A Systematic Review of Neuromonitoring Modalities in Children Beyond Neonatal Period After Cardiac Arrest

Prognostic Value of Brain Magnetic Resonance Imaging in Children After Out-of-Hospital Cardiac Arrest: Predictive Value of Normal Magnetic Resonance Imaging for a Favorable Two-Year Outcome

BACKGROUND

Determine the predictive value of brain magnetic resonance imaging (MRI) findings less than or equal to seven days post-pediatric out-of-hospital cardiac arrest (OHCA) for long-term outcomes.

METHODS

This retrospective single-center study included children (zero to 17 years) with OHCA admitted to a tertiary care hospital pediatrc intensive care unit from 2012 to 2020 who underwent brain MRI at most seven days postarrest. A neuroimaging scoring system was designed, using T1-, T2-, and diffusion-weighted images based on previously published scores and brain injury patterns. Extensive brain injury was defined as ≥50% cortex/white matter injury or four or more of nine predefined brain regions. Pediatric cerebral performance category (PCPC) scores were determined at hospital discharge and two years post-OHCA as part of routine follow-up care. Favorable neurological outcomes were defined as PCPC scores of 1 to 2 or no change from prearrest status.

RESULTS

Among 142 children, 56 had a brain MRI at less than or equal to seven days postarrest. Median arrest age was 3.3 years (first and third quartiles [Q1, Q3]: 0.6, 13.6), and 64% were male. Brain MRI was obtained four days post-OHCA (Q1, Q3: 3, 5). Normal brain MRI findings (i.e., negative test result) predicted favorable outcomes with 100% negative predictive value, whereas extensive injury (i.e., positive test result) predicted unfavorable outcomes and death with 100% positive predictive value.

CONCLUSIONS

A normal brain MRI at less than or equal to seven days postarrest predicts favorable neurological outcomes two years later, whereas extensive brain injury predicts unfavorable neurological outcomes or death at discharge and two years post-OHCA.

Practice patterns for acquiring neuroimaging after pediatric in-hospital cardiac arrest

AIMS

To determine which patient and cardiac arrest factors were associated with obtaining neuroimaging after in-hospital cardiac arrest, and among those patients who had neuroimaging, factors associated with which neuroimaging modality was obtained.

METHODS

Retrospective cohort study of patients who survived in-hospital cardiac arrest (IHCA) and were enrolled in the ICU-RESUS trial (NCT02837497).

RESULTS

We tabulated ultrasound (US), CT, and MRI frequency within 7 days following IHCA and identified patient and cardiac arrest factors associated with neuroimaging modalities utilized. Multivariable models determined which factors were associated with obtaining neuroimaging. Of 1000 patients, 44% had ≥ 1 neuroimaging study (US in 31%, CT in 18%, and MRI in 6% of patients). Initial USs were performed a median of 0.3 [0.1,0.5], CTs 1.4 [0.4,2.8], and MRIs 4.1 [2.2,5.1] days post-arrest. Neuroimaging timing and frequency varied by site. Factors associated with greater odds of neuroimaging were cardiac arrest in CICU (versus PICU), longer duration CPR, receiving ECMO post-arrest, and post-arrest care with targeted temperature management or EEG monitoring. US performance was associated with congenital heart disease. CT was associated with age ≥ 1-month, greater pre-arrest disability, and receiving CPR for ≥ 16 min. MRI utilization increased with pre-existing respiratory insufficiency and respiratory decompensation as arrest cause, and medical cardiac and surgical non-cardiac or trauma illness category. Overall, if neuroimaging was obtained, US was more common in CICU while CT/MRI were utilized more in PICU.

CONCLUSIONS

Practice patterns for acquiring neuroimaging after IHCA are variable and influenced by patient, cardiac arrest, and site factors.

Prediction of good neurological outcome after return of circulation following paediatric cardiac arrest: A systematic review and meta-analysis

AIM

To evaluate the ability of blood-biomarkers, clinical examination, electrophysiology, or neuroimaging, assessed within 14 days from return of circulation to predict good neurological outcome in children following out- or in-hospital cardiac arrest.

METHODS

Medline, EMBASE and Cochrane Trials databases were searched (2010-2023). Sensitivity and false positive rates (FPR) for good neurological outcome (defined as either 'no, mild, moderate disability or minimal change from baseline') in paediatric survivors were calculated for each predictor. Risk of bias was assessed using the QUIPS tool.

