- Email: [email protected]
Sustained low-efficiency dialysis with regional citrate anticoagulation in medical intensive care unit patients with liver failure: A prospective study
Journal of Critical Care xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Journal of Critical Care journal homepage: www.jccjournal.org
Sustained low-efficiency dialysis with regional citrate anticoagulation in medical intensive care unit patients with liver failure: a prospective study☆,☆☆ Tobias Lahmer, MD ⁎, Marlena Messer, MD, Sebastian Rasch, MD, Analena Beitz, MD, Christopher Schnappauf, MD, Roland M. Schmid, MD, Wolfgang Huber, MD II. Medizinische Klinik und Poliklinik, Klinikum rechts der Isar der Technischen Universität München, Munich, Germany
a r t i c l e
i n f o
Keywords: SLED Citrate anticoagulation Liver failure Critical care
s u m m a r y Purpose: Patients with liver failure requiring dialysis are at increased risk for citrate accumulation during sustained low-efficiency dialysis (SLED). The aim of this study was to evaluate the feasibilty of citrate SLED in critical ill patients with liver failure and investigate predictive parameters regarding citrate accumulation. Materials and methods: This is a prospective study in 24 medical intensive care unit patients with liver failure and a total of 43 SLED runs (maximum of 3 runs per patient) using citrate anticoagulation. Liver function was characterized before SLED using not only laboratory parameters but also determination of the plasma disappearance rate of indocyanine green. In addition, blood gas parameters as well total calcium and citrate in serum were measured at baseline and defined time points during SLED. Results: Accumulation of citrate could be observed in all SLED runs, which were nearly normalized until the end of SLED and 24 hours after SLED, respectively. However, the critical threshold of total calcium/ionized calcium on ratio of greater than 2.5 was exceeded in only 1 patient. Equalization of initial metabolic acidosis was possible without major disturbances of acid base and electrolyte status. Liver function parameters showed poor predicitve capabilities regarding citrate accumulation. Conclusions: Despite substantial accumulation of citrate in serum, SLED is save and feasible in patients with liver failure using a citrate anticoagulation. Careful monitoring of electrolytes and acid base status is mandatory to ensure patient safety. © 2015 Elsevier Inc. All rights reserved.
1. Introduction One of the most often complications in hospitalized patients is still an acute kidney injury (AKI), which significantly affects morbidity and mortality of intensive care unit (ICU) patients if renal replacement therapy (RRT) is required [1,2]. In most cases, continuous RRT is used in ICUs, which is often seen as the preferable RRT in critically ill AKI patients [3-5]. ☆ Competing interests: The author(s) declare that they have no competing interests. Declaration of interest: None of the authors have any potential financial conflict of interest related to this manuscript. None of the authors have got any funding or financial support regarding this manuscript. The manuscript has not been submitted or accepted elsewhere. All authors fulfill the criteria given in the authorship paragraph. No writing assistance other than copy editing was provided in the preparation of the manuscript. ☆☆ Authors' contribution: TL, MM, SR, AB, CS, RMS, and WH. Study design: TL, MM, SR, CS, RMS, and WH. Collected data: TL, MM, SR, AB, CS, RMS, and WH. Analyzed data: TL, MM, AB, SR, and WH. Wrote manuscript: TL, MM, AB, SR, CS, RMS, and WH. Study is approved by the Institutional Review Board of the Technical University of Munich, Munich, Germany. Written informed consent was obtained. ⁎ Corresponding author at: II. Medizinische Klinik und Poliklinik, Klinikum rechts der Isar der Technischen Universität München, Ismaninger Str. 22, 81675 Munich, Germany. Tel.: +49 89 4140 2267; fax: +49 89 4140 6243. E-mail address: [email protected] (T. Lahmer).
However, intermittent RRTs, so-called sustained low-efficiency dialysis (SLED), are more and more used in critically ill patients with AKI [6]. Sustained low-efficiency dialysis, with a running time of 8 to 12 hours, shares the advantages of a conventional intermittent (4 hours) and a continuous RRT (up to 72 hours) [7]. Moreover, over the last years, citrate has emerged as a safe and efficacious alternative to heparin for extracorporeal circuit anticoagulation [8]. Citrate chelates ionized calcium (Caion), the most important cofactor of the coagulation cascade. Thus, regional anticoagulation with citrate is a very effective anticoagulation method for hemodialysis. Some studies on continuous venovenous hemodialysis (CVVHD) detected a reduction of bleeding complications, a longer filter lifetime, and a possible reduction of mortality in ICU patients with a citrate anticoagulation [9]. Therefore, regional citrate anticoagulation has also been recommended by the recent Kidney Disease Improving Global Outcomes guidelines. However, some intensivists are still reluctant to adopt this technique. Reasons may be the complexity; need of customized citrate solutions/replacement fluids; fear of metabolic complications (eg, hypocalcemia and metabolic alkalosis); and difficulties in predicting and preventing citrate accumulation, especially in patients with impaired liver function [10]. Around 50% of citrate is removed as a complex bound with Caion through the dialyzer by diffusion/convection. However, citrate can partly
http://dx.doi.org/10.1016/j.jcrc.2015.06.006 0883-9441/© 2015 Elsevier Inc. All rights reserved.
Please cite this article as: Lahmer T, et al, Sustained low-efficiency dialysis with regional citrate anticoagulation in medical intensive care unit patients with liver failure..., J Crit Care (2015), http://dx.doi.org/10.1016/j.jcrc.2015.06.006
2
T. Lahmer et al. / Journal of Critical Care xxx (2015) xxx–xxx
enter the systemic circulation [11]. The hepatic citric acid cycle is the major site of citrate metabolization, which leads to the release of Caion into the systemic circulation. This is almost independent of renal function and urinary excretion [11,12]. However, in cases of impaired liver function with an also impaired citrate metabolism, the risk of citrate accumulation is rising. Drop of Caion due to complex binding between citrate and Caion requiring more calcium chloride substitution at the venous line of the extracorporeal circuit can result [12]. Moreover, this leads to an increase of total calcium (Catot) concentration, which is the sum of Caion, protein, and citrate-bound calcium, and an increased Catot/Caion ratio might be observed. After former studies, a serum Catot/Caion ratio greater than or equal to 2.5 might be a critical threshold of potential citrate accumulation [11,12]. Other side effects could be metabolic acidosis combined with an enlarged anion gap caused by reduced citric acid cycle production of bicarbonate out of citrate and accumulation of negative loaded citrate ions [8,9]. These possible side effects may be the reason why data on the feasibility of citrate CVVHD in liver failure patients are scarce. Moreover, there are only small data in critically ill patients with liver impairment using SLED [13,14]. Therefore, the aim of our study was to determine predictors for citrate accumulation and to investigate the feasibility of citrate anticoagulation in patients with impaired liver function.
2. Materials and methods 2.1. Patients Twenty-four ICU patients between aged 29 and 73 years with decompensated liver cirrhosis (18 patients) or acute liver failure (6 patients) who needed RRT were included in this study. A total of 43 SLED runs (maximum, 3 per patient) were analyzed in this study cohort. Liver cirrhosis was diagnosed either by histologic specimen, by ultrasound and/or computed tomography, and by clinical criteria for instance ascites or esophageal varices. We defined an acute liver failure as an abrupt loss of liver function without preexisting liver disease. Before SLED treatment, we used the model of end-stage liver disease score (MELD score), the Child-Pugh score in case of cirrhosis, and laboratory liver function parameters (aspartate aminotransferase [ASAT], alanine aminotransferase [ALAT], bilirubin, and prothrombin time) to characterize baseline liver function. In addition, plasma disappearance rate of indocyanine green was performed. Sequential Organ Failure Assessment (SOFA) score as well the Acute Physiology and Chronic Health Evaluation (APACHE) II score was calculated in accordance to demonstrate the severity of the underlying disease (Table 1). Patients with
Table 1 Overview of baseline liver function parameters
MELD score (points) Child-Pugh score APACHE II score SOFA score ICG-PDR Prothrombin time (%) INR Bilirubin (mg/dL) ASAT (U/L) ALAT (U/L) Lactate (mmol/L)
Mean/range/SD
Reference range
35/25-40/6.5 all C 33/24-40/3.5 16/8-19/2.5 5.1/1.5-24/4.3 38/16-66/13 2.2/1.3-4.3/0.9 18.4/2.1-46.5/14.5 331/25-1261/331 157/15-1047/157 2.1/0.8-9.7/1.7
Maximum, 40 points Maximum, 71 points Maximum, 24 points 18-25 70-120 b1.2 10-50 10-35 b2.4
ICG-PDR indicates indocyanine green-plasma disappearance rate; INR, international normalized ratio. To characterize baseline liver function, the MELD score and the plasma disappearance rate of indocyanine green (ICG-PDR) were calculated in each patient, and the Child-Pugh score only in patients with cirrhosis. The APACHE II score and the SOFA score as well as the laboratory parameters were determined immediately before the start of SLED treatment. Data were expressed as mean (highlighted), range, and SD.
severe alkalosis (pH N7.55) or acidosis (pH b 7.1) and deficiency of Caion (b0.9 mmol/L) were excluded from our study. In accordance to our Institutional Review Board of the Technical University of Munich, Germany, this study was approved, and written informed consent was obtained.
