Interaction between warfarin and levofloxacin: case series.
Key Words: warfarin, levofloxacin, interaction, hemopericardium, and retroperitoneal bleeding
Warfarin is used for multiple clinical conditions including atrial fibrillation, venous thromboembolism (VTE), certain mechanical prosthetic valves, and hypercoagulable states with concomitant VTE for its anticoagulant effect by inhibiting the vitamin K cycle. Levofloxacin, a fluoroquinolone antibiotic, has rapidly gained popularity among physicians for its broad-spectrum coverage, once daily dosing, and overall good side effect profile. (1-3) A few case reports thus far have suggested a drug-drug interaction between fluoroquinolone antibiotics and warfarin. (4-10) However, this interaction is not well understood due to the lack of prospective studies documenting this effect. We present a series of three cases where an apparent warfarin-levofloxacin interaction led to a significant rise in the INR and subsequent life-threatening bleeding complications, including one fatality.
Patient 1: Hemopericardium
A 70-year-old woman on warfarin for chronic atrial fibrillation presented to the emergency room (ER) with a 2-day history of nausea without vomiting and one day of right upper quadrant abdominal pain. She denied any fever, chills, chest pain, shortness of breath, or dark stools. She was on her last day of a 10-day course of levofloxacin for left lower lobe pneumonia. On physical examination, she was alert, oriented, and in mild distress. Her vital signs revealed a temperature of 38.5[degrees]C, pulse of 51/min, respiratory rate of 18/min, a blood pressure of 80/37 mm Hg, and an oxygen saturation of 98% on room air. She was anicteric and had no jugular venous distension (JVD). Chest was clear to auscultation. Her heart sounds were distant, and there were no murmurs, gallops, or rubs. She had mild epigastric and right upper quadrant tenderness with no Murphy sign. Her lower extremities had good distal pulses without edema. Neurologic examination was normal. Laboratory evaluation revealed a hemoglobin of 11.1 g/dL, white blood cell (WBC) count of 15,500 cells/mm3 with 77% segmented neutrophils. Her prothrombin time was 36.5 seconds and INR was 3.3. Chest x-ray (CXR) revealed cardiomegaly with increased pulmonary vascularity, but no infiltrates. She was admitted for hypotension with possible sepsis. The patient developed increasing dyspnea on exertion and by the third hospital day, her INR had increased to 7.3 despite warfarin being held (Table 1). A repeat CXR revealed bibasilar opacities with worsening cardiomegaly. Computed tomography scan showed bilateral pleural effusions with a large pericardial effusion (Fig. 1). An echocardiogram revealed normal systolic function but with evidence of impending pericardial tamponade. Before a pericardial window could be performed, the patient suffered a cardiac arrest requiring a bedside pericardiocentesis which yielded 850 cc of bloody fluid and a prolonged resuscitation effort. The patient developed multiorgan failure and severe anoxic brain injury. After 2 weeks in the medical intensive care unit, there was no neurologic improvement. Life-sustaining measures were withdrawn at the family's request in compliance with the patient's advance directives and the patient died.
Patient 2: Hemopericardium
An 80-year-old woman, a nursing home resident, on warfarin for chronic atrial fibrillation, presented to the ER with a 3-day history of nausea, vomiting, mild upper abdominal pain, and dyspnea on exertion. Two days before presentation, she completed a course of levofloxacin for a urinary tract infection. On physical examination, she was an obese, elderly woman in moderate distress with a temperature of 37[degrees]C, heart rate of 91 beats/min, respiratory rate of 24 respirations/min, and a blood pressure of 114/65 mm Hg. Her lungs had bibasilar crackles. Her cardiac examination revealed an irregularly irregular rhythm without murmurs, gallops, or rubs. She had mild epigastric tenderness and bilateral pitting pedal edema. Her neurologic examination was normal. Laboratory evaluation revealed a WBC count of 18,400 cells/[cm.sup.3], hemoglobin of 12.2 g/dL, prothrombin time >100 seconds, and INR >8.6 (Table 1). Abdominal ultrasound was normal, and CXR revealed a left lower lobe infiltrate. Computed tomography scan of the abdomen revealed a large pericardial effusion (Fig. 2) confirmed by transthoracic echocardiogram. There was no evidence of cardiac tamponade. The patient received fresh frozen plasma and vitamin K with improvement in her INR as noted in Table 1. A follow-up echocardiogram suggested early cardiac tamponade, and a pericardial window was performed revealing a large amount of bloody fluid within the pericardium. The patient was eventually discharged back to the nursing home in stable condition and warfarin was discontinued.
