Introduction: Chronic rhinitis (CR) and rhinosinusitis are prevalent conditions affecting people all over the world. Their exact relationship is still not fully understood. We sought to find out, whether CR is a risk factor for chronic rhinosinusitis (CRS) and which main subgroup or other factors could be predisposing. Methods: Patients with diagnosed CR between 2005 and 2010 were selected from the electronic medical record and were contacted by phone call. They were interviewed and screened for possible CRS using internationally approved questionnaires, e.g. NOSE-D and SNOT-20-GAV. Those with elevated scores were invited for a clinical examination. Results: Of 113 patients available for statistical analysis (48/65 = f/m), mean age of 52 ± 15 years, 13 patients were diagnosed with CRS. Extrapolated for the total cohort of 334, calculated prevalence was 9.5%. No statistical significantly higher probability of developing CRS for either main subgroup of CR was found. Age of onset, prior surgery of the nose, and use of topical nasal treatments were associated with the development of CRS in multivariate analyses (OR = 0.1, 3.2, and 3.2, respectively). Discussion/Conclusions: Only a small number of rhinitis patients developed CRS, questioning the paradigm of CR being a clear risk factor for CRS.

© 2022 The Author(s). Published by S. Karger AG, Basel

Introduction

Chronic rhinitis (CR) is affecting people all over the world with an estimated prevalence of around 30% of the total population including all age groups [1, 2]. It is accompanied by significant morbidity, a considerable financial burden, and leads to a reduced quality of life with decreased productivity, sleep impairment, and psychological morbidity [3-8]. There have been different approaches to categorize CR into subtypes, which are hindered by the high variability in both underlying pathophysiologic mechanisms (endotypes) and clinical presentations (phenotypes) [3, 9-11]. Further research in this field has shown that phenotypes/endotypes are dynamic, overlapping, and may evolve into one another. Hence, it is difficult to develop clear guidelines for its diagnosis and treatment [11]. Sinusitis, on the other hand, is known to be an inflammatory disease of the paranasal cavities. Since sinusitis and rhinitis coexist and a pathophysiological distinction between the nasal cavity and the sinuses is difficult, the term “rhinosinusitis” was adopted internationally.

Chronic rhinosinusitis (CRS) is reported in up to 10.9% of the western population [12] based on epidemiology (symptoms only) and 3–6% when combined with endoscopy and CT scan [13], and like rhinitis, it is also associated with major physical, emotional, and economic impact [14]. A lot of research has been done on endotypes to further understand the pathophysiology of CRS, but the pathogenesis is still not fully understood [15]. CR and CRS are associated in adults [9] and children [8], and there have been reports that CR is a risk factor for developing a rhinosinusitis [11]. It is commonly believed that a continuous transition between the two pathologies (from CR to CRS) takes place.

To date, most research on the topic was done in a retrospective fashion by investigating what common premorbidities patients with CRS had, and most prior relevant studies were limited by small sample size, short observation periods, or failure of standardized diagnosis of CRS [16, 17]. We chose a different set up by following patients with CR over time to assess whether they would develop CRS at a later timepoint. Besides identifying how many patients with CR would develop CRS, it was our goal to find subtypes/subgroups of CR or specific factors that could favor the transition into CRS. To the best of our knowledge, this is the first study to collect data in a linear fashion with a follow-up period of 10 years or more and standardized diagnosis of CRS in a cohort of over 100 patients. If the common belief of CR being a risk factor for CRS could be refuted, this would change patient care and education. It might reduce unnecessary anxieties and follow-up consultations.

Materials and MethodsStudy Design

This is a descriptive, retrospective cohort study of patients referring to the ENT Department of the University Hospital Zurich, Zurich, Switzerland, between 2005 and 2010 with diagnosed CR. After a follow-up period of 10 years or more, patients were contacted by phone call to inquire about their symptoms at present. In those patients where symptoms persisted, a clinical follow-up consultation was scheduled by the study team.

