Abstract

Aims

To assess the risk of death after the occurrence of different types of non-fatal events in patients with atrial fibrillation (AF). Antithrombotic therapies in AF have primarily focused on stroke prevention and bleeding. However, strokes and bleeds differ in severity, and the level of severity may differently impact mortality.

Methods and results

We analysed the risk of subsequent mortality after the occurrence of non-fatal vascular and bleeding events in the Atrial Fibrillation Clopidogrel Trial with Irbesartan for Prevention of Vascular Events (ACTIVE)-W trial. In the 3371 patients randomized to vitamin K antagonists and the 3335 patients randomized to clopidogrel plus aspirin in ACTIVE-W, the hazard ratio (HR) and 95% confidence intervals (95% CIs) for subsequent death associated with the occurrence of non-fatal stroke was 5.58 (95% CI 3.84–8.10, P < 0.0001). Both ischaemic (HR 5.29, 95% CI 3.53–7.93, P < 0.0001) and haemorrhagic strokes (HR 7.38, 95% CI 2.74–19.9, P < 0.0001) increased mortality, but transient ischaemic attacks did not. Disabling strokes (Rankin's score ≥3) increased mortality (HR 9.54; 95% CI 6.42–14.2, P< 0.0001), but non-disabling strokes did not. Severe bleeding increased mortality (HR 3.35, 95% CI 2.12–5.27, P < 0.0001), but major bleeding that was not severe according to the study definitions did not.

Conclusion

Non-fatal strokes increased mortality in ACTIVE-W, but non-disabling strokes did not. Among major bleeding events, only those also classified as severe increased mortality. Future research should emphasize the prevention of disabling strokes and severe bleeds and place less emphasis on non-disabling stroke or major bleeds that are not severe.

See page 2075 for the editorial comment on this article (doi:10.1093/eurheartj/ehp486)

Introduction

Atrial fibrillation (AF) is a common condition affecting 0.4% of the general population1 and >6% in people over 80 years.2–4 The prevalence of AF is increasing,5 even after adjustment for age.6,7 Atrial fibrillation is associated with reduced life expectancy.8,9 Atrial fibrillation increases the risk of thrombo-embolic stroke,8 which in turn increases the risk of death.10,11 This is reflected by the inclusion of stroke as a main outcome in many clinical trials testing antithrombotic therapies in AF.12–27 On the other hand, major haemorrhage has generally been of lesser concern and usually not included in the primary study outcomes.

This approach to clinical trial design has recently been challenged. Patients with AF may die of non-stroke-related vascular events, such as myocardial infarction (MI), because of associated co-morbidities.28 Risk factors for non-embolic stroke are common in AF, and as many as 25% of strokes in patients with AF are estimated to be non-cardioembolic, due to intrinsic cerebrovascular disease (present in about 12% of such patients29), other cardioembolic sources, or atheroembolism from the proximal aorta.30,31 The impact of different vascular events on mortality in AF patients is unknown. Stroke severity also may vary from non-disabling to very severe, and the extent to which severity of stroke in AF influences mortality is unknown.32

Therefore, we examined the risk of death after the occurrence of different types of non-fatal vascular events occurring in the Atrial Fibrillation Clopidogrel Trial with Irbesartan for Prevention of Vascular Events (ACTIVE)-W Trial, a large trial examining the effect of clopidogrel plus aspirin (C + A) vs. vitamin K antagonists (VKAs),33 with the purpose of understanding the clinical importance of different types of vascular events and of haemorrhage.

Methods

Details of the overall programme, study design, and the main results of the ACTIVE-W trial have been published.33 Patients eligible and willing to take a VKA were enrolled into ACTIVE-W, which compared C + A vs. VKAs. Patients were eligible for ACTIVE-W if they had electrocardiographic evidence of AF and also had at least one of the known risk factors for stroke. Patients were excluded if they had any of the following: requirement for VKAs (such as for prosthetic mechanical heart valves), documented peptic ulcer disease within the previous 6 months, prior intracerebral haemorrhage, significant thrombocytopenia (platelet count <50 × 109/L), or mitral stenosis. Patients randomized to VKAs could receive the VKA in use in their country and were monitored to keep the international normalized ratio (INR) between 2.0 and 3.0. Patients randomized to C + A received clopidogrel 75 mg once daily in addition to aspirin (recommended dose: 75–100 mg/day).33 The median follow-up for the whole of ACTIVE-W study was 1.28 years.

