Aims: Paroxysmal atrial fibrillation (PAF) is often asymptomatic but nonetheless harmful. We evaluated the performance of disease-related blood biomarkers and CHA2DS2-VASc score to discriminate for PAF in patients with continuous rhythm monitoring.
Methods and results: Clinical data and blood samples were obtained from patients with dual-chamber pacemakers selected according to the absence (no_AHRE) or presence of Atrial High-Rate Episodes (AHRE) >6 min in recent device history (case-control approach). We included 93 patients (n = 49 AHRE, n = 44 no_AHRE). In a subgroup with high AHRE burden and confirmed PAF 15 biomarkers were evaluated (n = 19 AHRE-AF vs. n = 20 no_AHRE). Significantly regulated biomarkers were then tested in all patients to distinguish no_AHRE from AHRE (receiver operating characteristics analysis). Hsp27, TGFβ1, cystatin C, matrix metalloproteinases MMP-2,-3,-9, albumin, and serum uric acid were not altered in the subgroup. Tissue inhibitors of metalloproteinases (TIMP) -1,-2,-4; NT-proANP, NT-proBNP, IL-6 and serum amyloid protein A were significantly different in AHRE vs. no_AHRE (subgroup and whole cohort), with best discriminatory performance for TIMP-4. Biomarkers performed better than CHADS2-VASc for AHRE discrimination. Intracardial electrograms and medical history from seven AHRE patients suggested atrial tachycardia and not AF (AHRE-AT). Four of the most relevant regulated biomarkers (TIMP-4, TIMP-2, SAA, NT-proBNP) behaved similarly in AHRE-AT and AHRE-AF. NT-proBNP >150 pg/mL indicated an odds ratio of 12.9 for AHRE. Combining two biomarkers significantly improved discrimination of AHRE.
Conclusion: TIMP-4, NT-proANP, NT-proBNP were strongest associated with PAF and AHRE. The discriminatory performance of CHADS2-VASc for PAF was increased by addition of selected biomarkers.
Keywords: Atrial high rate episodes; Biomarkers; CHA2DS2-VASc score; NT-proBNP; Paroxysmal atrial fibrillation; TIMPs.
Published on behalf of the European Society of Cardiology. All rights reserved. © The Author 2016. For permissions please email: [email protected].