Europace Advance Access originally published online on November 26, 2007
Europace 2008 10(1):3-5; doi:10.1093/europace/eum258
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
ATRIAL FIBRILLATION
Methodological considerations for interpretation of rates of major haemorrhage in studies of anticoagulant therapy for atrial fibrillation
1 Clinical Epidemiology Unit and Section of General Internal Medicine-Research Unit, Boston University School of Medicine, Boston, MA, USA; 2 Cardiovascular Medicine Division, Boston University School of Medicine, Boston, MA, USA
Major haemorrhage is a significant concern for elderly patients with atrial fibrillation (AF) taking anticoagulant therapy. Two recently published studies have demonstrated conflicting results with regard to the incidence of major haemorrhage on warfarin in this age group. The Birmingham Atrial Fibrillation Treatment of the Aged Study1
(BAFTA) randomized 973 patients with AF aged 75 years or more to warfarin or aspirin. During the 2.7 year mean follow-up, the incidence of major haemorrhage was nearly identical in the two study arms (1.9% per year for warfarin vs. 2.2% for those taking aspirin). In contrast, a single centre, inception cohort study identified a significantly greater incidence of major haemorrhage in elderly patients initiating warfarin for stroke prevention in AF:2 7.2% per year in a population with mean age 77 years, increased to 13.1% per year among those aged 80 and older.
How can such disparate rates be reconciled? The answers are found in the key elements of study design which guide the interpretation of event rates. These studies contrasting results demonstrate the importance of assessing study design, patient population (with particular regard to selection bias), length of observation, definition of major haemorrhage, specific anticoagulant, quality of anticoagulation control, and use of other risk-modifying therapy.
Randomized controlled trials are the most valid assessment of an intervention. Although often criticized for the highly selected nature of the study population, a trial's first priority is to ensure internal validity to facilitate an objective assessment of efficacy. Anticipated drop-out rates should be low and medication adherence rates should be high. Translating trial results into real-world practice can be challenging because the patient populations are often considerably different (external validity). Observational studies provide important insight into clinical practice, but these designs are also subject to bias. A source of bias that affects both randomized trials and observational studies is enrolment of prevalent users of warfarin opposed to those newly starting therapy. Bleeding events are more likely to occur early in the treatment course, and new users are often sicker at baseline and therefore more likely to stop therapy.3 Patients who remain on warfarin (prevalent users) are the ‘survivors’ and are a lower risk group compared with those individuals who ‘dropped out’. Trials and non-inception cohorts that predominantly enrol prevalent warfarin users will have lower event rates.
In BAFTA, 40% of subjects were taking warfarin at study entry; 100% of patients enrolled in the inception cohort study were newly starting warfarin. The latter study found the risk of haemorrhage was significantly increased during the first 90 days of therapy (IRR 3.31, 95% CI 1.51–7.25). In addition, bleeding and early warfarin termination occurred more often among patients with a CHADS2 score of 
3. This study demonstrated that higher risk patients have a disproportionate burden of adverse events, and that these events occur early following the initiation of warfarin. Thus, BAFTA may have been vulnerable to enrolling a population of patients who were not ‘warfarin naïve’. This higher proportion of warfarin prevalent subjects likely influenced the different major haemorrhage incident rates in these populations and emphasized the need for new-user designs in randomized controlled trials of antithrombotic therapy. Overall there were too few events in BAFTA to meaningfully assess this difference.
As is true of any potentially hazardous medication, the initial selection bias related to ‘candidacy’ for the drug is very difficult to overcome. This decision is often subjective and frequently based on relative rather than absolute contraindications to anticoagulant therapy. Trials often further restrict enrolment of patients perceived to be at the lowest risk for haemorrhage. The setting in which patients are identified also has critical implications: the period immediately post-hospital discharge is more risk-laden, for instance, than recruitment from routine care in an ambulatory setting. Patients identified from a hospitalization will likely also have a greater degree of comorbidities.
The patient populations enrolled in the two studies were distinctly different. Inclusion in BAFTA was restricted to patients for whom there was clinical uncertainty as to whether aspirin or warfarin should be used. This criterion therefore favoured enrolment of patients at lower risk of stroke and lower risk of haemorrhage. It is challenging to assert equipoise in aspirin vs. warfarin for patients with heart failure or high CHADS2 scores. Given the 50% likelihood of being randomized to warfarin, it also seems unlikely that these patients were deemed to be at high risk of haemorrhage. Exclusion criteria in BAFTA included: major haemorrhage in previous 5 years, intracranial haemorrhage, endoscopically proven peptic ulcer disease in the previous year, oesophageal varices, or surgery within the past 3 months. In contrast, by principle of study design, the inception cohort study had none of these exclusion criteria. Patients were deemed warfarin candidates independent of the study and placed on this anticoagulation regimen, regardless if they had any of the exclusion criteria articulated by the BAFTA investigators. In point of fact, 10% of patients had a history of gastrointestinal or other bleed. BAFTA enrolled 100% of its participants from the ambulatory setting, while the inception cohort study identified a third of its patients at hospital discharge. Twenty-eight percent of patients enrolled in BAFTA had a CHADS2 score of 3; this is in comparison with the 44% of patients in the cohort study aged 80 years or greater.
Because adverse events are more likely to occur early among new users of anticoagulant therapy, longer-term follow-up disproportionately reflects the experience of the warfarin ‘survivors’. The number of hemorrhagic events decreases over time while the person-year denominator continues to be enriched by prevalent users of warfarin.
