Advertisement

Processed EEG from depth of anaesthesia monitors and seizures: A scoping review

Open ArchivePublished:June 16, 2021DOI:https://doi.org/10.1016/j.seizure.2021.06.011

      Highlights

      • Processed EEG is a widely used monitoring adjunct in anaesthetics, that may have clinical utility in the detection of seizures.
      • Growing observational data shows it may be helpful in monitoring of burst suppression in refractory status epilepticus.
      • Variable patterns of responses of derived pEEG variables have been seen during seizures making clinical interpretation challenging.
      • Further randomised trials testing the use of pEEG in the detection of seizure and burst suppression are needed.

      Abstract

      Purpose

      Processed electroencephalogram (EEG) is used peri-operatively for monitoring depth of anaesthesia. Because these utilise EEG data, attempts have been made to investigate their use in diagnosing and monitoring seizures. This is important as formal EEG monitoring can be hard to obtain in many critical care environments. We undertook a scoping review of the evidence for using processed EEG (pEEG) from depth of anaesthesia monitors for this indication.

      Methods

      Medline, Psych INFO, and Embase were searched for peer-reviewed journals until 20 March 2021. Data and conclusions taken from the study of pEEG in both critical care and peri-operative settings have been included in a qualitative synthesis about the current evidence for the use of pEEG in the detection and monitoring of seizures.

      Results

      Searches yielded 8 observational studies, 1 randomised trial and 15 case reports in which the use of pEEG in critical care and peri-operative medicine was described. Most concerned the Bispectral Index (BIS) device. The majority of observational studies reported the use of BIS for optimisation of burst suppression in patients with refractory status-epilepticus (RSE), or in the comparison of pEEG data with conventional EEG during epileptic activity. Multiple case reports describe the application of pEEG in the presence of disorders of consciousness as a tool for detection of non-convulsive status-epilepticus, finding variable trends in the pEEG output.

      Conclusions

      Processed EEG may be helpful in monitoring pharmacologically induced burst suppression. Despite this, its use in the diagnosing or monitoring seizure activity is controversial and currently not evidenced, with numerous confounding variables that requires systematic assessment in future studies.

      Keywords

      Abbreviations:

      pEEG (Processed electroencephalogram), BIS (Bispectral Index), RSE (refractory status epilepticus), ECT (electro-convulsive therapy), NICE (National Institute for Care and Excellence), cEEG (continuous EEG), SR (Suppression Ratio)

      1. Introduction

      Electrographic monitoring is an important aspect of the management of epileptic seizures. This is particularly true in cases of status epilepticus, when continued seizure activity for more than five minutes, or recurrent events without recovery in between, are linked to a significant risk of cerebral damage [
      • Meldrum BS
      • Brierley JB.
      Prolonged epileptic seizures in primates: ischemic cell change and its relation to ictal physiological events.
      ,
      • Jenssen S
      • Gracely EJ
      • Sperling MR.
      How long do most seizures last? A systematic comparison of seizures recorded in the epilepsy monitoring unit.
      ], making the prevention of prolonged convulsion a priority. Those patients whose seizures are not terminated by initial therapies (benzodiazepines, rapid loading of anticonvulsant) require general anaesthesia and admission to critical care environment, and represent those most likely to suffer irreversible damage.
      Refractory status epilepticus occurs for a subset of patients who experience persistent seizure activity despite general anaesthesia. Advocated by NICE guidelines for UK practice in the monitoring of seizure activity [

      Overview | Epilepsies: diagnosis and management | Guidance | NICE n.d.

      ], EEGs are used in these cases to titrate sedating agents to the point of burst suppression (BS), an electrophysiological correlate with coma comprised of alternating isoelectric EEG periods followed by high voltage slow-waves [
      • Amzica F.
      What does burst suppression really mean?.
      ]. EEG requires specialist input to perform and interpret, and therefore is often limited to teaching hospitals with neuroscience departments in the UK, complicating the care of patients taken unwell away from these locations. A survey of continuous EEG availability within specialty ICUs world-wide, of which 75% were from the USA, found that rapid access and interpretation was available only in 32% [
      • Koffman L
      • Rincon F
      • Gomes J
      • Singh S
      • He Y
      • Ritzl E
      • et al.
      Continuous electroencephalographic monitoring in the intensive care unit: a cross-sectional study.
      ]. In cases of super-refractory status epilepticus (lasting greater than 24 hours), daily EEG is a minimum requirement and ideally continuous EEG monitoring is indicated [

      Overview | Epilepsies: diagnosis and management | Guidance | NICE n.d.

      ].

      2. Processed EEG and depth of anaesthesia monitors

      Processed EEG (pEEG) emerged within the field of anaesthetics as an automated, quantitative tool to assess depth of anaesthesia during surgical procedures. This culminated in the development of commercially available “depth of anaesthesia” monitors, now recommended by UK NICE guidelines for use in many anaesthetic settings [

      Overview | Depth of anaesthesia monitors – Bispectral Index (BIS), E-Entropy and Narcotrend-Compact M | Guidance | NICE n.d.

      ]. These use quantitative EEG algorithms to convert EEG output into a pEEG variable (numeric scale or letter category). This can used to guide anaesthetic dosing real time by an anaesthetist without training in neurophysiology, to prevent unintended operative awareness, and avoid overly deep anaesthesia associated with delirium and cognitive dysfunction on awakening [
      • Radtke FM
      • Franck M
      • Lendner J
      • Krüger S
      • Wernecke KD
      • Spies CD.
      Monitoring depth of anaesthesia in a randomized trial decreases the rate of postoperative delirium but not postoperative cognitive dysfunction.
      ,
      • Soehle M
      • Dittmann A
      • Ellerkmann RK
      • Baumgarten G
      • Putensen C
      • Guenther U.
      Intraoperative burst suppression is associated with postoperative delirium following cardiac surgery: A prospective, observational study.
      ].
      A number of proprietary pEEG tools are available, including the Entropy (GE Healthcare, USA), Narcotrend (Narcotrend-Gruppe, Germany), Neurosense (Neuro Wave Systems, USA), and SNAP II (Stryker, USA) devices, however most commonly mentioned in the literature is the Bispectral Index (BIS, Medtronic, Ireland) (see Table 1). BIS outputs include the “BIS” value and a sub-parameter the suppression ratio (SR). Both are dimensionless values calculated algorithmically from raw EEG data obtained from a small array of frontal electrodes (see Fig. 1, item A). These are scored on a continuum developed by analysing EEG data from many hundreds of alert and anaesthetised individuals. The “BIS” value is an overall indicator of awareness ranging from 0-100, with 0 correlating with a flat, isoelectric EEG trace, and scores of 90 and above expected in wakefulness (see Fig. 1, item D). During an anaesthetic, values between 40 and 60 are recommended, to minimise likelihood of awareness [
      • Myles PS
      • Leslie K
      • McNeil J
      • Forbes A
      • Chan MTV.
      Bispectral index monitoring to prevent awareness during anaesthesia: the B-aware randomised controlled trial.
      ], and avoid needlessly deep anaesthesia and associated side effects [
      • Soehle M
      • Dittmann A
      • Ellerkmann RK
      • Baumgarten G
      • Putensen C
      • Guenther U.
      Intraoperative burst suppression is associated with postoperative delirium following cardiac surgery: A prospective, observational study.
      ]. The suppression ratio (SR) is an estimate of the level of burst suppression (BS), describing the percentage of the preceding 63 second EEG trace that is isoelectric; a value of 0 having no isoelectric periods and 100 being entirely isoelectric. This correlates strongly in comparative studies to BS calculations taken from standard EEG recordings and compared against SR [
      • Arbour RB
      • Dissin J.
      Predictive value of the bispectral index for burst suppression on diagnostic electroencephalogram during drug-induced coma.
      ,
      • Cottenceau V
      • Petit L
      • Masson F
      • Guehl D
      • Asselineau J
      • Cochard J-F
      • et al.
      The Use of Bispectral Index to Monitor Barbiturate Coma in Severely Brain-Injured Patients with Refractory Intracranial Hypertension.
      ,
      • Riker RR
      • Fraser GL
      • Wilkins ML.
      Comparing the bispectral index and suppression ratio with burst suppression of the electroencephalogram during pentobarbital infusions in adult intensive care patients.
      ].
      Table 1Summary of devices used in highlighted studies. EEG = electroencephalogram; CDSA=colour density spectral array
      Processed EEG DeviceSourceOutput
      Bispectral index (BIS)BIS, Medtronic, IrelandNumeric value (BIS value) 0-100

      Suppression ratio

      CDSA

      Raw EEG trace – frontal montage
      NarcotrendNarcotrend-Gruppe, GermanyLetter based categorisation of anaesthesia depth (A-F)

      Raw EEG trace – summative montage
      SEDlineSEDline, Masimo, USAColour coded numeric value (Patient State Index – PSI) 0-100

      CDSA

      Raw EEG trace – frontal montage
      Table 2Study Characteristics grouped by setting (critical care setting/epilepsy; perioperative seizures) and study type (observational; case report).
      Critical care/epilepsy
      Observational Studies
      Author, Year, CountryStudy Design, N, patient characteristicsDevice usedIntention of studyRelevant findingsReview Conclusion
      Haesen et al, 2018 Belgium
      • Haesen J
      • Eertmans W
      • Genbrugge C
      • Meex I
      • Demeestere J
      • Vander Laenen M
      • et al.
      The validation of simplified EEG derived from the bispectral index monitor in post-cardiac arrest patients.
      Retrospective, 32, post-cardiac arrestBISComparison of raw EEG from BIS with standard EEGGood correlation simplified raw EEG from BIS device with standard EEG for burst suppression and status epilepticus, as determined by two neurophysiologists (r=0.810). Periodic discharges not detected in 75% of cases.Raw EEG data captured by BIS is able to demonstrate status epilepticus and burst suppression, but can miss periodic discharges. BIS value not examined in this study, limiting use to non-neurophysiology specialist
      You et al, 2017, South Korea
      • You KM
      • Suh GJ
      • Kwon WY
      • Kim KS
      • Ko SB
      • Park MJ
      • et al.
      Epileptiform discharge detection with the 4-channel frontal electroencephalography during post-resuscitation care.
      Prospective, 39, post-resuscitation careSEDlineDetection of epileptiform discharges in post-resuscitation careSEDline raw EEG had 100% sensitivity, specificity, PPV and NPV in detection epileptiform discharges compared to conventional EEG (epileptiform discharges found in 6 out of 39 patients). Good ability of two emergency physicians to detect epileptiform discharges (kappa value 0.71)Epileptiform discharges present on raw EEG trace from SEDline correlate with findings of conventional EEG.

