How accurate are Lyme disease tests?

Lyme disease tests look at whether the body makes antibodies against the Borrelia burgdorferi spirochete. Antibodies can be measured with ELISA and Western Blot test methods. There are ELISA and Western blot tests for many infectious diseases.

Some think these tests perform poorly for Lyme disease. Commercial laboratories have created their own version of these tests and use ‘in house’ criteria to define positive or negative results.

The current tests can come back negative in patients who nevertheless are infected with Lyme disease. Because of this possibility your medical history, exposure to ticks and current symptoms are important to share with your physician. In the context of your symptoms and tick exposure, the available tests for Lyme disease perform reasonably well. However, there are some limitations. It would be helpful for physicians to have other diagnostic tests available to make a more confident diagnosis. That is why scientists are looking for new ways to test for Lyme disease. How do these currently available tests work?


Antibodies are Y shaped proteins produced by plasma cells, a type of white blood cell. The immune system uses these antibodies to neutralize pathogens. The antibody recognizes a unique molecule of the pathogen called an antigen and binds to it.

The antibody tags the pathogen for attack by other parts of the immune system. It can also impair the function of the pathogen by inhibiting a part of the pathogen that is essential for its survival or glue multiple pathogens together. Antibodies are also called immunoglobulins and they appear in several forms.


Immunoglobulin M is the largest antibody and it is the first antibody to appear in the response to an infection. The spleen produces most of these first response immunoglobulins. Demonstrating IgM antibodies in a patient’s serum indicates recent infection with Lyme disease.


The immune system starts generating immunoglobulin G when the antibody response matures. These antibodies participate predominantly in the later phase of the immune response to infection. The production of IgG antibodies in Lyme disease is slow and can be measured after 4-6 weeks.


The ELISA test uses the serum of the patient and combines it with antigens of the Borrelia burgdorferi spirochete. If antibodies to the organism are present in the patient’s blood, they will bind to the antigen. These bound antibodies can then be detected when a second solution is added to the preparation.

This second solution has a color and binds to the antibodies of the patient. The measured color value in a given dilution is a measure of the amount of antibodies in the sample.

The ELISA measures all antibodies that a patient has produced against the antigens of Borrelia burgdorferi. It is not able to qualify to which antigens the patient’s immune system has produced a response. That is why the ELISA test is called a quantitative test.

It is however possible to test positive with an ELISA test even when you do not have Lyme disease. A Western Blot is used to verify the ELISA. This can help distinguish patients who have Lyme disease from those with other conditions.

Western Blot

When the ELISA test for Lyme disease returns positive a Western Blot is used to confirm the result. While the ELISA  is a quantitative test, the Western Blot is a qualitative test that looks specifically against which antigens of Borrelia burgdorferi the immune system has produced antibodies.

To do this, the antigens of the Borrelia burgdorferi spirochete are separated by molecular weight, expressed in Kilo Dalton (kDa). The separated antigens are then applied to a paper or nylon sheet. From this point on, the procedure is similar to that of the ELISA.

CDC criteria

The 2 tier testing protocol of the ELISA test confirmed by the Western Blot was established in 1994.

  • All serum specimens should be tested by a two-step process. In which the first step is a sensitive ELISA test. Specimens found positive or equivocal should be tested by a standardized Western Blot procedure. Specimens found to be negative by a sensitive ELISA need not to be tested further and reported negative.
  • When Western Blot is used in the first four weeks after disease onset both IgM and IgG procedures should be performed. In late Lyme disease the predominant antibody response is usually IgG. A positive IgM test result alone is not recommended for use in determining active disease in persons with illness of longer than one month duration.

Western Blot bands

  • An IgM Western Blot is considered positive if two of the following three bands are present: 24 kDa (Osp C)*, 39 kDa (BmpA) and 41kDa (flagella).
  • An IgG Western Blot is considered positive if five of the following ten bands are present: 18 kDa, 21 kDa (OspC) *, 28 kDa, 30 kDa, 39 kDa (BmpA), 41 kDa (Fla), 45 kDa, 58 kDa (not GroEL), 66 kDa, and 93 kDa.
    • The molecular weight of Osp C is dependent on the strain of Borrelia burgdorferi tested. 24kDa and 21kDa proteins referred to are the same.

Reporting of results

Equivocal or positive ELISA followed by a negative Western Blot is reported as negative. Equivocal or positive ELISA followed by a positive Western Blot is reported as positive.

