Incident vertebral fracture and longitudinal BMD change in Chinese postmenopausal women with early CKD: Peking Vertebral Fracture Study

Abstract

Early chronic kidney disease (CKD) and non-CKD individuals had similar morphometric vertebral fracture (mVF) incidence and longitudinal bone mineral density (BMD) change. CKD did not modify the association between BMD and incident mVF status. Patients with a higher baseline BMD had a higher longitudinal BMD loss in early CKD.

Purpose: The aim of this 5-year longitudinal cohort study was to compare the risk of incident morphometric vertebral fracture (mVF) and longitudinal bone mineral density (BMD) change between individuals with early chronic kidney disease (CKD) and those without CKD.

Methods: A total of 869 Chinese postmenopausal women were enrolled in the study. Serum creatinine levels were assessed using standard methods, and estimated glomerular filtration rate (eGFR) was calculated using the Chronic Kidney Disease Epidemiology Collaboration equation. Incident mVF was confirmed through lateral radiographs of the thoracolumbar spine. BMDs at the lumbar spine (LS) and femoral neck (FN) were measured using dual-energy X-ray absorptiometry. CKD was defined based on eGFR values: 60-89 mL/min/1.73m² for stage 2 (n = 511) and 30-59 mL/min/1.73m² for stage 3 (n = 92). The non-CKD group included individuals with an eGFR greater than or equal to 90 mL/min/1.73m².

Results: The incidence of mVF was not statistically different between individuals with early CKD and those without CKD (4.1% in non-CKD, 6.3% in CKD stage 2, and 7.6% in CKD stage 3; p = 0.348). Neither eGFR nor CKD status was significantly associated with incident mVF in the multivariate logistic regression analysis. Baseline BMD T-scores were negatively associated with incident mVF (LS T-score, OR = 0.603, 95% CI = 0.468-0.777; FN T-score, OR = 0.511, 95% CI = 0.350-0.746). No evidence of interaction between BMD T-scores and CKD status were identified (p = 0.284-0.785) . The differences in longitudinal BMD changes between non-CKD and CKD groups were comparable (FN BMD: -6.31 ± 7.20% in non-CKD, -5.07 ± 8.20% in CKD stage 2, and -4.49 ± 8.40% in CKD stage 3, p = 0.556; LS BMD: -1.38 ± 8.18% in non-CKD, -0.32 ± 7.14% in CKD stage 2, and 1.5 ± 6.97% in CKD stage 3, p = 0.406). Individuals with a higher baseline FN BMD showed a greater longitudinal FN BMD loss (r = -0.185, p < 0.001) .

Conclusions: Our study revealed that early CKD was not associated with an increased risk of incident mVF or greater longitudinal BMD loss. Moreover, CKD did not modify the association between BMD and the risk of incident mVF, suggesting that the management and prevention of fractures in early CKD should be approached similarly to those without CKD. Measurement of BMD appears to be crucial for predicting incident mVF risk and longitudinal bone loss in early CKD.

Keywords: Bone mineral density; Bone turnover markers; Early chronic kidney disease; Incident morphometric vertebral fracture.