Hemodynamic Stroke: Emerging Concepts, Risk Estimation, and Treatment.

Ischemic stroke can arise from the sudden occlusion of a brain-feeding artery by a clot (embolic), or local thrombosis. Hemodynamic stroke occurs when blood flow does not sufficiently meet the metabolic demand of a brain region at a certain time. This discrepancy between demand and supply can occur with cerebropetal arterial occlusion or high-grade stenosis but also arises with systemic conditions reducing blood pressure. Treatment of hemodynamic stroke is targeted toward increasing blood flow to the affected area by either systemically or locally enhancing perfusion. Thus, blood pressure is often maintained above normal values, and extra-intracranial flow augmentation bypass surgery is increasingly considered. Still, current evidence supporting the superiority of pressure or flow increase over conservative measures is limited. However, methods assessing hemodynamic impairment and identifying patients at risk of hemodynamic stroke are rapidly evolving. Sophisticated models incorporating clinical and imaging factors have been suggested to aid patient selection. In this narrative review, we provide current state-of-the-art knowledge about hemodynamic stroke, tools for assessment, and treatment options.

Increased Risk of Recurrent Stroke in Symptomatic Large Vessel Disease With Impaired BOLD Cerebrovascular Reactivity.

BACKGROUND: Impaired cerebrovascular reactivity (CVR) has been correlated with recurrent ischemic stroke. However, for clinical purposes, most CVR techniques are rather complex, time-consuming, and lack validation for quantitative measurements. The recent adaptation of a standardized hypercapnic stimulus in combination with a blood-oxygenation-level-dependent (BOLD) magnetic resonance imaging signal as a surrogate for cerebral blood flow offers a potential universally comparable CVR assessment. We investigated the association between impaired BOLD-CVR and risk for recurrent ischemic events. METHODS: We conducted a retrospective analysis of patients with symptomatic cerebrovascular large vessel disease who had undergone a prospective hypercapnic-challenged BOLD-CVR protocol at a single tertiary stroke referral center between June 2014 and April 2020. These patients were followed up for recurrent acute ischemic events for up to 3 years. BOLD-CVR (%BOLD signal change per mm Hg CO2) was calculated on a voxel-by-voxel basis. Impaired BOLD-CVR of the affected (ipsilateral to the vascular pathology) hemisphere was defined as an average BOLD-CVR, falling 2 SD below the mean BOLD-CVR of the right hemisphere in a healthy age-matched reference cohort (n=20). Using a multivariate Cox proportional hazards model, the association between impaired BOLD-CVR and ischemic stroke recurrence was assessed and Kaplan-Meier survival curves to visualize the acute ischemic stroke event rate. RESULTS: Of 130 eligible patients, 28 experienced recurrent strokes (median, 85 days, interquartile range, 5-166 days). Risk factors associated with an increased recurrent stroke rate included impaired BOLD-CVR, a history of atrial fibrillation, and heart insufficiency. After adjusting for sex, age group, and atrial fibrillation, impaired BOLD-CVR exhibited a hazard ratio of 10.73 (95% CI, 4.14-27.81; P<0.001) for recurrent ischemic stroke. CONCLUSIONS: Among patients with symptomatic cerebrovascular large vessel disease, those exhibiting impaired BOLD-CVR in the affected hemisphere had a 10.7-fold higher risk of recurrent ischemic stroke events compared with individuals with nonimpaired BOLD-CVR.

Safety of microneurosurgical interventions for superficial and deep-seated brain metastases: single-center cohort study of 637 consecutive cases.

