Diagnosis of Lyme disease

INTRODUCTION — Lyme disease is the most common tick-borne disease in the United States, Canada, and Europe [1-3]. It is a bacterial infection caused by six species in the spirochete family Borreliaceae. The taxonomy of these spirochetes is undergoing revision, and the genus name may be represented as either Borrelia or Borreliella. In either case, the abbreviation for the genus is "B" and stands for both terminologies in the discussion below.

In North America, infection is caused primarily by B. burgdorferi sensu stricto (hereafter called B. burgdorferi) and, less commonly, in a region of the upper Midwest, by B. mayonii. In Europe and Asia, infection is caused primarily by either B. afzelii or B. garinii, less commonly by B. burgdorferi, and rarely by B. spielmanii or B. bavariensis. There is a broad spectrum of manifestations, and severity of disease is due, in part, to differences in the infecting species.

The diagnosis of Lyme disease will be reviewed here. Specific considerations regarding the diagnosis of Lyme arthritis and neurologic Lyme disease, as well as topic reviews that discuss the microbiology, epidemiology, immunopathogenesis, clinical manifestations, treatment, and prevention of Lyme disease, are presented elsewhere.

●(See "Microbiology of Lyme disease".)

●(See "Epidemiology of Lyme disease".)

●(See "Immunopathogenesis of Lyme disease".)

●(See "Prevention of Lyme disease".)

●(See "Evaluation of a tick bite for possible Lyme disease".)

●(See "Clinical manifestations of Lyme disease in adults".)

●(See "Lyme disease: Clinical manifestations in children".)

●(See "Lyme carditis".)

●(See "Musculoskeletal manifestations of Lyme disease".)

●(See "Nervous system Lyme disease".)

●(See "Treatment of Lyme disease".)

CLINICAL SUSPICION FOR LYME DISEASE — The diagnosis of Lyme disease should be suspected in patients who are at risk of exposure to ticks carrying Lyme disease and have clinical manifestations that are consistent with Lyme disease. Once Lyme disease is suspected, the approach to diagnosis depends upon the stage of disease. (See 'Approach to diagnosis' below.)

Risk of exposure — Lyme disease is a spirochetal infection caused by Borrelia species (B. burgdorferi and B. mayonii in the United States, and primarily B. afzelii and B. garinii in Europe and Asia). The organisms are transmitted by the bite of infected Ixodes ticks (I. scapularis, I. pacificus, I. persulcatus, I. ricinus). (See "Microbiology of Lyme disease".)

Patients are at risk for Lyme disease if they are from or have travelled to an endemic area. In the United States, the primary endemic areas include Connecticut, Delaware, Maine, Maryland, Massachusetts, Minnesota, New Hampshire, New Jersey, New York, Pennsylvania, Rhode Island, Vermont, Virginia, West Virginia, and Wisconsin. State-specific information can be found on the United States Centers for Disease Control and Prevention website. In Europe, Lyme has been seen in many countries and is particularly prevalent in Central and Eastern Europe and southern parts of Nordic countries [4]. (See "Epidemiology of Lyme disease".)

Transmission is most likely to occur from May through October, with a peak in June in northeastern states. In addition, certain occupations and activities (eg, hiking, gardening) increase the likelihood of exposure. However, patients can still be at risk for Lyme disease if they are from an endemic area and do not report a specific exposure.

For clinicians in nonendemic areas for Lyme disease who see patients with possible Lyme disease, it is especially important to obtain a detailed travel and activity history, with inquiries about prior residences and prior clinical findings that might be consistent with manifestations of Lyme disease. Without this history, an erythema migrans (EM)-like skin lesion could be misdiagnosed as either a spider bite or as community-acquired cellulitis due to pyogenic bacteria. Such patients are often treated with a first-generation cephalosporin (eg, cephalexin), which is ineffective for treating Lyme disease [5,6].

Similarly, an appropriate travel history can help distinguish Lyme disease from Southern tick-associated rash illness (STARI), an illness principally reported in the southeast and south-central regions of the United States. In patients with STARI, a skin lesion indistinguishable from EM occurs following the bite of the Amblyomma americanum tick, not the Ixodes species tick. A. Americanum ticks have been identified as far north as Maine, so the range of A. americanum and I. scapularis ticks now overlap. (See "Treatment of Lyme disease", section on 'Early localized disease (single erythema migrans)'.)

Consistent clinical manifestations — The clinical manifestations of Lyme disease can generally be divided into three phases: early localized, early disseminated, and late disease (table 1). However, the clinical features of each stage can overlap, and some patients present in a later stage of Lyme disease without a history of prior signs or symptoms suggestive of earlier Lyme disease:

●Early localized disease is characterized by the appearance of the characteristic skin lesion, EM, with or without constitutional symptoms (picture 1). EM usually occurs within one month following the tick bite.

●Early disseminated disease is characterized by multiple EM lesions (that typically occur days to weeks after infection) and/or neurologic and/or cardiac findings (that typically occur weeks to months after infection). Some of these patients have no history of antecedent early localized Lyme disease.

●Late Lyme disease is typically associated with intermittent or persistent arthritis involving one or a few large joints, especially the knee (sometimes preceded by migratory arthralgias). Symptoms may develop months or longer after the initial infection and are most likely to be the presenting manifestation of the disease when early disease is asymptomatic or misdiagnosed. In untreated patients, objective manifestations of late Lyme disease, such as monoarticular or oligoarticular arthritis, may persist for several years. On rare occasion, late Lyme disease can also present with neurologic problems, such as a subtle encephalopathy or polyneuropathy.

