- Email: [email protected]
In vivo quantification of liver dialysis: Comparison of albumin dialysis and fractionated plasma separation
Journal of Hepatology 43 (2005) 451–457 www.elsevier.com/locate/jhep
In vivo quantification of liver dialysis: Comparison of albumin dialysis and fractionated plasma separation* Peter Krisper1,*, Bernd Haditsch1, Rudolf Stauber2, Aleksandra Jung3,4, Vanessa Stadlbauer2, Michael Trauner2, Herwig Holzer1, Daniel Schneditz3 1
Division of Nephrology and Hemodialysis, Department of Internal Medicine, Medical University Graz, Auenbruggerplatz 27, A-8036 Graz, Austria 2 Division of Gastroenterology and Hepatology, Department of Internal Medicine, Medical University Graz, Graz, Austria 3 Department of Systems Physiology, Medical University Graz, Graz, Austria 4 Department of Medical Physics, AGH University of Science and Technology, Krakow, Poland
Background/Aims: Artificial liver support represents a potentially useful option for the treatment of severe liver failure. A sufficient ‘dose’ might be crucial for such treatments to provide a survival benefit. The aim of this study was to compare in vivo efficiency and resulting delivered treatment dose of two commercially available devices that use different therapeutic principles: albumin dialysis (AD, MARSw) and fractionated plasma separation (FPS, Prometheusw). Methods: Eight patients with acute-on-chronic liver failure were treated alternately with AD and FPS. Thirty-two treatments at identical blood and dialysate flow rates were evaluated. Clearance and reduction ratio (a measure of delivered treatment dose) were compared for bilirubin subfractions, ammonia and urea. Results: FPS achieved significantly higher clearance for all measured protein-bound and water-soluble markers. This resulted in significantly higher reduction ratios for FPS compared to AD. Unconjugated bilirubin, a marker for strongly albumin-bound toxins, was influenced only by FPS. Conclusions: FPS provided a higher delivered treatment dose than a matching treatment with AD. Reduction ratios of bilirubin and urea should be reported in clinical studies on liver dialysis, since delivered dose is likely to be linked to the clinical effectiveness of extracorporeal liver support therapies. q 2005 European Association for the Study of the Liver. Published by Elsevier B.V. All rights reserved. Keywords: Liver failure; Liver support; Quantification; Bilirubin; Clearance; Reduction ratio; Albumin dialysis; Plasma separation; MARS; Prometheus
1. Introduction Several techniques of extracorporeal liver support (ELS) have been tested with varying success over the past 35 years for the treatment of life-threatening complications of acute or acute-on-chronic liver failure [1–5]. ‘Liver dialysis’ has
Received 23 October 2004; received in revised form 8 February 2005; accepted 18 February 2005; available online 12 May 2005 * The authors have declared that they received funding from drug companies involved in order to carry out their research. * Corresponding author. Tel.: C43 316 385 4373; fax: C43 316 385 4426. E-mail address: [email protected] (P. Krisper).
gained increasing attention since relatively simple detoxification devices with good biocompatibility were introduced for clinical use in 1999 (albumin dialysis, AD) [6] and 2003 (fractionated plasma separation, FPS) [7], respectively. These systems are capable of removing both water-solved and albumin-bound toxins, thought to play an important role in the pathogenesis of liver failure. Despite several encouraging case reports and small studies [8–12] indicating that ELS might reduce mortality in acute-onchronic liver failure [13,14] these techniques must still be considered experimental, since increased patient survival remains to be proven by larger ongoing trials. To evaluate the possible benefit of a treatment, it is essential to know the delivered dose of the treatment.
0168-8278/$30.00 q 2005 European Association for the Study of the Liver. Published by Elsevier B.V. All rights reserved. doi:10.1016/j.jhep.2005.02.038
30a 2a 3a 9a 106a 289 (OLT) 219 183 2 – 1 2 3 2 2 2 2 1 1 2 3 2 2 2 15 12 12 16 9 12 10 8 42 47 18 42 33 48 20 27 12 12 11 13 12 14 n.a. 12 2.3 4.4 0.6 2.5 1.8 5.9 1.1 1.0 103 51 70 76 37 63 34 186 M, MARS; P, Prometheus; OLT, orthotopic liver transplantation; CP, Child Pugh. a Fatal outcome.
2.8 2.8 2.8 2.9 2.9 2.5 3.3 2.4 Cirrhosis alcohol abuse Cirrhosis alcohol abuse Liver metastasis Cirrhosis alcohol abuse Cirrhosis alcohol abuse Cirrhosis alcohol abuse OLT rec. hepatitis C Cirrhosis alcohol abuse 2 3 4 5 6 7 8 9
m m m m f m m f
55 51 61 51 65 60 61 56
39.0 81.1 25.3 32.6 37.4 38.1 19.9 26.3
3.42 2.5 1.42 2.94 1.89 2.7 1.18 2.1
2 2 16 42 12 1.0 46 3.0 15.7 Cirrhosis Hep C, alcohol 1
f
53
Spont. bacterial peritonitis Alcoholic hepatitis Alcoholic hepatitis Liver resection Bleeding Alcoholic hepatitis Alcoholic hepatitis Post op. cholestasis Alcoholic hepatitis
3.62
P M SOFA score MELD score CP score Creatinine (mg/dl) Platelets (103/mm3) INR Albumin (mg/dl) Bilirubin (mg/dl) Precipitating event Underlying disease Age Sex ID
Table 1 Patient characteristics at time of inclusion
6a
P. Krisper et al. / Journal of Hepatology 43 (2005) 451–457 Follow-up days
452
This is true not only for drugs but also for other therapeutic approaches, like the removal of toxins. One can speculate that a certain treatment may basically have the potential to improve a clinical condition, but fails due to an insufficient dose. This strong connection between delivered dose and outcome has been demonstrated in the closely related field of hemodialysis, in the setting of both acute and chronic renal replacement therapy [15,16]. Quantification of dialysis treatments, therefore, has become a standard for accurate clinical research and patient care in nephrology [17]. Interestingly, the current literature on ELS only rarely describes ‘how much’ of the treatment under investigation has been delivered to the patient [18,19]. Our study applies the simple concepts of clearance and reduction ratio for quantification of liver dialysis. Markers were chosen according to some basic considerations: an appropriate substance should accumulate in liver failure, should be related to toxicity, eliminated by the treatment and easily available for analysis. Since ELS—like the normal liver itself—eliminates both albumin-bound and watersolved toxins, two appropriate markers are needed to quantify the treatment dose. Bilirubin—albeit virtually non-toxic in adults—seems to be a suitable and convenient marker for removal of albumin-bound toxins. It is noteworthy that urea is a well established marker in uremia, and is also not toxic per se. For hydrophilic substances, ammonia, which is directly related to hepatic encephalopathy [20], might be appropriate, although its reduction ratio might be less reliable due to its relatively high generation rate. The aim of this study was to compare two commercially available liver detoxification devices—i.e. albumin dialysis (AD, MARSw) and fractionated plasma separation (FPS, Prometheusw)—in vivo by their clearance as a measure of detoxification efficiency and by reduction ratio as a measure of delivered treatment dose. This study was not designed to investigate differences in clinical outcomes between the two devices.
