Feasibility of non-invasive neuro-monitoring during extracorporeal membrane oxygenation in children.

INTRODUCTION: Detection of neurological complications during extracorporeal membrane oxygenation (ECMO) may be enhanced with non-invasive neuro-monitoring. We investigated the feasibility of non-invasive neuro-monitoring in a paediatric intensive care (PIC) setting. METHODS: In a single centre, prospective cohort study we assessed feasibility of recruitment, and neuro-monitoring via somatosensory evoked potentials (SSEP), electroencephalography (EEG) and near infrared spectroscopy (NIRS) during venoarterial (VA) ECMO in paediatric patients (0-15 years). Measures were obtained within 24h of cannulation, during an intermediate period, and finally at decannulation or echo stress testing. SSEP/EEG/NIRS measures were correlated with neuro-radiology findings, and clinical outcome assessed via the Pediatric cerebral performance category (PCPC) scale 30 days post ECMO cannulation. RESULTS: We recruited 14/20 (70%) eligible patients (median age: 9 months; IQR:4-54, 57% male) over an 18-month period, resulting in a total of 42 possible SSEP/EEG/NIRS measurements. Of these, 32/42 (76%) were completed. Missed recordings were due to lack of access/consent within 24 h of cannulation (5/42, 12%) or PIC death/discharge (5/42, 12%). In each patient, the majority of SSEP (8/14, 57%), EEG (8/14, 57%) and NIRS (11/14, 79%) test results were within normal limits. All patients with abnormal neuroradiology (4/10, 40%), and 6/7 (86%) with poor outcome (PCPC ≥4) developed indirect SSEP, EEG or NIRS measures of neurological complications prior to decannulation. No study-related adverse events or neuro-monitoring data interpreting issues were experienced. CONCLUSION: Non-invasive neuro-monitoring (SSEP/EEG/NIRS) during ECMO is feasible and may provide early indication of neurological complications in this high-risk population.

Amplitude-reduction alert criteria and intervention during complex paediatric cervical spine surgery.

Objective: To determine the utility of widely used intraoperative neuromonitoring (IONM) alert criteria and intervention for predicting postoperative outcome following paediatric spinal surgery. Methods: Retrospective analysis of somatosensory evoked potentials (SSEP) and motor evoked potentials (MEP) in consecutive cervical spine fixations. An intervention protocol followed amplitude-reductions in SSEPs (≥50 %) and/or MEPs (≥80 %). Alert breaches were reversed when SSEP/MEP amplitude was restored to > 50 %/20 % of baseline. Sensorimotor function was assessed preoperatively and 3-months postoperatively via the Modified McCormick Scale score (MMS). We explored associations between postoperative outcome, demographic/surgical and IONM variables. Results: Forty-five procedures in 38 children (mean age:9 ±â€¯4 years;55 % female) were monitored, 42 %of which breached alert criteria. Instrumentation (6/19,32 %) and hypotension (5/19,26 %) were common causes for alert and the majority (13/19,68 %) were reversed following intervention. There was an association between pre- and post-MMS and the type of breach (p = 0.002). All children with worse postoperative MMS (3/38,8%) had irreversible breaches. Conclusions: IONM in this small sample accurately detected neurological injury. The majority of breaches reversed following an intervention protocol. Irreversible breaches frequently led to worse postoperative sensorimotor function. Significance: An intervention protocol which reversed IONM alerts never resulted in postoperative worsening of sensorimotor function.

Perioperative electroencephalography in cardiac surgery with hypothermic circulatory arrest: a narrative review.

OBJECTIVES: Cardiac surgery with hypothermic circulatory arrest (HCA) is associated with neurological morbidity of variable severity and electroencephalography (EEG) is a sensitive proxy measure of brain injury. We conducted a narrative review of the literature to evaluate the role of perioperative EEG monitoring in cardiac surgery involving HCA. METHODS: Medline, Embase, Central and LILACS databases were searched to identify studies utilizing perioperative EEG during surgery with HCA in all age groups, published since 1985 in any language. We aimed to compare EEG use with no use but due to the lack of comparative studies, we performed a narrative review of its utility. Two or more reviewers independently screened studies for eligibility and extracted data. RESULTS: Fourty single-centre studies with a total of 3287 patients undergoing surgery were identified. Most were observational cohort studies (34, 85%) with only 1 directly comparing EEG use with no use. EEG continuity (18, 45%), seizures (15, 38%) and electrocerebral inactivity prior to circulatory arrest (15, 38%) were used to detect, monitor, prevent and prognose neurological injury. Neurological dysfunction was reported in almost all studies and occurred in 0-21% of patients. However, the heterogeneity of reported clinical and EEG outcome measures prevented meta-analysis. CONCLUSIONS: EEG is used to detect cortical ischaemia and seizures and predict neurological abnormalities and may guide intraoperative cerebral protection. However, there is a lack of comparative data demonstrating the benefit of perioperative EEG monitoring. Use of a standardized methodology for performing EEG and reporting outcome metrics would facilitate the conduct of high-quality clinical trials.

