Application of Delta T1 maps for quantitative and objective assessment of extent of resection and survival prediction in glioblastoma.

Background and Objectives: Gross-total resection (GTR) and low residual tumor volume (RTV) have been associated with increased survival in glioblastoma. Largely due to the subjectivity involved, the determination of GTR and RTV remains difficult in the postoperative setting. In response, the objective of this study is to evaluate the clinical efficacy of an easy-to-use MRI metric, called delta T1 (dT1), to quantify extent of resection (EOR) and RTV, in comparison to radiologist impression, to predict overall survival (OS) in glioblastoma patients. Methods: 59 patients who underwent resection of glioblastoma were retrospectively identified. Delta T1 (dT1) images, automatically created from the difference between calibrated post- and pre-contrast T1-weighted images, were used to quantify EOR and RTV. Kaplan-Meier survival estimates were determined for EOR categories, an RTV cutoff of 5cm3 and radiologist interpretation of EOR. Multivariate Cox proportional hazard regression analysis was used to evaluate RTV and EOR along with effects related to sex, KPS, MGMT, and age on OS. Results: Kaplan-Meier analysis revealed a statistically significant difference in median OS for a dT1-determined RTV cutoff of 5 cm3 (P=.0024, HR=2.18 (1.232-3.856)), but not for radiological impression (P=0.666) or dT1-determined EOR (P=0.0803), which was limited to a comparison between partial and subtotal resections. Furthermore, when covariates were accounted for in multivariate Cox regression, significant differences in OS were retained for dT1-determined RTV. Additionally, a significantly strong yet short-term effect of MGMT methylation status on OS was revealed for each RTV and EOR model. Conclusion: The utility of dT1 maps to quantify EOR and RTV in glioblastoma and predict survival, suggests an emerging role for dT1s with relevance for intraoperative MRI, neuro-navigation and postoperative disease surveillance.

Noninvasive Autopsy-Validated Tumor Probability Maps Identify Glioma Invasion Beyond Contrast Enhancement.

BACKGROUND AND OBJECTIVES: This study identified a clinically significant subset of patients with glioma with tumor outside of contrast enhancement present at autopsy and subsequently developed a method for detecting nonenhancing tumor using radio-pathomic mapping. We tested the hypothesis that autopsy-based radio-pathomic tumor probability maps would be able to noninvasively identify areas of infiltrative tumor beyond traditional imaging signatures. METHODS: A total of 159 tissue samples from 65 subjects were aligned to MRI acquired nearest to death for this retrospective study. Demographic and survival characteristics for patients with and without tumor beyond the contrast-enhancing margin were computed. An ensemble algorithm was used to predict pixelwise tumor presence from pathological annotations using segmented cellularity (Cell), extracellular fluid, and cytoplasm density as input (6 train/3 test subjects). A second level of ensemble algorithms was used to predict voxelwise Cell, extracellular fluid, and cytoplasm on the full data set (43 train/22 test subjects) using 5-by-5 voxel tiles from T1, T1 + C, fluid-attenuated inversion recovery, and apparent diffusion coefficient as input. The models were then combined to generate noninvasive whole brain maps of tumor probability. RESULTS: Tumor outside of contrast was identified in 41.5% of patients, who showed worse survival outcomes (hazard ratio = 3.90, P < .001). Tumor probability maps reliably tracked nonenhancing tumor on a range of local and external unseen data, identifying tumor outside of contrast in 69% of presurgical cases that also showed reduced survival outcomes (hazard ratio = 1.67, P = .027). CONCLUSION: This study developed a multistage model for mapping gliomas using autopsy tissue samples as ground truth, which was able to identify regions of tumor beyond traditional imaging signatures.

Magnetic resonance imaging-derived relative cerebral blood volume characteristics in a case of pathologically confirmed neurocysticercosis: illustrative case.

