Interpretable Machine Learning Models Based on SHapley Additive exPlanations for Predicting the Risk of Cerebrospinal Fluid Leakage in Lumbar Fusion Surgery.

STUDY DESIGN: Retrospective study. OBJECTIVES: The objective of this investigation was to formulate and internally verify a customized machine learning (ML) framework for forecasting cerebrospinal fluid leakage (CSFL) in lumbar fusion surgery. This was accomplished by integrating imaging parameters and employing the SHapley Additive exPlanation (SHAP) technique to elucidate the interpretability of the model. SUMMARY OF BACKGROUND DATA: Given the increasing incidence and surgical volume of spinal degeneration worldwide, accurate predictions of postoperative complications are urgently needed. SHAP-based interpretable ML models have not been used for CSFL risk factor analysis in lumbar fusion surgery. METHODS: Clinical and imaging data were retrospectively collected from 3505 patients who underwent lumbar fusion surgery. Six distinct machine learning models were formulated: extreme gradient boosting (XGBoost), decision tree (DT), random forest (RF), support vector machine (SVM), Gaussian naive Bayes (GaussianNB), and K-nearest neighbors (KNN) models. Evaluation of model performance on the test dataset was performed using performance metrics, and the analysis was executed through the SHAP framework. RESULTS: CSFL was detected in 95 out of 3505 patients (2.71%). Notably, the XGBoost model exhibited outstanding accuracy in forecasting CSFLs, with high precision (0.9815), recall (0.6667), accuracy (0.8182), F1 score (0.7347), and AUC (0.7343). Additionally, through SHAP analysis, significant predictors of CSFL were identified, including ligamentum flavum thickness, zygapophysial joint degeneration grade, central spinal stenosis grade, decompression segment count, decompression mode, intervertebral height difference, Cobb angle, intervertebral height index difference, operation mode, lumbar segment lordosis angle difference, Meyerding grade of lumbar spondylolisthesis, and revision surgery. CONCLUSION: The combination of the XGBoost model with the SHAP is an effective tool for predicting the risk of CSFL during lumbar fusion surgery. Its implementation could aid clinicians in making informed decisions, potentially enhancing patient outcomes and lowering healthcare expenses. This study advocates for the adoption of this approach in clinical settings to enhance the evaluation of CSFL risk among patients undergoing lumbar fusion.

Natural language processing-driven artificial intelligence models for the diagnosis of lumbar disc herniation with L5 and S1 radiculopathy: a preliminary evaluation.

OBJECTIVE: To develop and validate natural language processing (NLP)-driven artificial intelligence (AI) models for the diagnosis of lumbar disc herniation (LDH) with L5 and S1 radiculopathy using electronic health records (EHRs). METHODS: EHRs of patients undergoing single-level Percutaneous Endoscopic Lumbar Discectomy (PELD) for the treatment of LDH at the L4/5 or L5/S1 level between June 1, 2013, and December 31, 2021, were collected. The primary outcome was LDH with L5 and S1 radiculopathy, which was defined as nerve root compression recorded in the operative notes. Datasets were created using the history of present illness (HPI) text and positive symptom (PS) text with radiculopathy (L5 or S1), respectively. The datasets were randomly split into a training set and a testing set in a 7:3 ratio. Two machine learning (ML) models, the Long Short-Term Memory (LSTM) network and Extreme Gradient Boosting (XGBoost), were developed using the training set. Performance evaluation of the models on the testing set was done using measures such as the receiver operating characteristic (ROC) curve, area under the curve (AUC), accuracy, recall, F1-score, and precision. RESULTS: The study included a total of 1681 patients, with 590 patients having L5 radiculopathy and 1091 patients having S1 radiculopathy. Among the four models developed, the LSTM model based on PS text showed the best discrimination in the testing set, with precision (0.9054), recall (0.9405), accuracy (0.8950), F1-score (0.9226), and AUC (0.9485). CONCLUSIONS: This study provides preliminary validation of the concept that NLP-driven AI models can be used for the diagnosis of lumbar disease using EHRs. This study could pave the way for future research that may develop more comprehensive and clinically impactful AI-driven diagnostic systems.

MiR-653-5p drives osteoarthritis pathogenesis by modulating chondrocyte senescence.

