TOPLINE

A retrospective study of opportunistic CT scans found that older age was linked to lower whole-bone CT-derived apparent volumetric bone mineral density (CT-vBMD) in all five bones analysed, most strongly in trabecular-rich regions.

METHODOLOGY

• Researchers retrospectively analysed 330 routine abdominal CT scans from adults aged 20-91 years, approximately balanced by sex, covering five bones (L4, L5, right proximal femur, right pelvic bone, and sacrum).
• Bones were segmented semiautomatically, mapped to bone-specific population templates, and internally calibrated to convert Hounsfield unit values to CT-vBMD.
• Mann-Whitney U tests with false discovery rate correction were used to compare whole-bone density across five age contrasts per bone (overlapping strata: 20-29, 20-39, 40-59, and 60-91 years) and between sexes.
• Element-wise maps were used to assess mean bone density and correlations with age and sex.
• Bone regions were sorted into three descriptive clusters based on mean CT-vBMD and age correlation (high-density age-independent, low-density age-sensitive, and high-density age-sensitive), not derived from a predictive model.

TAKEAWAY

• In all five bones, older age strata had lower whole-bone density than young-reference strata, and all 25 pairwise comparisons stayed significant after false discovery rate correction (adjusted q = 5.85 × 10^-26 to 2.85 × 10^-4).
• A low-density, age-sensitive, mostly trabecular cluster covered the vertebral bodies, lateral sacrum, ilium near the sacroiliac joint, and femoral neck spongiosa (Pearson correlation coefficient [r] = -0.45).
• Local age correlations (r) were commonly about -0.4 to -0.5 in the trabecular-rich portion of the proximal femur but only about -0.1 to -0.2 in the cortical-dominant femoral neck, and cortical bone elsewhere in the pelvis also showed markedly smaller age-related changes.
• None of the 25 female-male comparisons were significant after false discovery rate correction (adjusted q = 0.0744-0.921).

IN PRACTICE

“In routine practice, a spatial regional reference from opportunistic pelvic CT may help radiologists interpret low bone density more systematically, particularly in the sacrum, peri-acetabular region, and lumbar vertebral bodies, where visual assessment can be inconsistent,” the authors wrote.

“Such information should be viewed as an adjunctive screening signal that may prompt formal osteoporosis evaluation, not as a replacement for DXA [DEXA], FRAX [Fracture Risk Assessment Tool], or validated clinical thresholds,” they added.

SOURCE

The study was led by Kateřina Matoušová, Faculty of Health Studies, Technical University of Liberec, Liberec, Czechia. It was published online on October 20, 2026, in Skeletal Radiology.

LIMITATIONS

The study was retrospective and used scans from one tertiary care centre. The clinical indications, comorbidities, medications, and fracture history could not be linked to the anonymised scans. Calibration was internal rather than phantom-based, and missing contrast phase and dose details limited the interpretation of absolute density values. Sample sizes within age strata were small. The study did not assess fracture outcomes and could not show whether the patterns predicted fractures. Age groups were compared across different patients, not followed over time.

DISCLOSURES

The study was funded by the Czech Science Foundation. The authors reported having no conflicts of interest.