Pediatr Radiol DOI 10.1007/s00247-014-3208-5
ORIGINAL ARTICLE
Utility of CT-compatible EEG electrodes in critically ill children Nicholas S. Abend & Dennis J. Dlugos & Xiaowei Zhu & Erin S. Schwartz
Received: 11 March 2014 / Revised: 29 August 2014 / Accepted: 10 October 2014 # Springer-Verlag Berlin Heidelberg 2014
Abstract Background Electroencephalographic monitoring is being used with increasing frequency in critically ill children who may require frequent and sometimes urgent brain CT scans. Standard metallic disk EEG electrodes commonly produce substantial imaging artifact, and they must be removed and later reapplied when CT scans are indicated. Objective To determine whether conductive plastic electrodes caused artifact that limited CT interpretation. Material and methods We describe a retrospective cohort of 13 consecutive critically ill children who underwent 17 CT scans with conductive plastic electrodes during 1 year. CT images were evaluated by a pediatric neuroradiologist for artifact presence, type and severity. Results All CT scans had excellent quality images without artifact that impaired CT interpretation except for one scan in which improper wire placement resulted in artifact. Conclusion Conductive plastic electrodes do not cause artifact limiting CT scan interpretation and may be used in critically ill children to permit concurrent electroencephalographic monitoring and CT imaging. N. S. Abend : D. J. Dlugos Departments of Neurology and Pediatrics, The Children’s Hospital of Philadelphia, Perelman School of Medicine at the University of Pennsylvania, Philadelphia, PA, USA N. S. Abend (*) CHOP Neurology, CTRC 10th Floor, 3501 Civic Center Blvd., Philadelphia, PA 19104, USA e-mail: [email protected] X. Zhu : E. S. Schwartz Department of Radiology, The Children’s Hospital of Philadelphia, Perelman School of Medicine at the University of Pennsylvania, Philadelphia, PA, USA
Keywords Computed tomography . Electroencephalogram . Electrodes . Children
Introduction Continuous electroencephalographic monitoring (cEEG) of critically ill children is increasing [1] and has been shown to impact management for the majority of children who are monitored [2]. Non-convulsive seizures have been reported in 10–50% of critically ill children with acute encephalopathy [3], and high electrographic seizures burden has been associated with worse outcome even after adjustment for potential confounders related to acute encephalopathy aetiology and critical illness severity [3, 4]. Most electrographic seizures do not have any clinical correlate so identification requires cEEG and not merely close clinical observation [3]. These same critically ill children at risk for electrographic seizures may require frequent and urgent CT scans. Standard metallic disk EEG electrodes produce substantial beamhardening CT artifact that may reduce definitive imagingbased diagnosis and necessitate repeat transport of critically ill patients for neuroimaging. To avoid such artifact, EEG electrodes must be removed and later reapplied when CT scans are needed. Applying and removing EEG electrodes require specialized technologists and consumes time. At many centers, EEG technologists are not routinely available inhouse, so EEG electrodes may not be applied or reapplied, potentially leaving electrographic seizures undetected [1]. Additionally, the scalp scrubbing required for repeated application and removal of electrodes increases the risk of scalp injury and/or infection. These problems may limit the ability of critically ill children to undergo neuroimaging during cEEG and may make physicians less likely to obtain cEEG when they suspect urgent CT scans may be needed.
Pediatr Radiol
Conductive plastic electrodes are non-metallic, except for a thin silver epoxide coating. They could permit more efficient concurrent cEEG and CT imaging, but their impact on CT image quality has not been assessed in children. We aimed to determine whether conductive plastic electrodes cause artifacts limiting CT interpretation among a cohort of consecutive critically ill children who underwent CT scans with conductive plastic electrodes.
