Prediction Of Neurological Outcome Research Articles

Increased Glucose Variability During Long-Term Therapeutic Hypothermia as a Predictor of Poor Neurological Outcomes and Mortality: A Retrospective Study.

Therapeutic hypothermia (TH) is considered as a treatment option in patients with stroke and brain injury for controlling intracranial pressure. A relatively longer duration of TH is required in such patients than in cardiac arrest patients. We aimed to investigate blood glucose parameters during TH that predict unfavorable neurological outcomes and mortality in patients admitted to the neurological or neurosurgical intensive care unit (ICU). This retrospective study evaluated electronic medical records of patients admitted to the ICU from January 1, 2012, to May 20, 2017. A total of 103 patients were included in the analyses. Multivariable analyses revealed that age and glycemic variability (GV) index were significantly associated with poor neurological outcomes (odds ratio [OR] 1.09, 95% confidence interval [CI] 1.03-1.15, p = 0.002, and OR 1.04, 95% CI 1.02-1.06, p < 0.001, respectively); furthermore, cumulative input-output balance, sequential organ failure assessment score, and glucose variability index were associated with 90-day mortality (hazard ratio [HR] 1.13, 95% CI 1.05-1.21, p < 0.001; HR 1.20, 95% CI 1.04-1.38, p = 0.010; and HR 1.01, 95% CI 1.01-1.02, p < 0.001, respectively). Receiver operating characteristic curve analyses of the GV index for prediction of 90-day mortality and poor neurological outcomes revealed that the areas under the curves were 0.747 (95% CI 0.65-0.85) and 0.826 (95% CI 0.75-0.91), respectively. In conclusion, variability in glucose levels may be valuable for predicting 90-day mortality and poor neurological outcomes in patients undergoing long-term TH.

Circulating Levels of miR-574-5p Are Associated with Neurological Outcome after Cardiac Arrest in Women: A Target Temperature Management (TTM) Trial Substudy.

Purpose Postresuscitation neuroprognostication is guided by neurophysiological tests, biomarker measurement, and clinical examination. Recent investigations suggest that circulating microRNAs (miRNA) may help in outcome prediction after cardiac arrest. We assessed the ability of miR-574-5p to predict neurological outcome after cardiac arrest, in a sex-specific manner. Methods In this substudy of the Target Temperature Management (TTM) Trial, we enrolled 590 cardiac arrest patients for which blood samples were available. Expression levels of miR-574-5p were measured by quantitative PCR in plasma samples collected 48 h after cardiac arrest. The endpoint of the study was poor neurological outcome at 6 months (cerebral performance category scores 3 to 5). Results Eighty-one percent of patients were men, and 49% had a poor neurological outcome. Circulating levels of miR-574-5p at 48 h were higher in patients with a poor neurological outcome at 6 months (p < 0.001), both in women and in men. Circulating levels of miR-574-5p were univariate predictors of neurological outcome (odds ratio (OR) [95% confidence interval (CI)]: 1.5 [1.26-1.78]). After adjustment with clinical variables and NSE, circulating levels of miR-574-5p predicted neurological outcome in women (OR [95% CI]: 1.9 [1.09-3.45]), but not in men (OR [95% CI]: 1.0 [0.74-1.28]). Conclusion miR-574-5p is associated with neurological outcome after cardiac arrest in women.

Electroencephalogram and Somatosensory Evoked Potential Evaluation for Good and Poor Neurological Prognosis After Cardiac Arrest: A Prospective Multicenter Cohort Trial (ProNeCA)

Aim: Hypoxic-ischemic-encephalopathy is a severe and frequent neurological complication of successful cardiopulmonary-resuscitation after cardiac arrest. Prognosticating neurological outcomes in patients with hypoxic-ischemic-encephalopathy is challenging and recent guidelines suggest a multimodal approach. Only few studies have analyzed the prognostic power of the association between instrumental tests and, in addition, most of them were monocentric, retrospective and evaluating only poor outcome. Methods/design: We designed a multicenter prospective cohort study to assessing the prognostic power of the association of electroencephalogram and somatosensory evoked potentials for the prediction of both poor and good neurological outcomes at different times after cardiac arrest. Discussion: The results of our study will provide a high level of evidence for the use of neurophysiological evaluation in the current clinical practice.

