Diagnostic performance of quantitative ultrasound calcaneus measurement in case finding for osteoporosis in Thai postmenopausal women

Establish and validate the reliability of predictive models in bone mineral density by deep learning as examination tool for women

While FRAX with BMD could be more precise in estimating the fracture risk, DL-based models were validated to slightly reduce the number of under- and over-treated patients when no BMD measurements were available. The validated models could be used to screen for patients at a high risk of fracture and osteoporosis.

PURPOSE

Fracture risk assessment tool (FRAX) is useful in classifying the fracture risk level, and precise prediction can be achieved by estimating both clinical risk factors and bone mineral density (BMD) using dual X-ray absorptiometry (DXA). However, DXA is not frequently feasible because of its cost and accessibility. This study aimed to establish the reliability of deep learning (DL)-based alternative tools for screening patients at a high risk of fracture and osteoporosis.

METHODS

Participants were enrolled from the National Bone Health Screening Project of Taiwan in this cross-sectional study. First, DL-based models were built to predict the lowest T-score value in either the lumbar spine, total hip, or femoral neck and their respective BMD values. The Bland-Altman analysis was used to compare the agreement between the models and DXA. Second, the predictive model to classify patients with a high fracture risk was built according to the estimated BMD from the first step and the FRAX score without BMD. The performance of the model was compared with the classification based on FRAX with BMD.

RESULTS

Approximately 10,827 women (mean age, 65.4 ± 9.4 years) were enrolled. In the prediction of the lumbar spine BMD, total hip BMD, femoral neck BMD, and lowest T-score, the root-mean-square error (RMSE) was 0.099, 0.089, 0.076, and 0.68, respectively. The Bland-Altman analysis revealed a nonsignificant difference between the predictive models and DXA. The FRAX score with femoral neck BMD for major osteoporotic fracture risk was 9.7% ± 6.7%, whereas the risk for hip fracture was 3.3% ± 4.6%. Comparison between the classification of FRAX with and without BMD revealed the accuracy rate, positive predictive value (PPV), and negative predictive value (NPV) of 78.8%, 64.6%, and 89.9%, respectively. The area under the receiver operating characteristic curve (AUROC), accuracy rate, PPV, and NPV of the classification model were 0.913 (95% confidence interval: 0.904-0.922), 83.5%, 71.2%, and 92.2%, respectively.

CONCLUSION

While FRAX with BMD could be more precise in estimating the fracture risk, DL-based models were validated to slightly reduce the number of under- and over-treated patients when no BMD measurements were available. The validated models could be used to screen for patients at a high risk of fracture and osteoporosis.

Decision Tree Modeling for Osteoporosis Screening in Postmenopausal Thai Women

Osteoporosis is still a serious public health issue in Thailand, particularly in postmenopausal women; meanwhile, new effective screening tools are required for rapid diagnosis. This study constructs and confirms an osteoporosis screening tool-based decision tree (DT) model. Four DT algorithms, namely, classification and regression tree; chi-squared automatic interaction detection (CHAID); quick, unbiased, efficient statistical tree; and C4.5, were implemented on 356 patients, of whom 266 were abnormal and 90 normal. The investigation revealed that the DT algorithms have insignificantly different performances regarding the accuracy, sensitivity, specificity, and area under the curve. Each algorithm possesses its characteristic performance. The optimal model is selected according to the performance of blind data testing and compared with traditional screening tools: Osteoporosis Self-Assessment for Asians and the Khon Kaen Osteoporosis Study. The Decision Tree for Postmenopausal Osteoporosis Screening (DTPOS) tool was developed from the best performance of CHAID’s algorithms. The age of 58 years and weight at a cutoff of 57.8 kg were the essential predictors of our tool. DTPOS provides a sensitivity of 92.3% and a positive predictive value of 82.8%, which might be used to rule in subjects at risk of osteopenia and osteoporosis in a community-based screening as it is simple to conduct.

