Objective      To determine the biases among different series of Mindray auto-biochemical analyzers and subsequently provide experimental bases for selecting optimal domestic auto-biochemical analyzers.       Methods       According to Clinical and Laboratory Standards Institute (CLSI) guideline (EP9-A2)

Beckmann Coulter AU5400 was set as reference detection system

while Mindray auto-biochemical analyzers BS200

BS380

BS480 and BS820 were set as test detection systems

respectively. Fifteen conventional biochemical tests were selected for method comparison. Bias estimations were assessed by comparing determination coefficient (R2)

slope of regression equation

bias of medical decision level between the reference detection system and single test detection system

respectively.       Results      Fairly good correlations were seen between the reference detection system and the single test detection system

respectively. There were certain systematic biases between the reference and test systems. Only serum levels of total protein (TP)

albumin (ALB) and calcium (Ca) were out of comparability

while the rest of the 15 tests were within comparability when China’s health industry standards were set as quality goals. The quality goals of TP

ALB and CA were 5%

6% and 5%

respectively. The average bias of TP detected by BS200

BS380

BS480 and BS820 were 8.74%

4.59%

5.85% and 8.26%

respectively. While those of ALB were 12.66%

14.07%

11.66% and 12.99%

respectively

and those of CA were 6.79%

7.85%

7.7% and 8.96%

respectively. The average biases of TP

ALB and CA were greater in these 4 analyzer than quality goals at one or more medical decision levels.       Conclusion      The comparability between the detection system of Mindray series and the reference system of Beckmann Coulter AU 5400 is acceptable. Auto-biochemical analyzers of Mindray series deserve promotion for further use in basic level units.