Practical Guidelines for Lyme Testing

How to Best Use the Laboratory

The purpose of these guidelines is to assist the clinician in the evaluation of patients presenting with symptoms suggesting possible Lyme disease. An attempt is made to summarize what laboratory methods and techniques are available, and how to apply them to common, specific clinical presentations. General information is given regarding performance, criteria for test positivity, and on which tests may be most appropriate and helpful in various clinical settings. It should be emphasized that serologic tests can confirm prior exposure to Borrelia burgdorferi (seroconfirmation) but a positive test does not make the diagnosis of active infection.

False positive results are common for indirect ELISA. The use of screening serologies, e.g., testing all pregnant women to see if they might have exposed previously, in unwise in that the false positive rate will outweigh the true positive rate by a factor of well over 12 to 1. This is to say, in a population where the a priori pretest likelihood of Lyme disease is low, a positive test is vastly more likely to be a false positive.

The reader is referred to recent reviews on clinical(1,2,3,4) and laboratory(4,5) aspects of Lyme disease. The CDC recommends immunoblot confirmation of positive or equivocal ELISA results. Because of concerns over sensitivity, specificity, and overall test performance as a "confirmatory" strategy, The FDA has suggested that Western blotting (WB) should be viewed as providing adjunctive or supplemental information, because it lacks the performance characteristics of a true reliable confirmatory test. Both IgM and IgG WB in certain clinical situations lack sensitivity, and the IgM WB may show problematic specificity. Certainly, a "positive" IgM WB in a patient with chronic (greater than 1 month) symptoms, as the sole serologic finding, should NOT be considered sufficient laboratory evidence supporting a diagnosis of Lyme disease, "because the likelihood of a false-positive test is high for these individuals." (6)

These concepts are especially helpful in determining the stage of Borrelia burgdorferi (Bb) infection. WB definitely cannot be relied upon in the serologic confirmation of early infection. Patients with very early Lyme disease may be seronegative initially, especially be WB. It is quite common to see such patients demonstrate seroreactivity (even quite strikingly high titres by IgM, IgG, and IgA by antibody capture EIA) prior to the development of a positive WB. This latter pattern of findings (clearly positive EIA's with a negative WB) fails to meet CDC criteria for the laboratory confirmation of Lyme disease. Excellent reviews of the immunopathogenesis and immune effects of Lyme disease have recently been published. (7)

Information is organized and presented in accordance with generally accepted clinical stages:

I. Early Lyme Disease
II. Early Disseminated Lyme Disease
III. Late Lyme Disease
IV. Special Considerations

Most of the comments which follow pertaining to serologic test performance relate to antibody capture test methodologies as the proposed laboratory technique for enzyme linked immunoassay.(EIA) Capture EIA has been shown to be a relaible methodology to confirm early infection. (8) For this and Western immunoblotting, conventional pathologic strains of B. Burgdorferi are used as the test reagent. However, with the release of recombinant OspA Lyme disease vaccines, these test recommendations will become problematic, since sera from vaccinated subjects, in the absence of genuine Lyme disease, will test positive by all commonly used ELISA tests, and will also frequently show complex reactivity on Western immunoblotting which will not be straightforward to interpret. This is further addressed in the section on Vaccine Related Issues.

The term "positive" immunoblot requires caution since there exists some controversy as to which bands are specific for Bb infection, which are less specific, and how many of these bands need to be observed to conclude a test result is evidence for Bb infection. Treatment strategies and recommendations are not discussed.

I. Early Lyme Disease

Erythema Migrans (EM)

When this pathognomic skin lesion is present, the diagnosis of Lyme disease is reasonably certain even without laboratory confirmation.(8) Many clinicians do not seek serologic confiration in this situation however, since patients with EM are often seronegative A baseline serologic specimen may be invaluable in the patient who presents with subsequent symptoms, for follow-up evaluation. Early serologic specimens can be very useful, for example, in the evaluation of suspected treament failures, re-infections, and of concomitant or subsequent infection with other tick-borne pathogens such as Babesia microti or the agent of Human granulocytic ehrlichiosis (HGE). It has been suggested that more severe constitutional symptoms, in the setting of acute Lyme disease, should strongly suggest the possibility of co-infection. In the setting of EM the diagnosis of Lyme disease should not be ruled out by a negative serologic test, since in very early infection there may not have been sufficient time to develop an immune response. IgM is the first antibody isotype class to appear, followed by an isotype switch to IgG antibody production after a period of weeks. Furthermore, after B. burgdorferi exposure, early (even inadequate) treatment can render a patient permanently seronegative. Therefore, in early or treated patients, seropositivity is not a necessary condition for making the diagnosis of Lyme disease.

