Neurocritical Care最新文献

{"title":"Evaluating the Machine Learning Models in Predicting Intensive Care Unit Discharge for Neurosurgical Patients Undergoing Craniotomy: A Big Data Analysis.","authors":"Taghi Khaniyev, Efecan Cekic, Muhammet Abdullah Koc, Ilke Dogan, Sahin Hanalioglu","doi":"10.1007/s12028-025-02246-9","DOIUrl":"10.1007/s12028-025-02246-9","url":null,"abstract":"<p><strong>Background: </strong>Predicting intensive care unit (ICU) discharge for neurosurgical patients is crucial for optimizing bed sources, reducing costs, and improving outcomes. Our study aims to develop and validate machine learning (ML) models to predict ICU discharge within 24 h for patients undergoing craniotomy.</p><p><strong>Methods: </strong>The 2,742 patients undergoing craniotomy were identified from Medical Information Mart for Intensive Care dataset using diagnosis-related group and International Classification of Diseases codes. Demographic, clinical, laboratory, and radiological data were collected and preprocessed. Textual clinical examinations were converted into numerical scales. Data were split into training (70%), validation (15%), and test (15%) sets. Four ML models, logistic regression (LR), decision tree, random forest, and neural network (NN), were trained and evaluated. Model performance was assessed using area under the receiver operating characteristic curve (AUC), average precision (AP), accuracy, and F1 scores. Shapley Additive Explanations (SHAP) were used to analyze importance of features. Statistical analyses were performed using R (version 4.2.1) and ML analyses with Python (version 3.8), using scikit-learn, tensorflow, and shap packages.</p><p><strong>Results: </strong>Cohort included 2,742 patients (mean age 58.2 years; first and third quartiles 47-70 years), with 53.4% being male (n = 1,464). Total ICU stay was 15,645 bed days (mean length of stay 4.7 days), and total hospital stay was 32,008 bed days (mean length of stay 10.8 days). Random forest demonstrated highest performance (AUC 0.831, AP 0.561, accuracy 0.827, F1-score 0.339) on test set. NN achieved an AUC of 0.824, with an AP, accuracy, and F1-score of 0.558, 0.830, and 0.383, respectively. LR achieved an AUC of 0.821 and an accuracy of 0.829. The decision tree model showed lowest performance (AUC 0.813, accuracy 0.822). Key predictors of SHAP analysis included Glasgow Coma Scale, respiratory-related parameters (i.e., tidal volume, respiratory effort), intracranial pressure, arterial pH, and Richmond Agitation-Sedation Scale.</p><p><strong>Conclusions: </strong>Random forest and NN predict ICU discharge well, whereas LR is interpretable but less accurate. Numeric conversion of clinical data improved performance. This study offers framework for predictions using clinical, radiological, and demographic features, with SHAP enhancing transparency.</p>","PeriodicalId":19118,"journal":{"name":"Neurocritical Care","volume":" ","pages":"512-529"},"PeriodicalIF":3.6,"publicationDate":"2025-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12436523/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144012593","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Clinical Performance of the Brain4care System for Noninvasive Detection of Intracranial Hypertension.","authors":"Gustavo Frigieri, Thauan Leandro Gonçalves, Gabriela Nagai Ocamoto, Rodrigo de Ap Andrade, Bruno Cezar de Padua, Danilo Cardim","doi":"10.1007/s12028-025-02273-6","DOIUrl":"10.1007/s12028-025-02273-6","url":null,"abstract":"<p><strong>Background: </strong>Noninvasive methods for detecting intracranial hypertension (IH) are of growing importance in clinical settings. This study evaluates the clinical performance of the brain4care (B4C) System, which captures pulsatile cranial expansions that reveal a surrogate intracranial pressure (ICP) waveform and subsequently derives the P2/P1 ratio and time-to-peak (TTP) parameters to predict IH.</p><p><strong>Methods: </strong>This was a retrospective study conducted across multiple centers that included a total of 124 patients. Invasively monitored ICP and noninvasive B4C waveforms were recorded simultaneously from patients with acute brain injuries. Data were analyzed using specific cutoff values for the estimated P2/P1 ratio (ranging from 0.8 to 1.4) and TTP (at 0.3) to assess their diagnostic accuracy. Sensitivity and specificity for detecting IH (ICP > 20 mm Hg) were determined based on these metrics.