Materials and methods The Children’s Hospital of Philadelphia Institutional Review Board approved this study. As part of clinical practice, conductive plastic electrodes were used for all critically ill children undergoing clinically indicated cEEG between August 2010 and August 2011. The scalp near the electrodes was assessed daily per standard clinical care by EEG technologists. Subsequently, the electronic medical records of patients who underwent CT imaging with electrodes in place were reviewed for patient age and gender, diagnosis leading to ICU admission, indication for cEEG monitoring and indication for CT scanning. Twenty-one conductive plastic scalp surface electrodes (Ives EEG Solutions Inc., Manotick, Canada) were positioned according to the international 10–20 system and affixed with collodion adhesive. EEG data were acquired on a portable bedside computer networked to the hospital’s EEG server. Nursing staff or EEG technologists disconnected the harness from the EEG head box and reconnected it following CT imaging. EEG tracings were reviewed by a single pediatric electroencephalographer (N.S.A., 5 years of post-training experience). CT scans were performed with a Siemens Sensation 64 CT system (Erlangen, Germany) or a Neurologica Ceratome Portable CT system (Danvers, MA) using 5-mm contiguous axial whole-brain imaging or 0.75-mm helical contrast-enhanced vascular imaging. CT images were evaluated by a single pediatric neuroradiologist (E.S.S., 15 years of post-training experience) and scored regarding: (1) imaging findings, (2) artifact presence, (3) artifact type and (4) artifact severity. Type and severity were scored as (1) none, (2) minimal artifact from a single electrode but CT interpretable, (3) mild artifact from more than one electrode but CT interpretable, (4) moderate artifact from more than one electrode or other source but CT remained interpretable, and (5) substantial artifact from more than one electrode or other source and CT not interpretable.
Results Thirteen subjects underwent 17 CT scans with conductive plastic electrodes, including 15 whole-brain CTs, one
intracranial CT venogram and one intracranial CT angiogram. Clinical, electroencephalographic and radiologic data are shown in Table 1. The median age was 10 years (range: 2– 28 years), there were 9 males, and the most common diagnoses leading to ICU admission were hypoxic-ischemic encephalopathy, traumatic brain injury and epilepsy-related seizures. Clinical, electroencephalographic and radiologic data are shown in Table 1. There was no skin injury related to the electrodes. During a 6-hour period in which imaging occurred for each subject, 3 of the 13 subjects (23%) experienced electrographic seizures, and in two subjects these seizures constituted nonconvulsive status epilepticus. No scan had electrode-related artifact that impacted CT interpretation. Eight scans (seven whole-brain and one CT angiogram) had minimal or mild artifact that did not interfere with CT interpretation while eight scans had no artifact whatsoever (seven whole-brain and one CT venogram). One scan had moderate artifact that impacted CT interpretation, which was related to wires that were not disconnected during scanning and not the electrodes. While statistical analysis is not appropriate given our limited sample size, it appears that those scans performed at lower doses in younger children and/or those performed with a portable CT scanner were more likely to have artifact, albeit none that interfered with image interpretation. CT findings are provided in Table 1 and CT image examples are provided in Fig. 1.
Discussion In a retrospective cohort of critically ill children who underwent CT imaging with conductive plastic electrodes in place, the electrodes did not produce scalp injury and did not impair interpretation of CT images. Seizures including nonconvulsive status epilepticus occurred in 23% of subjects in the peri-imaging 6 h indicating that had clinicians decided to avoid EEG monitoring given the potential need for CT imaging the seizures may not have been identified and managed. These findings are consistent with studies investigating the use of electrodes in critically ill adults. One study described 23 critically ill adults who underwent CT scans with conductive plastic electrodes, and all had excellent image quality without significant CT artifacts [5]. A second study described no artifacts in one critically ill adult who underwent CT with conductive plastic electrodes [6]. To date, assessments of plastic electrodes had not occurred in children, and this led to our institution’s requirement that implementation be carried out along with assessment of the electrode’s impact on image quality. These data confirm the safety and utility of conductive plastic electrodes as an imaging-compatible electrode system in critically ill children.
