SEARCH

SEARCH BY CITATION

Keywords:

  • anticoagulation;
  • atrial fibrillation;
  • hemorrhage;
  • stroke;
  • warfarin

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. Disclosure of Conflict of Interests
  10. References
  11. Appendix

Summary. Background: Previous studies of anticoagulation for atrial fibrillation (AF) have predominantly occurred in academic settings or randomized trials, limiting their generalizability.Objective: To describe the management of patients with AF anticoagulated with warfarin in community-based practise.Methods: We enrolled 3396 patients from 101 community-based practises in 38 states. Data included demographics, comorbidities, and International Normalized Ratio (INR) values. Outcomes included time in therapeutic INR range (TTR), stroke, and major hemorrhage.Results: The mean TTR was 66.5%, but varied widely among patients: 37% had TTR above 75%, while 34% had TTR below 60%. The yearly rates of major hemorrhage and stroke were 1.90 per 100 person-years and 1.00 per 100 person-years. Four percent of patients (n = 127) were intentionally targeted to a lower INR, and spent 42.7% of time with an INR below 2.0, compared to 18.8% for patients with a 2.0–3.0 range (P < 0.001). Mean TTR for new warfarin users (57.5%) remained below that of prevalent users through the first six months. Patients with interruptions of warfarin therapy had lower TTR than all others (61.6% vs. 67.2%, P < 0.001), which corrected after deleting low peri-procedural INR values (67.0% vs. 67.4%, P = 0.73).Conclusions: Anticoagulation control varies widely among patients taking warfarin for AF. TTR is affected by new warfarin use, procedural interruptions, and INR target range. In this community-based cohort of predominantly prevalent warfarin users, rates of hemorrhage and stroke were low. The risk versus benefit of a lower INR target range to offset bleeding risk remains uncertain.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. Disclosure of Conflict of Interests
  10. References
  11. Appendix

The number of individuals with atrial fibrillation (AF) in the United States is projected to reach 7.5 million by the year 2020 [1]; many European countries can expect similar increases in the prevalence of AF due to aging of the population. Warfarin has been shown in randomized trials to reduce the risk of stroke in AF by 68% [2]. However, the effectiveness of warfarin is challenged by its variable dose response, narrow therapeutic window, and the need for frequent monitoring of the International Normalized Ratio (INR) [3]. Previous observational studies have explored anticoagulation care predominantly within the setting of large anticoagulation clinics [4–7], but little is known about anticoagulation care in community-based practise.

The objective of the Anticoagulation Consortium to Improve Outcomes Nationally (ACTION) Study was to better define patterns of care in community-based practise related to stroke prevention in AF, including use of lower INR target ranges and their effect on anticoagulation control. Ranges of 1.5–2.5 and 2.0–2.5 have been recommended as a possible strategy to offset bleeding risk, particularly among high-risk elderly patients and high-risk patients receiving antiplatelet agents in addition to warfarin following coronary intervention [8,9]. To date, there are no prospective studies documenting the effect of lower INR targets on percent time in the therapeutic INR range (TTR). However, previous studies do suggest that low INR values may confer an increased risk of stroke without reducing bleeding risk [10,11], raising concerns about this strategy. Other unreported areas of interest included frequency of INR testing, proportion of patients in community practise with stable INR control, time course to achieve control in patients new to warfarin, and effect of interruptions of therapy on INR control. A better understanding of these issues would provide for a more informed interpretation of time-in-range analyses, particularly in an era of quality measurement.

Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. Disclosure of Conflict of Interests
  10. References
  11. Appendix

Patients and practise sites

Methods for the study have been described in detail elsewhere [12,13]. Physician practises that were registered users of CoumaCare® software (Bristol-Myers Squibb, Princeton, NJ, USA) were invited to participate. CoumaCare® was freely available and widely used to assist with patient tracking, data entry, and record keeping, but did not include dosing algorithms or other forms of decision support. CoumaCare® provided a uniformity of data structure among the sites that made our study possible at a time when only 18% of US medical practises had an electronic medical record [14].

In total, 174 practises registered to participate and 101 sites had the technological capability and the review board approval necessary to proceed. All sites had at least one dedicated provider managing warfarin, usually within the setting of a community-based, physician group practise. For all centers, McKesson HBOC BioServices (Rockville, MD, USA), provided on-site training about how to recruit patients, obtain consent, transmit data, and report adverse events in accordance with federal regulatory requirements. McKesson is an independent healthcare services company that provides biomedical support services to the United States Government, industry, universities, and contract research organizations.

Patients were invited to participate by letter, clinic flyer, or in person (at the time of a routine appointment). To be eligible, patients had to be 18 years of age or older and provide written informed consent. In total, 6761 patients were enrolled in the ACTION study; enrollment began in April 2000 and follow-up ended in March 2002. The current report analyzes patients taking warfarin for AF, who represent 50.2% of the entire ACTION cohort.

Data collected included demographic information, indication for anticoagulation, medical diagnoses, INR target range, INR values, warfarin dose, and patient management notes. Missing data fields and data entry errors were flagged and resolved directly with the sites by McKesson HBOC. Any interval of 45 days or more without INR testing or any INR value >10 or <0.8 triggered a direct query from the data coordinating center. Resolution of the flag relating to the INR testing interval required validation of continued warfarin use and confirmation that the gap was not related to an adverse event.

