Pos0398 determinants of mortality and imminent re-fracture in patients hospitalized for severe osteoporotic fractures

Background Patients hospitalized for severe osteoporotic fractures are at increased risk of morbidity and mortality. It is recommended to improve their medical management by Fracture Liaisons Services (FLS) organization. Objectives Our aim was to assess the determinants of mortality and imminent refracture in those patients followed in FLS. Methods The CROSS study is a national, prospective, observational, multicenter study conducted in 12 centers with Fracture Liaison Services (FLS). Patients included were men and women above 60 years, hospitalized for a recent (less than 3 months) severe fragility fracture (hip, pelvis, humerus or vertebrae) that occurred after a low-energy trauma. Patients with either a non-severe fracture, a pathological fracture, a high trauma fracture or a per-prosthetic fracture were not included. At baseline and 2 years we have collected sociodemographic data, fracture event, bone risk factors, factors of falling, FRAX items, history of treatments, comorbidities, Charlson score and DXA measurement. To assess the risk factors for new severe fracture or death, a multivariate Cox proportional hazard multivariate analysis was performed. Results 895 patients were included in the cohort with the following fracture location distribution: clinical vertebrae 43.3%, hip 37.5%, pelvis 10.3%, and humerus 11.1%. Most patients were women (79%) with a median age of 81 years (71-85). 40% had a previous history of fragility fracture after 40 years. Only 17.7% received an anti-osteoporotic treatment in the 5 years prior baseline whereas 21.4% received a calcium supplementation and 43.6% received a vitamin D supplementation. At baseline 48% of patients had densitometric osteoporosis. Over the 2 years of follow-up (data completed for 95% of population), 116 severe fractures in 110 patients (12.9%) and 80 deaths (8.9%) occurred. 49.1% of patients were prescribed an antiosteoporotic treatment after the fracture (75% of bisphosphonates), that was initiated within the 3 months following fracture event in 63 % of the cases. Multivariate analysis showed that reduced spinal BMD (OR=24.6 CI 95% 2.84-247, p=0.027), recurrent falls (OR= 2.80 (1.11-6.65), p=0.023), antiosteoporotic treatment initiation (OR= 2.17, CI 95% 1.19-4.08, p=0.013) and increased age (OR=1,04 CI 95% 1,01-1.08, p=0.027) were significantly associated with the risk of a new severe fracture. Use of walking aids (RR=2.71 (CI 95% 1.15-6.39), p=0.02), diabetes (RR=3.70 (CI 95% 1.04 to 13.2), p=0.044), metastatic cancer (RR=15,5 (CI 95% 1.08 to 221), p=0.043) were positively associated with risk of death whereas antiosteoporotic initiation was negatively associated with this risk (OR=0.19 (CI 95% 0.07 to 0.49), p<0.001). Conclusion In patients hospitalized for severe osteoporotic fractures and managed in a FLS setting, initiation of an anti-osteoporotic treatment is associated with a decreased risk of mortality. In this population, a low BMD was a strong determinant of an imminent fracture risk. Acknowledgements This work received fundings of the French National Programme Hospitalier de Recherche Clinique (PHRC), of the French Society of Rheumatology and of the GRIO (Groupe de Recherche et d’Information sur les Ostéoporoses) Disclosure of Interests None Declared.