Subjects, Data Collection, and Inclusion Criteria

Cases enrolled at baseline were selected based on the following inclusion criteria: diagnosed CR (sneezing, nasal itching, rhinorrhea, or nasal congestion [18, 19] for ≥12 weeks), no known rhinosinusitis at onset. The exclusion of rhinosinusitis was done using nasal endoscopy and imaging where necessary. We collected data from the electronic medical record. This investigation was approved by the Swiss Ethics Committees on research involving humans (BASEC-Nr.: 2019-01426). Informed consent was obtained orally upon telephone interviews or written from those study participants who were seen for clinical examination. An overview of the exact enrollment process is shown in Figure 1.

Fig. 1.

Detailed overview of the enrolment process. It shows a detailed flowchart of the enrollment process, starting at the top left black box. It demonstrates how numbers for each group arose. In blue, it shows the initial cohort before our phone call. The red boxes show excluded patients due to incomplete data sets and in green those with definitive CRS diagnosis. The dashed dark red boxes with cursive text contain the extrapolated, expected missed numbers.

Our cohort consisted of female and male patients between the age of 18 and 99 years from different origins who had lived in Switzerland and had admitted to our outpatient clinic. Initially, a total of 627 patients were identified because the term “rhinitis” was found in their documented patients’ history. We excluded those patients who were diagnosed with CRS at onset and those patients with acute rhinitis and rhinosinusitis. A total of 334 patients met the inclusion criteria, six of whom had again been admitted to our clinic by 2019/2020 and had been diagnosed with CRS. The remaining 328 patients were contacted via telephone since it would have exceeded the scope of this study to invite every single patient for clinical examination. Therefore, patients were assessed during the phone call using the NOSE-D and the SNOT-20-GAV scores.

For all reached patients, we obtained a specific case history regarding nasal symptoms and whether patients had been exposed to gas, dust, or fumes or were diagnosed with asthma or aspirin-exacerbated respiratory disease (AERD/NERD). Furthermore, based on the knowledge of the major etiologic factors as proposed by Hellings et al. [2], patients were divided into four main subgroups: infectious, allergic, nonallergic, or mixed. The main subgroup of nonallergic rhinitis (NAR) was again subclassified into 6 minor subgroups: senile, gustatory, occupational, hormonal-induced, drug-induced, and idiopathic. Diagnosis of allergic rhinitis for the majority of patients was made based on history anamnestic, clinical features, on skin prick testing, and/or IgE testing. The remaining patients were diagnosed by other doctors, outside the University Hospital, and we relied on their diagnosis. Complementing these data, a personal history was obtained including regular use of a nasal spray. For each patient, a case report form was compiled.

Endpoints

The first aim of this study was to determine, whether and how many patients with CR would develop CRS after a follow-up period of more than 10 years in a chronological, descriptive fashion. The second goal was to identify whether there was a subgroup of CR patients prone to progress into a CRS. As a co-secondary endpoint, we wanted to know, whether specific determinants could be identified with higher probability of developing CRS.

Follow-Up Data Collection

Of the 328 contacted patients, 214 were not available, 114 were reached, and they agreed to participate. Two patients had meanwhile been diagnosed with CRS by another doctor, and the remaining 112 patients were assessed about persistent symptoms following a questionnaire, consisting of the NOSE-D (Nasal Obstruction Symptom Evaluation) and particularly the SNOT-20-GAV score (Sino-Nasal Outcome Test). Both scores are validated organ-specific, health-related quality-of-life measures for rhinologic patients [20-23]. The NOSE-D score was used to assess the severity of symptoms of nasal obstruction and the SNOT-20-GAV score to see whether patients could be suffering from symptoms complying with rhinosinusitis. We predefined a SNOT-20-GAV score of more than 12 as relevant as it was described by Baumann et al. [24]. In their collective, only 4.3% of patients with existing CRS scored lower than 12 points in a scale of the SNOT-20-GAV score, and the mean total score of patients without CRS was 12.6 points. Patients with elevated scores were invited for clinical follow-up examination, and all were examined by the same ENT doctor. A rhinosinusitis was diagnosed following the definition published by the European Position Paper On Rhinosinusitis and Nasal Polyps 2020 (EPOS 2020) [25]. CT scans were performed in 3 cases where CRSsNP was diagnosed.