2. Patients and methods 2.1. Patients The study protocol was approved by the local Ethics Committee. Written informed consent was obtained in advance. If the patient was unconscious at the time of inclusion, written informed consent was obtained after return to consciousness if applicable. The indication for ELS was established by a hepatologist (R.S., M.T.) in close cooperation with a nephrologist (P.K., H.H.). All 10 patients treated at our institution between March 2003 and April 2004 were eligible for this study.
2.2. Treatment Two different liver detoxification devices were used: MARSw (Teraklin, Rostock, Germany), which is based on the principle of albumin dialysis (AD) and adsorption [6] and Prometheusw (Fresenius Medical
P. Krisper et al. / Journal of Hepatology 43 (2005) 451–457
Heparin, epoprostenol (Flolanw, 4 ng/kg/min) or both were used for anticoagulation; activated partial prothrombin time (aPTT) was aimed to remain below 100 s. All patients were treated via a central venous catheter. Albumin was not administered during the treatment.
Care, Bad Homburg, Germany), which applies fractionated plasma separation (FPS) and adsorption [21,22]. Patients were randomly assigned to start with either AD or FPS and treated with the other device the following day. Treatments were performed by experienced nephrology personnel either in the intensive care unit or— with clinically stable patients—in the dialysis unit. Each treatment was performed for a duration of 6 h at identical blood and dialysate flows for all patients (200 and 300 mL/min, respectively), and the same dialysis machine (4008 H, Fresenius Medical Care) was used throughout the study. The flow in the secondary circuit was set according to the manufacturer’s recommendations (200 mL/min in AD, 300 mL/min in FPS). The dialysate contained glucose (1 g/L) and magnesium (0.75 mmol/L), while sodium, potassium and bicarbonate concentrations were adjusted to fit the patient’s needs.
A
M
160
Arterial blood pressure, heart rate and body weight were assessed. Ultrafiltration rates (UFR) and effective blood flow were obtained from the dialysis machine, and recirculation of the central venous catheter was checked at every treatment (Transonic HD01, Transonic Corp., Ithaca, New York). Before treatment, a blood sample was drawn directly from the central venous catheter. During the treatment, blood samples were drawn at 0.5, 1,
*
*
P
*
*
140 120 100 80
180
M
*
160
* Cl Urea [ml/min]
Cl Ammonia [ml/min]
2.3. Measurements
B
180
*
*
1
2
*
140 120 100 80
40
40 0
1
2
3
4
5
0
6
C 30
M
4
D 25
P
*
5
6
M
PlCl u -Bili [ml/min]
*
15 10
P
*
20
* 20
3
Time [h]
Time [h]
PlCl t- Bili [ml/min]
P
*
60
60
25
453
*
15
*
10
* *
5 0
5
–5
0 0
1
2
3
4
5
0
6
1
2
E 30
M
4
5
6
P
*
25
PlCl c- Bili [ml/min]
3
Time [h]
Time [h]
*
20
*
15 10 5 0 0
1
2
3
4
5
6
Time [h] Fig. 1. (A)–(E) Clearance as a measure of efficiency for MARSw (M, dotted line) and Prometheusw (P, solid line) at different time points of treatment. Blood clearance is given for water-soluble ammonia (A) and urea (B); plasma clearance for albumin-bound t-bili (C); u-bili (D); and c-bili (E). Note that P has significantly higher clearance than M for all investigated parameters. Values are means G SEM. An asterisk (*) signifies a significant difference between groups. t-Bili, u-bili, c-bili: total, unconjugated, and conjugated bilirubin; Cl, clearance; PlCl, plasma clearance.
P. Krisper et al. / Journal of Hepatology 43 (2005) 451–457
2, 4 and 6 h as paired samples from the ‘arterial’ (inflow) and ‘venous’ (outflow) lines of the extracorporeal circuit. Samples were centrifuged within 10 min and transported in a light-shielded container to the laboratory. Total, conjugated and unconjugated bilirubin (t-bili, c-bili and u-bili), ammonia and urea were measured by multilayer film technique (Vitros 950w, Ortho-Clinical Diagnostics, Raritan, NJ, USA). Hematocrit (Hct) was derived indirectly by an automated cell counter (Sysmexe XE 2100, Toa Medical Electronics, Kobe, Japan). Calibration followed the manufacturer’s instructions. Quality control measurements were performed twice daily using dedicated control material. The coefficient of variation for all methods did not exceed 5%.
2.4. Calculations Clearance (Cl) was calculated from paired arterial (a) and venous (v) samples and from effective blood flow (Qb): ClZ ð1K v=aÞ !Qb. Since bilirubin is cleared from the plasma fraction of whole blood, plasma clearance (PlCl) was determined for extracorporeal plasma flow according to the following relationship: PlClZ Cl !ð1K 0:01 !HctÞ. Reduction ratio (RR) was calculated from pre- and post-treatment concentrations: RRZ ð1K post=preÞ.
2.5. Statistical analysis Data from patients completing at least one pair of consecutive AD and FPS treatments were included for analysis. Results are given as meansGSD unless otherwise stated and were compared by the Mann–Whitney test. Changes in plasma concentration and clearance over time were assessed by the Friedmann’s test. A P!0.05 was considered significant. StatView 4.5 (Abacus Concepts, Berkeley, CA) was used for data analysis.
