TY - JOUR
AU - Chanard,, Jacques
AB - Abstract Background. Binding of polycationic unfractionated heparin onto the modified AN69 polyacrylonitrile membrane, whose surface electronegativity has been neutralized by layering polyethyleneimine (AN69ST), produces stable coating. We investigated whether the heparin-coated membrane was suitable for regular haemodialysis with low heparin doses. Methods. Sheep were instrumented for extracorporeal circulation perfusing a dialyser equipped with either the AN69ST or the original AN69 membrane. Dialysis sessions were performed after priming the dialyser with heparinized saline. The session was conducted without systemic administration of heparin. In chronic haemodialysis patients, the AN69ST membrane was tested for safety, clotting and thrombin generation according to protocols of 4-h haemodialysis sessions with tapered heparin doses. The goal was to define optimal heparin requirements with the heparin-coated membrane in the setting of continuous or intermittent administration of heparin. Both unfractionated and low molecular weight heparin (LMWH) (enoxaparin) were tested. Results. In sheep, systemic heparin-free haemodialysis was conducted for 6 h without clotting using the heparin-coated dialyser. In the same conditions, massive clotting was observed within 90 min of dialysis with the native AN69 membrane. In man, through kinetic measurements of activated partial thromboplastin time (APTT), heparin anti-Xa concentration and thrombin–anti-thrombin complexes levels (TAT), significant dialyser clotting was avoided when APTT and anti-Xa concentration at 180 min of dialysis, were maintained at >40 s and >0.2 IU/ml, respectively. With the AN69ST heparin-coated membrane, thrombin generation was reduced then suppressed, as compared with the original AN69, primed in the same conditions. Safety of haemodialysis conducted with the AN69ST heparin-coated membrane and low doses of unfractionated heparin (50% reduction of the reference dose) was validated by a survey of 2590 sessions in 32 patients. Doses of LMWH were also safely reduced by 50%. In addition, haemodialysis without systemic administration of heparin was possible with minor risk of clotting. Conclusion. During the rinsing phase, the ionic interactions between the new AN69ST polyacrylonitrile membrane and unfractionated heparin induce stable heparin coating. This allows a significant reduction of systemic anticoagulant requirements without increasing the risk of clotting, both in the experimental setting and in the chronic haemodialysis patients. Further studies are required to assess this advantage in patients with acute renal failure and at risk of bleeding and to reduce the metabolic consequences of long-term treatment with heparin. anticoagulation, biocompatibility, chronic renal failure, haemocompatibility, haemodialysis, heparin, polyacrylonitrile membrane Introduction Despite continuous improvement in haemodialysis technology, contact between blood and artificial dialysis membrane triggers activation of serum factors and blood nucleated cells, resulting in the so-called bioincompatibility of the extracorporeal circuit [1]. For example, complement activation still occurs with synthetic membranes but to a lesser extent than with cellulose membranes [2]. Coagulation is also activated and can be prevented with heparin, which is the most practicable and efficient anticoagulant used so far in chronic haemodialysis. Practical recommendations for the best anticoagulation strategy, particularly in patients at risk of bleeding, have been proposed [3–5]. Whether using unfractionated heparin or fractionated low molecular weight heparin (LMWH), these recommendations are based upon criteria of changes in global clotting tests [6] and pharmacokinetic parameters. Artificial membranes have been coated with heparin in order to reduce excessive systemic coagulation and thus the risk of bleeding while maintaining efficient anticoagulation in the extracorporeal circuit to avoid clotting. Heparin coating of artificial membranes has been investigated for years with questionable results, at least in the field of haemodialysis [7–11]. Coating mechanisms are not well known, but imply electric charge interactions. Heparin, which is strongly electronegative, exhibits binding affinity for cationic membranes whether the membrane surface is electropositive or neutral. In the case of neutral or near neutral membranes, such as cuprophane membrane, secondary grafting of heparin was possible through bridging with polyethylene glycol. Unfortunately, the ensuing composite membrane, with improved haemocompatibility and used in cardio-pulmonary surgery, was not suitable for long-term regular haemodialysis [10]. The polyacrylonitrile membrane, such as the AN69 membrane made of a mixture of polyacrylonitrile and sodium methallyl sulfonate, is an electronegatively charged surface [11,12]. This physical property associated to pH conditions plays a crucial role in the coagulation process through direct contact phase activation and kinins generation [11,13]. It has been demonstrated that reduced AN69 surface electronegativity following membrane coating with the polycationic polymer polyethyleneimine strongly