The clinical manifestations of Lyme disease are discussed in greater detail separately. (See "Clinical manifestations of Lyme disease in adults" and "Lyme disease: Clinical manifestations in children".)

When Lyme testing is not warranted — Serologic testing for Lyme disease should not be performed:

●For screening of asymptomatic patients who live in endemic areas

●For evaluation of patients who present only with nonspecific symptoms (eg, subacute or long-standing myalgias/arthralgias or fatigue)

We do not test these patients, since the use of serologic testing in populations with a low pre-test probability of Lyme disease results in a greater likelihood of false-positive test results than true-positive test results. As a consequence, asymptomatic patients and those with nonspecific articular or neurologic symptoms may be incorrectly diagnosed as having "chronic Lyme disease" and receive unnecessary antibiotics. In patients with noninflammatory musculoskeletal pain, cognitive complaints, fatigue, and/or irritability, alternative diagnoses should be considered, including chronic fatigue syndrome, fibromyalgia, myofascial pain syndrome, polymyalgia rheumatica, hypothyroidism, inflammatory myopathies, and chronic traumatic encephalopathy. (See "Clinical manifestations and diagnosis of fibromyalgia in adults" and "Clinical features and diagnosis of myalgic encephalomyelitis/chronic fatigue syndrome" and "Differential diagnosis of fibromyalgia" and "Overview of soft tissue musculoskeletal disorders", section on 'Myofascial pain syndrome' and "Clinical manifestations and diagnosis of polymyalgia rheumatica" and "Diagnosis of and screening for hypothyroidism in nonpregnant adults" and "Clinical manifestations of dermatomyositis and polymyositis in adults".)

In addition, we do not perform serologic testing for Lyme disease in asymptomatic patients after an Ixodes spp tick bite. The rationale for this approach is discussed elsewhere. (See "Evaluation of a tick bite for possible Lyme disease" and "Evaluation of a tick bite for possible Lyme disease", section on 'Serologic testing'.)

APPROACH TO DIAGNOSIS — The approach to diagnosis in patients with suspected Lyme disease depends upon the stage of disease, and it is also impacted by having a history of prior infection. (See 'Erythema migrans (single or multiple)' below and 'Early disseminated and late Lyme disease' below and 'Considerations in patients with prior Lyme disease' below.)

The United States Centers for Disease Control and Prevention (CDC) has developed a definition of Lyme disease (table 2) [7]. Although this definition was not intended to be used by clinicians to make a diagnosis of Lyme disease, as it is not comprehensive, the vast majority of cases of Lyme disease in practice do fulfill the case definition.

Erythema migrans (single or multiple) — The diagnosis of erythema migrans (EM) should be made on clinical grounds alone when at least one characteristic EM lesion is present in a patient who lives in or has recently traveled to an endemic area (picture 1 and picture 2) [8]. (See 'Risk of exposure' above and 'Consistent clinical manifestations' above.)

Serologic testing is neither required nor recommended in such patients, particularly if there is a single lesion, since the EM lesions often appear prior to development of a diagnostic, adaptive immune response, and patients who present with EM are often seronegative. In addition, serologic testing after treatment is generally not indicated since patients with EM who are treated early in their course may not develop an antibody response. Thus, a negative test result does not exclude the diagnosis. (See 'Limitations of serologic testing' below.)

However, if the cause of the skin lesion is in doubt, and empiric antimicrobial therapy is not administered, serologic testing is reasonable. Testing can be obtained at the time of presentation, and if negative, a repeat test can be performed two to three weeks later. In this setting, patients should be told to inform their provider if they develop clinical findings that become more suggestive of EM (eg, slowly expanding target-like rash); in that case, empiric treatment would be warranted. (See "Treatment of Lyme disease".)

Another way to diagnose early Lyme disease is by performing polymerase chain reaction (PCR) testing for B. burgdorferi on a skin punch biopsy of at least 2 mm diameter taken from the margin of the skin lesion [9]. However, this approach is not typically used since skin biopsies are not routinely performed in the primary care or urgent care setting, and PCR testing is not widely available for clinical use. (See 'Polymerase chain reaction' below.)

Although patients with EM may be spirochetemic, blood cultures for B. burgdorferi are not commercially available. (See 'Culture' below.)

Early disseminated and late Lyme disease — In patients with suspected early disseminated or late Lyme disease (table 1), serologic testing is usually warranted to support the diagnosis, except in patients with multiple erythema migrans, as described above. (See 'Erythema migrans (single or multiple)' above.)

Testing for antibodies to B. burgdorferi should be seen as an adjunct to the clinical diagnosis since, by itself, serologic testing can neither establish nor exclude the diagnosis of Lyme disease. A positive or negative serologic test result for Lyme disease simply changes the probability that a patient has been infected with B. burgdorferi. (See 'Ability to confirm active disease' below.)

In addition, there are special considerations in patients with a known history of prior Lyme disease, since antibodies (including immunoglobulin [Ig] M) can persist for years. (See 'Considerations in patients with prior Lyme disease' below.)

Serologic testing algorithms — A two-tiered conditional strategy is typically used to support the diagnosis of Lyme disease [10]. This section will review specific testing algorithms. A more detailed discussion of the types of tests is found below. (See 'Serologic tests' below.)

Choosing an algorithm — There are two options for two-tiered serologic testing algorithms. The traditional algorithm includes an initial enzyme immunoassay (EIA) or immunofluorescence assay (IFA) followed by a Western blot. Modified algorithms that include two sequential EIAs have been developed and validated; however, these modified algorithms may not be available in all clinical settings, and experience with them is more limited.

