This site is intended for health professionals only
Consultant Renal Physician/Honorary Senior Lecturer
UCL Centre for Nephrology
Royal Free and University College Medical School
Intermittent haemodialysis is the most common mode of renal replacement therapy for patients with end-stage kidney disease worldwide. For most patients anticoagulation is needed to prevent clotting in the extracorporeal circuit. Typically, unfractionated and low-molecular-weight heparins are used. Unfortunately, in a minority of patients, antibodies to heparin can develop, with a reported prevalence of 0–6%.(1) Interaction between heparins and platelet factor 4 can lead to conformational change in the platelet factor 4 molecule, with the exposure of novel antigenic epitopes, resulting in IgG isotype auto-antibody formation. These antibodies bind to heparin-platelet factor 4 complexes, activating platelets, leading to thrombocytopenia and arterial and venous thrombosis. Key management is to withdraw all heparins, as exposure even to minor amounts, as found in heparin locks for central venous access catheters, can result in death.(2) Once all heparins have been withdrawn, alternative systemic anticoagulants are required.
Heparin-induced thrombocytopenia (HIT) is more likely to occur in intensive care patients with acute kidney failure than stable thrice-weekly haemodialysis patients. In the ICU there are many potential causes of thrombocytopenia, and the “4Ts” scoring system has been developed as a way to assess the risk of developing HIT (see Table 1).(3) In a patient with moderate HIT risk, all heparins should be withdrawn while awaiting the result of an ELISA or platelet aggregation HIT antibody test. In the general medical and surgical patient, the risk of developing thrombosis in the first month after developing HIT is very high.(4) However, some patients develop the antibody without clinical sequelae. The term heparin-induced thrombocytopenia thrombosis syndrome (HITTS) is used to cover those with the antibody who develop clinical thrombosis.
Alternative systemic anticoagulants
UK guidelines recommend danaparoid, comprising a mixture of glycosaminoglycans, predominantly heparin sulphate, with additional dermatan and chondroitin sulphate, for patients with HIT.(5) Danaparoid probably acts by binding to heparin cofactor II and antithrombin, inhibiting the conversion of prothrombin to thrombin, and as such requires anti-Xa monitoring. Danaparoid is renally excreted, and as such has an increased t(½) in dialysis patients, to > 36 h. Although the molecular weight is 5.5 kDa, due to charge, very little is removed during dialysis and/or haemofiltration. Thus there is a potential for danaparoid to accumulate in kidney-failure patients.
For systemic anticoagulation, the danaparoid dose has to be reduced in patients with kidney failure. Dosages similar to those used for continuous renal replacement therapies (CRRT) can be utilised, with an initial bolus of 750 IU, then a continuous infusion, starting at 150 IU/h, and then adjusting the dose to maintain an anti-Xa activity of 0.35–0.6 IU/ml. In our clinical practice most patients require 100–150 IU/h.
In the laboratory there may be crossreactivity between danaparoid and HIT antibodies in up to 10% of cases, but there are few reports of clinical sequelae. In cases of bleeding due to overanticoagulation with danaparoid there is no specific antidote; activated factor VII concentrates may be needed. Dermatan sulphate, one of the constituents of danaparoid, was used some years ago for intermittent haemodialysis, using a 6 mg/kg loading dose, and for CRRT, using an initial 150 mg bolus followed by an infusion rate of 15 mg/h, titrated to achieve a aPTT ratio < 1.5.
Fondaparinux and idraparinux are synthetic pentasacharides which bind to antithrombin and are renally excreted. Fondaparinux is not licensed for haemodialysis, but preliminary data suggest that due to its prolonged t(½), an alternate-day dose of 2.5 mg in patients with kidney failure provides systemic anticoagulation, adequate for dialysis. The dose should be adjusted, as with danaparoid, according to anti-Xa activity. In the laboratory, crossreaction has been reported with HIT antibodies, but there are no reports of clinical sequelae. As with danaparoid there is no specific overanticoagulation antidote.
