Search Medications (Drugs):
Medications (Drugs) from A to Z:
Home > Medications (Drugs) > Warfarin > Interactions with Warfarin

Interactions with Warfarin

Contents

If you are currently being treated with any of the following medications, you should not use Warfarin without reading these interactions.

Moricizine

According to one case report, prothrombin time was elevated in a patient who was stable on warfarin and who then received moricizine. Controlled studies, however, have revealed only that serum warfarin half-life is decreased during coadministration. These studies have found no significant change in the prothrombin time. Similar effects may occur with other oral anticoagulants. Management consists of following the patient's prothrombin time or INR during coadministration. Patients should be advised to promptly report any signs of bleeding to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, vaginal bleeding, nosebleeds, bleeding of gums from brushing, unusual bleeding or bruising, red or brown urine, or red or black stools.

Acetaminophen (oral/rectal)

Acetaminophen (APAP) may potentiate the hypoprothrombinemic effect of warfarin and other oral anticoagulants, although data are somewhat conflicting and the mechanism of interaction is unknown. The interaction has generally been associated with prolonged ingestion of relatively high APAP dosages (greater than 1.3 g/day continuously for greater than 1 week) but not with brief, intermittent exposures of average doses. Reported increases in prothrombin time or INR from most studies were often small but statistically significant, although there have been isolated case reports citing bleeding episodes and clinically significant alterations in coagulation parameters. In contrast, one retrospective study found no significant effect of APAP 2000 to 2500 mg/day on the anticoagulant effect of phenprocoumon, and another study reported no effect of APAP 4 g/day for 2 weeks on single-dose warfarin pharmacokinetics and pharmacodynamics in healthy volunteers. Due to the lack of safer alternatives, acetaminophen is considered the analgesic and antipyretic drug of choice for patients receiving oral anticoagulant therapy. However, caution is recommended during concomitant therapy, particularly if high dosages of APAP are used for a prolonged period. Patients should be advised to promptly report any signs of bleeding to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, vaginal bleeding, nosebleeds, bleeding of gums from brushing, unusual bleeding or bruising, red or brown urine, or red or black stools.

Sucralfate (oral)

ADJUST DOSING INTERVAL: Sucralfate may interfere with the absorption of warfarin in some patients and may reduce its therapeutic effect. The mechanism is unknown and clinical data have been conflicting.

MANAGEMENT: Patients should be advised to take the anticoagulant at least 2 hours before or 6 hours after the sucralfate to minimize a potential interaction. Close observation for altered anticoagulant effect is recommended. Sucralfate may interact in the same manner with other oral anticoagulants as well.

Lanthanum carbonate

ADJUST DOSING INTERVAL: Theoretically, lanthanum carbonate may chelate with certain drugs in the gastrointestinal tract, resulting in reduced oral bioavailability of those drugs during coadministration. However, an in vitro study involving digoxin, enalapril, furosemide, metoprolol, phenytoin, and warfarin found no evidence that lanthanum carbonate forms insoluble complexes with these drugs in simulated gastric fluid. Studies in healthy subjects have also found no effect of lanthanum carbonate (1000 mg for 4 doses) on the absorption of a single dose of digoxin (0.5 mg), metoprolol (100 mg), or warfarin (10 mg).

MANAGEMENT: To minimize the potential for interaction, the product labeling recommends that drugs known to interact with antacids (e.g., ACE inhibitors, beta blockers, bisphosphonates, coumarin derivatives, digitalis glycosides, fluoroquinolones, iron, phenytoin, rifampin, tetracyclines, thyroid preparations, valproic acid) not be taken within 2 hours of administration of lanthanum carbonate.

Cilostazol

Cilostazol 100 mg twice daily for 7 or 13 days did not significantly affect the pharmacokinetics of single doses of warfarin 25 mg in healthy subjects. Activated partial thromboplastin time, prothrombin time, Ivy bleeding time, and platelet aggregation also did not change. There are no data on the effects of concomitant multiple dosing.

Clove

Clove and clove oil may potentiate the effects of anticoagulants, platelet inhibitors and thrombolytic agents, possibly increasing the risk of bleeding. Eugenol, the active constituent of clove and clove oil, has been shown to be an inhibitor of platelet activity. However, bleeding complications and interactions with hematologic agents have not been reported. Moreover, pharmacologic effects may be highly variable due to inconsistencies in formulation and potency of commercial herbal products. Patients should consult a healthcare provider before taking any herbal or alternative medicine. In patients who have used clove and clove oil extensively prior to receiving anticoagulation, antiplatelet or thrombolytic therapy, the potential for an interaction should be considered. Close clinical and laboratory observation for hematologic complications is recommended. Patients should be advised to promptly report any signs of bleeding to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, vaginal bleeding, nosebleeds, bleeding of gums from brushing, unusual bleeding or bruising, red or brown urine, or red or black stools.

Entacapone

Coadministration with entacapone has been shown to slightly increase plasma concentrations of the less pharmacologically active R-enantiomer of warfarin. The mechanism of interaction is unknown. In a study of 12 healthy subjects, entacapone (200 mg orally four times a day for 1 week) increased the mean steady-state area under the plasma concentration-time curve (AUC) of R-warfarin by 18% compared to placebo. Other pharmacokinetic parameters of R-warfarin, as well as pharmacokinetics of S-warfarin, were not significantly affected. Entacapone increased the INR by an average of 13% relative to placebo, although safety and tolerability variables were not different between treatment phases. These changes are unlikely to be of clinical significance.

Colestipol

Colestipol may decrease the absorption of warfarin and other anticoagulants. The therapeutic effect of warfarin may be reduced. However, colestipol also may interfere with vitamin K absorption. Prolonged prothrombin times have been reported in one patient. Patients on concomitant therapy should be monitored for alterations in anticoagulation.

Raloxifene (oral)

Concurrent administration of raloxifene and warfarin during a single-dose study did not alter the pharmacokinetic disposition of warfarin. However, a 10% reduction in prothrombin time (PT) was noted. The mechanism of action is not known. Clinical monitoring of PT or International Normalized Ratio (INR) is recommended when raloxifene therapy is initiated or discontinued in patients concurrently receiving warfarin. A similar reaction may occur with other oral anticoagulants.

Tamoxifen, Toremifene

CONTRAINDICATED: Tamoxifen may enhance the hypoprothrombinemic response to warfarin. Response to other anticoagulants may be affected as well. The mechanism is unknown but may be due to competition for the same metabolic pathways. Severe and/or fatal hemorrhages have been reported with concomitant use of coumarin derivatives. A similar interaction may occur with toremifene and other triphenylethylene antiestrogens.

MANAGEMENT: When used to reduce the incidence of breast cancer in high-risk women, tamoxifen is considered contraindicated in patients requiring concomitant coumarin anticoagulants. Other patients should be closely monitored for excessive anticoagulation and prolonged prothrombin times. Patients should be advised to notify their physicians if they experience any signs of excessive anticoagulation, such as unusual or prolonged bleeding, bruising, vomiting, change in stool or urine color, headache, dizziness, or weakness.

Mifepristone

CONTRAINDICATED: The concomitant use of anticoagulants may exacerbate mifepristone-induced bleeding.

MANAGEMENT: This combination is considered contraindicated by the U.S. manufacturer.

Methylphenidate, Methylphenidate (transdermal), Dexmethylphenidate

Data from an early study involving four subjects suggest that the coadministration with methylphenidate may prolong the half-life of ethyl biscoumacetate (a coumarin anticoagulant), presumably due to inhibition of its metabolism in the liver. However, a subsequent study of 12 healthy volunteers found little or no effect on ethyl biscoumacetate half-life or prothrombin time. Thus, available evidence does not seem to support a clinically significant drug interaction. Nevertheless, it may be appropriate to monitor the pharmacologic response to coumarin anticoagulants and the International Normalization Ratio (INR) more closely whenever methylphenidate (racemic) or dexmethylphenidate (the more pharmacologically active d-enantiomer) is added to or withdrawn from therapy, and the anticoagulant dosage adjusted as necessary. Patients should be advised to notify their physician if they experience any signs of excessive anticoagulation such as unusual or prolonged bleeding, bruising, vomiting, change in stool or urine color, headache, dizziness, or weakness.

Fluvastatin

Fluvastatin has been reported to increase the hypoprothrombinemic effect of warfarin. INR was altered, but no clinical bleeding occurred. A similar interaction has been reported for lovastatin and in some cases has led to clinically significant bleeding. It is recommended that the clinician monitor more closely for signs of bleeding if fluvastatin is started, fluvastatin is discontinued, or the dosage is changed while a patient is also taking oral anticoagulants. Pravastatin has not been shown to interact with warfarin.

Aspirin (oral), Aspirin (rectal)

GENERALLY AVOID: Aspirin, even in small doses, may increase the risk of bleeding in patients on oral anticoagulants by inhibiting platelet aggregation, prolonging bleeding time, and inducing gastrointestinal lesions. Analgesic/antipyretic doses of aspirin increase the risk of major bleeding more than low-dose aspirin; however bleeding has also occurred with low-dose aspirin.

MANAGEMENT: This combination, especially with analgesic/antipyretic aspirin doses, should generally be avoided unless the potential benefit outweighs the risk of bleeding. If concomitant therapy is used for additive anticoagulant effects, monitoring for excessive anticoagulation and overt and occult bleeding is recommended. The INR should be checked frequently and the dosage adjusted accordingly when aspirin is added to an anticoagulant regimen. Be cognizant that bleeding may occur without INR or prothrombin time increases. Patients should be advised to promptly report any signs of bleeding to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, vaginal bleeding, nosebleeds, bleeding of gums from brushing, unusual bruising, red or brown urine, or red or black stools. Patients should also be counseled to avoid any other over-the-counter oral or topical salicylate products.

Pentobarbital, Phenobarbital, Primidone, Secobarbital, Mephobarbital

GENERALLY AVOID: Barbiturates reduce the effects of oral anticoagulants by inducing their hepatic metabolism. Anticoagulant dosage requirements may be increased by 30% to 60%. After the barbiturate is discontinued, excessive anticoagulation and bleeding may occur if the anticoagulant dose is not reduced.

MANAGEMENT: Close attention should be paid to alterations in coagulation if a patient must take both drugs and for several weeks after the dose of the barbiturate is changed or discontinued. Alternatively, a benzodiazepine might be used instead of a barbiturate, if appropriate. Patients should be advised to promptly report any signs of blood clots (e.g. chest pain, shortness of breath, sudden loss of vision, or pain, redness or swelling in an extremity).

Charcoal

GENERALLY AVOID: Charcoal may reduce the absorption of many drugs and can absorb enterohepatically circulated drugs. Clinical utility may be the reduction either of the effects or of the toxicity of many drugs. Activated charcoal may absorb any therapeutic agents administered while it is in the gastrointestinal tract.

MANAGEMENT: The regular ingestion of charcoal should be avoided by patients requiring maintenance medications. If concomitant use is necessary, the dosage or route of administration may need to be altered.

Ubiquinone

GENERALLY AVOID: Coadministration with coenzyme Q10 (also known as ubiquinone or ubidecarenone) may reduce the hypoprothrombinemic effect of warfarin and other oral anticoagulants. The exact mechanism of interaction is unknown, although coenzyme Q10 is structurally related to menaquinone (vitamin K2) and may have procoagulant effects. There have been reports of decreased warfarin response in patients using coenzyme Q10 and a return to normal anticoagulation following its discontinuation. In contrast, a study consisting of 24 patients stabilized on warfarin for at least several months found no significant effect of coenzyme Q10 on INR or warfarin dosage compared to placebo, and no thromboembolic events were observed in the study. However, it is important to recognize that pharmacologic effects of herbal products may be highly variable due to inconsistencies in formulation and potency of commercial preparations.

MANAGEMENT: In general, patients should consult a healthcare provider before taking any herbal or alternative medicine. Given the potential for interaction and the narrow therapeutic index of oral anticoagulants, these drugs should preferably not be used with coenzyme Q10. If they are given together, the INR must be checked frequently and anticoagulant dosage adjusted accordingly, particularly following initiation, discontinuation or change of dosage of coenzyme Q10 in patients who are stabilized on their anticoagulant regimen.

Estradiol oral, Estradiol transdermal, Conjugated estrogens, Esterified estrogens, Estropipate, Chlorotrianisene, Estradiol topical (for use on skin), Estradiol vaginal (systemic), Estradiol injection, Estramustine, Estradiol (topical), Estradiol vaginal (local), Conjugated estrogens (vaginal)

GENERALLY AVOID: Concomitant therapy with estrogen-containing drugs may diminish the therapeutic effects of oral anticoagulants. Estrogens can increase the plasma levels of certain clotting factors such as fibrinogen, prothrombin, and factors VII and VIII in a dose-dependent manner, resulting in increased risk of thromboembolism, stroke, and/or myocardial infarction. The risk may be further increased by lifestyle choices such as smoking and lack of exercise.

MANAGEMENT: Use of estrogen-containing drugs should be avoided in patients receiving anticoagulant therapy unless benefits are anticipated to outweigh the risks. Close clinical and laboratory monitoring are recommended if the combination is prescribed. Patients should be advised to promptly notify their physician if they experience potential signs and symptoms of blood clots such as chest pain, shortness of breath, sudden loss of vision, and pain, redness or swelling in an extremity.

Imatinib (oral)

GENERALLY AVOID: Data from human liver microsome studies indicate that imatinib is a potent competitive inhibitor of the CYP450 2C9 isoenzyme. Although no formal clinical interaction studies have been performed, the coadministration of imatinib and warfarin is likely to result in elevated plasma concentrations and increased pharmacologic effects of warfarin due to inhibition of CYP450 2C9-mediated metabolism.

MANAGEMENT: The manufacturer recommends that patients treated with imatinib who require anticoagulation receive low molecular weight or standard heparin.

Dong quai

GENERALLY AVOID: Dong quai may potentiate the effects of anticoagulants, platelet inhibitors and thrombolytic agents, possibly increasing the risk of bleeding. Dong quai has been reported to exert an antithrombotic effect by inhibiting platelet activation and aggregation. In one patient, the addition of dong quai 565 mg once or twice daily to a stabilized regimen of warfarin resulted in a sudden increase in INR after 4 weeks. The INR returned to normal 4 weeks following discontinuation of the dong quai.

MANAGEMENT: Patients should consult a healthcare provider before taking any herbal or alternative medicine. In general, consumption of dong quai should preferably be avoided during use of coagulation-modifying agents. In patients who have used this herb extensively prior to receiving anticoagulation, antiplatelet or thrombolytic therapy, the potential for an interaction should be considered. Close clinical and laboratory observation for hematologic complications is recommended. Patients should be advised to promptly report any signs of bleeding to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, vaginal bleeding, nosebleeds, bleeding of gums from brushing, unusual bleeding or bruising, red or brown urine, or red or black stools.

Enoxaparin, Dalteparin, Danaparoid, Ardeparin, Tinzaparin, Fondaparinux (injectable)

GENERALLY AVOID: Drugs that can affect hemostasis such as dextran, platelet inhibitors, thrombin inhibitors, thrombolytic agents, or other anticoagulants may potentiate the risk of bleeding complications associated with the use of a low molecular weight heparin (LMWH), heparinoid, or fondaparinux. In patients receiving neuraxial anesthesia or spinal puncture, the risk of developing an epidural or spinal hematoma during LMWH, heparinoid, or fondaparinux therapy may also be increased by the concomitant use of other drugs that affect coagulation. The development of epidural and spinal hematoma can lead to long-term or permanent paralysis.

MANAGEMENT: In general, any agent that can enhance the risk of hemorrhage including other anticoagulants should be discontinued prior to initiation of LMWH, heparinoid, or fondaparinux therapy. If coadministration is necessary, it should be undertaken with caution and only after thorough assessment of risks and benefits. Close clinical and laboratory observation for bleeding complications is recommended. Patients undergoing neuraxial intervention and treated with these agents should also be monitored frequently for signs and symptoms of neurologic impairment such as midline back pain, sensory and motor deficits (numbness or weakness in lower limbs), and bowel or bladder dysfunction.

Desirudin

GENERALLY AVOID: Drugs that can affect hemostasis such as dextrans, systemic glucocorticoids, thrombolytic agents, or other anticoagulants may potentiate the risk of bleeding complications associated with the use of desirudin. In patients receiving neuraxial anesthesia or spinal puncture, the risk of developing an epidural or spinal hematoma during desirudin therapy may also be increased by the concomitant use of other drugs that affect coagulation. The development of epidural and spinal hematoma can lead to long-term or permanent paralysis.

MANAGEMENT: During prophylaxis of venous thromboembolism, concomitant treatment with heparins or dextrans is not recommended. The effects of desirudin and unfractionated heparins on prolongation of activated partial thromboplastin time (aPTT) are additive. In general, any agent which may enhance the risk of hemorrhage should be discontinued prior to initiation of desirudin therapy. If coadministration is necessary, it should be undertaken with caution and only after thorough assessment of risks and benefits. Close clinical and laboratory observation for bleeding complications is recommended. Patients undergoing neuraxial intervention and treated with these agents should also be monitored frequently for signs and symptoms of neurologic impairment such as midline back pain, sensory and motor deficits (numbness or weakness in lower limbs), and bowel or bladder dysfunction.

Clofibrate, Gemfibrozil, Fenofibrate

GENERALLY AVOID: Fibric acid derivatives may enhance the hypoprothrombinemic response to coumarin-type oral anticoagulants. The mechanism of action is unknown.

MANAGEMENT: Close clinical and laboratory monitoring for signs of excessive anticoagulation is recommended until the prothrombin time (PT) has stabilized if this combination is used. Some experts recommend decreasing the anticoagulant dose by approximately one-third to one-half in patients receiving concomitant coumarins and fibric acid derivatives with subsequent adjustment based on the PT. Patients should be advised to notify their physicians if they experience any signs of excessive anticoagulation, such as unusual or prolonged bleeding, bruising, vomiting, change in stool or urine color, headache, dizziness, or weakness.

Garlic

GENERALLY AVOID: Garlic may potentiate the effects of anticoagulants, platelet inhibitors and thrombolytic agents, possibly increasing the risk of bleeding. Garlic has been shown in some studies to have antithrombotic effects and may increase fibrinolysis, decrease platelet aggregation, and increase prothrombin time. Antiplatelet activity is present in garlic powder, aged garlic preparations, garlic oil, and fresh garlic (more so raw than when it is cooked). There have been isolated reports of bleeding complications associated with chronic, high dietary intake of garlic, as well as reports suggesting an interaction between warfarin and garlic resulting in increased INR.

MANAGEMENT: Patients should consult a healthcare provider before taking any herbal or alternative medicine. In general, consumption of garlic supplements and large amounts of garlic should preferably be avoided during use of coagulation-modifying agents. In patients who have used this herb extensively prior to receiving anticoagulation, antiplatelet or thrombolytic therapy, the potential for an interaction should be considered. Close clinical and laboratory observation for hematologic complications is recommended. Patients should be advised to promptly report any signs of bleeding to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, vaginal bleeding, nosebleeds, bleeding of gums from brushing, unusual bleeding or bruising, red or brown urine, or red or black stools.

Ginkgo

GENERALLY AVOID: Ginkgo may potentiate the effects of anticoagulants, platelet inhibitors and thrombolytic agents, possibly increasing the risk of bleeding. Ginkgolide B, a component of ginkgo, inhibits platelet-activating factor by displacing it from its receptor-binding site, resulting in reduced platelet aggregation. There have been isolated reports of bleeding complications and prolonged bleeding times associated with ginkgo ingestion that resolved following discontinuation of ginkgo use. Possible interactions with warfarin and aspirin have also been described in the medical literature. A patient stabilized on warfarin for five years developed intracerebral hemorrhage two months after starting ginkgo, and another who had been taking aspirin 325 mg/day for three years developed spontaneous bleeding of the iris into the anterior chamber of the eye one week after he began using ginkgo. In contrast, an investigative study found no significant effect of ginkgo pretreatment for 7 days on clotting status or the pharmacokinetics or pharmacodynamics of a single 25 mg dose of warfarin in 12 healthy volunteers. Another study consisting of 24 patients stabilized on warfarin for at least several months also found no significant effect of ginkgo on INR or warfarin dosage compared to placebo, and no major bleedings were observed in the study. However, it is important to recognize that pharmacologic effects of herbal products may be highly variable due to inconsistencies in formulation and potency of commercial preparations.