decreases the binding of high molecular weight kininogen and Hageman factor activation [14]. On the opposite, polyanionic heparin binds onto the modified membrane. We considered that such a binding is stable enough to sustain anticoagulation and to reduce systemic administration of heparin during dialysis. Prior to testing this hypothesis in chronically haemodialysed patients, we carried out both in vitro studies to quantify heparin binding to the new membrane (AN69ST membrane), and animal studies to confirm anticoagulation efficiency in dynamic conditions. In humans, the aim of the study was to find out whether the improved haemocompatibility of the AN69ST haemodialysis membrane used either as plate or fibre dialyser allowed heparin doses to be reduced without increasing the risk of extracorporeal clotting. Subjects and methods In vitro study Heparin binding onto AN69 and AN69ST membranes was measured in vitro. Heparinized saline solution (5000 IU/l) run at 100 ml/min for 10 min, in an open loop circuit perfusing dialyser either as plate-sheet or hollow-fibre membranes. Heparin adsorption was measured using the Azure-A colorimetric method. The amount of heparin adsorbed on the membrane was calculated from the difference in heparin concentrations contained in the perfusing solution, between baseline and study end. Results are expressed as International Units (IU) per square meter (m2). Unfractionated heparin and the LMWH enoxaparin, were used and polysulfone, triacetate, AN69 and AN69ST membranes were tested. Animal study The study was designed to compare the thrombogenic effect of AN69ST vs native AN69 dialysis membranes in a model of extracorporeal circulation mimicking haemodialysis in non-uraemic ‘Ile-de-France’ sheep (INRA, Lyon, France). The study was approved by the Animal Ethics Committee of the Ecole Nationale Vétérinaire de Lyon (France). Sheep weighing 50–70 kg were anaesthetized with i.v. sodium pentobarbital (15 mg/kg). Catheters were implanted percutaneously into a carotid artery and an internal jugular vein, and filled with unfractionated heparin solution. Following recovery from surgery, animals stayed free in their boxes. Haemodialysers were tested a few days later in awaken animals for continuous dialysis sessions with a minimal ultrafiltration flow rate of 16 ml/min. The filtrate was continuously returned into the jugular vein via the drip chamber access, in order to avoid fluid imbalance. The AN69ST membrane (Hospal, France) was tested as both plate (Crystal 4000 ST®) and hollow-fibre (Nephral 400 ST®) dialysers, both having a 1.6 m2 surface area. Performances were compared with the ‘native’ AN69 membrane of same exchange surface, used as plate (Crystal 4000®) or hollow-fibre (Nephral 400®) dialysers. Blood lines were identical in both series of experiments. Haemodynamic conditions were identical, too, with a blood flow rate of 300 ml/min perfusing the extracorporeal circuit which was primed with 2 l of saline containing 5000 IU/l of unfractionated heparin (Choay, France). Blood samples were withdrawn from the dialyser’s venous line, before dialysis, 10, 15, 30, 60 and every 10 min thereafter. The plasma activated partial thromboplastin time (APTT) of plasma was measured. Five sheep were tested twice using both membranes. The membrane order was allocated randomly. In all experiments APTT was ‘normal’ prior to dialysis initiation. Clinical study Patients. Thirteen chronically haemodialysed patients were included in an open prospective monocentric study whose aim was to find out the minimal systemic heparin dose in patients dialysed with the AN69ST heparin-binding membrane. The study was conducted according to national regulations and approved by the local Ethics Committee. Patients gave informed consent before enrolment. Inclusion criteria were as follows: patients aged over 18 years, haemodynamic stability during bicarbonate haemodialysis, use of non-reused highly permeable synthetic membrane, urea Kt/V greater than 1.2, absence of signs of patent denutrition and systemic inflammation with a serum albumin level >3.5 g/dl, and a plasma fibrinogen level lower than 4.5 g/l. Haematocrit was >30%. Non-inclusion criteria were: any haemostatic disorder favouring either bleeding or clotting, anti-vitamin K treatment, treatment with aspirin, dipyridamole or any drug likely to interfere with coagulation; congestive heart failure (stages III–IV, NY classification), or angina pectoris and treatment with angiotensin converting enzyme inhibitor. Haemodialysis strategy All patients were haemodialysed for 4 h, three times weekly. Blood and dialysate flow rates were set at 300 and 500 ml/min, respectively. Ultrafiltration was tailored according to patient’s fluid overload measured as inter-dialysis weight gain. The baseline anticoagulation protocol consisted in priming the dialyser with 2 l of heparinized saline (5000 IU/l). Heparinized saline was flushed out of the dialyser before connecting the venous line. A loading dose (4000 IU) of unfractionated heparin, followed by a bolus of 2000 IU after the second hour of the session were administered in the arterial line. Dialysis efficiency was monitored throughout sessions by means of ionic clearance measurements