WARFARIN & OTHER COUMARIN ANTICOAGULANTS
The clinical use of the coumarin anticoagulants began with the discovery of an anticoagulant substance formed in spoiled sweet clover silage which caused hemorrhagic disease in cattle. At the behest of local farmers, a chemist at the University of Wisconsin identified the toxic agent as bishydroxycoumarin. A synthesized derivative, dicumarol and its congeners, most notably warfarin (Wisconsin Alumni Research Foundation, with “arin” from cou-marin added; Figure 34–5), were initially used as rodenticides. In the 1950s warfarin (under the brand name Coumadin) was introduced as an antithrombotic agent in humans. Warfarin is one of the most commonly prescribed drugs, used by approxi-mately 1.5 million individuals, and several studies have indicated that the drug is significantly underused in clinical situations where it has proven benefit.
Warfarin is generally administered as the sodium salt and has 100% bioavailability. Over 99% of racemic warfarin is bound to plasma albumin, which may contribute to its small volume of dis-tribution (the albumin space), its long half-life in plasma (36 hours), and the lack of urinary excretion of unchanged drug. Warfarin used clinically is a racemic mixture composed of equal amounts of two enantiomorphs.
The levorotatory S-warfarin is four times more potent than the dextrorotatory R-warfarin. This observation is use-ful in understanding the stereoselective nature of several drug inter-actions involving warfarin.
Coumarin anticoagulants block the γ-carboxylation of several glutamate residues in prothrombin and factors VII, IX, and X as well as the endogenous anticoagulant proteins C and S (Figure 34–2, Table 34–1). The blockade results in incomplete coagulation factor molecules that are biologically inactive. The protein carboxylation reaction is coupled to the oxidation of vitamin K. The vitamin must then be reduced to reactivate it. Warfarin prevents reductive metabolism of the inactive vitamin K epoxide back to its active hydroquinone form (Figure 34–6). Mutational change of the responsible enzyme, vitamin K epox-ide reductase, can give rise to genetic resistance to warfarin in humans and especially in rats.
There is an 8- to 12-hour delay in the action of warfarin. Its anticoagulant effect results from a balance between partially inhib-ited synthesis and unaltered degradation of the four vitamin K-dependent clotting factors. The resulting inhibition of coagula-tion is dependent on their degradation half-lives in the circulation. These half-lives are 6, 24, 40, and 60 hours for factors VII, IX, X, and II, respectively. Larger initial doses of warfarin—up to about 0.75 mg/kg—hasten the onset of the anticoagulant effect. Beyond this dosage, the speed of onset is independent of the dose size. The only effect of a larger loading dose is to prolong the time that the plasma concentration of drug remains above that required for sup-pression of clotting factor synthesis. The only difference among oral anticoagulants in producing and maintaining hypoprothrom-binemia is the half-life of each drug.
Warfarin crosses the placenta readily and can cause a hemorrhagic disorder in the fetus. Furthermore, fetal proteins with γ-carboxyglutamate residues found in bone and blood may beaffected by warfarin; the drug can cause a serious birth defect characterized by abnormal bone formation. Thus, warfarin should never be administered during pregnancy. Cutaneous necrosis with reduced activity of protein C sometimes occurs during the first weeks of therapy. Rarely, the same process causes frank infarction of the breast, fatty tissues, intestine, and extremities. The patho-logic lesion associated with the hemorrhagic infarction is venous thrombosis, suggesting that it is caused by warfarin-induced depression of protein C synthesis.
Treatment with warfarin should be initiated with standard doses of 5–10 mg rather than the large loading doses formerly used. The initial adjustment of the prothrombin time takes about 1 week, which usually results in a maintenance dose of 5–7 mg/d. The prothrombin time (PT) should be increased to a level represent-ing a reduction of prothrombin activity to 25% of normal and maintained there for long-term therapy. When the activity is less than 20%, the warfarin dosage should be reduced or omitted until the activity rises above 20%.
