Detecting Cerebral Ischemia From Electroencephalography During Carotid Endarterectomy Using Machine Learning.

Monitoring cerebral neuronal activity via electroencephalography (EEG) during surgery can detect ischemia, a precursor to stroke. However, current neurophysiologist-based monitoring is prone to error. In this study, we evaluated machine learning (ML) for efficient and accurate ischemia detection. We trained supervised ML models on a dataset of 802 patients with intraoperative ischemia labels and evaluated them on an independent validation dataset of 30 patients with refined labels from five neurophysiologists. Our results show moderate-to-substantial agreement between neurophysiologists, with Cohen's kappa values between 0.59 and 0.74. Neurophysiologist performance ranged from 58-93% for sensitivity and 83-96% for specificity, while ML models demonstrated comparable ranges of 63-89% and 85-96%. Random Forest (RF), LightGBM (LGBM), and XGBoost RF achieved area under the receiver operating characteristic curve (AUROC) values of 0.92-0.93 and area under the precision-recall curve (AUPRC) values of 0.79-0.83. ML has the potential to improve intraoperative monitoring, enhancing patient safety and reducing costs.

Role of Intraoperative Electroencephalography in Predicting Postoperative Delirium in Patients Undergoing Cardiovascular Surgeries.

OBJECTIVE: To determine the utility of electroencephalography (EEG) in predicting postoperative delirium (POD) in patients who underwent cardiovascular surgeries with EEG monitoring. METHODS: A total of 1161 patients who underwent cardiovascular surgeries with EEG monitoring were included in the study, and their data were retrospectively reviewed. POD assessment was done utilizing Intensive Care Delirium Screening Checklist (ICDSC). Patients with a score of > 4 on ICDSC were diagnosed with POD. RESULTS: Of 1161 patients, 131 patients had EEG changes and 56 (42.74%) of 131 patients experienced POD. Of 1030 patients without EEG changes, 219 (21.26%) experienced POD. EEG showed specificity of 91.5% and negative predictive value of 78.7% in detecting POD. On multivariable analysis, EEG changes showed a strong association with POD (ORadj 1.97 CI (1.30-2.99), p = 0.001) with persistent EEG changes showing even a higher risk of developing POD (ORadj 2.65 (1.43-4.92), p = 0.002). CONCLUSION: EEG change has specificity of 91.5% emphasizing the need for its implementation as a diagnostic tool for predicting POD. Patients with POD are two times more likely to experience significant EEG changes, especially persistent EEG changes when undergoing cardiovascular surgeries. SIGNIFICANCE: Intraoperative EEG can detect POD, and EEG changes based therapeutic interventions can mitigate POD.

Intraoperative neuromonitoring as real-time diagnostic for cerebral ischemia in endovascular treatment of ruptured brain aneurysms.

OBJECTIVE: To evaluate the diagnostic accuracy of intraoperative neurophysiological monitoring (IONM) during endovascular treatment (EVT) of ruptured intracranial aneurysms (rIA). METHODS: IONM and clinical data from 323 patients who underwent EVT for rIA from 2014-2019 were retrospectively reviewed. Significant IONM changes and outcomes were evaluated based on visual review of data and clinical documentation. RESULTS: Of the 323 patients undergoing EVT, significant IONM changes were noted in 30 patients (9.29%) and 46 (14.24%) experienced postprocedural neurological deficits (PPND). 22 out of 30 (73.33%) patients who had significant IONM changes experienced PPND. Univariable analysis showed changes in somatosensory evoked potential (SSEP) and electroencephalogram (EEG) were associated with PPND (p-values: <0.001 and <0.001, retrospectively). Multivariable analysis showed that IONM changes were significantly associated with PPND (Odd ratio (OR) 20.18 (95%CI:7.40-55.03, p-value: <0.001)). Simultaneous changes in both IONM modalities had specificity of 98.9% (95% CI: 97.1%-99.7%). While sensitivity when either modality had a change was 47.8% (95% CI: 33.9%-62.0%) to predict PPND. CONCLUSIONS: Significant IONM changes during EVT for rIA are associated with an increased risk of PPND. SIGNIFICANCE: IONM can be used confidently as a real time neurophysiological diagnostic guide for impending neurological deficits during EVT treatment of rIA.

Diagnostic utility of different types of somatosensory evoked potential changes in pediatric idiopathic scoliosis correction surgery.

