Chronic rhinosinusitis is associated with increased risk of COVID-19 hospitalization.

OBJECTIVE: The rationale of the study was to examine the association between chronic rhinosinusitis (CRS) and COVID-19 hospitalization. STUDY
DESIGN: Retrospective cohort study.
SETTING: Cleveland Clinic hospital inpatient and outpatient.
METHODS: A retrospective chart review of patients that were tested for COVID-19 at Cleveland Clinic. The study took place between March 8, 2020 and May 15, 2020.
RESULTS: From a total of 23,282 Patients that underwent SARS-CoV-2 testing, 996 COVID-19 negative and 998 COVID-19 positive patients were included in the analysis. COVID-19 positive patients with chronic rhinosinusitis (CRS) were at higher risk for hospitalization compared to patients without CRS (39.2% vs 25.7%, p = 0.0486). There was no significant difference between the two groups in relation to ICU admission, mechanical ventilation, and death, After adjustment for covariates, our multivariate analysis showed that patients with chronic rhinosinusitis (CRS) were approximately 3.46 (OR = 3.19, 95% CI (1.12-10.68)) times more likely to be hospitalized compared to patients that have no CRS.
CONCLUSION: Our results demonstrated that patients with chronic rhinosinusitis are associated with higher risk of COVID-19 hospitalization, albeit no increased risk of mortality.

Introduction

Since the declaration of the coronavirus disease 2019 (COVID-19) pandemic caused by the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) in March 2020, the disease has spread rapidly throughout the world [1]. COVID-19 has a highly variable presentation from asymptomatic disease to severe acute respiratory distress and death. SARS-CoV-2 is a contagious virulent organism with an incubation period of 2 to 14 days with asymptomatic patients capable of transmitting the infection. The nasal and oral cavities are believed to be the most common portals of entry for the virus prior to its propagation into the lungs where the major comorbidities of the virus arise. Early reports regarding symptoms from China were mostly related to cough, fever, and fatigue, with symptoms of upper respiratory symptoms such as rhinorrhea, sore throat, and nasal congestion reported as very uncommon [2], [3], [4], [5], [6]. Early systematic reviews also reported symptoms of fever in 83.3%, cough 60.3%, and fatigue in 38% with no reports on impaired sense of smell or taste in these studies [6], [7]. It was not until April 23rd 2020 that the CDC (Centers for Disease Control and Prevention) has updated the symptoms to include “new loss of taste or smell” to the full list of COVID-19 symptoms [8]. Impaired sense of smell has been one of the early identified symptoms of COVID-19 as has been reported in multiple countries across the globe [9], [10], [11], [12], [13], [14], [15], [16], [17], [18], [19], [20], [21], [22], [23], [24], [25], [26], [27], [28], [29], [30], [31], [32]. There a wide variation in the reported loss of sense of smell with reported incidence as low as 5.1% and as large as 85.6% [33], [34]. This symptom is more commonly reported in mild to moderate COVID-19 and much less in severe COVID-19 [34].

In addition to the variable presentation of patients infected with COVID-19, one of the most important and puzzling issues is related to the ability to identify high risk patients that further go on and develop severe to critical COVID-19 requiring hospitalization, intensive care unit (ICU) admission, and mechanical ventilation. Determining inter-individual variation in disease severity has several implications including determining high risk patients and vaccine allocation. There are several known risk factors that have been linked to severe COVID-19 including advanced age and several comorbidities including chronic obstructive pulmonary disease (COPD), diabetes mellitus, hypertension, and obesity [35], [36]. Additionally, COPD was also found to be associated with increased health care utilization (hospitalization) but not in-hospital mortality [36]. Despite the clear evidence of viral replication and increased receptors in the nasal cavity, no studies have examined the role of upper airway comorbidities such as the presence or absence of chronic rhinosinusitis as it affects health care utilization and disease severity in patients with COVID-19.

Our study assesses the relationship between chronic rhinosinusitis and clinical outcomes with COVID-19 infection using the Cleveland Clinic COVID-19 registry.