BACKGROUND: Due to the unclear pathogenesis of osteoarthritis (OA), effective treatment for this ailment is presently unavailable. Accumulating evidence points to chondrocyte senescence as a key driver in OA development. This study aims to identify OA-specific microRNAs (miRNAs) targeting chondrocyte senescence to alleviate OA progression. METHODS: We screened and identified miRNAs differentially expressed in OA and normal cartilage, then confirmed the impact of miR-653-5p on chondrocyte functions and senescence phenotypes through in vitro experiments with overexpression/silencing. We identified interleukin 6 (IL-6) as the target gene of miR-653-5p and confirmed the regulatory influence of miR-653-5p on the IL-6/JAK/STAT3 signaling pathway through gain/loss-of-function studies. Finally, we assessed the therapeutic efficacy of miR-653-5p on OA using a mouse model with destabilization of the medial meniscus. RESULTS: MiR-653-5p was significantly downregulated in cartilage tissues and chondrocytes from OA patients. Overexpression of miR-653-5p promoted chondrocyte matrix synthesis and proliferation while inhibiting chondrocyte senescence. Furthermore, bioinformatics target prediction and the luciferase reporter assays identified IL-6 as a target of miR-653-5p. Western blot assays demonstrated that miR-653-5p overexpression inhibited the protein expression of IL-6, the phosphorylation of JAK1 and STAT3, and the expression of chondrocyte senescence phenotypes by regulating the IL-6/JAK/STAT3 signaling pathway. More importantly, the cartilage destruction was significantly alleviated and chondrocyte senescence phenotypes were remarkably decreased in the OA mouse model treated by agomiR-653-5p compared to the control mice. CONCLUSIONS: MiR-653-5p showed a significant decrease in cartilage tissues of individuals with OA, leading to an upregulation of chondrocyte senescence phenotypes in the articular cartilage. AgomiR-653-5p emerges as a potential treatment approach for OA. These findings provide further insight into the role of miR-653-5p in chondrocyte senescence and the pathogenesis of OA.

Circular RNA-GRIN2B Suppresses Neuropathic Pain by Targeting the NF-κB/SLICK Pathway.

The role of circular RNAs (circRNAs) in neuropathic pain is linked to the fundamental physiological mechanisms involved. However, the exact function of circRNAs in the context of neuropathic pain is still not fully understood. The functional impact of circGRIN2B on the excitability of dorsal root ganglion (DRG) neurons was investigated using siRNA or overexpression technology in conjunction with fluorescence in situ hybridization and whole-cell patch-clamp technology. The therapeutic efficacy of circGRIN2B in treating neuropathic pain was confirmed by assessing the pain threshold in a chronic constrictive injury (CCI) model. The interaction between circGRIN2B and NF-κB was examined through RNA pulldown, RIP, and mass spectrometry assays. CircGRIN2B knockdown significantly affected the action potential discharge frequency and the sodium-dependent potassium current flux (SLICK) in DRG neurons. Furthermore, knockdown of circGRIN2B dramatically reduced the SLICK channel protein and mRNA expression in vivo and in vitro. Our research confirmed the interaction between circGRIN2B and NF-κB. These findings demonstrated that circGRIN2B promotes the transcription of the SLICK gene by binding to NF-κB. In CCI rat models, the overexpression of circGRIN2B has been shown to hinder the progression of neuropathic pain, particularly by reducing mechanical and thermal hyperalgesia. Additionally, this upregulation significantly diminished the levels of the inflammatory cytokines IL-1ß, IL-6, and TNF-α in the DRG. Upon reviewing these findings, it was determined that circGRIN2B may mitigate the onset of neuropathic pain by modulating the NF-κB/SLICK pathway.

Deep Learning Model for Grading and Localization of Lumbar Disc Herniation on Magnetic Resonance Imaging.

BACKGROUND: Methods for grading and localization of lumbar disc herniation (LDH) on MRI are complex, time-consuming, and subjective. Utilizing deep learning (DL) models as assistance would mitigate such complexities. PURPOSE: To develop an interpretable DL model capable of grading and localizing LDH. STUDY TYPE: Retrospective. SUBJECTS: 1496 patients (M/F: 783/713) were evaluated, and randomly divided into training (70%), validation (10%), and test (20%) sets. FIELD STRENGTH/SEQUENCE: 1.5T MRI for axial T2-weighted sequences (spin echo). ASSESSMENT: The training set was annotated by three spinal surgeons using the Michigan State University classification to train the DL model. The test set was annotated by a spinal surgery expert (as ground truth labels), and two spinal surgeons (comparison with the trained model). An external test set was employed to evaluate the generalizability of the DL model. STATISTICAL TESTS: Calculated intersection over union (IoU) for detection consistency, utilized Gwet's AC1 to assess interobserver agreement, and evaluated model performance based on sensitivity and specificity, with statistical significance set at P < 0.05. RESULTS: The DL model achieved high detection consistency in both the internal test dataset (grading: mean IoU 0.84, recall 99.6%; localization: IoU 0.82, recall 99.5%) and external test dataset (grading: 0.72, 98.0%; localization: 0.71, 97.6%). For internal testing, the DL model (grading: 0.81; localization: 0.76), Rater 1 (0.88; 0.82), and Rater 2 (0.86; 0.83) demonstrated results highly consistent with the ground truth labels. The overall sensitivity of the DL model was 87.0% for grading and 84.0% for localization, while the specificity was 95.5% and 94.4%. For external testing, the DL model showed an appreciable decrease in consistency (grading: 0.69; localization: 0.66), sensitivity (77.2%; 76.7%), and specificity (92.3%; 91.8%). DATA CONCLUSION: The classification capabilities of the DL model closely resemble those of spinal surgeons. For future improvement, enriching the diversity of cases could enhance the model's generalization. TECHNICAL EFFICACY: Stage 2.

