Performance evaluation of LUMIPULSE G1200 autoimmunoanalyzer for the detection of serum hepatitis B virus markers.

BACKGROUND

We evaluated recently introduced automated immunoassay analyzer LUMIPULSE G1200 (Fujirebio, Inc., Tokyo, Japan) for detecting serologic hepatitis B virus (HBV) markers by comparison with the results by ARCHITECT i4000SR (Abbott, Abbott Park, IL).

METHODS

Precision performance was evaluated over 20 days. HBV surface antigen (HBsAg), HBV e antigen (HBeAg), antibodies to HBV core antigen (anti-HBc), antibodies to HBeAg (anti-HBe), and antibodies to HBsAg (anti-HBs) in a total of 1,000 serum samples were assessed by the two analyzers. Discrepant results were retested by COBAS e411 (Roche Diagnostics, Mannheim, Germany).

RESULTS

LUMIPULSE showed excellent precision performance of total imprecision less than 3.5% coefficient of variation. The qualitative results between the two analyzers were agreed with each other in 92.0-99.8% of the specimens according to the different HBV markers. The degrees of reactions for HBeAg were moderately correlated between the two analyzers (r = 0.60), and those of other HBV markers were well correlated (r = 0.80 or greater). However, there were 183 discrepancies among 1,000 cases, and most of them showed degree of reaction around the cutoff values.

CONCLUSIONS

LUMIPULSE G1200 showed well-concordant results with ARCITHECT for hepatitis B serologic tests. However, results near the cutoff values would need to be retested with other immunoassay or molecular methods, when the serological profiles of HBV markers are unusual or are not correlated to the clinical conditions of the patient, due to discrepancies between the immunoassay analyzers.