BACKGROUND: Neurocysticercosis (NCC) is a parasitic infection of the brain caused by ingesting water or food contaminated with tapeworm eggs. When it presents as a solitary mass, differentiation from a primary brain tumor on imaging can be difficult. Magnetic resonance imaging (MRI)-derived relative cerebral blood volume (rCBV) is a newer imaging technique used to identify areas of neovascularization in tumors, which may advance the differential diagnosis. OBSERVATIONS: A 25-year-old male presented after a seizure. Computed tomography (CT) and MRI demonstrated a partially enhancing lesion with microcalcifications and vasogenic edema. Follow-up rCBV assessment demonstrated mild hyperperfusion and/or small vessels at the lesional margins consistent with either an intermediate grade glioma or infection. Given the radiological equipoise, surgical accessibility, and differential diagnosis including primary neoplasm, metastatic disease, NCC, and abscess, resection was pursued. The calcified mass was excised en bloc and was confirmed as larval-stage NCC. LESSONS: CT or MRI may not always provide sufficient information to distinguish NCC from brain tumors. Although reports have suggested that rCBV may aid in identifying NCC, here the authors describe a case of pathologically confirmed NCC in which preoperative, qualitative, standardized rCBV findings raised concern for a primary neoplasm. This case documents the first standardized rCBV values reported in a pathologically confirmed case of NCC in the United States.

Symptomatic radionecrosis after postoperative but not preoperative stereotactic radiosurgery in a single patient: illustrative case.

BACKGROUND: Standard of care for brain metastases involves stereotactic radiosurgery (SRS). For cases that also require surgery because of lesion size, edema, or neurological symptoms, whether to provide pre- or postoperative SRS has become a prevalent debate. OBSERVATIONS: Herein, the unique case of a patient with brain metastases of the same pathology and similar size in two different brain locations at two different times is described. The patient underwent surgery with preoperative SRS for the first lesion and surgery with postoperative SRS for the second lesion. Although both treatments resulted in successful local control, the location that received postoperative SRS developed symptomatic and rapidly progressive radiation necrosis (RN) requiring a third craniotomy. LESSONS: Large randomized controlled trials are ongoing to compare pre- versus postoperative SRS for the treatment of symptomatic brain metastases (e.g., study NRG-BN012). Recent interest in preoperative SRS has emerged from its theoretical potential to decrease rates of postoperative RN and leptomeningeal disease. This valuable case in which both therapies were applied in a single patient with a single pathology and similar lesions provides evidence supportive of preoperative SRS.

Quantifying Hand Strength and Isometric Pinch Individuation Using a Flexible Pressure Sensor Grid.

Modulating force between the thumb and another digit, or isometric pinch individuation, is critical for daily tasks and can be impaired due to central or peripheral nervous system injury. Because surgical and rehabilitative efforts often focus on regaining this dexterous ability, we need to be able to consistently quantify pinch individuation across time and facilities. Currently, a standardized metric for such an assessment does not exist. Therefore, we tested whether we could use a commercially available flexible pressure sensor grid (Tekscan F-Socket [Tekscan Inc., Norwood, MA, USA]) to repeatedly measure isometric pinch individuation and maximum voluntary contraction (MVC) in twenty right-handed healthy volunteers at two visits. We developed a novel equation informed by the prior literature to calculate isometric individuation scores that quantified percentage of force on the grid generated by the indicated digit. MVC intra-class correlation coefficients (ICCs) for the left and right hands were 0.86 (p < 0.0001) and 0.88 (p < 0.0001), respectively, suggesting MVC measurements were consistent over time. However, individuation score ICCs, were poorer (left index ICC 0.41, p = 0.28; right index ICC -0.02, p = 0.51), indicating that this protocol did not provide a sufficiently repeatable individuation assessment. These data support the need to develop novel platforms specifically for repeatable and objective isometric hand dexterity assessments.

Large-scale brain networks and intra-axial tumor surgery: a narrative review of functional mapping techniques, critical needs, and scientific opportunities.

