Abstract
Background:
The purpose of this retrospective review was to determine how patient related factors and culture data affect neo-osteogenesis in patients with chronic rhinosinusitis (CRS) and patients with cystic fibrosis (CF) CRS.
Methods:
Information from a database associated with a large tertiary medical center was used to assess adult patients with CF CRS and non-CF CRS (Total n=102; CF CRS, n=31 and non-CF CRS, n=71). Radiologic evidence of neo-osteogenesis was measured using the Global Osteitis Scoring Scale (GOSS) and mucosal disease was assessed by the Lund-Mackay score (LMS) by 2 independent reviewers who were blinded to the patient’s disease state. Bacterial cultures were obtained endoscopically. Multiple logistic regression models were used to evaluate the effect of age, sex, and number of previous surgeries, CF, and culture species on the odds of neo-osteogenesis.
Results:
Fifty-one of the 102 patients (50%) met radiologic criteria for neo-osteogenesis. Sixty-nine patients (67.6%) with CF CRS and non-CF CRS had culture data. In the multiple logistic regression model, male gender was significantly associated with neo-osteogenesis (odds ratio [OR], 5.2; 95% confidence interval [CI], 1.68–17.86; p=0.006). Pseudomonas aeruginosa was not associated with neo-osteogenesis (OR, 3.12; CI, 0.84–12.80, p=0.097). Age, number of surgeries, CF, Staphylococcus aureus, and coagulase-negative staphylococcus were not statistically significant.
Conclusion:
To our knowledge, this is the first study to assess risk factors associated with neo-osteogenesis and patients with CF CRS. Interestingly, male gender was the only significant predictor of neo-osteogenesis.
Keywords: Chronic rhinosinusitis, cystic fibrosis, neo-osteogenesis, quality of life, imaging
Introduction
The exact pathophysiologic mechanisms contributing to chronic rhinosinusitis (CRS) are unclear, but there is strong evidence to suggest a combination of host and environmental factors lead to the development and persistence of CRS.1,2 The main inflammatory site in CRS is the nasal mucosa, but underlying bony changes have also been observed clinically, radiologically, and histologically.3 Neo-osteogenesis is the thickening and increased density of the underlying bone observed radiologically in association with persistent mucosal inflammation.3,4
Currently, there is no clear etiology for neo-osteogenesis associated with CRS. Although it commonly occurs alongside bacterial infection, it may represent an inflammatory response to the bacteria rather than a direct effect of infection.4,5,6 A recent paper by Huang, et al investigated predictive factors for neo-osteogenesis in non-CF CRS.7 In this study, colonization of Pseudomonas aeruginosa was significantly associated with neo-osteogenesis, while Staphylococcus was not, despite both pathogens being implicated in non-CF CRS. This finding suggests that P. aeruginosa is a strong independent risk factor for development of neo-osteogenesis, and neo-osteogenesis may be a product of the bacterial environment and related inflammatory state of the sinonasal mucosa.
Regardless of etiology, observation of neo-osteogenesis has been associated with persistent mucosal inflammation in CRS, increased disease severity of CRS, and less improvement in quality of life measures. Multiple previous studies have investigated the association between neo-osteogenesis and non-CF CRS outcomes.8,9,10One study showed that prior sinus surgery could contribute to the development of neo-osteogenesis and greater resistance to surgical treatment may be associated with the degree of neo-osteogensis.8 Another study showed that patients with neo-osteogenesis had significantly worse baseline CT, endoscopy, and olfactory score than patients without osteitis.9 Finally, in another study neo-osteogenesis was associated with the need for a course of oral corticosteroids in the 12 months post endoscopic sinus surgery.