A sensitive enzyme linked immunoassay (ELISA) such as an antibody capture test, is positive for IgM isotype antibodies against B. burgdorferi in at least 65-75% of patients with EM during this early phase.(9) In one series of 22 patients with culture positive EM, the IgM capture antibody test was positive in 91% of cases.(10) During convalescence this test will be positive in 90-95% of treated patients despite appropriate and curative antibiotic therapy.

IgG antibody isotype is typically negative in early EM patients, and IgG immunoblotting (Western Blotting) often shows 1 or a few non-specific bands (i.e. a negative test result), or it may be entirely non-reactive. Thus, a non-reactive IgG immunoblot cannot be taken as evidence against the existence of early Lyme disease.

The presence of significant IgA levels suggests early dissemination.(11,12) The presence of significant levels of concomitant IgA and IgM levels is virtually diagnostic of recently acquired B. burgdorferi infection. Again, the IgG capture ELISA or immunoblot may be positive or negative at this stage.

IgG antibody response develops after IgM and IgA and can appear even after a patient is successfully treated. For example, if the initial test is done in late EM, the isotype switch to IgG production may already be in progress. Therefore a modest level of IgG and a limited blot pattern in this setting does not imply treatment failure and is often seen in successfully treated patients.

After successful teatment, if a patient is followed serologically, IgM can rise for about one month, then slowly decrease. However, it may remain positive in many cases for greater than one year. Therefore if follow-up testing is performed, a persistent (but lower) level of IgM does not in itself imply persistent infection. IgA is expected to become negative, after adequate antibiotic therapy, over one to a few months.* Therefore, treatment failure might be suspected if, after antibiotic therapy, persisting or rising levels of B. burgdorferi specific IgA titres, or significantly rising IgM titres, or marked expansion of immunoblot bands, are demonstrated. Routine follow-up serologic tests to assess the efficacy of antiobiotic therapy are not routinely recommended, but may be of value in the evaluation of patients remaining symptomatic after treatment.

* IgA follow-up data on file - submitted for publication

Atypical Rash

In an endemic area, serologic testing is recommended when an atypical rash is suspected to be a presenting feature of Lyme disease.

In non-endemic areas or with low exposure risk, serologic testing may not be warranted. When an atypical rash is present and a patient is from a low endemic area or there exists a low clinical suspician for Lyme disease, one approach is to not treat, follow clinically, and perform acute (baseline) and convalescent serologies. Another approach is to obtain biopsy or skin aspirate specimen at the margin of the rash and test by PCR.** If atypical rash is caused by Borrelia burgdorferi infection, PCR is frequently positive and confirms the diagnosis. Since neither the PCR or serologic testing is 100% sensitive in the setting of early Lyme disease, combining the two may improve the diagnostic yield.

[**The recommendation for PCR assumes the following: the laboratory has taken the necessary precautions and implemented the necessary laboratory procedures and performs appropriate quality control and quality assurance to avoid amplicon contamination, which would yield false positive results.]

Viral-Like Symptoms - With or Without Atypical Rash

Depending on the prevalence of Lyme disease and other tick-borne infections in the geographic area in which the disease was suspected of being acquired, testing of baseline (acute) and convalescent sera for Lyme disease and possibly Babesia and/or Human Granulocytic Ehrlichiosis (HGE) and Human Monocytic Ehlrichiosis (HME) may be warranted. Whole blood PCR is available for Borrelia burgdorferi, Babesia microti, HGE, and HME. If an atypical rash is present, one can obtain skin aspirate or biopsy specimens at the margin of the rash for B. burgdorferi PCR testing.

