Evaluation of Regional Anticoagulation With Citrate in Extended Hemodialysis

The investigators conduct an open label randomized clinical trial with cross-over between the groups. The study will be performed in all the intensive care units (ICU) at the Heart Institute and the Cancer Institute, both hospitals related with the University of São Paulo Medical School, at the city of São Paulo, Brazil.

All inpatients over 18 years in the ICU with acute kidney injury (AKI) who are arranged for extended hemodialysis will be invited to participate in the study. After signing an informed consent formulary, the patients went to randomization between two groups: citrate or control therapy. The research team randomly drawn the patients using a box with twenty entries for each group, totalizing forty units. Thus, after the inclusion of forty patients, it is guaranteed that half of them will start in the citrate group and the other half in the control group. Afterwards, the box is refilled to its maximum capacity of forty units for the next draw.

The randomization process determine the type of anticoagulation that the patient will be submitted in the first dialysis session. Subsequently the subject will participate in the other group, alternating between the two modalities until the maximum of six extended dialysis for the protocol.

In the control group, the decision on the use of heparin or continuous saline infusion as an anticoagulation method will be based on the presence of contraindications to the heparin use, which would be I) platelet levels <150,000/mm3; II) active or recent bleeding; III) reduction in hemoglobin levels of more than 2.0g/dl in less than 24 hours; IV) invasive procedures or surgeries realized in the last 7 days or scheduled to the next 24 hours.

There may be a change in the choice of method (citrate versus control), if the nephrologist consultant team deems it necessary, due to demands of care, particularly if there is early or repeated coagulation of the system. The proper functioning of the hemodialysis catheter will be evaluated before the dialysis initiation, and in case of malfunction, it will be properly replaced before the procedure starts.

The study foresees the inclusion of 800 dialytic procedures. Half of them are expected to be performed in the citrate group and half in the control group. As each patient can perform up to six procedures, it is estimated that approximately 200 patients will participate in this study. Data in the other intermittent dialysis not included in this trial and performed in the intensive care setting during the same period will be registered.

Patients will be excluded if they present: I) acute liver failure; II) hemorrhagic stroke in the last 30 days; III) platelet levels below 20,000/mm3 and IV) active bleeding requiring transfusion of two or more red blood cell packs within 24 hours.

Decisions about the need for renal replacement therapy, modality and time will be made by the nephrologist attendant. Intermittent methods are preferred performed in stable patients with no vasoactive drug, or with low doses of these drugs (norepinephrine <0.2 mcg/kg/min and dobutamine <5 mcg/kg/min).

All slow extended dialysis (SLED) in this trial were performed in 6 to 8 hours using a single-pass batch machine (Genius 90 Therapy System® Fresenius Medical Care, Bad Homburg, Germany), initiated by a nephrology specialized nurse and managed by an exclusive attendant. In this device, blood and dialysate flow rates are equal and set in 180 ml/min. We used only high-flux polysulfone filter (Ultraflux® AV 600S, Fresenius Medical Care). In the SPB device, the dialysate is stored in an air-free 90 liters glass container (batch system). Dialysate solution after proper reconstitution has sodium 138 mEq/l, bicarbonate 32.6 mEq/l, magnesium 1.0 mEq/l, chloride 111 mEq/l, potassium 2.0 or 3.0 mEq/l and calcium 2.5 mEq/l (5.0 mg/dl). Dialysate temperature was set in 36-37°C.

Citrate dose was tailored to achieve 3 mmol per filtered blood liter using a citric acid solution 2.2% 800 ml bag (each 100 mL of the solution contains 730 mg of citric acid, 2.45 g of dextrose and 2.2g of sodium citrate). Calcium replacement began if the patient has a serum ionized calcium less than 4.4 mg/dl (or 1.10 mmol/l) before it starts or during the session. We obtained blood samples at every two hours to adjust the calcium infusion rate targeting serum calcium from 4.4 to 5.3 mg/dl (or 1.10 to 1.32 mmol/l), using a calcium solution with 10% calcium chloride solution (concentration at 0.133 mmol/ml). The post filter ionized calcium addressed was 2.4 to 2.8 mg/d (or 0.6 to 0.7 mmol/l). Furthermore, in the control group, the equipment receive continuous saline in a infusion rate of 180 ml per hour or unfractionated heparin 1,000 units in bolus at the initiation of the procedure and maintained at 500 units per hour in a specific syringe pump in the device.