Argatroban for Anticoagulation in Continuous Renal Replacement Therapy*

Argatroban for anticoagulation in continuous renal replacement therapy*

Andreas Link, MD; Matthias Girndt, MD; Simina Selejan, MD; Alexander Mathes, MD; Michael Bo¨hm, MD; Hauke Rensing, MD

Objective: Argatroban, a direct thrombin inhibitor, was evalu- during CRRT could be determined: a loading dose of 100 ␮g/kg ated for anticoagulation in continuous renal replacement therapy followed by a maintenance infusion rate (␮g/kg/min), which (CRRT) in critically ill patients with heparin-induced thrombocy- can be calculated from the scores as follows: for APACHE II: 2.15– ؊.81, p < 0.001); for SAPS II: 2.06–0.03 ؋؍ topenia type II and acute renal failure. The investigation focused 0.06 ؋ APACHE II (r ؊.8, p < 0.001); and for ICG-PDR: ؊0.35 ؉ 0.08 ؋؍ on predictors for the maintenance doses of argatroban with SAPS II (r -p < 0.001). The efficacy and safety of anticoag ,89. ؍ efficacy and safety of argatroban being secondary outcomes. ICG-PDR (r Design: Prospective, dose finding study. ulation during CRRT were determined by the steady state of blood Setting: Two intensive care units (medical and surgical) of a urea nitrogen (32.16 ؎ 18.02 mg/dL), mean filter patency at 24 hrs university hospital. (98%), and the rate of bleeding episodes. Only two patients devel- .with acute or oped minor bleeding; no patient developed severe bleeding episodes (30 ؍ Patients: Medical and surgical patients (n histories of heparin-induced thrombocytopenia type II and acute Conclusion: In critically ill patients with heparin-induced renal failure with necessity for CRRT. thrombocytopenia type II and necessity for CRRT critical illness Intervention: CRRT with argatroban for anticoagulation. scores (APACHE II, SAPS II) or ICG-PDR can help to predict the Measurements and Main Results: Critical illness severity required argatroban maintenance dose for anticoagulation. These scores Acute Physiology and Chronic Health Evaluation (APACHE)- predictors identify decreased argatroban dosing requirements II, Simplified Acute Physiology Score (SAPS) II, and the indocya- resulting in effective and safe CRRT. (Crit Care Med 2009; 37: nine green plasma disappearance rate (ICG-PDR) were correlated 105–110) to the argatroban maintenance doses. These diagnostic tools can KEY WORDS: continuous renal replacement therapy; heparin- help to identify patients with the necessity for decreased argatro- induced thrombocytopenia II syndrome; argatroban; multiple ban doses. The following recommendations for argatroban dosing organ failure

ontinuous renal replacement inhibitors. Although danaparoid has on argatroban under hemodialysis in- therapy (CRRT) is proven ef- been used in HIT-II for a long time sev- clude case reports and small studies only fective for critically ill pa- eral major drawbacks have been reported. in patients on intermittent hemodialysis tients with (acute) renal fail- Cross-reactivity with heparin/platelet fac- (12–14) or in patients after cardiovascu- ureC (1, 2). In patients with heparin- lar surgery with cardiopulmonary by- tor 4 antibodies in 15% of cases (5), pro- induced thrombocytopenia type II longation of half-life in renal failure up to passes (15). (HIT-II) requiring CRRT effective anti- 36–48 hours (6), and an increased risk In our study, critically ill patients with coagulation, it is necessary to avoid for bleeding events in 46% (7) have been HIT-II-syndromes or histories of HIT-II thrombembolic complications (3, 4). described. Direct thrombin inhibitors, in- and necessity for CRRT were assigned to The options for anticoagulant treat- cluding argatroban and lepirudin, differ an argatroban treatment. The investiga- ment in critically ill patients with in their mode of elimination. The major tion focused on predictors for the main- HIT-II requiring CRRT are limited and route of elimination of lepirudin