Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Nephrol Dial Transplant (2005) 20: 1889–1897 doi:10.1093/ndt/gfh915 Advance Access publication 31 May 2005 Original Article Prevention of clot formation during haemodialysis using the direct thrombin inhibitor melagatran in patients with chronic uraemia Per-Ola Attman1, Pia Ottosson1, Ola Samuelsson1, Ulf G. Eriksson2, Maria Eriksson-Lepkowska2 and Gunnar Fager2 1 Department of Nephrology, Sahlgrenska University Hospital, Göteborg and 2AstraZeneca R&D, Mölndal, Sweden Abstract Background. This study assessed the feasibility of replacing intravenous (i.v.) dalteparin with the direct thrombin inhibitor (DTI) melagatran administered via dialysis fluid in patients undergoing haemodialysis, and also examined the pharmacokinetics of melagatran with and without dialysis. Methods. During two 4 h haemodialysis sessions, 10 adult patients were administered i.v. dalteparin. During two subsequent sessions, melagatran was administered as an i.v. bolus before dialysis, and in the dialysis fluid. The pharmacokinetics of melagatran administered as a bolus before dialysis, and of i.v. melagatran during a dialysis-free day, were studied. Dialysis performances were evaluated from clinical criteria including clot formation in the dialyzer and bloodlines, pre-post dialyzer pressures and iohexol clearance. Anticoagulant efficacy was evaluated from dialysis success. Results. All dialysis sessions were successful, with no apparent difference in clot formation between the two treatments. Median iohexol clearance was similar with dalteparin (99–103 ml/min) and melagatran in the dialysis fluid (98–100 ml/min). There was no difference in pre- and post-dialyzer bloodline pressures between the two treatments. During dialysis sessions with melagatran in dialysis fluid, melagatran concentrations in plasma rapidly equilibrated to 70% of those in dialysis fluid. While the clearance of melagatran was low in patients with renal failure (mean±SD, 0.93±0.36 l/h), haemodialysis provided efficient clearance of melagatran (7.20±0.76 l/h). Melagatran clearance by dialysis (104±10 ml/min) was comparable to iohexol clearance. Conclusions. The DTI melagatran administered via dialysis fluid may provide sufficient anticoagulation for haemodialysis. Melagatran is rapidly cleared from Correspondence and offprint requests to: Per-Ola Attman, Department of Nephrology, Sahlgrenska University Hospital, SE 41345 Göteborg, Sweden. Email: [email protected] plasma by haemodialysis, suggesting that this method may be used to decrease drug levels in patients with renal impairment. Keywords: anticoagulant; chronic uraemia; haemodialysis; melagatran; thrombosis; ximelagatran Introduction Anticoagulation is a prerequisite for haemodialysis. Currently, individually titrated doses of heparin preparations are used to prevent clotting in the extracorporeal blood circuit. However, chronic administration of various heparins is associated with serious side effects. Annually, about 1% of uraemic patients on chronic haemodialysis develop heparin-induced thrombocytopaenia, which may cause severe bleeding [1,2]. Heparin, whether fractionated or unfractionated, activates lipoprotein lipase and this may contribute to the particular dyslipoproteinaemia and progressive atherosclerosis of chronic renal failure [3–5]. Long-term treatment with heparin may also accelerate osteoporosis and contribute to osteodystrophy seen in dialysis patients [3,6]. Replacement of unfractionated heparin with fractionated (low molecular weight) heparin may at most attenuate, but cannot eliminate, the long-term side effects of repeated heparin administration [3]. Arteriovenous (A-V) fistulas and central venous catheters, used to connect the extracorporeal circuit to the blood circulation, tend to clot and require intermittent acute fibrinolytic therapy or even chronic anticoagulation between dialysis sessions when they clot frequently. In the coagulation cascade, thrombin proteolytically cleaves fibrinogen to generate fibrin, which polymerizes to form the thrombus network. Thrombin is also a potent platelet activator and this may lead to the formation of platelet-rich arterial thrombi. ß The Author [2005]. Published by Oxford University Press on behalf of ERA-EDTA. All rights reserved. For Permissions, please email: [email protected] 1890 Hence, pharmacologic inhibition of thrombin may prevent development of thrombosis and clotting of extracorporeal blood circuits. Melagatran is a novel, low-molecular-weight (430 Da) thrombin inhibitor with predictable pharmacokinetics and pharmacodynamics for parenteral use [7]. Melagatran is the active form of the oral direct thrombin inhibitor ximelagatran. After oral administration, ximelagatran is rapidly absorbed and bioconverted via two intermediates to its active form, melagatran [7,8]. Ximelagatran has been shown to have predictable and stable pharmacokinetics with low intra- and inter-individual variability [9]. In young healthy volunteers, about 80% of systemically available melagatran