Predictive value of quantitative ultrasound parameters in individuals with chronic kidney disease: A population-based analysis of CARTaGENE.

Abstract

The use of calcaneal quantitative ultrasound (QUS) to predict fractures has not been well studied in early CKD populations. We compared the association of QUS with incidental fractures and its predictive properties in non-CKD and CKD individuals. Analysis of a prospective population-based survey of 40- to 69-year-old individuals recruited between 2009 and 2010. QUS parameters (stiffness index [SI], speed of sound [SOS], broadband attenuation [BUA]) were measured at baseline. Renal function was measured using baseline creatinine and was classified into CKD stages (non-CKD, stage 2, stage 3). Fracture incidence at any site or at major osteoporotic fracture sites for up to 7 years of follow-up was identified in administrative databases using a validated algorithm. The association (age-adjusted hazard ratio per standard deviation decrease in Cox models), discrimination (c-statistic) and calibration (standardized incidence ratio [SIR]) of QUS parameters with fracture outcomes was computed in each CKD stratum. We included 18,306 individuals (9,011 non-CKD; 8,595 CKD stage 2; 700 CKD stage 3). During a median follow-up of 70 months, we identified 782 fractures at any site and 326 major osteoporotic fractures. Although all QUS parameters (SI, SOS and BUA) were associated with any or major fracture incidence in non-CKD and CKD patients, the magnitude of these associations was lower for any fracture and for BUA. QUS parameters moderately discriminated incidental fractures across CKD strata but underestimated fracture incidence in CKD stage 3 even after adjustment for demographics and clinical risk factors. At a given QUS value, CKD stage 3 patients had higher fracture risk than non-CKD and CKD stage 2 patients. QUS parameters are associated with fracture incidence in both non-CKD and CKD but underestimate fracture incidence in individuals with early CKD.