Elsevier

Thrombosis Research

Volume 144, August 2016, Pages 21-26
Thrombosis Research

Full Length Article
The impact of new-onset cancer among veterans who are receiving warfarin for atrial fibrillation and venous thromboembolism

https://doi.org/10.1016/j.thromres.2016.05.028Get rights and content

Highlights

  • We studied a large sample of veterans receiving warfarin and diagnosed with new onset of cancer.

  • Subjects experienced significantly worse anticoagulation control after cancer onset.

  • Outcomes and warfarin control were poorest in the first 6 months after cancer onset.

  • Major bleeding and stroke rates were similar regardless of indication for warfarin.

Abstract

Background

A new cancer diagnosis adds significant complexity and uncertainty to the management of pre-existing warfarin therapy.

Objectives

To determine how new-onset cancer affects anticoagulation control and outcomes among patients who had been receiving warfarin for atrial fibrillation (AF) compared to patients who had been receiving warfarin for venous thromboembolism (VTE) prior to cancer diagnosis.

Patients/methods

This cohort study started with 122,875 veterans who had been receiving warfarin for at least six months from a VA Medical Center between 10/1/06 and 9/30/08. We identified patients with incident cancer during this interval, and excluded those with a prior cancer history. We analyzed percent time in therapeutic range (TTR) at 6 and 12-month intervals after cancer diagnosis compared to pre-cancer baseline, as well as crude rates of warfarin-relevant outcomes (stroke, major bleeding, mortality) between patients with AF and VTE.

Results

Among patients with new-onset cancer, patients anticoagulated for AF outnumbered those anticoagulated for VTE more than 2.5-fold. There were no significant differences in TTR by indication for warfarin in months 0–6 or 7–12 following cancer diagnosis, but TTR decreased significantly compared to the pre-cancer baseline for both groups in months 0–6. As expected, cancer patients with VTE had significantly worse mortality at six months and one year compared to cancer patients with AF.

Conclusion

Patients receiving chronic warfarin therapy who are newly diagnosed with cancer experience a significant decrease in TTR in the first 6 months after diagnosis, regardless of indication for anticoagulation. This effect appears to attenuate in months 7–12.

Section snippets

Background

Warfarin therapy remains a mainstay of treatment for both atrial fibrillation (AF) and venous thromboembolism (VTE) for thousands of Americans as a means to reduce incidence of embolic stroke and recurrent VTE, respectively [1]. However, a new cancer diagnosis adds complexity and uncertainty to the management of anticoagulation for these patients. Previous work by our group has shown that patients who receive warfarin for any reason who are diagnosed with cancer will have worse anticoagulation

Patient sample

All procedures were approved by the Bedford Veterans' Administration (VA) Institutional Review Board. This was a retrospective cohort study. We obtained encounter, demographic, laboratory and pharmacy fill data from the VA Clinical Data Warehouse and Medicare for 122,875 veterans who had ever received warfarin in a VA anticoagulation clinic from October 1, 2006 through September 30, 2008 (Fig. 1). Patients were deemed to have been receiving warfarin 1) while they were in possession of warfarin,

Anticoagulation control by indication for therapy

From the original 122,875 patients in our sample, 47,071 patients met inclusion criteria for warfarin use during the study period. After excluding 181 patients who were receiving warfarin for an indication other than AF or VTE, we had a sample of 46,890 patients who constituted our base analytic sample (Fig. 1). This sample of patients was 98.5% male, all over the age of 65, and 89.6% white race. Demographic profiles of patients with incident cancer were similar for each indication, except for

Discussion

In this sample of veterans who were receiving chronic warfarin therapy for either AF or VTE and were diagnosed with new-onset cancer, we have shown that the new cancer diagnosis significantly worsens warfarin control in the six-month period following diagnosis compared to the cancer-free cohort, regardless of indication for anticoagulation. This finding is especially pertinent when considering the lack of available clinical guidelines in the management of AF following cancer diagnosis [6], as

Conclusion

Among patients receiving warfarin who develop a new-onset cancer diagnosis, patients experience poorer anticoagulation control during the first six months after a new cancer diagnosis, a phenomenon which is similar between patients anticoagulated for AF or VTE, and which does not persist when examined 7–12 months after cancer onset. Given that patients anticoagulated for AF are more than 2.5 times as numerous as patients anticoagulated for VTE, future research should specifically address whether

Authorship addendum

D.B. Ambrus, MD, MSc: Primary role in conceptualization of the research question, guidance of data analysis and drafting of the manuscript.

J.I. Reisman, BA: Primary role in obtaining the data set, primary role in performing and guiding data analysis, and critical writing/revision of intellectual content.

A.J. Rose, MD, MSc: Role in conceptualization of the research question, design and guidance of data analysis, and critical writing/revision of intellectual content, obtaining funding, and study

References (18)

There are more references available in the full text version of this article.

Cited by (18)

  • Stroke risk assessment and atrial fibrillation (AF) in cancer patients

    2022, Cardiovascular Toxicity and Therapeutic Modalities Targeting Cardio-oncology: From Basic Research to Advanced Study
  • Association of anticoagulant-related bleeding events with cancer detection in atrial fibrillation: A systematic review and meta-analysis

    2021, Hellenic Journal of Cardiology
    Citation Excerpt :

    Michaels in 1962 was the first who noted the above association by presenting 3 case reports.22 The reported incidence of cancer in cohorts of patients under an NOAC or a VKA ranged between 1.2%-4.6%19,21 and 0.5%-5.7%,15,23 respectively. Warfarin is less often associated with GI bleeding than NOACs,24 because of the incomplete absorption of the NOACs as compared to warfarin across the GI mucosa and thus the potential for topical drug activity.25

  • Atrial fibrillation and cancer – An unexplored field in cardiovascular oncology

    2019, Blood Reviews
    Citation Excerpt :

    Due to the possibility of drug-drug interaction between anticoagulant treatment and various chemotherapies, certain vigilance is required as inadequate dosing and suboptimal therapy could result in an increased risk for thromboembolic and bleeding complications. In a cohort study with 122,875 veterans, it was shown that patients under warfarin therapy, who were newly diagnosed with cancer, experience a significant decrease in time in therapeutic range (TTR) in the first 6 months after diagnosis [75]. In a nested-case control study in a primary health care database, cancer was found as one of the strongest predictors of supra-therapeutic anticoagulation (INR ≥ 4) [76].

  • Atrial fibrillation in patients with active malignancy and use of anticoagulants: Under-prescription but no adverse impact on all-cause mortality

    2019, European Journal of Internal Medicine
    Citation Excerpt :

    This finding was observed in the whole population, as well as in analysis of subgroups. Our results are of clinical interest, even with the limitations of their observational nature, since according to published literature, patients with cancer taking warfarin have an increased risk (up to twofold) of major and fatal bleeding, as compared to non-cancer patients taking warfarin [40]. Therefore a potential impact of anticoagulation on the risk of bleeding and death is often assumed.

View all citing articles on Scopus

This study was supported by funding from the Veterans' Affairs Health Services Research and Development grant IIR-10-374. This article does not necessarily represent the official views or policies of the United States Department of Veterans Affairs.

View full text