P983
When should we measure biomarkers in patients with atrial fibrillation to predict recurrences?

No grants to declare

Atrial fibrillation (AF) in the most frequent arrhythmia. Remodeling and fibrosis phenomena are implicated in its development. ProBNP, galectin-3, ST2, urate, C reactive protein, troponin and fibrinogen are some biomarkers (BM) involved in these processes. 

The aim of our study was to assess the implication of these BM in the recurrence of AF after an electrical cardioversion (ECV) and if the relation depends on the moment that they are measured.

We included 115 patients with persistent AF who underwent an ECV. We performed a 6-months (M) -follow-up. We assessed our patients in a clinical, analytical (a blood sample was extracted pre-ECV and at 6M) and electrocardiographic (with holter at 3M and an ECG at 6M). Any documentation of AF during the follow-up was defined as a recurrence. We assessed the relation between basal BM and recurrence by Kaplan-Meier method and the relation between BM at 6M and recurrence by tStudent with normal variables and U-Mann-Withney with no-normal ones. We performed a ROC curve with those BM that show statistical significance in order to define the best cut-off point to classify our patients. We also made a logistic regression analysis to assess the existence of independent association between BM and recurrence.

71 patients had a recurrence (61.73%). These patients were older than those who did not, without any other significant clinical difference between both groups. None of the basal BM was related to recurrence during the follow-up. ProBNP and ST2 at 6M were significantly related to recurrence. ProBNP showed an area under the ROC curve of 0.774 (figure 1). The best cut-off point for proBNP was 300 pg/ml (sensibility (S) 72%, specificity (Sp) 72%) and for ST2 30.73 ng/ml (S 66% and Sp 58%). Only proBNP was independently related to recurrence (OR: 10.90; CI (3.87-30.75), P < 0.001.

BM measured pre-ECV do not predict recurrences in the follow-up, so they should not be used with that purpose. Those patients that suffered recurrence had higher levels of proBNP (distension BM) and ST2 (remodeling BM) measured at 6M. These 2 processes might be the most important pathophysiological factors in the maintenance of AF. The repeated measurement of these BM could lead the most accurate management in these patients.

Abstract Figure.