To the Editor We read with great interest the recent article by Clark et al.1 The use of sick controls was innovative and valuable to future studies evaluating international normalized ratio (INR) fluctuations in patients receiving long-term warfarin therapy. The authors did not identify a significantly increased proportion of patients with an INR above 5.0 among those receiving antimicrobials compared with sick controls. However, an analysis of their published data suggests that the difference observed between those receiving antimicrobials who inhibit cytochrome P450 metabolism, particularly the 2C9 enzyme (ie, metronidazole and trimethoprim-sulfamethoxazole) and sick controls was significant (8.6% vs 2.6%; P < .001), suggesting that although antimicrobials as a whole do not cause elevations in INR compared with sick controls, a select subset of highly potentiating antimicrobials will demonstrate a significant effect. We would be interested to know if the authors were able to perform this subgroup analysis and if other highly potentiating antimicrobials (eg, fluconazole, miconazole, voriconazole) were captured for analysis. Other studies have suggested that highly potentiating antimicrobials have a significant impact on INR.2- 5 If it appears that these highly potentiating antimicrobials contribute to a substantially increased INR in patients taking warfarin for chronic anticoagulation, then interventions in the ambulatory care setting are needed to reduce this risk in these patients.
Daniels LM, Barreto JN, Tosh PK. Supratherapeutic International Normalized Ratios in Warfarin-Treated Patients Who Receive a Highly Potentiating Antimicrobial Course. JAMA Intern Med. 2014;174(7):1200. doi:10.1001/jamainternmed.2014.1615