MANAGEMENT: Patients should consult a healthcare provider before taking any herbal or alternative medicine. In general, consumption of ginkgo should preferably be avoided during use of coagulation-modifying agents. In patients who have used this herb extensively prior to receiving anticoagulation, antiplatelet or thrombolytic therapy, the potential for an interaction should be considered. Close clinical and laboratory observation for hematologic complications is recommended. Patients should be advised to promptly report any signs of bleeding to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, vaginal bleeding, nosebleeds, bleeding of gums from brushing, unusual bleeding or bruising, red or brown urine, or red or black stools.

Ginseng

GENERALLY AVOID: Ginseng may reduce the blood concentrations and hypoprothrombinemic effect of warfarin. The mechanism of interaction is unknown. In one case report, use of Asian ginseng was associated with a reduction in INR in a patient who had been stabilized on warfarin for nine months. In a study of 20 healthy volunteers, subjects given American ginseng (1 g orally twice daily for 2 weeks) and warfarin (5 mg daily for 3 days) had a small, but statistically significant reduction (-0.19) in peak INR compared to subjects given placebo and warfarin. The peak plasma concentration (Cmax) and area under the concentration-time curve (AUC) of warfarin also were lower in the ginseng group, suggesting potential induction of warfarin metabolism via hepatic microsomal enzymes. All patients maintained a relatively consistent intake of dietary vitamin K and refrained from alcohol, tobacco products, and other medications throughout the study.

MANAGEMENT: Patients should consult a healthcare provider before taking any herbal or alternative medicine. In general, consumption of ginseng supplements and large amounts of ginseng should preferably be avoided during use of warfarin. The same precaution may be applicable during therapy with other oral anticoagulants, although clinical data are lacking.

St. John's wort

GENERALLY AVOID: Limited clinical data suggest that St. John's wort may reduce the plasma concentrations and/or hypoprothrombinemic effect of warfarin and other oral anticoagulants. The proposed mechanism is induction of CYP450 isoenzymes by constituents of St. John's wort. There have been case reports of patients stabilized on warfarin whose INR decreased following the addition of St. John's wort and returned to normal only after discontinuation of St. John's wort or an increase in anticoagulant dosage. In a study with 10 healthy volunteers, pretreatment with St. John's wort for 11 days resulted in a 17% decrease in the area under the plasma concentration-time curve (AUC) of unbound phenprocoumon (12 mg single dose) compared to placebo.

MANAGEMENT: In general, patients should consult a healthcare provider before taking any herbal or alternative medicine. Given the potential for interaction and the narrow therapeutic index of oral anticoagulants, these drugs should preferably not be used with St. John's wort. If they are given together, the INR must be checked frequently and anticoagulant dosage adjusted accordingly, particularly following initiation, discontinuation or change of dosage of St. John's wort in patients who are stabilized on their anticoagulant regimen.

Fenoprofen, Ketoprofen, Sulindac, Indomethacin, Tolmetin, Flurbiprofen, Ketorolac, Mefenamic acid, Nabumetone, Piroxicam, Diclofenac, Etodolac, Oxaprozin, Diflunisal, Meclofenamate, Meloxicam

GENERALLY AVOID: Nonsteroidal anti-inflammatory drugs (NSAIDs) may potentiate the hypoprothrombinemic effect and bleeding risk associated with oral anticoagulants. In a one-year observational study of a population of coumarin users, the relative risk of bleeding complications due to concomitant NSAID use was 5.8 compared to coumarin use alone. Some investigators suggest that the risk of hemorrhagic peptic ulcers in particular may be substantially increased, especially in elderly or debilitated patients. A retrospective epidemiologic study of patients aged 65 years or older reported a nearly 13-fold increase in the risk of developing hemorrhagic peptic ulcer disease in concurrent users of oral anticoagulants and NSAIDs compared with nonusers of either drug. Fatalities have been reported. The pharmacologic effects of NSAIDs that contribute to this interaction include gastrointestinal irritation, prolongation of prothrombin time, and inhibition of platelet adhesion and aggregation. In addition, various NSAIDs have also been shown to alter the pharmacokinetics of warfarin and other oral anticoagulants, resulting in increased INR or prothrombin time. However, some studies failed to demonstrate any evidence of an interaction.

MANAGEMENT: NSAIDs should be administered with oral anticoagulants only if benefit outweighs risk. The INR should be checked frequently and oral anticoagulant dosage adjusted accordingly, particularly following initiation or discontinuation of NSAIDs in patients who are stabilized on their anticoagulant regimen. Patients should be advised to promptly report any signs of unusual bleeding or bruising to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, vaginal bleeding, nosebleeds, bleeding of gums from brushing, red or brown urine, or red or black stools. Salicylates (except aspirin) appear to have less effect on coagulation and may be preferable in patients treated with oral anticoagulants. Nevertheless, caution is advised and close monitoring for gastrointestinal bleeding is recommended, particularly in elderly or debilitated patients.

Horse chestnut

GENERALLY AVOID: Orally administered horse chestnut may theoretically potentiate the effects of anticoagulants, possibly increasing the risk of bleeding. Horse chestnut contains the coumarin glycosides aesculetin, fraxin, and scopolin.

MANAGEMENT: Patients should consult a healthcare provider before taking any herbal or alternative medicine. Until more information is available, consumption of horse chestnut should preferably be avoided during use of anticoagulants. In patients who have used this herb extensively prior to receiving anticoagulation, the potential for an interaction should be considered. Close clinical and laboratory observation for hematologic complications is recommended. Patients should be advised to promptly report any signs of unusual bleeding or bruising to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, nosebleeds, bleeding of gums from brushing, red or brown urine, or red or black stools.

Devil's claw

GENERALLY AVOID: Purpura has been reported in a patient receiving warfarin and devil's claw. However, key information such as the patient's medical conditions, other concomitant medications, and the dosages and duration of use of each agent were not reported. The mechanism of interaction, if any, is unknown.

MANAGEMENT: Until further data are available, consumption of products containing devil's claw should preferably be avoided in patients receiving warfarin or other oral anticoagulants.

Rifampin, Rifabutin

GENERALLY AVOID: Rifampin may decrease the anticoagulant effect of warfarin by enhancing CYP450 hepatic microsomal enzyme metabolism of warfarin.

MANAGEMENT: Concurrent use should be avoided unless no alternatives are available. The INR or prothrombin time should be monitored closely during concomitant therapy and after rifampin is discontinued. Adjustments in warfarin dosage may be needed when rifampin dosage is added, discontinued, or changed. This interaction may occur with other oral anticoagulants and with rifabutin.

Sulfinpyrazone

GENERALLY AVOID: Sulfinpyrazone inhibits platelet function and inhibits the metabolism of warfarin and nicoumalone, potentiating their hypoprothrombinemic effects. Severe bleeding episodes have been reported. This interaction is expected with other coumarin derivatives also.

MANAGEMENT: If coumarin derivatives must be used with sulfinpyrazone, the patient should be monitored for excessive anticoagulation. However, it is best to avoid this combination or reduce the anticoagulant dosage by 20% to 50%. Patients should be advised to notify their physician if they experience any signs of excessive anticoagulation, such as unusual or prolonged bleeding, bruising, vomiting, change in stool or urine color, headache, dizziness, or weakness.

Bivalirudin

GENERALLY AVOID: The concomitant administration of bivalirudin and other anticoagulants may increase the risk of bleeding complications due to additive or synergistic effects on the clotting cascade. In clinical trials, the use of bivalirudin with heparin or warfarin in patients undergoing PTCA was associated with increased risks of major bleeding events. According to the manufacturer, results from drug interaction studies do not suggest pharmacodynamic interactions with low molecular weight heparin (LMWH). However, data are limited, and patients in clinical trials receiving LMWH were taken off the drug at least 8 hours prior to PTCA and administration of bivalirudin.

MANAGEMENT: Until further data are available, coadministration of bivalirudin and all anticoagulants should be undertaken with caution and only after thorough assessment of risks and benefits. Close clinical and laboratory observation for bleeding complications is recommended.

Salsalate, Magnesium salicylate

GENERALLY AVOID: Theoretically, salicylates may potentiate the effects of anticoagulants and increase the risk of bleeding. Salicylates interfere with the action of vitamin K and induce a dose-dependent alteration in hepatic synthesis of coagulation factors VII, IX and X, occasionally increasing the prothrombin time. While these effects are generally slight for most salicylates (except aspirin) at recommended dosages, they may be of clinical significance when combined with the inhibitory effects of anticoagulants on the clotting cascade. Moreover, salicylates are known to cause dose-related gastrointestinal bleeding, which may be complicated by anticoagulant therapy.

MANAGEMENT: Until further data are available, products containing salicylates, especially if given chronically or in high dosages, should preferably be avoided in patients receiving anticoagulants. Close clinical and laboratory observation for bleeding complications is recommended if concurrent therapy is necessary. The same precaution should be observed with the use of salicylate-related agents such as salicylamide because of their structural and pharmacological similarities. Ambulatory patients should be advised to promptly report any signs of bleeding to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, vaginal bleeding, nosebleeds, bleeding of gums from brushing, unusual bleeding or bruising, red or brown urine, or red or black stools. Patients should also be counseled to avoid any other over-the-counter salicylate products.

Tirofiban, Eptifibatide

GENERALLY AVOID: The risk of bleeding, including major bleeding (intracranial hemorrhage or a decrease in hemoglobin greater than 5 mg/dl) is significantly more likely when oral anticoagulants and glycoprotein IIb/IIIa inhibitors are given together.

MANAGEMENT: Some manufacturers of GP IIb/IIIa inhibitors consider their use contraindicated if patients have received oral anticoagulants within 7 days unless the prothrombin time is less than or equal to 1.2 times control. Bleeding precautions should be observed and patients should be closely monitored for clinical and laboratory evidence of adverse effects.

Glucagon

Glucagon may increase the risk of bleeding in patients on oral anticoagulants. The mechanism is unknown. Frequent monitoring of the prothrombin time or INR is recommended.

Green tea

Green tea in large amounts may antagonize the pharmacologic effect of warfarin and other related anticoagulants. Dried green tea leaves have been found to contain substantial amounts of vitamin K. Although brewed green tea is generally not considered a significant source of the vitamin, large amounts may contain sufficient vitamin K to interfere with the activity of some anticoagulants. A substantial decrease in the INR has been reported in a patient treated with warfarin after he began consuming large quantities (0.5 to 1 gallon daily) of green tea. While there is no need for patients treated with warfarin or related agents to avoid green tea, they should be advised that large quantities may diminish the effectiveness of their anticoagulation therapy.

Telmisartan

In an open-label, multiple-dose study consisting of 12 healthy volunteers, telmisartan (120 mg orally once daily) caused a slight but statistically significant decrease in the mean steady-state trough plasma concentration of warfarin (dosage individualized to attain INR between 1.2 and 1.8 before start of telmisartan). However, there were no changes in INR values or the overall safety profile of warfarin. The study's investigators theorize that telmisartan may have a stereoselective effect on the metabolism of the less potent R-enantiomer of warfarin, thereby reducing total plasma warfarin concentrations without affecting coagulation. Warfarin had no significant effect on the absorption or pharmacokinetic profile of telmisartan. Based on the results of this study, telmisartan may be administered safely with warfarin without the need for specific clinical intervention.

Ezetimibe

In a study of 12 healthy adult males, coadministration of ezetimibe (10 mg once daily) had no significant effect on the bioavailability of warfarin or prothrombin time. Increased International Normalized Ratio (INR) has been reported during postmarketing use of ezetimibe in patients who had it added to warfarin. However, most of these patients were also on other medications, so the relationship to ezetimibe is uncertain. Ezetimibe product labeling recommends that the INR be appropriately monitored if ezetimibe is used in patients receiving warfarin. The effect of ezetimibe on other oral anticoagulants is unknown.

Bicalutamide

In vitro studies have shown that bicalutamide can displace coumarin anticoagulants such as warfarin from protein-binding sites, theoretically resulting in enhanced anticoagulant effects. PT should be monitored closely when bicalutamide is initiated in patients already receiving coumarin anticoagulants. The anticoagulant dosage should be adjusted as needed. Patients should be advised to notify their physicians if they experience any signs of excessive anticoagulation, such as unusual or prolonged bleeding, bruising, vomiting, change in stool or urine color, headache, dizziness, or weakness.

Benazepril

Limited data suggest that benazepril may slightly decrease the anticoagulant effect of warfarin. The mechanism is not known and warfarin pharmacokinetics are not affected. The clinical significance is unknown. Monitoring of the prothrombin time or INR is recommended. Patients should be advised to promptly report any signs of blood clots (e.g. chest pain, shortness of breath, sudden loss of vision, or pain, redness or swelling in an extremity).

Acarbose

Limited data suggest that the anticoagulant effect of warfarin may be increased in patients also receiving acarbose. The International Normalization Ratio (INR) may increase out of the therapeutic range. The mechanism of action is not known, but may be related to increased absorption of warfarin. This interaction may occur with other oral anticoagulants although data is lacking. The clinician may consider closer INR and clinical monitoring in patients starting or stopping acarbose therapy while also receiving oral anticoagulants. The dose of the oral anticoagulant may need to be adjusted. Patients should be advised to promptly report any signs of bleeding to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, vaginal bleeding, nosebleeds, bleeding of gums from brushing, unusual bleeding or bruising, red or brown urine, or red or black stools.

Disopyramide

Limited data suggest that the anticoagulant effects of warfarin and dicumarol may be enhanced by disopyramide and decreased upon its withdrawal. The mechanism is unknown. The data are conflicting, and the impact of improved hemodynamic status on the anticoagulant effect has not been controlled for. Although no special precautions appear to be necessary, it is recommended that the PT or INR be followed when disopyramide is added or discontinued.

Daptomycin

Limited data suggest that there is no significant interaction between daptomycin and warfarin. In 16 healthy volunteers, coadministration of daptomycin (6 mg/kg once every 24 hours for 5 days) and warfarin (25 mg single oral dose) had no significant effect on the pharmacokinetics of either drug, and the INR was not significantly altered. However, given the limited nature of existing data and the high degree of interpatient variability with respect to warfarin pharmacokinetics, clinicians should consider monitoring the INR more closely for the first several days after initiating therapy with daptomycin in patients stabilized on their warfarin regimen. The same precaution is probably appropriate for other oral anticoagulants until further information is available.

Mesalamine (oral), Mesalamine (rectal), Balsalazide

MONITOR: A case has been reported of a likely interaction between warfarin and mesalamine resulting in failure to attain the hypoprothrombinemic effect of warfarin. The mechanism of this interaction is unknown. The clinician who reported this interaction speculated that mesalamine may have complexed with warfarin, preventing warfarin's intestinal absorption or metabolism. Whether rectally administered mesalamine may interfere with absorption is unclear. Data on other oral anticoagulants are lacking.

MANAGEMENT: Until more information is available, clinicians should closely monitor prothrombin time if an anticoagulant and mesalamine are administered concurrently. Patients should be advised to promptly report any signs of blood clots (e.g. chest pain, shortness of breath, sudden loss of vision, or pain, redness or swelling in an extremity).

Glucosamine, Chondroitin

MONITOR: A case report suggests that chondroitin and/or glucosamine may potentiate the hypoprothrombinemic effect of warfarin. The mechanism of interaction is unknown. Glucosamine is a chemical component of heparin, while chondroitin is a minor component of danaparoid and has been shown in vitro and in animals to prolong prothrombin time. However, adverse effects on coagulation and bleeding complications have not been reported in clinical studies using glucosamine and chondroitin dosages of up to 1.5 and 1.2 g/day, respectively. In the case report, a 69-year-old man who had been stabilized on warfarin 47.5 mg/week for four months demonstrated an increased INR (from 2.58 to 4.52) four weeks after initiating daily self-treatment with glucosamine hydrochloride (3 g) and chondroitin sulfate (2.4 g). The patient reported no other change in his medications or diet. After two weeks at a reduced warfarin dosage of 40 mg/week, the patient's INR returned to normal, and he remained stabilized at that dose for the next three months while he continued to take glucosamine and chondroitin.

MANAGEMENT: Patients should consult a healthcare provider before taking any herbal or alternative medicine. In patients treated with warfarin, the INR should be closely monitored following addition or discontinuation of chondroitin and glucosamine, and the warfarin dosage adjusted as necessary. The same precaution may be applicable during therapy with other oral anticoagulants, although clinical data are lacking. Patients should be advised to promptly report any signs of bleeding to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, vaginal bleeding, nosebleeds, bleeding of gums from brushing, unusual bleeding or bruising, red or brown urine, or red or black stools.

Troglitazone (oral)

MONITOR: A case report suggests that troglitazone may significantly enhance the effects of oral anticoagulants. The mechanism of this interaction may be inhibition by troglitazone of the CYP450 system which metabolizes warfarin, or displacement of warfarin from plasma proteins. This interaction may occur with other oral anticoagulants although data is lacking.

MANAGEMENT: Until more information is available, the dose of the oral anticoagulant should be titrated carefully if troglitazone is added to or discontinued from the drug therapy of a patient receiving an oral anticoagulant. Patients should be advised to promptly report any signs of bleeding to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, vaginal bleeding, nosebleeds, bleeding of gums from brushing, unusual bleeding or bruising, red or brown urine, or red or black stools.

Testolactone

MONITOR: According to the product labeling, testolactone may potentiate the hypoprothrombinemic effects of oral anticoagulants. The mechanism has not been described. The interaction has been reported with various androgens and anabolic steroids including danazol, oxymetholone, testosterone, methyltestosterone and stanozolol, resulting in bleeding complications and anticoagulant dosage reductions. However, it has not been reported with testolactone, a weak androgen and an aromatase inhibitor that is structurally distinct from true androgens.

MANAGEMENT: During concomitant therapy, the INR and/or PT should be monitored frequently and anticoagulant dosage adjusted accordingly, particularly following initiation or discontinuation of testolactone in patients who are stabilized on their anticoagulation regimen. Patients should be advised to promptly report any signs of bleeding to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, vaginal bleeding, nosebleeds, bleeding of gums from brushing, unusual bleeding or bruising, red or brown urine, or red or black stools.

Allopurinol

MONITOR: Allopurinol may inhibit the metabolism of warfarin, possibly enhancing its anticoagulant effect. Allopurinol may interact in a similar manner with other oral anticoagulants.

MANAGEMENT: Patients on concomitant therapy should be monitored for excessive anticoagulation. The INR should be checked frequently and dosage adjusted accordingly when allopurinol is added to an anticoagulant regimen. Patients should be advised to promptly report any signs of bleeding to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, vaginal bleeding, nosebleeds, bleeding of gums from brushing, unusual bleeding or bruising, red or brown urine, or red or black stools.

Bupropion (oral)

MONITOR: Altered PT or INR, infrequently associated with hemorrhagic or thrombotic complications, has been observed during coadministration of bupropion and warfarin. The mechanism of interaction is unknown. In patients taking bupropion for smoking cessation, physiological changes resulting from smoking cessation itself may alter the pharmacokinetics of warfarin. Tobacco smoke is an inducer of hepatic microsomal enzymes, thus smoking cessation may increase blood concentrations of warfarin and other drugs that are extensively metabolized due to de-induction of the enzymes.

MANAGEMENT: Caution is advised if warfarin is given with bupropion. The INR should be checked regularly and warfarin dosage adjusted accordingly, particularly following initiation, discontinuation or change of dosage of bupropion in patients who are stabilized on their warfarin regimen. The same precaution may be applicable during therapy with other oral anticoagulants, although clinical data are lacking. Patients should be advised to promptly report any signs of bleeding to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, vaginal bleeding, nosebleeds, bleeding of gums from brushing, unusual bleeding or bruising, red or brown urine, or red or black stools.

Dipyridamole

MONITOR: Although dipyridamole is often used in combination with oral anticoagulants in clinical practice, the potential for an increased risk of bleeding should be considered during coadministration. In general, dipyridamole does not appear to significantly alter prothrombin time or increase bleeding compared to warfarin alone. However, there have been isolated reports of enhanced hypoprothrombinemic response to warfarin in the presence of dipyridamole, or increased warfarin requirement following discontinuation of dipyridamole. The mechanism of interaction is unknown.

MANAGEMENT: Monitoring for signs of increased anticoagulation and bleeding may be advisable during coadministration of dipyridamole and oral anticoagulants. Patients should be advised to promptly report any signs of bleeding to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, vaginal bleeding, nosebleeds, bleeding of gums from brushing, unusual bleeding or bruising, red or brown urine, or red or black stools.