BAFTA's mean follow up was 2.7 years. However, the inception cohort was designed purposefully to truncate observation at 1 year of warfarin therapy. Had the exposure period been extended, the ‘surviving’ patients would have continued to contribute to the person-years in the denominator resulting in a lower reported rate of major haemorrhage.
Definition of major haemorrhage
Broad, less restrictive definitions of haemorrhage serve to inflate event rates. Reporting of aggregate bleeding rates (major plus minor) tends to deter use of warfarin in clinical practice. This is particularly problematic in AF given the increased disability and mortality related to stroke in AF. The inception cohort study used the definition of major haemorrhage promulgated by the International Society of Haemostasis and Thrombosis: fatal bleeding or symptomatic bleeding in a critical area or organ, such as intracranial, intraspinal, intraocular, retroperitoneal, intraarticular or pericardial, or intramuscular with compartment syndrome, and/or bleeding causing a fall in haemoglobin level of 20 g/L (1.24 mmol/L) or more, or leading to transfusion of two units of packed red blood cells.4
Therefore, the reported rate of 7.2% for major haemorrhage is not due to inclusion of minor events.
Characteristics of the anticoagulant drug
Both studies exclusively used warfarin and experienced nearly identical time above the therapeutic range (BAFTA 14% vs. 13%). Thus, the difference in bleeding rates was not attributable to different drug pharmacokinetics or suboptimal anticoagulation control. Adherence to warfarin in BAFTA was stated to be 67% which may have biased towards lower rates of bleeding in the intention to treat analysis. The investigators acknowledged this possibility, but noted no difference in the on-treatment analyses.
Use of other risk-modifying therapy
It is unclear in BAFTA if patients randomized to warfarin had access to over-the-counter aspirin. The hazards of dual therapy are now being more fully realized as this combination is being prescribed with increasing frequency to older patients.5,6 In the uncontrolled, unselected observational cohort study, 40% of patients were also taking aspirin for either primary or secondary prevention of cardiovascular disease. This higher than expected prevalence of dual therapy has also been found in other patient populations.7 Twelve of the 26 major bleeding events in the cohort study occurred on combination therapy.
These two studies are in marked contrast to one another in terms of study design, enrolment, patient characteristics and outcomes. BAFTA enrolled patients with a choice of anticoagulant therapies and then randomized them to warfarin or aspirin. In contrast, the inception cohort study identified a sicker population with more extensive comorbidities who had been placed on warfarin and then followed for the first year. If anything, the BAFTA population reflects the outcome of a healthier population. It would be interesting to know the rate of haemorrhage among the excluded patients in the BAFTA trial whose physicians deemed warfarin to be indicated. These patients are likely more comparable with those patients enrolled in the inception cohort study.
The higher rates of major haemorrhage in the inception cohort study were driven by the unselected nature of the study population, focus on the initiation period, higher proportion of patients identified at hospital discharge, restriction to 1-year follow-up, and high prevalence of concomitant use of aspirin. The cohort study highlights the clinical complexity of patients with AF encountered in routine practice and the need for aggressive strategies to optimize the benefit of anticoagulant therapy for stroke prevention in this vulnerable patient population.
Conflict of interest: none declared.
Footnotes
The opinions expressed in this article are not necessarily those of the Editors of Europace, the European Heart Rhythm Association or the European Society of Cardiology.*Corresponding author. Tel: +1 617 414 3743; fax: +1 617 414 4676. E-mail address: ehylek@bu.edu
References
[1] Mant J, Hobbs FD, Fletcher K, Roalfe A, Fitzmaurice D, Lip GY, et al, BAFTA investigators; Midland Research Practices Network. 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 randomised controlled trial. Lancet (2007) 370:460–1.[CrossRef][Web of Science][Medline]
[2] 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) 115:2689–96.
[3] Ray WA. Evaluating medication effects outside of clinical trials: new-user designs. Am J Epidemiol (2003) 158:915–20.
[4] Schulman S, Kearon C, On behalf of the Subcommittee on Control of Anticoagulation of the Scientific Standardization Committee of the International Society on Thrombosis and Haemostasis. Definition of major bleeding in clinical investigations of antihemostatic medicinal products in non-surgical patients. J Thromb Haemost (2005) 3:692–4.[CrossRef][Web of Science][Medline]
[5] Eikelboom JW, Hirsh J. Combined antiplatelet and anticoagulant therapy: clinical benefits and risks. J Thromb Haemost (2007) 5(Suppl. 1):255–63.[CrossRef][Web of Science][Medline]
[6] Dentali F, Douketis JD, Lim W, Crowther M. Combined aspirin-oral anticoagulant therapy compared with oral anticoagulant therapy alone among patients at risk for cardiovascular disease: a meta-analysis of randomized trials. Arch Intern Med (2007) 167:117–24.
[7] Johnson SG, Witt DM, Eddy TR, Delate T. Warfarin and antiplatelet combination use among commercially insured patients enrolled in an anticoagulation management service. Chest (2007) 131:1500–7.[CrossRef][Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
Related articles in Europace:
- Anticoagulation for elderly patients with atrial fibrillation: not to be neglected
- Jaspal S. Taggar and Gregory Y.H. Lip
Europace 2008 10: 1-2.[FREE Full Text]
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||