Repeat Surgery after Percutaneous Endoscopic Lumbar Discectomy for Adolescent Lumbar Disc Herniation: A Multicenter Observational Study.

OBJECTIVE: The reported date in the repeat surgical intervention for adolescent lumbar disc herniation (ALDH) after percutaneous endoscopic lumbar discectomy (PELD) was quite scarce. This study aims to introduce cases of repeat surgeries after PELD for ALDH and assess the incidence, chief causes, repeat surgery methods, and surgical outcomes of repeat surgeries after PELD for ALDH. METHODS: A retrospective multicenter observational study was conducted on patients undergoing repeat surgeries after PELD for ALDH at four tertiary referral hospitals from January 2014 through August 2022. The incidence of repeat surgeries, chief causes, strategies for repeat surgeries, and timing of repeat surgeries were recorded and analyzed. The clinical outcomes were evaluated by the Numeric Rating Scales (NRS) scores and the modified MacNab criteria. Statistical analyses were performed with the Wilcoxon signed-rank test. RESULTS: A total of 23 patients who underwent repeat surgeries after PELD for ALDH were included. The chief causes were re-herniation (homo-lateral re-herniation at the same level, new disc herniation of adjacent level). The repeat surgery methods were revision PELD, micro-endoscopic discectomy (MED), open discectomy and instrumented lumbar inter-body fusion. The NRS scores decreased significantly in follow-up evaluations and these scores demonstrated significant improvement at the last follow-up (p < 0.002). For the modified MacNab criteria, at the last follow-up, 18 patients (78.26%) had an excellent outcome, and the overall success rate was 86.95%. CONCLUSION: This study's data suggest that young patients who underwent repeat surgery improved significantly compared to baseline. The chief cause was re-herniation. Revision PELD was the main surgical procedure, which provides satisfactory clinical results in young patients who underwent repeat surgeries.

No Difference in Bone Tunnel Enlargement and Clinical Outcome between Cortical Suspension and Hybrid Femoral Fixation in Hamstring Anterior Cruciate Ligament Reconstruction.

OBJECTIVE: The best method for femoral fixation in anterior cruciate ligament reconstruction (ACLR) remains controversial. The study assesses the bone tunnel enlargement and clinical outcome in hamstring ACLR using cortical suspension or hybrid (cortical suspension and compression) femoral fixation. METHODS: From January 2010 to December 2021, 102 patients who underwent quadruple hamstring ACLR using cortical suspension (39 patients) or hybrid (63 patients) fixation on the femoral side were retrospectively analyzed. Clinical evaluation was conducted using the international knee documentation committee score, the Lysholm score, the Tegner activity level scale, the knee injury and osteoarthritis outcome score (quality of life score), the Lachman test, and the side-to-side difference by the KT-1000 arthrometer. The complications after the surgery were also evaluated. These data were compared at baseline and last follow-up. The diameters of the femoral tunnel were calculated at three sites: the width of the entrance of the femoral tunnel, 1 cm proximal to the entrance of the femoral tunnel and the largest diameter of the femoral tunnel on magnetic resonance imaging (MRI) coronal images. Bone tunnel widening data were contrasted between MRI images conducted at least 2 years and within 2 weeks after surgery. The morphology of bone tunnel enlargement was also observed and recorded. The categorical parameters were analyzed using the χ2-test and Fisher's exact test. The continuous variables conforming to a normal distribution were analyzed using Student's t-test, and the Mann-Whitney U-test was undertaken between the two groups without normal distribution. RESULTS: Both cortical suspension and hybrid femoral fixation in quadruple hamstring ACLR achieved significantly improved patient-reported outcome scores and knee stability compared to preoperative data. However, no significant differences were found between these two methods in clinical evaluations, postoperative complications, and patient-reported outcome scores. Although the mean diameter of the enlarged bone tunnel was lowered by an additional bioabsorbable interference screw fixation near the joint line, a statistically insignificant difference was found between the hybrid and cortical suspension fixation on the femoral side. There was no statistical difference in the distribution of enlarged bone tunnel morphology between groups. CONCLUSIONS: No significant difference was found in the bone tunnel enlargement and clinical outcome between cortical suspension and hybrid femoral fixation in ACLR using hamstring autograft.

Interferon Regulatory Factor 4 (IRF4) Plays a Key Role in Osteoblast Differentiation of Postmenopausal Osteoporosis.

