key: cord-0804771-vuvtt13q authors: Gelisse, Philippe; Rossetti, Andrea O.; Genton, Pierre; Crespel, Arielle; Kaplan, Peter W. title: How to carry out and interpret EEG recordings in COVID-19 patients in ICU? date: 2020-05-13 journal: Clin Neurophysiol DOI: 10.1016/j.clinph.2020.05.006 sha: 00f9ecf51476d19e638c118af55e0ef3614f5923 doc_id: 804771 cord_uid: vuvtt13q Abstract There are questions and challenges regarding neurologic complications in COVID-19 patients. EEG is a safe and efficient tool for the evaluation of brain function, even in the context of COVID-19. However, EEG technologists should not be put in danger if obtaining an EEG does not significantly advance diagnosis or change management in the patient. Not every neurologic problem stems from a primary brain injury: confusion, impaired consciousness that evolves to stupor and coma, and headaches are frequent in hypercapnic/hypoxic encephalopathies. In patients with chronic pulmonary disorders, acute symptomatic seizures have been reported in acute respiratory failure in 6%. The clinician should be aware of the various EEG patterns in hypercapnic/hypoxic and anoxic (post-cardiac arrest syndrome) encephalopathies as well as encephalitides. In this emerging pandemic of infectious disease, single-use subdermal EEG needle electrodes may be used in comatose patients. Full EEG montages using the 10-20 system, including a derivation for ECG, are strongly recommended so as to cover relevant temporal lobe regions. We recommend noting whether the patient is undergoing or has been placed prone, as well as noting the body and head position during the EEG recording (supine versus prone) to avoid overinterpretation of respiratory, head movement, electrode, muscle or other artifacts. There is slight elevation of intracranial pressure in the prone position. In non-comatose patients, the hyperventilation procedure should be avoided. At present, non-specific EEG findings and abnormalities should not be considered as being specific for COVID-19 related encephalopathy. 1. Subdermal single-use EEG needle electrodes can be used in comatose patients. 2. Full montages according to the 10-20 system with ECG derivations are strongly recommended. 3. Note if the patient undergoes prone positioning, note the patient and head position, along with patterns of breathing. EEG provides a readily available, safe, and effective means of evaluating brain function. Despite recent advances in neuroimaging, EEG remains essential in monitoring in neurology, especially in acute neurological conditions seen in the intensive care unit (ICU). With the current COVID-19 pandemic, there is increasing interest in the neurologic complications in these patients. Anosmia and dysgeusia, recently described (Lechien et al., 2020; Xydakis et al., 2020) , suggest the invasion of the olfactory nerve and the consequent risk of spread throughout the brain. Several human respiratory viruses are neurotropic and neuroinvasive, and it has been reported that the coronavirus family have a tropism for the CNS (Desforgues et al., 2019) . For COVID-19 patients, considering the current and expected evolution of the epidemicpandemic, updated and relevant appropriate practical guidelines for EEG recordings are warranted. 2.1. Cup or disk electrodes are commonly used in ICU for continuous EEG monitoring, but with infected comatose patients single-use subdermal EEG needle electrodes should be considered. These pose no problem with decontamination, and they do not require scalp abrasion. Above all, they can be applied rapidly (Caricato, 2018; Hopp and Atallah, 2018) . Disadvantages include higher impedance (Hopp and Atallah, 2018) and cost. 2.2. For the initial EEG, we recommend using of 19 scalp electrodes according to the 10-20 system, and a channel for ECG. Three temporal electrodes over each hemisphere are essential for evaluation of the temporal lobe and the inferior part of the frontal lobe. Reduced montages may result in difficulties in identifying artifacts, and especially in the failure to recognize lateralized periodic discharges (LPDs) of the temporal lobe or the inferior part of the frontal lobe. In cases of suspected encephalitis, assuming the involvement of the olfactory nerves, the site of predilection for LPDs is likely to be the temporal lobes, as in herpes simplex virus (HSV), emphasized in viral encephalitis 30 years ago by Giroud et al. (1990) (who proposed using sphenoidal electrodes to record deep periodic discharges, not seen with surface electrodes, to support typical findngs of HSV encephalitis, and to follow its evolution). Invasive electrodes seem no longer warranted in an ICU setting; additional temporal lobe lectrodes electrodes may be considered (Seeck et al., 2017) . 