In cystic fibrosis patients, the defect in mucociliary clearance and other mucosal abnormalities mediated by the CFTR mutations lead to higher incidence of CRS than the general population.11 Additionally, recent studies have suggested that the sinuses serve as a reservoir for multiple CF pathogens of the lower airway, such as P. aeruginosa.12 The relationship between CRS and neo-osteogenesis in patients with cystic fibrosis (CF) has not previously been investigated. Although it is known that Pseudomonas is a predominant organism causing disease in the sinuses and lungs of CF patients, it is unclear if patients with CF have a greater incidence of sinus neo-osteogenesis. Thus, we sought to investigate the interaction of multiple patient factors to help increase our knowledge of how clinical features of CRS and CF correlate with the pathophysiology of the disease state.
Patients and Methods
This study is a retrospective analysis of patients referred to a tertiary medical center and was the collaborative effort of the Departments of Otolaryngology, Pulmonology and Critical Care Medicine, and Microbiology & Immunology. The study was approved by the University of Minnesota Institutional Review Board. Sinus CT scans of 151 patients were reviewed. The indication for their CT was either prior diagnosis of CRS with and without nasal polyposis (n=102), CF-CRS(n=31), nasal obstruction, or pituitary tumor. The patients with diagnosis of nasal obstruction or pituitary tumor were used as controls (n=18).
Of these 151 patients, 98 had previously collected bacterial culture samples that were reviewed. Review of their record for demographic information included age, sex, number of sinus surgeries and CF status. Finally, subjective disease severity was assessed utilizing the SNOT 22 scores when available. All patients were 18 years of age or older. Patients were excluded if they were diagnosed with sinus disease related to other specific inflammatory or immune disorders, such as granulomatous polyangitis, sarcoidosis, or Churg-Strauss syndrome.
Radiologic scoring
Two independent otolaryngologists graded the patient’s sinus CT scans (one staff physician and senior author, one 3rd year resident). Both reviewers were blinded to disease state. After gathering the scores, a mean score was used for each sample. Two grading systems were used to assess each CT: the Lund-Mackay score (LMS) system and Global Osteitis Scoring Scale (GOSS).
The LMS was used to measure the degree of mucosal disease observed on CT. The five sinuses were scored (maxillary, frontal, sphenoid, posterior ethmoid, anterior ethmoid). For each sinus a score of 0–2 was used: 0= no opacification; 1= partial opacification; and 2 = complete opacification. In addition to the five sinuses, the ostiomeatal complex was scored 0 or 2: 0 = no opacification or 2 = fully obstructed. Thus, for the five paranasal sinuses and one ostiomeatal complex on each side, total score ranged from 0–24.13 The inter-rater reliability for LMS between the two reviewers was 0.94 (95% CI 0.87–0.97) when 23 patients were reviewed.
Reviewing the same CT scans, neo-osteogenesis was evaluated using the GOSS.8 This scale is a composite grading system that assesses bone thickness, extent of involvement in each sinus, and the number of sinuses involved. Each sinus is graded from Grade 0 to Grade 5; 5 being the most severe (Figure 1). The scores of all 10 sinuses are added together for a range of 0–50. A total score >5 is positive for neo-osteogenesis. This was classified into mild (5–20), moderate (21–35), and severe (>35). The inter-rater reliability for GOSS between the two reviewers was 0.77 (95% CI 0.54–0.89) for the same 23 patients that were referenced with the LMS scoring.
Figure 1:
Global Osteitis Scoring Scale (GOSS): Composite grading system that assesses bone thickness, extent of involvement for each sinus, and number of sinuses involved
Bacterial Culture
The sinus samples were acquired during surgery by aspiration of middle meatal and/or sinus secretions under endoscopic visualization. CT scans reviewed for the purpose of this study were taken on the date of surgery or within several months of the culture collection. If the patient was not evaluated in a timely manner after their CT scan secondary to patient or scheduling factors, culture data could not be obtained.
