Browse the corpus

This guideline pertains to the diagnosis and treatment of neurological manifestations of Lyme borreliosis in children and adults. The guideline also deals with aspects of chronic, non-specific symptoms associated with Lyme borreliosis, which are also subsumed under such terms as “post-treatment Lyme disease syndrome” (PTLDS), “chronic neuroborreliosis” and “Lyme encephalopathy” without a clear distinction being made between them. Twenty-two medical societies, 18 of which were members of the AWMF, the Robert Koch Institute and two patient organisations participated in its development. A systematic search and assessment of the literature for the first version of the S3 guideline “Lyme Neuroborreliosis” was carried out by the German Cochrane Centre Freiburg (Cochrane Germany) with significant input from RD. A systematic search and assessment of the literature for the updated guideline were again carried out by RD.

The previous S3 guideline “Lyme Neuroborreliosis” (AWMF register no. 030/071) was updated in accor d a nce with the methodological specifications of the Association of Scientific Medical Societies (AWMF). The update now includes the diagnosis and treatment of Lyme neuroborreliosis in children. There are still no analysable study data on the efficacy of combination antibiotic treatment. There are still no study data on the efficacy of chloroquine, carbapenems and metronidazole. A prospective, randomised clinical trial on the length of antibiotic treatment for early Lyme neuroborreliosis has now been published. It has found that treatment with doxycycline for 6 weeks provided no extra clinical benefit compared to treatment for 2 weeks [] . According to a pooled analysis with data from an earlier study [] , there is now level 1a evidence for the length of antibiotic treatment for early Lyme neuroborreliosis (section 5.2.1). A systematic review of the treatment for post-treatment Lyme disease syndrome (PTLDS) was carried out for the first time as part of the development of the guideline. It found that the analysed parameters quality of life, fatigue, depression and cognition did not respond to antibiotic treatment. There are currently no conclusive studies on the effectiveness of other treatment methods (section 4.3.5).

Steroids should not be administered alongside antibiotics to treat facial nerve paresis if there is probable or confirmed Lyme neuroborreliosis as per the diagnostic crite r i a (see section 3.10). Restrictions have been lifted on administering doxycycline to children under the age of 8 as part of the treatment of Lyme neuroborreliosis. Recent data show that this does not lead to a yellowing of the teeth [] , [] , [] , [] , [] (section 5.5).

The suspected clinical diagnosis of Lyme neuroborreliosis (cranial nerve deficits, meningitis/meningoradiculitis, encephalomyelitis) can be confirmed by the detection of i n flammatory changes in the cerebrospinal fluid in con junc tion with a borrelia-specific intrathecal antibody synthesis. Serology testing should only be ordered if there is sufficient clinical suspicion. Early Lyme neuroborreliosis should be treated with one of the following antibiotics: doxycycline, ceftriaxone, cefotaxime, penicillin G. Level of consensus: 94% (16/17 yes, 1 abstention) Late Lyme neuroborreliosis should be treated with one of the following antibiotics: doxycycline, ceftriaxone, cefotaxime, penicillin G. Doxycycline and beta-lactam antibiotics show similar efficacy and safety in regard to neurological symptoms in early Lyme neuroborreliosis [] , [] (section 5.2.2). Antibiotic treatment for early Lyme neuroborreliosis should be carried out over a period of 14 days. Antibiotic treatment for late Lyme neuroborreliosis should be carried out over a period of 14–21 days. Treatment success should be based on clinical symptoms. A systematic review found that the apparently high prevalence of persistent, non-specific, or atypical symptoms following a presumed case of Lyme neuroborreliosis, as reported in many studies, is largely attributable to methodological biases resulting from unclear case definitions [] .

Lyme borreliosis is the most common tick-borne infectious disease in Europe. A neurological manifestation occurs in 3–15% of infections and can manifest as polyradiculitis, meningitis and (rarely) encephalomyelitis. The disease can be treated with antibiotics.

This guideline is directed at physicians in private practices and clinics specialising in various fields of medicine who are directly or indirectly involved in treating Lyme neuroborreliosis in children and adults. The wide range of interdisciplinary medical fields dealing with Lyme neuroborreliosis is reflected in the number of medical societies involved and in the participation of the Robert Koch Institute (see the Guideline Report in Attachment 1 ). The guideline also acts as a source of information for patients and others interested in Lyme neuroborreliosis. These groups are represented in the consensus process by mandate holders from two patient and/or interest organisations (see the Guideline Report in ).

Differentiation of non-specific symptoms Cerebrospinal fluid (CSF) testing including antibody detection in CSF Effective use of molecular testing and culture tests Observation of the tick bite; information sheet for patients Diseases caused by relapsing fever Borrelia (Borrelia miyamotoi and Borrelia recurrentis) are not covered in this guideline. Questions relating to co-infections linked to tick-borne diseases are not covered in this guideline.