2.1.1. Sustained low-efficiency dialysis treatment We used the commercially available hemodialysis Genius singlepass batch dialysis system (Fresenius Medical Care, Bad Homburg, Germany) for SLED treatment. It provides 90 L of bicarbonate dialysate per dialysis session. The dialysis system uses a double-sided roller pump that generates equal blood, and dialysate flows up to a maximum of 350 mL/min. Moreover, the Genius system contents a closed dialysate tank of 90 L, in which fresh and spent dialysate are stored together without any mixing of both compounds. The ultrafiltered water out of the patient plasma is collected in an ultrafiltrate recipient. High-flux FX60 dialyzers (Fresenius Medical Care) were used in all sessions. For all SLED treatments, a blood flow of 150 mL/min, which is equal to dialysate flow, was used. The dialysate solution is a compound of an instant (HC-90; Fresenius Medical Care) and liquid (DS-90; Fresenius Medical Care) component. The instant component contains sodium chloride,sodium bicarbonate, and glucose. The liquid component contains potassium chloride, calcium chloride, and magnesium chloride. In all patients, the HC-90 concentration HC 31-90, which contains 40 mmol/L sodium, 3 mmol/L potassium, 1 mmol/L calcium, 0.5 mmol/L magnesium, 48 mmol/L chloride, 0.067 mmol/L citrate, and 2 mmol/L HCL, and the DS-90 concentration DS 135/35-90, which contains 95 mmol/L sodium, 35 mmol/L bicarbonate, 60 mmol/L chloride, and 5.5 mmol/L glucose, were used. Citrate solution was produced in the local hospital pharmacy (100 mL consists of 22 g tri natriumcitrate dihydrate and 8 g citrate monohydrate in aqua destillata). Calcium chloride solution was produced in the local hospital pharmacy (500 mmol/L, 73.5 g calcium chloride dihydrate in 1000 mL aqua destillata). Sodium citrate flow was started with 60 mL/h, and calcium chloride flow was started with 10 mL/h, respectively, and adapted according to the required limits of calcium between 0.35 and 0.45 mmol/L postfilter and 1.00 and 1.10 mmol/L in the patient's circulation. According to the study protocol, citrate and Catot in serum were measured at baseline just before the beginning of SLED treatment, after 45 L, 90-L dialysate turnover, and 24 hours. Blood gas analyses of the patient's circulation and Caion postfilter were measured at baseline, after 1 hour, after 3 hours, 45 L, 9 hours, 90 L, and after 24 hours. The total run time of SLED was approximatley 10.5 hours. Citrate levels were measured enzymatically by the citrate-lyase method (MVZ Labor Limbach, Heidelberg, Germany). In this method, citrate is metabolized to oxalacetate, and acetate is catalyzed by the enzyme citrate lyase. Oxalacetate is reduced to malate and lactate by the enzymes L-malate-dehydrogenase and L-lactate-dehydrogenase in a nicotinamide adenine nucleotide hydrogen-dependent manner. Nicotinamide adenine nucleotide hydrogen is the measured variable and is equivalent to the amount of citrate.
2.2. Statistical analysis In our study, we used for statistical analyses the IBM SPSS Statistics 21 (SPSS Inc, Chicago, IL). Descriptive statistics were expressed with mean ± SD and range for normally distributed continuous data. The t test was performed for paired samples. Wilcoxon signed rank test was performed for paired samples for normally distributed data and not normally distributed data, respectively. A P value below a significance level of 5% (P b .05) indicates statistical significance.
Please cite this article as: Lahmer T, et al, Sustained low-efficiency dialysis with regional citrate anticoagulation in medical intensive care unit patients with liver failure..., J Crit Care (2015), http://dx.doi.org/10.1016/j.jcrc.2015.06.006
T. Lahmer et al. / Journal of Critical Care xxx (2015) xxx–xxx
Basic kidney parameters and noradrenaline dosages during SLED were presented in Table 2. Noradrenaline changes during SLED treatment were not statistical significant. The mean length of ICU stay was 24 days (minimum, 6 days; maximum, 59 days). Nineteen patients died in the ICU. Amonge the 5 survivors, kidney function recoverd in only 1 patient; all other needed further hemodialysis treatment.
Table 2 Overview of basic kidney and SLED parameters Mean, range, SD Creatinine at hospital admission (mg/dL) Urea at hospital admission (mg/dL) Creatinine at ICU admission (mg/dL) Urea at ICU admission (mg/dL) 24-h Urine production (mL) at ICU admission Creatinine at first SLED treatment (mg/dL) Urea at first SLED treatment (mg/dL) Noradrenaline (μg/kg per hour) at SLED start Noradrenaline (μg/kg per hour) after 1-h SLED Noradrenaline (μg/kg per hour) after 3-h SLED Noradrenaline (μg/kg per hour) after 45-L dialysate turnover Noradrenaline (μg/kg per hour) after 9-h SLED Noradrenaline (μg/kg per hour) at SLED end
3
3.2, 1.4-7.1, 1.6 56, 20-159, 31 3.5, 1.4-7.1, 1.6 60, 21-159, 31 490 3.8, 1.3-7.1, 1.2 73, 28-159, 22 8.6, 0-24, 8.2 8.6, 0-24, 8.2 9, 2-13, 4.1 8.4, 0-29, 8.4 8.5, 2-18, 5.4 7.5, 0-24, 8.1
3.1.2. Acid base status and electrolyte balance during SLED treatment At baseline, pH was in the acidotic range with values less than 7.35 in 63% (27/43) of SLED runs (Table 3). During SLED treatment, the pH distribution shifted from the acidotic range toward equalized pH values. After 45-L dialysate turnover and at the end of the dialysis, the reference pH between 7.35 and 7.45 was achieved in 49% (22/43) and 66% (28/43) of running courses, respectively. In accordance with observed pH values, metabolic acidosis with bicarbonate values less than 22 mmol/L was observed in 76% (33/43) of SLED treatments at baseline. After 45-L dialysate turnover and at the end of the dialysis, balanced bicarbonate levels between 22 and 26 mmol/L could be achieved in 42% (18/43) and 52% (22/43) of SLED runs, respectively. However, after 24 hours of SLED treatment, pH was in the acidotic range with values less than 7.35 in 58% (25/43). In these SLED runs with acidotic bicarbonate values less than 22 mmol/L after 45-L dialysate turnover (25 patients) and at the end of the dialysis (21 patients), we observed a mean Catot/Caion ratio of 1.96 and mean citrate concentration of 84.7 mg/L and a mean Catot/Caion ratio of 1.86 and mean citrate conecentration of 51.9 mg/L, respectively. PCO2 might also influence pH and the bicarbonate level but remained stable during SLED treatment. At baseline, base excess (BE) was in the acidotic range (less than −2 mmol/L) in 100% (43/43) of SLED treatments with a mean BE of −6.8mmol/L. After 45-L dialysate turnover, BE was nearly normalized toward values between −2 and 3 mmol/L and −2 and 3 mmol/L with a mean BE of −3.8 mmol/L. After 90 L of dialysate turnover, the mean BE was −3,1 mmol/L. The anion gap was within the reference range in 49% (21/43) of SLED runs at baseline, in 42% (18/43) after 45-L dialysate turnover, and in 40% (17/43) of SLEDs after 90-L dialysate turnover. In accordance with the increase of bicarbonate, there was a trend toward a decrease in the anion gap with values less than 10 mmol/L in 60% (26/43) of treatments after 90-L dialysate turnover compared with 51% (22/43) at baseline. Regarding serum electrolytes, there was a slight trend toward hypocalcemia with Caion values less than 1.13 mmol/L in 48% (21/43) of SLED at baseline compared with 35% (15/43) after 90-L dialysate turnover. However, deficiency of Caion with a minimum Caion of 1 mmol/L was observed in not 1 case. The sodium balance was stable during SLED treatment, with sodium values being within the reference range of 135 to 145 mmol/L in 90% of runs after 90-L dialysate turnover. We observed mild hyperchloremia during SLED treatment.
Levels of creatinine (reference range, 0.7-1.3 mg/dL) and urea (reference range, 7-18 mg/dL) are depicted at hospital admission, at ICU admission, and at the beginning of the first SLED treatment. At ICU admission, all 24 study patients had impaired kidney function, meeting at least AKI stage II criteria. Catecholamine dosages are presented (micrograms per hour) at the start, after 45-L dialysate turnover, and the ending of SLED treatment. At baseline, no catecholamine therapy was necessary in 9 of the 43 SLED runs. Data were expressed as mean (highlighted), range, and SD.
3. Results 3.1. Patient characteristics The mean age of the 24 study patients was 59 years. Four patients were female. At baseline, 20 patients received catecholamine therapy, and all patients were on mechanical ventilation. Six patients had acute liver failure (4 patients due to septic shock and multiorgan failure and 2 patients with histologic proven alcoholic steatohepatitis). Eighteen patients had liver cirrhosis due to alcoholism, and 1 patient combined with hepatitis B. Patients were admitted to the ICU because of acute liver failure (6 patients), hepatorenal syndrome (6 patients), acute bleeding (7 patients), and spontaneous bacterial peritonitis (5 patients, in 1 patient Enterococcus faecium could be detected). Table 1 demonstrates the baseline parameters of liver function before the beginning of SLED treatment.