Methods

Cleveland clinic registry

The Cleveland Clinic Foundation (CCF) COVID-19 registry was used in this study [37]. The registry collects information on symptomatic patients presenting for COVID-19 testing. The variables that are collected by a dedicated research team include demographics, laboratory testing for COVID-19, symptoms, comorbidities, hospitalization data, medications, and mortality. The details regarding the characteristics of the registry were published previously [38], [39], [40], [41], [42]. Data was extracted from EHR (EpicR, Epic Systems Corporation, Wisconsin, USA) using an experienced research team using a predefined process [43].

Following Cleveland Clinic's Institutional Review Board (IRB), data on patients' demographics, symptoms, COVID-19 status, comorbidities, and outcomes (hospitalization, ICU admission, and mortality) were extracted from the CCF COVID-19 registry. Data extraction was performed manually by a trained research team streamlining all research projects at CCF using the COVID-19 registry. Additional variables were manually collected from the patient's electronic health records (EHR) including data on the presence or absence of impaired sense of smell, loss of taste, chronic rhinosinusitis, and history of adenoidectomy or tonsillectomy. Following the diagnosis of COVID-19, patients were enrolled in a home monitoring program that involved reporting their symptoms over a 14-day period by a registered nurse or through an app.

Participants

Data was collected on all patients that were tested for COVID-19 at CCF from March 8, 2020 until May 15, 2020. The registry started collecting data on symptomatic patients starting March 8, 2020. After April 15th 2020, the registry stopped collecting data on COVID-19 negative patients. We excluded patients younger than 18 years old. For the analyses, we randomly collected patients from a total of 23,282 patients to include 1000 COVID-19 positive patients, and 1000 COVID-19 negative patients. A random number was assigned to each patient in Excel using the RAND function. Rows were then sorted in ascending order of the number previously generated, and the first 1000 patients were selected for inclusion. This process was repeated for adults who tested negative for COVID-19. Data collected included demographics, COVID-19 testing results, symptoms, comorbidities, health care utilization, imaging, medications, ICU admission, mechanical ventilation, and mortality.

Statistical analysis

Descriptive statistics were presented as numbers and percentage for categorical variables with mean and standard deviation for continuous variables. Chi-square test was applied for group comparisons on categorical data, and the Fisher exact test was applied for data with an expected value <5. Group comparisons on continuous data, the Wilcoxon 2-sample test were used. Multivariate logistic regression models were applied for patient symptoms, demographic characteristics, and comorbidities that were significant from our bivariate analysis. Statistical analysis was performed using SAS JMP Pro statistical software (version 15.1; SAS Institute Inc., Cary, North Carolina).

Results

Demographics

Patient demographic information (age, gender, race), symptoms, smoking history, and comorbidities are reported in Table 1 . Patients with unmatching medical record numbers were excluded from the analysis with a remaining 996 COVID-19 negative and 998 COVID-19 positive patients included in the analysis. Overall, the demographics varied considerably between the COVID-19 positive and negative groups. The COVID-19 positive group was significantly older (mean 51.69 ± 0.59 vs 54.50 ± 0.6, p = 0.0009), and had a significantly greater percentage of Black patients (27.35% vs 19.26%, p < 0.0001). The COVID-19 negative group had significantly more smokers and comorbidities such as asthma, COPD, coronary artery disease, and heart failure.

Table 1

Demographics and characteristics of patients undergoing SARS-CoV-2 testing.

