Assessment of correlation between silent brain damages and clinical characteristics in patients with persistent atrial fibrillation and concomitant arterial hypertension

Abstract

Abstract Introduction Silent cerebral microembolic lesions (SML) are known to occur with higher prevalence in persistent AF and could be potential reason of cognitive decline. It seems reasonable to investigate which factors may be related with the SML occurrence in AF pts to implicate it in the future prophylaxis. Purpose To estimate association between brain MRI findings (incidence of silent cerebral microembolism and microbleedings) in patients with persistent atrial fibrillation and their clinical characteristics including concomitant arterial hypertension (AH) and CHA2DS2-VASc score. Methods Pts with persistent AF (n=69) were included in the study. All subjects underwent clinical assessment, thromboembolic and hemorrhagic risk calculation using CHA2DS2-VASc and HAS-BLED score with subsequent cranial MRI examination. The incidence of SML (registered on MRI like small focal hyperintense areas in the brain matter) and microbleedings were assessed. The pts were divided into groups: Gr1 (n=49) – pts with revealed SML, Gr2 (n=20) – without silent brain damages. Clinical characteristics of pts were compared regarding to presence of SML and microbleedings using Mann-Whitney U test. Spearman analysis and ROC- analysis were used for correlation assessment. Results The total incidence of SML was 71% (n=49). Isolated microbleedings were found in 13% (n=9) pts., 59.4% (n=41) pts were anticoagulant naïve. The most part of investigated pts (82.6%, n=57) had well controlled AH: 42 pts (85.7%) in Gr1 and 15 (75%) in Gr2, p=0.1. Persons with SML were significantly older compared to the pts without SML – 59.6±7.1 years vs 52.1±8.9 (p=0.005), had lower EF – 54.4±9.4% vs 58.5±6.8% (p=0,04), higher CHA2DS2-VASc score – 2 [1; 3] vs 1 [0; 1.5] and higher HAS-BLED score – 0 [0; 3] vs 1 [0; 2] (p=0.02). Notably, higher fasting plasma glucose level was registered in Gr1 in comparison with Gr2 – 6.34±2.4 mmol/l vs 5.4±0.5 mmol/l (p=0.008). Significant correlations were calculated between SML and CHA2DS2-VASc score (AUC – 0.875, CI: 0.773–0.977, p=0.006), SML and degree of AH (AUC – 0.658, CI: 0.512–0.803, p=0.039), SML and age (AUC – 0.757, CI: 0.62–0.895, p=0.001). The age cut-off value for revealing SML on brain MRI in pts with persistent AF was determined: 45.5 years (AUC – 0.757, Sensitivity – 97.9%, Specificity – 71.4%, p=0.001). Individuals with microbleedings had only significantly higher CHA2DS2-VASc score – 3 [1; 4] vs 1 [0; 4] (p=0.002) with weak positive correlation (r=0.32, n=69, p=0.009). Conclusion The correlation between brain SML and CHA2DS2-VASc score confirms the ability of that score to predict not only clinically apparent thromboembolic events but also silent microembolism in persistent AF. Brain SML in pts with persistent AF are associated with higher CHA2DS2-VASc score, higher fasting plasma glucose, degree of arterial hypertension. Notably that pts revealing SML on brain MRI were younger compared to the age using in CHA2DS2-VASc score. Funding Acknowledgement Type of funding source: None