Open reduction and C1C3 posterior Harms-Goel fixation for unstable Hangman's fracture: technical note.

Hangman's fracture occurs in the elderly following low kinetic energy fall from their height, or in the young during traffic accident. Classically described as bipedicular C2 fracture, Hangman's fracture results from oblique hyperextension-flexion vector forces which can lead to anterior dislocation of C1C2 complex over C3 vertebral body in case of associated damage to C2C3 disc and ligaments. Even though most cases of Hangman's fracture are not displaced (grade 1) and can be managed with orthopaedic treatment using cervical brace, highly displaced cases of Hangman's fractures (some grade 2, all grade 3) require surgical stabilization. Given the limited capabilities of reduction and the odds for mechanical failure of anterior C2C3 discectomy and fusion, we provide a reproducible method using a specific rod persuader on C1 screws aiming to realign and stabilize displaced cases of Hangman fracture using the C1C3 "Harms-Goel" procedure.

Conservative Treatments in the Management of Acute Painful Vertebral Compression Fractures: A Systematic Review and Network Meta-Analysis.

Importance: Osteoporotic vertebral compression fractures (VCFs) frequently cause substantial pain and reduced mobility, posing a major health problem. Despite the critical need for effective pain management to restore functionality and improve patient outcomes, the value of various conservative treatments for acute VCF has not been systematically investigated. Objective: To assess and compare different conservative treatment options in managing acute pain related to VCF. Data Sources: On May 16, 2023, 4 databases-PubMed, Embase, Scopus, and CINAHL-were searched. In addition, a gray literature search within Scopus and Embase was also conducted. Study Selection: Included studies were prospective comparative and randomized clinical trials that assessed conservative treatments for acute VCF. Data Extraction and Synthesis: Data extraction and synthesis were performed by 2 authors according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses for Network Meta-Analyses recommendations. A frequentist graph-theoretical model and a random-effects model were applied for the meta-analysis. Main Outcomes and Measures: Primary outcomes were short-term (4 weeks) pain during activity and long-term (latest available follow-up) nonspecified pain in patients with acute VCF. Results: The study included 20 trials, encompassing 2102 patients, and evaluated various interventions for managing VCF. Calcitonin (standardized mean difference [SMD], -4.86; 95% CI, -6.87 to -2.86) and nonsteroidal anti-inflammatory drugs (NSAIDs; SMD, -3.94; 95% CI, -7.30 to -0.58) were beneficial regarding short-term pain during activity compared with placebo. For long-term nonspecific pain management, bisphosphonates were associated with inferior pain outcomes compared with daily (SMD, 1.21; 95% CI, 0.11 to 2.31) or weekly (SMD, 1.13; 95% CI, 0.05 to 2.21) administration of teriparatide, with no treatment being superior to NSAIDs. The qualitative analysis of adverse events highlighted that typical adverse events associated with these medications were observed. Conclusions and Relevance: NSAIDs and teriparatide may be the preferred treatment options for pain management in acute osteoporotic VCF. Although calcitonin also proved to be beneficial, its safety profile and potential adverse effects restrict its widespread application. The limited evidence on braces and analgesics underscores the urgent need for future research.

Work intensity and fat mass percentage are associated with asymptomatic morphometric vertebral fractures in knee osteoarthritis patients: A cross-sectional study.

Knee osteoarthritis (OA) is a common condition with a prevalence of 365 million individuals globally, and it is an independent risk factor for falls and fractures, notably asymptomatic morphometric vertebral fractures (AMVF). The high prevalence of knee OA, the severity of AMVF, and their combined impacts on quality of life underscore the need for early detection, appropriate treatment and management. To address this, our cross-sectional study aims to identify potential predictive factors associated with AMVF in knee OA patients. Our cohort consisted of 76 patients diagnosed with knee OA, predominantly female (84.2%), of Malay ethnicity (84.2%), and obese (55.3%). In univariable analysis, significant association was found between occupation (moderate or heavy work) and AMVF (p<0.001). Diabetes mellitus comorbidity (p = 0.016) and fat mass percentage (p = 0.027) also demonstrated a significant association with AMVF in knee OA patients. Multivariable logistic regression analysis revealed that an increase in fat mass percentage resulted in decreased AMVF incidence (HR: 0.83, 95% CI: 0.70-0.97; p = 0.018), while occupation (moderate or heavy work) remained a highly significant predictor (HR: 57.76, 95% CI: 4.23-788.57; p = 0.002). These findings support the potential importance of considering occupational activities and body fat composition in managing AMVF among knee OA patients, but further research is required to establish causal relationships.

