Amplitude-Integrated Electroencephalography Predicts Outcome in Patients with Coma After Acute Brain Injury

Clinical and electroencephalogram characteristics of methylmalonic acidemia with MMACHC and MUT gene mutations

OBJECTIVE

This study investigated the clinical, imaging, and electroencephalogram (EEG) characteristics of methylmalonic acidemia (MMA) with nervous system damage as the primary manifestation.

METHODS

From January 2017 to November 2022, patients with nervous system injury as the main clinical manifestation, diagnosed with methylmalonic acidemia by metabolic and genetic testing, were enrolled and analyzed. Their clinical, imaging, and electroencephalogram data were analyzed.

RESULTS

A total of 18 patients were enrolled, including 15 males and 3 females. The clinical symptoms were convulsions, poor feeding, growth retardation, disorder of consciousness, developmental delay, hypotonia, and blood system changes. There were 6 cases (33%) of hydrocephalus, 9 (50%) of extracerebral space widened, 5 (27%) of corpus callosum thinning, 3 (17%) of ventricular dilation, 3 (17%) of abnormal signals in the brain parenchyma (frontal lobe, basal ganglia region, and brain stem), and 3 (17%) of abnormal signals in the lateral paraventricular. In addition, there were 3 cases (17%) of cerebral white matter atrophy and 1 (5%) of cytotoxic edema in the basal ganglia and cerebral peduncle. EEG data displayed 2 cases (11%) of hypsarrhythmia, 3 (17%) of voltage reduction, 12(67%) of abnormal discharge, 13 (72%) of abnormal sleep physiological waves or abnormal sleep structure, 1 (5%) of immature (delayed) EEG development, and 8 (44%) of slow background. There were 2 cases (11%) of spasms, 1 (5%) of atonic seizures, and 1 (5%) of myoclonic seizures. There were 16 patients (89%) with hyperhomocysteinemia. During follow-up, 1 patient was lost to follow-up, and 1 died. In total, 87.5% (14/16) of the children had varying developmental delays. EEG was re-examined in 11 cases, of which 8 were normal, and 3 were abnormal. Treatments included intramuscular injections of vitamin B12, L-carnitine, betaine, folic acid, and oral antiepileptic therapy. Acute treatment included anti-infective, blood transfusion, fluid replacement, and correcting acidosis. The other treatments included low-protein diets and special formula milk powder.

CONCLUSION

Methylmalonic acidemia can affect the central nervous system, leading to structural changes or abnormal signals on brain MRI. Metabolic screening and genetic testing help clarify the diagnosis. EEG can reflect changes in brain waves during the acute phase.

Nomogram for Early Prediction of Outcome in Coma Patients with Severe Traumatic Brain Injury Receiving Right Median Nerve Electrical Stimulation Treatment

BACKGROUND

Accurate outcome prediction can serve to approach, quantify and categorize severe traumatic brain injury (TBI) coma patients for right median electrical stimulation (RMNS) treatment, which can support rehabilitation plans. As a proof of concept for individual risk prediction, we created a novel nomogram model combining amplitude-integrated electroencephalography (AEEG) and clinically relevant parameters.

METHODS

This study retrospective collected and analyzed a total of 228 coma patients after severe TBI in two medical centers. According to the extended Glasgow Outcome Scale (GOSE), patients were divided into a good outcome (GOSE 3-8) or a poor outcome (GOSE 1-2) group. Their clinical and biochemical indicators, together with EEG features, were explored retrospectively. The risk factors connected to the outcome of coma patients receiving RMNS treatment were identified using Cox proportional hazards regression. The discriminative capability and calibration of the model to forecast outcome were assessed by C statistics, calibration plots, and Kaplan-Meier curves on a personalized nomogram forecasting model.

RESULTS

The study included 228 patients who received RMNS treatment for long-term coma after a severe TBI. The median age was 40 years, and 57.8% (132 of 228) of the patients were male. 67.0% (77 of 115) of coma patients in the high-risk group experienced a poor outcome after one year and the comparative data merely was 30.1% (34 of 113) in low-risk group patients. The following variables were integrated into the forecasting of outcome using the backward stepwise selection of Akaike information criterion: age, Glasgow Coma Scale (GCS) at admission, EEG reactivity (normal, absence, or the stimulus-induced rhythmic, periodic, or ictal discharges (SIRPIDs)), and AEEG background pattern (A mode, B mode, or C mode). The C statistics revealed that the nomograms' discriminative potential and calibration demonstrated good predictive ability (0.71).

CONCLUSION

Our findings show that the nomogram model using AEEG parameters has the potential to predict outcomes in severe TBI coma patients receiving RMNS treatment. The model could classify patients into prognostic groups and worked well in internal validation.

