Predicting postoperative outcomes in lumbar spinal fusion: development of a machine learning model.

BACKGROUND CONTEXT: Degenerative lumbar spondylolisthesis (DLS) is a prevalent spinal disorder, often requiring surgical intervention. Accurately predicting surgical outcomes is crucial to guide clinical decision-making, but this is challenging due to the multifactorial nature of postoperative results. Traditional risk assessment tools have limitations, and with the advent of machine learning, there is potential to enhance the precision and comprehensiveness of preoperative evaluations. PURPOSE: We aimed to develop a machine-learning algorithm to predict surgical outcomes in patients with degenerative lumbar spondylolisthesis (DLS) undergoing spinal fusion surgery, only using preoperative data. STUDY DESIGN: Retrospective cross-sectional study. PATIENT SAMPLE: Patients with DLS undergoing lumbar spinal fusion surgery. OUTCOME MEASURES: This study aimed to predict the occurrence of lower back pain (LBP) ≥4 on the numeric analogue scale (NAS) 2 years after surgery. LBP was evaluated as the average pain patients experienced at rest in the week before questioning. NAS ranges from 0 to 10, 0 representing no pain and 10 representing the worst pain imaginable. METHODS: We conducted a retrospective analysis of prospectively enrolled patients who underwent spinal fusion surgery for degenerative lumbar spondylolistheses at our institution in the United States between January 2016 and December 2018. The initial patient characteristics to be included in the training of the model were chosen by clinical expertise and through a literature review and included demographic characteristics, comorbidities, and radiologic features. The data was split into a training and validation datasets using a 60/40 split. Four different machine learning models were trained, including the modern XGBoost model, logistic regression, random-forest, and support vector machine (SVM). The models were evaluated according to the area under the curve (AUC) of the receiver operating characteristics (ROC) curve. An AUC of 0.7 to 0.8 was considered fair, 0.8 to 0.9 good, and ≥ 0.9 excellent. Additionally, a calibration plot and the Brier score were calculated for each model. RESULTS: A total of 135 patients (66% female) were included. A total of 38 (28%) patients reported LBP ≥ 4 after 2 years, representing the positive class. The XGBoost model demonstrated the best performance in the validation set with an AUC of 0.81 (95% CI 0.67-0.95). The other machine learning models performed significantly worse: with an AUC of 0.52 (95% CI 0.37-0.68) for the SVM, 0.56 (95% CI 0.37-0.76) for the logistic regression and an AUC of 0.56 (95% CI 0.37-0.78) for the random forest. In the XGBoost model age, composition of the erector spinae, and severity of lumbar spinal stenosis as were identified as the most important features. CONCLUSIONS: This study represents a novel approach to predicting surgical outcomes in spinal fusion patients. The XGBoost demonstrated a better performance compared with classical models and highlighted the potential contributions of age and paraspinal musculature atrophy as significant factors. These findings have important implications for enhancing patient care through the identification of high-risk individuals and modifiable risk factors. As the incorporation of machine learning algorithms into clinical decision-making continues to gain traction in research and clinical practice, our insights reinforce this trajectory by showcasing the potential of these techniques in forecasting surgical results.

Predicting conversion of ambulatory ACDF patients to inpatient: a machine learning approach.

BACKGROUND CONTEXT: Machine learning is a powerful tool that has become increasingly important in the orthopedic field. Recently, several studies have reported that predictive models could provide new insights into patient risk factors and outcomes. Anterior cervical discectomy and fusion (ACDF) is a common operation that is performed as an outpatient procedure. However, some patients are required to convert to inpatient status and prolonged hospitalization due to their condition. Appropriate patient selection and identification of risk factors for conversion could provide benefits to patients and the use of medical resources. PURPOSE: This study aimed to develop a machine-learning algorithm to identify risk factors associated with unplanned conversion from outpatient to inpatient status for ACDF patients. STUDY DESIGN/SETTING: This is a machine-learning-based analysis using retrospectively collected data. PATIENT SAMPLE: Patients who underwent one- or two-level ACDF in an ambulatory setting at a single specialized orthopedic hospital between February 2016 to December 2021. OUTCOME MEASURES: Length of stay, conversion rates from ambulatory setting to inpatient. METHODS: Patients were divided into two groups based on length of stay: (1) Ambulatory (discharge within 24 hours) or Extended Stay (greater than 24 hours but fewer than 48 hours), and (2) Inpatient (greater than 48 hours). Factors included in the model were based on literature review and clinical expertise. Patient demographics, comorbidities, and intraoperative factors, such as surgery duration and time, were included. We compared the performance of different machine learning algorithms: Logistic Regression, Random Forest (RF), Support Vector Machine (SVM), and Extreme Gradient Boosting (XGBoost). We split the patient data into a training and validation dataset using a 70/30 split. The different models were trained in the training dataset using cross-validation. The performance was then tested in the unseen validation set. This step is important to detect overfitting. The performance was evaluated using the area under the curve (AUC) of the receiver operating characteristics analysis (ROC) as the primary outcome. An AUC of 0.7 was considered fair, 0.8 good, and 0.9 excellent, according to established cut-offs. RESULTS: A total of 581 patients (59% female) were available for analysis. Of those, 140 (24.1%) were converted to inpatient status. The median age was 51 (IQR 44-59), and the median BMI was 28 kg/m2 (IQR 24-32). The XGBoost model showed the best performance with an AUC of 0.79. The most important features were the length of the operation, followed by sex (based on biological attributes), age, and operation start time. The logistic regression model and the SVM showed worse results, with an AUC of 0.71 each. CONCLUSIONS: This study demonstrated a novel approach to predicting conversion to inpatient status in eligible patients for ambulatory surgery. The XGBoost model showed good predictive capabilities, superior to the older machine learning approaches. This model also revealed the importance of surgical duration time, BMI, and age as risk factors for patient conversion. A developing field of study is using machine learning in clinical decision-making. Our findings contribute to this field by demonstrating the feasibility and accuracy of such methods in predicting outcomes and identifying risk factors, although external and multi-center validation studies are needed.