When laboratories offer both testing strategies, the choice between them should take into consideration the relative advantages for the individual patient. Specifically:

●For patients with suspected early disseminated Lyme disease who require testing (eg, those with cardiac or neurologic findings), a modified algorithm that uses two EIAs would be preferred, as this combination offers the greatest sensitivity and specificity in this setting.

●For patients with suspected late disease, the sensitivity and specificity of the traditional algorithm are equivalent to the modified algorithm; however, some experts prefer the traditional algorithm in this setting, since Western blot testing provides additional information that can be beneficial in the evaluation of more complex cases.

Descriptions of the individual algorithms are discussed below. (See 'Traditional two-tiered testing' below and 'Modified two-tiered algorithms' below.)

Traditional two-tiered testing — The traditional two-tiered testing algorithm starts with a sensitive enzyme immunoassay, such as a whole cell-based enzyme-linked immunosorbent assay (ELISA). In some cases, a different type of enzyme immunoassay (an IFA) is used [10]. If the immunoassay result is negative, the diagnosis of either early disseminated or late Lyme disease is unlikely and no further testing is performed. (See 'Enzyme-linked immunosorbent assays' below.)

If the immunoassay is positive, a more specific serologic test, the Western blot, is performed, typically with separate IgM and IgG blots. The IgM Western blot is considered positive if two of three particular bands are detected, and the IgG Western blot is considered positive if 5 of 10 particular bands are detected (table 3). (See 'Western blot' below.)

However, Western blot testing can be difficult to perform and interpret, and the IgM Western blot is prone to false-positive results. Furthermore, serologic testing alone can neither establish nor exclude the diagnosis of Lyme disease. Thus, clinicians must take into account the stage of disease when interpreting the Western blot results:

●Early disseminated Lyme disease – By the time patients with early disseminated Lyme disease present with clinical findings (multiple EM, lymphocytic meningitis, facial palsy, radiculoneuropathy, or carditis with heart block), they usually have detectable IgM and IgG antibodies to B. burgdorferi, and both the IgM and IgG Western blots are positive.

In a minority of patients, the IgG Western blot may be negative at the time of presentation (table 3). Although an isolated positive IgM Western blot (with a negative IgG Western blot) could be consistent with early disseminated Lyme disease in patients with symptoms of less than six to eight weeks' duration, that serologic pattern should not be used to support the diagnosis of Lyme disease in untreated patients with symptoms of early infection for longer than eight weeks [11]. If the patient has an isolated IgM Western blot after that time, the IgM test most likely is a false-positive result or, possibly, evidence of past treated infection. In the United States, the CDC recommends that a positive IgM response alone should not support the diagnosis of Lyme disease if symptoms have been present for more than four weeks, since the majority of patients will have a positive IgG Western blot after that time. However, in a small percentage of patients, it may take six to eight weeks for an IgG Western blot response to develop with reactivity with five or more spirochetal proteins (table 3) [12].

On rare occasions, both IgM and IgG Western blots are negative in early disseminated Lyme disease. This is most likely to occur in those who present with multiple EM (picture 3), since this can develop very early after infection. When this occurs, some patients may be treated empirically (eg, those with multiple EM from an endemic area). For those who are not treated empirically, repeat serologic testing about three weeks later may be helpful, since the Western blot should be positive by that time in most patients with untreated infection. (See 'Erythema migrans (single or multiple)' above.)

●Late Lyme disease – In patients with late disease (eg, Lyme arthritis), the IgG Western blot should be positive (table 3). In such patients, the IgM Western blot may or may not be positive and should not be used for diagnosis.

Modified two-tiered algorithms — Modified two-tiered algorithms that use novel ELISAs and/or different second step tests to validate a positive ELISA have been developed to improve the sensitivity and specificity of serologic testing for early Lyme disease [12].

●Algorithms using two enzyme immunoassays – Some modified algorithms use two enzyme immunoassays with different targets. Testing is consistent with Lyme disease if a patient has positive results on both tests. If the result of the second test is negative, Lyme disease is unlikely. In some cases, the initial test may be indeterminate; in this setting, a second test is still performed and, if positive, would suggest Lyme disease. However, test results must be correlated with the clinical scenario (eg, early disease versus past treated infection).

•VlsE1/pepC10 IgG/IgM test followed by second ELISA – A modified testing algorithm using an ELISA that detects IgG and/or IgM antibodies to the variable major protein-like sequence 1 (VlsE1) and pepC10 antigens from B. burgdorferi followed by a whole-cell ELISA (IgM, IgG or IgG/IgM) has been approved for use in the United States [13-15]. Data supporting the use of this approach include findings from a retrospective analysis of 356 samples that reported a sensitivity of 66 to 78 percent for early disease and 100 percent for late disease when the VlsE1/pepC10 IgG/IgM test was followed by second IgG/IgM ELISA [14].

•VlsE C6 ELISA to replace Western blot – Another strategy uses a whole cell-based ELISA followed by a C6 ELISA. (See 'Types of tests' below.)

This algorithm has been evaluated in several studies [16-18]. In one, serum samples were tested with the traditional ELISA followed by the VlsE C6 ELISA, the VlsE C6 ELISA alone, or the traditional ELISA followed by Western blot [17]. In patients with early Lyme disease, ELISA followed by VlsE C6 ELISA had a similar sensitivity as VlsE C6 ELISA alone (61 versus 64 percent), and both strategies had a higher sensitivity than ELISA followed by Western blot (48 percent). In patients with late Lyme disease, all three approaches had 100 percent sensitivity. The specificity of ELISA followed by VlsE C6 ELISA was equivalent to that of ELISA followed by Western blot (99.5 percent for each), and both two-tiered strategies were more specific than VlsE C6 ELISA alone (98.4 percent). The positive predictive value of the ELISA followed by the VlsE C6 ELISA was 70 percent, compared with 66 percent for traditional two-tiered testing and 43 percent for the VlsE C6 ELISA alone.

