Citrate Anticoagulation in Renal Replacement Therapy: Impact of a High Post-filter Calcium Target on Efficacy

Overview

Regional citrate anticoagulation (RCA) is the recommended method for anticoagulation in continuous renal replacement therapy (CRRT). However, the optimal post-filter ionized calcium (iCa) target level remains unclear. Currently, it is titrated to a post-filter iCa target ranging from 0.25 to 0.35 mmol/L, which is derived from a few underpowered trials. There are potential side effects associated with citrate administration, which may be increased in patient with liver failure and/or tissue dysoxia, such as alkalemia, acidemia, hypernatremia, hypocalcemia, hypomagnesemia, and citrate accumulation. Consequently, citrate anticoagulation is contraindicated in the most severe cases. The challenge is to use the minimum necessary dose of citrate to ensure both effective anticoagulation of the circuit and limit citrate administration to reduce the risks of metabolic complications and accumulation. This approach expands the indications for citrate, standardizes practice, and reduces financial costs. Investigators hypothesized that increasing the post-filter iCa target in RCA can limit the dose of citrate, thereby avoiding adverse effects (safety) without compromising the effectiveness of the treatment in preventing filter clotting. The aim of this study is to evaluate the impact of an increased post-filter iCa target from 0.25-0.35 to 0.35-0.45 mmol/L on the incidence of filter clotting for RCA-CRRT in critically ill patients. Investigators are designing a multicenter randomized controlled non-inferiority study.

RCA-CRRT will be ordered based on clinical indications and will be performed according to a standardized protocol (available as online supplementary material) in continuous veno-venous hemofiltration mode with the same system (Prismaflex®; Gambro-Baxter, Deerfield, IL, USA) and a 0.9 m2 high-flux AN69 membrane. Blood flow will be maintained between 120 and 180 mL/min according to the patient's ideal body weight. The prescribed dose of filtration will be 30 mL/kg/h to achieve a delivered dose of 20-25 mL/kg/h, following KDIGO guidelines. A citrated replacement solution (Regiocit®; Gambro-Baxter), containing 18.0 mmol/L of citrate, will be delivered continuously to the blood before the filter of the extracorporeal circuit. The rate of infusion of predilution replacement flow will be coupled to the blood flow, aiming for a stable citrate concentration in the extracorporeal circuit. The initial citrate dose will be 3.0 mmol/L of blood, and then citrate flow rate will be adjusted to the post-filter iCa target according to the protocol. Post-filter iCa will be measured on ABL90 FLEX PLUS™ (Radiometer Medical©, Copenhagen, Denmark) blood gas analyzer 15 minutes after any change in dose and then every 6 hours. Calcium chloride will be administered to the patient through a central line to maintain systemic-iCa within 1.00-1.30 mmol/L. Fluid removal rates will be left to the discretion of the attending physician in order to achieve optimal fluid balance. Additionally, metabolic monitoring will be carried out by a blood ionogram every 12 hours. The quantitative parameters will be presented as median and interquartile range \[IQR\], and comparisons will be made using either Student's t-test or the Mann-Whitney U test depending on whether the assumptions of the t-test are met or not. Categorical data will be reported as the number and percentage (%) and will be compared using Fisher's exact or chi-square test, as appropriate. The incidence of filter clotting will be expressed in absolute values (n) and percentage (%). Comparison between groups will be performed using Pearson's Chi-square test. The analysis of the primary endpoint will be conducted on a per-protocol basis as a first intention (the most conservative approach in a non-inferiority study) and on an intention-to-treat basis. Filter lifespan until clotting curves according to the post-filter iCa2 target will be plotted using the Kaplan-Meier method and compared using the log-rank test.