Predictive value of plasma neutrophil gelatinase-associated lipocalin in acute charcoal-burning carbon monoxide poisoning.

This study aimed to assess the feasibility of using the plasma neutrophil gelatinase-associated lipocalin (NGAL) level at the time of presentation in the emergency department (ED) to predict acute kidney injury (AKI) and the long-term neurological outcomes of acute charcoal-burning carbon monoxide (CO) poisoning. This retrospective study included 260 patients who suffered acute charcoal-burning CO poisoning. The median plasma NGAL concentration at the time of presentation in the ED after acute charcoal-burning CO poisoning was 78 (54-115) ng/ml. The NGAL level was an independent predictor of AKI development and could be used to stratify the severity of AKI. However, the area under the receiver operating characteristic curve (AUC) of the predictive model for AKI that included both the plasma NGAL level and clinical parameters was comparable to that of the predictive model including only the clinical parameters. The plasma NGAL level at the time of presentation in the ED was an independent factor predicting long-term neurological outcomes in patients who did not develop AKI. In these patients, the plasma NGAL level significantly improved the predictive accuracy of the model when used in combination with clinical parameters. In contrast, the plasma NGAL level was not associated with long-term neurological outcomes in patients who developed AKI. Measurement of the plasma NGAL level at the time ED presentation might improve the prediction of long-term neurological outcomes in patients who do not develop AKI after acute charcoal-burning CO poisoning. However, it might not offer additional benefit for AKI prediction compared to previously used markers.

Initial arterial pH as a predictor of neurologic outcome after out-of-hospital cardiac arrest: A propensity-adjusted analysis

BackgroundLower pH after out-of-hospital cardiac arrest (OHCA) has been associated with worsening neurologic outcome, with <7.2 identified as an “unfavorable resuscitation feature” in consensus treatment algorithms despite conflicting data. This study aimed to describe the relationship between decremental post-resuscitation pH and neurologic outcomes after OHCA. MethodsConsecutive OHCA patients treated with targeted temperature management (TTM) at multiple US centers from 2008 to 2017 were evaluated. Poor neurologic outcome at hospital discharge was defined as cerebral performance category ≥3. The exposure was initial arterial pH after return of spontaneous circulation (ROSC) analyzed in decremental 0.05 thresholds. Potential confounders (demographics, history, resuscitation characteristics, initial studies) were defined a priori and controlled for via ATT-weighting on the inverse propensity score plus direct adjustment for the linear propensity score. ResultsOf 723 patients, 589 (80%) experienced poor neurologic outcome at hospital discharge. After propensity-adjustment with excellent covariate balance, the adjusted odds ratios for poor neurologic outcome by pH threshold were: ≤7.3: 2.0 (1.0–4.0); ≤7.25: 1.9 (1.2–3.1); ≤7.2: 2.1 (1.3–3.3); ≤7.15: 1.9 (1.2–3.1); ≤7.1: 2.4 (1.4–4.1); ≤7.05: 3.1 (1.5–6.3); ≤7.0: 4.5 (1.8–12). ConclusionsNo increased hazard of progressively poor neurologic outcomes was observed in resuscitated OHCA patients treated with TTM until the initial post-ROSC arterial pH was at least ≤7.1. This threshold is more acidic than in current guidelines, suggesting the possibility that post-arrest pH may be utilized presently as an inappropriately-pessimistic prognosticator.