The study protocol was approved by the Western Institutional Review Board® (WIRB®) of Olympia, WA, USA, and by local review boards where they existed.

Study variables

Risk factors for stroke were extracted from the anticoagulation electronic record, and a CHADS2 stroke risk score was calculated for each patient [15]. TTR was calculated using linear interpolation, as described by Rosendaal et al. [16]. We also used the method described by Fihn et al. to calculate INR variability [4,17].

We defined a low INR target range as any range with an upper bound below 2.8. Visit text notes were reviewed for definitive evidence of new warfarin use upon study entry. Patients were considered new to warfarin if they had not yet taken any doses of warfarin as of the first clinic note, or were having their first INR measurement after starting warfarin. Frequency of testing was determined by calculating the mean interval between INR tests for each patient, excluding periods of 120 days or greater between INR measurements (periods when the patient was monitored elsewhere). All intentional interruptions of warfarin therapy for a procedure were documented through review of the anticoagulation clinic notes.

Adverse events

Ischemic stroke, systemic arterial embolism, and major hemorrhage were the adverse outcomes of interest. Major hemorrhage was defined as a fatal event, an event requiring hospitalization with transfusion of at least two units of packed red blood cells, or bleeding involving a critical anatomical site such as the cranium or the retroperitoneum. All patient progress notes were individually reviewed for evidence of adverse events; events were validated directly with the sites by McKesson.

Statistical analyses

We calculated TTR for each patient in the study and described the distribution of TTR among patients. We described the INR control of patients with low INR targets compared to patients with an INR target range of 2.0–3.0. We compared INR control of patients new to warfarin with the INR control of prevalent warfarin users, and described their progression toward stable control by week after inception of therapy. We also investigated the patient-level correlates of better or worse INR control, as measured by TTR.

We divided patients into four groups based on their mean interval between INR measurements and examined the relationship with TTR. We also examined the effect on TTR of excluding INR values within 14 days of a procedure. Finally, we computed unadjusted incidence rate ratios for ischemic stroke/systemic embolus and major hemorrhagic events, stratifying by INR target range, age ≥ 80 years, and new warfarin status.

Bivariate comparisons were performed using generalized estimating equations (GEEs), in order to account for intraclass correlation within sites of care. When appropriate, tests for linear trend were conducted across groups. Incidence rate ratios and their 95% confidence intervals (CIs) were conducted using a log–linear model (Poisson regression). Because multiple INR values for each patient violate standard assumptions of independent observations, a comparison of frequencies across INR categories was conducted using Monte Carlo methods. All analyses were performed using the R statistical package, version 2.2 (R Foundation, 2007).

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. Disclosure of Conflict of Interests
  10. References
  11. Appendix

Demographics and study groups

Of the 6761 total patients enrolled, 3396 were taking warfarin for AF (50.2%) and constituted our study cohort. Patients were drawn from 101 clinical practise sites within 38 states of the United States. Ninety-eight of the sites were community-based office practises (98.5% of patients) and three sites were designated anticoagulation clinics (1.5% of patients). Forty-three percent of the community-based sites were cardiology group practises. There were 2892 total person-years of follow-up. The mean age of participating patients was 74 years and 41.9% were female (Table 1).

Table 1.   Baseline patient characteristics stratified by International Normalized Ratio (INR) target range and prevalent warfarin status
Parameter Low target INR range (n = 127)Normal target INR range – new starts (n = 165)Normal target INR range – prevalent users (n = 3104)
  1. *P < 0.05, compared to prevalent warfarin users. †P < 0.001, compared to prevalent warfarin users.

Mean age (SD)78.1 (8.1)†72.9 (9.9)74.0 (9.3)
Female gender65 (51.2%)68 (41.2%)1300 (41.9%)
Race/Ethnicity
 White120 (94.5%)153 (92.7%)2909 (93.7%)
 Black2 (1.6%)3 (1.8%)42 (1.4%)
 Other5 (3.9%)9 (5.5%)153 (4.9%)
Coronary artery disease58 (45.7%)*57 (34.5%)927 (29.9%)
Stroke risk factors
 Age ≥7587 (68.5%)*79 (47.9%)1662 (53.5%)
 Diabetes mellitus19 (15.0%)23 (13.9%)521 (16.8%)
 Hypertension69 (54.3%)94 (57.0%)*1455 (46.9%)
 Heart failure27 (21.3%)32 (19.4%)722 (23.3%)
 Prior stroke13 (10.2%)10 (6.1%)2768 (10.8%)
CHADS2 stroke risk score
 012 (9.4%)27 (16.4%)483 (15.6%)
 145 (35.4%)59 (35.8%)1087 (35.0%)
 241 (32.3%)54 (32.7%)914 (29.4%)
 317 (13.4%)20 (12.1%)424 (13.7%)
 4 or greater12 (9.4%)5 (3.0%)196 (6.3%)

There were 127 patients (3.7%) with a low target INR range. These patients were not exclusively targeted below an INR of 2.0; 111 (87%) had an upper bound of 2.5, 2.6, or 2.7 and only 16 (13%) had a target range of 1.5–2.0. Patients with a low INR target range were older (78 vs. 74 years, P < 0.001) and more likely to have coronary artery disease (45.7% vs. 29.9%) compared to patients with an INR target range of 2.0–3.0. Patients newly starting warfarin (n = 165) were similar to prevalent users in terms of age and CHADS2 score.