Study outcomes and definitions

The primary outcome of ACTIVE-W was the first occurrence of any of the following vascular events: stroke, non-central nervous system (CNS) systemic embolism, MI, or vascular death. Strokes were classified into those that were ischaemic or primarily haemorrhagic, and the severity of stroke was measured with the modified Rankin score at the time of discharge from hospital or at 7 days after the event. Subdural haematomas were included as intracranial haemorrhages, but not classified as haemorrhagic strokes. Strokes were defined as disabling when the modified Rankin score was ≥3. Myocardial infarction was diagnosed according to American Heart Association/American College of Cardiology guidelines. Vascular death was any death clearly not due to non-vascular cause such as trauma. Major bleeding was defined as any bleeding requiring transfusion of at least 2U of red blood cells or equivalent of whole blood, or any bleeding defined as severe. Severe bleeding was major bleeding that was also associated with any of the following: death, drop in haemoglobin of at least 5 g/dL, significant hypotension with the need for inotropic agents, intraocular bleeding leading to significant loss of vision, bleeding requiring surgical intervention (other than vascular site repair), symptomatic intracranial haemorrhage, or requirement for a transfusion of at least 4U of blood. We here, therefore, indicate as majornon-severe those bleeding events that were major, but not severe, according to the previous definitions. In practice, these were bleeds with a need for transfusions of at least 2 U but <4 U of blood. Non-fatal events were defined for this analysis as those not causing death within 7 days. There were 14 and 15 fatal strokes on C + A and VKAs, respectively, and 7 and 11 fatal haemorrhages, respectively, in the ACTIVE-W cohort, as reported previously.33 For the current analysis, we eliminated the immediately fatal (<7 days) strokes and bleeds, since—in ACTIVE-W case report forms—investigators would sometimes check an event as fatal also when the time to death was >7 days. The number of fatal events omitted from analysis by the treatment group were: fatal strokes = 20 (9 on VKAs and 11 on C + A); fatal haemorrhages = 14 (8 on VKAs and 6 on C + A).

Transient ischaemic attacks (TIAs) were not among the outcomes of the study. They were diagnosed by local neurologists as acute episodes of temporary neurological dysfunction ascribed to a vascular occlusion to a particular area of the brain that persisted for <24 h, and captured in the database for further analyses. Patients were excluded from the analysis if they had a fatal stroke or a fatal bleed (see definitions above).

Statistical analysis

The primary analysis was based on a time-to-event model using the non-fatal event as a time-dependent covariate. Cox's regression analysis was used to quantify the risk of non-fatal outcomes on subsequent within-trial death. These regressions were adjusted for age, history of coronary artery disease, history of stroke, diabetes, history of heart failure, and study medication. Hazard ratios (HRs) and their 95% confidence interval (CI) were calculated using Cox's regression.

Extended Kaplan–Meier estimator curves were constructed to display the effect of the non-fatal event as the time-dependent risk on time to mortality.34 All analyses were done related to all-cause mortality.

The P-value threshold for significance used was a two-sided value of 0.05. Analyses were carried out with the SAS version 9.1 statistical software.

Results

The effect of non-fatal stroke on mortality

The overall median follow-up time for patients in ACTIVE-W was 1.28 years.33 For the present analysis, however, follow-up after first non-fatal events was shorter, namely 0.51 years after a stroke, 0.53 years after an MI, 0.37 years after a non-CNS embolism, 0.59 years after a major bleed, and 0.82 years after a minor bleed. Overall, the occurrence of non-fatal stroke increased the risk of death; HR (95% CI): 9.3 (6.3–13.7), P < 0.0001 (Figure 1). The increase in relative risk (RR) was significant for both haemorrhagic stroke and ischaemic stroke (Figure 1 and Table 1). Transient ischaemic attacks had no effect on mortality (Figure 1 and Table 1), despite being predictors of a subsequent stroke (HR 4.37, 95% CI 1.78–10.73, P = 0.001).

Table 1

Effect of non-fatal strokes, myocardial infarctions, and non-central nervous system systemic embolisms on subsequent mortality (over the 1.28 years of the median follow-up duration of the study)

Event With the event
 
Without the event
 
Effect of outcome
 
 Total Deaths Mortality (%) Total Deaths Mortality (%) Adjusted HR (95% CI) P-value 
Strokes (any) 139 31 22.3 6567 287 4.4 9.3 (6.3–13.7) <0.0001 
 Ischaemic 122 26 21.3 6584 292 4.4 8.2 (5.4–12.5) <0.0001 
 Haemorrhagic 13 30.8 6693 314 4.7 20.8 (7.6–56.5) <0.0001 
 Disabling 78 27 34.6 6628 291 4.4 14.3 (9.5–21.7) <0.0001 
 Non-disabling 58 6.9 6648 314 4.7 2.5 (0.93–6.78) 0.0686 

 
TIAs 87 3.5 6619 315 4.8 1.0 (0.33–3.3) 0.9455 
Myocardial infarctions 53 12 22.6 6653 306 4.6 7.3 (4.1–13.2) <0.0001 
Non-CNS systemic embolisms 21 14.3 6685 315 4.7 6.5 (2.1–20.6) 0.0014 
Event With the event
 
Without the event
 
Effect of outcome
 
 Total Deaths Mortality (%) Total Deaths Mortality (%) Adjusted HR (95% CI) P-value 
Strokes (any) 139 31 22.3 6567 287 4.4 9.3 (6.3–13.7) <0.0001 
 Ischaemic 122 26 21.3 6584 292 4.4 8.2 (5.4–12.5) <0.0001 
 Haemorrhagic 13 30.8 6693 314 4.7 20.8 (7.6–56.5) <0.0001 
 Disabling 78 27 34.6 6628 291 4.4 14.3 (9.5–21.7) <0.0001 
 Non-disabling 58 6.9 6648 314 4.7 2.5 (0.93–6.78) 0.0686 

 
TIAs 87 3.5 6619 315 4.8 1.0 (0.33–3.3) 0.9455 
Myocardial infarctions 53 12 22.6 6653 306 4.6 7.3 (4.1–13.2) <0.0001 
Non-CNS systemic embolisms 21 14.3 6685 315 4.7 6.5 (2.1–20.6) 0.0014 

TIAs, transient ischaemic attacks; CNS, central nervous system; HR, hazard ratio; CI, confidence interval.

