Archives of Osteoporosis最新文献

Summary

This study assessed osteoporosis risk factors in Armenian women using the FRAX calculator. A BMI below 30, corticosteroid use, prior fractures, thyroid disorders, and diabetes significantly increased fracture risk. Findings highlight the need for national osteoporosis guidelines tailored to Armenia.

Purpose

The aim of the study was to assess the prevalence of clinical risk factors for osteoporotic fractures, as defined by the Armenian-specific FRAX model, among women aged 40 years and older in the Republic of Armenia, and to evaluate their contribution to fracture risk.

Methods

A cross-sectional pilot study was conducted among 265 women aged ≥ 40 years who visited the Osteoporosis Center in Yerevan, Armenia, between September and October 2024. FRAX scores were calculated using Armenia-calibrated data. Participants were randomly selected, and data on FRAX risk factors were collected. Based on fracture probabilities, participants were stratified into low-, moderate-, and high-risk groups. Due to limited numbers in the low-risk group, low and moderate categories were merged for regression analysis. Logistic regression models were used to assess associations between clinical risk factors and high fracture risk.

Results

Among participants, 21.1% were classified as high risk. Multivariate logistic regression identified a body mass index < 30 (OR = 4.3, 95% CI 1.6–11.5), history of prior fractures (OR = 52.5, 95% CI 17.2–160.3), corticosteroid use (OR = 5.0, 95% CI 1.6–15.1), thyroid disease (OR = 8.8, 95% CI 3.1–25.1), and type 2 diabetes (OR = 3.6, 95% CI 1.1–11.0) as independent predictors of high fracture risk. Although age ≥ 60 was associated with increased risk in univariate analysis, it did not retain significance in the multivariate model.

Conclusion

The study findings highlight the importance of developing national osteoporosis prevention and management guidelines that are tailored to the specific characteristics of the local population. Additionally, further research involving larger sample sizes is needed to enhance understanding of the prevalence of these risk factors and to inform targeted public health strategies.

Summary

This study used explainable AI to improve the Danish FREM model for predicting one-year risk of major osteoporotic fractures in over 2.4 million individuals aged ≥ 45. A DART boosting algorithm improved performance (AUC 0.77), with explainable outputs aiding clinical interpretation and guiding referrals for fracture risk assessment.

Purpose

This study aimed to use explainable artificial intelligence to improve the Fracture Risk Evaluation Model (FREM), in the prediction of imminent (one-year) risk of major osteoporotic fractures (MOFs).

Methods

FREM_ML was trained and validated using complete registry data extracted for the Danish population ≥ 45 years without previous osteoporosis diagnoses or treatment (N = 2,438,140). A Dropouts meet multiple Additive Regression Tree (DART) boosting algorithm was used. Predictors of MOFs (2022), automatically extracted for the 15-year lookback period (2007–2021), included hospital diagnoses, filled medication prescriptions, days since the last redemption of medications specific to fall and osteoporosis risk, as well as markers of polypharmacy and multi-morbidity. Stratified analyses were carried out, and model outputs were evaluated in the context of explainable artificial intelligence (AI).

Results

FREM_ML displayed an overall area under the curve (95% confidence interval) of 0.77 (0.76, 0.77) – making it superior to previous versions of FREM. While age and sex were the most relevant predictors of MOF events, advanced feature engineering, including temporal information, contributed to model performance. Importantly, stratified analyses highlighted changing model performance across age groups and poorer prediction performance in males. Shapley Additive exPlanations values, a feature importance metric in explainable AI, facilitated clinical interpretation of relative MOF risk.

Conclusion

The publicly available FREM_ML boosting model, combined with explainable AI, may be an effective decision support approach in a physician’s referral of individuals at high imminent risk of fractures to dual-energy X-ray absorptiometry.

Summary

T-scores alone are inadequate for identifying hip fracture risk. Incorporating FRAX-HFP scores and femoral strength improves risk assessment. Tailored interventions are needed for different ethnicities, with a focus on females due to higher fracture risk. Sex-specific thresholds and targeted prevention strategies are essential for effective fracture prevention.