      Lack of assessment of PSI value may restrict application to non-specialist, however emergency physicians demonstrated to successfully interpret raw EEG data
      Hernandez-Hernandez et al, 2016, Spain
      • Hernández-Hernández MA
      • Fernández-Torre JL.
      Color density spectral array of bilateral bispectral index system: Electroencephalographic correlate in comatose patients with nonconvulsive status epilepticus.
      Retrospective cohort, 15, patients with NCSEBISComparison of BIS in NCSE versus sedation statesVariation in CDSA between NCSE and non-NCSE sedated states (NCSE confirmed on conventional EEG). Dynamic change in CDSA output administration of midazolam. High variability of BIS value during NCSE, likely due to EMG activityBIS CDSA output varies between NCSE and sedation states
      Musialowicz et al, 2010, Finland
      • Musialowicz T
      • Mervaala E
      • Kälviäinen R
      • Uusaro A
      • Ruokonen E
      • Parviainen I.
      Can BIS monitoring be used to assess the depth of propofol anesthesia in the treatment of refractory status epilepticus?.
      Prospective, 10, RSE ITU admissionsBISDetection of burst suppression using BIS in RSESignificantly lower BIS value during conventional EEG proven burst suppression. Close correlation between BIS value (r2 = -0.9, p < 0.001), suppression ratio (r2 = 0.88, p < 0.001) and EEG burst rate per minute. AUC for BIS value to detect BS on EEG 0.99. BIS values < 30 show 99% sensitivity and 98% specificity for BS pattern. Significantly higher BIS value during epileptiform activity (P<0.001)BIS value significantly changes with burst suppression periods and epileptiform discharges demonstrated on conventional EEG. Correlation between Suppression Ratio and EEG burst suppression parameters
      Arbour and Dissin, 2015
      • Arbour RB
      • Dissin J.
      Predictive value of the bispectral index for burst suppression on diagnostic electroencephalogram during drug-induced coma.
      Prospective cohort, 4, drug-induced coma for RSE or raised ICPBISCorrelation between EEG and BIS for burst suppressionStrong inverse correlation between EEG burst count and BIS SR (Spearman rank coefficient 0.8727)

      Strong correlation between BIS value versus quantitative burst count (Spearman's rank coefficient of 0.8810)
      BIS suppression rate correlates with conventional EEG parameters of burst suppression
      Prins et al 2007 Netherlands
      • Prins SA
      • de Hoog M
      • Blok JH
      • Tibboel D
      • Visser GH.
      Continuous noninvasive monitoring of barbiturate coma in critically ill children using the BispectralTM index monitor.
      Prospective, 8, drug-induced coma for RSE or traumatic brain injury (children)BISComparison of EEG and BIS for monitoring burst suppressionMean correlation between BIS suppression ratio and suppression rate on EEG r=0.68 (assessed 4/8 patients only)

      Two patients suppression ratio found to be either very high or very low precluding correlation
      Moderate correlates with conventional EEG burst suppression rates and BIS suppression ratio
      Riker et al, 2003
      • Riker RR
      • Fraser GL
      • Wilkins ML.
      Comparing the bispectral index and suppression ratio with burst suppression of the electroencephalogram during pentobarbital infusions in adult intensive care patients.
      USA
      Prospective cohort, 12, drug induced coma for raised ICPBISDetection of burst suppression in drug induced comaBIS value (r=0.90 P<0.001) and Suppression Ratio (r=-0.89 P<0.001) strongly correlated with number of bursts/min on conventional EEG confirming BS state. Lower BIS value when vasopressors required (P<0.001)BIS and SR correlate with burst suppression values on conventional EEG, in this case useful in titration of barbiturate therapy. BIS value might reflect cerebral perfusion.
      Case Reports
      Author, Year, CountryPresentationDevice usedScenarioRelevant findingsReview Conclusion
      Ntahe et al, 2018 France
      • Ntahe A.
      Early Diagnosis of nonconvulsive status epilepticus recurrence with raw EEG of a bispectral index monitor.
      Cardiac arrest secondary to pneumothoraxBISHypoxic brain injury and NCSENCSE demonstrated by standard EEG correlated with high voltage irregular morphology on raw EEG from BIS. BIS value demonstrated to be low during period of epileptiform activity (29-30)BIS derived raw EEG can demonstrate suspicious activity suggestive of NCSE. BIS value can be low during these events
      Smith et al 2015 UK
      • Smith M
      • Dobbs P
      • Eapen G.
      Abnormal bispectral index values associated with the presence of periodic lateralized epileptiform discharges.
      HSV encephalitisBISStatus epilepticusUnexpectedly high BIS values in comatose patient during periodic high amplitude spikes seen on raw EEG trace from BIS monitor. Periodic lateralised discharges found on conventional EEGHigh BIS values may be seen in comatose patients with periodic lateralised epileptic discharges
      Fernandez-Torre et al, 2012 Spain
      • Fernández-Torre JL
      • Hernández-Hernández MA.
      Utility of bilateral Bispectral index (BIS) monitoring in a comatose patient with focal nonconvulsive status epilepticus.
      Subarachnoid haemorrhageBISNSCE (proven on conventional EEG)CDSA showed stereotyped changes indicative of recurrent focal NCSE. Resolution of NCSE as proven with conventional EEG associated with change of CDSA. BIS value correlates with presence of NCSE (values not disclosed)CDSA correlates with NCSE as proven on conventional EEG
      Dahaba et al, 2010 Austria
      • Dahaba AA
      • Liu DW
      • Metzler H.
      Bispectral Index (BIS) monitoring of acute encephalitis with refractory, repetitive partial seizures (AERRPS).
      Encephalitis with refractory repetitive partial seizuresBISFocal and generalised seizures, induced burst suppressionElevated mean BIS value at seizure onset (94.5) with decrease after cessation of seizure episode (55.1). Correspondence of conventional EEG activity during seizure episodes. Decrease of BIS value to mean 11.9 and Suppression Ratio value to 65.5 with pharmacologically induced burst suppressionBIS value rises during seizure activity and returns to expected values for induced coma on cessation.

      Correlation between rising Suppression Ration value, falling BIS value and pharmacologically induced burst suppression
      Chamorro et al 2008 Spain
      • Chamorro C
      • Romera MA
      • Balandín B
      • Valdivia M.
      Nonconvulsive status and bispectral index.
      Unexplained cause of reduced GCSBIS2 cases of NCSEUnexpectedly high BIS values (>60) in comatose patients, leading to diagnosis of NCSEUnexpectedly high BIS value in a comatose patient may alert a clinician to the possibility of NCSE (unclear if confirmatory conventional EEG was performed in this case)
      Ohshima et al 2007 Japan
      • Ohshima N
      • Chinzei M
      • Mizuno K
      • Hayashida M
      • Kitamura T
      • Shibuya H
      • et al.
      Transient decreases in Bispectral Index without associated changes in the level of consciousness during photic stimulation in an epileptic patient.
      New diagnosis of generalised epilepsyBISPhotic stimulation in awake patientHigh voltage slow waves seen on conventional EEG during photic stimulation also seen on raw EEG output on BIS. Fall in BIS value to minimum 63 during photic stimulation and return to baseline after cessation. Reduction in above phenomena at repeat testing after commencing sodium valproate.Epileptiform discharges can be seen on BIS raw EEG output and reflected in BIS value in an awake patient with epilepsy. Attenuation in responses seen with anti-convulsant treatment.
      Tallach et al 2004 UK
      • Tallach RE
      • Ball DR
      • Jefferson P.
      Monitoring seizures with the Bispectral index.
      Refractory status epilepticusBISRefractory status epilepticusHigh BIS value (94) observed in patient during clinical seizure, with corresponding epileptic activity on raw EEG trace. Resolution with administration with propofol bolus, BIS value falling to 28. Patient non-rousable throughout, documented as having benzodiazepines and phenytoin prior to BIS value of 94Potential inappropriately high BIS value with fall following propofol administration and cessation of seizures. Caveat that change is BIS value could be due to propofol exposure alone.
      Jaggi et al 2003 USA
      • Jaggi P
      • Schwabe MJ
      • Gill K
      • Horowitz IN.
      Use of an anesthesia cerebral monitor bispectral index to assess burst-suppression in pentobarbital coma.
      Status epilepticus in childBISStatus epilepticusCorrelation between low BIS value (r=0.59), raised Suppression Ratio (r=-0.64 P<0.01) and bursts per minute on conventional EEG in pentobarbital induced burst suppressionLow BIS value and raised Suppression Ratio correlates with conventional EEG in pharmacologically induced burst suppression
      Perioperative seizures
      Observational Studies
      Author, Year, CountryStudy Design, N, patient characteristicsDevice usedIntention of studyRelevant findingsReview Conclusion
      Särkelä et al, 2007 Finland
      • Särkelä MOK
      • Ermes MJ
      • Van Gils MJ
      • Yli-Hankala AM
      • Jäntti VH
      • Vakkuri AP.
      Quantification of epileptiform electroencephalographic activity during sevoflurane mask induction.
      Prospective observational study, 60, gynaecological surgeryBISComparison of epileptiform EEG traces with BIS valueIncrease in BIS value during periodic discharges observed on 4-channel EEG – did not achieve statistical significance