3.1.1. Acute kidney injury To define AKI, we used the AKI Network classification. After this classification, 2 patients had AKI stage II and 22 patients from AKI stage III at ICU admission (Table 2). Normal renal function was not present in any of the 24 study patients at the time of ICU admission. Reasons for AKI were infection/sepsis (11 patients), hepatorenal syndrome (6 patients), and bleeding shock (7 patients). Table 3 Acid base status and electrolyte balance during SLED treatment
pH P O2 PCO2 Bicarbonate BE Sodium Potassium Cholride Caion Anion gap
Baseline
1h
3h
45 L
9h
90 L
24 h
7.29; 7.12 to 7.43; 0.05 86; 53 to 136; 17.2 45; 19 to 83; 15 19.8; 14.9 to 29.2; 3.2 −6.8; −11.2 to 1.5; 2.3 135; 120 to 150; 6.3 4.2; 3.4 to 5.2; 0.5 107; 95 to 124; 5.6 1.14; 0.9 to 1.49; 0.11 11; 5.2 to 22.1; 3.3
7.31; 7.11 to 7.55; 0.01 90; 60 to 148;19.8 43; 17 to 80;12.5 20.2; 15.3 to 29.6; 2.9 −5.6; −9.8 to 2; 1.8 135; 122 to 149; 5.2 4.1; 3.3 to 5.1; 0.5 107; 94 to 124; 5 1.10; 0.93 to 1.42; 0.1 11; 4.2 to 21.3; 3.5
7.21; 7.25 to 7.42; 0.05 89; 63 to 184; 18.8 44; 23 to 75;11.5 20.8; 14.9 to 30.2; 2.9 −4.7; −8.9 to 1.8; 1.4 135; 125 to 149; 5.1 4.1; 3.4 to 5.1; 0.4 106; 96 to 118; 4.5 1.09; 0.99 to 1.36; 0.07 10; 3.9 to 18.1; 3.2
7.31; 7.22 to 7.53; 0.09 91; 72 to 134; 14.8 44; 23 to 67; 11.2 20.7; 3 to 26; 3.5 −3,9; −6,7 ± 1; 15 136; 126 to 145; 4.1 3.9; 3.4 to 4.9; 0.4 107; 95 to 118; 4.1 1.07; 0.95 to 1.36; 0.07 9; 3.7 to 18.4; 3.2
7.33; 7.2 to 7.53; 0.08 87; 34 to 159; 19.8 43; 25 to 78; 11.1 21.4;15.4 to 24.2; 1.9 −3.6;−7 to 0.5; 1.2 135;126 to 143; 3.9 3.9; 3.2 to 4.7; 0.3 107; 95 to 117; 4 1.07; 0.94 to 1.23; 0.06 9; 4.2 to 18.2; 3.6
7.34; 7.15 to 7.5; 0.07 89; 56 to 160; 22.8 41; 22.6 to 68.4;10.6 22.1;14.1 to 32.6; 3.1 −3.2; −8.2 ± 5; 1.8 135; 126 to 151; 4.7 3.9;3,3 to 4.9; 0.35 107; 96 to 118; 4.5 1.14; 0.96 to 1.35; 0.08 10; 3.7 to 22.3; 3.8
7.34; 7.1 to 7.46; 0.07 80; 61 to 154; 15.3 41; 22.7 to 72.3; 9.9 21.1; 14.4 to 31.2; 3.9 −4,2; −8.2 ± 1.5; 1.9 134; 122 to 149; 4.9 4.1; 3.2 to 5.2; 0.5 107; 94 to 118; 4.3 1.14; 0.97 to 1.3; 0.08 10; 5.2 to 19.7; 3.4
Time course of pH, PO2, and PCO2 (millimeters of mercury); bicarbonate (millimolar); BE (millimolar); sodium (millimolar); potassium (millimolar); chloride (millimolar); and Caion (millimolar) and anion gap (millimolar) at baseline, after 1 hour, 3 hours, 45 L of dialysate turnover, 9 hours, and end of dialysis (90 L of dialysate turnover and 24 hours of SLED treatment time). Data are represented as mean (highlighted), range, and SD.
Please cite this article as: Lahmer T, et al, Sustained low-efficiency dialysis with regional citrate anticoagulation in medical intensive care unit patients with liver failure..., J Crit Care (2015), http://dx.doi.org/10.1016/j.jcrc.2015.06.006
4
T. Lahmer et al. / Journal of Critical Care xxx (2015) xxx–xxx
Table 4 Calcium and citrate levels during SLED treatment Baseline Total protein Albumin Catot Caion Ca postfilter Citrate Lactate
4.9; 3.3-6.4; 0.8 3.1; 1.9-4.6; 0.6 2.10; 1.74-2.78; 0.21 1.14; 0.9-1.49; 0.11
1h
1.10; 0.93-1.42; 0.1 0.43; 0.3-0.53; 0.05
3h
1.09; 0.99-1.36; 0.07 0.43; 0.36-0.48; 0.03
26.6; 5-41; 8.1 2.1; 0.8-9.7; 1.7
45 L
9h
5.1; 3.7-6.6; 0.8 3.1;1.8-4.6; 0.7 2.12; 1.84-2.55; 0.14 1.07; 0.95-1.36; 0.07 0.42; 0.35-0.52; 0.04 86.6; 31-142; 27.1 2; 0.5-9.5; 1.7
1.07; 0.94-1.23; 0.06 0.41; 0.34-0.52; 0.04
90 L
24 h
4.9; 3.5-6.5; 0.8 3.0;1.7-4.4; 0.7 2.14; 1.84-2.49; 0.8 1.14; 0.96-1.35; 0.08
4.8; 3.3-6.2; 0.8 2.9; 1.9-4.2; 0.6 2.10; 1.89-2.54; 0.1 1.14; 0.97-1.3; 0.08
47.1; 7-92; 17.6 2.1; 0.6-10.2; 1.5
27.6; 10-45; 7.3 2.2; 0.7-11; 1.6
Time course of total protein (deciliter); bicarbonate (millimolar); albumin (grams per deciliter); Catot (milligrams per deciliter), Caion (millimolar), and calcium postfilter (millimolar); citrate (12.5-25 mg/L); and lactate (mg/dL) at baseline, after 1 hour, 3 hours, 45 L of dialysate turnover, 9 hours, and end of dialysis (90 L of dialysate turnover and 24 hours of SLED treatment time). Data are represented as mean (highlighted), range, and SD.
3.1.3. Prediction of citrate accumulation in terms of Catot/Caion ratio greater than or equal to 2.5 In only 1 of 172 measurements determined during 43 SLED runs the Catot/Caion ratio exceeded the critical threshold of greater than or equal to 2.5, suggesting citrate accumulation during SLED treatment after 45-L dialysate turnover.
Furthermore, selected metabolic parameters, for example, BE, P = .357 or lactate, P = .461 are not statistical significant as predictors. An elevated citrate level before dialysis treatment is statistical significant (P = .039) as predictor for citrate accumulation.
3.1.4. Calcium and citrate levels during SLED treatment Measurement of citrate in serum demonstrated up to 6-fold elevated serum citrate levels after 45 L of dialysate turnover (86.6 mg/L [mean], 31-142 mg/L [range], 27.1 mg/L [SD]) compared with baseline citrate values (26.6 mg/L [mean], 5-41 mg/L [range], 8.1 mg/L [SD]) (see Table 4 and Fig. 1). After 45-L dialyste turnover, the citrate levels decreased significantly (47.1mg/L [mean], 7-92 mg/L [range], 17.6 mg/L [SD]; P b .05) and turned nearly normal citrate levels after 24 hours (27.6 mg/L [mean], 10-45 mg/L [range], 7.3 [SD]; P b .05). Calcium chloride flow was started with 10 mL/h and adapted according to the required limits of calcium between 0.35 and 0.45 mmol/L postfilter and 1.00 and 1.10 mmol/L in the patient's circulation. As presented in Table 4, the calcium parameters were within the required range. Moreover, changes during SLED treatment were not statistical significant.
Previous studies demonstrated the feasibility of citrate anticoagulation in CVVHD [9,13]. However, its use was usually restricted to patients without severe hepatic impairment. In contrast, there is no study that presents the feasibility of citrate anticoagulation in SLED dialysis in patients with severe liver failure. One reason might be that, in case of severe liver failure, citrate can accumulate, and development of metabolic acidosis and an increased anion gap might therefore be expected [13]. In contrast, in our study, a trend toward balanced pH and bicarbonate levels between 22 and 26 mmol/L in 42% (after 45 L) and 52% (after 90 L) vs 37% at baseline could be obsereved. Moreover, BE after 45 and 90 L was nearly normalized toward values between −2 and 3 mmol/L. The anion gap was within the reference range in 49% (21/43) of SLED runs at baseline, in 42% (18/43) after 45-L dialysate turnover, and in 40% (17/43) of SLED after 90-L dialysate turnover. In accordance with the increase of bicarbonate, there was a trend toward a decrease in the anion gap with values less than 10 mmol/L in 60% (26/43) of treatments after 90-L dialysate turnover compared with 51% (22/43) at baseline. All together, development of metabloic acidosis and after changes in the acid base status could not be observed during citrate anticoagulation in our study. Actually, we observed a prevailed shift from plasma acidification toward balanced acid base status over the SLED treatment time. This might be observed because of the more effective dialysis capacity in SLED than in CVVHD, with a higher dialysate turnover in a shorter period (10-12 vs up to 72 hours) [9]. Moreover, all patients in this study were mechanical ventilated, which influences the acid base status as well. In our study, a maximum of 6-fold increase of citrate in serum was measured. This result is difficult to interpret because, to the best of our knowledge, an upper normal or even toxic level of citrate in serum is not well established [15,16]. In fact, in the case of SLED, most of the citrate is removed by diffusion, with an average citrate reduction ratio that was numerically close to the urea reduction ratio, including patients with liver dysfunction [17,18]. Being a physiologic metabolite, citrate is probably not toxic itself but might induce metabolic disorders (especially hypocalcemia) due to complex binding between citrate and Caion [18-20]. A correlation between citrate in serum and the Catot/Caio ratio in critically ill patients without liver failure during CVVHD has previously been described [20]. This relationship between serum citrate levels and the Catot/Caion ratio in liver failure patients was also used in our study. This ratio with the critical threshold greater than or equal to 2.5 could be a helpful parameter to identify patients at risk for metabolic disturbances (eg, drop of Caion). Moreover, the citrate level itself with a missing cutoff value
3.1.5. Predictive parameters for citrate accumulation We analyzed the predictive capabilities of liver function parameters at baseline regarding as indicator for citrate accumulation. None of the liver function paramters were statistical significant as a predictor for citrate accumulation: bilirubin, P = .427; prothrombin time, P = .086; international normalized ratio, P = .092; ASAT, P = .152; and ALAT, P = .186. Even more, the plasma disappearance rate of indocyanine green showed no significance, P = .216.