is renal tenance doses of argatroban with efficacy include danaparoid and direct thrombin clearance. In anuria prolonged half-lives and safety of argatroban being secondary up to 50 hrs have been reported (8). Fur- outcomes. thermore, after 5 days 44% of the pa- *See also p. 342. tients developed lepirudin antibodies that METHODS From the Klinik fu¨r Innere Medizin III (AL, SS, MB,); decrease renal clearance, increase activ- Patients and Registration Klinik fu¨r Innere Medizin IV (MG); Klinik fu¨r Ana¨sthesi- ity, and thus, augment the risk of bleed- ologie und Intensivmedizin (AM, HR), Universita¨tsklini- With approval of the Institutional Review ing (9) contributing to major bleeding kum des Saarlandes; D-66421Homburg/Saar, Board, we evaluated all critically ill patients Germany. complications (10, 11). with necessity for CRRT and clinical signs or The authors have not disclosed any potential con- In contrast, argatroban is mainly elim- history of HIT-II syndromes on the medical flicts of interest. inated by hepatic metabolism. As such, For information regarding this article, E-mail: and surgical intensive care units of the Uni- [email protected] argatroban may be the more appropriate versity Hospital of the Saarland from January Copyright © 2008 by the Society of Critical Care direct thrombin inhibitor in critically ill 2006 to June 2007. HIT-II was defined clini- Medicine and Lippincott Williams & Wilkins patients with renal impairment and with cally as an unexplained decrease of more than DOI: 10.1097/CCM.0b013e3181932394 the necessity for CRRT. Prospective data 50% in platelet count from the patients base-

Crit Care Med 2009 Vol. 37, No. 1 105 line, or a platelet count lower than 100 ϫ Management and Monitoring of many) was injected central venously (0.5 mg/ 109/L while on heparin therapy, or by positive Anticoagulation kg). ICG is nearly exclusively eliminated in an local laboratory tests. HIT-II screening tests unconjugated form by the liver into the bile were performed at the University Hospital At the beginning of CRRT, all patients re- and does not undergo enterohepatic recircu- Laboratory using the particle gel immunoas- ceived argatroban at a dose of 100 ␮g/kg as a lation. Its elimination may be expressed as say (ID-HPF-4, DiaMed, Switzerland) for rapid bolus injection followed by a continuous infu- plasma disappearance rate (ICG-PDR) and can detection and the enzyme-linked immunosor- sion of 1 ␮g/kg/min. Argatroban was applied be measured noninvasively at the bedside by bent assay for discovering antibodies (IgG, into the extracorporeal circulation (prefilter transcutaneous pulse densitometry with a IgA, IgM) to heparin-platelet factor 4 com- injection/infusion). The monitoring of antico- finger clip (LiMON monitoring, Pulsion Med- plexes. Both HIT-II tests were done for all agulation was performed by repetitive mea- ical Systems, Germany). Normal values for patients. Critical illness was defined by com- surements of the systemic activated partial ICG-PDR are considered to be over 18%/min. monly used scores in intensive care medicine: thromboplastin time (aPTT) according to In general, ICG removal from the blood de- Acute Physiology and Chronic Health Evalua- standard methods. The anticoagulation regi- pends on liver blood flow, parenchymal cellu- tion II (APACHE-II, 16) and Simplified Acute men was designed to result in a 1.5- to 3-fold lar function, and biliary excretion (18, 19). Physiology Score II (SAPS-II, 17). Thus, crit- elevation of aPTT. The first measurement of ical illness and multiple organ dysfunction aPTT was done 30 minutes after starting syndromes were defined as a minimum CRRT, followed by further controls every 60 Statistical Analysis SAPS-II score of 30 points. minutes until a 1.5- to 3-fold elevated aPTT Demographical data and data on efficacy was reached. If aPTT was below the therapeu- and safety were summarized