Lyme disease, transmitted by tick bites, can be challenging to diagnose and treat. This comprehensive guide, based on official French medical society recommendations, explains how doctors use blood tests and spinal fluid analysis to confirm the infection, why a two-step testing approach is essential, and which antibiotics work best for different stages of the disease. The guidelines also address a critical concern for many patients: persistent symptoms after treatment. Importantly, the research shows that in 80% of patients with ongoing symptoms after suspected Lyme disease, another condition is eventually diagnosed, and repeated or prolonged antibiotic treatment is not recommended.
Lyme borreliosis, commonly known as Lyme disease, is caused by bacteria of the Borrelia burgdorferi sensu lato group, which are transmitted to humans through tick bites. In Europe, three main species can cause disease in humans: Borrelia afzelii, Borrelia garinii, and Borrelia burgdorferi sensu stricto. The disease can affect multiple organ systems, including the skin, nervous system, joints, heart, and eyes.
These guidelines were developed by a consortium of French scientific societies, including specialists in infectious diseases, dermatology, rheumatology, neurology, pediatrics, internal medicine, microbiology, psychiatry, cardiology, and general practice. They represent a comprehensive, evidence-based approach to diagnosing and treating Lyme disease and other tick-borne illnesses.
One of the most important messages from these guidelines is that diagnosis is not always straightforward. The accuracy of diagnostic tests depends heavily on the stage of the disease and the clinical presentation. Understanding these nuances can help patients navigate their care more effectively.
All national, European, and American evidence-based guidelines recommend a two-tier serology strategy for diagnosing Lyme disease. This approach involves two steps:
Almost all currently available ELISA tests include antigens from the three main European species pathogenic to humans, providing better sensitivity and specificity than first-generation ELISA tests. Several studies have demonstrated that one-tier (ELISA alone) and two-tier strategies (ELISA plus Western blot) have similar performance. However, no study has ever demonstrated the superiority of ELISA alone versus the two-tier strategy.
It is crucial to understand that no screening test is available for active Borrelia infection because asymptomatic seropositivity (having antibodies without symptoms) is common. In other words, a positive blood test alone is not sufficient to establish a diagnosis of Lyme disease. The test results must always be interpreted in the context of clinical symptoms and exposure history.
Detailed information about the characteristics and performance of commercialized serological tests, including the antigens used, study populations, and disease stages, is available from the French Agency for the Safety of Health Products (ANSM) and the national reference center for Borrelia (CNR). These resources help standardize testing and maximize diagnostic performance.
The performance of serological tests varies dramatically depending on the stage of the disease. This is one of the most important concepts for patients to understand.
During early localized manifestations of Lyme disease, specifically the characteristic rash called erythema migrans, Lyme serology has poor sensitivity of only 30–40%. This means that 60–70% of patients with the classic rash will test negative. Therefore, the guidelines clearly state that serological testing should not be performed when a patient presents with erythema migrans. The diagnosis at this stage should be based purely on clinical signs and symptoms.
Seroconversion (the development of detectable antibodies) occurs within approximately six weeks after symptoms onset, with IgG detection. Six weeks after symptoms begin, the serological test is associated with greater than 90% sensitivity and specificity. This means that by six weeks, the test correctly identifies more than 90% of infected patients and correctly rules out the disease in more than 90% of uninfected patients.
For early disseminated manifestations with neurological symptoms occurring within six weeks after the tick bite (early Lyme neuroborreliosis), the blood serological test may be negative. In these cases, the biological diagnosis should be based on the results of cerebrospinal fluid (CSF) analysis, which requires a lumbar puncture (spinal tap).
The sole persistence of IgM antibodies beyond six weeks should be considered a false positive result, because of the high risk of non-specific cross-reactions. This is an important point: IgM antibodies that remain positive for more than six weeks are likely not meaningful and should not be used to make a diagnosis.
The sensitivity of serological tests in late disseminated neurological, cutaneous, or joint manifestations is close to 100%, and very high IgG levels are common. Immunoblot testing almost always reveals IgG targeted against numerous B. burgdorferi sensu lato antigens. The ELISA serological test is rarely negative in patients presenting with late Lyme disease.
However, a few exceptional cases have been reported: two cases of acrodermatitis chronica atrophicans (ACA, a late skin manifestation) with atypical clinical presentation, one case of Lyme arthritis, and one case of seronegative late Lyme neuroborreliosis. Thus, a negative Lyme serological test result at this late stage of the disease should lead physicians to question the diagnostic hypothesis.
The sole presence of IgG, without IgM, is common in late manifestations of Lyme disease, even if the culture is positive (for example, 15–20% of ACA cases). High levels of antibodies can be observed in treated patients several years after recovery. The treatment should in that case not be resumed. Serological follow-up is therefore not recommended, and treatment success should be assessed based on clinical signs and symptoms.
