High Rates of Detection of Respiratory Viruses in the Nasal Washes and Mucosae of Patients with Chronic Rhinosinusitis

Abstract

Respiratory viral infections are often implicated as triggers of chronic rhinosinusitis (CRS) flare-ups. However, there is a paucity of respiratory viral surveillance studies in CRS patients, and such studies could elucidate the potential role of viruses in promoting symptoms and aggravating mucosal inflammation. Therefore, a prospective case-control study was conducted to determine the prevalence of respiratory viruses in CRS patients and non-CRS controls. Nasal lavage fluids and turbinate epithelial cells were collected prospectively from 111 CRS patients and 50 controls. Multiplex PCR was used to identify common respiratory viruses in both sample types and the infection rate was compared between groups. Respiratory viruses were detected in 50.5% of lavage samples and in 64.0% of scraping samples from CRS patients. The overall infection rate was significantly different in CRS patients and controls (odds ratio, 2.9 in lavage and 4.1 in scraping samples). Multiple viral infections were detected more frequently in lavage samples from CRS patients than those from controls (P < 0.01; odds ratio, 7.7). Rhinovirus was the most prevalent virus and the only virus with a significantly different infection rate in CRS patients and controls in both samples (odds ratio, 3.2 in lavage and 3.4 in scraping samples). This study detected a higher prevalence of respiratory viruses in CRS patients than controls, suggesting that there may be significant associations between inflammation of CRS and respiratory viruses, particularly rhinovirus. Further studies should investigate the exact role of highly prevalent respiratory viruses in CRS patients during symptomatic aggravation and ongoing mucosal inflammation.

INTRODUCTION

Chronic rhinosinusitis (CRS) is one of the most prevalent chronic illnesses, but the etiology and pathogenesis of CRS are not well understood (1). Clinically, respiratory viral infections are often implicated as triggers of CRS flare-ups (2, 3). Nevertheless, many researchers consider that there is insufficient evidence that respiratory viral infection is related to the pathogenesis of CRS, because viruses are virtually ubiquitous in the general population and they are not linked to persistent sinonasal inflammation. However, some experimental studies have suggested an important role for viral pathogens during the exacerbation and ongoing inflammation in CRS (4–8). In vitro studies using human sinonasal epithelial cells have demonstrated that viral infection or exposure to a synthetic viral analogue increases the expression of immune effectors (4, 5). Respiratory viral infections are also known to damage the nasal epithelial cells and cilia (6), cause dysfunctions of intercellular junctions such as tight junctions and adherens junctions (7), and increase bacterial adhesion (8).

Asthma and CRS are common airway diseases, which are linked to atopy and allergic inflammation, and potential parallels have been established in the pathogenesis of these conditions (9). A number of studies with asthmatic patients have demonstrated a very strong association between respiratory viral infection and acute exacerbations of asthma (10–12). A pilot study tested the presence of rhinovirus in the nasal cavity of CRS patients, and rhinoviruses were detected in 8/39 (21%) epithelial cell samples from CRS patients, whereas none were detected in the controls (13). This preliminary study suggested that rhinovirus may be associated with CRS pathogenesis, although the number of viral species tested and the number of patients enrolled were limited. There is also a paucity of respiratory viral surveillance studies in CRS patients, and such studies could elucidate the associations between mucosal inflammation of CRS and respiratory viruses. Thus, the prevalence of viral pathogens in CRS patients with no signs of acute viral infection was investigated in this prospective study using multiplex PCR assay to sensitively detect various respiratory viruses in nasal samples. After samples were collected from control subjects without CRS, the respiratory virus infection rate in CRS patients was compared with that in control subjects. These data might provide new insights into the putative role of respiratory viruses in promoting symptoms and aggravation of mucosal inflammation of CRS.

MATERIALS AND METHODS

Study population.