Specimens from patients with non-specific viral like illnesses are frequently sent for Lyme disease antibody testing. In such cases, serologic testing may reveal low levels of IgM B. burgdorferi antibodies or variable, occasionally borderline positive results on IgG immunoblots. These may be cross-reactive antibodies to Ehrlichia or other bacterial or viral infections. Therefore testing for related conditions which cross-react with Lyme disease antibody test is frequently considered. (e.g. HME, HGE, Babesia). These isolated IgM levels are cross reactive (i.e. false positive for B. burgdorferi infection) in an untreated patient when a follow-up serologic profile shows no isotype switch or expansion of immunoblot reactivity. Likewise an isolated IgG antibody capture or immunoblot result in an acutely ill patient is not suggestive of Lyme disease. The IgG reactivity in this setting may be a false positive or may represent remote Lyme disease - i.e., the pattern of positive IgG and negative IgM does not suggest recent infection.

No Rash, Non - Specific Symptoms

In the absence of a rash, but in the presence of viral-like illness, especially in the appropriate season and in an endemic area, the question arises as to whether to test for Lyme disease, or wait. In certain geographic areas, this clinical syndrome is as likely to be Babesia or Ehrlichia, or some non-specific viral syndrome, as B. burgdorferi. This is confounded by serologic cross-reactivity - i.e. patients with HGE or Babesiosis can present with reactive Lyme disease immunoblots, possibly due to cross reacting antibodies.(13) Therefore a panel of tests for tick-borne diseases may often be appropriate.

A laboratory work-up for tick-borne diseases is not usually recommended or helpful, if vague, chronic, non-specific complaints (e.g. chronic fatigue, other vague complaints or fibromyalgia) present in any geographic area, or if acute or chronic, non-specific complaints are present in a non-endemic area. (5,14) Thus, when the prior likelihood of Lyme disease is very low, serologic testing is rarely useful, may be falsely positive, and may lead to adverse consequences of overdiagnosis and overtreatment.(15)


II. Early Disseminated Lyme Disease

Early dissemination may occur at the time of EM infection, and even may occur early enough so that the seroreactivity has not yet emerged. By the time of the clinical manifestations mentioned below however, as well as in the patient with multiple EM lesions due to hematogenous dissemination, seroreactivity is usually present.

Early Neurologic Disease

Neurologic manifestations of this stage of Lyme disease include meningitis, facial and other cranial nerve palsies, and radiculoneuropathy. Serology can confirm a particular neurological syndrome as a manifestation of Lyme disease. Serology, when performed by antibody capture methodology, is virtually always positive in Lyme meningitis; however by some other ELISA methodologies it is possible to see the absence of serologic positivity at this stage.

Most 7th cranial nerve palsy (16) in most geographic areas is not due to Lyme disease. Serology is expected to be positive in this setting and will confirm B. burgdorferi as the etiologic agent. IgA is often present in this stage.(9) The first test of choice in all suspected neuroborreliosis is serology. There are very few well documented published cases of seronegative CNS Lyme disease. CSF evaluation is usually not done and is not required for diagnosis of cranial neuropathies due to Lyme disease. If done, CSF testing will typically demonstrate production of intrathecal antibodies.(17) Serum:CSF ratios of B. burgdorferi specific antibodies do not need to be corrected for total protein when capture assays are used. For IgG, a ratio of 1.5 or higher is significant. For IgM and IgA a CSF/serum ratio value of 0.5 is significant, although this testing has not yet been standardized.

The diagnosis of meningitis of any cause rests on CSF analysis. Paired serum/CSF for standard evaluation for antibody testing is useful. Even after antibiotic therapy, significant CSF antibody levels, if initially present (even with elevated ratios), can persist for months to years, and are of unknown significance. Both culture and PCR are available on CSF specimens and may be appropriate, as described in a subsequent section. Laboratory confirmation is essential in allowing the physician to determine whether or not B. burgdorferi is the etiology of lymphocytic meningitis.

Cardiac Disease

The most common cardiac manifestation of B. burgdorferi infection is heart block. Serology is the only available practical test. ELISA and immunoblot are expected to be significantly positive. The serologic pattern resembles that seen in Lyme meningitis, and typically demonstrates IgG antibody by ELISA and a moderately expanded response by Western Blot.


