Abstract
Objective:
Relapses are frequent and difficult to predict in ANCA-associated vasculitis (AAV), resulting in long-term use of immunosuppression. Although sinonasal disease is associated with relapse of AAV, detailed characterization of sinonasal symptoms is lacking. Using a patient-reported outcome, the 22-item SinoNasal Outcome Test (SNOT-22), we investigated the relationship between sinonasal symptoms and disease activity in AAV.
Methods:
This was a prospective, longitudinal study of AAV and healthy individuals. Relapse was defined as BVAS/WG greater than zero. Higher SNOT-22 scores indicate worse symptoms. Generalized estimating equation and Cox proportional hazard models evaluated the association between SNOT-22 and relapse.
Results:
There were 773 visits (106 active disease visits) from 168 patients with AAV and 51 controls. Median SNOT-22 at remission was higher in AAV vs. controls (20 vs. 5, p<0.001) and higher during active disease vs. remission (p<0.001). In all AAV, and particularly within granulomatosis with polyangiitis, higher SNOT-22 scores were observed months to years before relapse and were associated with increased risk of relapse (HR 2.7 [95% CI 1.2–6.2], p=0.02). Similar findings were seen when examining patients with vs. without sinonasal disease and after removing relapses limited to the ear, nose, and throat.
Conclusion:
A patient-reported outcome measure of sinonasal disease, the SNOT-22, not only changes with disease activity in AAV, but also is associated with a higher risk of relapse within 2 years. These findings support the possibility that the SNOT-22 score may enhance prediction of relapse and that persistent sinonasal disease may be important in the pathophysiology of relapse.
Keywords: patient reported outcomes, ANCA-associated vasculitis, granulomatosis with polyangiitis (GPA), risk factors, disease activity
Introduction
ANCA-associated vasculitis (AAV) is a group of small vessel vasculitides with multisystem involvement and includes granulomatosis with polyangiitis (GPA), eosinophilic granulomatosis with polyangiitis (EGPA), and microscopic polyangiitis (MPA). Relapses in AAV are frequent and difficult to predict, resulting in long-term use of immunosuppression in most patients. In particular, sinonasal inflammation is common and adversely impacts quality of life in patients with GPA or EGPA1,2. Although sinonasal disease has been associated with relapse in some studies of AAV, more detailed characterization of rhinologic symptoms is lacking in AAV3. In a longitudinal study of patients with GPA, the patient global assessment during visits immediately prior to relapse was higher than at other remission visits, suggesting that patient-reported measures of symptom burden may provide insight into subsequent disease flare4. Whether patient-reported outcomes (PROs) related to sinonasal disease similarly predict relapse in AAV is not known.
The 22-item SinoNasal Outcome Test (SNOT-22) instrument may be a promising tool to assess rhinologic disease and its impact on patients with AAV. SNOT-22 is a validated PRO designed to assess symptom burden and quality of life in chronic rhinosinusitis and has also been used to predict response to medical or surgical treatment5,6. Studies in vasculitis found that SNOT-22 and SF-36 scores correlate and that sinus symptoms have a large impact on quality of life1,2,7. In a study of sinonasal disease in patients with autoimmune diseases, including AAV, those with autoimmune disease had more severe subjective and objective disease and scored higher on all SNOT-22 domains compared to those with chronic rhinosinusitis without autoimmune disease8. Among patients with GPA, those with sinus involvement reported lower quality of life compared to those without sinus involvement2. Among patients with EGPA, even when systemic vasculitic disease activity was low, SNOT-22 scores indicated high ongoing sinonasal symptom burden with resulting negative impact on quality of life1. It is not known how SNOT-22 scores change throughout the course of disease relapse and remission in AAV.
This study investigated the longitudinal changes in SNOT-22 scores associated with disease activity in patients with AAV. We hypothesized that SNOT-22 scores not only increase during relapse, but also are associated with a higher risk of future relapse in patients with AAV, particularly those with sinonasal involvement.
Materials and Methods
Study Population
Patients were recruited into a prospective, longitudinal cohort study through the Penn Vasculitis Center at the University of Pennsylvania from 2016 to 2022. Patients with GPA, EGPA, and MPA were eligible if they met 1990 American College of Rheumatology classification criteria (GPA, EGPA)9,10 or 2012 Chapel Hill Consensus Conference criteria (MPA)11. Patients with MPA were included as disease comparators since MPA typically does not involve the sinonasal cavity. Healthy controls were participants without systemic vasculitis, other inflammatory disorders, allergic rhinitis, or nasal polyps. This study was approved by the Institutional Review Board of the University of Pennsylvania and all patients gave written informed consent.
Patient-reported outcome measure of sinonasal disease
The 22-item SinoNasal Outcome Test (SNOT-22) questionnaire was completed by patients at every visit. The SNOT-22 instrument is a 22-question patient reported outcome measure validated in chronic rhinosinusitis to assess symptom burden and quality of life (Supplementary Table 1)5. Each question is scored on a scale from 0–5 with higher scores indicating more severe symptoms. The questionnaire can be used as a total score (range 0–110) or divided into five subdomains including rhinologic (range 0–30), extra-nasal rhinologic (range 0–15), ear/facial (0–25), psychological (0–30), and sleep symptoms (0–30)12. Some questions are included in more than one subdomain. Higher scores are indicative of higher symptom burden. The minimal clinically important difference (MCID) in patients with medically-managed chronic rhinosinusitis is 8 for SNOT-22 total, and MCID values for rhinologic, extra-nasal rhinologic, ear/facial, psychological, and sleep symptom subdomain scores are 3.9, 2.5, 3.3, 3.4, and 2.9, respectively13. SNOT-22 was copyrighted in 2006 by Washington University of St. Louis, Missouri and was used in this study with permission.
Outcomes
Disease activity was determined by the Birmingham Vasculitis Activity Score for Wegener’s Granulomatosis14 (BVAS/WG), a validated outcome measure used in a majority of clinical studies and trials of AAV, particularly GPA. BVAS/WG > 0 indicates active disease and BVAS/WG = 0 indicates disease remission. In addition, physician global assessment (PhGA) on a 0–10 scale was obtained at each visit. Relapses with only ear, nose, or throat (ENT) involvement (‘ENT-limited relapse’) were defined by presence of active ENT disease according to the BVAS/WG ENT category (bloody nasal discharge/nasal crusting/ulcer, sinus involvement, swollen salivary gland, subglottic inflammation, conductive deafness, or sensorineural deafness) and absence of any other organ manifestation. Physicians were blinded to SNOT-22 scores when assessing disease activity.