Statistical Methods
Descriptive statistics were calculated for demographic variables. In order to determine the independent effect of various clinical factors on the presence of neo-osteogenesis, multiple logistic regression models were applied to the entire pooled cohort of CRS (with and without CF) to examine association between GOSS (presence of neo-osteogenesis) and the following: diagnosis of CF, number of previous surgeries, sex, and age. These covariates were selected on the basis of clinical interest, as well as those that had enough non-missing data to perform a reliable regression analysis. No model selection technique was used, as this was a descriptive analysis. This was repeated for the subset of patients with culture data comparing the association of the presence of neo-osteogenesis to all variables previously listed, as well as presence of P. aeruginosa, coagulase positive S. aureus and coagulase negative S. aureus. Finally, logistic regression models were used to analyze the CF-CRS patients and non-CF CRS patients separately. Analyses were conducted in R (V 3.4), and p-values less than 0.05 were considered statistically significant.
Results
Of the 151 patients in the study, 102 had non-CF CRS, 31 had CF CRS, and 18 were controls with no CRS. Patient demographic characteristics were similar across all three groups. The average age of patients in both CRS and control groups was 45 years old. Average age was slightly lower in the CF group at 32 years old. As expected, the CF-CRS and non-CF CRS groups both had a higher number of total surgeries than the control group. Approximately 50% of patients were male in each sample group (Table 1). The number of patients with neo-osteogenesis, defined as GOSS >5, was highest in the CF-CRS group (n=24, 77.4%), with fewer non-CF CRS patients having neo-osteogenesis (n=51, 50%), and nearly none in the control (n=1). On average, patients with CF-CRS had a higher mean LMS score than non-CF CRS. However, both CF-CRS and non-CF CRS with neo-osteogenesis had a higher average LMS score than patients without neo-osteogenesis (Table 2). After determining that these groups were appropriate for comparison we proceeded with the following analyses.
Table 1:
Summary of demographic data
| CF- CRS |
CRS | Control | |
|---|---|---|---|
| n | 31 | 102 | 18 |
| With neo-osteogenesis n (%) | 24 (77.4%) | 51 (50%) | 1 (5.6%) |
| Age, years mean (sd) | 31.8 (10.4) | 45.4 (15.3) | 44.8 (13.5) |
| Sex, M (%) | 16 (51.6%) | 48 (47.0%) | 9 (50%) |
| # of prev. surgeries, n (%) | |||
| 0 | 7 (22.6%) | 42 (41.2%) | 18 (100%) |
| 1 | 9 (29.0%) | 33 (32.4%) | 0 |
| 2 | 5 (16.1%) | 11 (10.8%) | 0 |
| 3 | 1 (3.2%) | 4 (3.9%) | 0 |
| 4+ | 9 (29.0%) | 12 (11.8%) | 0 |
| SNOT-22, n (non-missing) mean(sd) | 21 32.9 (27.9) |
76 36.3 (18.9) |
13 17.5 (15.7) |
| LMS, mean (sd) | 13.8 (4.9) | 10.5 (5.4) | 2.3 (2.3) |
| Cultures n (non-missing) | 26 | 71 | 1 |
| Total Cultures N positive (%) | 25 (96.2%) | 58 (81.7%) | 1 (100%) |
| Pseuodmonasaeruginosa N positive (%) | 16 (61.5%) | 26 (37.1%) | 0 (0%) |
| Staphylococcus aureus N positive (%) | 17 (65.4%) | 27 (39.1%) | 0 (0%) |
| Coagulase-negative Staph. N positive (%) | 7 (29.2%) | 19 (27.9%) | 0 (0%) |
Table 2:
Comparison of patients with and without neo-osteogenesis
| CF-CRS | CRS | Control | ||||
|---|---|---|---|---|---|---|
| Neo- osteogene sis |
Without N-O |
Neo- osteogene sis |
Without N-O |
Neo- osteogene sis |