Analysis of the impact of underlying diseases in the elderly on postoperative re-fractures after osteoporotic compression fractures.

BACKGROUND: Postoperative refracture of osteoporotic compression fractures in the elderly due to underlying illnesses is a complicated matter involving several variables. A multidisciplinary approach involving orthopedics, geriatrics, endocrinology, and rehabilitation medicine is necessary for an investigation of these issues. investigating the impact of older patients' underlying medical conditions on the refracture of osteoporotic compression fractures following surgery. METHODS: A retrospective analysis was conducted on 2383 patients between August 2013 and August 2023. 550 patients with comorbid geriatric underlying diseases were screened, 183 patients underwent refractories, and 367 patients were classified as non-refractories. The patients were then divided into two groups: those undergoing refractories and those not, and the underlying diseases of the patients in both groups were examined using ROC curves and unifactorial and multifactorial logistic regression analyses. RESULTS: Among the patients gathered, the frequency of re-fracture was 33.3%. A statistically significant difference was observed when re-fracture was linked to patients with long-term alcohol consumption, operated vertebrae ≤ 1, hypertension, COPD, diabetes mellitus, stroke sequelae, conservative treatment of coronary heart disease, trauma, mental abnormality, scoliosis, and chronic renal disease. Having hypertension decreased the risk of re-fracture (P = 0.018, OR = 0.548), while alcohol intake ≥ 10years (P = 0.003, OR = 2.165), mental abnormality (P < 0.001, OR = 4.093), scoliosis (P < 0.001, OR = 6.243), chronic kidney disease (P = 0.002, OR = 2.208), and traumatic injuries (P = 0.029, OR = 3.512) were the risk factors examined in a binary logistic regression analysis. The results of multiple linear stepwise regression analysis indicated that re-fracture was more influenced by scoliosis. CONCLUSIONS: Hypertensive disorders were protective factors against the formation of re-fracture, while alcohol intake usage for more than ten years, psychological abnormalities, scoliosis, chronic kidney disease, and trauma were risk factors. Scoliosis had the highest influence on re-fracture.

Synthetic lumbar MRI can aid in diagnosis and treatment strategies based on self-pix networks.

CT and MR tools are commonly used to diagnose lumbar fractures (LF). However, numerous limitations have been found in practice. The aims of this study were to innovate and develop a spinal disease-specific neural network and to evaluate whether synthetic MRI of the LF affected clinical diagnosis and treatment strategies. A total of 675 LF patients who met the inclusion and exclusion criteria were included in the study. For each participant, two mid-sagittal CT and T2-weighted MR images were selected; 1350 pairs of LF images were also included. A new Self-pix based on Pix2pix and Self-Attention was constructed. A total of 1350 pairs of CT and MR images, which were randomly divided into a training group (1147 pairs) and a test group (203 pairs), were fed into Pix2pix and Self-pix. The quantitative evaluation included PSNR and SSIM (PSNR1 and SSIM1: real MR images and Pix2pix-generated MR images; PSNR2 and SSIM2: real MR images and Self-pix-generated MR images). The qualitative evaluation, including accurate diagnosis of acute fractures and accurate selection of treatment strategies based on Self-pix-generated MRI, was performed by three spine surgeons. In the LF group, PSNR1 and PSNR2 were 10.884 and 11.021 (p < 0.001), and SSIM1 and SSIM2 were 0.766 and 0.771 (p < 0.001), respectively. In the ROI group, PSNR1 and PSNR2 were 12.350 and 12.670 (p = 0.004), and SSIM1 and SSIM2 were 0.816 and 0.832 (p = 0.005), respectively. According to the qualitative evaluation, Self-pix-generated MRI showed no significant difference from real MRI in identifying acute fractures (p = 0.689), with a good sensitivity of 84.36% and specificity of 96.65%. No difference in treatment strategy was found between the Self-pix-generated MRI group and the real MRI group (p = 0.135). In this study, a disease-specific GAN named Self-pix was developed, which demonstrated better image generation performance compared to traditional GAN. The spine surgeon could accurately diagnose LF and select treatment strategies based on Self-pix-generated T2 MR images.