16
28
8
10
1.5
11
2
14
5 10 6 2
2
3
4
5
6
7
8
9
10 11 12 13
Epilepsy Epilepsy exacerbation Encephalitis Epilepsy
ΔMS s/p convulsive seizures ΔMS ΔMS s/p convulsive seizures ΔMS s/p convulsive seizures 5 5 5 5
5
5
5
5 5
5
5
36.51 42.59 48.68 36.51
41.26
35.36
42.59
35.36 35.36
42.59
36.51
120/300 120/350 120/400 120/300
120/14*
120/12*
120/350
120/12* 120/12*
120/350
120/300
100/200 120/300
38.91 36.51
2 3 (from incorrectly placed wires, not electrodes) 0
2
0
0 0
1
1 0 1
Mod volume loss, basal ganglia mineralization Edema, IVH (electrodes 2 mentioned in report) Edema, craniectomy, hemorrhage 2 (artifact more prominent at craniectomy site) Normal 0 Normal 0 Cerebral edema 0 Brain normal, bilateral mastoid 2 opacification
Normal Cerebral edema Extensive cerebral edema, herniation Subdural hematoma, ventriculomegaly, shunt No sinovenous thrombosis Fracture, contusion, hemorrhage, shunt Interval craniotomy, edema, hemorrhage Diffuse, profound cerebral edema Normal Normal
Artifact severity b
b Artifact scored as 0=none, 1=minimal artifact from a single electrode but CT interpretable, 2=mild artifact from more than one electrode but CT interpretable, 3=moderate artifact from more than one electrode or other source but CT interpretable, and 4=substantial artifact from more than one electrode or other source and CT not interpretable
Some scans were performed on portable scanners (denoted with *) and have shorter rotation times (typically 2 s rather than 0.5 s for conventional scanners) and lower mA (since the gantry is 31.8 cm vs. 78 cm)
a
ΔMS altered mental status, HIE hypoxic-ischemic encephalopathy, CTA CT angiography, CTV CT venography, NCS non-convulsive seizure, NCSE non-convulsive status epilepticus, TBI traumatic brain injury
14 15 16 17
13
TBI
ΔMS
11
9 10
12
HIE - Cardiac Arrest
ΔMS
8
ΔMS s/p convulsive seizures Epilepsy exacerbation with sickle cell crisis ΔMS TBI
HIE - Cardiac Arrest
7
120/400
100/200 120/350 120/400
48.68
19.93 38.72 48.68
Slice thickness Dose CTDIvol CT parameters CT impression (mm) (mGy) (kVp/mAs) a
1 (CTA) 0.75 2 5 3 5
Scan
VPS Malfunction, 4 5 Symptomatic Seizures 5 (CTV) 0.75 TBI 6 5
ΔMS
ΔMS
Altered mental status s/p convulsive seizures
Altered mental status
HIE - Cardiac Arrest
HIE - Cardiac Arrest
15
1
Altered mental status
Acute encephalopathy etiology
Subject Age EEG monitoring (years) indication
Table 1 Clinical, EEG and imaging characteristics
Pediatr Radiol
Pediatr Radiol
Fig. 1 CT image of a 1.5-year-old girl following cardiac arrest undergoing EEG monitoring. Image (a) demonstrates artifact attributable not to the leads, but rather to the wires that inadvertently remained connected.
Image (b) demonstrates mild beam-hardening artifact under electrodes. Image (c) from a different patient demonstrates extensive artifact from metal scalp electrodes that nearly completely obscures the brain
There is increasing evidence that critically ill children need to undergo cEEG during the same acute time period when urgent neuroimaging may be required. In one study of 100 critically ill children who underwent cEEG, clinical changes include initiating or escalating anticonvulsants due to seizure identification in 43%, demonstration that a specific event was not seizure in 21%, and obtaining urgent neuroimaging that impacted clinical management in 3% [2]. Studies have reported that non-convulsive seizures and non-convulsive status epilepticus occur in 10-50% of critically ill children [3]. The largest of these studies was a recent retrospective multicenter observational study of cEEG that identified electrographic seizures in 30% of 550 critically ill children. The seizure burden was high enough to constitute electrographic status epilepticus in 38%. Further, only 35% of subjects had exclusively clinical-evident seizures indicating the majority of seizures would not be identified without cEEG [3]. Identifying these seizures may be important since several studies in critically ill children have reported an association between electrographic status epilepticus and worse long-term outcome [4]. While it remains unclear to what extent electrographic seizures are a biomarker of brain injury vs. a cause of secondary brain injury, several recent studies found that a high seizure burden was associated with worse clinical outcome, even after adjustment for potential confounders related to acute encephalopathy aetiology and critical illness severity [3, 4]. If seizures are to be identified, prolonged cEEG monitoring is generally needed, and may therefore span a time period when urgent neuroimaging may be required. A number of observational studies have indicated that only about 50% of patients with seizures are identified in the first hour of cEEG while 90% are identified in the first 24 h of monitoring [7]. Critically ill patients undergoing cEEG often have underlying acute encephalopathies that require urgent and sometimes repeat CT scans. For example, children with traumatic
