The value of quantitative analysis of radionuclide bone SPECT/CT imaging in vertebral compression fracture: a retrospective study.

BACKGROUND: Most patients with osteoporosis experience vertebral compression fracture (VCF), which significantly reduces their quality of life. These patients are at a high risk of secondary VCF regardless of treatment. Thus, accurate diagnosis of VCF is important for treating and preventing new fractures. We aimed to investigate the diagnostic and predictive value of quantitative bone imaging techniques for fresh VCF. METHODS: From November 2021 to March 2023, 34 patients with VCF were enrolled in this study, all of whom underwent routine 99mTc-MDP whole-body bone planar scan and local SPECT/CT imaging. The maximum standard uptake value (SUVmax) of 57 fresh VCF, 57 normal adjacent vertebrae, and 19 old VCF were measured. Based on the site of the fracture, fresh VCFs were regrouped into the intervertebral-type group and the margin-type group. Meanwhile, 52 patients who had no bone metastasis or VCFs in their bone scan were assigned to the control group. The SUVmax of 110 normal vertebral bodies and 10 old VCFs in the control group were measured. RESULTS: The median SUVmax of fresh VCF was 19.80, which was significantly higher than the SUVmax of other groups. The receiver operator characteristic (ROC) curve showed that the cut-off value of SUVmax was 9.925 for diagnosing fresh VCF. The SUVmax in the intervertebral-type group was significantly higher than that in the margin-type group (P = 0.04). The SUVmax of normal vertebrae was higher among patients than among the control group (P<0.01), but the CT HU value showed no significant difference. CONCLUSION: The quantitative technique of bone SPECT/CT has a significant value in diagnosing fresh VCF. It can also determine the severity of fractures. In addition, whether the SUVs of the vertebrae adjacent to the fractured vertebra can predict re-fracture deserves further studies.

Surgical management of traumatic spondyloptosis: a case series.

BACKGROUND: Traumatic spondyloptosis is a rare and severe spinal injury characterized by complete anterior translation of one vertebra over another, often resulting in debilitating neurological deficits. CASE PRESENTATION: We present two cases of traumatic spondyloptosis and elaborate on the clinical presentation, management, and follow-up improvement. The first case is a 30-year-old Nepalese man who sustained traumatic spondyloptosis following a blunt force injury to his back while engaged in tree-cutting activities. The patient presented with severe back pain, left lower limb paralysis, and neurological deficits (consistent with American Spinal Injury Association grade C). Radiographic evaluation revealed total anterior dislocation of the L4 vertebral body over L5, accompanied by fractures of the superior endplates of both vertebrae. The second case is a 35-year-old Nepalese female who presented with back pain and lower limb paralysis following a fall from a 300-m cliff, exhibiting tenderness and ecchymosis in the mid-back region. Radiological examination revealed D12 vertebra translation over L1 with fracture, categorized as American Spinal Injury Association grade A. Both cases were surgically managed and stabilized. CONCLUSION: These cases emphasize the importance of a comprehensive approach to trauma management as well as prompt recognition, and early surgical management in optimizing outcomes for patients with traumatic spondyloptosis.

When Is a Clay-Shoveler's Fracture, Not a Clay-Shoveler's Fracture?: A Case Report.

CASE: A 71-year-old man with a history of C5-7 anterior cervical discectomy and fusion (ACDF) sustained a C7 spinous process fracture after falling from a ladder. He was initially managed nonoperatively but developed anterolisthesis and kyphosis at C7-T1 with left hand weakness over the course of 11 days. Surgical treatment included spinous process wiring and C5-T3 posterior fusion. At 14-month follow-up, he demonstrated resolution of pain and returned motor function. CONCLUSION: The patient's ACDF likely created a longer lever arm, allowing the force of his fall to be concentrated at C7-T1. Patients with a suspected Clay-Shoveler's fracture require close follow-up.

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.

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.

Suboptimal osteoporosis care in hospitalized patients: a retrospective analysis of vertebral compression fractures detected on computed tomography.

