Causal assessment of CKD-MBD biomarker alterations on CKD progression through a g-formula analysis in the EQUAL study

INTRODUCTION: Chronic Kidney Disease-Mineral and Bone Disorder (CKD-MBD) may be both a cause and a consequence of CKD progression. Here, we examine the individual and joint effect of longitudinal circulating phosphate, parathyroid hormone, and calcium levels on estimated glomerular filtration rate (eGFR) decline, the risk of starting dialysis and the composite risk of death or dialysis, in a cohort of elderly patients with advanced CKD not on dialysis. Secondly, we explore whether these effects differ between men and women. METHODS: or less from six European countries. To avoid bias due to informative censoring, competing risks and time-varying confounding, we used the g-formula for causal inference. RESULTS: Isolated hyperphosphatemia and hyperparathyroidism were associated with a higher risk of CKD progression, whereas isolated hypercalcemia was not. The combination of hyperphosphatemia with hyperparathyroidism, and of hypocalcemia with hyperparathyroidism were also associated with significantly higher risk. The most adverse phenotype was the combination of hyperphosphatemia, hypocalcemia and hyperparathyroidism, which led to a mean difference in eGFR after three years of -3.71 ml/min (95% confidence intervals: -5.44, -2.15) and a 75% higher risk of starting dialysis (hazard ratio 1.75, 95% confidence interval 1.35-2.10), compared to having all biomarkers in their reference ranges. Most of this risk was attributable to hyperphosphatemia. Although the effect of abnormal mineral biomarkers on eGFR decline was similar between sexes, the associated risk of starting dialysis seemed stronger in men than in women. CONCLUSIONS: Among older men and women with advanced CKD not receiving dialysis, the presence of hyperphosphatemia, especially when combined with hypocalcemia and hyperparathyroidism, leads to an increased risk of CKD progression.