is excreted as unchanged compound via the kidneys [8] and the clearance of melagatran correlates to the renal function [10]. Melagatran and ximelagatran have been approved in several European countries for the prevention of thrombosis in conjunction with orthopaedic surgery. Potential side-effects of melagatran include those expected from inhibition of thrombin, i.e. bleeding tendency. There is no known antidote to reverse the action of melagatran. Studies in surgical and non-surgical patients have shown that treatment with ximelagatran or melagatran carries a low frequency of bleeding complications not different from that of other anticoagulant regimes, i.e. low molecular weight heparin (LMWH) or vitamin K antagonists [11]. Previous in vitro experiments confirmed that melagatran passes efficiently across dialysis membranes (unpublished observations). These results demonstrate that it should be possible to obtain anticoagulation in patients undergoing dialysis by addition of melagatran to the dialysis fluid and, more specifically, to the electrolyte-containing concentrate. Predictable diffusion across the membrane would rapidly provide suitable steady-state concentrations in these patients during dialysis. In a preliminary study in anaesthetized anuric (by renal arteriovenous ligation) pigs, 1 mmol/l melagatran was provided in the dialysis fluid during haemodialysis for 3 h [12]. The dialysis procedures were successful and there were no visible clots in the saline-perfused tubing and dialyzer after dialysis. Pressures in the dialysis circuit measured before and after the dialyzer together with iohexol clearance remained constant during the dialysis. This suggests that melagatran efficiently prevented clot formation in this model of haemodialysis. The primary objective of this exploratory and feasibility study was to investigate if intravenous anticoagulation can be successfully replaced by melagatran administered in the dialysis fluid for the prevention of clot formation in the extracorporeal circuit during haemodialysis in uraemic patients. A secondary objective was to characterize melagatran pharmacokinetics in uraemic patients undergoing haemodialysis and to assess its elimination during dialysis, as this may potentially be used in the event that reversal P.-O. Attman et al. Table 1. Pre-entry characteristics of 10 patients who completed the study according to protocol Age (years) Body weight (kg) Haemoglobin (g/l) Platelets (109/l) WBC (109/l) P-APTT (s) INR S-albumin (g/l) S-creatinine (mmol/l) Median Range 61 86 123 234 7.8 34 0.9 37 774 41–77 63–115 114–134 164–406 6–11.5 28–42 0.45–1.0 35–42 430–1090 INR, international normalized ratio; P-APTT, plasma activated partial thromboplastin time; S, serum; WBC, white blood cell count. of anticoagulation is necessary in patients with renal insufficiency. Subjects and methods Study design This was an open-label, fixed-sequence observational study in 10 uraemic patients on chronic haemodialysis (Table 1). After inclusion criteria were met (Visit 1), patients underwent two consecutive dialysis sessions (Visits 2 and 3) using their individually titrated standard dose (mean 5625, range 2500–7500 IU) of intravenous (i.v.) dalteparin (FragminÕ ; Pharmacia & Upjohn, Stockholm, Sweden) as anticoagulant immediately before start of dialysis. Between two subsequent dalteparin dialyses at Visit 4, eight patients agreed to receive 1 mg melagatran i.v. for the determination of melagatran clearance without dialysis. At the subsequent pharmacokinetic session (Visit 5), all 10 patients were given their usual dose of dalteparin and in addition an i.v. bolus of 1 mg (2.3 mmol) melagatran (AstraZeneca R&D, Mölndal, Sweden), both drugs given at start of dialysis. During the two last study dialysis sessions (Visits 6 and 7), melagatran was used as the only anticoagulant drug, and was provided both as an i.v. bolus just before start of dialysis to prevent early clotting and via the dialysis fluid to prevent later clot formation during dialysis. On the first occasion (Visit 6), doses were fixed to a 2 mg (4.6 mmol) i.v. bolus and a concentration of 0.5 mmol/l (0.21 mg/l) in the dialysis fluid. On the subsequent occasion (Visit 7), the investigator was allowed to decrease, maintain or increase this bolus dose (1, 2 or 3 mg) of melagatran and/or to decrease, maintain or increase melagatran in the dialysis fluid (0.25, 0.5 or 1.0 mmol/l) depending on early or late bleeding or clotting tendency, or to test if successful dialysis could be achieved also on a lower dose than previously used (Table 2). Patients Patients with chronic renal failure on chronic (>6 weeks) haemodialysis were eligible. Patients with significant bleeding or thrombotic disorder or treatment with drugs known to Melagatran for prevention of clot formation during haemodialysis Table 2. Melagatran pharmacokinetic data at the end of each dialysis session during which melagatran was added to the dialysis fluid Patient i.v. bolus Melagatran concentration dose (mg) Plasma at 4 h Dialysis fluid