Vertebral compression fractures (VCFs) are the most common osteoporotic fractures. Only 1/3 of patients with VCFs are clinically diagnosed. In our institution, the Fracture Liaison Service (FLS) was launched in 2017 to improve osteoporosis management for hospitalized patients. (1) To assess osteoporosis awareness among medical providers for emergency department (ED)/hospitalized patients aged 50 or greater; (2) To estimate the rate of FLS consults or referrals to primary care providers (FLS/PCP) by primary teams. A centralized radiology system was used to examine all thoracic and lumbar computed tomography (CT) scans conducted between June 1, 2017 and June 1, 2022. 449 studies were identified with the radiologic impression "compression fracture". 182 studies were excluded after manual chart review. 267 hospitalizations/ED visits with lumbar and/or thoracic spine CT scans were included. Referrals to FLS (26) or PCP (27) were made in 53 cases (~ 20% of the total). In the ED subgroup (131 hospitalizations), only 17 patients had FLS/PCP referrals. The "compression fracture" was mentioned in 227 (85%) discharge notes (any part), while "osteoporosis" was mentioned in only 74 (28%) hospitalizations. A statistically significant difference was found between the two groups when "osteoporosis" was mentioned in the "assessment and plan" section (p = 0.02). Our data show that the overall osteoporosis care for affected patients is suboptimal. Medical providers often overlook the presence of osteoporosis, leading to a lack of consultation with the FLS of referral to PCPs for further evaluation and treatment.

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.

Evaluation and analysis of risk factors for adverse events of the fractured vertebra post-percutaneous kyphoplasty: a retrospective cohort study using multiple machine learning models.

BACKGROUND: Adverse events of the fractured vertebra (AEFV) post-percutaneous kyphoplasty (PKP) can lead to recurrent pain and neurological damage, which considerably affect the prognosis of patients and the quality of life. This study aimed to analyze the risk factors of AEFV and develop and select the optimal risk prediction model for AEFV to provide guidance for the prevention of this condition and enhancement of clinical outcomes. METHODS: This work included 383 patients with primary osteoporotic vertebral compression fracture (OVCF) who underwent PKP. The patients were grouped based on the occurrence of AEFV postsurgery, and data were collected. Group comparisons and correlation analysis were conducted to identify potential risk factors, which were then included in the five prediction models. The performance indicators served as basis for the selection of the best model. RESULTS: Multivariate logistic regression analysis revealed the following independent risk factors for AEFV: kissing spine (odds ratio (OR) = 8.47, 95% confidence interval (CI) 1.46-49.02), high paravertebral muscle fat infiltration grade (OR = 29.19, 95% CI 4.83-176.04), vertebral body computed tomography value (OR = 0.02, 95% CI 0.003-0.13, P < 0.001), and large Cobb change (OR = 5.31, 95% CI 1.77-15.77). The support vector machine (SVM) model exhibited the best performance in the prediction of the risk of AEFV. CONCLUSION: Four independent risk factors were identified of AEFV, and five risk prediction models that can aid clinicians in the accurate identification of high-risk patients and prediction of the occurrence of AEFV were developed.

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.

[How well does artificial intelligence detects fractures in the cervical spine on CT?] / Hoe goed detecteert AI fracturen van de cervicale wervelkolom op CT?

OBJECTIVE: To compare diagnostic accuracy of artificial intelligence (AI) for cervical spine (C-spine) fracture detection on CT with attending radiologists. DESIGN: Retrospective, diagnostic accuracy study. METHODS: AI analyzed 2368 scans from patients screened for C-spine fracture with CT (2007-2014, fracture prevalence 9.3%). With the use of a validated reference standard, which includes information on injuries in need of stabilizing therapy (IST), diagnostic accuracy of AI and radiologists was calculated and subsequently compared. RESULTS: Median age was 48 years. AI detected 158/221 fractures and radiologists 195/221, with a sensitivity of respectively 71.5% and 88.2% (p<0.001). Specificity of the AI and the radiologists was comparable: 98.6% and 99.2% (p=0.07). Of the fractures undetected by AI, 30/63 were an IST versus 4/26 for radiologists. AI detected 22/26 scans with fractures undetected by radiologists. CONCLUSION: Compared to attending radiologists, AI has a lower sensitivity and misses more ISTs; however, it detected most fractures undetected by the radiologists, including ISTs.