Co-variates
History of sinonasal involvement was defined as having a history of at least one of the following prior to enrollment: bloody nasal discharge, nasal crusting/ulcer, nasal septal perforation, nasal collapse, or sinus disease. Clinical data was collected prospectively at every visit. Data included demographic information and self-reported race, disease history and ANCA type, presence of seasonal allergies, use of saline sinus rinse, and medication use (antibiotics, systemic glucocorticoids, non-glucocorticoid immunosuppressive therapy such as cyclophosphamide, rituximab, azathioprine, methotrexate, and mycophenolate mofetil). Current medication use was defined as use within 2 months for immunosuppressive therapy or antibiotics, administration within 8 months for rituximab, and use within 6 months for saline sinus rinses or 2 weeks for steroid or antibiotic sinus rinses. Serial ANCA testing was left to the discretion of the physician since routine monitoring of ANCA levels is not recommended in current guidelines15. The Vasculitis Damage Index (VDI)16 was used to assess damage.
Statistical Analysis
Demographics and disease characteristics were summarized using descriptive statistics and groups were compared using two-sample t-test or Wilcoxon rank sum test as appropriate. For our primary analysis, the association between SNOT-22 total or subdomain score and disease activity (dichotomized as active disease or remission) was assessed using generalized estimating equations which accounts for repeated measures within individuals. Models were adjusted for disease duration, concomitant seasonal allergies, and use of antibiotics, immunosuppressive medications, and sinus rinses. Subgroup analyses were performed according to disease type (GPA, EGPA, or MPA) and for patients with or without sinus disease. Sensitivity analyses were performed including systemic glucocorticoids as a covariate and using PhGA and continuous rather than dichotomous BVAS/WG scores as the measure of disease activity.
To determine if longitudinal changes in SNOT-22 scores differed based on development of relapse (relapsing vs non-relapsing AAV), we performed analyses including a visit-by-group (relapsing vs non-relapsing AAV) interaction among patients with GPA or EGPA and sinonasal disease. A sensitivity analysis was performed among all patients with AAV. Cox proportional hazards models were used to determine the relationship between SNOT-22 score during the first remission visit and risk of relapse according to disease type as our primary focus. Models were adjusted for ANCA type, disease duration, and prior sinonasal disease. Receiver-operating curves for relapse according to remission SNOT-22 total or subdomain score were generated and cut-points for SNOT-22 total and subdomain scores were generated using Youden Index (maximum sum of sensitivity and specificity)17,18. Relapse-free survival was assessed using the Kaplan-Meier method19. Sensitivity analyses included defining remission based on BVAS/WG=0 and prednisone dose 5mg or less, using second remission visit when available, performing subgroup analyses for all AAV according to presence or absence of sinonasal disease, and subgroup analysis excluding patients with ENT-limited relapse. In an exploratory analysis, we applied LASSO regression for variable selection among the 22 questions within the SNOT-22 questionnaire and tuning parameter (λ) was determined by 10-fold cross-validation. Analyses were conducted with Stata 18 (College Station, TX: StataCorp LLC) and R Statistical Software (v4.2.2; R Core Team 2022).
Results
Study patients
There were 773 study visits among 168 patients with AAV: 107 with GPA, 40 with EGPA, and 21 with MPA. An additional 51 healthy controls were included. The median time between visits was 3.9 months (interquartile range [IQR] 3.2, 6.0). Patients with AAV completed a median of 4 (IQR 2, 7) study visits and duration of enrollment was a median of 24 months (IQR 4, 27). There was a history of sinonasal involvement from AAV in 85%, 95%, and 0% of patients with GPA, EGPA, and MPA, respectively. Additional demographics, disease characteristics, and treatments at time of enrollment are shown in Table 1.
Table 1.
Demographics and disease characteristics at enrollment
| ANCA-associated vasculitis | Healthy controls | ||||
|---|---|---|---|---|---|
| All | GPA | EGPA | MPA | ||
| n=168 | n=107 | n=40 | n=21 | n=51 | |
| Demographics | |||||
| Age in years, mean (SD) | 55 (15.8) | 53 (16.3) | 56 (13.6) | 59 (16.5) | 59 (14.4) |
| Female | 98 (58%) | 60 (56%) | 27 (68%) | 11 (52%) | 28 (55%) |
| White race | 159 (95%) | 99 (93%) | 39 (98%) | 21 (100%) | 45 (88%) |
