Analysis of Borderline and Low Level Reactive Hepatitis B Virus Surface Antigen (HBsAg) by Different Automated HBsAg Assays

Hepatitis  B virus (HBV) surface antigen (HBsAg) has been used as the first-line marker to detect  HBV infection,  However,  HBsAg assay could  give false  positive result, The objective of this study  was to define the signal cutoff in borderline and low level reactive  HBsAg results in order  to interpret positive HBsAg results correctly. The retrospective study of data collected from  automated HBsAg assays at Clinical Immunology Unit, Department of Pathology, Ramathibodi Hospital from  January 2011 to May 2017  was conducted. The total 651 HBSAg test  results  that  Architect  HBsAg assay  showed  signal-to-cutoff (s/co)  ratios  0,9  to 20  were included  in this study.  The interpretation of HBsAg results was confirmed  by second automated HBsAg assays,  either  Elecsys HBsAg  II, Vitros  HBsAg or Vidas HBsAg. In addition,  the test results of HBsAg assay were compared  with HBV viral load. The second  HBsAg assays confirmed 59%  positive HBsAg, 18% intermediate HBsAg, and  23%  negative  HBsAg. When  Architect HBsAg showed  signal   10 s/co  ratios,  the second assays  confirmed  98%  positive HBsAg and  2% intermediate HBsAg  without  negative  HBsAg. The Architect HBsAg assay correlated with the second assays; Elecsys HBsAg II, Vitros HBsAg and  Vidas HBsAg, (r = 0.75,  0.65 and 0.60  respectively). However,  only 47% of total samples were positive for  HBV viral load. In conclusion, borderline and low level reactive  Architect  HBsAg results  that showed signal <5 s/co ratios  needed confirmation by the second assays before HBsAg interpretation and report. Architect HBsAg  test  results  that  showed  signal    10 s/co  ratios  could  be  precisely predicted  positive HBsAg results.