and descriptive tic range, the argatroban dose was increased statistics were computed (means Ϯ SD). For all Inclusion and Exclusion Criteria stepwise (in steps of 0.25 ␮g/kg/min). If aPTT patients the individual argatroban mainte- was more than 3-fold elevated, argatroban was Critically ill patients with manifestation or nance infusion rates were correlated to the discontinued for 2 consecutive hours and step- history of HIT-II and necessity for CRRT were initial cholinesterase levels, ICG-PDR, SAPS wise reduced (0.25 ␮g/kg/min). When the he- treated with argatroban after exclusion of clin- II, and APACHE II score (Pearson). Further- mofilter systems were routinely changed, the ically relevant bleeding signs, transient more a regression analysis was performed. A argatroban infusion was continued and no fur- thrombocytopenia, disseminated intravasal ␣-error of less than 0.05 was considered sta- ther boluses were given. In all surgical cases, coagulopathy, and preexisting chronic hepatic tistically significant. dysfunction (Child-Pugh C). Concomitant CRRT with argatroban started 24–48 hrs after treatment was performed according to best the surgical procedure. medical practice. The decision to anticoagu- RESULTS late with argatroban was made by the physi- Assessments for Anticoagulation, cian in charge according to clinical needs and Efficacy, and Safety Within 18 months, 34 critically ill pa- the inclusion and exclusion criteria. tients with clinical signs or history of Anticoagulation. For argatroban dose ti- HIT-II and necessity for CRRT were stud- tration, plasma aPTT was measured before, 30 ied. The incidence for the acute manifes- Continuous Renal Replacement minutes after starting CRRT, and every hour tation or history of HIT-II of our critically Therapy until a 1.5- to 3-fold aPTT was reached. In ill patients on medical and surgical inten- cases of a steady state, aPTT was measured sive care units during the study period Continuous venovenous hemodialysis was twice daily. All hemodialysis tubes and filters performed with a polysulfone high-flux hemo- were visually inspected for blood flow and ev- was 1.3%. A manifest HIT-II syndrome dialyzer using a roller-pump-driven hemofil- idence of clot formation. was diagnosed in 28 cases; a history of tration device manufactured by Fresenius AG Hemodialysis Efficacy. The hemodialysis HIT-II in 2 cases. These findings were (ADM 08; Bad Homburg, Germany) with a efficacy was assessed by the average fall in confirmed by laboratory tests in 26 cases, bicarbonate dialysate at a flow rate of 2000 blood urea nitrogen to steady state, which indicating a specificity of 86%. mL/hr. The blood flow was 100 mL/min. The occurred after a mean of 3.8 days after the Treatment with argatroban was per- ultrafiltration rate depended on clinical re- beginning of CRRT. formed on 30 patients. Four patients quirements (0–200 mL/hr). The hemodialysis Safety. Patients underwent physical exam- were excluded from the anticoagulation tubes and filters were renewed every 24 hrs. ination, hemodynamic measurements (blood regimen with argatroban because of pre- pressure, heart rate) and clinical laboratory existing chronic hepatic dysfunction (n ϭ tests (hemoglobin, hematocrit, sodium, potas- Argatroban Pharmacokinetics sium, platelets, red and white blood cells, 3) and disseminated intravascular coagu- ϭ urea, creatinine) daily on the intensive care lation (n 1) (Fig. 1). In those cases Argatroban (molecular weight 526.66 dal- units. Bleeding episodes were defined as actual lepirudin was used for anticoagulation. tone) is a rapidly acting direct thrombin in- bleeding noted. They were classified as minor None of the included patients had been hibitor with moderate protein binding (54%), when argatroban infusion had to be paused to diagnosed with acute or chronic hepatic primarily to albumin (20%) and ␣1-acid gly- restore coagulation and as