Atrial fibrillation (AF) is a common cardiac arrhythmia associated with an increased morbidity and mortality, and much of the AF-associated morbidity is secondary to a 5- to 6-fold increased risk of stroke and thromboembolism . The stroke risk in AF patients is not homogeneous, but depends on the presence of other underlying clinical conditions . To aid decision-making for thromboprophylaxis, several clinical risk stratification schemes have been developed using various risk factors and clinical characteristics. More recently, the CHA2DS2-VASc score has been proposed to complement the CHADS2 score in decision-making for oral anticoagulation, reflecting a risk factor-based approach to thromboprophylaxis .
AF patients also have a high incidence of other cardiovascular events (mainly acute coronary syndromes and vascular death) . The latter reflects the common presence of co-morbidities and risk factors (e.g. ischemic heart disease, hypertension or diabetes) in AF, which are also well-established risk factors for stroke – hence their inclusion in stratification risk scores. There is a high incidence of AF among patients with atherothrombosis and vice versa , and so for patients who have both AF and atherothrombosis, the risk of mortality is substantially higher when compared with patients without AF.
Accordingly, AF may well be a surrogate marker of vascular damage and subsequently atherothrombosis. In this setting, we have observed, in a ‘real life’ cohort of outpatients with AF under oral anticoagulation, a high incidence of cardiovascular events and mortality. Interestingly, we have demonstrated the ability of CHA2DS2-VASc to predict cardiovascular events and mortality among these patients .
Biomarkers may be helpful to refine assessment of AF risk. We have recently reported, in a large cohort of anticoagulated AF patients, that von Willebrand factor levels (an established biomarker of endothelial damage/dysfunction) were independent predictors for thrombotic, bleeding events and death . Similarly, we found a significant association during more than 2 years follow-up of detectable levels of hsTnT and adverse cardiovascular events and mortality, and specifically thrombotic stroke. Nonetheless, it remains uncertain whether these biomarkers bring new, different physiological factors into stroke prediction in AF patients or simply they are a measure of severity (or ‘disease burden’) of previously identified clinical predictors. AF risk factors are indicators of a biological process that relates to AF, while measured ‘biomarkers’ (whether blood tests, urine markers, imaging, etc.) are related to an AF-causing process (or consequence), but do not contribute by themselves to the biology of AF .
Providencia et al.  suggest that cardiac troponin I could reflect the prothrombotic state of the left atria as they found a significant relationship with the presence of transesophageal echocardiogram changes. However, is it cause or consequence? A very recent substudy from the RE-LY trial has reported that elevated levels of troponin I can frequently be detected in patients with AF and risk factors for stroke, and troponin I elevation is linked to an increased risk of stroke and mortality . Indeed, in a general population without apparent cardiovascular diseases, troponin I levels were greater in AF subjects than in non-AF subjects , with significant differences in left atrial diameter. Nonetheless, left atrial dilatation has been proposed as a reliable marker of left ventricular diastolic function, and thus, it may reflect increased left chamber pressure and ventricular remodeling.
In conclusion, increased levels of endothelial damage/dysfunction, ventricular remodeling or even systemic inflammation markers could only reflect the presence of a severe vascular disease, which is the link among AF, stroke and other severe cardiovascular events. Measurement of a ‘biomarker’ may offer some opportunity to refine the possible imprecision of risk scores based on clinical risk factors, but a balance is also needed between practicality, cost and user-friendliness in everyday clinical practice.