	COVID-19 negative	COVID-19 positive	P-value
N	996	998
Demographics
Age (in years) (SD)	51.69 ± 0.59	54.50 ± 0.60	0.0009
Female sex (%)	611 (61.28%)	519 (52.00%)	<0.0001
BMI (kg/m2) (SD)	30.51 ± 0.35	29.78 ± 0.44	0.57
Race (%)
White	661 (66.30%)	575 (57.62%)	<0.0001
Black	192 (19.26%)	273 (27.35%)	<0.0001
Symptoms
Cough	594 (72.26%)	714 (83.22%)	<0.0001
Fever	367 (49.33%)	529 (62.53%)	<0.0001
Fatigue	230 (43.15%)	352 (53.82%)	0.0003
Shortness of breath	376 (54.18%)	392 (48.10%)	0.0185
Diarrhea	133 (23.09%)	203 (28.67%)	0.0236
Loss of appetite	62 (13.87%)	186 (30.94%)	<0.0001
Vomiting	56 (10.24%)	47 (6.83%)	0.0379
Impaired sense of smell	18 (1.81%)	186 (18.69%)	<0.0001
Sore throat	262 (26.36%)	225 (22.59%)	0.0537
Nasal congestion	238 (23.94%)	247 (24.80%)	0.68
Rhinorrhea	172 (17.30%)	145 (14.56%)	0.098
Loss of taste	22 (2.21%)	215 (21.59%)	<0.0001
Comorbidities
Smoking	420 (47.89%)	326 (40.70%)	0.0032
COPD (emphysema)	84 (11.41%)	58 (7.31%)	0.0062
Asthma	221 (28.44%)	147 (18.19%)	<0.0001
Diabetes	202(25.96%)	206 (25.46%)	0.863
Hypertension	416 (51.23%)	429 (52.00%)	0.7668
Coronary artery disease	138 (18.25%)	112 (14.11%)	0.0265
Heart failure	115 (15.33%)	96 (12.09%)	0.0644
Cancer	149 (18.60%)	117 (14.18%)	0.0188
Immunosuppressive treatment	115 (14.67%)	86 (10.45%)	0.0127
Chronic rhinosinusitis	70 (7.05%)	52 (5.22%)	0.0932
Sinus surgery	17 (1.71%)	17 (1.71%)	1

Bolded text signifies P-values that are statistically significant.

Symptoms

Compared to the COVID-19 negative group, the COVID-19 positive group reported significantly more symptoms of cough (83.22% vs 72.26%, p < 0.0001), fever (62.53% vs 49.33%, p < 0.0001), fatigue (53.82% vs 43.15%, p < 0.0001), loss of appetite (30.94% vs 13.87%, p < 0.0001), loss of taste (21.59% vs 2.21%, p < 0.0001), impaired sense of smell (18.69% vs 1.81%, p < 0.0001), and diarrhea (28.67% vs 23.09%, p = 0.024). On the other hand, COVID-19 patients reported significantly more vomiting (10.24% vs 6.83%, p = 0.38), and shortness of breath (54.18% vs 48.10%, p = 0.019), compared to COVID-19 positive group. The difference in symptoms of sore throat, nasal congestion, and rhinorrhea did not reach statistical significance between the two groups. Among COVID-19 positive patients, the most common symptoms experienced were cough (83.22%), fever (62.53%), fatigue (53.82%), and shortness of breath (48.10%).

The most common reported comorbidities in the COVID-19 positive group included hypertension (52.00%), diabetes (25.46%), and asthma (18.19%). Chronic rhinosinusitis (CRS) was noted in 5.22% of COVID-19 positive patients. The demographic and clinical characteristics of COVID-19 positive patients with and without CRS are included in Table 2 . There was no statistically significant difference in age and race between the COVID-19 positive patients with and without CRS. In terms of symptoms, COVID-19 positive patients with CRS experienced more rhinorrhea (25.5% vs 14.1%, p = 0.041) compared to patients without CRS. The difference in other symptoms between the COVID-19 positive patients with CRS and without CRS did not reach statistical significance. COVID-19 positive patients with CRS were at higher risk for hospitalization compared to patients without CRS (38.46% vs 25.82%, p = 0.036). However, there was no significant difference between the two groups in relation to ICU admission, mechanical ventilation, and death (Table 2).

Table 2

Demographics and characteristics of COVID-positive patients with and without chronic rhinosinusitis.