      Tendency for BIS value not to be displayed during epileptiform activity, perhaps due to system considering these artefactual
      BIS value may increase during periodic discharges, but not shown to be statistically significant
      Stasiowski et al, 2019, Poland
      • Stasiowski MJ
      • Marciniak R
      • Duława A
      • Krawczyk L
      • Jałowiecki P.
      Epileptiform EEG patterns during different techniques of induction of general anaesthesia with sevoflurane and propofol: A randomised trial.
      Prospective randomised, 60, orthopaedic surgeryBISComparison of epileptiform discharge incidence between sevoflurane and propofolEpileptiform EEG patterns in 16/35 patients anaesthetised with sevoflurane (as captured by 4-lead conventional EEG). No epileptiform EEG patters seen in those anaesthetised with propofol. Variety of BIS values seen before and during onset of epileptiform EEG patterns; both excessively high, low and unchanged (range 11-97). Mean BIS value before and after seizures unchanged compared to similar time points in group without seizuresBIS value shown to vary in patients with onset of epileptiform EEG discharges, either excessively high, low or unchanged compared to expected levels during anaesthesia

      No mean difference seen in BIS value before or during epileptiform discharges compared to group without discharges at similar time point. This may have resulted from a cancelling out effect seen from alternative extremes of BIS value during discharges
      Case Reports
      Author, Year, CountrySurgeryDevice usedScenarioRelevant findingsReview Conclusion
      Berger-Estilita et al, 2019 Switzerland
      • Berger-Estilita J
      • Steck K
      • Vetter C
      • Seidel K
      • Krejci V
      • Hight D
      • et al.
      A case report of several intraoperative convulsions while using the Narcotrend monitor: Significance and predictive use.
      Elective resection glioblastomaNarcotrendIntraoperative tonic clonic seizuresRaw EEG extracted and analysed post-operatively, converted to Increase in mean beta power prior to tonic-clonic seizures Raw EEG used to enter burst suppression. Processed depth of anaesthesia value not assessed.Seizure onset demonstrated by Narcotrend device, however using raw EEG requiring specialist interpretation
      Iturri Clavero et al, 2015 Spain
      • Iturri Clavero F
      • Tamayo Medel G
      • de Orte Sancho K
      • González Uriarte A
      • Iglesias Martínez A
      • Martínez Ruíz A
      Use of BIS VISTATM bilateral monitor for diagnosis of intraoperative seizures, a case report.
      Resection of meningiomaBISIntraoperative tonic clonic seizureRise in BIS value from 30 to 45 during seizure. CDSA showed increased frequency and warmer coloursBIS value may rise during seizure with changes evident on CDSA display
      Galante et al, 2015 Brazil
      • Galante D
      • Fortarezza D
      • Caggiano M
      • de Francisci G
      • Pedrotti D
      • Caruselli M.
      Correlation of bispectral index (BIS) monitoring and end-tidal sevoflurane concentration in a patient with lobar holoprosencephaly.
      Orchidopexy in patient with known holoprosencephaly and epilepsyBISUnexpectedly low BIS value during surgery, intra-operative epileptiform dischargeBIS value found to be unexpectedly low (26.0) and persistent despite decrease of sevoflurane end-tidal concentration (26.5). Conventional EEG obtained demonstrated epileptiform activity at this time point. Patient recovered with no clinically evident seizure or sequelaeLow BIS value demonstrated during epileptiform activity demonstrated on conventional EEG, unresponsive to reduction in anaesthetic exposure.

      Confounded by potential idiosyncratic effect on structurally abnormal brain from anaesthetic
      Kim et al, 2013 Korea
      • Kim H
      • Kim SY.
      Pitfall of bispectral index during intraoperative seizure -a case report.
      Kidney transplantationBISIntraoperative tonic-clonic seizureBrief rise in BIS value during seizure (40 to 62), followed by subsequent fall to 0 for period of 5 minutes. Intermittently low BIS value following this for 30 minutes. Noted to have profound hyponatraemia.Rise in BIS value during seizure observed with markedly low value following cessation of convulsion.
      Ogawa et al, 2009 Japan
      • Ogawa S
      • Okutani R
      • Nakada K
      • Suehiro K
      • Shigemoto T.
      Spike-monitoring of anaesthesia for corpus callosotomy using bilateral bispectral index.
      Elective corpus callosotomy for childhood epilepsy syndromesBISDifference in BIS parameters pre/ post callosotomy (2 cases)Use of separate BIS monitors for each hemisphere. Variation in BIS value for each side dependent on occurrence of lateralised epileptic discharges. Loss of synchronisation of spikes on BIS monitor raw EEG signal post callosotomyDiffering lateralised BIS values dependent on lateralisation of epileptiform discharges
      Hamada et al, 2008 France
      • Hamada S
      • Laloë PA
      • Hausser-Hauw C
      • Fischler M.
      Seizure after aortic clamp release: a bispectral index pitfall.
      Acute mitral valve surgery for infective endocarditisBISIntraoperative seizureLeft lateralised epileptic discharges observed on intraoperative conventional EEG following release of aortic clamp. Delayed fall in BIS value to 0 following onset of prolonged discharge. Resolution of EEG changes and BIS value stabilised at normal anaesthetic range. Notably had unilateral right sided BIS electrodeProfound fall in BIS value in lateralised epileptic discharge. Complicated by use of single electrode on contralateral hemisphere to discharge. Authors disclose intra-operative confusion of BIS and BIS EMG values highlighting a potential pitfall.
      Chinzei et al 2004 Japan
      • Chinzei M
      • Sawamura S
      • Hayashida M
      • Kitamura T
      • Tamai H
      • Hanaoka K.
      Change in Bispectral Index during epileptiform electrical activity under sevoflurane anesthesia in a patient with epilepsy.
      ]
      Elective gastrectomy in patient with epilepsyBISIntra-operative EEG changesAbrupt transient periodic decreases in BIS (40 to 20) associated with occurrence of high-voltage, low frequency waves on raw EEG trace. Resolved with midazolam and barbiturate infusion.BIS value falls in concordance with transient epileptic phenomena on raw EEG trace, resolved with further sedation
      AUC; area under curve, BS; burst suppression, CDSA; colour density spectral array, ICA; internal carotid artery, ICP; intracranial pressure, NCSE; non-convulsive status epilepticus, RSE; refractory status epilepticus, ICP; intracranial pressure
      CDSA; colour density spectral array, HSV; herpes simplex virus, NCSE; non-convulsive status epilepticus,
      CDSA; colour density spectral array, SR; suppression ratio
      Fig 1:
      Fig. 1A BIS frontal electrode placement (as taken from Nunes et al 2012
      [
      • Nunes RR
      • Chave IMM
      • Alencar JCG de
      • Franco SB
      • Oliveira YGBR de
      • Menezes DGA de
      Bispectral index and other processed parameters of electroencephalogram: an update.
      ]
      ), B Image of BIS monitor output, BIS value shown as 76 (as adapted from Chi et al 2016
      [
      • Chi SI
      • Kim HJ
      • Seo K-S
      • Yang M
      • Chang J
      Use of ADMSTM during sedation for dental treatment of an intellectually disabled patient: a case report.
      ]
      ), C CDSA output comparing NCSE and sedation pattern; red and darker red tones consistent with NCSE pattern (as adapted from Hernandez-Hernandez et al 2016
      [
      • Hernández-Hernández MA
      • Fernández-Torre JL.
      Color density spectral array of bilateral bispectral index system: Electroencephalographic correlate in comatose patients with nonconvulsive status epilepticus.
      ]
      ), D Representation of raw EEG patterns at varying BIS level (as taken from Wildes et al 2016
      [
      • Wildes TS
      • Winter AC
      • Maybrier HR
      • Mickle AM
      • Lenze EJ
      • Stark S
      • et al.
      Protocol for the electroencephalography guidance of anesthesia to alleviate geriatric syndromes (ENGAGES) study: a pragmatic, randomised clinical trial.
      ]
      )
      Fig 2:
      Fig. 2Flow chart of the search and study selection process.
      This review explores the evidence to date on the use of proprietary, processed EEG depth of anaesthesia monitors for the detection and monitoring of seizure activity. Of particular interest is the use of “depth of anaesthesia” values (such as the “BIS” value) and their correlation with seizure activity (both clinical and neurophysiological). The ultimate question is whether processed EEG modalities may be able to replicate features of conventional EEG monitoring and represent an intuitive and easily interpreted paradigm for clinicians without neurophysiology training, such as a general neurologist or critical care doctor. The aim has been to produce a best evidence synthesis exploring what information is known in this area, whether there may be any utility to this approach and the required direction for further research.

      3. Methods

      This scoping review has followed the Preferred Reporting Items for Systematic Reviews and Meta-analysis and Centre for Reviews and Dissemination (PRISMA) guidance for undertaking reviews in healthcare, as far as relevant for the scope of this review.

      3.1 Search strategy

      A literature search of Medline, Psych-info and Embase databases was conducted in articles published from inception to 20 March 2021 to identify articles reporting studies of the use of depth of anaesthesia monitoring and seizure detection. The search was restricted to human studies and those published in English. Reference lists were searched for additional relevant articles. Search terms are available in Supplemental file 1. Searches were conducted independently by both authors after agreeing search terms and results combined. MS assessed the literature using the agreed search terms in 2018 and 2021. AG assessed the literature using these terms in 2020. Updated abstracts were added to the review process as they arose across the review period.

      3.2 Systematic search and review flow diagram

      3.2.1 Selection criteria and article screening

      Inclusion criteria were articles describing use of depth of anaesthesia monitors utilising pEEG in clinical scenarios involving the diagnosis or monitoring of seizure activity. This could include studies where incidental seizures occurred whilst pEEG was in place, for example during surgery. Additionally, articles assessing the use of pEEG devices for achieving and monitoring artificially induced burst suppression rhythms were also included. Instances when seizures were not confirmed by conventional EEG assessment or raw EEG trace were excluded (see discussion). Studies involving neuro-prognostication in brain injury without documentation of seizures were excluded, as were case reports documenting anaesthetic protocol using precautionary pEEG without actual occurrence of seizures.
      Eligible articles were identified by performing an initial screen of titles and abstracts, followed by a full article review of those that passed screening. All retrospective and prospective studies that met these criteria were selected. Study quality was not systematically assessed due to the heterogeneity of study content assessed to synthesise this best evidence review.