BACKGROUND: Postmenopausal osteoporosis (PMOP) is a prevalent disease, which features decreased bone mass, bone weakness and deteriorated bone microstructure in postmenopausal women. Although many factors have been revealed to contribute to the occurrence of PMOP, its mechanism remains undefined. This work aimed to identify significant changes in gene expression during PMOP formation and to examine the most valuable differential genes in postmenopausal osteoporosis versus the control group. METHODS: The GSE68303 dataset that contains 12 ovariectomize (OVX) experimental and 11 sham groups was downloaded and analyzed. The results indicated that interferon regulatory factor 4 (IRF4) might be a hub gene in the development of postmenopausal osteoporosis. Western blot and immunohistochemistry were carried out to evaluate IRF4 levels in thoracic vertebra extracts from OVX and Sham mice. To assess IRF4's impact on osteogenic differentiation in postmenopausal bone marrow mesenchymal stem cells (BM-MSCs), IRF4 overexpression (OV-IRF4) and knockdown (Sh-IRF4) plasmids were constructed. RESULTS: The results showed that comparing with the sham group, bone samples from the OVX group showed higher IRF4 expression. Alkaline phosphatase (ALP) staining revealed that IRF4 overexpression significantly inhibited ALP activity, while IRF4 knockdown promoted ALP activity in BM-MSCs. Simvastatin-treated OVX mice showed increased total bone volume/total tissue volume (BV/TV) and elevated Runx2 expression by immunohistochemical staining compared with the OVX group. CONCLUSIONS: This study demonstrated that IRF4 is associated with OVX induced osteoporosis, it can regulate bone stability by inhibiting the osteogenic differentiation BM-MSCs. This study may help enhance our understanding of the molecular mechanism of PMOP formation, providing new insights into estrogen defiance induced osteoporosis.

Prediction Model and Risk Factor Analysis of Adjacent Segment Disease After L4-5 Transforaminal Lumbar Interbody Fusion Through Preoperative Radiographic Features.

STUDY DESIGN: A retrospective study. OBJECTIVE: To investigate the risk of adjacent segment disease (ASD) after L4-5 transforaminal lumbar interbody fusion (TLIF) in patients diagnosed with lumbar spinal stenosis (LSS), a prediction model for ASD is established and validated. METHODS: A retrospective study was carried out on a sample of 290 patients who underwent L4-5 TLIF at Zhongda Hospital, Southeast University, from January 2015 to January 2021. The study collected baseline data and preoperative radiographic features of L3-4 and L5-S1. The determination of the outcome variable was based on X-ray results spanning over 24 months and JOA scores. Multivariate logistic regression was used to identify the risk factors in constructing a nomogram. RESULTS: Independent risk factors for L3-4 degeneration after TLIF included osteoarthritis of L3-4 facet joints, L3-4 foraminal stenosis, L4 upper endplate osteochondritis, L3-4 local lordosis angle, and L3-4 spinal stenosis. Independent risk factors for L5-S1 degeneration after TLIF included osteoarthritis of L5-S1 facet joints, L5-S1 intervertebral disc degeneration, L5-S1 spinal stenosis, L5-S1 coronal imbalance, and S1 upper endplate osteochondritis. A predictive model was developed. The AUC for the prediction models at L3-4 and L5-S1 were .945 and .956. The calibration curve demonstrated good consistency between the predicted and actual probabilities. The DCA curve indicated the clinical benefit and practical value of this predictive model. CONCLUSION: This study established nomograms for postoperative degeneration at L3-4 and L5-S1 based on selected preoperative radiographic features. These models provide a valuable auxiliary decision-making system for clinicians and aid in early surgical decisions.

Automated machine learning-based model for the prediction of pedicle screw loosening after degenerative lumbar fusion surgery.

The adequacy of screw anchorage is a critical factor in achieving successful spinal fusion. This study aimed to use machine learning algorithms to identify critical variables and predict pedicle screw loosening after degenerative lumbar fusion surgery. A total of 552 patients who underwent primary transpedicular lumbar fixation for lumbar degenerative disease were included. The LASSO method identified key features associated with pedicle screw loosening. Patient clinical characteristics, intraoperative variables, and radiographic parameters were collected and used to construct eight machine learning models, including a training set (80% of participants) and a test set (20% of participants). The XGBoost model exhibited the best performance, with an AUC of 0.884 (95% CI: 0.825-0.944) in the test set, along with the lowest Brier score. Ten crucial variables, including age, disease diagnosis: degenerative scoliosis, number of fused levels, fixation to S1, HU value, preoperative PT, preoperative PI-LL, postoperative LL, postoperative PT, and postoperative PI-LL were selected. In the prospective cohort, the XGBoost model demonstrated substantial performance with an accuracy of 83.32%. This study identified crucial variables associated with pedicle screw loosening after degenerative lumbar fusion surgery and successfully developed a machine learning model to predict pedicle screw loosening. The findings of this study may provide valuable information for clinical decision-making.

COL6A3 enhances the osteogenic differentiation potential of BMSCs by promoting mitophagy in the osteoporotic microenvironment.