Clinical significance of modified unilateral puncture percutaneous vertebroplasty guided by 3D- printed guides in the treatment of osteoporotic vertebral compression fractures: a retrospective study.

OBJECTIVE: To investigate the clinical significance of using 3D printing guides in modified unilateral puncture percutaneous vertebroplasty (PVP) for the treatment of osteoporotic vertebral compression fractures (OVCF), and to explore a new method for preventing paravertebral vein leakage during PVP in conjunction with a previous study of the optimal puncture-side bone cement/vertebral volume ratio(PSBCV/VV%). METHODS: This retrospective study analyzed 99 patients who underwent unilateral puncture PVP between January 2023 and December 2023. Patients were divided into a guide plate group (46 patients) and a conventional group (53 patients). The guide plate group underwent modified unilateral puncture PVP with the guidance of 3D printing guides, while the conventional group underwent unilateral puncture PVP using the conventional pedicle approach. The distribution of bone cement, surgical outcomes, and the occurrence of cement leakage into paravertebral veins were observed in both groups. RESULTS: The guide plate group had significantly shorter operating time and required fewer fluoroscopies compared to the conventional group. The amount of bone cement volume (BCV) used in the guide plate group was higher, but the amount of bone cement volume on the puncture side(PSBCV), the PSBCV/VV%, and the rate of paravertebral vein leakage were lower in the guide plate group compared to the conventional group (P < 0.05). Within each group, significant improvements in anterior vertebral margin height, Cobb angle, visual analog scale (VAS) score, and Oswestry Disability Index (ODI) were observed at 1 day and 1 month postoperatively compared to preoperative values (P < 0.05). CONCLUSION: Using 3D printing guides in modified unilateral puncture PVP is a safe and effective method for treating OVCF. And it has the advantages of short operation time, less fluoroscopy, even distribution of bone cement, and a low rate of paravertebral vein leakage.

An Innovative Deep Learning Approach to Spinal Fracture Detection in CT Images.

AIM: Spinal fractures, particularly vertebral compression fractures, pose a significant challenge in medical imaging due to their small-scale nature and blurred boundaries in Computed Tomography (CT) scans. However, advanced deep learning models, such as the integration of the You Only Look Once (YOLO) V7 model with Efficient Layer Aggregation Networks (ELAN) and Max-Pooling Convolution (MPConv) architectures, can substantially reduce the loss of small-scale information during computational processing, thus improving detection accuracy. The purpose of this study is to develop an innovative deep learning approach for detecting spinal fractures, particularly vertebral compression fractures, in CT images. METHODS: We proposed a novel method to precisely identify spinal injury using the YOLO V7 model as a classifier. This model was enhanced by integrating ELAN and MPConv architectures, which were influenced by the Receptive Field Learning and Aggregation (RFLA) small object recognition framework. Standard normalization techniques were utilized to preprocess the CT images. The YOLO V7 model, integrated with ELAN and MPConv architectures, was trained using a dataset containing annotated spinal fractures. Additionally, to mitigate boundary ambiguities in compressive fractures, a Theoretical Receptive Field (TRF) based on Gaussian distribution and an Effective Receptive Field (ERF) were used to capture multi-scale features better. Furthermore, the Wasserstein distance was employed to optimize the model's learning process. A total of 240 CT images from patients diagnosed with spinal fractures were included in this study, sourced from Ningbo No.2 Hospital, ensuring a robust dataset for training the deep learning model. RESULTS: Our method demonstrated superior performance over conventional object detection networks like YOLO V7 and YOLO V3. Specifically, with a dataset of 200 pathological images and 40 normal spinal images, our method achieved a 3% increase in accuracy compared to YOLO V7. CONCLUSIONS: The proposed method offers an innovative and more effective approach for identifying vertebral compression fractures in CT scans. These promising findings suggest the method's potential for practical clinical applications, highlighting the significance of deep learning in enhancing patient care and treatment in medical imaging. Future research should incorporate cross-validation and independent validation and test sets to assess the model's robustness and generalizability. Additionally, exploring other deep learning models and methods could further enhance detection accuracy and reliability, contributing to the development of more effective diagnostic tools in medical imaging.

