Diagnostic Performance of Proximal Humeral Cortical Thickness on Digital Radiography Compared with Dual-Energy X-Ray Absorptiometry Bone Mineral Density for Opportunistic Osteoporosis Screening in Postmenopausal Women: A Cross-Sectional Study at a Tertiary Referral Center in Indonesia

Abstract

Background: Osteoporosis causes systemic bone fragility in postmenopausal women. While dual-energy X-ray absorptiometry (DEXA) is the gold standard for bone mineral density (BMD) assessment, limited availability impedes routine screening in developing nations. Proximal humeral cortical thickness (HCT) on routine digital radiographs offers a practical opportunistic screening alternative.

Objective: To evaluate the diagnostic accuracy and optimal discriminatory threshold of radiogrammetric proximal HCT in differentiating osteoporosis compared with lumbar spine DEXA in postmenopausal Indonesian women.

Methods: A cross-sectional diagnostic accuracy study adhering to STARD 2015 guidelines was conducted at Dr. Mohammad Hoesin General Hospital, Palembang, Indonesia. Consecutive women aged ≥50 years undergoing upper extremity digital radiography and lumbar spine DEXA were enrolled. Proximal HCT was calculated as the mean of medial and lateral cortical thicknesses across two standardized diaphyseal levels. Diagnostic performance was evaluated using receiver operating characteristic (ROC) analysis.

Result: Thirty-four women (mean age: 67.65 ± 7.90 years) were evaluated; 9 (26.47%) had osteoporosis by DEXA. Combined HCT was significantly lower in osteoporotic than non-osteoporotic women (4.11 ± 0.87 mm vs 5.29 ± 0.75 mm, p = 0.001). ROC analysis showed an AUC of 0.858 (95% CI: 0.683–1.000, p = 0.002). The optimal Youden-derived cut-off of 4.380 mm yielded 77.78% sensitivity, 96.00% specificity, 87.50% PPV, 92.31% NPV, 19.45 positive likelihood ratio, 0.23 negative likelihood ratio, and 91.18% overall accuracy.

Conclusion: Radiogrammetric proximal HCT demonstrates high accuracy and outstanding specificity for osteoporosis screening. A threshold of 4.380 mm serves as a zero-marginal-cost tool to prioritize patients for confirmatory densitometry and early intervention.