Bleeding is the major concern for patients receiving oral anticoagulation therapy (OAT), especially those with histories of gastrointestinal ulcer. The aim of this study was to evaluate the efficacy and safety of OAT in patients with nonvalvular atrial fibrillation with histories of ulcer. A composite end point, including major adverse cardiac events or major bleeding, was compared between patients with AF with previous ulcers who were (OAT+; n = 200) and were not (OAT− n = 230) receiving OAT. During the follow-up period of 3.3 ± 2.7 years, 28 (14%) and 66 (29%) OAT+ and OAT− patients, respectively, had major adverse cardiac events (p = 0.001). Major bleeding occurred in 46 OAT+ patients (23%) and 25 OAT− patients (11%) (p = 0.001). There was no significant difference in the composite end point between OAT+ and OAT− patients (29% vs 36%, p = 0.08). The incidence of major bleeding was significantly lower, decreasing from 30% to 14%, when OAT began after endoscopic confirmation of ulcer healing (p = 0.02). OAT+ patients who achieved time in the therapeutic range ≥60% for international normalized ratio (2.0 to 3.0) demonstrated better cumulative survival free from the composite end point than OAT− patients (p = 0.01). In conclusion, OAT in patients with nonvalvular AF with histories of gastrointestinal ulcer made no difference in the composite end point compared to absence of OAT. In OAT+ patients, maintaining an optimal international normalized ratio reduced the composite end point, and the confirmation of ulcer healing reduced the incidence of bleeding.
Peptic ulcer disease is common, affecting 4 million patients in United States. Mortality from gastrointestinal (GI) bleeding ranges from 1% to 13%. In patients with nonvalvular atrial fibrillation (AF) with histories of GI ulcer disease, oral anticoagulation therapy (OAT) results in elevated bleeding risk. However, the risk for bleeding after OAT and optimal OAT in patients with AF with previous ulcers are poorly understood, because these patients have usually been excluded from studies of OAT. We hypothesized that OAT in patients with AF with previous GI ulcers would result in an increase in major bleeding and a decrease in thromboembolic events. Moreover, the increased bleeding risk might be outweighed by a decrease in thromboembolic events in patients who are maintained in the optimal international normalized ratio (INR) range. The aim of this study was to evaluate the clinical courses, including thromboembolic and bleeding events, of patients with AF with histories of GI ulcer according to whether or not they received OAT. We also evaluated the clinical outcomes of patients according to the quality of anticoagulation, represented as time in the therapeutic range (TTR) of INR.
Methods
The study protocol was approved by the institutional review board of Severance Cardiovascular Hospital, Seoul, Korea, and complied with the Declaration of Helsinki. From January 2001 to July 2011, using International Classification of Diseases, Ninth Revision, codes, we identified 810 consecutive patients with nonvalvular AF and GI ulcer disease. Patients with chronic liver disease (n = 162), thrombocytopenia (n = 57), endoscopic findings with malignancy, Mallory-Weiss tear, angiodysplasia or Dieulafoy’s lesion (n = 56), previous intracerebral hemorrhages (n = 12), and insufficient clinical data (n = 93) were excluded. Ultimately, we enrolled 430 patients in this study. AF was documented by 12-lead electrocardiography or 24-hour Holter recording. Peptic ulcer disease was confirmed by esophagogastroduodenoscopy.
Patients’ medical records were reviewed for information on age, gender, weight, height, drug therapy, and AF duration. Patient databases were searched to identify known or putative risk factors for ischemic stroke, heart failure, hypertension, diabetes mellitus, hyperlipidemia, tobacco use, and coronary heart disease. Heart failure was defined when hospitalized patients had appropriate symptoms (shortness of breath, fatigue, fluid retention, or any combination of these symptoms) and clinical signs of fluid retention (pulmonary or peripheral) with explainable abnormality of cardiac structure and function. The CHADS 2 (congestive heart failure, hypertension, age ≥75 years, diabetes mellitus, and previous stroke or transient ischemic attack [doubled]) risk index, CHA 2 DS 2 -VASc (congestive heart failure, hypertension, age ≥75 years [doubled], diabetes mellitus, previous stroke or transient ischemic attack [doubled], vascular disease, age 65 to 74 years, and sex category [female]) risk index, and the HAS-BLED (hypertension, abnormal renal or liver function, stroke, bleeding history or predisposition, labile INR, elderly [age >65 years], and drug and alcohol use concomitantly) score were evaluated. We divided patients into 2 groups, those receiving OAT (OAT+) and those not receiving OAT (OAT−).
