CHAPTER 28

ANTICOAGULATION

Miriam Rabkin, M.D.

Table of Contents

            Chronic oral anticoagulation therapy is used for the prevention of thromboembolic disorders among patients at high risk. Indications for long-term anticoagulation include atrial fibrillation, prosthetic heart valves, intracardiac thrombus, hypercoagulable states and prior thromboembolism. For many patients, anticoagulation can be initiated as well as maintained in the outpatient setting. More than a million Americans are treated with oral anticoagulants each year, and competence in this field is a basic part of a general medical practice. This chapter offers a brief review of the outpatient use of warfarin sodium and provides an introduction to local resources such as the Anticoagulation Clinic; there is also a short section on low molecular weight heparins and their use in the outpatient setting.

            Mechanisms

            Warfarin inhibits the synthesis of vitamin K-dependent clotting factors II, VII, IX and X, as well as post-translational modifications of the regulatory proteins C and S. Highly bioavailable, warfarin is quickly absorbed from the gastrointestinal tract, reaching peak blood levels in 1.5 hours in healthy volunteers.[i] Anticoagulation takes longer, however, as the biological effect of warfarin is determined by the clotting factor half-lives. Factor VII has a half-life of about six hours, and disappears the most rapidly. Prothrombin levels take the longest to fall, with a half-life of about 72 hours. Depending on the dose of warfarin used, it takes from two to seven days before an anticoagulant effect is observable. When warfarin is discontinued, coagulation returns to baseline in four to five days.[ii]

            It is important to understand that the dose-response relation of warfarin is not the same from person to person. Pharmacokinetic factors (such as absorption and clearance) and pharmacodynamic factors (such as individual reaction to given warfarin levels) play important roles. Drug interactions can potentiate or antagonize warfarin, and dietary vitamin K can influence its effects as well. The drug is metabolized in the liver, and hepatic failure further depresses clotting factor production. Fever and other hypermetabolic states increase responsiveness to warfarin by increasing the catabolism of vitamin K-dependent clotting factors. Warfarin has a narrow therapeutic index, and the variation in dose-response requires that patients be carefully monitored throughout the course of anticoagulation.

            Oral anticoagulation is appropriately monitored by following prothrombin time (PT), which increases as levels of factors II, VII and X fall. The assay employs thromboplastin, a preparation of tissue factor and phospholipid, which is added to recalcified citrated patient plasma. Thromboplastin sensitivity varies from lab to lab, and PTs calculated using different reagents are not interchangeable. The International Normalized Ratio (INR) compensates for differing thromboplastin sensitivity, and is reported with PT. Specific recommendations for monitoring PT and INR are found below. 

            Indications For Therapy

            Warfarin therapy can be initiated in the outpatient setting whenever rapid anticoagulation is unnecessary, most commonly in the context of chronic atrial fibrillation, intracardiac thrombus or indwelling central vein catheters. Other conditions, such as valve replacement, endovascular stent placement, deep vein thrombosis and pulmonary embolism require immediate anticoagulation, and heparin should precede warfarin use.

            Warfarin is contraindicated in patients with known bleeding diatheses, anatomic lesions that are likely to bleed (such as ulcers) and pregnancy. Patients who cannot adhere to a strict regimen and close follow-up are inappropriate candidates for oral anticoagulation, as are those who are predisposed to head trauma (alcoholics, epileptics and elderly patients with gait disorders). Age, hypertension, history of stroke and concurrent use of aspirin are not contraindications to warfarin use.[iii]^,[iv]^,[v]

            Skin necrosis following warfarin use has been reported in 0.01 percent to 0.1 percent of patients,[vi] and is attributed to protein C deficiency and the potential for a brief pro-coagulant effect of warfarin, as protein C (which has a short half-life) is inhibited before most of the clotting factors. Localized, painful lesions with sharply demarcated borders develop on the thighs, breasts or buttocks three to six days after warfarin is begun, and develop into full-thickness skin necrosis.[vii] Although this is a dreaded complication of warfarin therapy, routine screening for protein C deficiency is not currently recommended. Instead, a detailed history should exclude personal and family episodes of thromboembolic disease. If there is any suspicion that a patient may be at increased risk, he or she should be treated with subcutaneous heparin for the first five days of warfarin therapy, which will eliminate the risk of skin necrosis.