Lyme borreliosis is a multi-systemic inflammatory disease caused by an infection with spirochetes from the Borrelia burgdorferi sensu lato complex. These are transmitted in Germany through the bite of the Ixodes ricinus tick.

Lyme borreliosis is the most common vector-borne disease in the temperate climate zones of the northern hemisphere and is endemic. In North America, Lyme borreliosis is caused exclusively by the Borrelia species Borrelia burgdorferi sensu stricto, while in Europe B. a fzelii , B. bavariensis and B. garinii have also been found to be pathogenic to humans. In addition, the newly identified species Borrelia spielmanii has the potential of being pathogenic to humans. It has been detected in 4 out of 160 skin isolates (all from erythema migrans), but has so far not been linked to Lyme neuroborreliosis in Germany (72 CSF isolates) []. The pathogenic potential of the various Borrelia burgdorferi species varies []. After B. garinii OspA type 4 was classified as a new species of Borrelia bavariensis [], a re-evaluation of 242 human isolates from Germany [] revealed 21% of the 72 CSF isolates were B. afzelii , 22% B. bavariensis and 29% B. garinii . Of the 160 skin isolates, 67% were B. a fzelii , 12% B. bavariensis and 12% B. garinii ; i.e, only the skin isolates showed a clear prevalence of one species, namely B. afzelii .

No reliable figures are currently available on the incidence of Lyme borreliosis in the individual European countries. An analysis of reporting registries from six eastern German federal states found a strongly fluctuating incidence of 0.5 cases per 100,000 inhabitants depending on the region, compared to 138 cases per 100,000 inhabitants in the period from 2013 to 2017 []. Secondary data analyses of health insurance data on the basis of the ICD-10 coding A 69.2 (G) found a significantly higher incidence of 179 per 100,000 inhabitants with area-dependent fluctuations up to a factor of 16 (40–646 per 100,000 inhabitants) []. An earlier study using a similar methodology arrived at even higher case numbers, although the authors do not rule out the possibility that their case numbers were overestimated due to clinical misdiagnoses or miscoding []. In summary, the available epidemiological data are insufficient for drawing definitive conclusions. Data published to date in Germany suggest an incidence of Lyme borreliosis of between 60,000 to >200,000 cases/year.

According to a survey on reportable manifestations of Lyme borreliosis in nine federal states in Germany, acute neuroborreliosis (2.7%) is the second most common clinical manifestation after erythema migrans (95%). This is followed by Lyme arthritis (2.1%) []. In a prospective, population-based study conducted in the Würzburg area, 313 cases of Lyme borreliosis were identified over a 12-month period, corresponding to an incidence of 111 per 100,000 inhabitants. This resulted in the following manifestation rates []. 89% erythema migrans (erythema migrans linked to another organ manifestation in a further 3% of cases) Late Lyme neuroborreliosis (stage III) was not identified. According to one study, children have a higher risk of developing Lyme neuroborreliosis after a tick bite than adults, most likely because they are more frequently bitten on the head [].

Borrelia-specific antibodies are found in 3–20% of healthy individuals in Germany and Austria, depending on the endemic region and age group [], [], [], []. A seroprevalence of 20% was found in 964 (asymptomatic) Swiss orienteers; in asymptomatic blood donors this was 8% []. A cross-sectional German study of children and adolescents ranging in age from 1 to 17 years found an average seroprevalence of 4.8%. The relative probability of a positive antibody result was age-dependent and increased for every year of life by 6% in girls and by 11% in boys []. An elevated level of borrelia-specific IgG antibodies were found in 20% of men over the age of 60 [].

Studies of ticks in southern Germany revealed average infection rates of around 1% for larvae, 10% for nymphs and 20% for adults []. In addition to regional differences in rates of infected ticks (18–37% of adults and 5–12% of nymphs), there were also clear differences in the regional distribution of Borrelia species []. Infection rates in Switzerland were 5–7% depending on the region []. The population density of infected ticks also varies greatly from region to region, ranging from 2 to 58 per 100 m 2 in Switzerland. Across Europe, an analysis of publications from 24 countries revealed an average rate of infected ticks of 14.2% (range: 3.1–38.1%) []. In addition to Lyme borreliosis, ticks can transmit other infectious diseases including tick-borne encephalitis (TBE), human granulocytic anaplasmosis and rickettsiosis etc.

Lyme borreliosis is a multisystem disease that is transmitted through the bite of the Ixodes ricinus tick. It primarily affects the skin, nervous system and joints. Five species have so far been identified in Europe that are pathogenic to humans. There are no reliable figures on incidence (incidence from various surveys in Germany ranges from 60,000 to >200,000/year). The seroprevalence of Borrelia-specific antibodies is 3–20% and depends on region and age. Rates of infected ticks are area-dependent: 3–38% of adults, 5–12% of nymphs, 1% of larva.