One hundred twenty-two patients were recruited prospectively who had received endoscopic sinus surgery for CRS, which had not been controlled after optimal medical treatment, between August 2011 and March 2012 at the Asan Medical Center, Seoul, South Korea. All patients met the established diagnostic criteria for CRS (14). Patients with fungal sinusitis, current immunotherapy, aspirin desensitization, or cystic fibrosis were excluded from this study. Further reasons for exclusion were treatment with local or oral steroids within the previous 4 weeks. The controls were volunteers who required planned surgery for chronic tonsillitis or thyroid masses, but none of the controls had experienced CRS. The presence of sinusitis in the control group was ruled out based on their history, physical examination, plain sinus X-ray, and/or computed tomography (CT) imaging. All patients and control subjects were checked for viral upper respiratory infection (URIs) over a 4-week period. URI symptoms were assessed using the Jackson scale of eight symptoms: nasal discharge, nasal congestion, sneezing, cough, malaise, throat discomfort, fever/chills, and headache (15). Eleven CRS patients with symptoms or signs of viral URIs during the preceding 4 weeks were excluded from the study. Asthma was defined as a history of asthma symptoms (recurrent dyspnea, wheezing, cough, and chest tightness) associated with reversible airflow obstruction, measured by spirometry (16). Allergic rhinitis was defined by 2 or more allergic symptoms (watery rhinorrhea, nasal obstruction, nasal itching, and sneezing) for over 1 h on most days and a positive skin prick test or detectable serum-specific IgE (17). The study was approved by the Institutional Review Board of the Asan Medical Center (Seoul, South Korea). All adult participants and legally authorized representatives of children provided informed consent.

Collection of samples.

Nasal samples were collected from all participants from summer 2011 to spring 2012 (August 2011 to March 2012) for 8 months. The sample size was approximately equally distributed over the months, i.e., 13 to 14 samples/month from CRS patients and 7 to 8 samples/month from control subjects. To verify the presence of respiratory viruses in the study population, two high-sensitivity sampling techniques were used, i.e., nasal lavage and scrapes of the nasal epithelium from the inferior turbinate. Both methods were described in our previous study (13), where mucosal scraping was found to be more sensitive than nasal lavage for viral detection. In the operating room, nasal lavage samples were collected immediately after the induction of general anesthesia. A balloon catheter was placed posterior to the nasal cavity and choana, before sterile saline solution (0.9% NaCl) at room temperature was aerosolized into the nostrils using a needle-free syringe. This procedure was repeated alternately in the nostrils until 10 ml of lavage fluid was recovered. After lavage, the inferior turbinate epithelial cells were scraped from the medial aspect of the middle third of the inferior turbinate using a sterile disposable Rhino-probe mucosal curette (Arlington Scientific, Inc., Springville, UT). Scrape samples were removed immediately from the Rhino-probe cusp and transferred to plastic tubes containing 2 ml of sterile phosphate-buffered saline. The samples were frozen immediately after collection and stored at −70°C until use.

Identification of respiratory viruses.