INR control in patients with a low target INR range

Patients with INR targets below the standard range spent 42.7% of time with an INR below 2.0 (Table 2), compared to 18.8% for patients with the standard target range (P < 0.001). The mean INR value among patients with a low target range was 2.15, significantly lower than patients with a standard target range (2.47). Conversely, the low target group spent less time with an INR greater than 3.0 (4.8% vs. 13.6%, P < 0.001).

Table 2.   Warfarin management and anticoagulation control, stratified by International Normalized Ratio (INR) target range and prevalent warfarin status
Parameter Low target INR range (n = 127)Normal target INR range – new starts (n = 165)Normal target INR range – prevalent users (n = 3104)
  1. *P < 0.05, compared to prevalent warfarin users. †P < 0.001, compared to prevalent warfarin users. ‡P < 0.001 via Monte Carlo.

Follow-up and INR testing
 Mean days in database [SD]355 (127) 222 (140)†  332 (131)
 Mean number of INR values [SD] 16.2 (8.2)  15.9 (8.3)   16.3 (7.8)
 Mean INR values/month [SD]  1.40 (0.56)   2.76 (1.38)†    1.60 (0.83)
Intentional interruptions of warfarin for procedures per 100 patient-months  3.2   4.1    3.5
Frequency of INR by category‡
 1.9 or less851 (42.8%) 801 (30.9%)10767 (21.6%)
 2.0–3.0992 (49.9%)1358 (52.4%)31059 (62.0%)
 3.1–3.9114 (5.7%) 289 (11.1%) 6399 (12.8%)
 4.0 and above 32 (1.6%) 144 (5.6%) 1830 (3.7%)
Time in INR target range
 Time below range (%) 42.7%†  27.9%†   18.8%
 Time in range (%) 52.5%†  57.5%†   67.5%
 Time above range (%)  4.8%†  14.6%   13.6%
INR variability
 Mean INR value  2.15†   2.41*    2.47
 Standard deviation  0.53*   1.00†    0.68

INR control in patients new to warfarin

Compared to prevalent warfarin users with normal target INR ranges, patients new to warfarin (Table 2) had lower TTR (57.5% vs. 67.5%, P < 0.001), more time in the subtherapeutic range (27.9% vs. 18.8%, P < 0.001), and underwent more frequent testing (2.76 INR measurements per month vs. 1.60, P < 0.001). Table 3 shows the time to stable INR control over the first six months of therapy, compared with prevalent warfarin users. At six months, the INR control of the new warfarin group had yet to match that of the prevalent warfarin group.

Table 3.   Percentage of International Normalized Ratio (INR) values in the target range by consecutive week and month of warfarin therapy. The first four weeks are calculated separately only for the new starts group; beyond that, results are tabulated using 28-day months
Week of studyNew starts (n = 165)All others (n = 3104)
Number of INR values% INR values in-rangeNumber of INR values% INR values in-range
Index INR16521.2310460.4
Week 120040.0
Week 216648.2
Week 312751.2
Week 411155.0
Month 232456.5410661.3
Month 325858.1388861.1
Month 422454.0385262.9
Month 517555.4361863.3
Month 614654.1349862.2

Time in therapeutic INR range among the entire cohort

As is customary when calculating TTR [16], we did not interpolate between INR values separated by more than 56 days; 5.2% of person-time was not interpolated for this reason. For the entire cohort, the mean TTR was 66.5% (standard deviation 19.9%; median 68.3%, interquartile range 54.7–81.2%). TTR varied greatly among patients: 37% of the cohort achieved 75% or greater TTR (‘excellent’), 29% had a TTR between 60% and 75% (‘good’), and 34% had a TTR of 60% or less (‘poor’).

We compared demographics and comorbidities among these three groups with excellent, good, and poor TTR (Table 4). Mean age and CHADS2 risk score for stroke were similar in the three groups, but more patients with heart failure and/or coronary artery disease were among those with poor control. The group with excellent INR control had the lowest proportion of females (37.9%), followed by the good control group (42.6%), with the highest proportion of females in the poor control group (46.5%; P < 0.001 for trend).

Table 4.   Descriptive statistics for patients with poor (<60%), good (60–75%), and excellent (>75%) results for time in the therapeutic International Normalized Ratio range (TTR)
Parameter TTR < 60% (n = 1141)TTR 60%–75% (n = 1009) TTR > 75% (n = 1246)
  1. *P < 0.05, compared to >75% group. †P < 0.001, compared to >75% group.