brain injury often require CT scans, and have also been identified as being at increased risk for having seizures identified during cEEG [3]. Given that we have demonstrated conductive plastic electrodes produced no problematic CT artifact, there are several advantages to their use in critically ill children. First, removing EEG electrodes that are attached with adhesives such as collodion may take an EEG technologist or nurse 10–20 min, leading to a delay in CT imaging. Second, since technologists are not readily available 24 h per day at many centers [1], if electrodes are removed for CT imaging they may not be reapplied quickly, leading to long interruptions in cEEG. Seizures during this period may not be identified. A study of MRI-compatible electrodes in critically ill adults reported that 45% of patients with an MRI scan completed during the time technologists were unavailable at night also had seizures during this period, which would have been missed had patients undergone MRI scan that precluded the next several hours of cEEG [8]. Third, reapplication of electrodes may increase the risk of scalp abrasions. Fourth, both adhesives like collodion used to apply EEG electrodes and the acetone used to remove collodion are noxious to the patient, family and staff. Specialized rooms for application and removal of electrodes are often available in epilepsy monitoring units but cannot be used for critically ill patients where application and removal generally occur within the intensive care unit. Reducing the number of EEG electrode applications and removals reduces exposure. Fifth, physicians may defer imaging in patients undergoing cEEG, potentially leading to delays in obtaining useful radiographic data. In a study of critically ill adults, twice as many patients with MRIcompatible electrodes underwent MRI compared to those with MRI-incompatible electrodes [8], suggesting physicians may have been deferring imaging to continue EEG monitoring. Finally, in current practice, there may be delays in obtaining cEEG if a patient is considered likely to need CT imaging.
Pediatr Radiol
Studies have shown that about half of seizures are identified in the first 1 h of EEG monitoring [7] suggesting that even short delays due to these logistic concerns may lead to delay in seizure diagnosis in a substantial number of patients. Further study is needed to evaluate the safety and effectiveness of MRI-compatible electrodes in this population.
Conclusion Critically ill children with acute encephalopathy may benefit from cEEG during the same time period when they may also require urgent CT imaging. Conductive plastic electrodes produced no scalp injury and did not produce problematic CT artifact as long as the wires are disconnected appropriately.
Acknowledgements Dr. Abend is supported by a grant from the National Institutes of Health (K23NS076550). Conflicts of interest None
References 1. Sanchez SM, Carpenter J, Chapman KE et al (2013) Pediatric ICU EEG monitoring: current resources and practice in the United States and Canada. J Clin Neurophysiol 30:156–160 2. Abend NS, Topjian AA, Gutierrez-Colina AM et al (2011) Impact of continuous EEG monitoring on clinical management in critically ill children. Neurocrit Care 15:70–75 3. Abend NS, Arndt DH, Carpenter JL et al (2013) Electrographic seizures in pediatric ICU patients: cohort study of risk factors and mortality. Neurology 81:383–391 4. Wagenman KL, Blake TP, Sanchez SM et al (2014) Electrographic status epilepticus and long-term outcome in critically ill children. Neurology 4:396–404 5. Das RR, Lucey BP, Chou SH et al (2009) The utility of conductive plastic electrodes in prolonged ICU EEG monitoring. Neurocrit Care 10:368–372 6. Vulliemoz S, Perrig S, Pellise D et al (2009) Imaging compatible electrodes for continuous electroencephalogram monitoring in the intensive care unit. J Clin Neurophysiol 26:236–243 7. Abend NS, Gutierrez-Colina AM, Topjian AA et al (2011) Nonconvulsive seizures are common in critically ill children. Neurology 76:1071–1077 8. Mirsattari SM, Davies-Schinkel C, Young GB et al (2009) Usefulness of a 1.5 T MRI-compatible EEG electrode system for routine use in the intensive care unit of a tertiary care hospital. Epilepsy Res 84:28–32