One limitation of this algorithm compared with traditional testing is that both the whole cell-based ELISA and the C6 ELISA can be positive for patients infected with B. miyamotoi. This is in contrast to Western blot testing for B. burgdorferi, which is negative for patients with B. miyamotoi. Thus, this algorithm can result in a decrease in specificity if used in an area where the incidence of B. miyamotoi is high. Despite this limitation, this algorithm still has utility in the clinical setting, as both diseases are treated with the same antibiotics. A more detailed discussion of B. miyamotoi is found elsewhere. (See "Borrelia miyamotoi infection".)

Need for additional testing — In patients who have arthritis or certain neurologic findings (eg, aseptic meningitis, radiculoneuritis) suggestive of Lyme disease, serum antibody testing alone is often sufficient for diagnosis if the patient has risk factors for Lyme disease [9]. (See 'Risk of exposure' above and 'Consistent clinical manifestations' above and 'Early disseminated and late Lyme disease' above.)

However, if a more definitive diagnosis is required, testing of synovial fluid or cerebrospinal fluid (CSF) may be needed:

●For those with suspected Lyme arthritis, PCR testing of synovial fluid is the preferred test.

●For patients with neurologic symptoms, simultaneous samples of CSF and serum can be obtained to determine the CSF/serum antibody index. However, this test is usually positive only in patients with a CSF pleocytosis.

The decision to perform this testing must be determined on case-by-case basis. More detailed discussions regarding the diagnosis of Lyme disease in patients with musculoskeletal and neurologic manifestations of Lyme disease are presented elsewhere. (See "Musculoskeletal manifestations of Lyme disease", section on 'Diagnosis of Lyme arthritis' and "Nervous system Lyme disease".)

Considerations in patients with prior Lyme disease — Patients who were treated for prior Lyme disease with an appropriate treatment regimen can present with several different clinical scenarios. These include:

●Reinfection

●Post-treatment Lyme disease syndrome

●Post-infectious Lyme arthritis

Serologic testing is generally not helpful in patients with prior Lyme disease, since antibodies (including either IgG, IgM, or both) can persist for years. (See 'Persistence of antibodies' below.)

The approach to diagnosis for patients with these specific manifestations is discussed below. (See 'Reinfection' below and 'Post-treatment Lyme disease syndrome' below and 'Post-infectious arthritis' below.)

Persistence of antibodies — Most patients treated for early Lyme disease seroconvert over time, even after successful treatment. Depending upon which testing algorithm is used, this means that both the ELISA and Western blot are positive or that both EIA tests in the modified two-test approach are positive. However, for some patients, administering antibiotics in early Lyme disease can prevent seroconversion [19]. In such patients, the disease generally resolves as well. By contrast, patients with late Lyme disease are already seropositive before, and remain so after, treatment.

For those who do seroconvert, antibody titers to B. burgdorferi, as determined by ELISA, decline gradually after successful antibiotic treatment of Lyme disease [20,21], although positive responses often persist for years. Accurate measurement of the decline is best performed with serial serum samples tested together on the same plate, but this is usually difficult to perform as most commercial laboratories do not save previous samples. Antibodies measured using the VlsE C6 peptide ELISA test may decline more rapidly than other tests after treatment for infection, but this response may also persist for years after treatment [22,23].

The Western blot, a non-quantitative test, does not change much (or changes very slowly) after successful antibiotic treatment. In one study of 40 patients who had early Lyme disease 10 to 20 years earlier, 10 percent still had IgM responses to B. burgdorferi and 25 percent still had IgG reactivity by ELISA and Western blot [20]. Among 39 patients who had had Lyme arthritis 10 to 20 years earlier, 15 percent still had IgM responses and 62 percent still had IgG responses by two-tiered testing.

At this time, there is no information about how long test results may remain positive using the modified two-tiered testing approach (see 'Modified two-tiered algorithms' above). With this method, only a positive or negative result is reported, but presumably positive results may persist in some patients for years after successful treatment. Thus, this test is most useful in patients with primary infection and may not help when diagnosing reinfection.

Reinfection — Since many patients live in endemic areas, reinfection after antibiotic treatment of EM or another manifestation of early disease is not uncommon. The risk of reinfection seems to depend upon the patient's initial clinical presentation, although data are limited.

●If reinfection occurs, it typically occurs in those who previously had EM; reinfection is usually diagnosed when it presents as another EM lesion during another year [24].

●Patients who have neurologic or cardiac involvement may still become reinfected [25], but it appears they are less likely to do so compared with those who had EM. This may be due to a more developed antibody response in those with early disseminated versus early localized disease.

●Reinfection has not been reported after Lyme arthritis, which is usually associated with high antibody titers that typically persist for years and seem to be protective.

The diagnosis is typically based upon clinical findings (eg, rash consistent with EM in a patient who lives in an endemic area). Serologic testing is generally not helpful in patients with prior Lyme disease, since most patients seroconvert even after treatment with antibiotics, and antibodies (including IgM) can persist for years. New IgG blots may become positive with reinfection, although interpretation of these findings is challenging. (See 'Persistence of antibodies' above.)

