Spine Journal最新文献

{"title":"Multimodal Machine Learning for Predicting Perioperative Safety Indicators in Spinal Surgery.","authors":"Kyle Mani, Thomas Scharfenberger, Samuel N Goldman, Emily Kleinbart, Evan Mostafa, Rafael De La Garza Ramos, Mitchell S Fourman, Ananth Eleswarapu","doi":"10.1016/j.spinee.2025.03.021","DOIUrl":"https://doi.org/10.1016/j.spinee.2025.03.021","url":null,"abstract":"<p><strong>Background context: </strong>Machine learning (ML) algorithms can utilize the large amount of tabular data in electronic health records (EHRs) to predict peri-operative safety indicators. Integrating unstructured free-text inputs via natural language processing (NLP) may further enhance predictive accuracy.</p><p><strong>Purpose: </strong>To design and validate a pre-operative multi-modal machine learning architecture that integrates structured EHR data (patient demographics, comorbidities, and clinical covariates) with unstructured free-text inputs (past medical and surgical history, medications, and problem lists) via natural language processing (NLP). The multi-modal models aim to improve the prediction of peri-operative safety indicators compared to baseline ML models that only use structured tabular EHR data.</p><p><strong>Study design: </strong>Retrospective cohort study PATIENT SAMPLE: 1,898 patients admitted for elective or emergency spine surgery at four separate large urban academic spine centers during a five-year period from 2018-2023.</p><p><strong>Outcome measures: </strong>Numerical outputs between 0 to 1 corresponding to the likelihood of (I) extended length of stay (LOS), (II) 90-day reoperation, and (III) peri-operative intensive care unit (ICU) admission.</p><p><strong>Methods: </strong>We predicted the following safety indicators (I) extended length of stay (LOS), II (90-day reoperation, and (III) peri-operative intensive care unit (ICU) admission. The quanteda package for NLP within the R environment was utilized to preprocess free-text EHR inputs. The refined text was tokenized and transformed into numerical vectors using a bag-of-words approach and integrated with the tabular EHR data to create a document-feature matrix. Two extreme gradient boosted (XGBoost) ML models were trained: a base model utilizing only structured tabular EHR data and a combined multi-modal model that leveraged both combined structured tabular EHR data with numerical vectors derived from free-text NLP inputs. Hyperparameter tuning was performed via grid search, and the models were validated using 10-fold cross validation with an 80:20 training/testing split. Word clouds were generated for the free-text data and explainable artificial intelligence (XAI) techniques were employed for feature importance. Metrics calculated for model performance included Area Under the Receiving-Operating Characteristic Curve (AUC-ROC), Brier score, Calibration slope, Calibration Intercept, Precision, Recall and F1-Score.</p><p><strong>Results: </strong>1,898 patients (60.7% female) were extracted from January 2018 to September 2023, with a median age of 60.0 (IQR: 52.0 - 68.0) and median body mass index (BMI) of 30.3 kgm<sup>2</sup> (IQR: 26.3 - 34.6). Extended LOS was defined as ≥ 14.4 days, constituting 10.1% of all individuals. The median LOS for the entire cohort was 4.0 days (IQR: 2.0 - 7.0), while the 90-day reoperation rate was 10.54%, and the ICU admis","PeriodicalId":49484,"journal":{"name":"Spine Journal","volume":" ","pages":""},"PeriodicalIF":4.9,"publicationDate":"2025-03-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143755375","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Development of Quantitative Instrumentation for Cone of Economy Analysis: Bridging Radiographic and Clinical Measures.","authors":"Hsu Ming Chang, Soong Christina, Yeh Ting Jen, Chen Hsuan Yu","doi":"10.1016/j.spinee.2025.03.005","DOIUrl":"https://doi.org/10.1016/j.spinee.2025.03.005","url":null,"abstract":"<p><strong>Background context: </strong>Postural misalignment and compromised balance are major contributors to fall risk in the elderly, leading to significant physical injuries, reduced quality of life (QOL), and increased healthcare burdens. Evaluating postural stability is critical for fall prevention. The cone of economy (CoE) concept explores the range of motion of the center of mass and head required to maintain a stable upright posture, yet its measurement remains challenging due to the limitations of existing methods.