using the Diascan monitor (Hospal, France), which gives an accurate estimation of small molecule clearances, including urea clearance. Study design The patient was connected to the AN69ST dialyser. Two series of experiments were conducted first using a 1.48 m2 flat-sheet dialyser (Crystal ST®) and, then, a 1.65 m2 hollow fibre dialyser (Nephral ST®). Each series consisted in several steps of reduced heparin doses, as indicated in Table 1. Each step consisted in three consecutive sessions and was quoted in weeks (Wn). Unfractionated heparin was tested at baseline and during 7 weeks. During the last week (W7) except for priming, the session did not include any systemic administration of heparin. LMWH was only tested as pulse dose of 20 mg during the first week (baseline) and 10 mg during the second week. Table 1. Experimental protocols for tapering heparin doses during a 4 h session of haemodialysis with heparin-coated AN69-ST dialyser Weeks of experiment . Baseline 1 . . 2 . . 3 . . 4 . . 5 . . 6 . . 7 . . Intermittent administration (pulse) Unfractionated heparin doses (IU) given at: t0 t2 h t0 t2 h t0 t2 h t0 t2 h t0 t2 h t0 t2 h t0 t2 h Plate-sheet dialyser 4000 2000 4000 1000 4000 0 3000 0 2000 0 1000 0 0 0 Hollow-fibre dialyser 4000 2000 4000 1000 3000 1000 2000 1000 1000 1000 0 1000 0 0 LMW heparin (enoxaparin) Doses (mg) given with hollow fibre dialyser 20 – 10 – Continuous administrationa Unfractionated heparin: Pulse dose (IU) 2000 1000 500 0 0 Continuous infusion rate (IU/h) 500 500 500 500 0 Weeks of experiment . Baseline 1 . . 2 . . 3 . . 4 . . 5 . . 6 . . 7 . . Intermittent administration (pulse) Unfractionated heparin doses (IU) given at: t0 t2 h t0 t2 h t0 t2 h t0 t2 h t0 t2 h t0 t2 h t0 t2 h Plate-sheet dialyser 4000 2000 4000 1000 4000 0 3000 0 2000 0 1000 0 0 0 Hollow-fibre dialyser 4000 2000 4000 1000 3000 1000 2000 1000 1000 1000 0 1000 0 0 LMW heparin (enoxaparin) Doses (mg) given with hollow fibre dialyser 20 – 10 – Continuous administrationa Unfractionated heparin: Pulse dose (IU) 2000 1000 500 0 0 Continuous infusion rate (IU/h) 500 500 500 500 0 aWith hollow fibre dialyser only. Open in new tab Table 1. Experimental protocols for tapering heparin doses during a 4 h session of haemodialysis with heparin-coated AN69-ST dialyser Weeks of experiment . Baseline 1 . . 2 . . 3 . . 4 . . 5 . . 6 . . 7 . . Intermittent administration (pulse) Unfractionated heparin doses (IU) given at: t0 t2 h t0 t2 h t0 t2 h t0 t2 h t0 t2 h t0 t2 h t0 t2 h Plate-sheet dialyser 4000 2000 4000 1000 4000 0 3000 0 2000 0 1000 0 0 0 Hollow-fibre dialyser 4000 2000 4000 1000 3000 1000 2000 1000 1000 1000 0 1000 0 0 LMW heparin (enoxaparin) Doses (mg) given with hollow fibre dialyser 20 – 10 – Continuous administrationa Unfractionated heparin: Pulse dose (IU) 2000 1000 500 0 0 Continuous infusion rate (IU/h) 500 500 500 500 0 Weeks of experiment . Baseline 1 . . 2 . . 3 . . 4 . . 5 . . 6 . . 7 . . Intermittent administration (pulse) Unfractionated heparin doses (IU) given at: t0 t2 h t0 t2 h t0 t2 h t0 t2 h t0 t2 h t0 t2 h t0 t2 h Plate-sheet dialyser 4000 2000 4000 1000 4000 0 3000 0 2000 0 1000 0 0 0 Hollow-fibre dialyser 4000 2000 4000 1000 3000 1000 2000 1000 1000 1000 0 1000 0 0 LMW heparin (enoxaparin) Doses (mg) given with hollow fibre dialyser 20 – 10 – Continuous administrationa Unfractionated heparin: Pulse dose (IU) 2000 1000 500 0 0 Continuous infusion rate (IU/h) 500 500 500 500 0 aWith hollow fibre dialyser only. Open in new tab A second protocol using a 50% reduction of doses of unfractionated heparin, was implemented on 32 chronically haemodialysed patients for 6 months, as an open non-randomized study to evaluate the safety of the procedure. Partial or massive clotting in the dialyser or venous drip chamber was evaluated on a semi-quantitative visual scale. APTTs and heparin anti-Xa activity (CK Prest and chromogenic ROTACHROM method, Stago, France), respectively, were measured during the midweek session, before dialysis start and after 15, 60, 120, 180 and 240min of dialysis. Plasma thrombin generation was measured by thrombin–anti-thrombin (TAT) complex concentration (ELISA technique, Behring, Germany) before dialysis start and every 15 min thereafter, using dialysers equipped with either AN69ST or AN69 membrane, in a subgroup of 11 patients, for three sessions each. Haemodialysis was performed under similar technical conditions of priming, with pulse administration of a bolus of 4000 IU of unfractionated heparin at the beginning of the session and of 2000 IU at the second hour. Statistical analysis Results are expressed as mean ± standard deviation. Variances were compared with two-tailed, paired Student’s t-test. A P-value <5% was considered significant. Results In vitro study Figure 1 shows the maximal binding capacity of heparin by dialysis membranes. A 4-fold increase was measured for both unfractionated heparin and LMWH with the AN69ST membrane as compared with other membranes (P < 0.001). Using the LMWH enoxaparin, binding capacity did not differ. Fig. 1. Open in new tabDownload slide In vitro unfractionated heparin adsorption (white bars) and LMWH adsorption (black bars) during the priming step of haemodialysis on haemodialysis membranes: polysulfone (n = 5), cellulose triacetate (n = 5), AN69 (n = 3), AN69ST (n = 6). *P < 0.001: AN69 ST vs every membrane. Fig. 1. Open in new tabDownload slide In vitro unfractionated heparin adsorption (white bars) and LMWH adsorption (black bars) during the priming step of haemodialysis on haemodialysis membranes: polysulfone (n = 5), cellulose triacetate (n = 5), AN69 (n = 3), AN69ST (n = 6). *P < 0.001: AN69 ST vs every membrane. Animal study Under similar haemodynamic conditions, haemodialysis sessions with the AN69 membrane were stopped after ∼90 min of duration because of massive clotting in the extracorporeal circuit, but were sustained with the AN69ST membrane until the sixth hour, when they were voluntarily stopped. In all instances, except for the small amounts of heparin contained in the dialyser after rinsing, the session was conducted without any systemic administration of heparin (Figure 2). These encouraging results prompted us to test the improved haemocompatibility of the AN69ST membrane coated with unfractionated heparin prior to haemodialysis. Fig. 2. Open in new tabDownload slide Experimental haemodialysis in sheep after priming the dialyser with unfractionated heparin. Massive clotting of the extracorporeal circuit was observed within 1 h of blood perfusion with the native AN69 membrane but not with the AN69ST modified membrane, which binds heparin. With the latter membrane, the session was voluntarily stopped at 6 h. Fig. 2. Open in new tabDownload slide Experimental haemodialysis in sheep after priming the dialyser with unfractionated heparin. Massive clotting of the extracorporeal circuit was observed within 1 h of blood perfusion with the native AN69 membrane but not with the AN69ST modified membrane, which binds heparin. With the latter membrane, the session was voluntarily stopped at 6 h. Heparin dose-ranging during haemodialysis Clinical safety. Protocols were designed to test both the nature of heparin (unfractionated or fractionated), and the rheological constraints of the dialyser whether flat-sheet or hollow-fibre. Results of the visual evaluations of clotting in the extracorporeal circuit using unfractionated heparin are indicated in Tables 2 and 3. Table 2. Visual evaluation of clotting in the extracorporeal device after a 4 h dialysis session Total dose of heparin (IU) . 6000 . . 5000 . . 4000 . . 3000 . . 2000 . . 1000 . . 0 . . . Plate . Fibre . Plate . Fibre . Plate . Fibre . Plate . Fibre . Plate . Fibre . Plate . Fibre . Plate . Fibre . Sessions (n) 39 30 39 30 39 30 38 30 35 30 35 30 33 9 Massive clotting of the circuit (n) 0 0 0 0 0 0 1 1 0 0 1 0 1 0 Patchy clotting of the dialyser (n) 0 3 0 2 1 5 2 5 4 7 4 14 4 6 Partial clotting in the venous drip chamber (n) 2 0 6 0 11 3 20 9 22 20 26 24 19 5 Total dose of heparin (IU) . 6000 . . 5000 . . 4000 . . 3000 . . 2000 . . 1000 . . 0 . . . Plate . Fibre . Plate . Fibre . Plate . Fibre . Plate . Fibre . Plate . Fibre . Plate . Fibre . Plate . Fibre . Sessions (n) 39 30 39 30 39 30 38 30 35 30 35 30 33 9 Massive clotting of the circuit (n) 0 0 0 0 0 0 1 1 0 0 1 0 1 0 Patchy clotting of the dialyser (n) 0 3 0 2 1 5 2 5 4 7 4 14 4 6 Partial clotting in the venous drip chamber (n) 2 0 6 0 11 3 20 9 22 20 26 24 19 5 Two types of heparin-coated dialysers were used: plate-sheet (Crystal-ST®) and hollow-fibre (Nephral-ST®). Systemic pulse administration of unfractioned heparin was made at initiation and after 2 h of haemodialysis, as indicated in Table 1. Open in new tab Table 2. Visual evaluation of clotting in the extracorporeal device after a 4 h dialysis session Total dose of heparin (IU) . 6000 . . 5000 . . 4000 . . 3000 . . 2000 . . 1000 . . 0 . . . Plate . Fibre . Plate . Fibre . Plate . Fibre . Plate . Fibre . Plate . Fibre . Plate . Fibre . Plate . Fibre . Sessions (n) 39 30 39 30 39 30 38 30 35 30 35 30 33 9 Massive clotting of the circuit (n) 0 0 0 0 0 0 1 1 0 0 1 0 1 0 Patchy clotting of the dialyser (n) 0 3 0 2 1 5 2 5 4 7 4 14 4 6 Partial clotting in the venous drip chamber (n) 2 0 6 0 11 3 20 9 22 20 26 24 19 5 Total dose of heparin (IU) . 6000 . . 5000 . . 4000 . . 3000 . . 2000 . . 1000 . . 0 . . . Plate . Fibre . Plate . Fibre . Plate . Fibre . Plate . Fibre . Plate . Fibre . Plate . Fibre . Plate . Fibre . Sessions (n) 39 30 39 30 39 30 38 30 35 30 35 30 33 9 Massive clotting of the circuit (n) 0 0 0 0 0 0 1 1 0 0 1 0 1 0 Patchy clotting of the dialyser (n) 0 3 0 2 1 5 2 5 4 7 4 14 4 6 Partial clotting in the venous drip chamber (n) 2 0 6 0 11 3 20 9 22 20 26 24 19 5 Two types of heparin-coated dialysers were used: plate-sheet (Crystal-ST®) and hollow-fibre (Nephral-ST®). Systemic pulse administration of unfractioned heparin was made at initiation and after 2 h of haemodialysis, as indicated in Table 1. Open in new tab Table 3. Visual evaluation of clotting in the extracorporeal device after a 4 h dialysis session with the heparin-coated hollow fibre dialyser AN69-ST (Nephral-ST®) and continuous infusion of unfractionated heparin Week of experiment . Baseline . 1 . 2 . 3 . 4 . Doses of heparin pulse (IU) 2000 1000 500 0 0 Continuous infusion rate (IU/h) 500 500 500 500 0 Sessions (n) 27 24 26 27 24 Massive clotting of the circuit (n) 0 0 1 0 0 Patchy clotting of the dialyser (n) 2 2 1 4 7 Partial clotting in the venous drip chamber (n) 0 6 6 5 6 Week of experiment . Baseline . 