Association of fasting plasma glucose trajectory with lifetime risk of cardiovascular disease.

Abstract

Few studies have used multiple measurements of fasting plasma glucose (FPG) to examine the impact of long-term FPG trajectory patterns on lifetime risk of cardiovascular disease (CVD). We aimed to identify the long-term patterns in FPG trajectories and to estimate the lifetime risk of CVD according to FPG trajectories. Individuals free of CVD at index ages 35 (n = 72,324), 45 (n = 62,049), and 55 (n = 38,113) years were included. FPG concentrations were measured in 2006, 2008, and 2010. The FPG trajectories were identified by latent mixture modeling. The modified Kaplan-Meier method was used to calculate lifetime risk of CVD. We identified five distinct FPG trajectories and named them according to FPG range and changing pattern over time: low-stable, moderate-stable, moderate-increasing, elevated-decreasing, and elevated-stable. At index age 35 years, we documented 3110 CVD events in men during 371,219 person-years of follow-up and 357 CVD events in women during 107,447 person-years of follow-up. Among all participants, the elevated-stable FPG pattern experienced the highest lifetime risk of CVD (44.8%, 95% CI: 37.8-51.9%), low-stable pattern was lowest (24.3%, 95% CI: 23.3-25.2%). At index age 55 years, although the elevated-stable and elevated-decreasing FPG patterns had similar original FPG concentrations, individuals with elevated-decreasing pattern (30.0%, 95% CI: 23.9-36.1%) had approximately one-third less lifetime risk of CVD than those with elevated-stable pattern (43.6%, 95% CI: 31.8-55.3%). FPG trajectories were significantly associated with the lifetime risk of CVD. Both decrease in FPG over time and consistently lower FPG over 4 years were associated with lower lifetime risk of CVD.