BACKGROUND: Bone marrow mesenchymal stem cells (BMSCs) have been widely recognized as a highly promising option for cell-based tissue engineering therapy targeting osteoporosis. However, the osteogenic differentiation of BMSCs is impeded by the limited viability and diminished capacity for bone formation within the osteoporotic microenvironment. METHODS: In this study, the COL6A3 gene was confirmed through an extensive analysis of the preceding single-cell sequencing database. The generation of an inflammatory microenvironment resembling osteoporotic cell transplantation was achieved by employing lipopolysaccharide (LPS). A lentivirus targeting the COL6A3 gene was constructed, and a Western blotting assay was used to measure the marker proteins of osteogenesis, adipogenesis, and mitophagy. Immunofluorescence was utilized to observe the colocalization of mitochondria and lysosomes. The apoptosis rate of each group was evaluated using the TUNEL assay, and the mitochondrial membrane potential was assessed using JC-1 staining. RESULTS: This investigation discovered that the impaired differentiation capacity and decreased viability of BMSCs within the inflammatory microenvironment were markedly ameliorated upon overexpression of the specific COL6A3 gene. Moreover, the administration of COL6A3 gene overexpression successfully mitigated the inhibitory impacts of LPS on mitophagy and the expression of inflammatory mediators, specifically inducible nitric oxide synthase (iNOS) and cyclooxygenase-2 (COX-2), in BMSCs. To clarify the underlying mechanism, the role of mitophagy during the differentiation of COL6A3 gene-modified BMSCs in the inflammatory microenvironment was evaluated using the mitophagy inhibitor Mdivi-1. CONCLUSIONS: In the context of lipopolysaccharide (LPS) stimulation, COL6A3 enhances the differentiation of BMSCs into osteogenic and adipogenic lineages through the promotion of mitophagy and the maintenance of mitochondrial health. Our findings may provide a novel therapeutic approach utilizing stem cells in the treatment of osteoporosis.

Endoplasmic reticulum stress-mediated autophagy alleviates lipopolysaccharide-induced nucleus pulposus cell pyroptosis by inhibiting CHOP signaling in vitro.

Pyroptosis, a newly discovered pro-inflammatory programmed necrosis of cells, serves as an initiating and promoting event that leads to intervertebral disc (IVD) degeneration (IDD). Endoplasmic reticulum stress (ERS) and autophagy are vital regulatory mechanisms of cellular homeostasis, which is also closely related to IDD. However, the role and relationship of ERS and autophagy in the pyroptosis of nucleus pulposus cell (NPC) are not well understood. In this research, we aimed to elucidate the role and mechanism of ERS-C/EBP homologous protein (CHOP) in lipopolysaccharide (LPS)-induced cell pyroptosis and determine its interaction with autophagy. ERS and autophagy inducers or inhibitors were used or not in the preconditioning of rat NPCs. Cell viability, pyroptosis-related protein expression, caspase-1 activity assay, and enzyme-linked immunosorbent assay were performed to observe rat NPC pyroptosis after the treatment of LPS. Activation of the ERS pathway and autophagy were assessed by quantitative real-time PCR, western blot analyses, and immunofluorescence staining assay to classify the molecular mechanisms. Our results showed that LPS stimulation induced NPC pyroptosis with concomitant activation of the ERS-CHOP pathway and initiated autophagy. Activation of the ERS-CHOP pathway exacerbated rat NPC pyroptosis, whereas autophagy inhibited cell pyroptosis. LPS-induced cell pyroptosis and CHOP upregulation were negatively regulated by autophagy. LPS-induced autophagy was depressed by the ERS inhibitor but aggravated by the ERS inducer. Taken together, our findings suggested that LPS induced NPC pyroptosis by activating ERS-CHOP signaling and ERS mediated LPS-induced autophagy, which in turn alleviated NPC pyroptosis by inhibiting CHOP signaling.

Bioinformatics Analysis and Identification of Ferroptosis-Related Hub Genes in Intervertebral Disc Degeneration.

Approximately 80% of individuals encounter lower back pain (LBP), a prevalent clinical issue largely attributed to intervertebral disc degeneration (IDD). Ferroptosis is an iron-dependent lipid peroxidation-driven cell death, and there is growing evidence that ferroptosis plays an important role in various human diseases. However, the underlying mechanism of ferroptosis in IDD remains unclear. This study aims to reveal the potential hub genes and related pathways of ferroptosis in the pathogenesis and progression of IDD. In this study, we analyzed three microarray datasets from the GEO database. Additionally, we downloaded ferroptosis-related genes from FerrDb-V2 and extracted apoptosis-related genes from UniProt as a control to show the specificity of ferroptosis. Weighted gene co-expression network analysis (WGCNA) was performed to identify the IDD-related module genes. Then, ferroptosis-related genes and apoptosis-related genes were separately overlapped with the IDD-related module genes, resulting in the identification of 35 ferroptosis-related module genes (FRMG) and 142 apoptosis-related module genes (ARMG). Furthermore, we performed functional enrichment analysis and protein-protein interaction network, and Cytoscape along with CytoHubba was used to identify the hub genes. Finally, logistic regression models were constructed and identified two hub FRMGs (PTEN and EGFR) and one hub ARMG (CTNNB1), which could distinguish IDD patients from controls (P < 0.05). The areas under the ROC curves were 0.792 and 0.730, respectively, suggesting that ferroptosis is more specific than apoptosis in IDD. In conclusion, this study provided fresh perspectives on ferroptosis in the pathogenesis and progression of IDD that can be used to evaluate potential biomarker genes and therapeutic targets.

