Precision oncology in neurofibromatosis type 1: quantification of differential sensitivity to selumetinib in plexiform neurofibromas using single-cell RNA sequencing.

PURPOSE: Selumetinib is an FDA-approved targeted therapy for plexiform neurofibromas in neurofibromatosis type 1(NF1) with durable response rates seen in most, but not all patients. In this proof-of-concept study, we demonstrate single-cell RNA sequencing(scRNAseq) as a technique for quantifying drug response to selumetinib at the single cell level. METHODS: scRNAseq data from neurofibroma biopsies was obtained from a public genomics repository. Schwann cell populations were identified through standard clustering techniques and single-cell selumetinib sensitivity was quantified on a scale of 0(resistant) to 1(sensitive) based on the expression pattern of a 500 gene selumetinib sensitivity signature from the BeyondCell sensitivity library. RESULTS: A total of seven plexiform neurofibromas were included in our final analysis. The median absolute number of Schwann cells across samples was 658 cells (IQR: 1,029 cells, Q1-Q3: 135 cells to 1,163 cells). There was a statistically significant difference in selumetinib sensitivity profiles across samples (p < 0.001). The tumor with the highest median selumetinib sensitivity score had a median selumetinib sensitivity score of 0.64(IQR: 0.14, Q1-Q3: 0.59-0.70, n = 112 cells) and the tumor with the lowest median selumetinib sensitivity score had a median score of 0.37 (IQR: 0.21, Q1-Q3: 0.27-0.48, n = 1,034 cells). CONCLUSIONS: scRNAseq of plexiform neurofibroma biopsies reveals differential susceptibilities to selumetinib on a single cell level. These findings may explain the partial responses seen in clinical trials of selumetinib for NF1 and demonstrate the value of collecting scRNAseq data for future NF1 trials.

Outcomes of Acute Respiratory Failure in Patients With Cancer in the United States.

OBJECTIVE: To determine the epidemiological effect-magnitude and outcomes of patients with cancer vs those without cancer who are hospitalized with acute respiratory failure (ARF). PATIENTS AND METHODS: We reviewed hospitalizations within the National Inpatient Sample (NIS) database between January 1, 2016, and December 31, 2018. Patients were classified based on a diagnosis of solid-organ cancer, hematologic cancer, or no cancer. Noninvasive positive pressure ventilation (NIPPV) failure was defined as patients who initially received NIPPV and had progression to invasive mechanical ventilation. Weighted samples were used to derive population estimates. RESULTS: During the study period, there were an estimated 8,837,209 admissions with ARF in the United States, 8.9% (783,625) of which had solid-organ cancer and 2.0% (176,095) had hematologic cancers. Annually, 319,907 patients with cancer are admitted with ARF, with 27.3% (87,302) requiring invasive mechanical ventilation and 10.0% (31,998) requiring NIPPV. In-hospital mortality was higher in patients with cancer vs those without cancer (24.0% [76,813] vs 12.3% [322,465]; P<.001), and this proprotion persisted when stratified by the highest method of oxygen delivery. Patients with cancer had longer hospital length of stay (7.0 days [3.0 to 12.0 days] vs 5.0 days [3.0 to 10.0 days]; P<.001) and were more likely to have NIPPV failure (14.9% [3,992] vs 12.8% [41,875]). Compared with those with solid-organ cancer, patients with hematologic cancers experienced worse outcomes. The association between underlying cancer diagnosis and outcomes remained consistent when adjusted for age, sex, and comorbidities. CONCLUSION: In the United States, patients with cancer account for over 10% of ARF hospital admissions (959,720 of 8,837,209). They experience an approximately 2-fold higher mortality versus those without cancer. Those with hematologic cancers appear to experience worse outcomes than patients with solid-organ cancers.

High-Value Epilepsy Care in the United States: Predictors of Increased Costs and Complications from the National Inpatient Sample Database 2016-2019.