Lepirudin

MONITOR: Although their use often overlaps in clinical practice, the concomitant use of thrombin inhibitors and other anticoagulants may increase the risk of bleeding complications due to additive or synergistic effects on the clotting cascade. A small clinical trial (n=12) reported two minor bleeding events, but no pharmacokinetic interaction or serious adverse events after administration of low-dose argatroban and single-dose warfarin.

MANAGEMENT: Coadministration of thrombin inhibitors and all anticoagulants should be undertaken with caution and only after thorough assessment of risks and benefits. Current guidelines for transitioning from thrombin inhibitors to warfarin may be consulted. Close clinical and laboratory observation for bleeding complications is recommended.

Ticlopidine (oral)

MONITOR: Although therapy with oral anticoagulants and ticlopidine may overlap in clinical practice, the potential for additive pharmacodynamic effects and increased risk of bleeding, especially gastrointestinal bleeding, should be considered. The combination of warfarin and ticlopidine has also been associated with rare cases of cholestatic hepatitis. In addition, ticlopidine may inhibit the metabolism of R-warfarin by an unknown mechanism. Increases in R-warfarin concentrations have been reported during concomitant therapy; however, there were no significant changes in INR.

MANAGEMENT: The manufacturer recommends discontinuation of any anticoagulants before initiating ticlopidine therapy. If concomitant use cannot be avoided, patients should be closely monitored for signs of bleeding and the anticoagulant dosage should be adjusted as appropriate to maintain therapeutic aPTT/INR values. Patients should be advised to promptly report any signs of bleeding to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, vaginal bleeding, nosebleeds, bleeding of gums from brushing, unusual bruising, red or brown urine, or red or black stools.

Aminoglutethimide

MONITOR: Aminoglutethimide may significantly increase warfarin clearance by induction of CYP450 2C9 and/or 3A4 hepatic microsomal enzymes. In one study, warfarin clearance increased three- to fivefold. A similar effect may occur with other oral anticoagulants.

MANAGEMENT: The patient should be monitored for reduced anticoagulant effect when aminoglutethimide is added, discontinued, or the dosage is changed. Patients should be advised to promptly report any signs of blood clots (e.g. chest pain, shortness of breath, sudden loss of vision, or pain, redness or swelling in an extremity).

Methimazole, Propylthiouracil

MONITOR: Antithyroid drugs, by reducing the hyperthyroid state, may increase the hypoprothrombinemic response to oral anticoagulants. There have been reports of hypoprothrombinemia and bleeding in patients taking propylthiouracil or methimazole alone or combined with oral anticoagulants. Other antithyroid agents may also interact.

MANAGEMENT: Bleeding parameters should be monitored closely when antithyroid dosage is added, discontinued, or changed. Patients should be advised to promptly report any signs of bleeding to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, vaginal bleeding, nosebleeds, bleeding of gums from brushing, red or brown urine, or red or black stools.

Leflunomide (oral)

MONITOR: Available data suggest that leflunomide may potentiate the anticoagulant effect of warfarin. The proposed mechanism is leflunomide inhibition of CYP450 2C9, the isoenzyme responsible for the metabolic clearance of the more active S(-) enantiomer of warfarin. The interaction was suspected in a published report of a patient who developed increased INR shortly after switching from methotrexate to leflunomide for her rheumatoid arthritis. The patient's INR normalized after withholding of three warfarin doses and continual decrease of her warfarin dosage from 36 to 28 mg/week over the next three months while she continued taking leflunomide. In another report, a patient with stable INR developed gross hematuria requiring hospitalization after taking the second loading dose of leflunomide. The patient's INR declined following discontinuation of warfarin, administration of vitamin K 1 mg on the third day, and a decrease of the leflunomide dosage to 20 mg/day for maintenance. Warfarin was subsequently resumed at a lower dosage with no further incident. According to the author, the Committee on the Safety of Medicines in the U.K. had also received over 300 reports of elevated INR in patients treated with warfarin and leflunomide. No data are available for other oral anticoagulants, although at least one other coumarin derivative is known to be metabolized by CYP450 2C9.

MANAGEMENT: Given the potential for interaction and the high degree of interpatient variability with respect to warfarin metabolism, patients should be closely monitored during concomitant therapy with leflunomide. The INR should be checked frequently and warfarin dosage adjusted accordingly, particularly following initiation or discontinuation of leflunomide therapy in patients who are stabilized on their warfarin regimen. The same precaution may be applicable during therapy with other oral anticoagulants, although clinical data are lacking. Patients should be advised to promptly report any signs of bleeding to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, vaginal bleeding, nosebleeds, bleeding of gums from brushing, unusual bleeding or bruising, red or brown urine, or red or black stools.

Cyclosporine

MONITOR: Both increased and reduced effects of both agents have been reported with concomitant administration of cyclosporine and warfarin, although this interaction is not well established. The mechanism is unknown. A similar reaction may occur with other oral anticoagulants.

MANAGEMENT: Close monitoring of prothrombin time (PT) and/or International Normalized Ratio(INR), as well as cyclosporine blood concentrations and the patient's response is recommended if these agents are used together.

Carbamazepine (oral)

MONITOR: Carbamazepine may induce the metabolism of warfarin and reduce its anticoagulant effect. The mechanism is induction of hepatic isoenzyme metabolism by carbamazepine. Carbamazepine may interact with other oral anticoagulants in a similar fashion.

MANAGEMENT: INR should be monitored whenever carbamazepine is started or discontinued in a patient receiving an oral anticoagulant. Patients should be advised to promptly report any signs of blood clots (e.g. chest pain, shortness of breath, sudden loss of vision, or pain, redness or swelling in an extremity).

Propoxyphene

MONITOR: Case reports have suggested that propoxyphene may increase the hypoprothrombinemic effects of warfarin. Bleeding episodes have been reported. However, causality is unclear because a propoxyphene/acetaminophen combination product was involved. Controlled studies have not demonstrated an interaction between propoxyphene and warfarin. The mechanism is unknown.

MANAGEMENT: Until more information is available, if propoxyphene and an anticoagulant must be used together, observation for altered anticoagulant effect is indicated. Patients should be advised to promptly report any signs of bleeding to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, vaginal bleeding, nosebleeds, bleeding of gums from brushing, unusual bleeding or bruising, red or brown urine, or red or black stools.

Cefoxitin

MONITOR: Cefoxitin has been associated with an increase in prothrombin time and episodes of bleeding. These effects may potentiate the effects of warfarin and other oral anticoagulants.

MANAGEMENT: The patient should be closely observed for signs of bleeding and the INR/PT should be monitored if these agents are given together.

Ceftriaxone

MONITOR: Ceftriaxone has been associated with an increase in prothrombin time and episodes of bleeding. These effects may potentiate the effects of warfarin and other oral anticoagulants.

MANAGEMENT: The patient should be closely observed for signs of bleeding and the INR/PT should be monitored if these agents are given together.

Chloral hydrate, Chloral hydrate rectal

MONITOR: Chloral hydrate may temporarily potentiate the hypoprothrombinemic effect of warfarin and similar anticoagulants. The mechanism is protein binding displacement by trichloroacetic acid, a major metabolite of chloral hydrate, resulting in increased plasma concentrations of unbound (active) anticoagulant drug. However, since drug clearance is also increased with higher levels of unbound drug, the pharmacokinetic effects produced by chloral hydrate tend to balance out over time, resulting in little or no net effect on anticoagulation therapy in the long-term. The manufacturer of triclofos cautions that it may also interact with anticoagulants.

MANAGEMENT: Based on these observations, appropriate monitoring for excessive hypoprothrombinemia (INR/PT) is recommended for several days following the addition of chloral hydrate or triclofos to a stabilized regimen of warfarin or other oral anticoagulants. Consideration may be given to using other sedative/hypnotic agents that have not been shown to interact with oral anticoagulants such as benzodiazepines or diphenhydramine. Patients should be advised to promptly report any signs of bleeding to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, vaginal bleeding, nosebleeds, bleeding of gums from brushing, unusual bleeding or bruising, red or brown urine, or red or black stools.

Cholestyramine

MONITOR: Cholestyramine may decrease the absorption and increase the elimination of warfarin or other oral anticoagulants. Anticoagulant effect may be reduced. However, cholestyramine also may impair vitamin K absorption, and the anticoagulant effect of warfarin may be enhanced in some patients. Starting or stopping cholestyramine in the presence of chronic oral anticoagulant therapy may lead to altered PT or INR.

MANAGEMENT: Warfarin or other oral anticoagulants should be administered at least one hour before or four to six hours after cholestyramine. The INR or PT should be closely monitored. Patients should be advised to promptly report any signs of bleeding (e.g., pain, swelling, headache, dizziness, weakness, vaginal bleeding, nosebleeds, bleeding of gums from brushing, unusual bleeding or bruising, red or brown urine, or red or black stools) or clots (e.g. chest pain, shortness of breath, sudden loss of vision, or pain, redness or swelling in an extremity).

Cimetidine

MONITOR: Cimetidine inhibits the hepatic metabolism of warfarin, and may increase its anticoagulant effect over a one to two week period. This interaction appears to involve the inactive S-warfarin more than the active R-enantiomer. This effect may occur with other oral anticoagulants.

MANAGEMENT: If given together, the INR or PT should be monitored, and the lowest possible dose of cimetidine should be used. Another histamine-2 antagonist may be used with less risk of interaction. Patients should be advised to promptly report any signs of bleeding to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, vaginal bleeding, nosebleeds, bleeding of gums from brushing, unusual bleeding or bruising, red or brown urine, or red or black stools.

Cisapride (oral)

MONITOR: Cisapride may enhance the effects of oral anticoagulants. The mechanism is unknown. Increases in coagulation times have been reported following the initiation of cisapride therapy in patients stabilized on anticoagulant therapy. However, other studies have reported increased warfarin requirements in patients taking cisapride.

MANAGEMENT: It is recommended that the INR or PT be monitored if cisapride is added, cisapride is discontinued, or the dosage is adjusted in a patient receiving anticoagulants. Patients should be advised to promptly report any signs of bleeding to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, vaginal bleeding, nosebleeds, bleeding of gums from brushing, unusual bleeding or bruising, red or brown urine, or red or black stools.

Temsirolimus

MONITOR CLOSELY: According to the package labeling, patients receiving temsirolimus in combination with anticoagulation therapy may be at increased risk of developing intracerebral bleeding, including fatal outcomes. The mechanism of interaction has not been described.

MANAGEMENT: Patients receiving anticoagulants should be informed of the increased risk of developing intracerebral bleeding while on temsirolimus.

Heparin

MONITOR CLOSELY: Although therapy with heparin and oral anticoagulants often overlap in clinical practice, the potential for additive anticoagulant effects and increased risk of severe bleeding should be considered. Oral anticoagulants may prolong the activated partial thromboplastin time (aPTT) in patients receiving heparin, while heparin may prolong the International Normalized Ratio (INR) in patients receiving warfarin.

MANAGEMENT: During concurrent or overlapping therapy, close clinical and laboratory observation for hematologic complications is recommended. To minimize the interference of heparin on INR determinations, blood should not be drawn until at least 5 hours after the last intravenous heparin dose or 24 hours after the last subcutaneous heparin dose.

Testosterone injection, Testosterone topical (patches and gel), Testosterone buccal system, Stanozolol, Oxandrolone

MONITOR CLOSELY: Androgens and anabolic steroids may potentiate the hypoprothrombinemic response to oral anticoagulants and increase the risk of bleeding. The onset of interaction is generally observed within 2 to 3 days. The mechanism is unknown. There have been case reports of patients stabilized on oral anticoagulant therapy who developed bleeding complications following the addition of various androgenic agents including danazol, oxymetholone, testosterone, methyltestosterone, and stanozolol. In a clinical study (n=15), the concomitant administration of oxandrolone 10 to 20 mg per day increased the half-life of the more active S(-) enantiomer of warfarin from 26 to 48 hours and the systemic exposure (AUC) from 4.55 to 12.08 ng-hr/mL. R(-) warfarin half-life and AUC also increased. Microscopic hematuria and gingival bleeding were reported, and the warfarin dose had to be reduced by 80% to 85% to maintain the desired INR.

MANAGEMENT: During concomitant therapy, the INR and/or PT should be monitored closely and anticoagulant dosage adjusted accordingly, particularly following initiation, discontinuation or change of dosage of the androgenic agent in patients who are stabilized on their anticoagulation regimen. Significant anticoagulant dose reductions may be required. Some experts recommend avoiding this combination altogether. Patients should be advised to promptly report any signs of bleeding to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, vaginal bleeding, nosebleeds, bleeding of gums from brushing, unusual bleeding or bruising, red or brown urine, or red or black stools.

Dasatinib

MONITOR CLOSELY: Coadministration of dasatinib and drugs that interfere with platelet function or coagulation may potentiate the risk of bleeding complications. Treatment with dasatinib is associated with severe thrombocytopenia. In addition, dasatinib has been shown to induce platelet dysfunction in vitro. Severe central nervous system hemorrhages, including fatalities, occurred in 1% of patients receiving dasatinib in clinical trials. Severe gastrointestinal hemorrhage occurred in 7% of patients and generally required treatment interruptions and transfusions. Other cases of severe hemorrhage occurred in 4% of patients. Most bleeding events were associated with severe thrombocytopenia.

MANAGEMENT: Concomitant use of other medications that interfere with platelet function or coagulation should be considered cautiously in patients treated with dasatinib. Close clinical and laboratory observation for bleeding complications is recommended during therapy. A complete blood count (CBC) with differential and platelet count should be obtained prior to and at least weekly for the first 2 months, then monthly thereafter or as clinically indicated. Patients should be advised to promptly report any signs of bleeding to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, vaginal bleeding, nosebleeds, bleeding of gums from brushing, unusual bleeding or bruising, red or brown urine, or red or black stools.

Ibritumomab

MONITOR CLOSELY: Coadministration of ibritumomab and drugs that interfere with platelet function or coagulation may potentiate the risk of bleeding complications. Ibritumomab commonly causes severe and prolonged thrombocytopenia, and the risk is increased further in patients with mild thrombocytopenia at the initiation of therapy. Hemorrhage, including fatal cerebral hemorrhage, has occurred in a minority of patients in clinical studies. The median time to thrombocyte nadir was 7 to 9 weeks, and the median duration of thrombocytopenia was 22 to 35 days. However, in less than 5% of the cases, patients experienced severe cytopenia that extended beyond 12 weeks following administration of the ibritumomab therapeutic regimen. Platelet transfusions were given to 22% of patients in clinical trials.

MANAGEMENT: Caution is advised if ibritumomab is used in combination with drugs that interfere with platelet function or coagulation. Close clinical and laboratory observation for bleeding complications is recommended during and after ibritumomab therapy.

Sulfasalazine, Sulfadiazine, Sulfisoxazole

MONITOR CLOSELY: Coadministration with a sulfonamide may increase the plasma concentrations and hypoprothrombinemic effects of coumarin anticoagulants. The proposed mechanism is sulfonamide inhibition of coumarin metabolism via CYP450 2C9 and possibly also displacement of coumarin derivatives from plasma protein binding sites. The interaction has been reported in controlled studies in healthy volunteers and has been associated with significant bleeding and elevation of prothrombin time in several case reports. A retrospective cohort study of acutely ill warfarin patients at a U.S. Veteran's Affairs Medical Center also revealed a significantly increased risk and severity of overanticoagulation following initiation of sulfamethoxazole-trimethoprim (SMX-TMP) compared to a control agent, terazosin. Specifically, a mean INR increase of 1.76 was observed in the SMX-TMP group (n=16), compared to a mean decrease of 0.15 in the terazosin group (n=29). Elevations in INR beyond therapeutic levels were seen in 69% of SMX-TMP patients versus 5% of the terazosin patients, and INR elevations beyond 4 were seen in 44% of the SMX-TMP patients versus 0% of the terazosin patients. Within the SMX-TMP group, 38% showed a mean change in INR of 2 or more points, and 31% showed a rise of 5 or more points. Adverse bleeding events occurred in 13% of the SMX-TMP patients, while none occurred in the controls. These results are supported by a case-control study that found SMX-TMP to be one of only two medications to significantly increase the risk of overanticoagulation in previously stable outpatients treated with phenprocoumon or acenocoumarol. In that study, 300 outpatients at a Netherlands anticoagulant clinic who presented with an INR value greater than or equal to 6.0 (median value 6.8) were compared with 302 randomly selected matched controls with INR values within the target range (median value 3.2), and changes in the use of 87 potentially interacting drugs or drug classes in the four weeks prior to the index day were identified and analyzed. A course of SMX-TMP strongly increased the risk of overanticoagulation even after adjustment for potential confounding factors, particularly in patients treated with acenocoumarol. A follow-up study focusing on antibiotic use in outpatients treated with phenprocoumon or acenocoumarol at a different Netherlands anticoagulant clinic also identified SMX-TMP use as a risk factor for overanticoagulation, with the relative risk most strongly increased four days or more after start of the antibiotic.

MANAGEMENT: Patients receiving coumarin anticoagulants should be closely monitored during concomitant therapy with sulfonamides. The INR should be checked frequently and coumarin dosage adjusted accordingly, particularly following initiation or discontinuation of sulfonamide therapy in patients who are stabilized on their anticoagulant regimen. The same precaution may be applicable during therapy with other oral anticoagulants (e.g., indandiones), although clinical data are lacking. Patients should be advised to promptly report any signs of bleeding to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, vaginal bleeding, nosebleeds, bleeding of gums from brushing, unusual bleeding or bruising, red or brown urine, or red or black stools.

Erythromycin, Clarithromycin, Troleandomycin

MONITOR CLOSELY: Coadministration with clarithromycin or erythromycin may infrequently but substantially enhance the hypoprothrombinemic effect of warfarin and other coumarin anticoagulants. The exact mechanism of interaction is unknown. Data from clinical studies have not supported a significant, predictable pharmacodynamic or pharmacokinetic interaction in general. Although both macrolides are potent inhibitors of CYP450 3A4 and can inhibit metabolism of the R(+) enantiomer of warfarin, the overall effect on racemic warfarin pharmacokinetics appears to be minor. In 12 normal subjects, the clearance of warfarin (1 mg/kg single dose) decreased by an average of 14% following pretreatment with erythromycin 250 mg four times a day for 8 days. In a study of eight patients stabilized on warfarin, the addition of erythromycin led to only a small increase in the effect of warfarin. Nevertheless, a population-based cohort study focusing on antibiotic use in outpatients treated with phenprocoumon or acenocoumarol at a Netherlands anticoagulant clinic identified clarithromycin use as a risk factor for overanticoagulation (INR greater than or equal to 6), even after adjustment for potential confounding factors. There have also been case reports of increased prothrombin time or INR and/or serious bleeding complications in patients stabilized on coumarin therapy following the addition of erythromycin or clarithromycin. The UK Committee on Safety of Medicines reported on a woman taking warfarin who suffered a fatal cerebrovascular bleed 3 days after starting clarithromycin. However, other influences such as fever, infection, malnutrition, and other concomitant underlying conditions on clotting mechanisms and coumarin pharmacokinetics should also be considered.

MANAGEMENT: Given the potential for clinically significant interaction and even fatality in the occasional, susceptible patient, close monitoring is recommended if clarithromycin or erythromycin is prescribed during coumarin anticoagulant therapy. The INR should be checked frequently and coumarin dosage adjusted accordingly, particularly following initiation or discontinuation of macrolide therapy in patients who are stabilized on their anticoagulant regimen. Patients should be advised to promptly report any signs of bleeding to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, vaginal bleeding, nosebleeds, bleeding of gums from brushing, unusual bleeding or bruising, red or brown urine, or red or black stools. The same precaution may be applicable during therapy with other oral anticoagulants (e.g., indandiones) and other similar macrolides (e.g., troleandomycin), although clinical data are lacking.

Fluconazole

MONITOR CLOSELY: Coadministration with fluconazole may significantly increase the plasma concentrations and hypoprothrombinemic effect of warfarin. The mechanism is fluconazole inhibition of CYP450 2C9, the isoenzyme responsible for the metabolic clearance of the biologically more active S(-) enantiomer of warfarin. Additionally, fluconazole inhibits CYP450 2C19 and 3A4, which are responsible for the metabolism of R(+) warfarin. In six healthy, nonsmoking volunteers, coadministration of fluconazole (400 mg once daily for 14 days) and racemic warfarin (0.75 mg/kg single oral dose) resulted in an increase in the mean systemic exposure (AUC) of S(-) and R(+) warfarin by 184% and 108%, respectively, compared to administration of warfarin alone. Likewise, the mean plasma half-life of S(-) and R(+) warfarin increased by 175% and 111%, respectively, in the presence of fluconazole. All subjects also had markedly elevated prothrombin time (PT) during fluconazole coadministration compared to warfarin alone, including one whose PT increased to 34.2 seconds at 120 hours post-warfarin dose. In another study, 11 patients stabilized on warfarin all exhibited progressive PT increases following fluconazole 100 mg/day for 8 days. Mean PT increased from 15.8 seconds at baseline to 21.9 seconds on day 8, and fluconazole was stopped early in three patients due to high PTs. Clinically, the interaction has been associated with bleeding complications including intraocular and gastrointestinal hemorrhage and spinal epidural hematoma. No data are available for other oral anticoagulants, although at least one other coumarin derivative is known to be metabolized by CYP450 2C9.

