Published in

Public Library of Science, PLoS ONE, 8(17), p. e0273571, 2022

DOI: 10.1371/journal.pone.0273571

Links

Tools

Export citation

Search in Google Scholar

Blood-biomarkers and devices for atrial fibrillation screening: Lessons learned from the AFRICAT (Atrial Fibrillation Research In CATalonia) study

This paper is made freely available by the publisher.
This paper is made freely available by the publisher.

Full text: Download

Green circle
Preprint: archiving allowed
Green circle
Postprint: archiving allowed
Green circle
Published version: archiving allowed
Data provided by SHERPA/RoMEO

Abstract

Background and objective AFRICAT is a prospective cohort study intending to develop an atrial fibrillation (AF) screening program through the combination of blood markers, rhythm detection devices, and long-term monitoring in our community. In particular, we aimed to validate the use of NT-proBNP, and identify new blood biomarkers associated with AF. Also, we aimed to compare AF detection using various wearables and long-term Holter monitoring. Methods 359 subjects aged 65–75 years with hypertension and diabetes were included in two phases: Phase I (n = 100) and Phase II (n = 259). AF diagnosis was performed by baseline 12-lead ECG, 4 weeks of Holter monitoring (NuuboTM), and/or medical history. An aptamer array including 1310 proteins was measured in the blood of 26 patients. Candidates were selected according to p-value, logFC and biological function to be tested in verification and validation phases. Several screening devices were tested and compared: AliveCor, Watch BP, MyDiagnostick and Fibricheck. Results AF was present in 34 subjects (9.47%). The aptamer array revealed 41 proteins with differential expression in AF individuals. TIMP-2 and ST-2 were the most promising candidates in the verification analysis, but none of them was further validated. NT-proBNP (log-transformed) (OR = 1.934; p<0.001) was the only independent biomarker to detect AF in the whole cohort. Compared to an ECG, WatchBP had the highest sensitivity (84.6%) and AUC (0.895 [0.780–1]), while MyDiagnostick showed the highest specificity (97.10%). Conclusion The inclusion and monitoring of a cohort of primary care patients for AF detection, together with the testing of biomarkers and screening devices provided useful lessons about AF screening in our community. An AF screening strategy using rhythm detection devices and short monitoring periods among high-risk patients with high NT-proBNP levels could be feasible.