Lyme Disease Workup

The most widely used tests for Lyme disease are antibody detection tests, which can only ascertain that a patient has been exposed to B burgdorferi but cannot be used to confirm infection. In the presence of typical clinical manifestations and laboratory results suggestive of current activity (elevated synovial and spinal fluid cell counts), they can be used to confirm the clinical diagnosis. However, physicians must understand the limitations of serologic tests and that routine use of sequential serologic testing for use in individual patients with early Lyme disease should be discouraged.

Acute and convalescent-phase serologic testing has no role in Lyme disease. Erythema migrans should be treated in endemic areas without blood tests. Depending on the clinical situation, treatment should either be started once appropriate laboratory testing has begun drawn or after the results have confirmed the diagnosis. Because titers may remain elevated for extended periods (as can the positivity of Western blots), convalescent testing in not helpful.

Biopsy of dermatologic lesions suggestive of borrelial lymphocytoma or acrodermatitis chronica atrophicans in patients without a clear history of other symptoms suggestive of Lyme disease may be helpful. Biopsy of other skin lesions should be restricted to research settings.

Results of some laboratory studies may suggest some of the other co-infecting tick-borne pathogens such as ehrlichial or babesial species. Most patients with ehrlichiosis have elevated levels of hepatic transaminases, leukopenia, and/or thrombocytopenia. In addition, some patients have morulae (intracytoplasmic inclusions) in white blood cells as demonstrated on peripheral blood smears. Patients with babesiosis often are anemic (hemolytic type) and may have thrombocytopenia. Blood smears reveal the malarialike intraerythrocytic parasite in this disease as well, as shown below.

Most, but not all, patients with borrelial lymphocytoma are seropositive for antiborrelial antibodies. This is true for all early disseminated manifestations of Lyme disease. In addition, essentially all patients with acrodermatitis chronica atrophicans are seropositive for antiborrelial antibodies. Seriously question the diagnosis in seronegative patients.

In patients in whom significant ambiguity exists concerning the diagnosis after the history and physical examination are performed, several strategies can be used and are discussed in this section.

Empiric therapy approach

Erythema migrans is the only manifestation of Lyme disease in the United States for which clinical diagnosis should be made in the absence of laboratory confirmation. A patient with a significantly characteristic symptom with the appropriate history of possible exposure should be started on antibiotics after appropriate laboratory studies have been drawn.

According to US Centers for Disease Control and Prevention (CDC) surveillance criteria, patients presenting with a clinical picture compatible with early LD (ie, erythema migrans, constitutional flulike symptoms) and a history of exposure to an area in which tick exposure is likely do not require laboratory confirmation of the disease before receiving treatment. Thus, the CDC accepts physician-diagnosed erythema migrans of greater than 5 cm in size in its case definition (in clinical practice, erythema migrans can be smaller than this size. The 5-cm limit was designed for surveillance purposes.), and no tests, including serologic tests for Lyme disease, are indicated in these patients.

Complete blood cell (CBC) counts and erythrocyte sedimentation rates (ESRs) are often normal. Antibody titers B burgdorferi may be falsely negative at this early stage of disease and should not affect the decision to treat.

Empiric antibiotic therapy is reasonable if erythema migrans is moderately probable. Therapy using amoxicillin or doxycycline is usually is safe and inexpensive and can be initiated immediately. If the lesion is erythema migrans, prompt improvement (within a few days) and resolution is expected.

Observation

Observing the spontaneous evolution of the rash over several days is safe and universally available. In most patients with erythema migrans, some expansion of the rash is expected over 2-3 days without antibiotics. This is a reasonable alternative to immediate empiric therapy.

CDC 2-step approach

The CDC currently recommends a 2-step testing procedure consisting of a screening enzyme-linked immunoassay (ELISA) or immunofluorescent assay followed by a confirmatory Western immunoblot test on any samples with positive or equivocal results on ELISA. Thus, the recommendation from the CDC is as follows:

• The first step in patients with symptoms consistent with Lyme disease is to obtain an antibody titer. This can be either a total Lyme titer or separate immunoglobulin G (IgG) and immunoglobulin M (IgM) titers.
• The second step is to confirm positive titers with a Western blot. For IgM blots, if any 2 of the following 3 bands are positive, the test is positive: 23 kd, 39 kd, and 41 kd. For IgG blots, any 5 of the following bands are considered a positive test result: 18, 21, 28, 30, 39, 41, 45, 58, 66, and 93 kd.