BACKGROUND: For patients with medically refractory epilepsy, newer minimally invasive techniques such as laser interstitial thermal therapy (LITT) have been developed in recent years. This study aims to characterize trends in the utilization of surgical resection versus LITT to treat medically refractory epilepsy, characterize complications, and understand the cost of this innovative technique to the public. METHODS: The National Inpatient Sample database was queried from 2016 to 2019 for all patients admitted with a diagnosis of medically refractory epilepsy. Patient demographics, hospital length of stay, complications, and costs were tabulated for all patients who underwent LITT or surgical resection within these cohorts. RESULTS: A total of 6019 patients were included, 223 underwent LITT procedures, while 5796 underwent resection. Significant predictors of increased patient charges for both cohorts included diabetes (odds ratio: 1.7, confidence interval [CI]: 1.44-2.19), infection (odds ratio: 5.12, CI 2.73-9.58), and hemorrhage (odds ratio: 2.95, CI 2.04-4.12). Procedures performed at nonteaching hospitals had 1.54 greater odds (CI 1.02-2.33) of resulting in a complication compared to teaching hospitals. Insurance status did significantly differ (P = 0.001) between those receiving LITT (23.3% Medicare; 25.6% Medicaid; 44.4% private insurance; 6.7 Other) and those undergoing resection (35.3% Medicare; 22.5% Medicaid; 34.7% private Insurance; 7.5% other). When adjusting for patient demographics, LITT patients had shorter length of stay (2.3 vs. 8.9 days, P < 0.001), lower complication rate (1.9% vs. 3.1%, P = 0.385), and lower mean hospital ($139,412.79 vs. $233,120.99, P < 0.001) and patient ($55,394.34 vs. $37,756.66, P < 0.001) costs. CONCLUSIONS: The present study highlights LITT's advantages through its association with lower costs and shorter length of stay. The present study also highlights the associated predictors of LITT versus resection, such as that most LITT cases happen at academic centers for patients with private insurance. As the adoption of LITT continues, more data will become available to further understand these issues.

Optimal Implant Sizing Using Machine Learning Is Associated With Increased Range of Motion After Cervical Disk Arthroplasty.

BACKGROUND AND OBJECTIVES: Cervical disk arthroplasty (CDA) offers the advantage of motion preservation in the treatment of focal cervical pathology. At present, implant sizing is performed using subjective tactile feedback and imaging of trial cages. This study aims to construct interpretable machine learning (IML) models to accurately predict postoperative range of motion (ROM) and identify the optimal implant sizes that maximize ROM in patients undergoing CDA. METHODS: Adult patients who underwent CDA for single-level disease from 2012 to 2020 were identified. Patient demographics, comorbidities, and outcomes were collected, including symptoms, examination findings, subsidence, and reoperation. Affected disk height, healthy rostral disk height, and implant height were collected at sequential time points. Linear regression and IML models, including bagged regression tree, bagged multivariate adaptive regression spline, and k-nearest neighbors, were used to predict ROM change. Model performance was assessed by calculating the root mean square error (RMSE) between predicted and actual changes in ROM in the validation cohort. Variable importance was assessed using RMSE loss. Area under the curve analyses were performed to identify the ideal implant size cutoffs in predicting improved ROM. RESULTS: Forty-seven patients were included. The average RMSE between predicted and actual ROM was 7.6° (range: 5.8-10.1) in the k-nearest neighbors model, 7.8° (range: 6.5-10.0) in the bagged regression tree model, 7.8° (range: 6.2-10.0) in the bagged multivariate adaptive regression spline model, and 15.8° (range: 14.3-17.5°) in a linear regression model. In the highest-performing IML model, graft size was the most important predictor with RMSE loss of 6.2, followed by age (RMSE loss = 5.9) and preoperative caudal disk height (RMSE loss = 5.8). Implant size at 110% of the normal adjacent disk height was the optimal cutoff associated with improved ROM. CONCLUSION: IML models can reliably predict change in ROM after CDA within an average of 7.6 degrees of error. Implants sized comparably with the healthy adjacent disk may maximize ROM.

An Integrated Voice Recognition and Natural Language Processing Platform to Automatically Extract Thoracolumbar Injury Classification Score Features From Radiology Reports.