In recent years, a paradigm shift in neuroscience has been occurring from "localizationism," or the idea that the brain is organized into separately functioning modules, toward "connectomics," or the idea that interconnected nodes form networks as the underlying substrates of behavior and thought. Accordingly, our understanding of mechanisms of neurological function, dysfunction, and recovery has evolved to include connections, disconnections, and reconnections. Brain tumors provide a unique opportunity to probe large-scale neural networks with focal and sometimes reversible lesions, allowing neuroscientists the unique opportunity to directly test newly formed hypotheses about underlying brain structural-functional relationships and network properties. Moreover, if a more complete model of neurological dysfunction is to be defined as a "disconnectome," potential avenues for recovery might be mapped through a "reconnectome." Such insight may open the door to novel therapeutic approaches where previous attempts have failed. In this review, we briefly delve into the most clinically relevant neural networks and brain mapping techniques, and we examine how they are being applied to modern neurosurgical brain tumor practices. We then explore how brain tumors might teach us more about mechanisms of global brain dysfunction and recovery through pre- and postoperative longitudinal connectomic and behavioral analyses.

Quantitative assessments of finger individuation with an instrumented glove.

BACKGROUND: In clinical and research settings, hand dexterity is often assessed as finger individuation, or the ability to move one finger at a time. Despite its clinical importance, there is currently no standardized, sufficiently sensitive, or fully objective platform for these evaluations. METHODS: Here we developed two novel individuation scores and tested them against a previously developed score using a commercially available instrumented glove and data collected from 20 healthy adults. Participants performed individuation for each finger of each hand as well as whole hand open-close at two study visits separated by several weeks. Using the three individuation scores, intra-class correlation coefficients (ICC) and minimal detectable changes (MDC) were calculated. Individuation scores were further correlated with subjective assessments to assess validity. RESULTS: We found that each score emphasized different aspects of individuation performance while generating scores on the same scale (0 [poor] to 1 [ideal]). These scores were repeatable, but the quality of the metrics varied by both equation and finger of interest. For example, index finger intra-class correlation coefficients (ICC's) were 0.90 (< 0.0001), 0.77 (< 0.001), and 0.83 (p < 0.0001), while pinky finger ICC's were 0.96 (p < 0.0001), 0.88 (p < 0.0001), and 0.81 (p < 0.001) for each score. Similarly, MDCs also varied by both finger and equation. In particular, thumb MDCs were 0.068, 0.14, and 0.045, while index MDCs were 0.041, 0.066, and 0.078. Furthermore, objective measurements correlated with subjective assessments of finger individuation quality for all three equations (ρ = - 0.45, p < 0.0001; ρ = - 0.53, p < 0.0001; ρ = - 0.40, p < 0.0001). CONCLUSIONS: Here we provide a set of normative values for three separate finger individuation scores in healthy adults with a commercially available instrumented glove. Each score emphasizes a different aspect of finger individuation performance and may be more uniquely applicable to certain clinical scenarios. We hope for this platform to be used within and across centers wishing to share objective data in the physiological study of hand dexterity. In sum, this work represents the first healthy participant data set for this platform and may inform future translational applications into motor physiology and rehabilitation labs, orthopedic hand and neurosurgery clinics, and even operating rooms.

Synchronization of kinetic and kinematic hand tasks with electrocorticography and cortical stimulation during awake craniotomies.

Awake craniotomies provide unique and invaluable scientific opportunities for neurophysiological experimentation in consenting human subjects. While such experimentation carries a long history, rigorous reporting of methodologies focusing on synchronizing data across multiple platforms is not universally reported and often not translatable to across operating rooms, facilities, or behavioral tasks. Therefore, here we detail an intraoperative data synchronization methodology designed to work across multiple commercially available platforms to collect behavioral and surgical field videos, electrocorticography, brain stimulation timing, continuous finger joint angles, and continuous finger force production. Our technique was developed to be nonobstructive to operating room (OR) staff and generalizable to a variety of hand-based tasks. We hope that the detailed reporting of our methods will support the scientific rigor and reproducibility of future studies, as well as aid other groups interested in performing related experiments.