	No chronic rhinosinusitis	Chronic rhinosinusitis	P-value
N	942	52
Demographics
Age (in years) (SD)	54.36 ± 0.62	58.14 ± 2.27	0.107
Female sex (%)	481 (51.06%)	35 (67.31%)	0.084
BMI	29.86 ± 0.46	28.74 ± 1.81	0.558
Race (%)
White	542 (57.54%)	32 (61.54%)	0.665
Black	256 (27.18%)	16 (30.77%)	0.631
Symptoms
Cough	673 (83.50%)	38 (80.85%)	0.686
Fever	499 (62.77%)	19 (41.30%)	0.639
Fatigue	330 (53.66%)	20 (55.56%)	0.865
Shortness of breath	370 (48.30%)	20 (44.44%)	0.647
Diarrhea	192 (28.96%)	11 (26.19%)	0.861
Loss of appetite	177 (31.49%)	8 (24.24%)	0.382
Vomiting	45 (6.97%)	2 (5.13%)	1
Sore throat	211 (22.40%)	13 (25.00%)	0.613
Impaired sense of smell	171 (18.17%)	15 (28.85%)	0.055
Nasal congestion	229 (24.31%)	18 (34.62%)	0.100
Rhinorrhea	132 (14.01%)	13 (25.00%)	0.041
Loss of taste	198 (21.02%)	17 (32.69%)	0.046
Comorbidities
Smoking	299 (40.08%)	25 (50.00%)	0.1822
COPD (emphysema)	49 (6.63%)	9 (17.65%)	0.0087
Asthma	130 (17.22%)	17 (33.33%)	0.00076
Diabetes	191 (25.23%)	15 (30.00%)	0.503
Hypertension	399 (51.68%)	30 (58.82%)	0.386
Coronary artery disease	100 (13.48%)	12 (24.00%)	0.056
Heart failure	89 (11.99%)	7 (14.00%)	0.654
Cancer	110 (14.21%)	7 (14.29%)	1
Immunosuppressive treatment	70 (9.07%)	16 (32.65%)	<0.0001
Sinus surgery	4 (0.42%)	14 (25.49%)	<0.0001
Outcome
Hospitalized	237 (25.82%)	20 (38.46%)	0.0355
Mechanical ventilation	53 (50.48%)	4 (28.57%)	0.1587
ICU admission	84 (35.59%)	4 (20.00%)	0.22
Death	24 (4.60%)	1 (2.94%)	1

Bolded text signifies P-values that are statistically significant.

The demographic and clinical characteristics of COVID-19 positive patients with and without impaired sense of smell are included in Table 3 . The group that experienced impaired sense of smell was significantly younger than the group that did not experience impaired sense of smell (45.0 ± 1.07 vs 56.8 ± 0.68, p < 0.0001). There was no significant difference between the two groups in race. In terms of symptoms, patients that had impaired sense of smell reported significantly more upper respiratory symptoms of nasal congestion, rhinorrhea, and loss of taste (p < 0.0001). In addition, this subset of patients with impaired sense of smell had significantly lower comorbidities including hypertension, heart failure, and coronary artery disease (p < 0.0001). COVID-19 positive patients with impaired sense of smell were significantly less hospitalized (13.97% vs 29.29%, p < 0.0001), and there were no deaths in this group (0.00% vs 5.49%, p = 0.0223).

Table 3

Demographics and characteristics of patients with and without impaired sense of smell in COVID-19 positive patients.

	No impaired sense of smell	Impaired sense of smell	P-value
N	807	186
Demographics
Age (in years) (SD)	56.81 ± 0.68	45.02 ± 1.07	<0.0001
Female sex (%)	388 (48.08%)	127 (68.28%)	<0.0001
BMI (kg/m2) (SD)	29.40 ± 0.49	31.33 ± 1.07	0.096
Race (%)
White	469 (58.12%)	104 (55.91%)	0.62
Black	211 (26.15%)	61 (32.80%))	0.069
Symptoms
Cough	560 (83.04%)	150 (84.27%)	0.82
Fever	423 (63.61%)	102 (59.29%)	0.22
Fatigue	278 (53.88%)	72 (53.33%)	0.92
Shortness of breath	327 (51.01%)	63 (37.28%)	0.0018
Diarrhea	150 (27.42%)	53 (33.76%)	0.13
Loss of appetite	140 (30.30%)	45 (33.83%)	0.46
Vomiting	35 (6.57%)	12 (7.95%)	0.58
Sore throat	169 (20.94%)	55 (29.57%)	0.015
Nasal congestion	170 (21.01%)	76 (40.86%)	<0.0001
Rhinorrhea	98 (12.11%)	47 (25.27%)	<0.0001
Loss of taste	49 (6.06%)	166 (89.25%)	<0.0001
Comorbidities
Smoking	265 (41.47%)	61 (38.61%)	0.53
COPD (emphysema)	55 (8.63%)	3 (1.97%)	0.0029
Asthma	111 (17.16%)	36 (22.78%)	0.11
Diabetes	181 (27.76%)	25 (16.23%)	0.0028
Hypertension	377 (56.78%)	52 (32.91%)	<0.0001
Coronary artery disease	108 (16.90%)	4 (2.63%)	<0.0001
Heart failure	95 (14.87%)	1 (0.66%)	<0.0001
Cancer	104 (15.73%)	13 (8.07%)	0.012
Immunosuppressive treatment	71 (10.71%)	15 (9.55%)	0.77
Chronic sinusitis	37 (4.58%)	15 (8.06%)	0.066
Sinus surgery	15 (1.86%)	2 (1.08%)	0.75
Outcome
Hospitalized	232 (29.37%)	25 (13.97%)	<0.0001
Mechanical ventilation	53 (49.07%)	4 (36.36%)	0.53
ICU admission	80 (34.48%)	8 (32.00%)	1
Death	25 (5.49%)	0 (0.00%)	0.022

