Warfarin

Warfarin is a frequently used classic vitamin K antagonist (VKA) for the prevention of thromboembolic events. However, it is limited by its narrow therapeutic index; unpredictable pharmacokinetics and pharmacodynamics; multiple drug interactions; a need for frequent monitoring of levels; food interactions; adverse bleeding effects; and, unfortunately, the induction of hypercoagulable states. In addition to these limitations, physicians rely heavily on patient compliance in regard to dosage and frequent monitoring.

Warfarin is primarily metabolized by the CYP2C9 hepatic microsomal enzyme system. This system is inducible by many other medications and carries genetic variability that can alter activity ( Table 82-1 ). Warfarin is strongly protein bound, and it is the non–protein-bound fraction that is biologically active. The drug is water soluble and is highly absorbed after oral administration, mostly in the proximal small bowel. The biological half-life is 36 to 42 hours.

Warfarin interferes with the biosynthesis of the vitamin-K-dependent coagulation factors II, VII, IX, and X as well as the natural anticoagulant proteins C and S. Because of these contradictory effects, warfarin and other VKAs produce procoagulant and prothrombotic effects. The desired anticoagulant effects are delayed by approximately 36 to 72 hours depending on the clearance of normal clotting factors, particularly prothrombin, from the circulation.

Monitoring of warfarin levels is done by measurement of the international normalized rate (INR), defined as the ratio of the patient’s prothrombin time (PT) to a normal sample (control). A therapeutic INR is defined according to the indication for anticoagulation: For venous thromboembolism (VTE) prophylaxis, a therapeutic INR is typically in the range of 2.0 to 3.0. In the setting of mechanical heart valves, higher therapeutic goals for the INR are recommended.

Unfractionated Heparin

Unfractionated heparin (UFH) was discovered in 1916, and its first human trial was conducted in 1935. UFH potentiates the action of antithrombin III, inactivating thrombin and activated coagulation factors IX, X, XI, XII and plasmin, thereby preventing the conversion of fibrinogen to fibrin. Heparin is primarily metabolized by the liver, but it may be partially metabolized in the reticuloendothelial system. The elimination half-life of heparin when discontinued from a steady state is approximately 1 to 2 hours. Because anticoagulation with UFH can be challenging in the individual patient, it is common to use dosing protocols that guide dosing to reach a goal therapeutic activated partial thromboplastin time (aPTT) and thereafter to guide maintenance of the aPTT in the goal range. Importantly, the efficacy of this approach has not been truly tested.

UFH has been the traditional parenteral agent for anticoagulation. It is ubiquitous in the ICU for prevention of DVT and PE in diverse acute patient populations. UFH is also the preferred anticoagulant in severe renal failure (creatinine clearance <30 mL/min). The short half-life of UFH offers the advantage of quick reversal of anticoagulant effects when needed. UFH can be rapidly reversed with protamine sulfate (1 mg/100 U heparin). However, protamine can trigger anaphylaxis, particularly in patients with insulin-dependent diabetes and fish allergies. Specific ICU conditions that require the use of UFH have not been identified.

A rare but serious complication of heparin exposure is heparin-induced thrombocytopenia (HIT), a syndrome in which antibodies to the complex of heparin and platelet factor IV trigger platelet activation that causes major arterial and/or venous thrombosis. Treatment of this life-threatening complication is to terminate heparin exposure and to anticoagulate with a nonheparin alternative such as a direct thrombin inhibitor (DTI).

Low-Molecular-Weight Heparins

Heparin is a naturally occurring polysaccharide consisting of molecular chains of varying lengths or molecular weights. The low-molecular-weight heparins (LMWHs) are fractionated from heparin to yield only short polysaccharide chains. The main LMWHs in clinical practice are enoxaparin, dalteparin, and tinzaparin. Tinzaparin is currently not available in the United States. The advantages of LMWH relative to UFH are greater bioavailability, longer duration of anticoagulant action, fixed dosing, a lack of need for laboratory monitoring, and a lower risk of HIT. Multiple meta-analyses indicate that subcutaneous LMWH is more effective than UFH for the treatment of VTE, exhibiting higher rates of thrombus regression and lower rates of recurrent thrombosis, major bleeding, and mortality. Despite these presumed clinical advantages of LWMH, randomized trials of LMWH versus UFH for thromboprophylaxis in the ICU have yielded inconsistent results. A recently completed large, multicenter, randomized trial in 3754 ICU patients compared dalteparin with twice-daily UFH for thromboprophylaxis. There was no difference (hazard ratio 0.92; 95% confidence interval [CI], 0.68 to 1.23; P = .57) between groups in the primary outcome variable—the incidence of proximal leg DVT. However, dalteparin significantly lowered the incidence of PE (hazard ratio 0.51; 95% CI, 0.30 to 0.88; P = .01) and HIT (hazard ratio, 0.27; 95% CI, 0.08 to 0.98; P = .046). Importantly, UFH administered three times per day has been shown to be superior to twice-daily dosing, a fact not accounted for in the above trial.

LMWH can lower the risk of HIT. A recent meta-analysis identified a lower incidence of HIT in postoperative patients undergoing thromboprophylaxis with LMWH when compared with UFH (risk ratio 0.25; 95% CI, 0.07 to 0.82; P = .02). In the analysis for HIT complicated by VTE, LMWH was associated with an 80% risk reduction for this complication compared with UFH (risk ratio 0.20; 95% CI, 0.04 to 0.90; P = .04). Although these analyses are suggestive, further high-quality trials are essential. There are disadvantages to the use of LMWH in the ICU. These include variations in efficacy in obese patients and underweight elderly patients. Furthermore, the lack of a routine test to measure the effects of LMWH can be problematic in patients in whom bleeding is particularly dangerous. In addition, although dalteparin (5000 IU/day) did not bioaccumulate in critically ill patients with severe renal dysfunction (creatinine clearance <30 mL/min), other LMWHs that are known to be renally cleared have not been examined. The investigators demonstrated that dalteparin at a daily dose of 5000 IU did not bioaccumulate and that there was no excessive bleeding risk. Overall, further large randomized trials are required to evaluate for clinical superiority over UFH in the critically ill before LWMH will be more widely adopted for thromboprophylaxis in the ICU.

Intravenous Direct Thrombin Inhibitors

In contrast to the heparins, DTIs provide anticoagulation regardless of antithrombin III levels and have the ability to inhibit fibrin-bound thrombin, allowing more complete anticoagulant activity. DTIs also are more predictable because they do not bind to other plasma proteins. Currently available intravenous DTIs include lepirudin, desirudin, bivalirudin, and argatroban. The last two are used most commonly because they have a broader range of approved indications.

Bivalirudin is a hirudin analogue that directly binds to thrombin, leading to an anticoagulant effect within 5 minutes. This binding is reversible because of cleavage by thrombin, which leads to a short half-life (25 minutes) in patients with normal or mildly reduced renal function. Metabolism of the drug is primarily hepatic and proteolytic, but 20% of bivalirudin’s clearance is via the kidney, slightly prolonging the half-life in patients with moderate renal dysfunction (creatinine clearance of 30 to 59 mL/min). Monitoring of the anticoagulant effect can be performed with the activated clotting time (ACT). Although no rapid reversal agent exists, the drug is cleared by hemodialysis. The role of this agent in cardiac catheterization and cardiac surgery is emerging, but its value in the ICU remains undefined.

