Electronic Supplemental Material

Anticoagulation strategies in continuous renal replacement therapy: can the choice be evidence-based?

Oudemans-van Straaten HM, Wester JPJ, de Pont AC, Schetz MRC

Circuit survival time

Circuit survival time is generally used to judge the effect of anticoagulation. Comparison between studies is, however, hampered by the heterogeneity of patients, CRRT mode and dose, the complexity of factors influencing circuit survival and the practice of circuit change. Filters may be changed at a certain drop in ultrafiltrate flow, rise in transmembrane or prefilter pressure, clotting, or routinely. CRRT may also be interrupted because of procedures, recovering renal function, or withdrawal because of futility or death. Furthermore, before clotting in the filter occurs, permeability of the membrane declines, due to the deposition of proteins (‘clogging’). The decline in permeability is a gradual process, diminishing middle molecular permeability far before small molecular clearance. To measure the effect of an anticoagulant more precisely, and allow comparisons between studies, we need consensus criteria defining grades of loss of permeability.

There are several arguments to set an upper target for circuit survival time. Premature clotting is expensive and increases workload. Abrupt filter clotting causes loss of blood, while the routine change of filters prevents this loss. Furthermore, extremely long circuit survival may increase the risk of bacterial contamination of the circuit. This is reported in a continuous venovenous hemodialysis setting using ultra-pure (not sterile) dialysate [1]. Contamination started at a mean of about 50 h from start of therapy. Finally, the integrity of most silicone pump segments is guaranteed for no longer than 72 hours. The target for filter survival might be between 24 and 72 hours (Grade E).

Definition of bleeding

The relevance of bleeding depends on its severity, varying from slight oozing to hemorrhagic shock or intracranial bleeding. There is a large heterogeneity in the definition and grading of bleeding among the different studies. If bleeding is defined as transfusion requirement, blood loss in a clotted circuit, diagnostic sampling or critical illness-associated decreased erythropoiesis may contribute to this endpoint as well. If bleeding is defined as a decrease of hemoglobin, fluid balance may additionally influence the marker. Whether the reported bleeding is really caused by the anticoagulation or is an epiphenomenon related to underlying disease, is not always clear.

For valid comparisons between different anticoagulants, a well defined grading of bleeding is needed. Landefeld has validated a system to define minor and major bleeding [2]. A consensus definition of major bleeding has recently been presented by the Control of Anticoagulation Subcommittee of the International Society on Thrombosis and Haemostasis [3].

Monitoring of anticoagulation

Although monitoring of anticoagulation with thromboelastography provides additional information, the method does not appear to increase benefit in clinical practice [4-6].

Systemic anticoagulation

Uncontrolled studies evaluating the use of the different anticoagulants in continuous renal replacement therapy including > 10 patients are presented in Supplement Table 1 (S.T1) [7-11]. Relevant case reports and case series including less than 10 patients are mentioned in the reference list [12-14].

Alternative strategies in patients with heparin-induced thrombocytopenia (HIT)

Alternative strategies in patients with heparin-induced thrombocytopenia are summarized in S.T2. The choice depends on local availability and monitoring experience. It should be noted that HIT is a prothrombotic condition with increased in vivo thrombin generation. Therefore, if citrate is used for anticoagulation of the CRRT circuit, separate thromboprophylaxis should be applied. Inhibition of in vivo thrombin generation can be obtained via either direct inhibition of thrombin (IIa) (r-hirudin, bivaluridin or dermatan sulphate) or of anti-Xa (fondaparinux) or of both (nafamostat). The use of r-hirudin may be associated with severe complications (see main manuscript). Experience with the use of the specific anti-Xa inhibitor fondaparinux is not published yet. After some major bleeding experiences with r-hirudin, we (J.P.J. Wester and H.M. Oudemans-van Straaten) have lately used fondaparinux as a continuous intravenous infusion in addition to citrate in 15 patients with (suspected) HIT during CVVH. To attain a target anti-Xa of 0.25-0.35, we had to reduce the advised prophylactic dose of 2.5 mg/d after one to two days to 1.25 mg/d. Since HIT is associated with increased bleeding, the risk of bleeding should be weighed against the risk of thrombosis and filter clotting. Until now, there is no consensus about the degree of systemic anticoagulation to be attained in patient with HIT without manifest thrombosis. In case of major bleeding, discontinuation of systemic anticoagulation is recommended.