            Initiating Oral Anticoagulation

            Initiating warfarin therapy in the outpatient setting is only appropriate for conditions in which urgent anticoagulation is unnecessary and, therefore, no “loading dose” is required. Instead, patients should be started on an anticipated daily maintenance dose of 5 mg. Prothrombin time should be measured on the third day, and warfarin dosing decreased if the PT is supratherapeutic. If not, the PT should be measured twice a week for the first two weeks and - if it is stable and therapeutic - weekly for the next two weeks, every second week for two months and then monthly.

            Patient counseling is an important part of effective oral anticoagulation. Patients must understand the risks and benefits of warfarin use, as well as the importance of meticulous compliance. Written materials can enhance patient education, and are available in the AIM clinic. Specific goals of counseling include a clear understanding of the dangers of over-anticoagulation, the fact that diet and medication can change warfarin requirements and the need to be vigilant for signs of excess bruising or bleeding.

            Table 1 details some of the medications that can interfere with the anticoagulant effects of warfarin. It is good medical practice to check the PDR or the MicroMedex drug interaction program before initiating any medication in a patient taking warfarin. It is also a good idea to give a copy of this table to patients.

TABLE 1: Interactions with warfarin

Adapted and modified from reference 6.

            Goals of Oral Anticoagulation

            Clinical trials have established guidelines for oral anticoagulation, simplifying most target goals to two levels of intensity. For most indications, the goal is to keep patients INR between 2.0 and 3.0. The exceptions are patients with mechanical prosthetic heart valves, whose INR should stay between 2.5 and 3.5, and those with established hypercoagulable states such as (anticardiolipin syndrome) who have recurrent thrombosis (Table 2).

TABLE 2: Recommended therapeutic range for warfarin therapy

            Every published guideline to oral anticoagulation recommends monitoring PT and INR monthly for the entire duration of anticoagulation. Computer algorithms have been shown to be effective in patient management,[viii] as have specialized anticoagulation clinics. The AIM practice has an Anticoagulation Clinic, run by Beth Harding, ANP. Patients may be referred by calling her at 305-6262.

Duration of Oral Anticoagulation

            Determining the optimal duration of oral anticoagulation requires clinicians to balance the risks of hemorrhage with the protective antithrombotic effect of continued warfarin use, and has been the subject of intense study. Three recent prospective, randomized clinical trials have compared short and long courses of oral anticoagulation after an initial thromboembolic event, and all have found that anticoagulation for three to six months is associated with fewer thrombotic events and no more hemorrhagic events than regimens of four to six weeks.[ix]^,[x]^,[xi] Schulman et al.[xii] have also demonstrated that patients who have a second thromboembolic event do better with indefinite anticoagulation than with a six-month regimen; the rate of thromboembolic events was significantly lower in the cohort assigned to continued warfarin therapy (2.6 percent) than in the cohort assigned to six months of warfarin (20.7 percent).

            In patients with nonvalvular atrial fibrillation, the risk of stroke is approximately five percent per year. Five prospective clinical trials have demonstrated that warfarin therapy which prolongs INR to 2.0-3.0 reduces the risk of stroke by 68 percent with virtually no incidence of major bleeding.[xiii]^,[xiv]^,[xv]^,[xvi]^,[xvii] The exception is patients with “lone atrial fibrillation” - those under 60 with no history of previous thromboembolism, hypertension, diabetes, valvular heart disease or prosthesis, heart failure, thyroid disease, ischemic heart disease or echocardiographic risk factors. In these patients, anticoagulation (even with aspirin) may not be warranted,[xviii] although many would recommend daily aspirin therapy.