Interpretable Machine Learning Model to Predict Bone Cement Leakage in Percutaneous Vertebral Augmentation for Osteoporotic Vertebral Compression Fracture Based on SHapley Additive exPlanations.

STUDY DESIGN: Retrospective study. OBJECTIVES: Our objective is to create comprehensible machine learning (ML) models that can forecast bone cement leakage in percutaneous vertebral augmentation (PVA) for individuals with osteoporotic vertebral compression fracture (OVCF) while also identifying the associated risk factors. METHODS: We incorporated data from patients (n = 425) which underwent PVA. To predict cement leakage, we devised six models based on a variety of parameters. Evaluate and juxtapose the predictive performances relied on measures of discrimination, calibration, and clinical utility. SHapley Additive exPlanations (SHAP) methodology was used to interpret model and evaluate the risk factors associated with cement leakage. RESULTS: The occurrence rate of cement leakage was established at 50.4%. A binary logistic regression analysis identified cortical disruption (OR 6.880, 95% CI 4.209-11.246), the basivertebral foramen sign (OR 2.142, 95% CI 1.303-3.521), the fracture type (OR 1.683, 95% CI 1.083-2.617), and the volume of bone cement (OR 1.198, 95% CI 1.070-1.341) as independent predictors of cement leakage. The XGBoost model outperformed all others in predicting cement leakage in the testing set, with AUC of .8819, accuracy of .8025, recall score of .7872, F1 score of .8315, and a precision score of .881. Several important factors related to cement leakage were drawn based on the analysis of SHAP values and their clinical significance. CONCLUSION: The ML based predictive model demonstrated significant accuracy in forecasting bone cement leakage for patients with OVCF undergoing PVA. When combined with SHAP, ML facilitated a personalized prediction and offered a visual interpretation of feature importance.

Scientific Bibliometric and Visual Analysis of Studies on Autophagy in Intervertebral Disc Degeneration Based on Web of Science.

OBJECTIVE: To analyze the current research trends and potential mechanisms related to the role of autophagy in intervertebral disc degeneration (IVDD) and to provide new ideas for future research in this field. METHODS: All articles on IVDD and autophagy were retrieved and extracted from the Web of Science (WoS) core collection database. The results were evaluated and visualized using the bibliometric Web site, CiteSpace, and VOSviewer software, including annual articles published, countries, institutions, authors, journals, research areas, funding agencies, citations, and keywords. RESULTS: From January 1, 2011, to December 31, 2022, 323 reviews and original articles were included, and the overall trend in the number of articles was increasing rapidly. China and the United States were the countries with the most scientific research achievements. The 323 articles received a total number of citations of 6949, with an H index of 43 and an average citation of 21.51. The top publication country, institution, author, journal, research area, and funding agency were China, Huazhong University of Science and Technology, Cao Yang of Tongji Medical College, Oxidative Medicine and Cellular Longevity, cell biology, and National Natural Science Foundation of China, respectively. Most of the keywords were associated with the mechanisms and regulatory networks of autophagy. In addition, with increasing evidence showing the key role of autophagy in IVDD, therapy, signaling pathway, and mitophagy are emerging as new research hot spots that should be paid more attention. CONCLUSIONS: This study provided a scientific perspective on autophagy in IVDD and elucidated the current research status and hot spots in this field. The mechanism of autophagy and the application of regulating autophagy in the treatment of IVDD deserve further research.

Comparison of thoracolumbar versus non-thoracolumbar osteoporotic vertebral compression fractures in risk factors, vertebral compression degree and pre-hospital back pain.