      4. Results

      4.1 Literature search

      The literature search detected 126 citations, of which none were duplicates. One further paper was identified manually through reference screening. After screening titles and abstracts, 47 full-text articles were assessed for eligibility. 23 articles were excluded including those documenting induced seizures at ECT, 6 articles in which seizures were not confirmed with conventional EEG, one describing a research protocol and three examining pEEG in neuro-prognostication alone without correlation of pEEG variables during seizures(see Figure 2). Articles on ECT induced seizures have not been included in the systematic synthesis of this review and were excluded.

      4.2 Study characteristics

      Our systematic review searched the major libraries for articles documenting seizure events having data captured by depth of anaesthesia devices utilising pEEG. This yielded 8 cohort studies, 1 randomised trial and 15 case reports(see Table 2). These involved scenarios of peri-operative seizures [
      • Chi SI
      • Kim HJ
      • Seo K-S
      • Yang M
      • Chang J
      Use of ADMSTM during sedation for dental treatment of an intellectually disabled patient: a case report.
      ], or individuals in the critical care setting with either brain injury or status epilepticus [
      • Ntahe A.
      Early Diagnosis of nonconvulsive status epilepticus recurrence with raw EEG of a bispectral index monitor.
      ], and one involving an ambulatory patient with epilepsy. The majority of literature found concerned the Bispectral index system (BIS), with one found for SEDline and one for Narcotrend. No eligible studies were found involving the Entropy, SNAP or Neurosense systems. Three studies concerned patients of paediatric age with the remainder involving adults. Selected studies are presented and contextualised in Table 1.

      4.3 Critical care/epilepsy literature

      15 studies were identified that involved patients in a critical care setting experiencing seizures with processed EEG depth of anaesthesia monitoring in use. Eight papers were case reports documenting processed EEG parameters during clinically apparent or neuro-physiologically confirmed seizures [
      • Ntahe A.
      Early Diagnosis of nonconvulsive status epilepticus recurrence with raw EEG of a bispectral index monitor.
      ,
      • Fernández-Torre JL
      • Hernández-Hernández MA.
      Utility of bilateral Bispectral index (BIS) monitoring in a comatose patient with focal nonconvulsive status epilepticus.
      ,
      • Smith M
      • Dobbs P
      • Eapen G.
      Abnormal bispectral index values associated with the presence of periodic lateralized epileptiform discharges.
      ,
      • Dahaba A
      • Lui D
      • Metzler H.
      Bispectral index (BIS) monitoring of acute encephalitis with refractory, repetitive partial seizures (AERRPS).
      ,
      • Chamorro C
      • Romera MA
      • Balandín B
      • Valdivia M.
      Nonconvulsive status and bispectral index.
      ,
      • Ohshima N
      • Chinzei M
      • Mizuno K
      • Hayashida M
      • Kitamura T
      • Shibuya H
      • et al.
      Transient decreases in Bispectral Index without associated changes in the level of consciousness during photic stimulation in an epileptic patient.
      ,
      • Tallach RE
      • Ball DR
      • Jefferson P.
      Monitoring seizures with the Bispectral index.
      ,
      • Jaggi P
      • Schwabe MJ
      • Gill K
      • Horowitz IN.
      Use of an anesthesia cerebral monitor bispectral index to assess burst-suppression in pentobarbital coma.
      ].
      Seven relevant observational cohort studies were found of which five were prospective [
      • Arbour RB
      • Dissin J.
      Predictive value of the bispectral index for burst suppression on diagnostic electroencephalogram during drug-induced coma.
      ,
      • Riker RR
      • Fraser GL
      • Wilkins ML.
      Comparing the bispectral index and suppression ratio with burst suppression of the electroencephalogram during pentobarbital infusions in adult intensive care patients.
      ,
      • Hernández-Hernández MA
      • Fernández-Torre JL.
      Color density spectral array of bilateral bispectral index system: Electroencephalographic correlate in comatose patients with nonconvulsive status epilepticus.
      ,
      • Haesen J
      • Eertmans W
      • Genbrugge C
      • Meex I
      • Demeestere J
      • Vander Laenen M
      • et al.
      The validation of simplified EEG derived from the bispectral index monitor in post-cardiac arrest patients.
      ,
      • You KM
      • Suh GJ
      • Kwon WY
      • Kim KS
      • Ko SB
      • Park MJ
      • et al.
      Epileptiform discharge detection with the 4-channel frontal electroencephalography during post-resuscitation care.
      ,
      • Musialowicz T
      • Mervaala E
      • Kälviäinen R
      • Uusaro A
      • Ruokonen E
      • Parviainen I.
      Can BIS monitoring be used to assess the depth of propofol anesthesia in the treatment of refractory status epilepticus?.
      ,
      • Prins SA
      • de Hoog M
      • Blok JH
      • Tibboel D
      • Visser GH.
      Continuous noninvasive monitoring of barbiturate coma in critically ill children using the BispectralTM index monitor.
      ]. These attempted to answer varying hypotheses, chiefly comparison of epileptic activity pEEG data with conventional EEG or correlation with burst suppression rhythms as determined by conventional EEG techniques and specific scenarios. All studies used BIS except for one observational study which used SEDline [
      • Musialowicz T
      • Mervaala E
      • Kälviäinen R
      • Uusaro A
      • Ruokonen E
      • Parviainen I.
      Can BIS monitoring be used to assess the depth of propofol anesthesia in the treatment of refractory status epilepticus?.
      ].

      4.4 Peri-operative literature

      9 studies were identified that involved patients in a peri-operative setting with seizures experienced whilst depth of anaesthesia monitoring was in place. One of these was a randomised trial which examined the differences in epileptiform discharge rate between two types of anaesthetic (propofol versus sevoflurane) with data captured by a BIS system and a limited 4-lead conventional EEG [
      • Iturri Clavero F
      • Tamayo Medel G
      • de Orte Sancho K
      • González Uriarte A
      • Iglesias Martínez A
      • Martínez Ruíz A
      Use of BIS VISTATM bilateral monitor for diagnosis of intraoperative seizures, a case report.
      ]. Another was a prospective study examining correlation between epileptiform discharges on 4-lead EEG and the BIS system during anaesthetic for gynaecological surgery [
      • Särkelä MOK
      • Ermes MJ
      • Van Gils MJ
      • Yli-Hankala AM
      • Jäntti VH
      • Vakkuri AP.
      Quantification of epileptiform electroencephalographic activity during sevoflurane mask induction.
      ].
      7 case reports were identified documenting the pEEG parameters from depth of anaesthesia monitoring during clinical seizures or epileptiform appearances on EEG [
      • Iturri Clavero F
      • Tamayo Medel G
      • de Orte Sancho K
      • González Uriarte A
      • Iglesias Martínez A
      • Martínez Ruíz A
      Use of BIS VISTATM bilateral monitor for diagnosis of intraoperative seizures, a case report.
      ,
      • Berger-Estilita J
      • Steck K
      • Vetter C
      • Seidel K
      • Krejci V
      • Hight D
      • et al.
      A case report of several intraoperative convulsions while using the Narcotrend monitor: Significance and predictive use.
      ,
      • Kim H
      • Kim SY.
      Pitfall of bispectral index during intraoperative seizure -a case report.
      ,
      • Hamada S
      • Laloë PA
      • Hausser-Hauw C
      • Fischler M.
      Seizure after aortic clamp release: a bispectral index pitfall.
      ,
      • Chinzei M
      • Sawamura S
      • Hayashida M
      • Kitamura T
      • Tamai H
      • Hanaoka K.
      Change in Bispectral Index during epileptiform electrical activity under sevoflurane anesthesia in a patient with epilepsy.
      ,
      • Galante D
      • Fortarezza D
      • Caggiano M
      • de Francisci G
      • Pedrotti D
      • Caruselli M.
      Correlation of bispectral index (BIS) monitoring and end-tidal sevoflurane concentration in a patient with lobar holoprosencephaly.
      ,
      • Ogawa S
      • Okutani R
      • Nakada K
      • Suehiro K
      • Shigemoto T.
      Spike-monitoring of anaesthesia for corpus callosotomy using bilateral bispectral index.
      ]. These were during anaesthetic induction or the operative phase. The majority of studies use the BIS system, with one using Narcotrend [
      • Berger-Estilita J
      • Steck K
      • Vetter C
      • Seidel K
      • Krejci V
      • Hight D
      • et al.
      A case report of several intraoperative convulsions while using the Narcotrend monitor: Significance and predictive use.
      ].

      5. Discussion

      The relevant literature identified on the use of depth of anaesthesia monitor derived pEEG for diagnosing or monitoring seizures is diverse. This generally focused on two settings; operative anaesthesia and critical care. The context to the former is that depth of anaesthesia monitoring is already in place being used for its intended purpose (optimised titration of anaesthesia) and tended to document where seizures have arose either coincidentally or somewhat expectedly (in the case of neurosurgical procedures). The critical care literature is more exploratory in nature, having more observational studies relative to case reports. For case reports, depth of anaesthesia monitors may have been applied deliberately to monitor seizure activity, however many critical care doctors have an anaesthetic background and application to an anaesthetised patient could have occurred out of personal habit.
      The majority of literature consisted of case reports. These vary in their case content but tend to demonstrate a trend towards depth of anaesthesia pEEG parameters changing beyond expected values for anaesthesia during clinically apparent or purely electrophysiologically proven seizures. These are helpful, documenting a clinical narrative with associated changes in pEEG derived values and are a good starting point for justifying future systematic research. However, publication bias may be seen here with clinicians tending to publish instances where they have seen an intriguing trend of pEEG parameters fitting the apparent clinical narrative, but unconsciously neglecting to publish similar cases where this was not seen.
      Observational or interventional studies produced to date are heterogeneous and address varying hypotheses; few directly address the hypothesis that pEEG accurately diagnoses seizure activity. The evidence for this hypothesis so far has to be inferred via cross-referencing data presented in these studies, alternative to the author's intended purpose. An example of this is Stasiowski et al's work comparing the incidence of epileptiform discharges attributable to two different types of anaesthetic; the use of depth of anaesthesia monitoring existed as a secondary outcome, and the paper's main intention was not to question if the depth of anaesthesia monitor is a useful modality for diagnosing seizures, but rather which anaesthetic drug cause more epileptiform activity [
      • Stasiowski MJ
      • Marciniak R
      • Duława A
      • Krawczyk L
      • Jałowiecki P.
      Epileptiform EEG patterns during different techniques of induction of general anaesthesia with sevoflurane and propofol: A randomised trial.
      ]. Of interest is that considerable weight is placed on the BIS value in the answering this hypothesis, when the definitive ability of BIS systems to diagnose seizures has not yet been properly established.

