Non-Invasive Multimodal Neuromonitoring in Non-Critically Ill Hospitalized Adult Patients With COVID-19: A Systematic Review and Meta-Analysis

Abstract

IntroductionNeurological complications are frequent in patients with coronavirus disease-2019 (COVID-19). The use of non-invasive neuromonitoring in subjects without primary brain injury but with potential neurological derangement is gaining attention outside the intensive care unit (ICU). This systematic review and meta-analysis investigates the use of non-invasive multimodal neuromonitoring of the brain in non-critically ill patients with COVID-19 outside the ICU and quantifies the prevalence of abnormal neuromonitoring findings in this population.MethodsA structured literature search was performed in MEDLINE/PubMed, Scopus, Cochrane, and EMBASE to investigate the use of non-invasive neuromonitoring tools, including transcranial doppler (TCD); optic nerve sheath diameter (ONSD); near-infrared spectroscopy (NIRS); pupillometry; and electroencephalography (EEG) inpatients with COVID-19 outside the ICU. The proportion of non-ICU patients with CVOID-19 and a particular neurological feature at neuromonitoring at the study time was defined as prevalence.ResultsA total of 6,593 records were identified through literature searching. Twenty-one studies were finally selected, comprising 368 non-ICU patients, of whom 97 were considered for the prevalence of meta-analysis. The pooled prevalence of electroencephalographic seizures, periodic and rhythmic patterns, slow background abnormalities, and abnormal background on EEG was.17 (95% CI 0.04–0.29), 0.42 (95% CI 0.01–0.82), 0.92 (95% CI 0.83–1.01), and.95 (95% CI 0.088–1.09), respectively. No studies investigating NIRS and ONSD outside the ICU were found. The pooled prevalence for abnormal neuromonitoring findings detected using the TCD and pupillometry were incomputable due to insufficient data.ConclusionsNeuromonitoring tools are non-invasive, less expensive, safe, and bedside available tools with a great potential for both diagnosis and monitoring of patients with COVID-19 at risk of brain derangements. However, extensive literature searching reveals that they are rarely used outside critical care settings.Systematic Review Registration:www.crd.york.ac.uk/prospero/display_record.php?RecordID=265617, identifier: CRD42021265617.

The value of EEG attenuation in the prediction of outcome in COVID-19 patients

“…In a recent meta-analysis, the pooled prevalence of slow background abnormalities (theta and delta) in patients outside intensive care units (ICU) was 0.92 (95%CI 0.83–1.01, I 2 = 68.81%). The pooled prevalence of abnormal background in non-ICU patients was 0.95 (95%CI 0.88–1.09, I 2 = 44.98%) [ 20 ]. Diffuse slowing is the most frequent pattern also among our patients.…”

“…The most common finding was the presence of abnormal background activity, followed by slow background, rhythmic and periodic discharges, and electroencephalographic seizures. This may be explained by various factors: (1) patients with COVID-19 might be at higher risk of hypoxic and metabolic changes responsible for encephalopathy; (2) after the virus enters the cells, a strong inflammatory response followed by cytokine storms may alter cerebral permeability and hemodynamic, thus favouring encephalopathy and multiple organ failure with potential for EEG alterations; (3) seizures, although a prevalence comparable to the non-COVID-19 population may be indicative of new neurological complications [ 20 ].…”

The importance of monitoring cerebral oxygenation in non brain injured patients

“…NIRS has been evaluated to assess cerebral perfusion and autoregulation after cardiac arrest and detect episodes of cerebral desaturation showing a correlation between its values and severity of illness and with variable association between rSO 2 value and outcome [41,42]. Similarly, NIRS has shown to be useful to assess episodes of cerebral desaturation in patients with acute distress respiratory syndrome and COVID-19 during respiratory manipulations and the use of respiratory rescue therapies [7,[43][44][45][46], as well as in septic patients where cerebral desaturations were found to be predictors of neurological sequelae [47,48].…”

“…Among these, severe respiratory failure with or without extracorporeal membrane oxygenation (ECMO) [3,4], trauma, cardiac arrest, liver failure, intra-arterial thrombolysis during endovascular treatment [5,6], and sepsis are among the conditions where despite the absence of a primary cerebral damage, neurological complications are common and can affect patients' outcome [1]. The benefits of non invasive methods include safety, availability, and the provision of repeatable continuous data at the bedside, therefore helping clinicians detecting deterioration in neurologic function and earlier intervention [7].…”

Editorial: Personalized therapy in ARDS

“…Sedatives and anesthetics may impact patients' spontaneous breathing and MV duration. During the pandemic, metanalyses suggested that neuromonitoring patients with C-ARDS can be useful to guide sedation and detect those at risk of neurological complications ( 7 , 8 ). Tobar et al in a randomized double-blind trial, found that the use of a multiparameter electroencephalogram (EEG) protocol to guide sedation in C-ARDS as compared to no-EEG monitoring did not increase 30-day ventilator-free days, but reduced propofol administration and deeper sedation 5 days after randomization.…”