A machine learning based model for Out of Hospital cardiac arrest outcome classification and sensitivity analysis

BackgroundOut-of-hospital cardiac arrest (OHCA) affects nearly 400,000 people each year in the United States of which only 10% survive. Using data from the Cardiac Arrest Registry to Enhance Survival (CARES), and machine learning (ML) techniques, we developed a model of neurological outcome prediction for OHCA in Chicago, Illinois. MethodsRescue workflow data of 2639 patients with witnessed OHCA were retrieved from Chicago’s CARES. An Embedded Fully Convolutional Network (EFCN) classification model was selected to predict the patient outcome (survival with good neurological outcomes or not) based on 27 input features with the objective of maximizing the average class sensitivity. Using this model, sensitivity analysis of intervention variables such as bystander cardiopulmonary resuscitation (CPR), targeted temperature management, and coronary angiography was conducted. ResultsThe EFCN classification model has an average class sensitivity of 0.825. Sensitivity analysis of patient outcome shows that an additional 33 patients would have survived with good neurological outcome if they had received lay person CPR in addition to CPR by emergency medical services and 88 additional patients would have survived if they had received the coronary angiography intervention. ConclusionsML modeling of the complex Chicago OHCA rescue system can predict neurologic outcomes with a reasonable level of accuracy and can be used to support intervention decisions such as CPR or coronary angiography. The discriminative ability of this ML model requires validation in external cohorts to establish generalizability.

The prognostic value of optic nerve sheath diameter in patients with subarachnoid hemorrhage

BackgroundWe evaluated the role of optic nerve sheath diameter (ONSD) using brain computed tomography (CT) in predicting neurological outcomes of patients with subarachnoid hemorrhage (SAH).MethodsThis was a retrospective, multicenter, observational study of adult patients with SAH admitted between January 2012 and June 2017. Initial brain CT was performed within 12 h from onset of SAH, and follow-up brain CT was performed within 24 h from treatment of a ruptured aneurysm. Primary outcome was neurological status at 6-month follow-up assessed with the Glasgow Outcome Scale (GOS, 1 to 5).ResultsAmong 223 SAH patients, 202 (90.6%) survived until discharge. Of these survivors, 186 (83.4%) manifested favorable neurological outcomes (GOS of 3, 4, or 5). In this study, the ONSDs in the group of patients with poor neurological outcome were significantly greater than those in the favorable neurological outcome group (all p < 0.01). Intracranial pressure (ICP) was monitored in 21 (9.4%) patients during the follow-up CT. A linear correlation existed between the average ONSD and ICP in simple correlation analysis (r = 0.525, p = 0.036). Analysis of the receiver operating characteristic curve for prediction of poor neurological outcome showed that ONSD had considerable predictive value (C-statistics, 0.735 to 0.812). In addition, the performance of a composite of Hunt and Hess grade and ONSD was increasingly associated with poor neurological outcomes than the use of each marker alone.ConclusionsONSD measured with CT may be used in combination with clinical grading scales to improve prognostic accuracy in SAH patients.

Derivation and Validation of the SWAP Score for Very Early Prediction of Neurologic Outcome in Patients With Out-of-Hospital Cardiac Arrest

For patients with out-of-hospital cardiac arrest who receive cardiopulmonary resuscitation in an emergency department (ED), the early evaluation of their neurologic prognosis is essential for emergency physicians. The aim of this study is to establish a simple and useful assessment tool for rapidly estimating the prognosis of patients with out-of-hospital cardiac arrest after their arrival at an ED. A total of 852 patients admitted from January 1, 2015, to June 30, 2017, were prospectively registered and enrolled in thederivation cohort. Multivariate logistic regression on this cohort identified 4 independent factors associated with unfavorable outcomes: initial nonshockable rhythm (odds ratio [OR] 3.40; 95% confidence interval [CI] 1.58 to 7.32), no witness of collapse (OR3.19; 95% CI 1.51 to 6.75), older than 60 years (OR 3.65; 95% CI 1.64 to 8.09), and pH less than or equal to 7.00 (OR 3.27; 95% CI 1.42 to 7.54). The shockable rhythm-witness-age-pH (SWAP) score was developed and 1 point was assigned to each predictor. For a SWAP score of 4, the specificity was 97.14% (95% CI 91.62% to 100%) for unfavorable outcomes in the derivation cohort. For validation, we retrospectively collected data for 859 patients with out-of-hospital cardiac arrest from January 1, 2012, to December 31, 2014. A SWAP score of 4 was 100% specific (95% CI 99.9% to 100%) for unfavorable outcomes in the validation cohort. The SWAP score is a simple and useful predictive model that may provide information for the very early estimation of prognosis for patients with out-of-hospital cardiac arrest. Further research is required to integrate ultrasonographic findings and validate the SWAP score's application in other populations.