Pinless Electromagnetic Neuronavigation During Awake Craniotomies: Technical Pearls, Pitfalls, and Nuances.

BACKGROUND: Awake craniotomies are often performed with rigid pin fixation to support optical neuronavigation. Newer electromagnetic (EM) neuronavigation technology now enables unpinned cranial neurosurgery while maintaining robust intraoperative image guidance. Here, we share technical nuances, operative pearls, and lessons learned from our institutional experience using Curve EM neuronavigation during awake, unpinned craniotomies. METHODS: We describe our process for patient positioning, instrumentation setup, system registration, intraoperative navigation, and surgical adjunct use (e.g., intraoperative neuromonitoring and intraoperative magnetic resonance imaging) in detail. At each step, we provide pearls for success and tips for pitfall avoidance based on our experience. RESULTS: Ten patients underwent awake pinless intra-axial tumor resection using Curve EM neuronavigation from May 2021 to August 2022 with a single surgeon. Postoperative transient neurological deficits were seen in 8 of 10 cases (80.0%), as all resections were taken to functional margins. Of the 9 patients with a 3-month follow-up visit at the time of publication, all 9 (100%) had improved or stable preoperative symptoms. No surgical complications, clinically appreciable inaccuracies, intraoperative losses of registration, unexpected postoperative magnetic resonance imaging findings, or errors related to the use of EM neuronavigation occurred. CONCLUSIONS: The technical pearls outlined here will help interested neurosurgeons integrate EM neuronavigation into awake craniotomies. In our experience, using unpinned neuronavigation during awake cases provides many advantages to the patient, surgeon, and entire operative team. It has thus become the standard practice at our institution.

Case report: Fractional brain tumor burden magnetic resonance mapping to assess response to pulsed low-dose-rate radiotherapy in newly-diagnosed glioblastoma.

Background: Pulsed low-dose-rate radiotherapy (pLDR) is a commonly used reirradiation technique for recurrent glioma, but its upfront use with temozolomide (TMZ) following primary resection of glioblastoma is currently under investigation. Because standard magnetic resonance imaging (MRI) has limitations in differentiating treatment effect from tumor progression in such applications, perfusion-weighted MRI (PWI) can be used to create fractional tumor burden (FTB) maps to spatially distinguish active tumor from treatment-related effect. Methods: We performed PWI prior to re-resection in four patients with glioblastoma who had undergone upfront pLDR concurrent with TMZ who had radiographic suspicion for tumor progression at a median of 3 months (0-5 months or 0-143 days) post-pLDR. The pathologic diagnosis was compared to retrospectively-generated FTB maps. Results: The median patient age was 55.5 years (50-60 years). All were male with IDH-wild type (n=4) and O6-methylguanine-DNA methyltransferase (MGMT) hypermethylated (n=1) molecular markers. Pathologic diagnosis revealed treatment effect (n=2), a mixture of viable tumor and treatment effect (n=1), or viable tumor (n=1). In 3 of 4 cases, FTB maps were indicative of lesion volumes being comprised predominantly of treatment effect with enhancing tumor volumes comprised of a median of 6.8% vascular tumor (6.4-16.4%). Conclusion: This case series provides insight into the radiographic response to upfront pLDR and TMZ and the role for FTB mapping to distinguish tumor progression from treatment effect prior to redo-surgery and within 20 weeks post-radiation.

Understanding Variable Motor Responses to Direct Electrical Stimulation of the Human Motor Cortex During Brain Surgery.