A positive serology result does not distinguish an active infection from a serological scar. A positive serology result without any clinical signs and symptoms is either suggestive of a serological scar or an asymptomatic seroconversion indicating contamination but not active Lyme disease.
A Swiss longitudinal study conducted from 1986 to 1993 followed 305 patients infected with B. burgdorferi sensu lato who had a positive Lyme serology but no initial clinical signs. The study reported that more than 95% of these patients remained asymptomatic at seven years of follow-up. This finding underscores that a positive blood test alone, without symptoms, does not warrant treatment.
A prospective Scandinavian study showed that, in the absence of prophylaxis and within three months following a tick bite, 5.4% of patients (102 out of 1,886) achieved seroconversion, with clinical signs of Lyme disease developing in 39.2% of those cases (40 out of 102). This means that even after a confirmed tick bite, the vast majority of people do not develop Lyme disease.
For neurological manifestations, the guidelines recommend simultaneous quantification of specific anti-B. burgdorferi sensu lato IgG and total IgG in both the blood and the cerebrospinal fluid using ELISA testing. This allows calculation of the intrathecal synthesis index, which measures whether antibodies are being produced within the central nervous system itself. This is a strong indicator of neuroborreliosis (neurological Lyme disease).
Polymerase chain reaction (PCR) testing detects the DNA of B. burgdorferi sensu lato directly. The specificity of the test should be close to 100%, though this is not always the case depending on the manufacturer, as the targets, primers, and methods are neither standardized nor assessed.
A positive PCR result for B. burgdorferi sensu lato does not establish active infection. The PCR sensitivity varies depending on the disease stage and its localization. PCR testing is useful for difficult-to-establish diagnoses for cutaneous manifestations (PCR test on skin biopsy) or joint manifestations (PCR test on synovial fluid or synovial biopsy).
However, PCR is pointless in patients presenting with neurological manifestations for more than six weeks, due to poor sensitivity. Looking for B. burgdorferi sensu lato by PCR test in urine and blood samples is not recommended, as studies reported highly contradictory results.
Culture is the reference biological diagnostic method, with 100% specificity but limited sensitivity because of the small number of bacteria at the sampling sites. There is no healthy carriage of B. burgdorferi sensu lato: isolation of the bacterium indicates active Lyme disease.
Culture is performed in specialized laboratories. The culture medium is specific (called BSK), enriched, and it may easily be contaminated by commensal bacteria. Culture takes time, usually 2–8 weeks, and negative results are available only after three months.
Spirochetes (the corkscrew-shaped bacteria) cannot be detected by Gram staining at direct microscopic examination. A dark-field or phase-contrast microscope is required, or direct immunofluorescence should be used, though these have moderate sensitivity and specificity. Official identification of the bacterium is then performed by molecular biology. Microscopy can be used to interpret culture results, but direct microscopy on samples is not recommended because of its lack of specificity.
Histology (tissue examination under a microscope) is useful for the diagnosis of acrodermatitis chronica atrophicans and for differential diagnoses, but the result is not indicative of active Lyme disease.
A promising test measures the level of a protein called CXCL13 in the cerebrospinal fluid. This test has 89–97% sensitivity and 92–98% specificity in Lyme neuroborreliosis. However, it is still under evaluation and not yet part of routine clinical practice.
New tests will need to be assessed as part of prospective studies in future reference centers. Results will be published and evaluated by the ANSM and the CNRs for the relevant pathogens.
No radiologic lesion is indicative of Lyme disease. Imaging tests are mainly used to investigate a differential diagnosis, meaning they help rule out other conditions that might explain the symptoms.
Erythema migrans, the characteristic "bull's-eye" rash, spontaneously resolves without treatment within a few weeks. However, B. burgdorferi sensu lato may persist in the skin, and new manifestations may occur later on. An antibiotic therapy is therefore required, with documented efficacy, irrespective of symptoms duration before treatment.
Twenty studies compared several molecules with various treatment durations, dosing regimens, and outcomes. Antibiotics with proven efficacy include:
A meta-analysis suggested the absence of difference in efficacy and tolerability between molecules, with low rates of treatment failure (4% at 2 months, 2% at 12 months). Considering the continuum between early localized and disseminated manifestations and the efficacy of doxycycline in patients presenting with Lyme neuroborreliosis, doxycycline should be favored as first-line treatment.
Erythema migrans is usually cured after 7 to 13 days of antibiotic therapy. The recommended regimen is doxycycline for 14 days, with amoxicillin as an alternative. The specific dosing is typically doxycycline 200 mg per day.