3. Results Nine of 10 eligible patients entered the study. One patient could not be included for organizational reasons. The patients’ characteristics are listed in Table 1. Eight patients completed at least one pair of treatments, so that a total of 32 treatments (16 completed pairs) were available for analysis. Treatments were well tolerated and there were no severe adverse events that could be reasonably attributed to the treatment. Therapy was interrupted for up to 30 min twice during AD (leakage, clotting) and three times during FPS (clotting). Pre-treatment values, achieved effective blood flows (191G4 vs. 191G7 mL/min), plasma flows (140G8 vs. 142G8 mL/min) and ultrafiltration rates (155G209 vs. 193G243 mL/h) did not differ between FPS and AD. 3.1. Clearance as measure of efficiency To compare the efficiency of the two ELS devices, blood clearance for water-soluble and plasma clearance for albumin-bound substances was determined (Fig. 1). Clearance of all measured markers was higher with FPS than AD, although these differences lost significance beyond 4 h for tbili and c-bili. As expected, the clearance for u-bili, the more tightly albumin-bound fraction of bilirubin, was smaller than for the more loosely bound c-bili. With AD, an apparently ‘negative’ clearance for u-bili was even observed, as concentrations in venous line samples tended to be higher
than in samples from the arterial line. This can be explained by the effect of ultrafiltration-induced hemocentration of the venous sample, as indicated by a significant negative correlation between plasma clearance of u-bili and UFR in AD treatments: PlCl u-biliZ 0:7K 0:0093 !UFR (rZ0.52, P!0.0001). Clearance for urea remained stable throughout the treatment in both groups, as seen in hemodialysis. For ammonia—the second water-soluble marker evaluated— clearance significantly decreased over time with AD but remained constant with FPS. Clearance for bilirubin and its subfractions decreased during the treatment course with both devices, suggesting saturation of absorbers. 3.2. Reduction ratio as a measure of delivered dose To determine whether higher clearance led to higher delivered treatment dose, reduction ratios for albuminbound and water-soluble markers were calculated. Significantly higher reduction ratios for FPS compared to AD were found for bilirubin and urea: 37G7 vs. 28G6% for t-bili, 16G16 vs. K4G12% for u-bili, 52G12 vs. 43G7% for c-bili and 59G6 vs. 51G5 for urea (Fig. 2). Despite significantly higher ammonia clearance for FPS, there was no significant difference in the ammonia reduction ratio: 40G18 vs. 35G14% for FPS and AD, respectively. Relatively high rates of ammonia synthesis compared to elimination could contribute to this finding. Despite higher treatment efficiency and delivered dose with FPS, plasma levels did not differ significantly between FPS and AD (Fig. 3). This indicates that evaluation of plasma levels alone is not sufficient for accurate quantification of ELS.
4. Discussion Quantification of liver dialysis is a prerequisite for accurate clinical research. While an in vitro comparison of 70
Reduction Ratio (RR) [%]
454
M
P
*
*
60 50 40
*
30
*
20 10 0 –10
t-Bili
u-Bili
c-Bili
Ammonia
Urea
Fig. 2. Reduction ratio as a measure of delivered treatment dose for MARSw (M, light columns) and Prometheusw (P, dark columns). t-Bili, u-bili, c-bili: total, unconjugated, and conjugated bilirubin; values are meansGSEM. An asterisk (*) signifies a significant difference between the two groups.
P. Krisper et al. / Journal of Hepatology 43 (2005) 451–457
B 90
P
M
80
80
70
70
Urea [mg/dl]
Ammonia [µg/dl]
A 90
60 50
M
P
5
6
M
P
5
6
60 50 40
40
30
30 –1
0
1
2
3
4
5
6
–1
0
1
2
Time [h] C
455
3
4
Time [h]
25
D
P
M
5
20
u-Bili [mg/dl]
t-Bili [mg/dl]
4
15
3
2 1
10 –1
0 0
1
2
3
4
5
–1
6
0
1
2
E 20
c-Bili [mg/dl]
3
4
Time [h]
Time [h] M
P
15
10
5 –1
0
1
2
3
4
5
6
Time [h] Fig. 3. (A)–(E). Plasma levels (meanGSEM) for MARSw (M, dotted line) and Prometheusw (P, solid line) at different time points of treatment for ammonia (A); urea (B); t-bili (C); u-bili (D) and c-bili (E). No significant differences were found between the two groups. All measured markers decreased during the treatment with the exception of unconjugated bilirubin with M, which remained unchanged. Values are meansGSEM. t-Bili, ubili, c-bili: total, unconjugated, and conjugated bilirubin.
two procedures for AD has been published recently [23], this is the first study that applies standard quantification methods to evaluate different methods of ELS in vivo: MARSw—a variation of AD—and Prometheusw, which is based on FPS. Our results indicate that Prometheusw produced higher clearance for protein-bound and watersoluble markers. This resulted in higher delivered treatment dose compared to a matching treatment with MARSw. Clearance expresses the hypothetical volume per time unit (e.g. mL/min) that is completely ‘cleared’ by an organ (e.g. kidney or liver) or a device (e.g. dialyzer or
adsorber) and is a convenient measure for the efficiency of any detoxification method. For bilirubin, we observed a declining clearance during the treatment period with both systems, an expected effect due to adsorber saturation. Similar to hemodialysis, where saturation effects are absent, clearance of water-soluble substances remained stable, with the exception of a significant decrease in ammonia clearance in MARSw treatments. Higher treatment efficiency for albumin-bound markers with FPS than with AD may be explained by the way these
456
P. Krisper et al. / Journal of Hepatology 43 (2005) 451–457
systems take advantage of different principles of solute transport: in AD, solutes have to separate from the patient’s albumin in order to cross the albumin-impermeable membrane by diffusion, while in FPS, substrates reach the absorbers by diffusion and convection, as part of the patient’s albumin with the attached toxins is directly filtered into the secondary circuit. The higher clearance of watersoluble markers with Prometheusw may be attributed to the fact that this device places a high flux dialyzer directly into the bloodstream, while in MARSw, water-soluble substances partly bypass the low flux dialyzer, which is placed in the secondary plasma circuit. The differences in treatment efficiency resulted in higher delivered treatment dose by FPS than by AD as expressed by the reduction ratio of albumin-bound as well as hydrophilic markers. Mere evaluation of the patients’ blood concentrations would not have identified these differences between the two devices. Reduction ratios are easy to determine and should serve as a first approach to measuring the delivered dose of ELS. It must, however, be borne in mind that this concept also has its limitations, which may lead to either under- or overestimation of treatment dose. In vivo, multiple determinants in addition to removal rate influence the observable reduction of a substrate, including generation rates and transfer rates between multiple pools. Our study demonstrates this with the fact that with FPS, the reduction ratio for ammonia is clearly lower than for urea (40 vs. 59%), despite comparable clearance (141 vs. 147 mL/min). A higher generation rate in combination with rapid refilling from the extravascular compartments could explain this phenomenon. That is why for ammonia, a reduction ratio will considerably underestimate the treatment effect and probably limits its analytic usefulness. The urea reduction ratio could serve as an alternative and be used in combination with bilirubin as measure of treatment dose in ELS. Further, determination of a reduction ratio does not take into account possible post-treatment rebound of solutes, which would lead to an overestimation of delivered dose. In vivo, the phenomenon of rebound occurs almost universally, because substances usually are distributed in more than a single pool. In fact, this also seems to be the case for bilirubin and ammonia [24]. Such effects were of major importance in the accurate measurement of the dose of hemodialysis and have been addressed by urea kinetic modeling [25]. Bilirubin kinetic modeling could serve the same purpose for ELS in the future. Comparison of the technical efficiency of two devices— which was the purpose of this study—is only one application for quantification of ELS. To decide whether treatment intensity really has an impact on clinical outcome, in a first step, different patient groups—e.g. responders vs. non-responders, survivors vs. non-survivors—could be compared for delivered treatment dose. This might serve as the basis for optimization of treatment intensity, length and schedule.