PURPOSE: Microneurosurgical techniques have greatly improved over the past years due to the introduction of new technology and surgical concepts. To reevaluate the role of micro-neurosurgery in brain metastases (BM) resection in the era of new systemic and local treatment options, its safety profile needs to be reassessed. The aim of this study was to analyze the rate of adverse events (AEs) according to a systematic, comprehensive and reliably reproducible grading system after microneurosurgical BM resection in a large and modern microneurosurgical series with special emphasis on anatomical location. METHODS: Prospectively collected cases of BM resection between 2013 and 2022 were retrospectively analyzed. Number of AEs, defined as any deviations from the expected postoperative course according to Clavien-Dindo-Grade (CDG) were evaluated. Patient, surgical, and lesion characteristics, including exact anatomic tumor locations, were analyzed using uni- and multivariate logistic regression and survival analysis to identify predictive factors for AEs. RESULTS: We identified 664 eligible patients with lung cancer being the most common primary tumor (44%), followed by melanoma (25%) and breast cancer (11%). 29 patients (4%) underwent biopsy only whereas BM were resected in 637 (96%) of cases. The overall rate of AEs was 8% at discharge. However, severe AEs (≥ CDG 3a; requiring surgical intervention under local/general anesthesia or ICU treatment) occurred in only 1.9% (n = 12) of cases with a perioperative mortality of 0.6% (n = 4). Infratentorial tumor location (OR 5.46, 95% 2.31-13.8, p = .001), reoperation (OR 2.31, 95% 1.07-4.81, p = .033) and central region tumor location (OR 3.03, 95% 1.03-8.60) showed to be significant predictors in a multivariate analysis for major AEs (CDG ≥ 2 or new neurological deficits). Neither deep supratentorial nor central region tumors were associated with more major AEs compared to convexity lesions. CONCLUSIONS: Modern microneurosurgical resection can be considered an excellent option in the management of BM in terms of safety, as the overall rate of major AEs are very rare even in eloquent and deep-seated lesions.

Blood oxygenation-level dependent cerebrovascular reactivity imaging as strategy to monitor CSF-hemoglobin toxicity.

OBJECTIVES: Cell-free hemoglobin in the cerebrospinal fluid (CSF-Hb) may be one of the main drivers of secondary brain injury after aneurysmal subarachnoid hemorrhage (aSAH). Haptoglobin scavenging of CSF-Hb has been shown to mitigate cerebrovascular disruption. Using digital subtraction angiography (DSA) and blood oxygenation-level dependent cerebrovascular reactivity imaging (BOLD-CVR) the aim was to assess the acute toxic effect of CSF-Hb on cerebral blood flow and autoregulation, as well as to test the protective effects of haptoglobin. METHODS: DSA imaging was performed in eight anesthetized and ventilated sheep (mean weight: 80.4 kg) at baseline, 15, 30, 45 and 60 minutes after infusion of hemoglobin (Hb) or co-infusion with haptoglobin (Hb:Haptoglobin) into the left lateral ventricle. Additionally, 10 ventilated sheep (mean weight: 79.8 kg) underwent BOLD-CVR imaging to assess the cerebrovascular reserve capacity. RESULTS: DSA imaging did not show a difference in mean transit time or cerebral blood flow. Whole-brain BOLD-CVR compared to baseline decreased more in the Hb group after 15 minutes (Hb vs Hb:Haptoglobin: -0.03 ± 0.01 vs -0.01 ± 0.02) and remained diminished compared to Hb:Haptoglobin group after 30 minutes (Hb vs Hb:Haptoglobin: -0.03 ± 0.01 vs 0.0 ± 0.01), 45 minutes (Hb vs Hb:Haptoglobin: -0.03 ± 0.01 vs 0.01 ± 0.02) and 60 minutes (Hb vs Hb:Haptoglobin: -0.03 ± 0.02 vs 0.01 ± 0.01). CONCLUSION: It is demonstrated that CSF-Hb toxicity leads to rapid cerebrovascular reactivity impairment, which is blunted by haptoglobin co-infusion. BOLD-CVR may therefore be further evaluated as a monitoring strategy for CSF-Hb toxicity after aSAH.

Feasibility of glioblastoma tissue response mapping with physiologic BOLD imaging using precise oxygen and carbon dioxide challenge.

