Statistical analyses
Statistical analyses were performed using SPSS 25.0 (IBM Corp., Armonk, NY, USA). In the univariate analyses, the chi-squared test was used to compare categorical variables, while Student’s t-test was used for continuous variables. Logistic regression analysis was used to assess the risk factors for paranasal sinusitis. Additionally, the Cochran–Armitage trend test of SAS 9.4 (SAS Institute, Cary, NC, USA) was used to analyze the age-specific prevalence of paranasal sinusitis in smokers and non-smokers. A p-value <0.05 was considered statistically significant for all analyses.
Results
Among the 4813 participants who were investigated (2368 men and 2445 women; mean age, 53.1±10.1 years), paranasal sinusitis was diagnosed in 740 (15.4 %). Table 1 shows the characteristics of the study participants with and without paranasal sinusitis. The data show that sex was significantly associated with the presence of paranasal sinusitis, together with age, BMI, alcohol consumption, stroke, smoking, and hypertension.
Logistic regression analysis was performed to examine the risk factors associated with paranasal sinusitis (Table 2). When adjusted for age and sex (Model A), alcohol consumption and hypertension were not significantly associated with paranasal sinusitis. The final model (Model B) adjusted for all the factors that were found to be statistically significant in the univariate analysis, including age, sex, BMI, smoking, alcohol consumption, hypertension, and stroke—only smoking was significantly associated with the prevalence of paranasal sinusitis.
The age-specific prevalence of paranasal sinusitis in smokers and non-smokers is shown in Table 3. For all age groups, the prevalence was higher in smokers than in non-smokers (Figure 1). The prevalence of paranasal sinusitis significantly increased with age in non-smoking individuals, whereas no significant age-specific difference was found among smokers (trend test, non-smokers p<0.001; smokers p=0.499).
Table 4 shows a comparison of the prevalence of sinusitis affecting different regions between smokers and non-smokers; the anterior nasal sinuses were more often affected in smokers than in non-smokers, while no significant difference was found regarding the sphenoid sinus. Among the total study population, maxillary sinusitis was the most prevalent (13.4 %) while the prevalence was 19.1 % in smokers and 11.7 % in non-smokers.
Discussion
The present study investigated the prevalence of paranasal sinusitis, diagnosed using MRI, among a community population from Shanghai, China. Additionally, the epidemiological and clinical characteristics of sinusitis between smokers and non-smokers were investigated. Overall, the prevalence of paranasal sinusitis was found to be 15.4 % in adults aged 35–75 years. This prevalence increased concurrently with age, from 12.6 % (35–44 years) to 16.5 % (55–75 years); a peak was reached at 55 years, stabilizing thereafter. The Cochran–Armitage trend test further showed that there was a positive correlation between age and paranasal sinusitis in non-smokers, while the prevalence in smokers remained consistently high, at more than 20 % in each age group. Moreover, it was found that paranasal sinusitis more often involved the anterior nasal sinuses in smokers than in non-smokers.
In China, the first large-scale investigation concerning the epidemiology of CRS was reported in 2015. This investigation estimated a CRS prevalence of 8.0 % in the entire population and 8.2 % among adults aged 15–75 years 21, which was lower than the prevalence noted in the present study. This difference may be explained as follows: The previous study diagnosed CRS based on the results from a standardized questionnaire, whereas this study employed MRI for this purpose. The former method of diagnosis may result in some patients with unobvious symptoms being overlooked, while the latter is not able to distinguish the course or type of sinusitis effectively. Havas et al. reported abnormal CT and MRI findings of the paranasal sinuses in asymptomatic adults to be as high as 42.5 % 13. Kim et al. further confirmed that the prevalence and risk factors of CRS differed rather substantially when using different criteria, even in the same population22. Because of this difference, the investigation from 2015 may have underestimated the prevalence.
Compared to other studies that diagnosed paranasal sinusitis by means of MRI, the prevalence noted in our study was lower than that noted in studies from Denmark (31.7 %) 20, Japan (33.8 %)23, and Norway (66 %) 12. We believe that this variation could be due to the use of different diagnostic criteria and research techniques. Additionally, the prevalence of paranasal sinusitis is influenced by climate and race, being more likely to occur in colder climates and in people who are white 24-26. That said, in Japan, a Lund–Mackay score ≥4 was classified as an MRI abnormality that was suspected as sinusitis, whereas in Norway, opacifications that indicated mucosal thickenings, polyps, retention cysts, or fluid were recorded when measuring >1 mm. In our study, however, mucosal thickening of ≥5 mm was used as an indication of paranasal sinusitis, as per the Danish diagnostic criteria. By choosing a relatively high cut-off value, challenges regarding overdiagnosis due to the high sensitivity of MRI can be overcome, ensuring more accurate findings 20.
Among all the participants in this study, 49.2 % were men. We found that men were approximately 86 % more likely to have paranasal sinusitis than women (adjusted OR = 1.86, 95 % CI = 1.54–2.24), which is consistent with the results of studies from Asia 21, 23, 27. However, some surveys from developed Western countries found that women were at a higher risk of developing paranasal sinusitis6, 9, 28. This discrepancy may be related to differences in the living habits and level of awareness of paranasal sinusitis between populations from Eastern and Western countries.
At present, the relationship between the prevalence of paranasal sinusitis and age is somewhat controversial. In a previous study, Chen et al. found that the prevalence of CRS in Canadians older than 12 years increased with age, reaching a plateau after the age of 60 years28. Additionally, Hirsch et al. found that the prevalence of CRS peaked at 15.9 % between the ages of 50 and 59 years, subsequently dropping to 6.8 % after age 69 24, while Shi et al. found a higher prevalence in individuals aged 15–34 years21. In our study, we found that the overall prevalence of paranasal sinusitis in adults aged 35–75 years increased with age, reaching a peak at 55 years of age and stabilizing thereafter. The prevalence of paranasal sinusitis was found to have a strong positive correlation with age, particularly among non-smokers. This may be related to certain changes in the nasal anatomy and physiology that occurs with aging, such as reduced mucosal blood flow and impaired mucociliary function 29.
It is well known that smoking is closely related to respiratory diseases. Literature shows that tobacco smoke can adversely affect sinonasal mucociliary clearance, innate immune function, and olfactory mucosal metaplasia 8. Several clinical studies have confirmed that there is a correlation between smoking and paranasal sinusitis in the Asian population 8, 10, 30. Our research confirmed these findings, in that smoking is one of the significant independent risk factors in the development of paranasal sinusitis.
The location of paranasal sinusitis differs depending on its etiology. In contrast to the abundant literature regarding the correlation between smoking and paranasal sinusitis, there are relatively few studies assessing its most prevalent location. Interestingly, we found that in smokers, paranasal sinusitis more commonly occurs in the anterior nasal sinuses than in non-smokers, where it occurs most often in the maxillary sinus, followed by the ethmoid, frontal, and sphenoid sinuses. In smokers, the most common sites of sinusitis were found to be the maxillary sinus, followed by the sphenoid, frontal, and ethmoid sinuses. Kjaergaard et al. demonstrated that smokers have lower nasal cavity volumes and smaller minimum cross-sectional areas, achieve lower peak nasal inspiratory flow rates, and have a lower decongestive capacity of the nasal mucosa 31. Hence, we speculate that these nasal changes brought about by smoking are more likely to affect the mucociliary clearance movement and ventilation of the anterior nasal sinuses that open in the middle nasal passage. More research is needed to support this hypothesis.