Fig shows the percentage of patients with
Fig. 2 shows the percentage of patients with high D-dimer levels in relation to the CHA2DS2–VASc scores and BNP levels. None of the patients had a CHA2DS2–VASc score of 0. The BNP levels were elevated in 3 (38%) of the 8 patients with CHA2DS2–VASc scores of 1, and the D-dimer levels were elevated in 2 of these 3 patients. Among the patients with CHA2DS2–VASc scores of 1, none of the 5 patients whose BNP levels were not elevated had elevated D-dimer levels.
Discussion BNP levels have been shown to be elevated in patients with AF [11,12], even in those who have preserved left ventricular systolic function. As shown in Figs. 1 and 2, the D-dimer levels were elevated in some patients with low CHADS2 or CHA2DS2–VASc scores who did not have CHF but had elevated BNP levels. Using the RE-LY database, Hijazi et al.  recently reported that Ozagrel the annual rate of composite thromboembolic events was even higher in AF patients with high (top-quartile) NT-proBNP levels and CHADS2 scores of 0 or 1 than in patients with CHADS2 scores ≥3 and low (lowest-quartile) NT-proBNP levels. Although this study used D-dimer levels as a surrogate marker for thromboembolisms, our results were consistent with those of Hijazi et al. . It can be speculated that latent CHF with a possible hypercoagulable state was undiagnosed in patients with few clinical risk factors, although some other factors that might cause a hypercoagulable state other than CHF might coexist.
Limitations First, although high D-dimer levels≥0.5μg/mL could predict thromboembolic events in patients with AF during anticoagulation therapy , Receptor is not known whether these levels could be applied to patients who are not undergoing anticoagulant therapy. Second, we performed a post hoc analysis with a small number of patients; thus, there might have been selection biases. Therefore, drawing definite conclusions was not possible.
Conflict of interest