Direct thrombin inhibitors
These prevent thrombin from catalysing the conversion of fibrinogen to fibrin. Hirudin (lepirudin) irreversibly binds thrombin, and is renally excreted, so the t(½) is increased markedly in kidney dialysis patients (up to 50 hours), and even more so in anephric subjects. Continuous exposure to hirudin usually leads to antibody formation, and these antibodies extend the t½ still further. Thus hirudin tends to accumulate in kidney dialysis patients and dosing becomes difficult as some hirudin is removed during high flux dialysis or continuous haemodiafiltration (greater removal with polysulphone compared with polyamide membranes), although once antibodies have formed, the complexes are too big to be filtered.(6) Also, the relationship between plasma hirudin concentration and the activated partial thromboplastin time (aPTT) is not linear, and what seems to be a relatively small increase in aPTT may well represent a large increase in hirudin, with raised risk of haemorrhage. To overcome this problem a new generation of ecarin clotting tests (ECTs) has been developed to monitor hirudin activity. Due to the difficulty of dosing hirudin in kidney dialysis patients, we have changed from administering hirudin by infusion (0.01–0.015 mg/kg/h) to giving a 0.4 mg/kg (up to 50 mg) bolus, aiming to maintain an aPPT ratio of around 2.0, or plasma hirudin concentration of 0.6–1.4 µg/ml and an ECT of 60–80 s, then administersing further smaller boluses when the aPPT ratio has fallen to ≤ 1.0.
In cases of bleeding due to excess hirudin there is no antidote, and only anecdotal reports of successful use of factor VIIa concentrates. Hirudin can be cleared by haemodialfiltration, and plasma exchange has been used in cases of antihirudin antibodies. Some patients previously exposed to hirudin may suffer anaphylaxis when rechallenged.
Biliverdin is licensed for treating cardiac and not kidney patients, but has potential advantages: it is a reversible thrombin inhibitor with a much shorter t(½), as it is degraded by thrombin in plasma; it is removed by haemofiltration; and it appears to involve much less risk of antibody development.
Argatroban® has been available for treating HIT patients in the USA, and is being introduced into Europe (as Arganova® in the Netherlands, Argatra® in Germany and Austria and Novostan® in Scandanavia). Argatroban is synthesised from
L-arginine and is mainly hepatically metabolised, and its t½ is only moderately prolonged in kidney dialysis patients. A reversible thrombin inhibitor, it requires an initial bolus of 250 µg/kg, then a continuous infusion of 2 µg/kg/min (with lower doses needed in liver disease, 0.5 µg/kg/min), adjusted to maintain an aPTT ratio of 2.0–2.5. Other synthetic reversible direct thrombin inhibitors similar to melagatran are undergoing phase II and III clinical trials.
In Japan, the serine protease nafamostat maleate is also used to anticoagulate kidney dialysis patients with HITTS. It has a short t(½), measured in minutes, and after an initial bolus of 40 mg, an infusion of 20–40 mg/h is usually required to maintain an aPTT ratio of 2–2.5. Nafamostat has been occasionally reported to cause anaphylactic reactions, eosinophilia and bone marrow suppression. As with hirudin, the amount removed during haemofiltration/diafiltration depends on the dialyser membrane composition.
Patients with HITTS require systemic anticoagulation. Until recently in Europe the choice was between the heparinoid danaparoid and the irreversible direct thrombin inhibititor hirudin. The pharmacokinetics of hirudin, plus difficulty in monitoring, led the UK Haematology Standards Committee to recommend danaparoid. However, a newer generation of reversible synthetic direct thrombin inhibitors is being introduced, with others in development.
1. Davenport A. Management of heparin-induced thrombocytopenia during renal replacement therapy. Hemodial Int 2001;5:81-5.
2. Davenport A. Sudden collapse during haemodialysis due to immune-mediated heparin-induced thrombocytopenia. Nephrol Dial Transplant 2006;21:1721-4.
3. Warkentin TE, Cook DJ. Heparin, low molecular weight heparin, and heparin-induced thrombocytopenia in the ICU. Crit Care Clin 2005;21:513-29.
4. Napolitano L, Warkentin TE, Al-Mahameed A, Nasraway SA. Heparin-induced thrombocytopenia in the critical care setting: diagnosis and management. Crit Care Med 2006;34:2898-911.
5. Keeling D, Davidson S, Watson H. Haemostasis and Thrombosis Task Force of the British Committee for Standards in Haematology. The management of heparin-induced thrombocytopenia.
Br J Haematol 2006;133:259-69.
6. Benz K, Nauck MA, Bohler J, Fischer KG. Hemofiltration of recombinant hirudin by different hemodialyzer membranes: implications for clinical use. C J Am Soc Nephrol 2006;2:470-6.