2.3. The EEG technologist should note the position of the patient's head and body, relaxation of the head and neck muscles and the possible sources of extraneous electrical and medical device artifact. Concomitant video-recording is strongly recommended (see below). A prone (face-down) position is used for the treatment of acute respiratory distress syndrome to improve oxygenation with an average number of sessions of 4 (Guérin et al., 2013) , and patients may stay on their stomachs for 16 hours a day (WHOa, 2020). This procedure may require an increase in muscle relaxants or in sedation (Roche-Campo et al., 2011) , which could result in a more prolonged ICU stay. The prone position is rarely encountered by electroencephalographers (EEGer), and may result in difficulties in recognizing artifacts commonly seen in the posterior electrodes: in the supine position, occipital electrodes are particularly vulnerable to numerous artifacts, such as contact with the bed/pillow. When prone, these artifacts would be seen in the frontopolar leads that are in contact with the bed. Prone positioning effects on the EEG in adults for periods of 16 hours per day are unknown. There is a moderate elevation of intracranial pressure during prone positioning (Roth et al., 2014) , which is contraindicated in cases of raised intracranial pressure >30 mmHg or cerebral perfusion pressure <60 mmHg (Malhotra and Kacmarek, 2020) . During mechanical ventilation, respiration artifacts are common (Tatum, 2018) . Ventilatortriggered artifacts consist of periodic activities typically at about 20 breaths per minute, at regular intervals. In COVID-19 patients, the respiratory rate may be higher (WHOb, 2020), and should be checked by the EEG technologist and noted in the report. Irregular breath intervals are possible when the patient is not on the ventilator (Rampal et al., 2018) . These slight head movements produce waveforms that are seen over the posterior head regions (Tatum, 2018) , but periodic frontally dominant waveforms may be seen (Rampal et al., 2018) . Waveforms can be asymmetric. Water condensation within the tubing connected to the ventilator may produce artifacts that simulate intermittent polysharp-waves (Tatum et al., 2011) , and may appear to be lateralized if the ventilator connections lie in close proximity to the anterior head regions, mimicking LPDs (Tatum et al., 2011) . Yoo et al. reported one patient with irregular bursts of sharp activity time-locked to a gurgling sound independent of the ventilator rate due to movement of fluids within the upper respiratory tracts and/or the tube (Yoo et al., 2014) . Ballistocardiographic artifacts are more unusual. These result from movements of the head/body synchronous to cardiac activity (pulsatile force on the aortic arch) (Stern, 2015) . As it is a result of a slight movement of the head, this peculiar "pulse" artifact is also mainly seen on the posterior electrodes in contact with the bed in the supine position. The ECG helps identify this artifact. There are also many environmental/electric artifacts in ICU, such as 50/60-Hz notch interferences. Prone positioning effects on the EEG in adults for periods of 16 hours per day are unknown. There is a moderate elevation of intracranial pressure during prone positioning (Roth et al., 2014) , which is is contraindicated in cases of raised intracranial pressure >30 mmHg or cerebral perfusion pressure <60 mmHg (Malhotra & Kacmarek, 2020) . 2.4. EEG reactivity to auditory and nociceptive stimuli should be tested. In comatose patients, the absence of EEG reactivity is usually associated with poor outcome (Rossetti et al., 2017; André-Obadia et al., 2018) , and this probably also applies to comatose COVID-19 patients, especially if these are not deeply sedated. In cases of confusional states versus an isolated impairment of vigilance, reactivity can help differentiate metabolic/toxic/respiratory encephalopathies from nonconvulsive status epilepticus (Gélisse et al., 2019; Kaplan et al., 2020) (Table 1 ). In non-comatose patients, hyperventilation is not useful and should be not performed, as it may lead to cerebral vasoconstriction that can be potentially challenging for brain perfusion. 