Regional anticoagulation with citrate

The differences between the various settings described in the literature are discussed by separating the effect of citrate as an anticoagulant and citrate as a buffer. The different settings and solutions for metabolic control are presented in S.T4 {14-36].

Citrate as an anticoagulant. For optimal anticoagulation, citrate dose is adjusted to blood flow. The normal concentration of citrate in blood is about 0.05 mmol/l. A target concentration of 3-5 mmol/l should be reached in the extracorporeal circuit to obtain an ionized Ca (iCa) concentration of < 0.35 mmol/l. Circuit iCa is the best marker to be used for monitoring the level of citrate anticoagulation. Some centers routinely measure iCa in the extracorporeal circuit and adjust the citrate dose accordingly. Others prefer using a fixed relation between citrate dose and blood flow, and do not monitor circuit iCa, thereby simplifying the procedure and delivering a constant buffer load to the patient (on condition that the blood flow is not changed e.g. due to catheter problems) (see S.T24 and Table 6 main document). The wide variety in citrate dose used in the different studies is striking. Citrate is either infused as a separate solution or added to the replacement fluid in some of the predilution settings. The latter setting does not guarantee a fixed relation between citrate flow and blood flow, because the flow of the replacement fluid varies with filtrate flow and desired fluid removal. The varying citrate dose might affect circuit survival time negatively.

Citrate as a buffer. Even if citrate is infused as a separate solution, a wide variety of formulations is used. Citrate is generally administered as a tri-sodium citrate (TSC) solution, but concentrations range from 2 to 30%; the lower the concentration, the higher the volume to be infused and stored. The different solutions also vary in the proportion of TSC to citric acid. For example, in the acid-citrate-dextrose (ACD)-A solution, 14% of the cations consist of hydrogen, in the TSC solution used by Oudemans [26] this percentage is 10% and in other solutions, all the cations are sodium. As a result, the proportion of sodium and hydrogen varies in relation to the citrate content and this influences strong anion difference and buffer capacity. The generation of bicarbonate and strong anions is related to the conversion of TSC to citric acid, which is metabolized in the citric acid circle, leaving sodium and bicarbonate:

