Neurological manifestations in COVID-19 patients and their application in predicting fatal disease: A retrospective cohort study

Abstract

BackgroundTo explore the development of central nervous system (CNS) symptoms and clinical application in predicting the clinical outcomes of SARS-COV-2 patients.MethodsA retrospective cohort study was performed on the hospitalized patients with SARS-COV-2 recruited from four hospitals in Hubei Province, China from 18 January to 10 March 2020. The patients with CNS symptoms were determined. Data regarding clinical symptoms and laboratory tests were collected from medical records.ResultsOf 1268 patients studied, 162 (12.8%) had CNS symptoms, manifested as unconsciousness (71, 5.6%), coma (69, 5.4%), dysphoria (50, 3.9%), somnolence (34, 2.7%) and convulsion (3, 0.2%), which were observed at median of 14 (interquartile range 9–18) days after symptom onset and significantly associated with older age (OR = 5.71, 95% confidence interval [CI] 2.78–11.73), male (OR = 1.73, 95% CI 1.22–2.47) and preexisting hypertension (OR = 1.78, 95% CI 1.23–2.57). The presence of CNS symptoms could be predicted by abnormal laboratory tests across various clinical stages, including by lymphocyte counts of <0.93 × 109/L, LDH≥435 U/L and IL-6≥28.83 pg/L at 0–10 days post disease; by lymphocyte count<0.86 × 109/L, IL-2R ≥ 949 U/L, LDH≥382 U/L and WBC≥8.06 × 109/L at 11–20 days post disease. More patients with CNS symptoms developed fatal outcome compared with patients without CNS symptoms (HR = 33.96, 95% CI 20.87–55.16).ConclusionNeurological symptoms of COVID-19 were related to increased odds of developing poor prognosis and even fatal infection.