We evaluated major adverse cardiac events (MACE) and major bleeding events during follow-up. MACEs included ischemic stroke, myocardial infarction, pulmonary thromboembolism, or other systemic embolism. Ischemic stroke was defined as a neurologic deficit of sudden onset that persisted for >24 hours corresponding to a vascular territory in the absence of primary hemorrhage and was not explained by other causes, including trauma, infection, or vasculitis. Myocardial infarction was diagnosed according to the 2007 universal definition of the European Society of Cardiology, American College of Cardiology Foundation, American Heart Association, and World Heart Federation. Pulmonary embolism was diagnosed if there were ≥2 segmental defects without a ventilation defect on a ventilation-perfusion scan, positive angiographic findings, or documented evidence on computed tomography or magnetic resonance imaging of the chest. Other systemic embolism was defined as the appearance of acute ischemia, documented by angiography or surgery in the absence of atherosclerotic occlusion. Major bleeding was defined as any central nervous system bleeding, including intracranial hemorrhage, subdural hemorrhage, subarachnoid hemorrhage, epidural hemorrhage, any bleeding requiring transfusion of ≥2 U red blood cells or the equivalent of whole blood in 24 hours, or bleeding events that cause hypotension (systolic blood pressure <90 mm Hg), multiple-organ failure, or death. When a patient experienced MACEs and major bleeding events during the follow-up period, each event was counted respectively. However, when analyzing the Kaplan-Meier cumulative event-free survival of the composite end point, we counted the first event only.
The intensity of anticoagulation was determined according to INRs at each visit to the outpatient clinic and each emergency department visit and during hospital admission. Data on the first 4 weeks after the initiation of OAT were excluded from the analysis. All INR tests were performed in the same laboratory using the same reagents. To calculate TTR, INRs between the actual tests were estimated using the linear interpolation method as proposed by Rosendaal et al. This method assumes that the INRs between 2 consecutive measurements vary linearly. Patients with intervals ≥56 days between INR tests were excluded from the analysis.
Continuous variables that were normally distributed are reported as mean ± SD and were compared using Student’s t tests for parametric data and Mann-Whitney tests for nonparametric data. Categorical variables are reported as counts and percentages and were compared using chi-square or Fisher’s exact tests. We used Cox proportional-hazards modeling to compute hazard ratios (HRs) and 95% confidence intervals (CIs) and in a sensitivity analysis adjusted for baseline covariates including age, hypertension, congestive heart failure, stroke, and concomitant aspirin use that showed a difference between the study groups. Kaplan-Meier survival curves were plotted for the OAT+ and OAT− groups and compared using log-rank tests. SPSS version 18.0 (SPSS, Inc., Chicago, Illinois) was used to perform all statistical evaluations. A p value <0.05 was considered statistically significant.
Results
The comparison between excluded and included patients is listed in the supplementary table . Clinical characteristics of OAT+ and OAT− patients are listed in Table 1 . On average, OAT+ patients were younger than OAT− patients (p = 0.01). The prevalence rates of hypertension (p = 0.03), congestive heart failure (p = 0.008), and history of stroke (p = 0.007) were higher in OAT+ patients than in OAT− patients. However, there was no difference in CHADS 2 score (p = 0.42) or CHA 2 DS 2 -VASc score (p = 0.62) between the 2 groups. The HAS-BLED score was higher in OAT+ patients than in OAT− patients (p = 0.01). In OAT+ patients, the mean INR was 2.1 ± 0.5, and the mean TTR, assuming an optimal INR of 2.0 to 3.0, was 48.5 ± 17.8%. There was no difference in the follow-up duration between the 2 groups (3.5 ± 2.6 vs 3.2 ± 2.7 years, p = 0.23).
