1
|
Bitar R, Khan UM, Rosenthal ES. Utility and rationale for continuous EEG monitoring: a primer for the general intensivist. Crit Care 2024; 28:244. [PMID: 39014421 PMCID: PMC11251356 DOI: 10.1186/s13054-024-04986-0] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/06/2024] [Accepted: 06/09/2024] [Indexed: 07/18/2024] Open
Abstract
This review offers a comprehensive guide for general intensivists on the utility of continuous EEG (cEEG) monitoring for critically ill patients. Beyond the primary role of EEG in detecting seizures, this review explores its utility in neuroprognostication, monitoring neurological deterioration, assessing treatment responses, and aiding rehabilitation in patients with encephalopathy, coma, or other consciousness disorders. Most seizures and status epilepticus (SE) events in the intensive care unit (ICU) setting are nonconvulsive or subtle, making cEEG essential for identifying these otherwise silent events. Imaging and invasive approaches can add to the diagnosis of seizures for specific populations, given that scalp electrodes may fail to identify seizures that may be detected by depth electrodes or electroradiologic findings. When cEEG identifies SE, the risk of secondary neuronal injury related to the time-intensity "burden" often prompts treatment with anti-seizure medications. Similarly, treatment may be administered for seizure-spectrum activity, such as periodic discharges or lateralized rhythmic delta slowing on the ictal-interictal continuum (IIC), even when frank seizures are not evident on the scalp. In this setting, cEEG is utilized empirically to monitor treatment response. Separately, cEEG has other versatile uses for neurotelemetry, including identifying the level of sedation or consciousness. Specific conditions such as sepsis, traumatic brain injury, subarachnoid hemorrhage, and cardiac arrest may each be associated with a unique application of cEEG; for example, predicting impending events of delayed cerebral ischemia, a feared complication in the first two weeks after subarachnoid hemorrhage. After brief training, non-neurophysiologists can learn to interpret quantitative EEG trends that summarize elements of EEG activity, enhancing clinical responsiveness in collaboration with clinical neurophysiologists. Intensivists and other healthcare professionals also play crucial roles in facilitating timely cEEG setup, preventing electrode-related skin injuries, and maintaining patient mobility during monitoring.
Collapse
Affiliation(s)
- Ribal Bitar
- Department of Neurology, Massachusetts General Hospital, 55 Fruit St., Lunder 644, Boston, MA, 02114, USA
| | - Usaamah M Khan
- Department of Neurology, Massachusetts General Hospital, 55 Fruit St., Lunder 644, Boston, MA, 02114, USA
| | - Eric S Rosenthal
- Department of Neurology, Massachusetts General Hospital, 55 Fruit St., Lunder 644, Boston, MA, 02114, USA.
| |
Collapse
|
2
|
Parikh H, Sun H, Amerineni R, Rosenthal ES, Volfovsky A, Rudin C, Westover MB, Zafar SF. How many patients do you need? Investigating trial designs for anti-seizure treatment in acute brain injury patients. Ann Clin Transl Neurol 2024; 11:1681-1690. [PMID: 38867375 PMCID: PMC11251465 DOI: 10.1002/acn3.52059] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/21/2023] [Revised: 03/19/2024] [Accepted: 03/21/2024] [Indexed: 06/14/2024] Open
Abstract
BACKGROUND/OBJECTIVES Epileptiform activity (EA), including seizures and periodic patterns, worsens outcomes in patients with acute brain injuries (e.g., aneurysmal subarachnoid hemorrhage [aSAH]). Randomized control trials (RCTs) assessing anti-seizure interventions are needed. Due to scant drug efficacy data and ethical reservations with placebo utilization, and complex physiology of acute brain injury, RCTs are lacking or hindered by design constraints. We used a pharmacological model-guided simulator to design and determine the feasibility of RCTs evaluating EA treatment. METHODS In a single-center cohort of adults (age >18) with aSAH and EA, we employed a mechanistic pharmacokinetic-pharmacodynamic framework to model treatment response using observational data. We subsequently simulated RCTs for levetiracetam and propofol, each with three treatment arms mirroring clinical practice and an additional placebo arm. Using our framework, we simulated EA trajectories across treatment arms. We predicted discharge modified Rankin Scale as a function of baseline covariates, EA burden, and drug doses using a double machine learning model learned from observational data. Differences in outcomes across arms were used to estimate the required sample size. RESULTS Sample sizes ranged from 500 for levetiracetam 7 mg/kg versus placebo, to >4000 for levetiracetam 15 versus 7 mg/kg to achieve 80% power (5% type I error). For propofol 1 mg/kg/h versus placebo, 1200 participants were needed. Simulations comparing propofol at varying doses did not reach 80% power even at samples >1200. CONCLUSIONS Our simulations using drug efficacy show sample sizes are infeasible, even for potentially unethical placebo-control trials. We highlight the strength of simulations with observational data to inform the null hypotheses and propose use of this simulation-based RCT paradigm to assess the feasibility of future trials of anti-seizure treatment in acute brain injury.