Clinical application of a modified wiltse approach in middle and lower thoracic vertebrae: a case-control study of thoracic fracture patients.

BACKGROUND: The Wiltse approach has been extensively employed in thoracolumbar surgeries due to its minimal muscle damage. However, in the middle and lower thoracic spine, the conventional Wiltse approach necessitates the severance of the latissimus dorsi and trapezius muscles, potentially leading to muscular injury. Consequently, we propose a modified Wiltse approach for the middle and lower thoracic vertebrae, which may further mitigate muscular damage. METHODS: From May 2018 to April 2022, 60 patients with spinal fractures in the middle and lower thoracic vertebrae (T5-12) were enrolled in this study. Thirty patients underwent surgery using the modified Wiltse approach (Group A), while the remaining 30 patients received traditional posterior surgery (Group B). The observation indices included operation time, intraoperative blood loss, incision length, number of C-arm exposures, postoperative drainage, postoperative ambulation time, discharge time, as well as preoperative and postoperative Cobb's angle, percentage of anterior vertebral body height (PAVBH), visual analog scale (VAS) Score, and Oswestry disability index (ODI). RESULTS: Compared to the traditional posterior approach, the modified Wiltse approach demonstrated significant advantages in operation time, intraoperative blood loss, length of incision, postoperative ambulation time, postoperative drainage, and discharge time, as well as postoperative VAS and ODI scores. No significant differences were observed between the two groups in terms of number of C-arm exposures, postoperative Cobb's angle, or postoperative PAVBH. CONCLUSION: We propose a modification of the Wiltse approach for the middle and lower thoracic vertebral regions, which may further minimize muscular damage and facilitate the recovery of patients who have undergone surgery in the middle and lower thoracic vertebrae.

Jintiange capsule may have a positive effect on pain relief and functional activity in OVCF patients with PVA: A meta-analysis of randomized control trials.

BACKGROUND: This study aims to systematically evaluate the clinical efficacy and adverse reactions associated with Jintiange capsule (JTG capsule)-assisted percutaneous vertebral augmentation (PVA) in the treatment of osteoporotic vertebral compression fracture (OVCF). METHODS: A comprehensive search was conducted across multiple databases including PubMed, Cochrane Library, EMBASE, Web of Science Database, China Biomedical Database, China VIP Network, China National Knowledge Infrastructure, Wanfang, and VIP Chinese Journal databases until June 1, 2022. Manual searches were also performed in relevant journals. Randomized controlled trials investigating the efficacy of JTG capsule-assisted PVA in the treatment of OVCF were identified and selected for inclusion. The quality of the included studies was assessed using the Cochrane risk bias assessment tool and Jadad scale. Meta-analysis was conducted using Stata MP18 software. RESULTS: A total of 138 literatures were retrieved, and 12 RCTS were finally included after screening, involving 1099 patients. Overall, the quality of the included literature was low, and all the included literatures were randomized controlled experiments, among which 9 were grouped by random number table, and 3 did not specify the random assignment plan. The total effective rate of the experimental group was higher than that of the control group (relative ratio: 1.19, 95% confidence interval: 1.11, 1.26, P = .868, I2 = 0%). The heterogeneity of visual analog score, Oswestry disability index, bone mineral density (BMD) of lumbar vertebrae, BMD of femoral neck and bone-γ-carboxyglutamic acid-containing protein was high. The reasons for the high heterogeneity were the age of patients, the follow-up time and the small sample size. There is publication bias in visual analog score, Oswestry disability index scores, and lumbar spine bone mineral density, and we believe that publication bias may be related to selective reporting of positive results by the authors and selective publication of positive results by the publishers. CONCLUSION: JTG capsule has demonstrated promising outcomes in alleviating the pain experienced by OVCF patients following PVP. Additionally, it has shown efficacy in enhancing postoperative lumbar and back function. Furthermore, JTG capsule has been associated with improvements in postoperative vertebral BMD and serum bone-γ-carboxyglutamic acid-containing protein levels. These findings suggest that JTG capsule could potentially serve as a viable adjunctive treatment option for managing osteoporosis following PVA.

