Erectile dysfunction in atrial fibrillation: A risk factor for stroke or a reflection of stroke risk factors?

Atrial fibrillation (AF) confers a 5-fold increase in the risk of stroke and thromboembolism and a 2-fold increase in mortality due mainly to stroke and heart failure. Oral anticoagulation with well managed dose adjusted vitamin K antagonist or a non-vitamin K antagonist highly effective in reducing this risk. Unfortunately, despite oral thromboprophylaxis, adverse cardiovascular (CV) events in AF patients is approximately 4% [1]. In the current issue of “Cardiology Journal”, Szymanski et al. [2] report a higher prevalence of erectile dysfunction (ED) among older patients with increasing thromboembolic risk. In its present form the paper does little more than highlight the association between ED and older age since this is the only component of either the CHADS2 or CHA2SD2-VASc scores that is significantly different between older patients (defined as ≥ 65 by the authors) with and without ED and therefore driving the outcome in this population. In the younger group (defined as u003c 65 by the authors), the age difference is not significant enough to create a difference in the overall CHADS2 or CHA2SD2-VASc risk score between patients with and without ED. Also notable is the absence of a multivariate analysis to adjust for the difference in waist circumference since several measures of adiposity are negatively correlated with endothelial function and waist circumference reduction has been shown to be associated with improved endothelial function in certain groups [3]. Moreover, increased adiposity creates a procoagulant state and is linked to over expression of proinflammatory markers of endothelial dysfunction such as fibrinogen, C reactive protein and von Willebrand Factor (vWF), possibly via a mechanism involving the CD40 ligand [4, 5]. Furthermore, whilst it may be appropriate to consider ED as an additional marker of a prothrombotic state, it is difficult to see how a diagnosis of ED would influence the clinical decision-making process in patients over 65 with AF (other than to prompt treatment of the ED itself), who would already qualify for oral anticoagulation according to international guidelines. Also, it would seem counterintuitive that ED would add much predictive power to the CHA2SD2-VASc score given that women are at a higher overall risk of thromboembolic stroke than men for reasons which are unclear but may be related to hormone therapy and sex differences in hemodynamics and CV remodeling [6]. ED, like stroke, is an outcome of cumulative vascular risk factors including AF. The CHADS2 and the CHA2SD2-VASc scores have been shown to predict to thromboembolic risk in patients with AF, with the CHA2SD2-VASc score improving risk stratification at the ‘low risk’ end of the spectrum [7]. In addition, Roldan et al. [8] have previously reported a significant positive correlation between vWF and the CHADS2 score supporting the association between increasing thromboembolic risk and vascular dysfunction. The inherent threat to vascular health associated with the presence of these risk factors is common to all vascular territories so it is perhaps not surprising that accumulation of thromboembolic risk factors is associated with ED — itself, an independent marker of endothelial dysfunction that