Borrelial lymphocytoma is a less common skin manifestation of Lyme disease, appearing as a bluish-red nodule or plaque, most often on the ear lobe, nipple, or scrotum. A study of 144 patients with borrelial lymphocytoma found that antibiotic therapy was required for 14 patients (9.7%) because of initial treatment failure, defined as persistence of the lesion for more than one month after treatment, new signs of Lyme disease, or persistence of clear non-specific signs and symptoms. The main risk factor for treatment failure was the presence of signs suggestive of dissemination. The outcome was favorable for all patients at one year, with disappearance of the lymphocytoma within 21 days on average (range 10–30 days).
The first-line treatment of borrelial lymphocytoma is doxycycline (alternative: amoxicillin), at the same dosage as for erythema migrans, but for 21 days. Children can alternatively be treated with azithromycin for 10 days.
Recent guidelines from other countries recommend the use of amoxicillin 1 gram three times daily or doxycycline 200 mg/day for 21 days. Only the Belgian guidelines recommend a shorter treatment with doxycycline for 10 days or amoxicillin for 14 days.
The antibiotic therapy of Lyme neuroborreliosis has never been evaluated in placebo-controlled studies. The Cochrane Library retrieved seven European randomized studies, including one pediatric study, assessing penicillin G, cefotaxime, ceftriaxone, and doxycycline for 10 to 21 days.
An open-label randomized study did not report any difference between a 14-day regimen and a 28-day regimen with ceftriaxone. A retrospective cohort study of early Lyme disease manifestations, mainly cutaneous but also neurological, did not show any difference between treatment durations of less than 10 days and more than 16 days.
A non-inferiority, multicenter, randomized, placebo-controlled, blinded study is currently ongoing to compare two weeks with six weeks of doxycycline. A study demonstrated the non-inferiority of doxycycline (200 mg/day) versus ceftriaxone (2 g/day) for early disseminated manifestations. The studies included few late manifestations (10%), and specific analyses were not performed.
Adverse effects of ceftriaxone, mainly due to the parenteral (intravenous) route of administration or its broad spectrum, should lead physicians to favor doxycycline for the treatment of Lyme neuroborreliosis.
The evidence-based German guidelines recommend ceftriaxone or doxycycline during 14 days for early Lyme neuroborreliosis and during 14 to 21 days for late Lyme neuroborreliosis. A review of available pediatric data resulted in the same suggestions.
The British guidelines make a distinction between central and peripheral symptoms:
In summary, the recommended treatment for neuroborreliosis is doxycycline for 14 days if early, and 21 days if late, with ceftriaxone as an alternative.
Lyme arthritis typically presents as a swollen, painful joint, most commonly the knee. The treatment of articular manifestations of Lyme disease is based on doxycycline, ceftriaxone, or amoxicillin for 28 days. This longer duration reflects the need to eradicate bacteria that may be sequestered in joint tissues.
While the guidelines focus primarily on erythema migrans, neuroborreliosis, and arthritis, they also acknowledge other manifestations of Lyme disease, including:
For these manifestations, the same antibiotic principles apply, with treatment duration and route determined by the severity and location of the infection.
This is perhaps the most important section for many patients. Some patients present with persistent and pleomorphic (varied and changing) symptoms after documented or suspected Lyme disease. These symptoms may include fatigue, musculoskeletal pain, cognitive difficulties, and other nonspecific complaints.
The guidelines are clear and emphatic: Patients with persistent symptoms after appropriate treatment of Lyme disease should not be prescribed repeated or prolonged antibacterial treatment.
Critically, the research shows that another condition is eventually diagnosed in 80% of patients who present with persistent symptoms after documented or suspected Lyme disease. This means that the vast majority of patients with ongoing symptoms do not have an ongoing Borrelia infection, and their symptoms are due to something else.
This finding has profound implications for patient care. Rather than repeated courses of antibiotics, which can cause harm including antibiotic resistance, disruption of normal gut flora, and potential side effects, patients with persistent symptoms deserve a thorough evaluation for other potential causes of their symptoms.
The guidelines emphasize that treatment success should be assessed based on clinical signs and symptoms, not on laboratory values. High levels of antibodies can persist for years after successful treatment and do not indicate ongoing infection.
Ticks can transmit several other diseases besides Lyme disease. The guidelines provide diagnostic recommendations for these conditions as well.
Tick-borne encephalitis is a viral infection that can cause meningitis and encephalitis. Diagnosis is based on serology showing IgG and IgM seroconversion or increased IgG levels. Cross-reactions can occur with other arboviruses (viruses transmitted by arthropods) and yellow fever vaccine. PCR is only useful at the initial stage (viremia). CSF examination should show monocytosis (increased monocytes) and IgM. Urine PCR should not be performed.