      5.1 Burst suppression and processed EEG

      A number of observational studies examined the correlation between pEEG derived values and induced burst-suppression rhythms sought for neuroprotective coma. Although this does not technically constitute the detection of pathological seizures, this approach is of interest as a proxy for an abnormal brain rhythm that conceptually represents the inverse of epileptiform activity. The BIS derived Suppression Ratio (SR) can be correlated to extent of burst suppression, and has specific clinical utility for the treatment of status epilepticus or brain injury, representing a routinely available method for monitoring extent of neuroprotective coma. This may avoid need for the equipment and interpretative resources of conventional EEG, and widen access in many hospitals to diagnostic tools where this is not easily available.
      The five burst suppression studies showed correlation between states of burst suppression as confirmed on conventional EEG, and BIS parameters [
      • Arbour RB
      • Dissin J.
      Predictive value of the bispectral index for burst suppression on diagnostic electroencephalogram during drug-induced coma.
      ,
      • Riker RR
      • Fraser GL
      • Wilkins ML.
      Comparing the bispectral index and suppression ratio with burst suppression of the electroencephalogram during pentobarbital infusions in adult intensive care patients.
      ,
      • Haesen J
      • Eertmans W
      • Genbrugge C
      • Meex I
      • Demeestere J
      • Vander Laenen M
      • et al.
      The validation of simplified EEG derived from the bispectral index monitor in post-cardiac arrest patients.
      ,
      • Musialowicz T
      • Mervaala E
      • Kälviäinen R
      • Uusaro A
      • Ruokonen E
      • Parviainen I.
      Can BIS monitoring be used to assess the depth of propofol anesthesia in the treatment of refractory status epilepticus?.
      ,
      • Prins SA
      • de Hoog M
      • Blok JH
      • Tibboel D
      • Visser GH.
      Continuous noninvasive monitoring of barbiturate coma in critically ill children using the BispectralTM index monitor.
      ]. Uniformly, a low BIS value and a high suppression ratio was associated with this state, representing a relative paucity of EEG activity and increasing percentages of the trace being isoelectric. Statistical analysis confirmed a strong correlation between these BIS parameters and conventional EEG [
      • Riker RR
      • Fraser GL
      • Wilkins ML.
      Comparing the bispectral index and suppression ratio with burst suppression of the electroencephalogram during pentobarbital infusions in adult intensive care patients.
      ,
      • Musialowicz T
      • Mervaala E
      • Kälviäinen R
      • Uusaro A
      • Ruokonen E
      • Parviainen I.
      Can BIS monitoring be used to assess the depth of propofol anesthesia in the treatment of refractory status epilepticus?.
      ]. The study demonstrating the weakest link was perhaps influenced by a bimodal distribution of BIS values (being either very low or very high [
      • You KM
      • Suh GJ
      • Kwon WY
      • Kim KS
      • Ko SB
      • Park MJ
      • et al.
      Epileptiform discharge detection with the 4-channel frontal electroencephalography during post-resuscitation care.
      ]) and demonstrates some of the difficulties in using this value, as explored below. Similar trends for a low BIS value and rising suppression ratio during periods of burst suppression were also seen in several case reports [
      • Dahaba A
      • Lui D
      • Metzler H.
      Bispectral index (BIS) monitoring of acute encephalitis with refractory, repetitive partial seizures (AERRPS).
      ,
      • Jaggi P
      • Schwabe MJ
      • Gill K
      • Horowitz IN.
      Use of an anesthesia cerebral monitor bispectral index to assess burst-suppression in pentobarbital coma.
      ].

      5.2 Bispectral index value and seizures

      The bispectral index device was by far the most commonly used in the literature. All published studies demonstrated a change in BIS parameters during the occurrence of seizure activity. Two patterns of change in BIS value during or after the occurrence of seizures were frequently seen in the literature, predominantly in case reports. A rising BIS value during seizure was most commonly reported (9 studies where ictal BIS value was reported) [
      • Smith M
      • Dobbs P
      • Eapen G.
      Abnormal bispectral index values associated with the presence of periodic lateralized epileptiform discharges.
      ,
      • Dahaba A
      • Lui D
      • Metzler H.
      Bispectral index (BIS) monitoring of acute encephalitis with refractory, repetitive partial seizures (AERRPS).
      ,
      • Chamorro C
      • Romera MA
      • Balandín B
      • Valdivia M.
      Nonconvulsive status and bispectral index.
      ,
      • Tallach RE
      • Ball DR
      • Jefferson P.
      Monitoring seizures with the Bispectral index.
      ,
      • Iturri Clavero F
      • Tamayo Medel G
      • de Orte Sancho K
      • González Uriarte A
      • Iglesias Martínez A
      • Martínez Ruíz A
      Use of BIS VISTATM bilateral monitor for diagnosis of intraoperative seizures, a case report.
      ,
      • Särkelä MOK
      • Ermes MJ
      • Van Gils MJ
      • Yli-Hankala AM
      • Jäntti VH
      • Vakkuri AP.
      Quantification of epileptiform electroencephalographic activity during sevoflurane mask induction.
      ,
      • Kim H
      • Kim SY.
      Pitfall of bispectral index during intraoperative seizure -a case report.
      ,
      • Jung HJ
      • Lee JM.
      High BIS and low rSO2 during CPB: seizure?.
      ,
      • Elgueta MF
      • Vega P
      • Lema G
      • Clede L.
      Should we monitor with bispectral index in all patients at high risk for seizures in the operating room?.
      ], with a return to expected levels for anaesthesia, often with administration of further sedation and cessation of convulsion if present. Alternatively, a falling BIS value during epileptic activity was seen in 6 studies [
      • Ntahe A.
      Early Diagnosis of nonconvulsive status epilepticus recurrence with raw EEG of a bispectral index monitor.
      ,
      • Ohshima N
      • Chinzei M
      • Mizuno K
      • Hayashida M
      • Kitamura T
      • Shibuya H
      • et al.
      Transient decreases in Bispectral Index without associated changes in the level of consciousness during photic stimulation in an epileptic patient.
      ,
      • Hamada S
      • Laloë PA
      • Hausser-Hauw C
      • Fischler M.
      Seizure after aortic clamp release: a bispectral index pitfall.
      ,
      • Chinzei M
      • Sawamura S
      • Hayashida M
      • Kitamura T
      • Tamai H
      • Hanaoka K.
      Change in Bispectral Index during epileptiform electrical activity under sevoflurane anesthesia in a patient with epilepsy.
      ,
      • Galante D
      • Fortarezza D
      • Caggiano M
      • de Francisci G
      • Pedrotti D
      • Caruselli M.
      Correlation of bispectral index (BIS) monitoring and end-tidal sevoflurane concentration in a patient with lobar holoprosencephaly.
      ,
      • Bousselmi R
      • Lebbi A
      • Ferjani M.
      Bispectral index changes during generalised tonic-clonic seizures.
      ].
      This variability in BIS response was supported by the randomised control trial of anaesthetic agents, finding that BIS values during epileptiform EEG discharges could be inappropriately high, low or unchanged [
      • Stasiowski MJ
      • Marciniak R
      • Duława A
      • Krawczyk L
      • Jałowiecki P.
      Epileptiform EEG patterns during different techniques of induction of general anaesthesia with sevoflurane and propofol: A randomised trial.
      ]. Importantly, this included the largest sample of all highlighted papers (n=60), and found similar mean group BIS values between patients with and without epileptiform discharges. A single report found a persistently low BIS value following cessation of an intra-operative tonic-clonic seizure [
      • Kim H
      • Kim SY.
      Pitfall of bispectral index during intraoperative seizure -a case report.
      ]. This may have represented the post-ictal state, however ongoing non-convulsive epileptic activity was not excluded due to lack of conventional EEG monitoring.
      A key challenge in interpreting this literature is the variation in time-point at which the BIS value was recorded relative to seizure activity. Seizures themselves are not homogenous, with different lengths and a variable pre-symptomatic period of EEG change, making it hard to determine a universal data point in which BIS value should be compared between studies. Additionally, as a result of algorithmic processing a lag exists between the presence of seizure on raw EEG and change of BIS value. Filtering processes for signal quality control may cancel out the effect of non-sustained periods of epileptic activity, in order prioritise a steady depth of anaesthesia value as is the actual intention of the device. This might mean that short seizures are missed. Many seizures reported in the included case reports were not clinically apparent due to neuro-muscular blockade. This may have led to a difficulty with seizure recognition, but did allow ictal muscle activity during seizures as a cause of increased BIS value (previously reported [
      • Bruhn J
      • Bouillon TW
      • Shafer SL.
      Electromyographic activity falsely elevates the bispectral index.
      ]) to be removed as a confounder.
      Finally, although authors specifically report no change to anaesthetic dosing at the time of BIS recording, the ability of anaesthesia to co-linearly affect the BIS value as an outcome measure presents a challenge for interpretation common to all depth of anaesthesia systems. Only one study examined an awake patient, eliminating this confounder; Ohshima et al reported on BIS value changes seen in an awake ambulatory patient with a new diagnosis of epilepsy undergoing conventional EEG with photic stimulation [
      • Ohshima N
      • Chinzei M
      • Mizuno K
      • Hayashida M
      • Kitamura T
      • Shibuya H
      • et al.
      Transient decreases in Bispectral Index without associated changes in the level of consciousness during photic stimulation in an epileptic patient.
      ]. During stimulation, high voltage slow wave abnormalities appeared on EEG during photic stimulation which correlated with a fall in BIS value to 63. This returned to a normal value for an awake patient after cessation of photic stimuli.
      The variability in BIS value seen during seizures suggests that the reduction of the highly complex information that EEG conveys into a single value is too simplistic for fully representing all epileptic activity. It is possible that low frequency epileptiform EEG abnormalities cause a lowering of BIS value and high frequency discharges cause a raised value. This fits with the way in which the BIS value is calculated by determining the power of different component waveforms of the EEG. A reasonable approach would be to consider a contextually inappropriate variation in BIS value as a suggestion that epileptiform activity is occurring, however definitive confirmation still requires conventional EEG where there is lack of clinical signs.