1 . 2 . 3 . 4 . Doses of heparin pulse (IU) 2000 1000 500 0 0 Continuous infusion rate (IU/h) 500 500 500 500 0 Sessions (n) 27 24 26 27 24 Massive clotting of the circuit (n) 0 0 1 0 0 Patchy clotting of the dialyser (n) 2 2 1 4 7 Partial clotting in the venous drip chamber (n) 0 6 6 5 6 Open in new tab Table 3. Visual evaluation of clotting in the extracorporeal device after a 4 h dialysis session with the heparin-coated hollow fibre dialyser AN69-ST (Nephral-ST®) and continuous infusion of unfractionated heparin Week of experiment . Baseline . 1 . 2 . 3 . 4 . Doses of heparin pulse (IU) 2000 1000 500 0 0 Continuous infusion rate (IU/h) 500 500 500 500 0 Sessions (n) 27 24 26 27 24 Massive clotting of the circuit (n) 0 0 1 0 0 Patchy clotting of the dialyser (n) 2 2 1 4 7 Partial clotting in the venous drip chamber (n) 0 6 6 5 6 Week of experiment . Baseline . 1 . 2 . 3 . 4 . Doses of heparin pulse (IU) 2000 1000 500 0 0 Continuous infusion rate (IU/h) 500 500 500 500 0 Sessions (n) 27 24 26 27 24 Massive clotting of the circuit (n) 0 0 1 0 0 Patchy clotting of the dialyser (n) 2 2 1 4 7 Partial clotting in the venous drip chamber (n) 0 6 6 5 6 Open in new tab Using the flat-sheet dialyser (Crystal-ST®) and intermittent administration of heparin, tapering doses was safe until a cut-point concentration comprised between the total dose of 3000 and 4000 IU. With doses equal to or higher than 4000 IU, massive or patchy clotting of dialyser was observed in one of 117 sessions whereas it was observed in 17 of 139 sessions for doses equal to or lower than 3000 IU (P = 0.001). Using the hollow-fibre dialyser (Nephral-ST®) and intermittent administration of heparin, a cut-point comprised between 3000 and 2000 IU was considered optimal. Massive or patchy clotting of the dialyser was observed in one of 150 sessions for doses equal to or higher than 3000 IU and in five of 39 sessions for doses equal to or lower than 2000 IU (P = 0.008). No clotting was observed for 30 sessions with a single dose of LMWH (20 mg of enoxaparin) administered at the start of the session. Using a single dose of 10 mg, seven of 29 sessions were associated with patchy clotting in the circuit. In Table 3 the results of clotting in the extracorporeal circuit observed with continuous infusion of unfractionated heparin are shown. It appears that the initial pulse administration of heparin may be decreased at least by 50% without increasing the risk of clotting. Coagulation parameters As expected, a positive relationship was observed between heparin dose and the maximum increase of APTT and anti-Xa levels within 2 h following the first pulse administration (data not shown). Values obtained after 180 min of haemodialysis were considered as the set point governing the ensuing anticoagulation status until the end of the session (Figures 3 and 4). They must be >40 s for APTT level and >0.2 IU/ml level for anti-Xa level to avoid a high risk of clotting by the end of the session, in accordance with the results of visual evaluation of clotting. In order to minimize the risk of clotting we estimated that the optimal dose would be equal to or higher than the mean value resulting in APTT level ≥40 s plus 1 SD. Thus, appropriate protocols allowed administration of 2000 IU of unfractionated heparin at session start, followed by a second pulse of 1000 IU at the second hour for the Nephral ST® series and, 2000 and 2000 IU, respectively, for the Crystal ST® series. Fig. 3. Open in new tabDownload slide APTT measured after 180 min of haemodialysis with the heparin-binding AN69ST membrane according to doses of unfractionated heparin administered throughout the session. A total dose higher than or equal to 3000 IU per session was significantly associated with a mean APTT value >40 s (P < 0.001) (diamonds) hollow-fibre dialyser; (squares) flat-sheet dialyser. Fig. 3. Open in new tabDownload slide APTT measured after 180 min of haemodialysis with the heparin-binding AN69ST membrane according to doses of unfractionated heparin administered throughout the session. A total dose higher than or equal to 3000 IU per session was significantly associated with a mean APTT value >40 s (P < 0.001) (diamonds) hollow-fibre dialyser; (squares) flat-sheet dialyser. Fig. 4. Open in new tabDownload slide Anti-Xa level measured after 180 min of haemodialysis with the heparin-binding AN69ST membrane according to doses of unfractionated heparin administered throughout the session. A total dose higher than or equal to 3000 IU per session was significantly associated with anti-Xa level higher than 0.2 IU/ml (P < 0.001) (diamonds) hollow-fibre dialyser; (squares) flat-sheet dialyser. Fig. 4. Open in new tabDownload slide Anti-Xa level measured after 180 min of haemodialysis with the heparin-binding AN69ST membrane according to doses of unfractionated heparin administered throughout the session. A total dose higher than or equal to 3000 IU per session was significantly associated with anti-Xa level higher than 0.2 IU/ml (P < 0.001) (diamonds) hollow-fibre dialyser; (squares) flat-sheet dialyser. Thrombin generation kinetics, measured by the