ORIGINAL ARTICLE
Utility of CT-compatible EEG electrodes in critically ill children Nicholas S. Abend & Dennis J. Dlugos & Xiaowei Zhu & Erin S. Schwartz
Received: 11 March 2014 / Revised: 29 August 2014 / Accepted: 10 October 2014 # Springer-Verlag Berlin Heidelberg 2014
Abstract Background Electroencephalographic monitoring is being used with increasing frequency in critically ill children who may require frequent and sometimes urgent brain CT scans. Standard metallic disk EEG electrodes commonly produce substantial imaging artifact, and they must be removed and later reapplied when CT scans are indicated. Objective To determine whether conductive plastic electrodes caused artifact that limited CT interpretation. Material and methods We describe a retrospective cohort of 13 consecutive critically ill children who underwent 17 CT scans with conductive plastic electrodes during 1 year. CT images were evaluated by a pediatric neuroradiologist for artifact presence, type and severity. Results All CT scans had excellent quality images without artifact that impaired CT interpretation except for one scan in which improper wire placement resulted in artifact. Conclusion Conductive plastic electrodes do not cause artifact limiting CT scan interpretation and may be used in critically ill children to permit concurrent electroencephalographic monitoring and CT imaging. N. S. Abend : D. J. Dlugos Departments of Neurology and Pediatrics, The Children’s Hospital of Philadelphia, Perelman School of Medicine at the University of Pennsylvania, Philadelphia, PA, USA N. S. Abend (*) CHOP Neurology, CTRC 10th Floor, 3501 Civic Center Blvd., Philadelphia, PA 19104, USA e-mail: [email protected] X. Zhu : E. S. Schwartz Department of Radiology, The Children’s Hospital of Philadelphia, Perelman School of Medicine at the University of Pennsylvania, Philadelphia, PA, USA
Keywords Computed tomography . Electroencephalogram . Electrodes . Children
Introduction Continuous electroencephalographic monitoring (cEEG) of critically ill children is increasing [1] and has been shown to impact management for the majority of children who are monitored [2]. Non-convulsive seizures have been reported in 10–50% of critically ill children with acute encephalopathy [3], and high electrographic seizures burden has been associated with worse outcome even after adjustment for potential confounders related to acute encephalopathy aetiology and critical illness severity [3, 4]. Most electrographic seizures do not have any clinical correlate so identification requires cEEG and not merely close clinical observation [3]. These same critically ill children at risk for electrographic seizures may require frequent and urgent CT scans. Standard metallic disk EEG electrodes produce substantial beamhardening CT artifact that may reduce definitive imagingbased diagnosis and necessitate repeat transport of critically ill patients for neuroimaging. To avoid such artifact, EEG electrodes must be removed and later reapplied when CT scans are needed. Applying and removing EEG electrodes require specialized technologists and consumes time. At many centers, EEG technologists are not routinely available inhouse, so EEG electrodes may not be applied or reapplied, potentially leaving electrographic seizures undetected [1]. Additionally, the scalp scrubbing required for repeated application and removal of electrodes increases the risk of scalp injury and/or infection. These problems may limit the ability of critically ill children to undergo neuroimaging during cEEG and may make physicians less likely to obtain cEEG when they suspect urgent CT scans may be needed.
Pediatr Radiol
Conductive plastic electrodes are non-metallic, except for a thin silver epoxide coating. They could permit more efficient concurrent cEEG and CT imaging, but their impact on CT image quality has not been assessed in children. We aimed to determine whether conductive plastic electrodes cause artifacts limiting CT interpretation among a cohort of consecutive critically ill children who underwent CT scans with conductive plastic electrodes.