</p><p><strong>Results: </strong>The estimated P2/P1 ratio demonstrated a sensitivity of 92% and specificity of 19% at a threshold of 0.8, indicating high sensitivity for ruling out IH. At a ratio of 1.4, the specificity improved to 90%, suggesting its effectiveness for assessing IH. For TTP, a threshold of 0.3 was identified as the optimal cutoff, offering a specificity of 92%.</p><p><strong>Conclusions: </strong>The B4C System provides a viable, noninvasive approach to assessing IH. The study underscores the clinical utility of the P2/P1 ratio and TTP in detecting and ruling out IH, offering a significant alternative to invasive ICP monitoring methods.</p>","PeriodicalId":19118,"journal":{"name":"Neurocritical Care","volume":" ","pages":"628-635"},"PeriodicalIF":3.6,"publicationDate":"2025-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12436508/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144020334","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Grading Scores for Identifying Patients at Risk of Delayed Cerebral Ischemia and Neurological Outcome in Spontaneous Subarachnoid Hemorrhage: A Comparison of Receiver Operator Curve Analysis.","authors":"Alessia Degrassi, Caren Conticello, Hassane Njimi, Giacomo Coppalini, Fernando Oliveira, Alberto Diosdado, Marco Anderloni, Lise Jodaitis, Sophie Schuind, Fabio Silvio Taccone, Elisa Gouvêa Bogossian","doi":"10.1007/s12028-025-02270-9","DOIUrl":"10.1007/s12028-025-02270-9","url":null,"abstract":"<p><strong>Background: </strong>Numerous grading scales were proposed for subarachnoid hemorrhage (SAH) to assess the likelihood of unfavorable neurological outcomes (UO) and the risk of delayed cerebral ischemia (DCI). We aimed to validate the Hemorrhage, Age, Treatment, Clinical Status, and Hydrocephalus (HATCH) score and the VASOGRADE, a simple grading scale for prediction of DCI after aneurysmal SAH.</p><p><strong>Methods: </strong>This was a retrospective single-center study of patients with nontraumatic SAH (January 2016 to December 2021) admitted to the intensive care unit. We performed a receiver operating characteristic (ROC) curve analysis to assess the discriminative ability of the HATCH and the VASOGRADE to identify patients who had UO at 3 months (defined as Glasgow Outcome Scale score of 1-3), hospital mortality, and DCI and compared their performance with the World Federation of Neurosurgical Surgeons, the modified Fisher, the Sequential Organ Failure Assessment, and the Acute Physiology and Chronic Health Evaluation II scales. We performed a multivariate logistic regression analysis to assess the association between HATCH and UO at 3 months and between VASOGRADE and DCI.</p><p><strong>Results: </strong>We included 262 consecutive patients with nontraumatic SAH. DCI was observed in 82 patients (31.3%), whereas 78 patients (29.8%) died during hospital stay and 133 patients (51%) had UO at 3 months. HATCH was independently associated with UO (odds ratio 1.61, 95% confidence interval [CI] 1.36-1.90) and had an area under the ROC curve (AUROC) of 0.83 (95% CI 0.77-0.88), comparable to the Acute Physiology and Chronic Health Evaluation II (AUROC 0.84, 95% CI 0.79-0.89) and Sequential Organ Failure Assessment (AUROC 0.83, 95% CI 0.77-0.88).</p><p><strong>Conclusions: </strong>Hemorrhage, Age, Treatment, Clinical Status, and Hydrocephalus and VASOGARDE scores had a good performance to predict UO or in-hospital mortality and DCI, respectively; however, their performance did not outperform nonspecific routinely used scores.</p>","PeriodicalId":19118,"journal":{"name":"Neurocritical Care","volume":" ","pages":"616-627"},"PeriodicalIF":3.6,"publicationDate":"2025-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144030199","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"The Stroke Alert that Wasn't: Lessons Learned from Meningitis-Associated Vasospasm.","authors":"Julian L Moran, Erika J Sigman, Catherine S W Albin","doi":"10.1007/s12028-025-02291-4","DOIUrl":"10.1007/s12028-025-02291-4","url":null,"abstract":"","PeriodicalId":19118,"journal":{"name":"Neurocritical Care","volume":" ","pages":"695-698"},"PeriodicalIF":3.6,"publicationDate":"2025-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144208953","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Experimental Analysis of Echogenic Signals Appearing in Optic Nerve Ultrasound.","authors":"Hans-Christian Hansen, Jan-Peter Sperhake, Jakob Matschke, Benjamin Ondruschka, Knut Helmke","doi":"10.1007/s12028-024-02057-4","DOIUrl":"10.1007/s12028-024-02057-4","url":null,"abstract":"","PeriodicalId":19118,"journal":{"name":"Neurocritical