OBJECTIVES: Innovative physiologic MRI development focuses on depiction of heterogenous vascular and metabolic features in glioblastoma. For this feasibility study, we employed blood oxygenation level-dependent (BOLD) MRI with standardized and precise carbon dioxide (CO2) and oxygen (O2) modulation to investigate specific tumor tissue response patterns in patients with newly diagnosed glioblastoma. MATERIALS AND METHODS: Seven newly diagnosed untreated patients with suspected glioblastoma were prospectively included to undergo a BOLD study with combined CO2 and O2 standardized protocol. %BOLD signal change/mmHg during hypercapnic, hypoxic, and hyperoxic stimulus was calculated in the whole brain, tumor lesion and segmented volumes of interest (VOI) [contrast-enhancing (CE) - tumor, necrosis and edema] to analyze their tissue response patterns. RESULTS: Quantification of BOLD signal change after gas challenges can be used to identify specific responses to standardized stimuli in glioblastoma patients. Integration of this approach with automatic VOI segmentation grants improved characterization of tumor subzones and edema. Magnitude of BOLD signal change during the 3 stimuli can be visualized at voxel precision through color-coded maps overlayed onto whole brain and identified VOIs. CONCLUSIONS: Our preliminary investigation shows good feasibility of BOLD with standardized and precise CO2 and O2 modulation as an emerging physiologic imaging technique to detail specific glioblastoma characteristics. The unique tissue response patterns generated can be further investigated to better detail glioblastoma lesions and gauge treatment response.

Foundations of Feature Selection in Clinical Prediction Modeling.

Selecting a set of features to include in a clinical prediction model is not always a simple task. The goals of creating parsimonious models with low complexity while, at the same time, upholding predictive performance by explaining a large proportion of the variance within the dependent variable must be balanced. With this aim, one must consider the clinical setting and what data are readily available to clinicians at specific timepoints, as well as more obvious aspects such as the availability of computational power and size of the training dataset. This chapter elucidates the importance and pitfalls in feature selection, focusing on applications in clinical prediction modeling. We demonstrate simple methods such as correlation-, significance-, and variable importance-based filtering, as well as intrinsic feature selection methods such as Lasso and tree- or rule-based methods. Finally, we focus on two algorithmic wrapper methods for feature selection that are commonly used in machine learning: Recursive Feature Elimination (RFE), which can be applied regardless of data and model type, as well as Purposeful Variable Selection as described by Hosmer and Lemeshow, specifically for generalized linear models.

Machine Learning Algorithms in Neuroimaging: An Overview.

Machine learning (ML) and artificial intelligence (AI) applications in the field of neuroimaging have been on the rise in recent years, and their clinical adoption is increasing worldwide. Deep learning (DL) is a field of ML that can be defined as a set of algorithms enabling a computer to be fed with raw data and progressively discover-through multiple layers of representation-more complex and abstract patterns in large data sets. The combination of ML and radiomics, namely the extraction of features from medical images, has proven valuable, too: Radiomic information can be used for enhanced image characterization and prognosis or outcome prediction. This chapter summarizes the basic concepts underlying ML application for neuroimaging and discusses technical aspects of the most promising algorithms, with a specific focus on Convolutional Neural Networks (CNNs) and Generative Adversarial Networks (GANs), in order to provide the readership with the fundamental theoretical tools to better understand ML in neuroimaging. Applications are highlighted from a practical standpoint in the last section of the chapter, including: image reconstruction and restoration, image synthesis and super-resolution, registration, segmentation, classification, and outcome prediction.

Increased Ipsilateral Posterior Cerebral Artery P2-Segment Flow Velocity Predicts Hemodynamic Impairment.

BACKGROUND AND PURPOSE: Increased Transcranial Doppler flow velocity in the ipsilateral P2-segment of the posterior cerebral artery (PCA-P2: cm/second) is associated with recurrent cerebrovascular events in patients with unilateral internal carotid artery occlusion. However, its predictive value and correlation with hemodynamic impairment in an overall stroke patient cohort remains to be determined. METHODS: Transcranial doppler PCA-P2 flow velocity was measured in 88 patients with symptomatic unilateral steno-occlusive disease who also underwent blood oxygenation-level dependent cerebrovascular reactivity imaging (blood oxygenation-level dependent [BOLD]-cerebrovascular reactivity [CVR]). A multivariate linear regression was used to evaluate the independent correlation between the ipsilateral PCA-P2 flow velocity measurements and hemispheric BOLD-CVR. Follow-up BOLD-CVR imaging data, available in 25 patients, were used to evaluate the temporal evolution of the BOLD-CVR and PCA-P2 flow velocity association using a mixed-effect model. Furthermore, a transcranial doppler cutoff for hemodynamic failure stage 2 was determined. RESULTS: The ipsilateral systolic PCA-P2 flow velocity strongly correlated with hemispheric BOLD-CVR (R=0.79; R2=0.61), which remained unchanged when evaluating the follow-up data. Using a PCA-P2 systolic flow velocity cutoff value of 85 cm/second, patients with BOLD-CVR based hemodynamic failure stage 2 were diagnosed with an area under the curve of 95. CONCLUSIONS: In patients with symptomatic unilateral steno-occlusive disease, increased ipsilateral transcranial doppler PCA-P2 systolic flow velocity independently correlates with BOLD-CVR based hemodynamic failure. A cutoff value of 85 cm/second appears to indicate hemodynamic failure stage 2, but this finding needs to be validated in an independent patient cohort.