MANAGEMENT: Given the potential for interaction and the high degree of interpatient variability with respect to warfarin metabolism, patients should be closely monitored during concomitant therapy with fluconazole. The INR should be checked frequently and warfarin dosage adjusted accordingly, particularly following initiation or discontinuation of fluconazole in patients who are stabilized on their warfarin regimen. The same precaution may be applicable during therapy with other oral anticoagulants, although clinical data are lacking. Patients should be advised to promptly report any signs of bleeding to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, vaginal bleeding, nosebleeds, bleeding of gums from brushing, unusual bleeding or bruising, red or brown urine, or red or black stools.

Metronidazole

MONITOR CLOSELY: Coadministration with metronidazole may increase the plasma concentrations and hypoprothrombinemic effect of warfarin. The proposed mechanism is metronidazole inhibition of CYP450 2C9, the isoenzyme responsible for the metabolic clearance of the more active S(-) enantiomer of warfarin. The interaction was associated with significant bleeding and elevation of prothrombin time in two case reports. No data are available for other oral anticoagulants, although at least one other coumarin derivative is known to be metabolized by CYP450 2C9.

MANAGEMENT: Given the potential for interaction and the high degree of interpatient variability with respect to warfarin metabolism, patients should be closely monitored during concomitant therapy with metronidazole. The INR should be checked frequently and warfarin dosage adjusted accordingly, particularly following initiation or discontinuation of metronidazole in patients who are stabilized on their warfarin regimen. The same precaution may be applicable during therapy with other oral anticoagulants, although clinical data are lacking. Patients should be advised to promptly report any signs of bleeding to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, vaginal bleeding, nosebleeds, bleeding of gums from brushing, unusual bleeding or bruising, red or brown urine, or red or black stools.

Voriconazole

MONITOR CLOSELY: Coadministration with voriconazole may significantly increase the plasma concentrations and hypoprothrombinemic effect of warfarin. The mechanism is voriconazole inhibition of CYP450 2C9, 2C19 and 3A4, the isoenzymes responsible for the metabolic clearance of racemic warfarin. In healthy volunteers, coadministration of voriconazole (300 mg every 12 hours for 12 days) with warfarin (30 mg single dose on day 7) significantly increased maximum prothrombin time by approximately 2 times that observed with placebo in healthy subjects. Prothrombin times were still increased by a mean value of 5.4 seconds 144 hours post warfarin dose following coadministration with voriconazole.

MANAGEMENT: Patients should be closely monitored during concomitant therapy with warfarin and voriconazole. The INR should be checked frequently and warfarin dosage adjusted accordingly, particularly following initiation, discontinuation or change of dosage of voriconazole in patients who are stabilized on their anticoagulant regimen. Patients should be advised to promptly report any signs of bleeding to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, vaginal bleeding, nosebleeds, bleeding of gums from brushing, unusual bleeding or bruising, red or brown urine, or red or black stools. The same precaution may be applicable during therapy with other oral anticoagulants, although clinical data are lacking.

Nafcillin, Dicloxacillin

MONITOR CLOSELY: Dicloxacillin and nafcillin may diminish the hypoprothrombinemic effect of warfarin. The exact mechanism of interaction is unknown but may involve induction of hepatic warfarin metabolism by these penicillins. There have been case reports of patients who developed warfarin resistance following dicloxacillin or high-dose nafcillin therapy, resulting in increased warfarin requirement. In one case report, a 29-year-old male on long-term, stable anticoagulant therapy developed resistance to warfarin while receiving IV nafcillin 12 gm/day. His prothrombin time (PT) fell from 20 to 25 seconds down to 14 to 17 seconds despite doubling of his warfarin dosage. The warfarin half-life was 11 hours during nafcillin therapy and 44 hours eight months after discontinuation of nafcillin. In a retrospective review of patients enrolled in an anticoagulant clinic, investigators identified seven patients who had been stabilized on their warfarin regimen for at least 2 weeks prior to starting dicloxacillin therapy. A decrease in PT occurred approximately 4 to 5 days following initiation of dicloxacillin. The mean number of days from the initiation of dicloxacillin to the lowest recorded PT was 11.4 and the mean number of days from the last day of dicloxacillin therapy to the return of the PT to baseline was 22. The mean of the baseline and lowest recorded PTs were 18.2 and 15.1 seconds, respectively, representing a 17% decrease. Five patients required supplemental doses and/or dosage increases of warfarin. Another study found that dicloxacillin treatment led to a mean decline in PT of 9.1% (range 2% to 24%) in seven patients stabilized on warfarin for at least 2 weeks. No data are available for other oral anticoagulants.

MANAGEMENT: Coagulation parameters should be monitored more closely during coadministration of warfarin with dicloxacillin or nafcillin and for several weeks after discontinuation of the antibiotic. The same precaution may be applicable during therapy with other oral anticoagulants.

Fluorouracil (injection), Capecitabine

MONITOR CLOSELY: Fluorouracil and its prodrug capecitabine may significantly potentiate the hypoprothrombinemic effect of warfarin and other oral anticoagulants. The mechanism of the interaction is unknown. Altered coagulation parameters and/or bleeding have been reported in patients stabilized on a coumarin-derivative anticoagulant such as warfarin following the addition of fluorouracil or capecitabine. Based on reported cases, increases in INR or prothrombin time may occur within a couple days after initiation of chemotherapy and continue gradually over several months and even after stopping chemotherapy. Tegafur, another prodrug, may theoretically also interact.

MANAGEMENT: Patients should be closely monitored during concomitant therapy. The INR or PT should be checked frequently and anticoagulant dosage adjusted accordingly, particularly following initiation or discontinuation of fluorouracil or capecitabine therapy in patients who are stabilized on their anticoagulant regimen. Patients should be advised to promptly report any signs of bleeding to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, vaginal bleeding, nosebleeds, bleeding of gums from brushing, unusual bleeding or bruising, red or brown urine, or red or black stools.

Enoxacin, Lomefloxacin, Norfloxacin, Ofloxacin, Sparfloxacin, Levofloxacin, Grepafloxacin, Trovafloxacin, Moxifloxacin, Gatifloxacin, Gemifloxacin

MONITOR CLOSELY: Some quinolone antibiotics have been reported to potentiate the hypoprothrombinemic effect of warfarin and other coumarin anticoagulants. The exact mechanism is unknown but may involve inhibition of coumarin metabolism and/or depletion of certain clotting factors due to suppression of vitamin K-producing intestinal flora. A retrospective cohort study of acutely ill warfarin patients at a U.S. Veteran's Affairs Medical Center revealed an increased risk and severity of overanticoagulation following initiation of levofloxacin compared to a control agent, terazosin. Specifically, a mean INR increase of 0.85 was observed in the levofloxacin group (n=27), compared to a mean decrease of 0.15 in the terazosin group (n=29). Elevations in INR beyond therapeutic levels were seen in 33% of levofloxacin patients versus 5% of the terazosin patients, and INR elevations beyond 4 were seen in 19% of the levofloxacin patients versus 0% of the terazosin patients (the latter not statistically significant). Similarly, a population-based cohort study focusing on antibiotic use in outpatients treated with phenprocoumon or acenocoumarol at a Netherlands anticoagulant clinic identified norfloxacin use as a risk factor for overanticoagulation (INR greater than or equal to 6), even after adjustment for potential confounding factors. There have also been case reports of primarily elderly patients stabilized on warfarin who developed PT or INR increases and/or bleeding complications following the addition of a quinolone. Increased INR values have generally been observed within 2 to 16 days following initiation of quinolone therapy. Ciprofloxacin, enoxacin, gatifloxacin, levofloxacin, moxifloxacin, nalidixic acid, norfloxacin, ofloxacin, and trovafloxacin have been specifically implicated. Between 1987 and 1997, the U.S. Food and Drug Administration received 66 reported cases of suspected interaction involving ciprofloxacin alone. As of January 2004, Health Canada received 57 reports (levofloxacin 16; gatifloxacin 13; moxifloxacin 12; ciprofloxacin 10; norfloxacin 6), including four deaths involving gatifloxacin (2), ciprofloxacin (1), and levofloxacin (1). However, causality could not be established due to multiple confounding factors. In general, data from clinical studies with various quinolones have not supported a significant, predictable pharmacodynamic or pharmacokinetic interaction with warfarin. However, the potential for interaction in susceptible patients cannot be ruled out. Other influences such as fever, infection, malnutrition, and other concomitant underlying conditions on clotting mechanisms and warfarin pharmacokinetics should also be considered.

MANAGEMENT: Given the potential for clinically significant interaction and even fatality in the occasional, susceptible patient, close monitoring is recommended if a quinolone antibiotic is prescribed during coumarin anticoagulant therapy. The INR should be checked frequently and coumarin dosage adjusted accordingly, particularly following initiation or discontinuation of quinolone therapy in patients who are stabilized on their anticoagulant regimen. Patients should be advised to promptly report any signs of bleeding to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, vaginal bleeding, nosebleeds, bleeding of gums from brushing, unusual bleeding or bruising, red or brown urine, or red or black stools. The same precaution may be applicable during therapy with other oral anticoagulants (e.g., indandiones), although clinical data are lacking.

Miconazole vaginal, Miconazole topical

MONITOR CLOSELY: Systemically or even topically administered miconazole may increase the plasma concentrations and hypoprothrombinemic effect of warfarin. The proposed mechanism is miconazole inhibition of CYP450 2C9, the isoenzyme responsible for the metabolic clearance of the more active S(-) enantiomer of warfarin. There have been case reports of patients stabilized on warfarin who developed bleeding complications, bruising, and/or significantly increased prothrombin time (PT) or INR following the addition of miconazole. The interaction has also been reported with other oral anticoagulants and reportedly may occur up to 2 weeks after initiation of miconazole. Although the interaction is most likely to occur with systemic miconazole, it has been reported occasionally with oral gel and intravaginal formulations. In one case, a 52-year-old woman taking warfarin experienced hemorrhage of the right kidney after 12 days of using vaginal miconazole. Her PT and partial thromboplastin time were elevated upon hospitalization but returned to normal after discontinuation of miconazole use. There is also a reported case of an 80-year-old man stabilized on warfarin who developed a marked increase in his INR following application of a miconazole cream to his groin area for 2 weeks. Another 80-year-old man had a cerebral vascular accident following application of miconazole, although causality could not be established due to multiple medical problems and concurrent medications. A handful of cases involved oral miconazole gel, generally when it was applied to inflamed or compromised oral mucosa or when substantial amounts were swallowed following application, resulting in complications including epistaxis, spontaneous bruising, hematoma, hematuria, melena, and hypotension.

MANAGEMENT: Patients receiving warfarin or other oral anticoagulants should be closely monitored during concomitant therapy with miconazole. The INR should be checked frequently and anticoagulant dosage adjusted accordingly, particularly following initiation or discontinuation of miconazole therapy in patients who are stabilized on their anticoagulant regimen. Patients should be advised to promptly report any signs of bleeding to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, vaginal bleeding, nosebleeds, bleeding of gums from brushing, unusual bleeding or bruising, red or brown urine, or red or black stools. Although vaginally administered miconazole is generally less than 2% absorbed in healthy women of childbearing age, increased absorption may occur in the presence of atrophic vaginal epithelium. Therefore, the same precaution is applicable when vaginal formulations of miconazole is prescribed to women receiving oral anticoagulant therapy.

Tipranavir

MONITOR CLOSELY: Theoretically, tipranavir may potentiate the risk of bleeding in patients treated with agents that affect hemostasis such as anticoagulants, platelet inhibitors, thrombin inhibitors, thrombolytic agents, or agents that commonly cause thrombocytopenia. Tipranavir has been shown to inhibit human platelet aggregation in vitro at levels consistent with exposures observed in patients receiving tipranavir/ritonavir. Cases of intracranial hemorrhage, some fatal, have been reported during postmarketing use. However, many of these patients had other medical conditions or were receiving concomitant medications that may have caused or contributed to these events. No pattern of abnormal coagulation parameters has been observed in patients in general, or preceding the development of intracranial hemorrhage.

MANAGEMENT: Caution is advised if tipranavir/ritonavir is used in combination with other drugs that affect hemostasis. Patients should be advised to promptly report any signs of bleeding to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, vaginal bleeding, nosebleeds, bleeding of gums from brushing, unusual bleeding or bruising, red or brown urine, or red or black stools.

Aprepitant

MONITOR: Coadministration of aprepitant during warfarin therapy may result in clinically significant decreases in the INR or prothrombin time. The likely mechanism is aprepitant induction of CYP450 2C9, the isoenzyme responsible for metabolic clearance of the biologically more active S(-) enantiomer of warfarin. According to the product labeling, healthy subjects stabilized on chronic warfarin therapy given aprepitant (125 mg on day 1 and 80 mg/day on days 2 and 3) had a 34% decrease in S(-) warfarin trough concentration accompanied by a 14% decrease in the INR five days after completion of aprepitant dosing. There was no effect on the systemic exposure (AUC) of R(+) and S(-) warfarin determined on day 3. It is not known if the interaction occurs with other oral anticoagulants, although at least one other coumarin derivative is known to be metabolized by CYP450 2C9.

MANAGEMENT: In patients treated with warfarin, the INR should be closely monitored in the 2-week period, particularly at 7 to 10 days, following initiation of the 3-day regimen of aprepitant with each chemotherapy cycle or following a single 40 mg dose of aprepitant for the prevention of postoperative nausea and vomiting. The same precaution may be applicable during therapy with other oral anticoagulants, although clinical data are lacking.

Sorafenib

MONITOR: Coadministration of oral anticoagulants in combination with sorafenib may potentiate the risk of bleeding. Infrequent bleeding events or elevations in the INR have been reported in some patients taking warfarin while on sorafenib therapy. The mechanism of interaction is unknown, although an increased risk of bleeding may be associated with sorafenib alone. In a phase 3 clinical study, bleeding regardless of causality was reported in 15.3% of patients in the sorafenib group compared to 8.2% of patients in the placebo group. Sorafenib does not appear to significantly affect the pharmacokinetics of warfarin.

MANAGEMENT: Caution is advised if oral anticoagulants are prescribed with sorafenib. The INR should be checked frequently and anticoagulant dosage adjusted accordingly, particularly following initiation, discontinuation or change of dosage of sorafenib in patients who are stabilized on their anticoagulant regimen. Patients should be advised to promptly report any signs of bleeding to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, vaginal bleeding, nosebleeds, bleeding of gums from brushing, unusual bleeding or bruising, red or brown urine, or red or black stools.

Sildenafil (oral)

MONITOR: Coadministration of sildenafil with oral vitamin K antagonists in patients with pulmonary arterial hypertension has been associated with an increased risk of bleeding (primarily epistaxis) compared to administration without oral vitamin K antagonists. The incidence of epistaxis was 9% with oral vitamin K antagonists versus 2% with placebo.

MANAGEMENT: Caution is advised in patients with pulmonary arterial hypertension receiving sildenafil and oral vitamin K antagonists. The INR should be checked frequently and patients should be advised to promptly report any signs of bleeding to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, vaginal bleeding, nosebleeds, bleeding of gums from brushing, unusual bleeding or bruising, red or brown urine, or red or black stools.

Atazanavir (oral)

MONITOR: Coadministration with atazanavir may increase the plasma concentrations of warfarin. The proposed mechanism is atazanavir inhibition of CYP450 1A2 and 2C9, the isoenzymes responsible for the metabolic clearance of both the R(+) and S(-) enantiomers of warfarin, respectively. The interaction has not been studied but could conceivably lead to serious and/or life-threatening bleeding. No data are available for other oral anticoagulants, although at least one other coumarin derivative is known to be metabolized by CYP450 2C9.

MANAGEMENT: Given the potential for interaction and the high degree of interpatient variability with respect to warfarin metabolism, patients should be closely monitored during concomitant therapy with atazanavir. The INR should be checked frequently and warfarin dosage adjusted accordingly, particularly following initiation or discontinuation of atazanavir in patients who are stabilized on their warfarin regimen. The same precaution may be applicable during therapy with other oral anticoagulants. Patients should be advised to promptly report any signs of bleeding to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, vaginal bleeding, nosebleeds, bleeding of gums from brushing, unusual bleeding or bruising, red or brown urine, or red or black stools.

Azithromycin

MONITOR: Coadministration with azithromycin may occasionally enhance the hypoprothrombinemic effect of warfarin. The exact mechanism of interaction is unknown. Azithromycin does not inhibit CYP450 enzymes, and several studies have found no evidence of a significant, predictable interaction with warfarin. Nevertheless, a retrospective cohort study of acutely ill warfarin patients at a U.S. Veteran's Affairs Medical Center revealed an increased risk and severity of overanticoagulation following initiation of azithromycin compared to a control agent, terazosin. Specifically, a mean INR increase of 0.51 was observed in the azithromycin group (n=32), compared to a mean decrease of 0.15 in the terazosin group (n=29). Elevations in INR beyond therapeutic levels were seen in 31% of azithromycin patients versus 5% of the terazosin patients, and INR elevations beyond 4 were seen in 16% of the azithromycin patients versus 0% of the terazosin patients (the latter not statistically significant). There have also been case reports of patients stabilized on warfarin who developed PT or INR increases during or immediately following a course of azithromycin. One patient was hospitalized after coughing up blood and blood-streaked mucus two days following the completion of a 5-day course of azithromycin. Two other patients developed an intra-abdominal hematoma in association with elevated INRs shortly after the initiation of azithromycin therapy. The interaction has generally been observed within 8 days after initiation of azithromycin treatment, or 3 days after completion of a 5-day course. However, a causal relationship is difficult to establish in most cases due to confounding factors such as severe infection, fever, liver disease, hypoalbuminemia, anorexia, diarrhea, and other concomitant therapy.

MANAGEMENT: Given the potential for clinically significant interaction in the occasional, susceptible patient, close monitoring is recommended if azithromycin is prescribed during warfarin therapy. The INR should be checked frequently and warfarin dosage adjusted accordingly, particularly following initiation or discontinuation of azithromycin therapy in patients who are stabilized on their anticoagulant regimen. Patients should be advised to promptly report any signs of bleeding to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, vaginal bleeding, nosebleeds, bleeding of gums from brushing, unusual bleeding or bruising, red or brown urine, or red or black stools. The same precaution may be applicable during therapy with other oral anticoagulants, although clinical data are lacking.

Bortezomib

MONITOR: Coadministration with bortezomib may increase the plasma concentrations of drugs that are substrates of the CYP450 2C19 isoenzyme. The mechanism is decreased clearance due to inhibition of the isoenzyme by bortezomib.

MANAGEMENT: Caution is advised if bortezomib must be used concomitantly with medications that undergo metabolism by CYP450 2C19, particularly those with a narrow therapeutic range. Dosage adjustments as well as clinical and laboratory monitoring may be appropriate for some drugs whenever bortezomib is added to or withdrawn from therapy.

Bosentan

MONITOR: Coadministration with bosentan may decrease the plasma concentrations of warfarin and other oral anticoagulants. The mechanism is bosentan induction of CYP450 2C9 and 3A4, two isoenzymes that are responsible for the metabolic clearance of warfarin. According to the product labeling, bosentan (500 mg orally twice a day for 6 days) decreased the plasma concentrations of the biologically less active R(+) enantiomer of warfarin by 38% and the more active S(-) enantiomer by 29%. In clinical studies involving patients with pulmonary arterial hypertension treated with bosentan and warfarin, clinically relevant changes in INR or warfarin dosage were not observed (baseline vs. end of studies), and the need for warfarin dosage adjustments due to changes in INR or emergence of adverse events was similar among the bosentan and placebo groups. However, there has been a case report of a 35-year-old patient stabilized on warfarin for 3 months who had a decrease in INR 10 days after starting bosentan, which subsequently required a 64% increase in her warfarin requirement. No data are available for other oral anticoagulants, although at least one other coumarin derivative is known to be metabolized by CYP450 2C9.

