Neurosymptoms of COVID-19: results of cerebrospinal fluid and blood biomarkers and assessment of diagnostic efficacy of risk factors

Background Novel coronavirus disease (COVID-19) patients can exhibit acute neurosymptoms when infected with severe acute respiratory syndrome coronavirus 2 (SAS-COV-2), but he mechanism for the occurrence of neurosymptoms in COVID-19 patients are not yet clear. The purpose of this study was to identify potential causes of neurosymptoms in COVID-19 patients by exploring the characteristics of cerebrospinal fluid (CSF) and peripheral blood in COVID-19 patients with neurosymptoms. Methods A study was conducted on 40 COVID-19 patients with neurosymptoms (categorized into subgroups of encephalopathy, encephalitis, and other neurosymptoms). CSF biomarkers and serum cytokines were compared between neurosymptom subgroups and COVID-19-negative control group. Blood biomarkers were compared between neurosymptom subgroups and COVID-19-positive control group. Logistic regression analysis and receiver operating characteristic (ROC) analysis were used to detect the risk factors and evaluate the diagnostic performance of risk factors for neurosymptoms in COVID-19 patients. Results Compared with COVID-19-negative control, encephalopathy subgroup had significantly higher values of CSF to serum albumin ratio (QAlb) and CSF interleukin-6 (IL-6）（all P <0.05), encephalitis subgroup had significantly higher values of CSF total protein (TP), CSF albumin (Alb), QAlb, CSF white blood cell (WBC) count, and CSF IL-6 (all P <0.05) , other neurosymptom subgroup had significantly higher CSF TP ( P <0.05). In addition, serum IL-6 in all subgroups were higher than COVID-19-negative control ( P <0.05). Compared with COVID-19-positive control, all subgroups had significantly lower serum immunoglobulin G (IgG) levels ( P <0.05), significantly higher serum complement C3 (C3) levels ( P <0.05), and no differences in serum IL-6 concentrations were found between all subgroups and COVID-19-positive control ( P >0.05).Logistic regression analysis showed the levels of serum IgG and C3 might be risk factors for neurosymptoms in COVID-19 patients. The area under the curve (AUC) of serum IgG was 0.832 (95% CI 0.727~0.909， P <0.0001), with sensitivity of 80.00%, and specificity of 73.53%. The AUC of serum C3 was 0.768 (95% CI 0.655~0.858， P <0.0001), with sensitivity of 70.00%, and specificity 76.47%. Conclusion Immunological imbalance with decreased IgG levels and increased C3 levels in circulation may be key factors in the occurrence of neurosymptoms in COVID-19 patients.