Reliability in the assessment of paediatric somatosensory evoked potentials post cardiac arrest.

OBJECTIVE: To measure inter- and intra-rater agreement in the interpretation of cortical somatosensory evoked potential (SSEP) components following paediatric cardiac arrest (CA) in multi-professional neurophysiology teams. METHODS: Thirteen professionals blinded to patient outcome interpreted 96 SSEPs in paediatric patients 24-/48-/72-hours following CA. Of these, 34 were duplicates used to assess intra-rater agreement. Consistent interpretations (absent/present/indeterminate) between scientists (who record/identify SSEP components) and neurophysiologists (who provide prognostic SSEP interpretation) were expressed as percentages. Rates of agreement were calculated using Fleiss' kappa coefficient (K). RESULTS: Unanimous agreement between professionals was present in 40% (95%CI: 28-54%) of the interpreted SSEPs, with a K value of 0.62 (95%CI: 0.55-0.70) based on average agreement. Agreement was similar between neurophysiologists (K = 0.67; 95%CI: 0.57-0.77) and scientists (K = 0.62; 95%CI: 0.54-0.70) but lower in patients < 2 years old (K = 0.23; 95%CI: 0.14-0.33) and in those with poor outcome (K = 0.21; 95%CI: 0.07-0.35). No SSEP was unanimously interpreted as absent and 92% (95%CI: 89-95%) of duplicate SSEPs were interpreted consistently. CONCLUSION: Despite substantial agreement when interpreting prognostic SSEPs, this was significantly lower in children with poor outcome and of younger age. SIGNIFICANCE: Clinicians using SSEPs in the intensive care unit should be aware of the inter-rater variability when interpreting SSEPs as absent.

The Prognostic Value of Somatosensory Evoked Potentials in Children After Cardiac Arrest: The SEPIA Study.

PURPOSE: Absent cortical somatosensory evoked potentials (SSEPs) reliably predict poor neurologic outcome in adults after cardiac arrest (CA). However, there is less evidence to support this in children. In addition, targeted temperature management, test timing, and a lack of blinding may affect test accuracy. METHODS: A single-center, prospective cohort study of pediatric (aged 24 hours to 15 years) patients in which prognostic value of SSEPs were assessed 24, 48, and 72 hours after CA. Targeted temperature management (33-34°C for 24 hours) followed by gradual rewarming to 37°C was used. Somatosensory evoked potentials were graded as present, absent, or indeterminate, and results were blinded to clinicians. Neurologic outcome was graded as "good" (score 1-3) or "poor" (4-6) using the Pediatric Cerebral Performance Category scale 30 days after CA and blinded to SSEP interpreter. RESULTS: Twelve patients (median age, 12 months; interquartile range, 2-150; 92% male) had SSEPs interpreted as absent (6/12) or present (6/12) <72 hours after CA. Outcome was good in 7 of 12 patients (58%) and poor in 5 of 12 patients (42%). Absent SSEPs predicted poor outcome with 88% specificity (95% confidence interval, 53% to 98%). One patient with an absent SSEP had good outcome (Pediatric Cerebral Performance Category 3), and all patients with present SSEPs had good outcome (specificity 100%; 95% confidence interval, 51% to 100%). Absence or presence of SSEP was consistent across 24-hour (temperature = 34°C), 48-hour (t = 36°C), and 72-hour (t = 36°C) recordings after CA. CONCLUSIONS: Results support SSEP utility when predicting favorable outcome; however, predictions resulting in withdrawal of life support should be made with caution and never in isolation because in this very small sample there was a false prediction of unfavorable outcome. Further prospective, blinded studies are needed and encouraged.