Diagnostic and prognostic EEG analysis of critically ill patients: A deep learning study

Visual interpretation of electroencephalography (EEG) is time consuming, may lack objectivity, and is restricted to features detectable by a human. Computer-based approaches, especially deep learning, could potentially overcome these limitations. However, most deep learning studies focus on a specific question or a single pathology. Here we explore the potential of deep learning for EEG-based diagnostic and prognostic assessment of patients with acute consciousness impairment (ACI) of various etiologies. EEGs from 358 adults from a randomized controlled trial (CERTA, NCT03129438) were retrospectively analyzed. A convolutional neural network was used to predict the clinical outcome (based either on survival or on best cerebral performance category) and to determine the etiology (four diagnostic categories). The largest probability output served as marker for the confidence of the network in its prediction ("certainty factor"); we also systematically compared the predictions with raw EEG data, and used a visualization algorithm (Grad-CAM) to highlight discriminative patterns. When all patients were considered, the area under the receiver operating characteristic curve (AUC) was 0.721 for predicting survival and 0.703 for predicting the outcome based on best CPC; for patients with certainty factor ≥ 60 % the AUCs increased to 0.776 and 0.755 respectively; and for certainty factor ≥ 75 % to 0.852 and 0.879. The accuracy for predicting the etiology was 54.5 %; the accuracy increased to 67.7 %, 70.3 % and 84.1 % for patients with certainty factor of 50 %, 60 % and 75 % respectively. Visual analysis showed that the network learnt EEG patterns typically recognized by human experts, and suggested new criteria. This work demonstrates for the first time the potential of deep learning-based EEG analysis in critically ill patients with various etiologies of ACI. Certainty factor and post-hoc correlation of input data with prediction help to better characterize the method and pave the route for future implementations in clinical routine.

A Hybrid Approach for MS Diagnosis Through Nonlinear EEG Descriptors and Metaheuristic Optimized Classification Learning

Multiple sclerosis (MS), a disease of the central nervous system, affects the white matter of the brain. Neurologists interpret magnetic resonance images that are often complicated, time-consuming, and contradictory. Using EEG signals, this disease can be analyzed and diagnosed more accurately. However, it is imperative that MS be diagnosed by specialists using assistive technology. Until now, a few methods have been proposed in this field that are sometimes associated with different challenges in analysis. This paper presents a hybrid approach to MS diagnosis in order to decrease classification error rates. Using the proposed method, EEG descriptors in both the time and frequency domains are analyzed. After the feature extraction stage, a modified version of the ant colony optimization method (m-ACO) was used to select the appropriate subset of features. Then, the support vector machine is used to determine whether the disease exists. A metaheuristic algorithm adjusts the support vector machine's parameters to withstand overfitting challenges. Despite a limited number of input channels, significant classification accuracy has been achieved using wavelet analysis techniques, dividing all five subbands of EEG signals, signal windowing, and extracting efficient features from the data. Additionally, alpha, beta, and gamma bands of the signal are important, and the accuracy, sensitivity, and specificity levels are higher than 98.5%. Compared to similar MS diagnostic methods, the proposed method achieved significantly higher diagnostic accuracy. Application and implementation of this method can be effective in treating neurological diseases, including multiple sclerosis.

Standardized visual EEG features predict outcome in patients with acute consciousness impairment of various etiologies

BACKGROUND

Early prognostication in patients with acute consciousness impairment is a challenging but essential task. Current prognostic guidelines vary with the underlying etiology. In particular, electroencephalography (EEG) is the most important paraclinical examination tool in patients with hypoxic ischemic encephalopathy (HIE), whereas it is not routinely used for outcome prediction in patients with traumatic brain injury (TBI).

METHOD

Data from 364 critically ill patients with acute consciousness impairment (GCS ≤ 11 or FOUR ≤ 12) of various etiologies and without recent signs of seizures from a prospective randomized trial were retrospectively analyzed. Random forest classifiers were trained using 8 visual EEG features-first alone, then in combination with clinical features-to predict survival at 6 months or favorable functional outcome (defined as cerebral performance category 1-2).

RESULTS

The area under the ROC curve was 0.812 for predicting survival and 0.790 for predicting favorable outcome using EEG features. Adding clinical features did not improve the overall performance of the classifier (for survival: AUC = 0.806, p = 0.926; for favorable outcome: AUC = 0.777, p = 0.844). Survival could be predicted in all etiology groups: the AUC was 0.958 for patients with HIE, 0.955 for patients with TBI and other neurosurgical diagnoses, 0.697 for patients with metabolic, inflammatory or infectious causes for consciousness impairment and 0.695 for patients with stroke. Training the classifier separately on subgroups of patients with a given etiology (and thus using less training data) leads to poorer classification performance.

