Objective:To quantitatively evaluate the relationship between sleep duration and metabolic syndrome in adults in order to set up programs on prevention and treatment of metabolic syndrome in this population.Methods:Relevant studies were identified by systematically searching databases before October 2019. All statistical analyses were under the use of Stata 11.0.Results:A total of 656 319 participants including 150 638 cases with metabolic syndrome were involved in thes 38 articles. A U-shaped relationship between sleep duration and metabolic syndrome was noticed. For short and long sleep duration, the OR=1.11 (95% CI: 1.07-1.16) and 1.10 (95% CI: 1.03-1.18), respectively. Subgroup analyses on cross-sectional studies revealed that factors as men, aged under 60 years, being Asians or Caucasians would increase the risk of metabolic syndrome by 6%, 14%, 9%, and 24%, respectively for those with short sleep duration. Factors as aged 60 years and above, being black and with long sleep duration, would increase the risks of metabolic syndrome by 13% and 19%, respectively in women. In subgroup analyses on cohort studies, positive correlation between short sleep duration and metabolic syndrome was observed in both Asian ( RR=1.10, 95% CI: 1.07-1.13) and in Caucasians ( RR=1.56, 95% CI: 1.08-2.26) populations. Conclusions:Results of this study revealed an association between metabolic syndrome and the duration of sleep. We understand that sleep is a behavior that can be changed step by step, through adequate intervention programs, to reduce the risk of metabolic syndrome which has become an important public health issue.