Opportunistic computed tomography-based vertebral attenuation for osteoporosis screening

SUMMARY

Objective

The aim of this study was to evaluate opportunistic vertebral trabecular attenuation on routine unenhanced abdominal computed tomography for predicting osteoporosis and its relationship with dual-energy X-ray absorptiometry parameters.

Methods

This retrospective study inc luded 350 patients (mean age 68.5 years) who underwent both dual-energy X-ray absorptiometry and routine unenhanced abdominal computed tomography for non-oncologic clinical indications within a 3-month interval between 2022 and 2025. Patients were grouped into normal (n=119), osteopenia (n=117), and osteoporosis (n=114) based on L1–L4 T-scores. Two blinded radiologists measured Hounsfield Unit values using regions of interest on high-resolution 0.625-mm sagittal-reconstructed computed tomography images of the L1–L4 vertebral bodies. Interobserver agreement was assessed using intraclass correlation coefficients. Diagnostic performance was evaluated using receiver operating characteristic curve analysis.

Results

Lumbar Hounsfield Unit values decreased progressively from normal to osteoporosis groups (p<0.001). Mean L1–L4 attenuation values for normal, osteopenia, and osteoporosis groups were 148.99±5.54, 120.39±3.84, and 78.53±3.36 Hounsfield Unit, respectively (p<0.001). Interobserver reliability was excellent (intraclass correlation coefficient: 0.90–0.94). L1 Hounsfield Unit measurements showed the highest diagnostic performance for predicting osteoporosis (area under the curve: 0.913; 95%CI 0.870–0.955). An L1 cutoff of ≤95.50 Hounsfield Unit yielded 84.60% sensitivity and 86.40% specificity. Mean L1–L4 Hounsfield Unit also showed high accuracy (area under the curve: 0.881) at a ≤95.00 Hounsfield Unit cutoff (sensitivity: 80.80%, specificity: 77.30%).

Conclusion

Opportunistic sagittal lumbar Hounsfield Unit measurements, particularly at L1, provide high diagnostic accuracy and reproducibility for identifying osteoporosis on routine abdominal computed tomography without requiring additional imaging or radiation exposure. This protocol may facilitate early opportunistic screening for osteoporosis and support timely referral for dual-energy X-ray absorptiometry assessment.