SaO029THE BURDEN OF FRACTURES IN EARLY CHRONIC KIDNEY DISEASE: ANALYSIS OF CARTAGENE

INTRODUCTION: The association between ESRD and increased fracture risk is well known. Nevertheless, whether early CKD increases fracture risk is controversial, as most studies have focused on older adults and evaluated hip fracture only. Therefore, we aimed to evaluate the association between fracture incidence at any anatomical site and early CKD in middle-aged individuals. We also aimed to assess the effect of age and gender on this association and the role of CKD in fracture prediction. METHODS: Prospective analysis of CARTaGENE, a cohort of individuals from Quebec (Canada) aged 40 to 69 years recruited between 2009 and 2010. Individuals with eGFR above 30 ml/min/1,73m2are included. Early CKD is expressed continuously (eGFR levels using restricted cubic splines) or categorically (KDIGO Stages). Fracture incidence from recruitment to 2016 is identified using health administrative databases with validated algorithms. Associations between early CKD and fracture are assessed using Cox regression models adjusted for demographics, comorbidities, medication, bone mineral density, physical activity, and grip strength. The effect of age and sex on these associations is assessed using interaction terms. Predicted probabilities of fracture associated with early CKD and other traditional risk factors are computed for each gender at 45 and 65 years using previously built models. RESULTS: We included 19,391 individuals (51% women, mean age 54, mean eGFR 88 ml/min/1,73m2, 47% stage 2, 4% stage 3 CKD). 829 individuals had a fracture during the follow-up (380 non-CKD [4.0%], 400 CKD stage 2 [4.4%] and 49 CKD stage 3 [6.5%]). Decreased levels of eGFR and CKD stage 3 were associated with increased fracture incidence in unadjusted and adjusted models (Adjusted hazard ratio [HR] = 1.25 [1.05 to 1.49] for eGFR 60 vs 90 ml/min/1,73m2; HR = 1.65 [1.15 to 2.38] for eGFR 45 vs 90 ml/min/1,73m2; HR = 1.39 [1.02 to1.90] for CKD stage 3 vs non-CKD). eGFR levels of 75 to 120 ml/min/1,73m2 and CKD stage 2 were not associated with fractures. The cut-off for increased fracture risk was at 73 ml/min/1,73m2. Indeed, each 10 ml/min/1,73m2 reduction of eGFRwas linearly associated with fractures below but not above 73 ml/min/1,73m2 (HR = 1.22 [1.06 to 1.40] below; HR = 0.98 [0.92 to 1.05] above). CKD stage 3 was associated with increased fracture in younger individuals (HR=2.43 [1.27 to 4.64] at 45 years) but not in older individuals (HR= 1.11 [0.73 to 1.68] at 65 years; HR for interaction 0.68 [0.44 to 1.03]). Gender did not consistently modify the association between early CKD and fracture incidence. Predicted fracture probabilities for stage 3 CKD were greater than osteoporosis in younger but not in older individuals (6.6% for stage 3 CKD and 5.0% for osteoporosis at 45 years; 4.8% for stage 3 CKD and 8.6% for osteoporosis at 65 years). Compared to models including early CKD, models that excluded early CKD in fracture prediction had lower discrimination (p= 0.045) and underestimated fracture risk in individuals with stage 3 CKD (predicted/observed ratio of 0.77 without CKD and 1.00 to 1.04 with early CKD). CONCLUSIONS: In middle-aged adults, even early CKD is associated with increased fracture incidence, especially in younger individuals. Excluding early CKD in fracture prediction models resulted in lower discrimination and underestimation of fracture risk in individuals with CKD.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.