In the ICU setting, argatroban is an alternative parenteral DTI to bivalirudin. This drug also reversibly binds to an active site on thrombin, causing direct inhibition. The primary use of this DTI in the ICU setting is for anticoagulation in critically ill patients with HIT. Because of its rapid onset and short half-life, argatroban is typically given as an infusion. It is primarily metabolized by the liver; thus, dosing must be adjusted in patients with hepatic failure, but this is not necessary in renal dysfunction. Monitoring can be performed with either an ACT or aPTT. The goal aPTT is often 1.5 to 3 times that of the baseline value. Because argatroban affects thrombin-dependent coagulation tests, PT and INR may be altered, an important consideration when transitioning to warfarin. Formal studies of argatroban in critically ill patients have not been reported.

Parenteral Indirect Synthetic Factor Xa Inhibitors

Fondaparinux is an indirect factor Xa inhibitor that is a synthetic analog of a natural pentasaccharide contained in heparin and LMWH that interrupts the coagulation cascade upstream of thrombin. The lack of thrombin inhibition prevents rebound thrombin generation.

Fondaparinux has a half-life of 17 to 21 hours and is administered as a daily subcutaneous injection. Peak plasma concentrations are achieved within 2 hours of injection. A recent trial demonstrated that the bioavailability of fondaparinux after subcutaneous injection was not significantly affected by vasopressor therapy in critically ill patients. Clearance is significantly decreased in the setting of renal failure. The effects of fondaparinux can be followed with serial measurement of factor Xa activity, but routine monitoring is not recommended. The PT, INR, and aPTT typically are not affected. There is no specific reversal agent for fondaparinux, although recombinant factor VIIa may reverse its effects.

In a randomized controlled trial (N = 849 acute medical patients >60 years, 35 centers from 8 countries), fondaparinux as compared with placebo significantly reduced the risk of VTE by 46.7% from 10.5% to 5.6% ( P < .05) with no increase in bleeding risk. A meta-analysis (cumulative N > 13,000 medical and surgical patients, 8 randomized trials) confirmed a one-fifth reduction in mortality from VTE with fondaparinux as compared with the control groups of placebo or LMWH. Beyond thromboprophylaxis for VTE, a series of randomized controlled trials have established a role for fondaparinux in the management of ACSs managed with and without PCI. Fondaparinux may also have a role in the treatment of HIT in selected patients.

Oral Direct Thrombin Inhibitors

Dabigatran etexilate is an oral DTI. The oral form is a prodrug that is activated by nonspecific esterases. As discussed previously, DTIs inhibit free and fibrin-bound direct thrombin inhibition, an advantage over heparin, which is less effective at inhibiting the latter. Dabigatran has a rapid onset, reaching peak plasma concentrations within 1.5 hours, and a half-life of 12 to 14 hours. Adjustment for renal dysfunction is required because 80% of the drug undergoes renal elimination. There is no known hepatotoxicity.

The clinical use of dabigatran is expanding, and applications can occur in the ICU setting. Current indications include prophylaxis and treatment of VTE as well as stroke and thromboembolic prophylaxis in nonvalvular atrial fibrillation.

A large randomized trial compared twice-daily dabigatran with warfarin in atrial fibrillation patients (N > 18,000) who were at increased risk for stroke. In patients receiving a high dose (150 mg given twice daily), the incidence of stroke during the median 2-year follow-up was lower than warfarin, without an increased risk of major bleeding. Given this promising data on this new oral anticoagulant option, recent guidelines now endorse dabigatran as an alternative to warfarin in selected patients with atrial fibrillation (Class I recommendation; level of evidence B).

Monitoring is problematic: the thrombin time is oversensitive to dabigatran’s effect. The ecarin clotting time, another option, is not a routinely available test, and the absolute aPTT value does not correlate to actual concentrations of dabigatran. Reversal may require activated charcoal or hemodialysis. Fresh frozen plasma is not effective in reversing dabigatran-related bleeding, but clinical hemostasis can be achieved with activated prothrombin complex concentrates or recombinant factor VIIa. Because use in the critical care environment is limited and because it is an oral agent (subject to unpredictable absorption in the critically ill) that lacks a method for monitoring anticoagulant activity, use of dabigatran in the ICU requires further investigation.