The number of individual teeth decreases with age, resulting in a decrease in masticatory capacity, and is an important indicator of oral health. However, it is difficult to estimate the number of present teeth on the basis of age alone. We aimed to determine whether tooth retention could be estimated by the number of present teeth in middle-old-aged individuals. We used data from the Korean Genome and Epidemiology Study on Atherosclerosis Risk of Rural Areas in the Korean General Population (KoGES-ARIRANG). Subjects (40-75 years old) were invited to participate in a 3-year prospective follow-up survey conducted from 2010 to 2014. A total of 557 individuals (219 men and 338 women) took part in the study. Tooth retention was estimated from the number of present teeth by multivariate logistic regression analysis using SPSS v. 20.0. In total, 294 (52.8%) subjects retained teeth during a 3-year follow-up period. The number of present teeth and the proportion of subjects with complete tooth retention after 3 years decreased with increasing age. A greater number of present teeth in the baseline year was associated with complete tooth retention after 3 years in a greater proportion of subjects (p < 0.001). Logistic regression analysis revealed that the proportion of subjects with tooth retention in the Q4 quartile (28 teeth) was 9.17 times that in the Q1 quartile (less than 26 present teeth), even after adjusting for sociodemographic factors and oral health behavior in middle-aged individuals. In elderly individuals, tooth retention in the Q4 quartile (28 teeth) was 4.50 times that in the Q1 quartile (less than 17 teeth). The number of present teeth could be used to estimate tooth retention over a 3-year period. This highlights the importance of promoting oral health care in middle-aged individuals before tooth loss occurs.