ABSTRACT
Objective: To obtain neuroimaging on patients with COVID-19 using a low-field, portable magnetic resonance imaging (MRI) device. Background: Neuroimaging is a key step in the clinical evaluation of brain injury. Conventional MRI systems operate at high-strength magnetic fields (1.5-3T) that require patient transportation to access-controlled environments. During the COVID-19 pandemic, critically ill patients have had limited neuroimaging due to infection control and safety concerns. We report neuroimaging in patients with severe COVID-19 using a portable MRI device. Design/Methods: A 64mT point-of-care (POC) MRI was used to acquire neuroimaging in Yale New Haven Hospital ICUs from April 2020 through August 2020. COVID-19 patients with neurological symptoms and no MRI contraindications were scanned. Exams were acquired using a standard 110V/15A power outlet. Hospital rooms included vital signs monitors, ventilators, dialysis machines, and intravenous infusion pumps. Images were acquired by trained research staff, without the need for an MRI technician. POC MRI exams were interpreted by two boardcertified physicians (one neuroradiologist and one neurologist). Results: POC MRI exams were obtained on 22 ICU COVID-19 patients (19% female, ages 42-74 years, 86% mechanically ventilated). Glasgow Coma Scale and Richmond Agitation-Sedation Scale at time of scan were 7±3 and-3±2, respectively. T1-weighted (T1W), T2-weighted (T2W), fluid-attenuated inversion recovery (FLAIR), and diffusion-weighted imaging (DWI) sequences were obtained for all patients. Axial scan times were 4:54 minutes, 7:03 minutes, 9:31 minutes, and 9:04 minutes, respectively. Examination time was 35:40 minutes. Abnormal neuroimaging findings were observed in 10 patients: Intracranial hemorrhage (n=2), cerebral infarction (n=4), diffuse cerebral edema (n=1), and leukoencephalopathy (n=3). The device did not interfere with ICU equipment, and no significant adverse events occurred. Conclusions: We report the acquisition of neuroimaging using a low-field, portable MRI at the bedside of patients with severe COVID-19. This approach may hold promise for bedside assessment of neurological injury in settings with imaging access constraints.
ABSTRACT
BACKGROUND: Angiotensin-converting enzyme 2 (ACE2), a receptor targeted by the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), is highly expressed in the nasal mucosa. Chronic rhinosinusitis (CRS) shows diverse endotypes and is aggravated by viral infection. Whether viral stimulation and CRS endotype influence ACE2 expression remains unclear. We investigated the expression of ACE2 and the transmembrane protease, serine 2 (TMPRSS2), which mediate the entry of SARS-CoV-2 into cells, and assessed polyinosinic:polycytidylic acid (poly[I:C])-induced changes based on CRS endotype. METHODOLOGY: ACE2 and TMPRSS2 expression was evaluated based on CRS phenotype, endotype, and tissue type. Correlations between ACE2/TMPRSS2 expression and inflammatory mediators in nasal polyps (NP) were examined. Air-liquid interface culture experiments were performed to assess the effects of major cytokines or poly(I:C) stimulation on ACE2/TMPRSS2 expression in primary epithelial cells from healthy nasal mucosa, eosinophilic NP (ENP), and non-eosinophilic NP (NENP). RESULTS: In primary nasal epithelial cells, interleukin (IL)-13 decreased ACE2 expression but increased TMPRSS2. Eosinophilic CRS showed lower ACE2 expression than non-eosinophilic CRS, regardless of CRS phenotype. CRS endotype was an independent factor associated with ACE2/TMPRSS2 expression in NP. Serum and tissue eosinophilic marker levels were inversely correlated with ACE2 expression, whereas tissue neutrophilic marker levels and ACE2 expression were positively correlated in NP. ACE2 expression was suppressed in ENP tissues; however, a combination of poly(I:C) and IL-13 induced ACE2/TMPRSS2 upregulation in ENP. CONCLUSIONS: ENP tissues have lower ACE2 expression than NENP; however, viral stimulation promotes ACE2/TMPRSS2 upregulation in ENP.