Post-treatment Lyme disease syndrome — Post-treatment Lyme disease syndrome is the development of headache, diffuse pain, and/or perceived neurocognitive difficulties or fatigue following treatment of Lyme disease. Post-treatment Lyme disease syndrome can develop after any stage of disease (eg, EM, meningitis, or arthritis) and may persist for months or years. (See "Clinical manifestations of Lyme disease in adults", section on 'Post-treatment Lyme disease syndrome and chronic Lyme disease'.)

The diagnosis is based upon clinical criteria that include a prior history of Lyme disease treated with an accepted regimen and resolution or stabilization of the objective manifestations of Lyme disease. In addition, the onset of subjective symptoms (eg, fatigue, widespread musculoskeletal pain, complaints of cognitive difficulties) must have occurred within six months of the diagnosis of Lyme disease and persist (continuously or relapsing) for at least six months after completion of antimicrobial therapy (table 4).

There is no role for repeated serologic testing in the evaluation of patients with post-Lyme disease syndrome, since antibodies can persist for prolonged periods of time, despite successful treatment, and do not indicate active disease. (See 'Persistence of antibodies' above.)

Post-infectious arthritis — Patients with Lyme arthritis can develop a post-infectious arthritis despite appropriate antibiotic treatment. Such patients do not respond to further antibiotic therapy but do respond to disease-modifying antirheumatic drugs (DMARDs) [26]. Patients with suspected post-infectious arthritis should be evaluated in conjunction with a Lyme disease specialist.

Post-infectious Lyme arthritis is the development of a marked proliferative inflammatory synovitis resulting from an excessive, dysregulated proinflammatory immune response that develops during the infection and persists in the post-infectious period [27]. Typically, during the infectious period, the knee joint is more swollen with fluid and the percentage of polymorphonuclear leukocytes is very high (approximately 90 percent). By contrast, in patients with post-infectious Lyme arthritis, there is greater proliferation of synovial tissue, the joint is not as swollen with fluid, and the percentage of polymorphonuclear leukocytes declines. (See "Musculoskeletal manifestations of Lyme disease".)

The diagnosis of post-antibiotic Lyme arthritis is based upon the clinical findings of an inflammatory, proliferative synovitis in a patient who has received at least two courses of appropriate antibiotic therapy, including at least one course of intravenous therapy before making this diagnosis [9]. (See "Treatment of Lyme disease", section on 'Arthritis'.)

In patients with a known history of Lyme arthritis, further serologic and PCR testing are usually not helpful in diagnosing post-infectious arthritis. Lyme antibodies are expected to be positive in patients with Lyme arthritis, even after appropriate therapy [28]. In addition, B. burgdorferi DNA may persist in synovial fluid for weeks to months after treatment, thus the presence of DNA during that period does not rule out a post-infectious process [29]. Although Lyme testing may not be useful, ultrasound or magnetic resonance imaging (MRI) demonstrating synovitis and inflammation can support the diagnosis. (See "Imaging techniques for evaluation of the painful joint", section on 'Ultrasound' and "Imaging techniques for evaluation of the painful joint", section on 'Magnetic resonance imaging'.)

SEROLOGIC TESTS

Types of tests

Enzyme-linked immunosorbent assays — There are several different types of enzyme-linked immunosorbent assays (ELISAs). The most commonly used assay is a whole cell-based ELISA derived from the lysates of the organism. Newer variations of ELISA tests for anti-borrelial antibodies use purified antigens rather than whole-cell bacterial lysates and are more specific for B. burgdorferi [30]. One test that has been studied extensively is the VlsE C6 peptide ELISA.

●Whole cell-based enzyme linked immunosorbent assay (ELISA) – Whole cell-based ELISA tests are available for IgM (early), IgG (late), and combined IgM and IgG antibody detection. The combined preparation is typically used.

Since the antigen used in the whole cell-based ELISA test is derived from the lysates of the organism, cross-reaction with antibodies to other organisms due to similarity to B. burgdorferi antigens can occur. At least 5 percent of the normal population will test positive for antibodies to B. burgdorferi by whole cell-based ELISA due to cross-reacting antibodies elicited either by other infections or by the immune response to normal flora [31,32].

False-positive IgM testing is much more common than false-positive IgG testing, although both can occur.

•False-positive ELISA testing can be seen with patients with other borrelial diseases (eg, relapsing fever, B. miyamotoi disease), other spirochetal diseases (eg, syphilis, leptospirosis, pinta, yaws, gingivitis), other bacterial infections (eg, infective endocarditis), viral illnesses, and autoimmune diseases (eg, systemic lupus erythematosus and rheumatoid arthritis) [31-34].

•In addition, cross-reacting antibodies may be produced due to nonspecific activation of immunoglobulin production (eg, polyclonal B cell activation). Epstein-Barr virus and malaria can cause such activation, resulting in a positive ELISA for Lyme disease.

•Vaccination with the previously licensed Lyme disease vaccine also caused a positive result on whole cell-based ELISAs [35]. (See 'Interpretation of results in vaccine recipients' below.)

●VlsE C6 ELISA – The VlsE C6 peptide ELISA (or C6 test) measures IgG to the variable major protein-like sequence-expressed (VlsE) sixth invariant region (C6) peptide. IgG antibodies to the C6 invariant region develop early (within the first week), giving it comparable sensitivity to IgM ELISA tests with improved specificity [36]. However, there may be cross-reactivity in the setting of B. miyamotoi infection. (See "Borrelia miyamotoi infection", section on 'Diagnosis'.)

In some laboratories, the VlsE C6 peptide ELISA has been used as a standalone test. The benefits of the VlsE C6 peptide ELISA compared with traditional two-tiered serologic testing are the relative ease of testing and of interpreting the result, the earlier development of the IgG response, and the ability to standardize results between laboratories [36,37]. However, traditional two-tiered algorithms using a Western blot have slightly greater specificity compared with the VlsE C6 peptide ELISA; in addition, with the Western blot, the expansion of the antibody response to include greater numbers of positive bands gives information about the duration of infection. (See 'Western blot' below.)