MANAGEMENT: Patients should be closely monitored during concomitant therapy with oral anticoagulants and bosentan. The INR should be checked frequently and anticoagulant dosage adjusted accordingly, particularly following initiation, discontinuation, or change of dosage of bosentan in patients who are stabilized on their anticoagulant regimen.

Celecoxib (oral)

MONITOR: Coadministration with celecoxib may infrequently enhance the hypoprothrombinemic effect of warfarin. The proposed mechanism is competitive inhibition of the CYP450 2C9 metabolism of warfarin by celecoxib, as both drugs are substrates of this isoenzyme. A study involving 24 healthy volunteers found no effect of celecoxib (200 mg twice daily) on the prothrombin time or steady-state pharmacokinetics of warfarin. However, there have been isolated post-marketing case reports of elderly patients stabilized on warfarin who developed INR increases and/or bleeding complications following the addition of celecoxib. It is possible that patients with variant alleles resulting in reduced CYP450 2C9 metabolism (i.e., CYP450 2C9 poor metabolizers) may be at increased risk for this interaction, since they have reduced clearance of both celecoxib and warfarin.

MANAGEMENT: Given the potential for interaction and the high degree of interpatient variability with respect to warfarin metabolism, patients should be closely monitored during concomitant therapy with celecoxib. The INR should be checked frequently and warfarin dosage adjusted accordingly, particularly following initiation, discontinuation or change of dosage of celecoxib in patients who are stabilized on their warfarin regimen. The same precaution may be applicable during therapy with other oral anticoagulants, although clinical data are lacking. Patients should be advised to promptly report any signs of bleeding to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, vaginal bleeding, nosebleeds, bleeding of gums from brushing, unusual bleeding or bruising, red or brown urine, or red or black stools.

Fluvoxamine

MONITOR: Coadministration with fluvoxamine may increase the plasma concentrations and hypoprothrombinemic effect of warfarin and similar anticoagulants. The proposed mechanism is fluvoxamine inhibition of multiple CYP450 enzymes that are involved in the metabolism of oral anticoagulants. According to the manufacturer, coadministration of fluvoxamine (50 mg three times a day) and warfarin for two weeks resulted in a 98% increase in warfarin plasma concentrations and prolonged prothrombin times. In one case report, an elderly woman stabilized on warfarin developed an elevated INR shortly after citalopram was changed to fluvoxamine during a hospitalization. The INR elevation persisted for 7 days, even several days after fluvoxamine has been discontinued. The prolonged effect may be due to the relatively long half-life of fluvoxamine, particularly in the elderly.

MANAGEMENT: Caution is advised if fluvoxamine must be used with warfarin or other oral anticoagulants. The INR should be checked frequently and anticoagulant dosage adjusted accordingly, particularly following initiation, discontinuation or change of dosage of fluvoxamine in patients who are stabilized on their anticoagulant regimen. Patients should be advised to promptly report any signs of bleeding to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, vaginal bleeding, nosebleeds, bleeding of gums from brushing, unusual bleeding or bruising, red or brown urine, or red or black stools. Alternatively, fluvoxamine may be substituted with an antidepressant that is less likely to interfere with CYP450 metabolism, such as citalopram, sertraline, or venlafaxine.

Gefitinib

MONITOR: Coadministration with gefitinib has resulted in INR elevations and/or bleeding events in some patients taking warfarin. The mechanism of interaction has not been established, and it is not known whether other oral anticoagulants are similarly affected.

MANAGEMENT: Given the potential for interaction and the high degree of interpatient variability with respect to warfarin metabolism, patients should be closely monitored during concomitant therapy with gefitinib. The INR should be checked frequently and warfarin dosage adjusted accordingly, particularly following initiation or discontinuation of gefitinib in patients who are stabilized on their warfarin regimen. Patients should be advised to promptly report any signs of bleeding to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, vaginal bleeding, nosebleeds, bleeding of gums from brushing, unusual bleeding or bruising, red or brown urine, or red or black stools.

Orlistat

MONITOR: Coadministration with orlistat may potentiate the hypoprothrombinemic effect of oral anticoagulants. The proposed mechanism is orlistat reduction of the absorption of vitamin K, a fat-soluble vitamin and cofactor in the synthesis of blood clotting factors that are inhibited by oral anticoagulants. In addition, dietary changes often prompted by orlistat use may also alter vitamin K intake and contribute to the interaction. A case report describes a 66-year-old patient stabilized on warfarin who was found to have an increased INR during a follow-up visit 18 days after starting orlistat. A significant reduction in warfarin dosage was subsequently required to maintain INR in the therapeutic range while receiving orlistat.

MANAGEMENT: Patients should be closely monitored during concomitant therapy with oral anticoagulants and orlistat. The INR should be checked frequently and anticoagulant dosage adjusted accordingly, particularly following initiation or discontinuation of orlistat in patients who are stabilized on their anticoagulant regimen. Patients should be advised to promptly report any signs of bleeding to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, vaginal bleeding, nosebleeds, bleeding of gums from brushing, unusual bleeding or bruising, red or brown urine, or red or black stools.

Saquinavir, Ritonavir, Indinavir, Nelfinavir, Amprenavir, Fosamprenavir, Darunavir

MONITOR: Coadministration with protease inhibitors may alter the plasma concentrations and/or pharmacologic effects of oral anticoagulants. Both increases and decreases in INR have been reported. The exact mechanism of interaction is unknown. In a couple of cases, dramatic decreases in INR occurred following the addition of ritonavir. In one case, warfarin dosage had to almost double in order to maintain a therapeutic INR, and when ritonavir was discontinued months later, the INR rose rapidly and the warfarin dosage was significantly reduced. In the other, acenocoumarol dosage was progressively increased to 3 times the original dosage but failure to achieve the desired INR led to eventual withdrawal of ritonavir, whereupon INR returned to the previous level in 4 days. Nelfinavir was then substituted for ritonavir and also resulted in an interaction, but a therapeutic INR was achieved after a 210% increase in the acenocoumarol dosage. Another case report implicated indinavir in a similar interaction of a lesser magnitude. In contrast, a significant increase in INR occurred in one patient following a switch from efavirenz and abacavir to a regimen containing ritonavir, nelfinavir, zidovudine, and lamivudine. The patient's INR was low on days he skipped or lowered the ritonavir dosage and high on days in which he was compliant. In another instance, an interaction with saquinavir was thought to have caused the hypoprothrombinemia in a patient stabilized on warfarin, and a reduction of warfarin dosage by 20% was required.

MANAGEMENT: Caution is advised if oral anticoagulants must be used with protease inhibitors. The INR should be checked frequently and anticoagulant dosage adjusted accordingly, particularly following initiation, discontinuation or change of dosage of the protease inhibitor(s) in patients who are stabilized on their anticoagulant regimen.

Omeprazole, Lansoprazole, Rabeprazole, Pantoprazole, Esomeprazole

MONITOR: Coadministration with proton pump inhibitors (PPIs) has occasionally been associated with enhanced hypoprothrombinemic effect of warfarin. The exact mechanism is unknown but may involve PPI inhibition of CYP450 2C19 and/or 3A4, the isoenzymes partially responsible for the metabolic clearance of the biologically less active R(+) enantiomer of warfarin. There have been reports of increased INR and prothrombin time in patients receiving warfarin with various commercially available proton pump inhibitors. However, a significant pharmacokinetic interaction has not been reported. In one study, coadministration of omeprazole 20 mg/day and warfarin (individualized dosage) in 21 healthy, young men for 2 weeks resulted in a 12% increase in the mean plasma concentration of R(+) warfarin compared to coadministration with placebo. Plasma concentrations of the S(-) enantiomer were unaffected, and no clinically significant alterations in coagulation times were noted. Similar results were reported in a group of 28 patients on continuous therapy with warfarin given omeprazole 20 mg/day for 3 weeks. Additionally, another study reported no pharmacokinetic interaction between warfarin and pantoprazole.

MANAGEMENT: Given the potential for interaction and the high degree of interpatient variability with respect to warfarin metabolism, patients should be closely monitored during concomitant therapy with PPIs. The INR should be checked frequently and warfarin dosage adjusted accordingly, particularly following initiation, discontinuation or change of dosage of PPI in patients who are stabilized on their warfarin regimen. The same precaution may be applicable during therapy with other oral anticoagulants. Patients should be advised to promptly report any signs of bleeding to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, vaginal bleeding, nosebleeds, bleeding of gums from brushing, unusual bleeding or bruising, red or brown urine, or red or black stools.

Rifapentine

MONITOR: Coadministration with rifapentine may decrease the plasma concentrations of drugs that are substrates of the CYP450 2C8, 2C9, and/or 3A4 isoenzymes. The mechanism is accelerated clearance due to induction of those isoenzymes by rifapentine. Enzyme activities may be induced within 4 days of the first dose and return to normal 14 days after discontinuation of rifapentine. In vitro and in vivo enzyme studies have suggested rifapentine induction potential to be less than that of rifampin but greater than that of rifabutin. In addition, the magnitude of induction is dependent on dose and dosing frequency.

MANAGEMENT: When drugs that are known substrates of CYP450 2C8, 2C9, and/or 3A4 are coadministered with rifapentine, the possibility of a diminished therapeutic response to those drugs should be considered. Dosage adjustments as well as clinical and laboratory monitoring may be appropriate for some drugs, particularly those with a narrow therapeutic range, whenever rifapentine is added to or withdrawn from therapy.

Telithromycin

MONITOR: Coadministration with telithromycin may infrequently enhance the hypoprothrombinemic effect of warfarin. The exact mechanism of interaction is unknown. Although telithromycin is a potent inhibitor of CYP450 3A4 and can inhibit metabolism of the R(+) enantiomer of warfarin, the overall effect on racemic warfarin pharmacokinetics appears to be minor. According to the product labeling, no pharmacodynamic or pharmacokinetic effects on racemic warfarin were observed when telithromycin was coadministered in healthy subjects. Nevertheless, a case report describes a 73-year-old man stabilized on warfarin who developed an increased INR (from 3.1 to 11) and mild hemoptysis five days after starting telithromycin 800 mg/day. The INR returned to the therapeutic range 4 days after telithromycin was discontinued.

MANAGEMENT: Given the potential for clinically significant interaction in the occasional, susceptible patient, close monitoring is recommended if telithromycin is prescribed during warfarin therapy. The INR should be checked frequently and coumarin dosage adjusted accordingly, particularly following initiation or discontinuation of telithromycin therapy in patients who are stabilized on their anticoagulant regimen. Patients should be advised to promptly report any signs of bleeding to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, vaginal bleeding, nosebleeds, bleeding of gums from brushing, unusual bleeding or bruising, red or brown urine, or red or black stools. The same precaution may be applicable during therapy with other oral anticoagulants, although clinical data are lacking.

Tigecycline

MONITOR: Coadministration with tigecycline may increase the plasma concentrations of warfarin. The mechanism of interaction has not been described, although it seems to affect the biologically less active R(+) enantiomer of warfarin more than the S(-) enantiomer. In healthy subjects, coadministration of tigecycline (100 mg followed by 50 mg every 12 hours) and warfarin (25 mg single oral dose) resulted in a decrease in clearance of R(+) and S(-) warfarin by 40% and 23%, an increase in the peak plasma concentration (Cmax) by 38% and 43%, and an increase in the systemic exposure (AUC) by 68% and 29%, respectively. However, the hypoprothrombinemic effects of warfarin were not affected as measured by the INR. Warfarin did not alter the pharmacokinetics of tigecycline.

MANAGEMENT: Given the potential for interaction and the high degree of interpatient variability with respect to warfarin metabolism, patients should be closely monitored during concomitant therapy with tigecycline. The INR should be checked frequently and warfarin dosage adjusted accordingly, particularly following initiation or discontinuation of tigecycline in patients who are stabilized on their warfarin regimen. The same precaution may be applicable during therapy with other oral anticoagulants, although clinical data are lacking. Patients should be advised to promptly report any signs of bleeding to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, vaginal bleeding, nosebleeds, bleeding of gums from brushing, unusual bleeding or bruising, red or brown urine, or red or black stools.

Tinidazole

MONITOR: Coadministration with tinidazole may enhance the hypoprothrombinemic effect of warfarin and other oral anticoagulants. The interaction has been reported with metronidazole, another agent in the nitroimidazole class, and warfarin. Significant bleeding and elevation of prothrombin time have been observed. The proposed mechanism is metronidazole inhibition of CYP450 2C9, the isoenzyme responsible for the metabolic clearance of the more active S(-) enantiomer of warfarin. No data are available for tinidazole.

MANAGEMENT: Given its structural similarities to metronidazole, the potential for interaction with warfarin and other oral anticoagulants should be considered during coadministration with tinidazole. The INR should be checked frequently and anticoagulant dosage adjusted accordingly, particularly following initiation and for up to 8 days after discontinuation of tinidazole therapy in patients who are stabilized on their anticoagulant regimen. Patients should be advised to promptly report any signs of bleeding to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, vaginal bleeding, nosebleeds, bleeding of gums from brushing, unusual bleeding or bruising, red or brown urine, or red or black stools.

Tramadol

MONITOR: Coadministration with tramadol may potentiate the hypoprothrombinemic effect of warfarin. There have been rare reports of elevated prothrombin time or INR and bleeding in warfarin patients taking tramadol, and a return to normal anticoagulation following its discontinuation. The exact mechanism is unknown but may be related to reduced activity of CYP450 2D6, which can lead to increased plasma concentrations of tramadol. In a retrospective analysis of 17 cases of suspected interaction between tramadol and warfarin that were reported to the Swedish Adverse Drug Reactions Advisory Committee between 1995 and 2003, INR increases were observed between 1.5 to 60 days following initiation of tramadol therapy (mean 13 days). The recorded INR values ranged from 3.4 to greater than 8.5, and bleeding complications occurred in 6 patients (35%). In cases where follow-up information was available, INR values decreased within 1 to 11 days following tramadol withdrawal. In ten cases where genotypes were available for known polymorphisms of several CYP450 isoenzymes that are associated with reduced or absent enzyme activities, seven patients were found to carry a defective CYP450 2D6 allele. A further patient received concomitant drug treatments that may have resulted in CYP450 2D6 inhibition. Although not statistically significant, this finding suggests a higher prevalence of defective alleles compared to the general Swedish population average of 42.2%.

MANAGEMENT: Given the potential for interaction and the high degree of interpatient variability with respect to warfarin metabolism, patients should be closely monitored during concomitant therapy with tramadol. The INR should be checked frequently and warfarin dosage adjusted accordingly, particularly following initiation, discontinuation or change of dosage of tramadol in patients who are stabilized on their warfarin regimen. The same precaution may be applicable during therapy with other oral anticoagulants, although clinical data are lacking. Patients should be advised to promptly report any signs of bleeding to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, vaginal bleeding, nosebleeds, bleeding of gums from brushing, unusual bleeding or bruising, red or brown urine, or red or black stools.

Metronidazole vaginal, Metronidazole topical

MONITOR: Coadministration with vaginal metronidazole may enhance the hypoprothrombinemic effect of warfarin and other oral anticoagulants. The proposed mechanism is metronidazole inhibition of CYP450 2C9, the isoenzyme responsible for the metabolic clearance of the more active S(-) enantiomer of warfarin. The interaction has been reported with systemically administered metronidazole and warfarin. Significant bleeding and elevation of prothrombin time have been observed. No data are available for vaginally administered metronidazole. Although peak serum levels and systemic exposure are substantially lower than in systemic administration, the possibility of an interaction cannot be excluded.

MANAGEMENT: The potential for interaction with warfarin and other oral anticoagulants should be considered during coadministration with vaginal preparations of metronidazole. The INR should be checked frequently and anticoagulant dosage adjusted accordingly, particularly following initiation and discontinuation of metronidazole therapy in patients who are stabilized on their anticoagulant regimen. Patients should be advised to promptly report any signs of bleeding to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, vaginal bleeding, nosebleeds, bleeding of gums from brushing, unusual bleeding or bruising, red or brown urine, or red or black stools.

Valdecoxib

MONITOR: Coadministration with valdecoxib may increase the plasma concentrations and hypoprothrombinemic effect of warfarin and other oral anticoagulants. The exact mechanism of interaction is unknown, although it is possible that valdecoxib partially inhibits the metabolic clearance of warfarin via CYP450 2C9 and 2C19. In healthy subjects given warfarin with valdecoxib (40 mg twice a day for 7 days), the area under the plasma concentration-time curve (AUC) of the biologically less active R(+) enantiomer and that of the more active S(-) enantiomer of warfarin increased by 12% and 15%, respectively. Although the average International Normalization Ratio (INR) values were only slightly increased, the day-to-day variability in individual INR values was increased. Parecoxib, a prodrug of valdecoxib, may also interact.

MANAGEMENT: Patients should be closely monitored during concomitant therapy with oral anticoagulants and valdecoxib or parecoxib. The INR should be checked frequently and anticoagulant dosage adjusted accordingly, particularly following initiation, discontinuation or change of dosage of valdecoxib in patients who are stabilized on their anticoagulant regimen. Patients should be advised to promptly report any signs of bleeding to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, vaginal bleeding, nosebleeds, bleeding of gums from brushing, unusual bleeding or bruising, red or brown urine, or red or black stools.

Delavirdine

MONITOR: Concomitant use may increase warfarin levels. The probable mechanism is inhibition of CYP450 metabolism by delavirdine. Other anticoagulants may also be affected.

MANAGEMENT: Frequent monitoring of the INR and patient response and tolerance is strongly recommended. Patients should be advised to promptly report any signs of bleeding to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, nosebleeds, bleeding of gums from brushing, red or brown urine, or red or black stools. Dosage should be adjusted as appropriate.

Prednisolone, Dexamethasone, Hydrocortisone, Methylprednisolone, Prednisone, Cortisone, Triamcinolone (oral and injectable), Triamcinolone inhalation, Betamethasone, Fludrocortisone, Triamcinolone acetonide injection

MONITOR: Corticosteroids and adrenocorticotropic agents may alter the pharmacologic effects of oral anticoagulants. Both increased anticoagulant dosage requirements as well as bleeding and increased anticoagulant sensitivity have been reported during concomitant corticosteroid therapy. Proposed mechanisms include blood hypercoagulability or diminished vascular integrity induced by corticosteroids. Conversely, the risk of gastrointestinal bleeding associated with the ulcerogenic effect of corticosteroids may be potentiated by use of oral anticoagulants.

MANAGEMENT: Patients should be monitored for clinical and laboratory evidence of increased or decreased anticoagulant response during concurrent treatment with corticosteroids or adrenocorticotropic agents. They should be advised to promptly report any signs and symptoms of bleeding (e.g., pain, swelling, headache, dizziness, weakness, vaginal bleeding, nosebleeds, bleeding of gums from brushing, unusual bleeding or bruising, red or brown urine, or red or black stools) or of blood clots (e.g., chest pain, shortness of breath, sudden loss of vision, or pain, redness or swelling in an extremity).

Cyclophosphamide (oral/injection)

MONITOR: Cyclophosphamide may increase or decrease the anticoagulant activity of warfarin by an unknown mechanism. Since, in some cases, the anticoagulant effect of warfarin has markedly increased after cyclophosphamide has been discontinued, some experts have speculated that cyclophosphamide stimulates the hepatic metabolism of warfarin. According to this suggestion, discontinuing cyclophosphamide could decrease metabolism, increase plasma levels, and increase the anticoagulant effect of warfarin. Other oral anticoagulants may interact with cyclophosphamide in a similar manner.

MANAGEMENT: Careful monitoring of the INR during coadministration and after discontinuation is recommended. Patients should be advised to promptly report any signs of bleeding (e.g., pain, swelling, headache, dizziness, weakness, vaginal bleeding, nosebleeds, bleeding of gums from brushing, unusual bleeding or bruising, red or brown urine, or red or black stools) or blood clots (e.g. chest pain, shortness of breath, sudden loss of vision, or pain, redness or swelling in an extremity).

Dextrothyroxine

MONITOR: Dextrothyroxine may enhance the hypoprothrombinemic response to oral anticoagulants. The mechanism is unknown. If dextrothyroxine and an oral anticoagulant must be used together, the patient should be closely monitored for excessive anticoagulation, and the oral anticoagulant dosage should be reduced as needed.

MANAGEMENT: Patients should be advised to notify their physician if they experience any signs of excessive anticoagulation, such as unusual or prolonged bleeding, bruising, vomiting, change in stool or urine color, headache, dizziness, or weakness.

