Previous Article | Next Article 
Journal of Clinical Microbiology, June 2003, p. 2791, Vol. 41, No. 6
0095-1137/03/$08.00+0 DOI: 10.1128/JCM.41.6.2791.2003
Copyright © 2003, American Society for Microbiology. All Rights Reserved.
| LETTER TO THE EDITOR |
Improving the Specificity of Recombinant Immunoassays for Lyme Disease
 Top Letter References |
|---|
A recent paper by Hefty et al. (3) identified recombinant lipoproteins OspE and ElpB1 from Borrelia burgdorferi as potential aids in the serodiagnosis of Lyme disease. However, the use of recombinant FlaB and OspC, in addition to the above antigens, may have been prematurely rejected by the authors. The Hefty paper, similar to that of Gomes-Solecki et al. (2), considers a serologic response to any of a group of recombinant antigens indicative of B. burgdorferi infection. This method of data analysis attempts to maximize sensitivity but fails to maximally utilize the improved specificity available through recombinant techniques. Assuming that the antigens so selected are highly specific for B. burgdorferi infection, it is reasonable to presume that prior exposure to these select antigens is an essentially random event in normal hosts and in most nonborrelial disease states. Thus, it may be predicted that the rate of false-positive reactions to a panel of recombinant antigens should approximate the sum of the reactions to the individual antigens selected; it should also follow that antibody reactions to two or more antigens in the panel should occur no more frequently than by chance alone in normal hosts. The independence of the antibody responses to these select antigens in normal hosts can be tested directly as previously described by me (4).
The four antigens identified by Hefty as most important in the humoral response to early Lyme disease—OspE, OspC, ElpB1, and FlaB—have specificities of 0.921, 1.00, 0.968, and 0.921, respectively (D. R. Akins, personal communication). As predicted, false-positive reactions to at least one member of the panel were seen in 11 of 63 (17.5%) normal hosts, but simultaneous reactions to two or more of these antigens were seen in only 1 of 63 (1.6%) controls (Akins, personal communication). The observed rate of multiple false-positive antibody responses in normal hosts is nearly identical to the 1.1% rate predicted by the formula from reference 4, supporting the independence hypothesis. Antibody responses to two or more of these four recombinant antigens were observed in 19 of 21 patients with early Lyme disease, a sensitivity equal to the more limited panel of OspE and ElpB1. However, the specificity achieved by requiring two or more antibody responses is greater than that of alternative antigenic panels (Table 1). Consistent with the guidance provided by Biggerstaff (1), the likelihood ratios (both positive and negative) associated with using antigen combination 4 (Table 1) are superior to those seen with the alternatives (except for antigen combination 3). Only when the pretest risk of Lyme disease exceeds 64% does the accuracy of combination 3 exceed that of combination 4. Since most tests for Lyme disease are ordered in a setting of low (<10%) pretest probability (4), antigen combination 4 may be preferred under most clinical circumstances. Additional data are needed to assess test specificity in patients with potential cross-reacting conditions such as syphilis and connective tissue disorders.
|
View this table: [in a new window] |
TABLE 1. Test performance of various antigen combinations (adapted from Hefty et al.a)
|
|
TopLetterReferences |
|---|
- Biggerstaff, B. 2000. Comparing diagnostic tests: a simple graphic using likelihood ratios. Stat. Med. 19:649-663.[CrossRef][Medline]
- Gomes-Solecki, M. J., G. P. Wormser, M. Schriefer, G. Neuman, L. Hannafey, J. D. Glass, and R. J. Dattwyler. 2002. Recombinant assay for serodiagnosis of Lyme disease regardless of OspA vaccination status. J. Clin. Microbiol. 40:193-197.
[Abstract/Free Full Text] - Hefty, P. S., C. S. Brooks, A. M. Jett, G. L. White, S. K. Wikel, R. C. Kennedy, and D. R. Akins. 2002. OspE-related, OspF-related, and Elp lipoproteins are immunogenic in baboons infected with Borrelia burgdorferi and in human Lyme disease patients. J. Clin. Microbiol. 40:4256-4265.
[Abstract/Free Full Text] - Porwancher, R. 1999. A reanalysis of IgM western blot criteria for the diagnosis of early Lyme disease. J. Infect. Dis. 179:1021-1024.[CrossRef][Medline]
|
Richard Porwancher*
Department of Medicine St. Francis Medical Center 601 Hamilton Ave. Trenton, NJ 08629
|
||||||
|
* Phone: (609) 581-2000 Fax: (609) 581-5450 E-mail: porwancher{at}aol.com. |
Journal of Clinical Microbiology, June 2003, p. 2791, Vol. 41, No. 6
0095-1137/03/$08.00+0 DOI: 10.1128/JCM.41.6.2791.2003
Copyright © 2003, American Society for Microbiology. All Rights Reserved.
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»