Materials and methods The Children’s Hospital of Philadelphia Institutional Review Board approved this study. As part of clinical practice, conductive plastic electrodes were used for all critically ill children undergoing clinically indicated cEEG between August 2010 and August 2011. The scalp near the electrodes was assessed daily per standard clinical care by EEG technologists. Subsequently, the electronic medical records of patients who underwent CT imaging with electrodes in place were reviewed for patient age and gender, diagnosis leading to ICU admission, indication for cEEG monitoring and indication for CT scanning. Twenty-one conductive plastic scalp surface electrodes (Ives EEG Solutions Inc., Manotick, Canada) were positioned according to the international 10–20 system and affixed with collodion adhesive. EEG data were acquired on a portable bedside computer networked to the hospital’s EEG server. Nursing staff or EEG technologists disconnected the harness from the EEG head box and reconnected it following CT imaging. EEG tracings were reviewed by a single pediatric electroencephalographer (N.S.A., 5 years of post-training experience). CT scans were performed with a Siemens Sensation 64 CT system (Erlangen, Germany) or a Neurologica Ceratome Portable CT system (Danvers, MA) using 5-mm contiguous axial whole-brain imaging or 0.75-mm helical contrast-enhanced vascular imaging. CT images were evaluated by a single pediatric neuroradiologist (E.S.S., 15 years of post-training experience) and scored regarding: (1) imaging findings, (2) artifact presence, (3) artifact type and (4) artifact severity. Type and severity were scored as (1) none, (2) minimal artifact from a single electrode but CT interpretable, (3) mild artifact from more than one electrode but CT interpretable, (4) moderate artifact from more than one electrode or other source but CT remained interpretable, and (5) substantial artifact from more than one electrode or other source and CT not interpretable.
Results Thirteen subjects underwent 17 CT scans with conductive plastic electrodes, including 15 whole-brain CTs, one
intracranial CT venogram and one intracranial CT angiogram. Clinical, electroencephalographic and radiologic data are shown in Table 1. The median age was 10 years (range: 2– 28 years), there were 9 males, and the most common diagnoses leading to ICU admission were hypoxic-ischemic encephalopathy, traumatic brain injury and epilepsy-related seizures. Clinical, electroencephalographic and radiologic data are shown in Table 1. There was no skin injury related to the electrodes. During a 6-hour period in which imaging occurred for each subject, 3 of the 13 subjects (23%) experienced electrographic seizures, and in two subjects these seizures constituted nonconvulsive status epilepticus. No scan had electrode-related artifact that impacted CT interpretation. Eight scans (seven whole-brain and one CT angiogram) had minimal or mild artifact that did not interfere with CT interpretation while eight scans had no artifact whatsoever (seven whole-brain and one CT venogram). One scan had moderate artifact that impacted CT interpretation, which was related to wires that were not disconnected during scanning and not the electrodes. While statistical analysis is not appropriate given our limited sample size, it appears that those scans performed at lower doses in younger children and/or those performed with a portable CT scanner were more likely to have artifact, albeit none that interfered with image interpretation. CT findings are provided in Table 1 and CT image examples are provided in Fig. 1.
Discussion In a retrospective cohort of critically ill children who underwent CT imaging with conductive plastic electrodes in place, the electrodes did not produce scalp injury and did not impair interpretation of CT images. Seizures including nonconvulsive status epilepticus occurred in 23% of subjects in the peri-imaging 6 h indicating that had clinicians decided to avoid EEG monitoring given the potential need for CT imaging the seizures may not have been identified and managed. These findings are consistent with studies investigating the use of electrodes in critically ill adults. One study described 23 critically ill adults who underwent CT scans with conductive plastic electrodes, and all had excellent image quality without significant CT artifacts [5]. A second study described no artifacts in one critically ill adult who underwent CT with conductive plastic electrodes [6]. To date, assessments of plastic electrodes had not occurred in children, and this led to our institution’s requirement that implementation be carried out along with assessment of the electrode’s impact on image quality. These data confirm the safety and utility of conductive plastic electrodes as an imaging-compatible electrode system in critically ill children.