Assessment of serum biomarkers and coagulation/fibrinolysis markers for prediction of neurological outcomes of out of cardiac arrest patients treated with therapeutic hypothermia.

ObjectiveDespite increased survival in patients with cardiac arrest, it remains difficult to determine patient prognosis at the early stage. This study evaluated the prognosis of cardiac arrest patients using brain injury, inflammation, cardiovascular ischemic events, and coagulation/fibrinolysis markers collected 24, 48, and 72 hours after return of spontaneous circulation (ROSC).MethodsFrom January 2011 to December 2016, we retrospectively observed patients who underwent therapeutic hypothermia. Blood samples were collected immediately and 24, 48, and 72 hours after ROSC. Neuron-specific enolase (NSE), S100-B protein, procalcitonin, troponin I, creatine kinase-MB, pro-brain natriuretic protein, D-dimer, fibrin degradation product, antithrombin-III, fibrinogen, and lactate levels were measured. Prognosis was evaluated using Glasgow-Pittsburgh cerebral performance categories and the predictive accuracy of each marker was evaluated. The secondary outcome was whether the presence of multiple markers improved prediction accuracy.ResultsA total of 102 patients were included in the study: 39 with good neurologic outcomes and 63 with poor neurologic outcomes. The mean NSE level of good outcomes measured 72 hours after ROSC was 18.50 ng/mL. The area under the curve calculated on receiver operating characteristic analysis was 0.92, which showed the best predictive power among all markers included in the study analysis. The relative integrated discrimination improvement and category-free net reclassification improvement models showed no improvement in prognostic value when combined with all other markers and NSE (72 hours).ConclusionAlthough biomarker combinations did not improve prognostic accuracy, NSE (72 hours) showed the best predictive power for neurological prognosis in patients who received therapeutic hypothermia.

Multimodal assessment using early brain CT and blood pH improve prediction of neurologic outcomes after pediatric cardiac arrest

BackgroundEarly prediction of neurologic prognosis in children resuscitated from cardiac arrest is a major challenge. This study aimed to investigate the usefulness of a combined model based on brain computed tomography (CT) and initial blood gas analysis to predict neurologic prognoses in pediatric patients after cardiac arrest. MethodsWe retrospectively analyzed the medical records of patients resuscitated after cardiac arrest from 2000 to 2018. Patients aged one month to 18 years were included. Gray to white matter ratio (GWR), ambient cistern effacement (ACE), and blood gas analysis were studied. The primary outcome was neurological prognosis, which was evaluated using the Pediatric Cerebral Performance Category (PCPC) scale at discharge. ResultsOf 97 resuscitated patients, 64 brain CT images were available. Fourteen patients had a good neurologic outcome (PCPC 1–3) and 50 patients a poor neurologic outcome (PCPC 4–6). The multimodal model (AUC 0.897) containing GWR of basal ganglia (BG), ACE, and blood pH was found to be superior for predicting poor neurologic prognosis than single variable models (AUC of GWR-BG: 0.744, ACE: 0.804, pH: 0.747). Interestingly, we found the GWR-BG cutoff value for specificity 100% differed significantly between patients <4 years (cutoff value: 1.08, p = 0.04) and ≥4 years (cutoff value: 1.18, p = 0.004). ConclusionsThe combination of GWR-BG, ambient cistern effacement, and blood pH was found to usefully predict neurological outcome in children resuscitated from cardiac arrest. In addition, the cutoff value of GWR-BG for the prediction of neurologic outcome was found to increase with age.