Mean age (SD)73.8 (9.9)74.5 (9.1)74.2 (8.8)
Female gender46.5%†42.6%*37.9%
Race/Ethnicity
 White91.9%*93.6%95.4%
 Black2.4%*1.0%0.8%
 Other5.7%5.5%*4.7%
Coronary artery disease32.4%*31.6%28.3%
Stroke risk factors
 Age ≥7553.5%54.7%53.4%
 Diabetes mellitus17.5%16.4%15.9%
 Hypertension47.2%48.0%47.8%
 Congestive heart failure26.1%†22.5%20.5%
 Prior stroke12.0%9.1%10.4%
CHADS2 stroke risk score
 014.9%15.8%15.5%
 133.0%34.7%37.2%
 230.6%29.7%28.9%
 313.9%14.9%12.2%
 4 or greater7.5%5.0%6.2%

Frequency of INR testing

The mean interval between INR tests among all patients was 22.2 days (standard deviation 7.4). The relationship between testing interval and INR control is illustrated in Table 5. A longer interval between tests was strongly associated with increased TTR and reduced INR variability (P < 0.001 for trend for both). In particular, patients with the longest intervals had the best control: 21% of patients averaged greater than 28 days between INR measurements and had a mean TTR of 75.6%.

Table 5.   Time in therapeutic International Normalized Ratio (INR) range (TTR) and INR variability compared among groups stratified by mean interval between INR tests
Mean interval between INR tests (days) Number of patients (%) TTR (95% CI)* INR variability (95% CI) *
  1. *P < 0.001 for trend for both TTR and INR variability.

<14446 (13)51.6% (49.7, 53.5)1.22 (1.10, 1.33)
14–20.991041 (31)62.8% (61.7, 63.8)0.90 (0.85, 0.96)
21–27.991197 (35)69.9% (68.9, 70.9)0.53 (0.50, 0.55)
>28712 (21)75.6% (74.1, 77.1)0.33 (0.31, 0.35)

TTR and intentional interruptions of warfarin

Twenty-eight percent (n = 946) of patients had at least one intentional interruption of warfarin for a procedure. A subset of these patients (n = 425, 13%) had at least one INR below 1.5 recorded within 14 days of an interruption. These 425 patients had a mean TTR of 61.6% compared to the mean TTR of 67.2% for the remainder of the study cohort (P < 0.001). After deletion of all INR values within 14 days of a procedure, mean TTR in the two groups was 67.0% and 67.3%, no longer a statistically significant difference (P = 0.32).

Rates of adverse events

There were 55 major hemorrhagic events during 2892.1 person-years of observation, providing an incidence rate of 1.90 events/100 person-years (95% CI 1.46–2.48). Six (11%) events were intracranial hemorrhages and 36 (65%) were gastrointestinal hemorrhages. Major hemorrhage (Table 6) occurred more frequently among patients who were ≥80 years of age [unadjusted incidence rate ratio (IRR) 2.07; 95% CI 1.24–3.44] and among those with TTR less than 60% (unadjusted IRR 2.37; 95% CI 1.17–4.79). The low INR target group experienced nearly twice the rate of major hemorrhage despite spending significantly less time with an INR greater than 3.0; however, this difference was not statistically significant. Patients new to warfarin also appeared to be at higher risk of bleeding, but the difference was not statistically significant.

Table 6.   Incidence rates (IRs) and incidence rate ratios (IRRs) for major hemorrhage
Risk factorPatient-years of follow-upMajor bleedsIR per 100 person-yearsIRR (95% CI) unadjusted
  1. *P < 0.05 for comparison. TTR, time in therapeutic International Normalized Ratio range.

Patient new to warfarin
 No2798.2531.89
 Yes93.922.131.11 (0.27, 4.63)
Low target range
 No2780.6511.83
 Yes111.643.581.88 (0.66, 5.32)
Age ≥ 80
 No2015.0291.44
 Yes876.4262.972.07* (1.24, 3.44)
TTR ranges
 TTR > 75%1093.4141.28
 TTR 60–75%917.3141.531.20 (0.51, 2.81)
 TTR < 60%881.4273.062.37* (1.17, 4.79)

There were 29 ischemic stroke/systemic embolic events among 2892.1 person-years, an incidence rate of 1.00 event/100 person-years (95% CI 0.70–1.44). A limited number of events precluded a calculation of risk in subgroups. Of interest, among patients with a low target INR range, there were two stroke/embolic events in 111.6 patient-years, giving a crude incidence rate of 1.79 events/100 person-years. The unadjusted incidence rate ratio for such patients, compared to all others, was 1.85, but the difference was not statistically significant (95% CI 0.42–6.60).

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. Disclosure of Conflict of Interests
  10. References
  11. Appendix

Summary of main results and TTR variation among subjects

We report the results of a large, nationally representative, community-based cohort study of anticoagulation management and outcomes in patients with AF. The mean TTR in our sample (66.5%) is similar to that achieved in clinical trials in a recent meta-analysis (66.4%) [18]. TTR varied widely among individual patients in our study, with approximately one-third of patients below 60% and one-third above 75%. This distribution is strikingly similar to that recently reported from the warfarin-treated patients in the SPORTIF (Stroke Prevention Using an Oral Thrombin Inhibitor in Atrial Fibrillation) trials: 33% below 60%, 33% 60–75%, and 33% above 75%. These three groups were found to have different rates of adverse events such as hemorrhage and stroke [19]. Similarly, in our study, the group with the worst control had a higher rate of major hemorrhage than the group with the best control (unadjusted IRR 2.37).