| Variable | OAT | p Value | |
|---|---|---|---|
| Yes (n = 200) | No (n = 230) | ||
| Age (years) | 67.3 ± 9.7 | 69.7 ± 10.4 | 0.01 |
| Men | 118 (59%) | 148 (64%) | 0.26 |
| Hypertension | 152 (76%) | 149 (65%) | 0.03 |
| Diabetes mellitus | 91 (46%) | 122 (53%) | 0.12 |
| Heart failure | 91 (46%) | 76 (33%) | 0.008 |
| History of stroke/transient ischemic attack | 84 (42%) | 68 (30%) | 0.007 |
| Chronic kidney disease | 50 (25%) | 75 (33%) | 0.08 |
| CHADS 2 score | 2.8 ± 1.0 | 2.7 ± 1.0 | 0.42 |
| CHA 2 DS 2 -VASc score | 4.0 ± 1.3 | 3.9 ± 1.3 | 0.62 |
| HAS-BLED score | 5.7 ± 0.9 | 5.1 ± 0.6 | 0.01 |
| Ulcer | |||
| Gastric ulcer | 155 (78%) | 164 (72%) | 0.14 |
| Duodenal ulcer | 22 (11%) | 33 (14%) | 0.32 |
| Both | 23 (12%) | 33 (14%) | 0.15 |
| Medication | |||
| Aspirin | 69 (35%) | 101 (44%) | 0.04 |
| Clopidogrel ⁎ | 13 (7%) | 40 (17%) | 0.001 |
| Nonsteroidal anti-inflammatory drugs | 72 (36%) | 73 (32%) | 0.41 |
| Proton pump inhibitors | 124 (62%) | 124 (54%) | 0.11 |
| H2 blockers | 106 (53%) | 127 (55%) | 0.65 |
| Misoprostol | 6 (3%) | 9 (4%) | 0.61 |
⁎ Thirteen and 29 patients in the OAT+ and OAT− groups, respectively, were simultaneously prescribed aspirin.
The incidence rates of MACEs and major bleeding are listed in Table 2 . During the follow-up period, 28 OAT+ patients (14%) and 66 OAT− patients (29%) had MACEs. The incidence of MACE was significantly lower in OAT+ patients than in OAT− patients (p = 0.001). Seventeen OAT+ patients (9%) and 41 OAT− patients (18%) developed stroke. OAT decreased the unadjusted HR (relative risk 0.24, 95% CI 0.14 to 0.43, p <0.001) and adjusted HR (relative risk 0.22, 95% CI 0.13 to 0.41, p <0.001) for MACEs in patients with nonvalvular AF with previous GI ulcers. OAT+ patients had higher cumulative MACE-free survival than OAT− patients (p <0.001; Figure 1 ) .
| Variable | OAT | p Value | |
|---|---|---|---|
| Yes (n = 200) | No (n = 230) | ||
| MACEs | 28 (14%) | 66 (29%) | 0.001 |
| Stroke | 17 (9%) | 41 (18%) | 0.005 |
| Myocardial infarction | 8 (4%) | 14 (6%) | 0.33 |
| Pulmonary thromboembolism | 2 (1%) | 4 (2%) | 0.51 |
| Other systemic embolism | 1 (1%) ⁎ | 7 (3%) † | 0.05 |
| Major bleeding | 46 (23%) | 25 (11%) | 0.001 |
| Gastrointestinal tract | 32 (16%) | 20 (9%) | 0.02 |
| Any central nervous system bleeding | 9 (5%) | 5 (2%) | 0.19 |
| Genitourinary tract | 1 (1%) | 0 | 0.47 |
| Hematoma | 4 (2%) | 0 | 0.05 |
| Composite end point | 58 (29%) | 83 (36%) | 0.08 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