</p><p><strong>Purpose: </strong>This study introduces a novel apparatus for real-time measurement of both external and internal CoEs, offering a cost-effective alternative to expensive and complex traditional methods that require extensive data processing.</p><p><strong>Study design/setting: </strong>The study employed a controlled experimental design to develop and validate the proposed CoE measurement apparatus in a laboratory setting using a sample of healthy young adults.</p><p><strong>Methods: </strong>The proposed apparatus used two spherical measuring units to independently track pelvic and T1 vertebra motion, employing dual rotational magnetic encoders and a linear displacement sensor for precise 3D motion contour capture.</p><p><strong>Results: </strong>Validation experiments confirmed the system's reliability, achieving an average measurement error below 1.5 mm. The CoE is not an idealized cone but has an irregular conical shape, influenced by physiological factors (height and weight). The average range of sway (RoS) for external CoE at T1 was 42.7 (coronal), 47.6 (sagittal), and 12.5 cm (vertical), whereas that at the pelvic position was 14.3, 13.4, and 8 cm, respectively. The average RoS for internal CoE were smaller: 10.4 (coronal), 6.9 (sagittal), and 2 cm (vertical) at T1 and 8.2, 5.8, and 2.2 cm, respectively, at the pelvic position. The external CoE exhibited a larger RoS at the body's front, reflecting the foot's role in balancing forward shifts of gravity.</p><p><strong>Conclusions: </strong>Preliminary findings highlight a stronger correlation between external CoE and height than internal CoE, establishing a foundation for CoE research with implications for fall prevention and balance assessment.</p><p><strong>Clinical significance: </strong>This study introduces a reliable, cost-effective apparatus for real-time cone of economy (CoE) measurement, offering dynamic insights into postural stability, fall risk, and personalized balance assessment in clinical settings.</p>","PeriodicalId":49484,"journal":{"name":"Spine Journal","volume":" ","pages":""},"PeriodicalIF":4.9,"publicationDate":"2025-03-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143755373","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"International external validation of the SORG machine learning algorithm for predicting sustained postoperative opioid prescription after anterior cervical discectomy and fusion using a Taiwanese cohort of 1,037 patients.","authors":"Yu-Yung Chen, Hung-Kuan Yen, Jui-Yo Hsu, Ta-Chun Lin, Hao-Chen Lin, Chih-Wei Chen, Ming-Hsiao Hu, Olivier Q Groot, Joseph H Schwab","doi":"10.1016/j.spinee.2025.03.022","DOIUrl":"https://doi.org/10.1016/j.spinee.2025.03.022","url":null,"abstract":"&lt;p&gt;&lt;strong&gt;Background context: &lt;/strong&gt;Anterior cervical discectomy and fusion (ACDF) is widely performed for cervical spine disorders, with opioids commonly prescribed postoperatively for pain management. However, prolonged opioid use carries significant risks such as dependency and adverse health effects. Predictive models like the SORG machine learning algorithm (SORG-MLA) have been developed to forecast prolonged opioid use post-ACDF. External validation is essential to ensure their effectiveness across different healthcare settings and populations.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Purpose: &lt;/strong&gt;The study aimed to assess the generalizability of the SORG-MLA to a Taiwanese patient cohort for predicting prolonged opioid use after ACDF.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Study design: &lt;/strong&gt;Retrospective cohort study utilizing data from a tertiary care center in Taiwan.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Patient sample: &lt;/strong&gt;1,037 patients who underwent ACDF between 2010 and 2018 were included.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Outcome measures: &lt;/strong&gt;The primary outcome was sustained postoperative opioid prescription defined as continuous opioid use for at least 90 days following ACDF.