PURPOSE: To evaluate the diagnostic accuracy of intraoperative somatosensory evoked potential (SSEP) monitoring and types of SSEP changes in predicting the risk of postoperative neurological outcomes during correction surgery for idiopathic scoliosis (IS) in the pediatric age group (≤ 21 years). METHODS: Database review was performed to identify literature on pediatric patients with IS who underwent correction with intraoperative neuromonitoring. The sensitivity, specificity, and diagnostic odds ratio (DOR) of transient and persistent SSEP changes and complete SSEP loss in predicting postoperative neurological deficits were calculated. RESULTS: Final analysis included 3778 patients. SSEP changes had a sensitivity of 72.9%, specificity of 96.8%, and DOR of 102.3, while SSEP loss had a sensitivity of 41.8%, specificity of 99.3%, and DOR of 133.2 for predicting new neurologic deficits. Transient and persistent SSEP changes had specificities of 96.8% and 99.1%, and DORs of 16.6 and 59, respectively. CONCLUSION: Intraoperative SSEP monitoring can predict perioperative neurological injury and improve surgical outcomes in pediatric scoliosis fusion surgery. LEVEL OF EVIDENCE: Level 2. This journal requires that authors assign a level of evidence to each article. For a full description of these Evidence-Based Medicine ratings, please refer to the Table of Contents or the online Instructions to Authors www.springer.com/00266 .

Outcomes after Perioperative Transient Ischemic Attack Following Cardiac Surgery.

Perioperative transient ischemic attacks (PTIAs) are associated with significantly increased rates of postoperative complications such as low cardiac output, atrial fibrillation, and significantly higher mortality in cardiac procedures. The current literature on PTIAs is sparse and understudied. Therefore, we aim to understand the effects of PTIA on hospital utilization, readmission, and morbidity. Using data on all the cardiac procedures at the University of Pittsburgh Medical Center from 2011 to 2019, fine and gray analysis was performed to identify whether PTIAs and covariables correlate with increased hospital utilization, stroke, all-cause readmission, Major Adverse Cardiac and Cerebrovascular Events (MACCE), MI, and all-cause mortality. Logistic regression for longer hospitalization showed that PTIA (HR: 2.199 [95% CI: 1.416-3.416] increased utilization rates. Fine and gray modeling indicated that PTIA (HR: 1.444 [95% CI: 1.096-1.902], p < 0.01) increased the rates of follow-up all-cause readmission. However, PTIA (HR: 1.643 [95% CI: 0.913-2.956] was not statistically significant for stroke readmission modeling. Multivariate modeling for MACCE events within 30 days of surgery (HR: 0.524 [95% CI: 0.171-1.605], p > 0.25) and anytime during the follow-up period (HR: 1.116 [95% CI: 0.825-1.509], p > 0.45) showed no significant correlation with PTIA. As a result of PTIA's significant burden on the healthcare system due to increased utilization, it is critical to better define and recognize PTIA for timely management to improve perioperative outcomes.

Cost-benefit analysis of intraoperative neuromonitoring for cardiac surgery.

BACKGROUND: Intraoperative neuromonitoring (IONM) can detect large vessel occlusion (LVO) in real-time during surgery. The aim of this study was to conduct a cost-benefit analysis of utilizing IONM among patients undergoing cardiac surgery. METHODS: A decision-analysis tree with terminal Markov nodes was constructed to model functional outcome, as measured via the modified Rankin Scale (mRS), among 65-year-old patients undergoing cardiac surgery. Our cost-benefit analysis compares the use of IONM (electroencephalography and somatosensory evoked potential) against no IONM in preventing neurological complications from perioperative LVO during cardiac surgery. The study was performed over a lifetime horizon from a societal perspective in the United States. Base case and one-way probabilistic sensitivity analyses were performed. RESULTS: At a baseline LVO rate of 0.31%, the mean attributable lifetime expenditure for IONM-monitored cardiac surgeries relative to unmonitored cardiac surgeries was $1047.41 (95% CI, $742.12 - $1445.10). At a critical LVO rate of approximately 3.67%, the costs of both monitored and unmonitored cardiac surgeries were the same. Above this critical rate, implementing IONM became cost-saving. On one-way sensitivity analysis, variation in LVO rate from 0% - 10% caused lifetime costs attributable to receiving IONM to range from $1150.47 - $29404.61; variations in IONM cost, percentage of intervenable LVOs, IONM sensitivity, and mechanical thrombectomy cost exerted comparably minimal influence over lifetime costs. DISCUSSION: We find considerable cost savings favoring the use of IONM under certain parameters corresponding to high-risk patients. This study will provide financial perspective to policymakers, clinicians, and patients alike on the appropriate use of IONM during cardiac surgery.

Time-Series Aware Metrics for the Evaluation of Intraoperative Electroencephalography-Based Ischemia Detection.