Fig. 1. Boxplot presenting the citrate levels (milligrams per liter) in serum during SLED treatment.
4. Discussion
Please cite this article as: Lahmer T, et al, Sustained low-efficiency dialysis with regional citrate anticoagulation in medical intensive care unit patients with liver failure..., J Crit Care (2015), http://dx.doi.org/10.1016/j.jcrc.2015.06.006
T. Lahmer et al. / Journal of Critical Care xxx (2015) xxx–xxx
indicating intoxication during citrate accumulation may not be very helpful. However, only 1 patient reached the critical threshold after 45 L. Citrate accumulation comprises the risk of hypocalcemia due to complex binding with Caion [21-23]. In our study, no severe decrease in Caion was observed. This was probably prevented by regularly monitoring Caion in the patient' s circulation during the SLED treatment time. As documented above, a maximum of 6-fold increase of citrate in serum was measured. The maximum was reached after 45 L of dialysate turnover. After this timepoint, the citrate levels decreased statistically significant (P b .05) and were nearly normalized after 24 hours. We interpret this, that at the beginning of the dialysis, higher citrate levels have to be reached, until a stable anticoagulation is established. Moreover, SLED dialysis is very effective in combination with a highflux dialysate filter, which was used in this setting, compared with low flux filters used in CVVHD. This in combination with hemodiafiltration that might be the reason for the decreasing citrate levels. However, unaffected of citrate levels, the required limits of calcium between 0.35 and 0.45 mmol/L postfilter and 1.00 and 1.10 mmol/L in the patient's circulation during calcium chloride infusion could be reached. Even more changes during SLED treatment were not statistically significant. As documented above, several patients required noradrenaline; however, SLED has only a small effect on catecholmine changes during SLED treatment, which were not statistical significant. One of the aims of this study was to evaluate predictive capabilities of baseline liver function parameters regarding citrate accumulation [24-26]. We identified an elevated citrate level before dialyses as a predictor for citrate accumulation. This might be seen not by the first but maybe by after citrate dialysis. None of the established liver function parameters such as transaminases or bilirubin level showed appropriate predictive capabilities for citrate accumulation. Moreover, the citric acid cycle of the liver is oxygen dependent. The variety of reasons for elevated citrate, lactate levels, or other metabloic disorders can be caused by hypovolemia and hypoxia due to circulatory failure but also by liver failure itself. Another reason might be that all the observed patients were severe critically ill, which is expressed in the high APACHE II score with a multiorgan failure including not only the liver but also the kidneys and the lungs, organs with a great influence on possible metabolic disorders.
5. Conclusion Citrate accumulation in serum could be observed without major disturbances in the acid base status during SLED treatment. Thus, only a moderately correlation between an increase in the Catot/Caion ratio, with a threshold greater than or equal to 2.5 being indicative for citrate accumulation could be detected. However, established liver function parameters, such as transaminases and bilirubin showed poor predictive capabilities regarding prediction of citrate accumulation. To conclude, SLED is feasible and save in patients with impaired liver function. However, continous monitoring of the acid base status and calcium parameters is essential.
5
References [1] Bellomo R, Kellum JA, Ronco C. Acute kidney injury. Lancet 2012;380:756–66. [2] Druml W. Acute renal failure is not a “cute” renal failure! Intensive Care Med 2004; 30:1886–90. [3] Pannu N, Klarenbach S, Wiebe N, Manns B, Tonelli M, Alberta Kidney Disease Network. M: renal replacement therapy in patients with acute renal failure: a systematic review. JAMA 2008;299:793–805. [4] Ricci Z, Ronco C. Timing, dose and mode of dialysis in acute kidney injury. Curr Opin Crit Care 2011;17:556–61. [5] Schefold JC, von Haehling S, Pschowski R, Bender T, Berkmann C, Briegel S, et al. The effect of continuous versus intermittent renal replacement therapy on the outcome of critically ill patients with acute renal failure (CONVINT): a prospective randomized controlled trial. Crit Care 2014;18:R11. [6] Fliser D, Kielstein JT. Technology insight: treatment of renal failure in the intensive care unit with extended dialysis. Nat Clin Pract Nephrol 2006;2:32–9. [7] Schwenger V, Weigand MA, Hoffmann O, Dikow R, Kihm LP, Seckinger J, et al. Sustained low efficiency dialysis using a single-pass batch system in acute kidney injury - a randomized interventional trial: the REnal Replacement Therapy Study in Intensive Care Unit PatiEnts. Crit Care 2012;16:R140. [8] Morgera S. Regional anticoagulation with citrate: expanding its indications. Crit Care Med 2011;39:399–400. [9] Hetzel GR, Schmitz M, Wissing H, Ries W, Schott G, Heering PJ, et al. Regional citrate versus systemic heparin for anticoagulation in critically ill patients on continuous venovenous haemofiltration: a prospective randomized multicentre trial. Nephrol Dial Transplant 2011;26:232–9. [10] Disease K. Improving Global Outcomes (KDIGO) Acute Kidney Injury Work Group: KDIGO Clinical Practice Guideline for Acute Kidney Injury. Kidney Int 2012;2:1–138. [11] Simpson DP. Citrate excretion: a window on renal metabolism. Am J Physiol 1983; 244:F223–34. [12] Bauer E, Derfler K, Joukhadar C, Druml W. Citrate kinetics in patients receiving longterm hemodialysis therapy. Am J Kidney Dis 2005;46:903–7. [13] Schultheiß C, Saugel B, Phillip V, Thies P, Noe S, Mayr U, et al. Continuous venovenous hemodialysis with regional citrate anticoagulation in patients with liver failure: a prospective observational study. Crit Care 2012;16(4):R162. [14] Morath C, Miftari N, Dikow R, Hainer C, Zeier M, Morgera S, et al. Sodium citrate anticoagulation during sustained low efficiency dialysis (SLED) in patients with acute renal failure and severely impaired liver function. Nephrol Dial Transplant 2008;23(1):421–2 [Epub 2007 Oct 3]. [15] Diaz J, Acosta F, Parrilla P, Sansano T, Bento M, Cura S, et al. Citrate intoxication and blood concentration of ionized calcium in liver transplantation. Transplant Proc 1994;26:3669–70. [16] Meier-Kriesche HU, Finkel KW, DuBose Jr TD. Unexpected severe hypocalcemia during continuous venovenous hemodialysis with regional citrate anticoagulation. Am J Kidney Dis 1999;33:e8. [17] Meier-Kriesche HU, Gitomer J, Finkel K, DuBose T. Increased total to ionized calcium ratio during continuous venovenous hemodialysis with regional citrate anticoagulation. Crit Care Med 2001;29:748–52. [18] Berbece AN, Richardson RM. Sustained low-efficiency dialysis in the ICU: cost, anticoagulation, and solute removal. Kidney Int 2006;70:963–8. [19] Zender R, de Torrenté C, Schneider U. Enzymatic determination of citrate in plasma without deproteinization. Clin Chim Acta 1969;24:335–40. [20] Kramer L, Bauer E, Joukhadar C, Strobl W, Gendo A, Madl C. Citrate pharmacokinetics and metabolism in cirrhotic and noncirrhotic critically ill patients. Crit Care Med 2003;31:2450–5. [21] Hetzel GR, Taskaya G, Sucker C, Hennersdorf M, Grabensee B, Schmitz M. Citrate plasma levels in patients under regional anticoagulation in continuous venovenous hemofiltration. Am J Kidney Dis 2006;48:806–11. [22] Díaz J, Acosta F, Parrilla P, Sansano T, Contreras RF, Bueno FS, et al. Correlation among ionized calcium, citrate, and total calcium levels during hepatic tranplantation. Clin Biochem 1995;28:315–7. [23] Apsner R, Schwarzenhofer M, Derfler K, Zauner C, Ratheiser K, Kranz A. Impairment of citrate metabolism in acute hepatic failure. Wien Klin Wochenschr 1997;109:123–7. [24] Cooper GS, Bellamy P, Dawson NV, Desbiens N, Fulkerson Jr WJ, Goldman L, et al. A prognostic model for patients with end-stage liver disease. Gastroenterology 1997; 113:1278–88. [25] Pugh RN. Pugh's grading in the classification of liver decompensation. Gut 1992;33: 1583. [26] Zipprich A, Kuss O, Rogowski S, Kleber G, Lotterer E, Seufferlein T, et al. Incorporating indocyanin green clearance into the Model for End Stage Liver Disease (MELD-ICG) improves prognostic accuracy intermediate to advanced cirrhosis. Gut 2010;59:963–8.