Serum Neuron-Specific Enolase Level as Predictor of Neurologic Outcome after Aortic Surgery

This study aimed to evaluate the significance of serum neuron-specific enolase (NSE) level as a predictor of neurologic injury in thoracic aortic surgery. We neurologically assessed 60 consecutive patients who underwent thoracic aortic surgery for thoracic aortic aneurysm (n = 26) and aortic dissection (n = 34). Using moderate hypothermic circulatory arrest with antegrade cerebral perfusion, total arch replacement and hemiarch replacement were performed in 37 and 23 patients, respectively. Serum NSE levels in venous blood samples drawn before surgery and at 1 day after surgery were measured. Severity of neurologic injury was categorized as either uncomplicated (n = 48), temporary neurologic dysfunction (TND, n = 5), or permanent neurologic dysfunction (PND, n = 7). The extent of stroke was estimated on computed tomography or magnetic resonance imaging. The NSE level significantly differed among the three groups (PND > TND > uncomplicated) on the first postoperative day. Receiver-operating characteristic curve analysis showed that the cutoff value of NSE level was 34.14 ng/mL for neurologic injury (sensitivity, 0.769; specificity, 0.851) and 43.56 ng/mL for PND (sensitivity, 1.000; specificity, 0.963). The NSE level significantly correlated with the extent of stroke (r = 0.61, p < 0.001). Serum NSE level is a significant predictor of adverse neurologic outcomes and extent of stroke after thoracic aortic surgery.

Potential diagnostic and prognostic value of serum and cerebrospinal fluid biomarkers in traumatic spinal cord injury: A systematic review.

It remains unclear whether biomarkers in the serum or CSF can be used for diagnosis or prognosis of spinal cord injuries (SCI). Therefore, a systematic review was undertaken to evaluate the prognostic or diagnostic value of serum and CSF biomarkers in assessing the severity of SCI and the outcome of patients. Two independent reviewers summarized the human studies retrieved from the electronic databases of Medline, Embase, Scopus and ISI Web of Science until April 2018. Seventeen studies were included (1065 patients aged 16-94years old). Although the findings of the included studies suggest that inflammatory and structural proteins may be useful in assessing the severity of SCI and prediction of neurological outcome, the level of evidence is generally low. Given limitations to the available evidence, further investigation in this field is required using large prospective data sets with rigorous analysis of sensitivity, specificity and prediction.

Somatosensory Evoked Potentials for prediction of outcome in Pediatric Cardiac Intensive Care Unit

Somatosensory Evoked Potentials (SEPs) have been shown to provide prognostic information in both adults and children with acute brain injury. Aim of this study is to evaluate SEPs prospectively as a predictor of neurological outcome in children with ischemic brain injury. We performed SEPs in 29 consecutive patients with acute ischemic injury. SEPs were classified as bilaterally normal (SEPs-N/N), bilaterally absent (SEPs-A/A), unilaterally absent (SEPs-N/A). Ten pts have SEPs-N/N (outcome: 5 with normal outcome, 4 with epilepsy, 1 exitus); 3 pts have SEPs N/A (outcome: all with hemiparesis and one also with cognitive impairment); 15 pts with SEPs A/A (outcome: 9 exitus and 6 with tetraparesis and cognitive impairment). We suggest that SEPs may be a good predictor of neurological outcome in children with acute ischemic brain injury.