Patient 3: Retroperitoneal Hematoma
A 34-year-old woman with schizophrenia, migraine headaches, hypertension, epilepsy, peptic ulcer disease, and tobacco abuse was admitted to the hospital with a 3-day history of fever, shortness of breath, palpitations and tremors. Her neurologist had recently placed her on haloperidol and risperidone. Neuroleptic malignant syndrome was suspected. The patient had recently been hospitalized and diagnosed with a pulmonary embolus and was prescribed warfarin, although she did not take it after discharge from the hospital. On readmission, 3 mg warfarin, bridged with enoxaparin, was started. In addition, her urinalysis revealed a urinary tract infection, and she was started on levofloxacin. Laboratory evaluation on Day 5 of the hospital stay (2nd day of warfarin) revealed a high INR of 7 with a hemoglobin level of 7.5 g/dL (hemoglobin at admission was 13 g/dL) (Table 1). Physical examination revealed no obvious signs of bleeding and stool hemoccult testing was negative. Computed tomography scan of the abdomen revealed a large retro-peritoneal hematoma with involvement of the right psoas and iliacus muscles (Fig. 3). The patient was treated with fresh frozen plasma, vitamin K, and packed red blood cell transfusions. The patient's INR normalized and there was no further drop in hemoglobin. Warfarin, enoxaparin, and levofloxacin were discontinued. A follow-up CT scan and magnetic resonance imaging (MRI) of the abdomen revealed resolution of the retroperitoneal, right psoas, and iliacus muscle hematomas. The patient was discharged home in stable condition without warfarin.
Levofloxacin, a fluoroquinolone, is one of the most widely used antibiotics due to its broad spectrum coverage, (11) once-daily dosing, and generally favorable side effect profile. (3) Levofloxacin is the pure S-isomer of ofloxacin (12) and inhibits type II topoisomerases, including the bacterial topoisomerase IV and DNA gyrase. These enzymes are required for DNA replication, transcription, repair, and recombination. (13) Eighty percent of levofloxacin is eliminated unchanged in the urine. (11)
Warfarin, the mostly widely used vitamin-K antagonist, inhibits the enzyme epoxide reductase, a key enzyme in the vitamin K conversion cycle. This prevents the post-translational [gamma]-carboxylation of glutamate residues of coagulation factors II, VII, IX and X that are required for their activation. (14-16) Warfarin is a mixture of R- and S-enantiomers, with most of the anticoagulant activity attributable to S-warfarin. (17) Because of regioselective metabolism by P-450 isozymes in the liver, R-warfarin is preferentially metabolized by P-450 IA2 isozyme, and S-warfarin is predominantly metabolized by P-450 2C9 isozyme via oxidation. (10,14,16,18)
The anticoagulation effect of warfarin is monitored by following the INR and is maintained between 2.0 and 3.5 depending on the underlying clinical condition. (2,14) The risk of significant bleeding complications increases if the INR is not maintained within the upper limits of the therapeutic range. The anticoagulation effect of warfarin is influenced by various pharmacokinetic factors including total absorption, rate of metabolism, and individual body response to a given concentration of the drug. (14) Drugs that inhibit or induce P-450 isozymes can either potentiate or reduce the anticoagulation effect of warfarin, respectively.
Although significant interactions have been noted between warfarin and several of the fluoroquinolone antibiotics such as ciprofloxacin, ofloxacin, and norfloxacin, no prospective trials have shown a similar interaction between warfarin and levofloxacin. (7-9,19,20) Two prospective studies done by Liao et al in 1996 and Yamreudeewong et al in 2003 with small numbers of participants revealed no clinically significant interaction between warfarin and levofloxacin when used simultaneously. (1,3) Another retrospective cohort study done by McCall et al in 2005 showed no significant interaction between warfarin and levofloxacin. (2)
Multiple case reports and a case series, however, have suggested that a significant drug-drug interaction exists between warfarin and levofloxacin. Ravnan and Locke, Gheno and Cinetto, and Jones and Fugate all reported cases between 2001 and 2002 documenting potentiation of the anticoagulation effect of warfarin (prolongation of the INR) due to the concomitant use of levofloxacin. (4-6)
There are several proposed mechanisms by which fluoroquinolones may augment the anticoagulant effect of warfarin. First, with their broad spectrum of antibiotic activity, there is likely a reduction of vitamin K-producing bacteria in the gut, thus leading to a decrease in the vitamin K level in the body (4) Studies are inconsistent about the effect of fluoroquinolones on the gut bacteria with ciprofloxacin having more effect and ofloxacin having doubtful effects. (21-24) No studies have been reported specifically with levofloxacin in this regard, and, hence, more research is needed to confirm or deny this mechanism.