Finite Element Analysis of Bilateral Double Rod Constructs in Thoracolumbar Fractures.

AIM: To evaluate bilateral double rod contructs in thoracolumbar fractures in a Finite Element model MATERIAL and METHODS: A computed tomography of a 35-year old male have been chosen to create a vertebra model and 1/3 of the T12 was removed to create the burst fracture model. In model A, transpedicular polyaxial screws were inserted two levels above and two levels below the burst fracture. On each side the screws were connected with a single rod. In model B, the screws were connected with two rods on each side attached to two lateral connectors. A uniform 150 N axial load and 10 N/m torque was applied on the superior T10. RESULTS: ROM and von Mises stress nephrograms revealed that the bilateral double-rod construct is being the most rigid and that the force on the pedicle screws were significantly lower compared to model A. CONCLUSION: We believe that bilateral double-rod constructs for the stabilization of thoracolumbar fractures have a decreased load on pedicle screws and rods compared to the classic bilateral single rod stabilization system and can lower the risk of implant failure and the risk for secondary complications and revision surgery.

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.

The relationship between treatment-related changes in total hip BMD measured after 12, 18, and 24 mo and fracture risk reduction in osteoporosis clinical trials: the FNIH-ASBMR-SABRE project.

There is a strong association between total hip bone mineral density (THBMD) changes after 24 mo of treatment and reduced fracture risk. We examined whether changes in THBMD after 12 and 18 mo of treatment are also associated with fracture risk reduction. We used individual patient data (n = 122 235 participants) from 22 randomized, placebo-controlled, double-blind trials of osteoporosis medications. We calculated the difference in mean percent change in THBMD (active-placebo) at 12, 18, and 24 mo using data available for each trial. We determined the treatment-related fracture reductions for the entire follow-up period, using logistic regression for radiologic vertebral fractures and Cox regression for hip, non-vertebral, "all" (combination of non-vertebral, clinical vertebral, and radiologic vertebral) fractures and all clinical fractures (combination of non-vertebral and clinical vertebral). We performed meta-regression to estimate the study-level association (r2 and 95% confidence interval) between treatment-related differences in THBMD changes for each BMD measurement interval and fracture risk reduction. The meta-regression revealed that for vertebral fractures, the r2 (95% confidence interval) was 0.59 (0.19, 0.75), 0.69 (0.32, 0.82), and 0.73 (0.33, 0.84) for 12, 18, and 24 mo, respectively. Similar patterns were observed for hip: r2 = 0.27 (0.00, 0.54), 0.39 (0.02, 0.63), and 0.41 (0.02, 0.65); non-vertebral: r2 = 0.27 (0.01, 0.52), 0.49 (0.10, 0.69), and 0.53 (0.11, 0.72); all fractures: r2 = 0.44 (0.10, 0.64), 0.63 (0.24, 0.77), and 0.66 (0.25, 0.80); and all clinical fractures: r2 = 0.46 (0.11, 0.65), 0.64 (0.26, 0.78), and 0.71 (0.32, 0.83), for 12-, 18-, and 24-mo changes in THBMD, respectively. These findings demonstrate that treatment-related THBMD changes at 12, 18, and 24 mo are associated with fracture risk reductions across trials. We conclude that BMD measurement intervals as short as 12 mo could be used to assess fracture efficacy, but the association is stronger with longer BMD measurement intervals.

In this study, we looked at how changes in hip bone density over time relate to the risk of fractures in people taking osteoporosis medications. We analysed data from over 122 000 participants across 22 different clinical trials. We found that the increase in bone density measured after 12, 18, and 24 mo of treatment was linked to the risk of fractures. Specifically, greater improvements in bone density were associated with fewer fractures in the spine, hips, and other bones. Using statistical methods, we calculated the strength of this association. We discovered that the later, we measured BMD in people taking the medication, the stronger the link between improved bone density and reduced fracture risk became. Our findings suggest that bone density measurements after 12 mo of treatment could help predict how well a medication will prevent fractures. However, the best predictions came from bone density changes measured over longer periods.