Background: There is ongoing controversy concerning optimum anticoagulation and buffering in continuous venovenous haemofiltration (CVVH). Regional anticoagulation with trisodium citrate also acting as a buffer in the replacement fluid has several advantages and disadvantages over prefilter citrate administration alone. We analysed a large cohort of patients with acute kidney injury (AKI) treated by the former method and hypothesized that it is safe and efficacious.
Methods: Patients admitted at the intensive care unit with AKI and a high bleeding risk, without exclusion of liver disease, treated by CVVH with citrate in a custom-made replacement solution were prospectively included. Patient and CVVH characteristics, including citrate accumulation, were evaluated in outcome groups. A standardized mortality rate (SMR) was calculated using the simplified acute physiology score II.
Results: Ninety-seven patients were included; metabolic control was adequate and did not differ between outcome groups, apart from lower pH/bicarbonate in non-survivors. Citrate accumulation was proven in 9% and was timely identified. These patients had about threefold higher plasma transaminases and higher CVVH dose and mortality. The hospital mortality was 60% with a SMR of 1.1 (95% confidence interval 0.90-1.40): age and hyperlactatemia, rather than CVVH-characteristics and citrate accumulation, predicted mortality in multivariable analysis.
Conclusion: In critically ill, patients with AKI at high risk of bleeding, CVVH with citrate-containing replacement solution is safe and efficacious. The risk for citrate accumulation is 9% and best predicted by levels of transaminases. It carries, when citrate is discontinued, no attributable mortality.