Estimating Net Bone Formation Relative to Resorption Using Reference Bone Turnover Markers

Context: Individually, bone resorption or formation markers do not reflect bone balance. Objective: (1) Combine reference bone resorption [collagen type I C-telopeptide (CTX)] and formation [procollagen type I propeptide (PINP)] markers to estimate balance by creating a bone balance index (BBI); (2) examine associations of BBI, CTX, or PINP with bone mineral density (BMD) change. Design: Mixed effects linear regression quantified associations of BBI, CTX, or PINP with BMD change rate. Setting: Community-based cohort (Study of Women's Health Across the Nation). Participants: 535 women transitioning through menopause. Main Outcome Measures: Annualized percent change in lumbar spine (LS) and femoral neck (FN) BMD. Results: In mixed effects linear regression, adjusted for age, body mass index, race/ethnicity, menopause transition stage, and study site, more negative BBI (less favorable balance) related to more BMD loss. Each SD decrement in BBI was associated with a 0.26% greater decline in LS BMD annually and 0.42% greater decline in FN BMD annually (each P < .0001). In separate models, accounting for the previous covariates, greater CTX or PINP predicted more bone loss. Per SD increment in log2CTX, annual BMD decline was 0.37% greater (P < .0001) and 0.47% greater (P < .0001) at the LS and FN, respectively. Per SD increment in log2PINP, annual BMD loss was 0.22% (P < .0001, LS) and 0.19% (P = .01, FN) greater. Conclusion: When combined as a BBI, CTX and PINP estimate bone balance; more negative BBI predicts faster BMD loss. Separately, CTX or PINP reflect overall amount of bone turnover; greater CTX or PINP are related to more bone loss.