Several studies have evaluated the use of the VlsE C6 peptide ELISA. In a prospective study of 134 patients at various stages of Lyme disease, 89 patients with other illnesses, and 136 healthy controls, only 20 percent of patients with erythema migrans (EM; the initial skin lesion of Lyme disease) had positive results using either the VlsE C6 peptide ELISA or a traditional two-tiered algorithm, but both approaches had 100 percent sensitivity in patients with neurologic, cardiac, or joint manifestations of Lyme disease [36]. Specificity was 96 percent with the VlsE C6 peptide assay and 99 percent with two-tiered serologic testing.

The infecting strain of B. burgdorferi can affect the sensitivity of the different serologic tests for B. burgdorferi. In a study that assessed the sensitivity of a commercial VlsE C6 peptide ELISA in patients with EM, the results were stratified based upon which of the three B. burgdorferi subtypes (defined by ribosomal spacer typing) caused infection [38]. The VlsE C6 peptide ELISA had higher sensitivity compared with traditional two-tiered serologic testing in patients infected with two of the three B. burgdorferi subtypes.

The VlsE C6 ELISA has also been evaluated as part of a modified two-tiered algorithm, as described above. (See 'Modified two-tiered algorithms' above.)

●VlsE/PepC10 – An ELISA that detects IgG and/or IgM antibodies to the variable major protein-like sequence 1 (VlsE1) and pepC10 antigens from B. burgdorferi has been developed and is used as part of a modified two-tiered algorithm, as discussed above. (See 'Modified two-tiered algorithms' above.)

In one study that evaluated this assay as confirmatory test in 242 patients with Lyme disease at varying stages and 794 controls, this assay was equally specific (96 percent) to Western blot testing and more sensitive overall (75 versus 63 percent, respectively) [39].

Western blot — The Western blot (or immunoblot) allows detection of antibodies to individual components of the organism and thus provides more information regarding which antigens of B. burgdorferi are reacting with serum antibodies than a whole cell-based ELISA. Separate Western blots are performed to detect either IgM or IgG antibodies. Information learned from studies correlating cases of clinically diagnosed Lyme disease with patterns of reactivity on Western blot have enabled the United States Centers for Disease Control and Prevention (CDC) to recommend evidence-based criteria for Western blot interpretation (table 3) [40]. These criteria should be used in interpreting Western blot results.

Similar to the ELISA tests, false-positive IgM tests are seen on Western blot [41]. False-positive IgM results can be a result of over-reading of weak bands and the greater cross-reactivity and binding strength of IgM antibodies. In general, an isolated IgM (ie, IgG is negative) Western blot is not sufficient for the diagnosis of Lyme disease in a patient who has had symptoms for six to eight weeks or longer [42]. False-positive IgG Western blots are less common than false-positive IgM tests, but they also occur, particularly in patients with other specific medical conditions. The positive predictive value of all Lyme serologic testing is dependent upon the prior probability of disease and, as such, testing should not be used in patients with low prior probability of Lyme disease. (See 'Clinical suspicion for Lyme disease' above.)

A more detailed discussion of how to interpret the results of Western blot testing is found above. (See 'Early disseminated and late Lyme disease' above.)

Limitations of serologic testing

Ability to confirm active disease — Seropositivity indicates the presence of antibodies to the spirochete that causes Lyme disease. Seropositivity alone, however, is insufficient to make a diagnosis of active Lyme disease, since antibodies may persist for years after Lyme disease has been treated and cured. Thus, the diagnosis of extracutaneous Lyme disease depends upon having the appropriate clinical features in conjunction with seropositivity. (See 'Clinical suspicion for Lyme disease' above and 'Early disseminated and late Lyme disease' above.)

Lack of sensitivity in early disease — In Lyme disease, IgM antibodies to B. burgdorferi typically appear within one to two weeks, and IgG antibodies usually appear within two to six weeks following the onset of EM [36,40]. However, only 20 to 40 percent of patients with early localized Lyme disease (EM) are seropositive at the time of presentation using the two-tiered serologic method [36,43,44]. Thus, patients with skin lesions that are typical of EM do not require confirmatory serologic testing. (See 'Erythema migrans (single or multiple)' above.)

In comparison with the findings in early localized disease, most patients with early disseminated extracutaneous Lyme disease (eg, early neuroborreliosis or Lyme carditis) are seropositive [36]. However, on occasion, those with suspected early disseminated Lyme disease may be seronegative. If empiric therapy is not initiated, repeat serologic testing about three weeks later may be helpful, since the antibody response should be positive in most patients with untreated infection by that time. A more detailed discussion of how to interpret the results of serologic testing is found above. (See 'Serologic testing algorithms' above.)

Interlaboratory variation — There may be variability in the results of serologic tests performed in different laboratories [45-47]. Variability is typically greatest during the early stages of disease and particularly for IgM tests (both ELISA and Western blot). Inter- and intralaboratory concordance of serologic results improves during late-stage disease and approaches 100 percent for IgG testing in patients with disease for greater than three months when standard CDC criteria are used (table 2) [48].

These findings were illustrated in a study that compared the variability of different serologic tests at university-based, commercial-based, and specialty-based laboratories [47]. Sera from 37 patients with post-treatment Lyme disease syndrome, as well as 40 healthy controls without prior Lyme disease, were tested at four laboratories. There was no significant difference in the percent of positive ELISA and IgG western blots when CDC criteria were used. However, in one laboratory, the specificity declined significantly when their own in-house criteria were used to diagnose Lyme disease. Thus, results from laboratories that have established their own criteria for interpreting the Western blot should be viewed with skepticism. (See 'Western blot' above.)