Plasma at (mmol/l) 4 h (mmol/l) versus dialysis fluid (%) Visit 6: melagatran 0.5 mmol/l in dialysis fluid 2 2 0.5 0.380 3 2 0.5 0.408 4 2 0.5 0.348 7 2 0.5 0.297 10 2 0.5 0.374 11 2 0.5 0.337 1 2 0.5 0.442 5 2 0.5 0.350 9 2 0.5 0.390 8 2 0.5 0.380 76 82 70 59 75 67 88 70 78 76 10 0.371 0.040 0.297 0.377 0.442 11 10 74 8 59 75 88 11 n Mean SD Minimum Median Maximum CV Visit 7: melagatran 0.25, 0.5 or 1.0 mmol/l in dialysis fluid 2 2 1 0.609 61 3 2 1 0.716 72 4 2 1 0.706 71 7 2 1 0.613 61 10 2 1 0.665 67 11 2 1 0.628 63 1 1 0.25 0.226 90 5 1 0.25 0.175 70 9 1 0.25 0.184 74 8 3 0.5 0.417 42 Mean SD Minimum Median Maximum CV ND ND ND ND ND ND 67 12 42 68 90 18 CV, coefficient of variation; ND, not determined because of different dose levels. affect the coagulation system were excluded from the study. Eleven eligible patients (10 men, one woman) were included. One patient had a residual renal function with glomerular filtration rate (GFR) of 4 ml/min. The other patients had only minimal or no residual renal function (24 h urine output <500 ml). Blood access was obtained by A-V fistulas (n ¼ 8) or double-lumen central venous catheters (n ¼ 3). Signed informed consent to participate was obtained from all patients. The study was carried out according to the Declaration of Helsinki and approved by the Swedish Medical Products Agency and the Göteborg Ethics Committee. Ten patients completed the study according to protocol (Table 3). Eight of these patients agreed to participate in the optional pharmacokinetic session between dialyses (Visit 4). One 61-year-old man with past history of rheumatoid arthritis and amyloidosis withdrew from the study after the initial two dialyses with dalteparin and before any administration of melagatran because of septic arthritis in a prosthetic 1891 Table 3. Dialysis iohexol clearance (ml/min) by Visit Patient Ioxehol clearance Visit 2 Visit 3 Visit 6 Visit 7 1 2 3 4 5 7 8 9 10 11 106.3 93.6 106.5 101.9 104.5 91.0 109.5 103.7 102.9 98.1 116.9 89.0 117.0 98.4 98.7 83.8 108.1 105.1 96.3 99.6 119.2 81.5 98.5 81.1 86.7 83.9 104.1 109.4 101.0 111.3 107.4 71.1 77.5 81.3 97.0 98.3 95.0 113.9 103.5 114.0 n Mean SD Minimum Median Maximum 10 101.8 5.9 91.0 103.3 109.5 10 101.3 10.8 83.8 99.2 117.0 10 97.7 13.7 81.1 99.8 119.2 10 95.9 15.0 71.1 97.7 114.0 knee joint. He received long-term antibiotic therapy and was discharged from hospital after 13 days but without recovery at follow-up 11 weeks later. A causal relationship to the study drug was assessed as unlikely. Dialysis set-up All study dialyses were 4 h sessions. Before start of dialysis, the dialyzer was primed on both sides for 15 min with Biosol A 201.5/301.5 glucose (Pharmalink, Solna, Sweden) dialysis concentrate diluted 1 þ 34 using Gambro AK-200 haemodialysis equipment (Gambro, Lund, Sweden). The flow rate was adjusted to 500 ml/min on the dialysis fluid side and to 250 ml/min on the blood side of a low flux, low efficiency, biocompatible parallel flow dialyzer (Gambro, Lundia Pro 600) by separate pumps. Pressure in the bloodline was monitored on both sides of the dialyzer during the dialysis session using standard transducers (Peter von Berg Medizintechnik, Kirchsee, GmbH) connected to a batteryoperated transducer preamplifier (Kent Scientific Corporation/World Precision Instrument Sarasota, FL). Before each dialysis session the pressure transducers were calibrated. The mean pressure was shown on a digital display and recorded every 10 min. Before connection of the patient to the efferent and afferent bloodlines, 20 ml (300 mg/ml) iohexol was given intravenously to monitor dialyzer clearance during Visits 2, 3, 6 and 7. After completion of the dialysis session, the cannulae were removed from the A-V fistulas according to standard procedures and the duration of bleeding from puncture sites was recorded during compression. Permanent central venous catheters were disconnected and heparinized according to routine procedure. The investigator made an overall assessment whether or not the session had been successful and without complications. The disconnected bloodlines and dialyzer were cleared from blood by perfusion with saline and searched for visible clots. The results were reported separately for tubing and dialyzer and graded as no clot, suspected clot, evident clot or obstruction. 