Continuous intraoperative monitoring with electroencephalo2 graphy (EEG) is commonly used to detect cerebral ischemia in high-risk surgical procedures such as carotid endarterectomy. Machine learning (ML) models that detect ischemia in real time can form the basis of automated intraoperative EEG monitoring. In this study, we describe and compare two time-series aware precision and recall metrics to the classical precision and recall metrics for evaluating the performance of ML models that detect ischemia. We trained six ML models to detect ischemia in intraoperative EEG and evaluated them with the area under the precision-recall curve (AUPRC) using time-series aware and classical approaches to compute precision and recall. The Support Vector Classification (SVC) model performed the best on the time-series aware metrics, while the Light Gradient Boosting Machine (LGBM) model performed the best on the classical metrics. Visual inspection of the probability outputs of the models alongside the actual ischemic periods revealed that the time-series aware AUPRC selected a model more likely to predict ischemia onset in a timely fashion than the model selected by classical AUPRC.

Role of Intraoperative Neuromonitoring to Predict Postoperative Delirium in Cardiovascular Surgery.

OBJECTIVE: Postoperative delirium (POD) can occur in up to 50% of older patients undergoing cardiovascular surgery, resulting in hospitalization and significant morbidity and mortality. This study aimed to determine whether intraoperative neurophysiologic monitoring (IONM) modalities can be used to predict delirium in patients undergoing cardiovascular surgery. DESIGN: Adult patients undergoing cardiovascular surgery with IONM between 2019 and 2021 were reviewed retrospectively. Delirium was assessed multiple times using the Intensive Care Delirium Screening Checklist (ICDSC). Patients with an ICDSC score ≥4 were considered to have POD. Significant IONM changes were evaluated based on a visual review of electroencephalography (EEG) and somatosensory evoked potentials data and documentation of significant changes during surgery. SETTING: University of Pittsburgh Medical Center hospitals. PARTICIPANTS: Patients 18 years old and older undergoing cardiovascular surgery with IONM monitoring. MEASUREMENTS AND MAIN RESULTS: Of the 578 patients undergoing cardiovascular surgery with IONM, 126 had POD (21.8%). Significant IONM changes were noted in 134 patients, of whom 49 patients had delirium (36.6%). In contrast, 444 patients had no IONM changes during surgery, of whom 77 (17.3%) patients had POD. Upon multivariate analysis, IONM changes were associated with POD (odds ratio 2.12; 95% CI 1.31-3.44; p < 0.001). Additionally, baseline EEG abnormalities were associated with POD (p = 0.002). CONCLUSION: Significant IONM changes are associated with an increased risk of POD in patients undergoing cardiovascular surgery. These findings offer a basis for future research and analysis of EEG and somatosensory evoked potential monitoring to predict, detect, and prevent POD.

Electromyographic predictors of abducens nerve palsy after endoscopic skull base surgery.

OBJECTIVE: Recovery of abducens nerve palsy (ANP) after endoscopic endonasal skull base surgery (ESBS) has been shown to be potentially predicted by postoperative ophthalmological examination. Triggered electromyography (t-EMG) and free-run electromyography (f-EMG) activity provide an intraoperative assessment of abducens nerve function, but associations with long-term ANP outcomes have not been explored. The objective of this study was to describe intraoperative abducens EMG characteristics and determine whether these electrophysiological profiles are associated with immediately postoperative and long-term ANP outcomes after ESBS. METHODS: The authors conducted a 5-year (2011-2016) retrospective case-control study of patients who underwent ESBS in whom the abducens nerve was stimulated (t-EMG). Electrophysiological metrics were compared between patients with a new postoperative ANP (cases) and those without ANP (controls). Pathologies included chordoma, pituitary adenoma, meningioma, cholesterol granuloma, and chondrosarcoma. Electrophysiological data included the presence of abnormal f-EMG activity, t-EMG stimulation voltage, stimulation threshold, evoked compound muscle action potential (CMAP) amplitude, onset latency, peak latency, and CMAP duration at various stages of the dissection. Controls were selected such that pathologies were similarly distributed between cases and controls. RESULTS: Fifty-six patients were included, 26 with new postoperative ANP and 30 controls without ANP. Abnormal f-EMG activity (28.0% vs 3.3%, p = 0.02) and lack of response to stimulation (27% vs 0%, p = 0.006) were more frequent in patients with immediately postoperative ANP than in controls. Patients with immediately postoperative ANP also had a lower median CMAP amplitude (35.0 vs 71.2 µV, p = 0.02) and longer onset latency (5.2 vs 2.8 msec, p = 0.04). Comparing patients with transient versus persistent ANP on follow-up, those with persistent ANP tended to have a lower CMAP amplitude (12.8 vs 57 µV, p = 0.07) and higher likelihood of not responding to stimulation at the end of the case (45.5% vs 7.1%, p = 0.06). Abnormal f-EMG was not associated with long-term ANP outcomes. CONCLUSIONS: The presence of f-EMG activity, lack of CMAP response to stimulation, decreased CMAP amplitude, and increased CMAP onset latency were associated with immediately postoperative ANP. Long-term ANP outcomes may be associated with t-EMG parameters, including whether the nerve is able to be stimulated once identified and CMAP amplitude. Future prospective studies may be designed to standardize abducens nerve electrophysiological monitoring protocols to further refine operative and prognostic utility.

