Hands-on osteoporosis screening with digital x-ray radiogrammetry

Abstract

According to epidemiological studies about one third of women and one fifth of men over 50 years will experience a fragility fracture. These fractures account for substantial mortality, morbidity and health care cost. Osteoporosis is a silent disease and thus often goes undiagnosed for a long time. Second to trauma it is considered to be the main cause of fractures among elderly. Several methods to reduce fracture risk have been developed. Identifying individuals with high fracture risk who would most benefit from such measures is of utmost importance for cost-efficient fracture prevention. Dual-energy X-ray absorptiometry (DXA) is widely considered to be the gold standard for assessing bone mass, diagnosing osteoporosis and estimating fracture risk. However, access to DXA is limited and not everyone in need of an examination is able to have one. Other fracture risk prediction models have therefore been developed, e.g. questionnaire-based tools. Different bone mass measuring devices have also been invented, e.g. qualitative ultrasound, peripheral DXA and digital X-ray radiogrammetry (DXR). None of these methods has been investigated nor validated as much as DXA. The aim of this study was to investigate the clinical use of DXR, which uses hand radiographs to determine bone mass.

In paper I, we retrospectively analyzed already obtained radiographs from 8,257 patients with DXR and found that DXR was highly predictive for hip fractures in both women and men. Later we recruited study participants from the Swedish mammography screening program and sampled a prospective, population-based cohort, the so-called STOP cohort. In paper II, the cohort was described, and it was shown that self-reported information about established clinical risk factors for osteoporosis were significantly associated with DXR T-score. A subset of the STOP cohort based on those with the lowest bone mass for their age (Zscore) was studied in paper III. In this subset we found a high prevalence of DXA-verified osteoporosis. Underlying causes for secondary osteoporosis and risk factors for primary osteoporosis were also overrepresented. In paper IV the STOP cohort was matched with fracture data from the Swedish National Inpatient Register and fracture prediction with DXR-BMD with and without clinical risk factors was examined. DXR T-score was significantly associated with hip, major osteoporotic and any clinical fracture.

In summary DXR derived bone mass was associated with fracture risk and known clinical risk factors for osteoporosis. Further research should focus on longer follow-up of the STOP cohort and health economical assessments concerning potential clinical implementation of the method.