Limitations of Lyme disease tests

Serological tests do not become positive until an infected individual has had time to develop antibodies. This means that early acute Lyme disease, often characterized by an erythema migrans at the site of a tick bite, cannot be diagnosed by serology.

Not all patients develop an erythema migrans. In these patients the serological tests will not be able to confirm an infection for at least 4 weeks after becoming ill. There is an increased chance of persistent symptoms in patients with a delayed diagnosis.

Another problem is that serological tests cannot confirm a cure. Antibodies persist after the infection is gone. This means that your blood will continue to test positive months or years after treatment. In cases were symptoms return or persist despite earlier antibiotic therapy these tests are not useful to determine whether there is treatment failure.

Reinfected patients will already have antibodies against Borrelia burgdorferi. It can be difficult to use serological tests to confirm the reinfection. This is especially difficult when the patient does not remember having a tick bite or an erythema migrans but does present with symptoms that suggest disseminated infection.

In treatment failure and reinfection it might be useful to use serial Western Blots to document a rise in antibodies.

Correct use of Lyme disease tests

According to the CDC diagnostic tests are of clinical value only if they are used appropriately. The CDC makes the recommendation to estimate the pretest likelihood that a patient has Lyme disease by looking for objective signs of Lyme disease and a history of potential exposure to ticks.

This recommendation is meant to reduce the possibility that patients with a false positive test result receive unnecessary treatment. However it is important to note that what the CDC considers objective signs of Lyme disease is a very restrictive and narrow view of Lyme disease.


This recommendation can not only reduce the possibility patients with a false positive test result receive unnecessary treatment; it can also increase the possibility that patients with true Lyme disease receive a wrong diagnosis.

  1. It can erroneously give the impression that there is a false positive result in patients when a less well described expression of Lyme disease is dominant in the patient.
  2. A false-negative result can be interpreted as a true negative result in a patient with a less well described expression of Lyme disease in the favor of another diagnosis.

This risk is largest when patients present with symptoms of late neurological Lyme disease. Encephalopathy is a result of late neurological Lyme disease and is not included in the CDC case definition.

According to the researchers Fallon, Coyle and Logigian Lyme encephalopathy can result in symptoms such as depression, anxiety, word finding difficulties, memory and concentration problems, confusion and sleep disturbances.

Fallon, Omasits and Kohler have published about neuropsychiatric manifestations in patients with Lyme disease. Case studies suggest that Lyme disease can be associated with symptoms common to schizophrenia and bipolar disorder, including paranoia, delusions, olfactory, auditory and visual hallucinations, catatonia, and mania.

Kohler attempted to categorize the psychiatric symptoms by stage, listing depressive mood in early disease, organic personality disorder in the mid-stage and organic psychosis and dementia in the later stages of Lyme disease and Omasits stated that psychiatric manifestations of Lyme disease can be predominant in certain patients.

Another overlooked outcome of late neurological Lyme disease is dementia. Milkossy, Allen and McDonald published on the role of spirochetal infections and dementia. One study found 1,25% of 1.594 patients seen for dementia had positive anti-Borrelia antibodies in their spinal fluid. Those patients stabilized or had mild improvement of dementia after treatment with antibiotics.

Defining accuracy

It is problematic to determine the frequency of seroreactivity in patients with neurological, cardiological or joint symptoms because serological confirmation is part of the case definition of Lyme disease. Is it possible to avoid circular reasoning?

A medical history is carefully considered when choosing samples.  Patients should fit the clinical symptoms known to be associated with Lyme disease. However it remains difficult to determine whether such a well-defined patient group is representative for all Lyme patients.

In addition, there is doubt about the control groups in various studies that tried to determine the accuracy of serological tests for Lyme disease.  Lyme disease can cause encephalomyelitis that is hard to distinguish from MS in the late neurological stage. MS patients have been used as a control group to determine cross-reactivity of serological tests with other disorders. It is possible that the demyelization of MS patients with antibodies against Borrelia burgdorferi was caused by Lyme disease. Therefore their test results could have been true positives instead of false positives.

In other studies chronic fatigue syndrome patients where included as a control group. Chronic fatigue syndrome presents clinically in a similar way as post-Lyme syndrome. Symptoms of chronic fatigue syndrome overlap with Lyme encephalopathy and post-Lyme syndrome.

Although there are good indications that chronic fatigue syndrome has a different pathological cause than post-Lyme syndrome thanks to studies done at Columbia University, there are no tests to distinguish Lyme patients from chronic fatigue syndrome patients. How do you rule out the possibility that the false positives reported where true positives?

