Occurrence and potential implications of the hook effect in a Borrelia burgdorferi IgG antibody chemiluminescence immunoassay

Abstract

Purpose: Lyme borreliosis is diagnosed based on clinical symptoms and serological analysis, with high IgG antibody levels associated with late manifestations. Immunoassays with a broad detection range can exhibit a hook effect, leading to false-low results and potential misdiagnosis. This study aimed to investigate the occurrence of the high-dose hook effect and determine a dilution threshold to detect the hook effect and avoid overdilutions for the Liaison Borrelia IgG immunoassay. Methods: In a two-year period, 5639 patient samples analysed for Borrelia antibodies were screened at Karlstad Hospital, Sweden. Samples with IgG ≥ 75 AU/mL and < 75 AU/mL with detectable IgM underwent 1:10 dilution, with a further 1:50 dilution for samples with IgG > 240 AU/mL. A Gaussian Mixture Model was used to group samples with and without hook effect. Results: Of 389 samples eligible for dilution, 262 with IgG < 240 AU/mL were analysed. Of these, 70 (26.7%) showed hook effect, corresponding to 18% of included and 1.2% of all screened samples. Overdilution occurred in 58 (22.1%) diluted samples. Dilution thresholds of 98.7 and 96 AU/mL detected 95% of hook effect and overdilution samples, respectively. Conclusion: A substantial number of samples showed hook effect, which could lead to missed late Lyme borreliosis manifestations and inaccurate intrathecal index calculations. A dilution threshold of 98.7 AU/mL in the Liaison Borrelia IgG immunoassay effectively identified 95% of hook effect and avoided more than 95% of overdilutions.