Collapse
Affiliation(s)
- Harsh Parikh
- Department of BiostatisticsJohns Hopkins Bloomberg School of Public HealthBaltimoreMarylandUSA
| | - Haoqi Sun
- Department of NeurologyBeth Israel Deaconess Medical CenterBostonMassachusettsUSA
| | - Rajesh Amerineni
- Department of NeurologyMassachusetts General HospitalBostonMassachusettsUSA
| | - Eric S. Rosenthal
- Department of NeurologyMassachusetts General HospitalBostonMassachusettsUSA
| | | | - Cynthia Rudin
- Department of Computer ScienceDuke UniversityDukeNorth CarolinaUSA
| | - M. Brandon Westover
- Department of NeurologyBeth Israel Deaconess Medical CenterBostonMassachusettsUSA
| | - Sahar F. Zafar
- Department of NeurologyMassachusetts General HospitalBostonMassachusettsUSA
| |
Collapse
|
3
|
Parikh H, Sun H, Amerineni R, Rosenthal ES, Volfovsky A, Rudin C, Westover MB, Zafar SF. How Many Patients Do You Need? Investigating Trial Designs for Anti-Seizure Treatment in Acute Brain Injury Patients. MEDRXIV : THE PREPRINT SERVER FOR HEALTH SCIENCES 2023:2023.08.21.23294339. [PMID: 37662339 PMCID: PMC10473786 DOI: 10.1101/2023.08.21.23294339] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Grants] [Track Full Text] [Subscribe] [Scholar Register] [Indexed: 09/05/2023]
Abstract
Objectives Epileptiform activity (EA) worsens outcomes in patients with acute brain injuries (e.g., aneurysmal subarachnoid hemorrhage [aSAH]). Randomized trials (RCTs) assessing anti-seizure interventions are needed. Due to scant drug efficacy data and ethical reservations with placebo utilization, RCTs are lacking or hindered by design constraints. We used a pharmacological model-guided simulator to design and determine feasibility of RCTs evaluating EA treatment. Methods In a single-center cohort of adults (age >18) with aSAH and EA, we employed a mechanistic pharmacokinetic-pharmacodynamic framework to model treatment response using observational data. We subsequently simulated RCTs for levetiracetam and propofol, each with three treatment arms mirroring clinical practice and an additional placebo arm. Using our framework we simulated EA trajectories across treatment arms. We predicted discharge modified Rankin Scale as a function of baseline covariates, EA burden, and drug doses using a double machine learning model learned from observational data. Differences in outcomes across arms were used to estimate the required sample size. Results Sample sizes ranged from 500 for levetiracetam 7 mg/kg vs placebo, to >4000 for levetiracetam 15 vs. 7 mg/kg to achieve 80% power (5% type I error). For propofol 1mg/kg/hr vs. placebo 1200 participants were needed. Simulations comparing propofol at varying doses did not reach 80% power even at samples >1200. Interpretation Our simulations using drug efficacy show sample sizes are infeasible, even for potentially unethical placebo-control trials. We highlight the strength of simulations with observational data to inform the null hypotheses and assess feasibility of future trials of EA treatment.