Viral RNA was extracted from 140 μl of each respiratory sample using a QIAamp viral RNA kit (Qiagen, GmbH, Hilden, Germany), according to the manufacturer's instructions. To synthesize cDNA, reverse transcription was performed using a RevertAid first-strand cDNA synthesis kit (Fermentas, Burlington, ON, Canada) in a final volume of 20 μl containing 5 μM random hexamer primer, a 1 mM concentration of each deoxynucleoside triphosphate (dNTP), 20 U of RiboLock RNase inhibitor, 5× reaction buffer, and 200 U of RevertAid Moloney murine leukemia virus (M-MLV) reverse transcriptase. Each cDNA was subjected to multiplex PCR using a Seeplex RV15 ACE detection kit (Seeplex RV15; Seegene Inc., Seoul, South Korea), according to the manufacturer's instruction. Briefly, 3 μl of synthesized first-strand cDNA, 5× RV15 primer mix, 2× multiplex master mix (Taq DNA polymerase and dNTPs included), and 3 μl 8-methoxypsoralen (8-MOP) solution were mixed. The primer mixes contained the internal control template and primer pairs to validate the PCR. Three reactions (using primer mixes A, B, and C) were conducted for each sample, according to the kit instructions. The specific primer targets for each respiratory virus are shown in Table 1. The multiplex PCR detected rhinoviruses and enteroviruses, parainfluenza viruses 1, 2, 3, and 4, influenza viruses A and B, respiratory syncytial viruses (RSV) A and B, coronaviruses 229E/NL63 and OC43, adenovirus, human metapneumovirus (HMPV), and human bocaviruses 1, 2, 3, and 4. The PCR used the following conditions: initial denaturation at 94°C for 15 min; 40 cycles of 94°C for 30 s, 60°C for 1 min 30 s, and 72°C for 1 min 30 s; and a final extension at 72°C for 10 min. The multiplex PCR products were visualized by electrophoresis on 2% agarose gel.

Table 1.

Targets for the detection of respiratory viruses using the Seeplex respiratory detection assay

Target	Size in agarose gel (bp)
RV15 ACE detection (set A)
Internal control	850
Human adenovirus	534
Human coronavirus 229E/NL63	375
Human parainfluenza virus 2	264
Human parainfluenza virus 3	189
Human parainfluenza virus 1	153
RV15 ACE detection (set B)
Internal control	850
Human coronavirus OC43	578
Human rhinovirus A/B/C	394
Human respiratory syncytial virus A	269
Influenza A virus	206
Human respiratory syncytial virus B	155
RV15 ACE detection (set C)
Internal control	850
Human bocaviruses 1, 2, 3, and 4	579
Influenza B virus	455
Human metapneumovirus	351
Human parainfluenza virus 4	249
Human enterovirus	194

Statistical analysis.

Pearson's chi-squared tests and Fisher's exact tests (two-tailed) were used to compare virus detection rates in CRS patients and control subjects. Odds ratios (OR) and 95% confidence intervals (CI) were also calculated. Statistical analyses were performed using SPSS version 16.0 (SPSS, Inc., Chicago, IL), and a P value of <0.05 was regarded as significant.

RESULTS

From August 2011 to March 2012, the study enrolled 111 patients with CRS and 50 control subjects, and their demographic and clinical data are shown in Table 2. Among the 111 CRS patients, asthma was present in 11 (9.9%) patients and allergic rhinitis was present in 26 (23.4%) patients. Comparisons of case patients and control subjects showed that the two groups did not differ significantly in terms of age, gender, concomitant asthma, and allergic rhinitis (P > 0.05). This study included 62 (55.9%) CRS patients with nasal polyps and 49 (44.1%) CRS patients without nasal polyps. There was no association of asthma, allergic rhinitis, and nasal polyps with respiratory virus detection rate or multiple virus detection in 111 CRS patients, respectively (P > 0.05 for each). Additionally, there was no statistically significant difference related to the age of the patients in which viruses were detected, whether multiple viruses were detected, or the rate of detection (P > 0.05 for each).

Table 2.

Demographics and clinical data

Parameter	No. (%) of subjects		P value
	CRS patients (n = 111)	Control subjects (n = 50)
Age			>0.05a
≤17	18 (16.2)	6 (12.0)
>18	93 (83.8)	44 (88.0)
Gender			>0.05a
Male	67 (60.4)	24 (48.0)
Female	44 (39.6)	26 (52.0)
Asthma	11 (9.9)	4 (8.0)	>0.05b
Allergic rhinitis	26 (23.4)	9 (18.0)	>0.05a
No. of sinus operations	111 (100)	−
Primary (n = 1)	81 (73.0)
Revision (n ≥ 2)	30 (27.0)
Combined nasal pathology		−
Nasal polyps
+	62 (55.9)
−	49 (44.1)
No. of patients with:
Low CT score (3–12)c	74 (66.7)
High CT score (13–24)c	37 (33.3)
Deviated nasal septum
+	56 (50.5)
−	55 (49.5)
Inferior turbinate hypertrophy
+	52 (46.8)
−	59 (53.2)

^aChi-squared test.