In the absence of treatment, patients continue to produce IgM antibodies long after the initial infection. Thus, patients may have both IgM and IgG antibodies concurrently. If patients have IgM antibodies only on Western blot more than 6-8 weeks after putative exposure, this probably represents a false-positive result. Although some authors recommend other bands and bypassing antibody titers, no other testing recommendations are available from other national organizations.

In patients who have not been in endemic areas, the false-positive and false-negative rates of these tests reduce the likelihood that the predictive values of the results would be helpful.

Antibody testing using this 2-step process in patients with erythema migrans is not indicated, because the rash may develop before the antibodies.

Western blot testing without a titer is never recommended and should only be performed as a follow up of a recent positive titer without a Western blot. One should always order a “Lyme titer with reflex testing”. This ensures that a contemporaneous Western blot is always performed following a positive titer.

Newer serologic testing with the C6-peptide and VslE are promising new tests that may be more sensitive in patients with erythema migrans.^[23] However, because the treatment recommendation regarding erythema migrans is to treat without obtaining laboratory tests,^[24] there is no clear reason to perform this assay in clinical practice. Importantly, remember that whichever serologic test is performed, the principle of timing (see below) is still operant.^{[23, 25]} The C6 peptide test may be effective is differentiating STARI from Lyme disease.

Timing is important, as many patients at the erythema migrans stage are seronegative. In addition, numerous causes of false-positive ELISA test results exist, such as various infectious and immunologic diseases. Patients with past Lyme disease may be persistently seropositive. All previously vaccinated patients have positive ELISA test results.

In the United States, patients with extracutaneous involvement in the absence of treatment almost universally have positive titers.^[23] In Europe, negative serum titers have been reported in patients with neurologic Lyme disease that was confirmed by intrathecal antibody production.

Histology, culture, and PCR

Synovial fluid is usually inflammatory, with cell counts ranging from 500-98,000/µL reported. In adult patients, the fluid should also be examined for crystals to rule out gout and pseudogout.

Polymerase chain reaction (PCR) is growing in uses and availability. Nevertheless, PCR remains a research technique, in part because laboratories performing PCR tests must be meticulous in technique to minimize the likelihood of false-positive results. Most importantly, it is not readily available to most clinicians in routine practice. In addition, no large clinical series have been reported that assess the performance of the test in the nonresearch setting.

Serologic testing typically involves enzyme-linked immunosorbent assay (ELISA) or immunofluorescence assay (IFA). If results from either of these tests are positive or indeterminate, Western blotting should be performed to confirm Lyme disease.

Seroconversion can take as long as 6-8 weeks after a tick bite. The false-negative rate for ELISA is 32% in early disease. A variety of diseases, including Rocky Mountain spotted fever, syphilis, systemic lupus erythematosus, and rheumatoid arthritis, can cause false-positive results. Late disease is tentatively diagnosed when at least 1 objective clinical manifestation of disseminated disease is present and is supported by ELISA or IFA results. These tests have 89% sensitivity but only 72% specificity for detecting Lyme.

A positive result on Western blotting after ELISA or IFA is an indication for treatment in a patient with a consistent clinical picture. Likewise, a negative result is highly suggestive of a false-positive ELISA finding, and therapy is not indicated.

Inadequate antibiotic therapy for early Lyme disease may suppress the antibody response, potentially yielding a false-negative result on ELISA, IFA, or Western blotting. Serum concentrations of immunoglobulin M (IgM) antibodies usually peak 6-8 weeks after infection and disappear within 4-6 months, although levels sometimes remain elevated for several months or years. IgM titers are useful in evaluating early disease and are considered positive if 2 of the 3 most common bands associated with early disease (ie, 23, 39, or 41 kd) are present.