Care","volume":" ","pages":"682-684"},"PeriodicalIF":3.6,"publicationDate":"2025-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141860423","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"The Predictive Validity of the Full Outline of UnResponsiveness Score Compared to the Glasgow Coma Scale in the Intensive Care Unit: A Systematic Review.","authors":"Jaime E Schey, Monica Schoch, Debra Kerr","doi":"10.1007/s12028-024-02150-8","DOIUrl":"10.1007/s12028-024-02150-8","url":null,"abstract":"<p><p>The Full Outline of UnResponsiveness (FOUR) score was developed to overcome the limitations of the Glasgow Coma Scale (GCS) when assessing individuals with impaired consciousness. We sought to review the evidence regarding the predictive validity of the GCS and FOUR score in intensive care unit (ICU) settings. This review was prospectively registered in PROSPERO (CRD42023420528). Systematic searches of CINAHL, MEDLINE, and Embase were undertaken. Prospective observational studies were included if both GCS and FOUR score were assessed in adults during ICU admission and if mortality and/or validated functional outcome measure scores were collected. Studies were excluded if they exclusively investigated patients with traumatic brain injury. Screening, data extraction, and quality assessment using the Quality in Prognosis Studies tool were conducted by two reviewers. Twenty studies of poor to moderate quality were included. Many studies only included patients with neurological illness and excluded sedated patients, despite high proportions of intubated patients. The FOUR score achieved higher area under the receiver operating characteristic curve values for mortality prediction compared with the GCS, and the FOUR score achieved significantly higher area under the receiver operating characteristic curve values for predictions of ICU mortality. Both coma scales showed similar accuracy in predicting \"unfavorable\" functional outcome. The FOUR score appeared to be more responsive than the GCS in the ICU, as most patients with a GCS score of 3 obtained FOUR scores between 1 and 8 due to preserved brainstem function. The FOUR score may be superior to the GCS for predicting mortality in ICU settings. Further adequately powered studies with clear, reliable methods for assessment of index and outcome scores are required to clarify the predictive performance of both coma scales in ICUs. Inclusion of sedated patients may improve generalizability of findings in general ICU populations.</p>","PeriodicalId":19118,"journal":{"name":"Neurocritical Care","volume":" ","pages":"645-658"},"PeriodicalIF":3.6,"publicationDate":"2025-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12436514/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142576553","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A Novel, Interpretable Machine Learning Model to Predict Neurological Outcomes Following Venoarterial Extracorporeal Membrane Oxygenation.","authors":"Benjamin L Shou, Albert Leng, Preetham Bachina, Andrew Kalra, Alice L Zhou, Glenn Whitman, Sung-Min Cho","doi":"10.1007/s12028-025-02233-0","DOIUrl":"10.1007/s12028-025-02233-0","url":null,"abstract":"<p><strong>Background: </strong>We used machine learning models incorporating rich electronic medical record (EMR) data to predict neurological outcomes after venoarterial extracorporeal membrane oxygenation (VA-ECMO).</p><p><strong>Methods: </strong>This was a retrospective review of adult (≥ 18 years) patients undergoing VA-ECMO between 6/2016 and 4/2022 at a single center. The primary outcome was good neurological outcome, defined as a modified Rankin Scale score of 0 to 3, evaluated at hospital discharge. We extracted every measurement of 74 vital and laboratory values, as well as circuit and ventilator settings, from 24 h before cannulation through the entire duration of ECMO. An XGBoost model with Shapley Additive Explanations was developed and evaluated with leave-one-out cross-validation.</p><p><strong>Results: </strong>Overall, 194 patients undergoing VA-ECMO (median age 58 years, 63% male) were included. We extracted more than 14 million individual data points from the EMR. Of 194 patients, 39 patients (20%) had good neurological outcomes. Three models were generated: model A, which contained only pre-ECMO data; model B, which added data from the first 48 h of ECMO; and model C, which included data from the entire ECMO run. The leave-one-out cross-validation area under the receiver operator characteristics curves for models A, B, and C were 0.72, 0.81, and 0.90, respectively. The inclusion of on-ECMO physiologic, laboratory, and circuit data greatly improved model performance. Both modifiable and nonmodifiable variables, such as lower body mass index, lower age, higher mean arterial pressure, and higher hemoglobin, were associated with good neurological outcome.