A systematic review and meta-analysis of observational studies were performed to assess the dose–response associations between fruit or vegetable consumption and the chance of the metabolic syndrome (MetS). Studies on the association between fruit or vegetable consumption and the risk of the MetS published from January 1958 to 30 October 2018 were searched using the PubMed, MEDLINE and Embase databases, and the references of relevant articles were reviewed. Random-effects models were used to estimate the summary OR with 95 % CI for the MetS, and dose–response analysis was conducted to quantify the associations. Heterogeneity among studies was evaluated using Q and I2 statistics. A total of nine observational studies (seven cross-sectional studies and two cohort studies) were included in the meta-analysis. In a dose–response analysis of cohort studies and cross-sectional studies, the summary estimate of the MetS for an increase of 100 g/d in fruit consumption (nine studies) was 0·97 (95 % CI 0·95, 0·99; I2 = 26·7 %), whereas an increase of 100 g/d in vegetable consumption (nine studies) was not associated with a reduction in the MetS (OR 0·98; 95 % CI 0·96, 1·01; I2 = 54·6 %). In conclusion, an increased intake of fruit may reduce the risk of the MetS. For future research, prospective studies or randomised clinical trials are needed to identify the effects of fruits and vegetables by variety on the risk of the MetS.