1892 Measurement of dialyzer clearance For the determination of iohexol in plasma, arterial blood (6 ml) was drawn from efferent (arterial) tubing into heparinized test tubes every 30 min after the iohexol administration. Plasma was recovered after centrifugation (10 000 g for 5 min) and stored at 20 C until analysis. Iohexol in plasma was determined at the Laboratory of the Department of Nephrology (Sahlgrenska University Hospital, Göteborg, Sweden) using a Renalyser PRX 90 (Provalid AB, Lund, Sweden). Iohexol clearance, used as a marker of dialyzer function, was estimated as dose/area under the plasma concentration-versus-time curve (AUC) of iohexol, where the AUC of iohexol was calculated using the log-linear trapezoidal rule as described below for the pharmacokinetic evaluation of melagatran. Pharmacokinetic assessments Blood samples (2.7 ml) were obtained in citrate buffer (Venojectß; Terumo Europe N.V, Leuven, Belgium) for determination of plasma concentrations of melagatran. On Visit 4, when patients were administered melagatran as an i.v. bolus while not undergoing dialysis, venous blood samples (2.7 ml) were obtained predose and 10 and 30 min; and 1, 2, 3, 4, 6, 8, 10, 12 and 22 h postdose. During dialysis sessions on Visits 5, 6 and 7, the same blood samples (2.7 ml) were taken from the efferent (arterial) dialysis blood tubing predose and 10 and 30 min; and 1, 2, 3, and 4 h postdose. Catheters were flushed with physiological saline after each sampling, while the use of heparin was prohibited. Melagatran concentrations were determined at Bioanalytical Chemistry, AstraZeneca Research and Development, Mölndal, Sweden, using liquid chromatography-mass spectrometry with a limit of quantification of 0.010 mmol/l [13]. A dilution factor of 1.185 was used to adjust for the dilution of blood with citrate buffer in the sampling tubes. Noncompartmental methods were used to estimate pharmacokinetic variables of melagatran for the data obtained between dialysis sessions (Visit 4) and during dialysis (Visit 5). The AUC was calculated using the log-linear trapezoidal rule to the last measurable plasma concentration and extrapolated to infinity. Maximum plasma concentration (Cmax) was the observation at the first sampling time (10 min) after i.v. injection. The elimination half-life (t1/2) was calculated as 0.693/k, where k is the elimination rate constant estimated by linear least-squares regression of plasma concentrations versus time in the terminal phase of decline. Clearance of melagatran (CL) was calculated as dose/AUC. The clearance by dialysis (CLD) was estimated as (CL during the dialysis at Visit 5) – (CL between the dialysis sessions). The volume of distribution for steady-state condition (Vss) was calculated as (dose/AUC) (AUMC/AUC), where AUMC is the area under the first-moment curve calculated using the log-linear trapezoidal rule up to Clast and extrapolated to infinity. Results Clinical outcomes of dialysis sessions and dose adjustments For all 10 patients, dialysis sessions with dalteparin (Visits 2 and 3) and with melagatran i.v. added P.-O. Attman et al. to dialysis fluid (Visits 6 and 7) were judged by the investigators to be clinically successful. The two dalteparin dialyses were also regarded as successful in the eleventh patient, who later developed a septic gonitis and withdrew from the study before melagatran administration and completion of the study. There were few adverse events (AEs) during the study, especially when considering the uraemic condition of the study population and the average study duration of 3 months. Most AEs were of mild or moderate intensity. There were no serious AEs during or after administration of melagatran. Two serious AEs were reported in the periods after the first and the second baseline dialysis with dalteparin, respectively. Nine out of a total of 32 AEs started during treatment with melagatran. These were hypotension (n ¼ 3, different patients and treatment periods), headache (n ¼ 1), dyspepsia (n ¼ 1), hypoglycemia (n ¼ 1, diabetic patient), polyuria (n ¼ 1), parosmia (n ¼ 1) and taste perversion (n ¼ 1). For each one of the 10 patients who completed this study the individual dalteparin dose was kept constant throughout the study, whereas melagatran was provided at fixed doses via i.v. injection and dialysis fluid only at Visit 6 (cf Subjects and methods). At the last dialysis with melagatran (Visit 7), it was judged suitable by the investigator to maintain the i.v. bolus at 2 mg in six cases, to change to 1 mg in three cases, and to change to 3 mg in one case (Table 2). The melagatran concentration in dialysis fluid was increased to 1 mmol/l in six cases, decreased to 0.25 mmol/l in three cases and maintained at 0.5 mmol/l in one case (Table 2). Presence or absence of clots in tubing and dialyzers after Visit 6 provided reasons for the dose decision at Visit 7 (vide infra). Dialyzer clearance The plasma concentrations of iohexol versus time are shown in Figure 1 for the two dialysis sessions with dalteparin and the first dialysis session with melagatran in the dialysis fluid (Visit 6). Comparable observations were made at the second dialysis session with melagatran (7; data not shown). The (mean±SD) iohexol clearance values during the two dialysis sessions with melagatran in the dialysis fluid were similar to those observed during the two dialysis sessions with dalteparin (Table 3). Extracorporeal circuit pressure The pressures before and after the dialyzer during the two sessions with melagatran in the dialysis fluid were similar to those observed during the two sessions with dalteparin, and their pre- to post-dialyzer ratios remained constant and fully comparable throughout the 4 h sessions. Figure 2 shows the mean pressure ratio before/after dialyzer