This trial was designed as a multi-center randomized, controlled trial to recruit 202
patients who develop acute kidney injury (AKI) described as Kidney Disease: Improving Global
Outcomes (KDIGO) stage 2 (serum creatinine, 2.0 times the baseline level; urine output,
<0.5mL/kg/h for 6 or more hours) after cardiac surgery. Patients were randomized in a 1:1
ratio to 1 of the 2 treatment groups (Early vs. Delayed renal replacement therapy (RRT))
using a computerized system.
Sample size determination : power calculation were performed based on the primary end point
(operative mortality). The expected operative mortality in the control group with delayed
initiation of RRT was 55% based on the literature. Differences between treatment groups were
to be detected with a power of 80%, if the operative mortality of with early initiation of
RRT was 35% or less. The expected treatment effect of 20% was calculated on the mortality
differences between early and delayed RRT reported in previous studies. A required sample
size for the final analysis was 101 patients per treatment group, 202 patients in total
(level of significance, α = 0.05; type II error, β= 0.02; potential dropouts= 5%).
Early RRT was initiated within 6 hours of diagnosis of stage 2 AKI.
Delayed RRT was initiated if any one of the following absolute indications for RRT is present
serum urea level higher than 100mg/dL and/or with uremic encephalopathy
serum potassium level higher than 6mmol/L and/or with electrocardiography abnormalities
urine production lower than 0.3mL/kg/hr for 24 hours
pH of 7.15 or less and/or severe hypotension due to metabolic acidosis
organ edema in the presence of AKI resistant to diuretic treatment.
The primary end point is operative mortality (described as any death, regardless of cause,
occurring (1) within 30 days after surgery in or out of the hospital, and (2) after 30 days
during the same hospitalization subsequent to the operation).
The secondary end points included 90 day overall survival, cardiovascular mortality, RRT
dependence, and major adverse kidney events (MAKE), adverse events related to RRT or vascular
access, duration of mechanical ventilator support and intensive care unit stay, and hospital
length of stay.
RRT delivery
: Once RRT was initiated, both groups were treated using continuous venovenous
hemodiafiltration (CVVHDF) with identical settings. Initial target dose of hemodiafiltration
was 25 to 50mL/kg/hr depends on the decision of attending physician and further adjusted
according to the metabolic needs of the patient. Replacement fluid was delivered into the
extracorporeal circuit before the filter with a ratio of dialysate to replacement fluid of
1:1. Blood flow was maintained between 100 to 250mL/min. Regional anticoagulation with
nafamostat (dosage 20-50mg/hr) was used to prevent circuit clotting if necessary.
RRT was discontinued if renal recovery defined by urine output (>1mL/Kg/hr for 8 hours or
more or >1000mL/24h without diuretics; >2000mL/24h with diuretics) and creatinine clearance
(>20mL/min) occurred.
If cessation criteria were not fulfilled after 7 days, conversion to intermittent
hemodialysis would be considered.