The results of one study noted that differing sensitivity and specificity were found between various assays used to detect anti-Borrelia antibodies in patients suspected of having Lyme disease. False-positive results occurred in 7% of healthy controls in 2 of the 8 ELISA assays tested. This variability makes it very difficult to compare results from different laboratories, both among different patients and in individual patients.^[26]

In patients with a high probability of having early Lyme disease, IgM testing is 96% specific and 93% predictive. Immunoglobulin G (IgG) antibodies are typically detectable within 6-8 weeks after infection, peak within 4-6 months, and remain elevated indefinitely. In late-stage disease (>4-6 wk after infection), IgG results are more useful than IgM results and are considered diagnostic if 5 of 10 IgG bands common in late disease (ie, 18, 21, 28, 30, 39, 41, 45, 58, 66, or 93 kd) are present.

Careful consideration of both IgG and IgM antibodies is essential because the IgG response may be negative in as many as 50% of patients (particularly those with early disease), whereas a persistence of IgM antibodies can lead to false-positive findings in patients infected for more than 1 month who subsequently receive effective treatment. Of note, serologic results can remain positive years after adequate treatment and cannot be used to distinguish active from inactive disease.

Urine antigen testing has not been studied sufficiently. Because it has not been proven reliable or accurate, it should not be used as a diagnostic tool.

Culture is the usual method for confirming most bacterial infections. Although culturing B burgdorferi from skin biopsy specimens is possible, this is not practical in the usual clinical settings. Although a recent article from an endemic area reported positive culture results in 43.7% of patients with Lyme disease, this required culturing specifically for Lyme disease. In addition, all but 2 of the 213 patients met CDC criteria for Lyme disease and warranted treatment, regardless of culture results. Culturing the organism from joint fluid is rarely effective.

Approximately 60-80% of specimens isolated from the leading edge of a suspected erythema migrans lesion by means of saline-lavage needle aspiration or 2-mm punch biopsy reveal B burgdorferi. However, because the presence of a lesion along with a compatible history and clinical presentation are sufficient to initiate treatment, these skin biopsy procedures are seldom performed.

Erythema migrans

Histologic findings in erythema migrans are nonspecific, usually showing a perivascular cellular infiltrate consisting of lymphocytes, plasma cells, and histiocytes (see the following images). Occasionally, mast cells and neutrophils are seen. Central biopsies may show eosinophilic infiltrates consistent with a local reaction to an arthropod bite. Spirochetes occasionally may be identified using silver or antibody-labeled stains, although usually, a paucity of spirochetes is found in the tissues of patients with Lyme disease. In addition, cultures for B burgdorferi are positive in approximately 75% of patients.

Borrelial lymphocytoma

Histologic examination is recommended in patients with suspected borrelial lymphocytoma, when the location of the lesion or the clinical history is not clear to make a diagnosis. Borrelial lymphocytoma biopsy specimens show a dense dermal lymphocytic infiltrate with lymphoid follicles and pseudogerminal centers. Lymphocytes with both B-cell and T-cell markers, occasional macrophages, plasma cells, and eosinophils are seen.

Acrodermatitis chronica atrophicans

In acrodermatitis chronica atrophicans, biopsy specimens from early lesions show a lymphocytic dermal infiltrate, sometimes perivascular in location, with some vascular telangiectasia and lymphedema. Plasma cells also may be seen in the cellular infiltrate. Later lesions demonstrate epidermal thinning with loss of skin appendages. At this stage, plasma cells may be the only feature to distinguish acrodermatitis chronica atrophicans from morphea. The fibrotic nodules show fibrosis of the deeper dermis and sometimes, hyalinization of collagen bundles. B burgdorferi occasionally can be cultivated from the lesions; in one patient, cultivation was successful more than 10 years after the lesion's first appearance.

Spinal fluid should be obtained in patients with neurologic symptoms whose diagnosis is not obvious (eg, patients without erythema migrans). Unlike most bacterial infections in the spinal fluid, Lyme disease produces a pleocytosis characterized by mononuclear cells. In addition, spinal fluid levels of IgM and IgG antibodies to B burgdorferi should be measured, and an index of cerebrospinal fluid (CSF) to serum antibody (immunoglobulin-to-albumin ratio) should be calculated.^[27] This is particularly true in patients who have no other signs of Lyme disease.