In conclusion, Prometheusw—based on FPS—produced significantly higher clearance for protein-bound and watersoluble markers, which resulted in higher delivered treatment dose compared to a matching treatment with MARSw (AD). We want to emphasize that quantification of ELS is mandatory for accurate clinical research in this emerging field, and reduction ratios of bilirubin, urea and possibly ammonia should be reported in clinical studies on liver dialysis.
Acknowledgements The authors should like to thank G. Bergmann and G.J. Krejs, LKH-Univ. Klinikum Graz, for establishing an interdisciplinary study group on ELS in our hospital. Special thanks to the nursing staff of our dialysis unit for their dedicated patient care, to U. Leitner, C. Berger, S. Po¨tz and U. Spitzer for their painstaking sample collection and to W. Erwa for the analysis of plasma samples. We also thank A. Krause and E. Lamont for their assistance in preparing the manuscript. This study was supported in part by Fresenius Medical Care.
References [1] Abouna GM, Garry R, Hull C, Kirkley J, Walder DN. Pig-liver perfusion in hepatic coma. Lancet 1968;2:509–510. [2] O’Grady JG, Gimson AE, O’Brien CJ, Pucknell A, Hughes RD, Williams R. Controlled trials of charcoal hemoperfusion and prognostic factors in fulminant hepatic failure. Gastroenterology 1988;94:1186–1192. [3] Ash SR, Blake DE, Carr DJ, Carter C, Howard T, Makowka L. Clinical effects of a sorbent suspension dialysis system in treatment of hepatic coma (the BioLogic-DT). Int J Artif Organs 1992;15: 151–161. [4] Sauer IM, Zeilinger K, Pless G, Kardassis D, Theruvath T, Pascher A, et al. Extracorporeal liver support based on primary human liver cells and albumin dialysis—treatment of a patient with primary graft nonfunction. J Hepatol 2003;39:649–653. [5] Demetriou AA, Brown Jr RS, Busuttil RW, Fair J, McGuire BM, Rosenthal P, et al. Prospective, randomized, multicenter, controlled trial of a bioartificial liver in treating acute liver failure. Ann Surg 2004;239:660–667. [6] Stange J, Mitzner SR, Risler T, Erley CM, Lauchart W, Goehl H, et al. Molecular adsorbent recycling system (MARS): clinical results of a new membrane-based blood purification system for bioartificial liver support. Artif Organs 1999;23:319–330. [7] Rifai K, Ernst T, Kretschmer U, Bahr MJ, Schneider A, Hafer C, et al. Prometheus—a new extracorporeal system for the treatment of liver failure. J Hepatol 2003;39:984–990. [8] Mitzner SR, Stange J, Klammt S, Risler T, Erley CM, Bader BD, et al. Improvement of hepatorenal syndrome with extracorporeal albumin dialysis MARS: results of a prospective, randomized, controlled clinical trial. Liver Transpl 2000;6:277–286. [9] Sen S, Felldin M, Steiner C, Larsson B, Gillett GT, Olausson M, et al. Albumin dialysis and molecular adsorbents recirculating system (MARS) for acute Wilson’s disease. Liver Transpl 2002;8:962–967.
P. Krisper et al. / Journal of Hepatology 43 (2005) 451–457 [10] Heemann U, Treichel U, Loock J, Philipp T, Gerken G, Malago M, et al. Albumin dialysis in cirrhosis with superimposed acute liver injury: a prospective, controlled study. Hepatology 2002;36:949–958. [11] Jalan R, Sen S, Steiner C, Kapoor D, Alisa A, Williams R. Extracorporeal liver support with molecular adsorbents recirculating system in patients with severe acute alcoholic hepatitis. J Hepatol 2003;38:24–31. [12] Kramer L, Bauer E, Schenk P, Steininger R, Vigl M, Mallek R. Successful treatment of refractory cerebral oedema in ecstasy/cocaine-induced fulminant hepatic failure using a new high-efficacy liver detoxification device (FPSA-Prometheus). Wien Klin Wochenschr 2003;115:599–603. [13] Liu JP, Gluud LL, Als-Nielsen B, Gluud C. Artificial and bioartificial liver support systems for liver failure. Cochrane Database Syst Rev 2004;CD03628. [14] Jalan R, Sen S, Williams R. Prospects for extracorporeal liver support. Gut 2004;53:890–898. [15] Gotch FA, Sargent JA. A mechanistic analysis of the national cooperative dialysis study (NCDS). Kidney Int 1985;28:526–534. [16] Ronco C, Bellomo R, Homel P, Brendolan A, Dan M, Piccinni P, et al. Effects of different doses in continuous veno-venous haemofiltration on outcomes of acute renal failure: a prospective randomised trial. Lancet 2000;356:26–30. [17] National Kidney Foundation. K/DOQI clinical practice guidelines for hemodialysis adequacy, 2000. Am J Kidney Dis 2001;37: S7–S64.
457
[18] Klammt S, Stange J, Mitzner SR, Peszynski P, Peters E, Liebe S. Extracorporeal liver support by recirculating albumin dialysis: analysing the effect of the first clinically used generation of the MARSystem. Liver 2002;22:30–34. [19] McIntyre CW, Fluck RJ, Freeman JG, Lambie SH. Characterization of treatment dose delivered by albumin dialysis in the treatment of acute renal failure associated with severe hepatic dysfunction. Clin Nephrol 2002;58:376–383. [20] Vaquero J, Chung C, Cahill ME, Blei AT. Pathogenesis of hepatic encephalopathy in acute liver failure. Semin Liver Dis 2003;23: 259–269. [21] Falkenhagen D, Strobl W, Vogt G, Schrefl A, Linsberger I, Gerner FJ, et al. Fractionated plasma separation and adsorption system: a novel system for blood purification to remove albumin-bound substances. Artif Organs 1999;23:81–86. [22] Krause A. Prometheus—a new extracorporeal liver support therapy. In: Arroyo V, Forns X, Garcia-Paga´n JC, Rode´s J, editors. Progress in the treatment of liver diseases. Barcelona: Ars Medica; 2003. p. 437–443. [23] Sauer IM, Goetz M, Steffen I, Walter G, Kehr DC, Schwartlander R, et al. In vitro comparison of the molecular recirculation system (MARS) and single-pass albumin dialysis (SPAD). Hepatology 2004; 39:1408–1414. [24] Jung A, Krisper P, Haditsch B, Stauber R, Holzer H, Schneditz D. Bilrubin kinetics in extracorporeal liver support. Int J Artif Organs 2004;27:588 abstract. [25] Schneditz D, Daugirdas JT. Compartment effects in hemodialysis. Semin Dial 2001;14:271–277.
