We have read with interest the article by Róldan and colleagues  demonstrating that high sensitivity troponin T and interleukin-6 were significantly associated with prognosis (both all-cause mortality and adverse cardiovascular events) in patients with atrial fibrillation (AF), even after adjusting for CHADS2 score. Therefore, the authors have proposed that these markers may potentially be used for refining clinical risk stratification of atrial fibrillation.
However, this group claims that despite the possible coronary microvascular dysfunction there is no clear explanation for the association between high sensitivity troponin T and stroke. Moreover, the authors state that they ‘cannot distinguish between cardioembolic and ischemic stroke, and thus, the observed relationship between high sensitivity troponin T levels and stroke in AF may relate to widespread endothelial damage/dysfunction and local atherothrombosis, rather than to thromboembolism from the left atrium per se’.
We have recently published an investigation that may shed some light on this dilemma . Albeit using cardiac troponin I instead of T, we have observed a higher prevalence of changes on transesophageal echocardiogram (TEE) associated with a higher risk of thromboembolic events (left atrial appendage thrombus, dense spontaneous echocardiographic contrast and low flow velocities in the left atrial appendage) in patients with detectable levels of cardiac troponin I. Moreover, the higher the rise in troponin I, the higher the prevalence of these TEE changes. Furthermore, we also observed that adding cardiac troponin I to the CHADS2 and CHA2DS2-VASc scores improved their overall predictive ability for these markers of left atrial stasis.
It is known that more than 90% of thrombi in atrial fibrillation form in the left atrial appendage . As Róldan and colleagues state in their paper ‘thrombi formation in the left atrium fulfills Virchow's triad for thrombogenesis; such patients with a poorer contractile function and severe endothelial damage are perhaps more prone to thrombus formation and embolisation’. Therefore, we think that it is plausible to assume that this prothrombotic milleu in the left atrium in patients with small elevations of cardiac troponin I will make them more prone to thromboembolic events.
In conclusion, it seems to us that the observed relationship between high sensitivity troponin T levels and stroke in patients with AF that was described in this landmark paper is most likely the result of embolism from the left atrium.