Factors associated with loss of vertebral height and kyphosis correction after intermediate screws in short segment pedicular fixation for type-A fractures of the thoracolumbar spine: A retrospective study.

In this article, we attempted to identify risk factors affecting the loss of vertebral height and kyphosis correction on type A thoracolumbar fractures. Patients with type A thoracolumbar fractures who underwent short segments with intermediate screws at the fracture level management between 2017 and 2022 were included in this study. Clinical factors including patients' demographic characteristics (age, sex), history (smoking, hypertension and/or diabetes), value of height/kyphosis correction, the thoracolumbar injury classification and severity score (TLICS), the load sharing classification (LSC) scores and bone mineral density were collected. Correlation coefficient, simple linear regression analysis and multivariate regression analysis were performed to identify the clinical factors associated with the loss of vertebral height/kyphosis correction. Finally, 166 patients were included in this study. The mean height and kyphosis correction were 21.8% ±â€…7.5% and 9.9°â€…±â€…3.8°, respectively, the values of the loss were 6.5% ±â€…4.0% and 3.9°â€…±â€…1.9°, respectively. Simple linear regression analysis and multivariate regression analysis showed that age, value of height correction, LSC scores and bone mineral density were significantly associated with the loss of vertebral height and kyphosis correction (P < .01) We could draw the conclusion that patients with older age, lower bone mineral density, higher LSC scores and diabetes are at higher risk of vertebral height and kyphosis correction loss increase. For these patients, appropriate clinical measures such as long segment fixation, control of blood glucose, and increase of bone density must be taken to reduce the loss of correction.

[Vertebroplasty for chronic back pain due to osteoporotic vertebral compression fractures]. / Chronische rugpijn na osteoporotische wervelinzakking.

OBJECTIVE: Evidence regarding percutaneous vertebroplasty (PV) for chronic painful osteoporotic vertebral compression fractures (OVCFs) remains limited. To compare pain relief, quality of life, and disability between PV and active control (anesthetic infiltration) interventions for chronic OVCF. DESIGN: Randomized controlled trial. METHODS: This prospective randomized clinical trial was conducted between May 2013 and June 2019 in participants with pain due to OVCF lasting longer than 3 months with bone marrow edema present at MRI. Study participants were randomly assigned to undergo PV (n = 40) or active control intervention (n = 40). The primary outcome was pain severity, assessed with the visual analog scale (VAS) (range, 0-10) during 12 months after treatment. Secondary outcomes included Quality of Life Questionnaire of the European Foundation for Osteoporosis (QUALEFFO) score (range, 0-100) and Roland Morris Disability Questionnaire (RMDQ) score (range, 0-100). Outcomes were analyzed according to a longitudinal multilevel model used to test the difference between groups in change from baseline across follow-up. RESULTS: The mean age of the 80 participants (54 women) was 69 years ± 10 (SD) in the PV group and 71 years ± 10 in the active control group. VAS score was 7.6 (95% CI: 7.0, 8.2) in the PV group and 7.3 (95% CI: 6.9, 7.8) in the active control group at baseline (P = .47) and 3.9 (95% CI: 3.1, 4.8) and 5.1 (95% CI: 4.3, 6.0), respectively, at month 12 (P = .045). At month 12, the group difference from baseline was 1.3 (95% CI: 0.1, 2.6; P = .02) for VAS, 5.2 (95% CI: 0.9, 9.4; P = .02) for QUALEFFO, and 7.1 (95% CI: -3.3, 17.5; P = .18) for RMDQ, favoring the PV group. CONCLUSION: In the treatment of pain caused by chronic OVCFs, PV is more effective for pain relief and quality of life improvement than anesthetic injection alone, with similar improvement for disability between the groups.