Crossed Cerebellar Diaschisis in Patients With Symptomatic Unilateral Anterior Circulation Stroke Is Associated With Hemodynamic Impairment in the Ipsilateral MCA Territory.

BACKGROUND: In patients with steno-occlusive disease, recent findings suggest that hemodynamic alterations may also be associated with crossed cerebellar diaschisis (CCD) rather than a functional disruption alone. PURPOSE: To use a quantitative multiparametric hemodynamic MRI to gain a better understanding of hemodynamic changes related to CCD in patients with unilateral anterior circulation stroke. STUDY TYPE: Prospective cohort study. POPULATION: Twenty-four patients (25 datasets) with symptomatic unilateral anterior circulation stroke. FIELD STRENGTH/SEQUENCE: 3T/two sequences: single-shot (echo-planar imaging) EPI sequence and T2* gradient echo perfusion-weighted imaging study. ASSESSMENT: The presence of CCD was inferred from the cerebellar asymmetry index (CAI) of the blood oxygenation-level dependent cerebrovascular reactivity (BOLD-CVR) exam, which was calculated from the mean BOLD-CVR and standard deviation of the CAI of the healthy control group. For all perfusion-weighted (PW)-MRI parameters, the cerebellar and middle cerebral artery (MCA) territory asymmetry indices were calculated. STATISTICAL TESTS: Independent Student's t-test to compare the variables from the CCD positive(+) and CCD negative(-) groups and analysis of covariance (ANCOVA) to statistically control the effect of covariates (infarct volume and time since ischemia onset). RESULTS: CCD was present in 33% of patients. In the MCA territory of the affected hemisphere, BOLD-CVR was significantly more impaired in the CCD(+) group as compared to the CCD(-) group (mean BOLD-CVR ± SD [%BOLD signal/ΔmmHgCO2 ]: -0.03 ± 0.12 vs. 0.11 ± 0.13, P < 0.05). Moreover, the mean transit time (MTT) (asymmetry index (%) CCD(+) vs. CCD(-): 28 ± 23 vs. 4 ± 11, P < 0.05) and time to peak (TTP) (10 ± 10 vs. 2 ± 5, P < 0.05) in the MCA territory of the affected hemisphere were significantly prolonged, while cerebral blood volume was, on average, increased in the CCD(+) group (25 ± 15 vs. 4 ± 19, P < 0.05). DATA CONCLUSION: Our findings show that, in patients with symptomatic unilateral anterior circulation stroke, CCD is associated with hemodynamic impairment in the ipsilateral MCA territory, which further supports the concept of a vascular component of CCD. LEVEL OF EVIDENCE: 3 TECHNICAL EFFICACY STAGE: 3.

Impact of additional resection on new ischemic lesions and their clinical relevance after intraoperative 3 Tesla MRI in neuro-oncological surgery.

Intraoperative MRI (ioMRI) has become a frequently used tool to improve maximum safe resection in brain tumor surgery. The usability of intraoperatively acquired diffusion-weighted imaging sequences to predict the extent and clinical relevance of new infarcts has not yet been studied. Furthermore, the question of whether more aggressive surgery after ioMRI leads to more or larger infarcts is of crucial interest for the surgeons' operative strategy. Retrospective single-center analysis of a prospective registry of procedures from 2013 to 2019 with ioMRI was used. Infarct volumes in ioMRI/poMRI, lesion localization, mRS, and NIHSS were analyzed for each case. A total of 177 individual operations (60% male, mean age 45.5 years old) met the inclusion criteria. In 61% of the procedures, additional resection was performed after ioMRI, which resulted in a significantly higher number of new ischemic lesions postoperatively (p < .001). The development of new or enlarged ischemic areas upon additional resection could also be shown volumetrically (mean volume in ioMRI 0.39 cm3 vs. poMRI 2.97 cm3; p < .001). Despite the surgically induced new infarcts, mRS and NIHSS did not worsen significantly in cases with additional resection. Additionally, new perilesional ischemia in eloquently located tumors was not associated with an impaired neurological outcome. Additional resection after ioMRI leads to new or enlarged ischemic areas. However, these new infarcts do not necessarily result in an impaired neurological outcome, even when in eloquent brain areas.