Extrapolation for the Entire Group of 334 Patients (Refer to Fig. 1, Dotted Dark Red Boxes)

In our initial cohort of 334 patients, 6 had a documented follow-up and were diagnosed with CRS already before our telephone call. Excluding these patients from the study would have led to selection bias and thus lowering the incidence. Due to the fact that not all patients were available for phone call inquiries, we needed to fill the gap of missing data by extrapolation. Therefore, we extrapolated our data from the reached group of 114 without already documented follow-up to the entire group of 334. Five patients in the group of 16 interviewed patients were clinically diagnosed with CRS, accounting to 31.2%. This would mean that we would have missed 2 more CRS patients who were unavailable for clinical examination (7 patients). In the group of patients who were not contactable, we would have missed 17 CRS diagnoses (out of 214 patients). Adding the confirmed and missed diagnoses leads to our final count of CRS patients in the whole cohort as described in Figure 2. To verify whether this is appropriate, we used the Monte Carlo simulation method and repeatedly simulated independent random variables. Once the simulation was complete, the results were averaged together to provide an estimated number for our total cohort.

Fig. 2.

First endpoint. It illustrates the first endpoint. According to Figure 1 in the blue box, the number of our initial cohort, definitive CRS diagnosis in green, and expected missed CRS diagnosis in the dark red box.

Statistical Analysis

To assess our first goal, no specific statistical analysis was necessary. To test factors which could have a significant effect on developing rhinosinusitis, we used regression analysis, specifically logistic regression to analyze multiple determinants at the same time. Independent variables with different scale levels were compared, and the probability of developing CRS was assessed in consideration of all variables together. To test the overall model fit, we relied on the Akaike information criterion to evaluate the quality of the model through comparison of related models. All statistical testing was done with statistical software R (R Foundation, Vienna, Austria) with a p value of <0.05 to define statistical significance. We did not adjust p values for multiple comparisons due to the exploratory design of our study.

Results

For better comprehension of how patient enrollment took place, we refer to Figure 1. Of the 334 patients who met the inclusion criteria, 114 subjects were available for telephone interview and 7 of those patients had to be excluded due to incomplete data sets. 6 subjects from the initial cohort of 334 already had a confirmed, documented CRS diagnosis, and therefore did not have to be interviewed by telephone. They were included so that a total number of 113 patients (48 female, 65 male) remained with a mean age of 52 ± 15 years. The mean age of onset for diagnosis of CR was 38 ± 15 years. Patients with diagnosed rhinosinusitis had a mean age of onset for CR of 48 ± 13 years and for diagnosis of rhinosinusitis of 56 ± 16 years, diagnosis of a CRS came a median of 8 years after CR. Table 1 shows the demographics of the subjects in our cohort. The mean follow-up time was 13.6 years.

Table 1.

Demographics of our cohort

As mentioned above, prior to this study, 6 patients had referred to our outpatient clinic before the start of this study and were diagnosed with CRS (1 with AERD), 2 had been diagnosed with CRS by another doctor. We used the NOSE-D score only to assess the severity of nasal obstruction but not for the assessment of possible CRS. It therefore does not have any influence on our results. A total of 23 (20% of all replies) had a SNOT-20-GAV score >12 and were invited for a clinical follow-up examination including endoscopy. Seven did not want to come, due to the current pandemic, and therefore had to be excluded having incomplete data. Sixteen (14% of total sample patients) were then seen for clinical examination, and only five (31% of clinically examined patients) were diagnosed with CRS. As described in the methods section, we chose to extrapolate our data and to verify the correctness, independent random variables were simulated. This simulated value was 26.72 and statistical error calculation for a normally distributed collective (95% confidence interval) of 5–14, which is in line with the expected total number of 32 as stated above. Looking at the 107 interviewed patients with complete data sets, the prevalence was around 8% and 9.5% in the initial group of 334 patients.