for the two sessions with dalteparin and for the dialysis sessions with melagatran Melagatran for prevention of clot formation during haemodialysis 1893 A A 2.5 400 Mean filter pressure ratio (before/after) 350 lohexol (mg/l) 300 250 200 150 2.0 1.5 1.0 0.5 100 0 50 0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 Time (hours) 0 0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 B Time (hours) 2.5 B 2.0 Mean filter pressure ratio (before/after) 400 350 lohexol (mg/l) 300 250 200 1.5 1.0 0.5 150 0 100 0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 Time (hours) 50 0 0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 Time (hours) Fig. 1. Mean concentrations of iohexol in plasma (mg/l) with 95% CI versus time after: (A) dalteparin i.v. bolus (mean of Visits 2 and 3, n ¼ 10), and (B) melagatran 2 mg i.v. and 0.5 mmol/l in dialysis fluid (Visit 6, n ¼ 10). 0.5 mmol/l in the dialysis fluid. Similar results were observed at melagatran concentrations of 0.25 and 1.0 mmol/l (data not shown). Fig. 2. Dialyzer pressure given as pressure ratio before/after dialyzer during dialysis with: (A) dalteparin i.v. bolus (mean of Visits 2 and 3, n ¼ 20 observations for 10 patients), and (B) melagatran concentration of 0.5 mmol/l in dialysis fluid (Visits 6 þ 7, n ¼ 11 observations for 10 patients). the bolus dose of melagatran at Visit 7 was increased to 3 mg because of suspected clots in tubing and dialyzer at Visit 6. After dose adjustment at Visit 7, six patients showed suspected clots in tubing (n ¼ 3) and/or dialyzers (n ¼ 6). In the three patients with suspected clots in the tubing, one had evident and the other two suspected clots in the dialyzer. Evaluation of clot formation during the dialysis sessions Occluding thrombi were not observed in this study. Suspected clots in tubing and/or in the dialyzer were observed after both dalteparin dialyses in four of the 10 patients who completed the study (Visits 2 and 3). The same outcome was observed for the 10 patients who received dalteparin and melagatran i.v. during one dialysis (Visit 5). Suspected and/or evident clots were observed in tubing and/or dialyzers in eight of 10 patients after dialysis with fixed melagatran doses (Visit 6). Visually evident clots in dialyzers (n ¼ 4) or suspected clots in both tubing and dialyzers (n ¼ 2) at Visit 6 explained the increase in melagatran in the dialysis fluid during Visit 7. Three patients, who were given reduced melagatran bolus and dialysis fluid doses at Visit 7, had no clots in the tubing or in the dialyzer (n ¼ 2) or only suspected clots in the dialyzer (n ¼ 1) at Visit 6. In the tenth case, the melagatran concentration in dialysis fluid was maintained but Bleeding after percutaneous puncture of A-V fistulas The duration of bleeding at the dialysis access site (the puncture site in the A-V fistula) after removal of the fistula needle at the end of dialysis session was assessed in all patients, except for patients with a central venous catheter (patient nos 8, 10 and 11). Patient no. 2 had a central venous catheter at Visits 2 and 3 before his newly prepared A-V fistula could be used. On Visits 2 and 3 when the patients received dalteparin alone, the range of bleeding times was 3–9 min (Table 4). A similar range was observed on Visit 5 when the patients received an i.v. bolus of melagatran just after the start of the dialysis session in combination with pre-dialysis dalteparin. For Visits 6 and 7, when melagatran treatment was given, the bleeding times were within 4–9 min for the majority of the patients. Patient 4 had a longer bleeding time of 17 min on Visit 7 (melagatran concentration in 1894 P.-O. Attman et al. Table 4. Mean duration of bleeding (min) after percutaneous punctures of arteriovenous fistulas in connection with dialysis A 1.0 1 Duration of bleeding (min) n 2 3 5 6 7 6 6 7 7 7 Mean SD Minimum Maximum 6 7 6 7 9 1 2 1 2 5 4 3 4 5 4 7 9 8 11 17 Plasma concentration (µmol/l) 2 Visit 3 4 5 0.1 7 8 9 10 0.01 0 2 4 6 8 10 12 14 16 18 20 22 11 Time (hours) B A Patient Plasma concentration (µmol/l) 1 2 3 4 0.1 5 7 8 9 0.01 0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 10 11 1.0 Plasma concentration (µmol/l) 1.0 0.1 0.01 0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 Time (hours) Time (hours) Fig. 4. Individual plasma concentrations of melagatran versus time after (A) a single 1 mg intravenous bolus administered between dialysis sessions (Visit 4), and (B) a single 1 mg bolus dose administered just after the start of dialysis (Visit 5). B Plasma concentration (µmol/l) 1.0 0.1 0.01 0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 Time (hours) Fig. 3. Individual plasma concentrations of melagatran versus time after: (A) a single 2 mg intravenous bolus of melagatran followed by melagatran administered in dialysis fluid at a concentration of 0.5 mmol/l (Visit 6), and (B) after a single 1, 2 or 3 mg bolus dose of melagatran followed by melagatran administered in dialysis fluid at concentrations of 0.25, 0.5 or 1.0 mmol/l (Visit 7). dialysis fluid 1 