RESULTS

Thirty-five studies (2974 children) were included. The presence of any of the following had a FPR < 30% for predicting good neurological outcome with moderate (50-75%) or high (>75%) sensitivity: bilateral reactive pupillary light response within 12 h; motor component ≥ 4 on the Glasgow Coma Scale score at 6 h; bilateral somatosensory evoked potentials at 24-72 h; sleep spindles, and continuous cortical activity on electroencephalography within 24 h; or a normal brain MRI at 4-6d. Early (≤12 h) normal lactate levels (<2mmol/L) or normal s100b, NSE or MBP levels predicted good neurological outcome with FPR rate < 30% and low (<50%) sensitivity. All studies had moderate to high risk of bias with timing of measurement, definition of test, use of multi-modal tests, or outcome assessment heterogeneity.

CONCLUSIONS

Clinical examination, electrophysiology, neuroimaging or blood-biomarkers as individual tests can predict good neurological outcome after cardiac arrest in children. However, evidence is often low quality and studies are heterogeneous. Use of a standardised, multimodal, prognostic algorithm should be studied and is likely of added value over single modality testing.

Prediction of Survival After Pediatric Cardiac Arrest Using Quantitative EEG and Machine Learning Techniques

BACKGROUND AND OBJECTIVES

Early neuroprognostication in children with reduced consciousness after cardiac arrest (CA) is a major clinical challenge. EEG is frequently used for neuroprognostication in adults, but has not been sufficiently validated for this indication in children. Using machine learning techniques, we studied the predictive value of quantitative EEG (qEEG) features for survival 12 months after CA, based on EEG recordings obtained 24 hours after CA in children. The results were confirmed through visual analysis of EEG background patterns.

METHODS

This is a retrospective single-center study including children (0-17 years) with CA, who were subsequently admitted to the pediatric intensive care unit (PICU) of a tertiary care hospital between 2012 and 2021 after return of circulation (ROC) and were monitored using EEG at 24 hours after ROC. Signal features were extracted from a 30-minute EEG segment 24 hours after CA and used to train a random forest model. The background pattern from the same EEG fragment was visually classified. The primary outcome was survival or death 12 months after CA. Analysis of the prognostic accuracy of the model included calculation of receiver-operating characteristic and predictive values. Feature contribution to the model was analyzed using Shapley values.

RESULTS

Eighty-six children were included (in-hospital CA 27%, out-of-hospital CA 73%). The median age at CA was 2.6 years; 53 (62%) were male. Mortality at 12 months was 56%; main causes of death on the PICU were withdrawal of life-sustaining therapies because of poor neurologic prognosis (52%) and brain death (31%). The random forest model was able to predict death at 12 months with an accuracy of 0.77 and positive predictive value of 1.0. Continuity and amplitude of the EEG signal were the signal parameters most contributing to the model classification. Visual analysis showed that no patients with a background pattern other than continuous with amplitudes exceeding 20 μV were alive after 12 months.

DISCUSSION

Both qEEG and visual EEG background classification for registrations obtained 24 hours after ROC form a strong predictor of nonsurvival 12 months after CA in children.

Pediatric Extracorporeal Cardiopulmonary Resuscitation: Development of a Porcine Model and the Influence of Cardiopulmonary Resuscitation Duration on Brain Injury

Background The primary objective was to develop a porcine model of prolonged (30 or 60 minutes) pediatric cardiopulmonary resuscitation (CPR) followed by 22- to 24-hour survival with extracorporeal life support, and secondarily to evaluate differences in neurologic injury. Methods and Results Ten-kilogram, 4-week-old female piglets were used. First, model development established the technique (n=8). Then, a pilot study was conducted (n=15). After 80% survival was achieved in the final 5 pilot animals, a proof-of-concept randomized study was completed (n=11). Shams (n=6) underwent anesthesia only. Severe neurological injury was determined by a composite score of mitochondrial function, neuropathology, and cerebral metabolism: scale of 0-6 (severe: >3). Among 15 piglets in the pilot study, overall survival was 10 (67%); of the final 5, overall survival was 4 (80%). Eleven piglets were then randomized to 60 (CPR60, n=5) or 30 minutes of CPR (CPR30, n=5); 1 animal was excluded from prerandomization for intra-abdominal hemorrhage (10/11, 91% survival). Three of 5 animals in the CPR60 group had severe neurological injury scores versus 1 of 5 in the CPR30 group (P=0.52). During ECMO, CPR60 animals had lower pH (CPR60: 7.4 [IQR 7.4-7.4] versus CPR30: 7.5 [IQR 7.4-7.5], P=0.022), higher lactate (CPR60: 6.8 [IQR 6.8-11] versus CPR30: 4.2 [IQR 4.1-4.3] mmol/L; P=0.012), and higher ICP (CPR60: 19.3 [IQR 11.7-29.3] versus CPR30: 7.9 [IQR 6.7-9.3] mm Hg; P=0.037). Both groups had greater mitochondrial injury than shams (CPR60: P<0.001; CPR30: P<0.001). CPR60 did not differ from CPR30 in mitochondrial respiration, neuropathology, or cerebral metabolism. Conclusions A pediatric porcine model of extracorporeal cardiopulmonary resuscitation after 60 and 30 minutes of CPR consistently resulted in 24-hour survival with more severe lactic acidosis in the 60-minute cohort.

