Prognostic value of early electrographic biomarkers of epileptogenesis in high-risk ischaemic stroke patients

BACKGROUND AND PURPOSE

Post-stroke epilepsy (PSE) is frequent. Better prediction of PSE would enable individualized management and improve trial design for epilepsy prevention. The aim was to assess the complementary value of continuous electroencephalography (EEG) data during the acute phase compared with clinical risk factors currently used to predict PSE.

METHODS

A prospective cohort of 81 patients with ischaemic stroke who received early continuous EEG monitoring was studied to assess the association of early EEG seizures, other highly epileptogenic rhythmic and periodic patterns, and regional attenuation without delta (RAWOD, an EEG pattern of stroke severity) with PSE. Clinical risk factors were investigated using the SeLECT (stroke severity; large-artery atherosclerosis; early clinical seizures; cortical involvement; territory of middle cerebral artery) scores.

RESULTS

Twelve (15%) patients developed PSE. The presence of any of the investigated patterns was associated with a risk of epilepsy of 46%, with a sensitivity and specificity of 83% and 78%. The association remained significant after adjusting for the SeLECT score (odds ratio 18.8, interquartile range 3.8-72.7).

CONCLUSIONS

It was found that highly epileptogenic rhythmic and periodic patterns and RAWOD were associated with the development of PSE and complemented clinical risk factors. These findings indicate that continuous EEG provides useful information to determine patients at higher risk of developing PSE and could help individualize care.

Case report: Invasive neuromonitoring in status epilepticus induced hypoxic ischemic brain injury

Objectives

Literature on invasive neuromonitoring and bilateral decompressive craniectomies (BDC) in patients with refractory status epilepticus (RSE)-mediated hypoxic-ischemic brain injury (HIBI) is limited. Neuromonitoring can guide decision making and treatment escalation.

Methods and results

We report a case of a 17 years-old male who was admitted to our hospital's intensive care unit for RSE. HIBI was detected on neuroimaging on this patient's second day of admission after he developed central diabetes insipidus (DI). Invasive neuromonitoring revealed raised intracranial pressure (ICP) and brain hypoxia as measured by reduced brain tissue oxygen tension (PbtO2). Treatments were escalated in a tiered fashion, including administration of hyperosmolar agents, analgesics, sedatives, and a neuromuscular blocking drug. Eventually, BDC was performed as a salvage therapy as a means of controlling refractory ICP crisis in the setting of diffuse cerebral edema (DCE) following HIBI.

Discussion

SE-mediated HIBI can result in refractory ICP crisis. Neuromonitoring can help identify secondary brain injury (SBI), guide treatment strategies, including surgical interventions, and may lead to better outcomes.

Optimizing Electrode Configurations for Wearable EEG Seizure Detection Using Machine Learning

Epilepsy, a prevalent neurological disorder, profoundly affects patients' quality of life due to the unpredictable nature of seizures. The development of a reliable and user-friendly wearable EEG system capable of detecting and predicting seizures has the potential to revolutionize epilepsy care. However, optimizing electrode configurations for such systems, which is crucial for balancing accuracy and practicality, remains to be explored. This study addresses this gap by developing a systematic approach to optimize electrode configurations for a seizure detection machine-learning algorithm. Our approach was applied to an extensive database of prolonged annotated EEG recordings from 158 epilepsy patients. Multiple electrode configurations ranging from one to eighteen were assessed to determine the optimal number of electrodes. Results indicated that the performance was initially maintained as the number of electrodes decreased, but a drop in performance was found to have occurred at around eight electrodes. Subsequently, a comprehensive analysis of all eight-electrode configurations was conducted using a computationally intensive workflow to identify the optimal configurations. This approach can inform the mechanical design process of an EEG system that balances seizure detection accuracy with the ease of use and portability. Additionally, this framework holds potential for optimizing hardware in other machine learning applications. The study presents a significant step towards the development of an efficient wearable EEG system for seizure detection.

Etiology and prognostic significance of ictal EEG patterns in patients with non-convulsive seizures

