Warfarin, Cellulitis, and a Trimethoprim-Sulfamethoxazole Interaction in Reduced Renal Function
A 74-year-old woman with atrial fibrillation (CHA2DS2-VASc 5) has taken warfarin 5 mg daily for six years, with INRs between 2.2 and 2.8 for the past 18 months. She presents with a 4-day history of right lower-leg cellulitis; an urgent-care clinician prescribes trimethoprim-sulfamethoxazole DS twice daily for 10 days. Her creatinine is 1.6 mg/dL (eGFR 30 mL/min/1.73 m²). Five days later she returns with gingival bleeding and scattered ecchymoses. INR is 6.8, hemoglobin 10.9 g/dL (baseline 12.8 g/dL), platelets 118,000/µL; she is hemodynamically stable, alert, with no melena, hematuria, or headache.
1. Explain the mechanisms by which trimethoprim-sulfamethoxazole raises the INR in this patient, and why her renal function amplifies them.
Read the worked answer
- Enzyme inhibition. Sulfamethoxazole is a potent inhibitor of CYP2C9, the principal route of clearance for S-warfarin; trimethoprim adds weaker inhibition. Warfarin accumulates and the INR rises within days.
- Protein-binding displacement. Warfarin is more than 98% albumin-bound and sulfonamides are also highly bound; displacement raises the free fraction - a transient but real effect.
- Loss of vitamin K-producing gut flora. Suppression of menaquinone synthesis matters more when dietary intake is poor, as it often is in an older patient with an acute infection.
- Renal function. With an eGFR near 30, trimethoprim and sulfamethoxazole accumulate (trimethoprim also raises measured creatinine by blocking tubular secretion), so the inhibitory effect is amplified. The acute-phase response to infection itself suppresses synthesis of vitamin K-dependent factors.
The thrombocytopenia is a second concern: trimethoprim causes folate antagonism and marrow suppression, which compounds the bleeding risk beyond the INR alone.
2. She is hemodynamically stable with only mucosal and skin bleeding at an INR of 6.8. What is the appropriate immediate management, and what determines whether vitamin K is given?
Read the worked answer
- Stop the warfarin and stop the trimethoprim-sulfamethoxazole.
- For an INR between 4.5 and 10 without major bleeding, guidelines advise omitting warfarin with no routine vitamin K. Low-dose oral vitamin K (1–2.5 mg) is reasonable when there is added bleeding risk, and gingival bleeding plus new ecchymoses plus a falling hemoglobin places her toward that end of the spectrum.
- Do not give fresh frozen plasma or prothrombin complex concentrate; those are reserved for major bleeding or an urgent procedure.
- Recheck the INR in 24 hours, and look for bleeding that is not visible: the hemoglobin has dropped roughly 2 g/dL, so a stool guaiac, urinalysis, and a careful neurologic examination are warranted.
- Restart anticoagulation once the INR is in range and the precipitant has been removed, with a plan that accounts for the fact that she remains at high stroke risk.
The key judgement is that a supratherapeutic INR is not itself the emergency - bleeding is. Reversal is titrated to the bleeding, not to the number.
3. Choose a replacement antibiotic strategy and describe the monitoring plan, including any longer-term anticoagulation reconsideration.
Read the worked answer
- Pick an agent without CYP2C9 inhibition and dose it for an eGFR of about 30. Cephalexin is a reasonable choice for non-purulent cellulitis (no relevant CYP interaction; adjust dosing for reduced clearance).
- If MRSA is a concern (purulence, prior MRSA), doxycycline is often used but also perturbs the INR; linezolid is another option with its own monitoring requirements. There is no free lunch - any change needs follow-up.
- Check the INR at roughly 3 and 7 days after starting any new antibiotic and again after it is finished, because the interaction lasts as long as the drug is on board.
- This admission is the moment to ask whether warfarin is still the right anticoagulant for her. A direct oral anticoagulant with appropriate renal dosing would avoid the same cascade, provided she is a candidate and cost and adherence are addressed.
General principle: the antibiotic prescribed for a minor skin infection is the intervention most likely to cause a major bleed in this patient, so the drug choice and the monitoring interval are the prescription, not an afterthought.