Fractures and mortality prediction by opportunistic screening of osteoporosis
Related Experiment Video
Updated: Nov 1, 2026
Cortical Bone Assessment Using Ultrasonic Guided Waves: A Reproducibility Study in a Healthy Population
Published on: January 31, 2025
Christian Roux1, Antoine Rozes2, Aya Ajrouche2,3
1Université Paris Cité and Université Sorbonne Paris Nord, INSERM, INRAE, Center For Research in Epidemiology and Statistics (CRESS), F-75004 Paris and Assistance Publique-Hôpitaux de Paris, service de Rhumatologie, Hôpital Cochin F - 75014 PARIS, Paris, France. christian.roux@aphp.fr.
Abstract:
Opportunistic screening of osteoporosis can be used on computed tomography scans. In this study of 46,219 patients, results of this screening predict incident osteoporotic fractures necessitating hospitalization and mortality in a 3-year follow-up. These results suggest that opportunistic screening is a relevant method to optimise osteoporotic patients care.
Objectives:
The objectives were to assess the predictive accuracy of opportunistic screening of osteoporosis on CT scans on incident osteoporotic fractures necessitating hospitalization and mortality over 3 years.
Methods:
Multicentre observational retrospective cohort study with 3-year follow-up. Participants were patients aged 60 or older and with an inpatient CT scan involving the lumbar spine for any indication. A bone score was measured in a standardised region of interest of the L1 vertebral body, and a prevalent lumbar vertebral fractures (VF) were detected by a fully automated procedure. The main outcome measure was the first occurrence of osteoporotic fracture necessitating hospitalization (OFNH) ; secondary outcomes were location of OFNH and death.
Results:
46 219 patients (75.3 ± 9.3 years old) were included. The mean bone score was 127.4 ± 75.2 HU. A prevalent lumbar VF was detected in 27.3%. The 3-year crude incidence was 6.8% for OFNH and 27.9% for mortality. Areas under the ROC curves of the models predicting any OFNH and hip fractures from age, sex, bone score, and detection of a prevalent lumbar VF were 0.68 (0.67 to 0.69) and 0.71 (0.70 to 0.71) respectively. The crude cumulative incidence of OFNH was higher in patients with lumbar VF detected on the index scan: sub-hazard ratios were 2.22 (2.06 to 2.39) for any incident OFNH fracture, 2.18 (2.02-2.38) for non-vertebral, 2.34 (2.09-2.63) for hip, and 3.29 (2.45 to 4.4) for VF. Detection of the latter also predicted 3-year mortality, with an adjusted hazard ratio of 1.29 (1.25 to 1.34).
Conclusion:
Fully automated opportunistic screening, predicted incident osteoporotic fractures necessitating hospitalization and mortality over 3 years in patients aged 60 or older with an inpatient computed tomography scan involving the lumbar spine for any indication.
