A mathematical model of regional citrate anticoagulation in hemodialysis

Regional citrate anticoagulation (RCA) during hemodialysis (HD) has several advantages over heparin anticoagulation, but calcium (Ca) derangements are a major concern necessitating repeated monitoring of systemic ionized Ca (Ca(2+)). We developed a mathematical model of Ca and citrate (Ci) kinetics during RCA.

Using patient- and treatment-related parameters, including pre-HD serum Ca and protein concentrations, hematocrit, blood and dialysate flow rates, dialysate composition and access recirculation, the model computes all relevant aspects of RCA based on physicochemical, biochemical and physiological principles such as chemical Ca and Ci equilibria, transmembrane solute fluxes and Ci metabolic rate. The model was validated in 17 treatments using arterial Ci infusion, Citrasate dialysate, and no postdialyzer Ca substitution.

Measured and predicted systemic Ca(2+) before HD was 1.08 +/- 0.06 and 1.05 +/- 0.05 mmol/l, respectively (difference -0.03 +/- 0.046, 95% confidence interval, CI, -0.055 to -0.007), and at 15 min into the treatment 1.01 +/- 0.05 and 1.02 +/- 0.05 mmol/l, respectively (difference 0.012 +/- 0.054, 95% CI -0.015 to 0.04). At 15 min, the measured and predicted predialyzer Ca(2+) was 0.33 +/- 0.06 and 0.39 +/- 0.05 mmol/l, respectively (difference 0.06 +/- 0.03; 95% CI 0.044-0.077), and the measured and predicted postdialyzer Ca(2+) was 0.7 +/- 0.05 and 0.61 +/- 0.05 mmol/l, respectively (difference -0.09 +/- 0.04; 95% CI -0.11 to -0.07). Bland-Altman analysis showed no systematic bias in these predictions.

This novel model of RCA shows excellent accuracy in predicting systemic, pre- and postdialyzer Ca(2+) concentrations and may prove valuable in both research and clinical applications of RCA.