Association between renal function and bone mineral density in healthy postmenopausal Chinese women

Prediction of subsequent fragility fractures: application of machine learning

BACKGROUND

Machine learning (ML) has shown exceptional promise in various domains of medical research. However, its application in predicting subsequent fragility fractures is still largely unknown. In this study, we aim to evaluate the predictive power of different ML algorithms in this area and identify key features associated with the risk of subsequent fragility fractures in osteoporotic patients.

METHODS

We retrospectively analyzed data from patients presented with fragility fractures at our Fracture Liaison Service, categorizing them into index fragility fracture (n = 905) and subsequent fragility fracture groups (n = 195). We independently trained ML models using 27 features for both male and female cohorts. The algorithms tested include Random Forest, XGBoost, CatBoost, Logistic Regression, LightGBM, AdaBoost, Multi-Layer Perceptron, and Support Vector Machine. Model performance was evaluated through 10-fold cross-validation.

RESULTS

The CatBoost model outperformed other models, achieving 87% accuracy and an AUC of 0.951 for females, and 93.4% accuracy with an AUC of 0.990 for males. The most significant predictors for females included age, serum C-reactive protein (CRP), 25(OH)D, creatinine, blood urea nitrogen (BUN), parathyroid hormone (PTH), femoral neck Z-score, menopause age, number of pregnancies, phosphorus, calcium, and body mass index (BMI); for males, the predictors were serum CRP, femoral neck T-score, PTH, hip T-score, BMI, BUN, creatinine, alkaline phosphatase, and spinal Z-score.

CONCLUSION

ML models, especially CatBoost, offer a valuable approach for predicting subsequent fragility fractures in osteoporotic patients. These models hold the potential to enhance clinical decision-making by supporting the development of personalized preventative strategies.

Analysis of the relationship between serum alanine aminotransferase and body composition in Chinese women

Objective

To investigate the relationships between serum alanine aminotransferase (ALT) and body composition among postmenopausal women in China.

Methods

A cross‐sectional study was conducted with 776 postmenopausal women in China from May to July 2008. Clinical information was collected using a standardized questionnaire. Measures of body composition were obtained using dual X‐ray absorptiometry. Body lean mass and fat mass indices were calculated by dividing total body lean/fat weight (kg) by body height squared (kg/m²). Blood samples were collected to assess liver and renal functions and lipid profiles. Analysis of variance, Pearson correlations, and multiple regression were used to analyze the associations between serum ALT and body composition.

Results

We found negative relationships of serum ALT with age, menopause duration, and serum HDL‐C levels. Serum ALT was positively correlated with BMI, serum TG levels, and the lean mass index and fat mass index. In a multivariate model adjusted for age, menopause duration, serum TG, and HDL‐C levels, a 1‐unit increase in the fat mass index was associated with a 0.176 U/L increase in ALT (95% CI 0.020 to 0.050, P < 0.001).

Conclusion

Serum ALT was positively associated with the body fat mass index of postmenopausal women in China.

Incidence of osteoporosis and ambient air pollution in South Korea: a population-based retrospective cohort study

Background

This study investigated the associations between exposure to ambient air pollutants and the incidence of osteoporosis using the Korean National Insurance Service–National Sample Cohort.

Methods

This nationwide, population-based, retrospective cohort study included 237,149 adults aged ≥40 years that did not have a diagnosis of osteoporosis at baseline between January 1, 2003, and December 31, 2015. Osteoporosis was defined as claim codes and prescriptions of bisphosphonates or selective estrogen receptor modulators at least twice annually. After matching values for PM₁₀, NO₂, CO, and SO₂ during the 2002–2015 time period and PM_2.5 in 2015 with residential areas, the incidence of osteoporosis was analyzed using a Cox proportional hazards regression model according to the quartile of average yearly concentrations of pollutants.

Results

Overall 22.2% of the study subjects, 52,601 (male: 5.6%, female: 37.6%) adults in total, were newly diagnosed with osteoporosis and treated. Exposure to PM₁₀ was positively associated with incidence of osteoporosis (Q4: 1798 per 100,000 person-years vs. Q1: 1655 per 100,000 person-years). The adjusted hazard ratio (HR) with 95% confidence interval (CI) of Q4 in PM₁₀ was 1.034 (1.009–1.062). The effect of PM₁₀ on osteoporosis incidence was distinct in females (adjusted sub-HR: 1.065, 95% CI: 1.003–1.129), subjects aged < 65 years (adjusted sub-HR: 1.040, 95% CI: 1.010–1.072), and for residents in areas with low urbanization (adjusted sub-HR: 1.052, 95% CI: 1.019–1.087). However, there was no increase in osteoporosis based on exposure to NO₂, CO, SO₂, or PM_2.5.

Conclusions

Long-term exposure to PM₁₀ was associated with newly diagnosed osteoporosis in Korean adults aged ≥40 years. This finding can aid in policy-making that is directed to control air pollution as a risk factor for bone health.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12889-021-11866-7.