Total-to-ionized calcium ratio predicts mortality in continuous renal replacement therapy with citrate anticoagulation in critically ill patients

INTRODUCTION: Regional citrate anticoagulation is safe, feasible and increasingly used in critically ill patients on continuous renal replacement therapy (CRRT). However, in patients with hepatic or multi-organ dysfunction, citrate accumulation may lead to an imbalance of calcium homeostasis. The study aimed at evaluating the incidence and prognostic relevance of an increased total to ionized calcium ratio (T/I Ca(2+ )ratio) and its association to hepatic dysfunction. METHODS: We performed a prospective observational study on n = 208 critically ill patients with acute kidney injury (AKI) and necessity for CRRT with regional citrate anticoagulation (CRRT-citrate) between September 2009 and September 2011. Critical illness was estimated by Simplified Acute Physiology Score II; hepatic function was measured with indocyanine green plasma disappearance rate. After achieving a steady state of calcium homeostasis patients were classified into tertiles according to the T/I Ca(2+ )ratio (<2.0 versus 2.0 - 2.39 versus ≥2.4). RESULTS: The T/I Ca(2+ )ratio was determined as an independent predictor for 28-day mortality in critically ill patients with AKI on CRRT-citrate confirmed by receiver operating characteristics and multivariate analysis (Area under the curve 0.94 ± 0.02; p<0.001). A T/I Ca(2+ )ratio ≥2.4 independently predicted a 33.5-fold (p<0.001) increase in 28-day mortality-rate. There was a significant correlation between the T/I Ca(2+ )ratio and the hepatic clearance (p<0.001) and the severity of critical illness (p<0.001). The efficacy and safety of citrate anticoagulation, determined by blood urea nitrogen, mean filter patency and bleeding episodes, were not significantly different between the tertiles. CONCLUSIONS: In patients on CRRT-citrate T/I Ca(2+ )ratio is closely related to the clinical outcome and emerged as an independent predictor of 28-day mortality. Larger studies are required to define the cut-off and predictive value for the T/I Ca(2+ )ratio. This ratio is associated with hepatic and/or multi-organ dysfunction and therefore an important therapeutic target.