major when trans- disease before onset of critical illness. Pa- coprotein (34%). Its short elimation half-life fusion of blood or coagulation factors became tient characteristics, diagnosis underly- of 35–51 mins and major route of hepatic necessary. ing critical illness, and acute renal failure metabolism (70%) are important aspects for are shown in Table 1. use of argatroban in patients with renal fail- Assessments for Hepatic ure; ϳ14% of the unchanged drug is elimi- Function nated via feces and 16% is eliminated renally. Argatroban Infusion Rate and During hemodialysis, the dialytic clearance of Transaminases, total bilirubin, albumin, Anticoagulation argatroban was 0.3–0.7 mL/min/kg, suggest- and cholinesterase were measured. For dy- ing an increase of ϳ20% in argatroban clear- namic liver function, indocyanine green (ICG- Patients received an initial argatroban ance (15). PULSION, Pulsion Medical Systems, Ger- bolus of 100 ␮g/kg bolus at the start of

106 Crit Care Med 2009 Vol. 37, No. 1 Table 1. Baseline characteristics of argatroban treated patients

Argatroban in Continuous Characteristic Renal Replacement Therapy

n 30 Age (yrs) 67.1 Ϯ 14.2 men/women (n/n) 15/15 HIT-II diagnosis By history 2 By clinical signs 28 By laboratory test 26 Reasons for critical illness and acute renal failure Septic shock 9 Cardiogenic shock 10 Acute decompensation of a chronic renal failure 3 Acute ischemia (limb ischemia; mesenteric infarction) 2 Polytrauma 1 “open heart” surgery 5 Critical illness prognostic scores Acute Physiology and Chronic Health Evaluation II 26 (18; 34) Acute Physiology Score II 45 (30; 66) Dialysis efﬁcacy Activated partial thromboplastin time (sec) 82 Ϯ 16 CRRT duration (hr) 132 Ϯ 122 BUN Pretreatment BUN (mg/dL) 55.1 Ϯ 25.9 Steady-state BUN during continuous renal replacement 32.2 Ϯ 18.0 therapy (mg/dL) Figure 1. Study ﬂow chart for registered critically Filter patency at 24 hrs (%) 98 ill patients with acute manifestation or history of Liver function heparin-induced thrombocytopenia type-II (HIT) CHE*103 U/L 2.8 Ϯ 1.1 syndrome and acute renal failure with necessity Albumin (g/L) 25.5 Ϯ 6.3 for continuous renal replacement therapy. Thirty Bilirubin (mg/dL) 2.8 Ϯ 5.9 Ϯ patients were included and argatroban therapy Indocyanine green plasma disappearance rate (%/min) 14.6 3.9 Bleeding signs and thrombembolic complications was initiated; four patients were excluded. ICU, Major bleedings (patients, n) 0 intensive care unit; i.v., intravenous; aPTT, acti- Minor bleedings (patients, n) 2 vated partial thromboplastin time. Thrombembolic complications (patients, n) 1

Age, gender, reasons for critical illness and acute renal failure, critical illness prognostic scores CRRT followed by continuous infusion of (Acute Physiology and Chronic Health Evaluation II, Simplified Acute Physiology Score II), dialysis efficacy, liver function test (e.g., indocyanine green plasma disappearance rate), bleeding signs and 1 ␮g/kg/min. The bolus administration thrombembolic complications. Parameters are given as mean Ϯ SD respectively range (prognostic resulted in a peak response of anticoagu- score). lation with aPTT Ͼ140 seconds within 30 HIT-II, heparin-induced thrombocytopenia; CCRT, continuous renal replacement therapy; CHE, minutes without registration of any cholinesterase. bleeding signs. The following argatroban ␮ infusion rate of 1 g/kg/min resulted in ase activity (r ϭ .53, not significant, n ϭ an effective anticoagulation with 2- to Argatroban Infusion Rate in Critically Ill Patients 30, Fig. 2). At low argatroban dosages (0.1 3-fold elevated aPTT within 60 minutes. ␮g/kg/min), patients had a wide scatter of In 22 cases the infusion rate had to be Although none of our critically ill pa- albumin and cholinesterase levels. There- ␮ reduced (average rate 0.7 g/kg/min; tients included for argatroban anticoagu- fore, the adequate argatroban dose could ␮ range, 0.1–1.5 g/kg/min). In 5 cases the lation had a history of liver disease the not be predicted from these parameters. infusion rate was increased up to 1.2–1.5 clinical chemistry tests for hepatic func- The present data show that patients with ␮ g/kg/min. The median duration before tion were changed in many individuals. reduced clinical chemistry markers of patients reached their target aPTT was 5 Laboratory parameters for hepatic de- liver synthesis (albumin, cholinesterase) hours (range 0–14 hours). After achiev- struction such as aspartate aminotrans- required up to 90% less argatroban infu- ing therapeutically effective levels of an- terase and alanine aminotransterase had sion rate compared to the manufacturer’s ticoagulation (aPTT 70–90 seconds) the a wide range (318 U/L; minimum 10, recommendations. Using dynamic he- infusion rate of the continuous argatro- maximum 2168 U/L). Parameters for hepatic tests of liver clearance activity like ban application remained unchanged patic function of synthesis, e.g., cholines- ICG-PDR we measured a significant cor- during the whole treatment period. There terase ranged from normal to decreased relation to the effective argatroban was no evidence of loss of the argatroban levels (2.86 ϫ 103 U/L; minimum 0.87, maintenance infusion rate (r ϭ .89, p Ͻ effect during the whole CRRT period. Five maximum 5.48 ϫ 103 U/L). The mainte- 0.001, n ϭ 23, regression analysis: ar- patients with a CRRT were treated for more nance argatroban dose showed no statis- gatroban infusion rate [␮g/kg/min] ϭ than 10 days. In these cases no loss of effect tically significant correlation to parame- Ϫ0.35 ϩ 0.08 ϫ ICG-PDR, Fig. 3). In- of argatroban was observed and no increase ters of hepatic synthetic function, like dividuals with the need for low argatro- of the infusion rate was necessary. albumin concentration and cholinester- ban maintenance doses could not be

Crit Care Med 2009 Vol. 37, No. 1 107 analysis. We found a highly significant correlation between the continuous argatroban dose and APACHE II score (r ϭϪ.81, p Ͻ 0.001, n ϭ 30, regression analysis: argatroban infusion rate [␮g/kg/min] ϭ 2.15 Ϫ 0.06 ϫ APACHE II, Fig. 4) and SAPS II score (r ϭϪ.8, p Ͻ 0.001, n ϭ 30, regression analysis: argatroban infusion rate [␮g/kg/min] ϭ 2.06 Ϫ 0.03 ϫ SAPS II, Fig. 5). In patients with elevated APACHE II or SAPS II scores, argatroban infusion rates had to be considerably reduced compared with the manufacturer’s recommendations. Figure 2. Correlation between serum-cholinesterase and argatroban infusion rate during continuous renal replacement therapy in critically ill patients (r ϭ .53, n.s., n ϭ 30). N.S., not significant. Argatroban Efficacy and Safety During CRRT CRRT with argatroban was performed effectively, as evidenced by the average fall in blood urea nitrogen to steady state (32.16 Ϯ 18.02 mg/dL). The hemodialysis tubes and filters were renewed every 24 hours. In three cases a 24-hr period was interrupted because of the appearance of fibrin strands, early sludging or throm- bus material in the drip chamber of the dialyzing system. A cumulative number of 165 dialyzing systems were used, and only 2% of the systems had to be changed earlier than the 24-hr standard period. Two patients developed minor bleeding; argatroban infusion was discontin- Figure 3. Correlation between indocyanine green plasma disappearance rate (ICG-PDR) and argatroban infusion rate during continuous renal replacement therapy in critically ill patients (r ϭ .89, p Ͻ 0.001, n ϭ ued until restoration of coagulation. No 23). Regression analysis: Argatroban infusion rate [␮g/kg/min] ϭϪ0.35 ϩ 0.08 ϫ ICG-PDR. patient developed severe bleeding episodes or required transfusion. One thrombembolic complication during evaluation of the diagnosis and starting of the CRRT occurred. No new thromboem- bolic complications were detected during treatment with argatroban.