Blood microRNA-93 as an indicator for diagnosis and prediction of functional recovery of acute stroke patients

The present study evaluated the diagnostic and predictive potential of microRNA-93 in acute ischemic stroke (AIS) patients within 6 h of stroke onset and its regulation on microglial inflammation in vitro. Our results showed that the miR-93 levels in plasma and neutrophil detected by real-time PCR were evidently reduced in AIS patients, and Pearson's correlation analysis showed that miR-93 levels in plasma and neutrophils had a significant positive linear correlation. Moreover, miR-93 levels in plasma and neutrophils of stroke patients at time of admission were not correlated with infarct volume and NIHSS (National Institute of Health stroke scale) scores at admission, but neutrophil miR-93 levels were positively correlated with the Barthel Index 7 days after stroke. Importantly, miR-93 levels in plasma and neutrophil of AIS patients were negatively correlated with the expression of TNF-α and IL-10. Furthermore, in vitro treatment with miR-93 agomir decreased the OGD (Oxygen and glucose deprivation)–induced proliferation of BV2 microglial cells tested by Flow cytometry. We demonstrated that miR-93 in blood has a potential to facilitate the diagnosis and prediction of neurological outcomes of acute ischemic stroke, and is involved in inflammation possibly through targeting the proliferation of microglia.

Early serum miR-1297 is an indicator of poor neurological outcome in patients with aSAH

Objective: MiRNAs are important regulators of translation and have been described as biomarkers of a number of cardiovascular diseases, including stroke. The purpose of the study was to determine expression levels of serum miR-1297 in patients with aneurysmal subarachnoid hemorrhage (aSAH), and to assess whether miR-1297 was the prognostic indicator of aSAH. Methods: We treated 128 aSAH patients with endovascular coiling. The World Federation of Neurological Surgeons (WFNS) grades, Hunt–Hess grades, and modified Fisher scores were used to assess aSAH severity. Neurologic outcome was assessed using the Modified Rankin Scale (mRS) at 1-year post-aSAH. Serum was taken at various time points (24, 72, and 168 h, and 14 days). Serum samples from aSAH patients and healthy controls were subjected to reverse transcription (RT) quantitative real-time PCR (RT-qPCR). Results: A poor outcome at 1 year was associated with significantly higher levels of miR-1297 value at the four time points, higher WFNS grade, higher Hunt–Hess grade, and higher Fisher score. Serum miR-1297 levels were significantly higher in patients, compared with healthy controls. There were significant correlations of miR-1297 concentrations in serum with severity in aSAH. The AUCs of miR-1297 at the four time points for distinguishing the aSAH patients from healthy controls were 0.80, 0.94, 0.77, and 0.59, respectively. After multivariate logistic regression analysis, only miR-1297 at 24 and 72 h enabled prediction of neurological outcome at 1 year. Conclusion: Serum was an independent predictive factor of poor outcome at 1 year following aSAH. This result supports the use of miR-1297 in aSAH to aid determination of prognosis.

Abstract 351: Predictors of Neurological Outcome After Out-of-Hospital Cardiac Arrest With Ventricular Fibrillation Witnessed by Emergency Medical Service P.ersonnel., JCS-ReSS Study