Intraoperative BOLD-fMRI Cerebrovascular Reactivity Assessment.

Blood oxygenation-level dependent cerebrovascular reactivity (BOLD-CVR) has gained attention in recent years as an effective way to investigate CVR, a measure of the hemodynamic state of the brain, with high spatial and temporal resolution. An association between impaired CVR and diverse pathologies has been observed, especially in ischemic cerebrovascular diseases and brain gliomas. The ability to obtain this information intraoperatively is novel and has not been widely tested. We report our first experience with this intraoperative technique in vascular and oncologic neurosurgical patients, discuss the results of its feasibility, and the possible developments of the intraoperative employment of BOLD-CVR.

Agreement of novel hemodynamic imaging parameters for the acute and chronic stages of ischemic stroke: a matched-pair cohort study.

OBJECTIVE: In symptomatic patients with cerebrovascular steno-occlusive disease, impaired blood oxygenation level-dependent cerebrovascular reactivity (BOLD-CVR) and increased flow velocity of the P2 segment of the posterior cerebral artery (PCA-P2) on transcranial Doppler (TCD) ultrasonography have been introduced as emerging clinical imaging parameters to identify patients at high risk for recurrent ischemic events. Since hemodynamic physiology differs between the acute and chronic stages of ischemic stroke, the authors sought to investigate whether those parameters have merit for both the acute and chronic stages of ischemic stroke. METHODS: From a prospective database, patients who underwent BOLD-CVR and TCD examinations in the acute stroke stage (< 10 days) were matched to patients in the chronic stroke stage (> 3 months). A linear regression analysis for both groups was performed between ipsilateral PCA-P2 systolic flow velocity and BOLD-CVR of the ipsilateral (affected) hemisphere, the ipsilateral middle cerebral artery (MCA) territory, and the ipsilateral steal volume (i.e., paradoxical BOLD-CVR response). The resulting slopes and intercepts were statistically compared to evaluate differences between groups. RESULTS: Forty matched patient pairs were included. Regression analysis showed no significant difference for either the intercept (p = 0.84) or the slope (p = 0.85) between PCA-P2 flow velocity and BOLD-CVR as measured for the ipsilateral (affected) hemisphere. Similarly, no significant difference was seen between PCA-P2 flow velocity and BOLD-CVR of the ipsilateral MCA territory (intercept, p = 0.72; slope, p = 0.36) or between PCA-P2 flow velocity and steal volume (intercept, p = 0.59; slope, p = 0.34). CONCLUSIONS: The study results indicated that the relationship between ipsilateral PCA-P2 systolic flow velocity and BOLD-CVR remains the same during the acute and chronic stages of ischemic stroke. This provides further support that these novel hemodynamic imaging parameters may have merit to assess the risk for recurrent ischemic events for a wide ischemic stroke population. PCA-P2 systolic flow velocity, in particular, may be a highly practical screening tool, independent of ischemic stroke stage.

How to avoid pneumocephalus in deep brain stimulation surgery? Analysis of potential risk factors in a series of 100 consecutive patients.