Disulfiram

MONITOR: Disulfiram increases the anticoagulant effect of warfarin in most patients. The mechanism is unknown. A similar effect probably occurs with other oral anticoagulants.

MANAGEMENT: Close monitoring for clinical and laboratory evidence of altered anticoagulant effect is indicated when these drugs are used together. Patients should be advised to notify their physicians if they experience any signs of excessive anticoagulation, such as unusual or prolonged bleeding, bruising, vomiting, change in stool or urine color, headache, dizziness, or weakness.

Efavirenz

MONITOR: Efavirenz may increase or decrease plasma concentrations and therapeutic effects of warfarin. The mechanism is alteration of CYP450 hepatic metabolism of warfarin R- and S-isomers. Other anticoagulants may be affected, as well.

MANAGEMENT: Until more information is available, INR monitoring is recommended and patients should be closely observed for clinical response. Patients should be advised to notify their physician if they experience any signs of excessive anticoagulation (e.g., unusual or prolonged bleeding, bruising, vomiting, change in stool or urine color, headache, dizziness, or weakness).

Epoprostenol

MONITOR: Epoprostenol in combination with antiplatelet or anticoagulant agents may increase the risk of bleeding. However, in clinical trials patients received anticoagulants without increased bleeding. Warfarin (or another oral anticoagulant) is suggested to help prevent clotting in the epoprostenol central line. The mechanism of action is most likely due to platelet inhibition.

MANAGEMENT: If this patient receives antiplatelet agents or anticoagulants in combination with epoprostenol, consider monitoring more closely for signs of bleeding. Patients should be advised to promptly report any signs of bleeding to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, vaginal bleeding, nosebleeds, bleeding of gums from brushing, red or brown urine, red or black stools, or other unusual bleeding or bruising.

Erlotinib

MONITOR: Erlotinib may potentiate the hypoprothrombinemia effect of warfarin and other oral anticoagulants. The mechanism of the interaction is unknown. Altered coagulation parameters and/or gastrointestinal bleeding have been reported during clinical trials.

MANAGEMENT: Patients should be monitored during concomitant therapy. The INR or prothrombin time should be checked regularly for changes. Patients should be advised to promptly report any signs of bleeding to their doctor, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, vaginal bleeding, nosebleeds, bleeding of gums from brushing, red or brown urine, or red or black stools.

Felbamate

MONITOR: Felbamate has been reported to increase the anticoagulant effect of warfarin. The mechanism may be inhibition of warfarin metabolism via the CYP450 system in the liver. According to a single case report, the addition of felbamate necessitated a reduction of approximately 50% in warfarin dosage to maintain the target INR. However, causal effect has not been definitively established. Other oral anticoagulants could interact with felbamate in a similar manner.

MANAGEMENT: Close monitoring of the prothrombin time or the INR is recommended if felbamate and an oral anticoagulant must be used together. The anticoagulant dose may need to be decreased or temporarily stopped. Patients should be advised to notify their physician if they experience any signs of excessive anticoagulation, such as unusual or prolonged bleeding, bruising, vomiting, change in stool or urine color, headache, dizziness, or weakness.

Flutamide (oral)

MONITOR: Flutamide may enhance the hypoprothrombinemic response to warfarin by an unknown mechanism. In patients stabilized on warfarin, elevations in prothrombin time have been reported with the addition of flutamide to the drug regimen.

MANAGEMENT: Patients should be monitored for altered anticoagulation and advised to report any signs of bleeding to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, nosebleeds, bleeding of gums from brushing, bruising, red or brown urine, or red or black stools.

Gemcitabine

MONITOR: Gemcitabine may alter the pharmacodynamic response to warfarin resulting in an increased INR and risk of bleeding. The mechanism has not been fully determined. It has been suggested that warfarin metabolism or the synthesis of clotting factors may be decreased due to reversible hepatotoxicity induced by gemcitabine. Other anticoagulants may also be affected, although clinical data are lacking.

MANAGEMENT: Increased frequency of clinical monitoring of INR and liver function tests is recommended. Patients should be advised to promptly report any signs of unusual bleeding to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, bruising, increased menstrual flow, vaginal bleeding, nosebleeds, bleeding of gums from brushing, red or brown urine, or red or black stools.

Ginger

MONITOR: Ginger may potentiate the effects of anticoagulants, platelet inhibitors and thrombolytic agents, possibly increasing the risk of bleeding. Limited data suggest that ginger may decrease platelet aggregation via the inhibition of thromboxane synthetase, although some studies have found no effect on platelet function or thromboxane production or activity. Nevertheless, the interaction was suspected in a 76-year-old patient stabilized on coumarin therapy who developed epistaxis following use of ginger products (pieces of dried ginger, tea from ginger powder) for several weeks. Her INR was greater than 10 (target INR 2.0 to 3.0) and partial thromboplastin time (PTT) was 84.4 seconds (normal less than 35 seconds) upon hospital admission. INR and PTT values normalized after ginger intake was stopped and vitamin K given. In contrast, an investigative study found no significant effect of ginger pretreatment for 7 days on clotting status or the pharmacokinetics or pharmacodynamics of a single 25 mg dose of warfarin in 12 healthy volunteers.

MANAGEMENT: Patients should consult a healthcare provider before taking any herbal or alternative medicine. In patients who have used ginger and ginger supplements extensively prior to receiving anticoagulation, antiplatelet or thrombolytic therapy, the potential for an interaction should be considered. Close clinical and laboratory observation for hematologic complications is recommended. Patients should be advised to promptly report any signs of bleeding to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, vaginal bleeding, nosebleeds, bleeding of gums from brushing, unusual bleeding or bruising, red or brown urine, or red or black stools.

Griseofulvin

MONITOR: Griseofulvin has been reported to increase warfarin requirements in some previously stabilized patients. The mechanism may be induction of the hepatic metabolism of warfarin. Similar effects may occur with other oral anticoagulants.

MANAGEMENT: Patients on oral anticoagulants should be monitored for alterations in INR whenever griseofulvin therapy is added, discontinued, or changed.

Interferon alfa-2a, Interferon alfa-2b, Interferon alfa-n3 (injectable), Interferon gamma-1b, Interferon beta-1b, Interferon beta-1a, Interferon alfacon-1, Peginterferon alfa-2b, Peginterferon alfa-2a

MONITOR: Interferons may increase the plasma concentrations and pharmacologic effects of warfarin and other oral anticoagulants. At least one case of this interaction has been reported in the literature. The mechanism is unknown but may be related to interferon-induced suppression of hepatic CYP450 enzymes. Physiologic events and drugs that induce interferon synthesis (e.g., viral infection, influenza vaccination) have also been reported to decrease metabolism and potentiate the effects of certain drugs including warfarin.

MANAGEMENT: Caution is advised if oral anticoagulants are administered with an interferon. The INR should be monitored closely and anticoagulant dosage adjusted accordingly, particularly following initiation or discontinuation of interferon therapy in patients who are stabilized on their anticoagulant regimen. Patients should be advised to promptly report any signs of bleeding to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, vaginal bleeding, nosebleeds, bleeding of gums from brushing, unusual bleeding or bruising, red or brown urine, or red or black stools.

Venlafaxine (oral)

MONITOR: Isolated case reports have suggested that venlafaxine may potentiate the anticoagulant effects of warfarin. Increased prothrombin times and INRs and bleeding have been reported during concurrent use. Causality has not been established and the mechanism is unknown.

MANAGEMENT: The clinician may consider closer INR and clinical monitoring in patients starting or stopping venlafaxine therapy while also receiving oral anticoagulants. The dose of the oral anticoagulant may need to be adjusted. Patients should be advised to promptly report any signs of bleeding to their doctor, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, nosebleeds, bleeding of gums from brushing, red or brown urine, or red or black stools.

Isoniazid

MONITOR: Isoniazid may increase the anticoagulant effect of warfarin, perhaps by inhibition of warfarin metabolism. This effect has been reported with an isoniazid dose of 600 mg per day. Other oral anticoagulants may interact with isoniazid in a similar fashion. The proposed mechanism is inhibition of CYP450 2C9 hepatic metabolism by isoniazid.

MANAGEMENT: The patient's INR should be monitored, and the anticoagulant dosage should be adjusted if necessary, whenever isoniazid is initiated, discontinued or the dosage is changed. Patients should be advised to promptly report any signs of bleeding to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, vaginal bleeding, nosebleeds, bleeding of gums from brushing, unusual bruising, red or brown urine, or red or black stools.

Itraconazole

MONITOR: Itraconazole may enhance the hypoprothrombinemic effect of warfarin. The risk of bruising and hemorrhage may be increased. The mechanism may be related to inhibition of CYP450 hepatic metabolism. A similar effect may occur with other oral anticoagulants.

MANAGEMENT: During coadministration, frequent monitoring of the INR is recommended, and a reduction in oral anticoagulant dosage may be necessary. Patients should be advised to promptly report any signs of bleeding to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, vaginal bleeding, nosebleeds, bleeding of gums from brushing, unusual bleeding or bruising, red or brown urine, or red or black stools.

Ketoconazole

MONITOR: Ketoconazole may increase the hypoprothrombinemic effect of oral anticoagulants. The probable mechanism is inhibition of CYP450 hepatic metabolism by ketoconazole.

MANAGEMENT: Clinical and laboratory monitoring for enhanced hypoprothrombinemic effects is recommended. Patients should be advised to promptly report any signs of bleeding to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, nosebleeds, bleeding of gums from brushing, unusual bruising, red or brown urine, or red or black stools.

Omega-3 polyunsaturated fatty acids

MONITOR: Large doses of omega-3 polyunsaturated fatty acids (e.g., fish oil) may potentiate the hypoprothrombinemic effect of oral anticoagulants. The exact mechanism of interaction is unknown. Omega-3 polyunsaturated fatty acids may possess mild antiplatelet and hypocoagulant activities. In some studies, these substances have been shown to reduce thrombin generation and plasma levels of fibrinogen, prothrombin, and coagulation factors V, VII, and X. The interaction was suspected in a case report of a 67-year-old woman treated with warfarin for 1.5 years who exhibited an increase in INR from 2.8 the previous month to 4.3 approximately one week after doubling her fish oil dosage from 1000 to 2000 mg/day. Prior to the increase, her INR had been stable and therapeutic for 5 months while on warfarin 1.5 mg/day. The patient was advised to reduce her fish oil consumption to 1000 mg/day, while her warfarin dose was withheld for one day and then reduced to 1 mg alternating with 1.5 mg per day. Eight days later, her INR was subtherapeutic at 1.6, so the warfarin dosage was increased back to 1.5 mg/day. The patient's INR has been within therapeutic range since that time.

MANAGEMENT: In general, patients should consult a healthcare provider before taking any herbal or nutritional supplements. Patients receiving oral anticoagulant therapy in combination with omega-3 polyunsaturated fatty acids should be advised of the potential for increased hypoprothrombinemic effect. The INR should be checked frequently and anticoagulant dosage adjusted accordingly, particularly following initiation, discontinuation, or change of dosage of the fatty acids. Patients should be advised to promptly report any signs of bleeding to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, vaginal bleeding, nosebleeds, bleeding of gums from brushing, unusual bleeding or bruising, red or brown urine, or red or black stools.

Nevirapine

MONITOR: Limited clinical data suggest that nevirapine may reduce the hypoprothrombinemic effect of warfarin. There have been isolated case reports of patients stabilized on warfarin whose INR decreased following the addition of nevirapine and patients who were refractory to unusually high dosages of warfarin during treatment with antiretroviral regimens containing nevirapine. Only after discontinuation of nevirapine was coagulation maintained using standard dosages of warfarin. The exact mechanism by which this interaction may occur is unknown. Nevirapine has not been reported to induce CYP450 2C9 or 1A2, the predominant enzymatic pathways responsible for the metabolism of warfarin.

MANAGEMENT: Until further data are available, patients should be closely monitored during concomitant therapy. The INR should be checked frequently and warfarin dosage adjusted accordingly, particularly following initiation or discontinuation of nevirapine in patients who are stabilized on their warfarin regimen. Although data are lacking for other oral anticoagulants, the potential for a similar interaction should be considered.

Azathioprine, Mercaptopurine

MONITOR: Limited clinical data suggest that the purine analog antimetabolites, azathioprine and mercaptopurine, may reduce the hypoprothrombinemic response to warfarin and other oral anticoagulants. The mechanism has not been fully established but animal studies indicate that enhanced activation or synthesis of prothrombin may be involved. There have been case reports of patients stabilized on oral anticoagulant therapy whose prothrombin time or INR decreased following addition of the antimetabolites and returned to normal only after their discontinuation or an increase in anticoagulant dosage. Bleeding has been reported after discontinuation of the antimetabolite.

MANAGEMENT: Given the potential for interaction and the narrow therapeutic index of oral anticoagulants, patients should be closely monitored during concomitant therapy with azathioprine or mercaptopurine. The INR should be checked frequently and anticoagulant dosage adjusted accordingly, particularly following initiation or discontinuation of antimetabolite therapy in patients who are stabilized on their anticoagulant regimen. Patients should be advised to promptly report any signs of blood clots (e.g. chest pain, shortness of breath, sudden loss of vision, or pain, redness or swelling in an extremity).

Tolterodine

MONITOR: Limited clinical data suggest that tolterodine may potentiate the hypoprothrombinemic effect of warfarin in some patients. There have been case reports of patients stabilized on warfarin whose INR increased following the addition of tolterodine and returned to normal after discontinuation of tolterodine, temporary withholding of warfarin doses, or reduction of warfarin dosage. The mechanism is unknown. In 20 healthy volunteers, pretreatment with tolterodine (2 mg orally twice a day for 7 days) relative to placebo had no significant effect on the pharmacokinetics of R(+) and S(-) warfarin or prothrombin times and factor VII activity following a single 25 mg dose of warfarin administered on study day 4. Warfarin also had no effect on the pharmacokinetics of tolterodine or its pharmacologically active 5-hydroxymethyl metabolite.

MANAGEMENT: Given the potential for interaction and the high degree of interpatient variability with respect to warfarin metabolism, patients should be closely monitored during concomitant therapy with tolterodine. The INR should be checked frequently and warfarin dosage adjusted accordingly, particularly following initiation, discontinuation or change of dosage of tolterodine in patients who are stabilized on their warfarin regimen. The same precaution may be applicable during therapy with other oral anticoagulants, although clinical data are lacking. Patients should be advised to promptly report any signs of bleeding to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, vaginal bleeding, nosebleeds, bleeding of gums from brushing, unusual bleeding or bruising, red or brown urine, or red or black stools.

Ifosfamide, Mesna

MONITOR: Limited data (from three patients) reveal enhanced anticoagulant activity (significantly increased INR) during chemotherapy with ifosfamide/mesna in patients previously stable on warfarin. The mechanism is not known, but is suspected to be displacement of warfarin from protein-binding sites by ifosfamide metabolites.

MANAGEMENT: Careful and frequent monitoring of INR during coadministration is recommended. Also, close monitoring is recommended when chemotherapeutic agents are discontinued, because anticoagulant effect may be decreased and the risk of thromboembolism increased. It may also be advisable to monitor patients on other anticoagulants.

Isotretinoin (oral)

MONITOR: Limited data suggest that isotretinoin may reduce the hypoprothrombinemic effect of warfarin. The mechanism of interaction is unknown. A case report describes a 61-year-old patient stabilized on warfarin 2.5 mg/day for 2 to 3 years who experienced a decreased INR following initiation of cefpodoxime proxetil (200 mg orally twice a day) and isotretinoin (30 mg orally daily) for a facial eruption due to Klebsiella infection. An increase in the warfarin dosage to 3.75 mg/day was necessary to maintain his INR within therapeutic range. The INR remained stable at the new dosage after completion of cefpodoxime therapy but increased when isotretinoin was discontinued after 40 days. Warfarin dosage was subsequently reduced to the previous level.

MANAGEMENT: Until further information is available, INR should be monitored more closely whenever isotretinoin is added to or withdrawn from therapy, and the warfarin dosage adjusted as necessary. The same precaution may be applicable during therapy with other oral anticoagulants, although clinical data are lacking.

Neomycin

MONITOR: Limited data suggest that the effect of oral anticoagulants may be increased in patients also receiving oral neomycin. Increases in INR are possible. The exact mechanism is unknown but may involve depletion of certain clotting factors due to suppression of vitamin K-producing intestinal flora and inhibition of vitamin K absorption. Clinical data have been conflicting. Significant changes are not expected in patients with normal dietary intake of vitamin K.

MANAGEMENT: The clinician may consider closer INR and clinical monitoring in patients starting or stopping neomycin therapy while also receiving oral anticoagulants. The dose of the oral anticoagulant may need to be adjusted. Patients should be advised to promptly report any signs of bleeding to their doctor, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, nosebleeds, bleeding of gums from brushing, red or brown urine, or red or black stools.

Lovastatin, Red yeast rice

MONITOR: Lovastatin may enhance the hypoprothrombinemic effects of warfarin. Clinically significant alterations in anticoagulation have been reported. The mechanism is unknown.

MANAGEMENT: Caution is recommended if any HMG-CoA reductase inhibitor is used concomitantly with any oral anticoagulant. Close clinical and laboratory monitoring for altered anticoagulant effect is recommended. Patients should be advised to promptly report any signs of bleeding to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, vaginal bleeding, nosebleeds, bleeding of gums from brushing, unusual bleeding or bruising, red or brown urine, or red or black stools. Preliminary studies have shown that pravastatin and cerivastatin do not affect the anticoagulant response to warfarin.

Mitotane

MONITOR: Mitotane may decrease the effectiveness of orally administered anticoagulants. The mechanism is induction of hepatic microsomal enzymatic clearance. Data are available for warfarin.

MANAGEMENT: Close monitoring for clinical and laboratory evidence of altered anticoagulant efficacy is recommended. The dosage requirements for orally administered anticoagulants are expected to increase after mitotane is added. The clinician should be also be aware that the risk of bleeding from over-anticoagulation is increased when mitotane is withheld.

Chlorpropamide, Glipizide, Glyburide, Tolbutamide

MONITOR: Oral sulfonylureas may enhance or reduce the hypoprothrombinemic response to oral anticoagulants. The mechanism may be related to displacement from plasma protein binding sites. In addition, coumarin anticoagulants may cause an increase in blood levels of hypoglycemic agents, possibly by inhibiting their hepatic metabolism. Clinical data have been highly variable.

MANAGEMENT: The patient should be monitored for altered anticoagulation (PT/INR) and altered glycemic effect when either of these drugs is added to or removed from a patient's regimen. Patients should be advised to regularly monitor their blood sugar, counseled on how to recognize and treat hypoglycemia (e.g., headache, dizziness, drowsiness, nausea, tremor, hunger, weakness, or palpitations), and to promptly report any signs of bleeding (pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, vaginal bleeding, nosebleeds, bleeding of gums from brushing, red or brown urine, or red or black stools) to their physician.

Pegaspargase

MONITOR: Pegaspargase may lead to imbalances in coagulation factors predisposing the patient to bleeding or thrombosis.

MANAGEMENT: The clinician should exercise caution when coadministering any nonsteroidal anti-inflammatory agent or anticoagulant therapy with pegaspargase. Patients should be closely monitored for signs of bleeding during coadministration.

Amoxicillin, Carbenicillin, Oxacillin, Penicillin V, Bacampicillin, Cloxacillin, Penicillin G potassium injection, Ticarcillin

MONITOR: Penicillins may occasionally potentiate the risk of bleeding in patients treated with oral anticoagulants. The exact mechanism of interaction is unknown but may involve penicillin inhibition of platelet aggregation. In one study, defective platelet aggregation occurred with predictability in patients receiving penicillin G 24 million units/day, ampicillin 300 mg/kg/day, and methicillin 300 mg/kg/day. Other penicillins such as nafcillin, piperacillin, and ticarcillin have also been found to affect platelet function, and benzylpenicillin and carbenicillin have been reported to increase bleeding times and cause bleeding in the absence of an anticoagulant. There have been case reports describing increases in prothrombin time and INR as well as spontaneous bruising and bleeding in anticoagulated patients following initiation or completion of penicillin therapy. Although most cases have involved large, intravenous doses of some penicillins (e.g., carbenicillin, penicillin G, ticarcillin), the interaction has also been reported with regular, oral doses of amoxicillin and amoxicillin-clavulanate. In fact, a case-control study found amoxicillin-clavulanate to be one of only two medications to significantly increase the risk of overanticoagulation in previously stable outpatients treated with phenprocoumon or acenocoumarol. In that study, 300 outpatients at a Netherlands anticoagulant clinic who presented with an INR value greater than or equal to 6.0 (median value 6.8) were compared with 302 randomly selected matched controls with INR values within the target range (median value 3.2), and changes in the use of 87 potentially interacting drugs or drug classes in the four weeks prior to the index day were identified and analyzed. A course of amoxicillin-clavulanate increased the risk of overanticoagulation even after adjustment for potential confounding factors, particularly in patients treated with acenocoumarol. A follow-up study focusing on antibiotic use in outpatients treated with phenprocoumon or acenocoumarol at a different Netherlands anticoagulant clinic also identified amoxicillin use as a risk factor for overanticoagulation, with the relative risk most strongly increased four days or more after start of the antibiotic.