Background

We investigated the age-related trajectories of areal bone mineral density (aBMD), fracture risk assessment tool (FRAX)–based 10-year probability of hip fracture (FRAX-HFP), trochanteric soft tissue thickness (TSTT), and femoral strength in a multi-ethnic cohort of community-dwelling older adults in Singapore. We also examined the relationship between FRAX-HFP and femoral strength.

Methods

Dual-energy X-ray absorptiometry (DXA) scans were conducted for Singaporean older adults (n = 2235), enrolled in the Population Health and Eye Disease Profile in Elderly Singaporeans (PIONEER) study. aBMD and FRAX-HFP were recorded for the subjects. TSTT was derived from whole-body DXA scans. Femoral strength was derived from DXA-based 3D finite element models. Age-related trajectories were compared for three major ethnicities in Singapore. The relationship between FRAX-HFP and femoral strength was examined.

Results

The study included 2204 older adults (1224 females (73.71 ± 8.37 years), 980 males (73.45 ± 8.34 years)). Age-related trajectories for aBMD, FRAX-HFP, TSTT, and femoral strength indicated that Chinese ethnicity is at high risk for fracture, compared to Indians and Malays. Separately, FRAX-HFP identified 16% of males and 27% of females, and femoral strength identified 3% of males and 1% of females at risk. Both FRAX-HFP score and femoral strength identified 24% of males and 35% of females at risk.

Conclusion

Age-related trajectories for aBMD, FRAX-HFP, TSTT, and femoral strength were found to be consistent with the hip fracture trends in Singapore. FRAX-HFP and femoral strength identified different individuals at risk, indicating that each, either alone or combined with aBMD, could improve the ability to assess hip fracture risk.

Summary

Anti-osteoporosis medications (AOM) play a crucial role in the management of osteoporosis. The utilization of AOM in Iran has shown an overall increasing trend over time, albeit with fluctuations. However, underutilization is a fundamental challenge. Alendronate has consistently been the most widely used medication.

Purpose

Anti-osteoporosis medications (AOM) play a crucial role in the management of osteoporosis. This study investigates the patterns of consumption and expenditures of these agents in Iran.

Methods

The wholesale data of the Iran Food and Drug Administration on the consumption and costs of AOM from 2001 to 2021 was investigated. Annual utilization of medications was converted to defined daily doses (DDDs) per 1000 individuals per day (DID). Adequacy of treatment was also calculated based on the proportion of post-menopausal patients who received a DDD per day of AOMs. Data were analyzed using descriptive statistics and Excel software.

Results

The consumption and expenditure of AOM exhibited a fluctuation pattern. AOM utilization increased from 0.2 to 3.68 DID in 2021, indicating a rise of 19 times. Alendronate was consistently the most widely used medication, maintaining a share of > 90% from 2005 to 2015. With the increasing share of zoledronic acid and denosumab, the share of alendronate reduced in the last 5 years. In 2021, the AOM expenditure amounted to US$ 42.72 million purchasing power parities (PPP). In recent years, denosumab and teriparatide accounted for the majority of the expenditure. Majority of patients (84.25–97.54%) did not receive adequate treatment in the study period.

Conclusions

The utilization of AOM in Iran has shown an overall increasing trend over time, albeit with fluctuations. However, underutilization is a fundamental challenge. Given the impact of these medications on osteoporosis treatment and fracture reduction, it is imperative to identify patients and ensure timely and effective medication treatment.

Summary

Healthcare claims and survey data are increasingly used to assess the osteoporosis burden, but agreement and comparative validity of derived indicators are poorly documented. We show that no single data source can estimate the osteoporosis burden. Instead, coupling data sources allows assessing its burden and associated treatment and knowledge gaps.

Purpose

Healthcare claims data are increasingly used to assess the burden of osteoporosis and fragility fractures, although comparative evidence with other sources and especially self-reported data remains limited. Using the linkage of the French National Health Data System (SNDS) and Health Care and Insurance Survey (ESPS 2010-2014), we evaluated the agreement and comparative validity (concurrent and predictive) of several osteoporosis and clinical fragility fracture indicators and provided comprehensive estimates of their prevalence.