Transcranial Motor Evoked Potentials as a Predictive Modality for Postoperative Deficit in Cervical Spine Decompression Surgery - A Systematic Review and Meta-Analysis.

STUDY DESIGN: Systematic Review and Meta-analysis. OBJECTIVE: The purpose of this study was to evaluate whether transcranial motor evoked potential (TcMEP) alarms can predict postoperative neurologic complications in patients undergoing cervical spine decompression surgery. METHODS: A meta-analysis of the literature was performed using PubMed, Web of Science, and Embase to retrieve published reports on intraoperative TcMEP monitoring for patients undergoing cervical spine decompression surgery. The sensitivity, specificity, and diagnostic odds ratio (DOR), of overall, reversible, and irreversible TcMEP changes for predicting postoperative neurological deficit were calculated. A subgroup analysis was performed to compare anterior vs posterior approaches. RESULTS: Nineteen studies consisting of 4608 patients were analyzed. The overall incidence of postoperative neurological deficits was 2.58% (119/4608). Overall TcMEP changes had a sensitivity of 56%, specificity of 94%, and DOR of 19.26 for predicting deficit. Reversible and irreversible changes had sensitivities of 16% and 49%, specificities of 95% and 98%, and DORs of 3.54 and 71.74, respectively. In anterior procedures, TcMEP changes had a DOR of 17.57, sensitivity of 49%, and specificity of 94%. In posterior procedures, TcMEP changes had a DOR of 21.01, sensitivity of 55%, and specificity of 94%. CONCLUSION: TcMEP monitoring has high specificity but low sensitivity for predicting postoperative neurological deficit in cervical spine decompression surgery. Patients with new postoperative neurological deficits were 19 times more likely to have experienced intraoperative TcMEP changes than those without new deficits, with irreversible TcMEP changes indicating a much higher risk of deficit than reversible TcMEP changes.

Utility of transcranial motor-evoked potential changes in predicting postoperative deficit in lumbar decompression and fusion surgery: a systematic review and meta-analysis.

PURPOSE: The primary aim of this study was to evaluate whether TcMEP alarms can predict the occurrence of postoperative neurological deficit in patients undergoing lumbar spine surgery. The secondary aim was to determine whether the various types of TcMEP alarms including transient and persistent changes portend varying degrees of injury risk. METHODS: This was a systematic review and meta-analysis of the literature from PubMed, Web of Science, and Embase regarding outcomes of transcranial motor-evoked potential (TcMEP) monitoring during lumbar decompression and fusion surgery. The sensitivity, specificity, and diagnostic odds ratio (DOR) of TcMEP alarms for predicting postoperative deficit were calculated and presented with forest plots and a summary receiver operating characteristic curve. RESULTS: Eight studies were included, consisting of 4923 patients. The incidence of postoperative neurological deficit was 0.73% (36/4923). The incidence of deficits in patients with significant TcMEP changes was 11.79% (27/229), while the incidence in those without changes was 0.19% (9/4694). All TcMEP alarms had a pooled sensitivity and specificity of 63 and 95% with a DOR of 34.92 (95% CI 7.95-153.42). Transient and persistent changes had sensitivities of 29% and 47%, specificities of 96% and 98%, and DORs of 8.04 and 66.06, respectively. CONCLUSION: TcMEP monitoring has high specificity but low sensitivity for predicting postoperative neurological deficit in lumbar decompression and fusion surgery. Patients who awoke with new postoperative deficits were 35 times more likely to have experienced TcMEP changes intraoperatively, with persistent changes indicating higher risk of deficit than transient changes. LEVEL OF EVIDENCE II: Diagnostic Systematic Review.

Diagnostic Accuracy of Somatosensory Evoked Potentials and Electroencephalography During Endovascular Treatment of Unruptured Cerebral Aneurysms.