Please cite this article as: Lahmer T, et al, Sustained low-efficiency dialysis with regional citrate anticoagulation in medical intensive care unit patients with liver failure..., J Crit Care (2015), http://dx.doi.org/10.1016/j.jcrc.2015.06.006
Contents lists available at ScienceDirect
Journal of Critical Care journal homepage: www.jccjournal.org
Sustained low-efficiency dialysis with regional citrate anticoagulation in medical intensive care unit patients with liver failure: a prospective study☆,☆☆ Tobias Lahmer, MD ⁎, Marlena Messer, MD, Sebastian Rasch, MD, Analena Beitz, MD, Christopher Schnappauf, MD, Roland M. Schmid, MD, Wolfgang Huber, MD II. Medizinische Klinik und Poliklinik, Klinikum rechts der Isar der Technischen Universität München, Munich, Germany
a r t i c l e
i n f o
Keywords: SLED Citrate anticoagulation Liver failure Critical care
s u m m a r y Purpose: Patients with liver failure requiring dialysis are at increased risk for citrate accumulation during sustained low-efficiency dialysis (SLED). The aim of this study was to evaluate the feasibilty of citrate SLED in critical ill patients with liver failure and investigate predictive parameters regarding citrate accumulation. Materials and methods: This is a prospective study in 24 medical intensive care unit patients with liver failure and a total of 43 SLED runs (maximum of 3 runs per patient) using citrate anticoagulation. Liver function was characterized before SLED using not only laboratory parameters but also determination of the plasma disappearance rate of indocyanine green. In addition, blood gas parameters as well total calcium and citrate in serum were measured at baseline and defined time points during SLED. Results: Accumulation of citrate could be observed in all SLED runs, which were nearly normalized until the end of SLED and 24 hours after SLED, respectively. However, the critical threshold of total calcium/ionized calcium on ratio of greater than 2.5 was exceeded in only 1 patient. Equalization of initial metabolic acidosis was possible without major disturbances of acid base and electrolyte status. Liver function parameters showed poor predicitve capabilities regarding citrate accumulation. Conclusions: Despite substantial accumulation of citrate in serum, SLED is save and feasible in patients with liver failure using a citrate anticoagulation. Careful monitoring of electrolytes and acid base status is mandatory to ensure patient safety. © 2015 Elsevier Inc. All rights reserved.
1. Introduction One of the most often complications in hospitalized patients is still an acute kidney injury (AKI), which significantly affects morbidity and mortality of intensive care unit (ICU) patients if renal replacement therapy (RRT) is required [1,2]. In most cases, continuous RRT is used in ICUs, which is often seen as the preferable RRT in critically ill AKI patients [3-5]. ☆ Competing interests: The author(s) declare that they have no competing interests. Declaration of interest: None of the authors have any potential financial conflict of interest related to this manuscript. None of the authors have got any funding or financial support regarding this manuscript. The manuscript has not been submitted or accepted elsewhere. All authors fulfill the criteria given in the authorship paragraph. No writing assistance other than copy editing was provided in the preparation of the manuscript. ☆☆ Authors' contribution: TL, MM, SR, AB, CS, RMS, and WH. Study design: TL, MM, SR, CS, RMS, and WH. Collected data: TL, MM, SR, AB, CS, RMS, and WH. Analyzed data: TL, MM, AB, SR, and WH. Wrote manuscript: TL, MM, AB, SR, CS, RMS, and WH. Study is approved by the Institutional Review Board of the Technical University of Munich, Munich, Germany. Written informed consent was obtained. ⁎ Corresponding author at: II. Medizinische Klinik und Poliklinik, Klinikum rechts der Isar der Technischen Universität München, Ismaninger Str. 22, 81675 Munich, Germany. Tel.: +49 89 4140 2267; fax: +49 89 4140 6243. E-mail address: [email protected] (T. Lahmer).
However, intermittent RRTs, so-called sustained low-efficiency dialysis (SLED), are more and more used in critically ill patients with AKI [6]. Sustained low-efficiency dialysis, with a running time of 8 to 12 hours, shares the advantages of a conventional intermittent (4 hours) and a continuous RRT (up to 72 hours) [7]. Moreover, over the last years, citrate has emerged as a safe and efficacious alternative to heparin for extracorporeal circuit anticoagulation [8]. Citrate chelates ionized calcium (Caion), the most important cofactor of the coagulation cascade. Thus, regional anticoagulation with citrate is a very effective anticoagulation method for hemodialysis. Some studies on continuous venovenous hemodialysis (CVVHD) detected a reduction of bleeding complications, a longer filter lifetime, and a possible reduction of mortality in ICU patients with a citrate anticoagulation [9]. Therefore, regional citrate anticoagulation has also been recommended by the recent Kidney Disease Improving Global Outcomes guidelines. However, some intensivists are still reluctant to adopt this technique. Reasons may be the complexity; need of customized citrate solutions/replacement fluids; fear of metabolic complications (eg, hypocalcemia and metabolic alkalosis); and difficulties in predicting and preventing citrate accumulation, especially in patients with impaired liver function [10]. Around 50% of citrate is removed as a complex bound with Caion through the dialyzer by diffusion/convection. However, citrate can partly
http://dx.doi.org/10.1016/j.jcrc.2015.06.006 0883-9441/© 2015 Elsevier Inc. All rights reserved.
Please cite this article as: Lahmer T, et al, Sustained low-efficiency dialysis with regional citrate anticoagulation in medical intensive care unit patients with liver failure..., J Crit Care (2015), http://dx.doi.org/10.1016/j.jcrc.2015.06.006
2
T. Lahmer et al. / Journal of Critical Care xxx (2015) xxx–xxx
enter the systemic circulation [11]. The hepatic citric acid cycle is the major site of citrate metabolization, which leads to the release of Caion into the systemic circulation. This is almost independent of renal function and urinary excretion [11,12]. However, in cases of impaired liver function with an also impaired citrate metabolism, the risk of citrate accumulation is rising. Drop of Caion due to complex binding between citrate and Caion requiring more calcium chloride substitution at the venous line of the extracorporeal circuit can result [12]. Moreover, this leads to an increase of total calcium (Catot) concentration, which is the sum of Caion, protein, and citrate-bound calcium, and an increased Catot/Caion ratio might be observed. After former studies, a serum Catot/Caion ratio greater than or equal to 2.5 might be a critical threshold of potential citrate accumulation [11,12]. Other side effects could be metabolic acidosis combined with an enlarged anion gap caused by reduced citric acid cycle production of bicarbonate out of citrate and accumulation of negative loaded citrate ions [8,9]. These possible side effects may be the reason why data on the feasibility of citrate CVVHD in liver failure patients are scarce. Moreover, there are only small data in critically ill patients with liver impairment using SLED [13,14]. Therefore, the aim of our study was to determine predictors for citrate accumulation and to investigate the feasibility of citrate anticoagulation in patients with impaired liver function.
2. Materials and methods 2.1. Patients Twenty-four ICU patients between aged 29 and 73 years with decompensated liver cirrhosis (18 patients) or acute liver failure (6 patients) who needed RRT were included in this study. A total of 43 SLED runs (maximum, 3 per patient) were analyzed in this study cohort. Liver cirrhosis was diagnosed either by histologic specimen, by ultrasound and/or computed tomography, and by clinical criteria for instance ascites or esophageal varices. We defined an acute liver failure as an abrupt loss of liver function without preexisting liver disease. Before SLED treatment, we used the model of end-stage liver disease score (MELD score), the Child-Pugh score in case of cirrhosis, and laboratory liver function parameters (aspartate aminotransferase [ASAT], alanine aminotransferase [ALAT], bilirubin, and prothrombin time) to characterize baseline liver function. In addition, plasma disappearance rate of indocyanine green was performed. Sequential Organ Failure Assessment (SOFA) score as well the Acute Physiology and Chronic Health Evaluation (APACHE) II score was calculated in accordance to demonstrate the severity of the underlying disease (Table 1). Patients with
Table 1 Overview of baseline liver function parameters
MELD score (points) Child-Pugh score APACHE II score SOFA score ICG-PDR Prothrombin time (%) INR Bilirubin (mg/dL) ASAT (U/L) ALAT (U/L) Lactate (mmol/L)
Mean/range/SD
Reference range
35/25-40/6.5 all C 33/24-40/3.5 16/8-19/2.5 5.1/1.5-24/4.3 38/16-66/13 2.2/1.3-4.3/0.9 18.4/2.1-46.5/14.5 331/25-1261/331 157/15-1047/157 2.1/0.8-9.7/1.7
Maximum, 40 points Maximum, 71 points Maximum, 24 points 18-25 70-120 b1.2 10-50 10-35 b2.4
ICG-PDR indicates indocyanine green-plasma disappearance rate; INR, international normalized ratio. To characterize baseline liver function, the MELD score and the plasma disappearance rate of indocyanine green (ICG-PDR) were calculated in each patient, and the Child-Pugh score only in patients with cirrhosis. The APACHE II score and the SOFA score as well as the laboratory parameters were determined immediately before the start of SLED treatment. Data were expressed as mean (highlighted), range, and SD.
severe alkalosis (pH N7.55) or acidosis (pH b 7.1) and deficiency of Caion (b0.9 mmol/L) were excluded from our study. In accordance to our Institutional Review Board of the Technical University of Munich, Germany, this study was approved, and written informed consent was obtained.