Figure 1

The impact of the type of non-fatal stroke (all strokes, haemorrhagic strokes, and ischaemic strokes) and transient ischaemic attacks (TIAs) (left panel) and of non-fatal stroke severity [severe strokes and non-severe strokes according to definitions (see text) (right panel)] on subsequent mortality in ACTIVE-W. The hazard ratios (HR) for death and 95% confidence intervals associated with each definition are depicted, together with the P-values.

Figure 1

The impact of the type of non-fatal stroke (all strokes, haemorrhagic strokes, and ischaemic strokes) and transient ischaemic attacks (TIAs) (left panel) and of non-fatal stroke severity [severe strokes and non-severe strokes according to definitions (see text) (right panel)] on subsequent mortality in ACTIVE-W. The hazard ratios (HR) for death and 95% confidence intervals associated with each definition are depicted, together with the P-values.

Of the 139 strokes occurring during the trial, 56% (n = 78) were disabling (Rankin's score ≥ 3) and 42% (n = 58) were not (Rankin's score 1–2); 3 strokes were not classified in severity. There was an increase in the RR of death after a disabling stroke (HR 14.3, 95% CI 9.5–21.7, P < 0.0001), but not after a non-disabling stroke (HR 2.5, 95% CI 0.93–6.78, P = 0.07; Figure 1 and Table 1).

The impact of haemorrhage on mortality

The occurrence of any non-fatal bleeding increased the risk of death [HR (95% CI): 2.5 (1.8–3.5), P < 0.001]. Non-fatal major bleeding especially had an important impact on subsequent mortality [HR (95% CI): 4.2 (2.8–6.4), P < 0.001], whereas minor bleeding had a weaker, albeit still significant, effect [HR (95% CI): 1.6 (1.0–2.5, P = 0.04); Table 2]. A pre-specified classification of major bleeding events in ACTIVE designated these bleeds as severe or not. Only severe bleeding events had a significant impact on mortality (HR 5.7, 95% CI 3.6–9.1, P < 0.0001; Table 2 and Figure 2). Major, but non-severe, bleeds did not affect mortality significantly.

Table 2

Effect of non-fatal bleeding events on subsequent mortality (over the 1.28 years of the median follow-up duration of the study)

Event With the event
 
Without the event
 
Effect of outcome
 
 Total Deaths Mortality (%) Total Deaths Mortality (%) Adjusted HR (95% CI) -value 
Bleeding events (any) 593 46 7.8 6113 272 4.5 2.5 (1.8–3.5) <0.0001 
 Major 181 24 13.3 6526 294 4.5 4.2 (2.8–6.4) <0.0001 
  Major, severe 123 20 16.3 6583 298 4.5 5.7 (3.6–9.1) <0.0001 
  Major non-severe 58 6.9 6648 314 4.7 1.7 (0.64–4.7) 0.2827 
 Minor 412 22 5.3 6294 296 4.7 1.6 (1.0–2.5) 0.0361 
Event With the event
 
Without the event
 
Effect of outcome
 
 Total Deaths Mortality (%) Total Deaths Mortality (%) Adjusted HR (95% CI) -value 
Bleeding events (any) 593 46 7.8 6113 272 4.5 2.5 (1.8–3.5) <0.0001 
 Major 181 24 13.3 6526 294 4.5 4.2 (2.8–6.4) <0.0001 
  Major, severe 123 20 16.3 6583 298 4.5 5.7 (3.6–9.1) <0.0001 
  Major non-severe 58 6.9 6648 314 4.7 1.7 (0.64–4.7) 0.2827 
 Minor 412 22 5.3 6294 296 4.7 1.6 (1.0–2.5) 0.0361 

HR, hazard ratio; CI, confidence interval.

Figure 2

The impact of non-fatal bleeding events [all bleeding, severe bleeding, major non-severe bleeding, major bleeding, and minor bleeding according to definitions (see text)] on subsequent mortality in ACTIVE-W. The hazard ratio (HR) for death and 95% confidence intervals associated with each definition are depicted, together with the P-values.

Figure 2

The impact of non-fatal bleeding events [all bleeding, severe bleeding, major non-severe bleeding, major bleeding, and minor bleeding according to definitions (see text)] on subsequent mortality in ACTIVE-W. The hazard ratio (HR) for death and 95% confidence intervals associated with each definition are depicted, together with the P-values.

Figure 3 shows the subsequent mortality rate of patients experiencing non-fatal events compared with patients who had not had an event. This is shown for disabling and non-disabling strokes, as well as for severe haemorrhages and for major, but non-severe haemorrhages. The lack of effect of a major non-severe haemorrhage (as opposed to that of severe haemorrhage) on subsequent mortality was also evident at all time points throughout the duration of the follow-up (Figure 3).

