 |
|
Interaction of Duloxetine and Warfarin Causing Severe Elevation of International Normalized Ratio
To the Editor: We describe what we believe is the first report of a drug interaction between duloxetine hydrochloride and warfarin. The patient has provided consent for publication of this report, and it has received institutional review board approval.
Report of a Case
A 44-year-old woman, self-designated as white, homozygous for the factor V Leiden mutation, had been treated uneventfully for 1 year after an ischemic stroke with warfarin, 7.5 to 10 mg/d (mean international normalized ratio [INR], 2.2 [SD, 0.5]) (Figure). She had normal levels of protein C, protein S, homocysteine, and lipoprotein(a); homozygous wild-type genotypes for prothrombin (G20210A), MTHFR (C677T), PAI-1 (4G/5G), and platelet glycoprotein PL A1/A2; and no lupus anticoagulant or anticardiolipin antibodies. She had increased factor XI (179% [upper normal limit, 150%]) and plasminogen activator inhibitor activity (71.7 IU/mL [upper normal limit, 21.2]). She was being treated for poststroke seizures, migraine prophylaxis, panic disorder, and bronchospasm. She had no heart failure; thyroid, hepatic, or renal disease; or major weight loss or gain. She had a normal serum albumin–globulin ratio and reported taking no complementary or alternative medications.
|
|
|
|
Figure. Events and Laboratory Values Following Initiation of Duloxetine in a Patient Taking Warfarin
Mean international normalized ratio with warfarin, 7.5-10 mg/d, was 2.2 (SD, 0.5) (shaded region) for the year prior to starting duloxetine. Coagulation abnormalities before warfarin therapy included factor V Leiden homozygosity, high factor XI (179%; reference range, 50% to 150%), and high plasminogen activator inhibitor activity (71.7 IU/mL; reference range, 2 to 21.2 IU/mL).
*Upper limit of assay.
|
|
|
She had been taking a stable regimen of atorvastatin, 10 mg/d; warfarin, 7.5 to 10 mg/d; lamotrigine, 50 mg/d; topiramate, 200 mg/d; clonazepam, 2 mg/d; and albuterol extended-release tablets, 4 mg twice daily. She started taking duloxetine, a selective serotonin-norepinephrine reuptake inhibitor, 30 mg/d, on day 0 because of depression-associated sleeplessness (Figure).
She developed petechiae/purpura on day 55, with INR 5.0. Warfarin was stopped on day 58, but duloxetine was continued. Test results for thyroid, renal, liver, and platelet function and glucose level remained normal. There was an increase in INR to greater than 19 on day 85. There was no evidence for warfarin self-intoxication based on pill counts, discussions with family members, and absence of signs of a second warfarin supply.
On day 85 (27 days after the last warfarin dose), plasma warfarin level was 5.3 µg/mL (therapeutic range, 2-8 µg/mL). After treatment with intravenous vitamin K on day 85, INR briefly decreased but increased again to 6.4 on day 94. At that time (36 days after stopping warfarin), levels of vitamin K–dependent clotting factors II, VII, and X were critically low; fibrinogen level was normal.
On day 94, duloxetine was stopped, and by day 98 INR had decreased to 1.2, with factor II increasing to 48% and factor X to 54% (Figure). On day 105, INR was 0.9. Warfarin was restarted on day 110, and by day 140 INR was stable at 2.2, with a warfarin dosage of 7.5 to 10 mg/d; it remained stable through day 160 (Figure).
The score for the Naranjo algorithm1 (used to estimate the probability of adverse drug reactions) was 12, with scores of 9 or greater highly probable for an adverse drug reaction.
Comment
The time course and high score on the Naranjo algorithm support a causal link between the concomitant use of duloxetine and warfarin and the INR elevation. While the possibility of self-intoxication with warfarin cannot be completely excluded, the rapid normalization of the INR and factors II and X after stopping duloxetine and the normal INR response to restarted warfarin from days 110 to 160 suggest that this was not the explanation.