Background: Effective advanced life support is one of the important link in the chain of survival. In Japan, the emergency medical service (EMS) personnel can perform defibrillation, advanced airway management, intravenous access and administration of epinephrine as an advanced life support intervention for the treatment of out-of-hospital cardiac arrest (OHCA). However, whether these interventions performed by EMS improves neurological outcomes remains unclear. Objective: To evaluate predictors of favorable neurological outcome in patients suffering OHCA with ventricular fibrillation (VF) witnessed by an EMS personnel. Methods: The Fire and Disaster Management Agency (FDMA) of Japan developed a nationwide database of a prospective population-based cohort using an Utstein-style template for OHCA patients since January 2005. To evaluate data after the publication of Guideline2010, data from January 2011 to December 2015 of this database was used for the current analysis. A multivariate logistic-regression analysis was performed to assess factors associated with favorable neurological outcome (defined as Cerebral Performance Category 1 or 2) 1 month after cardiac arrest. Results: Of the 629,471 patients documented for the study period, 2,301 adult patients with an OHCA of cardiac origin and VF for the initial rhythm witnessed by an EMS personnel were included in the present analysis. The overall mortality was 49.6%. Rate of return of spontaneous circulation and favorable neurological outcome were 53.4% and 44.8%, respectively. High age (OR0.387, 95%CI0.316-0.472, p&lt;0.001), delayed defibrillation (OR0.598, 95%CI0.493-0.723, p&lt;0.001), advanced airway management (OR0.305, 95%CI0.223-0.413, p&lt;0.001), administration of epinephrine (OR0.356, 95%CI0.213-0.585, p&lt;0.001) and multiple attempts of defibrillation (OR0.484, 95%CI0.402-0.582, p&lt;0.001) were negatively associated with favorable neurological outcome. Conclusion: In patients with VF witnessed by EMS personnel, resuscitation efforts should simply focus on early defibrillation and CPR without advanced interventions.

The revised Cerebral Recovery Index improves predictions of neurological outcome after cardiac arrest

ObjectiveAnalysis of the electroencephalogram (EEG) background pattern helps predicting neurological outcome of comatose patients after cardiac arrest (CA). Visual analysis may not extract all discriminative information. We present predictive values of the revised Cerebral Recovery Index (rCRI), based on continuous extraction and combination of a large set of evolving quantitative EEG (qEEG) features and machine learning techniques. MethodsWe included 551 subsequent patients from a prospective cohort study on continuous EEG after CA in two hospitals. Outcome at six months was classified as good (Cerebral Performance Category (CPC) 1-2) or poor (CPC 3-5). Forty-four qEEG features (from time, frequency and entropy domain) were selected by the least absolute shrinkage and selection operator (LASSO) method and used in a Random Forests classification system. We trained and evaluated the system with 10-fold cross validation. For poor outcome prediction, the sensitivity at 100% specificity (Se100) and the area under the receiver operator curve (AUC) were used as performance of the prediction model. For good outcome, we used the sensitivity at 95% specificity (Se95). ResultsTwo hundred fifty-six (47%) patients had a good outcome. The rCRI predicted poor outcome with AUC = 0.94 (95% CI: 0.83–0.91), Se100 = 0.66 (0.65–0.78), and AUC = 0.88 (0.78–0.93), Se100 = 0.60 (0.51–0.75) at 12 and 24 h after CA, respectively. The rCRI predicted good outcome with Se95 = 0.72 (0.61–0.85) and 0.40 (0.30–0.51) at 12 and 24 h after CA, respectively. ConclusionsResults obtained in this study suggest that with machine learning algorithms and large set of qEEG features, it is possible to efficiently monitor patient outcome after CA. We also demonstrate the importance of selection of optimal performance metric to train a classifier model for outcome prediction. SignificanceThe rCRI is a sensitive, reliable predictor of neurological outcome of comatose patients after CA.

Prediction of neurological outcomes following the return of spontaneous circulation in patients with out-of-hospital cardiac arrest: Retrospective fast-and-frugal tree analysis