Na3citrate + 3H2CO3 « Citric acid (C6H8O7) + 3NaHCO3
Therefore, the net amount of buffer generated depends on the proportion of sodium and hydrogen in the citrate solution and is lower at higher hydrogen content. The amount of citrate entering the systemic circulation depends on the amount of citrate infused and its loss by filtration or diffusion, and thus on continuous renal replacement dose (CRRT). At a higher dose more citrate is removed and less enters the systemic circulation, influencing acid-base balance. Convective clearance of citrate approximates the filtration rate, while the sieving coefficient with dialysis is about 0.87 [23]. Others found that citrate clearance by ultrafiltration is equal to urea clearance and does not change when CVVH is switched to CVVHD [28]. Acid-base balance is subsequently determined by cellular uptake and the metabolism of citric acid. If citric acid accumulates, metabolic acidosis ensues, since acid accumulates. Therefore, during citrate CVVH, the composition of the citrate solution, the citrate infusion rate, the filtration or diffusion rate and the metabolism of citric acid all influence acid base balance.
Composition of the fluids. Systems for citrate anticoagulation further differ in the composition of the dialysis and replacement fluids. Citrate is generally infused as a separate sodium citrate solution. In some predilution options citrate is included in the replacement fluid which contains no other buffer [18,35]. The latter option strikes by its simplicity, however at higher CVVH dose or high metabolic acid production additional bicarbonate is needed [36]. If citrate is added to the replacement fluid, the strict relation between blood flow and citrate flow, necessary for stable anticoagulation, is not guaranteed if filtrate rate changes. The substitution fluid used for predilution should not contain calcium, but the dialysate can [34]. In each of the options, the sodium and buffer content of the dialysate and replacement fluid are different; they are calculated for the type and rate of citrate infusion, and the dose of CRRT. In general, additional bicarbonate is replaced if a higher CRRT dose is applied [27,37]. To compensate for additional loss of calcium and magnesium, bound to citrate, calcium is administered by a separate infusion into the systemic circulation. Although close metabolic control can be obtained if the protocol is strictly followed, metabolic alkalosis, metabolic acidosis and hypocalcemia can occur.
Adjustment of metabolic derangement. In general, different options for adjustment of metabolic alkalosis or acidosis are described, either by adjusting the bicarbonate concentration in the dialysis- or replacement fluid [17,19,26,27,29] or by varying the citrate infusion rate [20,31] (see S.T4). By adjusting the composition of the replacement fluid, a highly flexible and close metabolic control can be obtained without affecting the level of anticoagulation. By adjusting citrate flow, the level of anticoagulation may become compromised. To prevent metabolic alkalosis in case of a declining filtrate flow (more citrate enters the systemic circulation), it is advised to change the filter, if filtrate flow drops below a certain limit. Depending on blood flow and citrate flow, this limit has to be set in the local protocol. Several authors describe the less elegant escape of HCl infusion (S.T4). It is important to note that the removal of sodium and citrate also changes if blood flow is changed while citrate flow is kept constant, since the concentration of sodium and citrate in the filter changes. This effect is small and probably not clinically relevant. It is however mandatory to calculate solute balances for each combination of blood flow, citrate flow and filtrate flow before implementation of citrate anticoagulation in the ICU. A local protocol should be written and strictly followed.

Future perspective for early clotting

A potential perspective to prevent early clotting might be the use of tirofiban, a short-acting platelet glycoprotein IIb/IIIa antagonist, to counteract the platelet activating effect of UFH. Half-life is prolonged in renal failure, but ultrafiltration seems an effective means of elimination [38].

HIT

An interesting case report on the management of a patient with HIT is found in reference 39.

References

  1. Kanagasundaram NS, Larive AB, Paganini EP (2003) A preliminary survey of bacterial contamination of the dialysate circuit in continuous veno-venous hemodialysis. Clin Nephrol 59:47-55

2.  Landefeld CS, Anderson PA, Goodnough LT, Moir TW, Hom DL, Rosenblatt MW, Goldman L (1989) The bleeding severity index: validation and comparison to other methods for classifying bleeding complications of medical therapy. J Clin Epidemiol 42:711-708

3.  Schulman S, Kearon C (2005) Definition of major bleeding in clinincal investigations of antihemostatic medical products in non-surgical patients. J Thromb Haemost 3:692-694.

  1. Derrier M, Jambou P, Kaidomar M, Ichai C, Bayle J, Grimaud D (1995) Thromboelastography and monitoring of coagulation in patients undergoing continuous venovenous hemofiltration. Contrib Nephrol 116:159-62
  2. Holt AW, Bierer P, Glover P, Plummer JL, Bersten AD (2002) Conventional coagulation and thromboelastograph parameters and longevity of continuous renal replacement circuits. Intensive Care Med 28:1649-1655
  3. Baldwin I, Tan HK, Bridge N, Bellomo R (2000) A prospective study of thromboelastography (TEG) and filter life during continuous veno-venous hemofiltration. Ren Fail 22:297-306
  4. Gretz N, Quintel M, Ragaller M, Odenwalder W, Bender HJ, Rohmeiss P, Strauch M (1995) Low-dose heparinization for anticoagulation in intensive care patients on continuous hemofiltration..Contrib Nephrol 116:130-135

8.  Jeffrey RF, Khan AA, Douglas JT, Will EJ, Davison AM (1993) Anticoagulation with low molecular weight heparin (Fragmin) during continuous hemodialysis in the intensive care unit. Artif Organs 17:717-720

9.  Knight DW, Selwyn D, Girling K (2003) Low-molecular-weight heparin for anticoagulation during continuous venovenous hemofiltration Arch Int Med 163;981

10.  Lindhoff-Last E, Betz C, Bauersachs R (2001) Use of a low-molecular-weight heparinoid (danaparoid sodium) for continuous renal replacement therapy in intensive care patients. Clin Appl Thromb Hemost 7:300-304

11.  Fiaccadori E, Maggiore U, Rotelli C, Minari M, Melfa L, Cappe G, Cabassi A (2002) Continuous haemofiltration in acute renal failure with prostacyclin as the sole anti-haemostatic agent. Intensive Care Med 28:586-593

12.  Dager WE, White RH (2003) Argatroban for heparin-induced thrombocytopenia in hepato-renal failure and CVVHD. Ann Pharmacother 37:1232-1236

  1. Tang IY, Cox DS, Patel K, Reddy BV, Nahlik L, Trevino S, Murray PT (2005) Argatroban and renal replacement therapy in patients with heparin-induced thrombocytopenia. Ann Pharmacother 39:231-236
  2. Williamson DR, Boulanger I, Tardif M, Albert M, Gregoire G (2004) Argatroban dosing in intensive care patients with acute renal failure and liver dysfunction. Pharmacotherapy 24:409-414
  3. Mehta RL, McDonald BR, Ward DM (1991) Regional citrate anticoagulation for continuous arteriovenous hemodialysis. An update after 12 months. Contrib Nephrol 93:210-214
  4. Ward DM, Mehta RL (1993) Extracorporeal management of acute renal failure patients at high risk of bleeding. Kidney Int 43:S237-S244
  5. Aspner R, Druml W (1998) More on anticoagulation for continuous hemofiltration. Letter to the Editor. New Engl J Med 338:131-132

18.  Thoenen M, Schmid ER, Binswanger U, Schuepbach R, Aerne D, Schmidlin D (2002) Regional citrate anticoagulation using a citrate-based substitution solution for continuous venovenous hemofiltration in cardiac surgery patients. Wien Klin Wochenschr 114(3):108-114

  1. Hofmann RM, Maloney C, Ward DM, Becker BN (2002) A novel method for regional citrate anticoagulation in continuous venovenous hemofiltration (CVVHF). Ren Fail 24:325-335
  2. Gabutti L, Marone C, Colucci G, Duchini F, Schonholzer C (2002) Citrate anticoagulation in continuous venovenous hemodiafiltration: a metabolic challenge. Intensive Care Med 28:1419-1425
  3. Tobe SW, Aujla P, Walele AA, Oliver MJ, Naimark DM, Perkins NJ, Beardsall M (2003) A novel regional citrate anticoagulation protocol for CRRT using only commercially available solutions. J Crit Care 18:121-129
  4. Mitchell A, Daul AE, Beiderlinden M, Schafers RF, Heemann U, Kribben A, Peters J, Philipp T, Wenzel RR (2003) A new system for regional citrate anticoagulation in continuous venovenous hemodialysis (CVVHD). Clin Nephrol 59:106-114
  5. Swartz R, Pasko D, O'Toole J, Starmann B (2004) Improving the delivery of continuous renal replacement therapy using regional citrate anticoagulation. Clin Nephrol 61:134-143
  6. Morgera S, Scholle C, Voss G, Haase M, Vargas Hein O, Krausch D, Melzer C, Rosseau S, Zuckermann-Becker H, Neumayer HH (2004) Metabolic complications during regional citrate anticoagulation in continuous venovenous hemodialysis: single-center experience. Nephron Clin Pract 97:c131-136
  7. Monchi M, Berghmans D, Ledoux D, Canivet JL, Dubois B, Damas P (2004) Citrate vs. heparin for anticoagulation in continuous venovenous hemofiltration: a prospective randomized study. Intensive Care Med 30:260-265

26.  Oudemans-van Straaten HM (2004) Review and guidelines for regional anticoagulation with citrate in continuous hemofiltration. Neth J Crit Care 8: 146-156