Borrelia are transmitted through the bite of hard-bodied ticks (in Europe by the “castor bean tick” Ixodes ricinus). According to data from animal experiments, the risk of infection increases the longer blood meal. It is not possible to derive from current data the earliest point in time that an infection can occur, especially as the probability of transmission also appears to vary depending on the species []. The transmission mechanism of the Borrelia that survive in the tick’s intestine before the blood meal is very complex []. According to German studies, a seroconversion can be expected in 2.6–5.6% of those who have been bitten by a tick and disease will manifest in 0.3–1.4% [], [], []. A study conducted in western Switzerland found that the risk of becoming infected with Borrelia from a tick bite was just under 5% [].

It is very important to remove ticks early , before they have become engorged. The risk of Borrelia transmission increases with the length of time that the tick sucks []. Transmission in the first 12 hours was rarely observed in laboratory animals. After spending time in nature (garden, park, field, forest and meadows etc.) where contact with ticks is possible, the body should be checked for ticks that same evening; the head and neck of children should receive particular attention.

Ticks should be removed immediately using tick tweezers, a tick card or suitable tools in order to reduce the likelihood of a Borrelia transmission. If a suitable tool is not available, some authors recommend grasping hold of the animal between the thumb and forefinger without crushing it, then carefully pulling the animal vertically away from the skin, stretching out the skin, and then waiting up to 90 seconds until the tick releases itself. If parts of the sucking apparatus remain in the skin, they can be removed later using a needle or curettage []. If the head or the sucking apparatus is left in the skin, this poses no danger with regard to the transmission of Borrelia. The bodies of fully engorged nymphs and adult ticks should not be squeezed. Examining the removed tick for Borrelia is not recommended, as the detection of Borrelia in the tick is not sufficiently predictive of the transmission of Borrelia to the host and the development of disease. After the tick has been removed, the patient should be instructed to observe the bite site for the next 6 weeks (Appendix 6: Patient information after a tick bite in ).

According to an American study, the risk of infection can be reduced by taking a one-time, 200 mg prophylactic dose of doxycycline after a tick bite (87% efficacy) [], []. However, the results should be interpreted with caution, as only one follow-up was carried out after 6 weeks. Therefore, no conclusions can be drawn as to whether this is effective for late-stage infections. According to a meta-analysis, a single 200 mg prophylactic dose of doxycycline after a tick bite is effective (relative risk of 0.29 (95% CI: 0.14–0.60)); in contrast, neither prophylactic treatment with antibiotics over 10 days (amoxicillin, penicillin or tetracycline) nor a local prophylactic application of azithromycin were found to effective []. The authors conclude that 50 prophylactic treatments (95% CI: 25–100) are necessary to prevent one infection.

In view of the low risk of infection, a large number of unnecessary doxycycline doses would have to be administered in order to prevent one potential infection. Frequent administration of a prophylaxis could affect the intestinal flora, and the development of resistance is conceivable. For this reason, oral doxycycline prophylaxis is not recommended in Europe []. The prophylactic use of an antibiotic cream is also controversial. Animal studies with azithromycin cream show good prophylactic results [], []. A placebo-controlled study on its effectiveness in humans showed no prophylactic effect []. Therefore, this prophylactic treatment is also not recommended. Recommendations for preventing infection (Taken from the S2k guideline “Cutaneous Lyme Borreliosis”, AWMF Register No. 013/044, [].) To prevent tick bites, clothing should be worn that covers the body. The use of tick repellents can be recommended to a limited degree. After spending time outdoors where contact with ticks is possible, the skin should be checked for ticks no later than that evening. Ticks should be removed early to prevent Lyme borreliosis. The site of the bite should be observed for up to six weeks. The removed tick should not be analysed for Borrelia. Local or systemic prophylactic antibiotic treatment after a tick bite should not be carried out.

There is currently no vaccine that has been approved for use in humans. A vaccine with recombinant lipidated Osp A has been tested in the USA as part of a major study and has shown to be effective [], []. The vaccine was authorised in the USA in 1999, but was withdrawn from the market by the manufacturer in 2002. The reasons for this are not of a medical nature. Reports of adverse reactions to the vaccine in individuals with a genetic predisposition have been refuted by several qualified studies [], [], []. This monovalent vaccine is not suitable for use in Europe as it only protects against infection with B. burgdorferi sensu stricto and not against the genospecies B. afzelii and B. garinii , which are frequently found in Europe. A polyvalent OspA vaccine is currently being developed for Europe [], but approval is not expected in the foreseeable future. A 6-valent outer surface protein A vaccine is currently being investigated for efficacy, safety and tolerability as part of a phase 3 clinical trial, which is scheduled to finish in late 2024 (ClinicalTrials.gov Identifier: NCT05477524 ).