Without N-O |
|
| n (%) | 24 | 7 | 51 | 51 | 1 | 17 |
| Age, years mean (sd) | 33.2 (10.9) | 27.1 (7.3) | 42.2 (15.4) | 48.5 (14.7) | 55 (NA) | 44.2 (13.7) |
| Sex, M (%) | 15 (62.5%) | 1 (14.3%) | 30 (58.8%) | 18 (35.3%) | 0 (0%) | 9 (53.0%) |
| # of prev. surgeries, n (%) | ||||||
| 0 | 4 (16.7%) | 3 (42.9%) | 12 (23.5%) | 30 (58.8%) | 1 (100%) | 17 (100%) |
| 1 | 7 (29.2%) | 2 (28.6%) | 19 (37.3%) | 14 (27.5%) | 0 | 0 |
| 2 | 3 (12.5%) | 2 (28.6%) | 5 (9.8%) | 6 (11.8%) | 0 | 0 |
| 3 | 1 (4.2%) | 0 | 4 (7.8%) | 0 | 0 | 0 |
| 4+ | 9 (37.5%) | 0 | 11 (21.6%) | 1 (2.0%) | 0 | 0 |
| SNOT-22, n (non-missing) mean(sd) | 15 31.1 (17.2) |
6 37.3 (20.5) |
37 31.6 (15.7) |
39 40.7 (20.7) |
1 0 (NA) |
12 18.9 (15.5) |
| LMS, mean (sd) | 14.5 (4.3) | 11.6 (4.8) | 13.0 (4.6) | 8.1 (5.2) | 4 (NA) | 2.2 (2.3) |
| Cultures n (non-missing) | 20 | 6 | 40 | 31 | 0 | 1 |
| Total Cultures N positive (%) | 19 (95%) | 6 (100%) | 34 (85.0%) | 24 (77.4%) | 0 | 1 (100%) |
| Pseuodmonasaeruginosa N positive (%) | 13 (65%) | 3 (50%) | 19 (47.5%) | 7 (23.3%) | 0 | 0 |
| Staphylococcus aureus N positive (%) | 13 (65%) | 4 (66.7%) | 17 (42.5%) | 10 (34.5%) | 0 | 0 |
| Coagulase-negative Staph. N positive (%) | 5 (27.8%) | 2 (33.3%) | 11 (28.9%) | 8 (26.7%) | 0 | 0 |
To isolate variables related to neo-osteogenesis in all CRS patients, CF and non-CF patient data was analyzed using multiple logistic regression models. The variables analyzed include diagnosis of CF, number of surgeries, gender, and age. Because the purpose of this was to report the effect of the covariates in the presence of one another, no univariate analysis was performed with single variables, only multivariate analysis was performed. After adjusting for other predictors, those with CF have 4.9 (95% CI: 1.5, 18.7) times the odds of neo-osteogenesis than those without CF (p=0.013). Additionally, those with at least one previous surgery have 4.05 (95% CI: 1.6, 10.6) times the odds of neo-osteogenesis than those without previous surgeries (p=0.003). Male sex was found to be a predictor of neo-osteogenesis in this logistic regression model (OR 3.0, 95% CI: 1.2, 7.7, p=0.019), but age was not a significant predictor of neo-osteogenesis in the context of this model (Table 3).
Table 3:
Summary of variables related to neo-osteogenesis in all CRS patients
| CRS (n=102) | ||
|---|---|---|
| Variable | OR (95% CI) | p |
| Age (10 years) | 1.03 (0.7, 1.5) | 0.866 |
| Surgery ≥ 1 | 4.1 (1.6, 10.6) | 0.003 |
| CF | 4.9 (1.5, 18.7) | 0.013 |
| Sex (M) | 3.0 (1.2, 7.7) | 0.019 |
We then performed analyses using the culture data that was obtained. Ninety-eight of the patients with neo-osteogenesis (with and without CF) also had available culture data. Multiple logistic regression models were repeated using only patient data that included culture results to determine what factors were independent predictors of neo-osteogenesis.
This analysis of patients with neo-osteogenesis (with and without CF) found male sex was the only statistically significant predictor of neo-osteogenesis (OR 5.1, p=0.006). Positive culture of P. aeruginosa was not a significant predictor of neo-osteogenesis (OR 3.121, P=0.097). Neither coagulase negative nor coagulase positive S. aureus positive cultures predicted neo-osteogenesis. Diagnosis of CF, age of the patient, and number of surgeries were also not significant predictors of neo-osteogenesis in this subset of the sample.