Robotic Placement of 3 Trans-Sacral TransiIliac Screws Through a Single Corridor for U-Type Sacrum Fracture: A Case Report.

CASE: We present a case of robot-assisted placement of 3 trans-sacral transiliac screws through a single corridor for an unstable U-type sacral fracture in a 95-year-old woman. She had persistent pain and inability to mobilize with physical therapy. At 3-month follow-up, the patient had evidence of interval healing and stable hardware and was able to return to her prior functional baseline. CONCLUSION: We demonstrate successful utilization of robotics to place 3 trans-sacral transiliac screws in a single corridor for fixation of an unstable pelvic ring injury. This technique was used to overcome challenges with visualization and implant placement.

Odontoid Fracture with Locked Posterolateral Atlantoaxial Dislocation: A Case Report and Review of Literature.

CASE: Odontoid fractures with atlantoaxial dislocations are rare injuries. We report a case of a 41-year-old man with a Type 2 odontoid fracture with locket facet and posterolateral dislocation. He underwent single-stage C1-C4 posterior fixation and fusion, and at 2-year follow-up, he is symptom-free without any residual pain. Follow-up radiograph and CT scan show healed odontoid fracture with posterior fusion. CONCLUSION: This case highlights successful management of a complex odontoid fracture by a single-stage posterior surgery. Closed reduction is usually unsuccessful, and open reduction using posterior approach is preferable.

Effect of acupuncture anesthesia on inflammatory response and cellular immunity in patients with osteoporotic fractures.

Osteoporotic vertebral compression fractures (OVCFs) occur frequently in the elderly, with percutaneous vertebroplasty (PVP) being the major clinical treatment at present. How to improve the patient's surgical cooperation while ensuring surgical safety is the focus of clinical research. This study explores the influence of acupuncture anesthesia (AA) on the safety, inflammatory response, and cellular immunity of OVCF patients undergoing PVP, which may provide a more reliable safety guarantee for future treatment of OVCFs. The results showed that patients using AA had lower postoperative Visual Analogue Scale (VAS) scores and incidence of postoperative adverse reactions, a smaller anesthetic dosage, but an extended duration of anesthesia; moreover, the postoperative inflammatory response was markedly alleviated and the stability of T lymphocyte subsets was obviously enhanced. Therefore, AA has high clinical application value in PKP treatment of OVCFs in the future.

Subject-level spinal osteoporotic fracture prediction combining deep learning vertebral outputs and limited demographic data.