      5.3 BIS device - raw EEG and CDSA traces

      The BIS device allows the summated overall EEG wave form to be observed in real time, as well as exported for analysis. The former has proven useful to validate the degree of resolution possible with the limited montage available with the simplified BIS data. Haesen et al. demonstrated that raw EEG tracing from a BIS device was able to capture tracings diagnostic of status epilepticus, but missed observed periodic discharges that were captured on with concurrent conventional EEG [
      • Haesen J
      • Eertmans W
      • Genbrugge C
      • Meex I
      • Demeestere J
      • Vander Laenen M
      • et al.
      The validation of simplified EEG derived from the bispectral index monitor in post-cardiac arrest patients.
      ]. This has important implications on the potential resolution of pEEG, relying on fewer scalp electrodes and summating their output, with a high risk of missing abnormal activity.
      The latest BIS devices incorporate a colour density spectral array, visualising a power spectrum representation of the summated EEG activity (see Fig. 1, item C). Key discriminatory components that can be analysed in this are the range of colours and shape of constituent waveforms. The observational study by Hernandez-Hernandez et al. compared CDSA appearance in NCSE to patients with sedation only [
      • Hernández-Hernández MA
      • Fernández-Torre JL.
      Color density spectral array of bilateral bispectral index system: Electroencephalographic correlate in comatose patients with nonconvulsive status epilepticus.
      ]. This showed a difference in appearance between the two groups, with darker red colours indicating higher frequency waveforms. In one individual with epileptiform activity, CDSA reverted back to a pattern similar to the sedation only cases after administration of midazolam. Several case reports commented on the changes of CDSA output during seizures, with similar CDSA patterns seen [
      • Fernández-Torre JL
      • Hernández-Hernández MA.
      Utility of bilateral Bispectral index (BIS) monitoring in a comatose patient with focal nonconvulsive status epilepticus.
      ,
      • Iturri Clavero F
      • Tamayo Medel G
      • de Orte Sancho K
      • González Uriarte A
      • Iglesias Martínez A
      • Martínez Ruíz A
      Use of BIS VISTATM bilateral monitor for diagnosis of intraoperative seizures, a case report.
      ].

      5.4 Other depth of anaesthesia systems

      One observational study examined the use of the SEDline system, involving the assessment of raw EEG trace alone [
      • You KM
      • Suh GJ
      • Kwon WY
      • Kim KS
      • Ko SB
      • Park MJ
      • et al.
      Epileptiform discharge detection with the 4-channel frontal electroencephalography during post-resuscitation care.
      ]. Here two emergency physicians were able to reliably identify epileptiform discharges from this, although the extent of any previous coaching which they may or may not have received was not documented. One study constituted a case report documenting the Narcotrend system where raw EEG tracing was assessed without consideration of the proprietary pEEG output [
      • Berger-Estilita J
      • Steck K
      • Vetter C
      • Seidel K
      • Krejci V
      • Hight D
      • et al.
      A case report of several intraoperative convulsions while using the Narcotrend monitor: Significance and predictive use.
      ]. This demonstrated the ability for the raw EEG trace to detect the onset of intra-operative seizures. The literature to support the use of either of the above systems is currently very limited.

      5.5 Studies without conventional EEG correlation

      There are a number of studies which have reported the use of BIS without confirmation by conventional EEG or compatible raw EEG trace, and contribute less to our understanding of the clinical use of pEEG in seizure detection. These were excluded from the main literature synthesis but are worth consideration in understanding the challenges posed by the literature. Two reported use of BIS as an adjunct in the diagnosis of non-epileptic attacks, finding both high (BIS ~ 90) and low (BIS = 47) BIS values in cases with the clinical signs of non-epileptic attack, attributed to wakefulness and unintentional self-hypnosis/sleep respectively [
      • Donnelly B
      • Boyd V.
      Use of the bispectral index (BIS) monitor to aid in the diagnosis of pseudoseizures [4].
      ,
      • Sartorius A
      • Schmahl C.
      Bispectral index monitoring during dissociative pseudo-seizure.
      ]. A fall in BIS level during hypnosis and sleep have previously been documented [
      • Burkle CM
      • Jankowski CJ
      • Torsher LC
      • Rho EH
      • Degnim AC.
      Bis monitor findings during self-hypnosis.
      ,
      • Sleigh JW
      • Andrzejowski J
      • Steyn-Ross A
      • Steyn-Ross M.
      The bispectral index: a measure of depth of sleep?.
      ]. Both of these case reports cast concern on the ability of processed EEG be specific for the diagnosis of seizures, given the variable nature of the results and the overlap of BIS values with those seen in epileptiform seizures confirmed by conventional EEG.
      Four case studies reported the peri-operative BIS findings in the setting of clinical signs of possible seizure only (i.e. increase in lactate, prolonged post-operative drowsiness, tonic-clonic movements) [
      • Jung HJ
      • Lee JM.
      High BIS and low rSO2 during CPB: seizure?.
      ,
      • Elgueta MF
      • Vega P
      • Lema G
      • Clede L.
      Should we monitor with bispectral index in all patients at high risk for seizures in the operating room?.
      ,
      • Bousselmi R
      • Lebbi A
      • Ferjani M.
      Bispectral index changes during generalised tonic-clonic seizures.
      ,
      • Obara S
      • Kakinouchi K
      • Honda J
      • Noji Y
      • Hanayama C
      • Murakawa M.
      Dexmedetomidine administration in a patient with status epilepticus under color density spectral array monitoring.
      ]. One was a case report demonstrating a rise in the proprietary “PSI” value (Patient Sedation Index - conceptually similar to BIS value) as well as an increase in power appearances of the CDSA output [
      • Obara S
      • Kakinouchi K
      • Honda J
      • Noji Y
      • Hanayama C
      • Murakawa M.
      Dexmedetomidine administration in a patient with status epilepticus under color density spectral array monitoring.
      ], with change in CDSA to a lower power following administration of further sedation. Epileptiform appearing discharges could not be seen on the CDSA presented. All papers hypothesised that changes in pEEG output (a rise in BIS/PSI in three, and a fall in BIS in one) were related to peri-operative seizure, as evidenced by normalisation to expected values during anaesthetic following administration of further sedative. Given the inconsistency of these results, and lack of formal EEG confirmation of seizure activity, no conclusions can be made from these studies; indeed it is common for patients under sedation to experience non-seizure related muscle movements.

      5.6 Electro-convulsive therapy studies

      Twelve studies were identified concerning the use of the BIS system during anaesthesia for electro-convulsive therapy [
      • Gunawardane PO
      • Murphy PA
      • Sleigh JW.
      Bispectral index monitoring during electroconvulsive therapy under propofol anaesthesia.
      ,
      • Kayser S
      • Bewernick BH
      • Soehle M
      • Switala C
      • Gippert SM
      • Dreimueller N
      • et al.
      Degree of postictal suppression depends on seizure induction time in magnetic seizure therapy and electroconvulsive therapy.
      ,
      • Lemmens HJM
      • Levi DC
      • Debattista C
      • Brock-Utne JG.
      The timing of electroconvulsive therapy and bispectral index after anesthesia induction using different drugs does not affect seizure duration.
      ,
      • Ochiai R
      • Yamada T
      • Kiyama S
      • Nakaoji T
      • Takeda J.
      Bispectral index as an indicator of seizure inducibility in electroconvulsive therapy under thiopental anesthesia.
      ,
      • Pekel M
      • Postaci NA
      • Aytaç İ
      • Karasu D
      • Keleş H
      • Şen Ö
      • et al.
      Sevoflurane versus propofol for electroconvulsive therapy: effects on seizure parameters, anesthesia recovery, and the bispectral index.
      ,
      • Soehle M
      • Kayser S
      • Ellerkmann RK
      • Schlaepfer TE.
      Bilateral bispectral index monitoring during and after electroconvulsive therapy compared with magnetic seizure therapy for treatment-resistant depression.
      ,
      • Kranaster L
      • Hoyer C
      • Janke C
      • Sartorius A.
      Bispectral index monitoring and seizure quality optimization in electroconvulsive therapy.
      ,
      • White PF
      • Rawal S
      • Recart A
      • Thornton L
      • Litle M
      • Stool L.
      Can the bispectral index be used to predict seizure time and awakening after electroconvulsive therapy?.
      ,
      • Gombar S
      • Aggarwal D
      • Khanna AK
      • Gombar KK
      • Chavan BS.
      The bispectral electroencephalogram during modified electroconvulsive therapy under propofol anesthesia - relation with seizure duration and awakening.
      ,
      • White PF
      • Amos Q
      • Zhang Y
      • Stool L
      • Husain MM
      • Thornton L
      • et al.
      Anesthetic considerations for magnetic seizure therapy: a novel therapy for severe depression.
      ,
      • Nishihara F
      • Saito S.
      Pre-ictal bispectral index has a positive correlation with seizure duration during electroconvulsive therapy.
      ,
      • Nishihara F
      • Saito S.
      Adjustment of anaesthesia depth using bispectral index prolongs seizure duration in electroconvulsive therapy.
      ]; all observational studies with varied hypotheses. The majority intended to assess correlation between depth of anaesthesia and the duration of seizure, as a proxy for efficacy of the procedure. Several studies did not document a pEEG parameter during the induced-seizure (often focusing only on the value preceding seizure induction, reflecting the anaesthetic status); and those that did demonstrated variable BIS values. These studies are all extensively confounded by the method of anaesthesia in this context, with a short bolus given to render the patient unconscious to allow the seizure to be induced with minimal recovery time. Causality of any BIS value variation is near impossible to separate from the effect of the anaesthetic given the time scales involved.