Finally, patients with positive culture data, neo-osteogenesis, and data available regarding number or previous surgeries (n=67) were separated into two logistic regression models based on CF diagnosis (non-CF CRS n=43, CF-CRS n=24) (Table 4). When CF-CRS and non-CF CRS groups were analyzed separately, male sex remained the only significant predictive factor for both groups (OR 5.2, p=0.006). Interestingly, when compared with the non-CF CRS group, the CF-CRS group had a higher OR for S. aureus positive cultures (CF-CRS OR=2.79, p=0.652 vs. CRS OR=0.74, p=0.791). Additionally, the CF-CRS group had a lower OR for P. aeruginosa (CF-CRS OR=2.05, p=0.644 vs. CRS OR=4.66, p=0.087). Larger sample sizes would be required to determine the significance of these findings.
Table 4:
Summary of patients with neo-osteogenesis (with and without CF) that have positive culture data
| CRS (n=67) | ||
|---|---|---|
| Variable | OR (95% CI) | p |
| Age (10 years) | 1.04 (0.7, 1.6) | 0.865 |
| Surgery ≥ 1 | 1.5 (0.4, 5.1) | 0.518 |
| CF | 2.1 (0.4, 11.9) | 0.380 |
| Sex (M) | 5.2 (1.7, 17.9) | 0.006 |
| Pseuodmonasaeruginosa | 3.1 (0.8, 12.7) | 0.097 |
| Staphylococcus aureus | 1.0 (0.2, 4.0) | 0.976 |
| Coag Neg. Staph | 1.1 (0.3, 4.6) | 0.866 |
Discussion
The exact etiology of neo-osteogenesis in CRS is currently unclear, but recent studies have suggested that these bony changes represent an inflammatory response, potentially related to bacteria that are associated with the sinonasal mucosa.7, 6 Additionally, patients with CRS and neo-osteogenesis have lower quality of life measures and increased disease severity.8, 9 A better understanding of the factors related to development of neo-osteogenesis in CRS may help optimize treatment and prevention.
The objective of this study was to determine how the presence of neo-osteogenesis relates to CF CRS, the presence of specific bacterial species (e.g. Pseudomonas), and other patient factors. To our knowledge, there are no previous studies that look at the rates of neo-osteogenesis in CF-CRS or culture data in CF-CRS and its relation to neo-osteogenesis.
We saw a higher rate of neo-osteogenesis in patients with CF-CRS compared to patients with non-CF CRS (77% vs. 50%). The colonization of CF patients’ paranasal sinuses by P. aeruginosa is common.14 Previous studies have shown that the presence of P. aeruginosa is a strong independent risk factor for neo-osteogenesis in non-CF CRS.7 In our study, P. aeruginosa was not a significant independent risk factor for all patients with CRS (OR 3.12, p value = 0.097). Interestingly, when the CF-CRS group and non-CF CRS groups were analyzed separately, the CF CRS had a lower OR for P. aeruginosa than the non-CF CRS group (2.05 vs. 4.66), but this was not significant. This odds ratio had wide confidence intervals, suggesting a larger sample size would be necessary to determine if these trends are significant. When compared with the non-CF CRS group, the CF-CRS group had a higher OR for S. aureus positive cultures (2.79 vs. 0.74), but this was not significant. Similar to the previous study by Huang, et al., the presence S. aureus did not appear to have implications on the presence of neo-osteogenesis. Therefore, specific species may play a different role in neo-osteogenesis in the two different sinonasal environments.