In vivo quantification of liver dialysis: Comparison of albumin dialysis and fractionated plasma separation* Peter Krisper1,*, Bernd Haditsch1, Rudolf Stauber2, Aleksandra Jung3,4, Vanessa Stadlbauer2, Michael Trauner2, Herwig Holzer1, Daniel Schneditz3 1
Division of Nephrology and Hemodialysis, Department of Internal Medicine, Medical University Graz, Auenbruggerplatz 27, A-8036 Graz, Austria 2 Division of Gastroenterology and Hepatology, Department of Internal Medicine, Medical University Graz, Graz, Austria 3 Department of Systems Physiology, Medical University Graz, Graz, Austria 4 Department of Medical Physics, AGH University of Science and Technology, Krakow, Poland
Background/Aims: Artificial liver support represents a potentially useful option for the treatment of severe liver failure. A sufficient ‘dose’ might be crucial for such treatments to provide a survival benefit. The aim of this study was to compare in vivo efficiency and resulting delivered treatment dose of two commercially available devices that use different therapeutic principles: albumin dialysis (AD, MARSw) and fractionated plasma separation (FPS, Prometheusw). Methods: Eight patients with acute-on-chronic liver failure were treated alternately with AD and FPS. Thirty-two treatments at identical blood and dialysate flow rates were evaluated. Clearance and reduction ratio (a measure of delivered treatment dose) were compared for bilirubin subfractions, ammonia and urea. Results: FPS achieved significantly higher clearance for all measured protein-bound and water-soluble markers. This resulted in significantly higher reduction ratios for FPS compared to AD. Unconjugated bilirubin, a marker for strongly albumin-bound toxins, was influenced only by FPS. Conclusions: FPS provided a higher delivered treatment dose than a matching treatment with AD. Reduction ratios of bilirubin and urea should be reported in clinical studies on liver dialysis, since delivered dose is likely to be linked to the clinical effectiveness of extracorporeal liver support therapies. q 2005 European Association for the Study of the Liver. Published by Elsevier B.V. All rights reserved. Keywords: Liver failure; Liver support; Quantification; Bilirubin; Clearance; Reduction ratio; Albumin dialysis; Plasma separation; MARS; Prometheus
1. Introduction Several techniques of extracorporeal liver support (ELS) have been tested with varying success over the past 35 years for the treatment of life-threatening complications of acute or acute-on-chronic liver failure [1–5]. ‘Liver dialysis’ has
Received 23 October 2004; received in revised form 8 February 2005; accepted 18 February 2005; available online 12 May 2005 * The authors have declared that they received funding from drug companies involved in order to carry out their research. * Corresponding author. Tel.: C43 316 385 4373; fax: C43 316 385 4426. E-mail address: [email protected] (P. Krisper).
gained increasing attention since relatively simple detoxification devices with good biocompatibility were introduced for clinical use in 1999 (albumin dialysis, AD) [6] and 2003 (fractionated plasma separation, FPS) [7], respectively. These systems are capable of removing both water-solved and albumin-bound toxins, thought to play an important role in the pathogenesis of liver failure. Despite several encouraging case reports and small studies [8–12] indicating that ELS might reduce mortality in acute-onchronic liver failure [13,14] these techniques must still be considered experimental, since increased patient survival remains to be proven by larger ongoing trials. To evaluate the possible benefit of a treatment, it is essential to know the delivered dose of the treatment.
0168-8278/$30.00 q 2005 European Association for the Study of the Liver. Published by Elsevier B.V. All rights reserved. doi:10.1016/j.jhep.2005.02.038
30a 2a 3a 9a 106a 289 (OLT) 219 183 2 – 1 2 3 2 2 2 2 1 1 2 3 2 2 2 15 12 12 16 9 12 10 8 42 47 18 42 33 48 20 27 12 12 11 13 12 14 n.a. 12 2.3 4.4 0.6 2.5 1.8 5.9 1.1 1.0 103 51 70 76 37 63 34 186 M, MARS; P, Prometheus; OLT, orthotopic liver transplantation; CP, Child Pugh. a Fatal outcome.
2.8 2.8 2.8 2.9 2.9 2.5 3.3 2.4 Cirrhosis alcohol abuse Cirrhosis alcohol abuse Liver metastasis Cirrhosis alcohol abuse Cirrhosis alcohol abuse Cirrhosis alcohol abuse OLT rec. hepatitis C Cirrhosis alcohol abuse 2 3 4 5 6 7 8 9
m m m m f m m f
55 51 61 51 65 60 61 56
39.0 81.1 25.3 32.6 37.4 38.1 19.9 26.3
3.42 2.5 1.42 2.94 1.89 2.7 1.18 2.1
2 2 16 42 12 1.0 46 3.0 15.7 Cirrhosis Hep C, alcohol 1
f
53
Spont. bacterial peritonitis Alcoholic hepatitis Alcoholic hepatitis Liver resection Bleeding Alcoholic hepatitis Alcoholic hepatitis Post op. cholestasis Alcoholic hepatitis
3.62
P M SOFA score MELD score CP score Creatinine (mg/dl) Platelets (103/mm3) INR Albumin (mg/dl) Bilirubin (mg/dl) Precipitating event Underlying disease Age Sex ID
Table 1 Patient characteristics at time of inclusion
6a
P. Krisper et al. / Journal of Hepatology 43 (2005) 451–457 Follow-up days
452
This is true not only for drugs but also for other therapeutic approaches, like the removal of toxins. One can speculate that a certain treatment may basically have the potential to improve a clinical condition, but fails due to an insufficient dose. This strong connection between delivered dose and outcome has been demonstrated in the closely related field of hemodialysis, in the setting of both acute and chronic renal replacement therapy [15,16]. Quantification of dialysis treatments, therefore, has become a standard for accurate clinical research and patient care in nephrology [17]. Interestingly, the current literature on ELS only rarely describes ‘how much’ of the treatment under investigation has been delivered to the patient [18,19]. Our study applies the simple concepts of clearance and reduction ratio for quantification of liver dialysis. Markers were chosen according to some basic considerations: an appropriate substance should accumulate in liver failure, should be related to toxicity, eliminated by the treatment and easily available for analysis. Since ELS—like the normal liver itself—eliminates both albumin-bound and watersolved toxins, two appropriate markers are needed to quantify the treatment dose. Bilirubin—albeit virtually non-toxic in adults—seems to be a suitable and convenient marker for removal of albumin-bound toxins. It is noteworthy that urea is a well established marker in uremia, and is also not toxic per se. For hydrophilic substances, ammonia, which is directly related to hepatic encephalopathy [20], might be appropriate, although its reduction ratio might be less reliable due to its relatively high generation rate. The aim of this study was to compare two commercially available liver detoxification devices—i.e. albumin dialysis (AD, MARSw) and fractionated plasma separation (FPS, Prometheusw)—in vivo by their clearance as a measure of detoxification efficiency and by reduction ratio as a measure of delivered treatment dose. This study was not designed to investigate differences in clinical outcomes between the two devices.
