Chest, 2026 (SCI-Expanded, Scopus)
Background: Osteoporosis is an important cause for morbidity, but it remains underdiagnosed, and current screening programs for osteoporosis are underused. Research Question: Can low-dose CT (LDCT) imaging of the chest obtained for lung cancer screening identify patients with osteoporosis? Study Design and Methods: Patients who underwent LDCT imaging and dual-energy x-ray absorptiometry (DXA) within 4 months of each other were identified. CT imaging was used to measure vertebral trabecular bone attenuation (TBA) manually at the T7 through L1 vertebrae. Vertebral fractures seen on CT imaging were noted. DXA was used to obtain T-scores and Fracture Risk Assessment Tool (FRAX) scores. Mean CT imaging TBA values were correlated with DXA T-scores. Threshold CT imaging TBA values below which osteoporosis was present were reported. Results: One thousand three hundred thirty-six patients were included. The median age was 68 years (interquartile range, 64-72 years). Osteoporosis was diagnosed in 33.0% of the cohort based on DXA T-scores and FRAX scores. At least 1 vertebral fracture occurred in 11.9% of patients. Mean CT imaging TBA at the T7 through L1 vertebrae showed a moderate correlation with DXA T-score with an r coefficient of 0.573 (95% CI, 0.531-0.614). The area under the receiver operating characteristic curve of mean CT imaging TBA at the T7 through L1 vertebrae in diagnosing osteoporosis was 0.771 (95% CI, 0.744-0.798; P < .001). A CT imaging TBA at the T7 through L1 vertebrae threshold value of ≤ 130 Hounsfield units (HU) yielded a sensitivity of 90.5% and a specificity of 39.7%. A TBA at the T7 through L1 vertebrae threshold value of ≤ 80 HU yielded a specificity of 93.1%. The prevalence of vertebral fractures at T-score of ≤ –2.5 was 21%; a similar fracture prevalence occurred at a CT imaging TBA threshold of ≤ 110 HU. T-score values of ≤ –2.5 failed to identify 50.9% of vertebral fractures. Interpretation: CT imaging TBA is a promising tool for diagnosing osteoporosis using LDCT imaging obtained for lung cancer screening. This technique can be performed after minimal training without additional software, hardware, cost, or radiation exposure.