Our main finding is that in our cohort, only 13 had a confirmed diagnosis of CRS. Of those 13 patients, six had CRS with polyps (CRSwNP). Of the patients with CRSwNP, 4 had an allergic rhinitis at initial presentation.

In our cohort, 54 patients suffering from allergic rhinitis and of those 5 (9%) later suffered from CRS. In the group of NAR, including 71 patients, 7 (10%) developed CRS. Only two were diagnosed with an infectious rhinitis and both developed a CRSwNP. From the 25 with mixed forms, 2 (8%) developed CRS. Table 2 shows our findings after logistic regression analysis of the main subgroups of CR. In our cohort, we could not find a statistically significant difference for either main subgroup of CR, inter alia allergic rhinitis, or NAR to have a higher probability of developing CRS.

Table 2.

Logistic regression analysis of the main subgroups of CR in order to find out whether a main subgroup has a statistically significant higher probability of leading to rhinosinusitis

Logistic regression analyses are run also for determinants/factors as shown in Table 3. We found that patients with prior surgical treatment of the nose (e.g., septo-turbinoplasty) were statistically significant more likely to develop CRS and so were long-term use of topical nasal treatments and other medication. No statistical significance was found for patients with allergies or systemic disease, such as vasculitis.

Table 3.

Logistic regression analysis of different determinants to discover if a specific determinant is statistically more prone to develop rhinosinusitis

As shown in Table 3, the age of onset had a significant effect on the development of rhinosinusitis. Figure 3 shows the probability of rhinosinusitis as function of age of onset and of SNOT-20-GAV score. We determined the sensitivity and specificity for the SNOT-20-GAV score of 96% and 40%, respectively, at a threshold value of 0.5, which in our cohort was at 39.3 points.

Fig. 3.

Probability of rhinosinusitis as function of age of onset of rhinitis and of SNOT-20-GAV score. Scatterplots of probability of rhinosinusitis as function of age of onset (in years) on the left and of SNOT-20-GAV score (in points) on the right.

Discussion

The main finding of this study is that only a small portion of patients with CR developed CRS over a period of 10–15 years. In our cohort of 334 patients, we expect around 9.5% of patients to develop CRS and only 8% of patients in our sample of 107. When comparing these percentages separately to the reported overall epidemiologically based prevalence in the western population of 10.9% progression from CR to rhinosinusitis seems questionable. Though, when looking at the clinically based prevalence of 3–6%, we found a 1.5–2 times higher prevalence in our cohort of only CR patients. Due to the relatively heterogeneous sample group, we sought to compare our results to both reported prevalences. In addition, recent studies reported that the use of both the EPOS symptom criteria and radiographic findings leads to an estimated CRS prevalence of 3.6% or less. When compared to either radiographic or symptom criteria alone, the prevalence was >12% [26, 27].

All in all, exact comparison is challenging due to the wide range of reported prevalence because diagnosis of CR and rhinosinusitis has not always been standardized. In addition, our cohort cannot be directly compared to a general prevalence, as some of the patients were followed over several years, most likely overestimating the number of CRS patients. However, our findings raise the question whether CR should be considered a major risk factor for CRS.

The strength of this study, next to the large patient number and long follow-up, is that we did not purely rely on retrospective data but contacted all patients to learn about their symptoms at present. The population at risk was clinically examined. To the best of our knowledge, this is the first study to investigate rhinitis progression in a linear fashion with a follow-up period of at least 10 years in a cohort of more than 100 patients.