Bolded text signifies P-values that are statistically significant.

Regarding need for hospitalization in the COVID-19 positive group (Table 4 ), patients who were hospitalized reported significantly more shortness of breath (71.68% vs 38.70%, p < 0.0001) and fatigue (67.01% vs 47.97%, p < 0.0001). COVID-19 positive patients who did not require hospitalization reported significantly more upper respiratory symptoms including impaired sense of smell (21.57% vs 9.73%, p < 0.0001), sore throat (24.48% vs 17.90%, p = 0.0304), and loss of taste (23.08% vs 16.73%, p = 0.0336). In terms of comorbidities, hospitalized COVID-19 positive patients were more likely to have hypertension (73.17% vs 43.01%, p < 0.0001), diabetes (42.45% vs 18.51%, p < 0.0001), heart failure (25.32% vs 6.45%, p < 0.0001), coronary artery disease (24.05% vs 9.76%, p < 0.0001), asthma (22.36% vs 16.25%, p = 0.0440), cancer (21.22% vs 11.17%, p = 0.0003), COPD (12.82% vs 4.77%, p = 0.0002), history of tonsillectomy and/or adenoidectomy (20.70% vs 14.31%, p = 0.0219), and chronic rhinosinusitis (7.81% vs 4.35%, p = 0.0486).

Table 4

Demographics and clinical characteristics of hospitalized and non-hospitalized COVID-19 positive patients.

	Not hospitalized	Hospitalized	P-value
N	715	257
Demographics
Age (in years) (SD)	50.67 ± 0.70	65.30 ± 0.96	<0.0001
Female sex (%)	435 (60.67%)	159 (61.39%)	0.39
BMI (kg/m2) (SD)	29.41 ± 0.55	30.41 ± 0.74	0.32
Race (%)
White	417 (58.32%)	146(56.81%)	0.71
Black	170 (23.78%)	95 (36.96%)	<0.0001
Symptoms
Cough	508 (84.53%)	186 (79.83%)	0.12
Fever	382 (65.08%)	135 (56.96%)	0.029
Fatigue	212 (47.86%)	131 (66.84%)	<0.0001
Shortness of breath	219 (38.49%)	162 (72.00%)	<0.0001
Diarrhea	132 (27.39%)	66 (31.88%)	0.23
Loss of appetite	120 (28.85%)	63 (37.95%)	0.037
Vomiting	27 (5.78%)	18 (8.87%)	0.18
Impaired sense of smell	154 (21.63%)	25 (9.73%)	<0.0001
Sore throat	174 (24.40%)	46 (22.68%)	0.037
Nasal congestion	186 (26.09%)	55 (21.40%)	0.15
Rhinorrhea	112 (15.71%)	31 (12.06%)	0.18
Loss of taste	165 (23.08%)	43 (16.73%)	0.0334
Comorbidities
Smoking	194 (36.06%)	123 (50.00%)	0.0001
COPD (emphysema)	26 (4.77%)	30 (12.88%)	0.0002
Asthma	90 (16.25%)	53 (22.46%)	0.043
Diabetes	102 (18.51%)	104 (42.39%)	<0.0001
Hypertension	243 (43.01%)	179 (73.36%)	<0.0001
Coronary artery disease	53 (9.76%)	57 (24.42%)	<0.0001
Heart Failure	35 (6.45%)	60 (25.32%)	<0.0001
Cancer	63 (11.17%)	52 (21.32%)	0.0003
Immunosuppressive treatment	53 (9.36%)	32 (13.33%)	0.10
Chronic rhinosinusitis	32 (4.49%)	20 (7.78%)	0.052
Sinus surgery	9 (1.26%)	8 (3.11%)	0.091

