Long-term pulse pressure trajectories and risk of incident atrial fibrillation: the Tromsø Study

Abstract Background and Aims Sex-based differences in the association of long-term trends in pulse pressure with future risk of atrial fibrillation (AF) have been explored using data from the population-based Tromsø Study 1986–2016. Methods Women (n = 8331) and men (n = 7638) aged ≥20 years who attended at least two of the three Tromsø Study surveys conducted between 1986 and 2001 (the exposure period) were followed up for incident AF throughout 2016 (the follow-up period). Pulse pressure ≥60 mmHg was considered elevated. Group-based trajectory modelling and Cox regression were used for statistical analyses. Results Three long-term trajectory groups for pulse pressure were identified: Group 1 had normal pulse pressure throughout the exposure period, Group 2 had normal pulse pressure at the beginning and elevated pulse pressure at the end of the exposure period, and Group 3 had elevated pulse pressure throughout. Over the follow-up period, 568 (6.8%) women and 798 (10.5%) men developed AF. After adjustment for potential confounders at baseline, the long-term trajectory groups for elevated pulse pressure were associated with increased risk of AF in women, but not in men. In women, the adjusted hazard ratios of AF were 1.60 (95% confidence interval: 1.23, 2.09) for trajectory Group 2 and 2.78 (1.93, 4.02) for trajectory Group 3, compared with Group 1. Conclusions Long-term elevated pulse pressure trajectories were independently associated with increased risk of AF in women, but not in men. Our findings call for further investigations to understand the mechanisms behind these sex-based differences.