BACKGROUND: Accuracy of lead placement is the key to success in deep brain stimulation (DBS). Precise anatomic stereotactic planning usually is based on stable perioperative anatomy. Pneumocephalus due to intraoperative CSF loss is a common procedure-related phenomenon which could lead to brain shift and targeting inaccuracy. The aim of this study was to evaluate potential risk factors of pneumocephalus in DBS surgery. METHODS: We performed a retrospective single-center analysis in patients undergoing bilateral DBS. We quantified the amount of pneumocephalus by postoperative CT scans and corrected the data for accompanying brain atrophy by an MRI-based score. Automated computerized segmentation algorithms from a dedicated software were used. As potential risk factors, we evaluated the impact of trephination size, the number of electrode tracks, length of surgery, intraoperative blood pressure, and brain atrophy. RESULTS: We included 100 consecutive patients that underwent awake DBS with intraoperative neurophysiological testing. Systolic and mean arterial blood pressure showed a substantial impact with an inverse correlation, indicating that lower blood pressure is associated with higher volume of pneumocephalus. Furthermore, the length of surgery was clearly correlated to pneumocephalus. CONCLUSION: Our analysis identifies intraoperative systolic and mean arterial blood pressure as important risk factors for pneumocephalus in awake stereotactic surgery.

Crossed Cerebellar Diaschisis in Patients with Diffuse Glioma Is Associated with Impaired Supratentorial Cerebrovascular Reactivity and Worse Clinical Outcome.

Crossed cerebellar diaschisis (CCD) can be associated with impaired cerebrovascular reactivity (CVR) and poor clinical outcome, but whether this holds true for patients with diffuse glioma is unknown. With blood oxygenation level-dependent (BOLD)-CVR imaging, we determined the presence of CCD in patients with diffuse glioma and investigated its relationship with cerebrovascular reactivity and clinical outcome. For eighteen enrolled subjects (nineteen datasets) with diffuse glioma, CCD was deferred from BOLD-CVR using a predetermined cerebellar asymmetry index (CAI) cutoff value of 6.0%. A FET-PET study was done as a verification of the CCD diagnosis. BOLD-CVR values as well as clinical performance scores (i.e., Karnofsky performance score (KPS), disability rating scale (DRS), and modified Rankin scale (mRS)) by BOLD-CVR scan at 3-month clinical follow-up were assessed and compared for the CCD-positive and CCD-negative group. CCD was present in 26.3% of subjects and strongly associated with impaired BOLD-CVR of the affected (i.e., the hemisphere harboring the glioma) and unaffected supratentorial hemisphere (CCD(+) vs. CCD(-): 0.08 ± 0.11 vs. 0.18 ± 0.04; p = 0.007 and 0.08 ± 0.12 vs. 0.19 ± 0.04; p = 0.007, respectively). This finding was independent of tumor volume (p = 0.48). Furthermore, poorer initial (by scan) clinical performance scores at follow-up were found for the CCD(+) group. The presence of crossed cerebellar diaschisis in patients with diffuse glioma is associated with impaired supratentorial cerebrovascular reactivity and worse clinical outcome.

Feasibility of machine learning based predictive modelling of postoperative hyponatremia after pituitary surgery.

PURPOSE: Hyponatremia after pituitary surgery is a frequent finding with potential severe complications and the most common cause for readmission. Several studies have found parameters associated with postoperative hyponatremia, but no reliable specific predictor was described yet. This pilot study evaluates the feasibility of machine learning (ML) algorithms to predict postoperative hyponatremia after resection of pituitary lesions. METHODS: Retrospective screening of a prospective registry of patients who underwent transsphenoidal surgery for pituitary lesions. Hyponatremia within 30 days after surgery was the primary outcome. Several pre- and intraoperative clinical, procedural and laboratory features were selected to train different ML algorithms. Trained models were compared using common performance metrics. Final model was internally validated on the testing dataset. RESULTS: From 207 patients included in the study, 44 (22%) showed a hyponatremia within 30 days postoperatively. Hyponatremic measurements peaked directly postoperatively (day 0-1) and around day 7. Bootstrapped performance metrics of different trained ML-models showed largest area under the receiver operating characteristic curve (AUROC) for the boosted generalized linear model (67.1%), followed by the Naïve Bayes classifier (64.6%). The discriminative capability of the final model was assessed by predicting on unseen dataset. Large AUROC (84.3%; 67.0-96.4), sensitivity (81.8%) and specificity (77.5%) with an overall accuracy of 78.4% (66.7-88.2) was reached. CONCLUSION: Our trained ML-model was able to learn the complex risk factor interactions and showed a high discriminative capability on unseen patient data. In conclusion, ML-methods can predict postoperative hyponatremia and thus potentially reduce morbidity and improve patient safety.

Machine learning in neurosurgery: a global survey.