      6. Conclusion

      A growing body of literature has been published on the use of depth of anaesthesia processed EEG devices for the diagnosis or monitoring of seizures. Much of this is currently anecdotal, relying on case reports and few of the observational studies performed in this area directly address the question of their utility. As a result there is insufficient data to support the use of the processed EEG methods for the purpose of routine seizure detection or monitoring, in particular without traditional EEG confirmation given the current lack of supporting evidence.
      Well designed, systematic trials with confirmatory conventional EEG are needed to better address this question. A coherent and reproducible set of measurement time points will be needed; we suggest the value taken before, immediately at onset, mid episode, in the time immediately following cessation and finally some minutes later. Further exploration in awake patients with seizures will be insightful by removing the component of the anaesthetic, perhaps in the context of an epilepsy video-telemetry scenario which remains a highly controlled environment.
      Evidence for use in monitoring burst suppression is more developed with a number of studies experimentally examining this question and as such, depth of anaesthesia monitors with specific values such as suppression ratio (BIS) may help titrate anaesthetic dosage to maintain coma where continuous EEG monitoring is not readily available.
      Systems such as BIS may have utility in detecting NCSE in a comatose patient, by demonstrating a BIS value out of context with clinical picture (either excessively high or low). CDSA readout may act as a more sensitive correlate with epileptiform activity, however is relatively subjective to assess and would require standardised training. Additionally the literature on CDSA use is less extensive that for BIS value. However, given the current lack of evidence, processed EEG from depth of anaesthesia monitoring should not be relied on to rule out NCSE and a full conventional EEG should be sought where suspicion exists.

      Declaration of Competing Interest

      The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

      Funding

      This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

      Acknowledgments

      None

      References

        • Meldrum BS
        • Brierley JB.
        Prolonged epileptic seizures in primates: ischemic cell change and its relation to ictal physiological events.
        Arch Neurol. 1973; 28 (–7): 10https://doi.org/10.1001/archneur.1973.00490190028002
        • Jenssen S
        • Gracely EJ
        • Sperling MR.
        How long do most seizures last? A systematic comparison of seizures recorded in the epilepsy monitoring unit.
        Epilepsia. 2006; 47: 1499-1503https://doi.org/10.1111/j.1528-1167.2006.00622.x
      1. Overview | Epilepsies: diagnosis and management | Guidance | NICE n.d.

        • Amzica F.
        What does burst suppression really mean?.
        Epilepsy Behav. 2015; 49: 234-237https://doi.org/10.1016/j.yebeh.2015.06.012
        • Koffman L
        • Rincon F
        • Gomes J
        • Singh S
        • He Y
        • Ritzl E
        • et al.
        Continuous electroencephalographic monitoring in the intensive care unit: a cross-sectional study.
        J Intensive Care Med. 2020; 35: 1235-1240https://doi.org/10.1177/0885066619849889
      2. Overview | Depth of anaesthesia monitors – Bispectral Index (BIS), E-Entropy and Narcotrend-Compact M | Guidance | NICE n.d.