Despite careful attempts to determine the accuracy of serological tests for Lyme disease it is extremely difficult to come to definite conclusions about accuracy without a gold standard.

Sensitivity and specificity

In the Netherlands researchers evaluated 87 studies that reported on the sensitivity and specificity of ELISA and Western Blots.

There are differences with regards to Lyme disease testing between Europe and the United States. Mentioned sensitivity and specificity can differ with the tests available in the United States.

  • The sensitivity of the two-tier testing was highly heterogeneous in patients presenting with an erythema migrans. The average sensitivity was 50%.
  • The sensitivity of the two-tier testing had an average sensitivity of 77% in patients presenting with neurological Lyme disease.
  • In patients with a presentation of acrodermatitis chronicum migrans the average sensitivity of the two-tier testing was 97%.
  • For unspecified Lyme disease the average sensitivity was 73%.
  • The specificity was around 95% in studies with healthy controls but around 80% in cross-sectional studies.

Test sensitivity is the ability of a test to correctly identify those with the disease, whereas test specificity is the ability of the test to correctly identify those without the disease.

  • If 100 patients known to have a disease were tested, and 43 test positive, then the test has 43% sensitivity.
  • If 100 with no disease are tested and 96 return a negative result, then the test has 96% specificity.

It is not true that the most sensitive test will best rule out a disease and the most specific test will best rule in the disease because the diagnostic power of a test is determined by both sensitivity and specificity. It is important to take into account the correlation between the two.

The Bayesian theorem

Patient associations incorrectly apply data from early Lyme disease patients to people with later stages of the disease. This misrepresents the accuracy of Lyme disease tests.

You can’t make the assumption that when a test has a sensitivity of 50% the chance of the patient having the disease when they test positive is 50%. You need to apply Bayes theorem to get some perspective.

Bayes theorem in short:

  1. Take the prior probability of having the disease
  2. Multiply it by the probability of testing positive if you had the disease
  3. Divide that by the total probability of the event occurring: this is the combination of having the disease and testing positive plus the probability of not having the disease and being falsely identified.

The prior probability of having the disease is often the hardest part of this equation to figure out. A reasonable starting point is the frequency of the disease in the population but there is a risk of lapsing into circular reasoning again because of the concerns with Lyme disease tests.

If this sounds complicated, that is because it is.

Seronegative Lyme disease

In the late 1980’s scientists have held the opinion that it is possible for patients to have a weak antibody response to a Lyme infection. In 2018 most scientists have abandoned this idea. However some physicians believe it is possible that not all patients are able to meet the serological standard established in 1994.

Studies from the 1980’s

Patients often quote studies from the late 1980’s to support their argument that seronegative Lyme disease is real.

At that time the optimal treatment of patients with an erythema migrans was unknown. The antibiotics used most often were penicillin and tetracycline. Both of these antibiotics perform worse than the modern antibiotics of first choice: doxycycline and amoxicillin. Because of these reasons more patients developed late neurological Lyme disease symptoms as a result of treatment failure.

Treatment for early Lyme disease can abort the development of antibodies. To identify patients without antibodies that developed late neurological Lyme disease Raymond Dattwyler relied on measuring their cellular immune response to Borrelia burgdorferi.

Reasons for lack of antibody response

Some scientists and physicians question whether all patients are able to meet the serological standard established in 1994. I summarized some of their concerns:

  • Another strain of Borrelia infects the patient
  • Borrelia persister forms have different antigen expression
  • The rotation of OSP’s cause a variation in antigen expression
  • Interaction of OSP-A with B-cells causes failure to launch a sufficient antibody response
  • Cut-off is set too high
  • The serological tests lack sensitivity and specificity
  • The dilution is set too high


Lyme disease tests are not perfect. The diagnosis of Lyme disease is a clinical diagnosis based on the medical history, exposure to ticks, laboratory results and symptoms. Lyme disease is called the second “great imitator” after syphilis. Physicians may not be aware of all the symptoms that Lyme disease can cause and in conjunction with a false negative test this may result in a negative Lyme disease diagnosis in some patients.

Physicians have to rely on imperfect tests every day. When those tests are used correctly there is generally no cause for concern.  However, many diseases produce objective and tangible outcomes. Lyme disease can present itself more elusive, especially in the late neurological stage. It would be very helpful to have more tests available to give a physician more tools to make a confident diagnosis.