BACKGROUND: Thoracolumbar spine is at high risk of osteoporotic vertebral compression fractures (OVCF). This study aimed to identify the differences in risk factors, vertebral compression degree and back pain characteristics of thoracolumbar OVCF (TL-OVCF) and non-thoracolumbar OVCF (nTL-OVCF). METHODS: OVCF patients hospitalized in a spine center between June 2016 and October 2020 were retrospectively studied. Demographics, comorbidity, spine trauma, bone mineral density, duration of pre-hospital back pain, extent of vertebral marrow edema, and degree of vertebral compression of patients with nTL-OVCF were summarized and compared to those with TL-OVCF. RESULTS: A total of 944 patients with acute single-segment OVCF were included. There were 708 (75.0%) TL-OVCF located in T11-L2 and 236 (25.0%) nTL-OVCF in lower lumbar (L3-L5) and middle thoracic (T5-T10) spine. The female-male ratio was 4.1 in nTL-OVCF and differed not significantly from TL-OVCF. The middle thoracic OVCF were older and had higher comorbidity of coronary heart disease (21.3%) and cerebral infarction (36.3%) than TL-OVCF (12.1% and 20.6%). In nTL-OVCF the ratio of apparent spine trauma (44.9%) and pre-hospital back pain ≤ 1 week (47.5%) was lower than in TL-OVCF (66.9% and 62.6%). The T-score value of lumbar spine was - 2.99 ± 1.11, - 3.24 ± 1.14, - 3.05 ± 1.40 in < 70, 70-80, > 80 years old TL-OVCF and differed not significantly from nTL-OVCF. The lower lumbar OVCF had more cranial type of vertebral marrow edema (21.8%) and fewer concurrent lumbodorsal fasciitis (30.8%) than TL-OVCF (16.8% and 43.4%). In TL-OVCF the anterior-posterior vertebral height ratio was lower with back pain for > 4 weeks than for ≤ 1, 1-2, and 2-4 weeks. In nTL-OVCF the degree of vertebral compression differed not significantly with pre-hospital back pain for ≤ 1, 1-2, 2-4, and > 4 weeks. CONCLUSIONS: Thoracolumbar spine has 2-folds higher risk of OVCF than non-thoracolumbar spine. Non-thoracolumbar OVCF are not associated with female gender, apparent spine trauma or poor bone mineral density, but tend to maintain the degree of vertebral compression and cause longer duration of pre-hospital back pain.

Indirect Puncture Using a Novel Arc Puncture-Guided Device in Percutaneous Transforaminal Puncture on Goat Lumbar Spine Specimens.

OBJECTIVE: We investigated the indirect puncture method using a novel arc puncture-guided device in percutaneous transforaminal puncture to improve puncture accuracy and reduce the fluoroscopy, puncture, and operation times. METHODS: We have designed a novel arc puncture-guided device consisting of a 90° arc block and a 30° arc block. Punctures were performed on 8 fresh goat lumbar spine specimens. A senior doctor performed indirect punctures on the left side of the L2-L3, L3-L4, and L4-L5 levels using the novel device (group A) and on the right side of the L2-L3, L3-L4, and L4-L5 levels using the conventional method (group B). We recorded the fluoroscopy, puncture, and operation times. RESULTS: In group A, the first puncture could successfully reach the target after 1-3 punctures, and the one-time success rate of the second needle puncture was 91.67%. The total fluoroscopy time was 14.88 ± 0.99 minutes in group A and 16.08 ± 2.22 minutes in group B (P = 0.027). The puncture times were 3.00 ± 0.66 minutes in group A and 6.04 ± 2.13 minutes in group B (P < 0.01). The operation time was 273.75 ± 30.19 minutes in group A and 361.25 ± 69.57 minutes in group B (P < 0.01). The differences in fluoroscopy times, puncture times, and operation times between the 2 groups were statistically significantly (P < 0.05). CONCLUSIONS: Indirect puncture using the novel arc puncture-guided device for percutaneous transforaminal puncture can significantly improve puncture accuracy and reduce the fluoroscopy, puncture, and operation times. Indirect puncture using the novel device in percutaneous transforaminal endoscopic discectomy is a potential and practical puncture method.

Nucleus Pulposus Cells Induce M2 Polarization of RAW264.7 via CX3CL1/CX3CR1 Pathway and M2 Macrophages Promote Proliferation and Anabolism of Nucleus Pulposus Cells.

Background: The mechanisms underlying M2 macrophage polarization induced by nucleus pulposus (NP) cells are unclear. The effects that M2-polarized macrophages have on NP cells are also controversial. Methods: Transcriptome sequencing was performed to detect the gene change profiles between NP cells from ruptured intervertebral disc (IVD) and normal IVD. The main difference on biological activities between the two cell groups were analyzed by GO analysis and KEGG analysis. Virus transduction, flow cytometry, immunofluorescence, RT-PCR, western blot, CCK-8, TUNEL staining, and AO/EB staining were performed to explore the interactions between NP cells and RAW264.7 macrophages. Statistics were performed using SPSS26. Results: 801 upregulated and 276 downregulated genes were identified in NP cells from ruptured IVD in mouse models. According to GO and KEGG analysis, we found that the differentially expressed genes (DEG) were dominantly enriched in inflammatory response, extracellular matrix degradation, blood vessel morphogenesis, immune effector process, ossification, chemokine signaling pathway, macrophage activation, etc. CX3CL1 was one of the top 20% DEG, and we confirmed that both NP tissue and cells expressed remarkably higher level of CX3CL1 in mouse models (p < 0.001∗). Besides, we further revealed that both the recombinant CX3CL1 and NP cells remarkably induced M2 polarization of RAW264.7 (p < 0.001∗), respectively, while this effect was significantly reversed by si-CX3CL1 or JMS-17-2 (p < 0.001∗). Furthermore, we found that M2 macrophages significantly decreased the apoptosis rate (p < 0.001∗) and the catabolic gene levels (p < 0.001∗) of NP cells, while increased the viability, proliferation as well as the anabolic gene levels of NP cells (p < 0.01∗). Conclusions: Via regulating CX3CL1/CX3CR1 pathway, NP cells can induce the M2 macrophage polarization. M2 polarized macrophages can further promote NP cell viability, proliferation, and anabolism, while inhibit NP cell apoptosis and catabolism.