Figure 3

Late mortality following immediate survival from stroke or haemorrhage according to stroke or bleeding severity. Note the impact on later mortality of a disabling stroke and of severe haemorrhages, but not of a non-disabling stroke and of a major-non-severe bleeding event.

Figure 3

Late mortality following immediate survival from stroke or haemorrhage according to stroke or bleeding severity. Note the impact on later mortality of a disabling stroke and of severe haemorrhages, but not of a non-disabling stroke and of a major-non-severe bleeding event.

The impact of other non-fatal outcomes on mortality

The occurrence of MI was associated with a subsequent increase in mortality, the HR being 7.3 (95% CI 4.1–13.2, P < 0.0001; Table 1). A non-CNS embolism (HR 6.5, 95% CI 2.1–20.6, P = 0.0014) also was associated with increased mortality.

The impact of treatment on death and on non-fatal events having a subsequent impact on mortality

Having determined that disabling strokes, MI, and severe bleeding events were significantly and strongly (HR > 3.0) associated with subsequent increased mortality, we estimated the cumulative number of such events (which we now term ‘severe events’, having an important impact on mortality), with or without vascular death or total death (Figure 4), as a function of treatments tested in ACTIVE-W. Clopidogrel plus aspirin was associated with a significantly higher number of severe events (not including death) than VKAs [RR = 1.39 (1.09–1.77), P = 0.0084]. When vascular death was included in this analysis, the cumulative outcome of disabling stroke, MI, severe bleeding, and vascular death was significantly higher with C + A than with VKAs [RR = 1.28 (1.06–1.55), P = 0.0121; Figure 4]. There was a trend towards fewer outcomes of disabling stroke, MI, severe bleeding, and total death with VKAs compared with C + A [RR = 1.17 (0.98–1.40), P = 0.078].

Figure 4

The impact of therapy (clopidogrel + aspirin, C + A) vs. vitamin K antagonists (VKAs) (Kaplan–Meier curves) on the cumulative incidence of non-fatal events having a significant impact on mortality (disabling stroke, myocardial infarction, and severe bleeding) (left panel) and on the cumulative incidence of non-fatal events having a significant impact on mortality (disabling stroke, myocardial infarction, and severe bleeding) plus vascular death. RR indicates the relative risk along with its 95% confidence interval and significance (P-value).

Figure 4

The impact of therapy (clopidogrel + aspirin, C + A) vs. vitamin K antagonists (VKAs) (Kaplan–Meier curves) on the cumulative incidence of non-fatal events having a significant impact on mortality (disabling stroke, myocardial infarction, and severe bleeding) (left panel) and on the cumulative incidence of non-fatal events having a significant impact on mortality (disabling stroke, myocardial infarction, and severe bleeding) plus vascular death. RR indicates the relative risk along with its 95% confidence interval and significance (P-value).

Discussion

This report indicates that in the AF population recruited in the large ACTIVE-W study, the risk of death is increased by disabling strokes, MI, and severe bleeding events. However, non-disabling strokes and major non-severe bleeding events appear not to increase mortality. These findings emphasize the importance for clinicians to appropriately assess the risk–benefit ratio for treatments commonly used in AF and suggest that evaluation of the net benefits of antithrombotic agents in patients with AF should be based on the composite of outcomes that are more prognostically important rather than simply on the general category of ‘strokes’ and ‘bleeding events’. Such results may also have relevance in the planning of future clinical studies, both within and outside the AF setting, by refining the mortality predictions based on the characteristics of the entry populations.

Proper assessment of the risk–benefit ratio for commonly used therapies in atrial fibrillation

We found here that only some types of commonly considered ‘major’ outcomes have an effect on mortality. Severe haemorrhages, as defined in ACTIVE-W, impacted mortality, whereas some strokes did not. Our analysis was adjusted for a history of hypertension—which might have influenced the severity of both strokes and bleeding—in the model, although we recognize a limitation in this analysis, in that >80% of the ACTIVE-W population had such a history. Our analysis also was adjusted for other covariates likely affecting mortality, such as age, history of coronary artery disease, history of stroke, diabetes, history of heart failure, and study medications. In any case, our findings contrast previous belief that strokes are often disabling. Since clinicians prescribing antithrombotic treatments in their patients are constantly confronted with the problem of the choice between the risk of bleeding and the risk of thrombosis, knowing what events are really relevant and what are less relevant can help in decision-making. Doctors should also know that such evaluation may affect patients’ events not remotely in time, but relatively soon, since our analysis is based on the relatively short follow-up of little more than 1 year (1.28 years as median) allowed by our study. Such predictions may, therefore, have direct practical implications.