mmol/l). Patient 7 had longer bleeding times of 11 and 12 min for Visits 6 and 7, respectively (melagatran concentration in dialysis fluid 0.5 and 1 mmol/l). Pharmacokinetics of melagatran Stable plasma concentrations of melagatran were observed throughout both of the 4 h dialysis sessions with melagatran administered in the dialysis fluid (Figure 3). During Visit 6, when melagatran was administered in the dialysis fluid at a concentration of 0.5 mmol/l, mean plasma concentrations of melagatran were maintained between 0.317 and 0.371 mmol/l, a range that corresponded to 63–74% of the concentrations in the dialysis fluid (Table 2). Similar relationships between plasma concentrations of melagatran and melagatran concentrations in the dialysis fluid were observed during Visit 7, when melagatran was administered at concentrations of 0.25, 0.5 or 1.0 mmol/l (Figure 3 and Table 2). Comparison of melagatran plasma concentrations when melagatran was given as an i.v. bolus between dialysis sessions (Visit 4) with those when melagatran was given as an i.v. bolus just after the start of dialysis on Visit 5 showed that melagatran was rapidly eliminated via dialysis (Figure 4 and Table 5). When melagatran was administered intravenously between dialysis sessions, the mean plasma concentration was 0.494 mmol/l 10 min after injection. Melagatran plasma concentrations rapidly declined during the first 30 min after dose during the distribution phase and thereafter slowly declined to a mean of 0.043 mmol/l 22 h after injection (Figure 4). When melagatran was administered just after the start of dialysis, the mean plasma concentration was 0.169 mmol/l 10 min after injection. By 4 h, at the end of the dialysis session, mean plasma concentration of melagatran was 0.028 mmol/l (Figure 4). The fraction of the melagatran plasma concentration remaining at the end of the dialysis (mean±SD) was 17±4%. The clearance of melagatran (mean±SD) between dialysis sessions was 0.93±0.36 l/h compared with 7.20±0.76 l/h during dialysis (Table 5). The half-life (mean±SD) was 14.1±3.7 h between dialysis sessions compared with 2.0±0.3 h during dialysis. The clearance of melagatran by dialysis (mean±SD) for the eight patients having data both between dialysis sessions Melagatran for prevention of clot formation during haemodialysis 1895 Table 5. Melagatran pharmacokinetics after a single bolus 1 mg dose between dialyses (Visit 4) and after a single bolus 1 mg dose just after the start of dialysis (Visit 5) CLD (l/h) Vss (l) t1/2 (h) 0.67 1.09 0.53 0.95 0.95 0.75 0.79 1.71 – – – – – – – – 14.9 14.9 10.7 20.3 23.9 17.2 15.2 12.5 16.6 10.5 14.7 15.4 17.8 16.0 15.2 6.5 8 0.494 0.284 0.192 0.463 0.962 57 8 0.93 0.36 0.53 0.87 1.71 39 – – – – – – – 8 16.2 4.2 10.7 15.1 23.9 26 8 14.1 3.7 6.5 15.3 17.8 26 Visit 5: Melagatran given at the start of dialysis 1 0.315 0.260 2 0.392 0.156 3 0.304 0.166 4 0.333 0.168 5 0.333 0.153 7 0.317 0.132 8 0.378 0.198 9 0.310 0.179 10 0.269 0.143 11 0.322 0.135 7.39 5.95 7.67 7.01 6.99 7.36 6.17 7.51 8.65 7.25 5.28 6.58 6.48 6.04 6.41 5.42 6.72 6.94 14.1 21.7 20.0 21.1 21.3 20.0 15.4 19.3 21.2 24.0 1.4 2.6 1.9 2.2 2.2 1.9 1.8 1.9 1.7 2.3 10 7.20 0.76 5.95 7.31 8.65 11 8 6.23 0.60 5.28 6.45 6.94 10 10 19.8 3.0 14.1 20.6 24.0 15 10 2.0 0.3 1.4 1.9 2.6 17 Patient AUC (mmolh/l) Cmax (mmol/l) Visit 4: Melagatran given between dialysis sessions 2 3.475 0.496 3 2.145 0.962 4 4.378 0.817 5 2.449 0.280 7 2.460 0.192 8 3.096 0.193 9 2.938 0.430 10 1.363 0.584 n Mean SD Minimum Median Maximum CV n Mean SD Minimum Median Maximum CV 8 2.788 0.909 1.363 2.699 4.378 33 10 0.327 0.036 0.269 0.320 0.392 11 10 0.169 0.038 0.132 0.161 0.260 22 CL (l/h) AUC, area under the plasma concentration-versus-time curve; Cmax, maximum plasma concentration observed at the first sampling time after injection; CL, clearance; CLD, clearance by dialysis; CV, coefficient of variation; Vss, volume of distribution estimated for steady-state condition; t½ ¼ elimination half-life. and during dialysis was 6.23±0.60 l/h. Expressed in the same units as for iohexol clearance, melagatran clearance during dialysis was 104±10 ml/min; this was comparable to iohexol clearance, which was used as the marker for dialyzer function. Discussion This feasibility study showed that the low-molecularweight direct thrombin inhibitor melagatran can be safely administered from the dialysis fluid via the dialysis membrane to uraemic patients undergoing haemodialysis. It may successfully prevent thrombotic obstruction of the extracorporeal circuit to provide as efficient dialysis as with using conventional anticoagulation with fractionated heparin. More studies are needed to establish the cost-benefit balance of this concept. Furthermore, the optimal doses of melagatran must be explored further, but the overall conclusion about its usefulness applies to the range of doses used here. The plasma concentrations of melagatran during dialysis are strictly determined by the concentration in the dialysis fluid, with a predictable concentrationdependent degree of anticoagulation