ABSTRACT: Introduction: Zinc-?2-glycoprotein (ZAG) is a novel adipokine, which is involved in metabolic syndrome (MetS). This study aimed to investigate the relationship between serum ZAG and MetS in Chinese adults, who diagnosed according to the 2005 International Diabetes Federation (IDF) criteria. Methods: A group of 151 MetS patients, 84 patients with central obesity and 70 healthy controls were enrolled. General clinical information, serum samples were obtained from all subjects and serum ZAG levels were determined via the commercial ELISA kits. Results: Serum ZAG levels were the highest in the control group, then gradually decreased with the severity of the metabolic abnormalities increased (8.78 ± 1.66 ?g/mL for control vs. 8.37 ± 1.52 ?g/mL for central obesity vs. 7.98 ± 0.94 ?g/mL for MetS, P < 0.05). It was also decreased progressively with an increasing number of the MetS components (P for trend = 0.002). Additionally, serum ZAG/fat mass ratio was calculated and the similar changes were observed in the three groups (0.85 ± 0.53 ?g/mL/kg for control vs. 0.39 ± 0.10 ?g/mL/kg for central obesity vs. 0.36 ± 0.08 ?g/mL/kg for MetS, P < 0.05). In the multiple regression analysis, group was a strong independent factor contributing to serum ZAG levels (P < 0.001). Furthermore, compared with subjects with the highest tertile of ZAG, subjects in the lowest tertile of ZAG had 1.946-fold higher risk of MetS (95% CI 1.419-6.117, P = 0.004). This phenomenon still existed after controlling for age, gender (Model 1), ALP, AST, Cr, UA, Urea based on Model 1 (Model 2), grip strength, smoking, drinking, birth place, current address, education level, manual labor, and exercise frequency based on Model 2 (Model 3). Receiver operation characteristic (ROC) curve analysis revealed that serum ZAG might serve as a candidate biomarker for MetS (sensitivity 57.6%, specificity 70.0% and area under the curve 0.655), and serum ZAG/fat mass ratio showed improved diagnosis value accuracy, with ROC curve area of 0.951 (95% CI, 0.920-0.983, P < 0.001), and 90.7% sensitivity and 88.6% specificity. Conclusions: Serum ZAG levels were lowered in patients with MetS and central obesity. The decreased serum ZAG levels were associated with the increased risks of MetS. Serum ZAG, especially serum ZAG/fat mass ratio might be the candidate diagnostic biomarkers for MetS.