III. Late Lyme Disease

Arthritis

Inflammatory recurrent episodes of monoarthritis or oligoarthritis are the typical musculoskeletal manifestations of late Lyme disease. Serology is the best diagnostic method in this setting. There are no well documented cases of Lyme arthritis in North American patients in the absence of a positive serology, using the above mentioned test methodologies. A negative test generally rules out Lyme arthritis in an immunocompetent host. Patients with this late Lyme disease manifestation typically exhibit very high levels of IgG antibody by ELISA, and intensely expanded immunoblot patterns. There is a role of PCR testing on synovial fluid (18). Recent evidence suggests that PCR testing on synovial menbrane specimens may be more sensitive than testing on synovial fluid. (19)

Although a negative serology virtually rules out Lyme arthritis, in an immunocompetent untreated host, a positive serology does not absolutely assure that the arthritis is due to B. burgdorferi infection. The positive test may be either a false positive or it may reflect residual antibody from a remote episode of Lyme disease independent of the patient's presentation now. Serologic tests can be repeated some months after appropriate antibiotic therapy. The expected finding in a successfully treated patient would be no increase in ELISA titers, and no further expansion of reactivity on immunoblotting. PCR of joint fluid would also be expected to become negative, although the exact timing during or following treatment at which PCR reverts to negative needs to be better characterized.

Late Neurologic (CNS and Peripheral)

Serology is an essential first test. These patients usually have positive serologic tests, unless previously treated or immunosuppressed. We are not aware of any well documented published cases of CNS Lyme disease in the context of negative serologic results.

A positive finding in paired serum and CSF antibody testing generally diagnoses CNS Lyme disease. By antibody capture EIA, a positive result is defined as a ratio of CSF to serum IgG antibody levels of 1.5 or higher. In routine CSF studies, a lymphocytic pleocytosis is not expected.

PCR testing on CSF is available but not highly sensitive in this setting and has a lower sensitivity than antibody levels or ratios. We have not seen cases of neuroborreliosis diagnosed solely by PCR testing of CSF,
in the absence of intrathecal antibody concentration. Perhaps an explanation of this is that the spirochete does not seem to persist in CSF, seeming to have a higher affinity for certain tissues.

All CSF findings should be interpreted in the context of routine CSF studies and clinical features. Positive CSF ratios can persist over long periods in successfully treated patients. In the setting of negative serology and negative CSF antibodies, it is highly unlikely to see a positive PCR in CSF.(10)

Rare Late Skin Manifestations

Acrodermatitis chronicum is rare (especially in the United States); as in other late manifestations of B. burgdorferi infection, serologic tests are expected to be positive. In assisting in the diagnosis of this condition, the laboratory can also perform PCR testing on a skin aspirate or skin biopsy specimen.


IV. Special Consideration

Vaccine Related Issues

Recombinant OspA vaccines have recently been released to protect against B. burgdorferi infection. The vast majority of vaccinated patients will test positive for Lyme disease antibodies on currently available commercial ELISA test kits. By Western blotting, the typical pattern of reactivity in a vaccinated patient is the presence of a strong band at 30kd corresponding to OspA antibody, often accompanied by up to several other familiar bands, at both higher and lower molecular weights, of uncertain significance. The development of multiple bands after vaccination with recombinant OspA protein has been described and complicates the interpretation of Western blot results in vaccinated patients. (9,,24)

New generations of laboratory tests have been introduced to address this. Alternate strains of Bb are available which genetically lack the ability to express OspA, and which can be used as test reagents in both antibody capture EIA and Western blotting. All vaccine-induced reactivity, by both EIA and WB, is virtually eliminated by the use of these new OspA-free strains.(25)
Validation and further studies of test performance using this methodology are under way; these tests are proving highly useful for Lyme disease antibody testing in populations where vaccine use is widespread. A recent National Institutes of Allergy and Infectious Diseases Conference, held in collaboration with the CDC, highlighted this in making the "recommendation that, in the future, only bacterial antigens derived from OspA-deficient mutants of Borrelia burgdorferi be used in all diagnostic assays to circumvent false positive reactions likely to result from the widespread use of OspA vaccines." (26)

Staging Infection

Attempts have been made to learn about the stage of Bb infection, based upon the information provided above on the characteristic antibody profiles found in patients with early localized, early disseminated, and late Lyme disease. Knowledge of these typical patterns of immune response may allow one to estimate the timing or stage of infection from a set of antibody findings. This is best illustrated by first looking as sets of well-characterized patient groups, and studying their antibody profiles.

