The general determinants of the response to oral anticoagulant drugs are critically reviewed, and the specific physiologic, pathologic, and pharmacologic factors that increase and decrease the responsiveness to these agents are examined with needful scrupulosity. Because the experimental data in laboratory animals supporting the clinical application of anticoagulant drugs are controversial, their employment in patients is essentially empirical and based on clinical trials. The basic premise of oral anticoagulant therapy is that interference with hemostasis reduces morbidity and mortality from thromboembolic disorders. The latest experimental research indicates that blood coagulation involves the occurrence of both enzymatic and stoichiometric reactions in the conversion of circulating precursors to activated clotting factors and in the formation of complexes composed of one clotting factor acting as a prosthetic group for another, respectively. Thrombogenesis and coagulation are similar processes in that thrombus formation is a hemostatic event that occurs intravascularly. In the development of a thrombus the blood platelet is now known to occupy the quintessential position. Since the platelet plug forms the bulk of an arterial thrombus, the best therapeutic strategy should be antithrombotic agents that interfere with the adherence of platelets to vessel walls and to each other. Since the blood clot forms the bulk of a venous thrombus and a pulmonary embolus, the best therapeutic strategy should be anticoagulant drugs that interfere with the coagulation of blood. In practice, oral anticoagulants have little effect in arterial thrombotic disease, in contrast to their significant therapeutic effect in venous thrombotic disease and in emboli from mural thrombi of the heart.
The mechanism of action of oral anticoagulants proceeds from their competitive effect with vitamin K activity at a ribosomal level or at a subsequent step in which the conversion of peptide precursors to the vitamin K-dependent clotting factors is promoted by the vitamin and retarded by the anticoagulant. The pharmacodynamics in man of sodium warfarin, the most prescribed oral anticoagulant drug in the United States, include rapid gastrointestinal absorption and slow plasma elimination, a biologic effect lasting for days, a significantcorrelation between the plasma drug levels and the degree and duration of hypoprothrombinemia, a high degree of binding to plasma albumin, an apparent volume of distribution the size of the albumin space, and a lack of urinary excretion of the unchanged drug. Hydroxylation is the major pathway of warfarin and Tromexan metabolism in man and rat. Thermodynamic analysis of the warfarin-albumin binding process studied by equilibrium dialysis and by direct calorimetry shows an exothermic reaction with negative free energy, which suggests hydrogen bonding and hydrophobic bonding in the interaction. The lesser albumin binding, the absent anticoagulant effect, and the urinary excretion of the hydroxylated metabolites suggest that the addition by metabolism of a polar hydroxyl group onto the coumarin nucleus reduces its hydrophobic bonding to plasma albumin and hepatic receptor sites. The speed of onset of the "antithrombotic effect" of oral anticoagulants, the therapeutic reduction of factors II, IX, and X, is the same whether therapy is initiated with or without a loading dose of the drug. Although many modifications of the one-stage prothrombintest have been proposed, therapy is best controlled by the original method.
Many nonhereditary factors may cause increased responsiveness to oral anticoagulant drugs. A deficiency of one of the sources of vitamin K, from a lack of either dietary ingestion or intestinal synthesis, increases responsiveness, and a deficiency of both sources causes hypoprothrombinemia. The potentiation of the anticoagulant effect in patients with liver disease is not altered by administration of vitamin K. The increased responsiveness in patients with hypermetabolic states such as fever and hyperthyroidism probably results from increased catabolism of the vitamin K-dependent clotting factors. Many kinds of stress in animals, with or without concomitant anticoagulant therapy, result in hemorrhage. Anticoagulant drugs have been used surreptitiously to produce factitious disease, suicide, and even murder. Surreptitious noningestion of anticoagulants can be falsely interpreted as resistance to anticoagulants. The potential danger of moderate aspirin therapy, up to 3 g a day, during long-term therapy with anticoagulants results not from augmentation of the hypoprothrombinemia but from the local action on gastric mucosa and the systemic effect on hemostasis. Hemorrhagic complications of phenylbutazone therapy superimposed on an oral anticoagulant regimen are caused by induction of peptic ulceration, inhibition of hemostasis by platelet aggregation, and augmentation of the hypoprothrombinemia. Heparin increases the responsiveness to oral anticoagulants; this interaction can be only minimized by giving heparin intravenously and by drawing the blood for the prothrombin test just before the next heparin dose. Most well controlled studies indicate that sulfonamides and antibiotics have little effect on the prothrombin time of patients on longterm therapy with anticoagulants unless both dietary and intestinal sources of vitamin K are eliminated simultaneously. Compounds that augment the response to but not the plasma levels of oral anticoagulants do so by an obvious alteration of hepatic function (alkylated anabolic steroids), by a subtle alteration of hepatic function detectable only by a decreased rate of synthesis of the vitamin K-dependent clotting factors (perhaps clofibrate), or by a reduction of**lipid-soluble substances that may include vitamin K (several antilipemic agents). As ethyl alcohol seldom augments the hypoprothrombinemic effect of long-term anticoagulant therapy, the occasional use by patients of alcoholic beverages in moderation bears little risk.
The decreased responsiveness to oral anticoagulants in pregnant women results from increased activity of several clotting factors. Oral anticoagulants should be used cautiously during pregnancy because of their ability to cross the placental barrier, but they may be used liberally post partum, if indicated, as there is little evidence to indicate mammary transfer of the drug to the nursing infant. Barbiturates lessen the hypoprothrombinemic response through reduction of coumarin blood levels by enhancing metabolism (all oral forms) as well as by decreasing absorption of the anticoagulant (Dicumarol). The decreased response oral anticoagulants after treatment with diuretic drugs can result from concentration of the clotting factors or from a reduction of congestion of the liver, both as a result of the diuresis achieved. Uremia and uncomplicated renal disease usually do not increase the responsiveness to anticoagulants.
Hereditary factors can alter the responsiveness to oral anticoagulant drugs. However, there are no published data on any heritable condition associated an increased hypoprothrombinemic response to oral anticoagulant drugs. kinds of resistance to oral anticoagulants have been described. Resistance because of rapid metabolism of the drug occurs in normal rabbits from selection of the distribution extreme of anticoagulant metabolism, in rats from selective inbreeding, and in man from disease. In rats and in two human kindred resistance in the presence of normal drug metabolism has been described on the. of a genetic mutation of the vitamin K-anticoagulant receptor site that transmitted as a single autosomal gene controlling the dominant character of the resistance. What effect this resistance may have on survival is unknown. The of therapeutic action of these drugs may be a disadvantage to man, but relative resistance to compounds of this type, perhaps occasionally encountered environment, would be advantageous. The study of these genetic variants elucidate the normal mechanism of action of oral anticoagulant drugs.
Members of the medical profession may behold oral anticoagulants either as valuable or as valueless therapy (205). Though resistive to both extremes, we commend the study of these drugs to all students of pharmacology and to all practitioners of therapeutics. These agents can serve as a model system of pharmacologic action because of the ease of quantifying the anticoagulant effect and the drug itself in man. Perhaps the oral anticoagulant drugs have become a paragon among pharmaceuticals for probing the mechanisms of pharmacologic problems that beset our patients.
- 1970 by The Williams & Wilkins Co.