BACKGROUND: Recent technological advances have led to the development and implementation of machine learning (ML) in various disciplines, including neurosurgery. Our goal was to conduct a comprehensive survey of neurosurgeons to assess the acceptance of and attitudes toward ML in neurosurgical practice and to identify factors associated with its use. METHODS: The online survey consisted of nine or ten mandatory questions and was distributed in February and March 2019 through the European Association of Neurosurgical Societies (EANS) and the Congress of Neurosurgeons (CNS). RESULTS: Out of 7280 neurosurgeons who received the survey, we received 362 responses, with a response rate of 5%, mainly in Europe and North America. In total, 103 neurosurgeons (28.5%) reported using ML in their clinical practice, and 31.1% in research. Adoption rates of ML were relatively evenly distributed, with 25.6% for North America, 30.9% for Europe, 33.3% for Latin America and the Middle East, 44.4% for Asia and Pacific and 100% for Africa with only two responses. No predictors of clinical ML use were identified, although academic settings and subspecialties neuro-oncology, functional, trauma and epilepsy predicted use of ML in research. The most common applications were for predicting outcomes and complications, as well as interpretation of imaging. CONCLUSIONS: This report provides a global overview of the neurosurgical applications of ML. A relevant proportion of the surveyed neurosurgeons reported clinical experience with ML algorithms. Future studies should aim to clarify the role and potential benefits of ML in neurosurgery and to reconcile these potential advantages with bioethical considerations.

Feasibility and safety of intraoperative BOLD functional MRI cerebrovascular reactivity to evaluate extracranial-to-intracranial bypass efficacy.

Blood oxygenation level-dependent functional MRI cerebrovascular reactivity (BOLD-CVR) is a contemporary technique to assess brain tissue hemodynamic changes after extracranial- intracranial (EC-IC) bypass flow augmentation surgery. The authors conducted a preliminary study to investigate the feasibility and safety of intraoperative 3-T MRI BOLD-CVR after EC-IC bypass flow augmentation surgery. Five consecutive patients selected for EC-IC bypass revascularization underwent an intraoperative BOLD-CVR examination to assess early hemodynamic changes after revascularization and to confirm the safety of this technique. All patients had a normal postoperative course, and none of the patients exhibited complications or radiological alterations related to prolonged anesthesia time. In addition to intraoperative flow measurements of the bypass graft, BOLD-CVR maps added information on the hemodynamic status and changes at the brain tissue level. Intraoperative BOLD-CVR is feasible and safe in patients undergoing EC-IC bypass revascularization. This technique can offer immediate hemodynamic feedback on brain tissue revascularization after bypass flow augmentation surgery.

Staging Hemodynamic Failure With Blood Oxygen-Level-Dependent Functional Magnetic Resonance Imaging Cerebrovascular Reactivity: A Comparison Versus Gold Standard (15O-)H2O-Positron Emission Tomography.

BACKGROUND AND PURPOSE: Increased stroke risk correlates with hemodynamic failure, which can be assessed with (15O-)H2O positron emission tomography (PET) cerebral blood flow (CBF) measurements. This gold standard technique, however, is not established for routine clinical imaging. Standardized blood oxygen-level-dependent (BOLD) functional magnetic resonance imaging+CO2 is a noninvasive and potentially widely applicable tool to assess whole-brain quantitative cerebrovascular reactivity (CVR). We examined the agreement between the 2 imaging modalities and hypothesized that quantitative CVR can be a surrogate imaging marker to assess hemodynamic failure. METHODS: Nineteen data sets of subjects with chronic cerebrovascular steno-occlusive disease (age, 60±11 years; 4 women) and unilaterally impaired perfusion reserve on Diamox-challenged (15O-)H2O PET were studied and compared with a standardized BOLD functional magnetic resonance imaging+CO2 examination within 6 weeks (8±19 days). Agreement between quantitative CBF- and CVR-based perfusion reserve was assessed. Hemodynamic failure was staged according to PET findings: stage 0: normal CBF, normal perfusion reserve; stage I: normal CBF, decreased perfusion reserve; and stage II: decreased CBF, decreased perfusion reserve. The BOLD CVR data set of the same subjects was then matched to the corresponding stage of hemodynamic failure. RESULTS: PET-based stage I versus stage II could also be clearly separated with BOLD CVR measurements (CVR for stage I 0.11 versus CVR for stage II -0.03; P<0.01). Hemispheric and middle cerebral artery territory difference analyses (ie, affected versus unaffected side) showed a significant correlation for CVR impairment in the affected hemisphere and middle cerebral artery territory (P<0.01, R2=0.47 and P=0.02, R2= 0.25, respectively). CONCLUSIONS: BOLD CVR corresponded well to CBF perfusion reserve measurements obtained with (15O-)H2O-PET, especially for detecting hemodynamic failure in the affected hemisphere and middle cerebral artery territory and for identifying hemodynamic failure stage II. BOLD CVR may, therefore, be considered for prospective studies assessing stroke risk in patients with chronic cerebrovascular steno-occlusive disease, in particular because it can potentially be implemented in routine clinical imaging.