Secondly, the fluoroquinolones may cause displacement of warfarin from protein- binding sites, hence prolonging its activity. (4,7) Levofloxacin is less protein bound in the plasma (24-38%) (11) compared with warfarin (99%). (34) More evidence is required, especially with calculated affinities of both drugs to plasma protein alone and also in the presence of each other, to confirm or refute this theory.
Thirdly, the fluoroquinolone antibiotics, including levofloxacin, might inhibit P-450 2C9, the major enzyme involved in the metabolism of warfarin in the liver and hence delay its elimination. However, there are no studies indicating the effect of fluoroquinolones on P-450 2C9 isozyme. Instead, one of the more thoroughly studied fluoroquinolones, ciprofloxacin, has been shown to have an inhibitory effect on the P-450 1A2 isozyme (26) involved in the metabolism of R-warfarin. This might explain, at least in part, the observed drug-drug interaction since R-warfarin accounts for a small percentage of the antithrombotic effect of warfarin. Further research is needed to more clearly delineate the exact mechanisms and magnitude of the observed drug-drug interactions.
Our two cases of hemopericardium demonstrate an extremely serious potential complication of combining levofloxacin and warfarin with level of evidence of 2 (a probable interaction) as per the criteria for establishing a drug or food interaction with warfarin (Table 2). (27) We hypothesize that pericardial inflammation due to pneumonia might have increased the propensity for hemopericardium in the setting of significant supratherapeutic anticoagulation. A number of case reports of hemopericardium and cardiac tamponade exist in association with warfarin therapy alone in patients with predisposing factors such as myocardial infarction or necrosis, pericarditis, liver dysfunction, blunt or penetrating trauma, recent cardiac valve surgery, and free wall rupture of the left ventricle. (28-31)
In our third case, the concomitant use of enoxaparin may have contributed to the retroperitoneal hematoma and other described bleeding, giving a level of evidence of 3 (a possible interaction) as per the criteria as shown in Table 2. (27) However, the quick and unexpected rise in INR to 7 in less than 2 days with the use of a relatively low dose of warfarin highlights the possibility of an independent interaction between warfarin and levofloxacin.
A case-control study by Stroud et al showed no clinically significant bleeding associated with the simultaneous use of levofloxacin and warfarin. (32) In addition, a PubMed search did not yield reports of cases of significant bleeding such as hemopericardium or retroperitoneal bleeding from the interaction of levofloxacin with warfarin with or without predisposing factors. Therefore, this series of important bleeding complications should raise physicians' awareness of a possible interaction.
Our cases illustrate life-threatening bleeding complications associated with the simultaneous use of warfarin and levofloxacin. Although prospective and retrospective studies have failed to indicate a clinically significant increase in the risk of bleeding, those studies had small sample sizes making the interpretation of the results equivocal. Physicians should be aware of the possible drug-drug interaction between warfarin and levofloxacin and the fact that this may lead to serious complications. If a patient is taking warfarin, levofloxacin should not be used if an effective alternative antibiotic is available. Either empiric dosage adjustment of warfarin and/or more frequent monitoring of INR is needed should levofloxacin be used with warfarin.
The authors sincerely thank the staff of the Johnson City Medical Center Library and the Department of Radiology at Johnson City Medical Center for their help in retrieving articles and providing necessary x-rays used in this article.
1. Yamreudeewong W, Lower DL, Kilpatrick DM, et al. Effect of levofloxacin coadministration on the international normalized ratios during warfarin therapy. Pharmacotherapy 2003;23:333-338.