An example is shown in Figure 1, which are the laboratory results by capture EIA of a group of 10 patients, all untreated, who sequentially experienced physician diagnosed EM, then went on to get Lyme meningitis, and then went on to get Lyme arthritis. Six serum specimens were available on each patient during well defined time periods during their course of infection. Note the persistence of significant IgM levels well into late disease, and the steadily rising IgG levels as the patients clinically evolved into late Lyme disease. Figure 2 illustrates the WB patterns, with EIA below, on six sequential specimens from one of these individual patients. Here, note the evolution of an expnaded number and intensity of bands as the infection evolves.

These characteristic patterns have been repeatedly seen and confirmed in subsequently studied well characterized patient groups, and have assisted the laboratory in test interpretation and in contributing to clinical correlations in the context of a given set of serologic results.

Other Tests

There are other laboratory tests available or under investigation to assist the clinician in evaluating a patient for possible Lyme disease. The Lyme urinary antigen testing (LUAT) has not been validated and there are no prospective controlled studies of this test demonstrating any diagnostic usefulness. This test is not recommended for the diagnostic confirmation or management of Lyme disease. "Provocative testing", where samples of urine are taken before and after a course of antibiotic treatment, cannot be endorsed in the evaluation of such patients - this practice remains of no proven value.

A new assay has been proposed as a potential improvement for the serological confirmation of Bb infection. It's development is based on the fact that B. burgdorferi releases large quantities if its proteins. This suggested that in some circumstances specific antibodies might be bound in immune complexes and therefore sometimes undetectable in standard serologic assays. Thus an immune complex-based assay might be able to detect small amounts of antibodies, e.g. at the very earliest stages of infection. As well, since active production of specific immune complexes requires ongoing production of antigen derived from live organisms, persistence of immune complexes would imply persistence of infection. This new assay, called EMIBA (Enzyme-linked IgM-capture Immune complex Biotinylated-antigen Assay) was developed to detect IgM anti-B. burgdorferi in the immune complexes purified from Lyme disease patients' sera and to hopefully differentiate active infection from prior treated and cured infection with ongoing seropositivity. The theory is that the ongoing production of immune complexes depends on the ongoing presence of Borrelial antigens (i.e., ongoing infection) In preliminary studies, this assay had been promising in serologically confirming the diagnosis of early Lyme disease and has been able to differentiate active infection from persistence of serological reactivity in patients cured of their Lyme disease. If these initial studies are confirmed EMIBA may be useful in interpreting persistent seropositivity in patients with onging symptoms not necessarily classic for active Lyme disease.

A next generation assay is currently in development using peptide epitopes rather than whole Borrelia or even recombinant proteins as the source of antigen. Use of specific epitopes allows one to minimize false-positivity due to extraneous and irrelevant serum antibodies while focusing on the specific targets of the earliest humoral immune response to B. burgddorferi.

T-cell proliferative assays have been evaluated in limited patient populations(30), have not been validated in larger groups of patients, are not generally commercially available, and are still considered investigational at this time. False-positive and false-negative T-cell proliferative tests are observed.

In the evaluation of neuroborreliosis, other non-serologic studies have been proposed. Imaging procedures such as brain MRI scans or SPECT scans are often ordered, and are mainly of value in excluding other diagnoses as causes of presenting symptoms. SPECT scans in particular, although very popular and "in vogue" in some areas, have never been validated in well characterized patients. Their specificity and predictive value in the diagnosis of CNS Lyme disease are unproven and unknown. In other words, there are no SPECT scan findings specific for CNS Lyme disease, and some of the reported nonspecific findings have also been seen in patients with fibromyalgia or following cervical whiplash injuries.(31,32) In contrast, formal neuropsychological testing appears to show characteristic findings in patients with CNS Lyme disease(33) although these findings are certainly absolutely specific for neuroborreliosis. They occasionally have been used to follow the progress of patients over time after antibiotic therapy, and the absence of interval change can be used to re-assure the patient and allow the tailoring of a cognitive rehabilitation program for that individual.