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Methods: &lt;/strong&gt;The performance of the SORG-MLA in the validation cohort was assessed using discrimination measures (area under the receiver operating characteristic curve [AUROC] and the area under the precision-recall curve [AUPRC]), calibration, overall performance (Brier Score), and decision curve analysis. Comparing the validation cohort to the developmental revealed significant differences in demographic profiles, medicolegal frameworks, ethnic cultural contexts and key predictors of postoperative opioid use identified by the SORG-MLA. The Taiwanese cohort was characterized by an older age demographic, a lower proportion of female participants, higher smoking prevalence, higher incidence of preoperative myelopathy and radiculopathy, and more frequent use of antidepressants prior to surgery. Conversely, these patients were less likely to have extended preoperative opioid prescriptions beyond 180 days, undergo multi-level ACDF procedures, or be treated with concurrent medications such as Beta-2 agonists, Gabapentin, and ACE inhibitors. This study had no funding source or conflict of interests.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Results: &lt;/strong&gt;The model demonstrated good discriminative ability, with an AUROC of 0.78 and an AUPRC of 0.35. Calibration curves indicated that the model overestimated the risk of prolonged opioid use. This discrepancy may be attributed to the significantly higher incidence of sustained opioid consumption in the American development cohort, spanning from 2000 to 2018, which was threefold higher than that in the Taiwanese validation cohort between 2010 and 2018 (9.9% [270/2737] vs. 3.3% [34/1037]; p &lt; 0.01). The Brier score was 0.033, which improved upon the null model's score of 0.040, indicating robust overall performance. Decision curve analysis confirmed the model's clinical utility, demonstrating ","PeriodicalId":49484,"journal":{"name":"Spine Journal","volume":" ","pages":""},"PeriodicalIF":4.9,"publicationDate":"2025-03-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143755374","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Performance Comparison Between Hounsfield Units and DXA in Predicting Lumbar Cage Subsidence in the Degenerative Population.","authors":"Lindsay D Orosz, Kirsten A Schuler, Brandon J Allen, Wondwossen T Lerebo, Tarek Yamout, Rita T Roy, Thomas C Schuler, Christopher R Good, Colin M Haines, Ehsan Jazini","doi":"10.1016/j.spinee.2025.03.028","DOIUrl":"https://doi.org/10.1016/j.spinee.2025.03.028","url":null,"abstract":"<p><strong>Background context: </strong>Bone mineral density assessment is essential for surgical planning for most spine surgeries, but gold standard dual-energy x-ray absorptiometry (DXA) is affected by degeneration often resulting in falsely elevated scores. Studies of the opportunistic measurement of computed tomography (CT) Hounsfield units (HU) suggest lower CTHU values predict interbody cage subsidence, yet cutoff values vary and lack standardization.</p><p><strong>Purpose: </strong>This study aimed to determine if value CTHU<135 was associated with lumbar interbody cage subsidence and to compare the predictive performance of subsidence between CTHU and DXA.</p><p><strong>Study design/setting: </strong>Single-center, multi-surgeon, retrospective cohort study PATIENT SAMPLE: : Adult, circumferential lumbar fusions ≤ 5 interbody levels with DXA, CTs, radiographs, and at least 1 year of follow up.</p><p><strong>Outcome measures: </strong>CTHU at L1, lowest DXA T-score, and postoperative change in disc space height (cage migration) on radiographs METHODS: : Lowest DXA T-scores overall and of the lumbar spine were recorded and categorized, and L1 CTHUs were measured. Interbody fusions were analyzed for subsidence ≥ 2mm on radiographs by a validated, computer vision algorithmic approach. Analysis determined if an association existed between subsidence and CTHU<135 or DXA lowest T-score. Logistic regression analyzed the performance of predicting subsidence by each method.</p><p><strong>Results: </strong>The 127-patient cohort had 82.7% degenerative pathologies, 45.7% males, median age of 60 years, 2.4% osteoporosis on DXA, 44.1% CTHU<135, and 13.4% subsidence. CTHU<135 (p=0.004) and age (p=0.016) were significantly associated with subsidence, however DXA lowest T-score (p=0.550) was not. The odds of subsidence were significant if CTHU<135 for crude and adjusted (OR=4.0, 95% CI 1.2-13.9, p=0.029) comparisons. The odds of subsidence were not significant for DXA_any lowest T-score or DXA_spine lowest T-score (OR=1.8, 95% CI 0.6-4.9, p=0.284 and OR=1.1, 95% CI 0.3-4.1, p=0.920, respectively).</p><p><strong>Conclusion: </strong>CTHU<135 was associated with subsidence while DXA lowest T-score was not in this study of patients with degenerative pathologies. The odds of subsidence were 4.0 times higher for CTHU<135 after controlling for known risks, supporting this cutoff value. This study suggests that CTHU is a more reliable predictor of subsidence than DXA in this primarily degenerative population and is a useful tool for assessing bone quality at the region of interest when planning lumbar surgery.</p>","PeriodicalId":49484,"journal":{"name":"Spine Journal","volume":" ","pages":""},"PeriodicalIF":4.9,"publicationDate":"2025-03-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143755475","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"The Application Value of Intraoperative Neurophysiological Monitoring in Cervical Spinal Canal Stenosis Decompression Surgery.","authors":"Yongjie Zhang, Jialiang Li, Yang Yuan, Yuchen Wang, Dageng Huang, Huaguang Qi","doi":"10.1016/j.spinee.2025.03.029","DOIUrl":"https://doi.org/10.1016/j.spinee.2025.03.029","url":null,"abstract":"&lt;p&gt;&lt;strong&gt;Background context: &lt;/strong&gt;Although intraoperative neurophysiological monitoring (IONM) has been widely recognized and used in spine surgery, its characteristics vary for different types of spinal disorders, necessitating the development of tailored monitoring strategies. Cervical spinal stenosis presents complex clinical symptoms and carries significant surgical risks, creating a critical need to clarify the monitoring features, alert patterns, and their relationship with outcomes in such surgeries. A comprehensive assessment and the development of a refined IONM monitoring plan throughout the perioperative period is an important direction for future research.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Purpose: &lt;/strong&gt;This study aims to investigate the influencing factors of intraoperative neurophysiological monitoring (IONM) alarm events in patients with cervical spinal canal stenosis and to evaluate the predictive value of different IONM alarm patterns on neurological recovery following decompression surgery.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Design: &lt;/strong&gt;Retrospective study PATIENT SAMPLES: This analysis included 1,622 patients who underwent cervical spinal canal decompression surgery and had complete IONM monitoring data between February 2017 and December 2022.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Outcome measures: &lt;/strong&gt;The preoperative and postoperative neurological status of the patients was assessed using the modified Japanese Orthopaedic Association (mJOA) score. The primary IONM alarm indicators included somatosensory evoked potentials (SSEP) and transcranial motor evoked potentials (MEP), compared to the preoperative baseline.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Methods: &lt;/strong&gt;Logistic regression was employed to analyze the correlation between preoperative diagnostic risk factors and intraoperative alarm events. Additionally, a multifactorial interaction analysis was performed to determine the relationship between IONM changes and the reversibility of alarms with the six-month mJOA recovery rate.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Results: &lt;/strong&gt;Preoperative diagnoses of the ligamentum flavum hypertrophy and/or ossification of the posterior longitudinal ligament, combined with an mJOA score &lt;12, were identified as high-risk factors for intraoperative alarms. The sensitivity of alarms in the high-risk group was 100%, with a positive predictive value of 90.6%; in the low-risk group, the sensitivity was 91.7%, with a positive predictive value of 40.74%. Variance analysis indicated that the mJOA improvement rate at six months was significantly lower in patients with irreversible IONM alarms compared to those with reversible alarms. Interaction analysis suggested that the reversibility of intraoperative alarm events was a principal predictor of postoperative outcomes, while risk factors for alarms had predictive value only in patients with irreversible alarms.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Conclusions: &lt;/strong&gt;In patients with cervical spinal canal stenosis caused by disc degeneration, the presence of ligamentum flavum hypertrophy, ossific","PeriodicalId":49484,"journal":{"name":"Spine Journal","volume":" ","pages":""},"PeriodicalIF":4.9,"publicationDate":"2025-03-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143744325","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Adverse respiratory events during treatment with gabapentin and opioids among older adults with spine-related conditions: a propensity-matched cohort study in the US Medicare population.","authors":"Laura S Gold, Patrick J Heagerty, Ryan N Hansen, Janna L Friedly, Richard A Deyo, Michele Curatolo, Judith A Turner, Sean D Rundell, Jeffrey G Jarvik, Pradeep Suri","doi":"10.1016/j.spinee.2025.03.014","DOIUrl":"10.1016/j.spinee.2025.03.014","url":null,"abstract":"&lt;p&gt;&lt;strong&gt;Background context: &lt;/strong&gt;Recent work indicates no increased mortality risk with concurrent gabapentin and opioid use when using an active comparator control design. However, concurrent gabapentin and opioid prescriptions have been associated with greater risk of respiratory depression in some studies.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Purpose: &lt;/strong&gt;To compare the risk of respiratory events among Medicare enrollees with histories of spine-related diagnoses treated with gabapentin + opioids vs those treated with tricyclic antidepressants (TCA) or duloxetine + opioids. We hypothesized that enrollees treated with gabapentin + opioids would have increased risk of adverse respiratory events compared to those treated with an active control + opioids.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Study design/setting: &lt;/strong&gt;Propensity score-matched cohort study with an incident user, active comparator (TCA/duloxetine) control design. The primary analysis included those who concurrently (within 30 days) filled ≥1 incident gabapentin + ≥1 opioid or ≥1 incident TCA/duloxetine + ≥1 opioid prescription.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Patient sample: &lt;/strong&gt;U.S. Medicare beneficiaries with histories of spine-related diagnoses 2017-2019. People treated with gabapentin + opioids (n=66,860) were matched on demographic and clinical factors to people treated with TCAs/duloxetine + opioids (n=66,860).&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Outcome measures: &lt;/strong&gt;Time to a composite respiratory outcome consisting of mechanical ventilation, intubation, respiratory failure, pneumonia, or acute respiratory distress syndrome.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Methods: &lt;/strong&gt;Cox proportional hazard regression was used to estimate adjusted hazard ratios (aHRs) and 95% confidence intervals (95% CIs).&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Results: &lt;/strong&gt;Among 133,720 Medicare enrollees (median age 73.3 years; 66.9% female), 6277 (4.7%) experienced respiratory events before the end of follow-up. A total of 3469 (5.2%) of people who were treated with gabapentin + opioids (median initial dose/day of gabapentin was 300 mg) had respiratory events compared to 2808 (4.2%) of those treated with an active control + opioids. The increased risk in those treated with gabapentin + opioids was statistically significant after adjustment (HR 1.19; 95% CI 1.13, 1.25; p&lt;0.0001). The most common respiratory events were pneumonia (3.7% of people in the gabapentin + opioids group versus 3.0% of people in the TCA/duloxetine + opioids group) and respiratory failure (2.3% in the gabapentin + opioids group versus 1.8% in the TCA/duloxetine + opioids group). Results were similar in analyses (a) restricted to ≤30-day follow-up and (b) that required ≥2 fills of each prescription.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Conclusions: &lt;/strong&gt;While recent work indicates no increased mortality risk with concurrent gabapentin and opioid use in this population, the current findings suggest clinicians should exercise caution in prescribing gabapentin to older adults with spine conditions who are using opioids, due to possible impacts o","PeriodicalId":49484,"journal":{"name":"Spine Journal","volume":" ","pages":""},"PeriodicalIF":4.9,"publicationDate":"2025-03-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143744259","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"External validation of a machine learning prediction model for massive blood loss during surgery for spinal metastases: a multi-institutional study using 880 patients.","authors":"Daniël C de Reus, R Harmen Kuijten, Priyanshu Saha, Diego A Abelleyra Lastoria, Aliénor Warr-Esser, Charles F C Taylor, Olivier Q Groot, Darren Lui, Jorrit-Jan Verlaan, Daniel G Tobert","doi":"10.1016/j.spinee.2025.03.018","DOIUrl":"https://doi.org/10.1016/j.spinee.2025.03.018","url":null,"abstract":"&lt;p&gt;&lt;strong&gt;Background context: &lt;/strong&gt;A machine learning (ML) model was recently developed to predict massive intraoperative blood loss (&gt;2500mL) during posterior decompressive surgery for spinal metastasis that performed well on external validation within the same region in China.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Purpose: &lt;/strong&gt;We sought to externally validate this model across new geographic regions (North America and Europe) and patient cohorts.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Study design: &lt;/strong&gt;Multi-institutional retrospective cohort study PATIENT SAMPLE: We retrospectively included patients 18 years or older who underwent decompressive surgery for spinal metastasis across three institutions in the United States, the United Kingdom and the Netherlands between 2016 and 2022. Inclusion and exclusion criteria were consistent with the development study with additional inclusion of (1) patients undergoing palliative decompression without stabilization, (2) patients with multiple myeloma and lymphoma, and (3) patients who continued anticoagulants perioperatively.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Outcome measures: &lt;/strong&gt;Model performance was assessed by comparing the incidence of massive intraoperative blood loss (&gt;2,500mL) in our cohort to the predicted risk generated by the ML model. Blood loss was quantified in 7 ways (including the formula from the development study) as no gold standard exists, and the method in the development paper was not clearly defined. We estimated blood loss using the anesthesia report, and calculated it using transfusion data, and preoperative and postoperative hematocrit levels.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Methods: &lt;/strong&gt;The following five input variables necessary for risk calculation by the ML model were manually collected: tumor type, smoking status, ECOG score, surgical process, and preoperative platelet count. Model performance was assessed on overall fit (Brier score), discriminatory ability (area under the curve (AUC)), calibration (intercept & slope), and clinical utility (decision curve analysis (DCA)) for the total validation cohort, and for the North American and European cohorts separately. A sub-analysis, excluding the additional included patient groups, assessed the predictive model's performance with the same inclusion and exclusion criteria as the development cohort.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Results: &lt;/strong&gt;A total of 880 patients were included with a massive blood loss range from 5.3% to 18% depending on which quantification method was used. Using the most favorable quantification method, the predictive model overestimated risk in our total validation cohort and scored poorly on overall fit (Brier score: 0.278), discrimination (AUC: 0.631 [95%CI: 0.583, 0.680]), calibration, (intercept: -2.082, [95%CI: -2.285, -1.879]), slope: 0.283 [95%CI: 0.173, 0.393]), and clinical utility, with net harm observed in decision curve analysis from 20%. Similar poor performance results were observed in the sub-analysis excluding the additional included patients (n=676) and when ana","PeriodicalId":49484,"journal":{"name":"Spine Journal","volume":" ","pages":""},"PeriodicalIF":4.9,"publicationDate":"2025-03-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143744278","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"GPT4LFS (generative pre-trained transformer 4 omni for lumbar foramina stenosis): enhancing lumbar foraminal stenosis image classification through large multimodal models.","authors":"Elzat Elham-Yilizati Yilihamu, Fan-Shuo Zeng, Jun Shang, Jin-Tao Yang, Hai Zhong, Shi-Qing Feng","doi":"10.1016/j.spinee.2025.03.011","DOIUrl":"https://doi.org/10.1016/j.spinee.2025.03.011","url":null,"abstract":"&lt;p&gt;&lt;strong&gt;Background context: &lt;/strong&gt;Lumbar foraminal stenosis (LFS) is a common spinal condition that requires accurate assessment. Current magnetic resonance imaging (MRI) reporting processes are often inefficient, and while deep learning has potential for improvement, challenges in generalization and interpretability limit its diagnostic effectiveness compared to physician expertise.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Purpose: &lt;/strong&gt;The present study aimed to leverage a multimodal large language model to improve the accuracy and efficiency of LFS image classification, thereby enabling rapid and precise automated diagnosis, reducing the dependence on manually annotated data, and enhancing diagnostic efficiency.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Study design/setting: &lt;/strong&gt;Retrospective study conducted from April 2017 to March 2023.