Background: Lyme disease (LD) antibody testing currently involves a 2-tiered algorithm with a whole-cell sonicate (WCS) enzyme immunoassay (EIA), followed by IgM/IgG Western immunoblots. A single EIA using the C6 peptide of the Borrelia burgdorferi variable-major protein-like sequence expressed lipoprotein provides similar or better sensitivity but less specificity, compared with standard 2-tiered testing. Here, we investigated an alternative 2-tiered strategy, in which the first step remained a WCS EIA, but immunoblotting was replaced by a C6 EIA.
Methods: We determined the sensitivity of the 3 testing strategies with use of 91 serum samples from research study patients with LD and 78 serum samples from patients with LD whose samples were submitted to our hospital's clinical laboratory. Specificity was measured using 54 patients with other illnesses and 1246 healthy subjects from areas where the infection is endemic and nonendemic.
Results: The 2-EIA algorithm in early LD had similar sensitivity as C6 testing alone, and both strategies had better sensitivity than did standard 2-tiered testing (61% and 64%, respectively, vs 48%; P = .03 and P = .008). For late disease, all 3 strategies had 100% sensitivity. The specificity of the 2-EIA algorithm was equal to that of standard 2-tiered testing, and both 2-tiered strategies were more specific than C6 testing alone (for both, 99.5% vs 98.4%; P = .01). The positive predictive value of the 2-EIA algorithm was 70%, compared with 66% for standard 2-tiered testing and 43% for the C6 EIA alone.
Conclusions: The 2-EIA strategy matched the individual strengths of the C6 EIA and Western blotting, without the drawbacks. The 2 EIAs provided sensitivity comparable to that of the C6 EIA but maintained the specificity of standard 2-tiered testing.