A comparison of lyme disease serologic test results from 4 laboratories in patients with persistent symptoms after antibiotic treatment

Abstract

Background: As the incidence of Lyme disease (LD) has increased, a number of "Lyme specialty laboratories" have emerged, claiming singular expertise in LD testing. We investigated the degree of interlaboratory variability of several LD serologic tests-whole cell sonicate (WCS) enzyme-linked immunosorbent assay (ELISA), immunoglobulin M (IgM) and immunoglobulin G (IgG) Western blots (WBs), and an ELISA based on the conserved sixth region of variable major protein-like sequence expressed (C6)-that were performed at 1 university laboratory, 1 commercial laboratory, and 2 laboratories that specialize in LD testing.

Methods: Serum samples from 37 patients with posttreatment Lyme syndrome, as well as 40 medically healthy controls without prior LD, were tested independently at the 4 laboratories.

Results: In general, there was little difference among the laboratories in the percentage of positive test results on the ELISAs and IgG WBs, although the number of discordant results was often high. When in-house criteria for positivity were used at the 2 specialty laboratories, specificity at 1 laboratory declined considerably on both the IgM and IgG WBs. The Centers for Disease Control and Prevention (CDC) 2-tiered criteria improved overall concordance. At the 2 laboratories that performed the C6 ELISA, the percentage of positive tests was comparable to that of the WCS ELISA while providing higher specificity. The IgM WB performed poorly in our patient population of individuals with later-stage illness, a result consistent with previous studies.

Conclusions: Although there was surprisingly little difference among the laboratories in percentage of positive results on most assays using CDC criteria, interlaboratory variability was considerable and remains a problem in LD testing.

Keywords: Borrelia burgdorferi; Lyme; diagnosis; serology.