Follow-up Serologic Testing

After appropriate diagnosis and treatment, patients should be followed up clinically. In patients who become asymptomatic promptly with therapy and who exhibit an uncomplicated clinical resolution, follow-up laboratory testing is not routinely necessary or helpful. High levels of Bb antibodies - even high levels if IgM as well as IgG - can persist in successfully treated patients, even for years.(27) Thus, the clinician should not conclude that persistence of the antibody findings constitutes laboratory evidence of persistent infection, or that it implies the need for further courses of antibiotic therapy. This is a common misconception amongst some physicians caring for Lyme disease patients. Nor is there any evidence for rationale for serially following IgM WB results, as the sole evidence of Bb infection, in patients with chronic symptoms, as a guide to continued courses of antibiotics.

Follow-up testing can be of value in patients suspected of relapse or re-infection. If reliable sequential testing demonstrates rising titres by EIA (especially of multiple isotypes), or clear-cut expansion of new and more intense bands by WB, the clinician could interpret this as laboratory evidence supporting ongoing stimulation of the immune system from active infection. Such sequential test comparisons should not be made when tests are done in different laboratories because of the well-known interlaboratory variations in Lyme disease testing. (28) Ideally, to minimized test-to-test variation, and to optimize the validity of the comparisons, paired sera should be tested together on the same test rum in the laboratory.

Conclusion

Guidelines have been presented to assist the clinician in evaluating various clinical presentations for the possibility of Lyme disease, although all nuances and variations have not been covered. A basic understanding of available serologic and other test methods, including their performance characteristics and intrinsic limitations, will provide valuable information and contribute to the diagnostic evaluation in most cases. Methodologically sound and carefully interpreted laboratory medicine results can supplement and improve efficient and proper clinical assessment. Persisting or chronic symptoms following treatment for Lyme disease, as well as the emerging use of OspA Lyme disease vaccine, will challenge the clinician and laboratory alike. New test systems using OspA-free Bb strains, and possibly of immune-complex-based assays, have been developed or are being proposed to address this.

REFERENCES

1. Evans J. Lyme Disease. Current Opinion Rheum 1997;9(4):328-336

2. Sigal LH. Summary of the Sixth International Conference on Lyme Borreliosis. Arthritis Rheum 1995; 38:565- 569.

3. Steere AC. Lyme Disease: Diagnosis and Treatment of Lyme disease. in Kelley WN, Harris ED, Ruddy S, Sledge CB, editors. Textbook of Rheumatology 1993. Update #8:1-16.

4. Sigal LH. Pitfalls in the Diagnosis and Management of Lyme Disease. Arthritis Rheum 1998;41:195-203.

5. American College of Physicians. Guidelines for Laboratory Evaluation in the Diagnosis of Lyme Disease. Ann Int Med 1997;127:1106-1123.

6. Lyme disease - United States, 1993. MMWR Morb Mortal Wkly Rep 1994;43:564-71

7. dressler and steere paper on pred value/sensitivity of WB

8. Malane MS, Grant-Kels JM, Feder HM, Luger SW. Diagnosis of Lyme disease based on dermatologic manifestations. Ann Intern Med 1991;114:490-498

9. Berardi VP, Weeks KE, Steere AC. Serodiagnosis of early Lyme disease; Evaluation of IgM and IgG Antibody Responses by Using an Antibody Capture Enzyme Immunoassay. J Infect Dis 1988;158:754-760.

10. Unpublished Data, on file at Imugen, Inc.

11. Berardi VP, Seder RH, Accuracy of Diagnosis and Staging of Well-Characterized Lyme Disease Patients (CDC) by an Antibody Capture/Western Blot Panel, Compared with IgM and IgG ELISA, and IgM and IgG Western Blotting and a Flagellin-enriched ELISA. Second National Conference on Serologic Diagnosis of Lyme Disease; Dearborn, MI 1994 (abstract).

