Article Text

SAT0343 Severe prevalent vertebral fractures predict subsequent vertebral and non-vertebral fractures: A 3-year prospective study
  1. O. Bruyere1,
  2. C. Roux2,
  3. D. Nicolet1,
  4. J. Fechtenbaum2,
  5. R. Deroisy1,
  6. J.-Y. Reginster1
  1. 1Public Health, Epidemiology and Health Economics, University of Liège, Liège, Belgium
  2. 2Rheumatology, Cochin Hospital, Paris, France


Objectives To examine whether a prevalent severe vertebral fracture predicts subsequent fracture and to assess the effect of strontium ranelate to reduce fracture in patients with such prevalent severe vertebral fracture.

Methods Analysis of two 3-year randomized controlled trials investigating the clinical effect of strontium ranelate. Vertebral and major non-vertebral fractures (i.e. ribs sternum, wrist region, pelvic sacrum, collarbone, humerus, proximal femur) were assessed during the 3 years. At baseline on standard radiographs, each vertebra received a severity grade corresponding to either no fracture either a mild, moderate or severe fracture based on a semi-quantitative visual assessment. All women received calcium and vitamin D during the whole length of the study.

Results 6137 women aged mean (SD) 75.1 (6.1) years were included in this study. In the placebo group, after 3 years of follow-up, 43% of the women with a prevalent severe vertebral fracture had experienced a new vertebral fracture compared to only 12.5% of the women without any vertebral fracture (OR 3.46 [2.87-4.18]). The logistic regression analysis, adjusted for age, body mass index, femoral neck bone mineral density and number of prevalent vertebral fractures, showed that the presence of a severe vertebral fracture was significantly associated with new vertebral fractures over a 3-year follow-up period. In women with severe prevalent vertebral fracture (N=680), strontium ranelate was able to significantly reduce the risk of new vertebral fractures (OR 0.75 [0.61-0.91]).

Conclusions In summary, severe prevalent vertebral fracture is an independent predictor of new vertebral and non-vertebral fractures that could be taken into account in fracture risk management.

Disclosure of Interest O. Bruyere Grant/Research support from: Servier, C. Roux: None Declared, D. Nicolet: None Declared, J. Fechtenbaum: None Declared, R. Deroisy: None Declared, J.-Y. Reginster: None Declared

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.