Abstract
Osteoporosis can cause significant disability and cost to health services globally. We aim to compare risk fractures for both osteoporosis and fractures at the L1-L4 vertebrae (LV) and the neck of femurs (NOFs) in patients referred for DEXA scan in the North-West of England.
Data was obtained from 31546 patients referred for DEXA scan in the North-West of England between 2004 and 2011. Demographic data was retrospectively analysed using STATA, utilising chi-squared and t-tests. Logistical models were used to report odds ratios for risk factors included in the FRAX tool looking for differences between osteoporosis and fracture risk at the LV and NOFs.
In a study involving 2530 cases of LV fractures and 1363 of NOF fractures, age was significantly linked to fractures and osteoporosis at both sites, with a higher risk of osteoporosis at NOFs compared to LV. Height provided protection against fractures and osteoporosis at both sites, with a more pronounced protective effect against osteoporosis at NOFs. Weight was more protective for NOF fractures, while smoking increased osteoporosis risk with no site-specific difference. Steroids were unexpectedly protective for fractures at both sites, with no significant difference, while alcohol consumption was protective against osteoporosis at both sites and associated with increased LV fracture risk. Rheumatoid arthritis increased osteoporosis risk in NOFs and implied a higher fracture risk, though not statistically significant compared to LV. Results summarised in Table 1.
Our study reveals that established osteoporosis and fracture risk factors impact distinct bony sites differently. Age and rheumatoid arthritis increase osteoporosis risk more at NOFs than LV, while height and steroids provide greater protection at NOFs. Height significantly protects LV fractures, with alcohol predicting them. Further research is needed to explore risk factors’ impact on additional bony sites and understand the observed differences’ pathophysiology.
For any figures or tables, please contact the authors directly.