Collapse
Affiliation(s)
| | - Haoqi Sun
- Beth Israel Deaconess Medical Center, Department of Neurology
| | | | | | | | | | | | | |
Collapse
|
4
|
Ge W, Jing J, An S, Herlopian A, Ng M, Struck AF, Appavu B, Johnson EL, Osman G, Haider HA, Karakis I, Kim JA, Halford JJ, Dhakar MB, Sarkis RA, Swisher CB, Schmitt S, Lee JW, Tabaeizadeh M, Rodriguez A, Gaspard N, Gilmore E, Herman ST, Kaplan PW, Pathmanathan J, Hong S, Rosenthal ES, Zafar S, Sun J, Brandon Westover M. Deep active learning for Interictal Ictal Injury Continuum EEG patterns. J Neurosci Methods 2021; 351:108966. [PMID: 33131680 PMCID: PMC8135050 DOI: 10.1016/j.jneumeth.2020.108966] [Citation(s) in RCA: 3] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/14/2020] [Revised: 09/16/2020] [Accepted: 10/01/2020] [Indexed: 02/02/2023]
Abstract
OBJECTIVES Seizures and seizure-like electroencephalography (EEG) patterns, collectively referred to as "ictal interictal injury continuum" (IIIC) patterns, are commonly encountered in critically ill patients. Automated detection is important for patient care and to enable research. However, training accurate detectors requires a large labeled dataset. Active Learning (AL) may help select informative examples to label, but the optimal AL approach remains unclear. METHODS We assembled >200,000 h of EEG from 1,454 hospitalized patients. From these, we collected 9,808 labeled and 120,000 unlabeled 10-second EEG segments. Labels included 6 IIIC patterns. In each AL iteration, a Dense-Net Convolutional Neural Network (CNN) learned vector representations for EEG segments using available labels, which were used to create a 2D embedding map. Nearest-neighbor label spreading within the embedding map was used to create additional pseudo-labeled data. A second Dense-Net was trained using real- and pseudo-labels. We evaluated several strategies for selecting candidate points for experts to label next. Finally, we compared two methods for class balancing within queries: standard balanced-based querying (SBBQ), and high confidence spread-based balanced querying (HCSBBQ). RESULTS Our results show: 1) Label spreading increased convergence speed for AL. 2) All query criteria produced similar results to random sampling. 3) HCSBBQ query balancing performed best. Using label spreading and HCSBBQ query balancing, we were able to train models approaching expert-level performance across all pattern categories after obtaining ∼7000 expert labels. CONCLUSION Our results provide guidance regarding the use of AL to efficiently label large EEG datasets in critically ill patients.
Collapse
Affiliation(s)
- Wendong Ge
- Massachusetts General Hospital, Department of Neurology, Boston, MA, United States; Harvard Medical School, Boston, MA, United States
| | - Jin Jing
- Massachusetts General Hospital, Department of Neurology, Boston, MA, United States; Harvard Medical School, Boston, MA, United States
| | - Sungtae An
- Georgia Institute of Technology, College of Computing, Atlanta, GA, Georgia
| | | | | | - Aaron F Struck
- University of Wisconsin Madison Department of Neurology, United States
| | - Brian Appavu
- University of Arizona College of Medicine, Phoenix, United States
| | | | | | | | | | | | | | | | | | | | | | | | | | | | - Nicolas Gaspard
- Université Libre de Bruxelles, Hôpital Erasme and Yale University, Belgium
| | - Emily Gilmore
- Yale University, Yale New Haven Hospital, United States
| | - Susan T Herman
- Barrow Neurological Institute, Phoenix, AZ, United States
| | | | | | - Shenda Hong
- Georgia Institute of Technology, College of Computing, Atlanta, GA, Georgia
| | - Eric S Rosenthal
- Massachusetts General Hospital, Department of Neurology, Boston, MA, United States; Harvard Medical School, Boston, MA, United States
| | - Sahar Zafar
- Massachusetts General Hospital, Department of Neurology, Boston, MA, United States; Harvard Medical School, Boston, MA, United States
| | - Jimeng Sun
- University of Illinois at Urbana-Champaign, College of Computing, Champaign, IL, United States
| | - M Brandon Westover
- Massachusetts General Hospital, Department of Neurology, Boston, MA, United States; Harvard Medical School, Boston, MA, United States.