Clinical Conformity Between Heel Ultrasound and Densitometry in Postmenopausal Women: A Systematic Review

OBJECTIVES

To assess the conformity between heel ultrasound and densitometry, and the clinical application of densitometry T-score "gold standard" in quantitative ultrasound as a method of osteoporosis diagnosis in postmenopausal women.

METHODS

The study is a systematic review of studies published in the last 17 years in PubMed, NLM Gateway, Medline, Embase, and Cochrane Library. Calcaneal quantitative ultrasound sensitivity and specificity were analyzed with regard to densitometry measurements in postmenopausal women. In addition, we summarized the values of ultrasound T-scores, for which their accuracy in osteoporosis diagnosis is the highest.

RESULTS

The inclusion criteria met 15 research studies conducted on postmenopausal women. In 11 of them, the authors concluded that clinical conformity between heel ultrasound and densitometry is good. The recommended quantitative ultrasound T-score for osteoporosis diagnosis ranged between -1 and -3.65.

CONCLUSIONS

Heel ultrasound should be considered to be as accurate as densitometry in diagnosing osteoporosis. Nevertheless, it needs to have separate T-score ranges determined, because those used in densitometry are not adequate.

Is calcaneal quantitative ultrasound useful as a prescreen stratification tool for osteoporosis?

Calcaneal quantitative ultrasound (QUS) is attractive as a prescreening tool for osteoporosis, alternative to dual-energy X-ray absorptiometry. We investigated the literature of the usability of calcaneal QUS. We found large heterogeneity between studies and uncertainty about cutoff, device, and measured variable. Despite osteoporosis-related fractures being a major health issue, osteoporosis remains underdiagnosed. Dual-energy X-ray absorptiometry (DXA) of the hip or spine is currently the preferred method for diagnosis of osteoporosis, but the method is limited by low accessibility. QUS is a method for assessing bone alternative to DXA. The aim of this systematic review was to explore the usability of QUS as a prescreen stratification tool for assessment of osteoporosis. Studies that evaluated calcaneal QUS with DXA of the hip or spine as the gold standard was included. We extracted data from included studies to calculate number of DXAs saved and misclassification rates at cutoffs equal to high sensitivity and/or specificity. The number of DXAs saved and percentage of persons misclassified were measures of usability. We included 31 studies. Studies were heterogeneous regarding study characteristics. Analyses showed a wide spectrum of percentage of DXAs saved (2.7-68.8%) and misclassification rates (0-12.4%) depending on prescreen strategy and study characteristics, device, measured variable, and cutoff. Calcaneal QUS is potentially useful as a prescreen tool for assessment of osteoporosis. However, there is no consensus of device, variable, and cutoff. Overall, there is no sufficient evidence to recommend a specific cutoff for calcaneal QUS that provides a certainty level high enough to rule in or out osteoporosis. Calcaneal QUS in a prescreen or stratification algorithm must be based on device-specific cutoffs that are validated in the populations for which they are intended to be used.

Prevalence of osteoporosis in middle Anatolian population using calcaneal ultrasonography method

BACKGROUND

There are a few studies on the prevalence of osteoporosis in Turkey. Most of such studies have been performed using classical diagnostic methods. Quantitative calcaneal ultrasonography is a contemporary procedure used in the diagnosis of osteoporosis.

OBJECTIVE

The aim of this study is to determine the prevalence of the osteoporosis and risk factors in middle Anatolian population aged more than 40, using quantitative calcaneal ultrasonography.

SUBJECTS AND METHODS

The present study was carried out in rural and urban areas of primary health care units. Evaluation of skeletal status was based on calcaneal quantitative ultrasound measurement, carried out by the Sahara Clinical Bone Sonometer (Hologic Waltham, MA, USA).