Direct electrical stimulation of the brain is the gold standard technique used to define functional-anatomical relationships during neurosurgical procedures. Areas that respond to stimulation are considered "critical nodes" of circuits that must remain intact for the subject to maintain the ability to perform certain functions, like moving and speaking. Despite its routine use, the neurophysiology underlying downstream motor responses to electrical stimulation of the brain, such as muscle contraction or movement arrest, is poorly understood. Furthermore, varying and sometimes counterintuitive responses can be seen depending on how and where the stimulation is applied, even within the human primary motor cortex. Therefore, here we review relevant neuroanatomy of the human motor system, provide a brief historical perspective on electrical brain stimulation, explore mechanistic variations in stimulation applications, examine neurophysiological properties of different parts of the motor system, and suggest areas of future research that can promote a better understanding of the interaction between electrical stimulation of the brain and its function.

Generating artificial sensations with spinal cord stimulation in primates and rodents.

For patients who have lost sensory function due to a neurological injury such as spinal cord injury (SCI), stroke, or amputation, spinal cord stimulation (SCS) may provide a mechanism for restoring somatic sensations via an intuitive, non-visual pathway. Inspired by this vision, here we trained rhesus monkeys and rats to detect and discriminate patterns of epidural SCS. Thereafter, we constructed psychometric curves describing the relationship between different SCS parameters and the animal's ability to detect SCS and/or changes in its characteristics. We found that the stimulus detection threshold decreased with higher frequency, longer pulse-width, and increasing duration of SCS. Moreover, we found that monkeys were able to discriminate temporally- and spatially-varying patterns (i.e. variations in frequency and location) of SCS delivered through multiple electrodes. Additionally, sensory discrimination of SCS-induced sensations in rats obeyed Weber's law of just-noticeable differences. These findings suggest that by varying SCS intensity, temporal pattern, and location different sensory experiences can be evoked. As such, we posit that SCS can provide intuitive sensory feedback in neuroprosthetic devices.

Freezing of Gait in Parkinson's Disease: Invasive and Noninvasive Neuromodulation.

INTRODUCTION: Freezing of gait (FoG) is one of the most disabling yet poorly understood symptoms of Parkinson's disease (PD). FoG is an episodic gait pattern characterized by the inability to step that occurs on initiation or turning while walking, particularly with perception of tight surroundings. This phenomenon impairs balance, increases falls, and reduces the quality of life. MATERIALS AND METHODS: Clinical-anatomical correlations, electrophysiology, and functional imaging have generated several mechanistic hypotheses, ranging from the most distal (abnormal central pattern generators of the spinal cord) to the most proximal (frontal executive dysfunction). Here, we review the neuroanatomy and pathophysiology of gait initiation in the context of FoG, and we discuss targets of central nervous system neuromodulation and their outcomes so far. The PubMed database was searched using these key words: neuromodulation, freezing of gait, Parkinson's disease, and gait disorders. CONCLUSION: Despite these investigations, the pathogenesis of this process remains poorly understood. The evidence presented in this review suggests FoG to be a heterogenous phenomenon without a single unifying pathologic target. Future studies rigorously assessing targets as well as multimodal approaches will be essential to define the next generation of therapeutic treatments.

Operative Technique and Lessons Learned From Surgical Implantation of the NeuroPace Responsive Neurostimulation® System in 57 Consecutive Patients.

BACKGROUND: The Responsive Neurostimulation (RNS)® System (NeuroPace, Inc) is an implantable device designed to improve seizure control in patients with medically refractory focal epilepsy. Because it is relatively new, surgical pearls and operative techniques optimized from experience beyond a small case series have yet to be described. OBJECTIVE: To provide a detailed description of our operative technique and surgical pearls learned from implantation of the RNS System in 57 patients at our institution. We describe our method for frame-based placement of amygdalo-hippocampal depth leads, open implantation of cortical strip leads, and open installation of the neurostimulator. METHODS: We outline considerations for patient selection, preoperative planning, surgical positioning, incision planning, stereotactic depth lead implantation, cortical strip lead implantation, craniotomy for neurostimulator implantation, device testing, closure, and intraoperative imaging. RESULTS: The median reduction in clinical seizure frequency was 60% (standard deviation 63.1) with 27% of patients achieving seizure freedom at last follow up (median 23.1 mo). No infections, intracerebral hemorrhages, or lead migrations were encountered. Two patients experienced lead fractures, and four lead exchanges have been performed. CONCLUSION: The techniques set forth here will help with the safe and efficient implantation of these new devices.

Spinal cord stimulation and rehabilitation in an individual with chronic complete L1 paraplegia due to a conus medullaris injury: motor and functional outcomes at 18 months.

INTRODUCTION: Epidural electrical stimulation of the conus medullaris has helped facilitate native motor recovery in individuals with complete cervicothoracic spinal cord injuries (SCI). A theorized mechanism of clinical improvement includes supporting central pattern generators intrinsic to the conus medullaris. Because spinal cord stimulators (SCS) are approved for the treatment of neuropathic pain, we were able to test this experimental therapy in a subject with complete L1 paraplegia and neuropathic genital pain due to a traumatic conus injury. CASE PRESENTATION: An otherwise healthy 48-year-old male with chronic complete L1 paraplegia with no zones of partial preservation (ZPP) and intractable neuropathic genital pain presented to our group seeking nonmedical pain relief and any possible help with functional restoration. After extensive evaluation, discussion, and consent, we proceeded with SCS implantation at the conus and an intensive outpatient physical therapy regimen consistent with the recent SCI rehabilitation literature. DISCUSSION: Intraoperatively, no electromyography (EMG) could be elicited with epidural conus stimulation. At 18 months after implantation, his motor ZPPs had advanced from L1 to L5 on the left and from L1 to L3 on the right. Qualitative increases in lower extremity resting state EMG amplitudes were noted, although there was no consistent evidence of voluntary EMG or rhythmic locomotive leg movements. Three validated functional and quality of life (QoL) surveys demonstrated substantial improvements. The modest motor response compared to the literature suggests likely critical differences in the anatomy of such a low injury. However, the change in ZPPs and QoL suggest potential for neuroplasticity even in this patient population.

Toward Functional Restoration of the Central Nervous System: A Review of Translational Neuroscience Principles.

Injury to the central nervous system (CNS) can leave patients with devastating neurological deficits that may permanently impair independence and diminish quality of life. Recent insights into how the CNS responds to injury and reacts to critically timed interventions are being translated into clinical applications that have the capacity to drastically improve outcomes for patients suffering from permanent neurological deficits due to spinal cord injury, stroke, or other CNS disorders. The translation of such knowledge into practical and impactful treatments involves the strategic collaboration between neurosurgeons, clinicians, therapists, scientists, and industry. Therefore, a common understanding of key neuroscientific principles is crucial. Conceptually, current approaches to CNS revitalization can be divided by scale into macroscopic (systems-circuitry) and microscopic (cellular-molecular). Here we review both emerging and well-established tenets that are being utilized to enhance CNS recovery on both levels, and we explore the role of neurosurgeons in developing therapies moving forward. Key principles include plasticity-driven functional recovery, cellular signaling mechanisms in axonal sprouting, critical timing for recovery after injury, and mechanisms of action underlying cellular replacement strategies. We then discuss integrative approaches aimed at synergizing interventions across scales, and we make recommendations for the basis of future clinical trial design. Ultimately, we argue that strategic modulation of microscopic cellular behavior within a macroscopic framework of functional circuitry re-establishment should provide the foundation for most neural restoration strategies, and the early involvement of neurosurgeons in the process will be crucial to successful clinical translation.

Characterization of Perinatal Risk Factors and Complications Associated With Nonsyndromic Craniosynostosis.