^bFisher's exact test.

^cBased on the Lund-Mackay scoring system (41).

(i) Viral detection in nasal lavage samples.

PCR analysis was positive for one or more respiratory viruses in 56 (50.5%) of the nasal lavage samples from the 111 CRS patients and 14 (28.0%) of the control subjects; there were significant differences in the overall virus detection rates between the two groups (P < 0.01; Table 3). The CRS patients were 2.9-fold more likely to have been infected with viruses than control subjects without CRS based on the analysis of the nasal lavage fluid (OR = 2.9; 95% CI = 1.2 to 5.2). Interestingly, coinfections with >1 virus was detected in 27 (24.3%) of the CRS patients and two (4.0%) of the control subjects, and the coinfection rate was significantly different between the lavage samples from the two groups (P < 0.01, OR = 7.7; CI = 1.5 to 19.5). Nineteen CRS patients (17.1%) were infected with two species of virus, four patients (3.6%) were infected with three species of virus, and four patients (3.6%) were infected with more than four species of virus. The results of the PCR tests for each virus are shown in Table 4. Rhinoviruses were by far the most common respiratory viruses detected in the CRS and control participants. The PCR analysis was rhinovirus positive in 29 (26.1%) CRS samples and 5 (10.0%) control samples, and the detection rate was significantly different between the two groups (P < 0.05, OR = 3.2; CI = 1.2 to 8.8). The second most prevalent virus was parainfluenza virus in both groups, i.e., 26 samples (23.4%) were positive in the CRS group compared with 4 samples (8.0%) in the control group (P < 0.05, OR = 3.2; CI = 1.1 to 9.7). In the study group, 15 samples (13.5%) were positive for influenza, 12 (10.8%) for RSV, 12 (10.8%) for coronavirus 229E and OC43, 3 (2.7%) for adenovirus, 2 (1.8%) for enterovirus, 1 (1.0%) for HMPV, and 1 (1.0%) for human bocavirus. The detection rates for influenza, RSV, coronavirus 229E and OC43, adenovirus, enterovirus, HMPV, and human bocavirus did not differ between the CRS and control groups (P > 0.05).

Table 3.

Respiratory virus detection in nasal samples from CRS patients and control subjects

Respiratory viruses	No. (%) of positive samples from:
	CRS patients (n = 111)				Control subjects (n = 50)
	Lavage	Scraping	Overalla	Concordantb	Lavage	Scraping	Overalla	Concordantb
All viruses	56c (50.5)	71c (64.0)	84c (75.7)	27d (24.3)	13 (26.0)	15 (30.0)	19 (38.0)	4 (8.0)
Single virus	29 (26.1)	50d (45.0)	31 (27.9)	23 (20.7)	11 (22.0)	10 (20.0)	12 (24.0)	4 (8.0)
Multiple viruses (>1 species)	27c (24.3)	21 (19.0)	53c (47.7)	4 (3.6)	2 (4.0)	5 (10.0)	7 (14.0)	0 (0)
2 species	19 (17.1)	14 (12.4)	31 (27.9)	4 (3.6)	2 (4.0)	4 (8.0)	5 (10.0)	0 (0)
3 species	4 (3.6)	3 (2.7)	12 (10.8)	0 (0)	0 (0)	1 (2.0)	2 (4.0)	0 (0)
≥4 species	4 (3.6)	4 (3.6)	10 (9.0)	0 (0)	0 (0)	0 (0)	0 (0)	0 (0)

^aConsidered positive if either lavage or scraping was positive.

^bConsidered positive if both lavage and scraping were positive.

