Background The prevention of fragility fractures in patients with sarcoidosis (Sa) is a serious concern since risk factors are poorly understood and vitamin D (vitD) or calcium prescription is a matter of hesitation in these patients.
Objectives To evaluate factors increasing the risk of osteoporosis (OP) in Sa.
Methods We prospectively included 142 consecutive patients with histologically proven Sa. Biological and clinical parameters of Sa activity and severity, BMD (DXA, Lunar Prodigy) and vertebral fracture prevalence on lumbar and thoracic spine (with Genant’s semi-quantitative method of) were assessed.
Results 80 women (51 menopaused) and 62 men, (age 51 ±11.9), with a mean Sa duration of 9.5±7 years were included. 28 patients had never received corticosteroids (CCS), 33 had received vitD supplements, and 46 bisphosphonates, within the 6 months preceding the study. Fragility fractures had occurred in 23.5 % of the patients, (vertebral fractures in 13.7%). 40.1% had BMD T-score <-1DS in at least one site, (20 patients with OP). Nevertheless, the average BMD was not low (mean Tscore -0.44DS at the lumbar spine). VitD supplementation significantly increased serum 25(OH)vitD(p<0.0001)without any significant variation of the calcemia. In multivariate analysis, fracture (OR:3.69, 95%IC, 0.96-14.21, p=0.098), low calcium intake (OR:5.79, 95%IC, 1.82-18.42, p=0.001), and menopause (OR:3.32, 95%IC, 0.67-16.47, p=0.068), were predictive of a low BMD. Age (OR:1.09 95%IC, 1.0-1.18, p=0.021), the NYHA score(OR:3.28 95%IC, 0.79-13.61, p=0.089), the cumulative dose of CCS (OR:1.48, 95%IC, 1.06-2.05, p=0.023), and low BMD (OR:3.75, 95%IC, 0.91-15.36, p=0.093), were predictive of fracture. VitD showed a protective effect on BMD for serum levels between 10 and 20ng/ml (low BMD OR:0.17, 95%0.04-0.83 p=0.028). In a surprising way, values above this threshold were associated with an increased risk of fracture (OR:4.71, 95%IC, 1.01-22.01, p=0.052). Low 25(OH)vitD and not 1.25(OH)2D, was significantly correlated with disease activity parameters (disease flare, ACE, severity of lung involvement, erythrocyte sedimentation rate).
Conclusions Sa patients have a high risk of fracture despite not lowered BMD. There could be a bimodal relationship between VitD levels and fracture risk. Predictive factors of fracture, independent from BMD and reflecting the severity of Sa, should be considered when a preventive treatment of OP is planned
Acknowledgements This study was supported by the GRIO “group de recherche et d’information de l’osteoporose”.
Disclosure of Interest None Declared