Methods

Individual data from 5039 ESPS participants aged ≥ 25 years were linked to SNDS. Follow-up data included a health self-assessment in 2014 and 5-year occurrence of fractures and mortality. Prevalence was estimated for each indicator (self-reported in ESPS, diagnosis and treatment of osteoporosis, and clinical fragility fractures in SNDS) using several combinations and capture-recapture. Kappa statistics assessed agreement between indicators. Multivariate models evaluated determinants of disagreement between sources and associations of indicators with health outcomes and new fractures (concurrent and predictive validity).

Results

Prevalence estimated by capture-recapture was 7.6% versus 4.1% and 2.2% for self-reported and treated osteoporosis, respectively. Agreement between indicators was slight to moderate. Education, limitation in daily activities, and number of chronic conditions influenced agreement. SNDS indicators had better validity than self-reported osteoporosis, especially for predicting new fractures.

Conclusion

The French healthcare claims database provides valid indicators, although it is insufficient to assess and monitor the burden of osteoporosis in the general population. Coupling these indicators with self-reported data may help overcome these limitations and assess the treatment and knowledge gaps associated with osteoporosis.

Summary

Using nationwide data from 2006 to 2022, we found that vertebral fracture rates in Korea plateaued recently, with a slight decline during COVID-19, while one-year mortality remained unchanged. These findings highlight the growing burden of osteoporotic fractures and underscore the need for targeted prevention and management strategies.

Purpose

Our purposes were to evaluate the trends in the incidence and mortality of vertebral fractures and the effects of COVID-19 between 2006 and 2022 in Korea, using nationwide data from the National Health Insurance Service (NHIS).

Methods

A nationwide dataset was evaluated to identify all new visits to medical institutes for vertebral fractures in men and women aged 50 years or older between 2006 and 2022. Patients were defined using ICD-10 diagnosis codes combined with procedure codes to ensure specificity. The incidence and 1-year mortality rates were calculated. The age- and sex-standardized incidence rates were calculated using the 2020 Korean population as the reference, which demonstrated a plateauing trend in recent years.

Results

While the number of vertebral fractures increased, the incidence plateaued around 2010. Men and women respectively experienced 17,872 and 66,980 vertebral fractures in 2006 and 43,901 and 136,826 in 2022. The crude incidence of vertebral fractures changed from 319.4/100,000 to 408.5/100,000 person-years in men and 1011.1/100,000 to 1155.5/100,000 person-years in women between 2006 and 2022. The 1-year mortality rate after vertebral fractures gradually decreased from 6.5% in 2006 to 6.2% in 2021. There was no remarkable change of mortality during the outbreak of COVID-19. The overall mortality of vertebral fractures in men was about twice as high women as during the whole study period.

Conclusion

The incidence of vertebral fractures remained relatively stable after 2010 in Korea. During the outbreak of COVID-19, incidence of vertebral fractures decreased while the mortality after vertebral fractures was unaffected.

Summary

Rationale: Existing osteoporosis screening tools are inaccurate and inconvenient, prompting the need for a better alternative.

Main result: A machine learning tool (Gradient Boosting) with key factors (weight, age, height) outperformed OST (AUC 0.828 vs 0.781, p < 0.0001) in validation.

Significance

The validated, clinically applicable tool improves osteoporosis screening accessibility and accuracy.

Background

As the first “line of defence” for osteoporosis detection, existing screening tools have low accuracy and are inconvenient to use. Therefore, this study aims to develop a machine-learning-based, clinically applicable, and interpretable osteoporosis screening tool.

Methods

This study included 9405 American participants aged 50 years and older (with the average age of the osteoporosis population in the training set and test set being 72 ± 9 years and 73 ± 8 years, respectively). The study selected 13 clinically accessible indicators as candidate predictive variables, divided the data into a training set and a test set at a ratio of 7:3, used the Lasso for feature selection, compared six statistical and machine learning models, evaluated model performance through metrics such as the Area Under the Receiver Operating Characteristic Curve (AUC), Sensitivity, specificity, F1-score, decision curve, calibration curve, and clinical impact curve, employed the SHAP (Shapley Additive exPlanations) method to enhance model interpretability, and conducted external validation based on an independent dataset from the Second Hospital of Lanzhou University.

Results

“Weight,” “age,” and “height” are the most critical predictive factors. Gradient Boosting Machine (GB) showed optimal results, with training and test set AUC (0.850, 0.841), sensitivity (0.757, 0.737), specificity (0.793, 0.779), and F1-score (0.336, 0.316), respectively. External validation (3500 subjects) showed that the GB-based screening tool had an AUC of 0.828, which was significantly higher than that of the traditional Osteoporosis Self-Assessment Tool (OST, AUC = 0.781) via the DeLong test (z = 10.880, p < 0.0001).

Conclusion

A clinically applicable osteoporosis screening tool based on machine learning algorithms was developed and validated.

Summary

Around 9% of patients with previous bariatric surgery referred for osteoporosis treatment may have osteomalacia. Increased serum total alkaline phosphatase values should alert clinicians to this diagnosis since it requires a differential treatment approach, with some patients needing high doses of calcium or even parenteral vitamin D supplementation.

Introduction

The development of osteoporosis and fractures is a well-documented complication of bariatric surgery (BS). Nevertheless, subjects undergoing BS can also develop osteomalacia, which can be easily misdiagnosed as osteoporosis.

Objectives

To analyze the prevalence of osteomalacia and the main clinical characteristics of subjects with previous BS referred to a Rheumatology Department for osteoporosis treatment.

Methods

This retrospective study included a cohort of 46 subjects (aged 42–77 years) with previous BS referred for osteoporosis treatment. Clinical data were obtained from a review of medical records, including the type and time since surgery, treatment with calcium and/or vitamin D, and clinical, laboratory, radiologic, and densitometric data. Osteomalacia was diagnosed by compatible bone biopsy and/or clinical criteria (two of the following: low calcium, low phosphate, elevated total alkaline phosphatase [TAP], or suggestive radiology).

Results

Four out of 46 patients (8.7%) presented osteomalacia criteria; most were women (3/4) treated with malabsorptive surgery (from 4 to 13 years previously). All presented increased serum TAP and parathyroid hormone values, and most presented hypocalcemia and low vitamin D levels. Bone scan was compatible with osteomalacia in most subjects, and all subjects presented densitometric osteoporosis, with most developing fractures/pseudofractures after BS. No subject was referred to our department with clinical suspicion of osteomalacia.

Conclusions

Nearly 9% of patients with previous BS referred for osteoporosis treatment may have osteomalacia. Increased serum TAP values should alert clinicians to this diagnosis since it requires a differential treatment approach, with some patients needing high doses of calcium or even parenteral vitamin D supplementation.

Summary

Low bone mass (LBM), which can lead to osteoporosis, is often undetected and increases the risk of bone fractures. This study presents OsPenScreen, a deep learning model that can identify low bone mass early using standard chest X-rays (CXRs). By detecting low bone mass sooner, this tool helps prevent the disease progression to osteoporosis, potentially reducing health complications and treatment costs. OsPenScreen was validated across four external datasets and consistently performed well, showing its potential as a reliable, cost-effective solution for opportunistic early screening in CXR.

Purpose

Low bone mass, an often-undiagnosed precursor to osteoporosis, significantly increases fracture risk and poses a substantial public health challenge. This study aimed to develop and validate a deep learning model, OsPenScreen, for the opportunistic detection of low bone mass using routine chest X-rays (CXRs).

Methods

OsPenScreen, a convolutional neural network-based model, was trained on 77,812 paired CXR and dual-energy X-ray absorptiometry (DXA) datasets using knowledge distillation techniques. Validation was performed across four independent datasets (5,935 images) from diverse institutions. The model’s performance was assessed using area under the curve (AUC), accuracy, sensitivity, and specificity. Grad-CAM visualizations were employed to analyze model decision-making. Osteoporosis cases were pre-excluded by a separate model; OsPenScreen was applied only to non-osteoporotic cases.

Results

Our model achieved an AUC of 0.95 (95% CI: 0.94–0.97) on the external test datasets, with consistent performance across sex and age subgroups. The model demonstrated superior accuracy in detecting cases with significantly reduced bone mass and showed focused attention on weight-bearing bones in normal cases versus non-weight-bearing bones in low bone mass cases.

Conclusion

OsPenScreen represents a scalable and effective tool for opportunistic low bone mass screening, utilizing routine CXRs without additional healthcare burdens. Its robust performance across diverse datasets highlights its potential to enhance early detection, preventing progression to osteoporosis and reducing associated healthcare costs.