2.1.1. Sustained low-efficiency dialysis treatment We used the commercially available hemodialysis Genius singlepass batch dialysis system (Fresenius Medical Care, Bad Homburg, Germany) for SLED treatment. It provides 90 L of bicarbonate dialysate per dialysis session. The dialysis system uses a double-sided roller pump that generates equal blood, and dialysate flows up to a maximum of 350 mL/min. Moreover, the Genius system contents a closed dialysate tank of 90 L, in which fresh and spent dialysate are stored together without any mixing of both compounds. The ultrafiltered water out of the patient plasma is collected in an ultrafiltrate recipient. High-flux FX60 dialyzers (Fresenius Medical Care) were used in all sessions. For all SLED treatments, a blood flow of 150 mL/min, which is equal to dialysate flow, was used. The dialysate solution is a compound of an instant (HC-90; Fresenius Medical Care) and liquid (DS-90; Fresenius Medical Care) component. The instant component contains sodium chloride,sodium bicarbonate, and glucose. The liquid component contains potassium chloride, calcium chloride, and magnesium chloride. In all patients, the HC-90 concentration HC 31-90, which contains 40 mmol/L sodium, 3 mmol/L potassium, 1 mmol/L calcium, 0.5 mmol/L magnesium, 48 mmol/L chloride, 0.067 mmol/L citrate, and 2 mmol/L HCL, and the DS-90 concentration DS 135/35-90, which contains 95 mmol/L sodium, 35 mmol/L bicarbonate, 60 mmol/L chloride, and 5.5 mmol/L glucose, were used. Citrate solution was produced in the local hospital pharmacy (100 mL consists of 22 g tri natriumcitrate dihydrate and 8 g citrate monohydrate in aqua destillata). Calcium chloride solution was produced in the local hospital pharmacy (500 mmol/L, 73.5 g calcium chloride dihydrate in 1000 mL aqua destillata). Sodium citrate flow was started with 60 mL/h, and calcium chloride flow was started with 10 mL/h, respectively, and adapted according to the required limits of calcium between 0.35 and 0.45 mmol/L postfilter and 1.00 and 1.10 mmol/L in the patient's circulation. According to the study protocol, citrate and Catot in serum were measured at baseline just before the beginning of SLED treatment, after 45 L, 90-L dialysate turnover, and 24 hours. Blood gas analyses of the patient's circulation and Caion postfilter were measured at baseline, after 1 hour, after 3 hours, 45 L, 9 hours, 90 L, and after 24 hours. The total run time of SLED was approximatley 10.5 hours. Citrate levels were measured enzymatically by the citrate-lyase method (MVZ Labor Limbach, Heidelberg, Germany). In this method, citrate is metabolized to oxalacetate, and acetate is catalyzed by the enzyme citrate lyase. Oxalacetate is reduced to malate and lactate by the enzymes L-malate-dehydrogenase and L-lactate-dehydrogenase in a nicotinamide adenine nucleotide hydrogen-dependent manner. Nicotinamide adenine nucleotide hydrogen is the measured variable and is equivalent to the amount of citrate.
2.2. Statistical analysis In our study, we used for statistical analyses the IBM SPSS Statistics 21 (SPSS Inc, Chicago, IL). Descriptive statistics were expressed with mean ± SD and range for normally distributed continuous data. The t test was performed for paired samples. Wilcoxon signed rank test was performed for paired samples for normally distributed data and not normally distributed data, respectively. A P value below a significance level of 5% (P b .05) indicates statistical significance.
Please cite this article as: Lahmer T, et al, Sustained low-efficiency dialysis with regional citrate anticoagulation in medical intensive care unit patients with liver failure..., J Crit Care (2015), http://dx.doi.org/10.1016/j.jcrc.2015.06.006
T. Lahmer et al. / Journal of Critical Care xxx (2015) xxx–xxx
Basic kidney parameters and noradrenaline dosages during SLED were presented in Table 2. Noradrenaline changes during SLED treatment were not statistical significant. The mean length of ICU stay was 24 days (minimum, 6 days; maximum, 59 days). Nineteen patients died in the ICU. Amonge the 5 survivors, kidney function recoverd in only 1 patient; all other needed further hemodialysis treatment.
Table 2 Overview of basic kidney and SLED parameters Mean, range, SD Creatinine at hospital admission (mg/dL) Urea at hospital admission (mg/dL) Creatinine at ICU admission (mg/dL) Urea at ICU admission (mg/dL) 24-h Urine production (mL) at ICU admission Creatinine at first SLED treatment (mg/dL) Urea at first SLED treatment (mg/dL) Noradrenaline (μg/kg per hour) at SLED start Noradrenaline (μg/kg per hour) after 1-h SLED Noradrenaline (μg/kg per hour) after 3-h SLED Noradrenaline (μg/kg per hour) after 45-L dialysate turnover Noradrenaline (μg/kg per hour) after 9-h SLED Noradrenaline (μg/kg per hour) at SLED end
3
3.2, 1.4-7.1, 1.6 56, 20-159, 31 3.5, 1.4-7.1, 1.6 60, 21-159, 31 490 3.8, 1.3-7.1, 1.2 73, 28-159, 22 8.6, 0-24, 8.2 8.6, 0-24, 8.2 9, 2-13, 4.1 8.4, 0-29, 8.4 8.5, 2-18, 5.4 7.5, 0-24, 8.1
3.1.2. Acid base status and electrolyte balance during SLED treatment At baseline, pH was in the acidotic range with values less than 7.35 in 63% (27/43) of SLED runs (Table 3). During SLED treatment, the pH distribution shifted from the acidotic range toward equalized pH values. After 45-L dialysate turnover and at the end of the dialysis, the reference pH between 7.35 and 7.45 was achieved in 49% (22/43) and 66% (28/43) of running courses, respectively. In accordance with observed pH values, metabolic acidosis with bicarbonate values less than 22 mmol/L was observed in 76% (33/43) of SLED treatments at baseline. After 45-L dialysate turnover and at the end of the dialysis, balanced bicarbonate levels between 22 and 26 mmol/L could be achieved in 42% (18/43) and 52% (22/43) of SLED runs, respectively. However, after 24 hours of SLED treatment, pH was in the acidotic range with values less than 7.35 in 58% (25/43). In these SLED runs with acidotic bicarbonate values less than 22 mmol/L after 45-L dialysate turnover (25 patients) and at the end of the dialysis (21 patients), we observed a mean Catot/Caion ratio of 1.96 and mean citrate concentration of 84.7 mg/L and a mean Catot/Caion ratio of 1.86 and mean citrate conecentration of 51.9 mg/L, respectively. PCO2 might also influence pH and the bicarbonate level but remained stable during SLED treatment. At baseline, base excess (BE) was in the acidotic range (less than −2 mmol/L) in 100% (43/43) of SLED treatments with a mean BE of −6.8mmol/L. After 45-L dialysate turnover, BE was nearly normalized toward values between −2 and 3 mmol/L and −2 and 3 mmol/L with a mean BE of −3.8 mmol/L. After 90 L of dialysate turnover, the mean BE was −3,1 mmol/L. The anion gap was within the reference range in 49% (21/43) of SLED runs at baseline, in 42% (18/43) after 45-L dialysate turnover, and in 40% (17/43) of SLEDs after 90-L dialysate turnover. In accordance with the increase of bicarbonate, there was a trend toward a decrease in the anion gap with values less than 10 mmol/L in 60% (26/43) of treatments after 90-L dialysate turnover compared with 51% (22/43) at baseline. Regarding serum electrolytes, there was a slight trend toward hypocalcemia with Caion values less than 1.13 mmol/L in 48% (21/43) of SLED at baseline compared with 35% (15/43) after 90-L dialysate turnover. However, deficiency of Caion with a minimum Caion of 1 mmol/L was observed in not 1 case. The sodium balance was stable during SLED treatment, with sodium values being within the reference range of 135 to 145 mmol/L in 90% of runs after 90-L dialysate turnover. We observed mild hyperchloremia during SLED treatment.
Levels of creatinine (reference range, 0.7-1.3 mg/dL) and urea (reference range, 7-18 mg/dL) are depicted at hospital admission, at ICU admission, and at the beginning of the first SLED treatment. At ICU admission, all 24 study patients had impaired kidney function, meeting at least AKI stage II criteria. Catecholamine dosages are presented (micrograms per hour) at the start, after 45-L dialysate turnover, and the ending of SLED treatment. At baseline, no catecholamine therapy was necessary in 9 of the 43 SLED runs. Data were expressed as mean (highlighted), range, and SD.
3. Results 3.1. Patient characteristics The mean age of the 24 study patients was 59 years. Four patients were female. At baseline, 20 patients received catecholamine therapy, and all patients were on mechanical ventilation. Six patients had acute liver failure (4 patients due to septic shock and multiorgan failure and 2 patients with histologic proven alcoholic steatohepatitis). Eighteen patients had liver cirrhosis due to alcoholism, and 1 patient combined with hepatitis B. Patients were admitted to the ICU because of acute liver failure (6 patients), hepatorenal syndrome (6 patients), acute bleeding (7 patients), and spontaneous bacterial peritonitis (5 patients, in 1 patient Enterococcus faecium could be detected). Table 1 demonstrates the baseline parameters of liver function before the beginning of SLED treatment.