appearance of plasma TAT complexes, demonstrated the efficiency of heparin binding onto the membrane with sustained anticoagulant activity. Figure 5 illustrates mean thrombin generation induced by blood contact with either the AN69 membrane or the AN69ST membrane. Maximal TAT levels were reached at 60 and 240 min with AN69 and AN69ST membrane, respectively, giving confirmation of the improved haemocompatibility of the modified membrane. The Nephral ST® dialyser appeared less thrombogenic than the Crystal ST® dialyser. Fig. 5. Open in new tabDownload slide Mean thrombin generation measured as plasma TAT complexes during haemodialysis in patients (n = 11) treated with either native AN69 or heparin binding AN69ST dialysers. Continuous lines refer to heparin doses of 4000 + 1000 IU and dotted lines refer to heparin doses of 4000 + 2000 IU. (–) and (– – ) for AN69ST flat-sheet dialyser; (-) and (- - - -) for AN69ST hollow-fibre dialyser; (. . . .) for AN69 flat-sheet dialyser (with 4000 + 2000 IU heparin). Fig. 5. Open in new tabDownload slide Mean thrombin generation measured as plasma TAT complexes during haemodialysis in patients (n = 11) treated with either native AN69 or heparin binding AN69ST dialysers. Continuous lines refer to heparin doses of 4000 + 1000 IU and dotted lines refer to heparin doses of 4000 + 2000 IU. (–) and (– – ) for AN69ST flat-sheet dialyser; (-) and (- - - -) for AN69ST hollow-fibre dialyser; (. . . .) for AN69 flat-sheet dialyser (with 4000 + 2000 IU heparin). Clinical validation of the anticoagulation protocol using reduced doses of heparin Postulating safety of reduced doses of heparin for haemodialysis with the heparin-coated AN69ST membrane, 32 patients were included in an open pilot study for 6 months. These patients were regularly haemodialysed with the AN69ST membrane for 6 months, using regular doses of heparin (5000 IU/l for priming, 4000 IU for loading dose and 2000 IU at the second hour). During this period, no significant clotting in the extracorporeal circuit was observed. Doses of unfractionated heparin were then reduced by 50% during the following 6-month period. However, the priming step remained unchanged. No massive clotting occurred during 2590 sessions and blood staining of apparent fibres on the outer surface of the dialyser was negligible. Blood haemoglobin and erythropoietin requirements remained unchanged, from 118 ± 3 g/l at baseline to 115 ± 2 g/l at the end of the study for blood haemoglobin and from 8262 ± 1170 to 8631 ± 1137 IU/week for i.v. doses of erythropoietin alpha. Tapering doses of heparin did not alter membrane permeability to small molecular weight plasma molecules as measured by online ionic conductance and urea clearance throughout sessions (data not shown). Behaviour of patients without systemic administration of heparin A total of 66 haemodialysis sessions included in the protocol were performed without systemic administration of heparin under careful supervision: only one session was concluded by massive clotting, 17 sessions were associated with patchy clotting of the dialyser. These results demonstrate further the haemocompatibility of the AN69ST membrane. Discussion Blood flow through the extracorporeal circuit activates both the intrinsic and extrinsic coagulation pathways in such a manner that thrombin is formed and converts fibrinogen into fibrin clot. In turn, thrombin stimulates release of factors enhancing platelet adhesion and aggregation. Circulating platelets adhere to the membrane surface. Platelets are activated by high-shear stress, and interaction with adsorbed fibrinogen, thus activating prothrombin. Activated monocytes and polymorphonuclear neutrophils contribute to the coagulation cascade with increased expression of tissue factor and release of reactive oxygen species. Differences in blood thrombogenicity should be considered in the choice of the dialyser. Conflicting results have been published on the thrombogenic effect of cellulose and synthetic membranes, which is difficult to assess in the presence of heparin. These results probably reflect heterogeneity of anticoagulation methods [3,5,15]. The present study demonstrates the decreased thrombogenicity of the AN69ST membrane after priming with a heparin saline solution as compared with the native AN69 membrane in similar conditions. This effect was attributed partly to heparin binding onto the membrane. Binding did not alter the anticoagulant effect of the molecule, as the anticoagulant capacity of the heparin-coated membrane was maintained for 2 h of haemodialysis, reducing systemic heparin requirements. For protocols with intermittent administrations, a second bolus of unfractionated heparin remains mandatory. The half-life of the anticoagulant effect is comprised between 90 and 120 min. Precise evaluation of dialysis membrane haemocompatibility, flow configuration and anticoagulant regimen are prerequisites for the clinical application of dialysis systems minimizing blood activation. In vitro studies have demonstrated the dramatic changes in haemocompatibility induced by surface neutralization of electronegative charges of the basic polyacrylonitrile AN69 membrane with polyethyleneimine [12–14]. The ensuing membrane