Sarcopenia combined with FRAX probabilities improves fracture risk prediction in older Chinese men.

OBJECTIVES: To examine whether the inclusion of sarcopenia in prediction models adds any incremental value to fracture risk assessment tool (FRAX).

DESIGN, SETTING, AND PARTICIPANTS: Data from a prospective cohort of 4000 community-dwelling Chinese men and women aged 65 years and older with adjudicated fracture outcomes were analyzed.

MEASUREMENTS: At baseline, femoral neck bone mineral density (BMD) was assessed, as were the clinical risk factors included in FRAX, along with additional appendicular skeletal muscle mass, grip strength, and gait speed. Sarcopenia was defined according to the Asian Working Group for Sarcopenia algorithm. Incident fractures were documented during the follow-up period from 2001 to 2013.

RESULTS: Of 4000 participants, 565 experienced at least 1 type of incident fracture and 132 experienced a hip fracture during a follow-up of 10.2 years. Hazard ratios (HRs) for 1-unit increase in FRAX score without BMD in men were 1.12 [95% confidence interval (CI) 1.08-1.16] for all fractures combined and 1.19 (95% CI 1.13-1.27) for hip fracture, and in women were 1.04 (95% CI 1.03-1.06) for all fractures combined and 1.08 (95% CI 1.06-1.11) for hip fracture. Similar to results of the FRAX score without BMD, HRs for 1-unit increase in FRAX score with BMD in men were 1.04 (95% CI 1.03-1.06) for all fractures combined and 1.19 (95% CI 1.13-1.25) for hip fracture, and in women were 1.04 (95% CI 1.03-1.05) for all fractures combined and 1.06 (95% CI 1.05-1.08) for hip fracture. Sarcopenia was significantly associated with all fractures combined (Adjusted HR 1.87; 95% CI 1.30-2.68) and hip fracture (Adjusted HR 2.67; 95% CI 1.46-4.90) in men but not in women. The discriminative values for fracture, as measured by the area under the receiver operating characteristic curve, were 0.60-0.73 and 0.62-0.76 for FRAX without and with BMD, respectively. Adding sarcopenia did not significantly improve the discriminatory capacity over FRAX (P > .05). Using reclassification techniques, sarcopenia significantly enhanced the integrated discrimination improvement by 0.6% to 1.2% and the net reclassification improvement by 7.2% to 20.8% in men, but it did not contribute to predictive accuracy in women.

CONCLUSIONS: Sarcopenia added incremental value to FRAX in predicting incident fracture in older Chinese men.