CR and CRS affect a large proportion of the population worldwide. Since quality of life is heavily impaired with nasal symptoms, correct treatment as well as patient support and education are crucial. It is therefore important to investigate the pathophysiological pathways and the relationship. For example, reports on the relationship of allergic rhinitis, asthma, and CRS found different results. Meltzer et al. [28] reported that allergic rhinitis and rhinosinusitis may be caused by common inflammatory processes within a continuous airway of the nasal cavity and paranasal sinuses. Other projects have shown common genetic associations and that mucosal inflammation with dysregulation of the immune system and bacterial superantigens may play a role [29]. Furthermore, the contribution of allergies to the pathogenesis and symptoms of CRS has long been controversial [14, 30]. Suzuki et al. noted an increased prevalence of allergy (39%) in patients with CRS [31]. These findings were confirmed in some studies [32-35] but yet could not be reproduced by several others [36-38]. We would have expected to find an increased rate of CRS in our cohort of allergic individuals since 9 of the 13 patients with CRS had an allergy, but no statistically significant higher probability could be found. Interestingly, patients in our cohort suffering from allergic rhinitis also did not develop CRS significantly more often than other forms. Only 9% progressed to CRS. In contrast to these findings, Tan et al. [16], in a cohort of primary care patients with CRS, had found a higher prevalence of premorbid allergic rhinitis and CR, with percentages of 41–43% and 18–21%, respectively. In the group of patients with CRS, allergic and NAR were equally represented with 9% and 10% for the latter. Logistic regression analysis of our infectious rhinitis data did not show a significantly higher probability for this group to develop CRS even though the odds ratio clearly suggested differently. The odds ratio of this group in general has to be interpreted with care and is probably mainly due to the small sample size. Further research has to be done for this selected group of patients since the infectious/inflammatory pathophysiology could play a role in the development of CRS [39].

Different studies reported that NAR with eosinophilia syndrome (NARES) could act as a precursor to AERD/NERD since NARES patients frequently develop eosinophilic nasal polyps, bronchial hyperreactivity, and nonallergic asthma [11, 18, 40, 41]. Furthermore, AERD is thought to be a part of the local inflammatory drug-induced rhinitis endotype [11]. This was known when we started this study, and we wanted to further investigate this. Unfortunately, when looking at our cohort, these entities, e.g., AERD and NARES, were underrepresented, and so no meaningful statement can be made. But further focused investigation on the association of these entities can help in the understanding of the development of CRS. Another interesting report was published in 2008 by Gelinicik et al. [42], investigating whether allergic or NAR would predispose more to CRS. They started with CR patients like in our study and then saw these patients again for anamnestic and clinical follow-up. No predisposition was found for either form of CR, but they reported persistent rhinitis as a risk factor on its own. The follow-up period in this study was rather short.

Patients in our cohort with prior surgery of the nose (e.g., septo-, turbinoplasty) showed an increased probability to develop CRS. Studies analyzing the association of septal deviations and CRS reported an increased prevalence and severity of CRS, but this to date still is controversial [43, 44]. Both long-term use of topical treatment and prior surgery were predisposed to develop CRS in our cohort. This is most likely attributed to the fact that both therapies point to a higher disease burden rather than having a direct link with the intervention. Furthermore, medication use in general was a significant predisposition. This could be due to the fact that patients with confirmed CRS were older and therefore more likely to require regular medication (antihypertensives, etc.).

Recent reports found long-term breathing of polluted atmospheric air affecting tissue remodeling and therefore being factors in the multifactorial pathogenesis of CRS [45]. This is why we considered the inclusion of a variable which controls for the exposure to gases, dust, and fumes. But we had to remove it from all models as the dispersion was too big and sample size too small.

However, we observed that the older the patient was at onset of rhinitis, the higher the probability of developing rhinosinusitis over the course of time. From Figure 3, we may conclude that there is a relevant increase in probability at the age ≥40 years.