2. Patients and methods 2.1. Patients The study protocol was approved by the local Ethics Committee. Written informed consent was obtained in advance. If the patient was unconscious at the time of inclusion, written informed consent was obtained after return to consciousness if applicable. The indication for ELS was established by a hepatologist (R.S., M.T.) in close cooperation with a nephrologist (P.K., H.H.). All 10 patients treated at our institution between March 2003 and April 2004 were eligible for this study.
2.2. Treatment Two different liver detoxification devices were used: MARSw (Teraklin, Rostock, Germany), which is based on the principle of albumin dialysis (AD) and adsorption [6] and Prometheusw (Fresenius Medical
P. Krisper et al. / Journal of Hepatology 43 (2005) 451–457
Heparin, epoprostenol (Flolanw, 4 ng/kg/min) or both were used for anticoagulation; activated partial prothrombin time (aPTT) was aimed to remain below 100 s. All patients were treated via a central venous catheter. Albumin was not administered during the treatment.
Care, Bad Homburg, Germany), which applies fractionated plasma separation (FPS) and adsorption [21,22]. Patients were randomly assigned to start with either AD or FPS and treated with the other device the following day. Treatments were performed by experienced nephrology personnel either in the intensive care unit or— with clinically stable patients—in the dialysis unit. Each treatment was performed for a duration of 6 h at identical blood and dialysate flows for all patients (200 and 300 mL/min, respectively), and the same dialysis machine (4008 H, Fresenius Medical Care) was used throughout the study. The flow in the secondary circuit was set according to the manufacturer’s recommendations (200 mL/min in AD, 300 mL/min in FPS). The dialysate contained glucose (1 g/L) and magnesium (0.75 mmol/L), while sodium, potassium and bicarbonate concentrations were adjusted to fit the patient’s needs.
A
M
160
Arterial blood pressure, heart rate and body weight were assessed. Ultrafiltration rates (UFR) and effective blood flow were obtained from the dialysis machine, and recirculation of the central venous catheter was checked at every treatment (Transonic HD01, Transonic Corp., Ithaca, New York). Before treatment, a blood sample was drawn directly from the central venous catheter. During the treatment, blood samples were drawn at 0.5, 1,
*
*
P
*
*
140 120 100 80
180
M
*
160
* Cl Urea [ml/min]
Cl Ammonia [ml/min]
2.3. Measurements
B
180
*
*
1
2
*
140 120 100 80
40
40 0
1
2
3
4
5
0
6
C 30
M
4
D 25
P
*
5
6
M
PlCl u -Bili [ml/min]
*
15 10
P
*
20
* 20
3
Time [h]
Time [h]
PlCl t- Bili [ml/min]
P
*
60
60
25
453
*
15
*
10
* *
5 0
5
–5
0 0
1
2
3
4
5
0
6
1
2
E 30
M
4
5
6
P
*
25
PlCl c- Bili [ml/min]
3
Time [h]
Time [h]
*
20
*
15 10 5 0 0
1
2
3
4
5
6
Time [h] Fig. 1. (A)–(E) Clearance as a measure of efficiency for MARSw (M, dotted line) and Prometheusw (P, solid line) at different time points of treatment. Blood clearance is given for water-soluble ammonia (A) and urea (B); plasma clearance for albumin-bound t-bili (C); u-bili (D); and c-bili (E). Note that P has significantly higher clearance than M for all investigated parameters. Values are means G SEM. An asterisk (*) signifies a significant difference between groups. t-Bili, u-bili, c-bili: total, unconjugated, and conjugated bilirubin; Cl, clearance; PlCl, plasma clearance.
P. Krisper et al. / Journal of Hepatology 43 (2005) 451–457
2, 4 and 6 h as paired samples from the ‘arterial’ (inflow) and ‘venous’ (outflow) lines of the extracorporeal circuit. Samples were centrifuged within 10 min and transported in a light-shielded container to the laboratory. Total, conjugated and unconjugated bilirubin (t-bili, c-bili and u-bili), ammonia and urea were measured by multilayer film technique (Vitros 950w, Ortho-Clinical Diagnostics, Raritan, NJ, USA). Hematocrit (Hct) was derived indirectly by an automated cell counter (Sysmexe XE 2100, Toa Medical Electronics, Kobe, Japan). Calibration followed the manufacturer’s instructions. Quality control measurements were performed twice daily using dedicated control material. The coefficient of variation for all methods did not exceed 5%.
2.4. Calculations Clearance (Cl) was calculated from paired arterial (a) and venous (v) samples and from effective blood flow (Qb): ClZ ð1K v=aÞ !Qb. Since bilirubin is cleared from the plasma fraction of whole blood, plasma clearance (PlCl) was determined for extracorporeal plasma flow according to the following relationship: PlClZ Cl !ð1K 0:01 !HctÞ. Reduction ratio (RR) was calculated from pre- and post-treatment concentrations: RRZ ð1K post=preÞ.
2.5. Statistical analysis Data from patients completing at least one pair of consecutive AD and FPS treatments were included for analysis. Results are given as meansGSD unless otherwise stated and were compared by the Mann–Whitney test. Changes in plasma concentration and clearance over time were assessed by the Friedmann’s test. A P!0.05 was considered significant. StatView 4.5 (Abacus Concepts, Berkeley, CA) was used for data analysis.
