Bone Mineral Density as a Predictor of Subsequent Wrist Fractures: Findings From the Women's Health Initiative Study

Abstract

CONTEXT: Wrist fractures are common among postmenopausal women. Associations of bone mineral density (BMD) and 10-year predicted risk of major osteoporotic fracture (MOF) with wrist fractures are poorly characterized. OBJECTIVE: The objective was to examine associations between the Fracture Risk Assessment Tool (FRAX)-predicted risk of MOF, BMD, BMD change, and wrist fracture. DESIGN: This was a prospective observational study with a mean follow-up of 8.5 years. SETTING: This study included 40 US centers. PARTICIPANTS: A total of 11 392 participants from the Women's Health Initiative BMD Cohort aged 50-79 years at baseline were included in this study. INTERVENTIONS: None. MAIN OUTCOME: The goal was to measure incident wrist fracture. RESULTS: A FRAX-predicted MOF risk > /=9.3% identified 17% of the women aged < 65 years who subsequently experienced wrist fracture. Each one standard deviation lower BMD was associated with higher wrist fracture risk, with adjusted hazard ratio (95% confidence interval) of 1.66 (1.42-1.93) for femoral neck (FN) BMD and 1.45 (1.28-1.64) for lumbar spine BMD. Compared with FN BMD T score > /= -1.0, wrist fracture adjusted hazard ratios (95% confidence interval) were: 1.51 (1.06-2.16) for a T score between -1.01 and -1.49; 1.93 (1.36-2.72) for T score between -1.50 and -1.99; 2.52 (1.77-3.60) for a T score between -2.00 and -2.49; and 2.65 (1.78-3.95) for a T score < /= -2.5. Decrease in FN BMD between baseline and year 3 was associated with increased risk of subsequent wrist fracture; however, change in lumbar spine BMD was not. CONCLUSIONS: Lumbar spine and femoral neck BMDs were associated with incident wrist fracture, but the FRAX threshold recommended to identify screening candidates did not identify the majority of women who subsequently experienced wrist fracture. Improved understanding of determinants of wrist fractures is warranted.