An index of net bone formation relative to resorption is associated with incident fracture during the menopause transition and postmenopause

We previously created a bone balance index (BBI) that combines bone resorption and formation markers (CTX and PINP, respectively) to estimate bone balance. The bone balance index quantifies net bone formation relative to resorption; greater values indicate a more favorable balance. This study examined the association of the BBI with incident fracture across the menopause transition and tested each individual bone turnover marker (BTM) (either CTX or PINP) as a separate predictor of fracture. We included 451 participants from the Study of Women’s Health Across the Nation (SWAN) cohort study, with 2667 longitudinal observations from pre-, early peri-, late peri-, or post-menopause. Participants did not report fractures or the use of bone-active medications before their BTM assessments. Using repeated measures Cox proportional hazards regression, we examined the associations of the BBI, CTX, or PINP with 2 different fracture outcomes: incident fracture at any site or incident fracture at a major osteoporotic fracture (MOF) site (spine, hip, radius, or humerus). Women were censored at the first use of bone-beneficial medications during follow-up. Covariates were age, BMI, race/ethnicity, family history of hip fracture, cigarette use, diabetes mellitus, bone-detrimental medication use, menopause transition stage, FN BMD, and SWAN study site. The mean age across the BBI, CTX, and PINP assessments was 52 yr. Seventy-eight women sustained an incident fracture over a 15.1 yr of follow-up. Adjusted for covariates, each SD increment in the BBI was associated with a 20% smaller hazard for fracture at any site (hazard ratio [HR] [95% CI]: 0.80 [0.67, 0.97], p = .02) or a 25% lesser hazard for fracture at a MOF site (HR [95% CI]: 0.75 [0.61, 0.85]), p = .01). In contrast, when considered singly in adjusted analyses, neither CTX (p > .2) nor PINP (p > .1) was related to either fracture outcome. We conclude that BBI predicts fracture, independent of clinical and densitometric fracture risk factors.