Outcomes and mortality in atrial fibrillation

Most previous studies in AF have focused on stroke or the combination of stroke and systemic embolism,12–27 and not on mortality, as the primary outcome. This choice is due in part to understandable pathophysiological considerations, directly relating blood stasis in the left atrium with the occurrence of CNS and non-CNS systemic embolism and (ischaemic) stroke,10,11 and in part to the more frequent rate of strokes (most of which are non-fatal) and TIAs when compared with death, making comparisons of different therapies feasible even with a relatively small number of recruited patients. However, the severity of strokes may vary enormously in terms of functional consequences and prognostic importance, from strokes with no or little permanent impairment to disabling, or fatal. We found that TIAs have no impact on subsequent mortality, despite being themselves predictors of a subsequent stroke. One reason for this may lie in the somewhat uncertain diagnosis of TIAs, likely also including non-cerebrovascular events. However, also non-disabling strokes, which were 42% of all strokes in this study and which are of much less uncertain diagnosis, did not impact significantly on mortality. These findings suggest that non-disabling strokes (and TIAs) should not be grouped together with disabling strokes in outcome AF studies, because such two types have quite different consequences. Similar considerations pertain to major bleeding events, which here increased mortality only if ‘severe’, whereas even bleeding events termed as major, but not-severe did not affect prognosis. These, in the definitions adopted in ACTIVE, involve the need for transfusions of ≥2 U but <4 U of blood. The effect of non-fatal disabling strokes and of non-fatal severe bleeding events (as opposed to that of non-fatal non-disabling strokes and non-fatal minor or even major, but not severe, bleeding events) on subsequent death was constant over time, with a progressive separation of the Kaplan–Meier curves, indicating that both these types of prognostically relevant events started immediately to detrimentally affect prognosis and continued to affect subsequent death at a relatively constant rate thereafter.

Impact of non-stroke events on mortality

Since the risk factors for AF largely overlap with those for coronary artery disease and vascular disease in general (age, male gender, hypertension, and diabetes among others), it is important to consider that AF patients are likely to die from causes unrelated to thrombo-embolic stroke, including MI. Myocardial infarction did indeed occur in 0.9% of patients throughout the study follow-up and had a significant impact on mortality (HR 7.3, 4.1–13.2, P < 0.0001).

The impact of bleeding on mortality

We attempted at defining the relevance of bleeding severity for subsequent mortality and the differential effects of the two tested treatment strategies on such events, in the light of the recent increasing emphasis on bleeding as a determinant of prognosis in trials of antithrombotic agents.35,36 As for stroke, the overall category of bleeding encompasses widely different severities, with some bleeding events—even within the category defined as ‘major'—having little—if any—prognostic significance. Major bleeding events defined as severe (68% of all major bleeds) significantly increased mortality. Since the difference between what were here defined as severe bleeding (happening to have an impact on mortality) and majornon-severe bleeding (having no impact on mortality) essentially lies in the need of transfusions (≥2 U but <4 U of blood in the case of majornon-severe bleeding), our findings underscore the inappropriateness of including minor transfusion needs as a relevant outcome in AF trials. Here, the two treatment strategies tested had a differential impact on minor bleeding, which was significantly (and surprisingly) more frequent with C + A, but not on major or severe bleeding.33

Treatment impact on strokes in ACTIVE-W

The inappropriateness of grouping all strokes into one single category is further illustrated by the differential impact of the two treatment strategies tested in ACTIVE-W on stroke subtypes. Here, VKAs, with INR values in the therapeutic range (2.0–3.0) 63.8% of the time,33 were overall superior to C + A on total strokes, but their impact was larger on the less severe type of strokes. Vitamin K antagonists were actually associated with an increased incidence in primarily haemorrhagic strokes, likely not related to AF per se, and thereby likely caused by the adverse consequences of VKAs on haemostasis. This is in line with previous reports on the relative safety of antiplatelet therapies (even combining two antiplatelet agents such as C and A) from the standpoint of intracranial haemorrhage,37 whereas treatment with VKAs have long been known to be affected by a substantially increased number of this event.38

Vitamin K antagonists were superior to C + A in preventing all strokes (ACTIVE-W main findings). However, the benefit of VKAs over C + A decreases with increasing stroke severity (P = 0.036). The RR for non-disabling stroke (modified Rankin's score 0–2) was 2.49, but for disabling strokes (modified Rankin's score 3–6) or fatal strokes was only 1.47. Thus, compared with C + A, VKAs were significantly more likely to prevent a minor stroke than a more serious one.

Overall impact of vitamin K antagonists and clopidogrel plus aspirin on prognostically relevant outcomes and clinical implications

ACTIVE-W was terminated early because of a clear superiority of VKAs over C + A on the primary outcome of the study, which was the first occurrence of stroke, non-CNS systemic embolism, MI, or vascular death. However, there was no difference on mortality.33 The negative impact of VKAs on haemorrhagic strokes, which substantially increases the risk of death, certainly contributed to the neutral effects on mortality. However, the number of deaths in ACTIVE-W was only 318, and the 95% CI do not exclude a modest differential impact of VKAs vs. C + A on mortality. It was, therefore, of some importance to distinguish between a trial truly negative and a trial simply not powered enough to show differences in death rates between the two treatments tested, despite the sample size larger than for any of previous AF trials. When all non-fatal events subsequently substantially affecting mortality were grouped together, or all these were put together with total vascular death, VKAs were superior to C + A (to a statistically significant extent for the latter combined outcome). A similar trend, close to statistical significance, was apparent when total death was also included in the grouping. These analyses are exploratory, since they are the results of a post hoc analysis of the trial and do not ascertain whether VKAs were superior to C + A on mortality. They, however, show that VKAs, especially when adequately controlled,39 may be indeed superior to C + A on events prognostically related to mortality. The definite proof of an advantage of VKAs on mortality would have required a substantially larger and longer trial, which would not be interrupted due to a difference in strokes or other non-fatal events. It is unlikely that any such trial will ever be undertaken again, thus supporting the need of gathering the information presented here.

