To investigate potential interferences of two pre-analytical variables, the storage time and the vacutainer tube, on the quantification of 20 amino acids using a UHPLC-MS/MS method.Blood samples from 25 apparently healthy subjects were collected into duplicate sets of EDTA-2K, EDTA-3K, coagulation, heparin and citrate tubes, and stored in capped vacutainer tubes at 4°C for 6h, 12h and 24h before sample analysis. A UHPLC-MS/MS method was established for simultaneous quantification of 20 amino acids. ANOVA for repeated measurement was conducted based on the model of Mauchly's test of Sphericity. Student's t-test was applied for comparison between amino acid concentrations obtained from different vacutainer tubes, and consistency of the results was checked through Bland-Altman difference plots and Passing-Bablok regression analysis.Most of the 20 amino acids showed a least concentration fluctuation with storage time in heparin plasma, followed by EDTA-3K and citrate plasma. The amino acid concentrations were significantly lower in citrate plasma and slightly higher in serum, compared with those in heparin plasma. No fixed bias was observed for amino acid concentrations in EDTA and heparin plasma, but the differences were mostly of statistical significance. Amino acid concentrations in EDTA-3K plasma achieved a good consistency with those in heparin plasma by UHPLC-MS/MS analysis.Storage time and vacutainer tube were important variables for amino acid analysis. They should draw researchers' attention and then be controlled in good laboratory practice to reduce pre-analytical errors.