The Perils of Mixing Warfarin & Antibiotics: A Potentially Deadly Combination Lynn McNicoll, MD, FRCPC, AGSF Associate Professor of Medicine, Department of Medicine Warren Alpert Medical School of Brown University Consultant, Healthcentric Advisors Director of Education, Division of Geriatrics and Palliative Medicine, Rhode Island Hospital Disclosures none 1
Objectives At the conclusion of this session, participants will be able to: 1. List common complications of warfarin management when antibiotics are started 2. Describe management strategies for preventing complications from combining warfarin and antibiotics Warfarin Vitamin K antagonist decrease in vitamin K dependent clotting factors (II (prothrombin), VII, IX, X) Rat poison Many drug, diet, comorbidity, and genetic variables Narrow therapeutic window 97% Protein bound (mostly to albumen) and thus vulnerable to illness changes with albumen dropping as inflammatory marker 2
Warfarin Effectiveness Benchmarking rates of success are around 65% (meaning 65% of INR are within range of 2 3) 1/3 of time patients are out of range (too high or too low) Antibiotics are a huge factor in this variability as well as procedures (which often incur use of antibiotics) 3
Warfarin in the Nursing Home Warfarin is the top reason for ADE in the nursing homes as well as preventable ADE 20% of ADE are Hemorrhagic and 16% of preventable are hemorrhagic Warfarin and Antibiotics Common mechanisms of action for disruption A. Disruption of intestinal microflora B. Inhibition of hepatic CYP 2C9 (also 3A4 and 2C19) C. Changes in diet The infection itself can also cause problems such as reduced oral intake, diarrhea, altering metabolism 4
Warfarin and Antibiotics Most agents potentiate the effect of warfarin increasing the risk of bleeding complications Some agents inhibit the effect of warfarin increasing the risk of clotting Even topical agents are not completely safe INR changes usually occur within 3 days so checking an INR 5 or 7 days later will not help Basic Management Principles No clear guidelines available to follow Try to avoid antibiotics that are worst offenders If not, measuring INR in 2 3 days (especially if pt has not had a recent INR) Consider cutting dose of warfarin while on antibiotic especially if patient tends to have labile INR Concurrent use of tylenol, prednisone or tramadol, antiulcer (H2 Blockers/PPI) be extra careful if these are being added simultaneously If dose is increased or additional antibiotic is started, check INR again within a couple of days Don t forget to monitor closely once the antibiotic is stopped as well When in doubt, look it up (Epocrates or Lexicomp) 5
Potentiation 6
Inhibition 7
Methods Case control nested within a cohort 38,762 patients aged >65 on continuous warfarin Cases patients hospitalized for a primary diagnosis of bleeding Matched controls each case matched with 3 controls by age, race, sex, and indication for warfarin Matched controls had fewer comorbidities and less likely to be in a nursing home 8
Associated with all types of bleeding Recent exposure associated with highest risk Concurrent use of other medications increases risk Azole antifungals carry highest risk Followed by cotrimoxazole cephalosporins penicillins macrolides quinolones 9
Methods Retrospective cohort study in a VA outpatient clinic Patients aged >65 on stable warfarin given 1 prescription of antibiotic 2003 2011 205 patients studied 10
Times 1 and 2 were pre antibiotic, times 3 and 4 were postantibiotic. No significant difference by antibiotic Patients > 80 were more likely to have increase in INR compared to subjects 65 70 11
Methods Retrospective study of 22,272 veterans on warfarin between 2002 and 2008 High risk antibiotics included sulfamethoxazole, ciprofloxacin, levofloxacin, metronidazole, fluconazole, azithromycin, clarithromycin Low risk clindamycin, cephalexin Compared bleeding risk for patients receiving high risk versus low risk antibiotics 12
13
Factors associated with increased risk of bleeding Renal failure Heart valve indication High risk antibiotic especially azithromycin independently associated (others not significant) but also most prescribed antibiotic by far 25.9% 14
Methods Retrospective cohort study of ambulatory patients from 2005 to 2011 at Kaiser Permanente in Colorado Compared INR changes for those on warfarin Stratified into 3 groups 1. URTI on antibiotic 2. sick URTI but no antibiotic 3. stable not sick) Primary outcome is mean change in INR and rate of INR > 5 15
Those on antibiotics were older, female, > afib indication, > CV disease, more likely to be in range at baseline 16
Median of 6 days between index date and follow up INR for antibiotic group, less for sick group (4) Sig difference in mean INR change (0.08) Sig higher rate of INR >5 (3.2% vs 2.6% in sick and 1.2% in stable) Antibiotic and sick group had sig higher rate of INR > 3.5 No sig differences in sig outcomes (thrombosis, bleeding, mortality) 17
18
Antibiotics to avoid Azole antifungals Worst offender Clotrimoxazole Metronidazole Macrolides (erythromycin) Fluoroquinolones (ciprofloxacin) Safer Antibiotics = No or Little Interaction Reported Clindamycin Penicillin/Amoxicillin Cephalexin Nitrofurantoin Trimethoprim 19
Conclusions Be very vigilant especially if baseline INR >2.5, ill, high comorbidity burden Monitor within 3 days of starting, increasing, or stopping warfarin Implement systematic protocol for monitoring within your system Use lower risk antibiotics whenever possible 20