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Patient sample: &lt;/strong&gt;Sagittal T1-weighted MRI data for the lumbar spine were collected from 1,200 patients across three medical centers. A total of 810 patient cases were included in the final analysis, with data collected from seven different MRI devices.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Outcome measures: &lt;/strong&gt;Automated classification of LFS using the multi modal large language model. Accuracy, sensitivity, Specificity and Cohen's Kappa coefficient were calculated.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Methods: &lt;/strong&gt;An advanced multimodal fusion framework GPT4LFS was developed with the primary objective of integrating imaging data and natural language descriptions to comprehensively capture the complex LFS features. The model employed a pre-trained ConvNeXt as the image processing module for extracting high-dimensional imaging features. Concurrently, medical descriptive texts generated by the multimodal large language model GPT-4o and encoded and feature-extracted using RoBERTa were utilized to optimize the model's contextual understanding capabilities. The Mamba architecture was implemented during the feature fusion stage, effectively integrating imaging and textual features and thereby enhancing the performance of the classification task. Finally, the stability of the model's detection results was validated by evaluating classification task metrics, such as the accuracy, sensitivity, specificity, and Kappa coefficients.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Results: &lt;/strong&gt;The training set comprised 6,299 images from 635 patients, the internal test set included 820 images from 82 patients, and the external test set was composed of 930 images from 93 patients. The GPT4LFS model demonstrated an overall accuracy of 93.7%, sensitivity of 95.8%, and specificity of 94.5% in the internal test set (Kappa = 0.89,95% confidence interval (CI): 0.84-0.96, p&lt;.001). In the external test set, the overall accuracy was 92.2%, with a sensitivity of 92.2% and a specificity of 97.4% (Kappa = 0.88, 95% CI: 0.84-0.89, p&lt;.001). Both the internal and external test sets showed excellent consistency in the model. After the article is published, we will make the full code publicly available on GitHub.&lt;/p&gt;&lt;p&gt;&lt;s","PeriodicalId":49484,"journal":{"name":"Spine Journal","volume":" ","pages":""},"PeriodicalIF":4.9,"publicationDate":"2025-03-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143744279","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Can Treatment with Human Mesenchymal Stem Cells Rescue the Degenerative Disc Phenotype? An in Vitro Pilot Study of Induced Cytokine Expression.","authors":"Jonathan Dalton, Rajkishen Narayanan, Robert J Oris, Teeto Ezeonu, Evan Bradley, Jose A Canseco, Alexander R Vaccaro, John D Koerner, Dessislava Markova, Christopher Kepler","doi":"10.1016/j.spinee.2025.03.026","DOIUrl":"https://doi.org/10.1016/j.spinee.2025.03.026","url":null,"abstract":"&lt;p&gt;&lt;strong&gt;Background context: &lt;/strong&gt;Given the relatively low cell density in degenerative discs, strategies intended to bolster disc cellularity through stem cell injections have come into clinical use. Stem cell therapy is meant to provide a source of viable disc cells that can promote a healthy disc phenotype. Nevertheless, there is a limited understanding of the mechanisms through which stem cell therapy impacts degeneration.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Purpose: &lt;/strong&gt;The objectives of this pilot study were: 1) to evaluate gene expression changes associated with an in vitro induced degenerative phenotype in human nucleus pulposus (NP) cells, 2) to co-culture these degenerative NP cells with human mesenchymal stem cells (hMSCs) and investigate the impact this has on gene expression, 3) to investigate possible mechanisms by which hMSCs may impact the degenerative phenotype.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Study design: &lt;/strong&gt;Laboratory study.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Methods: &lt;/strong&gt;NP cells were isolated and cultured from patients undergoing anterior lumbar interbody fusion for degenerative disc disease. A degenerative phenotype was induced in cultured NP cells by treatment with an inflammatory protocol (10pg/ml IL-1β and 100pg/ml TNF-α) for 7 days. Gene expression of Treated NP cells was compared to Untreated NP cells via reverse transcriptase polymerase chain reaction. NP cells were then co-cultured with hMSCs in vitro and treated with the inflammatory protocol. Gene expression of Treated NP cells co-cultured with hMSCs was compared to Treated NP cells alone. Preliminary co-culture data demonstrated that IL-10 was uniquely and dramatically upregulated. Therefore, gene expression of Treated NP cells exposed to IL-10 for 24 hours was compared to Treated NP cells alone.