2.5. Simultaneous video recording is strongly advised. This is not a specific recommendation for COVID-19 patients, but video recording in this situation may help see the position of the body (supine versus prone position) and the position of the head (version to the right or to the left). Simultaneous video-EEG recording enables clinical correlations and can help to identify some artifacts with video and audio recordings (Yoo et al., 2014) . (Gélisse et al., 2019) . In advanced Alzheimer's disease, periodic complexes occasionally occur, but without anterior predominance (Gélisse et al., 2019) . Furthermore, elderly people also have increasing impairment of renal function contributing to delayed drug excretion. There is a high prevalence of obesity in severe acute respiratory syndrome, such as COVID-19 (Simonnet et al., 2020) . Obesity in the critically ill increases the risk of complications in several organ systems (Schetz et al., 2019) . in ICU should be used (2012) (Hirsch et al., 2013) ; these have been validated and allow easy generalization of EEG findings . 3.5. Electroencephalographers should have a solid knowledge of EEG patterns seen in the ICU, e.g., anoxic (post-cardiac arrest syndrome), toxic, metabolic, sepsis, and hypercapnic/hypoxic encephalopathies (Kaplan and Rossetti, 2011; Bauer et al., 2013; Stern, 2015; Gélisse et al., 2019; Kaplan et al., 2020) . Etiologies may be mixed (Fig. 1 Infectious encephalitides may show LPDs with periodicity of less of four seconds (so called "short periods") (Gaches, 1971; Dunand and Jallon 2002) . These are suggestive of HSV encephalitis, but can also be seen in other viral etiologies, e.g., herpes virus-6 limbic encephalitis (Gélisse et al., 2019) . In HSV encephalitis, these LPDs are seen over the temporal lobes and can spread to the homolateral frontal lobe (Fig. 3) . In early stages, there is slow polymorphic unilateral delta activity (Dunand and Jallon, 2002) . Periodic activities appear early in the course of disease and typically disappear after 15 days, replaced by diffusely abnormal patterns, when pathology may reveal widespread necrosis (Gaches et al., 1978) . EEG is thus helpful in identifying patterns of periodic discharges, subclinical seizures, non-convulsive status epilepticus, focal slowing (e.g., with strokes), and delayed recovery from anesthesia (sleep patterns). In the context of acute meningo-encephalitis, when periodic EEG bursts are seen bilaterally, and/or symmetrically, a non-HSV infection may be suspected (Fig. 4) . Acute disseminated encephalomyelitis (ADEM) is an inflammatory disorder, typically occurring a few days after a viral or bacterial infection, and is more common in children. The EEG may show bilateral slow-wave activity, more or less symmetrically (Dale et al., 2000; Hollinger et al., 2002; Kaplan and Rossetti, 2011) , sometimes with focal slowing and less frequently, with epileptiform discharges (Dale et al., 2000; Hollinger et al., 2002) . In children, anterior, bilateral, high-voltage and sometimes asymmetric complexes have been reported (Fig. 4) . Not every neurologic problem stems from a primary brain injury. In elderly patients with pneumonia, confusion is common. An isolated alteration of vigilance with an increasinga progressive change in consciousness ranging from somnolence to coma evoke an encephalopathy, and, in the context of COVID-19, hypercapnic/hypoxic encephalopathy. Recent studies show that 8% to 15% of patients with this infection report headache (Helms et al., 2020; Mao et al., 2020) , and this symptom was added to the list of COVID-19 symptoms (Centers for Disease Control and Prevention, 2020). However, headache is common with fever and with other flu-like symptoms. As mentioned above, headache as a consequence of hypercapnia was frequently reported (1/3 of the patients) in acute respiratory failure in chronic pulmonary patients, and acute symptomatic seizures may occur (Laxenaire-Aug et al., 1970) . Although much information may be gleaned from EEG, the value of this information in the diagnosis and management of the patient must be weighed against the COVID-19 risks of infection to the technologist. EEG findings in COVID-19 are in their early days and care should be given not to generalize or "typify" any such patterns to COVID-19. These guidelines recommendations may help the EEG technologist, EEGer and clinician in the use of EEG in this new, challenging time. Lateralized periodic discharges over the left temporal region. B, Herpes simplex encephalitis: 35-year-old man. Unusual headache for ten days and low-grade fever. Polymorphic delta activity over the right temporal lobe that spreads to the right frontal lobe. There is no periodic activity, maybe because of the subclinical evolution. Insets show fluid attenuated inversion recovery MRIs of the patients with typical mesiotemporal T2w signal hyperintensity. 