OBJECTIVE: Endovascular treatment (EVT) of unruptured cerebral aneurysms (UCA) offers a safer alternative to clipping. However, it is still associated with an increased risk for Postprocedural Neurological deficit (PPND). Prompt recognition using intraoperative neurophysiologic monitoring (IONM) and intervention can reduce the incidence and impact of new postoperative neurological complications. We aim to evaluate the diagnostic accuracy of IONM in predicting PPND after EVT of UCA. METHODS: We included 414 patients who underwent EVT for UCA from 2014 to 2019. The sensitivities, specificities, and diagnostic odds ratio of somatosensory evoked potentials and electroencephalography monitoring methods were calculated. We also determined their diagnostic accuracy using receiver operating characteristic plots. RESULTS: The highest sensitivity of 67.7% (95% confidence interval {CI}, 34.9%-90.1%) was obtained when either modality had a change. Simultaneous changes in both modalities have the highest specificity of 97.8% (95% CI, 95.8%-99.0%). The area under the receiver operating characteristic curve was 0.795 (95% CI, 0.655-0.935) for changes in either modality. CONCLUSIONS: IONM with somatosensory evoked potentials alone or in combination with electroencephalography has high diagnostic accuracy in detecting periprocedural complications and resultant PPND during EVT of UCA.

Minimally invasive extraocular cranial nerve electromyography.

OBJECTIVE: A reluctance to monitor extraocular cranial nerve (EOCN) function has restricted skull base surgery worldwide. Spontaneous and triggered electromyography (EMG) monitoring can be recorded intraoperatively to identify and assess potential cranial nerve injury. Determining the conductive function of EOCNs requires the collection of clear, reliable, and repeatable compound muscle action potentials (CMAPs) secondary to stimulation. EOCN EMG needle electrodes can, although infrequently, cause ocular morbidity including hematoma, edema, and scleral laceration. The aim of this study was to ascertain if minimally invasive 7-mm superficial needle electrodes would record CMAPs as well as standard 13-mm intraorbital electrodes. METHODS: Conventionally, the authors have monitored EOCN function with intraorbital placement of paired 13-mm needle electrodes into three extraocular muscles: medial rectus, superior oblique, and lateral rectus. A prospective case-control study was performed using shorter (7-mm) needle electrodes. A single minimally invasive electrode was placed superficially near each extraocular muscle and coupled with a common reference. CMAPs were recorded from the minimally invasive electrodes and compared with CMAPs recorded from the paired intraorbital electrodes. The presence or absence of CMAPs was analyzed and compared among EMG recording techniques. RESULTS: A total of 429 CMAPs were analyzed from 71 EOCNs in 25 patients. The experimental setup yielded 167 true-positive (39%), 106 false-positive (25%), 17 false-negative (4%), and 139 true-negative (32%) responses. These values were used to calculate the sensitivity (91%), specificity (57%), positive predictive value (61%), and negative predictive value (89%). EOCN electrodes were placed in 82 total eyes in 58 patients (CMAPs were obtained in 25 patients). Twenty-six eyes showed some degree of edema, bruising, or bleeding, which was transient and self-resolving. Three eyes in different patients had complications from needle placement or extraction including conjunctival hemorrhage, periorbital ecchymosis, and corneal abrasion, ptosis, and upper eyelid edema. CONCLUSIONS: Because of artifact contamination, 106 false-positive responses (25%), and 17 false-negative responses (4%), the minimally invasive EMG technique cannot reliably record CMAP responses intraoperatively as well as the intraorbital technique. Less-invasive techniques can lead to an inaccurate EOCN assessment and potential postoperative morbidity. EOCN palsies can be debilitating and lifelong; therefore, the benefits of preserving EOCN function outweigh the potential risks of morbidity from electrode placement. EMG monitoring with intraorbital electrodes remains the most reliable method of intraoperative EOCN assessment.

Role of Intraoperative Neurophysiological Monitoring in Preventing Stroke After Cardiac Surgery.