The wide variability of TTR among patients in our study suggests that some patients have an easier time achieving and maintaining stable anticoagulation than others. It is important to note that age was essentially the same across these TTR groups, suggesting that the increased rate of hemorrhage among elderly patients is not a result of more erratic INR control. The finding that male gender is associated with higher TTR remains to be explored. Further study is needed to elucidate the ability of high-quality clinical management to minimize variability of TTR among patients.

Patients with low target INR range

One of our most striking results is that patients with a low target INR range spent 42.7% of time with an INR below 2.0. As noted earlier, only 13% of these patients had a designated target of 1.5–2.0. Despite less time with an INR >3.0, these patients seemed to have a higher rate of bleeding than patients with normal target ranges, although this difference did not reach statistical significance. However, this reinforces the fact that these patients were indeed at elevated risk for bleeding, and underscores the complexity of managing anticoagulation in such patients. There also seemed to be a higher rate of thromboembolic events in the low INR target group, but the difference did not achieve statistical significance, probably because of a limited number of events. However, we did show that patients with low target INR ranges spend a great proportion of time with an INR below 2.0. Given the known increase in the risk of stroke when the INR is below 2.0 [10,11], patients and clinicians need to be cognizant of the potential trade-offs inherent to the use of a lower INR target range for stroke prevention in AF.

Patients new to warfarin

Our study helps to characterize the natural history of INR control in patients who are new to warfarin and suggests a persistent difference compared to prevalent warfarin users, even at six months post-initiation. This finding emphasizes the adherence and survivor biases that are intrinsic to longer-term use of warfarin and thus to cohorts of prevalent users.

Frequency of INR monitoring

We found that longer INR monitoring intervals were associated with improved INR control. In the United States, clinical guidelines currently recommend that INR testing occur at least every 28 days for all patients [3,9], but such recommendations are based on expert consensus rather than evidence. Our results suggest that clinicians are able to identify patients who can safely be tested less often, and call into question whether all patients must be tested with a fixed minimum frequency in clinical practise. Indeed, despite testing INR as seldom as every 12 weeks [20], British patients achieve TTR results of 65–70% in usual practise [21,22]; these results are at least as good as those achieved with the more frequent testing intervals common in US practise. Rather than a fixed maximum recall interval, recall intervals might be tailored to recent INR control [17,23]. Some software programs already provide optimized recall intervals based on recent INR control, most notably the Birmingham Anticoagulation Program for Primary Care (BAP-PC) [24].

Effect of interruptions of warfarin therapy on measurement of TTR

We found that measurement of TTR can be affected by low INR values because of intentional interruptions of therapy. In our sample, the deletion of INR values for 14 days before and after a procedure increased TTR from 62% to 67%, a clinically meaningful difference. This refinement improves the validity of TTR as a measurement of the quality of anticoagulation care, because the documentation of low INR values proximal to procedures does not imply poor care.

Rates of stroke and hemorrhage

The rates of adverse events in our study, 1.90 major hemorrhage and 1.00 stroke/systemic embolus per 100 person-years, are similar to other, relatively recent studies of patients with AF taking warfarin. For example, among a large cohort of patients taking warfarin for AF from Kaiser Permanente Northern California, Go et al. [7] found rates of 1.52 and 1.17 for major hemorrhage and stroke/systemic embolus, respectively. However, it should be noted that any cohort of predominantly prevalent warfarin users, such as ours, is enriched with patients able to tolerate warfarin therapy without complications, while patients new to warfarin may experience higher rates of adverse events [25].

Limitations

Our study has several limitations. First, the small number of stroke/systemic embolic events precluded a calculation of risk in key subgroups of patients. However, previous studies have already quantified the risk of stroke among many important subsets of patients [15]. Secondly, it is possible that healthier patients preferentially gave informed consent to participate in this study. However, the distribution of stroke risk factors and rates of adverse events (stroke, major hemorrhage) are similar to those found in previous studies [7], suggesting that our patients are representative. In addition, our finding that 34% of patients spent 60% or less time in the therapeutic range argues against biased patient selection.

Conclusions

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. Disclosure of Conflict of Interests
  10. References
  11. Appendix

Anticoagulation control in our cohort of community-based patients with AF was similar to participants in clinical trials, but the TTR of individual patients varied widely. Further studies are needed to investigate patient-level and site-level determinants of TTR. Our results concord with those of previous studies [7], suggesting that among prevalent users of warfarin, rates of major hemorrhage and ischemic stroke are low. Our study also highlights the complexity of antithrombotic management among patients deemed to be at highest risk of hemorrhage. Optimal management of these patients has yet to be determined.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. Disclosure of Conflict of Interests
  10. References
  11. Appendix

This study was funded by Bristol-Myers Squibb, the makers of Coumadin® brand warfarin. Bristol-Myers Squibb had no role in the design and conduct of the study, or in the collection, analysis, and interpretation of the data, or in the preparation, review, and approval of the manuscript. Dr Rose is supported by a career development grant from the United States Department of Veterans Affairs, Health Services Research and Development Service. The opinions expressed in this manuscript do not necessarily represent the views or policies of the Department of Veterans Affairs.