AimAlthough various quantitative methods have been developed for predicting neurological prognosis in patients with out-of-hospital cardiac arrest (OHCA), they are too complex for use in clinical practice. We aimed to develop a simple decision rule for predicting neurological outcomes following the return of spontaneous circulation (ROSC) in patients with OHCA using fast-and-frugal tree (FFT) analysis. MethodsWe performed a retrospective analysis of prospectively collected data archived in a multi-centre registry. Good neurological outcomes were defined as cerebral performance category (CPC) values of 1 or 2 at 28-day. Variables used for FFT analysis included age, sex, witnessed cardiac arrest, bystander cardiopulmonary resuscitation, initial shockable rhythm, prehospital defibrillation, prehospital ROSC, no flow time, low flow time, cause of arrest (cardiac or non-cardiac), pupillary light reflex, and Glasgow Coma Scale score after ROSC. ResultsAmong the 456 patients enrolled, 86 (18.9%) experienced good neurological outcomes. Prehospital ROSC (true = good), prompt or sluggish light reflex response after ROSC (true = good), and presumed cardiac cause (true = good, false = poor) were selected as nodes for the decision tree. Sensitivity, specificity, positive predictive value, and negative predictive value of the decision tree for predicting good neurological outcomes were 100% (42/42), 64.0% (119/186), 38.5% (42/109), and 100% (119/119) in the training set and 95.5% (42/44), 57.6% (106/184), 35.0% (42/120), and 98.1% (106/108) in the test set, respectively. ConclusionA simple decision rule developed via FFT analysis can aid clinicians in predicting neurological outcomes following ROSC in patients with OHCA.

Blood Lactate or Lactate Clearance: Which Is Robust to Predict the Neurological Outcomes after Cardiac Arrest? A Systematic Review and Meta-Analysis.

Aims Lactate and lactate clearance were supposed to be associated with cardiac arrest outcomes, but studies obtained different results. Thus, we conducted this meta-analysis to investigate the association between lactate or lactate clearance and neurological outcomes and their usefulness for prediction of neurological outcomes. Methods We conducted a systematic search in PubMed, Web of science, EMBASE, Medline, and Google Scholar until May 1, 2018, for relevant studies. Studies reporting lactate, lactate clearance on admission, or other time points after admission associated with neurological outcomes were included in our analysis. Pooled effect date was shown as weighed mean difference (WMD) and 95% confidence interval (CI). To measure the usefulness of lactate on admission to predict neurological outcomes, we also pooled the data of diagnostic test. Results 23 studies involving 6720 cardiac arrest (CA) patients were included. Results from our analysis indicated that patients with good neurological outcomes tended to have a lower lactate level on admission (WMD: -2.66 mmol/L, 95%CI: -3.39 to -1.93) and 12h, 24h, and 48h after admission (P<0.001). Furthermore, the pooled AUC for lactate level on admission to predict neurological outcomes was 0.77 (95%CI: 0.73-0.80). However, a significant association between lactate clearance and neurological outcomes was only found in 24h but not 12h lactate clearance rate. Conclusions Lactate levels on admission and all time points up to 48h were associated with neurological outcomes after CA, whereas the association between lactate clearance and neurological outcomes was not so stable. Lactate was a more robust surrogate marker than lactate clearance to predict neurological outcomes after CA.

Cushing response-based warning system for intensive care of brain-injured patients

ObjectiveCushing response (CR) is categorized. Wavelet transform (WT) and decision tree (DT) are utilized to analyze physiological signals from neurocritical patients. A warning model is built for recognition of CR, real-time evaluation of intracranial condition and prediction of neurological outcome. MethodsPhysiological signals of neurocritical patients are preprocessed by WT and compressed by linear regression. An algorithm labels each segment as pathological, physiological, negative or uncertain CR. The DT identifies CR. Continuous data input to the established DT predicts condition at that moment and following outcome. ResultsFrom 33 neurocritical patients, 422,524 sets of physiological signals were collected. The cross-validation scores of DT ranged from 0.562 to 0.579 with averaged accuracy rate 60.6% (3.5–98.1%). The model correctly predicted the outcome of the training group, 87.9% in accuracy. The ratios of pathological CR were 9.3 ± 16.6%, 74.2 ± 29.7% and 99.7 ± 0.3% in patients of good, coma and death groups, respectively. The prediction accuracy for a test set of 103 patients reached 81.6%. ConclusionsCushing response categorization helps in identifying critical conditions and predicting outcome. SignificanceA novel concept of four categories of Cushing response is proposed to represent broader ranges of intracranial change.