BACKGROUND: Perioperative stroke after cardiac surgical procedures carries significant morbidity. Dual intraoperative neurophysiological monitoring with electroencephalography (EEG) and somatosensory-evoked potentials detects cerebral hypoperfusion and predicts postoperative stroke in noncardiac procedures. We further evaluated preoperative risk factors and intraoperative neuromonitoring ability to predict postoperative stroke after cardiac operations. METHODS: All patients who underwent cardiac operations with intraoperative neurophysiological monitoring from 2009 to 2020 at a single academic medical center were retrospectively analyzed. Patients with circulatory arrest were excluded. Risks factors analyzed were sex, age, tobacco use, hypertension, diabetes mellitus, dyslipidemia, atrial fibrillation, prior cerebrovascular accident, cerebrovascular disease, antiplatelet/anticoagulant use, abnormal somatosensory-evoked potentials and EEG baselines, and significant somatosensory-evoked potentials and EEG change as well as their permanence. Patients were divided into 2 groups by 30-day postoperative stroke occurrence. Univariate and multivariate logistical regressions were used for postoperative stroke significant predictors, and Kaplan-Meier curves estimated survival. RESULTS: The study included 620 patients (67.6% men), mean age 65.1 ± 14.1 years, with stroke in 5.32%. In univariate analysis, diabetes (odds ratio [OR], 2.62) and permanence of EEG change (OR, 5.35) were each associated with increased postoperative stroke odds. In multivariate analysis, diabetes (OR, 2.64) and permanent EEG change (OR, 4.22) were independently significantly associated with postoperative stroke. Overall survival was significantly better for patients with no intraoperative neurophysiological monitoring changes (P < .005). CONCLUSIONS: Permanent EEG change and diabetes were significant postoperative stroke predictors in cardiac operations. Furthermore, overall survival out to 10 years postoperatively was significantly higher in the group without intraoperative neurophysiological monitoring changes, emphasizing its important predictive role.

Intraoperative neurophysiologic monitoring during aortic arch surgery.

OBJECTIVE: To evaluate the ability of intraoperative neurophysiologic monitoring (IONM) during aortic arch reconstruction with hypothermic circulatory arrest (HCA) to predict early (<48 hours) adverse neurologic events (ANE; stroke or transient ischemic attack) and operative mortality. METHODS: This was an observational study of aortic arch surgeries requiring HCA from 2010 to 2018. Patients were monitored with electroencephalogram (EEG) and somatosensory evoked potentials (SSEP). Baseline characteristics and postoperative outcomes were compared according to presence or absence of IONM changes, which were defined as any acute variation in SSEP or EEG, compared with baseline. Multivariable logistic regression analysis was used to assess the association of IONM changes with operative mortality and early ANE. RESULTS: A total of 563 patients underwent aortic arch reconstruction with HCA and IONM. Of these, 119 (21.1%) patients had an IONM change, whereas 444 (78.9%) did not. Patients with IONM changes had increased operative mortality (22.7% vs 4.3%) and increased early ANE (10.9% vs 2.9%). In multivariable analysis, SSEP changes were correlated with early ANE (odds ratio [OR], 4.68; 95% confidence interval [CI], 1.51-14.56; P = .008), whereas EEG changes were not (P = .532). Permanent SSEP changes were correlated with early ANE (OR, 4.56; 95% CI, 1.51-13.77; P = .007), whereas temperature-related SSEP changes were not (P = .997). Finally, any IONM change (either SSEP or EEG) was correlated with operative mortality (OR, 5.82; 95% CI, 2.72-12.49; P < .001). CONCLUSIONS: Abnormal IONM events during aortic arch reconstruction with HCA portend worse neurologic outcomes and operative mortality and have a negative predictive value of 97.1%. SSEP might be more sensitive than EEG for predicting early ANE, especially when SSEP changes are permanent.

Predictive Value of Multimodality Intraoperative Neurophysiological Monitoring During Cardiac Surgery.

INTRODUCTION: This study aimed to determine the ability of multimodality intraoperative neurophysiologic monitoring, including somatosensory evoked potentials (SSEP) and EEG, to predict perioperative clinical stroke and stroke-related mortality after open-heart surgery in high-risk patients. METHODS: The records of all consecutive patients who underwent coronary artery bypass grafting, and cardiac valve repair/replacement with high risk for stroke who underwent both SSEP and EEG recording at the University of Pittsburgh Medical Center between 2009 and 2015 were reviewed. Sensitivity and specificity of these modalities to predict in-hospital clinical strokes and stroke-related mortality were calculated. RESULTS: A total of 531 patients underwent open cardiac procedures monitored using SSEP and EEG. One hundred thirty-one patients (24.67%) experienced significant changes in either modality. Fourteen patients (2.64%) suffered clinical strokes within 24 hours after surgery, and eight patients (1.50%) died during their hospitalization. The incidence of in-hospital clinical stroke and stroke-related mortality among patients who experienced a significant change in monitoring compared with those with no significant change was 11.45% versus 1.75%. The sensitivity and specificity of significant changes in either SSEP or EEG to predict in-hospital major stroke and stroke-related mortality were 0.93 and 0.77, respectively. CONCLUSIONS: Intraoperative neurophysiologic monitoring with SSEP and EEG has high sensitivity and specificity in predicting perioperative stroke and stroke-related mortality after open cardiac procedures. These results support the benefits of multimodality neuromonitoring during cardiac surgery.