3.1.1. Acute kidney injury To define AKI, we used the AKI Network classification. After this classification, 2 patients had AKI stage II and 22 patients from AKI stage III at ICU admission (Table 2). Normal renal function was not present in any of the 24 study patients at the time of ICU admission. Reasons for AKI were infection/sepsis (11 patients), hepatorenal syndrome (6 patients), and bleeding shock (7 patients). Table 3 Acid base status and electrolyte balance during SLED treatment
pH P O2 PCO2 Bicarbonate BE Sodium Potassium Cholride Caion Anion gap
Baseline
1h
3h
45 L
9h
90 L
24 h
7.29; 7.12 to 7.43; 0.05 86; 53 to 136; 17.2 45; 19 to 83; 15 19.8; 14.9 to 29.2; 3.2 −6.8; −11.2 to 1.5; 2.3 135; 120 to 150; 6.3 4.2; 3.4 to 5.2; 0.5 107; 95 to 124; 5.6 1.14; 0.9 to 1.49; 0.11 11; 5.2 to 22.1; 3.3
7.31; 7.11 to 7.55; 0.01 90; 60 to 148;19.8 43; 17 to 80;12.5 20.2; 15.3 to 29.6; 2.9 −5.6; −9.8 to 2; 1.8 135; 122 to 149; 5.2 4.1; 3.3 to 5.1; 0.5 107; 94 to 124; 5 1.10; 0.93 to 1.42; 0.1 11; 4.2 to 21.3; 3.5
7.21; 7.25 to 7.42; 0.05 89; 63 to 184; 18.8 44; 23 to 75;11.5 20.8; 14.9 to 30.2; 2.9 −4.7; −8.9 to 1.8; 1.4 135; 125 to 149; 5.1 4.1; 3.4 to 5.1; 0.4 106; 96 to 118; 4.5 1.09; 0.99 to 1.36; 0.07 10; 3.9 to 18.1; 3.2
7.31; 7.22 to 7.53; 0.09 91; 72 to 134; 14.8 44; 23 to 67; 11.2 20.7; 3 to 26; 3.5 −3,9; −6,7 ± 1; 15 136; 126 to 145; 4.1 3.9; 3.4 to 4.9; 0.4 107; 95 to 118; 4.1 1.07; 0.95 to 1.36; 0.07 9; 3.7 to 18.4; 3.2
7.33; 7.2 to 7.53; 0.08 87; 34 to 159; 19.8 43; 25 to 78; 11.1 21.4;15.4 to 24.2; 1.9 −3.6;−7 to 0.5; 1.2 135;126 to 143; 3.9 3.9; 3.2 to 4.7; 0.3 107; 95 to 117; 4 1.07; 0.94 to 1.23; 0.06 9; 4.2 to 18.2; 3.6
7.34; 7.15 to 7.5; 0.07 89; 56 to 160; 22.8 41; 22.6 to 68.4;10.6 22.1;14.1 to 32.6; 3.1 −3.2; −8.2 ± 5; 1.8 135; 126 to 151; 4.7 3.9;3,3 to 4.9; 0.35 107; 96 to 118; 4.5 1.14; 0.96 to 1.35; 0.08 10; 3.7 to 22.3; 3.8
7.34; 7.1 to 7.46; 0.07 80; 61 to 154; 15.3 41; 22.7 to 72.3; 9.9 21.1; 14.4 to 31.2; 3.9 −4,2; −8.2 ± 1.5; 1.9 134; 122 to 149; 4.9 4.1; 3.2 to 5.2; 0.5 107; 94 to 118; 4.3 1.14; 0.97 to 1.3; 0.08 10; 5.2 to 19.7; 3.4
Time course of pH, PO2, and PCO2 (millimeters of mercury); bicarbonate (millimolar); BE (millimolar); sodium (millimolar); potassium (millimolar); chloride (millimolar); and Caion (millimolar) and anion gap (millimolar) at baseline, after 1 hour, 3 hours, 45 L of dialysate turnover, 9 hours, and end of dialysis (90 L of dialysate turnover and 24 hours of SLED treatment time). Data are represented as mean (highlighted), range, and SD.
Please cite this article as: Lahmer T, et al, Sustained low-efficiency dialysis with regional citrate anticoagulation in medical intensive care unit patients with liver failure..., J Crit Care (2015), http://dx.doi.org/10.1016/j.jcrc.2015.06.006
4
T. Lahmer et al. / Journal of Critical Care xxx (2015) xxx–xxx
Table 4 Calcium and citrate levels during SLED treatment Baseline Total protein Albumin Catot Caion Ca postfilter Citrate Lactate
4.9; 3.3-6.4; 0.8 3.1; 1.9-4.6; 0.6 2.10; 1.74-2.78; 0.21 1.14; 0.9-1.49; 0.11
1h
1.10; 0.93-1.42; 0.1 0.43; 0.3-0.53; 0.05
3h
1.09; 0.99-1.36; 0.07 0.43; 0.36-0.48; 0.03
26.6; 5-41; 8.1 2.1; 0.8-9.7; 1.7
45 L
9h
5.1; 3.7-6.6; 0.8 3.1;1.8-4.6; 0.7 2.12; 1.84-2.55; 0.14 1.07; 0.95-1.36; 0.07 0.42; 0.35-0.52; 0.04 86.6; 31-142; 27.1 2; 0.5-9.5; 1.7
1.07; 0.94-1.23; 0.06 0.41; 0.34-0.52; 0.04
90 L
24 h
4.9; 3.5-6.5; 0.8 3.0;1.7-4.4; 0.7 2.14; 1.84-2.49; 0.8 1.14; 0.96-1.35; 0.08
4.8; 3.3-6.2; 0.8 2.9; 1.9-4.2; 0.6 2.10; 1.89-2.54; 0.1 1.14; 0.97-1.3; 0.08
47.1; 7-92; 17.6 2.1; 0.6-10.2; 1.5
27.6; 10-45; 7.3 2.2; 0.7-11; 1.6
Time course of total protein (deciliter); bicarbonate (millimolar); albumin (grams per deciliter); Catot (milligrams per deciliter), Caion (millimolar), and calcium postfilter (millimolar); citrate (12.5-25 mg/L); and lactate (mg/dL) at baseline, after 1 hour, 3 hours, 45 L of dialysate turnover, 9 hours, and end of dialysis (90 L of dialysate turnover and 24 hours of SLED treatment time). Data are represented as mean (highlighted), range, and SD.
3.1.3. Prediction of citrate accumulation in terms of Catot/Caion ratio greater than or equal to 2.5 In only 1 of 172 measurements determined during 43 SLED runs the Catot/Caion ratio exceeded the critical threshold of greater than or equal to 2.5, suggesting citrate accumulation during SLED treatment after 45-L dialysate turnover.
Furthermore, selected metabolic parameters, for example, BE, P = .357 or lactate, P = .461 are not statistical significant as predictors. An elevated citrate level before dialysis treatment is statistical significant (P = .039) as predictor for citrate accumulation.
3.1.4. Calcium and citrate levels during SLED treatment Measurement of citrate in serum demonstrated up to 6-fold elevated serum citrate levels after 45 L of dialysate turnover (86.6 mg/L [mean], 31-142 mg/L [range], 27.1 mg/L [SD]) compared with baseline citrate values (26.6 mg/L [mean], 5-41 mg/L [range], 8.1 mg/L [SD]) (see Table 4 and Fig. 1). After 45-L dialyste turnover, the citrate levels decreased significantly (47.1mg/L [mean], 7-92 mg/L [range], 17.6 mg/L [SD]; P b .05) and turned nearly normal citrate levels after 24 hours (27.6 mg/L [mean], 10-45 mg/L [range], 7.3 [SD]; P b .05). Calcium chloride flow was started with 10 mL/h and adapted according to the required limits of calcium between 0.35 and 0.45 mmol/L postfilter and 1.00 and 1.10 mmol/L in the patient's circulation. As presented in Table 4, the calcium parameters were within the required range. Moreover, changes during SLED treatment were not statistical significant.
Previous studies demonstrated the feasibility of citrate anticoagulation in CVVHD [9,13]. However, its use was usually restricted to patients without severe hepatic impairment. In contrast, there is no study that presents the feasibility of citrate anticoagulation in SLED dialysis in patients with severe liver failure. One reason might be that, in case of severe liver failure, citrate can accumulate, and development of metabolic acidosis and an increased anion gap might therefore be expected [13]. In contrast, in our study, a trend toward balanced pH and bicarbonate levels between 22 and 26 mmol/L in 42% (after 45 L) and 52% (after 90 L) vs 37% at baseline could be obsereved. Moreover, BE after 45 and 90 L was nearly normalized toward values between −2 and 3 mmol/L. The anion gap was within the reference range in 49% (21/43) of SLED runs at baseline, in 42% (18/43) after 45-L dialysate turnover, and in 40% (17/43) of SLED after 90-L dialysate turnover. In accordance with the increase of bicarbonate, there was a trend toward a decrease in the anion gap with values less than 10 mmol/L in 60% (26/43) of treatments after 90-L dialysate turnover compared with 51% (22/43) at baseline. All together, development of metabloic acidosis and after changes in the acid base status could not be observed during citrate anticoagulation in our study. Actually, we observed a prevailed shift from plasma acidification toward balanced acid base status over the SLED treatment time. This might be observed because of the more effective dialysis capacity in SLED than in CVVHD, with a higher dialysate turnover in a shorter period (10-12 vs up to 72 hours) [9]. Moreover, all patients in this study were mechanical ventilated, which influences the acid base status as well. In our study, a maximum of 6-fold increase of citrate in serum was measured. This result is difficult to interpret because, to the best of our knowledge, an upper normal or even toxic level of citrate in serum is not well established [15,16]. In fact, in the case of SLED, most of the citrate is removed by diffusion, with an average citrate reduction ratio that was numerically close to the urea reduction ratio, including patients with liver dysfunction [17,18]. Being a physiologic metabolite, citrate is probably not toxic itself but might induce metabolic disorders (especially hypocalcemia) due to complex binding between citrate and Caion [18-20]. A correlation between citrate in serum and the Catot/Caio ratio in critically ill patients without liver failure during CVVHD has previously been described [20]. This relationship between serum citrate levels and the Catot/Caion ratio in liver failure patients was also used in our study. This ratio with the critical threshold greater than or equal to 2.5 could be a helpful parameter to identify patients at risk for metabolic disturbances (eg, drop of Caion). Moreover, the citrate level itself with a missing cutoff value
3.1.5. Predictive parameters for citrate accumulation We analyzed the predictive capabilities of liver function parameters at baseline regarding as indicator for citrate accumulation. None of the liver function paramters were statistical significant as a predictor for citrate accumulation: bilirubin, P = .427; prothrombin time, P = .086; international normalized ratio, P = .092; ASAT, P = .152; and ALAT, P = .186. Even more, the plasma disappearance rate of indocyanine green showed no significance, P = .216.
