ABSTRACT & COMMENTARYs
Warfarin may decrease morbidity and mortality in patients with kidney disease and atrial fibrillation after MI
By Deborah J. DeWaay, MD, FACP
Assistant Professor, Medical University of South Carolina, Charleston, SC
Dr. DeWaay reports no financial relationships in this field of study.
Patients with chronic kidney disease (CKD) are at increased risk of bleeding, atherosclerotic and thromboembolic events. These risks increase with progression of the CKD. Atrial fibrillation is common in patients with CKD. Warfarin is traditionally used in patients with atrial fibrillation to prevent thromboembolic disease. Observational studies analyzing the efficacy of warfarin in patients with CKD and atrial fibrillation have mixed results. This study sought to analyze outcomes of warfarin treatment in patients with CKD, cardiovascular disease and atrial fibrillation.
This study is an observational, multicenter, prospective cohort study that used the Swedish Web-System for Enhancement and Development of Evidence-Based Care in Heart Disease Evaluated According to Recommended Therapies (SWEDEHEART) registry (2003-2010) data. This registry uses data from all Swedish hospitals that give care to patients with acute cardiac disease. Patient past medical history was obtained from the registry and supplemented from the National Inpatient Registry, which includes data from all hospitalizations. Patient outcomes were stratified by renal function, which were categorized by eGFR (estimated glomerular filtration rate): those with eGFR higher than 60mL/min/1.73m2 were grouped together (normal renal function, CKD stage 1, or CKD stage 2); those with eGFR of 30-60mL/min/1.73m2 were defined as having moderate dysfunction (CKD stage 3); those with eGFR of 15-30mL/min/1.73m2 were defined as having severe dysfunction (CKD stage 4); and those with eGFR of <15mL/min/1.73m2 were defined as having end-stage renal disease (CKD stage 5). The study compared patients with a recent cardiovascular event and atrial fibrillation treated with warfarin vs. those that did not. Exposure to warfarin was defined as a prescription being given to the patient at discharge per the data in SWEDEHEART. The Swedish population registry was used to obtain mortality data. The following outcomes were analyzed: death, readmission for myocardial infarction (MI), stroke or bleeding within 1 year. The authors defined bleeding as anemia caused by bleeding, gastrointestinal bleeding or hemorrhagic stroke. Two sensitivity analyses were performed to minimize confounding variables and to calculate propensity scores in order to appropriately match patients.
More than 158,000 patients were admitted for MI during the period 2003 2010. Of these patients, over 34,000 had also had atrial fibrillation. Patients were excluded if they did not have information in the data set regarding their age, kidney function or warfarin treatment status. There were 24,317 patients who were included in the final analysis, and 21.8% of them had taken warfarin. There were no significant differences between the two groups in terms of basic demographics; however, the warfarin group was more likely to have heart failure, diabetes and stroke. The group that did not receive warfarin was more likely to have a clinical history of bleeding or hemorrhagic stroke, and 51.7% of patients had CKD III or higher.
The authors compared patients taking warfarin with those who did not with respect to a composite outcome of death, myocardial infarction, and ischemic stroke. A Cox regression analysis showed the following results, in terms of number of events per 100 person-years. For patients with an eGFR>60 (normal, CKD stage 1 or CKD stage 2), the event rate was 28.0 for warfarin vs. 36.1 for no warfarin, and adjusted hazard ratio (HR) was 0.73 (95% CI, 0.65 to 0.81); for CKD stage 3, the event rate was 48.5 for warfarin vs. 63.5 for no warfarin (HR 0.73, 95% CI 0.66 to 0.80); for CKD stage 4, the event rate was 84.3 for warfarin vs. 110.1 for no warfarin (HR 0.84, 95% CI 0.70-1.02); and for end-stage renal disease, the event rate was 83.2 for warfarin vs. 128.3 for no warfarin (HR 0.57, 95% CI 0.37-0.86). The bleeding risk for patients on warfarin compared with those not on warfarin, stratified by renal function, were as follows: For normal, CKD stage 1 or CKD stage 2, HR 1.10 [95% CI, 0.86-1.41]; for CKD stage 3, HR 1.04 [95% CI 0.81-1.33]; for CKD stage 4, HR 0.82 [95% CI 0.48-1.39]; and for end-stage renal disease, HR 0.52 [95% CI 0.16-1.65].
Patients with atrial fibrillation and CKD are very common. Hospitalists often start warfarin for patients with CKD and newly diagnosed atrial fibrillation. Although this practice is common, there has been little data as to the safety and efficacy of this practice in these patients. In addition, these patients tend to have multiple comorbidities, and they are not able to take the newer drugs for atrial fibrillation. One limitation of the study was that it is administrative data, which can be flawed. The definition of warfarin use was that the patient received a prescription at discharge. There were no data presented that the patients on warfarin were successfully anticoagulated based on an INR between 2 and 3. In addition, there were differences in comorbidities between the groups that were analyzed. However, the number of patients analyzed is impressive, and a randomized controlled trial would be incredibly challenging to do to answer this question. Based on this study, hospitalists should be comfortable prescribing warfarin for atrial fibrillation to their CKD patients in the absence of specific contraindications to warfarin or anticoagulation in general.