Bolded text signifies P-values that are statistically significant.

We also performed a multivariate logistic regression analysis predicting need for hospitalization adjusting for covariates that were shown to significantly correlate with impaired sense of smell including age, upper respiratory symptoms, lower respiratory symptoms, and comorbidities (Table 5 ). When adjusting for covariates, patients that reported shortness of breath were 4.34 (OR = 4.34, 95%CI (2.45–7.70), p < 0.0001) times more likely to be hospitalized compared to patients that had no shortness of breath. Patients with chronic rhinosinusitis were 3.46 (OR = 3.46, 95% CI (1.12–10.68)) times more likely to be hospitalized compared to patients that have no chronic rhinosinusitis. Patients that had impaired sense of smell were 0.28 (95% CI, 0.09–0.86) times less likely to be hospitalized compared to patients that did not have impaired sense of smell.

Table 5

Multivariate analysis of the factors predicting hospitalization.

Variable	Odds ratio	95% confidence interval		P-value
Age	0.95	0.93	0.97	<0.0001
Race: Black	2.33	1.23	4.41	0.0092
Fever	0.69	0.38	1.27	0.24
Fatigue	2.72	1.45	5.09	0.0018
Sore throat	0.57	0.29	1.09	0.090
Shortness of breath	4.34	2.45	7.70	<0.0001
Impaired sense of smell	0.28	0.09	0.86	0.026
Loss of taste	0.45	0.16	1.27	0.13
Loss of appetite	1.18	0.61	2.26	0.62
Smoking	1.30	0.74	2.29	0.36
COPD/emphysema	0.70	0.20	2.45	0.58
Asthma	1.07	0.53	2.13	0.85
Diabetes	2.00	1.05	3.82	0.034
Heart failure	4.57	1.53	13.60	0.0064
Chronic rhinosinusitis	3.46	1.12	10.68	0.030
Cancer	1.11	0.49	2.50	0.80
Hypertension	1.15	0.60	2.22	0.67
Coronary artery disease	1.56	0.63	3.89	0.34

Bolded text signifies P-values that are statistically significant.

Discussion

Using our CCF COVID-19 registry, we found that chronic rhinosinusitis (CRS) patients have an increased risk for hospitalization when compared to patients without CRS. Since the nasal and oral cavities are gateways for entry of the virus into the body, it is not surprising that upper airway pathologies may play a role in COVID-19 clinical outcomes. There are seven coronaviruses that cause infections in humans apart from SARS-CoV-2 [44]. Research has shown that the primary site of entry of coronaviruses depend on the type, with MERS-CoV infecting type II pneumocytes, SARS-CoV involving the lungs and SARS-CoV-2 infecting the nose and throat [45]. It is thought that goblet and ciliated cells are the targets that the virus infects [46] with nasal epithelial cells displaying the highest expression for ACE-2 receptors in the respiratory tree. The mechanism of entry into the host is through the interaction between its spike protein S1 and the host angiotensin-converting enzyme 2 (ACE2) receptor [47]. Transmembrane protease serine 2 (TMPRSS2) which is expressed in the host cells also plays a pivotal role in pathogenesis as it cleaves the viral S glycoprotein causing activation of viral particles [48]. In addition to nasal epithelial cells, recent studies have shown that ACE2 and TMPRSS2 are expressed by sustentacular and basal cells of the olfactory epithelium and are not expressed by olfactory sensory neurons [49] suggesting that olfactory impairment is related to the damage of the supporting cells rather than neurons.

