Abstract: Objective　To evaluate methodologically particle-enhanced turbidimetric immunoassay（PETIA）for determining quantitatively alpha-fetoprotein（AFP）. Methods　AFP was determined quantitatively by OLYMPUS AU5400 automatic biochemical analyzer， and the precision， reagent stability， detection low limit， sensitivity， linear range， interference and accuracy of this method were investigated. The correlation of fresh serum sample results between SIEMENS CENTAUR electro-chemiluminescence assay（ECLIA）and PETIA was analyzed. The reference interval was verified. Results　Testing 29.95 and 132.25 ng/mL patient samples， the within-run coefficients of variation（CV）were 1.67% and 1.52％，and the day-to-day CV were 7.17% and 7.84% respectively. The detection low limit and sensitivity were 0.52 and 0.53 ng/mL respectively. The linear correlation coefficient（r²）was 0.999 4 from 5.4 to 256 ng/mL， and the regression equation was Y=1.014 X +1.063 3. Both of the relative deviations of the high-level serum（5.96%）and low-level serum（8.40%）were lower than the evaluation criteria "target value±20%" of proficiency testing（PT） from National Center for Clinical Laboratories. There was a good correlation between PETIA with SIEMENS electro-chemiluminescence analyzer in 107 patients，Y_ECLIA=1.138 7X_PETIA+2.835 3（r²=0.992 1）. There was no obvious interference from hemolysis， jaundice and lipidemia. All the results of 20 healthy subject samples were included in the reference interval recommended by the manufacture. Conclusions　PETIA is a convenient， rapid， cheap， accurate and reliable method for determining AFP， particularly be suitable for large-scale health screening physical examination.

Key words: Particle-enhanced turbidimetric immunoassay, Alpha-fetoprotein, Methodology evaluation