Automated screening for vertebral fractures could improve outcomes. We achieved an AUC-ROC = 0.968 for the prediction of moderate to severe fracture using a GAM with age and three maximal vertebral body scores of fracture from a convolutional neural network. Maximal fracture scores resulted in a performant model for subject-level fracture prediction. Combining individual deep learning vertebral body fracture scores and demographic covariates for subject-level classification of osteoporotic fracture achieved excellent performance (AUC-ROC of 0.968) on a large dataset of radiographs with basic demographic data. PURPOSE: Osteoporotic vertebral fractures are common and morbid. Automated opportunistic screening for incidental vertebral fractures from radiographs, the highest volume imaging modality, could improve osteoporosis detection and management. We consider how to form patient-level fracture predictions and summarization to guide management, using our previously developed vertebral fracture classifier on segmented radiographs from a prospective cohort study of US men (MrOS). We compare the performance of logistic regression (LR) and generalized additive models (GAM) with combinations of individual vertebral scores and basic demographic covariates. METHODS: Subject-level LR and GAM models were created retrospectively using all fracture predictions or summary variables such as order statistics, adjacent vertebral interactions, and demographic covariates (age, race/ethnicity). The classifier outputs for 8663 vertebrae from 1176 thoracic and lumbar radiographs in 669 subjects were divided by subject to perform stratified fivefold cross-validation. Models were assessed using multiple metrics, including receiver operating characteristic (ROC) and precision-recall (PR) curves. RESULTS: The best model (AUC-ROC = 0.968) was a GAM using the top three maximum vertebral fracture scores and age. Using top-ranked scores only, rather than all vertebral scores, improved performance for both model classes. Adding age, but not ethnicity, to the GAMs improved performance slightly. CONCLUSION: Maximal vertebral fracture scores resulted in the highest-performing models. While combining multiple vertebral body predictions risks decreasing specificity, our results demonstrate that subject-level models maintain good predictive performance. Thresholding strategies can be used to control sensitivity and specificity as clinically appropriate.

The clinical effect of different vertebral body height restoration rates after percutaneous kyphoplasty for osteoporotic vertebral compression fractures.

OBJECTIVE: This study aimed to evaluate the clinical effect of different vertebral body heights restoration rate after percutaneous kyphoplasty (PKP) for the treatment of osteoporotic vertebral compression fractures (OVCF). METHODS: The patients were divided into two groups according to the height restoration rate of the anterior edge of the vertebral body fracture after PKP operation using X-Ray imaging. The group A was below 80%, and the group B was above 80%. Clinical preoperative and postoperative efficacy (1st day, 1st month, 6th month, and 12th month after surgery) were evaluated according to VAS, Oswestry Disability Index(ODI), Quality of Life Questionnaire of the European Foundation for Osteoporosis(QUALEFFO), and Back Pain Life Disorder Questionnaire(RQD). Simultaneously, the preoperative and postoperative local Cobb angles and changes in the injured vertebrae in the two groups were calculated and analyzed. RESULTS: The postoperative Cobb angle in group A was significantly higher than that in group B. The correction rate in group B was significantly better than that in group A. The VAS, ODI, QUALEFFO, and RQD scores of group B patients were significantly lower than those of patients in group A at each follow-up time point. The correlation coefficients of vertebral body height restoration rate and VAS, ODI, QUALEFFO, and RQD scores at the last follow-up were - 0.607 (P < 0.01), -0.625 (P < 0.01), -0.696 (P < 0.01), and - 0.662 (P < 0.01), respectively. CONCLUSIONS: The results of the correlation analysis between the vertebral body height restoration rate and the above clinical efficacy scores show that increasing the vertebral body anterior height restoration rate is beneficial for pain relief and improves the clinical efficacy of patients. Simultaneously, improving the height restoration rate of the anterior edge of the vertebral body and restoring the normal spinal structure is beneficial for reducing the incidence of refracture of the adjacent vertebral body.

Limitations of Self-Report Tools in Pain Evaluation of Patients With Vertebral Fractures.

Few studies have evaluated the usefulness and limitations of pain assessment using verbal communication tools for acute orthopedic diseases in older patients. The purpose of this study was to assess the rate of usage of the numerical rating scale (NRS), a verbal communication tool, and to identify the characteristics of patients in whom continuous assessment was impossible. We retrospectively examined electronic medical records of patients with acute vertebral fractures who had been admitted to our hospital between April 2018 and March 2020. Continuous pain assessment using the NRS was possible in 43.2% of hospitalized patients with the fractures. The factors preventing continuous pain assessment using the NRS were an advanced age and low Mini-Mental State Examination (MMSE) scores. Based on the receiver-operating characteristic curves, the cutoff age and MMSE score were >85.3 years and <22, respectively. Continuous NRS-based pain assessment is difficult in older adult patients or those with cognitive decline with acute vertebral fractures. In future, a simple observational assessment tool for patients with dementia should be introduced in acute medical care settings.

