Bone turnover correlates with bone quantity but not bone microarchitecture in chronic hemodialysis.

Abstract

In chronic hemodialysis, high-turnover bone disease was associated with decreased bone mineral density (BMD), poor bone quality (chemical and structural), and increased fracture risk. Our aim was to correlate bone turnover markers (BTMs) with bone microarchitecture measured by trabecular bone score (TBS) before and after correction for BMD. We measured lumbar spine (LS), femoral neck, and 1/3 radius BMD and LS TBS by dual X-ray absorptiometry in 81 patients on permanent hemodialysis. Bone turnover was assessed using serum parathyroid hormone, osteocalcin, C-terminal crosslaps of type 1 collagen, procollagen 1 N-terminal propeptide (P1NP), and alkaline phosphatase (ALP). No patient had any partial or total parathyroidectomy and no previous or current treatment with anti-osteoporotic drugs. All BTMs correlated significantly with each other. Univariate regressions showed significant negative correlations between BTMs and BMD (best r = - 0.53, between P1NP and 1/3 radius Z-score) or BTMs and TBS (best r = - 0.27, p < 0.05 between ALP and TBS T-score). TBS correlated significantly with BMD at all three sites (best r = 0.5, between LS BMD and TBS T-score). Multivariate regression showed that TBS, crude or adjusted, correlated with LS BMD. No model retained any of the BTMs as independent variables due to the better prediction of BMD and multicollinearity. We showed a progressively impaired bone microarchitecture with increasing bone turnover in chronic hemodialysis. However, this correlation is no longer present when controlling for bone mass. This suggests that impaired bone microarchitecture and increased fracture risk are dependent upon factors other than high bone turnover.