Abnormal uterine bleeding in anticoagulated patients by drug class: outcomes and management

Abstract

Background: Although abnormal uterine bleeding is a known adverse effect of anticoagulant drugs, true rates have not been widely studied. Society-backed recommendations and guidelines do not yet exist for prevention and management of abnormal uterine bleeding among anticoagulated patients.

Objective: This study aimed to describe the incidence of new-onset abnormal uterine bleeding among patients receiving therapeutic anticoagulation by anticoagulant class, and to evaluate gynecologic treatment patterns.

Study design: We conducted an institutional review board-waived retrospective chart review of female patients aged 18 to 55 years and prescribed therapeutic anticoagulants, including vitamin-K antagonists, low-molecular-weight heparins, and direct oral anticoagulants, in an urban hospital network from January 2015 through January 2020. We excluded patients with antecedent abnormal uterine bleeding and menopause. Associations between abnormal uterine bleeding, anticoagulant class, and other covariates were evaluated with Pearson chi-square and analysis-of-variance tests. The primary outcome, abnormal uterine bleeding odds by anticoagulant class, was modeled with logistic regression. Age, antiplatelet therapy, body mass index, and race were included in our multivariable model. Secondary outcomes included emergency department visits and treatment patterns.

Results: Of the 2479 patients who met the inclusion criteria, 645 were diagnosed with abnormal uterine bleeding after initiating therapeutic anticoagulation. After adjusting for age, race, body mass index, and concurrent use of antiplatelet therapy, those receiving all 3 classes of anticoagulants had higher odds of experiencing abnormal uterine bleeding (adjusted odds ratio, 2.63; confidence interval, 1.70-4.08; P<.001), whereas those taking only direct oral anticoagulants had the lowest odds (adjusted odds ratio, 0.70; confidence interval, 0.51-0.97; P=.032), with vitamin-K antagonists as the reference group. Race other than White was associated with higher odds of abnormal uterine bleeding, as was lower age. The most common hormone therapies used among patients with abnormal uterine bleeding were levonorgestrel intrauterine devices (7.6%; 49/645) and oral progestins (7.6%; 49/645). Sixty-eight patients (10.5%; 68/645) had an emergency department visit for abnormal uterine bleeding; 29.5% (190/645) of patients received a blood transfusion; 12.2% (79/645) began any pharmacologic therapy for bleeding; and 18.8% (121/645) underwent any gynecologic procedure.

Conclusion: Abnormal uterine bleeding occurs frequently among patients on therapeutic anticoagulation. Incidence in this sample varied considerably by anticoagulant class and race; use of single-agent direct oral anticoagulation carried the lowest risk. Important sequelae such as bleeding-related emergency department visits, blood transfusions, and gynecologic procedures were common. Balancing bleeding and clotting risk in patients on therapeutic anticoagulation requires a nuanced approach and should involve collaborative management between hematologists and gynecologists.

Keywords: direct oral anticoagulant; gonadotropin-releasing hormone agonist; heavy vaginal bleeding; hormonal therapy; intermenstrual bleeding; low-molecular-weight heparin; reproductive tract bleeding; therapeutic anticoagulation; uterine artery embolization; vitamin-K antagonist.