Long-term Changes In Proteomic Levels And The Risk Of Incident Atrial Fibrillation In Older Adults: The Atherosclerosis Risk In Communities (ARIC) Study

Background: Previous studies examining proteomic markers and atrial fibrillation (AF) have used protein measurements at one time point; however, few have studied these protein changes over time. Using data from the Atherosclerosis Risk in Communities (ARIC) study, we examined the association of long-term protein changes with AF in white and Black participants. Methods: Of 4877 plasma proteins measured at visits 3 (1993-1995) and 5 (2011-2013) using an aptamer-based proteomic profiling platform, we examined 17 proteins previously associated with AF. Protein changes between visits 3 and 5 were log base 2 transformed, corresponding to the log base 2 of the ratio [log 2 (protein5/protein3)]. AF was ascertained from hospital discharge codes, ECGs, and death certificates. We used Cox proportional hazards to assess the association of protein changes and AF. Results: This study included 3552 participants (mean age 75±5 years; 58% female; 17% Black) who had proteomic measures and were free of AF at visit 5. Over a mean follow-up of 6.4±2 years, 486 participants developed AF. After adjustment for clinical factors of AF, medication use, estimated glomerular filtration rate, and protein level at visit 5 (Model 2), changes in natriuretic peptides B and spondin-1 were associated with the risk of AF (HR: 1.57; 95%CI: 1.25-1.98; p-value=0.0001; and HR: 1.82; 95%CI: 1.23-2.70; p-value=0.0027, respectively) after Bonferroni correction. Although changes in NT-proBNP were not associated with AF risk, adjustment for visit 5 NT-proBNP attenuated associations for natriuretic peptides B and spondin-1 to non-significance. Conclusion: Examining protein changes over time may provide additional information about mechanistic pathways of AF development.