| Ever smoker | 62 (37%) | 43 (40%) | 13 (33%) | 6 (29%) | 23 (45%) |
| Number of visits | 4 [2, 7] | 5 [3, 7] | 4 [1, 7] | 3 [1, 6] | 1 [1, 2] |
| Disease Characteristics | |||||
| Duration of disease (years) | 3.0 [1.0, 6.0] | 3.5 [0.7, 6.3] | 2.6 [1.0, 5.0] | 3.3 [1.1, 6.8] | -- |
| Positive ANCA by ELISA, ever | |||||
| Anti-proteinase-3 | 77 (46%) | 73 (68%) | 1 (3%) | 3 (14%) | -- |
| Anti-myeloperoxidase | 57 (34%) | 28 (26%) | 11 (30%) | 18 (86%) | -- |
| Both | 1 (1%) | 0 | 1 (3%) | 0 | -- |
| None | 29 (17%) | 5 (5%) | 24 (65%) | 0 | -- |
| History of flare | 91 (54%) | 56 (52%) | 25 (63%) | 10 (48%) | -- |
| Flare involving sinonasal area | 55 (60%) | 39 (70%) | 16 (64%) | 0 | -- |
| History of rhinosinusitis from AAV | 129 (77%) | 91 (85%) | 38 (95%) | 0 | -- |
| Bloody nasal discharge | 78 (46%) | 63 (59%) | 14 (35%) | 0 | -- |
| Nasal crusting/ulcer | 79 (47%) | 63 (59%) | 16 (40%) | 0 | -- |
| Nasal septal perforation | 18 (11%) | 16 (15%) | 2 (5%) | 0 | -- |
| Nasal collapse | 17 (10%) | 17 (16%) | 0 | 0 | -- |
| Sinus involvement | 116 (69%) | 78 (73%) | 38 (95%) | 0 | -- |
| Vasculitis Damage Index score | 1 [1, 3] | 1 [0, 2] | 2 [1, 3] | 1 [0, 2] | |
| Current seasonal allergy symptoms | 39 (23%) | 18 (17%) | 17 (43%) | 4 (19%) | -- |
| Medications | |||||
| Oral glucocorticoid within 2 months | 87 (52%) | 51 (48%) | 26 (65%) | 10 (48%) | 1 (2%) |
| Rituximab within 8 months | 52 (31%) | 38 (36%) | 6 (15%) | 8 (38%) | 0 |
| Non-rituximab immunosuppressive medication within 2 months | 78 (46%) | 45 (42%) | 28 (70%) | 5 (24%) | 0 |
| Azathioprine | 34 (20%) | 21 (20%) | 10 (25%) | 3 (14%) | 0 |
| Methotrexate | 22 (13%) | 18 (17%) | 4 (10%) | 0 | 0 |
| Mycophenolate mofetil | 12 (7%) | 4 (4%) | 7 (18%) | 1 (5%) | 0 |
| Intravenous cyclophosphamide | 5 (3%) | 2 (2%) | 2 (5%) | 1 (5%) | 0 |
| Mepolizumab | 7 (4%) | 0 | 7 (18%) | 0 | 0 |
| Antibiotic use within 2 months | 1 (1%) | 1 (1%) | 0 (0%) | 0 | 0 |
| Nasal therapies | |||||
| Nasal glucocorticoid within 2 weeks | 33 (20%) | 14 (13%) | 19 (48%) | 0 | 2 (4%) |
| Any nasal antibiotic within 2 weeks | 11 (7%) | 9 (8%) | 2 (5%) | 0 | 0 |
| Nasal saline rinses within 6 months | 79 (47%) | 52 (50%) | 24 (60%) | 3 (14%) | 6 (12%) |
All values expressed as median [IQR] or n (%) unless otherwise specified.
Abbreviations: AAV, ANCA-associated vasculitis; ANCA, anti-neutrophil cytoplasmic antibody; EGPA, eosinophilic granulomatosis with polyangiitis; ELISA, enzyme-linked immunosorbent assay; GPA, granulomatosis with polyangiitis; IQR, interquartile range; MPA, microscopic polyangiitis; SD, standard deviation
Association of AAV type and SNOT-22 scores
At the first remission visit, SNOT-22 total score was higher for patients with AAV compared to healthy controls: median [IQR]: 20 [11, 38] vs. 5 [2, 16], p<0.001. SNOT-22 total scores at the first remission visit were higher in patients with vs. without history of sinonasal involvement prior to enrollment for all AAV (median [IQR]: 24 [12, 45] vs. 14 [9, 23], p=0.003) (Figure 1). When evaluating the individual SNOT-22 subdomains (rhinologic, extra-nasal rhinologic, ear/facial, psychological, and sleep), scores were elevated in patients with AAV vs. controls for each subdomain (Supplementary Figure 1). Compared to patients with GPA or EGPA, patients with MPA had lower scores in the subdomains specific to sinonasal disease (e.g., rhinologic, extra-nasal) but had similar scores in the psychological and sleep domains. Therefore, we found differences in SNOT-22 scores across clinical subgroups which were consistent with expectations.
Figure 1. SNOT-22 total scores by diagnosis and sinonasal involvement at first remission visit.
Bar graphs show median scores with 95% confidence intervals along with individual scores represented by overlying dot plot. Higher scores indicate greater disease burden. SNOT-22 total scores during remission were higher for patients with ANCA-associated vasculitis compared to healthy controls (p < 0.01). Patients with granulomatosis with polyangiitis (GPA) with sinonasal involvement had higher SNOT-22 total scores compared to patients with GPA without sinonasal involvement, though this was not statistically significant (p=0.06). EGPA, eosinophilic granulomatosis with polyangiitis; MPA, microscopic polyangiitis
Association of disease activity with SNOT-22 scores
Among the 773 visits in patients with AAV, there were a total of 106 active disease visits with a median BVAS/WG score of 3 (IQR 2, 5) and median PhGA score of 5 (IQR 3, 6). Among all active visits, 65% included active ENT disease (GPA 71%, EGPA 73%, MPA 0%) and the median SNOT-22 total score was 30 (IQR 17, 44)(Table 2).
Table 2.
Characteristics of active disease visits
| All AAV | GPA | EGPA | MPA | |
|---|---|---|---|---|
| n=168 | n=107 | n=40 | n=21 | |
| Number of study visits | n=773 | n=516 | n=177 | n=80 |
| Visits with active disease, n (% of total visits) | 106 (14%) | 75 (15%) | 22 (12%) | 9 (11%) |
| Visits with active ENT disease, n (% of active visits) | 69 (65%) | 53 (71%) | 16 (73%) | 0 |
| Visits with only ENT disease, n (% of active visits) | 20 (19%) | 16 (21%) | 4 (18%) | 0 |
| Time between visits (months) | 3.9 [3.2, 6.0] | 4.1 [3.2, 6.0] | 3.7 [3.2, 5.8] | 3.7 [3.2, 6.0] |
| Disease measures for visits with active disease | ||||
| SNOT-22 total, range 0–110 | 30 [17, 44] | 30 [16, 43] | 26 [15, 50] | 34 [25, 39] |
| Rhinologic, range 0–30 | 8 [4, 13] | 8 [4, 13] | 8 [4, 13] | 5 [4, 10] |
| Extra-nasal, range 0–15 | 4 [1, 7] | 3 [1, 7] | 4 [2, 9] | 4 [2, 6] |
| Ear/facial, range 0–25 | 4 [1, 7] | 4 [2, 8] | 3 [1, 6] | 3 [1, 5] |
| Psychologic, range 0–30 | 10 [4, 15] | 9 [3, 17] | 8 [1, 12] | 13 [12, 16] |
| Sleep, range 0–30 | 9 [3, 15] | 8 [3, 15] | 10 [3, 14] | 13 [9, 15] |
| BVAS/WG | 3 [2, 5] | 3 [2, 5] | 2 [2, 3] | 4 [3, 6] |
| General | 37 (35%) | 30 (40%) | 4 (18%) | 3 (33%) |
| Cutaneous | 8 (8%) | 8 (11%) | 0 | 0 |
| Mucous membranes/eyes | 6 (6%) | 5 (7%) | 0 | 1 (11%) |
| Ear, nose, throat | 70 (66%) | 54 (72%) | 16 (72%) | 0 |
| Cardiovascular | 0 | 0 | 0 | 0 |
| Gastrointestinal | 1 (1%) | 0 | 0 | 1 (11%) |
| Pulmonary | 23 (22%) | 21 (28%) | 1 (5%) | 1 (11%) |
| Renal | 8 (8%) | 4 (5%) | 0 | 4 (44%) |
| Nervous system | 15 (14%) | 5 (7%) | 6 (27%) | 4 (44%) |
| Other | 36 (34%) | 25 (33%) | 10 (45%) | 1 (11%) |
| Physician Global Assessment | 5 [3, 6] | 5 [3, 6] | 3 [3, 6] | 6 [4, 7] |
Values are expressed as median [IQR] or n (%) unless otherwise stated. Patients may contribute more than one active disease visit. Abbreviations: AAV, ANCA-associated vasculitis; BVAS/WG, Birmingham Vasculitis Activity Score/Wegener’s granulomatosis; EGPA, eosinophilic granulomatosis with polyangiitis; ENT, ear, nose, throat; GPA, granulomatosis with polyangiitis; IQR, interquartile range; MPA, microscopic polyangiitis
Compared to patients with AAV during times of remission, active disease was associated with higher SNOT-22 total scores for patients with AAV (β=6.5, 95% CI [3.6–9.4], p<0.001) and for each disease type (GPA: β=6.6, 95% CI [2.8–10.4], p<0.001; EGPA: β=7.2, 95% CI [2.8–11.5], p<0.001; MPA: β=9.9, 95% CI [1.6–18.3], p=0.02), after adjusting for potential confounders including nasal antibiotics/rinses, oral/intravenous antibiotics, immunosuppressive medications, and disease duration. Results were consistent when analyses were restricted to patients with sinonasal disease. Only the total score and psychological and sleep domains were significant when restricting to patients without sinonasal disease. In a sensitivity analysis, results were unchanged when current prednisone use was also included in the models. When evaluating disease activity using either PhGA or BVAS/WG as a continuous measure, greater levels of disease activity were associated with higher SNOT-22 total scores for all AAV (PhGA: β=1.4, 95% CI [0.8–2.1], p<0.001; BVAS/WG: β=1.6, 95% CI [0.7–2.5], p<0.001). Therefore, SNOT-22 is associated with active disease in AAV.
When SNOT-22 scores were divided into subdomains, active disease was associated with higher scores in all 5 subdomains for patients with AAV (p<0.01). In subgroup analyses according to AAV type, the SNOT-22 subdomains that were associated with active disease differed by AAV type. In GPA and EGPA, rhinologic and similar subdomains were significantly elevated during active disease vs. remission (p<0.05). However, in MPA only sleep and psychological subdomains were elevated during active disease vs. remission (p<0.05). Therefore, disease activity affects SNOT-22 subdomains differently among the three types of AAV.
Remission SNOT-22 score and risk of relapse within two years in ANCA-associated vasculitis
We investigated whether SNOT-22 scores at initial remission visit are associated with risk of relapse in AAV. Four patients (3 GPA, 1 EGPA) were excluded from this analysis because a remission visit was not available. Using Youden Index, the following cut-points were determined: SNOT-22 total 41, rhinologic 5, extra-nasal rhinologic 3, ear/facial 3, psychological 9, sleep 10. A higher SNOT-22 total score during remission was associated with increased risk of relapse for GPA (HR [95% CI]: 2.7 [1.2–6.2], p=0.02) but not for EGPA (HR [95% CI]: 1.5 [0.4–5.3], p=0.5) or MPA (Table 3, Figure 2, Supplementary Figure 2). Findings were similar after adjusting for ANCA type, disease duration, and prior sinonasal disease (Table 3).
Table 3.
Risk of relapse according to SNOT-22 scores during remission
| Granulomatosis with polyangiitis | Eosinophilic granulomatosis with polyangiitis | Microscopic polyangiitis | |||||||
|---|---|---|---|---|---|---|---|---|---|
| n = 103 | n=39 | n = 19 | |||||||
| Unadjusted model | HR | 95% CI | p-value | HR | 95% CI | p-value | HR | 95% CI | p-value |
| Total (cut-point 41) | 2.7 | 1.2–6.2 | 0.02 | 1.5 | 0.4–5.3 | 0.5 | 1 | - | - |
| Rhinologic (cut-point 5) | 3.6 | 1.3–9.7 | 0.01 | 1.1 | 0.3–4.3 | 0.9 | 7.4 | 0.5–122.3 | 0.2 |
| Extra-nasal (cut-point 3) | 3.7 | 1.3–10.8 | 0.02 | 2.0 | 0.4–9.4 | 0.4 | 3.8 | 0.2–62.4 | 0.3 |
| Ear/facial (cut-point 3) | 2.5 | 1.0–6.0 | 0.04 | 4.1 | 0.9–19.5 | 0.1 | 3.9 | 0.2–62.4 | 0.3 |
| Psychological (cut-point 9) | 3.3 | 1.4–7.8 | 0.01 | 1.7 | 0.5–6.0 | 0.4 | 0.9 | 0.1–13.5 | 0.9 |
| Sleep (cut-point 10) | 2.2 | 1.0–5.1 | 0.06 | 1.0 | 0.3–3.6 | 0.9 | 0 | 0 | 1.0 |
| Adjusted model | n = 72 | n = 30 | |||||||
| Total (cut-point 41) | 3.2 | 1.1–9.8 | 0.04 | 2.8 | 0.2–15.8 | 0.2 | insufficient observations | ||
| Rhinologic (cut-point 5) | 3.7 | 1.0–13.7 | 0.05 | 0.9 | 0.2–4.8 | 0.9 | |||
| Extra-nasal (cut-point 3) | 6.6 | 1.3–31.9 | 0.02 | 4.4 | 0.5–41.3 | 0.2 | |||
| Ear/facial (cut-point 3) | 2.7 | 0.9–8.5 | 0.09 | 10.5 | 1.2–95.1 | 0.04 | |||
| Psychological (cut-point 9) | 3.9 | 1.3–11.6 | 0.01 | 2.2 | 0.4–10.7 | 0.3 | |||
| Sleep (cut-point 10) | 2.3 | 0.8–6.3 | 0.11 | 1.8 | 0.4–9.2 | 0.5 | |||
All models adjusted for: ANCA type, disease duration, and history of sinonasal disease prior to enrollment; CI, confidence interval; HR, hazard ratio; SNOT-22, SinoNasal Outcome Test-22
Figure 2. SNOT-22 scores during remission and risk of relapse in GPA and EGPA.
Kaplan-Meier curves showing proportion free of relapse over time according to first remission SNOT-22 total and subdomain scores for patients with GPA or EGPA. Among patients with GPA, higher SNOT-22 total (A) and rhinologic subdomain scores (C) during remission were associated with increased risk of relapse. In contrast, among patients with EGPA, higher SNOT-22 total (B) and rhinologic subdomain scores (D) during remission were not associated with relapse risk. Cox models were adjusted for ANCA type, history of sinonasal disease, and disease duration. See Supplementary Figure 5 for remaining subdomains. CI, confidence interval; GPA, granulomatosis with polyangiitis; EGPA, eosinophilic granulomatosis with polyangiitis; HR, hazard ratio.
Additional sensitivity analyses were performed to test the robustness and aid in the interpretation of the primary analyses. Defining first remission visit based on lack of disease activity and prednisone dose of 5mg or less resulted in re-classification of the first remission visit for 27 patients but did not change results. Sensitivity analyses using the second remission visit when available did not change results. Results were similar for AAV with prior sinonasal disease (Supplementary Table 2). We also re-calculated cut-points among the patients with sinonasal disease which led to an increase in the rhinologic cut-point from 5 to 9, although this did not substantially change results (HR 2.2, 95% CI 1.0–4.8, p=0.049). Among patients with AAV without prior sinonasal disease, the rhinologic domain was significantly associated with future relapse (p = 0.01), while no associations were found among the other SNOT-22 subdomains (Supplementary Table 2, Supplementary Figure 3). Subgroup analyses excluded 10 patients with GPA whose relapses were all ‘ENT-limited relapse’ according to the ENT category on the BVAS/WG and continued to show that rhinologic and psychological subdomain scores were associated with future relapse (p= 0.04 and p=0.02, respectively).
As an exploratory analysis, lasso regression was used to identify a subset of SNOT-22 items that may be able to improve predictive performance. A new subset of questions was created consisting of questions 1, 3, 5, 7, 8, 10, 11, 15, 16, 19, 21, and 22 from the SNOT-22 instrument with cut-point of 18. This subset of 12 questions was significantly associated with a higher risk of relapse in GPA (HR 3.71 [95% CI 1.60–8.63], p=0.002) (Supplementary Figure 4). In summary, we found that SNOT-22 scores, particularly the rhinologic subdomain score, are associated with a 2–3-fold higher risk of relapse of AAV within 2 years even among patients without a prior history of sinonasal disease and even after excluding ENT-limited relapses.
Longitudinal SNOT-22 scores in relapsing vs. non-relapsing ANCA-associated vasculitis
We investigated whether longitudinal changes in SNOT-22 score would differ among patients who develop relapse (relapsing) versus those who remain in remission (non-relapsing). For this analysis, we used the first remission visit as the starting point and focused on the patients with AAV with history of sinonasal disease. In GPA, we found that even in the visits preceding onset of relapse, when patients were in clinical remission and months-to-years prior to relapse, the SNOT-22 total scores were significantly elevated in patients who developed a relapse (relapsing GPA) vs. those who remained in sustained remission (non-relapsing GPA) (Figure 3). Similarly, higher subdomain scores were observed several visits prior to and during relapse of GPA compared to the non-relapsing GPA group. For EGPA with sinonasal involvement, the differences in SNOT-22 scores between relapsing and non-relapsing groups were significant for rhinologic, extra-nasal, and ear/facial subdomains only at the time of relapse but not in visits prior to relapse, although sample size was smaller (Figure 2, Supplementary Figure 5). Subgroup analyses of GPA without prior sinonasal disease did not show significantly different SNOT-22 total or subdomain scores between relapsing and non-relapsing groups, including prior to relapse. Results were similar to the primary analysis even after excluding 10 patients with ‘ENT-limited relapses’ or evaluating all AAV regardless of sinonasal involvement (Supplementary Figure 6). Therefore, SNOT-22 scores remain persistently elevated for months-to-years prior to onset of relapse in patients with AAV and prior history of sinonasal disease.
Figure 3. Longitudinal SNOT-22 scores by visit in relapsing vs non-relapsing patients with GPA or EGPA and sinonasal disease.
Line plots with 95% CI of predicted values of SNOT-22 scores from regression models across consecutive visits. Analysis includes patients with sinonasal disease and at least one pre-relapse remission visit in the relapsing group. Relapse visits are aligned at visit 4 and indicated by a green square. Mean (SD) time between visits was 5.5 (3.4) months for GPA and 5.1 (3.6) months for EGPA. (A-B) SNOT-22 total scores, (C-D) rhinologic subdomain scores. In GPA, compared to non-relapsing group (dashed line), the relapsing group (solid green line) had higher SNOT-22 scores not just during relapse but also during the visits preceding relapse, months-to-years before onset of relapse. In contrast, among patients with EGPA, SNOT-22 scores were higher at the time of relapse but not during the preceding visits. Models were adjusted for immunosuppressive medications, antibiotics, nasal antibiotics or rinses, and disease duration. Between-group comparison: *p<0.05, **p<0.01. See Supplementary Figure 2 for remaining subdomains. GPA, granulomatosis with polyangiitis; EGPA, eosinophilic granulomatosis with polyangiitis
Discussion
This prospective cohort study of AAV examined the association between longitudinal changes in sinonasal symptoms with relapse of AAV using a patient-reported outcome measure, the SNOT-22. SNOT-22 scores discriminate active disease from remission among patients with AAV. Interestingly, among patients with GPA, SNOT-22 scores are elevated months-to-years prior to onset of systemic relapse but remain low in patients in sustained remission. Similarly, higher SNOT-22 scores during remission are associated with a 2–3-fold increased risk of future relapse in GPA, even among patients without history of sinonasal disease. Together, these findings support the possibility that a patient questionnaire could be used to predict relapse of GPA, which might inform therapeutic decision-making, and raises new hypotheses about the potential role of the upper airway in the pathophysiology of relapse of AAV.
Our results showed that SNOT-22 discriminates active disease from remission among patients with prior ENT involvement, which is consistent with a prior study. In a cross-sectional study of 44 patients with AAV (GPA and MPA), Cazzador et al compared SNOT-22 scores during active disease and found significantly higher scores for the group with sinonasal involvement. They also found higher scores in each subdomain except for sleep among the patients with sinonasal disease.20 In our study, patients with MPA, who generally do not have AAV-related sinonasal disease, also had higher SNOT-22 scores associated with active disease but these were primarily driven by higher scores in the sleep and psychological subdomains, suggesting that in MPA these domains are impacted by factors outside the sinonasal area. Overall, the differences in SNOT-22 found between the clinical subgroups support the content and construct validity of the instrument.
Interestingly, our study shows that higher SNOT-22 scores during remission in patients with AAV are associated with subsequent disease relapse. This relationship was most pronounced within the GPA subgroup, although analyses of other subgroups were limited by smaller sample sizes. Unexpectedly, even among patients without prior history of sinonasal disease, we found that higher SNOT-22 scores were associated with a higher risk of AAV relapse. Two potential explanations for why SNOT-22 predicts relapse even in those without known sinonasal disease include: (1) SNOT-22 is capturing other domains important to predicting relapse, or (2) the SNOT-22 instrument may be more sensitive than clinician assessment in detecting lower levels of sinonasal activity. The number of patients without sinonasal disease was small (n=37) so more studies are needed to validate these findings. These findings are consistent with previous work demonstrating that patient global assessment increases prior to GPA relapse, whereas physician-based measures do not change4.
Investigators have long searched for a reliable prognostic marker that identifies patients at high vs. low risk of relapse in AAV but, except for ANCA type, no prognostic biomarker is routinely used to inform treatment decision-making. As a result, many patients are maintained on long-term immunosuppression, sometimes indefinitely, to avoid relapse. Studies examining predictors of relapse in AAV reported a higher relapse risk among patients with PR3-ANCA type (vs MPO-ANCA) as well as prior ENT, GI, or lung involvement3,21. Our study builds upon these findings by identifying a high-risk subgroup among patients already with higher risk features (e.g., GPA, mostly PR3-ANCA positive, with a history of sinonasal disease). Further work is needed to determine the generalizability of these findings.
In addition to prognostic implications, this study raises new hypotheses about the potential role of the upper airway epithelium in the pathogenesis of relapse in AAV. We postulate that sinonasal symptoms, measured by the SNOT-22, reflect an aberrant nasal mucosa which may predispose to pro-inflammatory immune responses. A continuous and comprehensive measure such as the SNOT-22 may improve our ability to capture the spectrum of sinonasal disease in AAV, including those with low level or subclinical involvement. This may explain why the SNOT-22 score continued to predict relapse even among patients without clinically apparent sinonasal disease in the past. It is unclear whether higher SNOT-22 scores prior to relapse of AAV reflect active inflammation versus damage but these findings are consistent with studies demonstrating a dysfunctional nasal mucosal barrier in AAV. Chronic sinus congestion and nasal crusting are commonly reported despite resolution of other organ manifestations in AAV22. Studies of the nasal epithelium showed impaired ciliary function23 and higher cytokine levels24 associated with AAV, particularly in GPA. Studies of nasal epithelial gene expression identified signatures related to a disturbed epithelial barrier25,26 and multiple groups have found evidence of microbial dysbiosis in the sinonasal cavity of patients with GPA27–30. Seccia et al showed that patients with EGPA had evidence of ongoing sinonasal inflammation on nasal cytology despite systemic remission7. Future mechanistic studies are needed to unravel the relationships between sinonasal disease and relapse in AAV.
The strengths of this study include the use of a relatively large prospective cohort of patients with AAV, with and without prior sinonasal involvement, and healthy individuals. SNOT-22 scores and validated disease activity scores (BVAS/WG) were measured longitudinally, allowing for examination of not only the relapse visit, but also the visits preceding relapse. Limitations of this study include the difficulty in distinguishing between disease-related damage and active disease as well as non-disease-related symptoms (e.g., allergies, infection). Analyses were adjusted for current allergy symptoms and antibiotic and immunosuppressive use but adjustment for confounding may have been incomplete. Patients with chronically active sinonasal complaints might be more likely to be identified as having active disease, so there is potential for misclassification of disease status. However, in sensitivity analyses evaluating those without sinonasal disease (as determined by the clinician) and after exclusion of relapses limited to the sinonasal area, results were unchanged, suggesting that potential misclassification was not substantial enough to alter our primary findings. Future work that includes prospective evaluation of SNOT-22 scores in conjunction with nasal endoscopy to assess for active inflammation would be informative. An additional limitation is the small sample size of patients with EGPA and MPA. Unlike for GPA, among patients with EGPA, SNOT-22 scores were not elevated in the relapsing group during visits leading up to the relapse and high scores at remission were not associated with increased risk of disease relapse. It is unknown if these findings are a result of small sample size or if they represent true differences in SNOT-22 performance between disease groups.
There are several areas for future investigation of the SNOT-22 instrument in AAV. Our study focused on the total and pre-defined subdomain SNOT-22 scores which have been used in chronic rhinosinusitis. Future studies are needed to validate the cut-points identified in this study and determine if there is an optimal subset of questions which improves the performance of this PRO in AAV. Responsiveness of the instrument to immunosuppressive treatment is not well-understood and could inform both clinical management as well as serve as an outcome measure in clinical trials and studies. Future studies are needed to determine if incorporating the SNOT-22 questionnaire enhances clinical management and design of clinical trials (e.g., recruitment or stratification), and whether SNOT-22 scores provide added predictive value compared to currently recognized risk factors for relapse such as pulmonary involvement, patient global assessment, and ANCA level. Since serial ANCA testing is not recommended in current guidelines,15 availability of serial ANCA titers was limited in our cohort. We did not evaluate SNOT-22 scores or risk of relapse in relation to ANCA level, but this may be an area for future investigation. Lastly, some studies have modified the SNOT-22 instrument and included questions about nasal crusting, bleeding, and deformity, which are more specific for the AAV population2,33. There may be merit to investigating a modified, AAV-specific version of the instrument.
This study found that a patient-reported outcome measure of sinonasal disease, the SNOT-22, not only changes with disease activity in AAV, but also is associated with a higher risk of relapse in GPA, months to years prior to relapse. These findings support the possibility that: 1) the SNOT-22 score may be able to enhance our ability to predict relapse in a subgroup of high-risk patients (i.e., GPA with sinonasal disease), and 2) persistent sinonasal disease activity may play an important role in the pathophysiology of relapse. Future studies are needed to determine if the SNOT-22 score can be used for selective escalation or de-escalation of immunosuppressive therapy or assessment of treatment response.
Supplementary Material
Significance and Innovations.
Greater patient-reported sinonasal symptoms, measured by the SinoNasal Outcome Test (SNOT-22), are not only associated with disease activity but also associated with a higher risk of relapse within 2 years in ANCA-associated vasculitis (AAV).
Patient-reported outcomes may enhance monitoring of sinonasal disease and prediction of relapse in AAV.
These findings also suggest that persistent sinonasal disease may be important in the pathophysiology of relapse.
Grant or other funding support:
This study was supported by NIH-NIAMS (K23-AR071514, PI: Rhee) and NIH Rheumatology Research Training Grant T32-AR076951-04 (Romich).
Footnotes
Disclosures: see disclosures supplement
References
- 1.Latorre M, Baldini C, Seccia V, et al. Asthma Control and Airway Inflammation in Patients with Eosinophilic Granulomatosis with Polyangiitis. J Allergy Clin Immunol Pract. 2016;4(3):512–519. doi: 10.1016/j.jaip.2015.12.014 [DOI] [PubMed] [Google Scholar]
- 2.Srouji IA, Andrews P, Edwards C, Lund VJ. General and Rhinosinusitis-Related Quality of Life in Patients With Wegener???s Granulomatosis. Laryngoscope. 2006;116(9):1621–1625. doi: 10.1097/01.mlg.0000230440.83375.4b [DOI] [PubMed] [Google Scholar]
- 3.Safari S, Alesaeidi S, Pakzad B, Abbaspour S. Predictors of relapse in granulomatosis with polyangiitis: a multi-center study. Egypt Rheumatol Rehabil. 2022;49(1):59. doi: 10.1186/s43166-022-00160-y [DOI] [Google Scholar]
- 4.Tomasson G, Davis JC, Hoffman GS, et al. Brief Report: The Value of a Patient Global Assessment of Disease Activity in Granulomatosis With Polyangiitis (Wegener’s). Arthritis Rheumatol. 2014;66(2):428–432. doi: 10.1002/art.38248 [DOI] [PubMed] [Google Scholar]
- 5.Hopkins C, Gillett S, Slack R, Lund VJ, Browne JP. Psychometric validity of the 22-item Sinonasal Outcome Test. Clin Otolaryngol. 2009;34(5):447–454. doi: 10.1111/j.1749-4486.2009.01995.x [DOI] [PubMed] [Google Scholar]
- 6.Liu M, Liu J, Weitzel EK, Chen PG. The predictive utility of the 22-item sino-nasal outcome test (SNOT-22): A scoping review. Int Forum Allergy Rhinol. 2022;12(1):83–102. doi: 10.1002/ALR.22888 [DOI] [PubMed] [Google Scholar]
- 7.Seccia V, Baldini C, Latorre M, et al. Focus on the involvement of the nose and paranasal sinuses in Eosinophilic Granulomatosis with polyangiitis (Churg-Strauss Syndrome): Nasal cytology reveals infiltration of eosinophils as a very common feature. Int Arch Allergy Immunol. 2018;175(1–2):61–69. doi: 10.1159/000484602 [DOI] [PubMed] [Google Scholar]
- 8.Makary CA, Gill B, Parman B, et al. Subjective and Objective Measurements of Sinonasal Manifestations in Patients With Autoimmune Disorders. Laryngoscope. 2021;131(2):255–259. doi: 10.1002/lary.28693 [DOI] [PubMed] [Google Scholar]
- 9.Leavitt RY, Fauci AS, Bloch DA, et al. The American College of Rheumatology 1990 criteria for the classification of Wegener’s granulomatosis. Arthritis Rheum. 1990;33(8):1101–1107. doi: 10.1002/ART.1780330807 [DOI] [PubMed] [Google Scholar]
- 10.Masi AT, Hunder GG, Lie JT, et al. The American College of Rheumatology 1990 criteria for the classification of churg-strauss syndrome (allergic granulomatosis and angiitis). Arthritis Rheum. 1990;33(8):1094–1100. doi: 10.1002/ART.1780330806 [DOI] [PubMed] [Google Scholar]
- 11.Jennette JC, Falk RJ, Bacon PA, et al. 2012 revised International Chapel Hill Consensus Conference Nomenclature of Vasculitides. Arthritis Rheum. 2013;65(1):1–11. doi: 10.1002/ART.37715 [DOI] [PubMed] [Google Scholar]
