ObjectivesThe detection of low levels of antibodies against HBsAg (anti-HBs) below 10 IU/L in non-responders after a primary hepatitis B vaccination, is associated with seroconversion after revaccination. We compared the diagnostic performance of four anti-HBs assays in non-responders in their ability to differentiate between absence or presence of low levels of anti-HBs and propose a revaccination strategy guided by anti-HBs titres. MethodsNon-responders were revaccinated with Fendrix 20 μg at 0, 1 and 2 months. Anti-HBs titres were determined by Abbott Architect, Diasorin Liaison, Roche Cobas and Siemens ADVIA Centaur. Inter-assay agreement was evaluated with Cohen’s Kappa (k) in baseline samples between zero-responders without detectable antibodies and poor-responders with detectable antibodies < 10 IU/L. Seroconversion rates and geometric mean titres were analysed at 0, 1 and 3 months. A titre-based strategy (one revaccination dose and anti-HBs measurement followed by two more revaccination doses if required) was compared with the standard revaccination series of 3 doses. Results57 participants were included in the analysis. k was ≥ 0.65 for all assays except ADVIA (k ≤ 0.41). After one revaccination dose all assays detected a mean seroconversion rate in zero-responders of 42.9%, compared to 85.1% in poor-responders. The difference between zero- and poor-responders in seroconversion rate per assay was significant (p < 0.05). After three revaccination doses the mean seroconversion rate was 88.2% in zero-responders and 98.5% in poor-responders (p > 0.286 per assay). A titre-based strategy reduced the amount of revaccinations by 17% compared with the standard. ConclusionsAll assays demonstrated a comparable difference in seroconversion rate between zero- and poor-responders after one revaccination dose. The revaccination strategy could be optimised by differentiation between zero- and poor-responders followed by a titre-guided schedule.