Additionally, the underlying pathophysiology of CF may create an environment more favorable to certain inflammatory mediators that promote neo-osteogenesis. Male gender was the only significant predictive risk factor for neo-osteogenesis in CF and non-CF patients with CRS in our study. When controlling for all other factors, the odds of having neo-osteogenesis was 5.1 higher in males in all patients with CRS (p=0.006). Additionally, when non-CF CRS and CF-CRS were analyzed separately, male gender was again the only significant independent predictor for neo-osteogenesis. There is no obvious etiology for male gender being the only significant predictor of neo-osteogenesis. Potential causes include hormonal influence on bone growth, a previously unknown predisposition for males to have more severe sinonasal disease, or simply males not seeking treatment in a reasonable amount of time when compared with females. A recent study of health care utilization found females are more likely to seek treatment for CRS than males.15 This finding is certainly an area that would need to be further studied to investigate the connection between male gender and neo-osteogenesis.
The results of this study were limited by the retrospective nature, sample size, and culture data. Although we have a large database of CF, CRS, and control patients, the database was incomplete in terms of culture data and comorbities that may impact infection propensity. Many patients did not have culture data due to various reasons, such as being lost to follow up or being unable to tolerate the procedure in the office. These patients were excluded. Additionally, an insufficient number of patients had completed the SNOT-22 questionnaire. Due to this limitation it was not used as a covariate in the regression model. We were able to review 151 CT scans, but only 98 patients had complete culture data for analysis. We were only able to capture culture data at a single point in time. We acknowledge that culture data may shift with different sinonasal infections, especially in CF patients, and this is a limitation to the study. A larger cohort would have strengthened the analysis, especially predictors that were approaching significance. However, the sample size was large enough for primary analysis of CRS (excluding the culture data) because we had a sample size of 51 with neo-osteogenesis and 51 without neo-osteogenesis and only four covariates in the model.
Although the level of inter-rater reliability for GOSS (0.77) was not as high as LMS scoring (0.95), it would be expected to be more variable as aspects of GOSS are more subjective. However, 0.77 was determined by our biostatistician to be adequate for statistical analysis. Despite these limitations, valuable information was obtained regarding neo-osteogenesis and factors that predict this bony remodeling in the setting of CF and non-CF CRS that prompt further investigation. Future directions include potentially gathering microbiome data and comparing the presence and degree of neo-osteogenesis with the microbiome species. This analysis would provide more specific and detailed information into whether the microbiological environments of CF CRS and non-CF CRS were significantly different. Furthermore, data that analyzes microbiome composition could be used to direct more targeted therapy.
Conclusion
Previous studies have established that neo-osteogenesis in CRS is associated with worse quality of life and worse surgical outcomes. This retrospective study was the first to investigate factors that may contribute to neo-osteogenesis in CF patients. Overall, when all CRS patients (CF and non-CF) were analyzed, ≥1 surgery, male sex, and presence of CF were all significant independent risk factors for neo-osteogenesis. Further investigation to elucidate factors related to male sex and bacterial culture data may improve understanding of neo-osteogenesis in patients with non-CF CRS and CF CRS.
Acknowledgements:
Research reported in this publication was supported by NIH grant P30 CA77598 utilizing the Biostatistics and Bioinformatics Core shared resource of the Masonic Cancer Center, University of Minnesota and by the National Center for Advancing Translational Sciences of the National Institutes of Health Award Number UL1TR000114. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.
Jordan Dunitz, MD, has financial relationships with Hill-Rom, Vertex Pharmaceuticals, and Proteostasis Therapeutics Inc. This study was not funded by any of these corporations.
Footnotes
Disclosures: All other authors have no financial disclosures.