Sensitivity for different B. burgdorferi species and strains — ELISA and Western blot tests available in the United States are optimized for detection of antibodies to B. burgdorferi sensu stricto. Sensitivity of these tests for antibodies to strains of B. burgdorferi sensu lato present in Europe and Asia (such as B. garinii and B. afzelii) is often lower. The VlsE C6 epitope is relatively conserved between European and United States species, and thus VlsE C6 ELISAs may be a more appropriate test for detecting antibodies to all species of B. burgdorferi. Limited data also suggest that antibodies to a newly identified strain of Borrelia causing Lyme disease in the Midwestern United States, B. mayonii, may not be as reliably detected by the Western blot [49]. More detailed discussions of the VlsE ELISA are found above. (See 'Modified two-tiered algorithms' above and 'Enzyme-linked immunosorbent assays' above.)

Even among strains of B. burgdorferi sensu stricto in the United States, the sensitivity of the Western blot for detection may vary significantly. Borrelial strains may be classified by rRNA intergenic spacer type (RST). In culture-proven cases of infection, two-tiered testing was less sensitive than VlsE C6 ELISA testing in detecting antibodies to RST types 2 (37 percent positive versus 66.7 percent positive, respectively) and 3 (25 percent positive versus 75 percent positive) [38]. Detection of RST type 1 strains was equivalent. The lower sensitivity of the two-tiered approach was mainly due to decreased sensitivity of the Western blot.

Interpretation of results in vaccine recipients — LYMErix, the original Lyme disease vaccine interfered with the serologic diagnosis of Lyme disease by causing a positive ELISA [35,50,51]. However, this vaccine was withdrawn from the market in 2003 and is unlikely to explain a positive serologic response in patients being evaluated for Lyme disease. A new, second-generation OspA vaccine is currently undergoing clinical trials and may become available in the future. Although the impact on serologic testing has yet to be determined, the Western blot or the VlsE/PepC10 tests should not be impacted.

TESTS OF LIMITED OR UNCERTAIN VALUE — The United States Centers for Disease Control and Prevention (CDC) and the US Food and Drug Administration (FDA) have cautioned clinicians that some commercial laboratories are performing assays for Lyme disease, the accuracy and clinical usefulness of which have not been adequately established [52]. These tests include polymerase chain reaction (PCR) on inappropriate specimens such as blood and urine, urine antigen tests [53], immunofluorescent staining for cell wall-deficient forms of B. burgdorferi, and lymphocyte transformation tests.

Polymerase chain reaction — Polymerase chain reaction (PCR) has been used clinically to identify the presence of B. burgdorferi DNA in cerebrospinal fluid (CSF) or synovial fluid specimens [54-59]. PCR test results for B. burgdorferi are often positive in synovial fluid prior to antibiotic therapy [56]. There are important limitations to the use of PCR [60]:

●The accuracy of PCR is highly dependent upon the care used in sample collection and storage and the technique used in the assay.

●False-positive results are common. In particular, a positive CSF PCR is likely to be a false-positive result in patients who are seronegative for Lyme disease.

●CSF PCR has low sensitivity [61]. Thus, a negative PCR result does not exclude either neurologic Lyme disease or Lyme arthritis.

●PCR testing of specimens of CSF or synovial fluid for B. burgdorferi DNA in a reliable laboratory can add confirmatory information in seropositive patients. However, a positive PCR result with negative serum antibody results is likely to be a false-positive result, and a positive PCR result does not prove that the patient has active infection, since spirochetal DNA may persist long after spirochetal killing has occurred. PCR identifies DNA in the specimen; it does not differentiate between infection with a live organism and the presence of remnant DNA from a prior, but cured, infection. (See "Musculoskeletal manifestations of Lyme disease", section on 'Polymerase chain reaction testing' and "Nervous system Lyme disease", section on 'Serologic testing'.)

●False-negative PCR results can occur due to the presence of inhibitors of the DNA polymerase, such as hemoglobin or hyaluronic acid.

●PCR of the urine is of no proven value in the diagnosis or management of B. burgdorferi infection in humans.

Culture — Culture is not available in most clinical laboratories. However, in research studies, B. burgdorferi has been isolated from skin biopsy specimens, blood, and CSF [62,63]. Two commonly used mediums are the modified Kelly-Pettenkofer and Barbour-Stoenner-Kelly II.

In a study of 65 adults with erythema migrans (EM), the yield of blood cultures increased from 46 percent using traditional blood culture methods to 71 percent when quantitative PCR was performed on blood culture samples following incubation [64]. Of the 46 patients with a positive PCR assay, 42 (91 percent) had a positive PCR result within seven days after the cultures began incubating. An important limitation of this study is that negative control samples were not included, so it is not possible to determine whether the additional positive samples were true positives.

In a cross-sectional study in which blood cultures for B. burgdorferi were obtained in 213 untreated adults with EM, spirochetemia was present in 44 percent during the early phases (the first several weeks) [65]. Spirochetemic patients were significantly more likely to be symptomatic (89 versus 74 percent) and have multiple lesions (42 versus 15 percent) than those without spirochetemia, 8 of 35 asymptomatic patients (23 percent) with a single skin lesion had a positive blood culture. The risk for spirochetemia was present the day the patient noticed the lesion and continued for more than two weeks.