We aimed to study the ictal EEG patterns in patients with non-convulsive seizures (NCS) and their relationship with underlying etiology and patient outcome. We conducted a retrospective review of EEG studies from patients undergoing continuous EEG (cEEG) monitoring for indication of altered mental status with a suspicion of NCS. Ictal EEG findings of NCS were categorized as three patterns: focal or generalized epileptiform discharges (EDs) at frequencies >2.5 Hz (Pattern 1); EDs at frequencies of ≤2.5 Hz or rhythmic activity >0.5 Hz with spatiotemporal evolution (Pattern 2); and EDs with ≤2.5 Hz with subtle clinical correlate during the ictal EEG or clinical and EEG improvement after a trial of IV anti-seizure drugs (Pattern 3). Patients with anoxic brain injury were excluded from the study. Associations between ictal EEG patterns and underlying etiology and their impact on in-hospital mortality was measured. Of 487 patients included in the study, NCS was recorded on cEEG monitoring in 57 (12%). The ictal EEG Pattern 2 was the most commonly seen ictal EEG finding in our cohort of patients with NCS (70%, n=40/57), followed by Pattern 3 (15%, n=9/57) and Pattern 1(14%, n=8/57). In patients with acute brain injury, Pattern 2 (67%, n=27/40) was a commonly seen ictal EEG finding, whereas Pattern 1 (62% n=5/8) was seen in patients with underlying acute medical illness. No statistically significant difference was found between ictal EEG patterns and underlying neurological versus medical etiologies (p=0.27) or in-hospital mortality (p=0.5). Spatiotemporal evolution of epileptiform discharges at a lower frequency was the most commonly recorded ictal EEG pattern in our cohort. Further prospective studies with a larger sample size of patients with NCS may provide valuable clinical data that could be used to evaluate the etiologic correlate of the various ictal EEG patterns and their effect on outcome.

Serum neuron-specific enolase: a new tool for seizure risk monitoring after status epilepticus

BACKGROUND AND PURPOSE

There is a need for accurate biomarkers to monitor electroencephalography (EEG) activity and assess seizure risk in patients with acute brain injury. Seizure recurrence may lead to cellular alterations and subsequent neurological sequelae. Whether neuron-specific enolase (NSE) and S100-beta (S100B), brain injury biomarkers, can reflect EEG activity and help to evaluate the seizure risk was investigated.

METHODS

Eleven patients, admitted to an intensive care unit for refractory status epilepticus, who underwent a minimum of 3 days of continuous EEG concomitantly with daily serum NSE and S100B assays were included. At 103 days the relationships between serum NSE and S100B levels and two EEG scores able to monitor the seizure risk were investigated. Biochemical biomarker thresholds able to predict seizure recurrence were sought.

RESULTS

Only NSE levels positively correlated with EEG scores. Similar temporal dynamics were observed for the time courses of EEG scores and NSE levels. NSE levels above 17 ng/ml were associated with seizure in 71% of patients. An increase of more than 15% of NSE levels was associated with seizure recurrence in 80% of patients.

CONCLUSIONS

Our study highlights the potential of NSE as a biomarker of EEG activity and to assess the risk of seizure recurrence.

Treatment of Refractory and Super-refractory Status Epilepticus

Refractory and super-refractory status epilepticus (SE) are serious illnesses with a high risk of morbidity and even fatality. In the setting of refractory generalized convulsive SE (GCSE), there is ample justification to use continuous infusions of highly sedating medications-usually midazolam, pentobarbital, or propofol. Each of these medications has advantages and disadvantages, and the particulars of their use remain controversial. Continuous EEG monitoring is crucial in guiding the management of these critically ill patients: in diagnosis, in detecting relapse, and in adjusting medications. Forms of SE other than GCSE (and its continuation in a "subtle" or nonconvulsive form) should usually be treated far less aggressively, often with nonsedating anti-seizure drugs (ASDs). Management of "non-classic" NCSE in ICUs is very complicated and controversial, and some cases may require aggressive treatment. One of the largest problems in refractory SE (RSE) treatment is withdrawing coma-inducing drugs, as the prolonged ICU courses they prompt often lead to additional complications. In drug withdrawal after control of convulsive SE, nonsedating ASDs can assist; medical management is crucial; and some brief seizures may have to be tolerated. For the most refractory of cases, immunotherapy, ketamine, ketogenic diet, and focal surgery are among several newer or less standard treatments that can be considered. The morbidity and mortality of RSE is substantial, but many patients survive and even return to normal function, so RSE should be treated promptly and as aggressively as the individual patient and type of SE indicate.

Seizures, CSF neurofilament light and tau in patients with subarachnoid haemorrhage

OBJECTIVES

Patients with severe subarachnoid haemorrhage (SAH) often suffer from complications with delayed cerebral ischaemia (DCI) due to vasospasm that is difficult to identify by clinical examination. The purpose of this study was to monitor seizures and to measure cerebrospinal fluid (CSF) concentrations of neurofilament light (NFL) and tau, and to see whether they could be used for predicting preclinical DCI.

METHODS

We prospectively studied 19 patients with aneurysmal SAH who underwent treatment with endovascular coiling. The patients were monitored with continuous EEG (cEEG) and received external ventricular drainage (EVD). CSF samples of neurofilament light (NLF) and total tau (T-tau) protein were collected at day 4 and day 10. Cox regression analysis was applied to evaluate whether seizures and protein biomarkers were associated with DCI and poor outcome.