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Results: &lt;/strong&gt;Treated NP cells compared to Control NP cells showed upregulation of numerous pro-inflammatory cytokines, including CXCL5, IL-8, and IL-6 and downregulation of several anti-inflammatory cytokines, including IL-10. After co-culture of Treated NP cells with hMSCs, a significant increase in gene expression was identified in IL-10 (+15.34 fold), BMP-6 (+2.32 fold), and LIF (+2.14 fold). A significant decrease in gene expression (p &lt; 0.05) was seen in CCL7 (-2.03) and CXCL12 (-1.67). Exposure of Treated NP cells to IL-10 resulted in upregulation of COL-2 (+1.55 fold, p=0.013) and downregulation of IL-8 (-1.4 fold), CXCL-5 (-1.58 fold,), and MMP-3 (-2.02 fold).&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Conclusion: &lt;/strong&gt;This in vitro pilot study shows that co-culture of degenerative phenotype NP cells with hMSCs produces multiple gene regulatory changes associated with an anti-inflammatory phenotype. Additionally, exposure of degenerative phenotype NP cells to IL-10 produces gene regulation associated with both anti-inflammatory and pro-extracellular matrix effects.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Clinical significance: &lt;/strong&gt;These findings provide mechanistic support for the use of stem cell therapy as a strategy to decrea","PeriodicalId":49484,"journal":{"name":"Spine Journal","volume":" ","pages":""},"PeriodicalIF":4.9,"publicationDate":"2025-03-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143744273","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Spinal manifestations of diffuse large B-cell lymphoma.","authors":"Aymeric Amelot, Louis-Marie Terrier, Gabrielle Cognacq, Lotfi Benboubker, Christophe Destrieux, Ilyess Zemmoura, Patrick Francois, Mourad Aggad, Bertrand Mathon","doi":"10.1016/j.spinee.2025.03.025","DOIUrl":"https://doi.org/10.1016/j.spinee.2025.03.025","url":null,"abstract":"<p><strong>Background context: </strong>Spinal diffuse large B-cell lymphoma (DLBCL) can be divided into two categories: primary non-Hodgkin's lymphoma (PNHL) and metastases from disseminated DLBCL. Prognostic factors of spinal DLBCL metastases seem to differ from those of other spine metastases and PNHL, although the data in the literature remains scarce.</p><p><strong>Purpose: </strong>This study aims at investigating prognostic factors associated with overall survival (OS) in patients with spine DLBCL metastases.</p><p><strong>Study design: </strong>a retrospective study PATIENT SAMPLE: 371 patients treated for DLBCL, including 62 cases of spine DLBCL metastases OUTCOME MEASURES: Patient demographics were collected with survival.</p><p><strong>Methods: </strong>This study is based on consecutive prospective population of, between January 2015 and 2019.</p><p><strong>Results: </strong>The median age of the 371 patients was 68.4 years (range 19.1 to 94.0 years) and 58.8% were males (218 patients). The median OS for our whole series was 82.06 months (SD 11.2.), and 53.0 months (SD 41.2, p=0.622) for the 62 patients with spine DLBCL metastases. The mean duration between DLBCL diagnosis and development of spine metastases (SpM) was 9.0 months (range 0.0-160.8 months). Cox multivariate proportional hazard model identified ECOG <2 [HR: 0.O59, 95 % CI 0.019-0.075; p< 0.0001], age <40 years [HR: 0.206, 95 % CI 0.08-0.506; p =0.001], and IPI score ≤ 2 [HR: 0.472, 95 % CI 0.03-2.104; p =0.001] as predictors of longer survival. In contrast, age >80 years [HR: 2.198, 95 % CI 1.481-3.261; p < 0.0001], IPI score > 4 [HR: 3.232, 95 % CI 1.765-4.654; p =0.008] were independent poor prognostic factors of survival.</p><p><strong>Conclusion: </strong>Spinal lesions in DLBCL are metastatic in nature whereas spine PNHL, similar to multiple myeloma, appears to be a primary spinal malignancy. The main prognostic factors of DLBCL spine metastases are those of the primary disease itself, and should be considered before spinal surgery.</p>","PeriodicalId":49484,"journal":{"name":"Spine Journal","volume":" ","pages":""},"PeriodicalIF":4.9,"publicationDate":"2025-03-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143744324","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}