Although CSF cultures are positive in less than 10% of patients with apparent meningitis, intrathecal antibodies and a lymphocytic pleocytosis (approximately 100 cells/µL) are present in more than 80%. Patients with meningitis typically have elevated protein concentrations (>50 mg/dL) but normal glucose levels (45-80 mg/dL). Oligoclonal bands specific for B burgdorferi may be present.

Ongoing controversy surrounds the diagnosis of neurologic Lyme disease. One of the most important concepts to understand is that a positive Lyme disease serology in CSF does not mean that the person has neuroborreliosis. It could represent evidence of a previous infection or simply reflect potential leakage of serum antibodies across the blood-brain barrier.^[28] IgG and IgM antibodies may persist in CSF long after adequate treatment and in the absence of evidence of active neurologic disease.

Intrathecal anti-Borrelia antibody production is typically seen within 3-6 weeks of infection. Anti-Borrelia antibody CSF-to-serum index has been reported to show a 97% specificity and 75% sensitivity for the diagnosis of neuroborreliosis.^[29] A CSF-to-serum index greater than 1.0 suggests synthesis of antibody in the central nervous system (CNS).

It has been proposed that 4 of the following 5 criteria should be present in order to diagnose neuroborreliosis^[30] :

• No past history of neuroborreliosis
• CSF anti-B burgdorferiantibodies
• Positive anti-B burgdorferiantibody index
• Favorable clinical outcome after proper antibiotic therapy
• Absence of alternative diagnosis

Lyme multiplex PCR has not been standardized; therefore, it is not currently used in routine testing.

PCR is used to detect B burgdorferi DNA in the blood, CSF, urine, or synovial fluid within weeks of infection. The result is positive in approximately 30% of patients with active Lyme disease.

A notable disadvantage of PCR testing is the likelihood of false-negative results because of a sparsity of spirochetes in infected tissues. Likewise, inexperience with the PCR technique can yield false-positive findings when care is not taken to prevent contamination and when incorrect primers are used in preparing the specimen.

Although most PCR results become negative within 2 weeks of antimicrobial therapy, results can remain positive for years after apparent cure.

One of the most compelling uses of PCR may be in confirming persistent or recurrent disease, because a positive result is highly specific for exposure to B burgdorferi.

This test is used to detect highly specific borreliacidal antibodies that formed after exposure to B burgdorferi. The specificity of borreliacidal antibody test results far exceeds those of ELISA. One study of 572 patients showed a specificity of greater than 99%, although its sensitivity is marginal, at 72%. Although borreliacidal antibodies are detectable in the serum soon after infection, the number of antibodies increases with the duration and severity of illness; therefore, borreliacidal antibody testing is most useful in late disease. It should not be performed in patients who recently received antimicrobial therapy.

Electrocardiograms (ECGs) show fluctuating levels of atrioventricular block in patients with syncopal or near-syncopal symptoms secondary to Lyme carditis. In patients with possible exposure but without symptoms of myocardial ischemia, such changes should prompt further investigation for Lyme disease.

Abnormal electrophysiology results are often consistent with axonal degeneration in patients presenting with peripheral neuropathy in stage 3 disease.

Magnetic resonance images (MRIs) show abnormalities in approximately 15-20% of patients in the United States who have neurologic manifestations of Lyme disease. In European patients with CSF-confirmed Lyme disease, imaging findings have suggested that microvasculitis and macrovasculitis in the CNS may be responsible for neurologic sequelae and the MRI changes seen in patients with neuroborreliosis.

Punctate lesions of the periventricular white matter are common and resemble changes seen in demyelinating or inflammatory disorders. In an attempt to differentiate radiologic manifestations of neuroborreliosis and multiple sclerosis, one study proposed that occult brain tissue damage (seen by brain magnetization transfer and diffusion tensor magnetic resonance) are not common in neuroborreliosis, as opposed to multiple sclerosis.^[31] Space-occupying lesions have also been reported as a rare manifestation.

Functional brain imaging, such as single-photon emission computed tomography (CT) scanning, may contribute to the diagnosis of chronic neurologic Lyme disease.