MANAGEMENT: Caution is recommended if a penicillin is prescribed during oral anticoagulant therapy, especially in the elderly and patients with uremia or hepatic impairment. The INR should be checked frequently and anticoagulant dosage adjusted accordingly, particularly following initiation or discontinuation of penicillin therapy in patients who are stabilized on their anticoagulant regimen. Patients should be advised to promptly report any signs of bleeding to their doctor, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, vaginal bleeding, nosebleeds, bleeding of gums from brushing, unusual bleeding or bruising, red or brown urine, or red or black stools.

Vorinostat

MONITOR: Prolongation of prothrombin time and INR were observed during clinical trials in patients receiving vorinostat concomitantly with coumarin-derivative anticoagulants. The mechanism has not been described.

MANAGEMENT: Patients treated with a coumarin-derivative anticoagulant should be closely monitored during concomitant therapy with vorinostat. The INR should be checked frequently and anticoagulant dosage adjusted accordingly, particularly following initiation, discontinuation or change of dosage of vorinostat in patients who are stabilized on their anticoagulant regimen. Patients should be advised to promptly report any signs of bleeding to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, vaginal bleeding, nosebleeds, bleeding of gums from brushing, unusual bleeding or bruising, red or brown urine, or red or black stools.

Propafenone

MONITOR: Propafenone may increase anticoagulant serum concentrations and effects. The mechanism is inhibition of CYP450 2C9 metabolism. In healthy subjects, warfarin plasma concentrations increased 39% and prothrombin time increased 25% when propafenone was taken concomitantly.

MANAGEMENT: Management consists of monitoring PT or INR until a stable and safe anticoagulant dosage is found. Patients should be advised to promptly report any signs of bleeding to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, vaginal bleeding, nosebleeds, bleeding of gums from brushing, unusual bleeding or bruising, red or brown urine, or red or black stools.

Quinine

MONITOR: Quinine derivatives may induce hypoprothrombinemia and, when given concomitantly with oral anticoagulants, may put patients at risk for significant bleeding.

MANAGEMENT: INR should be monitored, and the dosage of oral anticoagulant adjusted as needed. Patients should be advised to notify their physicians if they experience any signs of excessive anticoagulation, such as unusual or prolonged bleeding, bruising, vomiting, change in stool or urine color, headache, dizziness, or weakness.

Influenza virus vaccine (injectable), Influenza virus vaccine (nasal)

MONITOR: Rare case reports have suggested that the administration of influenza vaccine to patients taking warfarin may result in increased INR and/or bleeding. The mechanism is unknown and causality was not clearly established. Subsequent studies have reported that influenza vaccine did not affect coagulation parameters or warfarin metabolism.

MANAGEMENT: It may be advisable to monitor anticoagulated patients for several days after administering influenza vaccine.

Ribavirin (oral)

MONITOR: Ribavirin may decrease the hypoprothrombinemic effect of warfarin. The mechanism of interaction is unknown. In an isolated case report, a 61-year-old man previously stabilized on 45 mg/week of warfarin required an approximately 40% increase in the anticoagulant dosage over a period of one month following addition of interferon alfa-2b and ribavirin therapy for hepatitis C. After discontinuation of interferon-ribavirin therapy a year later, the anticoagulant dosage decreased to 47.5 mg/week over a three-week period. A subsequent rechallenge with ribavirin appeared to confirm the interaction. The effect of ribavirin on other oral anticoagulants is unknown.

MANAGEMENT: Patients treated with warfarin should be closely monitored during concomitant therapy with ribavirin. The INR should be checked frequently (e.g., weekly) during the 4 weeks following initiation or discontinuation of ribavirin therapy in patients who are stabilized on their warfarin regimen, and the warfarin dosage adjusted accordingly. The same precaution may be applicable during therapy with other oral anticoagulants, although clinical data are lacking. Patients should be advised to promptly report any signs of bleeding to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, vaginal bleeding, nosebleeds, bleeding of gums from brushing, unusual bleeding or bruising, red or brown urine, or red or black stools.

Rofecoxib (oral)

MONITOR: Rofecoxib may potentiate the hypoprothrombinemic effect of warfarin and other oral anticoagulants as determined by the INR. In single- and multiple-dose studies in which healthy subjects received warfarin with rofecoxib or placebo, rofecoxib increased the INR by 5% to 11%, depending on the dosage used (12.5 mg to 50 mg). The higher INRs reported may be due to increased plasma concentrations of the biologically less active R(+) enantiomer of warfarin in the presence of rofecoxib. No significant changes were reported for the S(-) enantiomer. Although these results are not considered clinically significant, bleeding events associated with increases in prothrombin time have been reported in postmarketing use of the combination, particularly in the elderly.

MANAGEMENT: Patients should be closely monitored during concomitant therapy with oral anticoagulants and rofecoxib. The INR should be checked frequently and anticoagulant dosage adjusted accordingly, particularly following initiation, discontinuation or change of dosage of rofecoxib in patients who are stabilized on their anticoagulant regimen. Patients should be advised to promptly report any signs of unusual bleeding or bruising to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, vaginal bleeding, nosebleeds, bleeding of gums from brushing, red or brown urine, or red or black stools.

Ropinirole (oral)

MONITOR: Ropinirole may potentiate the hypoprothrombinemic effect of warfarin. The mechanism is unknown but may involve ropinirole displacement of warfarin from plasma protein binding sites. Alternatively, ropinirole may competitively inhibit the metabolism of the biologically less active R(+) enantiomer of warfarin via CYP450 1A2. In one case report, a 63-year-old patient who had been stabilized on warfarin 4 mg once daily for more than a year developed an increased INR nine days after initiation of ropinirole therapy for progression of Parkinson's disease. Other possible causes such as change in diet, alcohol use, or addition of over-the-counter or herbal products were ruled out. Warfarin was withheld for 4 days and restarted at 2 mg/day, then increased to 3 mg/day, whereupon the INR became therapeutic. However, the patient eventually required an increase back to 4 mg/day after ropinirole was discontinued due to gastrointestinal adverse effects.

MANAGEMENT: Given the potential for interaction and the high degree of interpatient variability with respect to warfarin metabolism, patients should be closely monitored during concomitant therapy with ropinirole. The INR should be checked frequently and warfarin dosage adjusted accordingly, particularly following initiation, discontinuation or change of dosage of ropinirole in patients who are stabilized on their warfarin regimen. Patients should be advised to promptly report any signs of bleeding to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, vaginal bleeding, nosebleeds, bleeding of gums from brushing, unusual bleeding or bruising, red or brown urine, or red or black stools.

Rosuvastatin

MONITOR: Rosuvastatin may enhance the hypoprothrombinemic effect of warfarin and other oral anticoagulants. According to the product labeling, administration of rosuvastatin to patients stabilized on warfarin resulted in clinically significant rises in the INR (greater than 4, baseline 2 to 3). The mechanism of interaction is unknown. Coadministration of rosuvastatin 40 mg and warfarin 20 mg did not alter the plasma concentrations of warfarin.

MANAGEMENT: Caution is recommended if rosuvastatin is used concomitantly with oral anticoagulants. The INR should be checked frequently and anticoagulant dosage adjusted accordingly, particularly following initiation, discontinuation or change of dosage of rosuvastatin in patients who are stabilized on their anticoagulant regimen. Patients should be advised to promptly report any signs of bleeding to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, vaginal bleeding, nosebleeds, bleeding of gums from brushing, unusual bleeding or bruising, red or brown urine, or red or black stools.

Clomipramine, Sertraline, Paroxetine, Sibutramine (oral), Citalopram (oral), Escitalopram

MONITOR: Serotonin reuptake inhibitors (SRIs) may potentiate the risk of bleeding in patients treated with agents that affect hemostasis such as anticoagulants, platelet inhibitors, thrombin inhibitors, thrombolytic agents, or agents that commonly cause thrombocytopenia. The tricyclic antidepressant, clomipramine, is also a strong SRI and may interact similarly. Serotonin release by platelets plays an important role in hemostasis, thus SRIs may alter platelet function and induce bleeding. Published case reports have documented the occurrence of bleeding episodes in patients treated with psychotropic agents that interfere with serotonin reuptake. Additional epidemiological studies have confirmed the association between use of these agents and the occurrence of upper gastrointestinal bleeding, and concurrent use of NSAIDs or aspirin was found to potentiate the risk. Preliminary data also suggest that there may be a pharmacodynamic interaction between SSRIs and oral anticoagulants that can cause an increased bleeding diathesis. Concomitant administration of paroxetine and warfarin, specifically, has been associated with an increased frequency of bleeding without apparent changes in the disposition of either drug or changes in the prothrombin time. Bleeding has also been reported with fluoxetine and warfarin, while citalopram and sertraline have been reported to prolong the prothrombin time of patients taking warfarin by about 5% to 8%.

MANAGEMENT: Caution is advised if SRIs or clomipramine are used in combination with other drugs that affect hemostasis. Close clinical and laboratory observation for hematologic complications is recommended. Patients should be advised to promptly report any signs of bleeding to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, vaginal bleeding, nosebleeds, bleeding of gums from brushing, unusual bleeding or bruising, red or brown urine, or red or black stools.

Fluoxetine

MONITOR: Several case reports, including one report of cerebral hemorrhage, suggest that fluoxetine may increase the anticoagulant effects of warfarin. Postulated mechanisms include displacement of warfarin from plasma protein-binding sites, inhibition of the CYP450 2C9 hepatic metabolism of warfarin by fluoxetine, and direct inhibitory effect of fluoxetine on hemostasis. Animal studies have also suggested that fluoxetine (given in very high doses) may increase prothrombin time in animals treated with warfarin. Two small studies of human volunteers, however, found no alteration in the prothrombin time in individuals treated with fluoxetine and warfarin. It is not known if other oral anticoagulants interact with fluoxetine in a similar manner.

MANAGEMENT: Close clinical and laboratory monitoring for altered anticoagulant effect is recommended if oral anticoagulants and fluoxetine must be used together. Patients should be advised to promptly report any signs of bleeding to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, vaginal bleeding, nosebleeds, bleeding of gums from brushing, unusual bleeding or bruising, red or brown urine, or red or black stools.

Levamisole

MONITOR: Significantly increased INR and prothrombin times have been reported in patients receiving concomitant warfarin, fluorouracil, and levamisole. Discontinuation of chemotherapy resulted in subtherapeutic anticoagulation. Reinstitution of chemotherapy resulted in a clinically significant elevation in the INR. The mechanism of this interaction is unknown and causality has not been clearly established. Other oral anticoagulants may interact with chemotherapy in a similar manner.

MANAGEMENT: If concomitant therapy is necessary, close monitoring of the INR or PT is recommended. Anticoagulant dose adjustments may be necessary. Patients should be advised to promptly report any signs of bleeding to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, nosebleeds, bleeding of gums from brushing, red or brown urine, red or black stools, or unusual bleeding or bruising.

Phenytoin (oral), Mephenytoin, Ethotoin

MONITOR: Some oral anticoagulants have been reported to increase phenytoin half-life and serum concentrations, possibly by inhibiting CYP450 metabolism. The addition of phenytoin to warfarin therapy has initially increased prothrombin time, but thereafter reduced anticoagulant effects. The mechanism is not known, but may be due to initial CYP450 metabolic inhibition, followed by isoenzyme induction. Data have been conflicting.

MANAGEMENT: Phenytoin serum concentrations and INR/PT should be monitored in patients receiving this combination and whenever the dosage is changed or discontinued. Patients should be advised to promptly report any signs of bleeding (e.g., pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, nosebleeds, bleeding of gums from brushing, red or brown urine, or red or black stools) or hydantoin toxicity (e.g., drowsiness, visual disturbances, change in mental status, seizures, nausea, or ataxia) to their physician.

Nortriptyline, Desipramine, Amitriptyline, Doxepin, Imipramine, Trimipramine, Amoxapine, Protriptyline

MONITOR: Some reports suggest that tricyclic antidepressants may increase or decrease the anticoagulant effect of warfarin by an unknown mechanism. Similar effects may occur with other oral anticoagulants; however, data have been conflicting.

MANAGEMENT: Until more information is available, it may be advisable to monitor the International Normalized Ratio (INR) or prothrombin time and clinical response if these agents are administered concomitantly. Patients should be advised to promptly report possible signs of bleeding (e.g., pain, swelling, headache, dizziness, weakness, nosebleeds, bleeding of gums from brushing, unusual bruising, red or brown urine, or red or black stools) or clots (e.g. chest pain, shortness of breath, sudden loss of vision, or pain, redness or swelling in an extremity).

Exenatide

MONITOR: Spontaneous reports of increased INR, sometimes accompanied by bleeding, have been associated with concomitant use of warfarin and exenatide. The mechanism of interaction is unknown. In a controlled study in healthy volunteers, the time to reach peak plasma concentration (Tmax) of warfarin was delayed by about two hours when warfarin was administered 30 minutes after exenatide. However, no clinically relevant effects on peak plasma concentration (Cmax) or systemic exposure (AUC) were observed.

MANAGEMENT: Until more data are available, caution is advised if exenatide is prescribed to patients treated with warfarin. The INR should be checked frequently and warfarin dosage adjusted accordingly, particularly following initiation or discontinuation of exenatide in patients who are stabilized on their warfarin regimen. The same precaution may be applicable during therapy with other oral anticoagulants, although clinical data are lacking. Patients should be advised to promptly report any signs of bleeding to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, vaginal bleeding, nosebleeds, bleeding of gums from brushing, unusual bleeding or bruising, red or brown urine, or red or black stools.

Suprofen ophthalmic, Diclofenac ophthalmic, Ketorolac ophthalmic, Flurbiprofen ophthalmic, Diclofenac topical, Bromfenac (ophthalmic), Nepafenac ophthalmic

MONITOR: Systemically and topically administered nonsteroidal anti-inflammatory drugs (NSAIDs) may potentiate the risk of bleeding in patients treated with anticoagulants and other drugs that affect hemostasis such as platelet inhibitors, thrombin inhibitors, thrombolytic agents, or agents that commonly cause thrombocytopenia. The pharmacologic effects of NSAIDs that contribute to this interaction include prolongation of prothrombin time and inhibition of platelet adhesion and aggregation.

MANAGEMENT: Caution is advised if NSAIDs are used in combination with other drugs that affect hemostasis. Close clinical and laboratory observation for hematologic complications is recommended. Patients should be advised to promptly report any signs of bleeding to their doctor, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, vaginal bleeding, nosebleeds, bleeding of gums from brushing, unusual bleeding or bruising, red or brown urine, or red or black stools.

Doxycycline, Tetracycline, Minocycline, Demeclocycline

MONITOR: Tetracycline antibiotics may enhance the action of oral anticoagulants by inhibiting the production of vitamin K producing bacteria in the GI tract.

MANAGEMENT: INR should be monitored whenever a tetracycline is started or stopped, and the patient should be observed for signs of bleeding. An adjustment of the anticoagulant dose may be necessary. Patients should be advised to notify their physicians if they experience any signs of excessive anticoagulation, such as unusual or prolonged bleeding, bruising, vomiting, change in stool or urine color, headache, dizziness, or weakness.

Nateglinide (oral)

MONITOR: The coadministration of nateglinide and drugs that are substrates of the CYP450 2C9 enzymatic pathway may result in elevated plasma concentrations of those drugs. The mechanism is decreased clearance due to inhibition of CYP450 2C9 activity by nateglinide.

MANAGEMENT: Caution is advised if nateglinide must be used concomitantly with medications that undergo metabolism by CYP450 2C9, particularly those with a narrow therapeutic range. A lower initial dosage, as well as clinical and laboratory monitoring, may be appropriate for some drugs.

Clopidogrel

MONITOR: The coadministration of oral anticoagulants with clopidogrel may increase the risk for bleeding. The safety of these agents in combination has not been established. A study of 43 patients with atrial fibrillation reported that average INR, R- and S-warfarin levels did not change significantly and bleeding did not occur when clopidogrel 75 mg/day was added to a stable warfarin regimen for 8 days.

MANAGEMENT: The manufacturer recommends to use caution if these two drugs are given concomitantly. The clinician may consider closer INR and clinical monitoring in patients receiving clopidogrel and oral anticoagulants. Patients should be advised to promptly notify their physician if they experience pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, nosebleeds, bleeding of gums from brushing, unusual bleeding or bruising, red or brown urine, or red or black stools.

Carboplatin, Etoposide (oral)

MONITOR: The combination of etoposide and carboplatin has been reported to potentiate the hypoprothrombinemic effect of warfarin in a patient stabilized on the anticoagulant. The patient presented 16 days after initiation of chemotherapy with suspected gastrointestinal bleeding and an INR of 12.6, compared to a baseline range of 1.15 to 2.11 over the previous 8 months. The exact mechanism of the interaction is unknown but may be related to protein binding displacement of warfarin. Etoposide is highly protein-bound in the plasma and has previously been reported to cause an increase in the INR within 24 hours of administration that lasted for approximately 3 days, consistent with the drug's relatively short half-life. The intermediate active platinum derived from carboplatin also binds to plasma proteins and has an elimination half-life of 4.9 to 5.8 days, which the investigators suggest may account for the delayed nature of the interaction in the reported case.

MANAGEMENT: Based on these observations, appropriate monitoring for excessive hypoprothrombinemia is recommended early on and later in time course following the addition of etoposide and carboplatin to a stabilized regimen of warfarin. The same precaution may be applicable during therapy with other oral anticoagulants, since these agents are all highly protein-bound in the plasma. Patients should be advised to promptly report any signs of bleeding to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, vaginal bleeding, nosebleeds, bleeding of gums from brushing, unusual bleeding or bruising, red or brown urine, or red or black stools.

Cefixime

MONITOR: The concomitant administration of cefixime may potentiate the effects of oral anticoagulants. Increases in INR with or without bleeding have been reported. The mechanism is unknown.

MANAGEMENT: Monitoring the patient for signs of bleeding and increases in INR may be advisable during coadministration. Patients should be advised to notify their physicians if they experience any signs of excessive anticoagulation, such as unusual or prolonged bleeding, bruising, coffee ground emesis, change stool or urine color, headache, dizziness, or weakness.

Sunitinib

MONITOR: The concomitant use of anticoagulants may theoretically increase the risk of sunitinib-induced bleeding. Severe and life-threatening tumour hemorrhage, hemoptysis, and pulmonary hemorrhage, and non-serious epistaxis have occurred during treatment with sunitinib.

MANAGEMENT: Monitoring for clinical and laboratory signs of bleeding is recommended during concomitant therapy, including INR/PT and complete blood counts. Patients should be advised to promptly report any potential signs of bleeding to their doctor, including nosebleeds, pain, swelling, or headache.

Terbinafine

MONITOR: The effect of terbinafine on oral anticoagulants, if any, is uncertain. A premarketing study conducted by the manufacturer demonstrated minimal alterations in the pharmacokinetics of a single dose of warfarin administered after 2 weeks of terbinafine compared to placebo. A postmarketing surveillance study of terbinafine in which 26 patients received concomitant treatment with warfarin also revealed no evidence of a drug interaction. However, there have been isolated case reports of INR changes (both increases and decreases) and gastrointestinal bleeding following the addition of terbinafine in patients stabilized on warfarin. A causal relationship has not been established, and the mechanism of the potential interaction is unknown. Terbinafine has not been shown to inhibit or induce CYP450 2C9 and 1A2, the predominant enzymatic pathways responsible for the metabolism of warfarin and similar anticoagulants.

MANAGEMENT: Until further data are available, it may be appropriate to monitor patients more closely during concomitant therapy. The INR should be checked frequently and anticoagulant dosage adjusted accordingly, particularly following initiation or discontinuation of terbinafine in patients who are stabilized on their anticoagulant regimen. Patients should be advised to promptly report any signs of bleeding to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, vaginal bleeding, nosebleeds, bleeding of gums from brushing, unusual bleeding or bruising, red or brown urine, or red or black stools.

