BACKGROUND: Incidental durotomy is a common complication of posterior lumbar spine surgery; however, effective and durable methods for primary repair remain elusive. Multiple existing techniques have previously been reported and extensively described, including sutured repair and the use of nonpenetrating titanium clips. The use of cranial aneurysm clips for primary repair of lumbar durotomy serves as a safe and effective alternative to obtain watertight closure of a dural tear. METHODS: We performed a retrospective review of patients at a single institution who underwent primary repair of an incidental lumbar durotomy with the use of an aneurysm clip during open posterior lumbar surgery between 2012 and 2023. Patient demographics, operative details, and postoperative metrics were collected and examined to evaluate the safety and efficacy of the novel technique. RESULTS: A total of 51 patients were included for analysis. Four patients underwent durotomy repair with an aneurysm clip alone, 27 patients were repaired with an aneurysm clip and fibrin glue, and 20 patients underwent repair with an aneurysm clip, fibrin glue, and a collagen dural substitute. Three patients (5.9%) reported headaches: 2 (3.9%) with pseudomeningocele and 1 (2%) with wound leakage. Two patients (3.9%) had treatment failure with a return to the operating room for repair of a cerebrospinal fluid leak. CONCLUSIONS: To the best of our knowledge, we report the largest series of patients undergoing primary repair of incidental durotomy with the use of an aneurysm clip. Use of an aneurysm clip is noted to be a safe, quick, and effective method of primary repair compared with existing repair techniques such as sutured repair or nonpenetrating titanium clips.
INTRODUCTION: Expandable devices such as interbody cages, vertebral body reconstruction cages, and intravertebral body expansion devices are frequently utilized in spine surgery. Since the introduction of expandable implants in the early 2000s, the variety of mechanisms that drive expansion and implant materials have steadily increased. By examining expandable devices that have achieved commercial success and exploring emerging innovations, we aim to offer an in-depth evaluation of the different types of expandable cages used in spine surgery and the underlying mechanisms that drive their functionality. AREAS COVERED: We performed a review of expandable spinal implants and devices by querying the National Library of Medicine MEDLINE database and Google Patents database from 1933 to 2024. Five major types of mechanical jacks that drive expansion were identified: scissor, pneumatic, screw, ratchet, and insertion-expansion. EXPERT OPINION: We identified a trend of screw jack mechanism being the predominant machinery in vertebral body reconstruction cages and scissor jack mechanism predominating in interbody cages. Pneumatic jacks were most commonly found in kyphoplasty devices. Critically reviewing the mechanisms of expansion and identifying trends among effective and successful cages allows both surgeons and medical device companies to properly identify future areas of development.
Tribology, an interdisciplinary field concerned with the science of interactions between surfaces in contact and their relative motion, plays a well-established role in the design of orthopedic implants, such as knee and hip replacements. However, its applications in spine surgery have received comparatively less attention in the literature. Understanding tribology is pivotal in elucidating the intricate interactions between metal, polymer, and ceramic components, as well as their interplay with the native human bone. Numerous studies have demonstrated that optimizing tribological factors is key to enhancing the longevity of joints and implants while simultaneously reducing complications and the need for revision surgeries in both arthroplasty and spinal fusion procedures. With an ever-growing and diverse array of spinal implant devices hitting the market for static and dynamic stabilization of the spine, it is important to consider how each of these devices optimizes these parameters and what factors may be inadequately addressed by currently available technology and methods. In this comprehensive review, the authors' objectives were twofold: 1) delineate the unique challenges encountered in spine surgery that could be addressed through optimization of tribological parameters; and 2) summarize current innovations and products within spine surgery that look to optimize tribological parameters and highlight new avenues for implant design and research.
Hypertension is a widely prevalent disease and uncontrolled hypertension predisposes affected individuals to severe adverse effects. Though the importance of controlling hypertension is clear, the multitude of therapeutic regimens and patient factors that affect the success of blood pressure control makes it difficult to predict the likelihood to predict whether a patient's blood pressure will be controlled. This project endeavors to investigate whether machine learning can accurately predict the control of a patient's hypertension within 12 months of a clinical encounter. To build the machine learning model, a retrospective review of the electronic medical records of 350,008 patients 18 years of age and older between January 1, 2015 and June 1, 2022 was performed to form model training and testing cohorts. The data included in the model included medication combinations, patient laboratory values, vital sign measurements, comorbidities, healthcare encounters, and demographic information. The mean age of the patient population was 65.6 years with 161,283 (46.1%) men and 275,001 (78.6%) white. A sliding time window of data was used to both prohibit data leakage from training sets to test sets and to maximize model performance. This sliding window resulted in using the study data to create 287 predictive models each using 2 years of training data and one week of testing data for a total study duration of five and a half years. Model performance was combined across all models. The primary outcome, prediction of blood pressure control within 12 months demonstrated an area under the curve of 0.76 (95% confidence interval; 0.75-0.76), sensitivity of 61.52% (61.0-62.03%), specificity of 75.69% (75.25-76.13%), positive predictive value of 67.75% (67.51-67.99%), and negative predictive value of 70.49% (70.32-70.66%). An AUC of 0.756 is considered to be moderately good for machine learning models. While the accuracy of this model is promising, it is impossible to state with certainty the clinical relevancy of any clinical support ML model without deploying it in a clinical setting and studying its impact on health outcomes. By also incorporating uncertainty analysis for every prediction, the authors believe that this approach offers the best-known solution to predicting hypertension control and that machine learning may be able to improve the accuracy of hypertension control predictions using patient information already available in the electronic health record. This method can serve as a foundation with further research to strengthen the model accuracy and to help determine clinical relevance.
Hypertension , Machine Learning , Male , Humans , Adolescent , Adult , Aged , Female , Retrospective Studies , Predictive Value of Tests , Comorbidity , Hypertension/diagnosis , Hypertension/drug therapy
BACKGROUND: Despite the expanding role of machine learning (ML) in health care and patient expectations for clinicians to understand ML-based tools, few for-credit curricula exist specifically for neurosurgical trainees to learn basic principles and implications of ML for medical research and clinical practice. We implemented a novel, remotely delivered curriculum designed to develop literacy in ML for neurosurgical trainees. METHODS: A 4-week pilot medical elective was designed specifically for trainees to build literacy in basic ML concepts. Qualitative feedback from interested and enrolled students was collected to assess students' and trainees' reactions, learning, and future application of course content. RESULTS: Despite 15 interested learners, only 3 medical students and 1 neurosurgical resident completed the course. Enrollment included students and trainees from 3 different institutions. All learners who completed the course found the lectures relevant to their future practice as clinicians and researchers and reported improved confidence in applying and understanding published literature applying ML techniques in health care. Barriers to ample enrollment and retention (e.g., balancing clinical responsibilities) were identified. CONCLUSIONS: This pilot elective demonstrated the interest, value, and feasibility of a remote elective to establish ML literacy; however, feedback to increase accessibility and flexibility of the course encouraged our team to implement changes. Future elective iterations will have a semiannual, 2-week format, splitting lectures more clearly between theory (the method and its value) and application (coding instructions) and will make lectures open-source prerequisites to allow tailoring of student learning to their planned application of these methods in their practice and research.
Education, Medical, Undergraduate , Students, Medical , Humans , Curriculum , Delivery of Health Care , Education, Medical, Undergraduate/methods , Feedback
OBJECTIVE: In clinical spine surgery research, manually reviewing surgical forms to categorize patients by their surgical characteristics is a crucial yet time-consuming task. Natural language processing (NLP) is a machine learning tool used to adaptively parse and categorize important features from text. These systems function by training on a large, labeled data set in which feature importance is learned prior to encountering a previously unseen data set. The authors aimed to design an NLP classifier for surgical information that can review consent forms and automatically classify patients by the surgical procedure performed. METHODS: Thirteen thousand two hundred sixty-eight patients who underwent 15,227 surgeries from January 1, 2012, to December 31, 2022, at a single institution were initially considered for inclusion. From these surgeries, 12,239 consent forms were classified based on the Current Procedural Terminology (CPT) code, categorizing them into 7 of the most frequently performed spine surgeries at this institution. This labeled data set was split 80%/20% into train and test subsets, respectively. The NLP classifier was then trained and the results demonstrated its performance on the test data set using CPT codes to determine accuracy. RESULTS: This NLP surgical classifier had an overall weighted accuracy rate of 91% for sorting consents into correct surgical categories. Anterior cervical discectomy and fusion had the highest positive predictive value (PPV; 96.8%), whereas lumbar microdiscectomy had the lowest PPV in the testing data (85.0%). Sensitivity was highest for lumbar laminectomy and fusion (96.7%) and lowest for the least common operation, cervical posterior foraminotomy (58.3%). Negative predictive value and specificity were > 95% for all surgical categories. CONCLUSIONS: Utilizing NLP for text classification drastically improves the efficiency of classifying surgical procedures for research purposes. The ability to quickly classify surgical data can be significantly beneficial to institutions without a large database or substantial data review capabilities, as well as for trainees to track surgical experience, or practicing surgeons to evaluate and analyze their surgical volume. Additionally, the capability to quickly and accurately recognize the type of surgery will facilitate the extraction of new insights from the correlations between surgical interventions and patient outcomes. As the database of surgical information grows from this institution and others in spine surgery, the accuracy, usability, and applications of this model will continue to increase.
Consent Forms , Natural Language Processing , Humans , Machine Learning , Laminectomy , Diskectomy
Decision-making in spine surgery is complex due to patients' heterogeneity and complexity of spinal pathologies and the various surgical options applied to a given pathology. Artificial intelligence/machine learning algorithms provide an opportunity to improve patient selection, surgical planning, and outcomes. The purpose of this article is to present the experience and applications of in spine surgery at 2 large academic health care systems.
BACKGROUND CONTEXT: Identifying optimal stratification techniques for subgrouping patients with low back pain (LBP) into treatment groups for the purpose of identifying optimal management and improving clinical outcomes is an important area for further research. PURPOSE: Our study aimed to compare performance of the STarT Back Tool (SBT) and 3 stratification techniques involving PROMIS domain scores for use in patients presenting to a spine clinic for chronic LBP. STUDY DESIGN: Retrospective cohort study. PATIENT SAMPLE: Adult patients with chronic LBP seen in a spine center between November 14, 2018 and May 14, 2019 who completed patient-reported outcomes (PROs) as part of routine care, and were followed up with completed PROs 1 year later. OUTCOME MEASURES: Four stratification techniques, including SBT, and 3 PROMIS-based techniques: the NIH Task Force recommended Impact Stratification Score (ISS), symptom clusters based on latent class analysis (LCA), and SPADE symptom clusters. METHODS: The 4 stratification techniques were compared according to criterion validity, construct validity, and prognostic utility. For criterion validity, overlap in characterization of mild, moderate, and severe subgroups were compared to SBT, which was considered the gold standard, using quadratic weighted kappa statistic. Construct validity compared techniques' ability to differentiate across disability groups defined by modified Oswestry LBP Disability Questionnaire (MDQ), median days in the past month unable to complete activities of daily living (ADLs), and worker's compensation using standardized mean differences (SMD). Prognostic utility was compared based on the techniques' ability to predict long-term improvement in outcomes, defined as improvement in global health and MDQ at 1-year. RESULTS: There were 2,246 adult patients with chronic LBP included in our study (mean age 61.0 [SD 14.0], 55.0% female, 83.4% white). All stratification techniques resulted in roughly a third of patients grouped into mild, moderate, and severe categories, with ISS and LCA demonstrating substantial agreement with SBT, while SPADE had moderate agreement. Construct validity was met for all techniques, with large effects demonstrated between mild and severe categories for differentiating MDQ, ADLs, and worker's compensation disability groups (SMD range 0.57-2.48). All stratification techniques demonstrated ability to detect improvement by 1-year, with severe groups experiencing the greatest improvement in multivariable logistic regression models. CONCLUSIONS: All 4 stratification techniques demonstrated validity and prognostic utility for subgrouping patients with chronic LBP based on risk of long-term disability. ISS and LCA symptom clusters may be the optimal methods given the improved feasibility of including only a few relevant PROMIS domains. Future research should investigate multidisciplinary treatment approaches to target mild, moderate, and severe patients based on these techniques.
Chronic Pain , Low Back Pain , Adult , Humans , Female , Middle Aged , Male , Low Back Pain/diagnosis , Low Back Pain/therapy , Retrospective Studies , Pain Measurement/methods , Activities of Daily Living , Syndrome , Disability Evaluation , Chronic Pain/diagnosis
Background: Prior work by our group developed a stratification tool based on four PROMIS domains for patients with low back pain (LBP). Our study aimed to evaluate the ability of our previously developed symptom classes to predict long-term outcomes, and determine whether there were differential treatment effects by intervention. Methods: This was a retrospective cohort study of adult patients with LBP seen in spine clinics in a large health system between November 14, 2018 and May 14, 2019 who completed patient-reported outcomes as part of routine care at baseline and again at 12-months follow-up. Latent class analysis identified symptom classes based on PROMIS domain scores (physical function, pain interference, social role satisfaction, and fatigue) that were ≥1 standard deviation worse (meaningfully worse) than the general population. The ability of the profiles to predict long-term outcomes at 12-months was evaluated through multivariable models. Differences in outcomes by subsequent treatments (physical therapy, specialist visits, injections, and surgery) were investigated. Results: There were 3,236 adult patients (average age 61.1 ± 14.2, 55.4% female) included in the study with three distinct classes identified: mild symptoms (n = 986, 30.5%), mixed (n = 798, 24.7%) with poor scores on physical function and pain interference but better scores on other domains, and significant symptoms (n = 1,452, 44.9%). The classes were significantly associated with long-term outcomes, with patients with significant symptoms improving the most across all domains. Utilization differed across classes, with the mixed symptom class receiving more PT and injections and significant symptom class receiving more surgeries and specialist visits. Conclusions: Patients with LBP have distinct clinical symptom classes which could be utilized to stratify patients into groups based on risk of future disability. These symptom classes can also be used to provide estimates of the effectiveness of different interventions, further increasing the clinical utility of these classes in standard care.
OBJECTIVE: Neurosurgery program websites serve as a valuable resource for applicants. However, each website exists in isolation, and it can be difficult to understand the general trends in U.S. neurosurgery resident demographics. In the present study, we collected data from program websites and analyzed the trends in the demographics of the current U.S. neurosurgery residents. METHODS: We used a program list obtained from the American Association of Medical Colleges Electronic Residency Application System to extract data from the current resident complement listed in each program's website, including program, year in program, medical school, sex (male vs. female), graduate and/or PhD degrees, and assessed the trends during 7 years of resident data using linear regression. RESULTS: We identified 116 neurosurgery residency programs in the United States, with 111 providing information on their current resident complement, yielding a dataset of 1599 residents. Of these 1599 residents, 348 (22%) were female, 301 (19%) had a graduate degree in addition to an MD or DO degree, 151 (9.4%) had a PhD degree, 300 (19%) had matched at the program affiliated with their medical school, and 121 (7.6%) had graduated from a foreign medical school. The proportion of matriculating female residents had increased an average of 2.1% annually (95% confidence interval, 0.6%-3.7%) from 2015 to 2021. The other demographic data had not changed significantly during the same period. CONCLUSIONS: In addition to summarizing the current resident demographics, our analysis identified a significant increase in the proportion of female residents between 2015 (15.1%) and 2021 (25.6%). This publicly available dataset should enable additional analyses of the evolution of neurosurgery resident demographics.
Internship and Residency , Neurosurgery , Male , Female , Humans , United States , Neurosurgery/education , Neurosurgeons , Schools, Medical
BACKGROUND: The increase in use of targeted systemic therapies in cancer treatments has catalyzed the importance of identifying patient- and tumor-specific somatic mutations, especially regarding metastatic disease. Mutations found to be most prevalent in patients with metastatic breast cancer include TP53, PI3K, and CDH1. OBJECTIVE: To determine the incidence of somatic mutations in patients with metastatic breast cancer to the spine (MBCS). To determine if a difference exists in overall survival (OS), progression-free survival, and progression of motor symptoms between patients who do or do not undergo targeted systemic therapy after treatment for MBCS. METHODS: This is a retrospective study of patients with MBCS. Review of gene sequencing reports was conducted to calculate the prevalence of various somatic gene mutations within this population. Those patients who then underwent treatment (surgery/radiation) for their diagnosis of MBCS between 2010 and 2020 were subcategorized. The use of targeted systemic therapy in the post-treatment period was identified, and post-treatment OS, progression-free survival, and progression of motor deficits were calculated for this subpopulation. RESULTS: A total of 131 patients were included in the final analysis with 56% of patients found to have a PI3K mutation. Patients who received targeted systemic therapies were found to have a significantly longer OS compared with those who did not receive targeted systemic therapies. CONCLUSION: The results of this study demonstrate that there is an increased prevalence of PI3K mutations in patients with MBCS and there are a significant survival benefit and delay in progression of motor symptoms associated with using targeted systemic therapies for adjuvant treatment.
Breast Neoplasms , Humans , Female , Breast Neoplasms/genetics , Breast Neoplasms/therapy , Breast Neoplasms/pathology , Retrospective Studies , Incidence , Mutation/genetics , Phosphatidylinositol 3-Kinases/genetics
OBJECTIVE: The US-based Spinal Laminectomy versus Instrumented Pedicle Screw (SLIP) trial reported improvement in disability following laminectomy with fusion versus laminectomy alone for patients with lumbar spondylolisthesis. Despite using similar methods, a concurrent Swedish trial investigating the same question did not reach the same conclusion. The authors performed a simulation-based analysis to elucidate potential causes of these divergent results. METHODS: The mean and standard deviation of the preoperative and 2-year postoperative Oswestry Disability Index (ODI) scores for each study group (laminectomy with fusion and laminectomy alone) were collected from the spondylolisthesis stratum of the Swedish trial and used to create a MATLAB simulator using linear transformations to predict postoperative ODI distributions. Applying this simulator to both varied and published preoperative ODI distributions from the SLIP trial, the authors simulated the results of the US-based trial using treatment effects from the Swedish study and compared simulated US results to those published in the SLIP trial. RESULTS: Simulated US results showed that as preoperative disability increased, the difference in postoperative ODI scores grew between treatment groups and increasingly favored laminectomy alone (p < 0.0001). In 100 simulations of a similarly sized US trial, the average mean change in ODI scores postoperatively was significantly higher than was published for laminectomy alone in the SLIP trial (-21.3 vs -17.9), whereas it was significantly lower than published for fusion (-16.9 vs -26.3). CONCLUSIONS: The expected benefit of surgical treatments for spondylolisthesis varied according to preoperative disability. Adapting Swedish-estimated treatment effects to the US context mildly overapproximated the improvement in postoperative disability scores for laminectomy, but more severely underapproximated the improvement reported for laminectomy and fusion in the SLIP trial. The observed heterogeneity between these studies is influenced more by patient response to fusion than response to laminectomy. This analysis paves the way for future studies on the impact of preoperative treatment group heterogeneity, differences in surgical methods, and empirical design on reported clinical benefits. Although bayesian reanalysis of published randomized controlled trial data is susceptible to biases that typically limit post hoc analyses, the authors' method offers a simple and cost-effective approach to improve the understanding of published clinical trial results and their implications for future studies.
Objective Radiofrequency ablation is widely utilized in otorhinolaryngology. It is used for ablation, coagulation and resection, and hemostasis. It causes tissue destruction through a chemical interaction of ions. The potential benefit is to cause less thermal injury to surrounding tissues compared with other coagulative tools. In this article, we present novel uses of radiofrequency ablation in endoscopic endonasal, and transcranial surgery. This is the first study to describe its use in transcranial cases. Design, Setting, and Participants This is a retrospective study of patients between 2016 and 2018 who underwent either endoscopic endonasal or transcranial surgery where radiofrequency ablation was used. Main Outcome Measures We looked at indication for usage, blood loss, postoperative imaging to identify any stroke or edema, and clinical outcomes of these patients. Results The radiofrequency device was used in eight endoscopic endonasal cases and four craniotomies. Four cases were for encephalocele repair and eight were for various intracranial pathologies. In endonasal encephalocele repair, the radiofrequency ablation helped in shrinking the herniated brain while minimizing thermal injury to the surrounding tissue. In tumors resection, the combination of ablation and coagulation effect was particularly effective for highly vascularized tumors. There were no vascular or major neurologic injuries. Postoperative periencephalocele edema was noted in one case. Conclusion Initial experience with radiofrequency ablation showed that it was a safe technique to use in both endonasal skull-base and transcranial procedures. It seemed particularly useful for highly vascularized tumors but a greater experience is needed to further clarify its role in these procedures.
BACKGROUND CONTEXT: Improving prognostic stratification for patients with low back pain (LBP) outside of a primary care setting has been identified as an important area for further research. PURPOSE: Our study aimed to identify clinical symptom classes of patients presenting to a spine clinic based on 4 Patient Reported Outcome Measurement Information System (PROMIS) domains and evaluate demographic and clinical differences across classes. STUDY DESIGN: An observational cross-sectional study of patients seen in spine centers at a large health system. PATIENT SAMPLE: Adult patients with LBP seen in a spine center between November 14, 2018 and May 14, 2019 who completed patient-reported outcomes as part of routine care. OUTCOME MEASURES: PROMIS physical function, pain interference, satisfaction with social roles and activities, and fatigue. METHODS: Latent class analysis identified symptom classes based on PROMIS domain scores ≥1 standard deviation worse (meaningfully worse) than the general population. A multivariable multinomial logistic regression model was constructed to evaluate differences in symptom classes based on demographics and socioeconomic characteristics. Lastly, the ability of the profiles to discriminate across levels of disability, based on the modified Oswestry Disability Questionnaire (ODI), was evaluated. RESULTS: There were 7,144 adult patients included in the study who visited spine clinics for a primary complaint of LBP and completed all 4 PROMIS domains (age 58.7±15.9, 54% female). Three distinct classes were identified. Class 1 ("Significant Symptoms," n=3238) had PROMIS scores that were meaningfully worse than the population average for all domains. Class 2 ("Mixed Symptoms," n=1366) had meaningfully worse scores on physical function and pain interference but average scores on other domains. Class 3 ("Mild Symptoms," n=2540) had average scores across all domains. Compared to patients in Class 3, those in Class 2 were more likely older, and those in Classes 1 and 2 were more likely to be divorced, have lower household income, and no employment. Level of disability was significantly different across each class (average (SD) ODI for Classes 1-3: 53.4 (14.3), 39.9 (12.5), 22.9 (12.1), p<.01). CONCLUSIONS: Patients presenting to specialty clinics for LBP demonstrate distinct clinical symptom classes which could be utilized to inform specific symptom-based treatment. Future research should evaluate the ability of these classes to predict long-term disability.
Low Back Pain , Patient Reported Outcome Measures , Adult , Aged , Cross-Sectional Studies , Female , Humans , Low Back Pain/diagnosis , Male , Middle Aged , Self Report , Spine
OBJECTIVE: In a healthcare landscape in which costs increasingly matter, the authors sought to distinguish among the clinical and nonclinical drivers of patient length of stay (LOS) in the hospital following elective lumbar laminectomy-a common spinal surgery that may be reimbursed using bundled payments-and to understand their relationships with patient outcomes and costs. METHODS: Patients ≥ 18 years of age undergoing laminectomy surgery for degenerative lumbar spinal stenosis within the Cleveland Clinic health system between March 1, 2016, and February 1, 2019, were included in this analysis. Generalized linear modeling was used to assess the relationships between the day of surgery, patient discharge disposition, and hospital LOS, while adjusting for underlying patient health risks and other nonclinical factors, including the hospital surgery site and health insurance. RESULTS: A total of 1359 eligible patients were included in the authors' analysis. The mean LOS ranged between 2.01 and 2.47 days for Monday and Friday cases, respectively. The LOS was also notably longer for patients who were ultimately discharged to a skilled nursing facility (SNF) or rehabilitation center. A prolonged LOS occurring later in the week was not associated with greater underlying health risks, yet it nevertheless resulted in greater costs of care: the average total surgical costs for lumbar laminectomy were 20% greater for Friday cases than for Monday cases, and 24% greater for late-week cases than for early-week cases ultimately transferred to SNFs or rehabilitation centers. A Poisson generalized linear model fit the data best and showed that the comorbidity burden, surgery at a tertiary care center versus a community hospital, and the incidence of any postoperative complication were associated with significantly longer hospital stays. Discharge to home healthcare, SNFs, or rehabilitation centers, and late-week surgery were significant nonclinical predictors of LOS prolongation, even after adjusting for underlying patient health risks and insurance, with LOSs that were, for instance, 1.55 and 1.61 times longer for patients undergoing their procedure on Thursday and Friday compared to Monday, respectively. CONCLUSIONS: Late-week surgeries are associated with a prolonged LOS, particularly when discharge is to an SNF or rehabilitation center. These findings point to opportunities to lower costs and improve outcomes associated with elective surgical care. Interventions to optimize surgical scheduling and perioperative care coordination could help reduce prolonged LOSs, lower costs, and, ultimately, give service line management personnel greater flexibility over how to use existing resources as they remain ahead of healthcare reforms.
STUDY DESIGN: Narrative review. OBJECTIVES: Artificial intelligence (AI) and machine learning (ML) have emerged as disruptive technologies with the potential to drastically affect clinical decision making in spine surgery. AI can enhance the delivery of spine care in several arenas: (1) preoperative patient workup, patient selection, and outcome prediction; (2) quality and reproducibility of spine research; (3) perioperative surgical assistance and data tracking optimization; and (4) intraoperative surgical performance. The purpose of this narrative review is to concisely assemble, analyze, and discuss current trends and applications of AI and ML in conventional and robotic-assisted spine surgery. METHODS: We conducted a comprehensive PubMed search of peer-reviewed articles that were published between 2006 and 2019 examining AI, ML, and robotics in spine surgery. Key findings were then compiled and summarized in this review. RESULTS: The majority of the published AI literature in spine surgery has focused on predictive analytics and supervised image recognition for radiographic diagnosis. Several investigators have studied the use of AI/ML in the perioperative setting in small patient cohorts; pivotal trials are still pending. CONCLUSIONS: Artificial intelligence has tremendous potential in revolutionizing comprehensive spine care. Evidence-based, predictive analytics can help surgeons improve preoperative patient selection, surgical indications, and individualized postoperative care. Robotic-assisted surgery, while still in early stages of development, has the potential to reduce surgeon fatigue and improve technical precision.
OBJECTIVE: Spinal cord astrocytoma (SCA) is a rare tumor whose epidemiology has not been well defined. The authors utilized the Central Brain Tumor Registry of the United States (CBTRUS) to provide comprehensive up-to-date epidemiological data for this disease. METHODS: The CBTRUS was queried for SCAs on ICD-O-3 (International Classification of Diseases for Oncology, 3rd edition) histological and topographical codes. The age-adjusted incidence (AAI) per 100,000 persons was calculated and stratified by race, sex, age, and ethnicity. Joinpoint was used to calculate the annual percentage change (APC) in incidence. RESULTS: Two thousand nine hundred sixty-nine SCAs were diagnosed in the US between 1995 and 2016, resulting in an average of approximately 136 SCAs annually. The overall AAI was 0.047 (95% CI 0.045-0.049), and there was a statistically significant increase from 0.051 in 1995 to 0.043 in 2016. The peak incidence of 0.064 (95% CI 0.060-0.067) was found in the 0- to 19-year age group. The incidence in males was 0.053 (95% CI 0.050-0.055), which was significantly greater than the incidence in females (0.041, 95% CI 0.039-0.044). SCA incidence was significantly lower both in patients of Asian/Pacific Islander race (AAI = 0.034, 95% CI 0.028-0.042, p = 0.00015) and in patients of Hispanic ethnicity (AAI = 0.035, 95% CI 0.031-0.039, p < 0.001). The incidence of WHO grade I SCAs was significantly higher than those of WHO grade II, III, or IV SCAs (p < 0.001). CONCLUSIONS: The overall AAI of SCA from 1995 to 2016 was 0.047 per 100,000. The incidence peaked early in life for both sexes, reached a nadir between 20 and 34 years of age for males and between 35 and 44 years of age for females, and then slowly increased throughout adulthood, with a greater incidence in males. Pilocytic astrocytomas were the most common SCA in the study cohort. This study presents the most comprehensive epidemiological study of SCA incidence in the US to date.