12. Davis S, Berardi VP, Luger S, Seder RH, Stack K. Correlation of B. burgdorferi-specific Serum IgA with Symptom Severity in Early Lyme Borreliosis. V International Conference on Lyme Borreliosis; Arlington, VA 1992 (abstract).

13. Wormser GP, Horowitz HW, Nowakowski J, McKenna D, Dumler JS, Varde S,Schwartz I, Carbonaro C, Aguero- Rosenfeld M. Positive Lyme disease Serology in Patients with Clinical and Laboratory Evidence of Human Granulocytic Ehrlichiosis. Am J Clin Pathol;1998 Oct;108(4):479-483.

14. Dinerman H, Steere A. Lyme Disease Associated with Fibromyalgia. Ann Int Med 1992;281-285.

15. Reid MC, Schoen RT, Evans J, Rosenberg JC, Horwitz RI. The Consequences of Overdiagnosis and Overtreatment of Lyme Disease; An Observational Study. Ann Intern Med 1998;128:354-362.

16. Kuoper H, Devriese P, Bartelt M, Vosk, Dankert J. Absence of Lyme Borreliosis Among Patients With Presumed
Bell's Palsy. Arch Neurol 1992;49:940-943.

17. Klaus H, Lebech AM. Lyme Neuroborreliosis: A New Sensitive Diagbnostic Assay for Intrathecal Synthesis of Borrelia burgdorferi - specific Immunoglobulin G,A, and M. Ann Neurol 1991;30:197-205.

18. Nocton J, Dressler F, Rutledge B, Rys P, Persing D, Steere A. Detection of Borrelia Burgdorferi DNA by Polymerase Chain Reaction in Synovial Fluid From Patients with Lyme Arthritis. N Engl J Med 1995;330:229-234.

19. Priem S, Burmester GR,. Karnradt T, Wolbart K, Rittig MG, and Krause A. Detection of Borrelia burgdorferi by polymerase chain reaction in synovial membrane, but not in synovial fluid, from patients with persisting Lyme arthritis after antibiotic therapy. Ann Rheum Dis 1998;57:118-121

20. Molloy PJ, Berardi VP, Weeks KE, Persing DH, Halling VW. Analysis of serologic responses to OspA Lyme Disease Vaccine. (abstract) Arthritis Rheum 1998; 41 No. 9(suppl):S130

20. Zhang Y, Mathiesen D, Kolbert CP, Anderson J, Schoen R, Fikrig E, and Persing D. Borrelia burgdorferi Enzyme- linked immunoabsorbant assay for discrimination of OspA vaccination from spirochete infection. J Clin Microbiol
1997;35:233-238

21. multiple bands manuscript, submitted

22. Zhang Y, Mathiesen D, Kolpert CP, Anderson J. Schoen R, Fikrig E, Persing D. Borrelia burgdorferi
enzyme-linked immunosorbant assay for discrimination of OspA vaccination from spirochete infection.
J Clin Microbiol 1997;35:233-238

23. Dressler F, Yoshinari NH, and Steere A. The T-cell proliferative assay in the diagnosis of Lyme disease.
Ann Intern Med 1991;115:533-539
24. Bicik I, Radanov BP, Schafer N, Dvorak J, Blum B, Weber B, Burger C, Von Schulthess CK, Buck A. PET with 18 flourodeoxy glucose and hexamethylpropylene amine oxime SPECT in late whiplash syndrome. Neuro 1998;Aug;51:345-50.

25. Mountz JM, Bradley LA, Modell JG, Alexander RW, Traina-Alexander M, Aaron LA, Stewart KE, Alarcon GS, Mountz JD. Fibromyalgia in Women. Abnormalities of Regional Blood Flow in the Thalamus and in the Caudate nucleus are Associated with Low Pain Thresholds. Arth Rheum 1995;38(7):926-38.

26. Krupp LB, Masur D, Schwartz J, Coyle PK, Langenbach LJ, Fernquist SK, Jandorf L, Halperin JJ.
Cognitive functioning in late Lyme borreliosis. Arch Neuro 1991;48:1125-1129


 


 220 Norwood Park South
Norwood, MA 02062
1-800-246-8436
1-781-255-0770