            As Diuguid discussed in a 1997 editorial,[xix] patients can be stratified into low, intermediate or high risk groups (table 3). Patients at low risk have transient risk factors for thromboembolic disease and require only a short (four to six week) course of anticoagulation. This group includes those receiving prophylaxis following high-risk surgery. Intermediate risk patients are those with medical risk factors such as an initial thromboembolic event (deep-vein thrombosis or pulmonary embolism); clinical trials have established that such patients should be anticoagulated for six months. High risk patients - those with recurrent thromboembolic disease, mechanical heart valves and chronic atrial fibrillation - should be anticoagulated indefinitely.

TABLE 3: Risk-stratifying patients who require anticoagulation

Adapted from reference 19.

            Complications of Oral Anticoagulation

            The risk for bleeding complications of chronic anticoagulation has been evaluated retrospectively in a multicenter study performed in 1993.[xx] In 1950 patient-years of observation, there were 226 serious, 31 life-threatening and 3 fatal bleeding events. The investigators also noted 1071 minor bleeding events, defined as symptoms reported to physicians (such as epistaxis) which did not result in dose-adjustment or further evaluation.

            Four variables were independently related to the risk of a first episode of bleeding; the presence of three or more comorbid conditions, highly variable INR, a shorter duration of anticoagulation, and prolonged INR (the investigators defined this as a prothrombin adjusted time of >2, which correlates to an INR of >4). Age, race, hypertension, history of stroke and indication for anticoagulation were not independent risk factors for bleeding complications.

            If a patient reports a bleeding complication, or if a routinely-ordered INR is supratherapeutic, warfarin dose adjustment may be necessary. An initial assessment of the patient’s complaint is required - minor epistaxis, bruising, hemorrhoidal bleeding and transient hematuria can be evaluated in clinic, while prolonged bleeding from any site should be referred to the emergency department. If self-limited minor bleeding complications occur, the PT should be checked and adjusted if supratherapeutic. If the INR is in the target range, no adjustment is necessary unless repeated minor bleeding complications have been noted. Dose adjustments should be calculated based on weekly (not daily) warfarin dosing - if an asymptomatic patient has an INR of 3.5 when their target INR is 2.0-3.0 the weekly dose should be reduced by 5 percent (from 35 mg to 33 mg, for example, 5mg x 5 days and 4mg x 2 days). Our policy is to prescribe only 5-mg tabs of coumadin and to adjust doses by 2.5 mg increments (i.e. by adding or subtracting a half-pill). This avoids confusion or error in telephone interactions, cross-coverage and refill situations.

Low molecular weight heparin

Low molecular weight (LMW) heparin was approved by the FDA in 1997, and preliminary clinical data support its safety and efficacy in specific settings.[xxi] Like unfractionated heparin, LMW heparin derivatives inactivate factor Xa. They have a much weaker effect on thrombin, however, and do not prolong the APTT. Dosing may be standardized by body weight and laboratory monitoring is not required. In contrast to warfarin, which may days to create an anticoagulant effect, LMW heparins have rapid onset and termination, making them particularly convenient in patients undergoing surgical procedures.[xxii] LWM heparins differ chemically and pharmacokinetically, and do not appear to be interchangeable;[xxiii]^,[xxiv] enoxaparin (Lovenox^TM) has been the most widely studied in clinical trials to date.

Data support the use of enoxaparin in the short-term prevention[xxv] and treatment of deep vein thrombosis, as well as in unstable angina.[xxvi] While there are a small number of encouraging trials suggesting that enoxaparin is also safe and effective in the secondary prophylaxis of venous thromboembolism,[xxvii] there are scant data regarding the use of LMW heparin for atrial fibrillation or prosthetic valves and no data about chronic use. Treatment failure has been reported in the setting of prosthetic heart valves;[xxviii] we do not recommend LMW for patients with atrial fibrillation or prosthetic valves. The ease of monitoring patients on enoxaparin may outweigh its expense and the inconvenience of subcutaneous administration, and it may be reasonable to use this agent for treatment and/or prophylaxis in select patients with venous thromboembolism.

Acknowledgment

We thank Dr. David Diuguid for his helpful comments and suggestions.