BACKGROUND: Many predictive models for estimating clinical outcomes after spine surgery have been reported in the literature. However, implementation of predictive scores in practice is limited by the time-intensive nature of manually abstracting relevant predictors. In this study, we designed natural language processing (NLP) algorithms to automate data abstraction for the thoracolumbar injury classification score (TLICS). METHODS: We retrieved the radiology reports of all Mayo Clinic patients with an International Classification of Diseases, 9th or 10th revision, code corresponding to a fracture of the thoracolumbar spine between January 2005 and October 2020. Annotated data were used to train an N-gram NLP model using machine learning methods, including random forest, stepwise linear discriminant analysis, k-nearest neighbors, and penalized logistic regression models. RESULTS: A total of 1085 spine radiology reports were included in our analysis. Our dataset included 483 compression, 401 burst, 103 translational/rotational, and 98 distraction fractures. A total of 103 reports had documented an injury of the posterior ligamentous complex. The overall accuracy of the random forest model for fracture morphology feature detection was 76.96% versus 65.90% in the stepwise linear discriminant analysis, 50.69% in the k-nearest neighbors, and 62.67% in the penalized logistic regression. The overall accuracy to detect posterior ligamentous complex integrity was highest in the random forest model at 83.41%. Our random forest model was implemented in the backend of a web application in which users can dictate reports and have TLICS features automatically extracted. CONCLUSIONS: We have developed a machine learning NLP model for extracting TLICS features from radiology reports, which we deployed in a web application that can be integrated into clinical practice.

Limitations in Evaluating Machine Learning Models for Imbalanced Binary Outcome Classification in Spine Surgery: A Systematic Review.

Clinical prediction models for spine surgery applications are on the rise, with an increasing reliance on machine learning (ML) and deep learning (DL). Many of the predicted outcomes are uncommon; therefore, to ensure the models' effectiveness in clinical practice it is crucial to properly evaluate them. This systematic review aims to identify and evaluate current research-based ML and DL models applied for spine surgery, specifically those predicting binary outcomes with a focus on their evaluation metrics. Overall, 60 papers were included, and the findings were reported according to the PRISMA guidelines. A total of 13 papers focused on lengths of stay (LOS), 12 on readmissions, 12 on non-home discharge, 6 on mortality, and 5 on reoperations. The target outcomes exhibited data imbalances ranging from 0.44% to 42.4%. A total of 59 papers reported the model's area under the receiver operating characteristic (AUROC), 28 mentioned accuracies, 33 provided sensitivity, 29 discussed specificity, 28 addressed positive predictive value (PPV), 24 included the negative predictive value (NPV), 25 indicated the Brier score with 10 providing a null model Brier, and 8 detailed the F1 score. Additionally, data visualization varied among the included papers. This review discusses the use of appropriate evaluation schemes in ML and identifies several common errors and potential bias sources in the literature. Embracing these recommendations as the field advances may facilitate the integration of reliable and effective ML models in clinical settings.

Characterizing T-cell dysfunction and exclusion signatures in malignant peripheral nerve sheath tumors reveals susceptibilities to immunotherapy.

PURPOSE: Malignant peripheral nerve sheath tumors (MPNSTs) are malignant tumors that arise from peripheral nerves and are the leading cause of mortality in Neurofibromatosis Type 1 (NF1). In this study, we characterized whether transcriptomic signatures of T-cell dysfunction (TCD) and exclusion (TCE) that inversely correlate with response to immune checkpoint blockade (ICB) immunotherapy exist in MPNSTs. METHODS: MPNST transcriptomes were pooled from Gene Expression Omnibus (GEO). For each sample, a tumor immune dysfunction and exclusion (TIDE) score, TCD and TCE subscores, and cytotoxic T-cell(CTL) level were calculated. In the TIDE predictive algorithm, tumors are predicted to have an ICB response if they are either immunologically hot (CTL-high) without TCD or immunologically cold (CTL-low) without TCE. TIDE scores greater than zero correspond with ICB nonresponse. RESULTS: 73 MPNST samples met inclusion criteria, including 50 NF1-associated MPNSTs (68.5%). The average TIDE score was + 0.41 (SD = 1.16) with 22 (30.1%) predicted ICB responders. 11 samples were CTL-high (15.1%) with an average TCD score of + 0.99 (SD = 0.63). Among 62 CTL-low tumors, 21 were predicted to have ICB response with an average TCE score of + 0.31(SD = 1.20). Age(p = 0.18), sex(p = 0.41), NF1 diagnosis (p = 0.17), and PRC2 loss(p = 0.29) were not associated with ICB responder status. CONCLUSIONS: Transcriptomic analysis of TCD and TCE signatures in MPNST samples reveals that a select subset of patients with MPNSTs may benefit from ICB immunotherapy.

Prediction Model for Neurogenic Bladder Recovery One Year After Traumatic Spinal Cord Injury.

INTRODUCTION: Neurogenic bladder is a common complication after spinal cord injury (SCI) that carries substantial burdens on the inflicted individual. The objective of this study is to build a prediction model for neurogenic bladder recovery 1 year after traumatic SCI. METHODS: We queried the National Spinal Cord Injury Model Systems database for patients with traumatic SCI who had neurogenic bladder at the time of injury. The primary outcome of interest was the complete recovery of bladder function at 1 year. Multiple imputations were performed to generate replacement values for missing data, and the final imputed data were used for our analysis. A multivariable odds logistic regression model was developed for complete bladder recovery at 1 year. RESULTS: We identified a total of 2515 patients with abnormal bladder function at baseline who had an annual follow-up. A total of 417 patients (16.6%) recovered bladder function in 1 year. Predictors of complete bladder recovery included the following baseline parameters: sacral sensation, American Spinal Injury Association (ASIA) impairment score, bowel function at baseline, voluntary sphincter contraction, anal sensation, S1 motor scores, and the number of days in the rehabilitation facility. The model performed with a discriminative capacity of 90.5%. CONCLUSIONS: We developed a prediction model for the probability of complete bladder recovery 1 year after SCI. The model performed with a high discriminative capacity. This prediction model demonstrates potential utility in the counseling, research allocation, and management of individuals with SCI.

Using machine learning to predict 30-day readmission and reoperation following resection of supratentorial high-grade gliomas: an ACS NSQIP study involving 9418 patients.

OBJECTIVE: High-grade gliomas (HGGs) are among the rarest yet most aggressive tumor types in neurosurgical practice. In the current literature, few studies have assessed the drivers of early outcomes following resection of these tumors and investigated their association with quality of care. The authors aimed to identify the clinical predictors for 30-day readmission and reoperation following HGG surgery using the American College of Surgeons (ACS) National Surgical Quality Improvement Project (NSQIP) database and sought to create web-based applications predicting each outcome. METHODS: Using the ACS NSQIP database, the authors conducted a retrospective, multicenter cohort analysis of patients who underwent resection of supratentorial HGGs between January 1, 2016, and December 31, 2020. Demographics and comorbidities were extracted. The primary outcomes were 30-day unplanned readmission and reoperation. A stratified 80:20 split of the available data was carried out. Supervised machine learning algorithms were trained to predict 30-day outcomes. RESULTS: A total of 9418 patients were included in our cohort. The observed rate of unplanned readmission within 30 days of surgery was 13.0% (n = 1221). In terms of predictors, weight, chronic steroid use, preoperative blood urea nitrogen level, and white blood cell count were associated with a higher risk of readmission. The observed rate of unplanned reoperation within 30 days of surgery was 5.2% (n = 489). In terms of predictors, increased weight, longer operative time, and more days between hospital admission and operation were associated with an increased risk of early reoperation. The random forest algorithm showed the highest predictive performance for early readmission (area under the curve [AUC] = 0.967), while the XGBoost algorithm showed the highest predictive performance for early reoperation (AUC = 0.985). Web-based tools for both outcomes were deployed (https://glioma-readmission.herokuapp.com/ and https://glioma-reoperation.herokuapp.com/). CONCLUSIONS: In this study, the authors provide the first nationwide analysis for short-term outcomes in patients undergoing resection of supratentorial HGGs. Multiple patient, hospital, and admission factors were associated with readmission and reoperation, confirmed by machine learning predicting patients' prognosis, leading to better planning preoperatively and subsequently improved personalized patient care.

Bariatric surgery reduces odds of perioperative complications after inpatient hysterectomy: Analysis from a national database, 2016 to 2018.

BACKGROUND: Increased body mass index is a known risk factor for increased adverse events post-hysterectomy. The effects of previous bariatric surgery on outcomes after inpatient hysterectomy are not well elucidated. METHODS: The 2016 to 2018 National Inpatient Sample was queried for patients who underwent hysterectomy using International Classification of Disease 10 Procedure Codes before a matched analysis was performed to neutralize the potential confounding effects of comorbidities, body mass index, and age. Patients were divided into the following 2 groups: a case group (those with a history of bariatric surgery) and a control group (those without a history of bariatric surgery). Patients in the respective groups were matched 1:2 by age, Elixhauser comorbidity score, and body mass index at the time of surgery to analyze the risk of complications and mean length of stay. RESULTS: When 1:2 case-control matching was performed, women with a history of bariatric surgery (N = 595) had significantly fewer complications and decreased mean length of stay than the non-bariatric group (N = 1,190), even after controlling for body mass index at the time of hysterectomy. CONCLUSIONS: When matched for age, body mass index, and comorbidity score, patients with previous bariatric surgery had fewer complications and shorter lengths of stay than patients without a history of bariatric surgery. Women with a body mass index ≥40 kg/m2 requiring non-urgent hysterectomy may benefit from undergoing bariatric surgery first.

That which is unseen: 3D printing for pediatric cerebrovascular education.

INTRODUCTION: Pediatric cerebrovascular lesions are very rare and include aneurysms, arteriovenous malformations (AVM), and vein of Galen malformations (VOGM). OBJECTIVE: To describe and disseminate a validated, reproducible set of 3D models for optimization of neurosurgical training with respect to pediatric cerebrovascular diseases METHODS: All pediatric cerebrovascular lesions treated at our institution with adequate imaging studies during the study period 2015-2020 were reviewed by the study team. Three major diagnostic groups were identified: aneurysm, AVM, and VOGM. For each group, a case deemed highly illustrative of the core diagnostic and therapeutic principles was selected by the lead and senior investigators for printing (CSG/JM). Files for model reproduction and free distribution were prepared for inclusion as Supplemental Materials. RESULTS: Representative cases included a 7-month-old female with a giant left MCA aneurysm; a 3-day-old male with a large, complex, high-flow, choroidal-type VOGM, supplied from bilateral thalamic, choroidal, and pericallosal perforators, with drainage into a large prosencephalic vein; and a 7-year-old male with a left frontal AVM with one feeding arterial vessel from the anterior cerebral artery and one single draining vein into the superior sagittal sinus CONCLUSION: Pediatric cerebrovascular lesions are representative of rare but important neurosurgical diseases that require creative approaches for training optimization. As these lesions are quite rare, 3D-printed models and open source educational materials may provide a meaningful avenue for impactful clinical teaching with respect to a wide swath of uncommon or unusual neurosurgical diseases.

Microfluidic 'brain-on chip' systems to supplement neurological practice: development, applications and considerations.

Among the greatest general challenges in bioengineering is to mimic human physiology. Advanced efforts in tissue engineering have led to sophisticated 'brain-on-chip' (BoC) microfluidic devices that can mimic structural and functional aspects of brain tissue. BoC may be used to understand the biochemical pathways of neurolgical pathologies and assess promising therapeutic agents for facilitating regenerative medicine. We evaluated the potential of microfluidic BoC devices in various neurological pathologies, such as Alzheimer's, glioblastoma, traumatic brain injury, stroke and epilepsy. We also discuss the principles, limitations and future considerations of BoC technology. Results suggest that BoC models can help understand complex neurological pathologies and augment drug testing efforts for regenerative applications. However, implementing organ-on-chip technology to clinical practice has some practical limitations that warrant greater attention to improve large-scale applicability. Nevertheless, they remain to be versatile and powerful tools that can broaden our understanding of pathophysiological and therapeutic uncertainties to neurological diseases.

In this paper, the authors describe the role of microfluidic 'brain-on-chip' systems as a tool to model and study the human brain. While animal studies have provided significant insights, they lack the complexity of human brain tissue in order to verify the effects of drugs on patients, study complex physiological pathways or personalize regenerative therapies. This makes studying diseases of complex human organs challenging. Microfluidics is a field of study that can address these challenges by developing sophisticated and miniaturized devices that can chamber human tissue. These devices could allow scientists to better study diseases on a model that is accurate and controllable, allowing researchers to better understand complex diseases, assess drug efficacy to specific areas of the brain and potentially accelerate the development of new therapies. Herein, we characterize the principles, development and challenges of microfluidics and the role they have served in different neurological diseases.

Alginate hydrogels: A potential tissue engineering intervention for intervertebral disc degeneration.

Intervertebral disc (IVD) degeneration is a major cause of low back pain and disability, affecting millions of people worldwide. Current treatments for IVD degeneration are limited to invasive surgery or pain management. Recently, there has been increasing interest in the use of biomaterials, such as alginate hydrogels, for the treatment of IVD degeneration. Alginate hydrogels are an example of such a biomaterial that is biocompatible and can be tailored to mimic the native extracellular matrix of the IVD. Derived from alginate, a naturally derived polysaccharide from brown seaweed that can be transformed into a gelatinous solution, alginate hydrogels are emerging in the field of tissue engineering. They can be used to deliver therapeutic agents, such as growth factors or cells, to the site of injury, providing a localized and sustained release that may enhance treatment outcomes. This paper provides an overview on the use of alginate hydrogels for the treatment of IVD degeneration. We discuss the properties of alginate hydrogels and their potential applications for IVD regeneration, including the mechanism against IVD degeneration. We also highlight the research outcomes to date along with the challenges and limitations of using alginate hydrogels for IVD regeneration, including their mechanical properties, biocompatibility, and surgical compatibility. Overall, this review paper aims to provide a comprehensive overview of the current research on alginate hydrogels for IVD degeneration and to identify future directions for research in this area.

The association between bone mineral density and proximal junctional kyphosis in adult spinal deformity: a systematic review and meta-analysis.

OBJECTIVE: Proximal junctional kyphosis (PJK) is a complication of surgical management for adult spinal deformity (ASD) with a multifactorial etiology. Many risk factors are controversial, and their relative importance is not fully understood. The authors aimed to elucidate the association between bone mineral density (BMD) and PJK. METHODS: A systematic literature search was performed using PubMed and Web of Science keywords of "Proximal Junctional Kyphosis [MeSH] OR Proximal Junctional Failure [MeSH]" AND "Bone Mineral Density [MeSH] OR Hounsfield Units [MeSH] OR DEXA [MeSH]" set to the date range of January 2002 to July 2022. Studies required a minimum of 10 patients and 12 months of follow-up. Articles were included if they were in the English language and presented a primary retrospective cohort that included a comparison of patients with and without PJK, as well as a radiographic biomarker for BMD, such as Hounsfield units (HU) or T-score. RESULTS: A total of 18 unique studies with 2185 patients who underwent ASD surgery were identified. Of these, 537 patients (24.6%) developed PJK. Eight studies provided T-scores that were amenable to comparison, which found that patients who developed PJK were found to have lower BMD T-scores by a mean of -0.69 (95% CI -0.88 to -0.50; I2 = 63.9%, p < 0.001). The HU at the UIV among patients with the PJK group (n = 101) compared with the non-PJK group (n = 156) was found to be significantly lower (mean difference -32.35, 95% CI -46.05 to -18.65; I2 = 28.7%, p < 0.001). CONCLUSIONS: This meta-analysis suggests that low preoperative BMD as measured by T-score and a diagnosis of osteoporosis were associated with higher postoperative PJK. Additionally, lower HU on CT at the UIV were found to be significant risk factors for postoperative PJK as well. These findings suggest that more attention to preoperative BMD is a risk factor for PJK among ASD patients is warranted.

Deep Learning Approaches for Glioblastoma Prognosis in Resource-Limited Settings: A Study Using Basic Patient Demographic, Clinical, and Surgical Inputs.

BACKGROUND: Glioblastoma (GBM) is the most common brain tumor in the United States, with an annual incidence rate of 3.21 per 100,000. It is the most aggressive type of diffuse glioma and has a median survival of months after treatment. This study aims to assess the accuracy of different novel deep learning models trained on a set of simple clinical, demographic, and surgical variables to assist in clinical practice, even in areas with constrained health care infrastructure. METHODS: Our study included 37,095 patients with GBM from the SEER (Surveillance Epidemiology and End Results) database. All predictors were based on demographic, clinicopathologic, and treatment information of the cases. Our outcomes of interest were months of survival and vital status. Concordance index (C-index) and integrated Brier scores (IBS) were used to evaluate the performance of the models. RESULTS: The patient characteristics and the statistical analyses were consistent with the epidemiologic literature. The models C-index and IBS ranged from 0.6743 to 0.6918 and from 0.0934 to 0.1034, respectively. Probabilistic matrix factorization (0.6918), multitask logistic regression (0.6916), and logistic hazard (0.6916) had the highest C-index scores. The models with the lowest IBS were the probabilistic matrix factorization (0.0934), multitask logistic regression (0.0935), and logistic hazard (0.0936). These models had an accuracy (1-IBS) of 90.66%; 90.65%, and 90.64%, respectively. The deep learning algorithms were deployed on an interactive Web-based tool for practical use available via https://glioblastoma-survanalysis.herokuapp.com/. CONCLUSIONS: Novel deep learning algorithms can better predict GBM prognosis than do baseline methods and can lead to more personalized patient care regardless of extensive electronic health record availability.

Peritumoral edema in meningiomas: pathophysiology, predictors, and principles for treatment

Meningiomas is a tumor of the meninges and is among the most common intracranial neoplasms in adults, accounting for over a third of all primary brain tumors in the United States. Meningiomas can be associated with peritumoral brain edema (PTBE) which if not managed appropriately can lead to poor clinical outcomes. In this review, we summarize the relevant pathophysiology, predictors, and principles for treatment of PTBE. The results of various case-reports and case-series have found that meningioma-associated PTBE have patterns in age, tumor size, and hormone receptor positivity. Our study describes how increased age, increased tumor size, tumor location in the middle fossa, and positive expression of hormone receptors, VEGF, and MMP-9 can all be predictors for worse clinical outcomes. We also characterize treatment options for PTBE such as glucocorticoids and VEGF inhibitors along with the ongoing clinical trials attempting to alleviate PTBE in meningioma cases. The trends summarized in this review can be used to better predict the behavior of meningioma-associated PTBE and establish prognosis models to identify at risk patients (AU)

Examining the relationship between severe persistent mental illness and surgical outcomes in women undergoing mastectomy for breast cancer.

BACKGROUND: Severe persistent mental illness (SPMI) is associated with worse outcomes in cancer patients. Less is known about the relationship between SPMI and surgical outcomes after mastectomy for breast cancer. METHODS: We selected patients with breast cancer and SPMI from the National Inpatient Sample (2016-2018) and used propensity score matching. We then used multivariate analysis, Kruskal-Wallis tests, and conditional logistic regression to compare demographics and outcomes. RESULTS: The study sample consisted of 670 patients: 536 without SPMI and 134 with SPMI. SPMI was associated with bilateral mastectomy (bilateral: 53% vs. unilateral: 42.7%, p = 0.033) and decreased frequency of breast reconstruction (p < 0.001). SPMI was associated with more extended hospitalization (4 days vs. 2 days, p < 0.001) and increased risk of developing post-procedural infection and sepsis (OR 2.909). CONCLUSIONS: SPMI is associated with bilateral mastectomy, more extended hospitalization, and increased risk for post-procedural infection and sepsis - suggesting the need for increased use of standardized screening tools to identify SPMI in patients and inform perioperative management correctly.

Peritumoral edema in meningiomas: pathophysiology, predictors, and principles for treatment.

Meningiomas is a tumor of the meninges and is among the most common intracranial neoplasms in adults, accounting for over a third of all primary brain tumors in the United States. Meningiomas can be associated with peritumoral brain edema (PTBE) which if not managed appropriately can lead to poor clinical outcomes. In this review, we summarize the relevant pathophysiology, predictors, and principles for treatment of PTBE. The results of various case-reports and case-series have found that meningioma-associated PTBE have patterns in age, tumor size, and hormone receptor positivity. Our study describes how increased age, increased tumor size, tumor location in the middle fossa, and positive expression of hormone receptors, VEGF, and MMP-9 can all be predictors for worse clinical outcomes. We also characterize treatment options for PTBE such as glucocorticoids and VEGF inhibitors along with the ongoing clinical trials attempting to alleviate PTBE in meningioma cases. The trends summarized in this review can be used to better predict the behavior of meningioma-associated PTBE and establish prognosis models to identify at risk patients.

Socioeconomic predictors of cost and length of stay for erythroderma: a cross-sectional analysis of the national inpatient sample.

Erythroderma is an uncommon but serious dermatologic disorder that often requires hospitalization for diagnosis and treatment. However, little is known about predictors influencing cost and patient outcomes. The present study sought to characterize the sociodemographic factors that predict patient outcomes and hospital cost. Data were obtained from the 2016-2018 National Inpatient Sample (NIS) provided by the Healthcare Cost and Utilization Project from the Agency for Healthcare Research and Quality for patients of any age with a primary or secondary diagnosis of exfoliative dermatitis. Regression analyses were performed to find predictors for hospital costs and patient outcomes, represented by the length of stay (LOS). Univariate analysis of LOS revealed urban teaching hospitals were associated with prolonged LOS (p = 0.023). Univariate analysis of hospital cost yielded the following factors associated with increased hospital cost: Black and Asian patients (p = .045), urban teaching hospitals (p = .035), and northeast or south geographic location (p = .004). Multivariable regression analysis revealed prolonged LOS was associated with female sex (p = .043) and large bed capacity (p = .044) while shorter LOS was associated with increased age (p = .025); lower hospital costs were associated with private-owned hospitals -  (p = .025). In patients diagnosed with erythroderma, there appear to be racial, economic, and geographic disparities for patients that lead to greater hospital costs and longer LOS.