</p><p><strong>Conclusions: </strong>An interpretable machine learning model from EMR-extracted data was able to predict neurological outcomes for patients undergoing VA-ECMO with excellent accuracy.</p>","PeriodicalId":19118,"journal":{"name":"Neurocritical Care","volume":" ","pages":"403-413"},"PeriodicalIF":3.6,"publicationDate":"2025-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143730661","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Aneurysmal Subarachnoid Hemorrhage Risk Assessment Model Identifies Patients for Safe Early Discharge at Day 15-The SAFE-SaH Score.","authors":"Eric E Kennison, Nick M Murray, Dave S Collingridge, Daniel Knox, Gabriel V Fontaine","doi":"10.1007/s12028-025-02236-x","DOIUrl":"10.1007/s12028-025-02236-x","url":null,"abstract":"<p><strong>Background: </strong>Patients with aneurysmal subarachnoid hemorrhage (aSAH) are often hospitalized for 21 days after aneurysm rupture due to the risk of complications. However, some never experience complications and are unlikely to benefit from a prolonged hospitalization. The aim of this study is to derive a risk assessment model (RAM) using data from the first 14 days of hospitalization to identify low-risk patients for early discharge, at day 15 or after.</p><p><strong>Methods: </strong>Patients ≥ 18 years old with an acute aSAH at a Comprehensive Stroke Center from 2017 to 2024 were included. Baseline demographics, aSAH grading scales, and in-hospital complications requiring intervention were characterized. Complications included: vasospasm, delayed cerebral ischemia (DCI), cerebral salt wasting (CSW), cerebral edema, seizures, arrhythmias, respiratory failure, and hydrocephalus. Binary logistic regression with leave-one-out cross validation (LOOCV) was used to identify an optimal RAM.</p><p><strong>Results: </strong>Of 165 patients, the mean Hunt Hess Score (HHS) was 2.5 (standard deviation, SD 1.2), modified Fisher Score (mFS) was 3.1 (SD 1), endovascular therapy was used for aneurysm securement in 73% of patients, and 54.5% of patients experienced complications during days 15-21. In bivariate analyses, days 0-14 variables associated with days 15 + complications were the following: HHS, mFS, middle cerebral artery (MCA) aneurysm, clinical or radiologic vasospasm, endovascular therapies, intraventricular hemorrhage, hydrocephalus, presence of external ventricular drain (EVD), mechanical ventilation, vasopressors, hypertonic solutions, antiseizure medications, milrinone, and fludrocortisone (all p < 0.05). LOOCV regression for a best fit RAM included six variables: Sum of Vasopressors, Artery (MCA aneurysm), Fludrocortisone, EVD, Scale (mFS and HHS), \"SAFE-SaH\" and had an area under the receiver operator characteristics curve of 0.90 (95% confidence interval 0.85-0.95), sensitivity of 0.94, specificity of 0.69, positive predictive value of 79%, and negative predictive value of 91% for predicting complications on day 15 + .</p><p><strong>Conclusions: </strong>This is the first ever RAM to incorporate clinical data from the first 14 days of hospitalization to identify patients with an aSAH at low risk for complications after day 14. With 94% sensitivity, the RAM classifies patients who will not have complications and may assist in earlier disposition on day 15 or after.</p>","PeriodicalId":19118,"journal":{"name":"Neurocritical Care","volume":" ","pages":"414-423"},"PeriodicalIF":3.6,"publicationDate":"2025-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143743317","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Longitudinal Improvement in Respiratory Function Following Plasma Exchange in Patients with Severe Myasthenia Gravis.","authors":"Panayiotis N Varelas, Ileana Lopez-Plaza, Ashar Ata, Mohammed F Rehman, Chandan Mehta, Riad Ramadan, Vaso Zisimopoulou","doi":"10.1007/s12028-025-02238-9","DOIUrl":"10.1007/s12028-025-02238-9","url":null,"abstract":"<p><strong>Background: </strong>There are no data on the effect size and timing of plasma exchange (PLEX) in patients with myasthenic crisis (MC).