In a recent issue of the Archives, one notes how the article on the Second Copenhagen Atrial Fibrillation, Aspirin, and Anticoagulation Study (AFASAK 2) by Gulløv et al1 and Albers'2 editorial conceal the most striking data from AFASAK 2, namely, that warfarin was not statistically significantly superior to aspirin alone for stroke prevention in atrial fibrillation. The cumulative primary event rates were 3.6% with aspirin and 2.8% with adjusted-dose warfarin sodium (P=.67). The introductory dependent clause in the "Conclusions" section of the Gulløv et al abstract— "Although the difference was insignificant . . ." —perhaps should be gist of the final conclusion vis-à-vis AFASAK 2's aspirin vs warfarin. The abstract's "Conclusions" section, however, cloaks the data regarding warfarin vs aspirin alone. No mention at all is made of the above findings. Albers' editorial likewise footnotes his position with no mention, for example, of another important recent meta-analysis about stroke, anticoagulation, and atrial fibrillation3 in which the margin between the benefit and harm for warfarin prophylaxis in patients with chronic nonvalvular atrial fibrillation was shown to be uncomfortably thin.
Bloom JM. Warfarin vs Aspirin and AFASAK 2. Arch Intern Med. 1999;159(9):1010. doi: