Article Text

Download PDFPDF

PTH-128 The Frax Algorithm is of Limited Utility in Predicting Osteoporosis in Coeliac Disease
  1. A Mattinson1,
  2. B Hauser2,
  3. H Gillett3,
  4. P Gillett3,
  5. C Goddard3,
  6. N Kennedy4,
  7. A Shand3,
  8. S Ralston2,
  9. J Satsangi4,
  10. P Riches2
  1. 1University of Edinburgh
  2. 2Rheumatic Diseses Unit, Institute of Genetics and Molecular Medicine
  3. 3Gastroenterology Department, Lothian University Hospitals NHS Trust
  4. 4Gastroenterology Department, Institute of Genetics and Molecular Medicine, Edinburgh, UK


Introduction Osteoporosis is the commonest complication of coeliac disease yet no reliable scoring system exists to guide patient selection for bone density measurement. The FRAX tool has been developed by the World Health Organisation to estimate fracture risk based on clinical factors and incorporates causes of secondary osteoporosis such as coeliac disease (1). We have analysed the utility of FRAX in identifying osteoporosis in a cohort of patients with coeliac disease.

Methods 170 patients were recruited from coeliac clinics between October 2011 and 2012. 17 patients in whom bone mineral density results were not available were excluded, yielding a final study population of 153. Information on clinical risk factors for osteoporosis were collected by questionnaire. Two-tailed independent student t-tests, Mann Whitney U test or Chi-square tests were applied as appropriate. Statistical analysis was performed on SPSS.

Abstract PTH-128 Table 1

The prevalence of osteoporosis in our cohort was 15% (23/153). The distribution of risk factors used in the FRAX algorithm are shown in table 1. Factors significantly associated with osteoporosis in our cohort included increasing age, reduced height, weight and history of glucocorticoid use. The median 10 year risk of major osteoporotic fracture was 6.7% (interquartile range 8.5). A ROC analysis of FRAX as a predictor of osteoporosis yielded an area under the curve of just 0.614.

Conclusion The FRAX algorithm is not a reliable predictor of osteoporosis. A screening threshold of > 10% 10 year risk of major fracture gives a sensitivity of 43% and specificity of 73% for detection of osteoporosis. A lower threshold of 5% 10 year risk only increases sensitivity to 78% at a cost to specificity of 59%. Further work in constructing specific risk predictors for osteoporosis in coeliac disease is required.

Disclosure of Interest None Declared.


  1. Kanis JA et al. (2008) FRAXTM and the assessment of fracture probability in men and women from the UK. Osteoporosis International 19: 385–397.

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.