CONCLUSIONS

While prognostication was best for patients with HIE and TBI, our study demonstrates that similar EEG criteria can be used in patients with various causes of consciousness impairment, and that the size of the training set is more important than homogeneity of ACI etiology.

[Value of amplitude-integrated EEG combined with Full Outline of Unresponsiveness in evaluating the prognosis of children with disturbance of consciousness]

OBJECTIVE

To study the value of amplitude-integrated EEG (aEEG), Full Outline of Unresponsiveness (FOUR), and Glasgow Coma Scale (GCS) in evaluating the prognosis of children with disturbance of consciousness in the pediatric intensive care unit (PICU).

METHODS

A total of 164 children with disturbance of consciousness who were admitted to the PICU of Children's Hospital Affiliated to Soochow University were enrolled as subjects. According to prognosis, they were divided into a poor prognosis group with 111 children and a good prognosis group with 53 children. The results of aEEG monitoring, FOUR score, and GCS score on days 1 and 5 of admission were collected. The association between evaluation methods and prognosis was analyzed. The receiver operating characteristic (ROC) curve was used to evaluate the value of aEEG, FOUR, and GCS in predicting prognosis.

RESULTS

The children with no improvement or abnormal aggravation of aEEG on day 5 tended to have a poor prognosis. The results of aEEG was positively correlated with prognosis (r=0.689, P<0.001), and FOUR and GCS were negatively correlated with prognosis (r=-0.655 and -0.554 respectively, P<0.001). The areas under the ROC curve (AUC) of aEEG, FOUR, and GCS were 0.894, 0.903, and 0.840 respectively, and there was no significant difference in the AUC between the three indices (P>0.05), while aEEG combined with FOUR had an AUC of 0.945, which was significantly larger than that of each index alone (P<0.05).

CONCLUSIONS

Both aEEG and FOUR can be used as effective tools to predict the prognosis of children with disturbance of consciousness, and a combination of aEEG and FOUR can improve the predictive value.

Prognostic value of amplitude-integrated EEG in neonates with high risk of neurological sequelae

OBJECTIVE

To determine the efficacy and the prognostic value of amplitude-integrated electroencephalography (aEEG) in term and near-term neonates with high risk of neurological sequelae.

METHODS

Infants of ≥35 weeks of gestation diagnosed with neonatal encephalopathy or with high risk of brain injury were included. All eligible infants underwent aEEG within 6 h after clinical assessment. The infants were followed up 12 months to evaluate neurological development.

RESULTS

A total of 250 infants were eligible, of which 85 had normal aEEG, 81 had mildly abnormal aEEG, and 84 had severely abnormal aEEG. Of these infants, 168 were diagnosed with different neonatal encephalopathies, 27 with congenital or metabolic diseases, and 55 with high risk of brain injury. In all, 22 infants died, 19 were lost to follow-up, and 209 completed the follow-up at 12 months, of which 62 were diagnosed with a neurological disability. Statistical analysis showed that severely abnormal aEEG predicted adverse neurological outcome with a sensitivity of 70.2%, a specificity of 87.1%, a positive predictive value of 75.6%, and a negative predictive value of 83.7%.

INTERPRETATION

aEEG can predict adverse outcomes in high-risk neonates and is a useful method for monitoring neonates with high risk of adverse neurological outcomes.

aEEG Use in Pediatric Critical Care-An Online Survey

Background: Evidence supporting continuous EEG monitoring in pediatric intensive care is increasing, but continuous full-channel EEG is a scarce resource. Amplitude-integrated EEG (aEEG) monitors are broadly available in children's hospitals due to their use in neonatology and can easily be applied to older patients. Objective: The aim of this survey was to evaluate the use of amplitude-integrated EEG in German and Swiss pediatric intensive care units (PICUs). Design: An online survey was sent to German and Swiss PICUs that were identified via databases provided by the German Pediatric Association (DGKJ) and the Swiss Society of Intensive Care (SGI). The questionnaire contained 18 multiple choice questions including the PICU size and specialization, indications for aEEG use, perceived benefits from aEEG, and data storage. Main results: Forty-three (26%) PICUs filled out the questionnaire. Two thirds of all interviewed PICUs use aEEG in non-neonates. Main indications were neurological complications or disease and altered mental state. Features assessed were mostly seizures and side differences, less frequently height of amplitude and background pattern. Interpretation of raw EEG also played an important role. All interviewees would appreciate the establishment of reference values for toddlers and children. Conclusions: aEEG is used in a large proportion of the interviewed PICUs. The wide-spread use without validation of data generates the need for further evaluation of this technique and the establishment of reference values for non-neonates.