Limitations

The observations from this study pertain to a selected AF population and may not necessarily be applied to the general patient population with this condition. This is a limit of the present study as well as of all studies specifying inclusion and exclusion criteria that may render the study population different from the real world. In addition, however, ACTIVE-W was done in a population eligible for therapy with VKAs and to a large extent warfarin-experienced. Confirmation of the different prognostic value of subtypes of strokes and bleeding events in different populations, e.g. in patients judged unsuitable to VKAs, is certainly worthwhile and is actually being planned in the ACTIVE-A cohort. Despite being one of the largest studies in AF conducted so far, the number of deaths in this study was still relatively small. Such an analysis should be, therefore, repeated in future studies to replicate the findings.

Funding

The ACTIVE Study Programme has been entirely funded by sanofi-aventis and Bristol Myers-Squibb.

Conflict of interest: R.D.C. has received honoraries and grant support from Sanofi-Aventis and Bristol Myers-Squibb; S.J.C. has received research grants and speaking fees from both Sanofi Aventis and Bristol Myers-Squibb; A.B. has received research grants and honoraria from Sanofi-Aventis; S.Y. has received research grants and speaking fees from both Sanofi Aventis and Bristol Myers-Squibb.

References

1
Ostranderld
J
Brandt
R
Kjelsberg
M
Epstein
F
Electrocardiographic findings among the adult population of a total natural community, Tecumseh, Michigan
Circulation
 , 
1965
, vol. 
31
 (pg. 
888
-
898
)
2
Flegel
KM
Shipley
MJ
Rose
G
Risk of stroke in non-rheumatic atrial fibrillation
Lancet
 , 
1987
, vol. 
1
 (pg. 
526
-
529
)
3
Wolf
PA
Abbott
RD
Kannel
WB
Atrial fibrillation as an independent risk factor for stroke: the Framingham Study
Stroke
 , 
1991
, vol. 
22
 (pg. 
983
-
988
)
4
Furberg
CD
Psaty
BM
Manolio
TA
Gardin
JM
Smith
VE
Rautaharju
PM
Prevalence of atrial fibrillation in elderly subjects (the Cardiovascular Health Study)
Am J Cardiol
 , 
1994
, vol. 
74
 (pg. 
236
-
241
)
5
Friberg
J
Buch
P
Scharling
H
Gadsbphioll
N
Jensen
GB
Rising rates of hospital admissions for atrial fibrillation
Epidemiology
 , 
2003
, vol. 
14
 (pg. 
666
-
672
)
6
Wolf
PA
Benjamin
EJ
Belanger
AJ
Kannel
WB
Levy
D
D'Agostino
RB
Secular trends in the prevalence of atrial fibrillation: The Framingham Study
Am Heart J
 , 
1996
, vol. 
131
 (pg. 
790
-
795
)
7
Tsang
TS
Petty
GW
Barnes
ME
O'Fallon
WM
Bailey
KR
Wiebers
DO
Sicks
JD
Christianson
TJ
Seward
JB
Gersh
BJ
The prevalence of atrial fibrillation in incident stroke cases and matched population controls in Rochester, Minnesota: changes over three decades
J Am Coll Cardiol
 , 
2003
, vol. 
42
 (pg. 
93
-
100
)
8
Risk factors for stroke and efficacy of antithrombotic therapy in atrial fibrillation. Analysis of pooled data from five randomized controlled trials
Arch Intern Med
 , 
1994
, vol. 
154
 (pg. 
1449
-
1457
)
9
Stewart
S
Hart
CL
Hole
DJ
McMurray
JJ
A population-based study of the long-term risks associated with atrial fibrillation: 20-year follow-up of the Renfrew/Paisley study
Am J Med
 , 
2002
, vol. 
113
 (pg. 
359
-
364
)
10
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
ACC/AHA/ESC 2006 guidelines for the management of patients with atrial fibrillation-executive summary: 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 (Writing Committee to Revise the 2001 Guidelines for the Management of Patients with Atrial Fibrillation)
Eur Heart J
 , 
2006
, vol. 
27
 (pg. 
1979
-
2030
)
11
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
Hunt
SA
Nishimura
R
Ornato
JP
Page
RL
Riegel
B
Priori
SG
Blanc
JJ
Budaj
A
Camm
AJ
Dean
V
Deckers
JW
Despres
C
Dickstein
K
Lekakis
J
McGregor
K
Metra
M
Morais
J
Osterspey
A
Tamargo
JL
Zamorano
JL
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 (Writing Committee to Revise the 2001 Guidelines for the Management of Patients With Atrial Fibrillation): developed in collaboration with the European Heart Rhythm Association and the Heart Rhythm Society
Circulation
 , 
2006
, vol. 
114
 (pg. 
e257
-
e354
)
12
Albers
GW
Diener
HC
Frison
L
Grind
M
Nevinson
M
Partridge
S
Halperin
JL
Horrow
J
Olsson
SB
Petersen
P
Vahanian
A
Ximelagatran vs warfarin for stroke prevention in patients with nonvalvular atrial fibrillation: a randomized trial
JAMA
 , 
2005
, vol. 
293
 (pg. 
690
-
698
)
13
Connolly
SJ
Laupacis
A
Gent
M
Roberts
RS
Cairns
JA
Joyner
C
Canadian Atrial Fibrillation Anticoagulation (CAFA) Study
J Am Coll Cardiol
 , 
1991
, vol. 
18
 (pg. 
349
-
355
)
14
Diener
HC
Stroke prevention using the oral direct thrombin inhibitor ximelagatran in patients with non-valvular atrial fibrillation. Pooled analysis from the SPORTIF III and V studies