Utility of deep neural networks in predicting gross-total resection after transsphenoidal surgery for pituitary adenoma: a pilot study.

OBJECTIVEGross-total resection (GTR) is often the primary surgical goal in transsphenoidal surgery for pituitary adenoma. Existing classifications are effective at predicting GTR but are often hampered by limited discriminatory ability in moderate cases and by poor interrater agreement. Deep learning, a subset of machine learning, has recently established itself as highly effective in forecasting medical outcomes. In this pilot study, the authors aimed to evaluate the utility of using deep learning to predict GTR after transsphenoidal surgery for pituitary adenoma.METHODSData from a prospective registry were used. The authors trained a deep neural network to predict GTR from 16 preoperatively available radiological and procedural variables. Class imbalance adjustment, cross-validation, and random dropout were applied to prevent overfitting and ensure robustness of the predictive model. The authors subsequently compared the deep learning model to a conventional logistic regression model and to the Knosp classification as a gold standard.RESULTSOverall, 140 patients who underwent endoscopic transsphenoidal surgery were included. GTR was achieved in 95 patients (68%), with a mean extent of resection of 96.8% ± 10.6%. Intraoperative high-field MRI was used in 116 (83%) procedures. The deep learning model achieved excellent area under the curve (AUC; 0.96), accuracy (91%), sensitivity (94%), and specificity (89%). This represents an improvement in comparison with the Knosp classification (AUC: 0.87, accuracy: 81%, sensitivity: 92%, specificity: 70%) and a statistically significant improvement in comparison with logistic regression (AUC: 0.86, accuracy: 82%, sensitivity: 81%, specificity: 83%) (all p < 0.001).CONCLUSIONSIn this pilot study, the authors demonstrated the utility of applying deep learning to preoperatively predict the likelihood of GTR with excellent performance. Further training and validation in a prospective multicentric cohort will enable the development of an easy-to-use interface for use in clinical practice.

Blood oxygen-level dependent functional assessment of cerebrovascular reactivity: Feasibility for intraoperative 3 Tesla MRI.

PURPOSE: To assess the feasibility of functional blood oxygen-level dependent (BOLD) MRI to evaluate intraoperative cerebrovascular reactivity (CVR) at 3 Tesla field strength. METHODS: Ten consecutive neurosurgical subjects scheduled for a clinical intraoperative MRI examination were enrolled in this study. In addition to the clinical protocol a BOLD sequence was implemented with three cycles of 44 s apnea to calculate CVR values on a voxel-by-voxel basis throughout the brain. The CVR range was then color-coded and superimposed on an anatomical volume to create high spatial resolution CVR maps. RESULTS: Ten subjects (mean age 34.8 ± 13.4; 2 females) uneventfully underwent the intraoperative BOLD protocol, with no complications occurring. Whole-brain CVR for all subjects was (mean ± SD) 0.69 ± 0.42, whereas CVR was markedly higher for tumor subjects as compared to vascular subjects, 0.81 ± 0.44 versus 0.33 ± 0.10, respectively. Furthermore, color-coded functional maps could be robustly interpreted for a whole-brain assessment of CVR. CONCLUSION: We demonstrate that intraoperative BOLD MRI is feasible in creating functional maps to assess cerebrovascular reactivity throughout the brain in subjects undergoing a neurosurgical procedure. Magn Reson Med 77:806-813, 2017. © 2016 International Society for Magnetic Resonance in Medicine.