DISCUSSION The anticoagulation of critically ill patients with HIT-II requiring CRRT is limited on danaparoid and the direct thrombin inhibitor lepirudin (3, 4) with high risks for bleeding events caused by drug accumulation because of their renal clearance. The new direct thrombin in- Figure 4. Correlation between Acute Physiology and Chronic Health Evaluation (APACHE) II score and hibitor argatroban is mainly cleared he- argatroban infusion rate during continuous renal replacement therapy in critically ill patients (r ϭ patically. Its recommendations are lim- Ϫ.81, p Ͻ 0.001; n ϭ 30). Regression analysis: Argatroban infusion rate [␮g/kg/min] ϭ 2.15–0.06 ϫ ited on patients with hemodynamic APACHE II. stability, the absence of chronic liver disease (Child-Pugh score Ͼ6), and in cases identiﬁed reliably by cholinesterase lev- Since scores such as APACHE II or of renal failure on an intermittent renal els since some of those with normal or SAPS II are validated for severity and replacement therapy (14). slightly reduced cholinesterase had im- mortality of critical illness, we com- In critically ill patients, hepatic clear- paired ICG clearances and low argatro- pared these scores to the maintenance ance is often impaired by shock-induced ban requirements. argatroban infusion rate by correlation reduced liver perfusion. Thus, a de-

108 Crit Care Med 2009 Vol. 37, No. 1 There is no specific antidote for direct thrombin inhibitors; however, effect measures typically return to baseline within 4–6 hrs after discontinuation of the argatroban therapy, and it has been suggested that recombinant factor VIIa may be a useful pharmacologic agent for reversing severe bleeding (29). The observed safety and efficacy of argartoban were achieved by targeting a high PTT range, daily dialyzer changes, low blood flow rates, and low filtratrion fractions. However, this first report on argatroban anticoagulation in critically ill patients with HIT and CRRT sets the stage for a Figure 5. Correlation between Simplified Acute Physiology Score (SAPS) II and argatroban infusion larger trial in these complex medical con- rate during continuous renal replacement therapy in critically ill patients (r ϭϪ.8, p Ͻ 0.001, n ϭ ditions. 30). Regression analysis: Argatroban infusion rate [␮g/kg/min] ϭ 2.06–0.03 ϫ SAPS II. In conclusion, this study is important for several reasons. First, the initial SAPS II evaluation and/or ICG-PDR measure- creased hepatic clearance is associated maintenance doses and ICG-PDR. They ment are suitable tools for the prediction with an increased elimation half-life of have investigated a population of criti- of the required argatroban maintenance argatroban. In critically ill patients with cally ill patients in whom almost 45% did dose in critically ill patients with HIT-II multiple organ dysfunction syndromes a not require CRRT whereas in our popu- syndromes and necessity for CRRT. Sec- dose reduced ten-fold (0.2 ␮g/kg/min) is lation all patients received CRRT. There- ond, the anticoagulation with argatroban reported to be sufficient and safe for ef- fore, hepatic clearance appears to be ef- is efficient and safe, but dose reductions fective anticoagulation (20–23). To eval- fective to estimate argatroban dosages in down to 0.1–1.0 ␮g/kg/min are required. uate reliable predictors for effective ar- patients with completely extinct kidney Third, argatroban appears to be a suitable gatroban doses in critically ill patients function. Nevertheless, ICG-PDR has not anticoagulant in critically ill patients “static” and “dynamic” hepatic tests and reached a widespread clinical application. with HIT syndrome undergoing CRRT. critical illness prognostic scores We therefore tested the relation between (APACHE II, SAPS II) were correlated