2. McCall KL, Scott JC, Anderson HG. Retrospective evaluation of a possible interaction between warfarin and levofloxacin. Pharmacotherapy 2005;25:67-73.
3. Liao S, Palmer M, Fowler C, et al. Absence of an effect of levofloxacin on warfarin pharmacokinetics and anticoagulation in male volunteers. J Clin Pharmacol 1996;36:1072-1077.
4. Jones CB, Fugate SE. Levofloxacin and warfarin interaction. Ann Pharmacother 2002;36:1554-1557.
5. Gheno G, Cinetto L. Levofloxacin-warfarin interaction. Eur J Clin Pharmacol 2001;57:427.
6. Ravnan SL, Locke C. Levofloxacin and warfarin interaction. Pharmacotherapy 2001;21:884-885.
7. Baciewicz AM, Ashar BH, Locke TW. Interaction of ofloxacin and warfarin. Ann Intern Med 1993;119:1223.
8. Leor J, Matetzki S. Ofloxacin and warfarin. Ann Intern Med 1988;109:761.
9. Ellis RJ, Mayo MS, Bodensteiner DM. Ciprofloxacin-warfarin coagulopathy: a case series. Am J Hematol 2000;63:28-31.
10. Artymowicz RJ, Cino BJ, Rossi JG, et al. Possible interaction between gatifloxacin and warfarin. Am J Health Syst Pharm 2002;59:1205-1206.
11. Fish DN, Chow AT. The clinical pharmacokinetics of levofloxacin. Clin Pharmacokinet 1997;32:101-119.
12. Peng MY. Randomized, double-blind, comparative study of levofloxacin and ofloxacin in the treatment of complicated urinary tract infections. J Microbiol Immunol Infect 1999;32:33-39.
13. Chung P, McNamara PJ, Campion JJ, et al. Mechanism-based pharmacodynamic models of fluoroquinolone resistance in Staphylococcus aureus. Antimicrob Agents Chemother 2006;50:2957-2965.
14. Hirsh J, Dalen J, Anderson DR, et al. Oral anticoagulants: mechanism of action, clinical effectiveness, and optimal therapeutic range. Chest 2001;119(1 Suppl):8S-21S.
15. Stafford DW. The vitamin K cycle. J Thromb Haemost 2005;3:1873-1878.
16. Ansell J, Hirsh J, Poller L, et al. The pharmacology and management of the vitamin K antagonists: the Seventh ACCP Conference on Antithrombotic and Thrombolytic Therapy. Chest 2004; 126 (3 Suppl):204S-233S.
17. Greenblatt DJ, von Moltke LL. Interaction of warfarin with drugs, natural substances, and foods. J Clin Pharmacol 2005;45:127-132.
18. Kaminsky LS, Zhang ZY. Human P450 metabolism of warfarin. Pharmacol Ther 1997;73:67-74.
19. Marchbanks CR. Drug-drug interactions with fluoroquinolones. Pharmacotherapy 1993;13(2 Pt 2):23S-28S.
20. De Sarro A, De Sarro G. Adverse reactions to fluoroquinolones. an overview on mechanistic aspects. Curr Med Chem 2001;8:371-384.
21. van Saene HK, Lemmens SE, van Saene JJ. Gut decontamination by oral ofloxacin and ciprofloxacin in healthy volunteers. J Antimicrob Chemother 1988;22(Suppl C):127-134.
22. Edlund C, Nord CE. A review on the impact of 4-quinolones on the normal oropharyngeal and intestinal human microflora. Infection 1988;16:8-12.
23. Midtvedt T. Influence of ofloxacin on the faecal flora. Drugs 1987;34 (Suppl 1):154-158.
24. Pecquet S, Andremont A. Tancrede C. Effect of oral ofloxacin on fecal bacteria in human volunteers. Antimicrob Agents Chemother 1987;31:124-125.
25. Micromedex Health Series. Warfarin. Thompson Micromedex Health Series. 2007. Ref Type: Electronic Citation.
26. Fuhr U, Strobl G, Manaut F, et al. Quinolone antibacterial agents: relationship between structure and in vitro inhibition of the human cytochrome P450 isoform CYP1A2. Mol Pharmacol 1993;43:191-199.
27. Wells PS, Holbrook AM, Crowther NR, et al. Interactions of warfarin with drugs and food. Ann Intern Med 1994;121:676-683.