This bacterial infection causes fever, headache, and muscle aches. Serology for IgG, IgM, or total immunoglobulins (TIg) is used. Diagnosis requires increased TIg or seroconversion, with sensitivity of 32% for initial testing. High levels of TIg have 74% sensitivity and 100% specificity. PCR on whole blood during the viremic stage has only 20% sensitivity. A thin blood film can detect morulae (clusters of bacteria inside white blood cells).
Babesiosis is a parasitic infection that affects red blood cells. Serology for IgG, IgM, or TIg is used, with diagnosis based on seroconversion or increased TIg. PCR on whole blood at the febrile stage is useful. A thin blood film can detect trophozoites (the parasitic form in red blood cells). Dark-field microscopy or phase-contrast microscopy should not be performed.
This relapsing fever-like illness is caused by a different Borrelia species. Diagnosis is made by PCR on whole blood at the febrile stage. Thin blood film and thick blood smear should not be performed.
This emerging infection is diagnosed by PCR on whole blood at the febrile stage. Thin blood film and thick blood smear should not be performed.
This rickettsial infection causes fever, rash, and a characteristic black spot (eschar) at the tick bite site. Serology for IgG and IgM shows seroconversion or increased TIg, with 100% sensitivity after Day 30. There is a risk of cross-reactions between species. PCR on the black spot is useful.
This bacterial infection can cause skin ulcers and swollen lymph nodes. Serology for IgG or TIg shows seroconversion or increased IgG levels, with 100% sensitivity after Day 30. IgM cross-reactions can occur. PCR on ulcer or lymph node samples has 75% sensitivity.
It is important to acknowledge the limitations of the evidence underlying these guidelines. The antibiotic therapy of Lyme neuroborreliosis has never been evaluated in placebo-controlled studies, meaning we lack the highest level of evidence for this treatment. The available studies included few late manifestations (only 10%), and specific analyses were not performed for these cases.
For serological testing, the sensitivity varies widely depending on the stage of disease, and no test can distinguish between an active infection and a past infection (serological scar). The performance characteristics of commercial tests vary, and standardization remains a challenge.
For other tick-borne diseases, the evidence base is even more limited, with many diagnostic tests having unknown sensitivity or specificity. The guidelines note that new tests will need to be assessed as part of prospective studies in future reference centers.
Finally, the finding that 80% of patients with persistent symptoms after suspected Lyme disease receive another diagnosis highlights the complexity of these cases and the need for thorough, multidisciplinary evaluation.
Based on these guidelines, here are the key recommendations for patients:
In early Lyme disease with erythema migrans, blood tests are negative in 60–70% of cases. Therefore, testing is not recommended at this stage. Diagnosis should be based on the rash and exposure history. Treatment with doxycycline for 14 days is highly effective.
Tests lacking sensitivity or specificity include thin blood film, dark-field microscopy, CD57+ cell levels, rapid diagnostic tests, and PCR in blood or urine. Also not recommended are lymphocyte transformation tests, xenodiagnosis, and CCL19 testing. These are not validated for clinical use.
Original Article Title: Lyme borreliosis and other tick-borne diseases. Guidelines from the French scientific societies
Authors: B. Jaulhac, A. Saunier, E. Caumes, K. Bouiller, J.F. Gehanno, C. Rabaud, S. Perrot, C. Eldin, T. de Broucker, F. Roblot, J. Toubiana, F. Sella, F. Vuillemet, C. Sordet, B. Fantin, G. Lina, C. Sobas, X. Gocko, J. Figoni, C. Chirouze, Y. Hansmann, V. Hentgen, P. Cathebras, M. Dieudonné, B. Bodaghi, J.P. Gangneux, B. Degeilh, H. Partouche, C. Lenormand, A. Sotto, A. Raffetin, J.J. Monsuez, C. Michel, N. Boulanger, C. Lemogne, P. Tattevin, and the French scientific societies (SFD, SFR, FFN, SFN, CNGE, CMG, SNFMI, SFM, CNPP-CNQSP, AFPBN, SPMPLLF, SFMT, SFC, SPD, GPIP, SOFREMIP, SFO, SFMM, SFP, CMIT, CNP-FFI, CNOG, SFETD, SPILF)
Journal: Médecine et maladies infectieuses (2019)
DOI: https://doi.org/10.1016/j.medmal.2019.05.001
Publication Date: Available online 2019 (Received March 30, 2019; Accepted May 7, 2019)
Note: This patient-friendly article is based on peer-reviewed research. It is intended for educational purposes and should not replace professional medical advice. Always consult with a qualified healthcare provider regarding diagnosis and treatment of any medical condition.
Use collapsible tabs for more detailed information that will help customers make a purchasing decision.
Ex: Shipping and return policies, size guides, and other common questions.