The therapeutic range for oral anticoagulant therapy is defined in terms of an international normalized ratio (INR). The INR is the prothrombin time ratio (patient prothrombin time/mean of normal prothrombin time for lab)ISI, where the ISI exponent refers to the International Sensitivity Index, and is dependent on the specific reagents and instruments used for the determination. The ISI serves to relate measured prothrombin times to a World Health Organization reference standard thromboplastin; thus the prothrombin times performed on different properly calibrated instruments with a variety of thromboplastin reagents should give the same INR results for a given sample. For most reagent and instrument combinations in current use, the ISI is close to 1, mak-ing the INR roughly the ratio of the patient prothrombin time to the mean normal prothrombin time. The recommended INR for prophylaxis and treatment of thrombotic disease is 2–3. Patients with some types of artificial heart valves (eg, tilting disk) or other medical conditions increasing thrombotic risk have a recom-mended range of 2.5–3.5.
Occasionally patients exhibit warfarin resistance, defined as progression or recurrence of a thrombotic event while in the therapeutic range. These individuals may have their INR target raised (which is accompanied by an increase in bleeding risk) or be changed to an alternative form of anticoagulation (eg, daily injec-tions of LMWH). Warfarin resistance is most commonly seen in patients with advanced cancers, typically of gastrointestinal origin (Trousseau’s syndrome). A recent study has demonstrated the superiority of LMWH over warfarin in preventing recurrent venous thromboembolism in patients with cancer.
The oral anticoagulants often interact with other drugs and with disease states. These interactions can be broadly divided into pharmacokinetic and pharmacodynamic effects (Table 34–2). Pharmacokinetic mechanisms for drug interaction with oral anticoagulants are mainly enzyme induction, enzyme inhibition, and reduced plasma protein binding. Pharmacodynamic mecha-nisms for interactions with warfarin are synergism (impaired hemostasis, reduced clotting factor synthesis, as in hepatic disease), competitive antagonism (vitamin K), and an altered physiologic control loop for vitamin K (hereditary resistance to oral anticoagulants).
The most serious interactions with warfarin are those that increase the anticoagulant effect and the risk of bleeding. The most dangerous of these interactions are the pharmacokinetic interactions with the mostly obsolete pyrazolones phenylbuta-zone and sulfinpyrazone. These drugs not only augment the hypoprothrombinemia but also inhibit platelet function and may induce peptic ulcer disease . The mecha-nisms for their hypoprothrombinemic interaction are a stereose-lective inhibition of oxidative metabolic transformation of S-warfarin (the more potent isomer) and displacement of albu-min-bound warfarin, increasing the free fraction. For this and other reasons, neither phenylbutazone nor sulfinpyrazone is in common use in the USA. Metronidazole, fluconazole, and trimethoprim-sulfamethoxazole also stereoselectively inhibit the metabolic transformation of S-warfarin, whereas amiodarone, disulfiram, and cimetidine inhibit metabolism of both enantiomorphs f warfarin. Aspirin, hepatic disease, and hyper-thyroidism augment warfarin’s effects—aspirin by its effect on platelet function and the latter two by increasing the turnover rate of clotting factors. The third-generation cephalosporins eliminate the bacteria in the intestinal tract that produce vitamin K and, like warfarin, also directly inhibit vitamin K epoxide reductase.
Barbiturates and rifampin cause a marked decrease of the anti-coagulant effect by induction of the hepatic enzymes that trans-form racemic warfarin. Cholestyramine binds warfarin in the intestine and reduces its absorption and bioavailability.
Pharmacodynamic reductions of anticoagulant effect occur with vitamin K (increased synthesis of clotting factors), the diuretics chlorthalidone and spironolactone (clotting factor concentration), hereditary resistance (mutation of vitamin K reactivation cycle molecules), and hypothyroidism (decreased turnover rate of clot-ting factors).
Drugs with no significant effect on anticoagulant therapy include ethanol, phenothiazines, benzodiazepines, acetamino-phen, opioids, indomethacin, and most antibiotics.
Excessive anticoagulant effect and bleeding from warfarin can be reversed by stopping the drug and administering oral or parenteral vitamin K1 (phytonadione), fresh-frozen plasma, prothrombin complex concentrates such as Bebulin and Proplex T, and recom-binant factor VIIa (rFVIIa). The disappearance of excessive effect is not correlated with plasma warfarin concentrations but rather with re-establishment of normal activity of the clotting factors. A modest excess of anticoagulant effect without bleeding may require no more than cessation of the drug. The warfarin effect can be rapidly reversed in the setting of severe bleeding with the administration of prothrombin complex or rFVIIa coupled with intravenous vitamin K. It is important to note that due to the long half-life of warfarin, a single dose of vitamin K or rFVIIa may not be sufficient.