making individual titration unnecessary. None of the small clots, suspected or evident, that were observed during dalteparin and melagatran dialyses sessions influenced the dialysis efficiency, dialyzer clearance of iohexol or extracorporeal pressures. The duration of bleeding at the dialysis access site in A-V fistulas following melagatran treatment was comparable to the results with dalteparin treatment in most patients. These results suggest that melagatran may be administered via the dialysis fluid to prevent clot formation during haemodialysis. Previous observations in acutely anuric pigs showed that they rapidly achieved a steady-state concentration level when melagatran was provided in the dialysis fluid during haemodialysis [10]. Initially no bolus dose of melagatran was given to the pigs before dialysis and we observed that thrombotic obstruction occurred very early in the extracorporeal circuit (unpublished observations). We attributed this to 1896 plasma melagatran concentrations having not reached a sufficient level during the very early stages of the dialysis experiments. With a half-life of 2 h during dialysis it takes 20 min to reach a plasma concentration of 10% of the expected steady state concentration. Therefore the pigs were given an i.v. bolus of melagatran immediately prior to start of dialysis, which effectively prevented early clotting. Hence, an analogous protocol including a bolus dose of melagatran before start of dialysis was used in this study. Melagatran added to dialysis fluid rapidly equilibrated across the dialysis membrane and was maintained at a stable level throughout the dialysis sessions. The plasma concentration of melagatran was maintained at 70% of that in the dialysis fluid regardless of its melagatran concentration (i.e., 0.25, 0.5 or 1.0 mmol/l). Clearance of melagatran administered during dialysis was 7 more rapid than that between dialysis sessions in these patients with severe renal failure. The dialysis clearance of melagatran had a low interindividual variability of 10% and was comparable to that of iohexol, which was used to assess dialyzer function as it is rapidly cleared by haemodialysis. These results demonstrate that melagatran was readily cleared during haemodialysis, and this suggests that dialysis may be used when rapid reversal of anticoagulation is necessary (e.g. suspected overdose, accumulation because of renal dysfunction, or bleeding). As the half-life of melagatran is relatively short for patients with normal renal function [14,15], dialysis would have the greatest impact for a patient with severely compromised renal function. For the uraemic patients studied, 80% of the administered i.v. dose of melagatran was eliminated during the 4 h dialysis period. The low melagatran clearance determined in the uraemic patients was expected, as renal excretion is the predominant route of elimination for systemic melagatran [7]. Compared with young healthy volunteers, the exposure of melagatran has been shown to be higher in older volunteers because of the age-related decrease in renal function [10]. In patients with normal or mildly to moderately impaired renal function receiving parenteral melagatran or oral ximelagatran treatment, renal function is the main determinant of interindividual variability in melagatran exposure [14,15]. In 12 patients with severe renal impairment (median GFR estimated as iohexol clearance: 13 ml/min), the melagatran exposure was increased by 4–5 compared to with that in healthy control volunteers with normal renal function [16]. For patients with congenital or acquired antithrombin III deficiency, heparin does not provide adequate anticoagulation during haemodialysis. Furthermore, it has been speculated that long-term heparin administration could lead to increased consumption and subsequent depletion of antithrombin III [17]. Ota et al. [18] recently reported successful anticoagulation with argatroban, a synthetic thrombin inhibitor in haemodialysis. In contrast to argatroban, melagatran is efficiently removed by haemodialysis. P.-O. Attman et al. A potential advantage of melagatran in the dialysis population is the long half-life between dialysis sessions. This may facilitate maintenance of patency of A-V fistulas and central dialysis catheters, and possibly reduce the need for separate anticoagulation between dialysis treatments. Another, albeit minor, advantage of the treatment modality used is that blood contact when administering anticoagulation is avoided. No serious bleeding complications were observed in this study during dialysis with melagatran or dalteparin. Based on previous studies in non-renal patients it could be expected that the use of melagatran as anticoagulant in haemodialysis should not result in increased frequency of such complications compared to that of conventional regimes including LMWH. In summary, the results of this exploratory feasibility study suggest that melagatran administration via dialysis fluid can be used for successful anticoagulation during haemodialysis. It