Cerebrovasc Dis
 , 
2006
, vol. 
21
 (pg. 
279
-
293
)
15
EAFT (European Atrial Fibrillation Trial) Study Group
Secondary prevention in non-rheumatic atrial fibrillation after transient ischaemic attack or minor stroke
Lancet
 , 
1993
, vol. 
342
 (pg. 
1255
-
1262
)
16
Ezekowitz
MD
Bridgers
SL
James
KE
Carliner
NH
Colling
CL
Gornick
CC
Krause-Steinrauf
H
Kurtzke
JF
Nazarian
SM
Radford
MJ
Warfarin in the prevention of stroke associated with nonrheumatic atrial fibrillation. Veterans Affairs Stroke Prevention in Nonrheumatic Atrial Fibrillation Investigators
. N Engl J Med
 , 
1992
, vol. 
327
 (pg. 
1406
-
1412
)
17
Gullov
AL
Koefoed
BG
Petersen
P
Pedersen
TS
Andersen
ED
Godtfredsen
J
Boysen
G
Fixed minidose warfarin and aspirin alone and in combination vs adjusted-dose warfarin for stroke prevention in atrial fibrillation: Second Copenhagen Atrial Fibrillation, Aspirin, and Anticoagulation Study
Arch Intern Med
 , 
1998
, vol. 
158
 (pg. 
1513
-
1521
)
18
Hellemons
BS
Langenberg
M
Lodder
J
Vermeer
F
Schouten
HJ
Lemmens
T
van Ree
JW
Knottnerus
JA
Primary prevention of arterial thromboembolism in non-rheumatic atrial fibrillation in primary care: randomised controlled trial comparing two intensities of coumarin with aspirin
BMJ
 , 
1999
, vol. 
319
 (pg. 
958
-
964
)
19
Morocutti
C
Amabile
G
Fattapposta
F
Nicolosi
A
Matteoli
S
Trappolini
M
Cataldo
G
Milanesi
G
Lavezzari
M
Pamparana
F
Coccheri
S
Indobufen versus warfarin in the secondary prevention of major vascular events in nonrheumatic atrial fibrillation. SIFA (Studio Italiano Fibrillazione Atriale) Investigators
Stroke
 , 
1997
, vol. 
28
 (pg. 
1015
-
1021
)
20
Olsson
SB
Stroke prevention with the oral direct thrombin inhibitor ximelagatran compared with warfarin in patients with non-valvular atrial fibrillation (SPORTIF III): randomised controlled trial
Lancet
 , 
2003
, vol. 
362
 (pg. 
1691
-
1698
)
21
Pengo
V
Zasso
A
Barbero
F
Banzato
A
Nante
G
Parissenti
L
John
N
Noventa
F
Dalla Volta
S
Effectiveness of fixed minidose warfarin in the prevention of thromboembolism and vascular death in nonrheumatic atrial fibrillation
Am J Cardiol
 , 
1998
, vol. 
82
 (pg. 
433
-
437
)
22
Perez-Gomez
F
Alegria
E
Berjon
J
Iriarte
JA
Zumalde
J
Salvador
A
Mataix
L
Comparative effects of antiplatelet, anticoagulant, or combined therapy in patients with valvular and nonvalvular atrial fibrillation: a randomized multicenter study
J Am Coll Cardiol
 , 
2004
, vol. 
44
 (pg. 
1557
-
1566
)
23
Petersen
P
Boysen
G
Godtfredsen
J
Andersen
ED
Andersen
B
Placebo-controlled, randomised trial of warfarin and aspirin for prevention of thromboembolic complications in chronic atrial fibrillation. The Copenhagen AFASAK study
Lancet
 , 
1989
, vol. 
1
 (pg. 
175
-
179
)
24
Stroke Prevention in Atrial Fibrillation II Study Investigators
Warfarin versus aspirin for prevention of thromboembolism in atrial fibrillation: Stroke Prevention in Atrial Fibrillation II Study
Lancet
 , 
1994
, vol. 
343
 (pg. 
687
-
691
)
25
Stroke Prevention in Atrial Fibrillation Study Investigators
Stroke Prevention in Atrial Fibrillation Study Final results
Circulation
 , 
1991
, vol. 
84
 (pg. 
527
-
539
)
26
Stroke Prevention in Atrial Fibrillation Study Investigators
Adjusted-dose warfarin versus low-intensity, fixed-dose warfarin plus aspirin for high-risk patients with atrial fibrillation: Stroke Prevention in Atrial Fibrillation III randomised clinical trial
Lancet
 , 
1996
, vol. 
348
 (pg. 
633
-
638
)
27
The Boston Area Anticoagulation Trial for Atrial Fibrillation Investigators
The effect of low-dose warfarin on the risk of stroke in patients with nonrheumatic atrial fibrillation
N Engl J Med
 , 
1990
, vol. 
323
 (pg. 
1505
-
1511
)
28
Wyse
DG
Gersh
BJ
Atrial fibrillation: a perspective: thinking inside and outside the box
Circulation
 , 
2004
, vol. 
109
 (pg. 
3089
-
3095
)
29
Kanter
MC
Tegeler
CH
Pearce
LA
Weinberger
J
Feinberg
WM
Anderson
DC
Gomez
CR
Rothrock
JF
Helgason
CM
Hart
RG
Carotid stenosis in patients with atrial fibrillation. Prevalence, risk factors, and relationship to stroke in the Stroke Prevention in Atrial Fibrillation Study
. Arch Intern Med
 , 
1994
, vol. 
154
 (pg. 
1372
-
1377
)
30
Bogousslavsky
J
Van Melle
G
Regli
F
Kappenberger
L
Pathogenesis of anterior circulation stroke in patients with nonvalvular atrial fibrillation: the Lausanne Stroke Registry
Neurology
 , 
1990
, vol. 
40
 (pg. 
1046
-
1050
)
31
Miller
VT
Rothrock
JF
Pearce
LA
Feinberg
WM
Hart
RG
Anderson
DC
Ischemic stroke in patients with atrial fibrillation: effect of aspirin according to stroke mechanism. Stroke Prevention in Atrial Fibrillation Investigators
. Neurology
 , 
1993
, vol. 
43
 (pg. 
32
-
36
)
32
Hankey
GJ
Long-term outcome after ischaemic stroke/transient ischaemic attack
Cerebrovasc Dis
 , 
2003
, vol. 
16
 