Correction to: Perioperative Stroke and Thirty-Day Hospital Readmission After Cardiac Surgeries: State Inpatient Database Study.

[This corrects the article DOI: 10.14740/jocmr4647.].

Risk of perioperative stroke and cerebral autoregulation monitoring: a systematic review.

BACKGROUND: Perioperative stroke, delirium, and cognitive impairment could be related to management and to variations in blood pressure control, cerebral hypoperfusion and raised blood volume. Cerebral autoregulation (CAR) is a mechanism to maintain cerebral perfusion through the control of the vascular tone and hemodynamic reactions in the circulation. OBJECTIVE: The present systematic review addresses the relationship between impaired CAR and perioperative stroke by evaluating the rate of neurological complications after surgery in studies in which perioperative CAR was tested or monitored. METHODS: We included randomized clinical trials and prospective observational studies. All studies had adjusted the relative risk, hazard ratio or 95% confidence interval (95%CI) values. These estimation effects were tested using random-effects models. Heterogeneity among the selected studies was assessed using the Higgins and Thompson I2 statistics. RESULTS: The Web of Science, PubMed and EMBASE electronic databases were searched to retrieve articles. A total of 4,476 studies published between 1983 and 2019 were analyzed, but only 5 qualified for the data extraction and were included in the final analysis. The combined study cohort comprised 941 patients who underwent CAR monitoring during surgical procedures. All studies provided information about perioperative stroke, which equated to 16% (158 of 941) of the overall patient population. CONCLUSION: The present meta-analysis showed evidence of the impact of CAR impairment in the risk of perioperative stroke. On the pooled analysis, blood fluctuations or other brain insults large enough to compromise CAR were associated with the outcome of stroke (odds ratio [OR]: 2.26; 95%CI: 1.54-2.98; p < 0.0001).

ANTECEDENTES: Acidente vascular cerebral (AVC) perioperatório, delírio e comprometimento cognitivo podem estar relacionados ao manejo e à variações no controle da pressão arterial, à hipoperfusão cerebral, e ao aumento do volume sanguíneo. A autorregulação cerebral (ARC) é um mecanismo para manter a perfusão cerebral por meio do controle do tônus vascular e das reações hemodinâmicas na circulação. OBJETIVO: Este trabalho aborda sistematicamente a determinação da relação entre a falha da ARC e o AVC perioperatório, com uma avaliação da taxa de complicações neurológicas pós-cirúrgicas em estudos em que a ARC perioperatória foi monitorada. MéTODOS: Esta revisão sistemática incluiu ensaios clínicos randomizados e estudos observacionais prospectivos. Todos os estudos ajustaram o risco relativo, a razão de risco ou os valores de intervalo de confiança de 95% (IC95%). Esses efeitos de estimativas foram testados usando modelos de efeitos aleatórios. A heterogeneidade entre os diferentes estudos foi avaliada por meio das estatísticas de Higgins e Thompson I2. RESULTADOS: As bases de dados eletrônicas Web of Science, PubMed e Embase foram pesquisadas para selecionar os artigos. Um total de 4.476 estudos publicados entre 1983 e 2019 foram analisados, mas apenas 5 se qualificavam para a extração de dados e foram incluídos na análise final. A coorte combinada do estudo foi composta por 941 pacientes. Todos os estudos forneceram informações sobre AVC perioperatório, o que equivaleu a 16% (158 de 941) da população geral de pacientes. CONCLUSãO: Esta meta-análise mostrou evidências do impacto do prejuízo da ARC no risco de AVC perioperatório. Na análise conjunta, as flutuações sanguíneas ou outros insultos cerebrais grandes o suficiente para comprometer a ARC foram associados ao resultado do AVC (razão de probabilidades [RP]: 2,26; IC95%: 1,54­2,98; p < 0,0001).