^cThe detection rate in CRS patients is significantly greater than that in controls (P < 0.01).

^dThe detection rate in CRS patients is significantly higher than that in controls (P < 0.05).

Table 4.

Identification of respiratory viruses detected in nasal samples from CRS patients and control subjects

Virus	No. (%) of positive samples from:
	CRS patients (n = 111)				Control subjects (n = 50)
	Lavage	Scraping	Overalla	Concordantb	Lavage	Scraping	Overalla	Concordantb
Rhinovirus	29c (26.1)	35d (31.4)	49d (44.0)	15c (13.5)	5 (10.0)	6 (12.0)	10 (20.0)	1 (2.0)
PIVe	26c (23.4)	24 (21.6)	44d (39.6)	6 (5.4)	4 (8.0)	7 (14.0)	9 (18.0)	2 (4.0)
PIV 1	14	7	20	1	0	2	2	0
PIV 2	11	10	19	2	1	0	1	0
PIV 3	11	9	21	0	1	1	2	0
PIV 4	1	1	2	0	0	1	1	0
Influenza A and B	15 (13.5)	14 (12.6)	24d (21.6)	5 (4.5)	2 (4.0)	3 (6.0)	5 (10.0)	0 (0)
RSVf	12 (10.8)	12 (10.8)	22c (19.8)	2 (1.8)	1 (2.0)	1 (2.0)	2 (4.0)	0 (0)
RSV A	9	9	17	1	1	1	2	0
RSV B	7	7	13	1	2	0	2	0
Coronavirus 229E and OC43	12 (10.8)	15 (13.5)	24 (21.6)	3 (2.7)	3 (6.0)	2 (4.0)	4 (8.0)	1 (2.0)
Adenovirus	3 (2.7)	2 (1.8)	5 (4.5)	0 (0)	0 (0)	0 (0)	0 (0)	0 (0)
Enterovirus	2 (1.8)	1 (1.0)	3 (2.8)	0 (0)	0 (0)	0 (0)	0 (0)	0 (0)
HMPVg	1 (1.0)	2 (1.8)	3 (2.8)	0 (0)	0 (0)	1 (2.0)	1 (2.0)	0
Human bocavirus	1 (1.0)	0 (0)	1 (1.0)	0 (0)	0 (0)	1 (2.0)	1 (2.0)	0

^aConsidered positive if either lavage or scraping was positive.

^bConsidered positive if both lavage and scraping were positive.

^cThe detection rate in CRS patients is significantly higher than that in controls (P < 0.05).

^dThe detection rate in CRS patients is significantly higher than that in controls (P < 0.01).

^eFour subtypes of parainfluenza virus (PIV) were detected as various combinations. PIV detected in one patient was counted as one regardless of how many subtypes the patient had.

^fTwo subtypes of respiratory syncytial virus (RSV) were detected as various combinations. RSV detected in one patient was counted as one regardless of how many subtypes the patient had.

^gHMPV, human metapneumovirus.

(ii) Viral detection in mucosal scrape samples.

In the mucosal scrape samples, the PCR analysis was positive for one or more respiratory viruses in 71 (64.0%) of the 111 CRS patients and 15 (30.0%) of the control subjects, and there were significant differences in the overall virus detection rates in the two groups (P < 0.01; Table 3). CRS patients were 4.1-fold more likely to be infected with viruses than control subjects according to the mucosal scrape samples (OR = 4.1; 95% CI = 2.0 to 8.5). Multiple viral infections were identified in 21 (19.0%) of the CRS patients and five (10.0%) of the control subjects. Fourteen CRS patients (12.4%) were infected with two species of virus, 3 patients (2.7%) were infected with three species of virus, and 4 patients (3.6%) were infected with more than four species of virus. In contrast to the nasal lavage fluid analysis, there were no significant differences in the coinfection rates in the two groups (P = 0.17). The results of the PCR tests for each virus are shown in Table 4. The PCR analysis was rhinovirus positive for 35 (31.4%) of the CRS group and 6 (12.0%) of the control group, and rhinovirus was the only respiratory virus that has a significantly different detection rate in CRS patients and control subjects (P < 0.01, OR = 3.4; CI = 1.3 to 8.7). In the CRS group, 24 (21.6%) were positive for parainfluenza, 14 (12.6%) for influenza, 12 (10.8%) for RSV, 15 (13.5%) for coronavirus 229E/NL63 and OC43, 2 (1.8%) for adenovirus, 1 (1.0%) for enterovirus, 2 (1.8%) for HMPV, and none for human bocavirus. There were no significant differences in the positive detection rates of parainfluenza, influenza, RSV, coronavirus 229E/NL63 and OC43, adenovirus, HMPV, and human bocavirus in the CRS and control groups (P > 0.05).