Disclosure of Conflict of Interests

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. Disclosure of Conflict of Interests
  10. References
  11. Appendix

The authors state that they have no conflict of interest.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. Disclosure of Conflict of Interests
  10. References
  11. Appendix
  • 1
    Miyasaka Y, Barnes ME, Gersh BJ, Cha SS, Bailey KR, Abhayaratna WP, Seward JB, Tsang TS. Secular trends in incidence of atrial fibrillation in Olmsted County, Minnesota, 1980 to 2000, and implications on the projections for future prevalence. Circulation 2006; 114: 11925.
  • 2
    Hart RG, Pearce LA, Aguillar MI. Meta-analysis: antithrombotic therapy to prevent stroke in patients who have non-valvular atrial fibrillation. Ann Intern Med 2007; 146: 85767.
  • 3
    Ansell J, Hirsh J, Poller L, Bussey H, Jacobson A, Hylek E. The pharmacology and management of the vitamin K antagonists: the Seventh ACCP Conference on Antithrombotic and Thrombolytic Therapy. Chest 2004;126(3 Suppl.):204S33S.
  • 4
    Fihn SD, McDonell M, Martin D, Henikoff J, Vermes D, Kent D, White RH. Risk factors for complications of chronic anticoagulation. A multicenter study. Warfarin Optimized Outpatient Follow-up Study Group. Ann Intern Med 1993; 118: 51120.
  • 5
    Palareti G, Leali N, Coccheri S, Poggi M, Manotti C, D’Angelo A, Pengo V, Erba N, Moia M, Ciavarella N, Devoto G, Berrettini M, Musolesi S. Bleeding complications of oral anticoagulant treatment: an inception-cohort, prospective collaborative study (ISCOAT). Italian Study on Complications of Oral Anticoagulant Therapy. Lancet 1996; 348: 4238.
  • 6
    Van Der Meer FJ, Rosendaal FR, Vandenbroucke JP, Briet E. Bleeding complications in oral anticoagulant therapy. An analysis of risk factors. Arch Intern Med 1993; 153: 155762.
  • 7
    Go AS, Hylek EM, Chang Y, Phillips KA, Henault LE, Capra AM, Jensvold NG, Selby JV, Singer DE. Anticoagulation therapy for stroke prevention in atrial fibrillation: how well do randomized trials translate into clinical practice? Jama 2003; 290: 268592.
  • 8
    Anderson JL, Adams CD, Antman EM, Bridges CR, Califf RM, Casey DE Jr, Chavey WE II, Fesmire FM, Hochman JS, Levin TN, Lincoff AM, Peterson ED, Theroux P, Wenger NK, Wright RS, Smith SC Jr, Jacobs AK, Halperin JL, Hunt SA, Krumholz HM, et al. ACC/AHA 2007 guidelines for the management of patients with unstable angina/non ST-elevation myocardial infarction: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines. Circulation 2007; 116: e148304.
  • 9
    Fuster V, Ryden LE, Cannom DS, Crijns HJ, Curtis AB, Ellenbogen KA, Halperin JL, Le Heuzey JY, Kay GN, Lowe JE, Olsson SB, Prystowsky EN, Tamargo JL, Wann S, Smith SC Jr, Jacobs AK, Adams CD, Anderson JL, Antman EM, Halperin JL, et al. ACC/AHA/ESC 2006 Guidelines for the Management of Patients with Atrial Fibrillation: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines and the European Society of Cardiology Committee for Practice Guidelines. Circulation 2006; 114: e257354.
  • 10
    Hylek EM, Skates SJ, Sheehan MA, Singer DE. An analysis of the lowest effective intensity of prophylactic anticoagulation for patients with non-rheumatic atrial fibrillation. N Engl J Med 1996; 8: 5406.
  • 11
    Hylek EM, Go AS, Chang Y, Jensvold NG, Henault LE, Selby JV, Singer DE. Effect of intensity of oral anticoagulation on stroke severity and mortality in atrial fibrillation. N Engl J Med 2003; 349: 101926.
  • 12
    Garcia D, Regan S, Crowther M, Hughes RA, Hylek EM. Warfarin maintenance dosing patterns in clinical practice: implications for safer anticoagulation in the elderly population. Chest 2005; 127: 204956.
  • 13
    Garcia DA, Regan S, Crowther M, Hylek EM. The risk of hemorrhage among patients with warfarin-associated coagulopathy. J Am Coll Cardiol 2006; 47: 8048.
  • 14
    Cherry DK, Burt CW, Woodwell DA. National Ambulatory Medical Care Survey: 2001 summary. Adv Data 2003; 11: 144.
  • 15
    Gage BF, Waterman AD, Shannon W, Boechler M, Rich MW, Radford MJ. Validation of clinical classification schemes for predicting stroke: results from the National Registry of Atrial Fibrillation. Jama 2001; 285: 286470.
  • 16
    Rosendaal FR, Cannegieter SC, Van Der Meer FJ, Briet E. A method to determine the optimal intensity of oral anticoagulant therapy. Thromb Haemost 1993; 69: 2369.
  • 17
    Kent DL, Vermes D, McDonell M, Henikoff J, Fihn SD. A model for planning optimal follow-up for outpatients on warfarin anticoagulation. Warfarin Optimal Outpatient Follow-up Study Group. Med Decis Making 1992; 12: 13241.
  • 18
    Van Walraven C, Jennings A, Oake N, Fergusson D, Forster AJ. Effect of study setting on anticoagulation control: a systematic review and metaregression. Chest 2006; 129: 115566.
  • 19
    White HD, Gruber M, Feyzi J, Kaatz S, Tse HF, Husted S, Albers GW. Comparison of outcomes among patients randomized to warfarin therapy according to anticoagulant control: results from SPORTIF III and V. Arch Intern Med 2007; 167: 23945.
  • 20
    Haemostasis and Thrombosis and Task Force for the British Committee for Standards in Haematology. Guidelines on oral anticoagulation: third edition. Br J Haematol 1998; 101: 37487.
  • 21
    Fitzmaurice DA, Murray ET, McCahon D, Holder R, Raftery JP, Hussain S, Sandhar H, Hobbs FD. Self management of oral anticoagulation: randomised trial. BMJ 2005; 331: 1057.
  • 22
    Mant J, Hobbs FD, Fletcher K, Roalfe A, Fitzmaurice D, Lip GY, Murray E. Warfarin versus aspirin for stroke prevention in an elderly community population with atrial fibrillation (the Birmingham Atrial Fibrillation Treatment of the Aged Study, BAFTA): a randomized controlled trial. Lancet 2007; 370: 493503.
  • 23
    Fihn SD, McDonell MB, Vermes D, Henikoff JG, Martin DC, Callahan CM, Kent DL, White RH. A computerized intervention to improve timing of outpatient follow-up: a multicenter randomized trial in patients treated with warfarin. National Consortium of Anticoagulation Clinics. J Gen Intern Med 1994; 9: 1319.
  • 24
    Van Walraven C, Oake N, Wells PS, Forster AJ. Burden of potentially avoidable anticoagulant-associated hemorrhagic and thromboembolic events in the elderly. Chest 2007; 131: 150815.
  • 25
    Hylek EM, Evans-Molina C, Shea C, Henault LE, Regan S. Major hemorrhage and tolerability of warfarin in the first year of therapy among elderly patients with atrial fibrillation. Circulation 2007; 115: 268996.

