Abstract
BACKGROUND AND OBJECTIVES: Circulating levels of anions that bind to magnesium and calcium are often altered in patients with CKD. However, it is unknown how these alterations affect the ionized fraction of magnesium and calcium. DESIGN, SETTING, PARTICIPANTS, & MEASUREMENTS: This cross-sectional study involved patients on maintenance hemodialysis and patients not on dialysis who visited the outpatient department of nephrology. We collected whole-blood samples to measure ionized magnesium and calcium concentrations. Adjusted anion gap was calculated as an integrative index of unmeasured anions. RESULTS: A total of 118 patients on hemodialysis and 112 patients not on dialysis were included. Although the prevalence of hypermagnesemia defined by total magnesium was much higher in patients on hemodialysis than in patients not on dialysis (69% versus 12%; P<0.001), the prevalence of hypermagnesemia defined by ionized magnesium did not differ significantly (13% versus 18%; P=0.28). Among patients on hemodialysis with high total magnesium, 83% had normal or low ionized magnesium. Consequently, the mean ionized fraction of magnesium in patients on hemodialysis was significantly lower than that in patients not on dialysis (51% versus 63%; P<0.001). Similarly, the mean ionized fraction of calcium in patients on hemodialysis was lower than that in patients not on dialysis (55% versus 56%; P<0.001). In patients on hemodialysis who had a higher adjusted anion gap than patients not on dialysis (mean [SD]: 14.1 [2.2] versus 5.1 [3.1]), the ionized fractions of magnesium and calcium were inversely associated with the adjusted anion gap. Furthermore, the anion gap significantly improved predictions of ionized magnesium and calcium in patients on hemodialysis. CONCLUSIONS: Anions that accumulate in patients on hemodialysis contribute to the lower ionized fraction of magnesium and calcium. Equations that incorporate the anion gap provide better predictions of ionized magnesium and calcium in patients on hemodialysis.