Background: Warfarin frequently causes increased healthcare utilization for major bleeding. Warfarin’s anticoagulant effect is measured by the international normalized ratio (INR). Elevated INRs are associated with an increased risk of bleeding. Currently, consensus guidelines based on low-quality evidence suggest treating patients who are not bleeding and have an INR ≥10 with oral vitamin K. In contrast, recommendations for patients with elevated INRs from 4.5-10 are to simply hold warfarin.
Aims: Assess the association between temporary discontinuation of warfarin with or without any over-the-counter or dietary vitamin K (conservative therapy) versus prescription vitamin K (vitamin K) and bleeding (ISTH definitions for major and clinically relevant non-major), any arterial or venous thromboembolism (TE), and all-cause mortality at 30 days after initial INR ≥10, and time to INR < 4.0.
Methods: This was a multi-center observational cohort study. Data were pooled using multivariable random-effects modeling for outcome analysis.
Results: Across four sites, 563 and 705 patients comprised the conservative and vitamin K groups, respectively (Table). There were 53 (9.4%) and 52 (7.4%) bleeding events in the conservative and vitamin K groups, respectively. TE events occurred in 4 (0.8%) and 11 (1.6%) of conservative and vitamin K group patients, respectively. Unadjusted odds ratios (with 95% confidence intervals) comparing conservative therapy to vitamin K at 30 days were (Figure): all bleeding (OR: 1.22 [0.77-1.92]), major bleeding (OR: 1.07 [0.55-2.09]), TE (OR: 0.45, [0.14-1.45]), and all-cause mortality (0.67 [0.46-0.97]). The mean difference in time to INR <4.0 was 0.77 days [95% CI 0.03-1.52, p=0.04] favoring vitamin K.
| University of Utah (n= 121) | University of Michigan (n=71) | Intermountain Healthcare (n=272) | Kaiser Permanente Colorado (n=809) | |||||
| VitK n=101 |
CT n=14 |
VitK n=48 |
CT n=21 |
Vit K n=224 |
CT n=48 |
VitK n=332 | CT n=477 |
|
| Mean Age, years (SD) | 62 (16) | 56 (12) | 58 (17) | 53 (15) | 74 (13) | 70 (14) | 72 (15) | 72 (15) |
| Male | 45% | 47% | 52% | 48% | 45% | 44% | 63% | 39% |
| White | 85% | 82% | 74% | 86% | 93% | 98% | 72% | 76% |
| >1 bleeding RF | 74% | 71% | 80% | 66% | 64% | 44% | n/a | n/a |
| Time to INR <4.0 mean days, (SD) | 2.5 (2.6) | 3.4 (2.3) | 4.5 (3.8) | 3.4 (2.7) | 3.2 (4.1) | 4.7 (3.1) | 2.2 (1.8) | 2.4 (2.3) |
| Vitk = Vitamin K, CT = Conservative Therapy, RF = risk factor, n/a = not available | ||||||||
Characteristics of Non-bleeding Patients with INRs >10 at each site stratified by treatment (Vitamin K versus Conservative Therapy)
Forest Plots for All Bleeding, Venous Thromboembolism and Mortality at 30 days, and Mean Difference to INR <4.0. Sites include University of Michigan (UM), University of Utah (UofU), Intermountain Healthcare (IHC), Kaiser Permanente Colorado (KPCO)
Conclusions: Compared to vitamin K, conservative therapy is associated with lower mortality and no differences in bleeding and TE and is thus a reasonable strategy for asymptomatic patients presenting with INRs ≥10. The difference in time to achieve an INR < 4.0 was statistically but not clinically different between groups.
To cite this abstract in AMA style:
Jones A, Vazquez S, Barnes G, Anderson C, Woller S, Stevens S, Clark N, Delate T, Crowthrm M, Witt D. Outcomes of Non-bleeding Patients on Warfarin with an INR > 10, who Received Vitamin K or Conservative Therapy [abstract]. Res Pract Thromb Haemost. 2021; 5 (Suppl 2). https://abstracts.isth.org/abstract/outcomes-of-non-bleeding-patients-on-warfarin-with-an-inr-10-who-received-vitamin-k-or-conservative-therapy/. Accessed October 2, 2023.« Back to ISTH 2021 Congress
ISTH Congress Abstracts - https://abstracts.isth.org/abstract/outcomes-of-non-bleeding-patients-on-warfarin-with-an-inr-10-who-received-vitamin-k-or-conservative-therapy/