RESULTS

There were 1437 individuals, 866 were females (60.3%) and 571 were males (39.7%), aged in between 40 and 89 years. Standardized prevalence rates using Segi's population were found to be 14.0% in total, 17.1% in female and 9.2% in male. Osteoporosis prevalence increased with increasing ages. The highest prevalence value was determined in the individuals older than 70 years of age. Osteoporosis was more common among women compared to men. Osteoporosis prevalence increased in consistent with a decrease in the amount of milk, yoghurt and cheese consumption.

CONCLUSION

Based on current study, it is concluded that osteoporosis is an important health problem. Requiring health education about life styles (activity and nutrition) to avoid and even control osteoporosis.

A nomogram for predicting osteoporosis risk based on age, weight and quantitative ultrasound measurement

INTRODUCTION

Quantitative ultrasound measurement (QUS) or clinical risk index alone are not reliable tools for the identification of women with osteoporosis. This study examined the prognostic value of combined QUS and clinical risk index for predicting osteoporosis risk in Thai women.

METHODS

The study was designed as a cross-sectional investigation with 300 women of Thai background, aged between 38 and 85 years (mean age: 58). Femoral neck bone mineral density (BMD) was measured by DXA (Hologic QDR-4500; Bedford, MA, USA). A femoral neck BMD T-scores <or= -2.5 was defined as "osteoporosis"; otherwise, "non-osteoporosis". QUS was measured by Achilles+ (GE Lunar, Madison, WI, USA) and converted to T-score. Three models for predicting osteoporosis were considered: model I included age, weight and QUS, model II included age and weight, and model III included only QUS. The prognostic performance among the models was assessed by the area under the receiver operating characteristic curve (AUC).

RESULTS

The prevalence of osteoporosis was 12.7% (n=38/300) by femoral neck BMD. Age, weight and QUS were each significantly associated with osteoporosis risk. The AUC+/-SE value for model I was 0.86 +/- 0.03, which was significantly higher (p=0.02) than that for model II (AUC = 0.80 +/- 0.04) or model III (AUC = 0.79 +/- 0.04). Based on the estimated parameters of model I, a nomogram was constructed for predicting osteoporosis.

CONCLUSION

These data suggest that the combination of QUS and age and weight could significantly improve the prognosis of osteoporosis in Asian women, and that the nomogram can assist primary care physicians in the identification of high-risk women.

Identifying osteoporosis in a primary care setting with quantitative ultrasound: relationship to anthropometric and lifestyle factors

Dual-energy X-ray absorptiometry (DXA), the "gold standard" for diagnosis of osteoporosis, is not recommended for population screening, and thus quantitative ultrasound (QUS) of the calcaneus is gaining popularity. The aim of the present study was to evaluate the relationship between QUS values and anthropometric and lifestyle factors, and to assess the diagnostic performance of QUS in predicting DXA-defined osteoporosis. Eight hundred and thirty-two women and 87 men aged 40-88 years were included in the study. Anthropometric measurements, the questionnaire, and QUS and DXA measurements were performed by trained physicians. Both QUS and DXA T-scores were lower for women than for men. Postmenopausal women had significantly lower QUS T-scores compared to premenopausal women (P < 0.001). Age over 50, female sex, sedentary lifestyle, fracture history, presence of chronic disease, and > or =5 years since menopause were associated with QUS T-scores lower than -1.00 by multivariate analysis. Low QUS T-scores were related to lack of direct sun exposure, high parity, fair skin color, and no education by univariate analysis (P < 0.005). A weak correlation was found between calcaneal QUS and DXA T-scores at lumbar spine (r = 0.310, P < 0.001) and femoral neck (r = 0.288, P < 0.001). The sensitivity and specificity of the QUS test were 73.7% and 57.4%, respectively, regarding the identification of osteoporotic patients. Lower QUS T-scores were associated with several osteoporotic risk factors, and the sensitivity and specificity of QUS for predicting DXA-defined osteoporosis were at optimum values at ages between 50 and 59 years. We conclude that, even though the accuracy of QUS for predicting DXA-defined osteoporosis is not remarkably high, it can be applied to identify subjects at risk in this age group in developing countries and rural districts who should be the focus of fracture prevention.