16
28
8
10
1.5
11
2
14
5 10 6 2
2
3
4
5
6
7
8
9
10 11 12 13
Epilepsy Epilepsy exacerbation Encephalitis Epilepsy
ΔMS s/p convulsive seizures ΔMS ΔMS s/p convulsive seizures ΔMS s/p convulsive seizures 5 5 5 5
5
5
5
5 5
5
5
36.51 42.59 48.68 36.51
41.26
35.36
42.59
35.36 35.36
42.59
36.51
120/300 120/350 120/400 120/300
120/14*
120/12*
120/350
120/12* 120/12*
120/350
120/300
100/200 120/300
38.91 36.51
2 3 (from incorrectly placed wires, not electrodes) 0
2
0
0 0
1
1 0 1
Mod volume loss, basal ganglia mineralization Edema, IVH (electrodes 2 mentioned in report) Edema, craniectomy, hemorrhage 2 (artifact more prominent at craniectomy site) Normal 0 Normal 0 Cerebral edema 0 Brain normal, bilateral mastoid 2 opacification
Normal Cerebral edema Extensive cerebral edema, herniation Subdural hematoma, ventriculomegaly, shunt No sinovenous thrombosis Fracture, contusion, hemorrhage, shunt Interval craniotomy, edema, hemorrhage Diffuse, profound cerebral edema Normal Normal
Artifact severity b
b Artifact scored as 0=none, 1=minimal artifact from a single electrode but CT interpretable, 2=mild artifact from more than one electrode but CT interpretable, 3=moderate artifact from more than one electrode or other source but CT interpretable, and 4=substantial artifact from more than one electrode or other source and CT not interpretable
Some scans were performed on portable scanners (denoted with *) and have shorter rotation times (typically 2 s rather than 0.5 s for conventional scanners) and lower mA (since the gantry is 31.8 cm vs. 78 cm)
a
ΔMS altered mental status, HIE hypoxic-ischemic encephalopathy, CTA CT angiography, CTV CT venography, NCS non-convulsive seizure, NCSE non-convulsive status epilepticus, TBI traumatic brain injury
14 15 16 17
13
TBI
ΔMS
11
9 10
12
HIE - Cardiac Arrest
ΔMS
8
ΔMS s/p convulsive seizures Epilepsy exacerbation with sickle cell crisis ΔMS TBI
HIE - Cardiac Arrest
7
120/400
100/200 120/350 120/400
48.68
19.93 38.72 48.68
Slice thickness Dose CTDIvol CT parameters CT impression (mm) (mGy) (kVp/mAs) a
1 (CTA) 0.75 2 5 3 5
Scan
VPS Malfunction, 4 5 Symptomatic Seizures 5 (CTV) 0.75 TBI 6 5
ΔMS
ΔMS
Altered mental status s/p convulsive seizures
Altered mental status
HIE - Cardiac Arrest
HIE - Cardiac Arrest
15
1
Altered mental status
Acute encephalopathy etiology
Subject Age EEG monitoring (years) indication
Table 1 Clinical, EEG and imaging characteristics
Pediatr Radiol
Pediatr Radiol
Fig. 1 CT image of a 1.5-year-old girl following cardiac arrest undergoing EEG monitoring. Image (a) demonstrates artifact attributable not to the leads, but rather to the wires that inadvertently remained connected.
Image (b) demonstrates mild beam-hardening artifact under electrodes. Image (c) from a different patient demonstrates extensive artifact from metal scalp electrodes that nearly completely obscures the brain
There is increasing evidence that critically ill children need to undergo cEEG during the same acute time period when urgent neuroimaging may be required. In one study of 100 critically ill children who underwent cEEG, clinical changes include initiating or escalating anticonvulsants due to seizure identification in 43%, demonstration that a specific event was not seizure in 21%, and obtaining urgent neuroimaging that impacted clinical management in 3% [2]. Studies have reported that non-convulsive seizures and non-convulsive status epilepticus occur in 10-50% of critically ill children [3]. The largest of these studies was a recent retrospective multicenter observational study of cEEG that identified electrographic seizures in 30% of 550 critically ill children. The seizure burden was high enough to constitute electrographic status epilepticus in 38%. Further, only 35% of subjects had exclusively clinical-evident seizures indicating the majority of seizures would not be identified without cEEG [3]. Identifying these seizures may be important since several studies in critically ill children have reported an association between electrographic status epilepticus and worse long-term outcome [4]. While it remains unclear to what extent electrographic seizures are a biomarker of brain injury vs. a cause of secondary brain injury, several recent studies found that a high seizure burden was associated with worse clinical outcome, even after adjustment for potential confounders related to acute encephalopathy aetiology and critical illness severity [3, 4]. If seizures are to be identified, prolonged cEEG monitoring is generally needed, and may therefore span a time period when urgent neuroimaging may be required. A number of observational studies have indicated that only about 50% of patients with seizures are identified in the first hour of cEEG while 90% are identified in the first 24 h of monitoring [7]. Critically ill patients undergoing cEEG often have underlying acute encephalopathies that require urgent and sometimes repeat CT scans. For example, children with traumatic