In another study, blood cultures were positive in 5 of 26 untreated patients (19 percent) with early disseminated Lyme disease with neurologic, cardiac, or musculoskeletal involvement [66]. Patients with a positive blood culture had a shorter duration of symptoms compared with those with a negative blood culture (mean 14 versus 125 days, respectively). These findings suggest that spirochetemia occurs early in the course of infection. Positive blood cultures were associated with the presence of a concomitant EM lesion; positive blood cultures occurred in four of nine patients (44 percent) with EM versus 1 of 17 patients (6 percent) without EM.

Urinary antigen testing — In general, most urinary antigen tests are of no proven value in the diagnosis or management of Lyme disease and are not recommended [53].

Testing ticks — Although testing of individual Ixodes ticks recovered from patients for B. burgdorferi by PCR is available commercially, most experts do not recommend testing of ticks, since the results do not affect clinical management [9]. If an Ixodes tick has been attached for less than 36 hours, it is unlikely to have transmitted disease, and no prophylaxis should be offered, even if PCR is positive. By contrast, if the tick has been attached for ≥36 hours, then prophylaxis is recommended, and a delay in initiation of prophylaxis while awaiting results of testing could be detrimental. (See "Evaluation of a tick bite for possible Lyme disease", section on 'Antimicrobial prophylaxis'.)

Other tests

●Immune complex detection – Standard serologic assays for antibodies to B. burgdorferi are commonly negative in early localized disease. One potential explanation for this finding is that anti-borrelial circulating antibodies are bound to their antigenic target [67,68]. The diagnostic value of these assays to detect antibody present in immune complexes remains unproven, and such tests have limited availability because they are labor intensive and not standardized or licensed.

●T cell proliferative response – T cell proliferative responses of human mononuclear cells to borrelial antigens have been detected in blood, synovial, and CSF samples from patients with Lyme disease [69-73]. Limitations of T cell testing include difficulties in performance and interpretation and the large number of false-positive results [72]. Thus, such testing is not recommended.

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Tick-borne infections (Lyme disease, ehrlichiosis, anaplasmosis, babesiosis, and Rocky Mountain spotted fever)".)

INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5^th to 6^th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10^th to 12^th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.

Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)

●Basics topics (see "Patient education: Lyme disease (The Basics)")

●Beyond the Basics topics (see "Patient education: Lyme disease symptoms and diagnosis (Beyond the Basics)" and "Patient education: Lyme disease prevention (Beyond the Basics)" and "Patient education: Lyme disease treatment (Beyond the Basics)")

SUMMARY AND RECOMMENDATIONS

●Approach to diagnosis – The diagnosis of Lyme disease should be suspected in patients who are at risk of exposure to ticks that carry Lyme disease and have clinical manifestations that are consistent with Lyme disease. Once Lyme disease is suspected, the approach to diagnosis depends upon the stage of infection. (See 'Clinical suspicion for Lyme disease' above.)

•Patients with erythema migrans – The diagnosis of erythema migrans (EM) should be made on clinical grounds alone when one or more characteristic EM lesion is present in a patient who lives in or has recently traveled to an endemic area (picture 1 and picture 2). (See 'Erythema migrans (single or multiple)' above.)

•Patients with early disseminated or late infection – In patients with suspected early disseminated (other than multiple EM) or late Lyme disease, serologic testing is warranted to support the diagnosis. However, testing for antibodies to B. burgdorferi should be seen as an adjunct to the clinical diagnosis, since, by itself, serologic testing can neither establish nor exclude the diagnosis of Lyme disease. (See 'Early disseminated and late Lyme disease' above.)

•No role for screening of asymptomatic patients or patients with nonspecific symptoms – Serologic testing for Lyme disease should not be performed to screen asymptomatic patients who live in endemic areas (even if they had a known Ixodes spp tick bite) or for evaluation of patients who present only with nonspecific symptoms (eg, sub-acute or long-standing myalgias/arthralgias or fatigue). (See 'When Lyme testing is not warranted' above.)

●Specific diagnostic tests

•Serologic tests – When testing is indicated, a two-tiered conditional strategy should be used to support the diagnosis of Lyme disease. Two-tiered testing can be performed using an initial enzyme-linked immunosorbent immunoassay (ELISA) or immunofluorescence assay, followed by either a Western blot (traditional two-tiered testing) or by a second ELISA that uses a different target (modified two-tiered testing). (See 'Serologic testing algorithms' above and 'Serologic tests' above.)

•Synovial and cerebrospinal tests – Testing of cerebrospinal or synovial fluid may be needed to support a diagnosis of Lyme disease in patients with evidence of aseptic meningitis, radiculoneuritis, or Lyme arthritis. This is discussed separately. (See "Nervous system Lyme disease", section on 'Diagnosis and differential diagnosis' and "Musculoskeletal manifestations of Lyme disease", section on 'Diagnosis of Lyme arthritis'.)

Use of laboratory tests other than ELISA, Western blot, and in limited cases, polymerase chain reaction, should not be used to support a diagnosis of Lyme disease. (See 'Tests of limited or uncertain value' above.)

●Limited role of testing in patients with prior Lyme disease – Some patients with prior Lyme disease may present with new symptoms after resolution of a previous infection or may have persistent symptoms that fail to improve despite appropriate treatment. Depending upon the clinical presentation, such patients may have reinfection with a new organism, post-treatment Lyme disease syndrome, or post-infectious arthritis. These diagnoses are often based upon epidemiologic and clinical findings. Serologic testing is rarely helpful in patients with prior Lyme disease since antibodies (including immunoglobulin [Ig]M) can persist for years. (See 'Considerations in patients with prior Lyme disease' above.)