Fig. 1. Boxplot presenting the citrate levels (milligrams per liter) in serum during SLED treatment.
4. Discussion
Please cite this article as: Lahmer T, et al, Sustained low-efficiency dialysis with regional citrate anticoagulation in medical intensive care unit patients with liver failure..., J Crit Care (2015), http://dx.doi.org/10.1016/j.jcrc.2015.06.006
T. Lahmer et al. / Journal of Critical Care xxx (2015) xxx–xxx
indicating intoxication during citrate accumulation may not be very helpful. However, only 1 patient reached the critical threshold after 45 L. Citrate accumulation comprises the risk of hypocalcemia due to complex binding with Caion [21-23]. In our study, no severe decrease in Caion was observed. This was probably prevented by regularly monitoring Caion in the patient' s circulation during the SLED treatment time. As documented above, a maximum of 6-fold increase of citrate in serum was measured. The maximum was reached after 45 L of dialysate turnover. After this timepoint, the citrate levels decreased statistically significant (P b .05) and were nearly normalized after 24 hours. We interpret this, that at the beginning of the dialysis, higher citrate levels have to be reached, until a stable anticoagulation is established. Moreover, SLED dialysis is very effective in combination with a highflux dialysate filter, which was used in this setting, compared with low flux filters used in CVVHD. This in combination with hemodiafiltration that might be the reason for the decreasing citrate levels. However, unaffected of citrate levels, the required limits of calcium between 0.35 and 0.45 mmol/L postfilter and 1.00 and 1.10 mmol/L in the patient's circulation during calcium chloride infusion could be reached. Even more changes during SLED treatment were not statistically significant. As documented above, several patients required noradrenaline; however, SLED has only a small effect on catecholmine changes during SLED treatment, which were not statistical significant. One of the aims of this study was to evaluate predictive capabilities of baseline liver function parameters regarding citrate accumulation [24-26]. We identified an elevated citrate level before dialyses as a predictor for citrate accumulation. This might be seen not by the first but maybe by after citrate dialysis. None of the established liver function parameters such as transaminases or bilirubin level showed appropriate predictive capabilities for citrate accumulation. Moreover, the citric acid cycle of the liver is oxygen dependent. The variety of reasons for elevated citrate, lactate levels, or other metabloic disorders can be caused by hypovolemia and hypoxia due to circulatory failure but also by liver failure itself. Another reason might be that all the observed patients were severe critically ill, which is expressed in the high APACHE II score with a multiorgan failure including not only the liver but also the kidneys and the lungs, organs with a great influence on possible metabolic disorders.
5. Conclusion Citrate accumulation in serum could be observed without major disturbances in the acid base status during SLED treatment. Thus, only a moderately correlation between an increase in the Catot/Caion ratio, with a threshold greater than or equal to 2.5 being indicative for citrate accumulation could be detected. However, established liver function parameters, such as transaminases and bilirubin showed poor predictive capabilities regarding prediction of citrate accumulation. To conclude, SLED is feasible and save in patients with impaired liver function. However, continous monitoring of the acid base status and calcium parameters is essential.
5
References [1] Bellomo R, Kellum JA, Ronco C. Acute kidney injury. Lancet 2012;380:756–66. [2] Druml W. Acute renal failure is not a “cute” renal failure! Intensive Care Med 2004; 30:1886–90. [3] Pannu N, Klarenbach S, Wiebe N, Manns B, Tonelli M, Alberta Kidney Disease Network. M: renal replacement therapy in patients with acute renal failure: a systematic review. JAMA 2008;299:793–805. [4] Ricci Z, Ronco C. Timing, dose and mode of dialysis in acute kidney injury. Curr Opin Crit Care 2011;17:556–61. [5] Schefold JC, von Haehling S, Pschowski R, Bender T, Berkmann C, Briegel S, et al. The effect of continuous versus intermittent renal replacement therapy on the outcome of critically ill patients with acute renal failure (CONVINT): a prospective randomized controlled trial. Crit Care 2014;18:R11. [6] Fliser D, Kielstein JT. Technology insight: treatment of renal failure in the intensive care unit with extended dialysis. Nat Clin Pract Nephrol 2006;2:32–9. [7] Schwenger V, Weigand MA, Hoffmann O, Dikow R, Kihm LP, Seckinger J, et al. Sustained low efficiency dialysis using a single-pass batch system in acute kidney injury - a randomized interventional trial: the REnal Replacement Therapy Study in Intensive Care Unit PatiEnts. Crit Care 2012;16:R140. [8] Morgera S. Regional anticoagulation with citrate: expanding its indications. Crit Care Med 2011;39:399–400. [9] Hetzel GR, Schmitz M, Wissing H, Ries W, Schott G, Heering PJ, et al. Regional citrate versus systemic heparin for anticoagulation in critically ill patients on continuous venovenous haemofiltration: a prospective randomized multicentre trial. Nephrol Dial Transplant 2011;26:232–9. [10] Disease K. Improving Global Outcomes (KDIGO) Acute Kidney Injury Work Group: KDIGO Clinical Practice Guideline for Acute Kidney Injury. Kidney Int 2012;2:1–138. [11] Simpson DP. Citrate excretion: a window on renal metabolism. Am J Physiol 1983; 244:F223–34. [12] Bauer E, Derfler K, Joukhadar C, Druml W. Citrate kinetics in patients receiving longterm hemodialysis therapy. Am J Kidney Dis 2005;46:903–7. [13] Schultheiß C, Saugel B, Phillip V, Thies P, Noe S, Mayr U, et al. Continuous venovenous hemodialysis with regional citrate anticoagulation in patients with liver failure: a prospective observational study. Crit Care 2012;16(4):R162. [14] Morath C, Miftari N, Dikow R, Hainer C, Zeier M, Morgera S, et al. Sodium citrate anticoagulation during sustained low efficiency dialysis (SLED) in patients with acute renal failure and severely impaired liver function. Nephrol Dial Transplant 2008;23(1):421–2 [Epub 2007 Oct 3]. [15] Diaz J, Acosta F, Parrilla P, Sansano T, Bento M, Cura S, et al. Citrate intoxication and blood concentration of ionized calcium in liver transplantation. Transplant Proc 1994;26:3669–70. [16] Meier-Kriesche HU, Finkel KW, DuBose Jr TD. Unexpected severe hypocalcemia during continuous venovenous hemodialysis with regional citrate anticoagulation. Am J Kidney Dis 1999;33:e8. [17] Meier-Kriesche HU, Gitomer J, Finkel K, DuBose T. Increased total to ionized calcium ratio during continuous venovenous hemodialysis with regional citrate anticoagulation. Crit Care Med 2001;29:748–52. [18] Berbece AN, Richardson RM. Sustained low-efficiency dialysis in the ICU: cost, anticoagulation, and solute removal. Kidney Int 2006;70:963–8. [19] Zender R, de Torrenté C, Schneider U. Enzymatic determination of citrate in plasma without deproteinization. Clin Chim Acta 1969;24:335–40. [20] Kramer L, Bauer E, Joukhadar C, Strobl W, Gendo A, Madl C. Citrate pharmacokinetics and metabolism in cirrhotic and noncirrhotic critically ill patients. Crit Care Med 2003;31:2450–5. [21] Hetzel GR, Taskaya G, Sucker C, Hennersdorf M, Grabensee B, Schmitz M. Citrate plasma levels in patients under regional anticoagulation in continuous venovenous hemofiltration. Am J Kidney Dis 2006;48:806–11. [22] Díaz J, Acosta F, Parrilla P, Sansano T, Contreras RF, Bueno FS, et al. Correlation among ionized calcium, citrate, and total calcium levels during hepatic tranplantation. Clin Biochem 1995;28:315–7. [23] Apsner R, Schwarzenhofer M, Derfler K, Zauner C, Ratheiser K, Kranz A. Impairment of citrate metabolism in acute hepatic failure. Wien Klin Wochenschr 1997;109:123–7. [24] Cooper GS, Bellamy P, Dawson NV, Desbiens N, Fulkerson Jr WJ, Goldman L, et al. A prognostic model for patients with end-stage liver disease. Gastroenterology 1997; 113:1278–88. [25] Pugh RN. Pugh's grading in the classification of liver decompensation. Gut 1992;33: 1583. [26] Zipprich A, Kuss O, Rogowski S, Kleber G, Lotterer E, Seufferlein T, et al. Incorporating indocyanin green clearance into the Model for End Stage Liver Disease (MELD-ICG) improves prognostic accuracy intermediate to advanced cirrhosis. Gut 2010;59:963–8.
Please cite this article as: Lahmer T, et al, Sustained low-efficiency dialysis with regional citrate anticoagulation in medical intensive care unit patients with liver failure..., J Crit Care (2015), http://dx.doi.org/10.1016/j.jcrc.2015.06.006