3. Results Nine of 10 eligible patients entered the study. One patient could not be included for organizational reasons. The patients’ characteristics are listed in Table 1. Eight patients completed at least one pair of treatments, so that a total of 32 treatments (16 completed pairs) were available for analysis. Treatments were well tolerated and there were no severe adverse events that could be reasonably attributed to the treatment. Therapy was interrupted for up to 30 min twice during AD (leakage, clotting) and three times during FPS (clotting). Pre-treatment values, achieved effective blood flows (191G4 vs. 191G7 mL/min), plasma flows (140G8 vs. 142G8 mL/min) and ultrafiltration rates (155G209 vs. 193G243 mL/h) did not differ between FPS and AD. 3.1. Clearance as measure of efficiency To compare the efficiency of the two ELS devices, blood clearance for water-soluble and plasma clearance for albumin-bound substances was determined (Fig. 1). Clearance of all measured markers was higher with FPS than AD, although these differences lost significance beyond 4 h for tbili and c-bili. As expected, the clearance for u-bili, the more tightly albumin-bound fraction of bilirubin, was smaller than for the more loosely bound c-bili. With AD, an apparently ‘negative’ clearance for u-bili was even observed, as concentrations in venous line samples tended to be higher
than in samples from the arterial line. This can be explained by the effect of ultrafiltration-induced hemocentration of the venous sample, as indicated by a significant negative correlation between plasma clearance of u-bili and UFR in AD treatments: PlCl u-biliZ 0:7K 0:0093 !UFR (rZ0.52, P!0.0001). Clearance for urea remained stable throughout the treatment in both groups, as seen in hemodialysis. For ammonia—the second water-soluble marker evaluated— clearance significantly decreased over time with AD but remained constant with FPS. Clearance for bilirubin and its subfractions decreased during the treatment course with both devices, suggesting saturation of absorbers. 3.2. Reduction ratio as a measure of delivered dose To determine whether higher clearance led to higher delivered treatment dose, reduction ratios for albuminbound and water-soluble markers were calculated. Significantly higher reduction ratios for FPS compared to AD were found for bilirubin and urea: 37G7 vs. 28G6% for t-bili, 16G16 vs. K4G12% for u-bili, 52G12 vs. 43G7% for c-bili and 59G6 vs. 51G5 for urea (Fig. 2). Despite significantly higher ammonia clearance for FPS, there was no significant difference in the ammonia reduction ratio: 40G18 vs. 35G14% for FPS and AD, respectively. Relatively high rates of ammonia synthesis compared to elimination could contribute to this finding. Despite higher treatment efficiency and delivered dose with FPS, plasma levels did not differ significantly between FPS and AD (Fig. 3). This indicates that evaluation of plasma levels alone is not sufficient for accurate quantification of ELS.
4. Discussion Quantification of liver dialysis is a prerequisite for accurate clinical research. While an in vitro comparison of 70
Reduction Ratio (RR) [%]
454
M
P
*
*
60 50 40
*
30
*
20 10 0 –10
t-Bili
u-Bili
c-Bili
Ammonia
Urea
Fig. 2. Reduction ratio as a measure of delivered treatment dose for MARSw (M, light columns) and Prometheusw (P, dark columns). t-Bili, u-bili, c-bili: total, unconjugated, and conjugated bilirubin; values are meansGSEM. An asterisk (*) signifies a significant difference between the two groups.
P. Krisper et al. / Journal of Hepatology 43 (2005) 451–457
B 90
P
M
80
80
70
70
Urea [mg/dl]
Ammonia [µg/dl]
A 90
60 50
M
P
5
6
M
P
5
6
60 50 40
40
30
30 –1
0
1
2
3
4
5
6
–1
0
1
2
Time [h] C
455
3
4
Time [h]
25
D
P
M
5
20
u-Bili [mg/dl]
t-Bili [mg/dl]
4
15
3
2 1
10 –1
0 0
1
2
3
4
5
–1
6
0
1
2
E 20
c-Bili [mg/dl]
3
4
Time [h]
Time [h] M
P
15
10
5 –1
0
1
2
3
4
5
6
Time [h] Fig. 3. (A)–(E). Plasma levels (meanGSEM) for MARSw (M, dotted line) and Prometheusw (P, solid line) at different time points of treatment for ammonia (A); urea (B); t-bili (C); u-bili (D) and c-bili (E). No significant differences were found between the two groups. All measured markers decreased during the treatment with the exception of unconjugated bilirubin with M, which remained unchanged. Values are meansGSEM. t-Bili, ubili, c-bili: total, unconjugated, and conjugated bilirubin.
two procedures for AD has been published recently [23], this is the first study that applies standard quantification methods to evaluate different methods of ELS in vivo: MARSw—a variation of AD—and Prometheusw, which is based on FPS. Our results indicate that Prometheusw produced higher clearance for protein-bound and watersoluble markers. This resulted in higher delivered treatment dose compared to a matching treatment with MARSw. Clearance expresses the hypothetical volume per time unit (e.g. mL/min) that is completely ‘cleared’ by an organ (e.g. kidney or liver) or a device (e.g. dialyzer or
adsorber) and is a convenient measure for the efficiency of any detoxification method. For bilirubin, we observed a declining clearance during the treatment period with both systems, an expected effect due to adsorber saturation. Similar to hemodialysis, where saturation effects are absent, clearance of water-soluble substances remained stable, with the exception of a significant decrease in ammonia clearance in MARSw treatments. Higher treatment efficiency for albumin-bound markers with FPS than with AD may be explained by the way these
456
P. Krisper et al. / Journal of Hepatology 43 (2005) 451–457
systems take advantage of different principles of solute transport: in AD, solutes have to separate from the patient’s albumin in order to cross the albumin-impermeable membrane by diffusion, while in FPS, substrates reach the absorbers by diffusion and convection, as part of the patient’s albumin with the attached toxins is directly filtered into the secondary circuit. The higher clearance of watersoluble markers with Prometheusw may be attributed to the fact that this device places a high flux dialyzer directly into the bloodstream, while in MARSw, water-soluble substances partly bypass the low flux dialyzer, which is placed in the secondary plasma circuit. The differences in treatment efficiency resulted in higher delivered treatment dose by FPS than by AD as expressed by the reduction ratio of albumin-bound as well as hydrophilic markers. Mere evaluation of the patients’ blood concentrations would not have identified these differences between the two devices. Reduction ratios are easy to determine and should serve as a first approach to measuring the delivered dose of ELS. It must, however, be borne in mind that this concept also has its limitations, which may lead to either under- or overestimation of treatment dose. In vivo, multiple determinants in addition to removal rate influence the observable reduction of a substrate, including generation rates and transfer rates between multiple pools. Our study demonstrates this with the fact that with FPS, the reduction ratio for ammonia is clearly lower than for urea (40 vs. 59%), despite comparable clearance (141 vs. 147 mL/min). A higher generation rate in combination with rapid refilling from the extravascular compartments could explain this phenomenon. That is why for ammonia, a reduction ratio will considerably underestimate the treatment effect and probably limits its analytic usefulness. The urea reduction ratio could serve as an alternative and be used in combination with bilirubin as measure of treatment dose in ELS. Further, determination of a reduction ratio does not take into account possible post-treatment rebound of solutes, which would lead to an overestimation of delivered dose. In vivo, the phenomenon of rebound occurs almost universally, because substances usually are distributed in more than a single pool. In fact, this also seems to be the case for bilirubin and ammonia [24]. Such effects were of major importance in the accurate measurement of the dose of hemodialysis and have been addressed by urea kinetic modeling [25]. Bilirubin kinetic modeling could serve the same purpose for ELS in the future. Comparison of the technical efficiency of two devices— which was the purpose of this study—is only one application for quantification of ELS. To decide whether treatment intensity really has an impact on clinical outcome, in a first step, different patient groups—e.g. responders vs. non-responders, survivors vs. non-survivors—could be compared for delivered treatment dose. This might serve as the basis for optimization of treatment intensity, length and schedule.