28. Miller RL. Hemopericardium with use of oral anticoagulant therapy. JAMA 1969;209:1362-1364.
29. Lee KS, Marwick T. Hemopericardium and cardiac tamponade associated with warfarin therapy. Cleve Clin J Med 1993;60:336-338.
30. Zipes DP, Braunwald E. Braunwald's Heart Disease: A Textbook of Cardiovascular Medicine, 7th ed. Philadelphia, W.B. Saunders, 2005.
31. Andreoli TE, Cecil RL. Cecil Essentials of Medicine, 6th ed. Philadelphia, W.B. Saunders, 2004.
32. Stroud LF, Mamdami MM, Kopp A, et al. The safety of levofloxacin in elderly patients on warfarin. Am J Med 2005;118:1417.
Raja S. Vadlamudi, MD, MPH, Roger D. Smalligan, MD, MPH, and Hassan M. Ismail, MD, MPH
From the Department of Internal Medicine, Quillen College of Medicine, East Tennessee State University, Johnson City, TN.
Reprint requests to Raja S. Vadlamudi, MD, MPH, Department of Internal Medicine, East Tennessee State University, VA-Building 1, Box 70622, Johnson City, TN 37614-0622. Email: email@example.com
Accepted February 28, 2007.
RELATED ARTICLE: Key Points
* Although prospective studies have revealed no significant drug-drug interaction between warfarin and levofloxacin, multiple case reports have noted prolongation of anticoagulant effect with no mention of bleeding complications when the two medications are combined.
* Bleeding complications resulting from drug-drug interactions between warfarin and levofloxacin can be serious and life threatening.
* Physicians should be wary of this interaction and be cautious in prescribing levofloxacin to patients taking warfarin.
Table 1. Laboratory findings of cases Dose of International normalized ratio (INR) warfarin Prior to Day of At time of Follow up after Cases (mg) admission admission discharge discharge Case 1 5 3.7 3.3 ** ** Case 2 5 2.8 >8.6 1.6 1.1 Case 3 5 1.2 2.0 1.2 NA Maximum INR Type of bleeding Cases reached complication Procedures used Case 1 >8.6 Hemopericardium Pericardiocentesis with pericardial window, FFP, vitamin k Case 2 >8.6 Hemopericardium Pericardiocentesis with pericardial window, FFP, vitamin K Case 3 7.0 Retroperitoneal FFP, vitamin K. bleeding with psoas PRBC transfusions muscle bleeding **. Not available as patient expired; NA, not available: FFP, fresh frozen plasma; PRBC, packed red blood cells. Table 2. Criteria for establishing a drug or food interaction with warfarin Level Criteria Case 1 Case 2 Case 3 A Was the timing pharmacologically Yes Yes Yes plausible? B Did results from the international Yes Yes Yes normalized ratio, prothrombin time, or thrombotest support the contention? C Were other potential factors affecting Yes Yes No warfarin pharmacokinetics or pharmacodynamics ruled out? D Was there other objective evidence (such No No No as warfarin blood levels)? E Was a dose-response relation shown for No No No the interacting drug? F Was the patient rechallenged and, if so, No No No did a similar response occur? G Did the same thing happen on previous No No No exposure to the drug? Yes, meeting the criteria: No, does not meet the criteria; potential factors taken into consideration for criteria C include liver disease, other drug therapy, and diet (notably dietary vitamin K intake). Level of Evidence: 1 = Highly Probable (A, B, and C, plus any one or more of D to G). 2 = Probable (A, B, plus one of more of C to G). 3 = Possible (A plus one or more of B to G). 4 = Doubtful (Any combination of B to G or A alone). Reprinted with permission from Wells PS, Holbrook AM, Crowther NR, et al. Interactions of warfarin with drugs and food. Ann Intern Med 1994;121:676-683.
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||Case Report|
|Author:||Vadlamudi, Raja S.; Smalligan, Roger D.; Ismail, Hassan M.|
|Publication:||Southern Medical Journal|
|Date:||Jul 1, 2007|
|Previous Article:||Wernicke encephalopathy complicating lymphoma therapy: case report and literature review.|
|Next Article:||Prolonged acenocoumarol treatment in patients with venous thromboembolic disease from a rural area.|