Risk of perioperative stroke and cerebral autoregulation monitoring: a systematic review / Risco de acidente vascular cerebral perioperatório e monitoramento de autorregulação cerebral: uma revisão sistemática

Abstract Background Perioperative stroke, delirium, and cognitive impairment could be related to management and to variations in blood pressure control, cerebral hypoperfusion and raised blood volume. Cerebral autoregulation (CAR) is a mechanism to maintain cerebral perfusion through the control of the vascular tone and hemodynamic reactions in the circulation. Objective The present systematic review addresses the relationship between impaired CAR and perioperative stroke by evaluating the rate of neurological complications after surgery in studies in which perioperative CAR was tested or monitored. Methods We included randomized clinical trials and prospective observational studies. All studies had adjusted the relative risk, hazard ratio or 95% confidence interval (95%CI) values. These estimation effects were tested using random-effects models. Heterogeneity among the selected studies was assessed using the Higgins and Thompson I2 statistics. Results The Web of Science, PubMed and EMBASE electronic databases were searched to retrieve articles. A total of 4,476 studies published between 1983 and 2019 were analyzed, but only 5 qualified for the data extraction and were included in the final analysis. The combined study cohort comprised 941 patients who underwent CAR monitoring during surgical procedures. All studies provided information about perioperative stroke, which equated to 16% (158 of 941) of the overall patient population. Conclusion The present meta-analysis showed evidence of the impact of CAR impairment in the risk of perioperative stroke. On the pooled analysis, blood fluctuations or other brain insults large enough to compromise CAR were associated with the outcome of stroke (odds ratio [OR]: 2.26; 95%CI: 1.54-2.98;p < 0.0001).

Resumo Antecedentes Acidente vascular cerebral (AVC) perioperatório, delírio e comprometimento cognitivo podem estar relacionados ao manejo e à variações no controle da pressão arterial, à hipoperfusão cerebral, e ao aumento do volume sanguíneo. A autorregulação cerebral (ARC) é um mecanismo para manter a perfusão cerebral por meio do controle do tônus vascular e das reações hemodinâmicas na circulação. Objetivo Este trabalho aborda sistematicamente a determinação da relação entre a falha da ARC e o AVC perioperatório, com uma avaliação da taxa de complicações neurológicas pós-cirúrgicas em estudos em que a ARC perioperatória foi monitorada. Métodos Esta revisão sistemática incluiu ensaios clínicos randomizados e estudos observacionais prospectivos. Todos os estudos ajustaram o risco relativo, a razão de risco ou os valores de intervalo de confiança de 95% (IC95%). Esses efeitos de estimativas foram testados usando modelos de efeitos aleatórios. A heterogeneidade entre os diferentes estudos foi avaliada por meio das estatísticas de Higgins e Thompson I2. Resultados As bases de dados eletrônicas Web of Science, PubMed e Embase foram pesquisadas para selecionar os artigos. Um total de 4.476 estudos publicados entre 1983 e 2019 foram analisados, mas apenas 5 se qualificavam para a extração de dados e foram incluídos na análise final. A coorte combinada do estudo foi composta por 941 pacientes. Todos os estudos forneceram informações sobre AVC perioperatório, o que equivaleu a 16% (158 de 941) da população geral de pacientes. Conclusão Esta meta-análise mostrou evidências do impacto do prejuízo da ARC no risco de AVC perioperatório. Na análise conjunta, as flutuações sanguíneas ou outros insultos cerebrais grandes o suficiente para comprometer a ARC foram associados ao resultado do AVC (razão de probabilidades [RP]: 2,26; IC95%: 1,54-2,98;p < 0,0001).

Predicting transient ischemic attack after carotid endarterectomy: The role of intraoperative neurophysiological monitoring.

OBJECTIVE: Transient ischemic attacks (TIA) after carotid endarterectomy (CEA) are not well-studied. We aimed to investigate the characteristics and the predictive role of intraoperative neurophysiological monitoring (IONM) in TIA post-CEA. METHODS: Patients who underwent CEA utilizing IONM from 2009-2020 were included. Analyses included TIA incidence, sensitivity, specificity, and predictive values of IONM, risk factor regression analyses, and mortality Kaplan Meier plots. RESULTS: Out of 2232 patients, 46 experienced TIA, 14 of which were within 24 hours of CEA (p < 0.01). Nine of these patients displayed significant IONM changes during CEA. The odds of TIA increased with somatosensory evoked potential (SSEP) changes (Odds Ratio (OR): 2.48 95% Confidence Interval (CI): 1.14-5.4), electroencephalogram (EEG) changes (OR: 2.65 95% CI: 1.22-5.77), and combined SSEP/EEG changes (OR: 2.98 95% CI: 1.17-7.55). Patients with TIA were less likely to be alive after an average of 4.3 years (OR: 0.5 95% CI: 0.26-0.96). CONCLUSIONS: The odds a patient will have TIA post-CEA are greater in patients with IONM changes. This risk is inversely related to the time post-CEA. SIGNIFICANCE: Changes in IONM during CEA predict postoperative TIA. Post-CEA TIA may increase long-term mortality, thus further research is needed to better elucidate clinical implications of postoperative TIA.