CRS is among the most common chronic diseases in the world, affecting about 15% of the U.S. population [50]. The formal criteria for diagnosis includes the presence of 2 or more of the following cardinal symptoms for 12 weeks: nasal obstruction, nasal discharge, facial pressure or hyposmia; and objective evidence of inflammation/purulence from the paranasal sinuses on computed tomography (CT) or nasal endoscopy [51]. All patients diagnosed with CRS in this study were diagnosed by an otolaryngologist adhering to the aforementioned diagnostic criteria. Histologically, CRS differs based on whether the inflammation is eosinophilic versus neutrophilic [52]. In terms of symptoms, COVID-19 positive CRS patients experienced more rhinorrhea (25.5% vs 14.01%, p = 0.041) compared to patients without CRS. Interestingly, impaired sense of smell as a major cardinal symptom of CRS was not found to be differentially different between CRS and non CRS COVID positive patients.

The increased risk for hospitalization in CRS patients with COVID-19 was not associated with increased ICU admission risk, mechanical ventilation, or mortality. The risk of hospitalization associated with CRS was demonstrated even when controlling for known comorbidities. Our multivariate logistic regression model predicting need for hospitalization showed that CRS patients are 3.46 (95% CI, 1.11–10.68) times more likely to be hospitalized compared to patients that have no CRS. Studies on the link of CRS to COVID-19 severity are limited and the findings have been mixed in nature. Wang et al. reported that CRS was not associated with severe COVID-19 infection, but their study excluded deceased patients with CRS that may have represented patients with severe COVID-19 infection [53]. Conversely, Lee et al. performed a national study in South Korea which demonstrated that CRS increased the risk of COVID-19 infection and severity of illness [54]. Patients with chronic sinusitis without nasal polyps were noted to be at increased risk [54].

Our observed increased risk of COVID-19 hospitalization could be explained by the variation in receptor densities in the nasal epithelium of CRS patients or the use of steroids as part of the standard treatment of this disease. The first line treatment of CRS is intranasal corticosteroids; however, it is unlikely that the use of intranasal corticosteroids affected the outcome of our COVID-19 patients. In a study on COVID-19 positive patients with COPD, inhaled corticosteroid use was found to be protective against hospitalization of COVID-19 infections [36]. Our observed worse outcomes in CRS patients are more likely to be explained by potential differences in ACE2 receptor densities in CRS patients. Several studies have found that patients with non-eosinophilic CRS had increased ACE2 and TMPRSS2 expression compared to eosinophilic CRS patients [55], [56]. Another study looked at the effect of eosinophilic vs neutrophilic inflammatory histopathological profile on COVID-19 in asthmatics. Patients with the neutrophilic asthma phenotype were found to be at higher risk for COVID-19 infection compared to the eosinophilic counterpart [57]. Due to the nature of our study, we were not able to determine the histopathological phenotype of our CRS patients and future studies are needed to determine the effect of the CRS inflammatory profile on COVID-19 infectivity and severity.

Our study has several limitations. First, it was conducted at the beginning of the COVID-19 pandemic in the U.S. There was less awareness for testing at that time particularly among the younger patients which may present milder forms of COVD-19. Hence, we believe that impaired sense of smell is underreported in our study. Second, we were not able to divide CRS into eosinophilic vs neutrophilic endotypes. Third, our study was retrospective in nature and we were not able to collect data on the use of intranasal steroids given their over the counter use. Fourth, our study included Cleveland Clinic patients in Northeast Ohio and Southeast Florida population, which limits the generalizability of our results.

Conclusion

Chronic rhinosinusitis is associated with increased risk of hospitalization due to COVID-19. Future studies are needed to examine the mechanistic bases and examine the relationship in a prospective fashion.

Author contributions

Firas Sbeih: Data collection, analysis, and initial draft.

Jorge Gutierrez: Data collection, analysis and initial draft.

George Saieed: Data collection, analysis, and initial draft.

Mohamad R. Chaaban: Design, analysis, initial draft and revise manuscript.

Declaration of competing interest

Dr. Mohamad Chaaban is a member of the Optinose advisory board. All other authors have no financial or other conflicts of interest to disclose.