Abstract Background The number of the patients with heart failure with preserved ejection fraction (HFpEF) is increasing as society ages, which indicates that the number of pre-HFpEF patients is increasing even more. To combat this sociomedical burden, it is critical to establish cost-effective and simple biomarkers to identify the high-risk population at early phase of the disease. Recent studies demonstrated the association between albuminuria and incident heart failure, but the prognostic impact of albuminuria on diastolic dysfunction, an early phenotype of HFpEF, has not been elucidated. Purpose To examine the impact of albuminuria on the diastolic function over time in the patients with cardiovascular risks. Methods One hundred and fifty-eight patients with at least one cardiovascular risk factor without history of heart failure were followed up for 5 years. Echocardiogram was conducted at baseline, year 3, and year 5. Diastolic dysfunction was defined as E/e’ >14. Albuminuria was defined as urine albumin-to-creatinine ratio (UACR) 30 ≧mg/gCre. The patients with ejection fraction <40% and those had diastolic dysfunction at baseline were excluded from the study. The relationship between albuminuria at baseline and diastolic function at year 3 and 5 were investigated. Results Among 158 subjects, 27 showed albuminuria at baseline and the prevalence of hypertension, dyslipidemia, and diabetes was not significantly different between patients with and without albuminuria. The baseline E/e’ tended to be higher in patients with albuminuria, although the difference was not statistically significant (Fig1: E/e’ = 10.0±2.4 vs. 9.4±2.2, p = 0.08) . However, the patients with albuminuria showed a steeper increase in E/e’ than those without albuminuria, and the significant difference was observed at year 3, that was even greater at year 5 (Fig1: E/e’ = 11.1±2.8 vs 10.0±2.6, p = 0.04 in year 3, E/e’ = 12.5±4.1 vs 10.3±2.9, p = 0.002 in year 5, respectively). Furthermore, log UACR at baseline showed significant correlation with E/e’ at year 5 (p = 0.02, r = 0.21) and normo-, micro-, and macroalbuminuria (UACR<30, 30-300, and >300mg/gCre, respectively) at baseline showed elevated E/e’ at year 5 in this order ( p = 0.01 for UACR<30 vs. 30-300mg/gCre, and p = 0.04 for UACR<30 vs. >300mg/gCre, reapectively ). The prevalence of incident diastolic dysfunction increased greater in patients with albuminuria throughout the follow-up period and the difference reached statistical significance at year 5 (Fig2: 18.5% vs. 9.2%, p = 0.28 for year 3, and 25.9% vs. 8.4%, p = 0.009 for year 5, respectively). Conclusion In the patients with cardiovascular risk factors, albuminuria was associated with development and progression of diastolic dysfunction over subsequent 5 years. Future prospective study is warranted to investigate the effectiveness of albuminuria-directed care to prevent development of HFpEF.