Nafamostat Mesylate Versus Regional Citrate Anticoagulation for Continuous Renal Replacement Therapy in Sepsis-Associated Acute Kidney Injury

Overview

Sepsis-associated acute kidney injury (SA-AKI) is common in critically ill patients and often requires continuous kidney replacement therapy (CRRT). The choice of blood thinner (anticoagulation) during CRRT affects how long the filter works and the risk of bleeding. Citrate is the current standard blood thinner, but it can cause metabolic problems in patients with shock or liver dysfunction. Nafamostat mesylate (NM) is a newer alternative with a very short half-life and local action, which may offer both effectiveness and safety. However, no large, high-quality study has directly compared NM with citrate in SA-AKI patients. This study aims to show that NM is not worse than citrate for a key outcome called MAKE30 (a combination of death, continued need for kidney replacement therapy, or persistent kidney dysfunction at 30 days). We will also compare filter life, kidney recovery, death rates, hospital stay, bleeding events, and other outcomes. This is a multicenter, randomized, single-blind, non-inferiority trial. A total of 1162 patients will be assigned equally to receive either NM or citrate during CRRT. Patients will not know which treatment they get, but healthcare providers will know. The study includes adults aged 18-90 with sepsis and severe acute kidney injury requiring CRRT for more than 48 hours, who have given informed consent. Key exclusions include active bleeding risk, severe liver failure, pregnancy, or participation in another trial within 3 months. The main outcome is MAKE30 at 30 days. Secondary outcomes include filter life, days off CRRT, death rates, length of stay, bleeding, and changes in organ failure scores. Safety monitoring will focus on metabolic problems and citrate accumulation. The study is designed to test whether NM is non-inferior to citrate with a margin of 5%. If non-inferiority is shown, we will also test if NM is superior. Analyses will follow intention-to-treat principles.

This is a multicenter, randomized, single-blind, active-controlled, parallel-group, non-inferiority trial. The study will enroll 1162 eligible patients with sepsis-associated acute kidney injury (SA-AKI) requiring continuous renal replacement therapy (CRRT). Patients will be randomized in a 1:1 ratio to receive either nafamostat mesylate (NM) or citrate anticoagulation during CRRT. Randomization and Blinding Randomization will be performed using an interactive web response system (IWRS) with stratification by study site and baseline cardiovascular SOFA score. Patients will be blinded to treatment assignment. Healthcare providers will not be blinded due to the practical differences in anticoagulation administration and monitoring. Statistical analyses will be conducted under blinded conditions. Anticoagulation Protocol Both groups will receive CRRT according to standard institutional protocols. Anticoagulation targets and dose adjustments follow predefined operational guidelines: NM group: Continuous infusion of nafamostat mesylate, adjusted to maintain target activated clotting time or other regional anticoagulation parameters as specified in the protocol. Citrate group: Regional citrate anticoagulation with calcium replacement, adjusted to maintain post-filter ionized calcium levels within target range. Key Outcome Definitions MAKE30: A composite of all-cause death, persistent renal dysfunction (serum creatinine ≥2 times baseline), or continued need for or new initiation of kidney replacement therapy within 30 days after randomization. Filter lifespan: Time from CRRT initiation to filter failure (clotting or transmembrane pressure threshold) or elective filter change. Bleeding events: Classified by severity according to international criteria. Sample Size The sample size is calculated for non-inferiority of NM versus citrate for the primary outcome MAKE30. Assuming a 50% event rate in the citrate group, a non-inferiority margin of 5% (absolute risk difference), two-sided alpha = 0.05, power = 80%, and accounting for 10% dropout, the required total sample size is 1162 patients (581 per group). Statistical Analysis The primary analysis will follow the intention-to-treat principle. Non-inferiority will be concluded if the upper limit of the two-sided 95% confidence interval for the risk difference (NM minus citrate) is less than 5%. If non-inferiority is established, superiority testing will be performed using the same confidence interval. Key secondary endpoints (e.g., filter lifespan, 30-day mortality, days off CRRT) will be adjusted for multiplicity using the Hochberg method. Other secondary endpoints will be analyzed at nominal significance levels. Secondary analyses include survival analysis (Kaplan-Meier, Cox regression), generalized linear models, and comparisons of means or proportions as appropriate. A per-protocol analysis will be conducted as a sensitivity analysis. Safety Monitoring Safety outcomes include metabolic complications (alkalosis, acidosis, electrolyte disturbances), citrate accumulation (assessed by total-to-ionized calcium ratio), and adverse events related to anticoagulation. Data Monitoring An independent Data and Safety Monitoring Board (DSMB) will oversee the trial and conduct periodic reviews of safety data.