can certainly provide a useful alternative for patients with heparin intolerance and a means to avoid long-term negative effects of chronic heparin administration in these patients. Although no serious side-effects were observed, additional studies are needed to establish risks and benefits of the treatment and optimal dosing in haemodialysis. Moreover, even if melagatran is an attractive alternative for dialysis patients with HIT, its use in other dialysis patients will depend on experience from further clinical trials and the costs in comparison with conventional anticoagulation. Finally, the current results indicate that melagatran is eliminated by haemodialysis, which suggests that this method may be used to decrease drug levels in patients with renal impairment. Acknowledgements. This study was sponsored by AstraZeneca R&D Mölndal (Mölndal, Sweden). The skilful secretarial assistance of Ann Stenman is gratefully acknowledged. The study dialysis sessions were expertly carried out by Margareta Dahlgren RN, Ulla-Britt Johansson RN and Hussain Quraishi. Conflict of interest statement. Ulf G. Eriksson, Maria ErikssonLepkowska and Gunnar Fager are employees of AstraZeneca. [See related Editorial by Flanigan, pp. 1789] References 1. Warkentin TE, Kelton JG. Heparin and platelets. Hemotol Oncol Clin North Am 1990; 4: 243–264 2. Samuelsson O, Amiral J, Attman P-O et al. Heparin-induced thrombocytopenia during continuous hemofiltration. Nephrol Dial Transplant 1995; 10: 1768–1771 3. Lai KN, Ho K, Cheung RCK et al. Effect of low molecular weight heparin on bone metabolism and hyperlipidemia in patients on maintenance hemodialysis. Int J Artif Organs 2001; 24: 447–455 4. Wiemer J, Winkler K, Baumstark M, März W, Scherberich JE. Influence of low molecular weight heparin compared to conventional heparin for anticoagulation during haemodialysis on low density lipoprotein subclasses. Nephrol Dial Transplant 2002; 17: 2231–2238 5. Liu G, Bengtsson-Olivecrona G, Ostergaard P, Olivecrona T. Low Mr heparin is as potent as conventional heparin in releasing Melagatran for prevention of clot formation during haemodialysis 6. 7. 8. 9. 10. 11. 12. lipoprotein lipase, but is less effective in preventing hepatic clearance of the enzyme. Biochem J 1991; 273: 747–752 Taal MW, Masud T, Green D, Cassidy MJD. Risk factors for reduced bone density in haemodialysis patients. Nephrol Dial Transplant 1999; 14: 1922–1928 Gustafsson D, Elg M. The pharmacodynamics and pharmacokinetics of the oral direct thrombin inhibitor ximelagatran and its active metabolite melagatran: a mini-review. Thromb Res 2003; 109: S9–S15 Eriksson UG, Bredberg U, Hoffmann KJ et al. Absorption, distribution, metabolism, and excretion of ximelagatran, an oral direct thrombin inhibitor, in rats, dogs, and humans. Drug Metab Dispos 2003; 31: 294–305 Eriksson UG, Bredberg U, Gisslén K et al. Pharmacokinetics and pharmacodynamics of ximelagatran, a novel oral direct thrombin inhibitor, in young healthy male subjects. Eur J Clin Pharmacol 2003; 59: 35–43 Johansson LC, Frison L, Logren U, Fager G, Gustafsson D, Eriksson UG. Influence of age on the pharmacokinetics and pharmacodynamics of ximelagatran, an oral direct thrombin inhibitor. Clin Pharmacokinet 2003; 42: 381–392 Bergqvist D. Bleeding profiles of anticoagulants, including the novel oral direct thrombin inhibitor ximelagatran: definitions, incidence and management. Eur J Haematol 2004; 73: 227–242 Carlsson S, Samuelsson O, Attman P-O, Zachrisson H, Fager G. Antithrombotic effect and pharmacokinetics of the 1897 13. 14. 15. 16. 17. 18. direct thrombin inhibitor melagatran administered in dialysate in a porcine model of haemodialysis. J Am Soc Nephrol 2004; 15: 638A. Larsson M, Logren U, Ahnoff M et al. Determination of melagatran, a novel direct thrombin inhibitor, in human plasma and urine by liquid chromatography-mass spectrometry. J Chromatogr B Biomed Sci Appl 2002; 766: 47–55 Eriksson UG, Mandema JW, Karlsson MO et al. Pharmacokinetics of melagatran and the effect on ex vivo coagulation time in orthopaedic surgery patients receiving subcutaneous melagatran and oral ximelagatran: A population model analysis. Clin Pharmacokinet 2003; 42: 687–701 Wolzt M, Wollbratt M, Svensson M, Wahlander K, Grind M, Eriksson UG. Consistent pharmacokinetics of the oral direct thrombin inhibitor ximelagatran in patients with nonvalvular atrial fibrillation and in healthy subjects. Eur J Clin Pharmacol 2003; 59: 537–543 Eriksson UG, Johansson S, Attman P-O et al. Influence of severe renal impairment on the pharmacokinetics and pharmacodynamics of oral ximelagatran and subcutaneous melagatran. Clin Pharmacokinet 2003; 42: 743–753 Brandt P, Jespersen K, Sorenson H. Antithrombin-III and platelets in haemodialysis patients. Nephron 1981; 28: 1–3 Ota K, Akizawa T, Hirasawa Y, Agishi T, Matsui N. Effects of argatroban as an anticoagulant for haemodialysis in patients with antithrombin III deficiency. Nephrol Dial Transplant 2003; 18: 1623–1630 Received for publication: 15.12.04 Accepted in revised form: 22.4.05