In conclusion, Prometheusw—based on FPS—produced significantly higher clearance for protein-bound and watersoluble markers, which resulted in higher delivered treatment dose compared to a matching treatment with MARSw (AD). We want to emphasize that quantification of ELS is mandatory for accurate clinical research in this emerging field, and reduction ratios of bilirubin, urea and possibly ammonia should be reported in clinical studies on liver dialysis.
Acknowledgements The authors should like to thank G. Bergmann and G.J. Krejs, LKH-Univ. Klinikum Graz, for establishing an interdisciplinary study group on ELS in our hospital. Special thanks to the nursing staff of our dialysis unit for their dedicated patient care, to U. Leitner, C. Berger, S. Po¨tz and U. Spitzer for their painstaking sample collection and to W. Erwa for the analysis of plasma samples. We also thank A. Krause and E. Lamont for their assistance in preparing the manuscript. This study was supported in part by Fresenius Medical Care.
References [1] Abouna GM, Garry R, Hull C, Kirkley J, Walder DN. Pig-liver perfusion in hepatic coma. Lancet 1968;2:509–510. [2] O’Grady JG, Gimson AE, O’Brien CJ, Pucknell A, Hughes RD, Williams R. Controlled trials of charcoal hemoperfusion and prognostic factors in fulminant hepatic failure. Gastroenterology 1988;94:1186–1192. [3] Ash SR, Blake DE, Carr DJ, Carter C, Howard T, Makowka L. Clinical effects of a sorbent suspension dialysis system in treatment of hepatic coma (the BioLogic-DT). Int J Artif Organs 1992;15: 151–161. [4] Sauer IM, Zeilinger K, Pless G, Kardassis D, Theruvath T, Pascher A, et al. Extracorporeal liver support based on primary human liver cells and albumin dialysis—treatment of a patient with primary graft nonfunction. J Hepatol 2003;39:649–653. [5] Demetriou AA, Brown Jr RS, Busuttil RW, Fair J, McGuire BM, Rosenthal P, et al. Prospective, randomized, multicenter, controlled trial of a bioartificial liver in treating acute liver failure. Ann Surg 2004;239:660–667. [6] Stange J, Mitzner SR, Risler T, Erley CM, Lauchart W, Goehl H, et al. Molecular adsorbent recycling system (MARS): clinical results of a new membrane-based blood purification system for bioartificial liver support. Artif Organs 1999;23:319–330. [7] Rifai K, Ernst T, Kretschmer U, Bahr MJ, Schneider A, Hafer C, et al. Prometheus—a new extracorporeal system for the treatment of liver failure. J Hepatol 2003;39:984–990. [8] Mitzner SR, Stange J, Klammt S, Risler T, Erley CM, Bader BD, et al. Improvement of hepatorenal syndrome with extracorporeal albumin dialysis MARS: results of a prospective, randomized, controlled clinical trial. Liver Transpl 2000;6:277–286. [9] Sen S, Felldin M, Steiner C, Larsson B, Gillett GT, Olausson M, et al. Albumin dialysis and molecular adsorbents recirculating system (MARS) for acute Wilson’s disease. Liver Transpl 2002;8:962–967.
P. Krisper et al. / Journal of Hepatology 43 (2005) 451–457 [10] Heemann U, Treichel U, Loock J, Philipp T, Gerken G, Malago M, et al. Albumin dialysis in cirrhosis with superimposed acute liver injury: a prospective, controlled study. Hepatology 2002;36:949–958. [11] Jalan R, Sen S, Steiner C, Kapoor D, Alisa A, Williams R. Extracorporeal liver support with molecular adsorbents recirculating system in patients with severe acute alcoholic hepatitis. J Hepatol 2003;38:24–31. [12] Kramer L, Bauer E, Schenk P, Steininger R, Vigl M, Mallek R. Successful treatment of refractory cerebral oedema in ecstasy/cocaine-induced fulminant hepatic failure using a new high-efficacy liver detoxification device (FPSA-Prometheus). Wien Klin Wochenschr 2003;115:599–603. [13] Liu JP, Gluud LL, Als-Nielsen B, Gluud C. Artificial and bioartificial liver support systems for liver failure. Cochrane Database Syst Rev 2004;CD03628. [14] Jalan R, Sen S, Williams R. Prospects for extracorporeal liver support. Gut 2004;53:890–898. [15] Gotch FA, Sargent JA. A mechanistic analysis of the national cooperative dialysis study (NCDS). Kidney Int 1985;28:526–534. [16] Ronco C, Bellomo R, Homel P, Brendolan A, Dan M, Piccinni P, et al. Effects of different doses in continuous veno-venous haemofiltration on outcomes of acute renal failure: a prospective randomised trial. Lancet 2000;356:26–30. [17] National Kidney Foundation. K/DOQI clinical practice guidelines for hemodialysis adequacy, 2000. Am J Kidney Dis 2001;37: S7–S64.
457
[18] Klammt S, Stange J, Mitzner SR, Peszynski P, Peters E, Liebe S. Extracorporeal liver support by recirculating albumin dialysis: analysing the effect of the first clinically used generation of the MARSystem. Liver 2002;22:30–34. [19] McIntyre CW, Fluck RJ, Freeman JG, Lambie SH. Characterization of treatment dose delivered by albumin dialysis in the treatment of acute renal failure associated with severe hepatic dysfunction. Clin Nephrol 2002;58:376–383. [20] Vaquero J, Chung C, Cahill ME, Blei AT. Pathogenesis of hepatic encephalopathy in acute liver failure. Semin Liver Dis 2003;23: 259–269. [21] Falkenhagen D, Strobl W, Vogt G, Schrefl A, Linsberger I, Gerner FJ, et al. Fractionated plasma separation and adsorption system: a novel system for blood purification to remove albumin-bound substances. Artif Organs 1999;23:81–86. [22] Krause A. Prometheus—a new extracorporeal liver support therapy. In: Arroyo V, Forns X, Garcia-Paga´n JC, Rode´s J, editors. Progress in the treatment of liver diseases. Barcelona: Ars Medica; 2003. p. 437–443. [23] Sauer IM, Goetz M, Steffen I, Walter G, Kehr DC, Schwartlander R, et al. In vitro comparison of the molecular recirculation system (MARS) and single-pass albumin dialysis (SPAD). Hepatology 2004; 39:1408–1414. [24] Jung A, Krisper P, Haditsch B, Stauber R, Holzer H, Schneditz D. Bilrubin kinetics in extracorporeal liver support. Int J Artif Organs 2004;27:588 abstract. [25] Schneditz D, Daugirdas JT. Compartment effects in hemodialysis. Semin Dial 2001;14:271–277.