Appendix

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. Disclosure of Conflict of Interests
  10. References
  11. Appendix

The following practises and directors participated in the study (the sites are listed in decreasing order of number of patients enrolled): Lutheran General Hospital, Niles, Illinois—W. Fried, M. Pubentz; Physicians, Inc., Lima, Ohio—D. Parker; Idaho Cardiology Associates, Boise, Idaho—F. Badke; North Clinic, Robbinsdale, Minnesota—V. Krug; Rockwood Clinic P.S., Main, Spokane, Washington—J. S. Pennock; Wenatchee Valley Clinic, Wenatchee, Washington—R. Kirby Primm, L. Vaughn; Framingham Heart Center, Framingham, Massachusetts—J. Dangel, S. R. Hewett; Clinic Pharmacy Consultants-Brainerd Medical Center, Brainerd, Minnesota—B. Twamley, R. Sorenson; Woodland Healthcare, Woodland, California—L. Smith, T. Fajerson; Cardiology, PC, Syracuse, New York—S. O’Donnell; Health Care American Corp, Bradenton, Florida—C. Hoffman; DuPage Medical Group, Department of Cardiology, Winfield, Illinois—N. Kinsley; Camino Medical Group, Sunnyvale, California—S. Edwards; Ohio Valley Heartcare, Evansville, Indiana—L. Janeira, J. Robb; Desert Medical Group/Oasis IPA, Palm Springs, California—H. F. Bellaci, J. Bellaci; Anchor Health Center, Naples, Florida—M. Means; Sutter Gould Medical Foundation, Modesto, California—J. E. Baker; Hannibal Clinic Inc., Hannibal, Missouri—L. Chalton; Saratoga Cardiology, Saratoga Springs, New York—R. Sheldon, D. Kandath; Lima Memorial Hospital, Lima, Ohio—C. L. Thompson, J. Recker; Staten Island University Hospital, Staten Island, New York—M. Howard; Jacksonville Cardiovascular Clinic, Jacksonville, Florida—R. A. Benson; River Valley Healthcare, Silvis, Illinois—K. Carroll; Family Physician Incorporated, N. Canton, Ohio—H. Marshall; Internal Medicine of Northern Michigan, Petoskey, Michigan—P. D. Blanchard; Redmond Internal Medicine, Redmond Oregon—D. Palmer, C. Gangan; Grove Hill Medical Center, New Britain, Connecticut—M. S. Werner; Olean Medical Group, Olean, New York—H. D. Storch, T. L. Buzzard; Internal Medicine Associates of Greenville, Greenville, South Carolina—J. S. Moore; Magan Medical Clinic, Covina, California—R. Sakamoto; Owatonna Clinic—Mayo Health System, Owatonna, Minnesota—T. Price; Dearborn Cardiology, Dearborn, Michigan—S. Dabbous; Westchester Medical Group, White Plains, New York—B. Newman; Central Cardiology Medical Clinic, Bakersfield, California—W. Nyitray; Salem Clinic, Salem, Oregon—M. Smith; East Carolina University, Greenville, North Carolina—C. Estrada; Northwest Primary Care Group, Milwaukie, Oregon—D. McAnulty, P. Devisser; The William W Backus Hospital, Norwich, Connecticut—S. Johnson; Jefferson City Medical Group, Jefferson City, Missouri—C. Balcer; Saint Louis University Department of Neurology, St Louis, Missouri—S. Cruz-Flores, E. Holzemer; Wellspan Health, Yorktowne, York, Pennsylvania—J. D. Horton; Mercy Medical Center, Canton, Ohio—M. Cudnik; Cardiovascular Group, Lawrenceville, Georgia—B. Craig-Allen; Asheville Cardiology Assoc, Asheville, North Carolina—W. Wharton, A. Moser; Cardiac Consultants Chartered, Bethesda, Maryland—L. Chappell; Valley