Early localised stage: An early Borrelia infection manifests in 80–90% of patients as local erythema migrans (early localised stage) [], []. General symptoms such as feeling unwell, arthralgia, myalgia, subfebrile temperatures or night sweats may occur a few days to weeks after a Borrelia infection []. Early disseminated stage: A disseminated infection can occur weeks to months after a tick bite (erythema migrans is only reported in around 25–50% of the acute cases of Lyme neuroborreliosis [], [], []). This predominantly affects the nervous system, joints and heart []. Late manifestations: In rare cases, a late or chronic manifestation can occur after months or years with involvement of the skin, nervous system and joints [], [], [], []. Information about tick bites reveals little about the time of infection, since unnoticed tick bites lead to infection in around two-thirds of all cases [], [], []. Therefore, in addition to the clinical picture, disease duration is being used more and more to classify Lyme neuroborreliosis [].

Garin-Bujadoux-Bannwarth syndrome (meningoradiculoneuritis) is the most common manifestation of acute Lyme borreliosis in adults in Europe after erythema migrans [], [], []. In Europe, isolated meningitis (without radicular symptoms) is predominantly observed in children [], [], [], [], []. The symptoms of radiculitis develop on average 4–6 weeks (maximum 1–18) after the tick bite or after the erythema migrans [], []. Segmental pain occurs first, which is worse and night and whose localisation can change. The pain is often initially localised in the extremity where the tick bite or erythema migrans had been observed [], []. The pain has a burning, piercing, stabbing or tearing nature and responds only mildly to conventional analgesics. It often peaks within a few hours or days. Three-quarters of patients develop neurological deficits after 1–4 weeks, and pareses are more frequent than sensory disorders [], []. Around 60% of patients with Bannwarth syndrome have cranial nerve deficits . All cranial nerves can be involved with the exception of the olfactory nerve.

The facial nerve is affected in over 80% of cases where there is cranial nerve involvement [], [], whereby a bilateral manifestation is observed (in around 1/3 of all cases) [], [], []. The sense of taste may not be affected. In unilateral cases, it can be difficult to differentiate from idiopathic facial nerve paresis; in some cases, however, symptoms or patient history (e.g. erythema migrans, radicular pain) can help diagnose Lyme neuroborreliosis. CSF testing can provide clarity here. Total recovery is observed in most cases within 1–2 months regardless of the severity of the facial nerve paresis. Residual symptoms or partial recovery with facial synkinesis (pathological, unintentional movements) occur in around 5–10% of patients [], [], []. Lyme neuroborreliosis can also affect the abducens nerve and very rarely the vestibulocochlear nerve, the optic nerve (optic neuritis, papilloedema), the cranial nerves (III and IV), the trigeminal nerve and the lower cranial nerves (IX–XII) [], [], [], []. It is uncertain whether isolated damage to the vestibulocochlear nerve occurs in the context of an acute Borrelia infection.

Polyneuropathy/polyneuritis as an expression of a Borrelia infection is linked to acrodermatitis chronica atrophicans (ACA) in 20% of patients in Europe []. Isolated polyneuropathies/polyneuritis without other clear symptoms of Lyme borreliosis have been described in 39–52% of American patients with Lyme borreliosis [], []. However, in 284 US patients with an aetiologically unexplained polyneuropathy, Lyme borreliosis was identified as the cause of the polyneuropathy in only one case (0.3%) following a diagnostic re-evaluation []. In contrast, there are only very few instances of polyneuropathy or polyneuritis in Europe with no link to ACA. No causal relationship can easily be made between neurological symptoms and a Borrelia infection in patients with polyneuropathy/polyneuritis whose blood has tested positive for Borrelia []. This is because Borrelia-specific antibodies are found in approx. 3–20% of healthy individuals, depending on the endemic region and age group [], [], []. Occupationally exposed risk groups, such as forestry workers, even have a seroprevalence of over 50% []. In such cases, the probability of a causal relationship depends on whether other clinical symptoms of Lyme borreliosis are present and whether other common causes of polyneuritis have been ruled out.

An involvement of the central nervous system is rare and occurs in around 3.3% of Lyme neuroborreliosis cases (95% CI 2.2–4.4%) [], [], []. Its onset is gradual and it is often chronic. The most common manifestation is myelitis with a spastic-ataxic gait and bladder dysfunction [], []. Symptoms can develop over days to several months. Some patients develop severe tetra- or paraparesis. Around 60% of patients with myelitis have additional signs of encephalitis and around 40% have cranial nerve involvement. Encephalitis has no clinical characteristics specific to the pathogen.