Allografts: expanding the surgeon's armamentarium.

In Germany, bone allografts are widely used and their application in clinics has increased over the years. Successful use of allografts depends on many factors such as the procurement, processing, sterilization and the surgeon's surgical experience. Tissue banks have provided safe and sterile allografts for decades ranging from hard to soft tissue. Allografts are obtained from various tissues such as bone, tendon, amniotic membrane, meniscus and skin. An advantage of allografts is their wide applicability that has never been limited by indication restrictions thus providing a huge benefit for surgeon's. The use of the correct allograft in different indications is extremely important. Thereby surgeons have access to various allograft forms such as mineralized, demineralized, freeze-dried, paste, powder, chips strips and putty. The vast options of allografts allow surgeon's to use allografts in indications they deem fit. Currently, the application of allografts is at the discretion of the expert surgeon. However, regulations are often changed locally or internationally and may impact/limit allograft use to certain indications. Here, we report the different indications where our peracetic acid (PAA) sterilised bone allografts were used as well as general literature on bone allograft use in other indications.

Gas within the Intervertebral Disc Does Not Rule Out Spinal Infection-A Case Series of 135 Patients with Spontaneous Spondylodiscitis.

Gas in the intervertebral disc is mainly associated with degenerative disc diseases and experts generally assume that it is unlikely in spinal infection. However, large-scale studies supporting this notion are lacking, which is why our study's aim was to analyze the prevalence of and factors associated with the occurrence of gas in patients with spontaneous spondylodiscitis. Patients presenting with spontaneous spondylodiscitis from 2006 to 2020 were included retrospectively. Exclusion criteria were previous interventions in the same spinal segment and missing imaging data. Clinical data were retrieved from electronic medical reports. Computed tomography (CT) scans were evaluated for the presence of intervertebral gas. Causative pathogens were identified from CT-guided biopsy, open biopsy, intraoperative tissue samples, and/or blood cultures. 135 patients with a mean age of 66.0 ± 13.7 years were included. In 93 patients (68.9%), a causative pathogen was found. Intervertebral gas was found in 31 patients (23.0%) in total and in 19 patients (20.4%) with positive microbiology. Patients with gas presented with significantly higher body temperatures (37.2 ± 1.1 vs. 36.8 ± 0.7 °C, p = 0.044) and CRP levels (134.2 ± 127.1 vs. 89.8 ± 97.3 mg/L, p = 0.040) on admission. As a considerable number of patients with spondylodiscitis showed intervertebral gas formation, the detection of intervertebral gas is not suited to ruling out spondylodiscitis but must be interpreted in the context of other imaging and clinical findings, especially in elderly patients.

Georg schmorl prize of the German spine society (DWG) 2021: Spinal Instability Spondylodiscitis Score (SISS)-a novel classification system for spinal instability in spontaneous spondylodiscitis.

PURPOSE: Even though spinal infections are associated with high mortality and morbidity, their therapy remains challenging due to a lack of established classification systems and widely accepted guidelines for surgical treatment. This study's aim therefore was to propose a comprehensive classification system for spinal instability based on the Spinal Instability Neoplastic Score (SINS) aiding spine surgeons in choosing optimal treatment for spontaneous spondylodiscitis. METHODS: Patients who were treated for spontaneous spondylodiscitis and received computed tomography (CT) imaging were included retrospectively. The Spinal Instability Spondylodiscitis Score (SISS) was developed by expert consensus. SINS and SISS were scored in CT-images by four readers. Intraclass correlation coefficients (ICCs) and Fleiss' Kappa were calculated to determine interrater reliabilities. Predictive validity was analyzed by cross-tabulation analysis. RESULTS: A total of 127 patients were included, 94 (74.0%) of which were treated surgically. Mean SINS was 8.3 ± 3.2, mean SISS 8.1 ± 2.4. ICCs were 0.961 (95%-CI: 0.949-0.971) for total SINS and 0.960 (95%-CI: 0.946-0.970) for total SISS. SINS yielded false positive and negative rates of 12.5% and 67.6%, SISS of 15.2% and 40.0%, respectively. CONCLUSION: We show high reliability and validity of the newly developed SISS in detecting unstable spinal lesions in spontaneous spondylodiscitis. Therefore, we recommend its use in evaluating treatment choices based on spinal biomechanics. It is, however, important to note that stability is merely one of multiple components in making surgical treatment decisions.