Care Health System, Pleasanton, California—N. Huynh; Bloomington Hospital, Bloomington, Indiana—K. Kalotta; Samaritan Anticoagulation Service, Corvallis, Oregon—R. Stockberger; Covenant Clinic, Waterloo, Iowa—D. Kohls; Dartmouth-Hitchcock Nashua, Nashua, New Hampshire—L. Cook; Cardiology Consultants, PC, Hamden, Connecticut—A. M. Radoff; Seventh Avenue Family Health Center, Fort Lauderdale, Florida—J. Berges; Diagnostic Cardiology, P.A., Jacksonville, Florida—P. D. Kuhlman; Norlanco Medical Associates, Elizabethtown, Pennsylvania—J. Rittenhouse; University of Texas Medical Branch, Galveston, Texas—H. von Marensdorff; Bend Memorial Clinic, Bend, Oregon—M. Hegewald; Memorial Primary Care Center, Hollywood, Florida—J. Beck; Batey Cardiovascular Center, Bradenton, Florida—D. Calabrila, E. J. Sanchez; Western Montana Clinic, Missoula, Montana—W. B. Bekemeyer, D. Ramsey; Winona Clinic, Winona, Minnesota—L. Tschumper; Cardiac Consultants, Lancaster, Pennsylvania—M. Lesko; Hattiesburg Clinic, Hattiesburg, Mississippi—A. J. Jackson; Bryn Mawr Medical Specialist Association, Bryn Mawr, Pennsylvania—H. Mayer; River Valley Healthcare, Moline, Illinois—B. Cady; Cardiovascular Group, Snellville, Georgia—L. Lesser; Medicor, Bridgewater, New Jersey—P. Saulino, C. Hartpence; Bond Clinic, P.A., Winter Haven, Florida—P. Lundsford. K. Bhatia; University of Cincinnati-Pharmacy Anticoagulation Services, Cincinnati, Ohio—J. McQueen; Senior Healthcare Center, Gainesville, Florida—M. L. Breeser; North Canton Medical Foundation, North Canton, Ohio—H. M. Schenker; Manor Family Health Center, Millersville, Pennsylvania—J. Ichter; Cardiology Associates of Central Florida, Ocala, Florida—L. McDaniel; Cardiovascular Associates Ltd, Chesapeake, Virgina—S. R. Jones; Woodburn Medical Clinic, Woodburn, Oregon—F. Golden; Rockwood Clinic North, Spokane, Washington—C. Laudenbach, J. S. Pennock; Wachspress, Shatkin & Rainear, Vineland, New Jersey—L. Assink; Chambersburg Hospital, Chambersburg, Pennsylvania—D. Grant; Wellspan Pharmacy-Dallastown, Dallastown, Pennsylvania—T. G. Williams; Pulmonary & Critical Care Associates, Ypsilanti, Michigan—W. F. Patton; Island Cardiac Specialist, Mineola, New York—P. Ragno; Portland Cardiovascular Institute 2, Portland, Oregon—R. Chelfky; River Valley Healthcare ACS, Bettendorf, Iowa—W. Langley; Consultants in Cardiology, Farmington Hills, Michigan—G. M. McKendrick; Portland Cardiovascular Institute 1, Portland, Oregon—R. Chelfky; Cleveland Clinic Florida, Weston, Florida—B. Fernandez; BiState Medical Consultants, St Louis, Missouri—P. M. Stein, C. B. Lomnel; Medical Consultants, PC, Muncie, Indiana—J. Bow; Cardiovascular Associates of South Florida, Coral Gables, Florida—J. S. Palmer; Parkway Cardiology Associates, Oak Ridge, Tennessee—S. Cooke; Northwest Georgia Diagnostic Clinic, Gainesville, Georgia—J. Jackson; Cardiovascular Associates, Kingsport, Tennessee—L. H. Cox; Heart Place, Dallas, Texas—C. N. Bowers; Rockwood Clinic, Spokane, Washington—C. Laudenbach; J. S. Pennock; Delaware Heart Group, Newark, Delaware—C. Bowens; Abilene Diagnostic Clinic, Abilene, Texas—P. Howard.