Sinus computed tomography findings in patients with COVID-19

ABSTRACT Objective: To analyze computed tomography scans of paranasal sinuses of a series of patients with coronavirus disease 2019, and correlate the findings with the disease. Methods: Computed tomography scans of 95 adult patients who underwent a polymerase chain reaction test for severe acute respiratory syndrome coronavirus 2 were analyzed. Clinical data were obtained from patients’ records and telephone calls. Paranasal sinus opacification was graded and compared according to severe acute respiratory syndrome coronavirus 2 positivity. Results: Of the patients 28 (29.5%) tested positive for severe acute respiratory syndrome coronavirus 2 (median age 52 [range 26-95] years) and 67 were negative (median age 50 [range 18-95] years). Mucosal thickening was present in 97.4% of maxillary sinuses, 80% of anterior ethmoid air cells, 75.3% of posterior ethmoid air cells, 74.7% of frontal sinuses, and 66.3% of sphenoid sinuses. Minimal or mild mucosal thickening (score 1)and normally aerated sinuses (score 0) corresponded to 71.4% and 21.3% of all paranasal sinuses, respectively. The mean score of each paranasal sinus among severe acute respiratory syndrome coronavirus 2 positive and negative patients was 0.85±0.27 and 0.87±0.38, respectively (p=0.74). Median paranasal sinus opacification score among severe acute respiratory syndrome coronavirus 2 positive patients was 9 (interquartile range 8-10) compared to 9 (interquartile range 5-10) in negative patients (p=0.89). There was no difference in mean score adjusted for age and sex. Nasal congestion was more frequent in severe acute respiratory syndrome coronavirus 2 positive than negative patients (p=0.05). Conclusion: Severe acute respiratory syndrome coronavirus 2 infection was associated with patient recall of nasal congestion, but showed no correlation with opacification of paranasal sinuses.


❚ INTRODUCTION
Coronavirus disease 2019 (COVID-19) caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) spread rapidly worldwide from its origin in China. Upper airway symptoms, such as nasal congestion and rhinorrhea, have been reported to be relatively rare when compared with pulmonary symptoms, (1)(2)(3) and few studies have addressed the effects of SARS-CoV-2 infection on the paranasal sinuses using computed tomography (CT) scans. (4)(5)(6) ❚ OBJECTIVE To analyze computed tomography scans of paranasal sinuses of a series of patients with coronavirus disease 2019, and correlate the findings with the disease.

❚ METHODS Patients
All patients undergoing paranasal sinus CT scans in our organization and who were tested for SARS-CoV-2 over an 11-week period, from March 14 to May 30, 2020, were eligible for inclusion. Emergency room, inpatients, and outpatients were included. Patient clinical and image data were reviewed retrospectively. All SARS-CoV-2 positive and a random sample of 30 SARS-CoV-2 negative patients were called by telephone. Those who responded were asked to recall whether they had anosmia, hyposmia and nasal congestion at the time of the CT scans. Severe acute respiratory syndrome coronavirus 2 positivity was determined from nasopharyngeal swabs collected for each patient and tested for SARS-CoV-2 using a realtime reverse transcription polymerase chain reaction (RT-PCR) assay.
The exclusion criteria were patients under 18 years of age; recent sinonasal surgery; sinonasal tumor; acute facial trauma.

Imaging analysis
All patients underwent paranasal sinus CT scans (16-to 640-slice CT scanners: Aquilion Prime™, Aquilion One™, Aquilion One Vision™, Toshiba Medical Systems Corporation, Ltda, Japan; PET-CT Discovery 600, GE Medical Systems, USA; and SOMATOM ® Definition, Siemens, Germany) with the following parameters: 35mA, 120kV, 0.5 to 0.6mm slice thickness. All CT images were independently analyzed by two head and neck radiologists, with 5 and 12 years of experience, who were blinded to RT-PCR results. Disagreements were resolved by consensus.

Statistical analysis
Frequencies and median scores in SARS-CoV-2 positive and negative patients were compared using χ 2 or Fisher's exact, Kruskal-Wallis, and Student's t tests (alpha <0.05), while differences in total opacification score and olfactory obstruction, adjusted for age and sex, were quantified by logistic and linear regression, respectively. This study was approved by the Ethics Committee of Hospital Israelita Albert Einstein (HIAE), CAAE: 31014720.9.0000.0071, number 4.002.511. The need for written patient consent was waived.
Few articles have addressed the effect of SARS-CoV-2 infection on the paranasal sinuses. Chung et al., (6) reported no signs of sinusitis on CT scans of six SARS-CoV-2 positive patients who had olfactory dysfunction. A recent article included 49 COVID-19 positive patients with anosmia, used the original Lund-Mackay scoring system, and showed partial opacification in less than 10% of maxillary, frontal and sphenoid sinuses, and the ethmoid air cells were normally aerated in all patients. (5) In a cohort of 16 COVID-19 positive patients with olfactory dysfunction, Lechien et al., (4) also found no significant mucosal thickening of paranasal sinuses, with a Lund-Mackay mean score of 0.8.
Our study included 95 COVID-19 positive and negative patients. We decided to use a modified Lund-Mackay scoring system to better distinguish mild mucosal thickening from almost completely opacified sinuses. Despite the use of a modified scoring system, our findings corroborated the findings of these previous studies. We found similar scores for both SARS-CoV-2 positive and negative patients, with scores zero and one being the most common, with a median score of nine for both SARS-CoV-2 positive and negative patients (out of a maximum score of 30; p=0.89). Indeed, scores of 13 or higher were more frequent in SARS-CoV-2 negative patients. Our findings show that SARS-CoV-2 infection is not associated with mucosal changes in the paranasal sinuses.
There is scarce evidence in the literature regarding the effects of common viral respiratory infections in the paranasal sinuses. Gwaltney et al., (7) found opacification varying from 87% in the maxillary sinuses to 39% in the sphenoid sinuses among patients with common cold. Alho, (8) using the Lund-Mckay scoring system, found CT mean scores from seven to ten among patients with fresh viral cold. Our results are similar to those obtained in these articles, both in SARS-CoV-2 positive and negative patients.
This was an observational study, therefore prone to selection bias if SARS-CoV-2 positive patients were more likely to undergo a CT scan for clinical reasons related to our outcomes. It also had a relatively small sample size which was determined by an 11-week period of recruitment rather than on any a priori knowledge of possible associations. Nasal congestion was assessed by asking participants to recall whether they had this symptom at the time of their CT scan, a method which is susceptible to recall bias. Clinical data such as the average duration of symptoms and the final diagnoses of patients who tested negative for SARS-CoV-2 infection were not available in most cases. Also, we did not extract follow-up information on the clinical course and eventual outcome of SARS-CoV-2 infection.
❚ CONCLUSION SARS-CoV-2 infection was associated with patient recall of nasal congestion, but showed no correlation with opacification of paranasal sinuses.