When a result was considered positive if either lavage or scraping was positive, there were significant differences in the overall and multiple virus detection rates between CRS patients and controls (P < 0.01 for each; Table 3). Rhinovirus, parainfluenza, influenza, and RSV in CRS patients were detected significantly greater than those in controls (P < 0.01 for rhinovirus, parainfluenza, and influenza and P < 0.05 for RSV; Table 4).

In concordant positive samples, which are positive both lavage and scraping, the overall virus detection rate in CRS patients was higher than that in control subjects (P < 0.05; Table 3). Like the data from mucosal scraping, only rhinovirus in CRS patients showed significantly higher detection rate than that in control subjects (P < 0.05; Table 4).

DISCUSSION

It is known that CRS has a multifactorial pathogenesis, and various inflammatory stimuli such as microbial and environmental factors may induce chronic inflammation at different stages of the disease and in different subgroups (18, 19). Respiratory viruses can cause florid sinonasal symptoms in CRS patients, which may have a long-lasting effect on the sinonasal mucosa (2, 3). The role of respiratory viruses in exacerbation and ongoing inflammation in the sinus mucosa has been demonstrated in a number of experimental studies (4–8, 20). However, few studies have evaluated the pathogenic impact of community-acquired respiratory viruses on patients with CRS. Thus, surveillance studies of viral respiratory infections may be the first step toward a better understanding of the relationship between inflammatory state of CRS and respiratory viral infection.

The present study found that a higher proportion of CRS patients had respiratory viruses in their nasal secretions and mucosae compared with non-CRS controls. Specifically, multiple viral infections were detected more frequently in CRS patients than the control subjects. These results suggest that chronic inflammatory conditions in the sinuses and nasal cavities of CRS patients may provide a favorable condition for infection with multiple respiratory viruses. In addition, these results also suggest that community-acquired respiratory viruses may play another important role in the aggravating inflammation of CRS, similar to bacteria and fungi (21, 22).

In this study, rhinovirus was the most prevalent virus in the lavage and mucosal scrape samples. The PCR analysis was rhinovirus positive for 29 (26.1%) of lavage samples and 35 (31.4%) mucosal scrape samples. Rhinovirus was the only respiratory virus that had a significantly different detection rate in CRS patients and control subjects in both samples (P < 0.05). The high prevalence of rhinovirus in CRS patients was also confirmed in our pilot study (13). Similar outcomes have been reported in studies of asthmatic patients. Rhinovirus RNA was found in 32.4% of children with asthma, suggesting that rhinovirus may be important in the pathogenesis of asthma (20). Many studies have demonstrated that a high proportion of asthma patients had respiratory viruses in their respiratory secretions, rhinoviruses were the only respiratory viruses associated with asthma exacerbations (23–26). Based on the current study, infections of the nasal cavity by rhinoviruses may have a significant role in the promoting symptoms and ongoing mucosal inflammation of CRS.