brain injury often require CT scans, and have also been identified as being at increased risk for having seizures identified during cEEG [3]. Given that we have demonstrated conductive plastic electrodes produced no problematic CT artifact, there are several advantages to their use in critically ill children. First, removing EEG electrodes that are attached with adhesives such as collodion may take an EEG technologist or nurse 10–20 min, leading to a delay in CT imaging. Second, since technologists are not readily available 24 h per day at many centers [1], if electrodes are removed for CT imaging they may not be reapplied quickly, leading to long interruptions in cEEG. Seizures during this period may not be identified. A study of MRI-compatible electrodes in critically ill adults reported that 45% of patients with an MRI scan completed during the time technologists were unavailable at night also had seizures during this period, which would have been missed had patients undergone MRI scan that precluded the next several hours of cEEG [8]. Third, reapplication of electrodes may increase the risk of scalp abrasions. Fourth, both adhesives like collodion used to apply EEG electrodes and the acetone used to remove collodion are noxious to the patient, family and staff. Specialized rooms for application and removal of electrodes are often available in epilepsy monitoring units but cannot be used for critically ill patients where application and removal generally occur within the intensive care unit. Reducing the number of EEG electrode applications and removals reduces exposure. Fifth, physicians may defer imaging in patients undergoing cEEG, potentially leading to delays in obtaining useful radiographic data. In a study of critically ill adults, twice as many patients with MRIcompatible electrodes underwent MRI compared to those with MRI-incompatible electrodes [8], suggesting physicians may have been deferring imaging to continue EEG monitoring. Finally, in current practice, there may be delays in obtaining cEEG if a patient is considered likely to need CT imaging.
Pediatr Radiol
Studies have shown that about half of seizures are identified in the first 1 h of EEG monitoring [7] suggesting that even short delays due to these logistic concerns may lead to delay in seizure diagnosis in a substantial number of patients. Further study is needed to evaluate the safety and effectiveness of MRI-compatible electrodes in this population.
Conclusion Critically ill children with acute encephalopathy may benefit from cEEG during the same time period when they may also require urgent CT imaging. Conductive plastic electrodes produced no scalp injury and did not produce problematic CT artifact as long as the wires are disconnected appropriately.
Acknowledgements Dr. Abend is supported by a grant from the National Institutes of Health (K23NS076550). Conflicts of interest None
References 1. Sanchez SM, Carpenter J, Chapman KE et al (2013) Pediatric ICU EEG monitoring: current resources and practice in the United States and Canada. J Clin Neurophysiol 30:156–160 2. Abend NS, Topjian AA, Gutierrez-Colina AM et al (2011) Impact of continuous EEG monitoring on clinical management in critically ill children. Neurocrit Care 15:70–75 3. Abend NS, Arndt DH, Carpenter JL et al (2013) Electrographic seizures in pediatric ICU patients: cohort study of risk factors and mortality. Neurology 81:383–391 4. Wagenman KL, Blake TP, Sanchez SM et al (2014) Electrographic status epilepticus and long-term outcome in critically ill children. Neurology 4:396–404 5. Das RR, Lucey BP, Chou SH et al (2009) The utility of conductive plastic electrodes in prolonged ICU EEG monitoring. Neurocrit Care 10:368–372 6. Vulliemoz S, Perrig S, Pellise D et al (2009) Imaging compatible electrodes for continuous electroencephalogram monitoring in the intensive care unit. J Clin Neurophysiol 26:236–243 7. Abend NS, Gutierrez-Colina AM, Topjian AA et al (2011) Nonconvulsive seizures are common in critically ill children. Neurology 76:1071–1077 8. Mirsattari SM, Davies-Schinkel C, Young GB et al (2009) Usefulness of a 1.5 T MRI-compatible EEG electrode system for routine use in the intensive care unit of a tertiary care hospital. Epilepsy Res 84:28–32