References:
- 1.Fokkens WJ, Lund VJ, Mullol J, et al. European Position Paper on Rhinosinusitis and Nasal Polyps 2012. Rhinol Suppl. 2012;23. [PubMed] [Google Scholar]
- 2.Lee S, Lane AP. Chronic Rhinosinusitis as a Multifactorial Inflammatory Disorder. Curr Infect Dis Rep. 2011;13(2):159–168. doi: 10.1007/s11908-011-0166-z [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Chiu AG. Osteitis in Chronic Rhinosinusitis. Otolaryngol Clin North Am. 2005;38(6):1237–1242. doi: 10.1016/j.otc.2005.07.007 [DOI] [PubMed] [Google Scholar]
- 4.Bhandarkar ND, Sautter NB, Kennedy DW, Smith TL. Osteitis in chronic rhinosinusitis: a review of the literature. Int Forum Allergy Rhinol. 2013;3(5):355–363. doi: 10.1002/alr.21118 [DOI] [PubMed] [Google Scholar]
- 5.Tuszynska A, Krzeski A, Postuba M, et al. Inflammatory cytokines gene expression in bone tissue from patients with chronic rhinosinusitis- a preliminary study. Rhinology. 2010;48(4):415–419. doi: 10.4193/Rhino09.087 [DOI] [PubMed] [Google Scholar]
- 6.Snidvongs K, Sacks R, Harvey RJ. Osteitis in Chronic Rhinosinusitis. Curr Allergy Asthma Rep. 2019;19(5):24. doi: 10.1007/s11882-019-0855-5 [DOI] [PubMed] [Google Scholar]
- 7.Huang Z, Hajjij A, Li G, Nayak JV, Zhou B, Hwang PH. Clinical predictors of neo-osteogenesis in patients with chronic rhinosinusitis. Int Forum Allergy Rhinol. 2015;5(4):303–309. doi: 10.1002/alr.21485 [DOI] [PubMed] [Google Scholar]
- 8.Georgalas C, Videler W, Freling N, Fokkens W. Global Osteitis Scoring Scale and chronic rhinosinusitis: a marker of revision surgery. Clin Otolaryngol. 2010;35(6):455–461. doi: 10.1111/j.1749-4486.2010.02218.x [DOI] [PubMed] [Google Scholar]
- 9.Bhandarkar ND, Mace JC, Smith TL. The impact of osteitis on disease severity measures and quality of life outcomes in chronic rhinosinusitis. Int Forum Allergy Rhinol. 2011;1(5):372–378. doi: 10.1002/alr.20068 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Sacks P-L, Snidvongs K, Rom D, Earls P, Sacks R, Harvey RJ. The impact of neo-osteogenesis on disease control in chronic rhinosinusitis after primary surgery. Int Forum Allergy Rhinol. 2013;3(10):823–827. doi: 10.1002/alr.21192 [DOI] [PubMed] [Google Scholar]
- 11.Loebinger MR, Bilton D, Wilson R. Upper airway 2: Bronchiectasis, cystic fibrosis and sinusitis. Thorax. 2009;64(12):1096–1101. doi: 10.1136/thx.2008.112870 [DOI] [PubMed] [Google Scholar]
- 12.Aanæs K Bacterial sinusitis can be a focus for initial lung colonisation and chronic lung infection in patients with cystic fibrosis. J Cyst Fibros. 2013;12:S1–S20. doi: 10.1016/S1569-1993(13)00150-1 [DOI] [PubMed] [Google Scholar]
- 13.Lund VJ, Kennedy DW. Staging for rhinosinusitis. Otolaryngol Head Neck Surg. 1997;117(3 Pt 2):S35–40. doi: 10.1016/S0194-59989770005-6 [DOI] [PubMed] [Google Scholar]
- 14.Johansen HK, Aanaes K, Pressler T, et al. Colonisation and infection of the paranasal sinuses in cystic fibrosis patients is accompanied by a reduced PMN response. J Cyst Fibros. 2012;11(6):525–531. doi: 10.1016/j.jcf.2012.04.011 [DOI] [PubMed] [Google Scholar]
- 15.Samuelson MB, Chandra RK, Turner JH, Russell PT, Francis DO. The Relationship between Social Determinants of Health and Utilization of Tertiary Rhinology Care. Am J Rhinol Allergy. 2017;31(6):376–381. doi: 10.2500/ajra.2017.31.4476 [DOI] [PMC free article] [PubMed] [Google Scholar]