Metabolism of warfarin involves several cytochrome P450 isoenzymes, including CYP1A2, CYP2C9, CYP2C19, and CYP3A4.2-3 Although duloxetine is metabolized by CYP2D6 and CYP1A2,4 the pharmacokinetics of theophylline (a CYP1A2 substrate) are not significantly altered by coadministration with duloxetine.5 Thus, duloxetine has not been thought to have a clinically significant effect on metabolism of CYP1A2 substrates. However, the events in this patient suggest a need to readdress whether duloxetine has more CYP1A2 effects than previously described or whether duloxetine may have unique metabolic properties that are not understood.
Duloxetine is highly protein-bound in human plasma (>90%). It could displace warfarin, which is also highly protein-bound, possibly resulting in a toxic effect. Fluvoxamine, a selective serotonin reuptake inhibitor, increases warfarin anticoagulant effect2; elevation of INR in patients whose warfarin is continued may persist for up to 2 weeks after cessation of fluvoxamine.6 In this patient, duloxetine strikingly potentiated warfarin effect, and this lasted for 36 days after warfarin was stopped. This combination of medications may need to be used with caution; patients taking both medications should have their INR monitored very closely.
Author Contributions: Dr Glueck had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: Glueck, Winiarska, Wang.
Acquisition of data: Glueck, Khalil, Winiarska, Wang.
Analysis and interpretation of data: Glueck, Khalil, Winiarska, Wang.
Drafting of the manuscript: Glueck, Winiarska, Wang.
Critical revision of the manuscript for important intellectual content: Glueck, Khalil, Winiarska, Wang.
Obtained funding: Glueck.
Administrative, technical, or material support: Glueck, Khalil, Winiarska, Wang.
Study supervision: Glueck.
Financial Disclosures: None reported.
Funding/Support: This work was supported in part by the Lipoprotein Research Fund of the Jewish Hospital of Cincinnati and by the Medical Research Council of the Jewish Hospital of Cincinnati.
Role of the Sponsor: The funding organizations had no role in the design and conduct of the study; the collection, management, analysis, and interpretation of the data; or the preparation, review, or approval of the manuscript.
Charles J. Glueck, MD
glueckch{at}healthall.com
Qasim Khalil, MD;
Magdalena Winiarska, MD;
Ping Wang, PhD
Cholesterol Center
Jewish Hospital
Cincinnati, Ohio
1. Naranjo CA, Busto U, Sellers EM, et al. A method for estimating the probability of adverse drug reactions. Clin Pharmacol Ther. 1981;30:239-245.
ISI
| PUBMED
2. Limke KK, Shelton AR, Elliott ES. Fluvoxamine interaction with warfarin. Ann Pharmacother. 2002;36:1890-1892.
ABSTRACT
3. Wadelius M, Sorlin K, Wallerman O, et al. Warfarin sensitivity related to CYP2C9, CYP3A5, ABCB1 (MDR1) and other factors. Pharmacogenomics J. 2004;4:40-48.
FULL TEXT
|
ISI
| PUBMED
4. Caccia S. Metabolism of the newest antidepressants: comparisons with related predecessors. IDrugs. 2004;7:143-150.
PUBMED
5. di Virgilio SN, Gonzales C, Knadler MP, et al. Effect of duloxetine on CYP1A2 mediated drug metabolism and the PK of theophylline [abstract]. Am Soc Clin Pharmacol Ther. 2002;71:63.
6. Yap KB, Low ST. Interaction of fluvoxamine with warfarin in an elderly woman. Singapore Med J. 1999;40:480-482.
PUBMED
Letters Section Editor: Robert M. Golub, MD, Senior Editor.
JAMA. 2006;295:1517-1518.
 CiteULike  Connotea  Del.icio.us  Digg  Reddit  Technorati  Twitter
What's this?
THIS ARTICLE HAS BEEN CITED BY OTHER ARTICLES
In Vitro Inhibition and Induction of Human Liver Cytochrome P450 Enzymes by Milnacipran
Paris et al.
Drug Metab. Dispos. 2009;37:2045-2054.
ABSTRACT
| FULL TEXT
A Case of Apical Ballooning Cardiomyopathy Associated with Duloxetine
Bergman et al.
ANN INTERN MED 2008;149:218-219.
FULL TEXT
Is there a place for {blacktriangledown}duloxetine?
DTB 2007;45:29-32.
ABSTRACT
| FULL TEXT
|