RESULTS

Seven patients developed DCI (37%), and 4 patients (21%) died within the first 2 months. Six patients (32%) had clinical seizures, and electrographic seizures were noted in one additional patient (4.5%). Increased tau ratio (proportion tau10/tau4) was significantly associated with DCI and hazard ratio [HR=1.33, 95% confidence interval (CI) 1.055-1.680. P = .016].

CONCLUSION

Acute symptomatic seizures are common in SAH, but their presence is not predictive of DCI. High values of the tau ratio in the CSF may be associated with development of DCI.

Extended EEG and non-convulsive status epilepticus: Benefit over routine EEG?

OBJECTIVE

EEG monitoring is increasingly used in critically ill patients, but impact on clinical outcome remains unclear. We aimed to investigate the benefit of repeated extended EEG in the prognosis of patients with non-convulsive status epilepticus (SE).

MATERIALS & METHODS

We retrospectively collected 29 consecutive patients with non-convulsive SE without coma, who underwent repeated extended EEG between 2013 and 2015. We compared these patients with an historical age-matched group of 58 patients managed between 2011 and 2013 with routine EEG only. We excluded patients treated with therapeutic coma for SE treatment. Outcome at hospital discharge was categorized as return to baseline conditions, new disability, and death.

RESULTS

Severity of SE was similar in the two groups, with similar proportion of potential fatal etiologies (58% in the extended EEG group vs 60%, P=.529), similar STESS scores (median was three in both groups, P=.714), and comparable acute hospitalization duration (median of 15 vs 11 days, P=.131). The extended EEG group received slightly more anti-epileptic drugs (median was three in both groups, P=.026). Distribution of the outcome categories at hospital discharge was similar (P=.129).

CONCLUSIONS

Extended EEG used for the management of non-convulsive status epilepticus does not seem to improve clinical outcome, but is associated with a higher number of prescribed anti-epileptic drugs. The benefit of continuous EEG monitoring in non-convulsive SE without coma SE should be addressed through a randomized trial.

Continuous EEG monitoring in the intensive care unit

The purpose and indications for continuous electroencephalography monitoring (CEEG) in intensive care unit (ICU) patients include seizure detection, monitoring the effects of treatment (including depth of sedation), grading and classification of EEG abnormalities, ischemia detection and prognostication. Practical considerations of ICU CEEG include: choice of montages (patterns of electrode placement and connections), EEG electrodes, recognition of artifacts, and the use of automated or computerized analysis. These aspects are reviewed, along with an identifcation of current advances and challenges for the future of CEEG in the ICU.

Nonconvulsive seizures in subarachnoid hemorrhage link inflammation and outcome

OBJECTIVE

Nonconvulsive seizures (NCSz) are frequent following acute brain injury and have been implicated as a cause of secondary brain injury, but mechanisms that cause NCSz are controversial. Proinflammatory states are common after many brain injuries, and inflammation-mediated changes in blood-brain barrier permeability have been experimentally linked to seizures.

METHODS

In this prospective observational study of aneurysmal subarachnoid hemorrhage (SAH) patients, we explored the link between the inflammatory response following SAH and in-hospital NCSz studying clinical (systemic inflammatory response syndrome [SIRS]) and laboratory (tumor necrosis factor receptor 1 [TNF-R1], high-sensitivity C-reactive protein [hsCRP]) markers of inflammation. Logistic regression, Cox proportional hazards regression, and mediation analyses were performed to investigate temporal and causal relationships.

RESULTS

Among 479 SAH patients, 53 (11%) had in-hospital NCSz. Patients with in-hospital NCSz had a more pronounced SIRS response (odds ratio [OR]=1.9 per point increase in SIRS, 95% confidence interval [CI]=1.3-2.9), inflammatory surges were more likely immediately preceding NCSz onset, and the negative impact of SIRS on functional outcome at 3 months was mediated in part through in-hospital NCSz. In a subset with inflammatory serum biomarkers, we confirmed these findings linking higher serum TNF-R1 and hsCRP to in-hospital NCSz (OR=1.2 per 20-point hsCRP increase, 95% CI=1.1-1.4; OR=2.5 per 100-point TNF-R1 increase, 95% CI=2.1-2.9). The association of inflammatory biomarkers with poor outcome was mediated in part through NCSz.

INTERPRETATION

In-hospital NCSz were independently associated with a proinflammatory state following SAH as reflected in clinical symptoms and serum biomarkers of inflammation. Our findings suggest that inflammation following SAH is associated with poor outcome and that this effect is at least in part mediated through in-hospital NCSz.