to the severity of critical illness scores, ACKNOWLEDGMENTS the argatroban maintenance infusion APACHE II (16) and SAPS II (17), and the rate. Monitoring of liver enzymes and he- argatroban maintenance infusion rate. We are indebted to Project Director patic protein synthesis have only limited For both scores, a strong correlation was Dr. S. Graeber (Institute for Medical Bi- diagnostic, therapeutic, and prognostic observed. In critically ill patients with ometry, Epidemiology and Medical Infor- value in hepatic dysfunction (24, 25). We necessity for CRRT our argatroban dosing matics, University of the Saarland, Ger- found no statistically significant correla- recommendations can be summarized as many) for statistical support and tion between the argatroban infusion rate follows: after a loading dose of 100 ␮g/kg, reviewing the manuscript. and cholinesterase. Our data show that the maintenance infusion rate may be patients with reduced hepatic protein calculated from the current ICG-PDR synthesis required up to 90% decrease of value, APACHE II value, or SAPS II value REFERENCES the argatroban infusion rate, but normal based on regression analysis. The use of hepatic protein synthesis does not ex- these regression equations could result in 1. Buchardi H: History and development of clude the need for a dose reduction. a more rapid target aPTT attainment and continuous renal replacement techniques. The dynamic hepatic clearance test lower cases of argatroban overdose. Kidney Int Suppl 1998; 66:S120–S124 ICG-PDR is currently a valuable tool for Efficacy of hemodialysis was demon- 2. Bellomo R, Ronco C: Continuous renal replacement therapy in intensive care unit. In- assessment of liver function in critically strated by an adequate decrease in blood tensive Care Med 1999; 25:781–789 ill patients and correlates well with the urea nitrogen and the high mean filter 3. Keeling D, Davidson S, Watson H: The man- outcomes in critcally ill patients (18, 26, patency at 24 hours (98%). Only minor agement of heparin-induced thrombocytope- 27). It has the advantage of noninvasive bleedings were observed in only 6.6% of nia. Br J Haematol 2006; 133:259–269 testing at the bedside by a transcutaneous the patients. This bleeding rate was com- 4. Oudemans-van Straaten MH, Wester JPJ, de system and providing results within a few parable to rates for argatroban-treated Pont ACJM, et al: Anticoagulation strategies minutes. We measured a significant cor- patients overall from the multicenter ret- in continuous renal replacement therapy: relation between the effective argatroban rospective studies (28). Can the choice be evidence based? Intensive infusion rate and ICG-PDR. Our data sug- One might argue that the study is lim- Care Med 2006; 32:188–202 gest that ICG-PDR can help to identify ited by the small sample size and missing 5. Magnani HN: Heparin-induced thrombocytopenia (HIT): An overview of 230 patients those patients who have decreased ar- measurements of argatroban concentra- treated with Orgaran (Org 10172). Thromb gatroban dosing requirements, presum- tions, its clearance and metabolites. As Haemost 1993; 70:554–561 ably related to occult hepatic dysfunc- described recently the dialytic clearance 6. Polkinghorne KR, McMahon LP, Becker GJ: tion. Beiderlinden et al (23) did not of argatroban with high-flux membranes Pharmocokinetic studies of dalteparin (Frag- observe a correlation between argatroban is clinically insignificant (ϳ20%) (15). min), enoxaparin (Clexane), and danaparoid