Association Between Quantitative Diffusion-Weighted Magnetic Resonance Neuroimaging and Outcome After Pediatric Cardiac Arrest

BACKGROUND AND OBJECTIVES

Diffusion MRI can quantify the extent of hypoxic-ischemic brain injury after cardiac arrest. Our objective was to determine the association between the adult-derived threshold of apparent diffusion coefficient (ADC) <650 × 10-6 mm2/s in >10% of brain tissue and an unfavorable outcome after pediatric cardiac arrest. Since ADC decreases exponentially as a function of increasing age, we determined the association between (1) having >10% of brain tissue below a novel age-dependent ADC threshold, and (2) age-normalized whole-brain mean ADC and unfavorable outcome.

METHODS

This was a retrospective study of patients aged ≤18 years who had cardiac arrest and a clinically obtained brain MRI within 7 days. The primary outcome was unfavorable neurologic status at hospital discharge based on the Pediatric Cerebral Performance Category score. ADC images were extracted from 3-direction diffusion imaging. We determined whether each patient had >10% of voxels with an ADC below prespecified thresholds. We computed the whole-brain mean ADC for each patient.

RESULTS

One hundred thirty-four patients were analyzed. Patients with ADC <650 × 10-6 mm2/s in >10% of voxels had 15 times higher odds (95% CI 5-65) of an unfavorable outcome compared with patients with ADC <650 × 10-6 mm2/s (area under the receiver operating characteristic curve [AUROC] 0.72 [95% CI 0.63-0.80]). These ADC criteria had a sensitivity and specificity of 0.49 and 0.94, respectively, and positive and negative predictive values of 0.93 and 0.52, respectively, for an unfavorable outcome. The age-dependent ADC threshold that yielded optimal sensitivity and specificity for unfavorable outcomes was <300 × 10-6 mm2/s below each patient's predicted whole-brain mean ADC. The sensitivity, specificity, and positive and negative predictive values for this ADC threshold were 0.53, 0.96, 0.96, and 0.54, respectively (odds ratio [OR] 26.4 [95% CI 7.5-168.3]; AUROC 0.74 [95% CI 0.66-0.83]). Lower age-normalized whole-brain mean ADC was also associated with an unfavorable outcome (OR 0.42 [0.24-0.64], AUROC 0.76 [95% CI 0.66-0.82]).

DISCUSSION

Quantitative diffusion thresholds on MRI within 7 days after cardiac arrest were associated with an unfavorable outcome in children. The age-independent ADC threshold was highly specific for predicting an unfavorable outcome. However, the specificity and sensitivity increased when using age-dependent ADC thresholds. Age-dependent ADC thresholds may improve prognostic accuracy and require further investigation in larger cohorts.

CLASSIFICATION OF EVIDENCE

This study provides Class III evidence that quantitative diffusion-weighted imaging within 7 days postarrest can predict an unfavorable clinical outcome in children.

Multimodal monitoring including early EEG improves stratification of brain injury severity after pediatric cardiac arrest

AIMS

Assessment of brain injury severity early after cardiac arrest (CA) may guide therapeutic interventions and help clinicians counsel families regarding neurologic prognosis. We aimed to determine whether adding EEG features to predictive models including clinical variables and examination signs increased the accuracy of short-term neurobehavioral outcome prediction.

METHODS

This was a prospective, observational, single-center study of consecutive infants and children resuscitated from CA. Standardized EEG scoring was performed by an electroencephalographer for the initial EEG timepoint after return of spontaneous circulation (ROSC) and each 12-h segment from the time of ROSC up to 48 h. EEG Background Category was scored as: (1) normal; (2) slow-disorganized; (3) discontinuous or burst-suppression; or (4) attenuated-featureless. The primary outcome was neurobehavioral outcome at discharge from the Pediatric Intensive Care Unit. To develop the final predictive model, we compared areas under the receiver operating characteristic curves (AUROC) from models with varying combinations of Demographic/Arrest Variables, Examination Signs, and EEG Features.

RESULTS

We evaluated 89 infants and children. Initial EEG Background Category was normal in 9 subjects (10%), slow-disorganized in 44 (49%), discontinuous or burst suppression in 22 (25%), and attenuated-featureless in 14 (16%). The final model included Demographic/Arrest Variables (witnessed status, doses of epinephrine, initial lactate after ROSC) and EEG Background Category which achieved AUROC of 0.9 for unfavorable neurobehavioral outcome and 0.83 for mortality.

CONCLUSIONS

The addition of standardized EEG Background Categories to readily available CA variables significantly improved early stratification of brain injury severity after pediatric CA.