Fracture prediction tools differ based on CKD status
Click Here to Manage Email Alerts
An analysis of three fracture risk assessment tools revealed that major osteoporotic fracture risk varied by chronic kidney disease status, according to findings published in the Journal of Bone and Mineral Research.
“Middle-aged individuals with early decrease of kidney function (CKD stage 3) have a higher risk of fracture,” Fabrice Mac-Way, MD, FRCPC, assistant professor in the department of medicine at Université Laval, Québec, Canada, told Healio. “The discrimination and calibration of FRAX and QFracture for prediction of fracture is similar among adults with early CKD and those without CKD, whereas Garvan tends to underestimate fracture risk in CKD.”
Mac-Way and colleagues analyzed data from 9,114 adults with stage 2 CKD (mean age, 56 years), 757 adults with stage 3 CKD (mean age, 63 years) and 9,522 adults without CKD (mean age, 51 years) who participated in CARTaGENE (CAG), a population-based survey of Quebec residents recruited between 2009 and 2010. Researchers assessed renal function with baseline creatinine levels, categorized according to Kidney Disease: Improving Global Outcomes (KDIGO) guidelines, and used the Fracture Risk Assessment Tool (FRAX), the QFracture algorithm and Garvan to predict 5-year fracture probabilities. Discrimination and calibration (using standardized incidence ratios before and after recalibration) were assessed for each CKD stage.
“Because dual X-ray absorptiometry BMD was not available in the CAG survey, FRAX and Garvan probabilities were computed using their version without BMD,” the researchers wrote. “Garvan probabilities of any fracture at 5 years were computed using the full published equation. QFracture 5-year major osteoporotic fracture probabilities were computed using the 2012 version, since the 2016 version was not available at the time this study was conducted. Because FRAX equations are not publicly available, FRAX 10-year major osteoporotic fracture probabilities were manually computed using Canadian FRAX Web Version 4.0 with age and BMI rounded to the nearest multiple of 5.”
Primary outcomes were incident fractures, identified via claim databases through March 2016. Hip fracture was not included in the analyses.
Within the cohort, 830 participants experienced any fracture during a median 70 months of follow-up, including 352 major osteoporotic fractures.
Researchers found that the FRAX and QFracture prediction models similarly discriminated major osteoporotic fracture risk among adults with and without CKD. HRs for major osteoporotic fracture prediction using FRAX were 1.89 for adults without CKD (95% CI, 1.63-2.2), 1.64 for adults with stage 2 CKD (95% CI, 1.41-1.91) and 1.76 for adults with stage 3 CKD (95% CI, 1.1-2.82). HRs for fracture prediction using QFracture were 1.9 for adults without CKD (95% CI, 1.62-2.22), 1.57 for adults with stage 2 CKD (95% CI, 1.35-1.82) and 1.86 for adults with stage 3 CKD (95% CI, 1.19-2.91).
In contrast, Garvan probabilities were associated with incident fractures at any site among participants without CKD and with CKD stage 2, but not among adults with CKD stage 3. HRs were 1.36 for adults without CKD (95% CI, 1.22-1.52), 1.34 for adults with stage 2 CKD (95% CI, 1.2-1.5) and 1.11 for adults with stage 3 CKD (95% CI, 0.79-1.55).
Researchers found that, before recalibration, FRAX globally overestimated fracture risk, whereas QFracture and Garvan globally underestimated fracture risk. After recalibration, FRAX and QFracture were adequately calibrated for major osteoporotic fracture across all CKD strata; however, Garvan tended to underestimate any fracture risk among adults with CKD stage 3 (standardized incidence ratio = 1.31; 95% CI, 0.95-1.81).
“It is important to consider the increased fracture risk even in individuals with only a slight decrease of kidney function and considered middle-aged,” Mac-Way said. “FRAX and QFracture seem adequate tools to predict fracture risk in this category of individuals. Clinicians should assess fracture risk in the CKD stage 3 population using adequate fracture prediction tools.”
Mac-Way said more research is needed on the use of fracture prediction tools in more advanced CKD and on the role and efficacy of osteoporosis therapy in CKD. – by Regina Schaffer
For more information:
Fabrice Mac-Way, MD, FRCPC, can be reached at CHU de Québec Research Center, L’Hôtel-Dieu de Québec Hospital, 10 McMahon, Quebec City (Quebec) G1R 2J6 Canada; email: fabrice.mac-way@mail.chuq.qc.ca.
Disclosures: The authors report no relevant financial disclosures.
Read more about
Click Here to Manage Email Alerts