17-year-old woman. Fever. Generalized tonicclonic seizure followed by severe impairment of awareness requiring hospitalization in ICU. Numerous periodic diffuse complexes occurring every 3-4 seconds. B, Acute Disseminated Encephalomyelitis: 6-year-old boy. Status epilepticus five days after pneumonia. EEG, eight days after the status. Diagnosis of mycoplasma pneumonia on PCR. MRI showed hyperintensities of the caudate and lentiform nuclei bilaterally, especially on the left hemisphere with normalization one month later. Eyes closed. On the left occipital region, the background is abnormal with irregular delta waves. These waves did not react on eyes opening. Frontal complexes evocative of encephalitis can be seen (*) with higher amplitude on the right side. Absence status epilepticus versus metabolic/toxic encephalopathies. The five key questions. Is it a confusional state or only a problem of vigilance? Is there a fluctuation of symptoms or change in consciousness from somnolence to coma? Is the EEG activity rhythmic or periodic? Is the EEG activity dynamic, showing spatiotemporal evolution, or relatively monomorphic? Is the EEG reactive to stimuli, wakefulness, sleep, arousal, or antiseizure drugs*? *Only consider an IV benzodiazepine test positive if both EEG and consciousness normalize. Recommendations for the use of electroencephalography and evoked potentials in comatose patients Lombardia CARe Researchers. Out-of-Hospital Cardiac Arrest during the Covid-19 Outbreak in Italy EEG patterns in hypoxic encephalopathies (post-cardiac arrest syndrome): fluctuations, transitions, and reactions Standardized EEG analysis to reduce the uncertainty of outcome prognostication after cardiac arrest Continuous Electroencephalography Monitoring in Adults in the Intensive Care Unit COVID-19) and Cardiovascular Disease Acute disseminated encephalomyelitis, multiphasic disseminated encephalomyelitis and multiple sclerosis in children Human coronaviruses and other respiratory viruses: underestimated opportunistic pathogens of the central nervous system? Critical Care EEG Monitoring Research Consortium. Interrater agreement for Critical Care EEG Terminology Neurology and Critical Care. Atlas of Electroencephalography PROSEVA Study Group. Prone positioning in severe acute respiratory distress syndrome Neurologic Features in Severe SARS-CoV-2 Infection American Clinical Neurophysiology Society's Standardized Critical Care EEG Terminology: 2012 version Acute disseminated encephalomyelitis in adults: a reappraisal of clinical, CSF, EEG, and MRI findings Electrodes and Montages An Electroencephalographic Voyage in search of Triphasic Waves (TWs) -the Sirens and Corsairs on the Encephalopathy/EEG horizon: a survey of TWs EEG patterns and imaging correlations in encephalopathy: encephalopathy part II Olfactory and gustatory dysfunctions as a clinical presentation of mild-to-moderate forms of the coronavirus disease (COVID-19): a multicenter European study Prone ventilation for adult patients with acute respiratory distress syndrome Neurologic Manifestations of Hospitalized Patients With Coronavirus Disease Electroencephalography and Artifact in the Intensive Care Unit Prone positioning in acute respiratory distress syndrome (ARDS): when and how Electroencephalography Predicts Poor and Good Outcomes After Cardiac Arrest: A Two-Center Study Does prone positioning increase intracranial pressure? A retrospective analysis of patients with acute brain injury and acute respiratory failure Early electroencephalography for outcome prediction of postanoxic coma: A prospective cohort study Obesity in the critically ill: a narrative review The standardized EEG electrode array of the IFCN High prevalence of obesity in severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) requiring invasive mechanical ventilation Atlas of EEG patterns Significance of triphasic waves in patients with acute encephalopathy: a nine-year cohort study Artifact: recording EEG in special care units Adult Electroencephalography Artifact Hospitalized Patients With 2019 Novel Coronavirus-Infected Pneumonia in Wuhan, China Clinical management of severe acute respiratory infection (SARI) when COVID-19 disease is suspected: Interim guidance V 1.2. WHO Reference Number COVID-19 v4. Operational Support & Logistics Disease Commodity Packages Smell and taste dysfunction in patients with COVID-19 Prevalence of comorbidities in the novel Wuhan coronavirus (COVID-19) infection: a systematic review and meta-analysis Respiratory artifact on EEG independent of the respirator