International Journal of Arrhythmia 2012;13(4): 34-36.
Untitled Document
ECG & EP CASES
Gastrointestinal Bleeding Caused by Complications in a Patient Receiving Warfarin Therapy
가톨릭대학교 의과대학 내과학교실 신 우 승
Woo-Seung Shin, MD, PhD Cardiology Division, Department of Internal Medicine, Catholic University College of Medicine, Seoul, Korea
Introduction
Anticoagulation therapy is effective for the
prevention and treatment of both arterial and
venous thromboembolism in patients with atrial
fibrillation (AF). This therapy produces bleeding
complications in an approximately dose-dependent
fashion; the risk of bleeding intensifies on
increasing doses or using multiple antithrombotic
agents. The risk of bleeding varies both with the anticoagulant type and patient characteristics. We
report a case of gastrointestinal bleeding caused by
complications in a patient undergoing warfarin
therapy.
Case
A 76-year-old woman presented to the
emergency department after 3~4 instances of
passing melena stools. She had been experiencing
epigastric discomfort that she found difficult to
describe; it was episodic in nature and mild in
intensity and there were no provocative or
palliative factors. She felt dizzy while standing and
fell to the ground, but did not lose consciousness. She was transported to the hospital by ambulance.
In 2006, she had developed atrial fibrillation and
cerebral infarction for which she was being treated
with warfarin. Her current medications included
low-dose aspirin (100 mg orally once daily),
extended-release diltiazem (180 mg orally once
daily), valsartan (80 mg orally once daily), and
warfarin (3.5 mg orally once daily). In the
emergency department, she was found to be
diaphoretic with an irregular pulse of 141 bpm and a
blood pressure of 80/50 mmHg. Abdominal
examination showed no abnormalities, but rectal
examination showed melena. Her electrocardiogram
showed atrial fibrillation with a heart rate of 141
bpm (Figure 1).
Two intravenous accesses were established for
the patient and she received crystalloids and was
observed in a monitored setting. Her laboratory
tests showed a hemoglobin level of 7.6 g/dL
(hemoglobin level at 1 month prior to presentation,
13.7 g/dL), a white blood cell count of 9,000/μL, and
a platelet count of 151×103/μL. Her prothrombin
time international normalized ratio (INR) was 8.04
and her urea level was 59 mg/dL; the electrolyte,
creatinine, and liver enzyme levels were normal.
She was transfused with 2 units of packed red
blood cells, given 5 mg of vitamin K subcutaneously
and was transfused with 1,000 U of prothrombin complex concentrate (a mixture of clotting factors
II, VII, IX, and X, and proteins C and S). After 12
hours, she was hemodynamically stable and
underwent a gastroduodenoscopy under conscious
sedation. Several variably sized gastric ulcers were
noted at the gastric antrum and pylorus (Figure 2).
Her hemoglobin level increased to 10.0 g/dL and her
INR was 1.4.
She was hospitalized for 7 days, during which she
received intravenous proton pump inhibitor (PPI)
therapy for 72 hours; thereafter, oral PPI therapy
was initiated. In addition, warfarin therapy without
aspirin was resumed.
Interruption of warfarin therapy for colonoscopy
had been recommended to the patient and the event
occurred 3 days after this interruption of therapy.
The patient began eating 24 hours after the
endoscopy and was discharged after 7 days.
Discussion
This patient had several risk factors for stroke
and the use of warfarin for long-term primary
prophylaxis of stroke would be strongly recommended
by consensus guidelines. However, the patient
showed evidence of acute bleeding (in the form of
melena). It is probable that a gastrointestinal ulcer
or malignancy was the cause of her presentation.
The risk for thromboembolic complications and
bleeding in AF varies in different medical
conditions.1 The CHADS2 score is a simple validated measure of risk. Not much data is available on
survival and bleeding rates in relation to the
CHADS2 risk score. Thus, the threshold at which
the benefit of oral anticoagulation for the
prevention of stroke in AF exceeds the risk for
bleeding is unclear. There are several other stroke
prediction measures, including the CHA2DS2-VASc
score, but they all have only a modest discrimination
ability for individual patients and are not very
different from the CHADS2 risk score in this
respect. International AF guidelines uniformly
recommend the use of warfarin for patients with a
CHADS2 score of 2 or higher. Warfarin is largely
underused because of concerns over the need for
systematic monitoring and the risk of bleeding
associated with its use.2 There is consequently a
need for new agents that can function as
alternatives to warfarin for long-term use in AF.
Dabigatran and rivaroxaban have been recently approved by the United States Food and Drug
Administration for stroke prevention in patients
with AF. Recent meta-analysis demonstrated that
the new oral anticoagulants lower the risk for
intracranial bleeding and may decrease the overall
risk for major bleeding events in patients with AF.3-5
Increasing CHADS2 scores are associated with
increased risk for stroke or systemic embolism,
major and intracranial bleeding, and death in
patients with AF who are treated with warfarin.
With respect to major bleeding complications, new
oral agents may be considered as alternatives in patients with AF.
Flaker GC, Reilly P, Yusuf S, Ezekowitz M, Wallentin L,
Brueckmann M, Connolly, S. Dabigatran etexilate versus warfarin
in patients with different types of atrial fibrillation: a RE-LY
subgroup analysis.
J Am Coll Cardiol.
2011;57:E62.