To clarify the association of serum kisspeptin levels in women with polycystic ovary syndrome (PCOS) by meta-analysis.Two English databases and two Chinese databases were searched for the relationship between kisspeptin and PCOS published from 2009. After the studies screening according to specific principles, we used STATA 12.0 for meta-analysis. Standardized mean difference (SMD) and its 95% confidence intervals (95% CIs) were used as the effect size and STATA 12.0 software was performed by this meta-analysis.Nine articles were included in the end, with a total of 1282 participants (699 patients and 583 controls). Heterogeneity between studies was statistically significant. Therefore, the random effects model was used to combine the effects. Meta-analysis showed statistically significant differences in serum kisspeptin levels between the PCOS patients and controls (SMD = 0.57, 95% CI [0.32, 0.82]), which indicated that there is a strong association between serum kisspeptin levels and PCOS. The source of high heterogeneity between the inclusion studies (I2  = 73.2%) might be due to the small sample size. The larger variation of kisspeptin concentration might be caused by different diagnosis criteria of PCOS and short half-time period of kisspeptin combined with nonstandard testing process.Serum kisspeptin levels in PCOS patients were higher than non-PCOS patients. It is a hint to indicate us that kisspeptin might be an independent biomarker of PCOS patients.© 2021 Japan Society of Obstetrics and Gynecology.