P064 Correlation between low Body Mass Index (BMI) and FRAX risk factors in patients presenting with a BMI below 20kg/m2

Bates, Nathan J and Amin, Hamzah and Bukhari, Marwan (2026) P064 Correlation between low Body Mass Index (BMI) and FRAX risk factors in patients presenting with a BMI below 20kg/m2. Rheumatology, 65 (Supple). ISSN 1462-0324

Abstract

Background/Aims Low BMI is a well-established risk factor for osteoporotic fractures. However, the relationship between low BMI and the risk factors utilised by the FRAX algorithm remains poorly understood. Female patients make up a significant proportion of the current data available and, consequently, the research into osteoporosis with its associated risks. We aimed to evaluate associations between BMI <20 and FRAX risk factors in men and women presenting with fractures. This project also aims to establish that FRAX is equally reliable in predicting fracture risk in the male population. Methods A retrospective observational study was performed. Participants were adult patients referred for a bone density assessment with BMI <20 referred to Morecambe Bay Hospital Trust, which includes Furness General Hospital, Royal Lancaster Infirmary, Millom Hospital and Westmorland General Hospital. The cohort included all adult patients who presented with a BMI <20 regardless of fracture status between 2004-2019. Data on FRAX risk factors (smoking, alcohol use, glucocorticoid exposure, family history of fracture, secondary osteoporosis, and prior fragility fractures) were collected alongside weight, height, and bone mineral density (BMD). The data was extracted from electronic health records. The value used for BMD was the average BMD from L1-L4 vertebrae and both hips after a DEXA scan. Statistical analyses of the continuous data using t-tests and categorical data using chi-square tests assessed correlations between fracture status and each variable. The significance threshold was set at p = <0.05. Results The cohort included 2,419 adult patients (313 men, 2,106 women). Among low BMI patients, 12.9% were male and 87.1% female. Significant associations with fracture status across both genders included lower weight (p = 0.014), shorter height (p = 0.004), reduced BMD (p = <0.001), history of prior fracture (p < 0.001), glucocorticoid use (p = 0.005), secondary osteoporosis (p = 0.005), and alcohol consumption (p < 0.001). There was no statistically significant difference in the risk factors between the male and female cohorts. Conclusion In adults with BMI <20 presenting with fractures, multiple FRAX risk factors, particularly low BMD, steroid use, alcohol intake, and prior fractures, showed significant correlations with fracture occurrence. The results also indicate that both genders followed similar risk patterns and, as a consequence is equally reliable for both, thus indicating that further stratification in clinical practice is unnecessary. These findings underscore the importance of comprehensive fracture risk assessment in underweight individuals using the FRAX algorithm. Disclosure N.J. Bates: None. H. Amin: None. M. Bukhari: None.