lessens the procoagulant potential of the membrane surface to initiate thrombin generation and then, activation of the coagulation cascade. Ex vivo models in humans have been developed to test membrane thrombogenicity in clinical dialysis conditions on a one to 50 scale, measuring thrombin generation through TAT complexes, prothrombin fragments 1–2, platelet β-thromboglobulin, and complement C3a [2]. For ethical reasons, this model has not been applied to uraemic patients, which explains why we tested heparin-coated dialysers with the AN69ST membrane in sheep, prior to clinical use in uraemic patients. In the animal with normal renal function, the heparin-coated dialyser was used for haemodialysis without systemic administration of unfractionated heparin. The session was uneventful until the sixth hour, whereas massive clotting in the extracorporeal circuit occurred within the first hour of haemodialysis using the native AN69 membrane that does not bind significant amounts of heparin during the priming procedure. Quite a similar result was observed in a subgroup of 66 sessions performed in haemodialysed patients after rinsing the AN69ST membrane with heparinized solution but without subsequent systemic administration of heparin. These surprising results prompted us to extend the technique on a long-term basis to six patients at high risk of bleeding. In short, three patients had cholesterol embol disease, two had intestinal angiomatosis and one Crohn disease. 770 sessions with a blood flow rate >300 ml/min and a mean duration of 3 h were performed. Massive coagulation was documented in 10 and partial clotting in 18. The mean number of haemodialysis sessions per patient was 128 (from 20 to 216 sessions) (submitted for publication). In haemodialysed patients receiving the same anticoagulation protocol (‘regular doses of heparin’), a session using the AN69 membrane induced plasma TAT complex generation after 15 min of membrane contact, and a plateau was reached after 1 h. In contrast, occurrence of TAT complexes was delayed until the sixtieth minute of haemodialysis using the AN69ST membrane and a plateau concentration was not obtained at the end of the session. A rough estimation of plasma TAT area-under-the-curve (Figure 5) showed that the thrombin generation rate was reduced by one-third with the AN69ST heparin-coated membrane. Both rheological factors and, possibly, the industrial process of dialysers may explain the difference in TAT generation observed between hollow fibre and flat-sheet dialysers. The better efficiency of the Nephral ST® dialyser series deserves additional studies to be established. In chronically haemodialysed patients, tapering heparin doses to at least 50% of ‘regular’ doses, did not alter dialysis efficiency nor increased the risk of clotting. This perspective was documented in this multi-step study. To prevent clotting, we postulated that APTT and anti-Xa levels might be higher than 40 s and 0.2 IU/ml, respectively, thresholds recommended in many institutions [16]. The validation of this reduced regimen was documented in a 6-month open study that gathered more than 2500 haemodialysis sessions performed with the AN69ST membrane. Regular doses of heparin were reduced by 50%. Neither massive clotting in the extracorporeal circuit nor worsening of anaemia or increased erythropoietin requirements was observed. Grafting polyethyleneimine onto the AN69 polyacrylonitrile dialysis membrane decreases surface electronegativity [12]. Consequently, the modified membrane, called AN69ST, has new surface properties, repelling cationic plasma molecules. High molecular weight kininogen adsorption is decreased and contact phase activation reduced, irrespective of the pH value [11]. Making serum β2-microglobulin more acidic through advanced glycation decreases it’s binding to the AN69ST membrane without altering its protein-sieving coefficient [17]. Polyethyleneimine coating of polyacrylonitrile membrane strongly binds anionic heparin, including it into the blood-derived membrane that coats the synthetic polymer. This result is at variance with polyethylene glycol-grafted membranes, which do not influence heparin requirements [16]. It is postulated that heparin binds to the AN69ST membrane through ionic interactions [18,19]. It can also be assumed that heparin lessens the polymer binding of various cells and plasma proteins, including coagulation factors, hence improving the haemocompatibility of the dialysis membrane [20]. To conclude, although the postulated mechanisms by which the AN69ST membrane binds heparin are not well elucidated, heparin coating of the polymer surface preserves its anticoagulant properties. It has also been demonstrated that membrane permeability and biocompatibility remain unchanged. Consequently, heparin requirements during haemodialysis are strongly reduced, at least by 50%, without increasing the risk of clotting in the extracorporeal circuit. This advantage should be tested in the near future in patients at increased risk of bleeding whether they suffer from acute renal failure or are chronically haemodialysed. In addition, the long-term effects of reduced heparin administration on erythropoietin requirements, which could minimize occult bleeding, and the hypothetical effects on renal osteodystrophy and dialysis-induced hypertriglyceridaemia that may be associated with lifelong administration of heparin should be investigated. This study was partly supported by a research grant by the Centre Hospitalier et Universitaire and Hospal R&D International, Meyzieu, France. Conflict of interest statement. J.