Risk vs. benefit of anticoagulation therapy in elderly patients with atrial fibrillation and documented ground-level falls (2015)

Böhm A, Michalek P, Slezak P, Stevlik J, Jackuliak P, Stevove M, Uher T

Introduction: Patients with atrial fibrillation have a five-fold increased risk of stroke, which can be effectively reduced with oral anticoagulant therapy. However, elderly patients with atrial fibrillation and tendency to ground-level falls are often deprived of the treatment for fear of bleeding. Clinicians prescribing oral anticoagulants often face a dilemma in these situations.

Purpose: Previous studies on this issue typically used “risk of falls” rather than documented falls or they got data about falls from hospital registries which usually provide unreliable information, because of the tendency not to record those events. This was the reason that led us to our “real world” research with personally acquired data.

Methods: Our study includes elderly patients from 2 medical institutions hospitalized for atrial fibrillation between the years 2008–2011 and treated with warfarin for at least 3 years. They were personally interviewed on the number of falls, episodes of spontaneous bleeding and bleeding outcomes, which were correlated with their medical reports. The Clopper-Pearson method was used for the exact two-sided confidence intervals for single proportion, and Fisher’s exact test for 2 by 2 contingency table analysis.

Results: Of the 204 patients monitored during 3 years, 23% (n=47) had a total of 94 falls. In both cohorts of patients with and without falls, the average CHA2DS2VASc score was 5 and HASBLED score was 3. Incidence of bleeding in the cohort with falls was significantly higher by 86.6% (95% confidence interval [CI] = 78.7% to 91.1%, p 0.0001), than in the cohort without falls. Incidence of minor bleeding (WHO grade 1) was also significantly higher in the cohort with falls by 16.9% ([CI] = 1.20% to 38.5%, p=0.0419), than in the cohort without falls. However, the incidence of severe bleeding (WHO grade 4) was significantly higher by 17.3% ([CI] = 6.17% to 37.8%, p=0.0023) in the cohort without falls, than in the cohort with falls. Incidence of severe bleeding (WHO grade 4) after a fall was 1.06% ([CI] = 0.03% to 5.79%).

Conclusions: Our study showed that the incidence of minor bleeding in patients on warfarin and documented falls is higher, but surprisingly incidence of severe bleeding is lower, than in patients without falls. This suggests that spontaneous bleeding is more dangerous than bleeding after a fall, indicating that HASBLED should be preferred over “the risk of falling”. Furthermore, the incidence of bleeding after a fall is quite low and according to our study we can recommend warfarin to patients with CHA2DS2VASc >3, despite the high risk of falls.