Crit Care Med 2009 Vol. 37, No. 1 109 sodium (Orgaran) in stable hemodialysis pa- tion in a patient requiring haemodialysis. 21. Williamson DR, Boulanger I, Tardif M, et al: tients. Am J Kidney Dis 2002; 40:990–995 Nephrol Dial Transplant 1995; 10:2137–2140 Argatroban dosing in intensive care patients 7. Lindhoff-Last E, Betz C, Bauersachs R: Use of 14. Murray PT, Reddy BV, Grossman EJ, et al: A with acute renal failure and liver dysfunc- a low-molecular-weight heparinoid (danap- prospective comparison of three argatroban tion. Pharmacotherapy 2004; 24:409–414 aroid sodium) for continuous renal replace- treatment regimes during hemodialysis in 22. Levine RL, Hurstling MJ, McCollum D: Ar- ment therapy in intensive care patients. Clin endstage renal disease. Kidney Int 2004; 66: gatroban therapy in heparin-induced throm- Appl Thromb Hemost 2001; 7:300–304 2446–2453 bocytopenia with hepatic dysfunction. Chest 8. Fischer KG: Hirudin in renal insufficiency. 15. Koster A, Buz S, Hetzer R, et al: Anticoagu- 2006; 129:1167–1175 Semin Thromb Hemost 2002; 28:467–482 lation with argatroban in patients with hep- 23. Beiderlinden M, Treschan TA, Go¨rlinger K, et 9. Eichler P, Friesen HJ, Lubenow N, et al: arin-induced thrombocytopenia antibodies al: Argatroban anticoagulation in critically ill Antihirudin antibodies in patients with hep- after cardiovascular surgery with cardiopul- patients. Ann Pharmacother 2007; 41: arin-induced thrombocytopenia treated with monary bypass: First results from the ARG- 749–754 lepirudin: Incidence, effects on aPTT, and E03 trial. J Thorac Cardiovasc Surg 2006; 24. Reichling JJ, Kaplan MM: Clinical use of se- 132:699–700 clinical relevance. Blood 2000; 96:2373–2378 rum enzymes in liver disease. Dig Dis Sci 16. Knaus W, Draper E, Wagner D, et al: APACHE 10. Kern H, Ziemer S, Kox WJ: Bleeding after 1988; 33:1601–1614 II: A severity of disease classification system. intermittent or continuous r-hirudin during 25. Shakil AO, Kramer D, Mazariegos GV, et al: Crit Care Med 1985; 13:818–832 CVVH. Intensive Care Med 1999; 25: Acute liver failure: Clinical features, outcome 17. Le Gall JR, Lemeshow S, Saulnier F: A new 1311–1314 analysis, and applicability of prognostic cri- simplified acute physiology score (SAPS II) 11. Muhl E, Siemens HJ, Kujath P, et al: Therapy teria. Liver Transpl 2000; 6:163–169 based on a European/North American multi- and monitoring of heparin-induced throm- center study. JAMA 1993; 270:2957–2963 26. Sakka SG: Assessing liver function. Curr bocytopenia type II in critically ill patients 18. Kimura S, Yoshioka T, Shibuya M, et al: Opin Crit Care 2007; 13:207–214 during continuous venovenous hemodiafil- Indocyanine green elimination rate detects 27. Sakka SG, Reinhart K, Meier-Hellmann A: tration: Comparison of aPTT and ecarin clot- hepatocellular dysfunction early in septic Prognostic value of the indocyanine green ting time for monitoring of r-hirudin ther- shock and correlates with survival. Crit Care plasma disappearance rate in critically ill pa- apy. J Intensive Care 2002; 17:34–40 Med 2001; 29:1159–1163 tients. Chest 2002; 122: 1715–1720 12. Matsuo T, Yamada T, Yamanashi R, et al: An- 19. Sakka SG, Reinhart K, Meier-Hellmann A: 28. Lewis BE, Wallis DE, Berkowitz SD, et al: ticoagulant therapy with MD805 of a hemodi- Prognostic value of the indocyanine green Argatroban anticoagulant therapy in patients alysis patient with heparin-induced thrombo- plasma disappearance rate in critically ill pa- with heparin-induced thrombocytopenia. cytopenia. Thromb Res 1990; 58:663–666 tients. Chest 2002; 122:1715–1720 Circulation 2001; 103:1838–1843 13. Koide M, Yamamoto S, Matsuo M, et al: An- 20. Arpino PA, Hallisey RK: Effect of renal func- 29. Alving BM: How I treat heparin-induced ticoagulation for heparin-induced thrombo- tion on the pharmacodynamics of argatro- thrombocytopenia and thrombosis. Blood cytopenia with spontaneous platelet aggrega- ban. Ann Pharmacother 2004; 38:25–29 2003; 101:31–37

110 Crit Care Med 2009 Vol. 37, No. 1