Elevated INRs, Elevated Insight: A Data-Driven Approach to Warfarin Use in Cardiothoracic Surgery

Introduction: Warfarin is a commonly used oral anticoagulant with a narrow therapeutic index and risk for supratherapeutic anticoagulation. In October 2017, Vizient implemented a lab-based metric tracking the percentage of adult inpatients receiving warfarin who develop an INR ≥ 5. Exclusions include therapy < 3 days, admission INR >3.5, and argatroban use. Though not tied to bleeding outcomes, the metric serves as a surrogate for anticoagulation safety. In FY24, 7% (24/343) of UK HealthCare inpatients on warfarin met the INR ≥ 5 threshold, with 50% of cases in cardiothoracic (CT) surgery patients—prompting investigation into patient-specific risk factors and dosing approaches.


Methods: This retrospective cohort quality improvement study identified CT surgery patients treated with warfarin in FY24 using EPIC SlicerDicer. Patients were stratified by INR outcome (≥5 vs. < 5). Endpoints included warfarin indication, valve type/location, redo sternotomy, cardiopulmonary bypass time, time to warfarin initiation, time to therapeutic INR, initial dose, INR trends, and hospital length of stay (LOS).


Results: Among 59 CT surgery patients, 11 (18.6%) developed INR ≥ 5. These patients received warfarin earlier (mean 2.7 vs. 4.3 days, p=0.03) and reached therapeutic INR faster (mean 1.3 vs. 2.6 days, p=0.04). Time from INR > 2 to discharge was longer in the INR ≥ 5 group (median 13 vs. 6 days, p=0.02). Starting dose did not predict INR ≥ 5 (p=0.41), and supratherapeutic INR occurred even with lower doses (2–2.5 mg). Initial dose also did not impact LOS or time to therapeutic INR (p>0.05). Valve type, bypass time, and number of doses were not significantly associated with INR ≥ 5 (all p>0.3), though mitral valve patients with INR ≥ 5 had the longest LOS (47 days).


Conclusions: Traditional risk factors for warfarin sensitivity—such as valve location, bypass time, and starting dose—did not predict INR ≥ 5. These findings suggest a standardized CT surgery-specific warfarin dosing protocol may be suitable across the broader post-op population, potentially improving anticoagulation safety and Vizient performance metrics.