Results
The baseline demographics of the study population are shown in Table 1. Among the 101 patients randomized 50 were randomized to dabigatran (49.5%) and 51 to warfarin (50.5%). The groups were balanced with the exception of coronary artery disease and treatment of coronary artery disease that was more prevalent in the warfarin treated group.
Of the patients enrolled in the study, 64% of those randomized to dabigatran and 61% of those randomized to warfarin completed the study. Supplemental Table 1 shows the reasons provided for discontinuing drug therapy.
Table 2 shows the baseline demographics of the study population that completed the study. In the dabigatran group, more patients had worse subtypes of AF (persistent and longstanding persistent) compared to the group treated with warfarin and a history of cancer was more common in the warfarin group.
The two year results data and secondary endpoints of Stroke or Transient ischemic attack (TIA), intracranial bleed, and changes from baseline scores on the mini-mental status evaluation and the Hachinski Ischemic Scale are shown in Table 3. No differences were observed in is rates of dementia or cognitive decline as measured by serial assessment. The incidence of new stroke and TIA were similar between groups and there were no significant changes detected in the MMSE or HIS surveys. During this period of time, 4 patients underwent a cardioversion, 1 received a pacemaker, none underwent an ablation, and 6 were start on antiarrhythmic drug therapy.
The survey scores of multiple tests for dementia, stroke, cognitive decline, and general function are shown in Table 4 and shown for 4 surveys in Figure 1. The scores were similar at baseline and 24 months without significant differences observed in the dabigatran versus warfarin groups amongst all surveys.
Finally, the results of the biomarker data are listed in Table 5 which were similar at baseline and 24 months that suggest similar efficacy of anticoagulation and consistent impact of it with both anticoagulation strategies. In addition, the data showed similar levels of Neprilysin and the single-nucleotide polymorphism rs6656401 and Complement Receptor-1gene that may have adversely impacted cerebral amyloid angiopathy risk. Additional APOE profiling was similar amongst groups and did not associate with abnormal cognitive survey scores at 24 months.
In the subgroup of patients that underwent MRI testing, there were no new strokes (clinical or subclinical) identified during the study follow-up or quantifiable change in volume (supplemental Table 2).