There is evidence of a bidirectional association between dental caries and anthropometric measures among children. This dual relationship has not been examined in the same population. The objectives of this study are (1) to examine the relationship between baseline caries and changes in weight and height; and (2) to assess whether baseline weight and height are associated with changes in dental caries in a sample of preschool Chinese children. Children were recruited from 15 kindergarten in Liaoning Province at baseline (8 in rural area and 7 in urban area), a total of 1,111 of children were included at baseline. The mean age of children at baseline was 50.82 months. Data were collected through clinical oral examination, assessment of anthropometric measures and structured questionnaire. Dental caries was assessed according World Health Organization (WHO) methods by one dentist. Sociodemographic and behaviour data were also collected. At follow-up, 772 children were included (attrition rate: 30%), dental caries and anthropometric measures were assessed again. Z-score for weight-for-age and height-for-age were calculated using the 2006 and 2007 WHO Child Growth Standards. The sum of decayed, missing and filled primary teeth (dmft) were used in the analysis. Multilevel analysis for longitudinal data was conducted to explore the relationship between z-score for weight-for-age and height-for-age, and dental caries among children. The median follow-up time was 10.12 months. There was a significant negative association between dmft at baseline and change in height-for-age. On the other hand, weight-for-age at baseline was negatively associated with change in dmft at follow-up. The findings suggest that dental caries impedes children's growth indicated by height for age. Low weight children appear to be more susceptible to dental caries in the same population.