Functional outcomes in unstable complete subaxial cervical spinal injuries with respiratory paralysis and/or hemodynamic instability.

INTRODUCTION: Subaxial cervical spine injuries (SCSI) can lead to disastrous consequences such as quadriplegia, with/without respiratory paralysis (RP) and hemodynamic instability (HDI). Till date, there is no literature available for reporting outcomes of SCSI patients specifically pertaining to those presenting with RP/HDI and ours is the first study to document the same. METHODS: Retrospective 6-year study from a tertiary trauma centre database including patients >/= 18 years of operated SCSI. Only patients with ASIA A grade with admission RP/HDI and unstable injuries (fractures, subluxations) were included. Patients with ASIA grade B and above, patients with non-osseous injuries (such as disc herniation, central cord syndrome etc.) were excluded. RESULTS: 24 cases were analysed. C5 and C6 levels were the commonest. Vertebral listhesis/subluxation was the predominant radiological finding. The mean age was 47.4 years (22-79 years) and all, except one were males. Fall from height and road traffic accident (RTA) were the most common mechanisms of injury. The most common surgery was anterior discectomy and fusion followed by corpectomy. The overall mortality rate was 22/24 (92)%. Cord edema and hemorrhage had significant association with survival. None of the grade A survivors with HDI/RP showed improvement. The mean FU duration was 18.5 months (range, 16.5-20.5 months). CONCLUSIONS: Subaxial ASIA A cervical spine injuries with pre-operative RP/HDI is an indicator for non-improvement. This is the first study documenting outcome in such patients. The mortality rate in these patients is very high and is an extremely poor prognostic factor for recovery. Hence, surgery in such patients need to be decided judiciously, especially in developing countries that has a significant financial impact on the family members.

Development of a deep learning model for detecting lumbar vertebral fractures on CT images: An external validation.

OBJECTIVE: To develop and externally validate a binary classification model for lumbar vertebral body fractures based on CT images using deep learning methods. METHODS: This study involved data collection from two hospitals for AI model training and external validation. In Cohort A from Hospital 1, CT images from 248 patients, comprising 1508 vertebrae, revealed that 20.9% had fractures (315 vertebrae) and 79.1% were non-fractured (1193 vertebrae). In Cohort B from Hospital 2, CT images from 148 patients, comprising 887 vertebrae, indicated that 14.8% had fractures (131 vertebrae) and 85.2% were non-fractured (756 vertebrae). The AI model for lumbar spine fractures underwent two stages: vertebral body segmentation and fracture classification. The first stage utilized a 3D V-Net convolutional deep neural network, which produced a 3D segmentation map. From this map, region of each vertebra body were extracted and then input into the second stage of the algorithm. The second stage employed a 3D ResNet convolutional deep neural network to classify each proposed region as positive (fractured) or negative (not fractured). RESULTS: The AI model's accuracy for detecting vertebral fractures in Cohort A's training set (n = 1199), validation set (n = 157), and test set (n = 152) was 100.0 %, 96.2 %, and 97.4 %, respectively. For Cohort B (n = 148), the accuracy was 96.3 %. The area under the receiver operating characteristic curve (AUC-ROC) values for the training, validation, and test sets of Cohort A, as well as Cohort B, and their 95 % confidence intervals (CIs) were as follows: 1.000 (1.000, 1.000), 0.978 (0.944, 1.000), 0.986 (0.969, 1.000), and 0.981 (0.970, 0.992). The area under the precision-recall curve (AUC-PR) values were 1.000 (0.996, 1.000), 0.964 (0.927, 0.985), 0.907 (0.924, 0.984), and 0.890 (0.846, 0.971), respectively. According to the DeLong test, there was no significant difference in the AUC-ROC values between the test set of Cohort A and Cohort B, both for the overall data and for each specific vertebral location (all P>0.05). CONCLUSION: The developed model demonstrates promising diagnostic accuracy and applicability for detecting lumbar vertebral fractures.