Efficacy and Safety Evaluation of Bilateral Pedicle Approach Combined with Positional Reduction for the Treatment of Osteoporotic Thoracolumbar Burst Fractures.

Background: Osteoporotic thoracolumbar burst fractures (OTLBF) pose challenges for vertebroplasty due to the risk of cement leakage and spinal injury resulting from the fracture of the posterior vertebra and spinal canal occupancy. It limits the application of vertebroplasty in these patients. Objective: This study investigates the efficacy and safety of a bilateral pedicle approach combined with postural reduction for treating OTLBF using vertebroplasty. Material and Methods: Thirteen patients (aged ≥ 65 years) with thoracolumbar fractures without neurological deficits underwent vertebroplasty. The fractures affected the anterior and middle columns of the vertebrae, with mild compression of the canal. Clinical symptoms, procedure effects, patient mobility, and pain were assessed before the procedure and between 1 day and 3 months post-procedure. Kyphosis correction, wedge angle, and height restoration were also measured. Results: Immediate improvements in pain and mobility were observed in all patients following vertebroplasty, with sustained improvements over 6 months. Significant improvements were observed between 1 day and 6 months post-procedure, with at least a 4-level reduction in pain after 6 months. No comorbidities were observed. Kyphosis correction, wedge angle, and height restoration were improved. In one patient, postoperative computed tomography revealed polymethylmethacrylate leakage into the disc space and paravertebral space through the endplate fracture site, while no intraspinal leakage was found in other patients. Conclusions: Although vertebroplasty is generally considered contraindicated in OTLBF patients with posterior body involvement, this study demonstrates successful and safe treatment without causing neurological deficits. Percutaneous vertebroplasty combined with body reduction may serve as an alternative method for treating OTLBF, effectively preventing major surgical complications. Furthermore, it offers superior kyphosis correction, vertebral body reduction, pain reduction, early mobilization, and pain relief for patients.

Comparison of acute single versus multiple osteoporotic vertebral compression fractures in radiographic characteristic and bone fragility.

BACKGROUND: Osteoporotic vertebral compression fractures (OVCF) are common in aged population with bone fragility. This study aimed to identify the radiographic and bone fragility characteristic of acute single and multiple OVCF. METHODS: OVCF patients hospitalized in a spine center between June 2016 and October 2020 were retrospectively studied. Demographics, comorbidity, bone mineral density, spine trauma, duration of pre-hospital back pain, anatomical location and distribution pattern of OVCF, extent of vertebral marrow edema, and degree of vertebral compression of patients with multi-segment vertebral fractures (MSVF) were summarized and compared to those with single segment vertebral fractures (SSVF). RESULTS: A total of 1182 patients with 1530 acute fractured vertebrae were included. There were 944 SSVF (79.9%) and 238 MSVF (20.1%) simultaneously involving two (MSVF-2) or three and more vertebra (MSVF-3/m). The Female-Male ratio was 4.4 and differed not significantly between SSVF and MSVF. Females in SSVF were younger than males while MSVF-2 tended to occur in older females. L1, T12, and L2 were the three most frequently fractured vertebra and MSVF involved more vertebra in thoracic and lumbar spine. 31.1% in MSVF-2 and 83.1% in MSVF-3/m had at least two vertebral fractures in adjacent. The fractured thoracolumbar vertebra in MSVF was less compressed than that in SSVF. Apparent spine trauma was reported by 61.4% of SSVF, 44.1% of MSVF-2, and 36.3% of MSVF-3/m, while early hospitalization with pre-hospital back pain ≤ 1 week was 58.9% in SSVF, 45.3% in MSVF-2, and 25.9% in MSVF-3/m. Only females aged 70-80 years old in MSVF-3/m showed lower baseline bone mineral density than in MSVF-2 and SSVF. MSVF were not associated with increased comorbidity of hypertension, diabetes, coronary heart disease, cerebral infarction, and chronic pulmonary disease. CONCLUSIONS: 20% of acute OVCF can involve multiple vertebra without significant spine trauma or lower baseline bone mineral density. Multiple OVCF tend to occur in adjacent vertebra with less thoracolumbar vertebral compression but longer duration of pre-hospital back pain.