Opportunistic L1 Attenuation on 100-kVp LDCT Predicts Ten-Year Fragility Fracture Risk in Postmenopausal Women in a Self-Paid Health Check-up Cohort.

Rationale

AND

Objectives

Low-dose computed tomography (LDCT) is standard for lung cancer screening, yet validated fracture risk thresholds for 100-kVp protocols have not been established in health check-up cohorts. We evaluated whether L1 vertebral attenuation on 100-kVp LDCT predicts 10-year fragility fracture incidence.

Materials And Methods

This retrospective cohort study included 1464 women age ≥50 years (median follow-up: 10.1 years). L1 Hounsfield Units (HU) were measured on 100-kVp LDCT by two independent readers. References value for normal and low bone mineral density (BMD) was derived from a separate calibration cohort (n = 246) with paired LDCT-Dual-energy X-ray absorptiometry (DXA) data (AUC 0.841). Endpoints included incident vertebral fractures and composite major fragility fractures.

Results

HU = 160 (AUC = 0.841) stratified the whole group into normal BMD (>160 HU; n = 952) and low BMD (≤160 HU; n = 512). At 10 years, cumulative vertebral fracture incidence was 38.2 vs. 2.8% in low vs. normal BMD groups; composite fractures reached 48.1% vs. 6.3%. The incidence rate ratio was 3.46 (95% CI: 2.62-4.57) for vertebral fractures and 2.72 (95% CI: 2.14-3.46) for all fragility fractures. On multivariable Cox regression adjusting for age, body mass index, and comorbidities, low BMD (≤160 HU) remained an independent predictor of vertebral fracture (adjusted HR: 2.54; 95% CI: 1.33-4.85; p = 0.005).

Conclusion

L1 attenuation ≤160 HU on 100-kVp LDCT identified postmenopausal women at three-fold higher fragility fracture risk, with performance comparable to Fracture Risk Assessment Tool without BMD. In this healthy screenee cohort, L1 HU serves as the primary bone health signal, enabling early identification of at-risk individuals within existing LDCT workflows at no additional cost.