Improving the delivery of continuous renal replacement therapy using regional citrate anticoagulation

Aims: Regional citrate anticoagulation during acute renal replacement therapy (RRT) effectively prevents extracorporeal thrombosis and avoids bleeding risk. There have been a number of citrate anticoagulation protocols published; but a simple and predictable scheme with standardized components and procedures, as well as clearly defined citrate pharmacokinetics, is needed for continuous RRT (CRRT) that is now used frequently in the critical care setting. The present study sets forth methodology with standardized blood flow and dialysate composition, and with citrate and calcium infusions that are quantitatively linked to extracorporeal blood flow rate - a predictable and easily replicated CRRT paradigm. Materials and methods: CRRT using continuous venovenous hemofiltration with dialysis (CVVHD) was standardized using 150 - 200 ml/min blood flow, calcium-free dialysate with only moderate sodium (135 mEq/l) and bicarbonate (28 mEq/l) concentrations, and ultrafiltration limited to that needed for overall fluid balance in the intensive care unit. Citrate infusion (ACD-A solution) into the extracorporeal blood and calcium repletion in blood returned to the patient were proportional to blood flow. Anticoagulation was accomplished by keeping extracorporeal ionized calcium below 0.4 mM/l. Filter performance, citrate removal and changes in calcium, sodium and alkali were evaluated longitudinally. Results: CVVHD using this protocol delivered urea clearance exceeding 2 l/h (48 l/d) when filter function was sustained. Filter longevity was markedly improved using citrate when compared with standard heparin anticoagulation, and nursing time spent on initiating and troubleshooting CRRT was approximately halved using this protocol. Sieving coefficients for urea, creatinine and citrate were approximately 0.9 and were sustained through nearly 3 days of filter use. Citrate clearance and removal were quantitatively linked to dialysate and ultrafiltration flow, resulting in 35 - 50% direct removal of the citrate-calcium chelate and reduced systemic citrate load. Serum tonicity and acid-base status were not problematic. The only notable side effect was modest calcium accumulation that necessitated reduction in calcium repletion rate. Conclusions: CVVHD is well suited to regional citrate anticoagulation. The present protocol is straightforward and predictable, with minor metabolic consequences that can be anticipated and adjusted. These results commend regional citrate anticoagulation to wider application.