-L. Renaux is an employee of HOSPAL Company as Scientific Affairs Manager. None of the remaining authors declare any conflicts of interest. References 1 Clark WR, Hamburger RJ, Lysaght J. Effect of membrane composition and structure on solute removal and biocompatibility in hemodialysis. Kidney Int 1999 ; 56 : 2005 –2015 2 Seyfert UT, Helmling E, Hauck W, Skroch D, Albert W. Comparison of blood biocompatibility during haemodialysis with cuprophane and polyacrylonitrile membranes. Nephrol Dial Transplant 1991 ; 6 : 428 –434 3 Lindhout T. Biocompatibility of extracorporeal blood treatment. Selection of haemostatic parameters. Nephrol Dial Transplant 1994 ; 9 : 83 –85 4 Ouseph R, Ward RA. Anticoagulation for intermittent hemodialysis. Semin Dial 2000 ; 13 : 181 –187 5 Sagedal S, Hartmann A, Sundstrom K, Bjornsen S, Brosstad F. Anticoagulation intensity sufficient for hemodialysis does not prevent activation of coagulation and platelets. Nephrol Dial Transplant 2001 ; 16 : 987 –993 6 Sultan Y, London GM, Goldfarb B, Toulon P, Marchais SJ. Activation of platelets, coagulation and fibrinolysis in patients on long-term haemodialysis: influence of cuprophan and polyacrylonitrile membrane. Nephrol Dial Transplant 1990 ; 5 : 362 –368 7 Ourum E, Mollnes TE, Fosse E et al. Complement and granulocyte activation in two different types of heparinized extracorporeal circuits. J Thorac Cardiovasc Surg 1995 ; 110 : 1623 –1632 8 Fukutomi M, Kobayashi S, Niwaya K, Hamada Y, Kitamura S. Changes in platelet, granulocyte and complement activation during cardiopulmonary bypass using heparin-coated equipment. Artif Organs 1996 ; 20 : 767 –776 9 Wright MJ, Woodrow G, Umpleby S, Hull S, Brownjohn AM, Turney JH. Low thrombogenicity of polyethylene glycol-grafted cellulose membranes does not influence heparin requirements in hemodialysis. Am J Kidney Dis 1999 ; 34 : 36 –42 10 Wendel HP, Ziemer G. Coating techniques to improve the hemocompatibility of artificial devices used for extracorporeal circulation. Eur J Cardiothorac Surg 1999 ; 16 : 342 –350 11 Renaux JL, Thomas M, Crost T, Loughraieb N, Vantard G. Activation of the kallikrein-kinin system in hemodialysis: role of membrane electronegativity, blood dilution and pH. Kidney Int 1999 ; 55 : 1097 –1103 12 Moachon N, Boullanger C, Fraud S, Vial E, Thomas M, Quash G. Influence of the charge of low molecular weight proteins on their efficacy of filtration and/or adsorption on dialysis membranes with different intrinsic properties. Biomaterials 2002 ; 23 : 651 –658 13 Valette P, Thomas M, Dejardin P. Adsorption of low molecular weight proteins to hemodialysis membranes: experimental results and stimulations. Biomaterials 1999 ; 20 : 1621 –1634 14 Thomas M, Valette P, Mausset AL, Dejardin P. High molecular weight kininogen adsorption on hemodialysis membranes: influence of pH and relationship with contact phase activation of blood plasma-influence of pre-treatment with poly(ethyleneimine). Int J Artif Organs 2000 ; 23 : 20 –26 15 Mulvihill J, Crost T, Renaux JL, Cazenave JP: Evaluation of hemodialysis membrane biocompatibility by parallel assessment in an ex vivo model in healthy volunteers. Nephrol Dial Transplant 1997 ; 12 : 1968 –1973 16 Wright MJ, Woodrow G, Umpleby S, Hull Shaila, Brownjohn AM, Turney JH. Low thrombogenicity of polyethylene glycol-grafted cellulose membrane does not influence heparin requirements in hemodialysis. Am J Kidney Dis 1999 ; 34 : 36 –42 17 Randoux C, Gillery P, Georges N, Lavaud S, Chanard J. Filtration of native and glycated β2-microglobulin by charged and neutral dialysis membranes. Kidney Int 2001 ; 60 : 1571 –1577 18 Lonneman G, Schindler R, Luftt V, Mahiout A, Shaldon S, Koch KM: The role of plasma coating on the permeation of cytokine-inducing substances through dialyser membranes. Nephrol Dial Transplant 1995 ; 10 : 207 –211 19 Hofbauer R, Moser D, Frass M. Effect of anticoagulation on blood membrane interaction during hemodialysis. Kidney Int 1999 ; 56 : 1578 –1583 20 Cheung AK, Parker C, Wilcox L, Janatova J. Activation of complement by hemodialysis membranes: polyacrylonitrile binds more C3a than cuprophan. Kidney Int 1990 ; 37 : 1055 –1059 Author notes 1Service of Nephrology and Transplantation and Association Régionale pour la Promotion de la Dialyse à Domicile (ARPDD), University Hospital, Reims, 2Ecole Nationale Vétérinaire de Lyon, Marcy l’Etoile and 3Hospal International, Lyon, France European Renal Association–European Dialysis and Transplant Association
TI - Optimal anticoagulation strategy in haemodialysis with heparin-coated polyacrylonitrile membrane
JF - Nephrology Dialysis Transplantation
DO - 10.1093/ndt/gfg272
DA - 2003-10-01
UR - https://www.deepdyve.com/lp/oxford-university-press/optimal-anticoagulation-strategy-in-haemodialysis-with-heparin-coated-ongwl2cEaV
SP - 2097
EP - 2104
VL - 18
IS - 10
DP - DeepDyve
ER -