The exact pathomechanism of CRS and its association with CR is still not fully understood, so further research has to be done. But our findings show no association between any main subgroup of CR or CR as such with the probability of developing CRS. This could also be due to the fact that the treatment of our patients prevented them from developing CRS since the greater part of patients in our cohort still was using nasal topical treatments. If that was the case, this would strengthen the value of correct long-term therapy of nasal symptoms.

Study Limitations

This study was set up in a retrospective fashion. Our cohort consists of only CR patients, which constitutes a selection bias. We do not know the number of patients with rhinosinusitis in the non-reached group or for those with elevated SNOT-20-GAV score, who did not want to come for clinical examination. This shortcoming was corrected by extrapolation, as described above. We had to exclude patients because of incomplete data, though these patients were excluded randomly, so no selection bias arose. To accurately predict the development of CRS from rhinitis, we not only applied a mathematical model but also combined data from known CRS diagnoses and potentially missed diagnoses. Although extrapolation does introduce some source of bias, we feel that it is more accurate than limiting the sample size to those who could be reached.

We deliberately chose not to add a control group without CR for comparison due to the retrospective nature of the study and the lack of an unbiased population of patients visiting our department. As already mentioned in the results, sample sizes of some of the main subgroups of CR were small, and therefore, statistical analysis was impaired.

A large proportion of patients was still applying topical nasal steroids. Therefore, we cannot rule out that some patients progressed from CR to CRS yet being well controlled and thus having a SNOT-20-GAV score below 12. These patients would have been missed if they did not know their CRS diagnosis. We strongly believe that this is of low clinical relevance. In addition, topical nasal steroids require a prescription and thus regular checks by a physician, so the patients would know a change in their diagnosis that was inquired upon every telephone call.

Conclusion

Only a small number of rhinitis patients developed CRS, questioning the paradigm of rhinitis being a clear risk factor for CRS. None of the rhinitis subgroups showed a statistically significant higher odds ratio, and therefore, no specific population at risk could be identified. Only age of onset as well as high disease burden seem to negatively impact on the development of CRS.

Statement of Ethics

This investigation was approved by the Swiss Ethics Committees on research involving humans (BASEC-Nr.: 2019-01426). Informed consent was obtained orally upon telephone interviews or written from those study participants who were seen for clinical examination with approval as indicated above.

Conflict of Interest Statement

All authors have no conflict of interest to disclose.

Funding Sources

No funding was received for this study.

Author Contributions

Bram van Schie and Joel J. Vavrina carried out the experiment and fabricated the data sample. Bram van Schie wrote the manuscript with support from Michael B. Soyka. Bram van Schie and Michael B. Soyka conceived the original idea. Michael B. Soyka supervised the project.

Data Availability Statement

All data generated or analyzed during this study are included in this article. Further inquiries can be directed to the corresponding author.

This article is licensed under the Creative Commons Attribution-NonCommercial 4.0 International License (CC BY-NC). Usage and distribution for commercial purposes requires written permission. Drug Dosage: The authors and the publisher have exerted every effort to ensure that drug selection and dosage set forth in this text are in accord with current recommendations and practice at the time of publication. However, in view of ongoing research, changes in government regulations, and the constant flow of information relating to drug therapy and drug reactions, the reader is urged to check the package insert for each drug for any changes in indications and dosage and for added warnings and precautions. This is particularly important when the recommended agent is a new and/or infrequently employed drug. Disclaimer: The statements, opinions and data contained in this publication are solely those of the individual authors and contributors and not of the publishers and the editor(s). The appearance of advertisements or/and product references in the publication is not a warranty, endorsement, or approval of the products or services advertised or of their effectiveness, quality or safety. The publisher and the editor(s) disclaim responsibility for any injury to persons or property resulting from any ideas, methods, instructions or products referred to in the content or advertisements.