Reduction in Stroke and Thromboembolic Events in AF
Reduction in Stroke and Thromboembolic Events in AF
Of the 9,088 apixaban-assigned and 9,052 warfarin-assigned patients who received at least 1 dose of study drug, a total of 6,809 patients from the apixaban group and 6,588 from the warfarin group completed study drug at the end of trial (Figure)). Baseline characteristics of these patients were generally well balanced (Table I). For patients who completed double-blind treatment, 83% converted to open-label VKAs, and 8%, to another oral anticoagulant (such as dabigatran) (Table II).
(Enlarge Image)
Figure.
Diagram of early permanent and final study drug discontinuations.
Permanent premature study drug discontinuation occurred in 2,279 patients (25.1%) in the apixaban group, with 3.6% of the discontinuations due to death and 7.5% due to adverse events, versus 2,464 (27.2%) patients in the warfarin group, with 3.9% due to death and 8.1% due to adverse events.
At the end of the trial, there were 21 strokes or systemic emboli (4.02%/year) and 26 major bleeding (4.97%/year) events in the apixaban group and 5 strokes or systemic emboli (0.99%/year) and 10 major bleeding (1.97%/year) events in the warfarin group with a consistently higher rate per time interval and with most of the imbalance between groups being after the first week (Table III). A similar pattern of excess events in the apixaban (14 events) compared with the warfarin group (2 events) was observed when the analysis was restricted to the 84% of patients who completed treatment and received VKA therapy. In the 17% of patients not transitioned to a VKA early after the end of the trial, there were 7 events in the apixaban group and 3 in the warfarin group.
After adjusting for major predictors of the primary efficacy and safety outcomes, the results were similar with a 4.10-fold (95% CI 1.54–10.86) excess of stroke or systemic embolism and a 2.56-fold (95% CI 1.23–5.30) excess of major bleeding in the apixaban group.
Among patients who permanently discontinued study drug during the trial, there was no increase in the rate of events in the apixaban versus warfarin group (Table IV). In fact, there were numerically more events in the warfarin group, particularly in the first week after discontinuation (Table IV).
To provide another experience of initiation versus continuation of VKA therapy, we compared clinical outcomes in the warfarin-assigned patients at the beginning of the trial among the 3,888 warfarin-naive (initiation of warfarin) patients versus the 5,193 warfarin-experienced (continuation of warfarin) patients for the first 30 days. Patients starting warfarin (warfarin naive assigned to warfarin) had a higher rate of stroke or systemic embolism (5.41%/year) than warfarin-experienced patients (1.42%/year). This pattern was not seen with apixaban, where events were similar at approximately 1.5%/year in both the warfarin-naive and warfarin-experienced populations. Bleeding showed a similar, although less striking, excess in the first 300 days in the warfarin-naive population: 5.18%/year major bleeding rate in warfarin-naive versus 3.12%/year among warfarin-experienced patients. Bleeding with apixaban was approximately 2%/year in both warfarin-experienced and warfarin-naive patients.
Results
Of the 9,088 apixaban-assigned and 9,052 warfarin-assigned patients who received at least 1 dose of study drug, a total of 6,809 patients from the apixaban group and 6,588 from the warfarin group completed study drug at the end of trial (Figure)). Baseline characteristics of these patients were generally well balanced (Table I). For patients who completed double-blind treatment, 83% converted to open-label VKAs, and 8%, to another oral anticoagulant (such as dabigatran) (Table II).
(Enlarge Image)
Figure.
Diagram of early permanent and final study drug discontinuations.
Permanent premature study drug discontinuation occurred in 2,279 patients (25.1%) in the apixaban group, with 3.6% of the discontinuations due to death and 7.5% due to adverse events, versus 2,464 (27.2%) patients in the warfarin group, with 3.9% due to death and 8.1% due to adverse events.
Clinical Events After Discontinuation and During Transition at the End of Trial
At the end of the trial, there were 21 strokes or systemic emboli (4.02%/year) and 26 major bleeding (4.97%/year) events in the apixaban group and 5 strokes or systemic emboli (0.99%/year) and 10 major bleeding (1.97%/year) events in the warfarin group with a consistently higher rate per time interval and with most of the imbalance between groups being after the first week (Table III). A similar pattern of excess events in the apixaban (14 events) compared with the warfarin group (2 events) was observed when the analysis was restricted to the 84% of patients who completed treatment and received VKA therapy. In the 17% of patients not transitioned to a VKA early after the end of the trial, there were 7 events in the apixaban group and 3 in the warfarin group.
After adjusting for major predictors of the primary efficacy and safety outcomes, the results were similar with a 4.10-fold (95% CI 1.54–10.86) excess of stroke or systemic embolism and a 2.56-fold (95% CI 1.23–5.30) excess of major bleeding in the apixaban group.
Permanent Study Drug Discontinuation During the Trial
Among patients who permanently discontinued study drug during the trial, there was no increase in the rate of events in the apixaban versus warfarin group (Table IV). In fact, there were numerically more events in the warfarin group, particularly in the first week after discontinuation (Table IV).
Pattern of Events at the Beginning of the Trial According to Prior Warfarin use
To provide another experience of initiation versus continuation of VKA therapy, we compared clinical outcomes in the warfarin-assigned patients at the beginning of the trial among the 3,888 warfarin-naive (initiation of warfarin) patients versus the 5,193 warfarin-experienced (continuation of warfarin) patients for the first 30 days. Patients starting warfarin (warfarin naive assigned to warfarin) had a higher rate of stroke or systemic embolism (5.41%/year) than warfarin-experienced patients (1.42%/year). This pattern was not seen with apixaban, where events were similar at approximately 1.5%/year in both the warfarin-naive and warfarin-experienced populations. Bleeding showed a similar, although less striking, excess in the first 300 days in the warfarin-naive population: 5.18%/year major bleeding rate in warfarin-naive versus 3.12%/year among warfarin-experienced patients. Bleeding with apixaban was approximately 2%/year in both warfarin-experienced and warfarin-naive patients.
Source...