have limited evaluation of future symptom progression), and for a few patients, the data may have been affected by adjuvant treatments such as ATAD after ESS. Lastly, because we are a tertiary care referral center with specialization in caring for patients with NSAID-ERD, our cohort of patients may have more severe disease than the general population of those with NSAID-ERD, as a result of which our findings may not be generalizable to all patients with NSAID-ERD. Despite these limitations, understanding patient perceptions of CRSwNP recurrence, in addition to impacts on QoL, is of foundational importance in understanding how to treat this complex disease.

Patients with NSAID-ERD often have aggressive sinonasal disease and frequently require revision ESS.⁹ Our study highlights the importance of assessing symptom burden and QoL to identify patients at risk of recurrence; many of these patients may benefit from treatment with a biologic or ATAD. Recurrence of CRSwNP is often first identified as increased sinonasal symptoms, primarily nasal congestion and hyposmia or anosmia. These findings are important for clinicians and patients to recognize, as they offer potential early indicators of recurrent disease that can prompt earlier intervention for adjuvant therapies such as higher intranasal corticosteroid doses, ATAD, or biologics. Our study highlights the importance of patient education regarding early signs of disease recurrence and overall disease burden associated with NSAID-ERD.

Disclosure statement

Supported by the National Institutes of Health (grants U19AI095219, K23AI139352, K24AI180296, T32AI007306, and R21AI182809) and by generous contributions from the Vinik and Kaye families.

Disclosure of potential conflict of interest: K. M. Buchheit has served on scientific advisory boards for AstraZeneca, Regeneron, Sanofi-Genzyme, Eli Lilly, and GlaxoSmithKline and has received research funding from Regeneron. T. M. Laidlaw has served on scientific advisory boards for GlaxoSmithKline, AstraZeneca, Sanofi-Genzyme, Regeneron, and Eli Lilly and has received research funding from Sanofi. The rest of the authors declare that they have no relevant conflicts of interest.