BACKGROUND: Certain intrauterine risk factors are known to increase the risk of premature cranial suture fusion and may cause complications during birth. Some of these risk factors may be modifiable. Therefore, the authors sought to characterize the institutional patterns of prenatal risk factors and perinatal complications in nonsyndromic craniosynostosis patients compared to normal births from the surrounding area to identify areas for possible intervention or prevention. METHODS: The medical records of all infants with nonsyndromic craniosynostosis and full birth records born at Duke University Health System from 2006 to 2017 were retrospectively reviewed. Maternal comorbidities, prenatal risk factors, and perinatal complications were collected. The North Carolina State Center for Health Statistics was queried for perinatal statistics from Durham county and the Northeastern Perinatal Care Region to represent a control cohort of normal births from the same time period and region. The primary outcome investigated was the incidence of prenatal risk factors and complications at birth associated with premature fusion of cranial sutures. RESULTS: Eighty births with nonsyndromic craniosynostosis were included in this study. The majority of these patients were males (61.7%) and born via cesarean section (55.0%). Intrauterine growth restriction occurred in 10.0% and head trauma during delivery occurred in 2.5%. Twinning (14.8% vs 3.6%, P < 0.0001), cesarean births (55.5% vs 30.0%, P < 0.0001), and breech presentation (17.3% vs 3.2%, P < 0.0001) were significantly more common in craniosynostosis patients. Prenatally, mothers of craniosynostosis infants had higher incidence of gestational diabetes (13.5% vs 5.0%, P < 0.0001) and oligohydramnios (6.1% vs 1.3%, P < 0.0001) compared to regional controls. CONCLUSION: This study demonstrates that premature suture fusion is associated with prenatal risk factors such as gestational diabetes and oligohydramnios. Continued research into potentially modifiable prenatal risk factors and more refined prenatal diagnostic tools has the potential to reduce both the incidence of premature suture fusion and the sequelae of birth complications in this population.

Neurological Outcomes After Surgical or Conservative Management of Spontaneous Spinal Epidural Abscesses: A Systematic Review and Meta-Analysis of Data From 1980 Through 2016.

STUDY DESIGN: This is a meta-analysis. OBJECTIVE: Perform a systematic review and quantitative meta-analysis of neurological outcomes from all available spinal epidural abscess (SEA) literature published between 1980 and 2016. SUMMARY OF BACKGROUND DATA: Current literature on SEAs lacks large-scale data characterizing prognostic factors and surgical indications. MATERIALS AND METHODS: PubMed was queried for studies reporting neurological outcomes from patients undergoing conservative or surgical management for spontaneous SEA. Inclusion criteria included outcomes data measured ≥6 months after presentation, ≥10 human subjects, and diagnosis by magnetic resonance imaging or Computed tomography-myelogram. Where available, demographic data, abscess location, comorbidities, pretreatment neurological deficits, treatment methods, bacterial speciation, and complications were extracted from each study. Potential outcome predictors represented by continuous variables were compared using student t test and categorical variables were compared using the Pearson χ test. Variables identified as potentially associated with outcome (P≤0.05) were subjected to meta-analysis using Cochran-Mantel-Haenszel testing to calculate odds ratios (ORs) and 95% confidence intervals (CIs). RESULTS: In total, 808 patients were analyzed from 20 studies that met inclusion criteria. 456 (56.3%) patients were treated with surgery and antibiotics, and 353 (43.7%) patients were managed with antibiotics alone. Neither surgical intervention (OR=1.01, 95% CI=0.40-2.59), lumbosacral location (OR=1.51, 95% CI=0.23-9.79), nor neurological deficit on presentation (OR=0.88, 95% CI=0.40-1.92) were significantly associated with good (stable or improved) or bad (worsened) neurological outcome, whereas delayed surgery was significantly associated with bad outcome (OR=0.01, 95% CI=0.02-0.62) and cervicothoracic location approached significance for predicting bad outcome (OR=0.41, 95% CI=0.15-1.09). CONCLUSIONS: Current literature does not definitively support or oppose surgical intervention in all SEA cases. Therefore, until better evidence exists, the decision to operate must be made on an individual case-by-case basis with the goals of preventing neurological decline, obtaining source control after failed conservative treatment, or restoring spinal stability.