</p><p><strong>Methods: </strong>We retrospectively analyzed measurements of forced vital capacity (FVC) and negative inspiratory force (NIF) in the days before and after PLEX (administered every other day) in patients with MC admitted to a tertiary hospital over 4 years. For multiple measurements in one day, the average value was used. The day immediately before the first treatment with PLEX was considered baseline. Using time as a continuous or categorical variable in mixed-effects multiple linear regressions, we estimated predicted values for these tests.</p><p><strong>Results: </strong>Twenty-two patients (mean age 67.3 years, 51.9% male patients) with 27 MC episodes and 508 measurements (234 FVC and 274 NIF; from 5 days before to 20 days after PLEX) were included. Presence of antibodies was detected in 70.4%. Intubation and mechanical ventilation occurred in 36.6% of patients. The mean number of PLEX was 5.1 (range 3-11). NIF values decreased before the first PLEX but increased after by on average 1 cm H₂O/day (95% confidence interval [CI] 0.68-1.32, p < 0.001). FVC fluctuated before the first PLEX but then increased by on average 51.2 mL/day (95% CI 35.8-66.1, p < 0.001). The maximum increase in NIF occurred during the day of the first PLEX (9.2 cm H₂O, 95% CI 3.3-15.1, p = 0.002) and rather slowed after day 10. FVC increase compared to baseline became significant the second day after the first PLEX (287 mL, 95% CI 7.5-567.6, p = 0.04) and continued overall to increase (with fluctuations) up to day 17.</p><p><strong>Conclusions: </strong>Significant increases in bedside respiratory measurements are observed as soon as the first PLEX day but with more variability on FVC than NIF, which may either reflect more FVC technique inconsistencies or more consistent effect of the treatment on NIF.</p>","PeriodicalId":19118,"journal":{"name":"Neurocritical Care","volume":" ","pages":"458-466"},"PeriodicalIF":3.6,"publicationDate":"2025-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143780636","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Electroencephalographic Findings Add Prognostic Value to Clinical Features Associated with Mortality on Venoarterial Extracorporeal Support.","authors":"Alyson Baker, Ekta Shah, Amy Ouyang, Maya Silver, Stuart R Tomko, Kristin Guilliams, Ahmed S Said, Réjean M Guerriero","doi":"10.1007/s12028-025-02248-7","DOIUrl":"10.1007/s12028-025-02248-7","url":null,"abstract":"<p><strong>Background: </strong>The objective of this study was to identify clinical and continuous electroencephalogram (cEEG) variables associated with outcomes of pediatric venoarterial (V-A) extracorporeal membrane oxygenation support (ECMO).</p><p><strong>Methods: </strong>We conducted a retrospective single-center study of pediatric patients on V-A ECMO between January 1, 2015, and September 30, 2020. Serial clinical and cEEG variables were collected to assess the relationship of pre- and on-ECMO variables with hospital mortality in patients who underwent cEEG monitoring.</p><p><strong>Results: </strong>Ninety-four patients undergoing V-A ECMO had cEEG monitoring, with a hospital mortality of 43%. Nonsurvivors had significantly lower pH and higher lactate levels prior to ECMO initiation. Nineteen (20%) had seizures, with 7 (7%) developing status epilepticus. In the first 24 h patients were on ECMO, unfavorable background score and lack of cEEG variability or reactivity were associated with mortality. A multivariable model investigating in-hospital mortality that included pH and lactate level 2 h prior to ECMO initiation, presence of electrographic seizures, and asymmetry on cEEG as variables, had an area under the receiver operating characteristic curve (AUROC) of 0.8 (95% confidence interval [CI] 0.74-0.86, p < 0.02). The model for on-ECMO mortality (ECMO nonsurvivors) that included pH 2 h prior to ECMO initiation, presence of electrographic seizures, and lack of variability/reactivity at any point on cEEG as variables had an AUROC of 0.85 (95% CI 0.8-0.9, p < 0.001).</p><p><strong>Conclusions: </strong>These data demonstrate an association of evolving pre-ECMO impaired tissue oxygenation and on-ECMO neurophysiologic impairment, measured by cEEG, with mortality. They provide preliminary evidence that the timing of ECMO initiation, in relation to worsening tissue oxygenation, should be investigated further, and cEEG may be used to evaluate the potential impact on both neurologic injury and mortality.</p>","PeriodicalId":19118,"journal":{"name":"Neurocritical Care","volume":" ","pages":"530-540"},"PeriodicalIF":3.6,"publicationDate":"2025-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143972486","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}