| |
Collapse
|
5
|
Jing J, d'Angremont E, Ebrahim S, Tabaeizadeh M, Ng M, Herlopian A, Dauwels J, Brandon Westover M. Rapid annotation of seizures and interictal-ictal-injury continuum EEG patterns. J Neurosci Methods 2020; 347:108956. [PMID: 33099261 DOI: 10.1016/j.jneumeth.2020.108956] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.3] [Reference Citation Analysis] [Abstract] [Key Words] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/24/2020] [Revised: 09/16/2020] [Accepted: 09/18/2020] [Indexed: 10/23/2022]
Abstract
BACKGROUND Manual annotation of seizures and interictal-ictal-injury continuum (IIIC) patterns in continuous EEG (cEEG) recorded from critically ill patients is a time-intensive process for clinicians and researchers. In this study, we evaluated the accuracy and efficiency of an automated clustering method to accelerate expert annotation of cEEG. NEW METHOD We learned a local dictionary from 97 ICU patients by applying k-medoids clustering to 592 features in the time and frequency domains. We utilized changepoint detection (CPD) to segment the cEEG recordings. We then computed a bag-of-words (BoW) representation for each segment. We further clustered the segments by affinity propagation. EEG experts scored the resulting clusters for each patient by labeling only the cluster medoids. We trained a random forest classifier to assess validity of the clusters. RESULTS Mean pairwise agreement of 62.6% using this automated method was not significantly different from interrater agreements using manual labeling (63.8%), demonstrating the validity of the method. We also found that it takes experts using our method 5.31 ± 4.44 min to label the 30.19 ± 3.84 h of cEEG data, more than 45 times faster than unaided manual review, demonstrating efficiency. COMPARISON WITH EXISTING METHODS Previous studies of EEG data labeling have generally yielded similar human expert interrater agreements, and lower agreements with automated methods. CONCLUSIONS Our results suggest that long EEG recordings can be rapidly annotated by experts many times faster than unaided manual review through the use of an advanced clustering method.
Collapse
Affiliation(s)
- Jin Jing
- Massachusetts General Hospital, Boston, MA, United States; Nanyang Technological University, Singapore, Singapore
| | | | - Senan Ebrahim
- Massachusetts General Hospital, Boston, MA, United States
| | | | - Marcus Ng
- University of Manitoba, Winnipeg, MB, Canada
| | - Aline Herlopian
- Yale University School of Medicine, New Haven, CT, United States
| | | | | |
Collapse
|
6
|
Foreman B. Neurocritical Care: Bench to Bedside (Eds. Claude Hemphill, Michael James) Integrating and Using Big Data in Neurocritical Care. Neurotherapeutics 2020; 17:593-605. [PMID: 32152955 PMCID: PMC7283405 DOI: 10.1007/s13311-020-00846-1] [Citation(s) in RCA: 5] [Impact Index Per Article: 1.3] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 12/19/2022] Open
Abstract
The critical care environment drives huge volumes of data, and clinicians are tasked with quickly processing this data and responding to it urgently. The neurocritical care environment increasingly involves EEG, multimodal intracranial monitoring, and complex imaging which preclude comprehensive human synthesis, and requires new concepts to integrate data into clinical care. By definition, Big Data is data that cannot be handled using traditional infrastructures and is characterized by the volume, variety, velocity, and variability of the data being produced. Big Data in the neurocritical care unit requires rethinking of data storage infrastructures and the development of tools and analytics to drive advancements in the field. Preprocessing, feature extraction, statistical inference, and analytic tools are required in order to achieve the primary goals of Big Data for clinical use: description, prediction, and prescription. Barriers to its use at bedside include a lack of infrastructure development within the healthcare industry, lack of standardization of data inputs, and ultimately existential and scientific concerns about the outputs that result from the use of tools such as artificial intelligence. However, as implied by the fundamental theorem of biomedical informatics, physicians remain central to the development and utility of Big Data to improve patient care.
Collapse
Affiliation(s)
- Brandon Foreman
- Department of Neurology & Rehabilitation Medicine, University of Cincinnati Medical Center, 231 Albert Sabin Way, Cincinnati, OH, 45267-0517, USA.
- Collaborative for Research on Acute Neurological Injuries (CRANI), University of Cincinnati, Cincinnati, OH, USA.
| |
Collapse
|