Suppl. 1)
(pg. 
14
-
19
)
33
Connolly
S
Pogue
J
Hart
R
Pfeffer
M
Hohnloser
S
Chrolavicius
S
Pfeffer
M
Hohnloser
S
Yusuf
S
Clopidogrel plus aspirin versus oral anticoagulation for atrial fibrillation in the Atrial fibrillation Clopidogrel Trial with Irbesartan for prevention of Vascular Events (ACTIVE W): a randomised controlled trial
Lancet
 , 
2006
, vol. 
367
 (pg. 
1903
-
1912
)
34
Snapinn
S
Jiang
Q
Iglewicz
B
Illustrating the impact of a time-varying covariate with an extended Kaplan-Meier estimator
Am Stat
 , 
2005
, vol. 
59
 (pg. 
301
-
307
)
35
Yusuf
S
Mehta
SR
Chrolavicius
S
Afzal
R
Pogue
J
Granger
CB
Budaj
A
Peters
RJ
Bassand
JP
Wallentin
L
Joyner
C
Fox
KA
Comparison of fondaparinux and enoxaparin in acute coronary syndromes
N Engl J Med
 , 
2006
, vol. 
354
 (pg. 
1464
-
1476
)
36
Stone
GW
McLaurin
BT
Cox
DA
Bertrand
ME
Lincoff
AM
Moses
JW
White
HD
Pocock
SJ
Ware
JH
Feit
F
Colombo
A
Aylward
PE
Cequier
AR
Darius
H
Desmet
W
Ebrahimi
R
Hamon
M
Rasmussen
LH
Rupprecht
HJ
Hoekstra
J
Mehran
R
Ohman
EM
Bivalirudin for patients with acute coronary syndromes
N Engl J Med
 , 
2006
, vol. 
355
 (pg. 
2203
-
2216
)
37
Patrono
C
Bachmann
F
Baigent
C
Bode
C
De Caterina
R
Charbonnier
B
Fitzgerald
D
Hirsh
J
Husted
S
Kvasnicka
J
Montalescot
G
Garcia Rodriguez
LA
Verheugt
F
Vermylen
J
Wallentin
L
Priori
SG
Alonso Garcia
MA
Blanc
JJ
Budaj
A
Cowie
M
Dean
V
Deckers
J
Fernandez Burgos
E
Lekakis
J
Lindahl
B
Mazzotta
G
Morais
J
Oto
A
Smiseth
OA
Morais
J
Deckers
J
Ferreira
R
Mazzotta
G
Steg
PG
Teixeira
F
Wilcox
R
Expert consensus document on the use of antiplatelet agents. The task force on the use of antiplatelet agents in patients with atherosclerotic cardiovascular disease of the European society of cardiology
Eur Heart J
 , 
2004
, vol. 
25
 (pg. 
166
-
181
)
38
De Caterina
R
Husted
S
Wallentin
L
Agnelli
G
Bachmann
F
Baigent
C
Jespersen
J
Kristensen
SD
Montalescot
G
Siegbahn
A
Verheugt
FW
Weitz
J
Anticoagulants in heart disease: current status and perspectives
Eur Heart J
 , 
2007
, vol. 
28
 (pg. 
880
-
913
)
39
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
, vol. 
349
 (pg. 
1019
-
1026
)

Supplementary data

Comments

0 Comments