The presence of respiratory viruses but the absence of symptoms or signs of viral URIs in the CRS participants may be explained in several ways. It may reflect persistent asymptomatic infection, the asymptomatic period before the development of symptomatic infection, or the presence of the virus after symptomatic infection. However, nasal samples were obtained from participants who did not show URI symptoms during the preceding 4 weeks, and none of the subjects developed URI symptoms after postoperative sample collection in the hospital. Therefore, the detection of viruses in the nasal samples may have been due to the subclinical persistence of respiratory viruses from previous infections. A few studies have provided evidence of persistent respiratory viral infections, where HRV can replicate in the cells of the lower respiratory tract and persist for over a year (27–29). A previous study by our group also demonstrated persistent infections with parainfluenza virus for up to 3 months in patients with olfactory dysfunctions after virus infections (30).

A major strength of this study was that the prevalence of respiratory virus was assessed using two sampling techniques: nasal lavage and mucosal scrapes from the inferior turbinate. Although respiratory samples for detection of respiratory viruses are generally collected by nasopharyngeal swabs or nasopharyngeal aspirate in adults (31), nasal lavage fluid and scraped inferior turbinate epithelial cells have been used to detect respiratory virus in upper respiratory tract (27, 30, 32–34). This multitechnique approach to the detection of respiratory viruses improved the sensitivity, and the results of this study confirmed the prevalence of respiratory viruses in the nasal cavity of CRS patients. However, we considered mucosal scraping as more clinically significant and specific method than nasal lavage fluid, since the turbinate epithelial cells may better reflect sinus mucosa. In addition, nasal lavage fluid might have an increased chance of trapping viruses floating in ambient air, thus resulting in higher false-positive rates, and of containing a more mixed viral infection than the scraping samples. Another strength of this study was the use of very sensitive multiplex PCR assays (Seeplex RV15; Seegene Inc., Seoul, South Korea) for a full range of respiratory viruses. Other studies have also shown that this multiplex PCR approach is more sensitive for detecting respiratory viruses compared with conventional methods (35, 36).

This study had some limitations. First, there were differences in the baseline number of case patients and control subjects (111 CRS patients and 50 control subjects). In addition, the control subjects were heterogeneously diagnosed and composed of patients with chronic tonsillitis (n = 24) or thyroid masses (n = 26). However, great efforts were made to ensure the adequate selection of control subjects, and it was confirmed that none of control subjects had experienced sinusitis based on their history, physical examination, a plain sinus X-ray, and/or CT imaging. The viral detection rate in the control samples was also similar to that of asymptomatic control subjects in another study (37). Thus, the different number of subjects in the case group and the heterogeneous control group may not have confounded our findings. Second, the site of sample collection was a limitation in this study. Lavage samples were collected from the nasal cavity, whereas mucosal scrapings were taken from the inferior turbinate epithelia. Thus, both samples were not collected from the sinus mucosae. The previous study reported that no respiratory viruses were found in any sinus mucosa samples of CRS patients and normal subjects (38). Therefore, it may be necessary to determine whether respiratory viruses invade the sinus mucosa to understand the relationship of CRS pathogenesis to respiratory viruses. Third, although enterovirus was detected in only a few CRS patients in the present study, it has potential cross-reactivity with rhinovirus in most PCR assay. However, cross-reactivity could be ignored in this study, because enterovirus contributes relatively little to URI in the immunocompetent subjects and sensitivity of Seeplex RV15 to the respiratory virus detection showed compatible with that of reverse transcriptase PCR (39, 40).

In summary, this study detected a higher prevalence of respiratory virus infection in CRS patients than in controls. Of the various respiratory viruses, rhinovirus was by far the most prevalent virus and the only virus that has a significantly different detection rate in two groups. These results suggest that there may be strong associations between mucosal inflammation of CRS and respiratory viral infections, particularly rhinovirus infection. Further studies should investigate the etiologic role of high detection rate of respiratory viruses in CRS patients during symptomatic aggravation and ongoing mucosal inflammation.