Glimepiride

MONITOR: The hypoglycemic action of glimepiride may be potentiated by highly protein-bound drugs. The mechanism is displacement of glimepiride from protein-binding sites, resulting in higher plasma concentrations of unbound glimepiride available to act on pancreatic beta cells.

MANAGEMENT: Patients should be monitored closely for hypoglycemia during concomitant therapy, particularly following addition of a highly protein-bound drug to a stabilized glimepiride regimen. Patients should be advised to regularly monitor their blood sugar, counseled on how to recognize and treat hypoglycemia (e.g., headache, dizziness, drowsiness, nausea, tremor, hunger, weakness, or palpitations) and to notify their physician if it occurs. Likewise, such patients should be observed for loss of glycemic control when the highly protein-bound drug is withdrawn.

Clozapine

MONITOR: The manufacturer reports that clozapine may displace warfarin from plasma protein-binding sites. Increased levels of unbound warfarin could result and could increase the risk of hemorrhage. Clozapine may interact in a similar manner with other oral anticoagulants.

MANAGEMENT: If clozapine and an anticoagulant must be used together, caution is recommended, as is close monitoring for clinical and laboratory evidence of altered anticoagulant effect. Patients should be advised to promptly report any signs of bleeding to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, vaginal bleeding, nosebleeds, bleeding of gums from brushing, unusual bleeding or bruising, red or brown urine, or red or black stools.

Iloprost

MONITOR: Theoretically, iloprost may potentiate the risk of bleeding in patients treated with agents that affect hemostasis such as anticoagulants, platelet inhibitors, thrombin inhibitors, thrombolytic agents, or agents that commonly cause thrombocytopenia. Iloprost inhibits platelet aggregation. In clinical trials, hemoptysis occurred in 5% of the patients receiving iloprost compared to 2% of patients receiving placebo.

MANAGEMENT: Caution is advised if iloprost is used in combination with other drugs that affect hemostasis. Close clinical and laboratory observation for hematologic complications is recommended. Patients should be advised to promptly report any signs of bleeding to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, vaginal bleeding, nosebleeds, bleeding of gums from brushing, unusual bleeding or bruising, red or brown urine, or red or black stools.

Pentosan polysulfate sodium

MONITOR: Theoretically, pentosan polysulfate sodium may potentiate the risk of bleeding in patients treated with agents that affect hemostasis such as anticoagulants, platelet inhibitors, thrombin inhibitors, thrombolytic agents, or agents that commonly cause thrombocytopenia. Pentosan polysulfate sodium is a weak anticoagulant with approximately 1/15 the activity of heparin. Increased prothrombin time, increased partial thromboplastin time, and bleeding complications including ecchymosis, epistaxis, and gum hemorrhage have been reported in association with its use.

MANAGEMENT: Caution is advised if pentosan polysulfate sodium is used in combination with other drugs that affect hemostasis. Close clinical and laboratory observation for hematologic complications is recommended. Patients should be advised to promptly report any signs of bleeding to their doctor, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, vaginal bleeding, nosebleeds, bleeding of gums from brushing, unusual bleeding or bruising, red or brown urine, or red or black stools.

Modafinil

MONITOR: Theoretically, the coadministration with modafinil may increase the plasma concentrations of warfarin and possibly other coumarin anticoagulants. Exposure of human hepatocytes to modafinil in vitro has produced concentration-dependent suppression of CYP450 2C9, the isoenzyme that metabolizes warfarin, but no other indication of CYP450 2C9 suppression has been observed. In a study in which 28 healthy subjects received warfarin (5 mg single dose) alone and after treatment with either modafinil (200 mg for 7 days followed by 400 mg for 21 days) or placebo, the pharmacokinetics of warfarin were not significantly altered following modafinil compared to placebo. However, the potential for more subtle interactions after chronic warfarin administration cannot be ruled out.

MANAGEMENT: Given the potential for interaction and the high degree of interpatient variability with respect to anticoagulant metabolism, patients should be closely monitored during concomitant therapy with modafinil. The INR should be checked regularly for the first several months of coadministration and thereafter whenever modafinil dosing is changed or stopped. Patients should be advised to promptly report any signs of bleeding to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, vaginal bleeding, nosebleeds, bleeding of gums from brushing, unusual bleeding or bruising, red or brown urine, or red or black stools.

Treprostinil

MONITOR: Theoretically, treprostinil may potentiate the risk of bleeding in patients treated with agents that affect hemostasis such as anticoagulants, platelet inhibitors, thrombin inhibitors, thrombolytic agents, or agents that commonly cause thrombocytopenia. Treprostinil inhibits platelet aggregation. However, patients received concomitant anticoagulants and nonsteroidal anti-inflammatory agents without increased bleeding in clinical trials.

MANAGEMENT: Caution is advised if treprostinil is used in combination with other drugs that affect hemostasis. Patients should be advised to promptly report any signs of bleeding to their doctor, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, vaginal bleeding, nosebleeds, bleeding of gums from brushing, unusual bleeding or bruising, red or brown urine, or red or black stools.

Levothyroxine, Liothyronine, Thyroid (desiccated)

MONITOR: Thyroid hormones increase the catabolism of the clotting factors dependent on vitamin K. The hypoprothrombinemic response to oral anticoagulants may be enhanced. No significant effects are expected when warfarin is added to a patient stabilized on thyroid replacement.

MANAGEMENT: INR or prothrombin time should be monitored closely when adding, discontinuing, or changing thyroid hormone dosage. Patients should be advised to notify their physician if they experience any signs of excessive anticoagulation, such as unusual or prolonged bleeding, bruising, vomiting, change in stool or urine color, headache, dizziness, or weakness.

Trastuzumab (injectable)

MONITOR: Trastuzumab may potentiate the hypoprothrombinemic effect of warfarin, although data are limited and the mechanism is unknown. There have been isolated post-marketing case reports of patients stabilized on warfarin whose INR increased several weeks following the addition of trastuzumab, accompanied by signs of excessive anticoagulation such as epistaxis, bruisability, and prolonged bleeding from cuts. A search on a large patient database conducted by the manufacturer of trastuzumab found several bleeding events among 469 women with HER2-overexpressing metastatic breast cancer who were randomized to receive chemotherapy with or without trastuzumab. Although trastuzumab patients had a higher crude incidence of bleeding events, which were generally mild to moderate in severity (with nosebleeds being the most common), they were also seen more frequently and received treatment significantly longer than controls. After adjusting for time on study, the incidence rates were similar for both groups, with or without anticoagulants.

MANAGEMENT: Given the potential for interaction and the high degree of interpatient variability with respect to warfarin metabolism, patients should be closely monitored during concomitant therapy with trastuzumab. The INR should be checked frequently and warfarin dosage adjusted accordingly, particularly following initiation, discontinuation or change of dosage of trastuzumab in patients who are stabilized on their warfarin regimen. The same precaution may be applicable during therapy with other oral anticoagulants, although clinical data are lacking. Patients should be advised to promptly report any signs of bleeding to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, vaginal bleeding, nosebleeds, bleeding of gums from brushing, unusual bleeding or bruising, red or brown urine, or red or black stools.

Valproic acid, Divalproex sodium

MONITOR: Valproic acid may transiently potentiate the hypoprothrombinemic effect of warfarin and other oral anticoagulants. The mechanism is protein binding displacement by valproic acid, resulting in increased plasma concentrations of unbound (active) anticoagulant drug. In an in vitro study, valproic acid increased the unbound fraction of warfarin by up to 33%. The clinical significance of this effect is unknown. The interaction was suspected in one case report, although a causal relationship was not established. Because drug clearance is also increased with higher levels of unbound drug, the pharmacokinetic effect produced by valproic acid is expected to balance out over time, resulting in little or no net effect on anticoagulant pharmacokinetics in the long-term. However, valproic acid appears to cause dose-related thrombocytopenia. This effect should be considered in patients treated concomitantly with oral anticoagulants.

MANAGEMENT: Based on the proposed transient nature of the pharmacokinetic interaction, appropriate monitoring for excessive hypoprothrombinemia is recommended during the first week following the addition of valproic acid to a stabilized regimen of warfarin or other oral anticoagulants. Patients should be advised to promptly report any signs of bleeding to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, vaginal bleeding, nosebleeds, bleeding of gums from brushing, unusual bleeding or bruising, red or brown urine, or red or black stools.

Vitamin E

MONITOR: Vitamin E may potentiate the effects of warfarin and other related anticoagulants and increase the risk of bleeding. Vitamin E is thought to inhibit the oxidation of reduced vitamin K and interfere with the functions of vitamin K-dependent clotting factors. Although conflicting information exists regarding the effect of vitamin E on coagulation, available data suggest a potential interaction with oral anticoagulants at vitamin E dosages exceeding 400 international units per day.

MANAGEMENT: The INR should be monitored closely and anticoagulant dosage adjusted accordingly during concomitant therapy, particularly following initiation or discontinuation of vitamin E therapy in patients who are stabilized on an oral anticoagulant. Patients should be advised to notify their physicians if they experience any signs of excessive anticoagulation, such as unusual or prolonged bleeding, bruising, vomiting, change in stool or urine color, headache, dizziness, or weakness.

Phytonadione

MONITOR: Vitamin K may antagonize the hypoprothrombinemic effect of oral anticoagulants. Vitamin K is a cofactor in the synthesis of blood clotting factors that are inhibited by oral anticoagulants, thus administration of vitamin K can reverse the action of oral anticoagulants.

MANAGEMENT: Intake of vitamin K through supplements or diet should not vary significantly during oral anticoagulant therapy. The International Normalization Ratio (INR) should be monitored and anticoagulant dosage adjusted as necessary.

Zafirlukast

MONITOR: Zafirlukast may increase warfarin concentrations and anticoagulant effects. The mechanism is inhibition of CYP450 2C9 hepatic metabolism of S-warfarin. Other oral anticoagulants may also be affected.

MANAGEMENT: If given together, the INR or PT should be monitored closely and the warfarin dose should be adjusted accordingly. Patients should be advised to promptly report any signs of bleeding to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, vaginal bleeding, nosebleeds, bleeding of gums from brushing, unusual bleeding or bruising, red or brown urine, or red or black stools.

Zileuton

MONITOR: Zileuton is believed to inhibit the hepatic metabolism of warfarin, and may increase its anticoagulant effect. This interaction appears to stereospecifically inhibit metabolism of the R-enantiomer of warfarin.

MANAGEMENT: If these drugs must be used together, the INR or PT should be monitored closely, and the warfarin dose adjusted accordingly. Patients should be advised to promptly report any signs of bleeding to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, vaginal bleeding, nosebleeds, bleeding of gums from brushing, unusual bleeding or bruising, red or brown urine, or red or black stools.

Nitazoxanide

Nitazoxanide may increase the potential for toxicity of other highly plasma protein bound drugs. The mechanism of action is competition for plasma protein binding sites. Tizoxanide, its active metabolite, is over 99.9% plasma protein bound. The clinical significance is unknown. Caution is recommended when highly protein bound drugs with narrow therapeutic indices are administered concurrently. Additionally, total plasma concentrations should be interpreted cautiously, because a greater amount of free drug may be present.

Chlordiazepoxide

One case report describes increased warfarin requirements when chlordiazepoxide was added to the regimen. However, controlled trials with many patients have not reported an interaction. Patients on oral anticoagulants should be monitored for alterations in INR whenever a change in drug regimen involves chlordiazepoxide.

Trazodone

One case report has suggested that trazodone may reduce prothrombin and partial thromboplastin times. The mechanism and clinical significance are unknown. Similar effects may occur with other oral anticoagulants. The patient's PT or INR should be observed for alterations. An increase in oral anticoagulant dosage may be necessary. Patients should be advised to promptly report any signs of blood clots (e.g. chest pain, shortness of breath, sudden loss of vision, or pain, redness or swelling in an extremity).

Olsalazine

Patients treated concomitantly with olsalazine and warfarin have experienced an increase in prothrombin time. The clinical significance is unknown. During concomitant therapy, the INR should be monitored. Other oral anticoagulants may interact with olsalazine in a similar manner.

Pravastatin

Pravastatin has not been reported to affect the response to warfarin. However, some other HMG-CoA reductase inhibitors have been shown to lead to an enhanced hypoprothrombinemic response. Some patients have developed bleeding (i.e., hematuria, epistaxis). The clinical significance of this interaction has not been fully evaluated. Patients should be monitored for altered hypoprothrombinemic response to warfarin or other oral anticoagulants when pravastatin dosage is added, discontinued or changed. Patients should be advised to promptly report any signs of bleeding to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, vaginal bleeding, nosebleeds, bleeding of gums from brushing, unusual bleeding or bruising, red or brown urine, or red or black stools.

Simvastatin

Simvastatin may slightly increase the anticoagulant response to warfarin. The mechanism may be warfarin displacement from protein binding sites. The clinical significance of this interaction has not been established, however, clinical monitoring of patient warfarin response and tolerance is recommended. Simvastatin may interact in a similar way with other oral anticoagulants. In addition, a case study has reported reversible rhabdomyolysis and acute renal failure 7 days following initiation of warfarin in a patient stabilized on simvastatin. Pravastatin, another HMG-CoA reductase inhibitor, does not appear to affect anticoagulation. Patients should be advised to promptly report any signs of bleeding to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, vaginal bleeding, nosebleeds, bleeding of gums from brushing, unusual bleeding or bruising, red or brown urine, or red or black stools.

Chamomile, Fenugreek

Some herbs such as chamomile and fenugreek have coumarin constituents. Theoretically, ingesting large quantities of these herbs may potentiate the risk of bleeding in patients treated with agents that affect hemostasis such as anticoagulants, platelet inhibitors, thrombin inhibitors, thrombolytic agents, or agents that commonly cause thrombocytopenia. However, their effects on the coagulation system have not been studied, and bleeding complications have not been reported in humans. Moreover, pharmacologic effects may be highly variable due to inconsistencies in formulation and potency of commercial herbal products. In one patient, the addition of boldo and fenugreek to a stabilized regimen of warfarin resulted in an increase in INR. The INR returned to normal 1 week following discontinuation of both products but increased again when the patient resumed usage, which subsequently led to a 15% reduction in the weekly warfarin dosage. It is not certain whether the effects on INR are due to boldo or fenugreek, or a combination of both products. Patients should consult a healthcare provider before taking any herbal or alternative medicine. In patients who have used chamomile or fenugreek extensively prior to receiving anticoagulation, antiplatelet or thrombolytic therapy, the potential for an interaction should be considered. Close clinical and laboratory observation for hematologic complications is recommended. Patients should be advised to promptly report any signs of bleeding to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, vaginal bleeding, nosebleeds, bleeding of gums from brushing, unusual bleeding or bruising, red or brown urine, or red or black stools.

Furosemide, Bumetanide, Ethacrynic acid, Torsemide

Some loop diuretics may displace warfarin or other oral anticoagulants from plasma protein binding sites. Plasma warfarin concentrations and warfarin effects may be increased. Data are available for furosemide and ethacrynic acid only. This interaction appears to be of minor clinical importance. Close monitoring of the INR is recommended, particularly if diuretic dosage must be high. Patients should be advised to promptly report any signs of bleeding to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, vaginal bleeding, nosebleeds, bleeding of gums from brushing, unusual bleeding or bruising, red or brown urine, or red or black stools.

Propranolol, Timolol, Penbutolol

Some oral beta-blockers may increase serum levels of oral anticoagulants and enhance anticoagulant effects. The mechanism is unknown . Data have been conflicting. Case reports exist for propranolol, while acebutolol, atenolol, betaxolol, bisoprolol, carvedilol, esmolol, metoprolol, and pindolol reportedly do not interact. Hypoprothrombinemic effects are expected to be minimal. It is recommended that PT or INR be monitored when a beta-blocker is added to or deleted from the medical regimen of a patient receiving an oral anticoagulant. Patients should be advised to promptly report any signs of bleeding to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, vaginal bleeding, nosebleeds, bleeding of gums from brushing, unusual bleeding or bruising, red or brown urine, or red or black stools.

Spironolactone

Spironolactone may cause diuresis and hemoconcentration of clotting factors. The effects of some anticoagulants may be decreased. Data are available for warfarin. The INR or PT should be monitored, and oral coagulant dosage should be increased as needed.

Tamsulosin (oral)

Studies evaluating a potential drug interaction between warfarin and tamsulosin have provided inconclusive results. The manufacturer recommends caution if these agents are used together. Clinical and INR/PT monitoring is recommended. Patients should be advised to promptly report any signs of bleeding to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, vaginal bleeding, nosebleeds, bleeding of gums from brushing, unusual bleeding or bruising, red or brown urine, or red or black stools.

Atorvastatin

The manufacturer reports that atorvastatin had no clinically significant effect on prothrombin time when given to patients also receiving chronic warfarin therapy. The clinical utility of the lack of this interaction has not been fully evaluated, and caution should still be exercised if these drugs are coadministered. Lovastatin has been shown to alter the hypoprothrombinemic response to warfarin. Theoretically, no interaction should occur with other oral anticoagulants and atorvastatin, although data is lacking.

Capsicum

Theoretically, capsicum may potentiate the risk of bleeding in patients treated with agents that affect hemostasis such as anticoagulants, platelet inhibitors, thrombin inhibitors, thrombolytic agents, or agents that commonly cause thrombocytopenia. Animal and limited human data suggest that capsicum may increase fibrinolytic activity, lower plasma fibrinogen levels, increase plasma antithrombin III levels, inhibit platelet aggregation, and prolong bleeding time. However, bleeding complications and interactions with hematologic agents have not been reported. Moreover, pharmacologic effects of capsicum preparations may be highly variable due to inconsistencies in formulation and potency of commercial herbal products. Patients should consult a healthcare provider before taking any herbal or alternative medicine. In patients who have used capsicum extensively prior to receiving anticoagulation, antiplatelet, or thrombolytic therapy, the potential for an interaction should be considered. Clinical and laboratory observation for hematologic complications is recommended. Patients should be advised to promptly report any signs of bleeding to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, vaginal bleeding, nosebleeds, bleeding of gums from brushing, unusual bleeding or bruising, red or brown urine, or red or black stools.

Feverfew

Theoretically, feverfew extracts may potentiate the effects of anticoagulants, platelet inhibitors, and thrombolytic agents, possibly increasing the risk of bleeding. In vitro studies have shown feverfew to interfere with hemostasis and platelet aggregation by neutralizing platelet sulfhydryl groups as well as preventing prostaglandin synthesis. However, bleeding complications and interactions with hematologic agents have not been reported. Moreover, pharmacologic effects may be highly variable due to inconsistencies in formulation and potency of commercial herbal products. Patients should consult a healthcare provider before taking any herbal or alternative medicine. In patients who have used feverfew extensively prior to receiving anticoagulation, antiplatelet, or thrombolytic therapy, the potential for an interaction should be considered. Clinical and laboratory observation for hematologic complications is recommended. Patients should be advised to promptly report any signs of bleeding to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, vaginal bleeding, nosebleeds, bleeding of gums from brushing, unusual bleeding or bruising, red or brown urine, or red or black stools.

Ascorbic acid (vitamin C)

The use of ascorbic acid has been implicated in causing warfarin resistance. However, controlled clinical trials have not demonstrated clinically important variations in prothrombin times. The possibility of an interaction may be considered if warfarin resistance is encountered in patients taking high doses of ascorbic acid. Other anticoagulants may be affected as well, although data are lacking.

Chlorothiazide, Chlorthalidone, Hydrochlorothiazide, Indapamide, Metolazone

Thiazide diuretics may reduce intravascular volume. Higher plasma levels of clotting factors and decreased anticoagulant effect may result; however the effects may not be clinically significant. Management consists of monitoring coagulation parameters during the initial phase of coadministration. Patients should be advised to promptly report any signs of blood clots (e.g. chest pain, shortness of breath, sudden loss of vision, or pain, redness or swelling in an extremity).

Tolcapone

Tolcapone may theoretically interfere with the CYP450 2C9 hepatic metabolism of oral anticoagulants. Based on interaction studies with tolcapone and other drugs, however, the clinical impact may be insignificant. The manufacturer recommends close monitoring of coagulation parameters if these drugs are coadministered. Patients should be advised to promptly report any signs of bleeding to their physician, including pain, swelling, headache, dizziness, weakness, prolonged bleeding from cuts, increased menstrual flow, vaginal bleeding, nosebleeds, bleeding of gums from brushing, unusual bleeding or bruising, red or brown urine, or red or black stools.

Advertisement