Recommendations for Patients with High Return to Sports Expectations after TKA Remain Controversial.

(1) Background: Improved surgical techniques and implants in total knee arthroplasty (TKA) have led to broader indications for surgical interventions of osteoarthritis of the knee. There is a growing young and active patient subgroup with high return to sports (RTS) expectations after TKA. The current lack of evidence regarding RTS capacity in this patient cohort, requires the consolidation of experts' opinions and experiences to address the special needs among these patients. The aim of this study was to assess current expert opinions in regard to preoperative patient assessment, surgical technique and decision-making and patient counseling for these patients. (2) Methods: We performed a survey among surgeons specialized in arthroplasty with a questionnaire designed to assess current recommendations, surgical techniques, and implant preferences as well as patient counseling in patients with high expectations for RTS after TKA. (3) Results: The majority of surgeons are in favor of return to low-impact sports after TKA within 3 to 6 months. Some even recommend return to high-impact sports. Despite improvement of surgical techniques and implants, we observed no clear preference for a single surgical technique or implant specification in active patients. (4) Conclusions: Current evidence for sports-associated complications after TKA is scarce. Despite a growing array of surgical techniques and implants, the available literature is still controversial with no single surgical technique or TKA design distinguishing itself clearly from others. Surgeons' recommendations are mostly based on their experience and training. Nonetheless, we observed growing faith in modern implants with some surgeons even recommending high-impact sports after TKA.

BRD4 prevents the accumulation of R-loops and protects against transcription-replication collision events and DNA damage.

Proper chromatin function and maintenance of genomic stability depends on spatiotemporal coordination between the transcription and replication machinery. Loss of this coordination can lead to DNA damage from increased transcription-replication collision events. We report that deregulated transcription following BRD4 loss in cancer cells leads to the accumulation of RNA:DNA hybrids (R-loops) and collisions with the replication machinery causing replication stress and DNA damage. Whole genome BRD4 and γH2AX ChIP-Seq with R-loop IP qPCR reveals that BRD4 inhibition leads to accumulation of R-loops and DNA damage at a subset of known BDR4, JMJD6, and CHD4 co-regulated genes. Interference with BRD4 function causes transcriptional downregulation of the DNA damage response protein TopBP1, resulting in failure to activate the ATR-Chk1 pathway despite increased replication stress, leading to apoptotic cell death in S-phase and mitotic catastrophe. These findings demonstrate that inhibition of BRD4 induces transcription-replication conflicts, DNA damage, and cell death in oncogenic cells.

Enhanced efficacy of combined temozolomide and bromodomain inhibitor therapy for gliomas using targeted nanoparticles.

Effective treatment for glioblastoma (GBM) is limited by the presence of the blood-brain barrier (BBB) and rapid resistance to single agent therapies. To address these issues, we developed a transferrin-functionalized nanoparticle (Tf-NP) that can deliver dual combination therapies. Using intravital imaging, we show the ability of Tf-NPs to traverse intact BBB in mice as well as achieve direct tumor binding in two intracranial orthotopic models of GBM. Treatment of tumor-bearing mice with Tf-NPs loaded with temozolomide and the bromodomain inhibitor JQ1 leads to increased DNA damage and apoptosis that correlates with a 1.5- to 2-fold decrease in tumor burden and corresponding increase in survival compared to equivalent free-drug dosing. Immunocompetent mice treated with Tf-NP-loaded drugs also show protection from the effects of systemic drug toxicity, demonstrating the preclinical potential of this nanoscale platform to deliver novel combination therapies to gliomas and other central nervous system tumors.

Identifying molecular markers for the sensitive detection of residual atypical teratoid rhabdoid tumor cells.

Atypical teratoid rhabdoid tumor (AT/RT), a rare and highly malignant tumor entity of the central nervous system that presents in early childhood, has a poor prognosis. AT/RTs are characterized by biallelic inactivating mutations of the gene SMARCB1 in 98% of patients; these mutations may serve as molecular markers for residual tumor cell detection in liquid biopsies. We developed a marker-specific method to detect residual AT/RT cells. Seven of 150 patient samples were selected, each with a histological and genetically ascertained diagnosis of AT/RT. Tumor tissue was either formalin fixed or fresh frozen. DNA was extracted from the patients' peripheral blood leukocytes (PBL) and cerebrospinal fluid (CSF). Multiplex ligation-dependent probe amplification, DNA sequencing, and fluorescence in situ hybridization were used to characterize the tumors' mutations. Residual tumor cell detection used mutation-specific primers and real-time PCR. The detection limit for the residual tumor cell search was 1-18%, depending on the quality of the template provided. The residual tumor cell search in PBL and CSF was negative for all seven patients. The SMARCB1 region of chromosome 22 is prone to DNA double-strand breaks. The individual breakpoints and breakpoint-specific PCR offer the option to detect minimal residual tumor cells in CSF or blood. Even if we did not detect minimal residual tumor cells in the investigated material, proof of principle for this method was confirmed.