        • Radtke FM
        • Franck M
        • Lendner J
        • Krüger S
        • Wernecke KD
        • Spies CD.
        Monitoring depth of anaesthesia in a randomized trial decreases the rate of postoperative delirium but not postoperative cognitive dysfunction.
        Br J Anaesth. 2013; 110https://doi.org/10.1093/bja/aet055
        • Soehle M
        • Dittmann A
        • Ellerkmann RK
        • Baumgarten G
        • Putensen C
        • Guenther U.
        Intraoperative burst suppression is associated with postoperative delirium following cardiac surgery: A prospective, observational study.
        BMC Anesthesiol. 2015; 15https://doi.org/10.1186/s12871-015-0051-7
        • Myles PS
        • Leslie K
        • McNeil J
        • Forbes A
        • Chan MTV.
        Bispectral index monitoring to prevent awareness during anaesthesia: the B-aware randomised controlled trial.
        Lancet. 2004; 363: 1757-1763https://doi.org/10.1016/S0140-6736(04)16300-9
        • Arbour RB
        • Dissin J.
        Predictive value of the bispectral index for burst suppression on diagnostic electroencephalogram during drug-induced coma.
        J Neurosci Nurs. 2015; 47: 113-122https://doi.org/10.1097/JNN.0000000000000124
        • Cottenceau V
        • Petit L
        • Masson F
        • Guehl D
        • Asselineau J
        • Cochard J-F
        • et al.
        The Use of Bispectral Index to Monitor Barbiturate Coma in Severely Brain-Injured Patients with Refractory Intracranial Hypertension.
        Anesth Analg. 2008; 107: 1676-1682https://doi.org/10.1213/ane.0b013e318184e9ab
        • Riker RR
        • Fraser GL
        • Wilkins ML.
        Comparing the bispectral index and suppression ratio with burst suppression of the electroencephalogram during pentobarbital infusions in adult intensive care patients.
        Pharmacotherapy. 2003; 23: 1087-1093https://doi.org/10.1592/phco.23.10.1087.32766
        • Nunes RR
        • Chave IMM
        • Alencar JCG de
        • Franco SB
        • Oliveira YGBR de
        • Menezes DGA de
        Bispectral index and other processed parameters of electroencephalogram: an update.
        Rev Bras Anestesiol. 2012; 62: 111-117https://doi.org/10.1590/s0034-70942012000100014
        • Chi SI
        • Kim HJ
        • Seo K-S
        • Yang M
        • Chang J
        Use of ADMSTM during sedation for dental treatment of an intellectually disabled patient: a case report.
        J Dent Anesth Pain Med. 2016; 16: 217https://doi.org/10.17245/jdapm.2016.16.3.217
        • Hernández-Hernández MA
        • Fernández-Torre JL.
        Color density spectral array of bilateral bispectral index system: Electroencephalographic correlate in comatose patients with nonconvulsive status epilepticus.
        Seizure. 2016; 34: 18-25https://doi.org/10.1016/j.seizure.2015.11.001
        • Wildes TS
        • Winter AC
        • Maybrier HR
        • Mickle AM
        • Lenze EJ
        • Stark S
        • et al.
        Protocol for the electroencephalography guidance of anesthesia to alleviate geriatric syndromes (ENGAGES) study: a pragmatic, randomised clinical trial.
        BMJ Open. 2016; 6e011505https://doi.org/10.1136/bmjopen-2016-011505
        • Ntahe A.
        Early Diagnosis of nonconvulsive status epilepticus recurrence with raw EEG of a bispectral index monitor.
        Case Rep Crit Care. 2018; 2018: 1-4https://doi.org/10.1155/2018/1208401
        • Fernández-Torre JL
        • Hernández-Hernández MA.
        Utility of bilateral Bispectral index (BIS) monitoring in a comatose patient with focal nonconvulsive status epilepticus.
        Seizure. 2012; 21: 61-64https://doi.org/10.1016/j.seizure.2011.09.001
        • Smith M
        • Dobbs P
        • Eapen G.
        Abnormal bispectral index values associated with the presence of periodic lateralized epileptiform discharges.
        J Neurosurg Anesthesiol. 2015; 27: 73-74https://doi.org/10.1097/ANA.0000000000000069
        • Dahaba A
        • Lui D
        • Metzler H.
        Bispectral index (BIS) monitoring of acute encephalitis with refractory, repetitive partial seizures (AERRPS).
        Minerva Anastesiol. 2010; 76: 298-301
        • Chamorro C
        • Romera MA
        • Balandín B
        • Valdivia M.
        Nonconvulsive status and bispectral index.
        Crit Care Med. 2008; 36: 2218-2219https://doi.org/10.1097/CCM.0b013e31817c473c
        • Ohshima N
        • Chinzei M
        • Mizuno K
        • Hayashida M
        • Kitamura T
        • Shibuya H
        • et al.
        Transient decreases in Bispectral Index without associated changes in the level of consciousness during photic stimulation in an epileptic patient.
        Br J Anaesth. 2007; 98: 100-104https://doi.org/10.1093/bja/ael309
        • Tallach RE
        • Ball DR
        • Jefferson P.
        Monitoring seizures with the Bispectral index.
        Anaesthesia. 2004; 59: 1033-1034https://doi.org/10.1111/j.1365-2044.2004.03953.x
        • Jaggi P
        • Schwabe MJ
        • Gill K
        • Horowitz IN.
        Use of an anesthesia cerebral monitor bispectral index to assess burst-suppression in pentobarbital coma.
        Pediatr Neurol. 2003; 28: 219-222https://doi.org/10.1016/S0887-8994(02)00633-1
        • Haesen J
        • Eertmans W
        • Genbrugge C
        • Meex I
        • Demeestere J
        • Vander Laenen M
        • et al.
        The validation of simplified EEG derived from the bispectral index monitor in post-cardiac arrest patients.
        Resuscitation. 2018; 126: 179-184https://doi.org/10.1016/j.resuscitation.2018.01.042
        • You KM
        • Suh GJ
        • Kwon WY
        • Kim KS
        • Ko SB
        • Park MJ
        • et al.
        Epileptiform discharge detection with the 4-channel frontal electroencephalography during post-resuscitation care.
        Resuscitation. 2017; 117: 8-13https://doi.org/10.1016/j.resuscitation.2017.05.016
        • Musialowicz T
        • Mervaala E
        • Kälviäinen R
        • Uusaro A
        • Ruokonen E
        • Parviainen I.
        Can BIS monitoring be used to assess the depth of propofol anesthesia in the treatment of refractory status epilepticus?.
        Epilepsia. 2010; 51: 1580-1586https://doi.org/10.1111/j.1528-1167.2009.02514.x
        • Prins SA
        • de Hoog M
        • Blok JH
        • Tibboel D
        • Visser GH.
        Continuous noninvasive monitoring of barbiturate coma in critically ill children using the BispectralTM index monitor.
        Crit Care. 2007; 11https://doi.org/10.1186/cc6138
        • Iturri Clavero F
        • Tamayo Medel G
        • de Orte Sancho K
        • González Uriarte A
        • Iglesias Martínez A
        • Martínez Ruíz A
        Use of BIS VISTATM bilateral monitor for diagnosis of intraoperative seizures, a case report.
        Rev Esp Anestesiol Reanim. 2015; 62: 590-595https://doi.org/10.1016/j.redar.2015.03.006
        • Särkelä MOK
        • Ermes MJ
        • Van Gils MJ
        • Yli-Hankala AM
        • Jäntti VH
        • Vakkuri AP.
        Quantification of epileptiform electroencephalographic activity during sevoflurane mask induction.
        Anesthesiology. 2007; 107: 928-938https://doi.org/10.1097/01.anes.0000291444.68894.ee
        • Berger-Estilita J
        • Steck K
        • Vetter C
        • Seidel K
        • Krejci V
        • Hight D
        • et al.
        A case report of several intraoperative convulsions while using the Narcotrend monitor: Significance and predictive use.
        Med (United States). 2019; 98https://doi.org/10.1097/MD.0000000000018004
        • Kim H
        • Kim SY.
        Pitfall of bispectral index during intraoperative seizure -a case report.
        Korean J Anesthesiol. 2013; 65: 449-452https://doi.org/10.4097/kjae.2013.65.5.449
        • Hamada S
        • Laloë PA
        • Hausser-Hauw C
        • Fischler M.
        Seizure after aortic clamp release: a bispectral index pitfall.
        J Cardiothorac Vasc Anesth. 2008; 22: 119-121https://doi.org/10.1053/j.jvca.2007.01.016
        • Chinzei M
        • Sawamura S
        • Hayashida M
        • Kitamura T
        • Tamai H
        • Hanaoka K.
        Change in Bispectral Index during epileptiform electrical activity under sevoflurane anesthesia in a patient with epilepsy.
        Anesth Analg. 2004; 98: 1734-1736https://doi.org/10.1213/01.ANE.0000117282.72866.26
        • Galante D
        • Fortarezza D
        • Caggiano M
        • de Francisci G
        • Pedrotti D
        • Caruselli M.
        Correlation of bispectral index (BIS) monitoring and end-tidal sevoflurane concentration in a patient with lobar holoprosencephaly.
        Braz J Anesthesiol (English Ed. 2015; 65: 379-383https://doi.org/10.1016/j.bjane.2014.07.003
        • Ogawa S
        • Okutani R
        • Nakada K
        • Suehiro K
        • Shigemoto T.
        Spike-monitoring of anaesthesia for corpus callosotomy using bilateral bispectral index.
        Anaesthesia. 2009; 64: 776-780https://doi.org/10.1111/j.1365-2044.2009.05917.x
        • Stasiowski MJ
        • Marciniak R
        • Duława A
        • Krawczyk L
        • Jałowiecki P.
        Epileptiform EEG patterns during different techniques of induction of general anaesthesia with sevoflurane and propofol: A randomised trial.
        Anaesthesiol Intensive Ther. 2019; 51: 21-34https://doi.org/10.5603/AIT.a2019.0003
        • Jung HJ
        • Lee JM.
        High BIS and low rSO2 during CPB: seizure?.
        J Anesth. 2018; 32: 786https://doi.org/10.1007/s00540-018-2530-8
        • Elgueta MF
        • Vega P
        • Lema G
        • Clede L.
        Should we monitor with bispectral index in all patients at high risk for seizures in the operating room?.
        Rev Esp Anestesiol Reanim. 2013; 60: 469-471https://doi.org/10.1016/j.redar.2012.06.014
        • Bousselmi R
        • Lebbi A
        • Ferjani M.
        Bispectral index changes during generalised tonic-clonic seizures.
        Anaesthesia. 2013; 68: 1084-1085https://doi.org/10.1111/anae.12427
        • Bruhn J
        • Bouillon TW
        • Shafer SL.
        Electromyographic activity falsely elevates the bispectral index.
        Anesthesiology. 2000; 92: 1485-1487https://doi.org/10.1097/00000542-200005000-00042
        • Donnelly B
        • Boyd V.
        Use of the bispectral index (BIS) monitor to aid in the diagnosis of pseudoseizures [4].
        Br J Anaesth. 2006; 96: 538-539https://doi.org/10.1093/bja/ael041
        • Sartorius A
        • Schmahl C.
        Bispectral index monitoring during dissociative pseudo-seizure.
        World J Biol Psychiatry. 2009; 10: 603-605https://doi.org/10.1080/15622970701530933
        • Burkle CM
        • Jankowski CJ
        • Torsher LC
        • Rho EH
        • Degnim AC.
        Bis monitor findings during self-hypnosis.
        J Clin Monit Comput. 2005; 19: 391-393https://doi.org/10.1007/s10877-005-6539-9
        • Sleigh JW
        • Andrzejowski J
        • Steyn-Ross A
        • Steyn-Ross M.
        The bispectral index: a measure of depth of sleep?.
        Anesth Analg. 1999; 88: 659-661https://doi.org/10.1097/00000539-199903000-00035
        • Obara S
        • Kakinouchi K
        • Honda J
        • Noji Y
        • Hanayama C
        • Murakawa M.
        Dexmedetomidine administration in a patient with status epilepticus under color density spectral array monitoring.
        JA Clin Rep. 2019; 5: 12https://doi.org/10.1186/s40981-019-0234-1
        • Gunawardane PO
        • Murphy PA
        • Sleigh JW.
        Bispectral index monitoring during electroconvulsive therapy under propofol anaesthesia.
        Br J Anaesth. 2002; 88: 184-187https://doi.org/10.1093/bja/88.2.184
        • Kayser S
        • Bewernick BH
        • Soehle M
        • Switala C
        • Gippert SM
        • Dreimueller N
        • et al.
        Degree of postictal suppression depends on seizure induction time in magnetic seizure therapy and electroconvulsive therapy.
        J ECT. 2017; 33: 167-175https://doi.org/10.1097/YCT.0000000000000425
        • Lemmens HJM
        • Levi DC
        • Debattista C
        • Brock-Utne JG.
        The timing of electroconvulsive therapy and bispectral index after anesthesia induction using different drugs does not affect seizure duration.
        J Clin Anesth. 2003; 15: 29-32https://doi.org/10.1016/S0952-8180(02)00477-4
        • Ochiai R
        • Yamada T
        • Kiyama S
        • Nakaoji T
        • Takeda J.
        Bispectral index as an indicator of seizure inducibility in electroconvulsive therapy under thiopental anesthesia.
        Anesth Analg. 2004; 98: 1030-1035https://doi.org/10.1213/01.ANE.0000105874.50605.3C
        • Pekel M
        • Postaci NA
        • Aytaç İ
        • Karasu D
        • Keleş H
        • Şen Ö
        • et al.
        Sevoflurane versus propofol for electroconvulsive therapy: effects on seizure parameters, anesthesia recovery, and the bispectral index.
        Turkish J Med Sci. 2016; 46: 756-763https://doi.org/10.3906/sag-1502-110
        • Soehle M
        • Kayser S
        • Ellerkmann RK
        • Schlaepfer TE.
        Bilateral bispectral index monitoring during and after electroconvulsive therapy compared with magnetic seizure therapy for treatment-resistant depression.
        Br J Anaesth. 2014; 112: 695-702https://doi.org/10.1093/bja/aet410
        • Kranaster L
        • Hoyer C
        • Janke C
        • Sartorius A.
        Bispectral index monitoring and seizure quality optimization in electroconvulsive therapy.
        Pharmacopsychiatry. 2013; 46: 147-150https://doi.org/10.1055/s-0032-1331748
        • White PF
        • Rawal S
        • Recart A
        • Thornton L
        • Litle M
        • Stool L.
        Can the bispectral index be used to predict seizure time and awakening after electroconvulsive therapy?.
        Anesth Analg. 2003; 96: 1636-1639https://doi.org/10.1213/01.ANE.0000066018.13553.08
        • Gombar S
        • Aggarwal D
        • Khanna AK
        • Gombar KK
        • Chavan BS.
        The bispectral electroencephalogram during modified electroconvulsive therapy under propofol anesthesia - relation with seizure duration and awakening.
        J ECT. 2011; 27: 114-118https://doi.org/10.1097/YCT.0b013e3181df4ebb
        • White PF
        • Amos Q
        • Zhang Y
        • Stool L
        • Husain MM
        • Thornton L
        • et al.
        Anesthetic considerations for magnetic seizure therapy: a novel therapy for severe depression.
        Anesth Analg. 2006; 103: 76-80https://doi.org/10.1213/01.ane.0000221182.71648.a3
        • Nishihara F
        • Saito S.
        Pre-ictal bispectral index has a positive correlation with seizure duration during electroconvulsive therapy.
        Anesth Analg. 2002; 94: 1249-1252https://doi.org/10.1097/00000539-200205000-00037
        • Nishihara F
        • Saito S.
        Adjustment of anaesthesia depth using bispectral index prolongs seizure duration in electroconvulsive therapy.
        Anaesth Intensive Care. 2004; 32: 661-665https://doi.org/10.1177/0310057x0403200509
        • Dahaba AA
        • Liu DW
        • Metzler H.
        Bispectral Index (BIS) monitoring of acute encephalitis with refractory, repetitive partial seizures (AERRPS).
        Minerva Anestesiol. 2010; 76: 298-301
        • Ohshima N
        • Chinzei M
        • Mizuno K
        • Hayashida M
        • Kitamura T
        • Shibuya H
        • et al.
        Transient decreases in Bispectral Index without associated changes in the level of consciousness during photic stimulation in an epileptic patient.
        Br J Anaesth. 2007; 98: 100-104https://doi.org/10.1093/bja/ael309