- 12.Deconde AS, Mace JC, Bodner T, et al. SNOT-22 quality of life domains differentially predict treatment modality selection in chronic rhinosinusitis. Int Forum Allergy Rhinol. 2014;4(12):972–979. doi: 10.1002/alr.21408 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Chowdhury NI, Mace JC, Bodner TE, et al. Does Medical Therapy Improve SinoNasal Outcomes Test–22 Domain Scores? An Analysis of Clinically Important Differences. Laryngoscope. 2019;129(1):31–36. doi: 10.1002/lary.27470 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Stone JH, Hoffman GS, Merkel PA, et al. A disease-specific activity index for Wegener’s granulomatosis: Modification of the Birmingham Vasculitis Activity Score. Arthritis Rheum. 2001;44(4):912–920. doi: [DOI] [PubMed] [Google Scholar]
- 15.Chung SA, Langford CA, Maz M, et al. 2021 American College of Rheumatology/Vasculitis Foundation Guideline for the Management of Antineutrophil Cytoplasmic Antibody–Associated Vasculitis. Arthritis Care Res (Hoboken). 2021;73(8):1088–1105. doi: 10.1002/acr.24634 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Exley AR, Bacon PA, Luqmani RA, et al. Development and initial validation of the Vasculitis Damage Index for the standardized clinical assessment of damage in the systemic vasculitides. Arthritis Rheum. 1997;40(2):371–380. doi: 10.1002/ART.1780400222 [DOI] [PubMed] [Google Scholar]
- 17.Youden WJ. Index for rating diagnostic tests. Cancer. 1950;3(1):32–35. doi: [DOI] [PubMed] [Google Scholar]
- 18.Ewald B. Post hoc choice of cut points introduced bias to diagnostic research. J Clin Epidemiol. 2006;59(8):798–801. doi: 10.1016/J.JCLINEPI.2005.11.025 [DOI] [PubMed] [Google Scholar]
- 19.Kaplan EL, Meier P. Nonparametric Estimation from Incomplete Observations. J Am Stat Assoc. 1958;53(282):457–481. doi: 10.2307/2281868 [DOI] [Google Scholar]
- 20.Cazzador D, Padoan R, Colangeli R, et al. Health-Related Quality of Life in Patients With ANCA-Associated Vasculitis and Sinonasal Involvement: A Single-Center Cross-Sectional Study. J Clin Rheumatol. 2022;28(1):e89–e94. doi: 10.1097/RHU.0000000000001630 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Hogan SL, Falk RJ, Chin H, et al. Predictors of relapse and treatment resistance in antineutrophil cytoplasmic antibody-associated small-vessel vasculitis. Ann Intern Med. 2005;143(9). doi: 10.7326/0003-4819-143-9-200511010-00005 [DOI] [PubMed] [Google Scholar]
- 22.Hoffman GS, Kerr GS, Leavitt RY, et al. Wegener granulomatosis: an analysis of 158 patients. Ann Intern Med. 1992;116(6):488–498. doi: 10.7326/0003-4819-116-6-488 [DOI] [PubMed] [Google Scholar]
- 23.Ullrich S, Gustke H, Lamprecht P, et al. Severe impaired respiratory ciliary function in Wegener granulomatosis. Ann Rheum Dis. 2009;68(6):1067–1071. doi: 10.1136/ARD.2008.096974 [DOI] [PubMed] [Google Scholar]
- 24.Wohlers J, Breucker K, Podschun R, et al. Aberrant cytokine pattern of the nasal mucosa in granulomatosis with polyangiitis. Arthritis Res Ther. 2012;14(5). doi: 10.1186/AR4041 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Laudien M, Häsler R, Wohlers J, et al. Molecular signatures of a disturbed nasal barrier function in the primary tissue of Wegener’s granulomatosis. Mucosal Immunol. 2011;4(5):564–573. doi: 10.1038/MI.2011.9 [DOI] [PubMed] [Google Scholar]
- 26.Grayson PC, Steiling K, Platt M, et al. Defining the nasal transcriptome in granulomatosis with polyangiitis (Wegener’s). Arthritis Rheumatol (Hoboken, NJ). 2015;67(8):2233–2239. doi: 10.1002/ART.39185 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Rhee RL, Sreih AG, Najem CE, et al. CHARACTERIZATION OF THE NASAL MICROBIOTA IN GRANULOMATOSIS WITH POLYANGIITIS. Ann Rheum Dis. 2018;77(10):1448. doi: 10.1136/ANNRHEUMDIS-2018-213645 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Rhee RL, Lu J, Bittinger K, et al. Dynamic Changes in the Nasal Microbiome Associated With Disease Activity in Patients With Granulomatosis With Polyangiitis. Arthritis Rheumatol (Hoboken, NJ). 2021;73(9):1703–1712. doi: 10.1002/ART.41723 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Wagner J, Harrison EM, Martinez Del Pero M, et al. The composition and functional protein subsystems of the human nasal microbiome in granulomatosis with polyangiitis: a pilot study. Microbiome. 2019;7(1). doi: 10.1186/S40168-019-0753-Z [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Stegeman CA, Cohen Tervaert JW, Sluiter WJ, Manson WL, De Jong PE, Kallenberg CGM. Association of chronic nasal carriage of Staphylococcus aureus and higher relapse rates in Wegener granulomatosis. Ann Intern Med. 1994;120(1):12–17. doi: 10.7326/0003-4819-120-1-199401010-00003 [DOI] [PubMed] [Google Scholar]
- 31.Tomasson G, Grayson PC, Mahr AD, LaValley M, Merkel PA. Value of ANCA measurements during remission to predict a relapse of ANCA-associated vasculitis--a meta-analysis. Rheumatology (Oxford). 2012;51(1):100–109. doi: 10.1093/RHEUMATOLOGY/KER280 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Finkielman JD, Merkel PA, Schroeder D, et al. Antiproteinase 3 antineutrophil cytoplasmic antibodies and disease activity in Wegener granulomatosis. Ann Intern Med. 2007;147(9):611–619. doi: 10.7326/0003-4819-147-9-200711060-00005 [DOI] [PubMed] [Google Scholar]
- 33.Srouji I, Lund V, Andrews P, Edwards C. Rhinologic Symptoms and Quality-of-Life in Patients with Churg-Strauss Syndrome Vasculitis. Am J Rhinol. 2008;22(4):406–409. doi: 10.2500/ajr.2008.22.3204 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.