Accuracy of IgM Immunoblotting to Confirm the Clinical Diagnosis of Early Lyme Disease
- 14 October 1996
- journal article
- research article
- Published by American Medical Association (AMA) in Archives of internal medicine (1960)
- Vol. 156 (18) , 2105-2109
- https://doi.org/10.1001/archinte.1996.00440170121013
Abstract
Background: A 2-test approach for the serologic diagnosis of Lyme disease has recently been proposed. A positive or equivocal result on a first-stage test (eg, an enzyme immunoassay) is followed by a Western immunoblot test. For a sample to be considered seropositive for Lyme disease, the immunoblot result must be positive. Objectives: To assess the accuracy of IgM immunoblotting for detection of early Lyme disease and to establish interpretative criteria for a commercially available immunoblot assay. Methods: Serum samples from 44 patients with erythema migrans were tested by an IgM immunoblot assay. All patients were culture-positive for Borrelia burgdorferi. Serum samples from 2 different control groups were also tested. Interpretative criteria were developed using receiver operating characteristic curves. Results: The presence of any 2 IgM bands was found to be the optimal criterion for a positive test result, and in patients with illness of less than 7 days' duration, this was significantly more sensitive than the criterion of any 2 of the 3 specific bands defined by the Centers for Disease Control and Prevention/Association of State and Territorial Public Health Laboratory Directors Lyme Disease Workgroup (P<.05). Specificity of the criterion of any 2 bands was 100% for 1 group of controls but only 96% for the more clinically relevant control group; this small difference had a large impact on the positive predictive value in populations at low risk for Lyme disease. Conclusions: Using a commercially available immunoblot test kit, the presence of any 2 IgM bands is proposed as a positive result. The predictive value of a positive IgM immunoblot result, however, is poor in patients with minimal clinical evidence for Lyme disease. Arch Intern Med. 1996;156:2105-2109This publication has 0 references indexed in Scilit: