Amounts of Bile Acids and Bilirubin Removed During Single‐Pass Albumin Dialysis in Patients With Liver Failure

To the Editor, In severe liver failure the accumulation of several toxins including bilirubin and bile acids has been blamed to account for distant organ dysfunctions (1) and of further liver impairment (2). Accordingly, extracorporeal dialysis has long been proposed in liver failure, as a symptomatic treatment to remove toxins from the patient. The purpose is to provide a bridge, either to liver transplantation, or hepatic regeneration. As the toxins accumulated in liver failure are mainly bound to albumin, adding albumin in the dialysate has been shown to enhance the clearance rate of those toxins. To date, two albumin dialysis systems have been developed, using a dialysate in a closed circuit which regenerates the albumin (3). They require special training and expertise and a specific circuit involving sorbent columns. Effective albumin dialysis can be performed by using a conventional continuous venovenous hemodialyzer, but with a dialysate being enriched with albumin. This system, called single pass albumin dialysis (SPAD), has been initially described in clinical use by Seige et al. in 1999 and also uses the bound-solute dialysis principle but in a more simple and less expansive design (4). It can be performed on any standard hemodialyzer with a high-flux dialyzing polysulfone membrane and a standard dialyzing solution enriched with human albumin. Although in vitro studies confirmed the feasibility of such albumin dialysis (5) clinical studies report case reports (6) or more recently short cohorts of patients (7–9). In this study, our purpose was to measure the plasmatic decrease in total bilirubin and bile acids induced by SPAD session in 14 patients and to evaluate its correlation with the actual extracted amount. We performed a retrospective analysis of 14 consecutive patients treated with SPAD for acute liver failure (ALF-n = 2) or acute on chronic liver failure (AoCLF-n = 12). All patients were informed of the procedure, and as a retrospective observational study any Institutional Review Board approval was not searched for. The first three sessions were analyzed when carried out (n = 37). Beside the albumin dialysis, all patients received standard medical therapy for liver failure, including, when necessary, N-acetyl cysteine, anti-oxidant vitamins, terlipressin or norepinephrine in case of hepatorenal syndrome, 40 mg per day of solumedrol in case of proved alcoholic hepatitis, and systemic antibiotics in case of sepsis. Indications for SPAD were stage II or more hepatic encephalopathy and/or a total blood bilirubin level of at least 300 mmol/L. SPAD was conducted with a conventional venovenous hemodialyzer without ultrafiltration (HOSPAL Gambro, Lund, Sweden). The blood flow and dialysate rates were 150 mL/min and 1000 mL/h, respectively. Dialysate solutions were prepared to obtain a final concentration of 3.2% albumin. Each treatment lasted 10 h corresponding to 10 L of albumin dialysis. Measurement of toxins in the serum and in the effluent dialysate allowed us to calculate clearances rates and extracted amounts. Six patients were listed for liver transplantation, and were eventually transplanted after 3, 4, 5, 21, 22, and 45 days of treatment.A transplanted patient died 34 days after transplantation from Escherichia coli septicemia. Only two patients, suffering from acute alcoholic hepatitis and toxic hepatitis, survived without transplantation. They were discharged from the intensive care unit (ICU) at 16 and 63 days, after six and one SPAD sessions, respectively. Ammonia levels were not modified by the SPAD sessions, whereas changes in urea and creatinine plasmatic levels, although significant were tenuous (Table 1). The prothrombin ratio, factor V and fibrinogen levels were not modified by SPAD sessions. Conversely, platelet levels decreased slightly but significantly from 64 (34–118) g/L before SPAD to 58 (32–104) g/L after SPAD (median [1–3 quartile]). Total serum bilirubin and bile acids levels decreased significantly after each SPAD session of 19 (11–24) % and 33 (24–45) % from the initial values, respectively. The plasma clearance rates were 4.0 (2.0–5.0) mL/min and 14.2 (9.4–18.5) mL/min for the total bilirubin and bile acids, respectively. The total extracted amounts Therapeutic Apheresis and Dialysis 15(5):504–510 doi: 10.1111/j.1744-9987.2011.00980,00977,00978,00975.x © 2011 The Authors Therapeutic Apheresis and Dialysis © 2011 International Society for Apheresis